摘要:

Critical Review Development and Operating Characteristics of Micro Flow Injection Analysis Systems Based on Electroosmotic Flow A Review S. J. Haswell School of Chemistry, University of Hull, Hull, UK HU6 7RX Summary of Contents Introduction Fabrication Pump Design Injector Design Reactor Design Detector Design Future Trends Conclusion References Keywords: Micro flow injection; electroosmotic flow; miniaturisation; micro total analytical system ( mTAS) Introduction Over the past 50 years, the design of instrumentation for the measurement of chemical, biological and physical parameters has brought analytical chemistry to a highly automated and technologically advanced science.It would be naive, however, to believe that the chemistries and physics associated with present day analytical measurements, are in any way limited by current technology. On the contrary, laboratory based analysis has been almost exclusively designed, for the ergonomic needs of the human worker.There can be little doubt that this preoccupation with the physical size and operation of equipment, for laboratory based measurements, has significantly influenced the teaching and practice of analytical chemistry. If one considers the fundamental chemical and physical phenomena on which measurements are based, then it is apparent that one only requires a few molecules or atoms to be present for quantitative measurement to occur. In short, the physics and engineering of measurement science can be considered as oversized and in many cases over engineered to meet the needs of the chemical measurement.Whilst instrumental design has relentlessly moved towards automation, the realization that the quality of analytical data can be profoundly influenced by factors such as sampling, sample storage, sample pretreatment and matrix interferences has led to the concept of the total analytical approach. This more holistic view of analysis, in which traceability and uncertainty prediction can be carried out, is often embodied in the concept of the ‘total analytical system’ (TAS).1 There are, however, numerous disadvantages to TAS, including slow sample processing, lack of selectivity and high reagent consumption;2 however, many such problems may be overcome through system miniaturisation.The term micro TAS (mTAS) was first used at the Transducers 89 Conference.3 In its simples sense, mTAS involves the miniaturisation of all the functions found in an analytical method for example, pumps, valves, flow manifolds, mixing and reaction chambers, phase and analyte separation, detectors, control and communication electronics.One of the more exciting prospects of the mTAS concept is the suggestion that the entire chemical measurement laboratory could be miniaturised onto a device of a few square centimetres.4 This type of miniaturisation has become possible largely through the adoption of microfabrication techniques developed in general by the microelectronics industry, and one can consider miniaturisation of chemical reactors and their corresponding instrumental requirements to be in a similar position to the microelectronics industry 30 years ago.Interestingly, it was from the heart of the microelectronics industry that Terry et al.5 were one of the first to demonstrate that integrated circuit (IC) technology could be used to fabricate a miniature GC instrument using a 5 in silicon wafer.Although the work of Terry and co-workers served to point the way, it was not until the advent of capillary electrophoresis (CE)1,6–8 and in particular, the exploitation of electroosmotic flow (EOF), that the realization of mTAS came about some 10 years after the miniature GC work was first reported. Stephen J. Haswell is a Senior Lecturer in Analytical Chemistry at the University of Hull, UK. Following a period of five years working in the food and plastics industries, he undertook a Ph.D.in the area of atomic spectroscopy graduating in 1983 from the University of Plymouth. His current research activities are in the areas of mFIA, microwave enhanced reaction chemistry, trace elemental speciation and chemometrics. He is author of over 100 research papers, books and patents and presenter of numerous national and international lectures on his research activities. The subject of this present review has become a great interest of the author over the past five years.Analyst, January 1997, Vol. 122(1R–10R) 1RA significant research base has now built up in the area of mTAS9–14 and the development of mFIA over the past 5 years can be attributed to key research carried out by groups based in Basle, Switzerland,14 Texas Tech. University, USA,15 University of Alberta, Canada,16 the Oak Ridge National Laboratory, Tennessee, USA,17 and in a very modest way in the author’s own laboratory.18 What follows is a review of the features present in a relatively small, but rapidly expanding research base, which can be exploited to develop micro flow injection analysis (mFIA) systems, based on electroosmotic flow, to produce reliable sensor-type devices which incorporate mechanical and chemical robustness.Fabrication The fabrication of mFIA systems has attracted various approaches in recent years, ranging from manifolds based around CE-type fused-silica capillaries,1 to modular micro pump systems.19,20 This particular section, however, will focus specifically on fabrication methodology which leads to the production of a single integrated device, designed to use electroosmotic mobility for sample and reagent pumping.18 The discussion will focus almost exclusively on micro machined monolithic device fabrication, which adopts standard IC photolithographic, wet etching and bonding techniques to produce a planar structure.21,22 As the techniques used are based essentially on IC technology, they represent well established methodology, outlined schematically in Fig. 1. If one considers the current ‘state-of-the-art’ in IC fabrication, then it is clear that the challenge of producing mFIA manifolds can in no way be considered as technologically demanding. From what follows, it will become clear that mFIA device fabrication can be relatively simple and offer a reasonably high degree of flexibility in systems design. The choice of a suitable substrate into which the channels of a mFIA manifold can be etched is closely related to factors such as the fabrication method, the analytical chemistries involved and the proposed pumping, injection and detection systems.The obvious candidate given the close relationship with IC technology is silicon,23 which is amenable to the fabrication of structures in the nano- and micro-metre range. This crystalline material can be obtained in a very pure form, at relatively low cost, and offers good mechanical and chemical properties.Whilst silicon may seem an ideal microengineering candidate for mFIA fabrication, it is by nature a conductor and so surface modifications will be required if electroosmotic pumping of samples and reagents is to be achieved. This can be carried out by coating the silicon with SiO2 or Si3N4 to produce the required surface chemistries. Harrison et al.16 described the fabrication of such a manifold, in which the properties of silicon and oxide/nitride-modified layered surfaces were evaluated.They reported that the operating voltage achievable with such devices will be limited by the quality of the oxide insulating film produced. Nevertheless, the results are sufficient to demonstrate that silicon-based devices using a sandwich structure (i.e., oxide–nitride–oxide) can sustain potentials in the range 400–1200 V before dielectric breakdown, which is certainly sufficient to generate EOF mobility for mFIA systems. Although silicon does offer the attraction of producing high precision engineered mFIA manifolds, with well characterised surfaces and the potential to integrate control and detection on to one substrate, other materials such as glass24 and silica (quartz)25 are also amenable to IC based fabrication techniques and will produce surfaces eminently suitable for EOF pumping.Judging from the reports in the literature, glass and silica have proved to be the most popular materials for device fabrication to date, but metal, plastic and ceramic substrates are also possible candidates for the manufacture of mFIA systems.26,27 In general, three basic concepts are required for the preparation of mFIA systems: (i) a suitably prepared substrate, (ii) photolithographic equipment and (iii) wet etching and bonding facilities.28 Prepared substrates based on glass, silica and silicon can be obtained commercially from photomask producers (e.g., Alignrite, Wales, UK); these come ready coated with a metal film (0.1 mm) such as chromium over which is spun a positive photoresist layer (0.7 mm). These plates, which are typically 152.4 mm square (5 3 5 in) and 3 mm thick, can be used directly for manifold pattern transfer using photolithographic methods. A mask or negative of the final channel patterns required for the mFIA system can be produced using CAD computer software.29,30 The mask pattern can be transferred to the photoresist film on the substrate, using basic photographic development equipment.For channels larger than 1 mm, visible or UV light can be used to transfer the pattern from the mask to the photoresist (exposure times being approximately 5 ms); however, to obtain good line definition and submicro patterning, X-ray or electron-beam (e-beam) photolithography will be required. Where facilities permit, e-beam photolithography offers at present one of the most flexible ways of producing mFIA manifold designs; however, for most applications simple basic photographic equipment will suffice.Once the pattern has been transferred to the photoresist film on the substrate, the plate can be developed and wet etched. If one uses the large plates as described, then multiple devices can be prepared; in our laboratory, for example, 25 manifolds of various geometries and channel widths are produced from one plate (Fig. 2).18 Workers in the field tend to prepare their own plates rather than using a commercial source and, although this adds additional steps to the fabrication process, it is not too technically demanding.29,30 As indicated previously, these are typically made from silica or glass on to which a thin metal film (0.05–0.1 mm) of chromium, gold or a combination is produced, using sputtering or chemical vapour deposition (CVD) methods.The metal film is then spin coated with a layer of positive photoresist (approximately 0.5–2 mm). Fig. 1 Schematic diagram of the steps involved in the anistropic etching technique for the fabrication of mFIA devices. The arrow in Step 1 represents radiation. 2R Analyst, January 1997, Vol. 122Following the photolithographic transfer of the manifold pattern, the exposed photoresist and corresponding metal film are removed using commercially available reagents. The prepared plates may then be placed in an oven at 120 °C for 72 h to harden the photoresist and remove solvent residues from the plate.18 It should be noted that although the presence of a metal film on the glass is important for controlling the degree of surface etching that will occur, the exposed substrate will experience undercutting or sideways etching, which will proceed at a rate of approximately 2 : 1.Thus, as the etch goes down to 1 mm it spreads laterally 2 mm at the edges, producing channels, for which the width always exceeds the depth.31 Clearly, for a given substrate, the choice of the initial photomask line width and the time of etch will influence the size of the final channels produced.Various reagents have been used for the wet etching of the substrate, but in general for glass and silica, hot (70 °C) dilute HF–NH4F (1% HF + 5% NH4F in water)18,29 will give an etch rate of approximately 0.3–0.5 mm min21, whereas dilute HF– HNO3 produces an etch rate of between 0.5 and 0.8 mm min21 in Pyrex.30 Silicon etching requires the use of reagents such as ethylenediaminepyrocatechol (115 °C, etch rate 2.5 mm min21) or KOH.16 Although indicative figures for etch rates can be obtained,31,32 it is strongly recommended that sample strips of the substrate are tested in the selected etchant to establish the rate of etching for a given system.It should be stressed that the final size of the mFIA channels will depend on the initial width of the lines on the photomask, the substrate material and mode of etching. During the etching process, agitation of the substrate or etchant is advisable otherwise asymmetric channel formation, i.e., uneven etching on the bottom edge or side of the channel, may occur.It is usual, however, to obtain a channel profile that is wider at the top than the bottom. This effect, associated with undercutting, will vary as a function of the etch time, substrate type and reagents used. The surface quality of the etch is generally related to the types of material used. Good quality silica, for example, gives a well defined etch, whereas Pyrex or borosilicate glass can produce a rougher surface owing to the crystalline structure of the material.The real effect or significance of the surface properties in mFIA channels has not yet been fully characterised, but clearly the more controlled the etch, the more precise and smaller will be the channels that one can produce. What is not apparent from experimental results is the real influence that surface topography may have on the mobility and reactivity of reagents in a mFIA system.Rough or poorly defined surfaces and intersections, of one or more channel, will increase the effect of turbulence and in turn promote dispersion, which in some instances may prove to be an advantage where mixing is required. This factor is particularly important in capillary systems where the Reynolds number is lower than the transitional values of 2000 or 2300 indicating laminar flow, thus minimal mixing will dominate.33–35 However, research indicates that at low Reynolds numbers microfluidic mixing cannot be simply classified as a laminar or turbulent model.36 What is clear is that as channel sizes become smaller, surface effects will inevitably become more significant, which in turn will have an impact on fabrication and material specifications.Manz and Simon37 demonstrated that for a 10-fold decrease in channel size, a 1000-fold decrease in reagent consumption and a 100-fold decrease in related time variables would be obtained. Pressure requirements of such a system will, however, increase by a factor of 100, but as indicated later this does not effect the voltage requirements for EOF.Once an etched base has been produced, the top of the channels need to be sealed. Various bonding methods have been suggested, including glueing, low temperature bonding or annealing, high temperature fusion and anodic bonding.38 The use of an adhesive in such systems can pose problems due to channel plugging and, although this may be overcome to some extent by using photosensitive dry films laminated on to the substrate surface as a protective coating,39 the technique has found little real use in mFIA.Thermal bonding or annealing of top plates represents one of the simplest and therefore most widely used methods in device fabrication. These methods usually involve heating the substrate and top plate, which may be under slight pressure,30 in an oven or furnace to near the upper annealing temperature.For glass and silica this is usually between 500 and 600 °C, whereas silicon will require higher temperatures of around 900 °C. The period of heating varies between 48 and 96 h, after which the device is slowly cooled (48–72 h) to minimize physical stress. For glass substrates, hydrolysing the surface with dilute NH3–H2O2 followed by heating at 500 °C for 24 h29 or 575 °C for 96 h18 has been demonstrated to achieve good reliable bonding.Slightly more complex temperature programmes have been suggested, but these are essentially slight modifications to the same basic method.30 In order to keep the fabrication of a mFIA manifold as simple as possible and to increase the bonding success rate, it is advisable to use the same material (i.e., similar thermal expansion coefficients) for both the base and top of the device, so minimizing stress features. The fabrication of multilayer devices can become complex and techniques using intermediate layers have been described in which more complex structures such as pumps and valves are required.40,41 One technique which has found particular favour with workers fabricating microvalves, micropumps and silicon devices generally is anodic bonding.42,43 In this process, ions such as sodium and oxygen are thought to migrate at elevated temperatures to electrodes placed on the outer surface of one of the layers, so increasing the electrostatic surface charge.The temperature and applied voltages used will depend on the materials in question, but for glass these are found to be 200–400 °C at 30–300 V and for silica 700–800 °C at 30 V applied for a period of 45 min. It might be necessary when employing anodic bonding to treat the surface with HF prior to bonding. Holes or ports made in the top plate are commonly used as reagent or sample feeds into the manifold and can be generated either before18 or after29,30 the bonding process.The holes can simply be produced by using mechanical18 or ultrasonic16 drilling of the substrate and vary in diameter from 0.5–2 mm. At present, laser ablation, which offers an attractive method for hole production, has not been employed, but no doubt the technique will find its way into the literature before long. Plastic Fig. 2 Example of the photomask design used to produce five mFIA manifolds designs with variable line or channel widths.The line widths shown are A and E, 50, B, 30 and C, 10 mm. Analyst, January 1997, Vol. 122 3Rreservoirs are usually glued to hold liquids and support the platinum electrodes when required. A photograph of a completed device used for phosphate and nitrite determination is shown in Fig. 3. To date, only planar mFIA systems have been reported based on the fabrication techniques described. The fabrication of three dimensional systems is, however, an attractive prospect and no doubt stacked systems will be produced as more sophisticated chemistries are exploited.Such systems based on micropumps have been reported in which multi-layers or modules are clamped or stuck together to produce a complete device.19,20 Once a complete mFIA manifold has been fabricated for use with EOF pumping, the success of the bonding process can be checked by filling the channels with a buffer44 or weak acid18 and plotting the current–voltage relationship.This plot is generally found to be linear up to 10 kV in glass and silica, after which point dielectric breakdown of the substrate occurs. Any deviation in linearity at lower applied voltages is indicative of a device failure, associated with incomplete bonding. Pump Design In order to achieve reliable flow dynamics in an FIA manifold, which may contain numerous reagent streams, pulse free, constant flow characteristics must be maintained. It is not surprising therefore, to find that mFIA systems also call for pulse free, variable nl min21–ml min21 flow control. Although a direct syringe or piston pump can be used with mFIA systems,45,46 the closest equivalent to a conventional peristaltic FIA pump is the so called micropump, which usually takes the form of a pulsating one-way valve driven typically by a piezoelectric device.36,45,47–50 Based on microengineering technology, micropumps have been employed in mFIA manifolds consisting of channels 100 mm wide, 10 mm deep and 4 cm in total length.47 When such small channel dimensions are employed, the hydrodynamics of the system can produce a pressure drop of 0.7 atm at a flow rate of 0.5 ml min21.However, diaphragm devices, commonly using microchemical silicon membranes, are only reliable up to backpressures of 0.2 atm and so high pressure pumps or large channel dimensions are therefore required. High pressure pump designs using nickel rather than silicon check values are now being developed using low temperature bonding techniques, incorporating intermediate photoresist layers to adhere the valve to the body of the pump.47 These devices, which use a piezoelectric disc glued to the outer surface of the pump housing, operate at many hundred Hertz, requiring around 300 V to produce flow rates in the range 20–300 ml min21.In an elaborate example, four such pumps have been used to drive reagents through a three-dimensional system constructed from silicon and glass at a flow rate of 1 ml min21, the channel dimensions being 600 mm wide and 200 mm deep in order to accommodate pressure effects.19 Diaphragm micropumps do have an important role to play in applications where electroosmotic pumping is not applicable, owing to sample or reagent chemistries.It is essential, however, that micropumps remain chemically inert to the reagents and samples with which they may come into contact. In addition to piezoelectrically driven pumps, ultrasonic51 and other more exotic electrically activated pumps have also been described.36,52–54 Electrophoresis embodies two basic electrokinetic components, electroosmotic and electrophoretic flow.These two components combine to give the total flow or velocity (n) in the following way: total flow rate = n = (meo + mep)E (1) where meo = electroosmotic mobility, mep = electrophoretic mobility and E = applied electric field. At relatively low electric field strengths, electroosmotic mobility represents the larger component of the two processes; however, as the field strengths are increased the influence of the electrophoretic component (migration of analyte ions) increases also.For mFIA, where bulk mobility is desired, one finds that field strengths around 300–400 V cm21 are usually adequate; however, if separation is required, as in CE, then field strengths greater than 1 kV cm21 are typically required. It is common for mFIA systems based on EOF to operate with field strengths of the order of 80–300 V cm21, at which little or no analyte separation is observed The generation of an EOF to pump reagents and samples through a mFIA system is subject, however, to certain physico-chemical limitations.Firstly, in order to generate the EOF one must use a material which will yield negatively charged groups on the surface or walls of the channels when placed in contact with an appropriate liquid. Secondly, the liquid phase must dissociate to some extent in order to generate counter positive ions (notably H+ ions).The combination of the negatively charged surface (typically SiO2) and the H+ ions in solution will form a diffuse double layer (sometimes referred to as the Gouy or Helmholtz layer). This diffuse double layer acts as a parallel-plate electric capacitor whose plates are d cm apart each carrying a charge e per cm2. The zeta potential (x) will be the potential difference between the plates, given by the general equation x = 4ped/er (2) where er is the relative permittivity of the medium between the hypothetical plates. If an electrical field is now applied through the liquid phase, ions will migrate to their respective electrodes dominated by the positive ions (H+) moving to the negative or ground electrode.As the ions move, they induce a drag on the bulk of the liquid, which in turn results in a corresponding net transfer of the solution to the negative electrode.Thus, in EOF the mobility of the solution is from the anode (positive electrode) to the cathode (negative electrode) or ground. In practice, the formation of the double layer is limited to the pH range 4–10. At lower pH values the cationic population becomes so high that the EOF is overrun by conductive flow and at pH values greater than 10 the cation population becomes too low to sustain the double layer. The zeta potential that is generated when the double layer forms will be influenced by changes in pH and ionic strength of the solutions in the channel, and this can affect the corresponding flow rate.Clearly, as one of the main attributes of a pump in FIA is to maintain a reproducible flow rate, the pH and ionic strength of the reagents Fig. 3 A fully assembled mFIA device showing the mounted plastic reservoirs which act as wells for the reagents and samples and support the platinum electrodes required for EOF pumping. The device is shown mounted in a rig which allows fibre optics (seen either side of the block) to be coupled into relevant sections or channels of the manifold. 4R Analyst, January 1997, Vol. 122used in a mFIA manifold need to be controlled, through the use of buffer systems. As the ionic strength increases, for example, a counter-ion effect will prevent the ions from migrating through the solution independent of the EOF. In general, it is therefore preferable to keep the total ionic strength of reagents and samples as low as possible.Once an EOF has been generated, the flow establishes a flat, trapezoidal profile notably different from the parabolic (bullet) shape commonly associated with conventional FIA. Closer investigation of the mFIA profile reveals that the edges of the profile (i.e., closest to the wall surface) are slightly in advance of the bulk owing to the drag effect and the extent of this effect will be a function of the solution viscosity.55 The fact that the flow profile, which offers minimal band broadening, does not contain the parabolic flow characteristics of a pressure or hydrodynamically pumped system, does have implications with respect to the mixing of reagents in the mFIA reactor, where some dispersion will be necessary to achieve the desired chemical reactions.Thus EOF-pumped mFIA can be considered to be more akin to segmented flow analysis rather than conventional FIA, in its flow characteristics.In an EOF-pumped mFIA system, the flow rate can be effectively controlled by varying the applied voltage as follows: n flow rate v = m (3) L where n = applied voltage (V), L = length of the channel (m) and m = sum of the electroosmotic and electrophoretic mobility component under different conditions of pH and ionic strength. The EOF and hence the flow rate in mFIA, will be unaffected by capillary diameter up to channel sizes approaching 250 mm, after which drag effects of the bulk solution may cause serious disruption to the EOF.The EOF produced in a channel can be considered, in the electrical sense, as a resistor, thus Joule heat will occur in the capillary. However, the almost negligible bulk properties of components in a channel, relative to the substrate, will ensure an effective dissipation of heat, which rarely poses a serious problem in mFIA. If, however, the resistance is allowed to increase, for example when a highly viscous or an immobile solvent is present in a channel, heating can occur.It should be stressed that an EOF can only be established if the double layer is formed, and this requires ions or dipoles to be present in the liquid stream. In the case, for example, where organic compounds or solvents may be present, such conditions may not be met. For example Zheng and Dasgupta55 described some initial studies, using silica capillaries, to evaluate the suitability of an EOF in a mFIA system for carrying out in-line analyte phase separation or solvent extraction.They described the use of a 50 cm 3 7.5 mm id polyimide coated silica capillary to investigate various aqueous ionic complexes, based on well characterised cationic, anionic and neutral ion-pair chemistries. A quaternary ammonium salt, tetrabutylammonium perchlorate, was added to chloroform as a supporting electrolyte to assist in the mobilization process56 and to catalyse phase transfer.57 The results clearly indicated that organic solvent pumping was possible, when modified with the quaternary ammonium salt, but that migration of the ammonium ion occurred, creating a positive front end to the solvent slug.Bleeding of ions into the secondary aqueous buffer from the organic phase was also reported. Furthermore, they suggest that a thin interfacial layer of buffer is generated between the organic solvent and the capillary walls which enables the EOF to be generated. Investigations into solvent extraction indicated that the ion-pair complexes studied were either extracted into the organic phase or accommodated in the aqueous phase ahead of the organic slug interface.Not only were the findings of Zheng and Dasgupta a significant contribution in terms of demonstrating phase extraction and organic solvent mobility using EOF in a mFIA system, but they also clearly indicate that the migration of ions within a solvent under the influence of an electric field could lead to the development of gradient and separation techniques, complementary to CE, such as selective solvent extractions and matrix modification. Interestingly, the use of micellar electrokinetic capillary chromatography, described by Moore et al.,58 suggests that multiphase systems could be developed for mFIA applications, thus offering significant advantages for organic solvent based chemistries.Although it is possible to use EOF as the primary pumping mechanism in a mFIA manifold, it may be preferable or even necessary to isolate the pumping mechanism from the injector, reactor and detector components of the system.Such occasions might be, for example, when the chemistry of the method or detector system is not compatible with the electrical field required for direct EOF pumping. One such system has been described by Dasgupta and Liu45 in which a section of polyimide-coated fused-silica tubing (40 cm 3 75 mm id) was connected to a second capillary, via an isolating membrane. In the first channel a 2 mm borax buffer is pumped by direct EOF, with a field strength of around 40 V cm21.As the flow in the pumped capillary was electrically isolated from the second capillary it produced a hydrodynamic effect, sufficient to create a flow in the second channel, which could be varied between 1 nl min21 and 100 ml min21. In this case the hydrodynamic effect of the EOF is exploited in a mFIA manifold without the need for a direct electrical field. Injector Design The introduction or injection of a sample into an FIA manifold can generally be classified into two general types.The first is the timed or gated injection in which a sample is drawn into the FIA manifold, usually through a sampling probe, for a controlled period of time, after which the flow is switched back to the carrier stream. In this way, a variable volume can be injected into a manifold as a function of time. The second and more usual method of injection in FIA, is the introduction of a constant sample or reagent volume into a mobile carrier in a way that affords minimal flow dispersion.The most common approach for such systems is to use a rotary or slide valve, which contains a sample loop of a defined volume. It would therefore seem appropriate to have both these forms of sample injection available in mFIA systems, and this is indeed found to be the case. The injection of a sample into a mFIA manifold can be performed using hydrodynamic/pneumatic pressure control,46 miniaturised valves1 or electrokinetic mobility based on EOF.15,18,59–62 Even the lowest volumes obtainable with traditionally based rotary-type valves63 are clearly too large for practical use in mFIA.However, Liu and Dasgupta1 recently described the use of commercially available valves with an injection volume of 60 nl. It should be stressed, however, that most of the work relating to the introduction of small sample volumes ( < 20 nl) into capillary systems has been focused on CE methodology rather than addressing mFIA systems.Of particular relevance to mFIA systems is the development of the so-called valveless injection method, which uses EOF control in conjunction with specific capillary channel geometries.64,65 The two simplest approaches to sample injection based on EOF are first to pump a sample for a given period of time into a flow channel and second to fill a defined volume (equivalent of a sample loop) built into the mFIA manifold.In the former case, Zheng and Dasgupta55 described a modified CE system in which an organic solvent was loaded for a given period of time into a mFIA system consisting of a 50 cm 3 75 mm id polyimide Analyst, January 1997, Vol. 122 5Rcoated fused-silica capillary. The solvent was introduced into the system by placing the capillary electrode in the solvent reservoir at 15 kV for 10 s. Following the injection step, the capillary electrode was switched to a borate buffer which subsequently pumped the solvent through the capillary to the detector.Using this approach, the authors reported an RSD of 0.51% associated with migration effects and 1.7% for the peak, suggesting that surface tension, viscosity and flow rate will influence this particular mode of injection. It should be noted that in this work, the authors were using the method described for introducing a solvent into the mFIA system for the purpose of solvent extraction.The second approach to sample introduction, based on defined volumes using EOF, falls into two categories, the Xand Z-type injections (Fig. 4). Of the two injection geometries, the X or cross design, has been most widely investigated. Depending on the geometry and electrical field used, the sample injection can be classified as ‘floating’ (gated) or ‘pinched’ (discrete), with the former occasionally being referred to as the continuous mode.59,60,64,65 Experimentally, the simplest of these two modes is the ‘floating’ method, as it only requires one pair of electrodes (Fig. 5). In this case the sample is pumped by an EOF from the sample reservoir (A the positive electrode) to the sample waste reservoir (B the negative electrode) along channel AB crossing part of channel CD. Note that the channel AB will have been filled prior to the sample introduction with a suitable buffer. The reservoirs C and D contain no electrodes and therefore have no applied field, hence their potential is floating relative to the field in channel AB.As the sample passes across the intercept of AB and CD, diffusion and eddy effects will allow some of the sample to migrate in the direction of both C and D [Fig. 5(a)]. Subsequently, on placing the electrodes in reservoirs C and D the field can be made to run in the CD direction, so any sample molecules in the intercept will be pumped towards D, the grounded reservoir [Fig. 5(b)], allowing the sample to pass via a detector on its way. Clearly, we can see that the leakage or migration of the sample solution into channel CD during injection and the possibility of dragging the stationary sample from channels A and B into D as a function of flow [Fig. 5(c)] will lead to an uncontrolled volume injection. Examples of this effect have been reported.59,60,64 The leakage observed into the main channel may not be a serious problem for certain applications and clearly it offers a simple method of operation, in which only one pair of electrodes are required. However, the quality of etch, surface topography and geometry of the channels concerned will influence the degree of diffusion using the ‘floating’ injection method.The need, especially in CE, to have more precise control over the sample volume injected has lead to the development of the so-called ‘pinch’ method (Fig. 5). In this case, the sample is once again pumped under an EOF from reservoir A to B, but this time reservoirs C and D are not electrically floating, but like A are given a positive potential, relative to the waste reservoir B [Fig. 5(d)]. The effect of this is to draw buffer from channels C and D into B along with the sample from A. Under these flow dynamics, the sample gains a pinched or trapezoidal shape at the intercept of channels AB and CD. When the flow is redirected towards D [Fig. 5(e)], the sample volume, which may be only a few picolitres, is of a more precisely defined volume, less affected by diffusion effects.Jacobson et al.60 have reported improvements in the RSD from 2.7 to 1.7% for a 90 pl injection volume using the ‘pinch’ method, compared with the ‘floating’ approach. In this particular study, the applied voltages based on a 1 kV power supply were reservoir A 90, B 0, C 90 and D 100%, giving channel field strengths of 270, 20, 400 and 690 V cm21 in A, B, C and D, respectively.Clearly, operating the manifold in a multi-electrode configuration will reduce dispersion effects at intersections and allow improved control of the sample injection volume. The concept of the ‘push-back distance’ associated with surface diffusion has been used66 to define the migration rate and shape of flow patterns at X-type intercepts. This approach to sample injection has considerable potential in controlling sample volumes as a function of applied field. The larger volumes (9–22 nl) of the Z-type injection (Fig. 6) are introduced by pumping the sample between two reservoirs across a defined volume in a Z geometry. In the example shown, the sample is pumped from reservoir C or D to the waste reservoir D or E, so filling a defined volume in channel AX. The loaded sample can then be reacted with reagents in channel AX and moved to a detector or be monitored in situ. The Z mode of injection produces a larger volume than that obtained by the X mode, which in turn will be less influenced by the diffusional effects described previously.18,67 Recent work in the author’s laboratory using the Z injection technique has indicated that Fig. 4 Discrete volume injection using (a) the X and (b) the Z approach. Fig. 5 Flow and dispersion characteristics of floating and pinched modes of injection. 6R Analyst, January 1997, Vol. 122pinch control offers no improvement in precision over the floating method.68 The ability to matrix modify a sample in an injector system could be an attractive option in mFIA and techniques such as synchronized cyclic capillary electrophoresis22,69 or sample fractionation67 may offer considerable scope for such methodology.In these systems selected, analyte fractions can be separated from a more complex and possibly interfering matrix, owing to the variation in migration and flow direction under electrophoretic conditions. The switching of electrodes can be used to move components along channels and across channel intersections akin to shunting railway trucks, until the analytes required become isolated from matrix or interfering components. This aspect of sample manipulation is potentially very exciting for mFIA system development and will no doubt be exploited in future applications.Reactor Design Having satisfactorily injected a sample into an FIA system, the next objective is usually to present the sample to a detector in an appropriate form as quickly as possible, using tube or channel geometries that minimize the dispersion of the sample slug.In mFIA systems sample diffusion is known to be a function of the square root of the time after the injection of a sample.69 There remains one additional and often complex step, however, that of achieving the appropriate chemistry or biochemistry necessary for the detection of the analyte of interest. This process typically requires the addition of at least one reagent and can include more complex steps such as in-line solid-phase extraction and multiphase separations.70 Indeed, the complexity of the FIA manifold is only limited by the imagination or ingenuity of the analyst.The first and perhaps the most important step in reactor design is a consideration of the physical and chemical characteristics that occur when two solutions combine at a channel junction in a mFIA manifold. From the previous section, it is clear that even at Reynolds numbers less than 2000 some turbulent mixing takes place at channel intercepts and that the electrical potentials at which the channels are held can significantly influence the interfacial or mixing zone. Turbulent mixing, for example, has been reported in channels 5.2 mm deep and 57 mm wide.71 The characterization of channel intersections indicates that leakage or diffusion of reagents from a ‘floating’ side stream into a channel in which EOF is present is around 2–5%, depending on the viscosity and flow rates.64 This bleed occurs as a result of hydrodynamic effects, in which the molecules in the pumped stream entrain the stationary side channel molecules owing to frictional and eddy properties in a Venturi- or Bernoulli-type effect.What is interesting with such systems is that the flow characteristics in the main channel will have a flat, trapezoidal profile whereas the side channel, whose flow will be pressure driven, will have a parabolic profile.Hence the profiles of the merging streams may have a significant effect on the mixing characteristics at such an interface. As one of the main objectives of reagent addition in an FIA system is to induce mixing and so achieve the required reaction, some consideration needs to be given to this aspect of pumping in mFIA systems. In a valuable study, Seiler et al.44 demonstrated that the individual electrical resistances in a series of interconnecting channels can be used to predict the flow characteristics, in a similar way to hydrodynamic estimations in conventional FIA.The basic concept considers the intercepting channels as electrical resistors and evokes Kirchhoff’s rules,72 to predict the net flow of current which can be attributed to the flow dynamics of the device. One of the important findings of the work by Seiler et al.44 is that by controlling channel voltages one is able to manipulate sample or buffer dilutions by adjusting flow rates, and this offers considerable scope for stopped-flow or reverseflow operations. This effect was also recently demonstrated in our laboratory using a manifold previously described for phosphate analysis.18 Figure 6(a) illustrates how a sample of orthophosphate, held in reservoir D, is injected into channel AX using the Z technique, by pumping the sample to reservoir E.Thus the slug in channel AX defined by the intercepts DE represents the injected sample which reacts with ammonium heptamolybdate (0.01 m) in the presence of ascorbic acid (0.05 m) to produce molybdenum blue.Detection of the molybdenum blue is achieved by measuring the absorbance at 744 nm in a spectrophotometer coupled to the manifold by a fibre-optic system attached along channel AX (optical path 2 cm). Through the selection of different injection volumes [i.e., C–D (9 nl), D– E (13 nl) and C–E (22 nl)] and variation in the positive voltage applied at reservoirs A and B relative to D or C, the absorbance signal was observed to decrease as the intercept ‘pinch’ potential applied across A–E and B–E became sufficient to dilute the formation of the coloured complex at each end of the injection slug.68 In effect, the flow rate in channel A–B was being increased as a function of voltage, so diluting the flow from channels D–E, C–E or C–D into channel AX.Results obtained using this approach (Fig. 7) indicated that calibration based on one standard is possible and that the technique will have an important role to play in the development of future methodology.Following the mixing of the reagent and analyte streams, some period of time is usually required to produce sufficient product, either for further reactions or for subsequent detection. Conventionally, this hold time is effected by using a coil or knotted reactor and/or a stopped-flow mode, in order to minimize dispersion whilst achieving the required reaction chemistries.In mFIA it may be that owing to the more efficient interfacial nature of the mixing, i.e., the need for less bulk mixing, reactions rates may increase; however, reactions are still likely to require a finite period of time for product production. A serpentine-type pattern or structure offers a very simple, but effective, use of space for extending channel length and is compatible with photolithiographic fabrication techniques. The flow characteristics of serpentine channels have been studied and whilst disruption to the electroosmotic flow Fig. 6 Controlled dilution experiments using a pinch voltage. Analyst, January 1997, Vol. 122 7Rwas observed at the 90° bends, no significant band broadening has been found.60 The particular design used had a total channel length of 17 cm operating with a field strength of 700 V cm21 and was used to perform electrophoretic separations. It would seem that using such an approach, time-dependent mFIA reactions could be accommodated in a physically small area with minimal dispersion through the use of serpentine or spiral channels.61 Modification to the surface properties to induce selective flow characteristics, given the limitations of producing a double layer and hence EOF, has attracted some interesting approaches. These include the use of reversed-phase hydrophobic polymers, 73 surfactants,74 silanals and quaternary ammonia groups.75 In addition, the covalent bonding or immobilisation of glucose oxidase has been used to develop a glucose mFIA method based on a serpentine geometry using a 260 cm long 3 100 mm wide 3 70 mm diameter reactor.76 One of the interesting extensions of the mFIA technology described is the development of electrochromatographic (EC) separations in which capillaries are packed with small particles (approximately 3 mm) on which efficient separation can be achieved.77 Using EOF as the primary pumping mechanism, EC can be mediated with reference to the zeta potential and hence separation is potentially possible through selection of the appropriate surface properties of a packing material.Thus the walls of the capillary act to facilitate the primary pumping mechanism and the packed chromatographic material offers enhanced electrophoretic separations. In addition to surface modification, the physical effect of field flow fractionation may also be incorporated as a separation process in capillary systems.78 In this case, an external force, e.g., magnetism or gravity, is used to pull fractions to the wall of a channel; on removing the force, a gradual diffusion of the fractions in the sample occurs back into the flowing stream, usually based on molecular size.79,80 As yet field flow fractionation has not been widely studied in mFIA systems, but the possibility of using field flow fractionation and modifying the zeta potential using a magnetic field might be an interesting subject for future research. Detector Design Much of the research reported to date in the area of mFIA has focused on characterising the physical processes of reagent mobilisation and mixing in microchannel manifolds.Such work has relied heavily on imaging techniques, using, for example, charged coupled device (CCD) cameras60 and microscopybased techniques.30,44,61,81 In this section, consideration is given to the design of detectors suitable for direct integration into EOF based mFIA systems. The major detector systems used in conventional FIA are based on optical absorption/emission and electrochemical techniques,70 and this is also found to be the case for mFIA systems.82,83 Clearly, if electroosmotic based pumping is employed in the mFIA manifold, then some care is required with the design of electrochemical detectors, but this does not pose any real serious limitations.83 Not surprisingly, optical detection has proved to be the most attractive approach to on-device detection and various examples can be found in the literature,15,40,45,67 mainly associated with CE detectors.84 One of the less obvious advantages of miniaturisation is the ability to introduce detector systems not readily available to conventional larger flow systems, such as mass-selective devices of the surface acoustic wave type.85 Clearly, the design of a flow cell in terms of physical volume and optical geometry is very important in miniature systems if sensitivity, reliability and robustness are to be achieved. The incorporation of fibre optics as part of an optical detection system has proved to be of value, particularly with the advent of fibres with diameters of the order of 0.1 mm.86 In the area of optical detectors, some effort has gone into developing axially rather than perpendicularly oriented measurement cells.87 Increasing the volume of the flow cell in a perpendicular viewing axis has been used to increase the optical pathlength,88 whilst the use of multiple reflections axially in a flow cell has also been evaluated.89 The first example led to severe dispersion effects and a subsequent loss in sensitivity and the second approach, based on a silicon device, suffered from signal loss due to scatter and absorption of light at the silicon surface.One interesting approach to absorbance measurements is to consider the mFIA channel as an extension of a fibre optic system.18,90 In this way, total internal reflection of light may be achieved and any photoactive species spatially present in the channel will undergo interaction with the photons present, to produce either a direct absorption measurement or a subsequent fluorescence effect.As with most absorption methods, it would be preferable to use a dual-channel system to improve stability and reduce scatter. Although a long pathlength is appealing for absorption measurements, emission-based techniques would benefit from a small spatial volumes in which the emission effect can be concentrated, and volumes less than 1 nl have been suggested.89 The size of the detector cell is clearly related to the concentration and sensitivity of the method in question and a flow cell of 15 ml has been reported to be adequate for chemiluminescent measurements of glucose and lactate in human serum.91 More recently Liang, et al.92 have described a UV cell with a parallel flow optical path of 120–140 mm for absorbance and fluorescence detection at the end of a CE column, which would be most suitable for mFIA applications.Sequential detector arrays are an attractive approach in sensor design and lend themselves well to mFIA systems.82,93 One such system82 has been described, based on electrochemical detectors, for liquid chromatographic separations of catecholamines and consisted of four photolithographically prepared ISFET sensors aligned sequentially in a 5 mm silicon channel 100 mm wide and 70 mm deep.The total volume of the detector was 20 ml. Another device,93 using a peristaltic pump to control flow rates, consisted of nine 5 mm ISFET sensors housed in a 15 ml cell, used for pH, potassium and calcium determinations in biological fluids. Although neither of these systems was used in mFIA manifolds, they do illustrate the ability to develop array detectors compatible with mFIA technology. One of the most exciting prospects for mFIA detector design is the ability to incorporate their fabrication into one integrated device.For example, miniature mass spectrometers (3 3 3 3 3 mm) have been produced based on fabrication technology that would be ideally suited for mFIA.94 Miniature spectroscopic systems are also becoming available95,96 and future developments in spectrometers incorporating opto-electronic systems Fig. 7 Calibration based on 100 ppb PO4 with dynamic applied pinch voltage. The equivalent absorbance values for 100, 50 and 25 ppb phosphate are indicated on the x-axis. 8R Analyst, January 1997, Vol. 122will undoubtedly bring valuable complementary technology for future mFIA detector design. Future Trends The preceding sections have tried to focus on the basic concepts and developments relating to mFIA systems based on EOF. Some indication has been given of the likely areas where current and future research may prove to be of great value. These include the fabrication of devices where there is considerable potential for the construction of stacked or three-dimensional systems, possibly using cold bonding and direct laser etching techniques.The mobilisation of reagents and analytes, based on EOF, requires more complete characterisation if flows and mixing effects at intersecting channels are to be effectively exploited. The most important development if mFIA technology is to be fully realised is the fabrication of self contained operational systems with proven application robustness.The close relationship of mFIA technology with separation techniques such as capillary electrophoretic and micro-electrochromatographic separations may well lead to some form of hybride system in the near future. The area perhaps where the greatest advances in mFIA-based technology will be most readily realised is the biotechnology sector, where methodology and applications are complementary to miniature systems. Already examples are emerging of applications in DNA fragment analysis97 and immunoassay methodology.98 Developments, however, need to be focused not only on the integrated device which may be encapsulated and equipped with telecommunication for remote operation, but also to include interfacing to existing measurement systems such as MS or NMR, which would benefit from some form of sample pretreatment.One area which has not yet been considered in mFIA systems is a return to the early gasphase work started by Terry et al.5 Clearly there are some exciting possibilities in gas and multiphase systems yet to be realised.Conclusion Where has mFIA and more generally mTAS got to? In 1991, Manz et al.13 questioned whether the developments in mTAS were ‘a look into next century’s technology or just a fashionable craze’. In their concluding remarks, the authors made some general comments relating to the uptake or acceptance of mTAS technology, ‘namely that changes are required in the political and cultural opinions of analytical work if appropriate financial support is to be forthcoming and that the research base must grow worldwide to foster both competitive and collaborative research’.Further, they identified that ‘the market acceptance of the technology must be embraced through the design and production of mTAS concepts’. Clearly, these requirements have been only partially fulfilled. An examination of the literature indicates that there has been a positive growth in the research base and this will obviously support the fundamental development of the science.What is less obvious is the political and more importantly the economic will to support the development of the technology, and this may yet be seen to be the most seriously limitation to the growth of the science. Although miniaturisation is conceptually appealing, there remain some important technical obstacles to overcome, such as the introduction of ‘real’ samples and the ability to deal with suspended particles.These limitations are not beyond the scope of present day membrane technology, and should pose no serious hindrance to the advancement of the science. Developments in the field of mTAS since 1991 clearly point to the technique forming the basis of future methodology applicable to a wide range of applications ranging from measurement science to chemical synthesis in which mFIA based on EOF flow will play a significant role.The limitation in releasing the considerable potential that micro reactor technology can offer resides not in the technological challenge but in the imagination of our minds and only requires us to realize it. References 1 Liu, S., and Dasgupta, P. K., Anal. Chim. Acta, 1993, 283, 739. 2 Bogue, R., Lab. Equip. Dig., 1995, February, 14. 3 Manz, A., Effenhauser, C. S., Burggraf, N., Verpoorte, E. J. M., Raymond, D. E., and Widmer, H. M., Analusis Mag., 1994, 22, M25. 4 Craston, D., and Cowen, S., Chem.Br., 1996, October, 31. 5 Terry, S. C., Jarman, J. H., and Angell, J. B., IEEE Trans. Electron. Devices, 1979, ED-26, 1880. 6 Capillary Electrophoresis: Theory and Practice, ed. Grossman, P. D., and Colburn, J. C., Academic Press, New York, 1992. 7 Mikkers, F. E. P., Everaerts, F. M., and Verheggen, Th. P. E. M., J. Chromatogr., 1979, 169, 11. 8 Jorgenson, J. W., and Lukacs, K. D., Anal. Chem., 1981, 53, 1298. 9 Whalen-Pechersen, E. K., and Jurs, P.C., Anal. Chem., 1981, 53, 2184. 10 Manz, A., Graber, N., and Widmer, H. M., Sens. Actuators B, 1990, 1, 244. 11 Advances in Chromatography, ed. Brown, P. R., and Grushka, E., Marcel Dekker, New York, 1993, pp. 1–66. 12 Micro Total Analysis Systems, ed. van den Berg, A., and Bergveld, P., Kluwer, Dordrecht, 1995. 13 Manz, A., Fettinger, J. C., Verpoorte, E. J. M., L�udi, H., Widmer, H. M., and Harrison, D. J., TrAC, Trends Anal. Chem. (Pers. Ed.), 1991, 5, 144. 14 Manz, A., Harrison, D.J., Verpoorte, E. M. J., Fettinger, J. C., L�udi, H., and Widmer, H. M., Chimia, 1991, 45, 103. 15 Lui, S., and Dasgupta, P. K., Anal. Chim. Acta, 1992, 268, 1. 16 Harrison, D. Glavina, P. G., and Manz, A., Sens. Actuators B, 1993, 10, 107. 17 Jacobson, S. C., Hergenr�oder, R., Koulng, L. B., and Ramsey, J. M., Anal. Chem., 1994, 66, 2369. 18 Daykin, R. N. C., and Haswell, S. J., Anal. Chim. Acta, 1995, 313, 155. 19 Verpoorte, E. J. M., Van der Schoot, B. H., Jeanneret, S., Manz, A., and de Rooij, N.F., in Interfacial Design and Chemical Sensing, ACS Symposium Series, American Chemical Society, Washington, DC, 1994, ch. 21, p. 244. 20 Schomburg, W. K., Vollmer, J., B�ustgens, J., Fahrenberg, J., Hein, H., and Manz, W., J. Micromech. Microeng., 1994, 4, 1. 21 Sensors: a Comprehensive Survey, ed. Grandke, T., and Ko, W. H., VCH, Weinheim, 1989, vol. 1, p. 107. 22 Manz, A., Verpoorte, E. M. J., Effenhauser, C. S., Burggraf, N., Raymond, D.E. and Widmer, H. M., Fresenius’ J. Anal. Chem., 1994, 348, 567. 23 Woffenbuttel, R. F., Sens. Actuators A, 1992, 30, 109. 24 Altmann, J. P., Das Neue Lehrbuch de Glus�alzerei, Gentner, Stuttgart, 1963, p. 29. 25 Daniel, J. S., and Delapierre, G., J. Micromech. Microeng., 1991, 1, 187. 26 Becker, E. W., Ehrfeld, W., Hagmann, P., Maner, A., and M�uchmeyer, D., Microelectron. Eng., 1986, 4, 35. 27 Knapp, J., Andreae, G., and Petersohn, D., Sens. Actuators A, 1990, 21–23, 1080. 28 Kovacs, G. T. A., Petersen, K., and Albin, M., Anal. Chem., 1996, 68, 407A. 29 Jacobson, S. C., Koutny, L. B., Herenr�oder, R., Moore, A. W., Jr., and Ramsey, J. M., Anal. Chem., 1994, 66, 3472. 30 Fan, Z., and Harrison, D. J., Anal. Chem., 1994, 66, 177. 31 Elliot, D. J., Integrated Circuit Fabrication Technology, McGraw- Hill, New York, 1982. 32 Thin Film Processes, ed. Vossen, J. L., and Keren, W., Academic Press, San Diego, 1978, ch. V-1, p. 401. 33 van der Linden, W. E., TrAC, Trends Anal.Chem. (Pers. Ed.), 1987, 6, 37. 34 Tyssen, R., Anal. Chim. Acta, 1980, 114, 71. 35 Hungerford, J., Thesis, University of Washington, 1986. 36 Gravesen, P., Branebjerg, J., and Jensen, O. S., J. Micromech. Microeng., 1993, 3, 168. 37 Manz, A., and Simon, N., Anal. Chem., 1987, 59, 74. Analyst, January 1997, Vol. 122 9R38 Micro Total Analysis Systems, ed. van den Berg, A., and Bergveld, P., Klewer, Dordrecht, 1995, p. 165. 39 Shoji, S., and Esashi, M., Sens.Actuators B, 1992, 8, 205. 40 Field, L. A., and Muller, R. S., Sens. Actuators A, 1990, 21–23, 935. 41 Fung, C. D., Cheung, P. W., and Fleming, D. G., Microachining and Micropacking of Transducers, Elsevier, Amsterdam, 1985. 42 Harz, M., J. Micromech. Microeng., 1992, 2, 161. 43 Esashi, M., Nakano, A., Shoji, S., and Hebiguchi, H., Sens. Actuators A, 1990, 21–23, 931. 44 Seiler, K., Fan, Z. H., Fluri, K., and Harrison, D. J., Anal. Chem., 1994, 66, 3485. 45 Dasgupta, P.K., and Liu, S., Anal. Chem., 1994, 66, 1792. 46 Li, S. F. Y., Capillary Electrophoresis, Elsevier, New York, 1992. 47 Shaji, S., Esashi, M., van der Schoot, B., and de Rooij, N. F., Sens. Actuators A, 1992, 32, 335. 48 Horowitz, P., and Hill, W., The Art of Electronics, Cambridge University Press, Cambridge, 2nd edn., 1989, p. 368. 49 van der Schoot, B. H., Jeanneret, S., van den Berg, A., and de Rooij, N. F., Anal. Methods Instrum., 1993, 1, 38. 50 Stewart, K. K., Beecher, G.R., and Hare, P. E., Fed. Proc., Fed. Am. Soc. Exp. Biol., 1974, 33, 1434. 51 Moriney, R. M., White, R. M., and Howe, R. T., Proc. MEMS’91, Nara, Japan, 1991, p. 277. 52 Richter, A., Plettner, A., Hofmann, K. A., and Sandmeier, H., Proc. MEMS ’91, Nara, Japan, 1991, p. 271. 53 Osada, Y., Okuzzaki, H., and Hari, H., Nature (London), 1992, 355, 242. 54 Fuhr, G., Fiedler, S., M�uller, T., Schnell, T., Glasser, H., Lisec, T., and Wagner, B., Sens. Actuators A, 1994, 41–42, 230. 55 Zheng, H. J., and Dasgupta, P. K., Anal. Chem., 1994, 66, 3997. 56 Adams, R. N., Electrochemistry at Solid Electrodes, Marcel Dekker, New York, 1969. 57 Masson, D., Magdassi, S., and Sasson, Y., J. Org. Chem., 1990, 55, 2714. 58 Moore, A. W., Jr., Jacobson, S. C., and Ramsey, J. M., Anal. Chem., 1995, 67, 4184. 59 Jacobson, S. C., Hergenr�oder, R., Koutny, L. B., and Ramsey, J. M., Anal. Chem., 1994, 66, 1114. 60 Jacobson, S. C., Hergenr�oder, R., Koutny, L. B., Warmack, R. J., and Ramsey, J.M., Anal. Chem., 1994, 66, 1107. 61 Seiler, K., Harrison, D. J., and Manz, A., Anal. Chem., 1993, 65, 1481. 62 Manz, A., Effenhauser, C. S., Burggraf, N., Harrison, D. J., Seiler, K., and Fluri, K., J. Micromech. Microeng., 1994, 4, 257. 63 Tseida, T., Mizuno, T., and Akiyama, J., Anal. Chem., 1987, 59, 799. 64 Harrison, D. J., Manz, A., Fan, Z., L�udi, H., and Widmer, H. M., Anal. Chem., 1992, 54, 1926. 65 Manz, A., Harrison, D. J., Verpoorte, E. M. J., Fellinger, J.C., Paulus, A., L�udi, A., and Widmer, H. M., J. Chromatogr., 1992, 593, 253. 66 Effenhauser, C. S., Paulus, A., Manz, A., and Widmer, H. M., Anal. Chem., 1994, 66, 2949. 67 Effenhauser, C. S., Manz, A., and Widmer, H. M., Anal. Chem., 1995, 67, 2284. 68 Hasser, A., MSc Thesis, University of Hull, 1996. 69 Burggraf, N., Manz, A., Verpoorte, E. M. J., Effenhauser, C. S., Widmer, H. M., and de Rooij, N. F., Sens. Actuators B, 1994, 20, 103. 70 Ruzicka, J., and Hansen, E.H., Flow Injection Analysis, Wiley, New York, 2nd edn., 1988. 71 Halliday, D., and Resnick, R., Physics, Wiley, New York, 1960. 72 Jacobson, S. C., Koutny, L. B., Hergenr�oder, R., Moore, A. W., Jr., and Ramsey, J. M., Anal. Chem., 1994, 66, 3472. 73 Monning, C. A., and Jorgenson, J. W., Anal. Chem., 1991, 63, 802. 74 Yao, X. W., Wu, D., and Regnier, F. E., J. Chromatogr., 1993, 636, 21. 75 Smith, J. T., and Rassl, Z. E., J. High Resolut. Chromatogr., 1992, 15, 573. 76 Marakami, Y., Suda, M., Yokogama, K., Tukeuchi, T., Tamiya, E., and Kurube, I., Microchem., 1994, 49, 319. 77 Tsuda, T., Nomura, K., and Nakagawa, G., J. Chromatogr., 1982, 248, 241. 78 Giddings, J. C., Science, 1993, 260, 1456. 79 Widmer, H. M., and Koch-Kellner, L., Chimia., 1989, 43, 320. 80 Koch, L., Koch, T., and Widmer, H. M., J. Chromatogr., 1990, 517, 395. 81 Raymond, D. E., Manz, A., and Widmer, H. M., Anal. Chem., 1994, 66, 2858. 82 Sudfa, M., Sakuhara, R., and Karube, I., Appl.Biochem. Biotechnol., 1994, 41, 3. 83 Haber, C., Sclvestri, I., B�o�osli, S., and Seman, W., Chimia, 1991, 45, 117. 84 Chervet, J. P., Ursem, M., Salzmann, J. P., and Vannoort, R. W., LC– GC, 1989, 7, 515. 85 Wohltjen, J., Sens. Actuators, 1984, 5, 305. 86 Ton, W., Zhang-You, S., Smith, S., Birnbaum, D., and Kopelman, R., Science., 1992, 258, 778. 87 Xi, X., and Yeing, F. S., Anal. Chem., 1990, 62, 1580. 88 Xue, Y., and Yeung, E. S., Anal. Chem., 1994, 66, 3575. 89 Verpoorte, E.M. J., Manz, A., L�udi, H., Bruno, A. E., Maystre, F., Kruttiger, B., Widmer, H. M., van der Schoot, B. H., and de Rooij, N. F., Sens. Actuators B, 1992, 6, 66. 90 Weigl, B. H., and Woelfbeis, O. S., Anal., Chem., 1994, 66, 3323. 91 Suda, M., Sakuchara, T., and Karube, I., 2nd Bioengineering Symposium, Tsukuba, Japan, 1992, p. 113. 92 Liang, Z., Chiem, N., Ocvirk, G., Tang, T., Fluri, K., and Harrison, D. J., Anal. Chem., 1996, 68, 1040. 93 Arquint, P., Koudelka-Hep, M., van der Schoot, B.H., van der Wal, P., and de Rooij, N. F., Clin. Chem., 1994, 40, 1805. 94 Micro Total Analysis Systems, ed. van den Berg, A., and Bergveld, P., Kluwer, Dordrecht, 1995, p. 299. 95 Conzen, J. P., B�urch, J., and Ache, H. J., Appl. Spectrosc., 1993, 47, 753. 96 Ache, H. J., Chem. Ind., 1993, 1, 40. 97 Jacobson, S. C., and Ramsey, J. M., Anal. Chem., 1996, 68, 720. 98 Koutny, L. B., Schmalzing, D., Taylor, T. A., and Fuchs, M., Anal. Chem., 1995, 68, 18. Paper 6/06289J Received September 16, 1996 Accepted November 18, 1996 10R Analyst, January 1997, Vol. 122 Critical Review Development and Operating Characteristics of Micro Flow Injection Analysis Systems Based on Electroosmotic Flow A Review S. J. Haswell School of Chemistry, University of Hull, Hull, UK HU6 7RX Summary of Contents Introduction Fabrication Pump Design Injector Design Reactor Design Detector Design Future Trends Conclusion References Keywords: Micro flow injection; electroosmotic flow; miniaturisation; micro total analytical system ( mTAS) Introduction Over the past 50 years, the design of instt of chemical, biological and physical parameters has brought analytical chemistry to a highly automated and technologically advanced science.It would be naive, however, to believe that the chemistries and physics associated with present day analytical measurements, are in any way limited by current technology. On the contrary, laboratory based analysis has been almost exclusively designed, for the ergonomic needs of the human worker. There can be little doubt that this preoccupation with the physical size and operation of equipment, for laboratory based measurements, has significantly influenced the teaching and practice of analytical chemistry.If one considers the fundamental chemical and physical phenomena on which measurements are based, then it is apparent that one only requires a few molecules or atoms to be present for quantitative measurement to occur.In short, the physics and engineering of measurement science can be considered as oversized and in many cases over engineered to meet the needs of the chemical measurement. Whilst instrumental design has relentlessly moved towards automation, the realization that the quality of analytical data can be profoundly influenced by factors such as sampling, sample storage, sample pretreatment and matrix interferences has led to the concept of the total analytical approach.This more holistic view of analysis, in which traceability and uncertainty prediction can be carried out, is often embodied in the concept of the ‘total analytical system’ (TAS).1 There are, however, numerous disadvantages to TAS, including slow sample processing, lack of selectivity and high reagent consumption;2 however, many such problems may be overcome through system miniaturisation. The term micro TAS (mTAS) was first used at the Transducers 89 Conference.3 In its simples sense, mTAS involves the miniaturisation of all the functions found in an analytical method for example, pumps, valves, flow manifolds, mixing and reaction chambers, phase and analyte separation, detectors, control and communication electronics. One of the more exciting prospects of the mTAS concept is the suggestion that the entire chemical measurement laboratory could be miniaturised onto a device of a few square centimetres.4 This type of miniaturisation has become possible largely through the adoption of microfabrication techniques developed in general by the microelectronics industry, and one can consider miniaturisation of chemical reactors and their corresponding instrumental requirements to be in a similar position to the microelectronics industry 30 years ago.Interestingly, it was from the heart of the microelectronics industry that Terry et al.5 were one of the first to demonstrate that integrated circuit (IC) technology could be used to fabricate a miniature GC instrument using a 5 in silicon wafer. Although the work of Terry and co-workers served to point the way, it was not until the advent of capillary electrophoresis (CE)1,6–8 and in particular, the exploitation of electroosmotic flow (EOF), that the realization of mTAS came about some 10 years after the miniature GC work was first reported.Stephen J. Haswell is a Senior Lecturer in Analytical Chemistry at the University of Hull, UK.Following a period of five years working in the food and plastics industries, he undertook a Ph.D. in the area of atomic spectroscopy graduating in 1983 from the University of Plymouth. His current research activities are in the areas of mFIA, microwave enhanced reaction chemistry, trace elemental speciation and chemometrics. He is author of over 100 research papers, books and patents and presenter of numerous national and international lectures on his research activities.The subject of this present review has become a great interest of the author over the past five years. Analyst, January 1997, Vol. 122(1R–10R) 1RA significant research base has now built up in the area of mTAS9–14 and the development of mFIA over the past 5 years can be attributed to key research carried out by groups based in Basle, Switzerland,14 Texas Tech. University, USA,15 University of Alberta, Canada,16 the Oak Ridge National Laboratory, Tennessee, USA,17 and in a very modest way in the author’s own laboratory.18 What follows is a review of the features present in a relatively small, but rapidly expanding research base, which can be exploited to develop micro flow injection analysis (mFIA) systems, based on electroosmotic flow, to produce reliable sensor-type devices which incorporate mechanical and chemical robustness. Fabrication The fabrication of mFIA systems has attracted various approaches in recent years, ranging from manifolds based around CE-type fused-silica capillaries,1 to modular micro pump systems.19,20 This particular section, however, will focus specifically on fabrication methodology which leads to the production of a single integrated device, designed to use electroosmotic mobility for sample and reagent pumping.18 The discussion will focus almost exclusively on micro machined monolithic device fabrication, which adopts standard IC photolithographic, wet etching and bonding techniques to produce a planar structure.21,22 As the techniques used are based essentially on IC technology, they represent well established methodology, outlined schematically in Fig. 1. If one considers the current ‘state-of-the-art’ in IC fabrication, then it is clear that the challenge of producing mFIA manifolds can in no way be considered as technologically demanding. From what follows, it will become clear that mFIA device fabrication can be relatively simple and offer a reasonably high degree of flexibility in systems design.The choice of a suitable substrate into which the channels of a mFIA manifold can be etched is closely related to factors such as the fabrication method, the analytical chemistries involved and the proposed pumping, injection and detection systems. The obvious candidate given the close relationship with IC technology is silicon,23 which is amenable to the fabrication of structures in the nano- and micro-metre range.This crystalline material can be obtained in a very pure form, at relatively low cost, and offers good mechanical and chemical properties. Whilst silicon may seem an ideal microengineering candidate for mFIA fabrication, it is by nature a conductor and so surface modifications will be required if electroosmotic pumping of samples and reagents is to be achieved. This can be carried out by coating the silicon with SiO2 or Si3N4 to produce the required surface chemistries.Harrison et al.16 described the fabrication of such a manifold, in which the properties of silicon and oxide/nitride-modified layered surfaces were evaluated. They reported that the operating voltage achievable with such devices will be limited by the quality of the oxide insulating film produced. Nevertheless, the results are sufficient to demonstrate that silicon-based devices using a sandwich structure (i.e., oxide–nitride–oxide) can sustain potentials in the range 400–1200 V before dielectric breakdown, which is certainly sufficient to generate EOF mobility for mFIA systems.Although silicon does offer the attraction of producing high precision engineered mFIA manifolds, with well characterised surfaces and the potential to integrate control and detection on to one substrate, other materials such as glass24 and silica (quartz)25 are also amenable to IC based fabrication techniques and will produce surfaces eminently suitable for EOF pumping.Judging from the reports in the literature, glass and silica have proved to be the most popular materials for device fabrication to date, but metal, plastic and ceramic substrates are also possible candidates for the manufacture of mFIA systems.26,27 In general, three basic concepts are required for the preparation of mFIA systems: (i) a suitably prepared substrate, (ii) photolithographic equipment and (iii) wet etching and bonding facilities.28 Prepared substrates based on glass, silica and silicon can be obtained commercially from photomask producers (e.g., Alignrite, Wales, UK); these come ready coated with a metal film (0.1 mm) such as chromium over which is spun a positive photoresist layer (0.7 mm).These plates, which are typically 152.4 mm square (5 3 5 in) and 3 mm thick, can be used directly for manifold pattern transfer using photolithographic methods. A mask or negative of the final channel patterns required for the mFIA system can be produced using CAD computer software.29,30 The mask pattern can be transferred to the photoresist film on the substrate, using basic photographic development equipment.For channels larger than 1 mm, visible or UV light can be used to transfer the pattern from the mask to the photoresist (exposure times being approximately 5 ms); however, to obtain good line definition and submicro patterning, X-ray or electron-beam (e-beam) photolithography will be required. Where facilities permit, e-beam photolithography offers at present one of the most flexible ways of producing mFIA manifold designs; however, for most applications simple basic photographic equipment will suffice.Once the pattern has been transferred to the photoresist film on the substrate, the plate can be developed and wet etched. If one uses the large plates as described, then multiple devices can be prepared; in our laboratory, for example, 25 manifolds of various geometries and channel widths are produced from one plate (Fig. 2).18 Workers in the field tend to prepare their own plates rather than using a commercial source and, although this adds additional steps to the fabrication process, it is not too technically demanding.29,30 As indicated previously, these are typically made from silica or glass on to which a thin metal film (0.05–0.1 mm) of chromium, gold or a combination is produced, using sputtering or chemical vapour deposition (CVD) methods. The metal film is then spin coated with a layer of positive photoresist (approximately 0.5–2 mm).Fig. 1 Schematic diagram of the steps involved in the anistropic etching technique for the fabrication of mFIA devices. The arrow in Step 1 represents radiation. 2R Analyst, January 1997, Vol. 122Following the photolithographic transfer of the manifold pattern, the exposed photoresist and corresponding metal film are removed using commercially available reagents. The prepared plates may then be placed in an oven at 120 °C for 72 h to harden the photoresist and remove solvent residues from the plate.18 It should be noted that although the presence of a metal film on the glass is important for controlling the degree of surface etching that will occur, the exposed substrate will experience undercutting or sideways etching, which will proceed at a rate of approximately 2 : 1.Thus, as the etch goes down to 1 mm it spreads laterally 2 mm at the edges, producing channels, for which the width always exceeds the depth.31 Clearly, for a given substrate, the choice of the initial photomask line width and the time of etch will influence the size of the final channels produced.Various reagents have been used for the wet etching of the substrate, but in general for glass and silica, hot (70 °C) dilute HF–NH4F (1% HF + 5% NH4F in water)18,29 will give an etch rate of approximately 0.3–0.5 mm min21, whereas dilute HF– HNO3 produces an etch rate of between 0.5 and 0.8 mm min21 in Pyrex.30 Silicon etching requires the use of reagents such as ethylenediaminepyrocatechol (115 °C, etch rate 2.5 mm min21) or KOH.16 Although indicative figures for etch rates can be obtained,31,32 it is strongly recommended that sample strips of the substrate are tested in the selected etchant to establish the rate of etching for a given system.It should be stressed that the final size of the mFIA channels will depend on the initial width of the lines on the photomask, the substrate material and mode of etching.During the etching process, agitation of the substrate or etchant is advisable otherwise asymmetric channel formation, i.e., uneven etching on the bottom edge or side of the channel, may occur. It is usual, however, to obtain a channel profile that is wider at the top than the bottom. This effect, associated with undercutting, will vary as a function of the etch time, substrate type and reagents used. The surface quality of the etch is generally related to the types of material used.Good quality silica, for example, gives a well defined etch, whereas Pyrex or borosilicate glass can produce a rougher surface owing to the crystalline structure of the material. The real effect or significance of the surface properties in mFIA channels has not yet been fully characterised, but clearly the more controlled the etch, the more precise and smaller will be the channels that one can produce.What is not apparent from experimental results is the real influence that surface topography may have on the mobility and reactivity of reagents in a mFIA system. Rough or poorly defined surfaces and intersections, of one or more channel, will increase the effect of turbulence and in turn promote dispersion, which in some instances may prove to be an advantage where mixing is required. This factor is particularly important in capillary systems where the Reynolds number is lower than the transitional values of 2000 or 2300 indicating laminar flow, thus minimal mixing will dominate.33–35 However, research indicates that at low Reynolds numbers microfluidic mixing cannot be simply classified as a laminar or turbulent model.36 What is clear is that as channel sizes become smaller, surface effects will inevitably become more significant, which in turn will have an impact on fabrication and material specifications. Manz and Simon37 demonstrated that for a 10-fold decrease in channel size, a 1000-fold decrease in reagent consumption and a 100-fold decrease in related time variables would be obtained.Pressure requirements of such a system will, however, increase by a factor of 100, but as indicated later this does not effect the voltage requirements for EOF. Once an etched base has been produced, the top of the channels need to be sealed. Various bonding methods have been suggested, including glueing, low temperature bonding or annealing, high temperature fusion and anodic bonding.38 The use of an adhesive in such systems can pose problems due to channel plugging and, although this may be overcome to some extent by using photosensitive dry films laminated on to the substrate surface as a protective coating,39 the technique has found little real use in mFIA.Thermal bonding or annealing of top plates represents one of the simplest and therefore most widely used methods in device fabrication.These methods usually involve heating the substrate and top plate, which may be under slight pressure,30 in an oven or furnace to near the upper annealing temperature. For glass and silica this is usually between 500 and 600 °C, whereas silicon will require higher temperatures of around 900 °C. The period of heating varies between 48 and 96 h, after which the device is slowly cooled (48–72 h) to minimize physical stress. For glass substrates, hydrolysing the surface with dilute NH3–H2O2 followed by heating at 500 °C for 24 h29 or 575 °C for 96 h18 has been demonstrated to achieve good reliable bonding.Slightly more complex temperature programmes have been suggested, but these are essentially slight modifications to the same basic method.30 In order to keep the fabrication of a mFIA manifold as simple as possible and to increase the bonding success rate, it is advisable to use the same material (i.e., similar thermal expansion coefficients) for both the base and top of the device, so minimizing stress features. The fabrication of multilayer devices can become complex and techniques using intermediate layers have been described in which more complex structures such as pumps and valves are required.40,41 One technique which has found particular favour with workers fabricating microvalves, micropumps and silicon devices generally is anodic bonding.42,43 In this process, ions such as sodium and oxygen are thought to migrate at elevated temperatures to electrodes placed on the outer surface of one of the layers, so increasing the electrostatic surface charge.The temperature and applied voltages used will depend on the materials in question, but for glass these are found to be 200–400 °C at 30–300 V and for silica 700–800 °C at 30 V applied for a period of 45 min. It might be necessary when employing anodic bonding to treat the surface with HF prior to bonding. Holes or ports made in the top plate are commonly used as reagent or sample feeds into the manifold and can be generated either before18 or after29,30 the bonding process.The holes can simply be produced by using mechanical18 or ultrasonic16 drilling of the substrate and vary in diameter from 0.5–2 mm. At present, laser ablation, which offers an attractive method for hole production, has not been employed, but no doubt the technique will find its way into the literature before long. Plastic Fig. 2 Example of the photomask design used to produce five mFIA manifolds designs with variable line or channel widths. The line widths shown are A and E, 50, B, 30 and C, 10 mm.Analyst, January 1997, Vol. 122 3Rreservoirs are usually glued to hold liquids and support the platinum electrodes when required. A photograph of a completed device used for phosphate and nitrite determination is shown in Fig. 3. To date, only planar mFIA systems have been reported based on the fabrication techniques described. The fabrication of three dimensional systems is, however, an attractive prospect and no doubt stacked systems will be produced as more sophisticated chemistries are exploited.Such systems based on micropumps have been reported in which multi-layers or modules are clamped or stuck together to produce a complete device.19,20 Once a complete mFIA manifold has been fabricated for use with EOF pumping, the success of the bonding process can be checked by filling the channels with a buffer44 or weak acid18 and plotting the current–voltage relationship.This plot is generally found to be linear up to 10 kV in glass and silica, after which point dielectric breakdown of the substrate occurs. Any deviation in linearity at lower applied voltages is indicative of a device failure, associated with incomplete bonding. Pump Design In order to achieve reliable flow dynamics in an FIA manifold, which may contain numerous reagent streams, pulse free, constant flow characteristics must be maintained.It is not surprising therefore, to find that mFIA systems also call for pulse free, variable nl min21–ml min21 flow control. Although a direct syringe or piston pump can be used with mFIA systems,45,46 the closest equivalent to a conventional peristaltic FIA pump is the so called micropump, which usually takes the form of a pulsating one-way valve driven typically by a piezoelectric device.36,45,47–50 Based on microengineering technology, micropumps have been employed in mFIA manifolds consisting of channels 100 mm wide, 10 mm deep and 4 cm in total length.47 When such small channel dimensions are employed, the hydrodynamics of the system can produce a pressure drop of 0.7 atm at a flow rate of 0.5 ml min21. However, diaphragm devices, commonly using microchemical silicon membranes, are only reliable up to backpressures of 0.2 atm and so high pressure pumps or large channel dimensions are therefore required.High pressure pump designs using nickel rather than silicon check values are now being developed using low temperature bonding techniques, incorporating intermediate photoresist layers to adhere the valve to the body of the pump.47 These devices, which use a piezoelectric disc glued to the outer surface of the pump housing, operate at many hundred Hertz, requiring around 300 V to produce flow rates in the range 20–300 ml min21. In an elaborate example, four such pumps have been used to drive reagents through a three-dimensional system constructed from silicon and glass at a flow rate of 1 ml min21, the channel dimensions being 600 mm wide and 200 mm deep in order to accommodate pressure effects.19 Diaphragm micropumps do have an important role to play in applications where electroosmotic pumping is not applicable, owing to sample or reagent chemistries.It is essential, however, that micropumps remain chemically inert to the reagents and samples with which they may come into contact.In addition to piezoelectrically driven pumps, ultrasonic51 and other more exotic electrically activated pumps have also been described.36,52–54 Electrophoresis embodies two basic electrokinetic components, electroosmotic and electrophoretic flow. These two components combine to give the total flow or velocity (n) in the following way: total flow rate = n = (meo + mep)E (1) where meo = electroosmotic mobility, mep = electrophoretic mobility and E = applied electric field.At relatively low electric field strengths, electroosmotic mobility represents the larger component of the two processes; however, as the field strengths are increased the influence of the electrophoretic component (migration of analyte ions) increases also. For mFIA, where bulk mobility is desired, one finds that field strengths around 300–400 V cm21 are usually adequate; however, if separation is required, as in CE, then field strengths greater than 1 kV cm21 are typically required.It is common for mFIA systems based on EOF to operate with field strengths of the order of 80–300 V cm21, at which little or no analyte separation is observed The generation of an EOF to pump reagents and samples through a mFIA system is subject, however, to certain physico-chemical limitations. Firstly, in order to generate the EOF one must use a material which will yield negatively charged groups on the surface or walls of the channels when placed in contact with an appropriate liquid.Secondly, the liquid phase must dissociate to some extent in order to generate counter positive ions (notably H+ ions). The combination of the negatively charged surface (typically SiO2) and the H+ ions in solution will form a diffuse double layer (sometimes referred to as the Gouy or Helmholtz layer). This diffuse double layer acts as a parallel-plate electric capacitor whose plates are d cm apart each carrying a charge e per cm2.The zeta potential (x) will be the potential difference between the plates, given by the general equation x = 4ped/er (2) where er is the relative permittivity of the medium between the hypothetical plates. If an electrical field is now applied through the liquid phase, ions will migrate to their respective electrodes dominated by the positive ions (H+) moving to the negative or ground electrode. As the ions move, they induce a drag on the bulk of the liquid, which in turn results in a corresponding net transfer of the solution to the negative electrode. Thus, in EOF the mobility of the solution is from the anode (positive electrode) to the cathode (negative electrode) or ground.In practice, the formation of the double layer is limited to the pH range 4–10. At lower pH values the cationic population becomes so high that the EOF is overrun by conductive flow and at pH values greater than 10 the cation population becomes too low to sustain the double layer.The zeta potential that is generated when the double layer forms will be influenced by changes in pH and ionic strength of the solutions in the channel, and this can affect the corresponding flow rate. Clearly, as one of the main attributes of a pump in FIA is to maintain a reproducible flow rate, the pH and ionic strength of the reagents Fig. 3 A fully assembled mFIA device showing the mounted plastic reservoirs which act as wells for the reagents and samples and support the platinum electrodes required for EOF pumping. The device is shown mounted in a rig which allows fibre optics (seen either side of the block) to be coupled into relevant sections or channels of the manifold. 4R Analyst, January 1997, Vol. 122used in a mFIA manifold need to be controlled, through the use of buffer systems. As the ionic strength increases, for example, a counter-ion effect will prevent the ions from migrating through the solution independent of the EOF.In general, it is therefore preferable to keep the total ionic strength of reagents and samples as low as possible. Once an EOF has been generated, the flow establishes a flat, trapezoidal profile notably different from the parabolic (bullet) shape commonly associated with conventional FIA. Closer investigation of the mFIA profile reveals that the edges of the profile (i.e., closest to the wall surface) are slightly in advance of the bulk owing to the drag effect and the extent of this effect will be a function of the solution viscosity.55 The fact that the flow profile, which offers minimal band broadening, does not contain the parabolic flow characteristics of a pressure or hydrodynamically pumped system, does have implications with respect to the mixing of reagents in the mFIA reactor, where some dispersion will be necessary to achieve the desired chemical reactions.Thus EOF-pumped mFIA can be considered to be more akin to segmented flow analysis rather than conventional FIA, in its flow characteristics.In an EOF-pumped mFIA system, the flow rate can be effectively controlled by varying the applied voltage as follows: n flow rate v = m (3) L where n = applied voltage (V), L = length of the channel (m) and m = sum of the electroosmotic and electrophoretic mobility component under different conditions of pH and ionic strength. The EOF and hence the flow rate in mFIA, will be unaffected by capillary diameter up to channel sizes approaching 250 mm, after which drag effects of the bulk solution may cause serious disruption to the EOF.The EOF produced in a channel can be considered, in the electrical sense, as a resistor, thus Joule heat will occur in the capillary. However, the almost negligible bulk properties of components in a channel, relative to the substrate, will ensure an effective dissipation of heat, which rarely poses a serious problem in mFIA.If, however, the resistance is allowed to increase, for example when a highly viscous or an immobile solvent is present in a channel, heating can occur. It should be stressed that an EOF can only be established if the double layer is formed, and this requires ions or dipoles to be present in the liquid stream. In the case, for example, where organic compounds or solvents may be present, such conditions may not be met. For example Zheng and Dasgupta55 described some initial studies, using silica capillaries, to evaluate the suitability of an EOF in a mFIA system for carrying out in-line analyte phase separation or solvent extraction.They described the use of a 50 cm 3 7.5 mm id polyimide coated silica capillary to investigate various aqueous ionic complexes, based on well characterised cationic, anionic and neutral ion-pair chemistries. A quaternary ammonium salt, tetrabutylammonium perchlorate, was added to chloroform as a supporting electrolyte to assist in the mobilization process56 and to catalyse phase transfer.57 The results clearly indicated that organic solvent pumping was possible, when modified with the quaternary ammonium salt, but that migration of the ammonium ion occurred, creating a positive front end to the solvent slug.Bleeding of ions into the secondary aqueous buffer from the organic phase was also reported. Furthermore, they suggest that a thin interfacial layer of buffer is generated between the organic solvent and the capillary walls which enables the EOF to be generated.Investigations into solvent extraction indicated that the ion-pair complexes studied were either extracted into the organic phase or accommodated in the aqueous phase ahead of the organic slug interface. Not only were the findings of Zheng and Dasgupta a significant contribution in terms of demonstrating phase extraction and organic solvent mobility using EOF in a mFIA system, but they also clearly indicate that the migration of ions within a solvent under the influence of an electric field could lead to the development of gradient and separation techniques, complementary to CE, such as selective solvent extractions and matrix modification.Interestingly, the use of micellar electrokinetic capillary chromatography, described by Moore et al.,58 suggests that multiphase systems could be developed for mFIA applications, thus offering significant advantages for organic solvent based chemistries.Although it is possible to use EOF as the primary pumping mechanism in a mFIA manifold, it may be preferable or even necessary to isolate the pumping mechanism from the injector, reactor and detector components of the system. Such occasions might be, for example, when the chemistry of the method or detector system is not compatible with the electrical field required for direct EOF pumping. One such system has been described by Dasgupta and Liu45 in which a section of polyimide-coated fused-silica tubing (40 cm 3 75 mm id) was connected to a second capillary, via an isolating membrane.In the first channel a 2 mm borax buffer is pumped by direct EOF, with a field strength of around 40 V cm21. As the flow in the pumped capillary was electrically isolated from the second capillary it produced a hydrodynamic effect, sufficient to create a flow in the second channel, which could be varied between 1 nl min21 and 100 ml min21. In this case the hydrodynamic effect of the EOF is exploited in a mFIA manifold without the need for a direct electrical field.Injector Design The introduction or injection of a sample into an FIA manifold can generally be classified into two general types. The first is the timed or gated injection in which a sample is drawn into the FIA manifold, usually through a sampling probe, for a controlled period of time, after which the flow is switched back to the carrier stream. In this way, a variable volume can be injected into a manifold as a function of time.The second and more usual method of injection in FIA, is the introduction of a constant sample or reagent volume into a mobile carrier in a way that affords minimal flow dispersion. The most common approach for such systems is to use a rotary or slide valve, which contains a sample loop of a defined volume. It would therefore seem appropriate to have both these forms of sample injection available in mFIA systems, and this is indeed found to be the case.The injection of a sample into a mFIA manifold can be performed using hydrodynamic/pneumatic pressure control,46 miniaturised valves1 or electrokinetic mobility based on EOF.15,18,59–62 Even the lowest volumes obtainable with traditionally based rotary-type valves63 are clearly too large for practical use in mFIA. However, Liu and Dasgupta1 recently described the use of commercially available valves with an injection volume of 60 nl.It should be stressed, however, that most of the work relating to the introduction of small sample volumes ( < 20 nl) into capillary systems has been focused on CE methodology rather than addressing mFIA systems. Of particular relevance to mFIA systems is the development of the so-called valveless injection method, which uses EOF control in conjunction with specific capillary channel geometries.64,65 The two simplest approaches to sample injection based on EOF are first to pump a sample for a given period of time into a flow channel and second to fill a defined volume (equivalent of a sample loop) built into the mFIA manifold.In the former case, Zheng and Dasgupta55 described a modified CE system in which an organic solvent was loaded for a given period of time into a mFIA system consisting of a 50 cm 3 75 mm id polyimide Analyst, January 1997, Vol. 122 5Rcoated fused-silica capillary. The solvent was introduced into the system by placing the capillary electrode in the solvent reservoir at 15 kV for 10 s.Following the injection step, the capillary electrode was switched to a borate buffer which subsequently pumped the solvent through the capillary to the detector. Using this approach, the authors reported an RSD of 0.51% associated with migration effects and 1.7% for the peak, suggesting that surface tension, viscosity and flow rate will influence this particular mode of injection. It should be noted that in this work, the authors were using the method described for introducing a solvent into the mFIA system for the purpose of solvent extraction.The second approach to sample introduction, based on defined volumes using EOF, falls into two categories, the Xand Z-type injections (Fig. 4). Of the two injection geometries, the X or cross design, has been most widely investigated. Depending on the geometry and electrical field used, the sample injection can be classified as ‘floating’ (gated) or ‘pinched’ (discrete), with the former occasionally being referred to as the continuous mode.59,60,64,65 Experimentally, the simplest of these two modes is the ‘floating’ method, as it only requires one pair of electrodes (Fig. 5). In this case the sample is pumped by an EOF from the sample reservoir (A the positive electrode) to the sample waste reservoir (B the negative electrode) along channel AB crossing part of channel CD. Note that the channel AB will have been filled prior to the sample introduction with a suitable buffer.The reservoirs C and D contain no electrodes and therefore have no applied field, hence their potential is floating relative to the field in channel AB. As the sample passes across the intercept of AB and CD, diffusion and eddy effects will allow some of the sample to migrate in the direction of both C and D [Fig. 5(a)]. Subsequently, on placing the electrodes in reservoirs C and D the field can be made to run in the CD direction, so any sample molecules in the intercept will be pumped towards D, the grounded reservoir [Fig. 5(b)], allowing the sample to pass via a detector on its way.Clearly, we can see that the leakage or migration of the sample solution into channel CD during injection and the possibility of dragging the stationary sample from channels A and B into D as a function of flow [Fig. 5(c)] will lead to an uncontrolled volume injection. Examples of this effect have been reported.59,60,64 The leakage observed into the main channel may not be a serious problem for certain applications and clearly it offers a simple method of operation, in which only one pair of electrodes are required.However, the quality of etch, surface topography and geometry of the channels concerned will influence the degree of diffusion using the ‘floating’ injection method. The need, especially in CE, to have more precise control over the sample volume injected has lead to the development of the so-called ‘pinch’ method (Fig. 5). In this case, the sample is once again pumped under an EOF from reservoir A to B, but this time reservoirs C and D are not electrically floating, but like A are given a positive potential, relative to the waste reservoir B [Fig. 5(d)]. The effect of this is to draw buffer from channels C and D into B along with the sample from A. Under these flow dynamics, the sample gains a pinched or trapezoidal shape at the intercept of channels AB and CD.When the flow is redirected towards D [Fig. 5(e)], the sample volume, which may be only a few picolitres, is of a more precisely defined volume, less affected by diffusion effects. Jacobson et al.60 have reported improvements in the RSD from 2.7 to 1.7% for a 90 pl injection volume using the ‘pinch’ method, compared with the ‘floating’ approach. In this particular study, the applied voltages based on a 1 kV power supply were reservoir A 90, B 0, C 90 and D 100%, giving channel field strengths of 270, 20, 400 and 690 V cm21 in A, B, C and D, respectively.Clearly, operating the manifold in a multi-electrode configuration will reduce dispersion effects at intersections and allow improved control of the sample injection volume. The concept of the ‘push-back distance’ associated with surface diffusion has been used66 to define the migration rate and shape of flow patterns at X-type intercepts. This approach to sample injection has considerable potential in controlling sample volumes as a function of applied field.The larger volumes (9–22 nl) of the Z-type injection (Fig. 6) are introduced by pumping the sample between two reservoirs across a defined volume in a Z geometry. In the example shown, the sample is pumped from reservoir C or D to the waste reservoir D or E, so filling a defined volume in channel AX. The loaded sample can then be reacted with reagents in channel AX and moved to a detector or be monitored in situ.The Z mode of injection produces a larger volume than that obtained by the X mode, which in turn will be less influenced by the diffusional effects described previously.18,67 Recent work in the author’s laboratory using the Z injection technique has indicated that Fig. 4 Discrete volume injection using (a) the X and (b) the Z approach. Fig. 5 Flow and dispersion characteristics of floating and pinched modes of injection. 6R Analyst, January 1997, Vol. 122pinch control offers no improvement in precision over the floating method.68 The ability to matrix modify a sample in an injector system could be an attractive option in mFIA and techniques such as synchronized cyclic capillary electrophoresis22,69 or sample fractionation67 may offer considerable scope for such methodology. In these systems selected, analyte fractions can be separated from a more complex and possibly interfering matrix, owing to the variation in migration and flow direction under electrophoretic conditions.The switching of electrodes can be used to move components along channels and across channel intersections akin to shunting railway trucks, until the analytes required become isolated from matrix or interfering components. This aspect of sample manipulation is potentially very exciting for mFIA system development and will no doubt be exploited in future applications. Reactor Design Having satisfactorily injected a sample into an FIA system, the next objective is usually to present the sample to a detector in an appropriate form as quickly as possible, using tube or channel geometries that minimize the dispersion of the sample slug.In mFIA systems sample diffusion is known to be a function of the square root of the time after the injection of a sample.69 There remains one additional and often complex step, however, that of achieving the appropriate chemistry or biochemistry necessary for the detection of the analyte of interest.This process typically requires the addition of at least one reagent and can include more complex steps such as in-line solid-phase extraction and multiphase separations.70 Indeed, the complexity of the FIA manifold is only limited by the imagination or ingenuity of the analyst. The first and perhaps the most important step in reactor design is a consideration of the physical and chemical characteristics that occur when two solutions combine at a channel junction in a mFIA manifold. From the previous section, it is clear that even at Reynolds numbers less than 2000 some turbulent mixing takes place at channel intercepts and that the electrical potentials at which the channels are held can significantly influence the interfacial or mixing zone.Turbulent mixing, for example, has been reported in channels 5.2 mm deep and 57 mm wide.71 The characterization of channel intersections indicates that leakage or diffusion of reagents from a ‘floating’ side stream into a channel in which EOF is present is around 2–5%, depending on the viscosity and flow rates.64 This bleed occurs as a result of hydrodynamic effects, in which the molecules in the pumped stream entrain the stationary side channel molecules owing to frictional and eddy properties in a Venturi- or Bernoulli-type effect.What is interesting with such systems is that the flow characteristics in the main channel will have a flat, trapezoidal profile whereas the side channel, whose flow will be pressure driven, will have a parabolic profile.Hence the profiles of the merging streams may have a significant effect on the mixing characteristics at such an interface. As one of the main objectives of reagent addition in an FIA system is to induce mixing and so achieve the required reaction, some consideration needs to be given to this aspect of pumping in mFIA systems. In a valuable study, Seiler et al.44 demonstrated that the individual electrical resistances in a series of interconnecting channels can be used to predict the flow characteristics, in a similar way to hydrodynamic estimations in conventional FIA.The basic concept considers the intercepting channels as electrical resistors and evokes Kirchhoff’s rules,72 to predict the net flow of current which can be attributed to the flow dynamics of the device. One of the important findings of the work by Seiler et al.44 is that by controlling channel voltages one is able to manipulate sample or buffer dilutions by adjusting flow rates, and this offers considerable scope for stopped-flow or reverseflow operations.This effect was also recently demonstrated in our laboratory using a manifold previously described for phosphate analysis.18 Figure 6(a) illustrates how a sample of orthophosphate, held in reservoir D, is injected into channel AX using the Z technique, by pumping the sample to reservoir E.Thus the slug in channel AX defined by the intercepts DE represents the injected sample which reacts with ammonium heptamolybdate (0.01 m) in the presence of ascorbic acid (0.05 m) to produce molybdenum blue. Detection of the molybdenum blue is achieved by measuring the absorbance at 744 nm in a spectrophotometer coupled to the manifold by a fibre-optic system attached along channel AX (optical path 2 cm). Through the selection of different injection volumes [i.e., C–D (9 nl), D– E (13 nl) and C–E (22 nl)] and variation in the positive voltage applied at reservoirs A and B relative to D or C, the absorbance signal was observed to decrease as the intercept ‘pinch’ potential applied across A–E and B–E became sufficient to dilute the formation of the coloured complex at each end of the injection slug.68 In effect, the flow rate in channel A–B was being increased as a function of voltage, so diluting the flow from channels D–E, C–E or C–D into channel AX.Results obtained using this approach (Fig. 7) indicated that calibration based on one standard is possible and that the technique will have an important role to play in the development of future methodology. Following the mixing of the reagent and analyte streams, some period of time is usually required to produce sufficient product, either for further reactions or for subsequent detection. Conventionally, this hold time is effected by using a coil or knotted reactor and/or a stopped-flow mode, in order to minimize dispersion whilst achieving the required reaction chemistries. In mFIA it may be that owing to the more efficient interfacial nature of the mixing, i.e., the need for less bulk mixing, reactions rates may increase; however, reactions are still likely to require a finite period of time for product production.A serpentine-type pattern or structure offers a very simple, but effective, use of space for extending channel length and is compatible with photolithiographic fabrication techniques.The flow characteristics of serpentine channels have been studied and whilst disruption to the electroosmotic flow Fig. 6 Controlled dilution experiments using a pinch voltage. Analyst, January 1997, Vol. 122 7Rwas observed at the 90° bends, no significant band broadening has been found.60 The particular design used had a total channel length of 17 cm operating with a field strength of 700 V cm21 and was used to perform electrophoretic separations.It would seem that using such an approach, time-dependent mFIA reactions could be accommodated in a physically small area with minimal dispersion through the use of serpentine or spiral channels.61 Modification to the surface properties to induce selective flow characteristics, given the limitations of producing a double layer and hence EOF, has attracted some interesting approaches. These include the use of reversed-phase hydrophobic polymers, 73 surfactants,74 silanals and quaternary ammonia groups.75 In addition, the covalent bonding or immobilisation of glucose oxidase has been used to develop a glucose mFIA method based on a serpentine geometry using a 260 cm long 3 100 mm wide 3 70 mm diameter reactor.76 One of the interesting extensions of the mFIA technology described is the development of electrochromatographic (EC) separations in which capillaries are packed with small particles (approximately 3 mm) on which efficient separation can be achieved.77 Using EOF as the primary pumping mechanism, EC can be mediated with reference to the zeta potential and hence separation is potentially possible through selection of the appropriate surface properties of a packing material.Thus the walls of the capillary act to facilitate the primary pumping mechanism and the packed chromatographic material offers enhanced electrophoretic separations. In addition to surface modification, the physical effect of field flow fractionation may also be incorporated as a separation process in capillary systems.78 In this case, an external force, e.g., magnetism or gravity, is used to pull fractions to the wall of a channel; on removing the force, a gradual diffusion of the fractions in the sample occurs back into the flowing stream, usually based on molecular size.79,80 As yet field flow fractionation has not been widely studied in mFIA systems, but the possibility of using field flow fractionation and modifying the zeta potential using a magnetic field might be an interesting subject for future research.Detector Design Much of the research reported to date in the area of mFIA has focused on characterising the physical processes of reagent mobilisation and mixing in microchannel manifolds. Such work has relied heavily on imaging techniques, using, for example, charged coupled device (CCD) cameras60 and microscopybased techniques.30,44,61,81 In this section, consideration is given to the design of detectors suitable for direct integration into EOF based mFIA systems.The major detector systems used in conventional FIA are based on optical absorption/emission and electrochemical techniques,70 and this is also found to be the case for mFIA systems.82,83 Clearly, if electroosmotic based pumping is employed in the mFIA manifold, then some care is required with the design of electrochemical detectors, but this does not pose any real serious limitations.83 Not surprisingly, optical detection has proved to be the most attractive approach to on-device detection and various examples can be found in the literature,15,40,45,67 mainly associated with CE detectors.84 One of the less obvious advantages of miniaturisation is the ability to introduce detector systems not readily available to conventional larger flow systems, such as mass-selective devices of the surface acoustic wave type.85 Clearly, the design of a flow cell in terms of physical volume and optical geometry is very important in miniature systems if sensitivity, reliability and robustness are to be achieved.The incorporation of fibre optics as part of an optical detection system has proved to be of value, particularly with the advent of fibres with diameters of the order of 0.1 mm.86 In the area of optical detectors, some effort has gone into developing axially rather than perpendicularly oriented measurement cells.87 Increasing the volume of the flow cell in a perpendicular viewing axis has been used to increase the optical pathlength,88 whilst the use of multiple reflections axially in a flow cell has also been evaluated.89 The first example led to severe dispersion effects and a subsequent loss in sensitivity and the second approach, based on a silicon device, suffered from signal loss due to scatter and absorption of light at the silicon surface.One interesting approach to absorbance measurements is to consider the mFIA channel as an extension of a fibre optic system.18,90 In this way, total internal reflection of light may be achieved and any photoactive species spatially present in the channel will undergo interaction with the photons present, to produce either a direct absorption measurement or a subsequent fluorescence effect.As with most absorption methods, it would be preferable to use a dual-channel system to improve stability and reduce scatter.Although a long pathlength is appealing for absorption measurements, emission-based techniques would benefit from a small spatial volumes in which the emission effect can be concentrated, and volumes less than 1 nl have been suggested.89 The size of the detector cell is clearly related to the concentration and sensitivity of the method in question and a flow cell of 15 ml has been reported to be adequate for chemiluminescent measurements of glucose and lactate in human serum.91 More recently Liang, et al.92 have described a UV cell with a parallel flow optical path of 120–140 mm for absorbance and fluorescence detection at the end of a CE column, which would be most suitable for mFIA applications.Sequential detector arrays are an attractive approach in sensor design and lend themselves well to mFIA systems.82,93 One such system82 has been described, based on electrochemical detectors, for liquid chromatographic separations of catecholamines and consisted of four photolithographically prepared ISFET sensors aligned sequentially in a 5 mm silicon channel 100 mm wide and 70 mm deep.The total volume of the detector was 20 ml. Another device,93 using a peristaltic pump to control flow rates, consisted of nine 5 mm ISFET sensors housed in a 15 ml cell, used for pH, potassium and calcium determinations in biological fluids. Although neither of these systems was used in mFIA manifolds, they do illustrate the ability to develop array detectors compatible with mFIA technology. One of the most exciting prospects for mFIA detector design is the ability to incorporate their fabrication into one integrated device.For example, miniature mass spectrometers (3 3 3 3 3 mm) have been produced based on fabrication technology that would be ideally suited for mFIA.94 Miniature spectroscopic systems are also becoming available95,96 and future developments in spectrometers incorporating opto-electronic systems Fig. 7 Calibration based on 100 ppb PO4 with dynamic applied pinch voltage. The equivalent absorbance values for 100, 50 and 25 ppb phosphate are indicated on the x-axis. 8R Analyst, January 1997, Vol. 122will undoubtedly bring valuable complementary technology for future mFIA detector design. Future Trends The preceding sections have tried to focus on the basic concepts and developments relating to mFIA systems based on EOF. Some indication has been given of the likely areas where current and future research may prove to be of great value.These include the fabrication of devices where there is considerable potential for the construction of stacked or three-dimensional systems, possibly using cold bonding and direct laser etching techniques. The mobilisation of reagents and analytes, based on EOF, requires more complete characterisation if flows and mixing effects at intersecting channels are to be effectively exploited. The most important development if mFIA technology is to be fully realised is the fabrication of self contained operational systems with proven application robustness.The close relationship of mFIA technology with separation techniques such as capillary electrophoretic and micro-electrochromatographic separations may well lead to some form of hybride system in the near future. The area perhaps where the greatest advances in mFIA-based technology will be most readily realised is the biotechnology sector, where methodology and applications are complementary to miniature systems. Already examples are emerging of applications in DNA fragment analysis97 and immunoassay methodology.98 Developments, however, need to be focused not only on the integrated device which may be encapsulated and equipped with telecommunication for remote operation, but also to include interfacing to existing measurement systems such as MS or NMR, which would benefit from some form of sample pretreatment.One area which has not yet been considered in mFIA systems is a return to the early gasphase work started by Terry et al.5 Clearly there are some exciting possibilities in gas and multiphase systems yet to be realised. Conclusion Where has mFIA and more generally mTAS got to? In 1991, Manz et al.13 questioned whether the developments in mTAS were ‘a look into next century’s technology or just a fashionable craze’. In their concluding remarks, the authors made some general comments relating to the uptake or acceptance of mTAS technology, ‘namely that changes are required in the political and cultural opinions of analytical work if appropriate financial support is to be forthcoming and that the research base must grow worldwide to foster both competitive and collaborative research’.Further, they identified that ‘the market acceptance of the technology must be embraced through the design and production of mTAS concepts’. Clearly, these requirements have been only partially fulfilled. An examination of the literature indicates that there has been a positive growth in the research base and this will obviously support the fundamental development of the science.What is less obvious is the political and more importantly the economic will to support the development of the technology, and this may yet be seen to be the most seriously limitation to the growth of the science. Although miniaturisation is conceptually appealing, there remain some important technical obstacles to overcome, such as the introduction of ‘real’ samples and the ability to deal with suspended particles.These limitations are not beyond the scope of present day membrane technology, and should pose no serious hindrance to the advancement of the science. Developments in the field of mTAS since 1991 clearly point to the technique forming the basis of future methodology applicable to a wide range of applications ranging from measurement science to chemical synthesis in which mFIA based on EOF flow will play a significant role.The limitation in releasing the considerable potential that micro reactor technology can offer resides not in the technological challenge but in the imagination of our minds and only requires us to realize it. References 1 Liu, S., and Dasgupta, P. K., Anal. Chim. Acta, 1993, 283, 739. 2 Bogue, R., Lab. Equip. Dig., 1995, February, 14. 3 Manz, A., Effenhauser, C.S., Burggraf, N., Verpoorte, E. J. M., Raymond, D. E., and Widmer, H. M., Analusis Mag., 1994, 22, M25. 4 Craston, D., and Cowen, S., Chem. Br., 1996, October, 31. 5 Terry, S. C., Jarman, J. H., and Angell, J. B., IEEE Trans. Electron. Devices, 1979, ED-26, 1880. 6 Capillary Electrophoresis: Theory and Practice, ed. Grossman, P. D., and Colburn, J. C., Academic Press, New York, 1992. 7 Mikkers, F. E. P., Everaerts, F. M., and Verheggen, Th. P. E. M., J. Chromatogr., 1979, 169, 11. 8 Jorgenson, J. W., and Lukacs, K. D., Anal. Chem., 1981, 53, 1298. 9 Whalen-Pechersen, E. K., and Jurs, P. C., Anal. Chem., 1981, 53, 2184. 10 Manz, A., Graber, N., and Widmer, H. M., Sens. Actuators B, 1990, 1, 244. 11 Advances in Chromatography, ed. Brown, P. R., and Grushka, E., Marcel Dekker, New York, 1993, pp. 1–66. 12 Micro Total Analysis Systems, ed. van den Berg, A., and Bergveld, P., Kluwer, Dordrecht, 1995. 13 Manz, A., Fettinger, J. C., Verpoorte, E. J. M., L�udi, H., Widmer, H.M., and Harrison, D. J., TrAC, Trends Anal. Chem. (Pers. Ed.), 1991, 5, 144. 14 Manz, A., Harrison, D. J., Verpoorte, E. M. J., Fettinger, J. C., L�udi, H., and Widmer, H. M., Chimia, 1991, 45, 103. 15 Lui, S., and Dasgupta, P. K., Anal. Chim. Acta, 1992, 268, 1. 16 Harrison, D. J., Glavina, P. G., and Manz, A., Sens. Actuators B, 1993, 10, 107. 17 Jacobson, S. C., Hergenr�oder, R., Koulng, L. B., and Ramsey, J. M., Anal. Chem., 1994, 66, 2369. 18 Daykin, R.N. C., and Haswell, S. J., Anal. Chim. Acta, 1995, 313, 155. 19 Verpoorte, E. J. M., Van der Schoot, B. H., Jeanneret, S., Manz, A., and de Rooij, N. F., in Interfacial Design and Chemical Sensing, ACS Symposium Series, American Chemical Society, Washington, DC, 1994, ch. 21, p. 244. 20 Schomburg, W. K., Vollmer, J., B�ustgens, J., Fahrenberg, J., Hein, H.,z, W., J. Micromech. Microeng., 1994, 4, 1. 21 Sensors: a Comprehensive Survey, ed. Grandke, T., and Ko, W. H., VCH, Weinheim, 1989, vol. 1, p. 107. 22 Manz, A., Verpoorte, E. M. J., Effenhauser, C. S., Burggraf, N., Raymond, D. E. and Widmer, H. M., Fresenius’ J. Anal. Chem., 1994, 348, 567. 23 Woffenbuttel, R. F., Sens. Actuators A, 1992, 30, 109. 24 Altmann, J. P., Das Neue Lehrbuch de Glus�alzerei, Gentner, Stuttgart, 1963, p. 29. 25 Daniel, J. S., and Delapierre, G., J. Micromech. Microeng., 1991, 1, 187. 26 Becker, E. W., Ehrfeld, W., Hagmann, P., Maner, A., and M�uchmeyer, D., Microelectron.Eng., 1986, 4, 35. 27 Knapp, J., Andreae, G., and Petersohn, D., Sens. Actuators A, 1990, 21–23, 1080. 28 Kovacs, G. T. A., Petersen, K., and Albin, M., Anal. Chem., 1996, 68, 407A. 29 Jacobson, S. C., Koutny, L. B., Herenr�oder, R., Moore, A. W., Jr., and Ramsey, J. M., Anal. Chem., 1994, 66, 3472. 30 Fan, Z., and Harrison, D. J., Anal. Chem., 1994, 66, 177. 31 Elliot, D. J., Integrated Circuit Fabrication Technology, McGraw- Hill, New York, 1982. 32 Thin Film Processes, ed.Vossen, J. L., and Keren, W., Academic Press, San Diego, 1978, ch. V-1, p. 401. 33 van der Linden, W. E., TrAC, Trends Anal. Chem. (Pers. Ed.), 1987, 6, 37. 34 Tyssen, R., Anal. Chim. Acta, 1980, 114, 71. 35 Hungerford, J., Thesis, University of Washington, 1986. 36 Gravesen, P., Branebjerg, J., and Jensen, O. S., J. Micromech. Microeng., 1993, 3, 168. 37 Manz, A., and Simon, N., Anal. Chem., 1987, 59, 74. Analyst, January 1997, Vol. 122 9R38 Micro Total Analysis Systems, ed. van den Berg, A., and Bergveld, P., Klewer, Dordrecht, 1995, p. 165. 39 Shoji, S., and Esashi, M., Sens. Actuators B, 1992, 8, 205. 40 Field, L. A., and Muller, R. S., Sens. Actuators A, 1990, 21–23, 935. 41 Fung, C. D., Cheung, P. W., and Fleming, D. G., Microachining and Micropacking of Transducers, Elsevier, Amsterdam, 1985. 42 Harz, M., J. Micromech. Microeng., 1992, 2, 161. 43 Esashi, M., Nakano, A., Shoji, S., and Hebiguchi, H., Sens. Actuators A, 1990, 21–23, 931. 44 Seiler, K., Fan, Z. H., Fluri, K., and Harrison, D. J., Anal. Chem., 1994, 66, 3485. 45 Dasgupta, P. K., and Liu, S., Anal. Chem., 1994, 66, 1792. 46 Li, S. F. Y., Capillary Electrophoresis, Elsevier, New York, 1992. 47 Shaji, S., Esashi, M., van der Schoot, B., and de Rooij, N. F., Sens. Actuators A, 1992, 32, 335. 48 Horowitz, P., and Hill, W., The Art of Electronics, Cambridge University Press, Cambridge, 2nd edn., 1989, p. 368. 49 van der Schoot, B. H., Jeanneret, S., van den Berg, A., and de Rooij, N. F., Anal. Methods Instrum., 1993, 1, 38. 50 Stewart, K. K., Beecher, G. R., and Hare, P. E., Fed. Proc., Fed. Am. Soc. Exp. Biol., 1974, 33, 1434. 51 Moriney, R. M., White, R. M., and Howe, R. T., Proc. MEMS’91, Nara, Japan, 1991, p. 277. 52 Richter, A., Plettner, A., Hofmann, K. A., and Sandmeier, H., Proc. MEMS ’91, Nara, Japan, 1991, p. 271. 53 Osada, Y., Okuzzaki, H., and Hari, H., Nature (London), 1992, 355, 242. 54 Fuhr, G., Fiedler, S., M�uller, T., Schnell, T., Glasser, H., Lisec, T., and Wagner, B., Sens. Actuators A, 1994, 41–42, 230. 55 Zheng, H. J., and Dasgupta, P. K., Anal. Chem., 1994, 66, 3997. 56 Adams, R. N., Electrochemistry at Solid Electrodes, Marcel Dekker, New York, 1969. 57 Masson, D., Magdassi, S., and Sasson, Y., J. Org. Chem., 1990, 55, 2714. 58 Moore, A. W., Jr., Jacobson, S. C., and Ramsey, J. M., Anal. Chem., 1995, 67, 4184. 59 Jacobson, S. C., Hergenr�oder, R., Koutny, L. B., and Ramsey, J. M., Anal. Chem., 1994, 66, 1114. 60 Jacobson, S. C., Hergenr�oder, R., Koutny, L. B., Warmack, R. J., and Ramsey, J. M., Anal. Chem., 1994, 66, 1107. 61 Seiler, K., Harrison, D. J., and Manz, A., Anal. Chem., 1993, 65, 1481. 62 Manz, A., Effenhauser, C. S., Burggraf, N., Harrison, D. J., Seiler, K., and Fluri, K., J. Micromech. Microeng., 1994, 4, 257. 63 Tseida, T., Mizuno, T., and Akiyama, J., Anal. Chem., 1987, 59, 799. 64 Harrison, D. J., Manz, A., Fan, Z., L�udi, H., and Widmer, H. M., Anal. Chem., 1992, 54, 1926. 65 Manz, A., Harrison, D. J., Verpoorte, E. M. J., Fellinger, J. C., Paulus, A., L�udi, A., and Widmer, H. M., J. Chromatogr., 1992, 593, 253. 66 Effenhauser, C. S., Paulus, A., Manz, A., and Widmer, H. M., Anal. Chem., 1994, 66, 2949. 67 Effenhauser, C. S., Manz, A., and Widmer, H. M., Anal. Chem., 1995, 67, 2284. 68 Hasser, A., MSc Thesis, University of Hull, 1996. 69 Burggraf, N., Manz, A., Verpoorte, E. M. J., Effenhauser, C. S., Widmer, H. M., and de Rooij, N. F., Sens. Actuators B, 1994, 20, 103. 70 Ruzicka, J., and Hansen, E. H., Flow Injection Analysis, Wiley, New York, 2nd edn., 1988. 71 Halliday, D., and Resnick, R., Physics, Wiley, New York, 1960. 72 Jacobson, S. C., Koutny, L. B., Hergenr�oder, R., Moore, A. W., Jr., and Ramsey, J. M., Anal. Chem., 1994, 66, 3472. 73 Monning, C. A., and Jorgenson, J. W., Anal. Chem., 1991, 63, 802. 74 Yao, X. W., Wu, D., and Regnier, F. E., J. Chromatogr., 1993, 636, 21. 75 Smith, J. T., and Rassl, Z. E., J. High Resolut. Chromatogr., 1992, 15, 573. 76 Marakami, Y., Suda, M., Yokogama, K., Tukeuchi, T., Tamiya, E., and Kurube, I., Microchem., 1994, 49, 319. 77 Tsuda, T., Nomura, K., and Nakagawa, G., J. Chromatogr., 1982, 248, 241. 78 Giddings, J. C., Science, 1993, 260, 1456. 79 Widmer, H. M., and Koch-Kellner, L., Chimia., 1989, 43, 320. 80 Koch, L., Koch, T., and Widmer, H. M., J. Chromatogr., 1990, 517, 395. 81 Raymond, D. E., Manz, A., and Widmer, H. M., Anal. Chem., 1994, 66, 2858. 82 Sudfa, M., Sakuhara, R., and Karube, I., Appl. Biochem. Biotechnol., 1994, 41, 3. 83 Haber, C., Sclvestri, I., B�o�osli, S., and Seman, W., Chimia, 1991, 45, 117. 84 Chervet, J. P., Ursem, M., Salzmann, J. P., and Vannoort, R. W., LC– GC, 1989, 7, 515. 85 Wohltjen, J., Sens. Actuators, 1984, 5, 305. 86 Ton, W., Zhang-You, S., Smith, S., Birnbaum, D., and Kopelman, R., Science., 1992, 258, 778. 87 Xi, X., and Yeing, F. S., Anal. Chem., 1990, 62, 1580. 88 Xue, Y., and Yeung, E. S., Anal. Chem., 1994, 66, 3575. 89 Verpoorte, E. M. J., Manz, A., L�udi, H., Bruno, A. E., Maystre, F., Kruttiger, B., Widmer, H. M., van der Schoot, B. H., and de Rooij, N. F., Sens. Actuators B, 1992, 6, 66. 90 Weigl, B. H., and Woelfbeis, O. S., Anal., Chem., 1994, 66, 3323. 91 Suda, M., Sakuchara, T., and Karube, I., 2nd Bioengineering Symposium, Tsukuba, Japan, 1992, p. 113. 92 Liang, Z., Chiem, N., Ocvirk, G., Tang, T., Fluri, K., and Harrison, D. J., Anal. Chem., 1996, 68, 1040. 93 Arquint, P., Koudelka-Hep, M., van der Schoot, B. H., van der Wal, P., and de Rooij, N. F., Clin. Chem., 1994, 40, 1805. 94 Micro Total Analysis Systems, ed. van den Berg, A., and Bergveld, P., Kluwer, Dordrecht, 1995, p. 299. 95 Conzen, J. P., B�urch, J., and Ache, H. J., Appl. Spectrosc., 1993, 47, 753. 96 Ache, H. J., Chem. Ind., 1993, 1, 40. 97 Jacobson, S. C., and Ramsey, J. M., Anal. Chem., 1996, 68, 720. 98 Koutny, L. B., Schmalzing, D., Taylor, T. A., and Fuchs, M., Anal. Chem., 1995, 68, 18. Paper 6/06289J Received September 16, 1996 Accepted November 18, 1996 10R Analy

ISSN:0003-2654    

DOI:10.1039/a606289j

出版商:RSC    

年代:1997

数据来源: RSC

摘要:

Effect of Storage on the Recovery of Different Types of Pesticides Using a Solid-phase Extraction Method C. de la Colina, F. S�anchez-Rasero, G. Dios, E. Romero and A. Pe�na* Estaci�on Experimental del Zaid�ýn (CSIC), Profesor Albareda, 1, E-18008 Granada, Spain Recoveries of different pesticide groups after storage either on C18 cartridges or as dried residues from organic solutions, and their analysis by gas chromatography with electron capture and flame photometric detection, were studied.Two storage temperatures, 4 and 218 °C, and three storage periods, 3, 7 and 30 d, were considered. The effect of storage temperature and storage time on the recovery of 27 pesticides in water was investigated. In general, the pesticide recoveries were !70% after 30 d of storage at 218 °C on C18 cartridges. Exceptions included captan and folpet. The storage of the dried residues generally did not affect the pesticide recovery when kept at 218 °C for up to 30 d. Keywords: Sample handling; sample storage; pesticide stability; water; solid-phase extraction; gas chromatography For analytical data to be valid, they should reflect the concentration of pesticides at the time of sampling, but often in a laboratory, owing to temporary shortages of personnel, problems with or breakdown of analytical equipment or sudden unexpected requirements for equipment to be used for other work, samples must be stored for variable periods before their analysis.For this purpose, and also when samples must be transported to other national or foreign laboratories, it is essential to know how long the content of a sample may remain unchanged.Therefore, stability studies of pesticides and organic contaminants in water and other matrices are important. In a recent report from the US Environmental Protection Agency (EPA),1 in which 147 pesticides in water samples were checked for stability for at least 14 d at 4 °C, 26 pesticides were removed from the list because of a 100% loss, even after biological inhibition of the water microorganisms. Research on the storage of pesticides in different water samples has been undertaken,2–5 that confirmed the instability of many pesticides in natural waters, depending on microbial degradation, hydrolysis and photolysis.To compensate for this instability, various stabilizers or preservatives have been added to aqueous samples, e.g., methanol,6 dichloromethane,7 dilute acid solutions1,8 or HgCl2,1,9 or the sample has been freezedried10 to retard the decomposition of the constituent chemicals.An alternative could be the preservation of the pesticides retained in solid-phase extraction (SPE) cartridges or discs, which are being increasingly used in environmental laboratories, 11 a possibility already used for hydrocarbon samples and some pesticides.9–17 The compounds of interest have been effectively preserved from microbial degradation for up to 54 d on these adsorbents12 and photolysis is avoided because the discs and cartridges are usually stored in the dark.It has been speculated that this preservative effect is a result of the trapping of the organics within the lattice structure of the adsorbent, although protection against hydrolysis is still in question.15,17 In addition, the processed samples may be preserved as organic extracts.1 It was reported that analytes generally remained stable in stored sample extracts, although no information was given about how long and for which analytes this kind of storage was effective.In this work, 27 pesticides, including nine organophosphorus, six organochlorine, a carbamate, a pyrethroid and another seven pesticides of miscellaneous groups, were studied at concentrations between 20 and 100 ng l21. Two ways of storing the different pesticides present in water samples were considered: on C18 SPE cartridges and as organic extracts concentrated to dryness.Experimental Reagents All the pesticides were of > 98.5% purity and all the solvents were of pesticide residue analysis grade. C18 extraction cartridges (J. T. Baker, Phillipsburg, NJ, USA) with 500 mg of packing material were used. Cartridge Storage Water samples of 1 l, obtained from a Milli-Q purification system (Millipore, Bedford, MA, USA), were fortified with 20–100 ng of the chemicals (see Table 1). The cartridge was conditioned by rinsing it with the eluents in reverse order of elution, then with methanol and water.20 After the passage of the 1 l fortified water sample, the cartridge was air dried for 30 min.Solid-phase cartridges on which pesticides had been retained were wrapped in Parafilm and stored, at 4 or 218 °C, for 3, 7 and 30 d. At the end of the storage period they were eluted with ethyl acetate and isooctane and evaporated to dryness under a gentle stream of nitrogen. The final residue was taken up with 1 ml of hexane and the internal standard, bromophos, was added before injection into the gas chromatographs.21 Concentrated Extract Storage A mixture of the pesticides in the amounts shown in Table 1 was added to 4 ml of a mixture of the eluents [ethyl acetate– isooctane (1 : 1)] and then concentrated to dryness under a gentle stream of nitrogen. The glass tubes containing the dried eluates were fitted with a glass stopper, covered with Parafilm and stored at the same temperatures and for the same periods of time as used for the cartridges.The dried residue was treated as for the cartridge. For both storage treatments, cartridges and concentrated extracts, the remainder of the analytical process was finished rapidly with injection into the chromatograph on the same day. Statistical Analysis A completely randomized design with three replications of the whole analytical procedure, with three injections per replicate, of two storage temperatures and three storage times was Analyst, January 1997, Vol. 122 (7–11) 7employed for each storage treatment. Mean percentage recoveries were calculated and separated by Fisher’s least significant difference (LSD) at the 0.05 level of significance. Analytical Methodology All the samples were quantified by GC.20,21 Organophosphorus pesticides were determined using a Hewlett-Packard (Avondale, PA, USA) gas chromatograph with a flame photometric detector (FPD), provided with an HP-1 capillary column (12 m 3 0.2 mm id, 0.33 mm film thickness) with the following oven temperature programme: 45 °C (1 min), increased at 30 °C min21 to 170 °C (2 min), at 4 °C min21 to 200 °C (2 min) and at 20 °C min21 to 270 °C (2 min).The injector and detector temperatures were 250 and 275 °C, respectively. The other pesticides were determined using a Hewlett- Packard gas chromatograph with an electron capture detector (ECD), in which an Ultra-2 capillary column (25 m 3 0.32 mm id, 0.17 mm film thickness) was installed, with the following oven temperature programme: 160 °C (1 min), increased at 4 °C min21 to 230 °C (2 min) and at 20 °C min21 to 280 °C (6 min).The injector temperature was 250 °C and the detector temperature was 300 °C. Results and Discussion The pesticides included in this study, were tested at concentrations lower than or equal to the maxima allowed by EU legislation. Such concentrations are at least 102–103 times more dilute than the values previously reported in some storage studies using extraction discs15,16 or cartridges.17 Therefore, the amounts used in this study approach legal conditions and avoid higher concentrations which can influence degradation, as has already been reported for some organophosphorus and carbamate pesticides in soil columns.22 Fig. 1 shows the separation of a pesticide standard solution, with the different detection methods, at the concentrations indicated in Table 1. Cartridge Storage The recoveries obtained for the pesticides retained on the C18 cartridges are given in Table 2 for the pesticides in which no interaction between storage time and temperature was encountered and in Table 3 for the chemicals with a significant interaction between the storage factors studied.As can be seen for the effect of temperature (Table 2), only alachlor, capA-DDE and chlorpyrifos-ethyl show significant differences for the two temperatures studied. Nevertheless, for most of these chemicals only small differences in the recoveries for the two storage temperatures are observed (@10%).Captan, and the chemically related folpet, exhibited an important reduction in recovery, higher for the freezer (218 °C) than for the refrigerator (4 °C), for which no explanation has Table 1 Pesticide amounts used for both storage treatments and their physico-chemical properties. Data from ref. 18 Amount Vapour Water added/ pressure/ solubility/ Pesticide ng* mPa Log Kow mg l21 Trifluralin 20 9.5 4.0† 0.221 Lindane 20 5.6 3.7† 7.3 Triallate 40 16 na‡ 4 Alachlor 80 2.9 na 242 Captan 40 1.3 2.5† 3.3 Folpet 40 1.3 3.1 1 o,pA-DDE 20 na na na p,pA-DDE 20 0.025 5.7–7.0† na Oxyfluorfen 20 0.0267 4.5 0.116 o,pA-DDT 20 na 5.8† na p,pA-DDT 20 na 6.2–6.9† na Bromopropylate 20 0.011 5.4 < 0.5 Dicofol 40 0.053 4.3 0.8 Tetradifon 20 3.231025 4.6 0.08 Deltamethrin 40 0.002 4.6 < 0.0002 Dimethoate 100 1.1 0.7 23800 Fonofos 100 28 3.9 13 Diazinon 100 12 3.3 60 Formothion 100 0.113 na 2600 Fenitrothion 100 18 3.4 21 Malathion 100 5.3 2.9† 145 Fenthion 100 0.74 4.8 4.2 Chlorpyrifos 100 2.7 5.3† 1.4 Methidathion 100 0.25 2.2 200 Phosmet 100 0.065 3.0 25 Azinphos-methyl 100 0.18 3.0 28 Phosalone 100 < 0.067 4.3† 1.7 * The same amount was added to 1 l of water or to 4 ml of the desorption solution. † Data from ref. 19. ‡ Not available. Table 2 Effect of storage temperature and storage time on recoveries of pesticides retained on C18 cartridges Storage Storage temperature/°C* time/d† Pesticide 218 4 LSD‡ 3 7 30 LSD‡ Trifluralin 90 92 86 89 101 5.1 Lindane 96 97 95 97 98 Triallate 89 86 96 86 82 6.6 Atachlor 94 98 2.7 96 95 97 Captan 58 73 9.1 74 66 58 11.2 Folpet 38 50 50 41 47 o,pA-DDE 69 64 4.0 63 72 65 4.9 p,pA-DDE 67 66 64 67 68 Oxyfluorfen 85 84 81 84 88 3.5 o,pA-DDT 67 69 63 71 70 4.9 p,pA-DDT 70 69 68 71 70 Bromopropylate 84 83 84 84 84 Dicofol 87 86 85 86 87 Tetradifon 95 98 100 96 94 4.0 Diazinon 87 87 89 89 84 3.8 Fenitrothion 94 95 99 97 88 4.0 Malathion 94 94 98 98 87 3.6 Chlorpyriphos-ethyl 82 79 2.4 85 83 72 2.9 Phosmet 95 96 103 87 97 9.7 * Average recovery for nine observations (three injections per observation).† Average recovery for six observations (three injections per observation). ‡ LSD = least significant difference (P < 0.05). The LSD values are provided for those compounds with a significant difference between the means. Table 3 Effect of temperature and time of storage on recovery of pesticides retained on C18 cartridges Recovery (%)* Refrigerator Freezer (4 °C) (218 °C) Pesticide 3 d 7 d 30 d 3 d 7 d 30 d LSD† Fonofos 75 68 55 79 64 65 7.9 Methidathion 105 98 89 101 103 99 6.9 Azinphos-methyl 121 102 129 119 114 167 17.8 Phosalone 102 96 87 110 93 98 6.9 * Average recovery for three observations (three injections per observation). † See Table 2. 8 Analyst, January 1997, Vol. 122been found. In a previous study, stability problems with captan were reported (54% recovery on a C18 disc at 4 °C and 32% at 218 °C after 30 d of storage) and ascribed to hydrolysis under the C18 packing of the SPE disc15 and, later, to volatilization,17 although the vapour pressure of this fungicide ( < 1.3 mPa, Table 1) is lower than that of some other pesticides included in this study, which do not show a loss.In addition, the captan recovery is also affected by the storage time (Table 2), so this method of preservation is not appropriate for this pesticide.When captan was stored on extraction discs,15 removal of residual water was recommended. The length of the storage affects several compounds, apart from captan, as can be seen in Table 2. In general, for the pesticides affected (trifluralin, triallate, o,pA-DDE, oxyfluorfen, o,pA-DDT, tetradifon, diazinon, fenitrothion, malathion, chlorpyrifos- ethyl and phosmet), the recovery decreases with increasing storage time, but remains within a ±10–15% variation for the whole length of the storage period.This variation range is within that commonly found in other storage studies.15,17 In addition, the final recovery after 30 d remains over 70% (the minimum recovery required by the EPA regulations23), except for DDEs, with recoveries of !65%, and captan and folpet. Low recoveries have been also reported for DDE and DDT retained on Empore filters for up to 4 weeks, from waters spiked at 50 ng l21.14 Fonofos, methidathion, azinphos-methyl and phosalone showed a significant interaction between temperature and length of storage (Table 3) and therefore both factors cannot be considered individually.For fonofos, methidathion and phosalone, a trend of loss on storage in the refrigerator after 30 d was observed. For fonofos, its storage on a cartridge is not recommended, because its recovery is already below 70% after 7 d at the two temperatures investigated. A complete loss of fonofos, which is the most volatile pesticide included in the present study (Table 1), has been reported for storage in similar cartridges at 4 °C after 1.5 months.17 In the contrast, the same group17 reported complete recovery of fonofos when stored at 220 °C for 8 months.Our results show a 14% loss after 1 month of storage in the freezer. For azinphos-methyl, which is not completely separated from phosalone in the chromatographic column, an artifact appears after 30 d of storage and therefore its quantification was inaccurate.Fig. 2(A) shows the separation of the pesticide mixture with electron capture detection (ECD) after storage in the cartridge for 30 d at 218 °C. The recoveries of the fungicides captan and folpet are clearly affected. The additional peaks correspond to the organophosphorus pesticides that can be detected by ECD. As indicated previously, several organophosphorus pesticides included in the present study (azinphos-methyl, fenitrothion, malathion, diazinon and phosmet) have been eliminated from the US EPA survey list owing to their instability;1 the first three are nevertheless included in the 76/464/EEC Council Directive List of Pesticides to be monitored in the aquatic environment.23 In addition, phosmet has been repeatedly reported to be unstable on storage in water.2,5 These pesticides could be stabilized by storing them on the cartridge packing at 4 or 218 °C for at least 30 d.Nevertheless, certain losses have been observed for some of the pesticides, especially those with low partition coefficients, which confirms the results of Lacorte et al.,17 who pointed to hydrolysis and microbial degradation as the main factors causing instability with this kind of storage. Concentrated Extract Storage The recoveries obtained for the pesticides stored in the dry extract are given in Table 4 for the pesticides in which no interaction between storage time and temperature was encountered and in Table 5 for the chemicals with a statistically significant interaction.For this storage treatment, three additional organophosphorus and one pyrethroid pesticide not considered in the cartridge storage were included: dimethoate, formothion, fenthion and deltamethrin. These compounds could only be recovered between 12 and 59% from water samples after the whole SPE Fig. 1 Gas chromatogram of a standard pesticide solution, at the concentrations shown in Table 1, with electron capture (ECD) and flame photometric detection (FPD).TR, trifluralin; L, lindane; TL, triallate; A, alachlor; IS, bromophos (internal standard); CP, captan; FP, folpet; OAE, o,pA-DDE; PAE, p,pADDE; OX, oxyfluorfen; OAT, o,pA-DDT; PAT, p,pA-DDT; BP, bromopropylate; DF, dicofol; TF, tetradifon; DT, deltamethrin; D, dimethoate; FO, fonofos; DZ, diazinon; FR, formothion; FN, fenitrothion; M, malathion; FT, fenthion; CL, chlorpyriphos; MT, methidathion; PM, phosmet; AZ, azinphos-methyl; and PS, phosalone.Analyst, January 1997, Vol. 122 9process,20 which did not allow an appropriate study of their stability when retained on the cartridge packing. Pesticides stored after evaporation to dryness in general give good recoveries. In this experiment, no real extraction in water was carried out and only the effect of storage on the simulated eluate was studied. Table 4 indicates that for this storage factor, higher recoveries were obtained on storage at 218 °C.Seven out of the 20 compounds listed are affected by storage temperature. Nevertheless, the difference in the recovery for the two temperatures is !10% only for captan, folpet, dimethoate, formothion and fenthion. In Fig. 2(B), decrease in recovery of the last three pesticides after storage for 30 d at 4 °C is shown. Captan and folpet, which already showed instability during cartridge storage, together with the last three organophosphorus pesticides, especially fenthion, which were not included previously, appear to be difficult to stabilize also in this dried form.For captan, the reason cannot be hydrolysis, because no water was present. Volatilization is also possible, but again captan, folpet and fenthion, with differences in recoveries > 25% for the two temperatures considered, have low vapour pressures (Table 1). Molecular structure may be responsible for the instability, as has already been reported for fenthion.5 The effect of storage time for the same chemicals is also shown in Table 4.Many pesticides show a statistically significant difference among the storage periods investigated but, for most of them, the recoveries fall within a 10% variation. Only for dimethoate, formothion and fenthion was the loss Table 4 Effect of storage temperature and storage time on recoveries of pesticides stored after their concentration to dryness Recovery (%) Storage Storage temperature/°C* time/d† Pesticide 218 4 LSD‡ 3 7 30 LSD‡ Alachlor 97 95 101 88 100 3.1 Captan 99 76 5.6 91 90 82 6.8 Folpet 101 74 7.3 93 87 83 o,pA-DDE 97 94 106 89 91 4.0 p,pA-DDE 98 96 1.8 100 89 102 2.2 Oxyfluorfen 96 94 96 96 93 2.7 o,pA-DDT 100 99 98 93 107 2.7 p,pA-DDT 101 98 99 92 108 3.9 Bromopropylate 104 101 101 97 111 6.8 Dicofol 105 98 5.3 97 99 108 6.5 Tetradifon 103 100 100 98 107 5.3 Deltamethrin 102 97 98 96 106 5.5 Dimethoate 93 80 9.9 86 98 75 12.1 Formothion 121 102 7.2 140 117 76 8.8 Malathion 105 105 103 100 111 3.9 Fenthion 74 33 8.0 64 53 43 9.9 Methidathion 101 99 106 99 96 5.0 Phosmet 107 108 109 104 111 Azinphos-methyl 102 103 104 103 101 Phosalone 106 107 105 101 114 5.4 * Average recovery for nine observations (three injections per observation).† Average recovery for six observations (three injections per observation). ‡ See Table 2. Fig. 2 Gas chromatograms of the pesticides stored A, in the cartridge for 30 d at 218 °C, with ECD, and B, in the concentrated extract at 4 °C for 30 d, with FPD.Abbreviations as in Fig. 1. Table 5 Effect of temperature and length of storage on recovery of pesticides stored after their concentration to dryness Recovery (%)* Refrigerator Freezer (4 °C) (218 °C) Pesticide 3 d 7 d 30 d 3 d 7 d 30 d LSD† Trifluralin 84 62 44 96 86 109 6.5 Lindane 78 58 47 93 84 97 5.9 Triallate 87 68 63 96 87 94 4.7 Fonofos 82 65 44 97 87 97 6.9 Diazinon 91 76 49 100 93 97 7.0 Fenitrothion 100 97 88 103 97 100 3.5 Chlorpyriphos-ethyl 101 95 86 98 94 107 6.3 * Average recovery for three observations (three injections per observation).† See Table 2. 10 Analyst, January 1997, Vol. 122> 10%. Nevertheless, except for fenthion, their recoveries were > 70%, which complies with EPA regulations.23 Table 5 shows the trends of a reduction in recovery with time and storage temperature which apply to all of the compounds. For trifluralin, lindane, triallate, fonofos and diazinon, the loss is already clear after 7 d of storage at 4 °C, and is more evident after 30 d with recoveries, in many cases, around 50%. If their vapour pressures are considered, a clear relationship between this property (Table 1) and pesticide loss can be observed.A high degree of correlation was found between vapour pressure and pesticide loss after 30 d at 4 °C for fonofos, triallate and chlorpyrifos. The losses of diazinon, trifluralin and lindane are higher than expected and those of fenitrothion are lower, if the only reason considered to be involved in the process was volatility.For diazinon, an imidoyl phosphate, it has been reported3 that the oxygen–aromatic moiety linkage may be activated and cleaved under both acidic and alkaline conditions to form the corresponding carbonyl compounds with a doublebond shift. This functional group is also present in the structure of chlorpyrifos, so the losses must be due to a mixture of both effects.In contrast, all the pesticides present in Table 5 show recoveries between 94 and 117% at 218 °C after 30 d of storage. It can be concluded from these results that the storage of dried eluates in the refrigerator may be problematic for volatile compounds, as already reported for chlormephos and dichlorvos (7600 and 1600 mPa, respectively) when stored in water samples5 or in disposable SPE cartridges.17 Nevertheless, each case should be studied carefully to verify the observed tendencies.The comparison with physico-chemical data provided in Table 1 indicates that, in general, the most volatile compounds are affected by storage in dried form in the refrigerator, although at 218 °C the decrease in recovery is almost negligible for all of them except for fenthion, which showed great instability, independently of the storage conditions. The storage of the dried eluates in the freezer at 218 °C seems to be a good alternative for these compounds, except for fenthion. Stability problems with triallate, tetradifon, diazinon, fenitrothion, malathion, chlorpyrifos-ethyl and fonofos, which could not be properly controlled when retained on the C18 cartridges, could be solved by storing them as dried eluates for up to 30 d at 218 °C.Conclusions Although immediate extraction and analysis are the best way of obtaining the most accurate residue data, this is not always possible. Therefore, two different storage conditions, corresponding to different stages in the over-all SPE process, were investigated. The results indicated that storage of different pesticide classes on the C18: silica gel surface of solid-phase cartridges for 30 d at 4 or 218 °C is effective for 17 out of the 23 pesticides studied.Triallate, captan, fenitrothion, malathion, chlorpyriphos and fonofos are affected. These pesticides, except for captan and fonofos, can be safely stored in the cartridges for up to 3 or 7 d, and their recoveries after 30 d are still > 70%.All of them, including captan and fonofos, could alternatively be stored for up to 30 d at 218 °C in the dried extract. For this kind of storage, only formothion and fenthion showed considerable losses along the storage period studied. Formerly, only a limited number of 1 or 2 l bottles could be stored in the laboratory. For both treatments considered here, owing to the small size of the cartridges and glass tubes, the space required for their storage has been reduced and a large number of samples can be preserved in a conventional freezer.The authors thank the Comisi�on Interministerial de Ciencia y Tecnolog�ýa (CICYT) for financial support (Project No. NAT91- 0407). Ma. D. Maroto is acknowledged for technical assistance and Ma. D. Mingorance is thanked for her assistance with the statistical analysis. References 1 Munch, D. J., and Frebis, C. P., Environ. Sci. Technol., 1992, 26, 921. 2 Ripley, B. D., Wilkinson, R.J., and Chau, A. S. Y., J. Assoc. Off. Anal. Chem., 1974, 57, 1033. 3 Chau, A. S. Y., Ripley, B. D., and Kawahara, F., in Analysis of Pesticides in Water, ed. Chau, A. S. Y., and Afghan, B. K., CRC Press, Boca Raton, FL, 1982, vol. II, pp. 61–154. 4 Barcel�o, D., Chiron, S., Lacorte, S., Mart�ýnez, E., Salau, J. S., and Hennion, M. C., TrAC, Trends Anal. Chem., (Pers. Ed.), 1994, 13, 352. 5 Lartiges, S. B., and Garrigues, P. P., Environ. Sci. Technol., 1995, 29, 1246. 6 Liska, I., Brower, E. R., Ostheimer, A. G. L., Lingeman, H., and Brinkman, U. A. Th.. Environ. Anal. Chem., 1992, 47, 267. 7 Bourne, S., J. Environ. Sci. Health, 1978, B13, 75. 8 Chau, A. S. Y., and Thomson, K., J. Assoc. Off. Anal. Chem., 1978, 61, 1481. 9 Lopez-Avila, V., Wesselman, R., and Edgell, K., J. Assoc. Off. Anal. Chem., 1990, 73, 276. 10 Barcel�o, D., House, W. A., Maier, E. A., and Griepink, B., Int. J. Environ. Anal. Chem., 1994, 57, 237. 11 Font, G., Ma�nes, J., Molt�o, J.C., and Pic�o, Y., J. Chromatogr., 1993, 642, 135. 12 Green, D. R., and Le Pape, D., Anal. Chem., 1987, 59, 699. 13 Berkane, K., Caissie, G. E., and Mallet, V. N., J. Chromatogr., 1977, 139, 386. 14 Tomkins, B. A., Merriweather, R., Jenkins, R. A., and Bayne, C. K., J. Assoc. Off. Anal. Chem. Int., 1992, 75, 1091. 15 Senseman, S. A., Lavy, T. L., Mattice, J. D., Myers, B. M., and Skulman, B. W., Environ. Sci. Technol., 1993, 27, 516. 16 Johnson, W. J., Lavy, T.L., and Senseman, S. A., J. Environ. Qual., 1994, 23, 1027. 17 Lacorte, S., Ehresmann, N., and Barcel�o, D., Environ. Sci. Technol., 1995, 29, 2834. 18 The Pesticide Manual. A World Compendium, British Crop Protection Council, Thornton Heath, UK, 8th edn., 1987. 19 Noble, A., J. Chromatogr., 1993, 642, 3. 20 de la Colina, C., S�anchez-Rasero, F., Dios, G., Romero, E., and Pe�na, A., Analyst, 1995, 120, 1723. 21 de la Colina, C., Pe�na, A., Mingorance, M. D., and S�anchez- Rasero, F., J.Chromatogr. A, 1996, 733, 275. 22 Belisle, A. A., and Swineford, D. M., Environ. Toxicol. Chem., 1988, 7, 749. 23 Barcel�o, D., J. Chromatogr., 1993, 643, 117. Paper 6/05275D Received July 29, 1996 Accepted September 30, 1996 Analyst, January 1997, Vol. 122 11 Effect of Storage on the Recovery of Different Types of Pesticides Using a Solid-phase Extraction Method C. de la Colina, F. S�anchez-Rasero, G. Dios, E. Romero and A. Pe�na* Estaci�on Experimental del Zaid�ýn (CSIC), Profesor Albareda, 1, E-18008 Granada, Spain Recoveries of different pesticide groups after storage either on C18 cartridges or as dried residues from organic solutions, and their analysis by gas chromatography with electron capture and flame photometric detection, were studied.Two storage temperatures, 4 and 218 °C, and three storage periods, 3, 7 and 30 d, were considered. The effect of storage temperature and storage time on the recovery of 27 pesticides in water was investigated.In general, the pesticide recoveries were !70% after 30 d of storage at 218 °C on C18 cartridges. Exceptions included captan and folpet. The storage of the dried residues generally did not affect the pesticide recovery when kept at 218 °C for up to 30 d. Keywords: Sample handling; sample storage; pesticide stability; water; solid-phase extraction; gas chromatography For analytical data to be valid, they should reflect the concentration of pesticides at the time of sampling, but often in a laboratory, owing to temporary shortages of personnel, problems with or breakdown of analytical equipment or sudden unexpected requirements for equipment to be used for other work, samples must be stored for variable periods before their analysis.For this purpose, and also when samples must be transported to other national or foreign laboratories, it is essential to know how long the content of a sample may remain unchanged. Therefore, stability studies of pesticides and organic contaminants in water and other matrices are important.In a recent report from the US Environmental Protection Agency (EPA),1 in which 147 pesticides in water samples were checked for stability for at least 14 d at 4 °C, 26 pesticides were removed from the list because of a 100% loss, even after biological inhibition of the water microorganisms. Research on the storage of pesticides in different water samples has been undertaken,2–5 that confirmed the instability of many pesticides in natural waters, depending on microbial degradation, hydrolysis and photolysis.To compensate for this instability, various stabilizers or preservatives have been added to aqueous samples, e.g., methanol,6 dichloromethane,7 dilute acid solutions1,8 or HgCl2,1,9 or the sample has been freezedried10 to retard the decomposition of the constituent chemicals. An alternative could be the preservation of the pesticides retained in solid-phase extraction (SPE) cartridges or discs, which are being increasingly used in environmental laboratories, 11 a possibility already used for hydrocarbon samples and some pesticides.9–17 The compounds of interest have been effectively preserved from microbial degradation for up to 54 d on these adsorbents12 and photolysis is avoided because the discs and cartridges are usually stored in the dark.It has been speculated that this preservative effect is a result of the trapping of the organics within the lattice structure of the adsorbent, although protection against hydrolysis is still in question.15,17 In addition, the processed samples may be preserved as organic extracts.1 It was reported that analytes generally remained stable in stored sample extracts, although no information was given about how long and for which analytes this kind of storage was effective.In this work, 27 pesticides, including nine organophosphorus, six organochlorine, a carbamate, a pyrethroid and another seven pesticides of miscellaneous groups, were studied at concentrations between 20 and 100 ng l21.Two ways of storing the different pesticides present in water samples were considered: on C18 SPE cartridges and as organic extracts concentrated to dryness. Experimental Reagents All the pesticides were of > 98.5% purity and all the solvents were of pesticide residue analysis grade. C18 extraction cartridges (J. T. Baker, Phillipsburg, NJ, USA) with 500 mg of packing material were used.Cartridge Storage Water samples of 1 l, obtained from a Milli-Q purification system (Millipore, Bedford, MA, USA), were fortified with 20–100 ng of the chemicals (see Table 1). The cartridge was conditioned by rinsing it with the eluents in reverse order of elution, then with methanol and water.20 After the passage of the 1 l fortified water sample, the cartridge was air dried for 30 min. Solid-phase cartridges on which pesticides had been retained were wrapped in Parafilm and stored, at 4 or 218 °C, for 3, 7 and 30 d.At the end of the storage period they were eluted with ethyl acetate and isooctane and evaporated to dryness under a gentle stream of nitrogen. The final residue was taken up with 1 ml of hexane and the internal standard, bromophos, was added before injection into the gas chromatographs.21 Concentrated Extract Storage A mixture of the pesticides in the amounts shown in Table 1 was added to 4 ml of a mixture of the eluents [ethyl acetate– isooctane (1 : 1)] and then concentrated to dryness under a gentle stream of nitrogen. The glass tubes containing the dried eluates were fitted with a glass stopper, covered with Parafilm and stored at the same temperatures and for the same periods of time as used for the cartridges.The dried residue was treated as for the cartridge. For both storage treatments, cartridges and concentrated extracts, the remainder of the analytical process was finished rapidly with injection into the chromatograph on the same day.Statistical Analysis A completely randomized design with three replications of the whole analytical procedure, with three injections per replicate, of two storage temperatures and three storage times was Analyst, January 1997, Vol. 122 (7–11) 7employed for each storage treatment. Mean percentage recoveries were calculated and separated by Fisher’s least significant difference (LSD) at the 0.05 level of significance. Analytical Methodology All the samples were quantified by GC.20,21 Organophosphorus pesticides were determined using a Hewlett-Packard (Avondale, PA, USA) gas chromatograph with a flame photometric detector (FPD), provided with an HP-1 capillary column (12 m 3 0.2 mm id, 0.33 mm film thickness) with the following oven temperature programme: 45 °C (1 min), increased at 30 °C min21 to 170 °C (2 min), at 4 °C min21 to 200 °C (2 min) and at 20 °C min21 to ctor and detector temperatures were 250 and 275 °C, respectively. The other pesticides were determined using a Hewlett- Packard gas chromatograph with an electron capture detector (ECD), in which an Ultra-2 capillary column (25 m 3 0.32 mm id, 0.17 mm film thickness) was installed, with the following oven temperature programme: 160 °C (1 min), increased at 4 °C min21 to 230 °C (2 min) and at 20 °C min21 to 280 °C (6 min).The injector temperature was 250 °C and the detector temperature was 300 °C.Results and Discussion The pesticides included in this study, were tested at concentrations lower than or equal to the maxima allowed by EU legislation. Such concentrations are at least 102–103 times more dilute than the values previously reported in some storage studies using extraction discs15,16 or cartridges.17 Therefore, the amounts used in this study approach legal conditions and avoid higher concentrations which can influence degradation, as has already been reported for some organophosphorus and carbamate pesticides in soil columns.22 Fig. 1 shows the separation of a pesticide standard solution, with the different detection methods, at the concentrations indicated in Table 1. Cartridge Storage The recoveries obtained for the pesticides retained on the C18 cartridges are given in Table 2 for the pesticides in which no interaction between storage time and temperature was encountered and in Table 3 for the chemicals with a significant interaction between the storage factors studied.As can be seen for the effect of temperature (Table 2), only alachlor, captan, o,pA-DDE and chlorpyrifos-ethyl show significant differences for the two temperatures studied. Nevertheless, for most of these chemicals only small differences in the recoveries for the two storage temperatures are observed (@10%). Captan, and the chemically related folpet, exhibited an important reduction in recovery, higher for the freezer (218 °C) than for the refrigerator (4 °C), for which no explanation has Table 1 Pesticide amounts used for both storage treatments and their physico-chemical properties. Data from ref. 18 Amount Vapour Water added/ pressure/ solubility/ Pesticide ng* mPa Log Kow mg l21 Trifluralin 20 9.5 4.0† 0.221 Lindane 20 5.6 3.7† 7.3 Triallate 40 16 na‡ 4 Alachlor 80 2.9 na 242 Captan 40 1.3 2.5† 3.3 Folpet 40 1.3 3.1 1 o,pA-DDE 20 na na na p,pA-DDE 20 0.025 5.7–7.0† na Oxyfluorfen 20 0.0267 4.5 0.116 o,pA-DDT 20 na 5.8† na p,pA-DDT 20 na 6.2–6.9† na Bromopropylate 20 0.011 5.4 < 0.5 Dicofol 40 0.053 4.3 0.8 Tetradifon 20 3.231025 4.6 0.08 Deltamethrin 40 0.002 4.6 < 0.0002 Dimethoate 100 1.1 0.7 23800 Fonofos 100 28 3.9 13 Diazinon 100 12 3.3 60 Formothion 100 0.113 na 2600 Fenitrothion 100 18 3.4 21 Malathion 100 5.3 2.9† 145 Fenthion 100 0.74 4.8 4.2 Chlorpyrifos 100 2.7 5.3† 1.4 Methidathion 100 0.25 2.2 200 Phosmet 100 0.065 3.0 25 Azinphos-methyl 100 0.18 3.0 28 Phosalone 100 < 0.067 4.3† 1.7 * The same amount was added to 1 l of water or to 4 ml of the desorption solution.† Data from ref. 19. ‡ Not available. Table 2 Effect of storage temperature and storage time on recoveries of pesticides retained on C18 cartridges Storage Storage temperature/°C* time/d† Pesticide 218 4 LSD‡ 3 7 30 LSD‡ Trifluralin 90 92 86 89 101 5.1 Lindane 96 97 95 97 98 Triallate 89 86 96 86 82 6.6 Atachlor 94 98 2.7 96 95 97 Captan 58 73 9.1 74 66 58 11.2 Folpet 38 50 50 41 47 o,pA-DDE 69 64 4.0 63 72 65 4.9 p,pA-DDE 67 66 64 67 68 Oxyfluorfen 85 84 81 84 88 3.5 o,pA-DDT 67 69 63 71 70 4.9 p,pA-DDT 70 69 68 71 70 Bromopropylate 84 83 84 84 84 Dicofol 87 86 85 86 87 Tetradifon 95 98 100 96 94 4.0 Diazinon 87 87 89 89 84 3.8 Fenitrothion 94 95 99 97 88 4.0 Malathion 94 94 98 98 87 3.6 Chlorpyriphos-ethyl 82 79 2.4 85 83 72 2.9 Phosmet 95 96 103 87 97 9.7 * Average recovery for nine observations (three injections per observation).† Average recovery for six observations (three injections per observation). ‡ LSD = least significant difference (P < 0.05). The LSD values are provided for those compounds with a significant difference between the means. Table 3 Effect of temperature and time of storage on recovery of pesticides retained on C18 cartridges Recovery (%)* Refrigerator Freezer (4 °C) (218 °C) Pesticide 3 d 7 d 30 d 3 d 7 d 30 d LSD† Fonofos 75 68 55 79 64 65 7.9 Methidathion 105 98 89 101 103 99 6.9 Azinphos-methyl 121 102 129 119 114 167 17.8 Phosalone 102 96 87 110 93 98 6.9 * Average recovery for three observations (three injections per observation). † See Table 2. 8 Analyst, January 1997, Vol. 122been found. In a previous study, stability problems with captan were reported (54% recovery on a C18 disc at 4 °C and 32% at 218 °C after 30 d of storage) and ascribed to hydrolysis under the C18 packing of the SPE disc15 and, later, to volatilization,17 although the vapour pressure of this fungicide ( < 1.3 mPa, Table 1) is lower than that of some other pesticides included in this study, which do not show a loss.In addition, the captan recovery is also affected by the storage time (Table 2), so this method of preservation is not appropriate for this pesticide. When captan was stored on extraction discs,15 removal of residual water was recommended. The length of the storage affects several compounds, apart from captan, as can be seen in Table 2.In general, for the pesticides affected (trifluralin, triallate, o,pA-DDE, oxyfluorfen, o,pA-DDT, tetradifon, diazinon, fenitrothion, malathion, chlorpyrifos- ethyl and phosmet), the recovery decreases with increasing storage time, but remains within a ±10–15% variation for the whole length of the storage period. This variation range is within that commonly found in other storage studies.15,17 In addition, the final recovery after 30 d remains over 70% (the minimum recovery required by the EPA regulations23), except for DDEs, with recoveries of !65%, and captan and folpet.Low recoveries have been also reported for DDE and DDT retained on Empore filters for up to 4 weeks, from waters spiked at 50 ng l21.14 Fonofos, methidathion, azinphos-methyl and phosalone showed a significant interaction between temperature and length of storage (Table 3) and therefore both factors cannot be considered individually.For fonofos, methidathion and phosalone, a trend of loss on storage in the refrigerator after 30 d was observed. For fonofos, its storage on a cartridge is not recommended, because its recovery is already below 70% after 7 d at the two temperatures investigated. A complete loss of fonofos, which is the most volatile pesticide included in the present study (Table 1), has been reported for storage in similar cartridges at 4 °C after 1.5 months.17 In the contrast, the same group17 reported complete recovery of fonofos when stored at 220 °C for 8 months.Our results show a 14% loss after 1 month of storage in the freezer. For azinphos-methyl, which is not completely separated from phosalone in the chromatographic column, an artifact appears after 30 d of storage and therefore its quantification was inaccurate. Fig. 2(A) shows the separation of the pesticide mixture with electron capture detection (ECD) after storage in the cartridge for 30 d at 218 °C.The recoveries of the fungicides captan and folpet are clearly affected. The additional peaks correspond to the organophosphorus pesticides that can be detected by ECD. As indicated previously, several organophosphorus pesticides included in the present study (azinphos-methyl, fenitrothion, malathion, diazinon and phosmet) have been eliminated from the US EPA survey list owing to their instability;1 the first three are nevertheless included in the 76/464/EEC Council Directive List of Pesticides to be monitored in the aquatic environment.23 In addition, phosmet has been repeatedly reported to be unstable on storage in water.2,5 These pesticides could be stabilized by storing them on the cartridge packing at 4 or 218 °C for at least 30 d.Nevertheless, certain losses have been observed for some of the pesticides, especially those with low partition coefficients, which confirms the results of Lacorte et al.,17 who pointed to hydrolysis and microbial degradation as the main factors causing instability with this kind of storage.Concentrated Extract Storage The recoveries obtained for the pesticides stored in the dry extract are given in Table 4 for the pesticides in which no interaction between storage time and temperature was encountered and in Table 5 for the chemicals with a statistically significant interaction. For this storage treatment, three additional organophosphorus and one pyrethroid pesticide not considered in the cartridge storage were included: dimethoate, formothion, fenthion and deltamethrin.These compounds could only be recovered between 12 and 59% from water samples after the whole SPE Fig. 1 Gas chromatogram of a standard pesticide solution, at the concentrations shown in Table 1, with electron capture (ECD) and flame photometric detection (FPD). TR, trifluralin; L, lindane; TL, triallate; A, alachlor; IS, bromophos (internal standard); CP, captan; FP, folpet; OAE, o,pA-DDE; PAE, p,pADDE; OX, oxyfluorfen; OAT, o,pA-DDT; PAT, p,pA-DDT; BP, bromopropylate; DF, dicofol; TF, tetradifon; DT, deltamethrin; D, dimethoate; FO, fonofos; DZ, diazinon; FR, formothion; FN, fenitrothion; M, malathion; FT, fenthion; CL, chlorpyriphos; MT, methidathion; PM, phosmet; AZ, azinphos-methyl; and PS, phosalone.Analyst, January 1997, Vol. 122 9process,20 which did not allow an appropriate study of their stability when retained on the cartridge packing.Pesticides stored after evaporation to dryness in general give good recoveries. In this experiment, no real extraction in water was carried out and only the effect of storage on the simulated eluate was studied. Table 4 indicates that for this storage factor, higher recoveries were obtained on storage at 218 °C. Seven out of the 20 compounds listed are affected by storage temperature. Nevertheless, the difference in the recovery for the two temperatures is !10% only for captan, folpet, dimethoate, formothion and fenthion. In Fig. 2(B), decrease in recovery of the last three pesticides after storage for 30 d at 4 °C is shown. Captan and folpet, which already showed instability during cartridge storage, together with the last three organophosphorus pesticides, especially fenthion, which were not included previously, appear to be difficult to stabilize also in this dried form. For captan, the reason cannot be hydrolysis, because no water was present.Volatilization is also possible, but again captan, folpet and fenthion, with differences in recoveries > 25% for the two temperatures considered, have low vapour pressures (Table 1). Molecular structure may be responsible for the instability, as has already been reported for fenthion.5 The effect of storage time for the same chemicals is also shown in Table 4. Many pesticides show a statistically significant difference among the storage periods investigated but, for most of them, the recoveries fall within a 10% variation.Only for dimethoate, formothion and fenthion was the loss Table 4 Effect of storage temperature and storage time on recoveries of pesticides stored after their concentration to dryness Recovery (%) Storage Storage temperature/°C* time/d† Pesticide 218 4 LSD‡ 3 7 30 LSD‡ Alachlor 97 95 101 88 100 3.1 Captan 99 76 5.6 91 90 82 6.8 Folpet 101 74 7.3 93 87 83 o,pA-DDE 97 94 106 89 91 4.0 p,pA-DDE 98 96 1.8 100 89 102 2.2 Oxyfluorfen 96 94 96 96 93 2.7 o,pA-DDT 100 99 98 93 107 2.7 p,pA-DDT 101 98 99 92 108 3.9 Bromopropylate 104 101 101 97 111 6.8 Dicofol 105 98 5.3 97 99 108 6.5 Tetradifon 103 100 100 98 107 5.3 Deltamethrin 102 97 98 96 106 5.5 Dimethoate 93 80 9.9 86 98 75 12.1 Formothion 121 102 7.2 140 117 76 8.8 Malathion 105 105 103 100 111 3.9 Fenthion 74 33 8.0 64 53 43 9.9 Methidathion 101 99 106 99 96 5.0 Phosmet 107 108 109 104 111 Azinphos-methyl 102 103 104 103 101 Phosalone 106 107 105 101 114 5.4 * Average recovery for nine observations (three injections per observation).† Average recovery for six observations (three injections per observation). ‡ See Table 2. Fig. 2 Gas chromatograms of the pesticides stored A, in the cartridge for 30 d at 218 °C, with ECD, and B, in the concentrated extract at 4 °C for 30 d, with FPD. Abbreviations as in Fig. 1. Table 5 Effect of temperature and length of storage on recovery of pesticides stored after their concentration to dryness Recovery (%)* Refrigerator Freezer (4 °C) (218 °C) Pesticide 3 d 7 d 30 d 3 d 7 d 30 d LSD† Trifluralin 84 62 44 96 86 109 6.5 Lindane 78 58 47 93 84 97 5.9 Triallate 87 68 63 96 87 94 4.7 Fonofos 82 65 44 97 87 97 6.9 Diazinon 91 76 49 100 93 97 7.0 Fenitrothion 100 97 88 103 97 100 3.5 Chlorpyriphos-ethyl 101 95 86 98 94 107 6.3 * Average recovery for three observations (three injections per observation).† See Table 2. 10 Analyst, January 1997, Vol. 122> 10%. Nevertheless, except for fenthion, their recoveries were > 70%, which complies with EPA regulations.23 Table 5 shows the trends of a reduction in recovery with time and storage temperature which apply to all of the compounds. For trifluralin, lindane, triallate, fonofos and diazinon, the loss is already clear after 7 d of storage at 4 °C, and is more evident after 30 d with recoveries, in many cases, around 50%. If their vapour pressures are considered, a clear relationship between this property (Table 1) and pesticide loss can be observed.A high degree of correlation was found between vapour pressure and pesticide loss after 30 d at 4 °C for fonofos, triallate and chlorpyrifos. The losses of diazinon, trifluralin and lindane are higher than expected and those of fenitrothion are lower, if the only reason considered to be involved in the process was volatility. For diazinon, an imidoyl phosphate, it has been reported3 that the oxygen–aromatic moiety linkage may be activated and cleaved under both acidic and alkaline conditions to form the corresponding carbonyl compounds with a doublebond shift.This functional group is also present in the structure of chlorpyrifos, so the losses must be due to a mixture of both effects. In contrast, all the pesticides present in Table 5 show recoveries between 94 and 117% at 218 °C after 30 d of storage. It can be concluded from these results that the storage of dried eluates in the refrigerator may be problematic for volatile compounds, as already reported for chlormephos and dichlorvos (7600 and 1600 mPa, respectively) when stored in water samples5 or in disposable SPE cartridges.17 Nevertheless, each case should be studied carefully to verify the observed tendencies.The comparison with physico-chemical data provided in Table 1 indicates that, in general, the most volatile compounds are affected by storage in dried form in the refrigerator, although at 218 °C the decrease in recovery is almost negligible for all of them except for fenthion, which showed great instability, independently of the storage conditions.The storage of the dried eluates in the freezer at 218 °C seems to be a good alternative for these compounds, except for fenthion. Stability problems with triallate, tetradifon, diazinon, fenitrothion, malathion, chlorpyrifos-ethyl and fonofos, which could not be properly controlled when retained on the C18 cartridges, could be solved by storing them as dried eluates for up to 30 d at 218 °C.Conclusions Although immediate extraction and analysis are the best way of obtaining the most accurate residue data, this is not always possible. Therefore, two different storage conditions, corresponding to different stages in the over-all SPE process, were investigated. The results indicated that storage of different pesticide classes on the C18: silica gel surface of solid-phase cartridges for 30 d at 4 or 218 °C is effective for 17 out of the 23 pesticides studied.Triallate, captan, fenitrothion, malathion, chlorpyriphos and fonofos are affected. These pesticides, except for captan and fonofos, can be safely stored in the cartridges for up to 3 or 7 d, and their recoveries after 30 d are still > 70%. All of them, including captan and fonofos, could alternatively be stored for up to 30 d at 218 °C in the dried extract. For this kind of storage, only formothion and fenthion showed considerable losses along the storage period studied.Formerly, only a limited number of 1 or 2 l bottles could be stored in the laboratory. For both treatments considered here, owing to the small size of the cartridges and glass tubes, the space required for their storage has been reduced and a large number of samples can be preserved in a conventional freezer. The authors thank the Comisi�on Interministerial de Ciencia y Tecnolog�ýa (CICYT) for financial support (Project No. NAT91- 0407). Ma. D. Maroto is acknowledged for technical assistance and Ma. D. Mingorance is thanked for her assistance with the statistical analysis. References 1 Munch, D. J., and Frebis, C. P., Environ. Sci. Technol., 1992, 26, 921. 2 Ripley, B. D., Wilkinson, R. J., and Chau, A. S. Y., J. Assoc. Off. Anal. Chem., 1974, 57, 1033. 3 Chau, A. S. Y., Ripley, B. D., and Kawahara, F., in Analysis of Pesticides in Water, ed. Chau, A. S. Y., and Afghan, B. K., CRC Press, Boca Raton, FL, 1982, vol. II, pp. 61–154. 4 Barcel�o, D., Chiron, S., Lacorte, S., Mart�ýnez, E., Salau, J. S., and Hennion, M. C., TrAC, Trends Anal. Chem., (Pers. Ed.), 1994, 13, 352. 5 Lartiges, S. B., and Garrigues, P. P., Environ. Sci. Technol., 1995, 29, 1246. 6 Liska, I., Brower, E. R., Ostheimer, A. G. L., Lingeman, H., and Brinkman, U. A. Th., Int. J. Environ. Anal. Chem., 1992, 47, 267. 7 Bourne, S., J. Environ. Sci. Health, 1978, B13, 75. 8 Chau, A. S. Y., and Thomson, K., J. Assoc. Off. Anal. Chem., 1978, 61, 1481. 9 Lopez-Avila, V., Wesselman, R., and Edgell, K., J. Assoc. Off. Anal. Chem., 1990, 73, 276. 10 Barcel�o, D., House, W. A., Maier, E. A., and Griepink, B., Int. J. Environ. Anal. Chem., 1994, 57, 237. 11 Font, G., Ma�nes, J., Molt�o, J. C., and Pic�o, Y., J. Chromatogr., 1993, 642, 135. 12 Green, D. R., and Le Pape, D., Anal. Chem., 1987, 59, 699. 13 Berkane, K., Caissie, G. E., and Mallet, V. N., J. Chromatogr., 1977, 139, 386. 14 Tomkins, B. A., Merriweather, R., Jenkins, R. A., and Bayne, C. K., J. Assoc. Off. Anal. Chem. Int., 1992, 75, 1091. 15 Senseman, S. A., Lavy, T. L., Mattice, J. D., Myers, B. M., and Skulman, B. W., Environ. Sci. Technol., 1993, 27, 516. 16 Johnson, W. J., Lavy, T. L., and Senseman, S. A., J. Environ. Qual., 1994, 23, 1027. 17 Lacorte, S., Ehresmann, N., and Barcel�o, D., Environ. Sci. Technol., 1995, 29, 2834. 18 The Pesticide Manual. A World Compendium, British Crop Protection Council, Thornton Heath, UK, 8th edn., 1987. 19 Noble, A., J. Chromatogr., 1993, 642, 3. 20 de la Colina, C., S�anchez-Rasero, F., Dios, G., Romero, E., and Pe�na, A., Analyst, 1995, 120, 1723. 21 de la Colina, C., Pe�na, A., Mingorance, M. D., and S�anchez- Rasero, F., J. Chromatogr. A, 1996, 733, 275. 22 Belisle, A. A., and Swineford, D. M., Environ. Toxicol. Chem., 1988, 7, 749. 23 Barcel�o, D., J. Chromatogr., 1993, 643, 117. Paper 6/05275D Received July 29, 1996 Accepted September 30, 1996 Analys

ISSN:0003-2654    

DOI:10.1039/a605275d

出版商:RSC    

年代:1997

数据来源: RSC

摘要:

Enzymic Digestion–High-pressure Homogenization Prior to Slurry Introduction Electrothermal Atomic Absorption Spectrometry for the Determination of Selenium in Plant and Animal Tissues Yanxi Tan and William D. Marshall* Department of Food Science and Agricultural Chemistry, Macdonald Campus of McGill, 21 111 Lakeshore Road, Ste.-Anne-de-Bellevue, Qu�ebec, Canada H9X 3V9. E-mail: marshall@agradm.lan.mcgill.ca Homogenization, in combination with partial enzymic digestion with a crude protease alone or admixed with lipase or cellulase, was investigated as a means of releasing Se residues from zoological and botanical matrices prior to slurry introduction ETAAS.Preliminary timed trials with two zoological certified reference materials (CRMs), one botanical CRM and one animal feed indicated that Se release became quantitative with 4–8 h of digestion, that homogenization prior to digestion increased the initial rate of analyte release, but that homogenization post-digestion and immediately prior to ETAAS did not significantly improve the precision of replicate digestions.Storage of the crude botanical digests at 4 °C for 5 d resulted in quantitative recoveries of Se from each of the digests. Storage at 4 °C for 10 d of 4 and 8 h lipase/protease digests of six other CRMs resulted in quantitative recoveries of Se unless their certified concentrations were appreciably less than the levels determined in control digests containing the enzyme(s) alone.Apparently, Se residues were transferred virtually quantitatively to the liquid phase of the digested suspension and showed no tendency to segregate during the 10 d of storage. Eight other mixtures of ground plant matter (0.13 @ [Se] @ 1.31 mg g21), formulated as animal feed supplements, behaved identically when stored post-digestion. The technique was also applied successfully to freeze-dried fresh and boiled fish tissues The principal advantages of the enzymic digestion procedure are its simplicity and lack of operator intervention.Keywords: Selenium determination; enzymic digestion; slurry introduction electrothermal atomic absorption spectrometry; botanical and zoological certified reference materials Conventional sample preparation of biological materials prior to atomic spectrometry involves complete solubilization of the analyte and matrix, which is achieved typically by oxidative mineralization of the organic matter and solubilization of the residue in a suitable solvent.1–4 Even for microwave-assisted digestions, whereas complete dissolution can usually be achieved by a suitable choice of digestion conditions, complete decomposition of the organic matrix in biological/botanical samples is appreciably more difficult.Often complete mineralization is achieved only with supplemental treatment of the digested matrix with H2O2 or even HF.5 These digestion procedures can be labour intensive, time consuming and prone to contamination errors.In consequence, there is a continuing interest in the development of simplified sample preparation techniques. The preparation/introduction of slurried samples continues to attract considerable attention because of the ease with which quasi-stable preparations can be generated and their compatibility with conventional liquid handling techniques. Within the general field of solid sampling analysis,6–11 it is the use of slurried samples which has become the most popular approach to trace element determination.Direct atomization from the solid state can provide excellent sensitivity, but the interpretation of results can be complicated by molecular absorptions and/or scattering from the matrix, which can produce sufficiently large background signals to overwhelm the compensation capabilities of common deuterium background correction systems. Additional difficulties can include sample inhomogeneities, the requirement for repeated microweighings and the lack of suitable calibration standards and techniques.A variety of sample pre-treatment procedures and additives12 –18 have been described and evaluated for the production of quasi-stable suspensions of samples prior to analysis by atomic spectrometry. Alternatively, suspensions with a tendency to segment rapidly have been reproducibly sampled by using ultrasonic agitation,19 air or argon20 bubbling, vortex mixing21 or magnetic stirring.22 Partial digestion procedures to produce carbonaceous slurries have also been successfully applied to the analysis, by ICP-AES, of a series of standard reference materials of biological origin.23 Various alkylammonium hydroxide formulations have been used extensively to solubilize tissues,24–27 particularly those of zoological origin. Recently, high-pressure homogenization has been evaluated for the preparation of quasi-stable dispersions suitable for FTIR28 or ETAAS.29,30 The advantages of this approach to sample preparation were the ease and speed of the slurry preparation, which required less than 1 min, and the fact that analyte metals were quantitatively extracted into the liquid phase during the preparation so that no analyte segmentation was detected within the slurry even after standing for several days.Certified reference materials (CRMs), frozen liver and kidney and dried animal feeds of botanical origin were analysed successfully for Cd, Cr, Cu, Mn, Ni and Pb but not for Se.The principal limitation of the high-pressure homogenization technique was the amounts of contaminating analyte metals introduced into the sample by the homogenization operation. Contamination was reduced appreciably, but not eliminated entirely, by capping the flat face of the stainless-steel homogenizing valve with a ruby disc.30 The objectives of this work were (i) to evaluate the efficiencies of other materials as caps to reduce further the levels of contamination introduced by the high-pressure processing and (ii) to develop efficient alternative slurry preparation techniques for the determination of Se in biological materials.Although there have been few reports of the determination of Se in slurried samples,31–34 recent reports35,36 indicate that the approach is promising for this analyte. Prolonged enzymic digestion with a crude protease fraction has been used37 to liberate component selenoamino acids from proteins.This approach seemed promising as a pre-homogenization sample preparation. Analyst, January 1997, Vol. 122 (13–18) 13Experimental Reagents TRIS was purchased from Aldrich (Milwaukee, WI, USA) and aqueous Se solution (1000 mg ml21, traceable to NIST primary standard) was purchased from SCP Chemical (St.-Laurent, Qu�ebec, Canada). Samples CRMs were purchased from the National Research Council of Canada (NRCC) or the US National Institute of Science and Technology (NIST).Samples of animal diet mixtures destined for a zoo were chosen to contain a variety of plant and animal materials, including timothy grass, bamboo leaves, whole smelts, cricket chow and a mixture (contents unspecified) formulated for panda bears. Animal feed supplements were composed of mixed forage crops. Sample Preparation An accurately weighed sample (approximately 0.2 g) of CRM, dried feed or supplements (ground, to pass a 0.5 mm screen, in a Tecator Cyclotec sample mill; Tecator, H�ogan�as, Sweden) was added directly to 10 ml of ethanol–0.03 m TRIS (1 + 19 v/v, pH 7.5) containing either 20 mg of crude protease (Type XIV; Sigma St.Louis, MO, USA), 20 mg of protease plus 20 mg of lipase (Type VII; Sigma) or 20 mg protease plus 20 mg of cellulase (Cellulysin; Calbiochem–Novabiochem, La Jolla, CA, USA). The resulting suspension was then processed through the 20 ml capacity flat valve homogenizer (EmulsiFlex Model EFB3; Avestin, Ottawa, ON, Canada), capable of developing 138 MPa when provided with compressed air (689.4 kPa).Each slurried sample was re-processed through the homogenizer three more times. The homogenates, in 50 ml Erlenmeyer flasks, were then digested at 37 °C with gentle agitation for 4–8 h. Homogenizer The valve stem of the screw-cap assembly of the homogenizer was modified by gluing a polished 4 mm diameter 32 mm thick disctured from a 6–12 mm diameter sphere of tungsten carbide (Spex, Metuchen, NJ, USA), zirconia (Optimize Technologies, Portland, OR, USA), sapphire (from an HPLC piston) or polymethacrylate (Spex).Sample + enzyme suspension was transferred to the sample chamber via the inlet port, which was then sealed with a fine-threaded screw-cap. The stainless-steel piston (connected to a pneumatic multiplier) then forced the fluid through an aperture and the homogenate was collected from the sample outlet. Each sample was re-processed three more times with the valve stem retracted slightly to provide a slightly larger gap setting.Selenium Determinations (Hydride Generation or Fluorescence) Feed samples were dried to constant mass and ground to pass a 1 mm screen. Accurately weighed aliquots of ground feed or freeze-dried fish tissue (approximately 2 g) were digested at room temperature in a perchlorate fume-hood with 25 ml of 70% HNO3–HClO4 (4 + 1 v/v) until gas evolution had ceased, then heated at 80 °C until a clear yellow solution was obtained.The resulting strong acid digests were analysed by HGAAS38 or fluorescence of the piazselenol derivative39 after conversion of the analyte residues in to SeIV. ETAAS Selenium determinations were performed using a hot injection technique on a Varian (Palo Alto, CA, USA) Model 300 ETAAS system equipped with an autosampler, pyrolytic graphite-coated platform graphite tubes, a conventional Se hollow-cathode lamp and Zeeman-effect background correction.Ashing–atomization curves were generated for Se standard in the presence and absence of co-injected biological sample. At temperatures < 2300 °C, the Se atomization signal was broadened by the presence of biological materials but was sharpened (and did not tail) for atomizations at 2400 °C. In the presence of the palladium–citric acid modifier, no loss in the Se signal was observed at an ashing temperature @1400 °C. Analytical operating parameters are presented in Table 1.Calibration ETAAS quantification was performed by both the method of external standards (ES) and by standard additions (SA). ES consisting of appropriately diluted processed reagent blank and up to four levels of standard were prepared automatically by the sample introduction device. The background-corrected peakarea response, resulting from three replicate injections of each diluted standard, was used to define the best-fit regression equation. For SA calibrations, 10 ml aliquots of processed fluid were amended with 2, 5 or 10 ml of aqueous standard chosen to result in a range of peak areas including signals which were half and at least twice the signal for the unamended sample.The data were modelled by least-squares linear regression. Quantification was performed by dividing the intercept on the ordinate of the regression equation by the slope of the equation and the overall standard error of the estimate (SEest) was calculated from SEest = (SE2 y-int + SE2 slope)1/2 Results and Discussion Preliminary experiments were directed to evaluating the influence of different capping materials on the levels of contaminating metals introduced into the homogenates during processing. It was postulated that exposed stainless-steel surfaces within the valve homogenizer, particularly the flat face of the demountable valve head, were the principle sites responsible for the contamination.Further, capping the valve head with an inert surface capable of withstanding the impact of the jet of fluid exiting the homogenizing orifice might reduce the levels of contamination appreciably.It has been reported40 previously that zirconium oxide beads used to reduce the particle size and to mix particulate solids introduced appreciable levels of Fe, Cr and Al but that silicon nitride or boron carbide provides good abrasion resistance and offers little likelihood of Table 1 Furnace operating parameters for determinations of selenium Parameter* Value Wavelength/nm 196.0 Lamp current/A 10 Slit width/nm 1.0 Injection temperature/°C 60 Pre-injection Yes Temperature of last dry step (10 s)/°C 250 Charring sequence 10 s ramp to 1400 °C, 40 s hold Cooling None Atomization 0.6 s ramp to 2400 °C, 5.0 s hold Measurement time/s 5.6 Chemical modifier 5 ml (0.5% m/m Pd + 2.5% m/m critic acid) for 10 ml sample * Each step of the furnace programme (with the exception of the read step) was performed in the presence of argon purge gas (3 l min21). 14 Analyst, January 1997, Vol. 122contamination. However, even for the relatively lower pressure requirements of pistons and check valves for HPLC, sapphire, ruby and zirconium oxide are preferred over other ceramic materials for their superior resistance. Separate discs composed of zirconia, tungsten carbide or polymethacrylate, which had been manufactured by grinding and polishing a 6–12 mm diameter grinding ball, were glued temporarily to the flat face of the demountable valve head.Similarly, the sapphire disc was generated from a used HPLC piston. Solvent mixture (20 ml) was homogenized four successive times (in the presence or absence of the test disc) prior to ETAAS analysis for Al, As, Cd, Cr, Cu, Fe, Mn, Ni, Pb and Se. Analyte concentrations (Table 2) were expressed as if the solvent had contained 0.100 g of sample. The heavy metal content of the homogenized fluid was lowered appreciably in all cases.Nonetheless, contamination remained appreciable for several elements, even in the presence of the polymethacrylate or the sapphire cap. Previous attempts to determine Se in biological materials by slurry introduction ETAAS of high-pressure homogenates were not successful in our hands using a variety of furnace programmes, yet there was no evidence of any analyte loss prior to the atomization stage as judged by the signal graphics software, which provided a continuous display of the Se signal over the course of the furnace programme. Since a high proportion of the analyte element in biological materials is considered to be protein bound, it was decided to evaluate partial enzymic hydrolysis as a means of liberating bound analyte residues.Arbitrarily, it had been decided initially to attempt to develop a single combination of mixed enzymes which it was hoped would be applicable to all sample matrices. Previous studies41 had indicated that a combination of crude proteases and lipases efficiently hydrolysed avian egg yolk.Initial studies were limited to this combination of enzymes. The mixture of crude enzymes was suspended in 10 ml of TRIS buffer (pH 7.5), then passed sequentially four times through the polymethacrylate-capped homogenizer to furnish a digestion control homogenate. Relative to a distilled, de-ionized water blank, this ‘zoological’ control sample contained 0.11 mg g21 ± 12.6% and 0.10 mg g21 ± 11.3% after 4 and 8 h of digestion respectively (Table 3), when it was assumed that the digests had contained 0.200 g of sample.Similarly, a control homogenate composed of protease alone contained 0.044 mg g21 ± 12.6% after 4 h. A crude cellulase was substituted for the lipase in the enzyme mixture and the digestions were performed in analogous fashion to furnish alternative enzymic digestion control samples. The ‘botanical’ control samples contained 0.048 mg g21 ± 14.2% and 0.051 mg g21 ± 12.0% after 4 and 8 h of digestion respectively (Table 3), and the solvent blank + protease alone contained 0.044 mg g21 ± 12.6%, again assuming that the digests had contained 0.200 g of sample.Thus, virtually all of the Se in control digests originated with the lipase and/or the protease. Based on a 3 RSD criterion, the corresponding method limit of detection (LOD) for digestions with mixed protease lipase, with protease cellulase and with protease alone were 0.020, 0.010 and 0.010 mg g21, respectively.In preliminary experiments, three biological CRMs and one animal feed, suspended in 10 ml TRIS buffer, were digested with a combination of protease and lipase for up to 16 h at 37 °C. Table 2 Apparent analyte concentrations (mg g21 sample) in 20 ml of solvent mixture following various mixing treatments. Concentrations in the homogenized fluid are expressed as if the solvent mixture had contained 0.100 g of sample Treatment Al As Cd Cr Cu Fe Pb Mn Ni Se Unhomogenized solvent blank 0.32 n.d.* n.d.n.d. n.d. 0.40 n.d. n.d. n.d. n.d. Four successive homogenizations with: Stainless-steel head 42.12 0.02 4.53 15.0 56.94 1.38 2.31 3.57 n.d. Polymethacrylate cap 7.92 n.d. 0.02 1.80 n.d. 5.20 n.d. 0.10 n.d. n.d. Ruby cap 21.52 n.d. 0.03 4.02 0.70 13.99 0.28 0.39 0.11 n.d. Sapphire cap 3.64 n.d. n.d. 3.65 0.80 4.90 n.d. 0.10 n.d. n.d. Tungsten carbide cap 15.32 n.d. 0.04 4.40 1.40 38.6 2.00 0.40 0.10 n.d. Zirconia cap 0.42 n.d. 0.02 4.00 1.20 15.7 n.d. 0.20 0.10 n.d. * n.d. = None detected above the mean background signal for ethanol–0.03 m TRIS (1 + 19 v/v, pH 7.5) solvent. Table 3 Selenium concentrations (mg g21) (±1 RSD based on three replicate samples) in certified reference materials determined immediately after 4 or 8 h of enzymic digestion or following digestion plus 10 d of storage 4 h digestion + 8 h digestion + Certified Matrix 4 h digestion 8 h digestion 10 d storage 10 d storage concentration Solvent blank + protease + lipase 0.11 ± 12.6% 0.10 ± 11.3% 0.12 ± 10.2% 0.10 ± 13.6% Solvent blank + protease 0.044 ± 12.6% Solvent blank + protease + cellulase 0.048 ± 14.2% 0.051 ± 12.0% 0.040 ± 23.2% 0.045 ± 12.2% Oyster Tissue* 2.18 ± 12.7% 2.27 ± 11.6% 2.16 ± 11.4% 2.15 ± 12.6% 2.21 ± 0.24 DORM-2* 1.38 ± 9.2% 1.35 ± 11.6% 1.34 ± 6.6% 1.33 ± 11.0% 1.40 ± 0.090 Bovine Muscle* 0.067 ± 22.6% 0.066 ± 20.5% 0.067 ± 20.2% 0.070 ± 21.4% 0.076 ± 0.010 Apple Leaves† 0.043 ± 12.7% 0.045‡ ± 19.9% 0.047 ± 12.7% 0.041 ± 12.7% 0.050 ± 0.009 Corn Bran† 0.034‡ ± 12.5% 0.036 ± 14.9% 0.033 ± 15.3% 0.036 ± 16.5% 0.045 ± 0.008 Corn Stalk† n.d.§ n.d.§ 0.025‡ ± 47.0% 0.011‡ ± 52.2% 0.016 ± 0.008 * Reported concentrations have been corrected for the [Se] in the protease + lipase control homogenate.† Reported concentrations have been corrected for the [Se] in the protease + cellulase control homogenate. ‡ No [Se] above the LOD (0.010 mg g21) was detected in one of the three aliquots. § No [Se] above the LOD (0.010 mg g21) was detected in any of the three aliquots.Analyst, January 1997, Vol. 122 15Each suspension was homogenized immediately prior or directly after digestion, then analysed by ETAAS. The results are presented in Figs. 1 and 2. The TORT-1 results and the animal feed results (triangular symbols in Fig. 1 and Fig. 2, respectively) have been displaced by 0.4 h for clarity of presentation. For all four substrates, homogenization prior to digestion (closed symbols) generally resulted in higher recoveries relative to homogenization post-digestion (open symbols), although the differences were only rarely statistically significant.Moreover, the differences tended to decrease with longer digestion times. Presumably, homogenization initially exposed more of the protein component to the enzyme. On the other hand, homogenization post-digestion and immediately prior to ETAAS did not significantly improve the precision of the determination (as judged by the RSD associated with three replicate measurements performed on each of three digests). In general, there was a gradual but small improvement in precision with increased length of digestion (more evident with the plant samples in Fig. 2). For both the DORM-1 and the TORT-1 marine tissue samples in Fig. 1, 4 h of digestion at 37 °C were sufficient to liberate the Se quantitatively, whereas recoveries from the plant samples were quantitative only after 8 h of digestion.After sampling for ETAAS, the plant digests were stored at 4 °C for 5 d and then re-analysed. No effort was made to resuspend solid materials; instead, a portion of each supernatant fraction was transferred directly to the sampling cup. The recovery of Se from each of the supernatant fractions was quantitative (Fig. 3), indicating that the crude protease was active at the storage temperature and that there was no apparent segmentation of the Se residues between the liquid and solid phases of the crude digest.Surprisingly, the short-term repeatability of the procedure was not improved by the storage, as evidenced by the RSD associated with replicates. Repeatability continued to be improved for longer digestions at 37 °C. Three replicate aliquots of each of six other certified reference materials were homogenized and then digested for 4 or 8 h prior to ETAAS. The results, corrected for the Se concentration in the appropriate zoological or botanical control digest, are presented in Table 3.Whereas digestion of the three marine CRM homogenates provided estimates which were not significantly different from the certified Se concentrations (despite the higher Se concentration in the zoological control) the lower concentrations in the botanical CRMs resulted in estimates which, occasionally, were not different from the control concentrations. In the latter cases, the certified Se concentrations were less than the Se concentrations in control digests.Likewise, three aliquots (approximately 0.2 g) of each of eight dried, ground feed supplements consisting of mixed forage crops were suspended in 10 ml of TRIS buffer, homogenized and then digested with the protease–cellulase combination for either 4 or 8 h prior to Se determination by ETAAS. Again, there was good agreement between the results (Table 4) for slurry introduction ETAAS following 4 or 8 h of enzymic digestion and a single fluorescence determination Fig. 1 Variation in percentage recovery of Se (±1 RSD based on three replicate samples) from DORM-1 or TORT-1 certified reference materials versus hours of enzymic digestion with protease plus lipase prior to (open symbols) or after (closed symbols) high-pressure homogenization. For clarity of presentation, the TORT-1 results have been displaced by 0.4 h. Fig. 2 Variation in percentage recovery of Se (±1 RSD based on three replicate samples) from Durham wheat flour CRM or a ground animal feed sample versus hours of enzymic digestion with protease plus cellulase prior to (open symbols) or after (closed symbols) high-pressure homogenization. For clarity of presentation, the animal feed results have been displaced by 0.4 h.Fig. 3 Influence of storage subsequent to slurry preparation on percentage recovery of Se (±1 RSD based on three replicate samples) in flour CRM or a ground animal feed sample.For clarity of presentation, the animal feed results have been displaced by 0.4 h. 16 Analyst, January 1997, Vol. 122following strong acid digestion and piazselenol formation. Storage of the digests for 10 d at 4 °C did not change the measured concentrations of analyte (Table 4). In only one feed supplement (in which the Se concentration was appreciably less than that in the botanical control digest) were the results of the two methods discordant. No matrix effects for Se determinations in any of the samples were detected.The slopes of the best-fit regression lines for standard additions to homogenized protease–lipase or protease– cellulase enzyme suspensions in TRIS buffer, in the presence or absence of DORM-1, TORT-1, wheat flour, corn bran or apple leaf CRM, or to five of the feed supplements varied by less than 11% (RSD) provided that calibrations and determinations were performed on the same day. This observation suggested that a single calibration by standard addition(s) to the enzyme mixtures would suffice for the determination of Se in any of the samples.A single calibration curve generated by adding aqueous Se standard to the botanical control homogenate was then used to determine the Se content of freeze-dried freshwater fish fillets which had been frozen fresh or boiled to simulate cooking following common native practice. Aliquots of the freeze-dried materials were digested enzymically for 4 or 8 h and then analysed by ETAAS or digested with strong acids and then analysed by HGAAS (Table 5).Boiling the fillet prior to freezedrying did not inhibit the enzymic release of Se residues from the matrix but apparently lowered the Se concentration in the cooked product. There were no significant differences between the results after 4 and 8 h of digestion or between ETAAS and HGAAS results. However, the precision associated with replicate enzymic digestion–ETAAS Se determinations (mean RSD nearly 15 ± 2%) was appreciably worse than the precision associated with hydride generation determinations (mean 7 ± 4%) but typical of the replicate determinations of other experiments (mean RSD for the 36 determinations in Table 3 14.4 ± 0.3% and for 29 of the determinations in Table 4 12.6 ± 0.5%).Thus, the short-term repeatability (i) was not adversely affected by the use of the single calibration curve but (ii) can be expected to be degraded relative to other conventional procedures for Se determination.The principal advantages of the enzymic digestion procedure are the simplicity and speed relative to conventional unassisted acid digestions and that they can be performed unattended. The conditions of digestion do not appear to be critical and there was no tendency for the liberated Se residues to segregate within the resulting suspensions. Feed samples and feed supplements and determinations of their Se content by fluorescence of their piazselenol derivatives were generously supplied by E.R. Chavez, McGill University. Samples of fish fillets and determinations of their Se content by HGAAS were kindly donated by H. M. Chan, McGill University. Financial support in the form of an operating grant from the Natural Science and Engineering Research Council of Canada (NSERC) is gratefully acknowledged. References 1 Novozamski, I., van der Lee, H. J., and Houba, V. J. G., Microchim. Acta, 1995, 119, 183. 2 Sansoni, B., and Panday, V. K., in Analytical Techniques for Heavy Metals in Biological Fluids, ed. Fachetti, S., Elsevier, Amsterdam, 1983, p. 91. 3 Mincwewski, J., Chwastowska, J., and Dybczynski, R., Separation and Preconcentration Methods in Inorganic Trace Analysis, Ellis Horwood, Chichester, 1982. 4 Matusiewicz, J., and Sturgeon, R. E., Prog. Anal. Spectrosc., 1989, 12, 21. 5 Reid, H. J., Greenfield, S., and Edmonds T. E., Analyst, 1995, 120, 1543. 6 Langmyhr, F.J., Analyst, 1979, 104, 993. 7 Langmyhr, F. J., Prog. Anal. At. Spectrosc., 1985, 8, 193. 8 Miller-Ihli, N. J., Anal. Chem., 1992, 64, 965A. 9 de Benzo, Z. A., Velosa, M., Ceccarelli, C., de la Guardia, M., and Salvador, A., Fresenius’ J. Anal. Chem., 1991, 339, 235. 10 Bendicho, C., and de Loos-Vollebregt, M. T. C., J. Anal. At. Spectrom., 1991, 6, 353. 11 Miller-Ihli, N. J., Fresenius’ J. Anal. Chem., 1993, 345, 482. 12 Stephen, S. C., Littlejohn, D., and Ottaway, J.M., Analyst, 1985, 110, 1147. 13 Thompson, D. D., and Allen, R. J., At. Spectrosc., 1981, 2, 53. 14 Madrid, Y., Bonilla, M., and Camara, C., J. Anal. At. Spectrom., 1989, 4, 167. 15 L�opez Garc�ýa, I., Ortiz Sobejano, F., and Hern�andez C�ordoba, M., Analyst, 1991, 116, 517. 16 Hoenig, M., and Hoeyweghen, P. V., Anal. Chem., 1986 58, 2614. 17 Albers, D., and Sacks, R., Anal. Chem., 1987, 59, 593. Table 4 Selenium concentrations (mg g21) in dried ground plant materials as determined by fluorescence (single measurement) or by ETAAS* (±1 RSD for triplicate determinations of three replicate samples) immediately after 4 or 8 h of enzymic digestion or following enzymic digestion plus 10 d of storage Strong acid 4 h digestion + 8 h digestion + digestion + Sample 4 h digestion 8 h digestion 10 d storage 10 d storage fluorescence Solvent blank + protease + cellulase 0.048 ± 14.2% 0.051 ± 12.0% 0.040 ± 23.3% 0.045 ± 12.2% 9–3 1.29 ± 10.6% 1.24 ± 9.9% 1.28 ± 8.1% 1.25 ± 9.0% 1.31 6–3 1.31 ± 12.3% 1.19 ± 9.0% 1.25 ± 9.1% 1.20 ± 8.9% 1.23 7–3 1.04 ± 12.3% 1.15 ± 9.0% 1.11 ± 6.6% 1.18 ± 6.5% 1.17 546–3 0.79 ± 8.2% 0.79 ± 11.1% 0.80 ± 9.6% 0.79 ± 10.5% 0.81 158–5 0.53 ± 9.2% 0.57 ± 8.9% 0.49 ± 11.0% 0.55 ± 7.2% 0.59 299–5 0.44 ± 13.7% 0.42 ± 8.2% 0.45 ± 18.2% 0.40 ± 10.7% 0.43 314–5 0.14 ± 16.0% 0.14 ± 13.2% 0.14 ± 20.8% 0.13 ± 10.7% 0.13 307–5 0.021 ± 23.5% 0.022 ± 24.1% 0.022 ± 24.2% 0.021 ± 20.9% 0.03 * Reported concentrations have been corrected for the [Se] in the protease + cellulase control homogenate.Table 5 Selenium concentrations (mg g21) (±1 SEE*) in freeze-dried fresh or boiled fish fillet following 4 or 8 h of enzymic digestion with protease and lipase and ETAAS Hydride Sample 4 h digestion 8 h digestion generation† Lake trout (boiled) 3.56 ± 20.9% 3.48 ± 14.1% 3.60 ± 8.0% Lake trout (fresh) 1.86 ± 11.9% 2.12 ± 13.2% 1.69 ± 10.1% Northern pike (fresh) 2.48 ± 16.6% 2.51 ± 12.5% 2.75 ± 2.9% * SEE, standard error of estimate based on three replicate determinations of three separate digests.† ±1 RSD based on duplicate determinations. Analyst, January 1997, Vol. 122 1718 Tsalev, D. L., Slaveykova, V. I., and Mandjunkov, P. B., Spectrochim. Acta, Rev., 1990, 13, 225. 19 Miller-Ihli, N. J., J. Anal. Atom. Spectrom., 1989, 4, 295. 20 Bendicho, C., and de Loos-Vollebregt, M. T. C., Spectrochim. Acta, Part B, 1990, 45, 679. 21 Miller-Ihli, N. J., J. Anal. At. Spectrom., 1988, 3, 73. 22 Lynch, S., and Littlejohn, D., J. Anal. At. Spectrom., 1989, 4, 157. 23 Fagioli, F., Landi, S., Locatelli, C., Righini, F., Settimo, R., and Magarini, R., J. Anal. At. Spectrom., 1990, 5, 519. 24 Hansen, D. L., and Bush, E. T., Anal. Biochem., 1967, 18, 320. 25 Jackson, A. J., Michael, L. M., and Schumacher, H. J., Anal. Chem., 1972, 44, 1064. 26 Murthy, L., Menden, E. E., Eller, P. M., and Petering, H. G., Anal. Biochem., 1973, 53, 365. 27 Uchida, T., Isoyama, H., Yamada, K., Oguchi, K., Nakagawa, G., Sugie, H., and Iida, C., Anal.Chim. Acta, 1992, 256, 277. 28 Dion, B., Ruzbie, M., van de Voort, F. R., Ismail, A. A., and Blais, J. S., Analyst, 1994, 119, 1765. 29 Tan, Y., Marshall, W. D, and Blais, J.-S., Analyst, 1996, 121, 483. 30 Tan, Y., Blais, J.-S., and Marshall, W. D., Analyst, 1996, 121, 1419. 31 Ebdon, L., and Perry, H. G. M., J. Anal. At. Spectrom. 1988, 3, 131. 32 Bradshaw, D., and Slavin, W., Spectrochim. Acta Part B, 1989, 44, 1245. 33 Wagley, D., Schmiedel, G., Mainka, E., and Ache, H. J., At. Spectrosc., 1989, 10, 106. 34 Bendicho, C., and Sancho, A., At. Spectrosc., 1993, 14, 187. 35 Cabrera, C., Lorenzo, M. L., and Lopez, M. C., J. AOAC Int., 1995, 78, 1061. 36 L�opez-Garc�ýa, I., Vi�nas, P., Campillo, N., and Hern�andez- C�ordoba, M., J. Agric. Food Chem.. 1996, 44, 836. 37 Gilon, N., Astruc, A., Astruc, M., and Potin-Gautier, M., Appl. Organomet. Chem., 1995, 9, 623. 38 Dedina, J. and Tsalev, D.L., Hydride Generation Atomic Absorption Spectrometry (Chemical Analysis, vol. 130), ed. Wineforder, J. D., and Kolthoff, I. M., Wiley, Chichester, 1995. 39 Johansson, K., Luo, X., and Olin, A., Talanta, 1995, 42, 1979. 40 Miller-Ihli, N. J., At. Spectrosc., 1992, 13, 1. 41 Forsyth, D. S., and Marshall, W. D., Environ. Sci. Technol.. 1986, 20, 1033. Paper 6/05880I Received August 27, 1996 Accepted October 15, 1996 18 Analyst, January 1997, Vol. 122 Enzymic Digestion–High-pressure Homogenization Prior to Slurry Introduction Electrothermal Atomic Absorption Spectrometry for the Determination of Selenium in Plant and Animal Tissues Yanxi Tan and William D.Marshall* Department of Food Science and Agricultural Chemistry, Macdonald Campus of McGill, 21 111 Lakeshore Road, Ste.-Anne-de-Bellevue, Qu�ebec, Canada H9X 3V9. E-mail: marshall@agradm.lan.mcgill.ca Homogenization, in combination with partial enzymic digerotease alone or admixed with lipase or cellulase, was investigated as a means of releasing Se residues from zoological and botanical matrices prior to slurry introduction ETAAS.Preliminary timed trials with two zoological certified reference materials (CRMs), one botanical CRM and one animal feed indicated that Se release became quantitative with 4–8 h of digestion, that homogenization prior to digestion increased the initial rate of analyte release, but that homogenization post-digestion and immediately prior to ETAAS did not significantly improve the precision of replicate digestions. Storage of the crude botanical digests at 4 °C for 5 d resulted in quantitative recoveries of Se from each of the digests.Storage at 4 °C for 10 d of 4 and 8 h lipase/protease digests of six other CRMs resulted in quantitative recoveries of Se unless their certified concentrations were appreciably less than the levels determined in control digests containing the enzyme(s) alone. Apparently, Se residues were transferred virtually quantitatively to the liquid phase of the digested suspension and showed no tendency to segregate during the 10 d of storage.Eight other mixtures of ground plant matter (0.13 @ [Se] @ 1.31 mg g21), formulated as animal feed supplements, behaved identically when stored post-digestion. The technique was also applied successfully to freeze-dried fresh and boiled fish tissues The principal advantages of the enzymic digestion procedure are its simplicity and lack of operator intervention.Keywords: Selenium determination; enzymic digestion; slurry introduction electrothermal atomic absorption spectrometry; botanical and zoological certified reference materials Conventional sample preparation of biological materials prior to atomic spectrometry involves complete solubilization of the analyte and matrix, which is achieved typically by oxidative mineralization of the organic matter and solubilization of the residue in a suitable solvent.1–4 Even for microwave-assisted digestions, whereas complete dissolution can usually be achieved by a suitable choice of digestion conditions, complete decomposition of the organic matrix in biological/botanical samples is appreciably more difficult.Often complete mineralization is achieved only with supplemental treatment of the digested matrix with H2O2 or even HF.5 These digestion procedures can be labour intensive, time consuming and prone to contamination errors.In consequence, there is a continuing interest in the development of simplified sample preparation techniques. The preparation/introduction of slurried samples continues to attract considerable attention because of the ease with which quasi-stable preparations can be generated and their compatibility with conventional liquid handling techniques. Within the general field of solid sampling analysis,6–11 it is the use of slurried samples which has become the most popular approach to trace element determination.Direct atomization from the solid state can provide excellent sensitivity, but the interpretation of results can be complicated by molecular absorptions and/or scattering from the matrix, which can produce sufficiently large background signals to overwhelm the compensation capabilities of common deuterium background correction systems. Additional difficulties can include sample inhomogeneities, the requirement for repeated microweighings and the lack of suitable calibration standards and techniques.A variety of sample pre-treatment procedures and additives12 –18 have been described and evaluated for the production of quasi-stable suspensions of samples prior to analysis by atomic spectrometry. Alternatively, suspensions with a tendency to segment rapidly have been reproducibly sampled by using ultrasonic agitation,19 air or argon20 bubbling, vortex mixing21 or magnetic stirring.22 Partial digestion procedures to produce carbonaceous slurries have also been successfully applied to the analysis, by ICP-AES, of a series of standard reference materials of biological origin.23 Various alkylammonium hydroxide formulations have been used extensively to solubilize tissues,24–27 particularly those of zoological origin.Recently, high-pressure homogenization has been evaluated for the preparation of quasi-stable dispersions suitable for FTIR28 or ETAAS.29,30 The advantages of this approach to sample preparation were the ease and speed of the slurry preparation, which required less than 1 min, and the fact that analyte metals were quantitatively extracted into the liquid phase during the preparation so that no analyte segmentation was detected within the slurry even after standing for several days.Certified reference materials (CRMs), frozen liver and kidney and dried animal feeds of botanical origin were analysed successfully for Cd, Cr, Cu, Mn, Ni and Pb but not for Se.The principal limitation of the high-pressure homogenization technique was the amounts of contaminating analyte metals introduced into the sample by the homogenization operation. Contamination was reduced appreciably, but not eliminated entirely, by capping the flat face of the stainless-steel homogenizing valve with a ruby disc.30 The objectives of this work were (i) to evaluate the efficiencies of other materials as caps to reduce further the levels of contamination introduced by the high-pressure processing and (ii) to develop efficient alternative slurry preparation techniques for the determination of Se in biological materials.Although there have been few reports of the determination of Se in slurried samples,31–34 recent reports35,36 indicate that the approach is promising for this analyte. Prolonged enzymic digestion with a crude protease fraction has been used37 to liberate component selenoamino acids from proteins.This approach seemed promising as a pre-homogenization sample preparation. Analyst, January 1997, Vol. 122 (13–18) 13Experimental Reagents TRIS was purchased from Aldrich (Milwaukee, WI, USA) and aqueous Se solution (1000 mg ml21, traceable to NIST primary standard) was purchased from SCP Chemical (St.-Laurent, Qu�ebec, Canada). Samples CRMs were purchased from the National Research Council of Canada (NRCC) or the US National Institute of Science and Technology (NIST).Samples of animal diet mixtures destined for a zoo were chosen to contain a variety of plant and animal materials, including timothy grass, bamboo leaves, whole smelts, cricket chow and a mixture (contents unspecified) formulated for panda bears. Animal feed supplements were composed of mixed forage crops. Sample Preparation An accurately weighed sample (approximately 0.2 g) of CRM, dried feed or supplements (ground, to pass a 0.5 mm screen, in a Tecator Cyclotec sample mill; Tecator, H�ogan�as, Sweden) was added directly to 10 ml of ethanol–0.03 m TRIS (1 + 19 v/v, pH 7.5) containing either 20 mg of crude protease (Type XIV; Sigma St.Louis, MO, USA), 20 mg of protease plus 20 mg of lipase (Type VII; Sigma) or 20 mg protease plus 20 mg of cellulase (Cellulysin; Calbiochem–Novabiochem, La Jolla, CA, USA). The resulting suspension was then processed through the 20 ml capacity flat valve homogenizer (EmulsiFlex Model EFB3; Avestin, Ottawa, ON, Canada), capable of developing 138 MPa when provided with compressed air (689.4 kPa).Each slurried sample was re-processed through the homogenizer three more times. The homogenates, in 50 ml Erlenmeyer flasks, were then digested at 37 °C with gentle agitation for 4–8 h. Homogenizer The valve stem of the screw-cap assembly of the homogenizer was modified by gluing a polished 4 mm diameter 32 mm thick disc manufactured from a 6–12 mm diameter sphere of tungsten carbide (Spex, Metuchen, NJ, USA), zirconia (Optimize Technologies, Portland, OR, USA), sapphire (from an HPLC piston) or polymethacrylate (Spex). Sample + enzyme suspension was transferred to the sample chamber via the inlet port, which was then sealed with a fine-threaded screw-cap.The stainless-steel piston (connected to a pneumatic multiplier) then forced the fluid through an aperture and the homogenate was collected from thle outlet. Each sample was re-processed three more times with the valve stem retracted slightly to provide a slightly larger gap setting.Selenium Determinations (Hydride Generation or Fluorescence) Feed samples were dried to constant mass and ground to pass a 1 mm screen. Accurately weighed aliquots of ground feed or freeze-dried fish tissue (approximately 2 g) were digested at room temperature in a perchlorate fume-hood with 25 ml of 70% HNO3–HClO4 (4 + 1 v/v) until gas evolution had ceased, then heated at 80 °C until a clear yellow solution was obtained. The resulting strong acid digests were analysed by HGAAS38 or fluorescence of the piazselenol derivative39 after conversion of the analyte residues in to SeIV. ETAAS Selenium determinations were performed using a hot injection technique on a Varian (Palo Alto, CA, USA) Model 300 ETAAS system equipped with an autosampler, pyrolytic graphite-coated platform graphite tubes, a conventional Se hollow-cathode lamp and Zeeman-effect background correction.Ashing–atomization curves were generated for Se standard in the presence and absence of co-injected biological sample. At temperatures < 2300 °C, the Se atomization signal was broadened by the presence of biological materials but was sharpened (and did not tail) for atomizations at 2400 °C. In the presence of the palladium–citric acid modifier, no loss in the Se signal was observed at an ashing temperature @1400 °C. Analytical operating parameters are presented in Table 1.Calibration ETAAS quantification was performed by both the method of external standards (ES) and by standard additions (SA). ES consisting of appropriately diluted processed reagent blank and up to four levels of standard were prepared automatically by the sample introduction device. The background-corrected peakarea response, resulting from three replicate injections of each diluted standard, was used to define the best-fit regression equation. For SA calibrations, 10 ml aliquots of processed fluid were amended with 2, 5 or 10 ml of aqueous standard chosen to result in a range of peak areas including signals which were half and at least twice the signal for the unamended sample.The data were modelled by least-squares linear regression. Quantification was performed by dividing the intercept on the ordinate of the regression equation by the slope of the equation and the overall standard error of the estimate (SEest) was calculated from SEest = (SE2 y-int + SE2 slope)1/2 Results and Discussion Preliminary experiments were directed to evaluating the influence of different capping materials on the levels of contaminating metals introduced into the homogenates during processing.It was postulated that exposed stainless-steel surfaces within the valve homogenizer, particularly the flat face of the demountable valve head, were the principle sites responsible for the contamination. Further, capping the valve head with an inert surface capable of withstanding the impact of the jet of fluid exiting the homogenizing orifice might reduce the levels of contamination appreciably.It has been reported40 previously that zirconium oxide beads used to reduce the particle size and to mix particulate solids introduced appreciable levels of Fe, Cr and Al but that silicon nitride or boron carbide provides good abrasion resistance and offers little likelihood of Table 1 Furnace operating parameters for determinations of selenium Parameter* Value Wavelength/nm 196.0 Lamp current/A 10 Slit width/nm 1.0 Injection temperature/°C 60 Pre-injection Yes Temperature of last dry step (10 s)/°C 250 Charring sequence 10 s ramp to 1400 °C, 40 s hold Cooling None Atomization 0.6 s ramp to 2400 °C, 5.0 s hold Measurement time/s 5.6 Chemical modifier 5 ml (0.5% m/m Pd + 2.5% m/m critic acid) for 10 ml sample * Each step of the furnace programme (with the exception of the read step) was performed in the presence of argon purge gas (3 l min21). 14 Analyst, January 1997, Vol. 122contamination. However, even for the relatively lower pressure requirements of pistons and check valves for HPLC, sapphire, ruby and zirconium oxide are preferred over other ceramic materials for their superior resistance. Separate discs composed of zirconia, tungsten carbide or polymethacrylate, which had been manufactured by grinding and polishing a 6–12 mm diameter grinding ball, were glued temporarily to the flat face of the demountable valve head.Similarly, the sapphire disc was generated from a used HPLC piston. Solvent mixture (20 ml) was homogenized four successive times (in the presence or absence of the test disc) prior to ETAAS analysis for Al, As, Cd, Cr, Cu, Fe, Mn, Ni, Pb and Se. Analyte concentrations (Table 2) were expressed as if the solvent had contained 0.100 g of sample. The heavy metal content of the homogenized fluid was lowered appreciably in all cases.Nonetheless, contamination remained appreciable for several elements, even in the presence of the polymethacrylate or the sapphire cap. Previous attempts to determine Se in biological materials by slurry introduction ETAAS of high-pressure homogenates were not successful in our hands using a variety of furnace programmes, yet there was no evidence of any analyte loss prior to the atomization stage as judged by the signal graphics software, which provided a continuous display of the Se signal over the course of the furnace programme.Since a high proportion of the analyte element in biological materials is considered to be protein bound, it was decided to evaluate partial enzymic hydrolysis as a means of liberating bound analyte residues. Arbitrarily, it had been decided initially to attempt to develop a single combination of mixed enzymes which it was hoped would be applicable to all sample matrices. Previous studies41 had indicated that a combination of crude proteases and lipases efficiently hydrolysed avian egg yolk.Initial studies were limited to this combination of enzymes. The mixture of crude enzymes was suspended in 10 ml of TRIS buffer (pH 7.5), then passed sequentially four times through the polymethacrylate-capped homogenizer to furnish a digestion control homogenate. Relative to a distilled, de-ionized water blank, this ‘zoological’ control sample contained 0.11 mg g21 ± 12.6% and 0.10 mg g21 ± 11.3% after 4 and 8 h of digestion respectively (Table 3), when it was assumed that the digests had contained 0.200 g of sample.Similarly, a control homogenate composed of protease alone contained 0.044 mg g21 ± 12.6% after 4 h. A crude cellulase was substituted for the lipase in the enzyme mixture and the digestions were performed in analogous fashion to furnish alternative enzymic digestion control samples. The ‘botanical’ control samples contained 0.048 mg g21 ± 14.2% and 0.051 mg g21 ± 12.0% after 4 and 8 h of digestion respectively (Table 3), and the solvent blank + protease alone contained 0.044 mg g21 ± 12.6%, again assuming that the digests had contained 0.200 g of sample.Thus, virtually all of the Se in control digests originated with the lipase and/or the protease. Based on a 3 RSD criterion, the corresponding method limit of detection (LOD) for digestions with mixed protease lipase, with protease cellulase and with protease alone were 0.020, 0.010 and 0.010 mg g21, respectively.In preliminary experiments, three biological CRMs and one animal feed, suspended in 10 ml TRIS buffer, were digested with a combination of protease and lipase for up to 16 h at 37 °C. Table 2 Apparent analyte concentrations (mg g21 sample) in 20 ml of solvent mixture following various mixing treatments. Concentrations in the homogenized fluid are expressed as if the solvent mixture had contained 0.100 g of sample Treatment Al As Cd Cr Cu Fe Pb Mn Ni Se Unhomogenized solvent blank 0.32 n.d.* n.d.n.d. n.d. 0.40 n.d. n.d. n.d. n.d. Four successive homogenizations with: Stainless-steel head 42.12 0.02 4.53 15.0 56.94 1.38 2.31 3.57 n.d. Polymethacrylate cap 7.92 n.d. 0.02 1.80 n.d. 5.20 n.d. 0.10 n.d. n.d. Ruby cap 21.52 n.d. 0.03 4.02 0.70 13.99 0.28 0.39 0.11 n.d. Sapphire cap 3.64 n.d. n.d. 3.65 0.80 4.90 n.d. 0.10 n.d. n.d. Tungsten carbide cap 15.32 n.d. 0.04 4.40 1.40 38.6 2.00 0.40 0.10 n.d.Zirconia cap 0.42 n.d. 0.02 4.00 1.20 15.7 n.d. 0.20 0.10 n.d. * n.d. = None detected above the mean background signal for ethanol–0.03 m TRIS (1 + 19 v/v, pH 7.5) solvent. Table 3 Selenium concentrations (mg g21) (±1 RSD based on three replicate samples) in certified reference materials determined immediately after 4 or 8 h of enzymic digestion or following digestion plus 10 d of storage 4 h digestion + 8 h digestion + Certified Matrix 4 h digestion 8 h digestion 10 d storage 10 d storage concentration Solvent blank + protease + lipase 0.11 ± 12.6% 0.10 ± 11.3% 0.12 ± 10.2% 0.10 ± 13.6% Solvent blank + protease 0.044 ± 12.6% Solvent blank + protease + cellulase 0.048 ± 14.2% 0.051 ± 12.0% 0.040 ± 23.2% 0.045 ± 12.2% Oyster Tissue* 2.18 ± 12.7% 2.27 ± 11.6% 2.16 ± 11.4% 2.15 ± 12.6% 2.21 ± 0.24 DORM-2* 1.38 ± 9.2% 1.35 ± 11.6% 1.34 ± 6.6% 1.33 ± 11.0% 1.40 ± 0.090 Bovine Muscle* 0.067 ± 22.6% 0.066 ± 20.5% 0.067 ± 20.2% 0.070 ± 21.4% 0.076 ± 0.010 Apple Leaves† 0.043 ± 12.7% 0.045‡ ± 19.9% 0.047 ± 12.7% 0.041 ± 12.7% 0.050 ± 0.009 Corn Bran† 0.034‡ ± 12.5% 0.036 ± 14.9% 0.033 ± 15.3% 0.036 ± 16.5% 0.045 ± 0.008 Corn Stalk† n.d.§ n.d.§ 0.025‡ ± 47.0% 0.011‡ ± 52.2% 0.016 ± 0.008 * Reported concentrations have been corrected for the [Se] in the protease + lipase control homogenate.† Reported concentrations have been corrected for the [Se] in the protease + cellulase control homogenate. ‡ No [Se] above the LOD (0.010 mg g21) was detected in one of the three aliquots.§ No [Se] above the LOD (0.010 mg g21) was detected in any of the three aliquots. Analyst, January 1997, Vol. 122 15Each suspension was homogenized immediately prior or directly after digestion, then analysed by ETAAS. The results are presented in Figs. 1 and 2. The TORT-1 results and the animal feed results (triangular symbols in Fig. 1 and Fig. 2, respectively) have been displaced by 0.4 h for clarity of presentation. For all four substrates, homogenization prior to digestion (closed symbols) generally resulted in higher recoveries relative to homogenization post-digestion (open symbols), although the differences were only rarely statistically significant.Moreover, the differences tended to decrease with longer digestion times. Presumably, homogenization initially exposed more of the protein component to the enzyme. On the other hand, homogenization post-digestion and immediately prior to ETAAS did not significantly improve the precision of the determination (as judged by the RSD associated with three replicate measurements performed on each of three digests).In general, there was a gradual but small improvement in precision with increased length of digestion (more evident with the plant samples in Fig. 2). For both the DORM-1 and the TORT-1 marine tissue samples in Fig. 1, 4 h of digestion at 37 °C were sufficient to liberate the Se quantitatively, whereas recoveries from the plant samples were quantitative only after 8 h of digestion.After sampling for ETAAS, the plant digests were stored at 4 °C for 5 d and then re-analysed. No effort was made to resuspend solid materials; instead, a portion of each supernatant fraction was transferred directly to the sampling cup. The recovery of Se from each of the supernatant fractions was quantitative (Fig. 3), indicating that the crude protease was active at the storage temperature and that there was no apparent segmentation of the Se residues between the liquid and solid phases of the crude digest.Surprisingly, the short-term repeatability of the procedure was not improved by the storage, as evidenced by the RSD associated with replicates. Repeatability continued to be improved for longer digestions at 37 °C. Three replicate aliquots of each of six other certified reference materials were homogenized and then digested for 4 or 8 h prior to ETAAS. The results, corrected for the Se concentration in the appropriate zoological or botanical control digest, are presented in Table 3.Whereas digestion of the three marine CRM homogenates provided estimates which were not significantly different from the certified Se concentrations (despite the higher Se concentration in the zoological control) the lower concentrations in the botanical CRMs resulted in estimates which, occasionally, were not different from the control concentrations. In the latter cases, the certified Se concentrations were less than the Se concentrations in control digests. Likewise, three aliquots (approximately 0.2 g) of each of eight dried, ground feed supplements consisting of mixed forage crops were suspended in 10 ml of TRIS buffer, homogenized and then digested with the protease–cellulase combination for either 4 or 8 h prior to Se determination by ETAAS.Again, there was good agreement between the results (Table 4) for slurry introduction ETAAS following 4 or 8 h of enzymic digestion and a single fluorescence determination Fig. 1 Variation in percentage recovery of Se (±1 RSD based on three replicate samples) from DORM-1 or TORT-1 certified reference materials versus hours of enzymic digestion with protease plus lipase prior to (open symbols) or after (closed symbols) high-pressure homogenization.For clarity of presentation, the TORT-1 results have been displaced by 0.4 h. Fig. 2 Variation in percentage recovery of Se (±1 RSD based on three replicate samples) from Durham wheat flour CRM or a ground animal feed sample versus hours of enzymic digestion with protease plus cellulase prior to (open symbols) or after (closed symbols) high-pressure homogenization.For clarity of presentation, the animal feed results have been displaced by 0.4 h. Fig. 3 Influence of storage subsequent to slurry preparation on percentage recovery of Se (±1 RSD based on three replicate samples) in flour CRM or a ground animal feed sample.For clarity of presentation, the animal feed results have been displaced by 0.4 h. 16 Analyst, January 1997, Vol. 122following strong acid digestion and piazselenol formation. Storage of the digests for 10 d at 4 °C did not change the measured concentrations of analyte (Table 4). In only one feed supplement (in which the Se concentration was appreciably less than that in the botanical control digest) were the results of the two methods discordant.No matrix effects for Se determinations in any of the samples were detected. The slopes of the best-fit regression lines for standard additions to homogenized protease–lipase or protease– cellulase enzyme suspensions in TRIS buffer, in the presence or absence of DORM-1, TORT-1, wheat flour, corn bran or apple leaf CRM, or to five of the feed supplements varied by less than 11% (RSD) provided that calibrations and determinations were performed on the same day. This observation suggested that a single calibration by standard addition(s) to the enzyme mixtures would suffice for the determination of Se in any of the samples.A single calibration curve generated by adding aqueous Se standard to the botanical control homogenate was then used to determine the Se content of freeze-dried freshwater fish fillets which had been frozen fresh or boiled to simulate cooking following common native practice. Aliquots of the freeze-dried materials were digested enzymically for 4 or 8 h and then analysed by ETAAS or digested with strong acids and then analysed by HGAAS (Table 5).Boiling the fillet prior to freezedrying did not inhibit the enzymic release of Se residues from the matrix but apparently lowered the Se concentration in the cooked product. There were no significant differences between the results after 4 and 8 h of digestion or between ETAAS and HGAAS results. However, the precision associated with replicate enzymic digestion–ETAAS Se determinations (mean RSD nearly 15 ± 2%) was appreciably worse than the precision associated with hydride generation determinations (mean 7 ± 4%) but typical of the replicate determinations of other experiments (mean RSD for the 36 determinations in Table 3 14.4 ± 0.3% and for 29 of the determinations in Table 4 12.6 ± 0.5%).Thus, the short-term repeatability (i) was not adversely affected by the use of the single calibration curve but (ii) can be expected to be degraded relative to other conventional procedures for Se determination. The principal advantages of the enzymic digestion procedure are the simplicity and speed relative to conventional unassisted acid digestions and that they can be performed unattended.The conditions of digestion do not appear to be critical and there was no tendency for the liberated Se residues to segregate within the resulting suspensions. Feed samples and feed supplements and determinations of their Se content by fluorescence of their piazselenol derivatives were generously supplied by E.R. Chavez, McGill University. Samples of fish fillets and determinations of their Se content by HGAAS were kindly donated by H. M. Chan, McGill University. Financial support in the form of an operating grant from the Natural Science and Engineering Research Council of Canada (NSERC) is gratefully acknowledged. References 1 Novozamski, I., van der Lee, H. J., and Houba, V. J. G., Microchim.Acta, 1995, 119, 183. 2 Sansoni, B., and Panday, V. K., in Analytical Techniques for Heavy Metals in Biological Fluids, ed. Fachetti, S., Elsevier, Amsterdam, 1983, p. 91. 3 Mincwewski, J., Chwastowska, J., and Dybczynski, R., Separation and Preconcentration Methods in Inorganic Trace Analysis, Ellis Horwood, Chichester, 1982. 4 Matusiewicz, J., and Sturgeon, R. E., Prog. Anal. Spectrosc., 1989, 12, 21. 5 Reid, H. J., Greenfield, S., and Edmonds T. E., Analyst, 1995, 120, 1543. 6 Langmyhr, F. J., Analyst, 1979, 104, 993. 7 Langmyhr, F. J., Prog. Anal. At. Spectrosc., 1985, 8, 193. 8 Miller-Ihli, N. J., Anal. Chem., 1992, 64, 965A. 9 de Benzo, Z. A., Velosa, M., Ceccarelli, C., de la Guardia, M., and Salvador, A., Fresenius’ J. Anal. Chem., 1991, 339, 235. 10 Bendicho, C., and de Loos-Vollebregt, M. T. C., J. Anal. At. Spectrom., 1991, 6, 353. 11 Miller-Ihli, N. J., Fresenius’ J. Anal. Chem., 1993, 345, 482. 12 Stephen, S. C., Littlejohn, D., and Ottaway, J.M., Analyst, 1985, 110, 1147. 13 Thompson, D. D., and Allen, R. J., At. Spectrosc., 1981, 2, 53. 14 Madrid, Y., Bonilla, M., and Camara, C., J. Anal. At. Spectrom., 1989, 4, 167. 15 L�opez Garc�ýa, I., Ortiz Sobejano, F., and Hern�andez C�ordoba, M., Analyst, 1991, 116, 517. 16 Hoenig, M., and Hoeyweghen, P. V., Anal. Chem., 1986 58, 2614. 17 Albers, D., and Sacks, R., Anal. Chem., 1987, 59, 593. Table 4 Selenium concentrations (mg g21) in dried ground plant materials as determined by fluorescence (single measurement) or by ETAAS* (±1 RSD for triplicate determinations of three replicate samples) immediately after 4 or 8 h of enzymic digestion or following enzymic digestion plus 10 d of storage Strong acid 4 h digestion + 8 h digestion + digestion + Sample 4 h digestion 8 h digestion 10 d storage 10 d storage fluorescence Solvent blank + protease + cellulase 0.048 ± 14.2% 0.051 ± 12.0% 0.040 ± 23.3% 0.045 ± 12.2% 9–3 1.29 ± 10.6% 1.24 ± 9.9% 1.28 ± 8.1% 1.25 ± 9.0% 1.31 6–3 1.31 ± 12.3% 1.19 ± 9.0% 1.25 ± 9.1% 1.20 ± 8.9% 1.23 7–3 1.04 ± 12.3% 1.15 ± 9.0% 1.11 ± 6.6% 1.18 ± 6.5% 1.17 546–3 0.79 ± 8.2% 0.79 ± 11.1% 0.80 ± 9.6% 0.79 ± 10.5% 0.81 158–5 0.53 ± 9.2% 0.57 ± 8.9% 0.49 ± 11.0% 0.55 ± 7.2% 0.59 299–5 0.44 ± 13.7% 0.42 ± 8.2% 0.45 ± 18.2% 0.40 ± 10.7% 0.43 314–5 0.14 ± 16.0% 0.14 ± 13.2% 0.14 ± 20.8% 0.13 ± 10.7% 0.13 307–5 0.021 ± 23.5% 0.022 ± 24.1% 0.022 ± 24.2% 0.021 ± 20.9% 0.03 * Reported concentrations have been corrected for the [Se] in the protease + cellulase control homogenate.Table 5 Selenium concentrations (mg g21) (±1 SEE*) in freeze-dried fresh or boiled fish fillet following 4 or 8 h of enzymic digestion with protease and lipase and ETAAS Hydride Sample 4 h digestion 8 h digestion generation† Lake trout (boiled) 3.56 ± 20.9% 3.48 ± 14.1% 3.60 ± 8.0% Lake trout (fresh) 1.86 ± 11.9% 2.12 ± 13.2% 1.69 ± 10.1% Northern pike (fresh) 2.48 ± 16.6% 2.51 ± 12.5% 2.75 ± 2.9% * SEE, standard error of estimate based on three replicate determinations of three separate digests. † ±1 RSD based on duplicate determinations. Analyst, January 1997, Vol. 122 1718 Tsalev, D. L., Slaveykova, V. I., and Mandjunkov, P. B., Spectrochim. Acta, Rev., 1990, 13, 225. 19 Miller-Ihli, N. J., J. Anal. Atom. Spectrom., 1989, 4, 295. 20 Bendicho, C., and de Loos-Vollebregt, M. T. C., Spectrochim. Acta, Part B, 1990, 45, 679. 21 Miller-Ihli, N. J., J. Anal. At. Spectrom., 1988, 3, 73. 22 Lynch, S., and Littlejohn, D., J. Anal. At. Spectrom., 1989, 4, 157. 23 Fagioli, F., Landi, S., Locatelli, C., Righini, F., Settimo, R., and Magarini, R., J. Anal. At. Spectrom., 1990, 5, 519. 24 Hansen, D. L., and Bush, E. T., Anal. Biochem., 1967, 18, 320. 25 Jackson, A. J., Michael, L. M., and Schumacher, H. J., Anal. Chem., 1972, 44, 1064. 26 Murthy, L., Menden, E. E., Eller, P. M., and Petering, H. G., Anal. Biochem., 1973, 53, 365. 27 Uchida, T., Isoyama, H., Yamada, K., Oguchi, K., Nakagawa, G., Sugie, H., and Iida, C., Anal. Chim. Acta, 1992, 256, 277. 28 Dion, B., Ruzbie, M., van de Voort, F. R., Ismail, A. A., and Blais, J. S., Analyst, 1994, 119, 1765. 29 Tan, Y., Marshall, W. D, and Blais, J.-S., Analyst, 1996, 121, 483. 30 Tan, Y., Blais, J.-S., and Marshall, W. D., Analyst, 1996, 121, 1419. 31 Ebdon, L., and Perry, H. G. M., J. Anal. At. Spectrom. 1988, 3, 131. 32 Bradshaw, D., and Slavin, W., Spectrochim. Acta Part B, 1989, 44, 1245. 33 Wagley, D., Schmiedel, G., Mainka, E., and Ache, H. J., At. Spectrosc., 1989, 10, 106. 34 Bendicho, C., and Sancho, A., At. Spectrosc., 1993, 14, 187. 35 Cabrera, C., Lorenzo, M. L., and Lopez, M. C., J. AOAC Int., 1995, 78, 1061. 36 L�opez-Garc�ýa, I., Vi�nas, P., Campillo, N., and Hern�andez- C�ordoba, M., J. Agric. Food Chem.. 1996, 44, 836. 37 Gilon, N., Astruc, A., Astruc, M., and Potin-Gautier, M., Appl. Organomet. Chem., 1995, 9, 623. 38 Dedina, J. and Tsalev, D. L., Hydride Generation Atomic Absorption Spectrometry (Chemical Analysis, vol. 130), ed. Wineforder, J. D., and Kolthoff, I. M., Wiley, Chichester, 1995. 39 Johansson, K., Luo, X., and Olin, A., Talanta, 1995, 42, 1979. 40 Miller-Ihli, N. J., At. Spectrosc., 1992, 13, 1. 41 Forsyth, D. S., and Marshall, W. D., Environ. Sci. Technol.. 1986, 20, 1033. Paper 6/05880I Received August 27, 1996 Accepted October 15, 1996 18 Analyst, Janu

ISSN:0003-2654    

DOI:10.1039/a605880i

出版商:RSC    

年代:1997

数据来源: RSC

摘要:

Assessment of Dowex 1-X8-based Anion-exchange Procedures for the Separation and Determination of Ruthenium, Rhodium, Palladium, Iridium, Platinum and Gold in Geological Samples by Inductively Coupled Plasma Mass Spectrometry Ian Jarvis*a, Marina M. Totland†a and Kym E. Jarvisb a School of Geological Sciences, Kingston University, Penrhyn Road, Kingston upon Thames, Surrey, UK KT1 2EE b NERC ICP-MS Facility, Centre for Analytical Research in the Environment, Imperial College at Silwood Park, Buckhurst Road, Ascot, Berkshire, UK SL5 7TE Synthetic multielement solutions of the platinum group metals (PGE: Ru; Rh; Pd; Ir; Pt) and gold, with analysis by ICP-AES and ICP-MS, have been used to study the behaviour of the precious metals on Dowex 1-X8 resin.Simple solutions of precious-metal chlorocomplexes showed near-complete adsorption ( > 99%) of most elements, and only minor breakthrough of Ru and Ru ( Å 5%). Solutions pre-treated with acid mixtures typically used to decompose geological samples, demonstrated that perchloric acid adversely affects the adsorption of the PGEs on the resin.Solutions treated with HF–HNO3–HCl maintained good retention of Ir, Pt, Au ( > 99%), Pd ( > 94%) and Ru ( > 90%), but displayed significant loss (up to 40%) of Rh. A two-step procedure was necessary to elute the precious metals from the resin: 0.3 mol l21 thiourea prepared in 0.1 mol21 HCl removed Ru, Pd, Pt, Au, and some Rh: 12 mol l21 HCl eluted remaining Rh and all Ir.Recoveries ranged from 50 to 100%. At low levels, the determination of PGE and Au in the thiourea fraction by ICP-MS was compromised by high levels of total dissolved solids (TDS), which necessitated dilution of the eluate prior to analysis. The TDS was reduced by decomposing thiourea with HNO3 and removing SO4 22 by precipitation of BaSO4, but this led to lower and more erratic results, and increased contamination. An assessment of the optimised procedure employing geological reference materials PTM-1, PTC-1 and SARM7, indicated that acceptable results should be attainable for ICP-MS determination of most elements in geological samples containing high concentrations ( > 1 mg g21) of the PGE, for which decomposition of thiourea is unneccessary.The addition of a decomposition step led to low recovery of all elements except Ir, which was present entirely in the HCl eluate. The method is viable for the determination of Ir in a range of geological materials, but modifications will be required if it is to be extended to the other precious metals.Keywords: Platinum group element; gold; cation exchange chromatography; geological sample; inductively coupled plasma mass spectrometry The commonest methods1–3 used for the determination of the platinum group elements (PGE: Ru; Rh; Pd; Os; Ir; Pt) and gold are based on fire assay, which involves fusing 10–100 g samples with large amounts of alkali flux and collecting the PGE into a nickel sulfide or, for Au, a lead button.The nickel sulfide button is usually dissolved in acid to remove the nickel and sulfur, leaving a residue of PGE sulfides which may be analysed directly by neutron activation analysis (NAA). More commonly, the precipitate is taken into solution for analysis by AAS, direct coupled plasma-atomic emission spectroscopy (DCP-AES), or ICP-AES. Although fire assay is used extensively, it is highly labour intensive, and one of its most serious limitations is a dependence of the quality of results on the experience of the analyst.A more robust method would clearly be advantageous. Fire assay is also not well suited to the analysis of small ( < 5 g) samples, which may be required for some geological studies, such as investigations of precious metal mobility in sediment cores.4 We have previously described5 a new combined microwave digestion–minifusion method with analysis by ICP-MS, that yields quantitative data for Ru, Rh, Pd, Ir, Pt and Au in mineralised samples, but is limited by modest lower limits of determination in samples of 0.2–1 mg g21 (ppm).The objective of the present study was to investigate anion-exchange separation procedures for the PGE and Au, that would allow the determination of these elements at the lower levels (1–10 ng g21; ppb) found in most geological materials. Similar techniques have been employed by others6–8 with varying success, to determine individual or small groups of precious metals in a variety of sample types.Dowex 1-X8 is a strong base anion-exchanger with a styrene divinyl benzene polymer skeleton to which tertiary ammonium groups have been bound.7 Dowex 1-X8 exhibits a high selectivity for the PGE and Au, and it has been shown7,9 that it will remove these elements quantitatively from HCl solutions. However, the resin has not been used previously to simultaneously separate and preconcentrate the entire group of PGE and Au from solutions of geological materials, prior to their analysis by ICP-MS.The aims of this study were to: (a) develop experimental conditions under which 100% of the PGE and Au present in solution would be absorbed onto the Dowex 1-X8 resin, with minimal adsorption of competing ions; (b) find a set of conditions to elute 100% of the adsorbed PGE and Au; (c) develop a methodology to quantify the PGE and Au at ng ml21 levels in the eluate using ICP-MS.The first objective imposes limitations on the solution chemistry of the digested geological materials that are to undergo separation, while the second is known7 to be difficult to achieve from strong base cationexchange resins. Osmium was not included in the study because its volatility precludes its determination in most solutions of geological materials prepared for other precious metal analysis. 2,3,10 Experimental Instrumentation A JY 70 Plus (Jobin-Yvon, Longjumeau, France) inductively coupled plasma atomic emission spectrometer, located at † Present address: Atomic Energy of Canada Ltd., Chalk River Laboratories, Chalk River, Ontario, Canada KOJ 1JO.Analyst, January 1997, Vol. 122 (19–26) 19Kingston University, was used to determine the PGE and Au at high levels ( > 50 ng ml21) during method development. A VG Elemental PlasmaQuad PQ2 Plus (Fisons Instruments, Winsford, Cheshire, UK) ICP-MS, formerly at Royal Holloway University of London, was employed for lower level ( < 50 ng ml21) work.Operating conditions for the two instruments are given in Tables 1 and 2. In both cases, an external drift monitor (generally, a calibration standard solution) was used to correct for signal changes with time. Spectral lines used for PGE and Au determination by ICPAES were selected experimentally. Tables of spectral data11–13 were used to identify the most sensitive wavelengths for each element.A 1 mol l21 ARISTAR HCl (Merck, Lutterworth, UK) blank solution, a 1 mg ml21 solution of the analyte in the same matrix, and a multi-element solution containing all of the precious metals at 1 mg ml21 (both prepared using aliquots of single-element 1000 mg ml21 precious metal standard solutions, supplied in 20% v/v HCl; Johnson Matthey, Royston, UK), were each scanned14 across a 0.12 nm window, centred on the spectral line selected. The emission lines chosen (Table 3) were the most sensitive lines having no interference from other elements in the group.As only simple solutions of the precious metals were to be analysed by ICP-AES, interferences from other elements were not considered. Subsequently, calibration was achieved using five synthetic multi-element PGE and Au standards, prepared using aliquots of the 1000 mg ml21 singleelement standard solutions. Solutions analysed were generally in 0.1–1 mol l21 ARISTAR HCl; matrix effects were minimised by matching the acidity and salt content of samples and standard solutions.Detection limits (Table 3) calculated as 3s standard deviations of eleven determinations of a 1 mol l21 ARISTAR HCl blank, demonstrate that ICP-AES is well suited to the quantification of PGE and Au at concentrations in excess of 50 ng ml21. The ICP-MS was optimised using 59Co and 238U to give maximum sensitivity whilst minimising interferences, particularly refractory oxide species. The PGE and Au lie in two distinct parts of the mass range, 96–110 and 184–198 u.Some isotopes have an isobaric overlap and these were not used for analysis. The mass spectrometer was scanned from 98–200 u, with masses between 112–179 u being skipped to maximise the integration time spent on PGE and Au isotopes. During data processing, the existence of polyatomic or oxide interferences was determined by comparing isotopic ratios for elements with their theoretical values; only isotopes free from all spectroscopic interferences (Table 3) were used for quantification. External calibration of the ICP-MS for PGE and Au determinations was accomplished using multielement standards prepared in dilute HCl from 1000 mg ml21 single-element standard solutions; in all cases, the acid concentration was matched to the samples.A single standard (ranging between 20 and 200 ng ml21, depending on the expected concentration in the samples) and a blank were used as calibration points.It was noted that some elements displayed a significant memory effect (Au and, to a lesser extent, Pd), and extended washout times were necessary to avoid sample carryover. For example, following a solution containing 200 ng ml21 of the PGE and Au, a 3–4 min washout (using maximum pumping speed) with 1 mol l21 HCl was required to minimise the background level of Au. To avoid cross-contamination between high- and lowconcentration solutions, the integrals for all elements were monitored throughout the analytical run, and wash times were extended after solutions containing high levels of the precious metals had been analysed.Blanks were measured to establish the background level of these elements throughout the run, and blank corrections were applied where necessary. The measured detection limits for all elements by ICP-MS (Table 3) were better than 0.22 ng ml21. It is worth noting that since this study was completed, a newer instrument fitted with an Enhanced Performance Interface (Fisons Instruments) has been installed, which yields detection limits for the PGE and Au between 0.006 and 0.05 ng ml21.However, these do not translate directly to improved lower limits of determination for samples, because the new instrument displays poorer tolerance of total dissolved solids (TDS), necessitating higher dilution factors for solutions. Nevertheless, with these exceptionally low detection limits, ICP-MS is ideally suited to the determination of low concentrations of PGEs and Au in simple solutions.ICPMS is also far less prone to interferences than most other instrumental techniques.15,16 However, limitations on the types of solution analysed are imposed by the level of TDS; upper limits of approximately 0.2% TDS may be aspirated into an ICP-MS instrument without causing significant instrumental drift and/or matrix effects, so solutions had to be diluted to an appropriate level prior to analysis.Anion-exchange Experiments The Dowex 1-X8 resin (Bio-Rad Laboratories, Hemel Hempstead, UK) used here was nominally in chloride form (i.e., with chlorine counter-ions) and graded from 200–400 mesh (74–37 mm). However, the resin supplied contained a much Table 1 ICP-AES operating parameters for PGE and Au determination Instrument Jobin-Yvon JY 70 Plus Plasma power 1000 W Reflected power < 5 W Coolant gas flow 12 l min21 Auxiliary gas flow 0 l min21 Sheath gas flow 0.2 l min21 Nebuliser gas flow 0.35 l min21 Nebuliser Meinhard TR-50-C1 concentric glass Sample uptake 1 ml min21 Spray chamber JY Scott-type, double pass Spectrometer Czerny–Turner monochromator Grating Holographic 3600 grooves mm21 Integration period 10 s Table 2 ICP-MS operating parameters for PGE and Au determination Instrument VG PlasmaQuad PQ2 Plus Plasma power 1300 W Reflected power < 1 W Coolant gas flow 14 l min21 Auxiliary gas flow 0.5 l min21 Nebuliser gas flow 0.75 l min21 Nebuliser de Galan high dissolved solids Sample uptake rate 0.5 ml min21 Spray chamber Surrey design, single-pass, water-cooled at 4 °C Scan regions 98–111 and 180–200 u Table 3 Detection limits for the PGE and Au by ICP-AES and ICP-MS.Values calculated as 3s standard deviations for 11 determinations of a 1 mol l21 ARISTAR HCl blank ICP-AES ICP-MS Wavelength/ Detection limit/ Mass Detection limit/ Element nm ng ml21 (u) ng ml21 Ru 245.66 14 101 0.22 Rh 233.48 11 103 0.03 Pd 340.46 13 105 0.17 Ir 224.27 9.0 193 0.07 Pt 214.42 29 195 0.11 Au 242.80 5.9 197 0.06 20 Analyst, January 1997, Vol. 122wider range of grain-sizes than the nominal fraction, including a significant amount of very fine particles, possibly produced during packaging and handling.These were removed by slurrying the resin with deionised water, allowing the resin to settle for a short time, and then decanting off the suspended fines with the supernatant. This was repeated at least three times to remove most of the fine material.The resin was batch-cleaned prior to use by slurrying it with 6 mol l21 AnalaR HCl (Merck), allowing it to stand for 10–20 min, and then decanting off the acid. This procedure was repeated at least three times, or until no further discolouration of the acid was observed. After pouring off the last portion of the cleaning acid, the resin was slurried with 1 mol l21 ARISTAR HCl, and stored ready for use. For each ion-exchange experiment, following loading on the column, the resin was conditioned by passing 100 ml (later increased to 500 ml) 1 mol l21 ARISTAR HCl through the column.Only new resin was used for this work. To evaluate possible anion-exchange procedures, two parameters were measured: (1) percentage of the PGE and Au not adsorbed onto the resin, termed ‘breakthrough’; (2) percentage of the adsorbed metals eluted from the column, termed ‘recovery’. Unless stated otherwise, Merck ARISTAR reagents and ultra-pure (better than 18 MW cm) deionised water were used throughout this study.Adsorption of the PGEs and Au on Dower 1-X8 resin The PGEs must be present in appropriate oxidation states to ensure strong adsorption onto Dowex 1-X8 resin,7 and the charge of the PGE-complex may change depending on the ligands attached to the metal. Preliminary evaluation of the resin was conducted using stable solutions of PGE anionic chlorocomplexes, supplied at 1000 mg ml21 and stored in 20% v/v HCl solutions (Johnson Matthey, Royston, UK). Breakthrough of each PGE and Au was determined using synthetic multielement solutions produced by diluting aliquots of 1000 mg ml21 standards with 1 mol l21 HCl to yield 500 mg and 1 mg spikes of each element, diluted to 10 ml.Column dimensions were chosen based on previous studies6–9,17–23 to hold a settled resin bed of 1 cm diameter, 10 cm long. The percentage breakthrough of each element was determined by analysing the 1 mol l21 HCl eluate (25 ml) collected as the spike solutions were loaded onto the column.Solutions generated during experiments with the 500 mg spike were sufficiently concentrated to enable analysis by ICP-AES. This high level was chosen to determine whether the capacity of the resin was likely to be exceeded in normal use. The proportion of each element not retained on the column was small: < 0.1% Pd, Pt, Au; 0.2% Ir; 4.6% Rh; 7.3% Ru. A level of precious metals much closer to that expected for geological materials (1 mg) was used to evaluate breakthrough at lower concentrations. In this instance, absolute breakthrough was expected to be very small, so ICP-MS was used as the analytical finish.Results showed no significant breakthrough of Pd, Pt, Au ( < 0.1%), Ir (0.1%) and Rh (0.8%), and < 10% breakthrough of Ru from the 1 mg spike. While the above results verified the adsorption characteristics of Dowex 1-X8 described in the literature, they represent a simplified situation.The chemical state of the PGE and Au following digestion of geological materials5,24 may not be identical to those found in standard solutions. The effects of digestion procedures on the chemical state of individual PGE and Au needed to be established experimentally. Knowledge of the exact chemical state of each element is not necessary, provided that the PGE and Au are completely adsorbed onto the resin.It is not possible to measure breakthrough of the PGE and Au directly with geological materials, because the high concentrations of matrix elements passing through the columns preclude the determination of low levels of precious metals in the eluate. To estimate possible breakthrough when real samples are separated by this method, synthetic solutions were treated in an identical manner to geological samples. In this way, the effects of various acids used in digestion procedures could be evaluated. An HF–HClO4-based acid attack is one of the most commonly used methods to digest geological samples.24–29 To assess the affects of these acids on column retention, a 1 mol l21 HCl solution containing 500 mg of PGE and Au was treated as follows: (1) 8 ml of 16 mol l21 HNO3, 4 ml of 29 mol l21 HF, 4 ml of 12 mol l21 HClO4 were added in a PTFE beaker, evaporated at 200 °C to incipient dryness, a further 2 ml of HClO4 (12 mol l21) were added, evaporated to near-dryness, and the final solution made to 10 ml with 1 mol l21 HCl; (2) 4 ml of 16 mol l21 HNO3, 4 ml of 29 mol l21 HF, 4 ml of 12 mol l21 HClO4 were added and evaporated to incipient dryness, 8 ml of 12 mol21 HCl were added and evaporated to neardryness (twice), and the final solution also made to 10 ml with 1 mol l21 HCl.Breakthrough from these solutions were measured by ICP-AES from 1 cm diameter, 10 cm long columns of Dowex 1-X8. Breakthrough increases significantly following acid pre-treatment (Table 4).Losses at this stage were reduced when the solution was evaporated twice with 12 mol l21 HCl, instead of HClO4. Nevertheless, breakthrough of Rh and Ir remained very high, 58 and 18%, respectively. A comparison made using different elution volumes of the HClO4- evaporated solution (Table 4), demonstrated that the amount of breakthrough increased dramatically when 65 ml of 1 mol l21 HCl was eluted compared to 25 ml; in the former case, no Rh was retained on the resin.The level of breakthrough and the dependence on elution volume are clearly unacceptable for this separation method, precluding the use of HClO4-based digests prior to cation-exchange chromatography. A series of 1 mol l21 HCl solutions spiked with PGE and Au were evaporated in PTFE beakers at 100 °C on a hotplate, in the presence of 10 ml of 29 mol l21 HF, 15 ml of 12 mol l21 HCl and 5 ml of 16 mol l21 HNO3. This acid mixture has been demonstrated5 to be effective for the digestion of preciousmetal- bearing geological samples.Three masses of PGE and Au spike (10, 1, 0.1 mg) showed greatly reduced losses following the application of this acid mixture (Table 5), and subsequent evaporation (twice) with 4 ml of 12 mol l21 HCl to ensure conversion of elements to chloride form. Final solutions were approximately 10 ml of 1 mol l21 HCl; 25 ml of 1 mol l21 HCl were eluted and analysed by ICP-MS. Breakthrough was considered to be acceptably low ( < 10%) for Ru, Pd, Ir, Pt and Au, but the loss of Rh was high (27–40%).Elution conditions Dowex 1-X8 resin has a high affinity for the PGE and Au in dilute HCl solutions, but increased acid concentrations favour their removal.7 Distribution coefficients fall significant with Table 4 Breakthrough (%) of 500 mg of PGE and Au on Dowex 1-X8 anionexchange resin following pretreatments with perchloric acid. HClO4 = evaporated with HNO3, HF, HClO4; HClO4 added and evaporated.HCl = evaporated with HNO3, HF, HClO4; HCl added and evaporated. Both final solutions in 1 mol l21 HCl HClO4 HCl No pretreatment Element 25 ml 65 ml 25 ml 25 ml Ru 12 18 6.1 7.3 Rh 80 100 58 4.6 Pd 0.4 1.0 < 0.1 < 0.1 Ir 9.4 31 18 0.2 Pt 47 53 2.0 < 0.1 Au 0.4 1.0 < 0.1 < 0.1 Analyst, January 1997, Vol. 122 21increasing acid molarity, although there remains significant adsorption of some PGE even in concentrated HCl. Exceptions are Ir3+ and Rh3+ which have distribution coefficients of < 1 in 12 mol l21 HCl, and Ru4+, which is slightly higher at 10.7 The inference from these observations is that concentrated HCl will elute only part of the Ru, Rh and Ir adsorbed on the column; increasing the concentration of the counter ion in the system will not be sufficient to elute the entire group of elements.Thiourea has been successfully used to elute Pd, Pt and Au from strong base anion-exchange resins;7 selected PGE and Au are either reduced and/or complexed by thiourea, and the resulting complexes have low affinity for the resin.Elution experiments were conducted using 500 mg of each PGE and Au loaded in 10 ml of 1 mol l21 HCl. These solutions are coloured, and it was possible to get a quick assessment of elution conditions by observing the migration of coloured bands on the columns. Using this approach, it became apparent that although a portion of the PGE and Au could be eluted with thiourea, some remained on the column.It was concluded that a two-step procedure was necessary to elute the PGE and Au from a Dowex 1-X8 column. Three 25 ml aliquots of 0.3 mol l21 thiourea (4.7 g AnalaR thiourea dissolved in 200 ml of deionised water, acidified to 0.1 mol l21 HCl using 1.7 ml of 12 mol l21 HCl; selected based on work by Korkisch7) were eluted through the Dowex 1-X8 columns used for the breakthrough study. Solutions from the 500 mg experiments were diluted four-fold prior to ICP-AES analysis; 1 mg solutions were diluted 40-fold and analysed by ICP-MS.Dilution was necessary to reduce the high levels of TDS from the thiourea, to levels acceptable for each technique. 16 Aliquots of 25 ml of 12 mol l21 HCl were then eluted through each column, collected, diluted 10-fold to reduce the acid concentration, and analysed by ICP-AES or ICP-MS. A total of 100 ml of 12 mol l21 HCl was collected for the 500 mg experiment, and 125 ml from the 1 mg spike.Elution profiles (Fig. 1) show that most Ru, Pd, Pt and Au are eluted by 75 ml of 0.3 mol l21 thiourea, while Rh is only partly eluted and Ir remains bound to the resin. Remaining Rh and Au are completely eluted from the column with 12 mol l21 HCl, and Ir is removed by this eluent. Elution profiles are instructive for determining the rate of elution of elements and the relative efficacy of eluents, but overall recovery is the critical measure of the usefulness of a procedure.Recoveries for the 500 mg spike were 92% for Ru, and better than 97% for all other elements. Slightly lower recoveries ( Å 85%) were measured for Ir and Rh at the 1 mg level. High recoveries of gold were caused by carry-over effects during analysis; these were minimised in later experiments by increasing washout times and uptake rates during the wash period. To optimise our procedure, various parameters were studied including the: effect of temperature on elution efficiency; volume of eluent required; possibility of combining thiourea and 12 mol l21 HCl into a single step.The volume of thiourea required to elute the PGEs and Au is dependent on the rate of formation of thiourea complexes. These are easily identified in concentrated ( > 100 mg ml21) solutions because they form strong colours. Experiments showed that the rate of formation of the thiourea complex varied between elements; the colour of solutions began to change almost immediately upon the addition of individual PGE and Au (each treated separately) to a 0.3 mol l21 thiourea in 0.1 mol l21 HCl.Colours continued to change (orange Ru solutions changed to green and then blue) when left standing for 1 h. Some elements precipitated when left to stand for several hours, not surprisingly, since thiourea has been used 30 to quantitatively precipitate some of the PGE and gold, and facilitate their separation. However, the formation of insoluble thiourea complexes was considered to be unlikely during ion-exchange, because the level of PGE and Au in solution would be very low, and precipitation only occurs from concentrated solution.Furthermore, fresh thiourea is continually added to the column, so the equilibrium for formation of these complexes is always shifted towards dissociation. Our experiments indicated that the recovery of the PGE and Au from a Dowex 1-X8 column is governed by both kinetic and thermodynamic factors.These were evaluated further by increasing the elution temperature using a column with a jacket through which heated water was passed. A solution of 1 mol l21 HCl containing 500 mg of each of the PGEs and Au was loaded onto the column at room temperature, and the eluate analysed to determine the concentration of each element remaining on the column. The column temperature was raised to 50 °C and a solution of 0.3 mol l21 thiourea in 0.1 mol l21 HCl, also heated to 50 °C, was eluted through the column (the eluate was collected as three 25 ml fractions), followed by the usual concentrated HCl elution step.A control experiment was run at room temperature ( Å 20 °C). The only improvement observed for elution at 50 °C was the complete recovery of Rh in the thiourea fraction (Fig. 2). The recovery of Ir in the thiourea fraction did not increase, so a 12 mol l21 HCl elution step was still required; indeed, elution of Ir with concentrated HCl seemed to be hindered by previous elution with heated thiourea (Fig. 2). Table 5 Breakthrough (%) of PGE and Au on Dowex 1-X8 anion-exchange resin after evaporated with HNO3, HF and HCl Mass of each PGE and Au Element 10 mg 1 mg 0.1 mg Ru 4.8 9.0 9.5 Rh 27 40 36 Pd 1.8 6.0 6.0 Ir < 0.1 < 0.1 < 0.1 Pt < 0.1 2.3 < 0.1 Au < 0.1 0.4 0.5 Fig. 1 Elution profiles (cumulative % recovery) for 500 mg (squares) and 1 mg (circles) spikes of the PGEs and Au from a 10 cm long, 1 cm diameter Dowex 1-X8 column using a two-step elution procedure. 22 Analyst, January 1997, Vol. 122A manually operated two-step elution is more cumbersome and requires more operator attention than a single-step procedure. Attempts were made to find experimental conditions that would enable the PGE and Au to be eluted using a single solvent. When 0.3 mol l21 thiourea in 12 mol l21 HCl was used as an eluent (Fig. 2), total recovery from 125 ml of eluent was generally lower than that achieved by the two-step method.Consequently, a two-step elution operated at room temperature was judged to yield the best results. Ion-exchange column dimensions The 1 cm diameter and 10 cm long Dowex 1-X8 resin bed employed in our initial studies, has been used by previous workers.7 However, the high capacity of the resin demonstrated by minimum breakthrough of PGE and Au at the 500 mg level, suggested that it might be possible to reduce column length while retaining good absorption on the resin.The recovery of the PGE and Au from a shorter, 1 cm diameter, 5 cm long column was examined. The volume of reagents used was varied to establish the effect of each reagent on the overall recovery of each element. Two experiments were undertaken: (a) 50 ml of 0.3 mol l21 thiourea in 0.1 mol l21 HCl, then 150 ml of 12 mol l21 HCl; (b) 75 ml of 0.3 mol l21 thiourea in 0.1 mol l21 HCl, followed by 75 ml of 12 mol l21 HCl. The eluate was collected in 25 or 50 ml fractions and analysed by ICP-AES. Ruthenium, Rh, Pd, Pt and Au were completely recovered by both procedures (Fig. 3). The recovery of Ir, which is primarily eluted in the 12 mol l21 HCl fraction, is highly dependent on eluate volume, 73% was recovered in 75 ml, increasing to 89% in 150 ml, so the larger volume of concentrated HCl was used in all further experiments. Results for the shorter column of Dowex 1-X8 showed that recoveries were not improved by reducing the length of the resin bed, while breakthrough experiments demonstrated that adsorption efficiency decreased (around 8% Ru and 18% Rh were not retained), so a 10 cm long column was confirmed as being optimum.Cleaning Dowex 1-X8 resin When using ion-exchange resins, procedures are needed to clean new, and if possible regenerate previously used, resin. A common approach is to clean the resin with reagents used for the elution step (in this case, thiourea and 12 mol l21 HCl). This approach is commonly used in chromatographic techniques to ensure that no additional contamination is obtained from the resin when the solvent is changed.The percentage breakthrough of the PGE and Au was compared after the resin had been cleaned with: (a) 0.3 mol l21 thiourea in 0.1 mol l21 HCl; and (b) 6 mol l21 HCl. Breakthrough was assessed as previously (called here a single pass), as well as for a double pass of the sample solution. For the double pass, the eluate containing the PGE and Au from a single pass through the column was collected and passed through the column again.The second eluate was then analysed (the same final volume was eluted for both experiments). Comparing the levels of breakthrough for single passes (Table 6), shows that there was significantly increased loss of Rh (from 5 to 17%) from resin pre-treated with thiourea, compared to that washed only with HCl. The loss of Ru was Fig. 2 Elution profiles (cumulative % recovery) for 500 mg spikes of the PGE and Au from a 10 cm long, 1 cm diameter Dowex 1-X8 column using different elution conditions.Fig. 3 Elution profiles (cumulative % recovery) for 500 mg spikes of the PGE and Au from a short (5 cm long, 1 cm diameter) Dowex 1-X8 column using : (a) 50 ml of 0.3 mol l21 thiourea in 0.1 mol l21 HCl, then 150 ml of 12 mol l21 HCl (circles); (b) 75 ml of 0.3 mol l21 thiourea in 0.1 mol l21 HCl, followed by 75 ml of 12 mol l21 HCl (squares). Table 6 Breakthrough (%) of 500 mg PGE and Au for Dowex 1-X8 resin cleaned with thiourea solution or HCl Thiourea HCl Element Single pass Double pass Single pass Ru 6.6 13 7.3 Rh 17 26 4.6 Pd 0.4 0.2 < 0.1 Ir < 0.1 0.4 0.2 Pt < 0.1 0.2 < 0.1 Au < 0.1 < 0.1 < 0.1 Analyst, January 1997, Vol. 122 23approximately the same in both cases ( Å 7%), while there was no significant breakthrough of Pd, Ir, Pt and Au. The level of breakthrough increased significantly with a double pass on the thiourea-cleaned column. This may have been caused by the initial eluate removing residual thiourea trapped on the resin, which then eluted additional Ru, Rh, and some Pt on the second pass.Breakthrough of small amounts of Ir cannot be attributed to this mechanism, but might have been caused by a change in oxidation state from Ir4+ to Ir3+ during the procedure. It was observed that as the first eluate was passed through the column a second time, the coloured band of PGE and Au moved rapidly down the column.This means that breakthrough is very sensitive to small changes in elution volume. Clearly, only resin which had not been in contact with thiourea provides a reliable anion-exchange medium, so reuse of resin cannot be recommended. This is not a serious limitation, since the cost of Dowex 1-X8, rather than analytical-grade (AG) resin, is not high. Blank levels of each PGE and Au were determined for Dowex 1-X8 resin after batch cleaning with 6 mol l21 HCl.The column was prepared as previously and eluted with three fractions of 0.3 mol l21 thiourea in 0.1 mol l21 HCl, followed by five 25 ml fractions of 12 mol l21 HCl. The solutions were analysed by ICP-MS and the total amount of each element was obtained by summing the fractions. Small amounts (ng) of Ru (18), Rh (8), Pd (40), Ir (14) and Pt (22) were found, which would only be significant if the method was applied to the determination of very low levels of precious metals.Concentrations of Au (490 ng) were higher, illustrating the need to carefully clean the resin prior to use. This level of Au was not significant for the 500 mg experiments, but the cleaning step was lengthened to include preconditioning of the column with 500 ml of 1 mol l21 HCl for later, low-level work. Determination of the PGE and Au at low levels Relatively high levels of the PGE and Au were generally used during the initial development of the ion-exchange method.This enabled ICP-AES to be used as the analytical finish, which is more precise and less prone to analytical problems caused by high and variable levels of TDS in solutions than ICP-MS. Experiments on solutions containing 1 mg spikes, however, necessitated diluting the thiourea fraction to reduce the level of TDS presented to the instrument. Diluted samples were below the lower limit of quantitation for ICP-AES, and close to those for ICP-MS, so the ion-exchange method developed so far would only be applicable to geological samples containing high ( > 1 mg g21) PGE and Au contents (calculated based on a 1 g sample size).There was a requirement, therefore, to concentrate the precious metals in the eluate (the volume eluted could not be reduced, since it was the minimum necessary to completely elute the PGE and Au), and/or eliminate the need for dilution to reduce the TDS and acid concentrations. The TDS content of the thiourea eluate could be reduced by decomposing the thiourea with nitric acid: H2NCSNH2 + HNO3 » NH3 + COx + SOx (1) where x = 1, 2, 3, or 4, as appropriate.The ammonia, carbon monoxide and carbon dioxide are lost by volatilisation. The main species remaining in solution after decomposition with HNO3 is sulfate (SO4 22). Unfortunately, sulfate solutions are not well suited to analysis by ICP-MS because the Ni sampling cone rapidly degrades, causing signal instability during analysis. Two possibilities were considered for the analysis of these solutions by ICP-MS: (1) use of a Pt-tipped sampling cone, which is not damaged by low levels of SO4 22; (2) removal of the sulfate before analysis.The first option was not feasible because Pt was one of the elements being sought, so precipitation of BaSO4 was investigated as a means of removing excess sulfate from solution. The solubility of BaSO4 is 2.2 mg ml21 in cold water.31 It was calculated that 75 ml of 0.3 mol l21 thiourea would produce 0.225 mol 121 SO4 22, assuming that there is complete conversion of S in thiourea to SO4 22.This estimate is almost certainly too high, since it takes no account of the loss of volatile lower-order sulfur–oxygen species during the decomposition step, so a maximum of 5.5 g of BaCl2·2H2O is needed to precipitate the sulfate. The following procedure was used to decrease the TDS of the thiourea fraction: (1) each 0.3 mol l21 thiourea solution was reduced to Å 10 ml in a 100 ml Pyrex beaker by evaporation at 95 °C on a sand bath, and allowed to cool; (2) fuming AnalaR HNO3 was added dropwise (2–3 ml) until effervescing ceased; (3) the resulting solution was evaporated at 95 °C to < 1 ml, to remove excess HNO3, and then diluted to 5–10 ml with deionised water; (4) AnalaR BaCl2 (5.5 g BaCl2·2H2O dissolved in Å 30 ml of H2O) was added, and the solution stirred to ensure complete precipitation; (5) the BaSO4 precipitate was removed by vacuum filtration through a Whatman (Whatman, Maidstone, Kent, UK) 0.45 mm cellulose nitrate filter membrane, using a large diameter (47 mm) filter funnel; (6) the filtered solution was evaporated on a sandbath at 95 °C to Å 5 ml, transferred into a 10 ml calibrated flask, and made up to volume with 0.5 mol l21 HCl.To evaluate the effectiveness of the HNO3–BaSO4 method, solutions of thiourea were spiked with known amounts of the PGE and Au, and treated as above. ICP-AES analyses of the treated solutions showed high levels of Ba (up to 1000 mg ml21) and S (approximately 400 mg ml21), indicating incomplete conversion of the sulfur in the thiourea to sulfate ions.Although the level of Ba and S was below the 2000 mg ml21 TDS limit imposed for solutions being analysed by ICP-MS, the presence of such high levels of individual elements caused signal suppression, necessitating the need for all thiourea fractions to be analysed by the standard additions method.This enabled more accurate analyses to be obtained, but quadrupled the number of solutions that had to be processed. Results for duplicate 1 mg experiments were in poor agreement (Table 7), making assessment of the results difficult. In general, PGE and Au in the spiked thiourea solutions showed moderate to good (70–100%) recoveries of Rh, Pd, Ir, Pt and Au at the 10 and 1 mg levels, but low values (50–70%) were obtained from the 0.1 mg spike, except for Au which was completely recovered in all three cases.Ruthenium recovery ranged from 70% for 10 mg, to 53% for the 0.1 mg solution. In a second experiment, 10, 1 and 0.1 mg PGE and Au solutions prepared in 1 mol l21 HCl were loaded onto Dowex 1-X8 columns, the initial eluate was collected to determine the breakthrough, and the precious metals were eluted with 75 ml of thiourea solution and 125 ml of 12 mol l21 HCl. The thiourea fraction was treated using the HNO3–BaSO4 method, while the concentrated HCl fraction was evaporated to incipient dryness and made up to 10 ml in 0.5 mol l21 HCl.The initial eluate, treated thiourea and HCl fractions were all analysed separately by ICP-MS. Recoveries were calculated by summing analyses Table 7 Recovery (%) of the PGE and Au from thiourea solutions. ICP-MS determinations after decomposition in HNO3 and removal of sulfate by precipitation of BaSO4. Averages and standard deviations for the 1 mg spike are based on determination of two solutions Mass in spike/mg Element 10 1 0.1 Ru 69 67 ± 7 53 Rh 85 67 ± 12 53 Pd 87 76 ± 22 57 Ir 99 96 ± 8 69 Pt 103 95 ± 13 72 Au 110 110 ± 29 100 24 Analyst, January 1997, Vol. 122of the thiourea and HCl fractions, and expressing results as a percentage of the amount in the original spike. The 1 mg experiment was performed in triplicate to establish the reproducibility of the method. The best results (Table 8) were obtained from the 1 mg spikes, with combined breakthrough and recovery of Ru, Rh, Pd and Pt generally totalling > 90%.However, the standard deviation of the measured recovery was relatively high, around 2–20%. The recovery of Au was 70% (with no breakthrough), while 67% Ir was eluted and 6% lost through breakthrough. The range of results obtained, however, included one run with 90% recovery of all PGE and Au. Lower recoveries were observed for all elements except Ru and Pt for the 10 mg solutions (Table 8). The 0.1 mg experiment yielded recoveries of > 100% for all elements, indicating contamination during handling of these solutions. The variable, and generally low, recovery of the PGE and Au from the thiourea solution following treatment with HNO3 and BaCl2 is attributed to coprecipitation and/or occlusion of the PGE and Au with the BaSO4 precipitate.Contamination was also encountered while determining the level of PGE and Au eluted from the Dowex 1-X8 resin in a blank run. Blank levels of the PGE and Au obtained for two columns are given in Table 9.Method blanks were also obtained by processing a 1 mol l21 HCl solution in an identical method to a geological sample (incorporating a digestion step using the method of Totland et al.,5 and anion-exchange), and analysing the eluent after decomposing the thiourea; again, two sets of results are presented because of the large difference obtained. Runs with high blank levels of the PGE and Au, were generally caused by high concentrations in the thiourea fraction.However, analysis of 75 ml of thiourea solution processed using the HNO3–BaSO4 method (Table 9), indicated low PGE and Au concentrations in the reagents. The highly variable blank levels, therefore, were not caused by contamination or interferences arising from the reagents used, but were probably a result of the extensive handling of solutions required in the procedure. This makes subtraction of a true blank difficult. Geological Reference Materials Although developed using synthetic solutions of the PGE and Au, the low levels of breakthrough and good recovery of several elements, indicated that the method should be applicable to the separation and determination of these elements in geological materials.To assess this, a study was undertaken using geological reference materials. Nickel copper matte PTM-1 (CCRMP, Canadian Certified Reference Materials Project, Energy Mines and Resources, Ottawa, Canada) contains relatively high levels of the PGE and Au, ranging from 0.34 mg g21 Ir to 5.8 mg g21 Pt, so this material was chosen to evaluate the basic anion-exchange procedure described above.In this case, the thiourea and 12 mol l21 HCl fractions could simply be diluted prior to analysis by ICP-MS. Three reference materials were used to evaluate the procedure employing the decomposition of thiourea: CCRMP materials PTM-1 and PTC-1 (sulfide flotation concentrate); Council for Mineral Technology (MINTEK, South African Bureau of Standards, Pretoria, South Africa) platinum ore, SARM7.Samples were prepared using a microwave aciddigestion procedure, described in detail elsewhere.5 Briefly, the method employs 1 g samples and acid digestion with 20 ml of aqua regia and 10 ml of 29 of mol l21 HF in Ultem-jacketed Teflon PFA sealed-vessels, heated at elevated pressure (200 psi; Å 1.4 MPa) in an MDS-2000 microwave oven (CEM Corporation, Matthews, NC, USA).Samples are subsequently evaporated to near-dryness, digested in 1 mol l21 HCl, filtered, and the insoluble residues fused with small amounts of 1 + 1 Na2O2 + Na2CO3 (silicate samples) or Na2O2 (sulfides), before being dissolved in 1 mol l21 HCl. Filtrate and dissolved residue solutions are combined to give 10–20 ml of 1 mol l21 HCl, which is suitable for loading directly onto the anion-exchange column. Data for PTM-1 obtained by direct analysis of the thiourea fraction following anion-exchange separation were (mg g21): Ru 0.3; Rh 1.3; Pd 9.8; Ir 0.3; Pt 6.4; Au 2.6. When compared to results (Table 10) obtained following acid digestion and Table 9 Blank values (ng) obtained from Dowex 1-X8 columns following digestion of thiourea and preconcentration of HCl eluents prior to analysis by ICP-MS.Method blank includes a microwave digestion procedure.5 Values for a decomposed thiourea blank are included for comparison Column blank Method blank Element A B A B Thiourea blank Ru < 2 190 36 70 < 2 Rh < 0.3 150 < 0.3 39 < 0.3 Pd < 2 120 11 42 1.5 Ir 10 180 35 120 < 0.7 Pt 370 180 77 230 1.2 Au < 0.6 450 17 280 0.7 Table 8 Breakthrough and recovery (%) of the PGEs and Au after anionexchange separation.ICP-MS determinations following decomposition of thiourea and preconcentration of the HCl eluents. Averages and standard deviations for the 1 mg spike are based on three replicates Mass in spike/mg 10 1 0.1 Break- Re- Break- Re- Break- Re- Element through covery through covery through covery Ru 4.8 84 9.0 ± 2.3 79 ± 2 9.5 140 Rh 27 41 40 ± 15 49 ± 14 36 120 Pd < 0.1 65 2.3 ± 4.0 87 ± 11 < 0.1 180 Ir 1.8 50 6.0 ± 3.6 67 ± 22 6.0 200 Pt < 0.1 96 0.4 ± 0.2 97 ± 14 < 0.1 230 Au < 0.1 38 < 0.1 70 ± 18 < 0.1 107 Table 10 Results for geological reference materials (mg g21) obtained following acid digestion, alkali fusion and anion-exchange separation with decomposition of thiourea, compared to acid digestion and fusion only,5 and reference values Element Ion exchange Digestion Reference PTC-1— Ru 0.29 ± 0.11 0.50 ± 0.07 0.65 Rh 0.30 ± 0.15 0.480 ± 0.089 0.62 ± 0.70 Pd 2.3 ± 1.2 11.1 ± 1.2 12.7 ± 0.7 Ir 0.21 ± 0.02 0.11 ± 0.01 0.1 Pt 2.40 ± 0.09 1.70 ± 0.14 3.0 ± 0.2 Au 0.52 ± 0.12 0.38 ± 0.21 0.65 ± 0.10 PTM-1— Ru 0.36 ± 0.06 0.670 ± 0.029 0.5 Rh 0.33 ± 0.06 0.940 ± 0.025 0.9 ± 0.2 Pd 6.1 ± 0.6 7.60 ± 0.12 8.1 ± 0.7 Ir 0.38 ± 0.22 0.35 ± 0.04 0.3 Pt 4.5 ± 0.5 4.90 ± 0.08 5.8 ± 0.4 Au 0.90 ± 0.02 1.500 ± 0.045 1.8 ± 0.2 SARM7— Ru 0.19 ± 0.11 0.360 ± 0.027 0.430 ± 0.057 Rh 0.049 ± 0.003 0.230 ± 0.007 0.240 ± 0.013 Pd 1.30 ± 0.14 1.230 ± 0.095 1.530 ± 0.032 Ir 0.11 ± 0.03 0.110 ± 0.016 0.074 ± 0.012 Pt 2.90 ± 0.37 3.40 ± 0.30 3.740 ± 0.045 Au 0.170 ± 0.013 0.290 ± 0.094 0.310 ± 0.015 Analyst, January 1997, Vol. 122 25fusion of the insoluble residue without an ion-exchange step,5 and with reference values, these data demonstrate acceptable, if marginally high, recovery of Rh, Pd, Ir and Pt.The value for Ru is low, but the level of Ru in the thiourea solution was close to the limit of detection for the ICP-MS, making the assessment inconclusive. Gold yielded a high value, suggesting a continuing contamination problem. Results for three preparations of PTM-1, and duplicate preparations of PTC-1 and SARM7 (Table 10), obtained following anion-exchange with decomposition of thiourea show, with a few exceptions, low recoveries of Ru, Rh, Pd, Pt and Au when compared to digestion only and reference data.These elements are eluted in the thiourea fraction, and it is believed that the treatment used to reduce the TDS was the cause of the poor recovery due, at least in part, to occlusion of a portion of the PGE and Au in the BaSO4 precipitate. This conclusion is supported by the complete recovery of most elements in PTM-1 when the thiourea fraction was analysed directly.Furthermore, Ir data (Table 10) are in good agreement with reference values. Iridium is eluted entirely with the 12 mol l21 HCl fraction, producing a simple matrix that poses no analytical difficulties by ICP-MS. Conclusions Our experiments demonstrate that the PGE and Au may be quantitatively adsorbed onto Dowex 1-X8 anion-exchange resin, and eluted using a two-stage procedure: thiourea to elute most Ru, Rh, Pd, Pt, Au; concentrated HCl to complete elution of these elements, and to elute all Ir.Evaluation of the procedure using geological reference materials showed encouraging results. In particular, the method has been successfully applied to the separation and determination of Ir in three rock reference materials by ICP-MS. The application of our method to the entire group of PGE and Au is limited principally by difficulties associated with analysis of the thiourea fraction. The extra dilution required for direct analysis of this eluate by ICP-MS, leads to limits of quantitation in samples26 of around 1 mg g21 for Ru, Rh, Pd, Pt and Au, which are similar to those achievable5 without separation from matrix elements.Reduction of the TDS in the first eluate was undertaken by decomposing thiourea with fuming HNO3, causing the loss by volatilisation of NH3 and CO2. However, high concentrations of sulfate ions remaining in solutions prevented their analysis by ICP-MS, because of the risk of corroding the Ni sampling cone.Removal of sulfate by precipitation with Ba was of limited success, producing erratic and generally low values for elements eluted in this fraction. It is concluded that precipitation is unsuitable for the analytical method, because the potential for coprecipitation and/or occlusion of the PGE and Au is too high and unpredictable. Although the use of isotope dilution methods could be used to compensate for low recoveries of Ru, Pd, Ir and Pt,32 extensive handing required at this stage led to sporadic contamination and difficulties in producing reliable procedural blanks, which is less easily addressed.To apply our method to the separation and determination of low levels of the PGE and Au, an alternative method for analysing the thiourea fraction is required. Potential ways to achieve this include electrothermal vaporisation or flow injection ICP-MS. These techniques may be used to directly analyse solutions with high levels of TDS, but their development is non-trivial and is beyond the scope of this study.Funding by RTZ Mining and Exploration Ltd. and enthusiastic support from Drs. C. Carlon and N. Badham (RTZ) are gratefully acknowledged. The operation of the ICP-MS laboratory as an analytical facility, located at Imperial College Centre for Analytical Research in the Environment, is supported by the UK Natural Environment Research Council (NERC). References 1 Hall, G. E. M., and Bonham-Carter, G. F., J. Geochem.Explor., 1988, 30, 255. 2 Van Loon, J. C., and Barefoot, R. R., Analytical Methods for Geochemical Exploration, Academic Press, San Diego, CA, 1989. 3 Van Loon, J. C., and Barefoot, R. R., Determination of the Precious Metals—Selected Instrumental Methods, Wiley, Chichester, 1991. 4 Colodner, D. C., Boyle, E. A. Edmond, J. M., and Thomson, J., Nature, 1992, 358, 402. 5 Totland, M. M., Jarvis, I., and Jarvis, K. E., Chem. Geol., 1995, 124, 21. 6 Ali-Bazi, S. J., and Chow, A., Talanta, 1984, 31, 815. 7 Korkisch, J., Handbook of Ion Exchange Resins: Their Application in Inorganic Analytical Chemistry, CRC Press, Boca Raton, FL, 1989, vol. 3. 8 Marhol, M., in Comprehensive Analytical Chemistry, ed. Svehla, G., Wilson and Wilson’s, Prague, 1982, vol. XIV, p. 580. 9 Korkisch, J., and Klakl, H., Talanta, 1968, 15, 339. 10 Totland, M. M., Jarvis, I., and Jarvis, K. E., Chem. Geol., 1993, 104, 175. 11 Boumans, P. W. J. M., Line Coincidence Tables for Inductively Coupled Plasma Atomic Emission Spectrometry, 2nd edn., Pergamon, Oxford, 1984, vol. 2. 12 Boumans, P. W. J. M., Line Coincidence Tables for Inductively Coupled Plasma Atomic Emission Spectrometry, 2nd edn., Pergamon, Oxford, 1984, vol. 1. 13 Winge, R. K., Fassel, V. A., Paterson, V. J., and Floyd, M. A., Inductively Coupled Plasma-Atomic Emission Spectroscopy—An Atlas of Spectral Information, Elsevier, Amsterdam, 1985. 14 Totland, M. M., PhD Thesis, Kingston University, Kingston upon Thames, 1993. 15 Jarvis, K. E., Gray, A. L., and Houk, R. S., Handbook of Inductively Coupled Plasma Mass Spectrometry, Blackie, Glasgow, 1992. 16 Jarvis, I., and Jarvis, K. E., Chem. Geol., 1992, 95, 1. 17 Branch, C. H., and Hutchison, D., J. Anal. At. Spectrom., 1986, 1, 433. 18 Busch, D. D., Prospero, J. M., and Naumann, R. A., Anal. Chem., 31, 884. 19 De Laeter, J. R., and Mermelengas, N., Geostand. Newsl., 1978, 2, 9. 20 Hodge, V., Stallard, M., Koide, M., and Goldberg, E.D., Anal. Chem., 1986, 58, 616. 21 Kraus, K. A., Nelson, F., and Smith, G. W., J. Phys. Chem., 1954, 58, 11. 22 Morgan, J. W., Anal. Chim. Acta, 1965, 32, 8. 23 Petrie, R. K., and Morgan, J. W., J. Radioanal. Chem., 1982, 74, 15. 24 Totland, M. M., Jarvis, I., and Jarvis, K. E., Chem. Geol., 1992, 95, 35. 25 Chao, T. T., and Sanzolone, R. F., J. Geochem. Explor., 1992, 44, 65. 26 Jarvis, I., in Handbook of Inductively Coupled Plasma Mass Spectrometry, ed. Jarvis, K. E., Gray, A.L., and Houk, R. S., Blackie, Glasgow, 1992, pp. 172–224. 27 Potts, P. J., A Handbook of Silicate Rock Analysis, Blackie, London, 1987. 28 Potts, P. J., in Analysis of Geological Materials, ed. Riddle, C., Marcel Dekker, New York, 1993, pp. 123–220. 29 Sulcek, Z., and Povondra, P., Methods of Decomposition in Inorganic Analysis, CRC Press, Boca Raton, FL, 1989. 30 Singh, S., Mathur, S. P., Thakur, R. S., and Lal, K., Orient. J. Chem., 1987, 3, 203. 31 CRC Handbook of Chemistry and Physics, ed.Weast, R. C., Astle, M. J., and Beyer, W. H., CRC Press, Boca Raton, FL, 68th edn., 1987. 32 Enzweiler, J., Potts, P. J., and Jarvis, K. E., Analyst., 1995, 120, 1391. Paper 6/06169I Received September 9, 1996 Accepted November 1, 1996 26 Analyst, January 1997, Vol. 122 Assessment of Dowex 1-X8-based Anion-exchange Procedures for the Separation and Determination of Ruthenium, Rhodium, Palladium, Iridium, Platinum and Gold in Geological Samples by Inductively Coupled Plasma Mass Spectrometry Ian Jarvis*a, Marina M.Totland†a and Kym E. Jarvisb a School of Geological Sciences, Kingston University, Penrhyn Road, Kingston upon Thames, Surrey, UK KT1 2EE b NERC ICP-MS Facility, Centre for Analytical Research in the Environment, Imperial College at Silwood Park, Buckhurst Road, Ascot, Berkshire, UK SL5 7TE Synthetic multielement solutions of the platinum group metals (PGE: Ru; Rh; Pd; Ir; Pt) and gold, with analysis by ICP-AES and ICP-MS, have been used to study the behaviour of the precious metals on Dowex 1-X8 resin.Simple solutions of precious-metal chlorocomplexes showed near-complete adsorption ( > 99%) of most elements, and only minor breakthrough of Ru and Ru ( Å 5%). Solutions pre-treated with acid mixtures typically used to decompose geological samples, demonstrated that perchloric acid adversely affects the adsorption of the PGEs on the resin. Solutions treated with HF–HNO3–HCl maintained good retention of Ir, Pt, Au ( > 99%), Pd ( > 94%) and Ru ( > 90%), but displayed significant loss (up to 40%) of Rh.A two-step procedure was necessary to elute the precious metals from the resin: 0.3 mol l21 thiourea prepared in 0.1 mol21 HCl removed Ru, Pd, Pt, Au, and some Rh: 12 mol l21 HCl eluted remaining Rh and all Ir. Recoveries ranged from 50 to 100%. At low levels, the determination of PGE and Au in the thiourea fraction by ICP-MS was compromised by high levels of total dissolved solids (TDS), which necessitated dilution of the eluate prior to analysis.The TDS was reduced by decomposing thiourea with HNO3 and removing SO4 22 by precipitation of BaSO4, but this led to lower and more erratic results, and increased contamination. An assessment of the optimised procedure employing geological reference materials PTM-1, PTC-1 and SARM7, indicated that acceptable results should be attainable for ICP-MS determination of most elements in geological samples containing high concentrations ( > 1 mg g21) of the PGE, for which decomposition of thiourea is unneccessary.The addition of a decomposition step led to low recovery of all elements except Ir, which was present entirely in the HCl eluate. The method is viable for the determination of Ir in a range of geological materials, but modifications will be required if it is to be extended to the other precious metals. Keywords: Platinum group element; gold; cation exchange chromatography; geological sample; inductively coupled plasma mass spectrometry The commonest methods1–3 used for the determination of the platinum group elements (PGE: Ru; Rh; Pd; Os; Ir; Pt) and gold are based on fire assay, which involves fusing 10–100 g samples with large amounts of alkali flux and collecting the PGE into a nickel sulfide or, for Au, a lead button.The nickel sulfide button is usually dissolved in acid to remove the nickel and sulfur, leaving a residue of PGE sulfides which may be analysed directly by neutron activation analysis (NAA).More commonly, the precipitate is taken into solution for analysis by AAS, direct coupled plasma-atomic emission spectroscopy (DCP-AES), or ICP-AES. Although fire assay is used extensively, it is highly labour intensive, and one of its most serious limitations is a dependence of the quality of results on the experience of the analyst. A more robust method would clearly be advantageous.Fire assay is also not well suited to the analysis of small ( < 5 g) samples, which may be required for some geological studies, such as investigations of precious metal mobility in sediment cores.4 We have previously described5 a new combined microwave digestion–minifusion method with analysis by ICP-MS, that yields quantitative data for Ru, Rh, Pd, Ir, Pt and Au in mineralised samples, but is limited by modest lower limits of determination in samples of 0.2–1 mg g21 (ppm).The objective of the present study was to investigate anion-exchange separation procedures for the PGE and Au, that would allow the determination of these elements at the lower levels (1–10 ng g21; ppb) found in most geological materials. Similar techniques have been employed by others6–8 with varying success, to determine individual or small groups of precious metals in a variety of sample types. Dowex 1-X8 is a strong base anion-exchanger with a styrene divinyl benzene polymer skeleton to which tertiary ammonium groups have been bound.7 Dowex 1-X8 exhibits a high selectivity for the PGE and Au, and it has been shown7,9 that it will remove these elements quantitatively from HCl solutions.However, the resin has not been used previously to simultaneously separate and preconcentrate the entire group of PGE and Au from solutions of geological materials, prior to their analysis by ICP-MS. The aims of this study were to: (a) develop experimental conditions under which 100% of the PGE and Au present in solution would be absorbed onto the Dowex 1-X8 resin, with minimal adsorption of competing ions; (b) find a set of conditions to elute 100% of the adsorbed PGE and Au; (c) develop a methodology to quantify the PGE and Au at ng ml21 levels in the eluate using ICP-MS.The first objective imposes limitations on the solution chemistry of the digested geological materials that are to undergo separation, while the second is known7 to be difficult to achieve from strong base cationexchange resins.Osmium was not included in the study because its volatility precludes its determination in most solutions of geological materials prepared for other precious metal analysis. 2,3,10 Experimental Instrumentation A JY 70 Plus (Jobin-Yvon, Longjumeau, France) inductively coupled plasma atomic emission spectrometer, located at † Present address: Atomic Energy of Canada Ltd., Chalk River Laboratories, Chalk River, Ontario, Canada KOJ 1JO.Analyst, January 1997, Vol. 122 (19–26) 19Kingston University, was used to determine the PGE and Au at high levels ( > 50 ng ml21) during method development. A VG Elemental PlasmaQuad PQ2 Plus (Fisons Instruments, Winsford, Cheshire, UK) ICP-MS, formerly at Royal Holloway University of London, was employed for lower level ( < 50 ng ml21) work. Operating conditions for the two instruments are given in Tables 1 and 2. In both cases, an external drift monitor (generally, a calibration standard solution) was used to correct for signal changes with time.Spectral lines used for PGE and Au determination by ICPAES were selected experimentally. Tables of spectral data11–13 were used to identify the most sensitive wavelengths for each element. A 1 mol l21 ARISTAR HCl (Merck, Lutterworth, UK) blank solution, a 1 mg ml21 solution of the analyte in the same matrix, and a multi-element solution containing all of the precious metals at 1 mg ml21 (both prepared using aliquots of single-element 1000 mg ml21 precious metal standard solutions, supplied in 20% v/v HCl; Johnson Matthey, Royston, UK), were each scanned14 across a 0.12 nm window, centred on the spectral line selected.The emission lines chosen (Table 3) were the most sensitive lines having no interference from other elements in the group. As only simple solutions of the precious metals were to be analysed by ICP-AES, interferences from other elements were not considered.Subsequently, calibration was achieved using five synthetic multi-element PGE and Au standards, prepared using aliquots of the 1000 mg ml21 singleelement standard solutions. Solutions analysed were generally in 0.1–1 mol l21 ARISTAR HCl; matrix effects were minimised by matching the acidity and salt content of samples and standard solutions. Detection limits (Table 3) calculated as 3s standard deviations of eleven determinations of a 1 mol l21 ARISTAR HCl blank, demonstrate that ICP-AES is well suited to the quantification of PGE and Au at concentrations in excess of 50 ng ml21. The ICP-MS was optimised using 59Co and 238U to give maximum sensitivity whilst minimising interferences, particularly refractory oxide species.The PGE and Au lie in two distinct parts of the mass range, 96–110 and 184–198 u. Some isotopes have an isobaric overlap and these were not used for analysis. The mass spectrometer was scanned from 98–200 u, with masses between 112–179 u being skipped to maximise the integration time spent on PGE and Au isotopes.During data processing, the existence of polyatomic or oxide interferences was determined by comparing isotopic ratios for elements with their theoretical values; only isotopes free from all spectroscopic interferences (Table 3) were used for quantification. External calibration of the ICP-MS for PGE and Au determinations was accomplished using multielement standards prepared in dilute HCl from 1000 mg ml21 single-element standard solutions; in all cases, the acid concentration was matched to the samples.A single standard (ranging between 20 and 200 ng ml21, depending on the expected concentration in the samples) and a blank were used as calibration points. It was noted that some elements displayed a significant memory effect (Au and, to a lesser extent, Pd), and extended washout times were necessary to avoid sample carryover.For example, following a solution containing 200 ng ml21 of the PGE and Au, a 3–4 min washout (using maximum pumping speed) with 1 mol l21 HCl was required to minimise the background level of Au. To avoid cross-contamination between high- and lowconcentration solutions, the integrals for all elements were monitored throughout the analytical run, and wash times were extended after solutions containing high levels of the precious metals had been analysed. Blanks were measured to establish the background level of these elements throughout the run, and blank corrections were applied where necessary.The measured detection limits for all elements by ICP-MS (Table 3) were better than 0.22 ng ml21. It is worth noting that since this study was completed, a newer instrument fitted with an Enhanced Performance Interface (Fisons Instruments) has been installed, which yields detection limits for the PGE and Au between 0.006 and 0.05 ng ml21. However, these do not translate directly to improved lower limits of determination for samples, because the new instrument displays poorer tolerance of total dissolved solids (TDS), necessitating higher dilution factors for solutions.Nevertheless, with these exceptionally low detection limits, ICP-MS is ideally suited to the determination of low concentrations of PGEs and Au in simple solutions. ICPMS is also far less prone to interferences than most other instrumental techniques.15,16 However, limitations on the types of solution analysed are imposed by the level of TDS; upper limits of approximately 0.2% TDS may be aspirated into an ICP-MS instrument without causing significant instrumental drift and/or matrix effects, so solutions had to be diluted to an appropriate level prior to analysis.Anion-exchange Experiments The Dowex 1-X8 resin (Bio-Rad Laboratories, Hemel Hempstead, UK) used here was nominally in chloride form (i.e., with chlorine counter-ions) and graded from 200–400 mesh (74–37 mm).However, the resin supplied contained a much Table 1 ICP-AES operating parameters for PGE and Au determination Instrument Jobin-Yvon JY 70 Plus Plasma power 1000 W Reflected power < 5 W Coolant gas flow 12 l min21 Auxiliary gas flow 0 l min21 Sheath gas flow 0.2 l min21 Nebuliser gas flow 0.35 l min21 Nebuliser Meinhard TR-50-C1 concentric glass Sample uptake 1 ml min21 Spray chamber JY Scott-type, double pass Spectrometer Czerny–Turner monochromator Grating Holographic 3600 grooves mm21 Integration period 10 s Table 2 ICP-MS operating parameters for PGE and Au determination Instrument VG PlasmaQuad PQ2 Plus Plasma power 1300 W Reflected power < 1 W Coolant gas flow 14 l min21 Auxiliary gas flow 0.5 l min21 Nebuliser gas flow 0.75 l min21 Nebuliser de Galan high dissolved solids Sample uptake rate 0.5 ml min21 Spray chamber Surrey design, single-pass, water-cooled at 4 °C Scan regions 98–111 and 180–200 u Table 3 Detection limits for the PGE and Au by ICP-AES and ICP-MS.Values calculated as 3s standard deviations for 11 determinations of a 1 mol l21 ARISTAR HCl blank ICP-AES ICP-MS Wavelength/ Detection limit/ Mass Detection limit/ Element nm ng ml21 (u) ng ml21 Ru 245.66 14 101 0.22 Rh 233.48 11 103 0.03 Pd 340.46 13 105 0.17 Ir 224.27 9.0 193 0.07 Pt 214.42 29 195 0.11 Au 242.80 5.9 197 0.06 20 Analyst, January 1997, Vol. 122wider range of grain-sizes than the nominal fraction, including a significant amount of very fine particles, possibly produced during packaging and handling.These were removed by slurrying the resin with deionised water, allowing the resin to settle for a short time, and then decanting off the suspended fines with the supernatant. This was repeated at least three times to remove most of the fine material. The resin was batch-cleaned prior to use by slurrying it with 6 mol l21 AnalaR HCl (Merck), allowing it to stand for 10–20 min, and then decanting off the acid.This procedure was repeated at least three times, or until no further discolouration of the acid was observed. After pouring off the last portion of the cleaning acid, the resin was slurried with 1 mol l21 ARISTAR HCl, and stored ready for use. For each ion-exchange experiment, following loading on the column, the resin was conditioned by passing 100 ml (later increased to 500 ml) 1 mol l21 ARISTAR HCl through the column. Only new resin was used for this work.To evaluate possible anion-exchange procedures, two parameters were measured: (1) percentage of the PGE and Au not adsorbed onto the resin, termed ‘breakthrough’; (2) percentage of the adsorbed metals eluted from the column, termed ‘recovery’. Unless stated otherwise, Merck ARISTAR reagents and ultra-pure (better than 18 MW cm) deionised water were used throughout this study. Adsorption of the PGEs and Au on Dower 1-X8 resin The PGEs must be present in appropriate oxidation states to ensure strong adsorption onto Dowex 1-X8 resin,7 and the charge of the PGE-complex may change depending on the ligands attached to the metal.Preliminary evaluation of the resin was conducted using stable solutions of PGE anionic chlorocomplexes, supplied at 1000 mg ml21 and stored in 20% v/v HCl solutions (Johnson Matthey, Royston, UK). Breakthrough of each PGE and Au was determined using synthetic multielement solutions produced by diluting aliquots of 1000 mg ml21 standards with 1 mol l21 HCl to yield 500 mg and 1 mg spikes of each element, diluted to 10 ml.Column dimensions were chosen based on previous studies6–9,17–23 to hold a settled resin bed of 1 cm diameter, 10 cm long. The percentage breakthrough of each element was determined by analysing the 1 mol l21 HCl eluate (25 ml) collected as the spike solutions were loaded onto the column. Solutions generated during experiments with the 500 mg spike were sufficiently concentrated to enable analysis by ICP-AES.This high level was chosen to determine whether the capacity of the resin was likely to be exceeded in normal use. The proportion of each element not retained on the column was small: < 0.1% Pd, Pt, Au; 0.2% Ir; 4.6% Rh; 7.3% Ru. A level of precious metals much closer to that expected for geological materials (1 mg) was used to evaluate breakthrough at lower concentrations. In this instance, absolute breakthrough was expected to be very small, so ICP-MS was used as the analytical finish.Results showed no significant breakthrough of Pd, Pt, Au ( < 0.1%), Ir (0.1%) and Rh (0.8%), and < 10% breakthrough of Ru from the 1 mg spike. While the above results verified the adsorption characteristics of Dowex 1-X8 described in the literature, they represent a simplified situation. The chemical state of the PGE and Au following digestion of geological materials5,24 may not be identical to those found in standard solutions.The effects of digestion procedures on the chemical state of individual PGE and Au needed to be established experimentally. Knowledge of the exact chemical state of each element is not necessary, provided that the PGE and Au are completely adsorbed onto the resin. It is not possible to measure breakthrough of the PGE and Au directly with geological materials, because the high concentrations of matrix elements passing through the columns preclude the determination of low levels of precious metals in the eluate. To estimate possible breakthrough when real samples are separated by this method, synthetic solutions were treated in an identical manner to geological samples. In this way, the effects of various acids used in digestion procedures could be evaluated.An HF–HClO4-based acid attack is one of the most commonly used methods to digest geological samples.24–29 To assess the affects of these acids on column retention, a 1 mol l21 HCl solution containing 500 mg of PGE and Au was treated as follows: (1) 8 ml of 16 mol l21 HNO3, 4 ml of 29 mol l21 HF, 4 ml of 12 mol l21 HClO4 were added in a PTFE beaker, evaporated at 200 °C to incipient dryness, a further 2 ml of HClO4 (12 mol l21) were added, evaporated to near-dryness, and the final solution made to 10 ml with 1 mol l21 HCl; (2) 4 ml of 16 mol l21 HNO3, 4 ml of 29 mol l21 HF, 4 ml of 12 mol l21 HClO4 were added and evaporated to incipient dryness, 8 ml of 12 mol21 HCl were added and evaporated to neardryness (twice), and the final solution also made to 10 ml with 1 mol l21 HCl.Breakthrough from these solutions were measured by ICP-AES from 1 cm diameter, 10 cm long columns of Dowex 1-X8. Breakthrough increases significantly following acid pre-treatment (Table 4). Losses at this stage were reduced when the solution was evaporated twice with 12 mol l21 HCl, instead of HClO4. Nevertheless, breakthrough of Rh and Ir remained very high, 58 and 18%, respectively.A comparison made using different elution volumes of the HClO4- evaporated solution (Table 4), demonstrated that the amount of breakthrough increased dramatically when 65 ml of 1 mol l21 HCl was eluted compared to 25 ml; in the former case, no Rh was retained on the resin. The level of breakthrough and the dependence on elution volume are clearly unacceptable for this separation method, precluding the use of HClO4-based digests prior to cation-exchange chromatography.A series of 1 mol l21 HCl solutions spiked with PGE and Au were evaporated in PTFE beakers at 100 °C on a hotplate, in the presence of 10 ml of 29 mol l21 HF, 15 ml of 12 mol l21 HCl and 5 ml of 16 mol l21 HNO3. This acid mixture has been demonstrated5 to be effective for the digestion of preciousmetal- bearing geological samples. Three masses of PGE and Au spike (10, 1, 0.1 mg) showed greatly reduced losses following the application of this acid mixture (Table 5), and subsequent evaporation (twice) with 4 ml of 12 mol l21 HCl to ensure conversion of elements to chloride form.Final solutions were approximately 10 ml of 1 mol l21 HCl; 25 ml of 1 mol l21 HCl were eluted and analysed by ICP-MS. Breakthrough was considered to be acceptably low ( < 10%) for Ru, Pd, Ir, Pt and Au, but the loss of Rh was high (27–40%). Elution conditions Dowex 1-X8 resin has a high affinity for the PGE and Au in dilute HCl solutions, but increased acid concentrations favour their removal.7 Distribution coefficients fall significant with Table 4 Breakthrough (%) of 500 mg of PGE and Au on Dowex 1-X8 anionexchange resin following pretreatments with perchloric acid.HClO4 = evaporated with HNO3, HF, HClO4; HClO4 added and evaporated. HCl = evaporated with HNO3, HF, HClO4; HCl added and evaporated. Both final solutions in 1 mol l21 HCl HClO4 HCl No pretreatment Element 25 ml 65 ml 25 ml 25 ml Ru 12 18 6.1 7.3 Rh 80 100 58 4.6 Pd 0.4 1.0 < 0.1 < 0.1 Ir 9.4 31 18 0.2 Pt 47 53 2.0 < 0.1 Au 0.4 1.0 < 0.1 < 0.1 Analyst, January 1997, Vol. 122 21increasing acid molarity, although there remains significant adsorption of some PGE even in concentrated HCl. Exceptions are Ir3+ and Rh3+ which have distribution coefficients of < 1 in 12 mol l21 HCl, and Ru4+, which is slightly higher at 10.7 The inference from these observations is that concentrated HCl will elute only part of the Ru, Rh and Ir adsorbed on the column; increasing the concentration of the counter ion in the system will not be sufficient to elute the entire group of elements.Thiourea has been successfully used to elute Pd, Pt and Au from strong base anion-exchange resins;7 selected PGE and Au are either reduced and/or complexed by thiourea, and the resulting complexes have low affinity for the resin. Elution experiments were conducted using 500 mg of each PGE and Au loaded in 10 ml of 1 mol l21 HCl.These solutions are coloured, and it was possible to get a quick assessment of elution conditions by observing the migration of coloured bands on the columns. Using this approach, it became apparent that although a portion of the PGE and Au could be eluted with thiourea, some remained on the column. It was concluded that a two-step procedure was necessary to elute the PGE and Au from a Dowex 1-X8 column. Three 25 ml aliquots of 0.3 mol l21 thiourea (4.7 g AnalaR thiourea dissolved in 200 ml of deionised water, acidified to 0.1 mol l21 HCl using 1.7 ml of 12 mol l21 HCl; selected based on work by Korkisch7) were eluted through the Dowex 1-X8 columns used for the breakthrough study.Solutions from the 500 mg experiments were diluted four-fold prior to ICP-AES analysis; 1 mg solutions were diluted 40-fold and analysed by ICP-MS. Dilution was necessary to reduce the high levels of TDS from the thiourea, to levels acceptable for each technique. 16 Aliquots of 25 ml of 12 mol l21 HCl were then eluted through each column, collected, diluted 10-fold to reduce the acid concentration, and analysed by ICP-AES or ICP-MS. A total of 100 ml of 12 mol l21 HCl was collected for the 500 mg experiment, and 125 ml from the 1 mg spike. Elution profiles (Fig. 1) show that most Ru, Pd, Pt and Au are eluted by 75 ml of 0.3 mol l21 thiourea, while Rh is only partly eluted and Ir remains bound to the resin.Remaining Rh and Au are completely eluted from the column with 12 mol l21 HCl, and Ir is removed by this eluent. Elution profiles are instructive for determining the rate of elution of elements and the relative efficacy of eluents, but overall recovery is the critical measure of the usefulness of a procedure. Recoveries for the 500 mg spike were 92% for Ru, and better than 97% for all other elements. Slightly lower recoveries ( Å 85%) were measured for Ir and Rh at the 1 mg level.High recoveries of gold were caused by carry-over effects during analysis; these were minimised in later experiments by increasing washout times and uptake rates during the wash period. To optimise our procedure, various parameters were studied including the: effect of temperature on elution efficiency; volume of eluent required; possibility of combining thiourea and 12 mol l21 HCl into a single step. The volume of thiourea required to elute the PGEs and Au is dependent on the rate of formation of thiourea complexes. These are easily identified in concentrated ( > 100 mg ml21) solutions because they form strong colours.Experiments showed that the rate of formation of the thiourea complex varied between elements; the colour of solutions began to change almost immediately upon the addition of individual PGE and Au (each treated separately) to a 0.3 mol l21 thiourea in 0.1 mol l21 HCl. Colours continued to change (orange Ru solutions changed to green and then blue) when left standing for 1 h.Some elements precipitated when left to stand for several hours, not surprisingly, since thiourea has been used 30 to quantitatively precipitate some of the PGE and gold, and facilitate their separation. However, the formation of insoluble thiourea complexes was considered to be unlikely during ion-exchange, because the level of PGE and Au in solution would be very low, and precipitation only occurs from concentrated solution.Furthermore, fresh thiourea is continually added to the column, so the equilibrium for formation of these complexes is always shifted towards dissociation. Our experiments indicated that the recovery of the PGE and Au from a Dowex 1-X8 column is governed by both kinetic and thermodynamic factors. These were evaluated further by increasing the elution temperature using a column with a jacket through which heated water was passed. A solution of 1 mol l21 HCl containing 500 mg of each of the PGEs and Au was loaded onto the column at room temperature, and the eluate analysed to determine the concentration of each element remaining on the column.The column temperature was raised to 50 °C and a solution of 0.3 mol l21 thiourea in 0.1 mol l21 HCl, also heated to 50 °C, was eluted through the column (the eluate was collected as three 25 ml fractions), followed by the usual concentrated HCl elution step. A control experiment was run at room temperature ( Å 20 °C).The only improvement observed for elution at 50 °C was the complete recovery of Rh in the thiourea fraction (Fig. 2). The recovery of Ir in the thiourea fraction did not increase, so a 12 mol l21 HCl elution step was still required; indeed, elution of Ir with concentrated HCl seemed to be hindered by previous elution with heated thiourea (Fig. 2). Table 5 Breakthrough (%) of PGE and Au on Dowex 1-X8 anion-exchange resin after evaporated with HNO3, HF and HCl Mass of each PGE and Au Element 10 mg 1 mg 0.1 mg Ru 4.8 9.0 9.5 Rh 27 40 36 Pd 1.8 6.0 6.0 Ir < 0.1 < 0.1 < 0.1 Pt < 0.1 2.3 < 0.1 Au < 0.1 0.4 0.5 Fig. 1 Elution profiles (cumulative % recovery) for 500 mg (squares) and 1 mg (circles) spikes of the PGEs and Au from a 10 cm long, 1 cm diameter Dowex 1-X8 column using a two-step elution procedure. 22 Analyst, January 1997, Vol. 122A manually operated two-step elution is more cumbersome and requires more operator attention than a single-step procedure.Attempts were made to find experimental conditions that would enable the PGE and Au to be eluted using a single solvent. When 0.3 mol l21 thiourea in 12 mol l21 HCl was used as an eluent (Fig. 2), total recovery from 125 ml of eluent was generally lower than that achieved by the two-step method. Consequently, a two-step elution operated at room temperature was judged to yield the best results. Ion-exchange column dimensions The 1 cm diameter and 10 cm long Dowex 1-X8 resin bed employed in our initial studies, has been used by previous workers.7 However, the high capacity of the resin demonstrated by minimum breakthrough of PGE and Au at the 500 mg level, suggested that it might be possible to reduce column length while retaining good absorption on the resin.The recovery of the PGE and Au from a shorter, 1 cm diameter, 5 cm long column was examined. The volume of reagents used was varied to establish the effect of each reagent on the overall recovery of each element.Two experiments were undertaken: (a) 50 ml of 0.3 mol l21 thiourea in 0.1 mol l21 HCl, then 150 ml of 12 mol l21 HCl; (b) 75 ml of 0.3 mol l21 thiourea in 0.1 mol l21 HCl, followed by 75 ml of 12 mol l21 HCl. The eluate was collected in 25 or 50 ml fractions and analysed by ICP-AES. Ruthenium, Rh, Pd, Pt and Au were completely recovered by both procedures (Fig. 3). The recovery of Ir, which is primarily eluted in the 12 mol l21 HCl fraction, is highly dependent on eluate volume, 73% was recovered in 75 ml, increasing to 89% in 150 ml, so the larger volume of concentrated HCl was used in all further experiments.Results for the shorter column of Dowex 1-X8 showed that recoveries were not improved by reducing the length of the resin bed, while breakthrough experiments demonstrated that adsorption efficiency decreased (around 8% Ru and 18% Rh were not retained), so a 10 cm long column was confirmed as being optimum.Cleaning Dowex 1-X8 resin When using ion-exchange resins, procedures are needed to clean new, and if possible regenerate previously used, resin. A common approach is to clean the resin with reagents used for the elution step (in this case, thiourea and 12 mol l21 HCl). This approach is commonly used in chromatographic techniques to ensure that no additional contamination is obtained from the resin when the solvent is changed. The percentage breakthrough of the PGE and Au was compared after the resin had been cleaned with: (a) 0.3 mol l21 thiourea in 0.1 mol l21 HCl; and (b) 6 mol l21 HCl.Breakthrough was assessed as previously (called here a single pass), as well as for a double pass of the sample solution. For the double pass, the eluate containing the PGE and Au from a single pass through the column was collected and passed through the column again. The second eluate was then analysed (the same final volume was eluted for both experiments).Comparing the levels of breakthrough for single passes (Table 6), shows that there was significantly increased loss of Rh (from 5 to 17%) from resin pre-treated with thiourea, compared to that washed only with HCl. The loss of Ru was Fig. 2 Elution profiles (cumulative % recovery) for 500 mg spikes of the PGE and Au from a 10 cm long, 1 cm diameter Dowex 1-X8 column using different elution conditions. Fig. 3 Elution profiles (cumulative % recovery) for 500 mg spikes of the PGE and Au from a short (5 cm long, 1 cm diameter) Dowex 1-X8 column using : (a) 50 ml of 0.3 mol l21 thiourea in 0.1 mol l21 HCl, then 150 ml of 12 mol l21 HCl (circles); (b) 75 ml of 0.3 mol l21 thiourea in 0.1 mol l21 HCl, followed by 75 ml of 12 mol l21 HCl (squares).Table 6 Breakthrough (%) of 500 mg PGE and Au for Dowex 1-X8 resin cleaned with thiourea solution or HCl Thiourea HCl Element Single pass Double pass Single pass Ru 6.6 13 7.3 Rh 17 26 4.6 Pd 0.4 0.2 < 0.1 Ir < 0.1 0.4 0.2 Pt < 0.1 0.2 < 0.1 Au < 0.1 < 0.1 < 0.1 Analyst, January 1997, Vol. 122 23approximately the same in both cases ( Å 7%), while there was no significant breakthrough of Pd, Ir, Pt and Au. The level of breakthrough increased significantly with a double pass on the thiourea-cleaned column. This may have been caused by the initial eluate removing residual thiourea trapped on the resin, which then eluted additional Ru, Rh, and some Pt on the second pass.Breakthrough of small amounts of Ir cannot be attributed to this mechanism, but might have been caused by a change in oxidation state from Ir4+ to Ir3+ during the procedure. It was observed that as the first eluate was passed through the column a second time, the coloured band of PGE and Au moved rapidly down the column. This means that breakthrough is very sensitive to small changes in elution volume. Clearly, only resin which had not been in contact with thiourea provides a reliable anion-exchange medium, so reuse of resin cannot be recommended.This is not a serious limitation, since the cost of Dowex 1-X8, rather than analytical-grade (AG) resin, is not high. Blank levels of each PGE and Au were determined for Dowex 1-X8 resin after batch cleaning with 6 mol l21 HCl. The column was prepared as previously and eluted with three fractions of 0.3 mol l21 thiourea in 0.1 mol l21 HCl, followed by five 25 ml fractions of 12 mol l21 HCl.The solutions were analysed by ICP-MS and the total amount of each element was obtained by summing the fractions. Small amounts (ng) of Ru (18), Rh (8), Pd (40), Ir (14) and Pt (22) were found, which would only be significant if the method was applied to the determination of very low levels of precious metals. Concentrations of Au (490 ng) were higher, illustrating the need to carefully clean the resin prior to use. This level of Au was not significant for the 500 mg experiments, but the cleaning step was lengthened to include preconditioning of the column with 500 ml of 1 mol l21 HCl for later, low-level work.Determination of the PGE and Au at low levels Relatively high levels of the PGE and Au were generally used during the initial development of the ion-exchange method. This enabled ICP-AES to be used as the analytical finish, which is more precise and less prone to analytical problems caused by high and variable levels of TDS in solutions than ICP-MS.Experiments on solutions containing 1 mg spikes, however, necessitated diluting the thiourea fraction to reduce the level of TDS presented to the instrument. Diluted samples were below the lower limit of quantitation for ICP-AES, and close to those for ICP-MS, so the ion-exchange method developed so far would only be applicable to geological samples containing high ( > 1 mg g21) PGE and Au contents (calculated based on a 1 g sample size). There was a requirement, therefore, to concentrate the precious metals in the eluate (the volume eluted could not be reduced, since it was the minimum necessary to completely elute the PGE and Au), and/or eliminate the need for dilution to reduce the TDS and acid concentrations.The TDS content of the thiourea eluate could be reduced by decomposing the thiourea with nitric acid: H2NCSNH2 + HNO3 » NH3 + COx + SOx (1) where x = 1, 2, 3, or 4, as appropriate. The ammonia, carbon monoxide and carbon dioxide are lost by volatilisation.The main species remaining in solution after decomposition with HNO3 is sulfate (SO4 22). Unfortunately, sulfate solutions are not well suited to analysis by ICP-MS because the Ni sampling cone rapidly degrades, causing signal instability during analysis. Two possibilities were considered for the analysis of these solutions by ICP-MS: (1) use of a Pt-tipped sampling cone, which is not damaged by low levels of SO4 22; (2) removal of the sulfate before analysis.The first option was not feasible because Pt was one of the elements being sought, so precipitation of BaSO4 was investigated as a means of removing excess sulfate from solution. The solubility of BaSO4 is 2.2 mg ml21 in cold water.31 It was calculated that 75 ml of 0.3 mol l21 thiourea would produce 0.225 mol 121 SO4 22, assuming that there is complete conversion of S in thiourea to SO4 22. This estimate is almost certainly too high, since it takes no account of the loss of volatile lower-order sulfur–oxygen species during the decomposition step, so a maximum of 5.5 g of BaCl2·2H2O is needed to precipitate the sulfate.The following procedure was used to decrease the TDS of the thiourea fraction: (1) each 0.3 mol l21 thiourea solution was reduced to Å 10 ml in a 100 ml Pyrex beaker by evaporation at 95 °C on a sand bath, and allowed to cool; (2) fuming AnalaR HNO3 was added dropwise (2–3 ml) until effervescing ceased; (3) the resulting solution was evaporated at 95 °C to < 1 ml, to remove excess HNO3, and then diluted to 5–10 ml with deionised water; (4) AnalaR BaCl2 (5.5 g BaCl2·2H2O dissolved in Å 30 ml of H2O) was added, and the solution stirred to ensure complete precipitation; (5) the BaSO4 precipitate was removed by vacuum filtration through a Whatman (Whatman, Maidstone, Kent, UK) 0.45 mm cellulose nitrate filter membrane, using a large diameter (47 mm) filter funnel; (6) the filtered solution was evaporated on a sandbath at 95 °C to Å 5 ml, transferred into a 10 ml calibrated flask, and made up to volume with 0.5 mol l21 HCl.To evaluate the effectiveness of the HNO3–BaSO4 method, solutions of thiourea were spiked with known amounts of the PGE and Au, and treated as above. ICP-AES analyses of the treated solutions showed high levels of Ba (up to 1000 mg ml21) and S (approximately 400 mg ml21), indicating incomplete conversion of the sulfur in the thiourea to sulfate ions.Although the level of Ba and S was below the 2000 mg ml21 TDS limit imposed for solutions being analysed by ICP-MS, the presence of such high levels of individual elements caused signal suppression, necessitating the need for all thiourea fractions to be analysed by the standard additions method. This enabled more accurate analyses to be obtained, but quadrupled the number of solutions that had to be processed. Results for duplicate 1 mg experiments were in poor agreement (Table 7), making assessment of the results difficult.In general, PGE and Au in the spiked thiourea solutions showed moderate to good (70–100%) recoveries of Rh, Pd, Ir, Pt and Au at the 10 and 1 mg levels, but low values (50–70%) were obtained from the 0.1 mg spike, except for Au which was completely recovered in all three cases. Ruthenium recovery ranged from 70% for 10 mg, to 53% for the 0.1 mg solution. In a second experiment, 10, 1 and 0.1 mg PGE and Au solutions prepared in 1 mol l21 HCl were loaded onto Dowex 1-X8 columns, the initial eluate was collected to determine the breakthrough, and the precious metals were eluted with 75 ml of thiourea solution and 125 ml of 12 mol l21 HCl.The thiourea fraction was treated using the HNO3–BaSO4 method, while the concentrated HCl fraction was evaporated to incipient dryness and made up to 10 ml in 0.5 mol l21 HCl. The initial eluate, treated thiourea and HCl fractions were all analysed separately by ICP-MS.Recoveries were calculated by summing analyses Table 7 Recovery (%) of the PGE and Au from thiourea solutions. ICP-MS determinations after decomposition in HNO3 and removal of sulfate by precipitation of BaSO4. Averages and standard deviations for the 1 mg spike are based on determination of two solutions Mass in spike/mg Element 10 1 0.1 Ru 69 67 ± 7 53 Rh 85 67 ± 12 53 Pd 87 76 ± 22 57 Ir 99 96 ± 8 69 Pt 103 95 ± 13 72 Au 110 110 ± 29 100 24 Analyst, January 1997, Vol. 122of the thiourea and HCl fractions, and expressing results as a percentage of the amount in the original spike. The 1 mg experiment was performed in triplicate to establish the reproducibility of the method. The best results (Table 8) were obtained from the 1 mg spikes, with combined breakthrough and recovery of Ru, Rh, Pd and Pt generally totalling > 90%. However, the standard deviation of the measured recovery was relatively high, around 2–20%. The recovery of Au was 70% (with no breakthrough), while 67% Ir was eluted and 6% lost through breakthrough. The range of results obtained, however, included one run with 90% recovery of all PGE and Au.Lower recoveries were observed for all elements except Ru and Pt for the 10 mg solutions (Table 8). The 0.1 mg experiment yielded recoveries of > 100% for all elements, indicating contamination during handling of these solutions. The variable, and generally low, recovery of the PGE and Au from the thiourea solution following treatment with HNO3 and BaCl2 is attributed to coprecipitation and/or occlusion of the PGE and Au with the BaSO4 precipitate. Contamination was also encountered while determining the level of PGE and Au eluted from the Dowex 1-X8 resin in a blank run.Blank levels of the PGE and Au obtained for two columns are given in Table 9. Method blanks were also obtained by processing a 1 mol l21 HCl solution in an identical method to a geological sample (incorporating a digestion step using the method of Totland et al.,5 and anion-exchange), and analysing the eluent after decomposing the thiourea; again, two sets of results are presented because of the large difference obtained.Runs with high blank levels of the PGE and Au, were generally caused by high concentrations in the thiourea fraction. However, analysis of 75 ml of thiourea solution processed using the HNO3–BaSO4 method (Table 9), indicated low PGE and Au concentrations in the reagents.The highly variable blank levels, therefore, were not caused by contamination or interferences arising from the reagents used, but were probably a result of the extensive handling of solutions required in the procedure. This makes subtraction of a true blank difficult. Geological Reference Materials Although developed using synthetic solutions of the PGE and Au, the low levels of breakthrough and good recovery of several elements, indicated that the method should be applicable to the separation and determination of these elements in geological materials.To assess this, a study was undertaken using geological reference materials. Nickel copper matte PTM-1 (CCRMP, Canadian Certified Reference Materials Project, Energy Mines and Resources, Ottawa, Canada) contains relatively high levels of the PGE and Au, ranging from 0.34 mg g21 Ir to 5.8 mg g21 Pt, so this material was chosen to evaluate the basic anion-exchange procedure described above.In this case, the thiourea and 12 mol l21 HCl fractions could simply be diluted prior to analysis by ICP-MS. Three reference materials were used to evaluate the procedure employing the decomposition of thiourea: CCRMP materials PTM-1 and PTC-1 (sulfide flotation concentrate); Council for Mineral Technology (MINTEK, South African Bureau of Standards, Pretoria, South Africa) platinum ore, SARM7. Samples were prepared using a microwave aciddigestion procedure, described in detail elsewhere.5 Briefly, the method employs 1 g samples and acid digestion with 20 ml of aqua regia and 10 ml of 29 of mol l21 HF in Ultem-jacketed Teflon PFA sealed-vessels, heated at elevated pressure (200 psi; Å 1.4 MPa) in an MDS-2000 microwave oven (CEM Corporation, Matthews, NC, USA).Samples are subsequently evaporated to near-dryness, digested in 1 mol l21 HCl, filtered, and the insoluble residues fused with small amounts of 1 + 1 Na2O2 + Na2CO3 (silicate samples) or Na2O2 (sulfides), before being dissolved in 1 mol l21 HCl.Filtrate and dissolved residue solutions are combined to give 10–20 ml of 1 mol l21 HCl, which is suitable for loading directly onto the anion-exchange column. Data for PTM-1 obtained by direct analysis of the thiourea fraction following anion-exchange separation were (mg g21): Ru 0.3; Rh 1.3; Pd 9.8; Ir 0.3; Pt 6.4; Au 2.6. When compared to results (Table 10) obtained following acid digestion and Table 9 Blank values (ng) obtained from Dowex 1-X8 columns following digestion of thiourea and preconcentration of HCl eluents prior to analysis by ICP-MS.Method blank includes a microwave digestion procedure.5 Values for a decomposed thiourea blank are included for comparison Column blank Method blank Element A B A B Thiourea blank Ru < 2 190 36 70 < 2 Rh < 0.3 150 < 0.3 39 < 0.3 Pd < 2 120 11 42 1.5 Ir 10 180 35 120 < 0.7 Pt 370 180 77 230 1.2 Au < 0.6 450 17 280 0.7 Table 8 Breakthrough and recovery (%) of the PGEs and Au after anionexchange separation.ICP-MS determinations following decomposition of thiourea and preconcentration of the HCl eluents. Averages and standard deviations for the 1 mg spike are based on three replicates Mass in spike/mg 10 1 0.1 Break- Re- Break- Re- Break- Re- Element through covery through covery through covery Ru 4.8 84 9.0 ± 2.3 79 ± 2 9.5 140 Rh 27 41 40 ± 15 49 ± 14 36 120 Pd < 0.1 65 2.3 ± 4.0 87 ± 11 < 0.1 180 Ir 1.8 50 6.0 ± 3.6 67 ± 22 6.0 200 Pt < 0.1 96 0.4 ± 0.2 97 ± 14 < 0.1 230 Au < 0.1 38 < 0.1 70 ± 18 < 0.1 107 Table 10 Results for geological reference materials (mg g21) obtained following acid digestion, alkali fusion and anion-exchange separation with decomposition of thiourea, compared to acid digestion and fusion only,5 and reference values Element Ion exchange Digestion Reference PTC-1— Ru 0.29 ± 0.11 0.50 ± 0.07 0.65 Rh 0.30 ± 0.15 0.480 ± 0.089 0.62 ± 0.70 Pd 2.3 ± 1.2 11.1 ± 1.2 12.7 ± 0.7 Ir 0.21 ± 0.02 0.11 ± 0.01 0.1 Pt 2.40 ± 0.09 1.70 ± 0.14 3.0 ± 0.2 Au 0.52 ± 0.12 0.38 ± 0.21 0.65 ± 0.10 PTM-1— Ru 0.36 ± 0.06 0.670 ± 0.029 0.5 Rh 0.33 ± 0.06 0.940 ± 0.025 0.9 ± 0.2 Pd 6.1 ± 0.6 7.60 ± 0.12 8.1 ± 0.7 Ir 0.38 ± 0.22 0.35 ± 0.04 0.3 Pt 4.5 ± 0.5 4.90 ± 0.08 5.8 ± 0.4 Au 0.90 ± 0.02 1.500 ± 0.045 1.8 ± 0.2 SARM7— Ru 0.19 ± 0.11 0.360 ± 0.027 0.430 ± 0.057 Rh 0.049 ± 0.003 0.230 ± 0.007 0.240 ± 0.013 Pd 1.30 ± 0.14 1.230 ± 0.095 1.530 ± 0.032 Ir 0.11 ± 0.03 0.110 ± 0.016 0.074 ± 0.012 Pt 2.90 ± 0.37 3.40 ± 0.30 3.740 ± 0.045 Au 0.170 ± 0.013 0.290 ± 0.094 0.310 ± 0.015 Analyst, January 1997, Vol. 122 25fusion of the insoluble residue without an ion-exchange step,5 and with reference values, these data demonstrate acceptable, if marginally high, recovery of Rh, Pd, Ir and Pt. The value for Ru is low, but the level of Ru in the thiourea solution was close to the limit of detection for the ICP-MS, making the assessment inconclusive.Gold yielded a high value, suggesting a continuing contamination problem. Results for three preparations of PTM-1, and duplicate preparations of PTC-1 and SARM7 (Table 10), obtained following anion-exchange with decomposition of thiourea show, with a few exceptions, low recoveries of Ru, Rh, Pd, Pt and Au when compared to digestion only and reference data. These elements are eluted in the thiourea fraction, and it is believed that the treatment used to reduce the TDS was the cause of the poor recovery due, at least in part, to occlusion of a portion of the PGE and Au in the BaSO4 precipitate.This conclusion is supported by the complete recovery of most elements in PTM-1 when the thiourea fraction was analysed directly. Furthermore, Ir data (Table 10) are in good agreement with reference values. Iridium is eluted entirely with the 12 mol l21 HCl fraction, producing a simple matrix that poses no analytical difficulties by ICP-MS. Conclusions Our experiments demonstrate that the PGE and Au may be quantitatively adsorbed onto Dowex 1-X8 anion-exchange resin, and eluted using a two-stage procedure: thiourea to elute most Ru, Rh, Pd, Pt, Au; concentrated HCl to complete elution of these elements, and to elute all Ir.Evaluation of the procedure using geological reference materials showed encouraging results. In particular, the method has been successfully applied to the separation and determination of Ir in three rock reference materials by ICP-MS.The application of our method to the entire group of PGE and Au is limited principally by difficulties associated with analysis of the thiourea fraction. The extra dilution required for direct analysis of this eluate by ICP-MS, leads to limits of quantitation in samples26 of around 1 mg g21 for Ru, Rh, Pd, Pt and Au, which are similar to those achievable5 without separation from matrix elements.Reduction of the TDS in the first eluate was undertaken by decomposing thiourea with fuming HNO3, causing the loss by volatilisation of NH3 and CO2. However, high concentrations of sulfate ions remaining in solutions prevented their analysis by ICP-MS, because of the risk of corroding the Ni sampling cone. Removal of sulfate by precipitation with Ba was of limited success, producing erratic and generally low values for elements eluted in this fraction. It is concluded that precipitation is unsuitable for the analytical method, because the potential for coprecipitation and/or occlusion of the PGE and Au is too high and unpredictable. Although the use of isotope dilution methods could be used to compensate for low recoveries of Ru, Pd, Ir and Pt,32 extensive handing required at this stage led to sporadic contamination and difficulties in producing reliable procedural blanks, which is less easily addressed. To apply our method to the separation and determination of low levels of the PGE and Au, an alternative method for analysing the thiourea fraction is required. Potential ways to achieve this include electrothermal vaporisation or flow injection ICP-MS. These techniques may be used to directly analyse solutions with high levels of TDS, but their development is non-trivial and is beyond the scope of this study. Funding by RTZ Mining and Exploration Ltd. and enthusiastic support from Drs. C. Carlon and N. Badham (RTZ) are gratefully acknowledged. The operation of the ICP-MS laboratory as an analytical facility, located at Imperial College Centre for Analytical Research in the Environment, is supported by the UK Natural Environment Research Council (NERC). References 1 Hall, G. E. M., and Bonham-Carter, G. F., J. Geochem. Explor., 1988, 30, 255. 2 Van Loon, J. C., and Barefoot, R. R., Analytical Methods for Geochemical Exploration, Academic Press, San Diego, CA, 1989. 3 Van Loon, J. C., and Barefoot, R. R., Determination of the Precious Metals—Selected Instrumental Methods, Wiley, Chichester, 1991. 4 Colodner, D. C., Boyle, E. A. Edmond, J. M., and Thomson, J., Nature, 1992, 358, 402. 5 Totland, M. M., Jarvis, I., and Jarvis, K. E., Chem. Geol., 1995, 124, 21. 6 Ali-Bazi, S. J., and Chow, A., Talanta, 1984, 31, 815. 7 Korkisch, J., Handbook of Ion Exchange Resins: Their Application in Inorganic Analytical Chemistry, CRC Press, Boca Raton, FL, 1989, vol. 3. 8 Marhol, M., in Comprehensive Analytical Chemistry, ed. Svehla, G., Wilson and Wilson’s, Prague, 1982, vol. XIV, p. 580. 9 Korkisch, J., and Klakl, H., Talanta, 1968, 15, 339. 10 Totland, M. M., Jarvis, I., and Jarvis, K. E., Chem. Geol., 1993, 104, 175. 11 Boumans, P. W. J. M., Line Coincidence Tables for Inductively Coupled Plasma Atomic Emission Spectrometry, 2nd edn., Pergamon, Oxford, 1984, vol. 2. 12 Boumans, P. W. J. M., Line Coincidence Tables for Inductively Coupled Plasma Atomic Emission Spectrometry, 2nd edn., Pergamon, Oxford, 1984, vol. 1. 13 Winge, R. K., Fassel, V. A., Paterson, V. J., and Floyd, M. A., Inductively Coupled Plasma-Atomic Emission Spectroscopy—An Atlas of Spectral Information, Elsevier, Amsterdam, 1985. 14 Totland, M. M., PhD Thesis, Kingston University, Kingston upon Thames, 1993. 15 Jarvis, K. E., Gray, A. L., and Houk, R. S., Handbook of Inductively Coupled Plasma Mass Spectrometry, Blackie, Glasgow, 1992. 16 Jarvis, I., and Jarvis, K. E., Chem. Geol., 1992, 95, 1. 17 Branch, C. H., and Hutchison, D., J. Anal. At. Spectrom., 1986, 1, 433. 18 Busch, D. D., Prospero, J. M., and Naumann, R. A., Anal. Chem., 31, 884. 19 De Laeter, J. R., and Mermelengas, N., Geostand. Newsl., 1978, 2, 9. 20 Hodge, V., Stallard, M., Koide, M., and Goldberg, E. D., Anal. Chem., 1986, 58, 616. 21 Kraus, K. A., Nelson, F., and Smith, G. W., J. Phys. Chem., 1954, 58, 11. 22 Morgan, J. W., Anal. Chim. Acta, 1965, 32, 8. 23 Petrie, R. K., and Morgan, J. W., J. Radioanal. Chem., 1982, 74, 15. 24 Totland, M. M., Jarvis, I., and Jarvis, K. E., Chem. Geol., 1992, 95, 35. 25 Chao, T. T., and Sanzolone, R. F., J. Geochem. Explor., 1992, 44, 65. 26 Jarvis, I., in Handbook of Inductively Coupled Plasma Mass Spectrometry, ed. Jarvis, K. E., Gray, A. L., and Houk, R. S., Blackie, Glasgow, 1992, pp. 172–224. 27 Potts, P. J., A Handbook of Silicate Rock Analysis, Blackie, London, 1987. 28 Potts, P. J., in Analysis of Geological Materials, ed. Riddle, C., Marcel Dekker, New York, 1993, pp. 123–220. 29 Sulcek, Z., and Povondra, P., Methods of Decomposition in Inorganic Analysis, CRC Press, Boca Raton, FL, 1989. 30 Singh, S., Mathur, S. P., Thakur, R. S., and Lal, K., Orient. J. Chem., 1987, 3, 203. 31 CRC Handbook of Chemistry and Physics, ed. Weast, R. C., Astle, M. J., and Beyer, W. H., CRC Press, Boca Raton, FL, 68th edn., 1987. 32 Enzweiler, J., Potts, P. J., and Jarvis, K. E., Analyst., 1995, 120, 1391. Paper 6/06169I Received September 9, 1996 Accepted November 1, 1996 26 Analyst, January 1997, Vol. 122

ISSN:0003-2654    

DOI:10.1039/a606169i

出版商:RSC    

年代:1997

数据来源: RSC

摘要:

Micro-homogeneity of Candidate Reference Materials Characterized by Particle Size and Homogeneity Factor Determination Thomas-Michael Sonntag and Matthias Rossbach* Institute of Applied Physical Chemistry, Research Centre J�ulich, KFA, 52425 J�ulich, Germany The IAEA Analytical Quality Control Services (AQCS) have made available two single-cell algal materials, IAEA-392 and IAEA-393, and an urban dust, IAEA-396, to study their use for analytical sample sizes in the milligram range and below.Solid sampling Zeemans effect AAS was applied to the determination of trace elements on the bases of 1026–1023 g amounts of the selected materials. The comparability of the mean values and the reproducibility of successive measurements is being evaluated in order to compare relative homogeneity factors for many elements in the investigated materials. From the reported results it seems that the algal materials IAEA-392 and IAEA-393 are extremely homogeneous biological materials for a number of elements with an extraordinarily sharp particle size distribution below 1025 m.A similar situation seems to hold for the urban dust material IAEA-396, which had been air-jet milled to a particle size distribution of around 4 3 1026 m. The introduction of these materials as CRMs with very small amounts needed to determine the certified concentrations will help to meet the needs of micro-analytical techniques for natural matrix reference materials.Keywords: Homogeneity; biological reference materials; particle size distribution; solid sampling atomic absorption spectrometry Certified Reference Materials (CRMs) from various producers are commonly used in trace analysis for method development and quality control purposes. Analytical techniques are becoming more and more sensitive and require smaller amounts of sample. Particularly the so-called ‘micro-analytical techniques’ such as m-PIXE, laser ablation MS, solid sampling AAS and Xray fluorescence techniques are especially sensitive to sample homogeneity.Determinations rely in some cases almost entirely on the availability of CRMs of proven homogeneity in the range of milligrams or below. Some of the producers of CRMs state in their certificates, however, that certified values are valid only if more than 100 mg of the material is used for analysis. Besides the fact that precious material is wasted if a homogeneous material is consumed in such a large quantity for the determination of a single element in one analytical run, a more precise evaluation of the homogeneity with regard to the element (or compound) considered in a given material would be desirable.The precision of an analytical measurement is a direct function of the stability of the instrument’s response and the material’s homogeneity, i.e., detecting the same chemical composition in successive aliquots. With decreasing mass of the aliquot the statistical probability of finding the same distribution of particles with identical overall composition decreases.Assuming a certain degree of heterogeneity of natural matrix materials, the homogeneity of a reference material at a level of 1026–1023 g aliquots is a direct function of its particle size distribution.1 Trace elements and compounds tend to be unevenly distributed throughout a biological or other environmental material. In the course of reference material production, homogenization of such materials is a critical step and generally results in a finely ground powder.The experimental determination of elementspecific homogeneity factors is tedious and time consuming. In many cases it is assumed from an exemplary survey analysis of one or two elements (with sometimes inadequate methods and/ or irrelevant sample mass) that all the certified elements behave in the same way. In order to be on the safe side, CRM producers generally quote the homogeneity of their materials to be satisfactory only at a high to very high (100–500 mg) sample mass.2 Precisely determined sampling constants3 or homogeneity factors for individual elements, which enable analysts to trace the material-inherent variability of concentrations to the level of material consumed for a particular analytical technique, would help to increase considerably the precision of results from such techniques and help to establish a higher degree of reliability of the results.The accurate applicability for quantitative analysis of micro-analytical techniques in biomedical, forensic, archaeological and palaeontological investigations would be greatly enhanced. Downscaling of sample preparation techniques and of digestion instruments would be feasible. If smaller amounts of reference materials could be used for quality assurance in various applications, these valuable materials could be used more extensively and the production of new materials could be concentrated on a greater diversity of matrices rather than on replacing exhausted materials.Experimental Three batches of single-cell green algae (Chlorella sp.) grown in cultivation media with different concentration levels of a number of environmentally relevant elements were produced at the Czech University of Agriculture, Prague, in large amounts.4 After harvesting 30–50 kg of fresh material with elevated, environmental and low levels of trace elements, the materials were air sprayed for drying.Without further grinding, the materials with elevated levels (IAEA-393) and the material with environmental levels (IAEA-392) were distributed among the participating laboratories of the first IAEA-AQCS meeting on ‘Reference materials for micro-analytical nuclear techniques’ held in Zagreb, Croatia, in December 1994.5 The third material, single-cell algae with low levels of trace elements (IAEA-391), was distributed only in early 1996 and could therefore not be fully implemented in this intercomparison.A second candidate reference material, ‘urban dust’, was distributed at the same time in two versions: one batch had been sieved through a mesh size of 70 31026 m (IAEA-396A/S) and the other was air-jet milled material (IAEA-396A/M). Both materials originate from a collection of air particulate matter extracted from the air conditioning filters of a Vienna hospital and should reflect the average loading of air particulates from an industrialized urban environment. The material was collected in Analyst, January 1997, Vol. 122 (27–31) 27a large enough amount and is intended to be used as a future air filter reference material.6 Solid sampling Zeeman-effect atomic absorption spectrometry (SS-ZAAS) is a well established technique within the AAS family of techniques.7,8 It allows the introduction of small amounts (0.02–20 mg) of solid materials for electrothermal atomization in a graphite furnace and single-element determination. Zeeman splitting of the specific absorption lines is performed by a strong magnetic field applied at the emission lamp and the instrument (SM 30, Firma Gr�un Optic, Wetzlar, Germany) is automatically tuned for zero adjustment and optimum sensitivity.Calibration was carried out using appropriate reference materials (NIES No. 9, sargassum for algae, and NIST SRM 1648, urban particulate matter for the IAEA dust materials). As it turned out that the peak height of the absorption signals gave more consistent results than the integrated peak area, peak height was used throughout the evaluation of our experiments for quantification.Whether this is due to a specific software problem or is a special feature of the direct solid sampling techniques could not be investigated. The technique is rapid (no sample digestion), inexpensive and sensitive for a large number of elements.9 The laser particle analyser Analysette 22 (Fritsch, Idar- Oberstein, Germany) contains an He–Ne laser and a detector system with 31 units each containing 10 channels.Particles between 0.1 and 1250 3 1026 m can be determined in intervals of 250 3 1026 m with a resolution of 310 channels. About 100 mg of dry sample are dispersed in propan-2-ol and agitated in a ultrasonic bath for about 10 min. The suspension is then pumped through the meast cell of the instrument with a built-in ultrasonic agitator.The diffraction pattern of the scattered laser light is used to calculate the particle size and the mean distribution of particle sizes is recorded.10 Results Particle Size Distributions In Figs. 1–5, the determined particle size distributions of the five materials [algae elevated level (IAEA-393), algae environmental level (IAEA-392), algae low level (IAEA-391), urban dust sieved (IAEA-396A/S) and urban dust jet milled (IAEA-396A/M)] are shown. From Figs. 1–5, it is clear that all materials exhibit fairly sharp particle size distributions with 90% of the particles smaller than 7, 15, 4.5, 20 and 25 mm for the respective materials.This finding points to very homogeneous materials where small amounts for analysis could show good reproducibility of replicate measurements11 provided that no systematic errors hamper the quantification. The systematic error of the technique was tested in performing replicate measurements with liquid standard solutions.It could be shown that SS-ZAAS does not add more than 3% to the overall standard deviation and is therefore well suited for this kind of homogeneity study. Homogeneity Factor and Sampling Constants According to Ingamells and Swizer,12 a sampling constant can be given by Ks = R2m (1) where Ks = sampling constant, R = relative standard deviation and m = mean sample mass (mg). This constant was originally intended to be used in geological sampling of larger amounts than are generally used for analysis, notably in microanalysis. Fig. 1 Particle size distribution in IAEA-393, algae elevated level. Fig. 2 Particle size distribution in IAEA-392, algae environmental level. Fig. 3 Particle size distribution in IAEA-391, algae low level. Fig. 4 Particle size distribution in IAEA-396A/M, urban dust air-jet milled. 28 Analyst, January 1997, Vol. 122Kurf�urst et al.13 therefore took the square root of this factor to calculate the relative homogeneity factor, HE: HE = sHOMAmºº (2) where sHOM = relative standard deviation and m = mean sample mass (mg).The difference between the two factors can be illustrated graphically as shown in Figs. 6 and 7. Whereas Ingamell and Switzer’s factor ends up with very large numbers at moderate sample mass, Kurf�urst et al.’s factor is suitable particularly for the description of low sample masses but not so much for higher sample masses. For application of milligram sample inputs to the SS-ZAAS instrument we preferred Kurf�urst et al.’s approach to describe the relative homogeneity as HE in mg1 2.11,14 Homogeneity Testing The systematic error of the SS-ZAAS method was checked using liquid standard solutions for several elements.It was found that, related to the absolute mass of element introduced into the graphite furnace, the reproducibility of repetitive measurements varied only between 2 to 3%. The pipetting error can be estimated (by weighing) to be < 1%. Weighing of the solid materials using a Sartorius 4503 microbalance with a maximum loading of 4.1 g and a weighing error of d = ±0.001 mg assures a maximum weighing inaccuracy of < 1% at a sample mass of 0.1 mg.A few examples of the reproducibility of repetitive (20) measurements of the investigated materials using the SS-ZAAS approach are shown in Figs. 8–11. It can be seen that the reproducibility of repetitive measurements varies with the element analysed in a particular sample material and the standard deviation varies with the mass of the sample analysed Fig. 5 Particle size distribution in IAEA-396A/S, urban dust sieved. Fig. 6 Proportionality between sampling constant and sample mass at different fixed standard deviations. Fig. 7 Proportionality between the relative homogeneity factor HE and the sample mass at different fixed standard deviations. Fig. 8 Lead homogeneity in IAEA 393, algae elevated level, mean sample mass 0.193 mg. Fig. 9 Copper homogeneity in IAEA-392, algae environmental level, mean sample mass 0.458 mg.Fig. 10 Copper homogeneity in IAEA 396A/M, urban dust, mean sample mass 0.34 mg. Analyst, January 1997, Vol. 122 29(e.g., Cu in dust at a sample mass of 0.34 mg is reproducible with RSD = 5.5% and at a level of 0.08 mg with RSD = 10%). In order to achieve relative homogeneity factors related to the sample mass used for analysis, a number of measurement series all following the same routine (20 measurements, same calibration, same wavelength) were carried out and the results were plotted [RSD versus sample mass (mg)] as shown in Figs. 12 and 13. Fitting the parabolic curves yields a function of x depending on a factor times m21 2. The factor is the relative homogeneity factor of Kurf�urst et al. (HE) and it is related to this particular element’s homogeneity in the investigated material. In Table 1, the HE values for the four materials are given for all the elements investigated.The Cd concentration in IAEA-392 is very low and close to the detection limit of the method (0.015 ± 0.006 mg kg21) at a sample mass of 1.3 mg. For this reason, only one measurement series of 20 measurements could be carried out. Within these 20 measurements one outlier of about three times the mean of all other results was recorded. This value was not rejected and hence the RSD reached about 50%. Cadmium therefore cannot be considered to be homogeneously distributed in this particular material (IAEA-392, algae, environmental level).Lead concentrations could not be determined at all in this material at such a sample mass level. Increasing the sample mass by a factor of ten resulted in a more sensitive determination. Discussion and Conclusion It was shown that the particle size distributions of all five materials are exceptionally narrow and peak at very low particle sizes, compared with the normal range of other biological reference materials.From this finding, the assumption of good to very good homogeneity for trace elements can be drawn. Using the approach of Kurf�urst et al. a numerical description of element-specific homogeneity of the materials was experimentally elaborated. As Kurf�urst and co-workers pointed out,13–16 HE < 10 indicates very good homogeneity. All the materials investigated for the investigated elements show HE values in this range (with the exception of Cd in IAEA-392) and can be considered suitable as reference materials for micro-analytical techniques analysing samples with masses between 0.1 and 10 mg.For techniques using even smaller amounts of sample, natural matrix reference materials still have to be investigated. As the relative homogeneity factors HE are determined in an accurate way for individual elements in a given material, they can be used to assign the uncertainty of a certified concentration which is due only to material heterogeneity. Today the uncertainty given for certified reference materials is a composite of several uncertainties, such as (i) the systematic error of the analytical technique, (ii) bias from different standardizations of all the techniques used for certification and (iii) the materials’ inherent heterogeneity.By assigning a certified value with the homogeneity factor related to the sample mass consumed in analysis it is possible to calculate a mass-dependent uncertainty which derives strictly from the quality of the material itself.Hence systematic errors and bias of analytical techniques will be easier to recognize and the analytical process will be more transparent. The accurate determination of element-specific homogeneity factors for certified reference materials is essential for all CRMs to be used for quality control in micro-analytical trace element determinations. Suitable techniques such as INAA and SSZAAS with no sample digestion and accurate control over the total sample mass analysed are available and should be applied regularly in the course of certification of CRMs for the quantification of trace element distributions in natural matrix materials.The ‘true value’ of an element in a matrix is not a very useful parameter unless reliable information on the probability of obtaining exacue in repetitive measurements (repro- Fig. 11 Copper homogeneity IAEA 396, urban dust air-jet mulled, mean sample mass 0.08 mg. Fig. 12 RSD versus sample mass for Pb in IAEA 393, algae elevated level. The data points can be fitted as f(x) + 4.68 m21 2. Fig. 13 RSD versus sample mass for Cu in IAEA 393, algae elevated level. The data points can be fitted as f(x) + 8.67 m21 2. Table 1 Relative homogeneity factors, HE, for individual elements in the four materials investigated No. of Material measurements HE (Cu) HE (Cd) HE (Pb) IAEA-392 20 8.9 45.3 —* IAEA-393 80 8.7 4.0 4.7 IAEA-396A/M 20 3.0 4.6 6.8 IAEA-396A/S 20 8.9 8.4 6.7 * Pb in IAEA-392 was below the limit of determination. 30 Analyst, January 1997, Vol. 122ducibility) of the same material is attached to it. This probability is clearly dependent on the number of particles with potentially differing concentrations in an analytical aliquot and hence on the sample mass consumed for analysis. Hitherto certified values were given with a conservative estimate (amounting to 50% in some cases) of the overall uncertainty. The empirical statement ‘a minimum sample mass of 250 mg of the dried material .. . is necessary for any certified value . . . to be valid within the stated uncertainty’ is misleading and may only have a commercial background. Precisely determined individual homogeneity factors for each element would therefore help CRM users to (i) save precious material, (ii) determine the systematic error of the applied analytical technique and (iii) test the reliability of the sample preparation techniques more accurately and should therefore be endorsed also by CRM producers.We are grateful for and greatly appreciate the intensive work of D. Koglin in particle size analysis. M.R. acknowledges the financial support of the Bundesminister f�ur Umwelt, Naturschutz und Reaktorsicherheit, Bonn, and the Bundesumweltamt, Berlin. Many thanks are due to the IAEA (Dr. R. Zeisler, Dr. V. Valkovic) for letting us participate in this interesting AQCS programme. References 1 Rossbach, M., Ostapczuk, P., Schladot, J.D., and Emons, H., UWSFZ. Umweltchem. � Okotox., 1995, 7(6), 365. 2 Pauwels, J., and Vandecasteele, C., Fresenius’ J. Anal. Chem., 1993, 345, 121. 3 Chatt, A., Jayawickreme C. K., and McDowell, L. S., Fresenius’ J. Anal. Chem., 1990, 338, 399. 4 Mader, P., Stejskalova, I., and Slamova, A., Fresenius’ J. Anal. Chem., 1995, 352, 131. 5 Report of the Research Co-ordination Meeting on Reference Materials for Microanalytical Nuclear Techniques, IAEA/AL/083, IAEA, Vienna, 1994. 6 Zeisler, R., personal communication. 7 L�ucker, E., K�onig, H., Gabriel, W., and Rosopulo, A., Fresenius’ J. Anal. Chem., 1992, 342, 941. 8 Mohl, C., Grobecker, K. H., and Stoeppler, M., Fresenius’ J. Anal. Chem., 1987, 328, 413. 9 Kurf�urst, U., Fresenius’ Z. Anal. Chem., 1982, 313, 97. 10 Friedrich, H., and Mansour, A., Nachr. Chem. Tech. Lab., 1995, 43, 87. 11 Sonntag, Th.-M., Homogenit�atsstudien f�ur ausgew�ahlte Elemente in verschiedenen Materialien der Bank f�ur Umweltproben und der Internationalen Atomenergiebeh�orde mit der Solid-sampling-Zeeman- Atomabsorptionsspektrometrie, Diplomarbeit, Fachhochschule Aachen, Abteilung J�ulich, 1996. 12 Ingamells, C. O., and Swizer, P., Talanta, 1973, 20, 547. 13 Kurf�urst, U., Grobecker, K.-H., and Stoeppler, M., Trace Elem., 1984, 3, 591. 14 Stoeppler, M., Kurf�urst, U., and Grobecker, K.-H., Fresenius’ J. Anal. Chem., 1985, 322, 687. 15 Kurf�urst, U., Fresenius’ Z. Anal. Chem., 1983, 315, 304. 16 Kurf�urst, U., Fresenius’ Z. Anal. Chem., 1983, 316, 1. Paper 6/05396C Received August 1, 1996 Accepted October 21, 1996 Analyst, January 1997, Vol. 122 31 Micro-homogeneity of Candidate Reference Materials Characterized by Particle Size and Homogeneity Factor Determination Thomas-Michael Sonntag and Matthias Rossbach* Institute of Applied Physical Chemistry, Research Centre J�ulich, KFA, 52425 J�ulich, Germany The IAEA Analytical Quality Control Services (AQCS) have made available two single-cell algal materials, IAEA-392 and IAEA-393, and an urban dust, IAEA-396, to study their use for analytical sample sizes in the milligram range and below.Solid sampling Zeemans effect AAS was applied to the determination of trace elements on the bases of 1026–1023 g amounts of the selected materials. The comparability of the mean values and the reproducibility of successive measurements is being evaluated in order to compare relative homogeneity factors for many elements in the investigated materials.From the reported results it seems that the algal materials IAEA-392 and IAEA-393 are extremely homogeneous biological materials for a number of elements with an extraordinarily sharp particle size distribution below 1025 m. A similar situation seems to hold for the urban dust material IAEA-396, which had been air-jet milled to a particle size distribution of around 4 3 1026 m. The introduction of these materials as CRMs with very small amounts needed to determine the certified concentrations will help to meet the needs of micro-analytical techniques for natural matrix reference materials.Keywords: Homogeneity; biological reference materials; particle size distribution; solid sampling atomic absorption spectrometry Certified Reference Materials (CRMs) from various producers are commonly used in trace analysis for method development and quality control purposes. Analytical techniques are becoming more and more sensitive and require smaller amounts of sample.Particularly the so-called ‘micro-analytical techniques’ such as m-PIXE, laser ablation MS, solid sampling AAS and Xray fluorescence techniques are especially sensitive to sample homogeneity. Determinations rely in some cases almost entirely on the availability of CRMs of proven homogeneity in the range of milligrams or below. Some of the producers of CRMs state in their certificates, however, that certified values are valid only if more than 100 mg of the material is used for analysis.Besides the fact that precious material is wasted if a homogeneous material is consumed in such a large quantity for the determination of a single element in one analytical run, a more precise evaluation of the homogeneity with regard to the element (or compound) considered in a given material would be desirable. The precision of an analytical measurement is a direct function of the stability of the instrument’s response and the material’s homogeneity, i.e., detecting the same chemical composition in successive aliquots.With decreasing mass of the aliquot the statistical probability of finding the same distribution of particles with identical overall composition decreases. Assuming a certain degree of heterogeneity of natural matrix materials, the homogeneity of a reference material at a level of 1026–1023 g aliquots is a direct function of its particle size distribution.1 Trace elements and compounds tend to be unevenly distributed throughout a biological or other environmental material.In the course of reference material production, homogenization of such materials is a critical step and generally results in a finely ground powder. The experimental determination of elementspecific homogeneity factors is tedious and time consuming. In many cases it is assumed from an exemplary survey analysis of one or two elements (with sometimes inadequate methods and/ or irrelevant sample mass) that all the certified elements behave in the same way.In order to be on the safe side, CRM producers generally quote the homogeneity of their materials to be satisfactory only at a high to very high (100–500 mg) sample mass.2 Precisely determined sampling constants3 or homogeneity factors for individual elements, which enable analysts to trace the material-inherent variability of concentrations to the level of material consumed for a particular analytical technique, would help to increase considerably the precision of results from such techniques and help to establish a higher degree of reliability of the results.The accurate applicabimicro-analytical techniques in biomedical, forensic, archaeological and palaeontological investigations would be greatly enhanced. Downscaling of sample preparation techniques and of digestion instruments would be feasible. If smaller amounts of reference materials could be used for quality assurance in various applications, these valuable materials could be used more extensively and the production of new materials could be concentrated on a greater diversity of matrices rather than on replacing exhausted materials.Experimental Three batches of single-cell green algae (Chlorella sp.) grown in cultivation media with different concentration levels of a number of environmentally relevant elements were produced at the Czech University of Agriculture, Prague, in large amounts.4 After harvesting 30–50 kg of fresh material with elevated, environmental and low levels of trace elements, the materials were air sprayed for drying.Without further grinding, the materials with elevated levels (IAEA-393) and the material with environmental levels (IAEA-392) were distributed among the participating laboratories of the first IAEA-AQCS meeting on ‘Reference materials for micro-analytical nuclear techniques’ held in Zagreb, Croatia, in December 1994.5 The third material, single-cell algae with low levels of trace elements (IAEA-391), was distributed only in early 1996 and could therefore not be fully implemented in this intercomparison.A second candidate reference material, ‘urban dust’, was distributed at the same time in two versions: one batch had been sieved through a mesh size of 70 31026 m (IAEA-396A/S) and the other was air-jet milled material (IAEA-396A/M).Both materials originate from a collection of air particulate matter extracted from the air conditioning filters of a Vienna hospital and should reflect the average loading of air particulates from an industrialized urban environment. The material was collected in Analyst, January 1997, Vol. 122 (27–31) 27a large enough amount and is intended to be used as a future air filter reference material.6 Solid sampling Zeeman-effect atomic absorption spectrometry (SS-ZAAS) is a well established technique within the AAS family of techniques.7,8 It allows the introduction of small amounts (0.02–20 mg) of solid materials for electrothermal atomization in a graphite furnace and single-element determination.Zeeman splitting of the specific absorption lines is performed by a strong magnetic field applied at the emission lamp and the instrument (SM 30, Firma Gr�un Optic, Wetzlar, Germany) is automatically tuned for zero adjustment and optimum sensitivity. Calibration was carried out using appropriate reference materials (NIES No. 9, sargassum for algae, and NIST SRM 1648, urban particulate matter for the IAEA dust materials). As it turned out that the peak height of the absorption signals gave more consistent results than the integrated peak area, peak height was used throughout the evaluation of our experiments for quantification. Whether this is due to a specific software problem or is a special feature of the direct solid sampling techniques could not be investigated.The technique is rapid (no sample digestion), inexpensive and sensitive for a large number of elements.9 The laser particle analyser Analysette 22 (Fritsch, Idar- Oberstein, Germany) contains an He–Ne laser and a detector system with 31 units each containing 10 channels. Particles between 0.1 and 1250 3 1026 m can be determined in intervals of 250 3 1026 m with a resolution of 310 channels. About 100 mg of dry sample are dispersed in propan-2-ol and agitated in a ultrasonic bath for about 10 min.The suspension is then pumped through the measurement cell of the instrument with a built-in ultrasonic agitator. The diffraction pattern of the scattered laser light is used to calculate the particle size and the mean distribution of particle sizes is recorded.10 Results Particle Size Distributions In Figs. 1–5, the determined particle size distributions of the five materials [algae elevated level (IAEA-393), algae environmental level (IAEA-392), algae low level (IAEA-391), urban dust sieved (IAEA-396A/S) and urban dust jet milled (IAEA-396A/M)] are shown.From Figs. 1–5, it is clear that all materials exhibit fairly sharp particle size distributions with 90% of the particles smaller than 7, 15, 4.5, 20 and 25 mm for the respective materials. This finding points to very homogeneous materials where small amounts for analysis could show good reproducibility of replicate measurements11 provided that no systematic errors hamper the quantification.The systematic error of the technique was tested in performing replicate measurements with liquid standard solutions. It could be shown that SS-ZAAS does not add more than 3% to the overall standard deviation and is therefore well suited for this kind of homogeneity study. Homogeneity Factor and Sampling Constants According to Ingamells and Swizer,12 a sampling constant can be given by Ks = R2m (1) where Ks = sampling constant, R = relative standard deviation and m = mean sample mass (mg).This constant was originally intended to be used in geological sampling of larger amounts than are generally used for analysis, notably in microanalysis. Fig. 1 Particle size distribution in IAEA-393, algae elevated level. Fig. 2 Particle size distribution in IAEA-392, algae environmental level. Fig. 3 Particle size distribution in IAEA-391, algae low level. Fig. 4 Particle size distribution in IAEA-396A/M, urban dust air-jet milled. 28 Analyst, January 1997, Vol. 122Kurf�urst et al.13 therefore took the square root of this factor to calculate the relative homogeneity factor, HE: HE = sHOMAmºº (2) where sHOM = relative standard deviation and m = mean sample mass (mg). The difference between the two factors can be illustrated graphically as shown in Figs. 6 and 7. Whereas Ingamell and Switzer’s factor ends up with very large numbers at moderate sample mass, Kurf�urst et al.’s factor is suitable particularly for the description of low sample masses but not so much for higher sample masses.For application of milligram sample inputs to the SS-ZAAS instrument we preferred Kurf�urst et al.’s approach to describe the relative homogeneity as HE in mg1 2.11,14 Homogeneity Testing The systematic error of the SS-ZAAS method was checked using liquid standard solutions for several elements. It was found that, related to the absolute mass of element introduced into the graphite furnace, the reproducibility of repetitive measurements varied only between 2 to 3%.The pipetting error can be estimated (by weighing) to be < 1%. Weighing of the solid materials using a Sartorius 4503 microbalance with a maximum loading of 4.1 g and a weighing error of d = ±0.001 mg assures a maximum weighing inaccuracy of < 1% at a sample mass of 0.1 mg. A few examples of the reproducibility of repetitive (20) measurements of the investigated materials using the SS-ZAAS approach are shown in Figs. 8–11. It can be seen that the reproducibility of repetitive measurements varies with the element analysed in a particular sample material and the standard deviation varies with the mass of the sample analysed Fig. 5 Particle size distribution in IAEA-396A/S, urban dust sieved. Fig. 6 Proportionality between sampling constant and sample mass at different fixed standard deviations. Fig. 7 Proportionality between the relative homogeneity factor HE and the sample mass at different fixed standard deviations.Fig. 8 Lead homogeneity in IAEA 393, algae elevated level, mean sample mass 0.193 mg. Fig. 9 Copper homogeneity in IAEA-392, algae environmental level, mean sample mass 0.458 mg. Fig. 10 Copper homogeneity in IAEA 396A/M, urban dust, mean sample mass 0.34 mg. Analyst, January 1997, Vol. 122 29(e.g., Cu in dust at a sample mass of 0.34 mg is reproducible with RSD = 5.5% and at a level of 0.08 mg with RSD = 10%).In order to achieve relative homogeneity factors related to the sample mass used for analysis, a number of measurement series all following the same routine urements, same calibration, same wavelength) were carried out and the results were plotted [RSD versus sample mass (mg)] as shown in Figs. 12 and 13. Fitting the parabolic curves yields a function of x depending on a factor times m21 2. The factor is the relative homogeneity factor of Kurf�urst et al.(HE) and it is related to this particular element’s homogeneity in the investigated material. In Table 1, the HE values for the four materials are given for all the elements investigated. The Cd concentration in IAEA-392 is very low and close to the detection limit of the method (0.015 ± 0.006 mg kg21) at a sample mass of 1.3 mg. For this reason, only one measurement series of 20 measurements could be carried out. Within these 20 measurements one outlier of about three times the mean of all other results was recorded. This value was not rejected and hence the RSD reached about 50%.Cadmium therefore cannot be considered to be homogeneously distributed in this particular material (IAEA-392, algae, environmental level). Lead concentrations could not be determined at all in this material at such a sample mass level. Increasing the sample mass by a factor of ten resulted in a more sensitive determination. Discussion and Conclusion It was shown that the particle size distributions of all five materials are exceptionally narrow and peak at very low particle sizes, compared with the normal range of other biological reference materials. From this finding, the assumption of good to very good homogeneity for trace elements can be drawn.Using the approach of Kurf�urst et al. a numerical description of element-specific homogeneity of the materials was experimentally elaborated. As Kurf�urst and co-workers pointed out,13–16 HE < 10 indicates very good homogeneity.All the materials investigated for the investigated elements show HE values in this range (with the exception of Cd in IAEA-392) and can be considered suitable as reference materials for micro-analytical techniques analysing samples with masses between 0.1 and 10 mg. For techniques using even smaller amounts of sample, natural matrix reference materials still have to be investigated. As the relative homogeneity factors HE are determined in an accurate way for individual elements in a given material, they can be used to assign the uncertainty of a certified concentration which is due only to material heterogeneity.Today the uncertainty given for certified reference materials is a composite of several uncertainties, such as (i) the systematic error of the analytical technique, (ii) bias from different standardizations of all the techniques used for certification and (iii) the materials’ inherent heterogeneity. By assigning a certified value with the homogeneity factor related to the sample mass consumed in analysis it is possible to calculate a mass-dependent uncertainty which derives strictly from the quality of the material itself. Hence systematic errors and bias of analytical techniques will be easier to recognize and the analytical process will be more transparent.The accurate determination of element-specific homogeneity factors for certified reference materials is essential for all CRMs to be used for quality control in micro-analytical trace element determinations.Suitable techniques such as INAA and SSZAAS with no sample digestion and accurate control over the total sample mass analysed are available and should be applied regularly in the course of certification of CRMs for the quantification of trace element distributions in natural matrix materials. The ‘true value’ of an element in a matrix is not a very useful parameter unless reliable information on the probability of obtaining exactly this value in repetitive measurements (repro- Fig. 11 Copper homogeneity IAEA 396, urban dust air-jet mulled, mean sample mass 0.08 mg. Fig. 12 RSD versus sample mass for Pb in IAEA 393, algae elevated level. The data points can be fitted as f(x) + 4.68 m21 2. Fig. 13 RSD versus sample mass for Cu in IAEA 393, algae elevated level. The data points can be fitted as f(x) + 8.67 m21 2. Table 1 Relative homogeneity factors, HE, for individual elements in the four materials investigated No.of Material measurements HE (Cu) HE (Cd) HE (Pb) IAEA-392 20 8.9 45.3 —* IAEA-393 80 8.7 4.0 4.7 IAEA-396A/M 20 3.0 4.6 6.8 IAEA-396A/S 20 8.9 8.4 6.7 * Pb in IAEA-392 was below the limit of determination. 30 Analyst, January 1997, Vol. 122ducibility) of the same material is attached to it. This probability is clearly dependent on the number of particles with potentially differing concentrations in an analytical aliquot and hence on the sample mass consumed for analysis.Hitherto certified values were given with a conservative estimate (amounting to 50% in some cases) of the overall uncertainty. The empirical statement ‘a minimum sample mass of 250 mg of the dried material . . . is necessary for any certified value . . . to be valid within the stated uncertainty’ is misleading and may only have a commercial background. Precisely determined individual homogeneity factors for each element would therefore help CRM users to (i) save precious material, (ii) determine the systematic error of the applied analytical technique and (iii) test the reliability of the sample preparation techniques more accurately and should therefore be endorsed also by CRM producers.We are grateful for and greatly appreciate the intensive work of D. Koglin in particle size analysis. M.R. acknowledges the financial support of the Bundesminister f�ur Umwelt, Naturschutz und Reaktorsicherheit, Bonn, and the Bundesumweltamt, Berlin. Many thanks are due to the IAEA (Dr. R. Zeisler, Dr. V. Valkovic) for letting us participate in this interesting AQCS programme. References 1 Rossbach, M., Ostapczuk, P., Schladot, J. D., and Emons, H., UWSFZ. Umweltchem. � Okotox., 1995, 7(6), 365. 2 Pauwels, J., and Vandecasteele, C., Fresenius’ J. Anal. Chem., 1993, 345, 121. 3 Chatt, A., Jayawickreme C. K., and McDowell, L. S., Fresenius’ J. Anal. Chem., 1990, 338, 399. 4 Mader, P., Stejskalova, I., and Slamova, A., Fresenius’ J. Anal. Chem., 1995, 352, 131. 5 Report of the Research Co-ordination Meeting on Reference Materials for Microanalytical Nuclear Techniques, IAEA/AL/083, IAEA, Vienna, 1994. 6 Zeisler, R., personal communication. 7 L�ucker, E., K�onig, H., Gabriel, W., and Rosopulo, A., Fresenius’ J. Anal. Chem., 1992, 342, 941. 8 Mohl, C., Grobecker, K. H., and Stoeppler, M., Fresenius’ J. Anal. Chem., 1987, 328, 413. 9 Kurf�urst, U., Fresenius’ Z. Anal. Chem., 1982, 313, 97. 10 Friedrich, H., and Mansour, A., Nachr. Chem. Tech. Lab., 1995, 43, 87. 11 Sonntag, Th.-M., Homogenit�atsstudien f�ur ausgew�ahlte Elemente in verschiedenen Materialien der Bank f�ur Umweltproben und der Internationalen Atomenergiebeh�orde mit der Solid-sampling-Zeeman- Atomabsorptionsspektrometrie, Diplomarbeit, Fachhochschule Aachen, Abteilung J�ulich, 1996. 12 Ingamells, C. O., and Swizer, P., Talanta, 1973, 20, 547. 13 Kurf�urst, U., Grobecker, K.-H., and Stoeppler, M., Trace Elem., 1984, 3, 591. 14 Stoeppler, M., Kurf�urst, U., and Grobecker, K.-H., Fresenius’ J. Anal. Chem., 1985, 322, 687. 15 Kurf�urst, U., Fresenius’ Z. Anal. Chem., 1983, 315, 304. 16 Kurf�urst, U., Fresenius’ Z. Anal. Chem., 1983, 316, 1. Paper 6/05396C Received August 1, 1996 Accepted October 21, 199

ISSN:0003-2654    

DOI:10.1039/a605396c

出版商:RSC    

年代:1997

数据来源: RSC

摘要:

Quantitative Analysis of Sulfated Calcium Carbonates Using Raman Spectroscopy and X-ray Powder Diffraction Christos G. Kontoyannis*†, Malvina G. Orkoula and Petros G. Koutsoukos ICE/HT-FORTH and Department of Pharmacy, University of Patras, University Campus, GR 26500 Patras, Greece A non-destructive method based on the use of Raman spectroscopy (RS) for the determination of the percentage of gypsum in sulfated marble is presented. The Raman spectra of well mixed powder samples of calcite–aragonite, calcite–gypsum and gypsum–aragonite pairs of mixtures were recorded and the characteristic bands at 280 cm21 for calcite, 205 cm21 for aragonite and 412 cm21 for gypsum were used as the basis for the quantitative analysis of specimens in which the most stable calcium carbonate phases, calcite and aragonite, were present.The detection limits were found to be 0.3 mol% for calcite, 0.5 mol% for aragonite and 0.6 mol% for gypsum. For samples containing only one calcium carbonate phase the use of the strong and sharp Raman band at 1085 cm21, common for aragonite and calcite, together with the intensity of the Raman peak at 1006 cm21 for gypsum, yielded lower detection limits: calcite 0.1, aragonite 0.1 and gypsum 0.05 mol%.The analysis by RS was compared with X-ray powder diffraction (XRD). In this analysis, the calibration curves were constructed using the relative intensities corresponding to the 113, the 111 and the 121– reflections of the calcite, aragonite and gypsum, respectively.The detection limits for calcite, aragonite and gypsum were 4, 5 and 1–2 mol%, respectively. The potential of using RS for a point-by-point analysis (‘mapping’) of a surface by focusing the laser beam on the selected spots was also demonstrated on a marble sample removed from Athens National Garden, exposed in the open air. Keywords: Raman spectroscopy; calcium carbonate polymorphs; gypsum; X-ray powder diffraction Marble is a metamorphic rock formed from limestone by geological processes involving high temperatures and high pressures.Limestone, in turn, is a calcium carbonate rock built up from the sedimentation of marine organisms over millions of years. Marble sulfation has been the subject of many investigations in the recent past, ranging from studies of field samples1,2 to laboratory investigations using synthetic atmospheric environments enriched in SO2.3–5 Atmospheric pollution has been accused of causing the deterioration of calcium carbonate stones and several mechanisms, leading to the conversion of CaCO3 into CaSO4·2H2O, have been proposed.2,6–8 The development of a technique for monitoring the progress of marble sulfation on important cultural monuments in a nondestructive manner is needed.Commonly used methods include the employment of classical elemental analysis (CEA),6 scanning electron microscopy (SEM),1,6,9 X-ray powder diffraction (XRD),1,9 petrographic microscopy (PM)6,9 and atomic absorption spectrometry (AAS).6 Some of these methods are used for identification purposes only (SEM, PM), others are destructive for the sample (CEA, XRD, SEM, PM, AAS) and others are suitable for elemental analysis only and not for the determination of the species present (CEA, AAS).The potential presence of the two commonly encountered crystal phases of calcium carbonate, calcite and aragonite, in the specimen10,11 complicates the analytical problem.The identification of these two phases and the determination of the respective percentages can be accomplished through the use of vibrational spectroscopic techniques such as infrared (IR) and Raman spectroscopy (RS). Although the IR determination of the calcite to aragonite ratio has been reported,12,13 disadvantages of this method include broadness of inorganic absorption bands and specimen preparation involving grinding or pelleting, which can lead to the conversion of aragonite into calcite.RS is a non-destructive technique which has already been used for the identification of calcium carbonate phases14 with the potential for in situ application using fibre optics. This method however, to our knowledge, has not yet been applied to the quantitative analysis of sulfated calcareous rocks. In this work, the possibility of using RS as a non-destructive technique for the determination of the gypsum content in marble, was investigated and the results were compared with those obtained by application of quantitative XRD analysis.15 Experimental Preparation of Chemicals and Samples Pure aragonite crystals were prepared by the simultaneous dropwise addition of 5 ml of a solution of 1 mol l21 Ca(NO3)2 (Ferak, Berlin, Germany) at 90 °C and 5 ml of 1 mol l21 (NH4)2CO3 (Ferak) at 45 °C into 200 ml of triply distilled water at 95 °C.The solution, during precipitation, was saturated with CO2 by bubbling the gas through the slurry. The crystals, in the form of a slurry, were filtered (Millipore, Bedford, MA, USA; 0.22 mm) and washed with triply distilled water at 90 °C and with absolute ethanol at room temperature.The powder was dried at 80 °C for 1 h and stored in a desiccator. Calcite powder was prepared as follows: 1 l of 1 mol l21 (NH4)2CO3 solution was added dropwise to 1 l of 1 mol l21 Ca(NO3)2 solution and stirred magnetically at ambient temperature. The suspension was incubated in the mother liquor for 15 d.Next, it was filtered through membrane filters and washed with triply distilled water at 70 °C. The crystals were dried at 120 °C for 2 d and stored in a desiccator. Gypsum powder was prepared by adding 1 l of 0.1 mol l21 Na2CO3 solution (Merck, Darmstadt, Germany) to 1 l of 0.1 mol l21 Ca(NO3)2 solution stirred magnetically at 70 °C. Next, the slurry was filtered, washed with triply distilled water, resuspended and aged at 70 °C for 15 d with stirring.It was then filtered again, washed, dried at 120 °C for 2 d and stored in a dessicator. The calcite, aragonite and gypsum crystals were characterized by IR spectroscopy, XRD, and SEM (JEOL, Tokyo, Japan; JSM 5200). † Present address: Institute of Chemical Engineering and High Temperature Chemical Processes, University Campus, P.O. Box 1414, GR 26500, Patras, Greece. Analyst, January 1997, Vol. 122 (33–38) 33In order to construct the calibration curves, carefully weighed mixtures of calcite–aragonite, aragonite–gypsum and gypsum– calcite, ranging from 0 to 100 mol% purity, were prepared from the respective solids.The solid mixtures were thoroughly mixed mechanically. The homogeneity of the mixed powders was verified by obtaining several Raman spectra for each mixture, focusing the laser beam at randomly selected parts of the surface. Instrumentation Raman spectroscopy (system configuration) Raman spectra were excited by focusing 488 nm radiation from a 4 W Spectra-Physics (San Jose, CA, USA) argon laser on the marble sample and on the synthetically prepared mixtures.The plasma lines were removed from the laser beam by using a small monochromator as a filter. A cylindrical lens, with focal length 127 mm, was used to focus the laser line on the sample. The scattered light was collected at an angle of 90° and analysed with a Spex Industries (Edison, NJ, USA) Model 1403, 0.85 m double monochromator equipped with an RCA photomultiplier cooled to 220 °C and EG&G Ortec (Oak Ridge, TN, USA) photon-counting and electronic amplifier.The power of the incident laser beam was about 200 mW on the sample surface. Typical spectral resolution and time constant were 0.3 cm21 and 3 s, respectively. The system was interfaced with a computer. X-ray diffractometry X-ray powder diffraction analysis was performed with a Philips (Eindhoven, The Netherlands, Model 1830/40 instrument) on finely powdered samples using Cu Ka radiation (40 kV and 30 mA) and an Ni filter with a scanning speed of 0.005° 2q s21.The time constant was set at 2 s. Results and Discussion Theory for Construction of Raman Calibration Curve The Raman spectra of calcite, aragonite and gypsum are shown in Fig. 1. The characteristic Raman bands due to the lattice vibration mode at 280, 205 and 412 cm21 for calcite, aragonite and gypsum, respectively, and the n1 internal mode (symmetric stretching) at 1085 cm21 for both calcite and aragonite16,17 and at 1006 cm21 for gypsum16 are easily distinguished.The spectra are in good agreement with earlier reports.18 It is obvious that the Raman bands in the spectral region between 170 and 450 cm21 should be used for the quantitative analysis of a mixture of calcite, aragonite and gypsum since the strong peak at 1085 cm21, which is attributed to the symmetric C–O stretching, is common for all calcium carbonate phases. On the other hand, when the specimen contains only one CaCO3 phase and gypsum, the 1085 and 1006 cm21 bands can also be used for quantitative analysis.In this case a lower detection limit is expected, since the analysis is based on stronger and sharper peaks. In the present work both possibilities were addressed by constructing Raman calibration curves using the bands of the spectral regions of interest, i.e., 170–450 and 970–1110 cm21. The purpose was to find an easy and reliable method for calculating each ingredient’s percentage and, therefore, peak heights were used and not integrated intensities of the bands.The intensity of a Raman line depends on a number of factors, including incident laser power, frequency of scattered radiation, absorptivity of the materials involved in the scattering and the response of the detection system. Thus, the measured Raman intensity, I(n), can be represented by19 I(n) = I0 K(n) C (1) where I0 is the intensity of the exciting laser radiation, n is the Raman shift, K(n) is a factor which includes the frequencydependent terms (the over-all spectrophotometer response, the self-absorption of the medium and the molecular scattering properties) and C is the concentration of the Raman-active species.Differences in the measured intensities of the various spectra obtained from the same specimen, which are attributed to factors such as the intensity variation of I0 or the positioning of the sample, suggested the use of relative factors within each spectrum, e.g., ratio of band intensities characteristic for each component.Thus, in a spectrum obtained from a sample in which several species are present, the ratio of the intensities of two peaks attributed to different compounds should be given by IA IB = KA KB ¥ xA xB (2) where the subscripts A and B indicate the different components and xA/xB is the molar fraction ratio of the two species. Eqn. (2) is valid only when there is no chemical interaction between the substances present in the sample.From eqn. 2, it is apparent that a plot of IA/IB versus xA/xB should yield a straight line with slope KA/KB. In the frequency range of the spectral measurements, the over-all spectrometer response may be considered to be constant, hence the K ratios are dependent only on the scattering parameter associated with each band, assuming that no significant absorption of the exciting radiation occurs. Consequently, the values assigned to the K ratio may be used regardless of the Raman spectroscopic system, provided that 488 nm radiation from an argon laser is used.Raman Calibration Lines For an aragonite–calcite mixture, eqn. (2) may be rewritten for the 280 cm21 peak of calcite and the 205 cm21 peak of aragonite as follows: Ic 280 Ia 205 = Kc 280 Ka 205 ¥ xc xa (3) where the subscripts c and a represent the calcite and the aragonite crystal phases, respectively, and the superscripts 280 and 205 are the wavenumbers of the respective Raman bands.The plot of Ic 280/Ia 205 versus xc/xa is shown in Fig. 2. The Fig. 1 Raman spectra of the synthetically prepared A, calcite, B, aragonite and C, gypsum. 34 Analyst, January 1997, Vol. 122equation for the calibration line was obtained by linear regression of the experimental data: Ic 280 Ia 205 = 1.646 ¥ xc xa (4) The correlation coefficient (r), was 0.999 997 and the standard deviation (s), for the slope and the intercept were 1.4 3 1023 and 4.9 3 1022, respectively. Assuming that xa + xc = 1, the RSD for xa = 0.4 was found to be 3.3%, and the detection limits (DL) for calcite and aragonite were calculated to be of the order of 0.3 and 0.5%, respectively.Similarly, calibration lines for calcite–gypsum and aragonite –gypsum mixtures were constructed using the Raman bands at 280 and 1085 cm21 for calcite, 412 and 1006 cm21 for gypsum and 205 and 1085 cm21 for aragonite.Typical spectra of the various mixtures are shown in Fig. 3. The cumulative information for the RS calibration lines appear in Table 1. Calculation of the Molar Fraction of Calcite, Aragonite and Gypsum in a Sample Using the Raman Calibration Curves Two-component systems The gypsum–calcite and the gypsum–aragonite mixtures belong to this category. The stronger Raman bands at 1085 cm21 for either the calcite or aragonite phases and at 1006 cm21 for gypsum can be used [eqn. (7) and (8) in Table 1].For the calcite–gypsum mixtures and assuming that xg + xc = 1, eqn. (7) can be transformed to xg = 0.721Ig 1006 Ic 1085 + 0.721Ig 1006 (9) Similarly, for the aragonite–gypsum system, eqn. (8) can be rewritten as xg = 0.748Ig 1006 Ia 1085 + 0.748Ig 1006 (10) Three-component systems This case applies to sulfated calcareous rocks in which both calcium carbonate phases are present together with gypsum. The calibration lines used in the two-component system case cannot be used since both aragonite and calcite exhibit a strong peak at the same frequency, 1085 cm21. Eqns.(4), (5) and (6) in Table 1 were used instead. Assuming that xa + xg + xc = 1, Fig. 2 Raman calibration line for calcite–aragonite mixtures. Table 1 Cumulative results for the RS calibration curves Eqn. s of s of No. Calibration line r slope intercept RSD (%) DL (mol%) I c 280 xc 3.3 Calcite 0.3 (4) = 1.646 3 0.999 997 1.4 3 1023 4.9 3 1022 (xa = 0.4) Aragonite 0.5 I a 205 xa I c 280 xc 3.3 Calcite 0.3 (5) = 1.912 3 0.999 94 8.7 3 1023 6 3 1022 (xg = 0.4) Gypsum 0.6 Ig 412 xg I a 205 xa 3.0 Aragonite 0.5 (6) = 1.171 3 0.999 08 2.3 3 1022 3.7 3 1022 (xg = 0.4) Gypsum 0.6 I g 412 xg I c 1085 xc 2.0 Calcite 0.1 (7) = 0.721 3 0.999 97 2.2 3 1023 1.4 3 1022 (xg = 0.4) Gypsum 0.05 I g 1006 xg I a 1085 xa 1.7 Aragonite 0.1 (8) = 0.748 3 0.999 99 1.7 3 1023 1.1 3 1022 (xg = 0.4) Gypsum 0.05 I g 1006 xg Fig. 3 Raman spectra of: A, 20 mol% calcite–80 mol% gypsum; B, 80 mol% calcite–20 mol% gypsum; C, sulfated marble sample; D, 20 mol% aragonite–80 mol% gypsum; E, 80 mol% aragonite–20 mol% gypsum; and F, 40 mol% calcite–30 mol% aragonite–30 mol% gypsum.Analyst, January 1997, Vol. 122 35the molar fractions in a sample may be determined from the following relationships: xa = 1.646Ia 205 Ic 280 +1.646Ia 205 +1.927Ig 412 (11) xc = Ic 280 1.646Ia 205 ¥ xa (12) xg = 1.171Ig 412 Ia 205 ¥ xa (13) The validity of these expressions was tested on the spectrum recorded from a powder mixture consisting of 40 mol% calcite, 30 mol% aragonite and 30 mol% gypsum [Fig. 3(F)]. The results were calcite 39.7, aragonite 30.9 and gypsum 29.4 mol%. The deviation of the results obtained was within the experimental error (s in Table 1). It should be noted that the analytical methodology presented here does not depend on the simultaneous existence of the calcium carbonate phases and the gypsum, since there is neither chemical interaction between these species nor overlap of the corresponding bands in the Raman spectra.As a result, the relative intensities used in the analysis are not affected. Moreover, as may be seen from eqn. (1), the intensity of the Raman bands depends on the concentration of the investigated species alone. If an additional compound, besides gypsum, aragonite and/or calcite, is also present and provided that this does not contribute to the RS signal at the proposed frequencies and that there is no chemical interaction among the species present, eqns.(4)–(8) are still valid and the ratio of the gypsum to the calcium carbonate phases can be determined. Theory for Construction of XRD Calibration Curve The XRD spectra of calcite, aragonite and gypsum are shown in Fig. 4. The calcite spectrum exhibits two major peaks associated with the 104 and 113 reflections. Unfortunately, the former coincides with a gypsum peak, so only the intensity of the 113 reflection was used.The peaks attributed to the 121 – and 111 reflections of gypsum and aragonite, respectively, were also used for the quantitative analysis. If the sample is a uniform mixture of two components and extinction and microabsorption effects are neglected, it can be shown that20 IA IB = L ¥ xA xB (14) where L is a proportionality constant which depends on the component, the diffraction line and the mass absorption coefficient of the species present. IA/IB represents the ratio of the intensities of two selected diffraction lines in a mixture of two substances, and xA/xB is the molar fraction ratio of the two substances. A plot of IA/IB versus xA/xB should yield a straight line with an intercept of zero.XRD Calibration Lines Calibration curves for calcite–aragonite, aragonite–gypsum and gypsum–calcite binary mixtures were constructed. For gypsum–aragonite mixtures the plot of Ia 111/Ig 121– versus xa/xg, where the subscripts a and g represent gypsum and aragonite, respectively, and the superscripts the XRD reflections, as shown in Fig. 5. The equation for the calibration line was obtained by linear regression of the experimental data: Ia 111 Ig 121 = 0.18 ¥ xa xg (15) The correlation coefficient (r) was 0.9993 and the s for the slope and the intercept were 3 3 1022 and 2.3 3 1022, respectively. Assuming that xa + xg = 1, the RSD for xg = 0.4 was found to be 12.2%, and DL for gypsum and aragonite were calculated to be of the order of 1–2 and 5%, respectively.Similarly, calibration lines were constructed for calcite– gypsum and aragonite–calcite mixtures using the XRD reflection peaks of 113, 121 – and 111 for calcite, gypsum and aragonite, respectively. Representative spectra are shown in Fig. 6. The cumulative information for the XRD calibration lines is given in Table 2. Calculation of the Molar Fraction of Calcite, Aragonite and Gypsum in a Ternary Sample using the XRD Calibration Curves Assuming that xa + xg + xc = 1 and by employing eqns.(15), (16) and (17) in Table 2, the molar fractions in a sample in which calcite, aragonite and gypsum are present can be determined using the following relationships: xa = 1.13Ia 111 1.13Ia 111 + Ic 113 + 0.20Ig 121 (18) Fig. 4 XRD spectra of the synthetically prepared A, calcite, B, aragonite and C, gypsum. Fig. 5 XRD calibration line for gypsum–aragonite mixtures. 36 Analyst, January 1997, Vol. 122xc = Ic 113 1.13Ia 111 ¥ xa (19) xg = 0.18Ig 121 Ia 111 ¥ xa (20) The validity of the expressions derived for the XRD technique was also tested on a spectrum recorded from the same ternary powder mixture as used for testing the Raman calibration curves, Fig. 6(C). The percentages determined were calcite 41.9, aragonite 30.8 and gypsum 27.3 mol%. The deviation of the results obtained was within the experimental error (s in Table 2). As mentioned for the RS method, the presence of a fourth compound does not affect the validity of the derived equations provided that there is no chemical interaction among the species present or overlap of the XRD peaks used for the analysis with the XRD lines of the additional species.Application to a Sulfated Marble Sample A marble sample taken from Athens National Garden was tested for gypsum using the techniques described here. Raman spectra were excited from several points of the marble surface. Only calcite and gypsum were present [Fig. 3(C)]. Application of eqn. (9) yielded gypsum concentrations between 0 and 8.5 mol%. An external layer from sample surface was removed mechanically and the XRD spectrum was recorded [Fig. 6(D)]. No detectable gypsum was found. This result was expected because only the average percentage from the removed surface layer can be observed with the XRD technique, and the possibility of having material from the inner marble layer, consisting of pure calcite only, in the powder collected from the surface is not trivial.Hence the percentage of gypsum in the XRD-tested material was below the detection limit. Comparison Between RS and XRD Both techniques were used for the determination of the percentage of gypsum on a calcium carbonate surface, and RS exhibited certain advantages over the XRD method: (a) RS was non-destructive for the sample and less time consuming; (b) reliable point-by-point analysis (‘mapping’) of the surface was accomplished using RS, whereas XRD yielded only the average percentage of the bulk, ground powder sample; and (c) from the comparison of the calibration line statistics (Tables 1 and 2) it can be seen that RS exhibited lower SD and lower DL than the XRD calibration curves.Conclusions Methods based on RS and XRD for quantitative determination of the transformation of the surface of monuments into gypsum, known as marble deterioration, were developed. Calibration curves from mixtures of calcite and aragonite (the most stable phases of calcium carbonate), and gypsum were constructed.The much lower detection limits given by RS, the fact that it is non-destructive and the potential use of the technique for chemical mapping of the marble surface are among the major advantages of the RS over powder XRD. The authors are indebted to Professor G.N. Papatheodorou for helpful suggestions and for providing the experimental facilities. They thank Dr. P. Klepetsanis for kindly supplying the gypsum powder.Partial support of this work by the GSRT EPET II Program (contract No. 368/11-1-95) is gratefully acknowledged. References 1 Camuffo, D., Del Monte, M., Sabbioni, C., and Vittori, O., Atmos. Environ., 1982, 16, 2253. 2 Ross, M., McGee, E. S., and Ross, D. R., Am. Mineral., 1989, 74, 367. 3 Skoulikidis, T., and Charalambous, D., Br. Corros. J., 1981, 16, 70. 4 Johanson, L. G., Lindqvist, O., and Mangio, R., Durability Build. Mater., 1988, 5, 439. 5 Gauri, K.L., Chowdhury, A. N., Kulshreshta, N. P., and Punuru, A. R., Stud. Conserv., 1989, 34, 201. 6 Verges-Belmin, V., Atmos. Environ., 1994, 28, 295. 7 Van Houte, G., Rodrique, L., Genet, M., and Delmon, B., Environ. Sci. Technol., 1981, 15, 327. 8 Lipfert, F. W., Atmos. Environ., 1989, 23, 415. 9 Gauri, K. Lal, and Holdren G. C., Jr., Environ. Sci. Technol., 1981, 15, 386. 10 Berner, R. A., Am. J. Sci., 1966, 264, 1. Table 2 Cumulative results for the XRD calibration lines Eqn.s of s of No. Calibration line r slope intercept RSD (%) DL (mol%) Ia 111 xa 12.2 Aragonite 5 (15) = 0.18 3 0.9993 3 3 1022 2.3 3 1022 (xg = 0.4) Gypsum 1–2 Ig 121 – xg Ic 113 xc 13.3 Calcite 4 (16) = 0.18 3 0.9853 1.5 3 1022 2.7 3 1022 (xg = 0.4) Gypsum 1–2 Ig 121 – xg Ic 113 xc 11.2 Calcite 4 (17) = 1.13 3 0.9991 2.4 3 1022 4.2 3 1022 (xa = 0.4) Aragonite 5 Ia 111 xa Fig. 6 XRD spectra of: A, 20 mol% calcite–80 mol% gypsum; B, 80 mol% calcite–20 mol% gypsum; C, 40 mol% calcite–30 mol% aragonite– 30 mol% gypsum; and D, sulfated marble sample.Analyst, January 1997, Vol. 122 3711 Hacker, B. R., Kirby, S. H., and Bohlen, S. R., Science, 1992, 258, 110. 12 Compere, E. L., and Bates, J. M., Limnol. Oceanogr., 1973, 18, 326. 13 Xyla, A., and Koutsoukos, P. G., J. Chem. Soc., Faraday Trans. 1, 1989, 85, 3165. 14 Herman, R. G., Bogdan, C. E., Sommer, A. J., and Simpson, D. R., Appl. Spectrosc., 1987, 41, 437. 15 Silk, S. T., and Lewin, S. Z., Adv.X-ray Anal., 1971, 14, 29. 16 Griffith, P. G., in Spectroscopy of Inorganic-based Materials, eds. Clark, R. J. H., and Hester, R. E., Wiley, Chichester, 1987, pp. 137 and 151. 17 Behrens, G., Kuhn, L. T., Ubic R., and Heuer, A. H., Spectrosc. Lett., 1995, 28, 983. 18 Degen, A., and Newman, G. A., Spectrochim. Acta, Part A, 1993, 49, 859. 19 Strommen, D., and Nakamoto, K., in Laboratory Raman Spectroscopy, Wiley, New York, 1984, pp. 71–75. 20 Whiston, C., in X-ray Methods, Wiley, New York, 1987, p. 113.Paper 6/06167B Received September 9, 1996 Accepted September 24, 1996 38 Analyst, January 1997, Vol. 122 Quantitative Analysis of Sulfated Calcium Carbonates Using Raman Spectroscopy and X-ray Powder Diffraction Christos G. Kontoyannis*†, Malvina G. Orkoula and Petros G. Koutsoukos ICE/HT-FORTH and Department of Pharmacy, University of Patras, University Campus, GR 26500 Patras, Greece A non-destructive method based on the use of Raman spectroscopy (RS) for the determination of the percentage of gypsum in sulfated marble is presented. The Raman spectra of well mixed powder samples of calcite–aragonite, calcite–gypsum and gypsum–aragonite pairs of mixtures were recorded and the characteristic bands at 280 cm21 for calcite, 205 cm21 for aragonite and 412 cm21 for gypsum were used as the basis for the quantitative analysis of specimens in which the most stable calcium carbonate phases, calcite and aragonite, were present.The detection limits were found to be 0.3 mol% for calcite, 0.5 mol% for aragonite and 0.6 mol% for gypsum. For samples containing only one calcium carbonate phase the use of the strong and sharp Raman band at 1085 cm21, common for aragonite and calcite, together with the intensity of the Raman peak at 1006 cm21 for gypsum, yielded lower detection limits: calcite 0.1, aragonite 0.1 and gypsum 0.05 mol%. The analysis by RS was compared with X-ray powder diffraction (XRD).In this analysis, the calibration curves were constructed using the relative intensities corresponding to the 113, the 111 and the 121– reflections of the calcite, aragonite and gypsum, respectively. The detection limits for calcite, aragonite and gypsum were 4, 5 and 1–2 mol%, respectively. The potential of using RS for a point-by-point analysis (‘mapping’) of a surface by focusing the laser beam on the selected spots was also demonstrated on a marble sample removed from Athens National Garden, exposed in the open air.Keywords: Raman spectroscopy; calcium carbonate polymorphs; gypsum; X-ray powder diffraction Marble is a metamorphic rock formed from limestone by geological processes involving high temperatures and high pressures. Limestone, in turn, is a calcium carbonate rock built up from the sedimentation of marine organisms over millions of years. Marble sulfation has been the subject of many investigations in the recent past, ranging from studies of field samples1,2 to laboratory investigations using synthetic atmospheric environments enriched in SO2.3–5 Atmospheric pollution has been accused of causing the deterioration of calcium carbonate stones and several mechanisms, leading to the conversion of CaCO3 into CaSO4·2H2O, have been proposed.2,6–8 The development of a technique for monitoring the progress of marble sulfation on important cultural monuments in a nondestructive manner is needed.Commonly used methods include the employment of classical elemental analysis (CEA),6 scanning electron microscopy (SEM),1,6,9 X-ray powder diffraction (XRD),1,9 petrographic microscopy (PM)6,9 and atomic absorption spectrometry (AAS).6 Some of these methods are used for identification purposes only (SEM, PM), others are destructive for the sample (CEA, XRD, SEM, PM, AAS) and others are suitable for elemental analysis only and not for the determination of the species present (CEA, AAS).The potential presence of the two commonly encountered crystal phases of calcium carbonate, calcite and aragonite, in the specimen10,11 complicates the analytical problem. The identification of these two phases and the determination of the respective percentages can be accomplished through the use of vibrational spectroscopic techniques such as infrared (IR) and Raman spectroscopy (RS).Although the IR determination of the calcite to aragonite ratio has been reported,12,13 disadvantages of this method include broadness of inorganic absorption bands and specimen preparation involving grinding or pelleting, which can lead to the conversion of aragonite into calcite.RS is a non-destructive technique which has already been used for the identification of calcium carbonate phases14 with the potential for in situ application using fibre optics. This method however, to our knowledge, has not yet been applied to the quantitative analysis of sulfated calcareous rocks.In this work, the possibility of using RS as a non-destructive technique for the determination of the gypsum content in marble, was investigated and the results were compared with those obtained by application of quantitative XRD analysis.15 Experimental Preparation of Chemicals and Samples Pure aragonite crystals were prepared by the simultaneous dropwise addition of 5 ml of a solution of 1 mol l21 Ca(NO3)2 (Ferak, Berlin, Germany) at 90 °C and 5 ml of 1 mol l21 (NH4)2CO3 (Ferak) at 45 °C into 200 ml of triply distilled water at 95 °C.The solution, during precipitation, was saturated with CO2 by bubbling the gas through the slurry. The crystals, in the form of a slurry, were filtered (Millipore, Bedford, MA, USA; 0.22 mm) and washed with triply distilled water at 90 °C and with absolute ethanol at room temperature. The powder was dried at 80 °C for 1 h and stored in a desiccator. Calcite powder was prepared as follows: 1 l of 1 mol l21 (NH4)2CO3 solution was added dropwise to 1 l of 1 mol l21 Ca(NO3)2 solution and stirred magnetically at ambient temperature. The suspension was incubated in the mother liquor for 15 d.Next, it was filtered through membrane filters and washed with triply distilled water at 70 °C. The crystals were dried at 120 °C for 2 d and stored in a desiccator. Gypsum powder was prepared by adding 1 l of 0.1 mol l21 Na2CO3 solution (Merck, Darmstadt, Germany) to 1 l of 0.1 mol l21 Ca(NO3)2 solution stirred magnetically at 70 °C.Next, the slurry was filtered, washed with triply distilled water, resuspended and aged at 70 °C for 15 d with stirring. It was then filtered again, washed, dried at 120 °C for 2 d and stored in a dessicator. The calcite, aragonite and gypsum crystals were characterized by IR spectroscopy, XRD, and SEM (JEOL, Tokyo, Japan; JSM 5200). † Present address: Institute of Chemical Engineering and High Temperature Chemical Processes, University Campus, P.O. Box 1414, GR 26500, Patras, Greece.Analyst, January 1997, Vol. 122 (33–38) 33In order to construct the calibration curves, carefully weighed mixtures of calcite–aragonite, aragonite–gypsum and gypsum– calcite, ranging from 0 to 100 mol% purity, were prepared from the respective solids. The solid mixtures were thoroughly mixed mechanically. The homogeneity of the mixed powders was verified by obtaining several Raman spectra for each mixture, focusing the laser beam at randomly selected parts of the surface.Instrumentation Raman spectroscopy (system configuration) Raman spectra were excited by focusing 488 nm radiation from a 4 W Spectra-Physics (San Jose, CA, USA) argon laser on the marble sample and on the synthetically prepared mixtures. The plasma lines were removed from the laser beam by using a small monochromator as a filter. A cylindrical lens, with focal length 127 mm, was used to focus the laser line on the sample.The scattered light was collected at an angle of 90° and analysed with a Spex Industries (Edison, NJ, USA) Model 1403, 0.85 m double monochromator equipped with an RCA photomultiplier cooled to 220 °C and EG&G Ortec (Oak Ridge, TN, USA) photon-counting and electronic amplifier. The power of the incident laser beam was about 200 mW on the sample surface. Typical spectral resolution and time constant were 0.3 cm21 and 3 s, respectively. The system was interfaced with a computer.X-ray diffractometry X-ray powder diffraction analysis was performed with a Philips (Eindhoven, The Netherlands, Model 1830/40 instrument) on finely powdered samples using Cu Ka radiation (40 kV and 30 mA) and an Ni filter with a scanning speed of 0.005° 2q s21. The time constant was set at 2 s. Results and Discussion Theory for Construction of Raman Calibration Curve The Raman spectra of calcite, aragonite and gypsum are shown in Fig. 1. The characteristic Raman bands due to the lattice vibration mode at 280, 205 and 412 cm21 for calcite, aragonite and gypsum, respectively, and the n1 internal mode (symmetric stretching) at 1085 cm21 for both calcite and aragonite16,17 and at 1006 cm21 for gypsum16 are easily distinguished.The spectra are in good agreement with earlier reports.18 It is obvious that the Raman bands in the spectral region between 170 and 450 cm21 should be used for the quantitative analysis of a mixture of calcite, aragonite and gypsum since the strong peak at 1085 cm21, which is attributed to the symmetric C–O stretching, is common for all calcium carbonate phases.On the other hand, when the specimen contains only one CaCO3 phase and gypsum, the 1085 and 1006 cm21 bands can also be used for quantitative analysis. In this case a lower detection limit is expected, since the analysis is based on stronger and sharper peaks. In the present work both possibilities were addressed by constructing Raman calibration curves using the bands of the spectral regions of interest, i.e., 170–450 and 970–1110 cm21.The purpose was to find an easy and reliable method for calculating each ingredient’s percentage and, therefore, peak heights were used and not integrated intensities of the bands. The intensity of a Raman line depends on a number of factors, including incident laser power, frequency of scattered radiation, absorptivity of the materials involved in the scattering and the response of the detection system.Thus, the measured Raman intensity, I(n), can be represented by19 I(n) = I0 K(n) C (1) where I0 is the intensity of the exciting laser radiation, n is the Raman shift, K(n) is a factor which includes the frequencydependent terms (the over-all spectrophotometer response, the self-absorption of the medium and the molecular scattering properties) and C is the concentration of the Raman-active species. Differences in the measured intensities of the various spectra obtained from the same specimen, which are attributed to factors such as the intensity variation of I0 or the positioning of the sample, suggested the use of relative factors within each spectrum, e.g., ratio of band intensities characteristic for each component.Thus, in a spectrum obtained from a sample in which several species are present, the ratio of the intensities of two peaks attributed to different compounds should be given by IA IB = KA KB ¥ xA xB (2) where the subscripts A and B indicate the different components and xA/xB is the molar fraction ratio of the two species.Eqn. (2) is valid only when there is no chemical interaction between the substances present in the sample. From eqn. 2, it is apparent that a plot of IA/IB versus xA/xB should yield a straight line with slope KA/KB. In the frequency range of the spectral measurements, the over-all spectrometer response may be considered to be constant, hence the K ratios are dependent only on the scattering parameter associated with each band, assuming that no significant absorption of the exciting radiation occurs.Consequently, the values assigned to the K ratio may be used regardless of the Raman spectroscopic system, provided that 488 nm radiation from an argon laser is used. Raman Calibration Lines For an aragonite–calcite mixture, eqn. (2) may be rewritten for the 280 cm21 peak of calcite and the 205 cm21 peak of aragonite as follows: Ic 280 Ia 205 = Kc 280 Ka 205 ¥ xc xa (3) where the subscripts c and a represent the calcite and the aragonite crystal phases, respectively, and the superscripts 280 and 205 are the wavenumbers of the respective Raman bands.The plot of Ic 280/Ia 205 versus xc/xa is shown in Fig. 2. The Fig. 1 Raman spectra of the synthetically prepared A, calcite, B, aragonite and C, gypsum. 34 Analyst, January 1997, Vol. 122equation for the calibration line was obtained by linear regression of the experimental data: Ic 280 Ia 205 = 1.646 ¥ xc xa (4) The correlation coefficient (r), was 0.999 997 and the standard deviation (s), for the slope and the intercept were 1.4 3 1023 and 4.9 3 1022, respectively.Assuming that xa + xc = 1, the RSD for xa = 0.4 was found to be 3.3%, and the detection limits (DL) for calcite and aragonite were calculated to be of the order of 0.3 and 0.5%, respectively. Similarly, calibration lines for calcite–gypsum and aragonite –gypsum mixtures were constructed using the Raman bands at 280 and 1085 cm21 for calcite, 412 and 1006 cm21 for gypsum and 205 and 1085 cm21 for aragonite.Typical spectra of the various mixtures are shown in Fig. 3. The cumulative information for the RS calibration lines appear in Table 1. Calculation of the Molar Fraction of Calcite, Aragonite and Gypsum in a Sample Using the Raman Calibration Curves Two-component systems The gypsum–calcite and the gypsum–aragonite mixtures belong to this category.The stronger Raman bands at 1085 cm21 for either the calcite or aragonite phases and at 1006 cm21 for gypsum can be used [eqn. (7) and (8) in Table 1]. For the calcite–gypsum mixtures and assuming that xg + xc = 1, eqn. (7) can be transformed to xg = 0.721Ig 1006 Ic 1085 + 0.721Ig 1006 (9) Similarly, for the aragonite–gypsum system, eqn. (8) can be rewritten as xg = 0.748Ig 1006 Ia 1085 + 0.748Ig 1006 (10) Three-component systems This case applies to sulfated calcareous rocks in which both calcium carbonate phases are present together with gypsum.The calibration lines used in the two-component system case cannot be used since both aragonite and calcite exhibit a strong peak at the same frequency, 1085 cm21. Eqns. (4), (5) and (6) in Table 1 were used instead. Assuming that xa + xg + xc = 1, Fig. 2 Raman calibration line for calcite–aragonite mixtures. Table 1 Cumulative results for the RS calibration curves Eqn. s of s of No.Calibration line r slope intercept RSD (%) DL (mol%) I c 280 xc 3.3 Calcite 0.3 (4) = 1.646 3 0.999 997 1.4 3 1023 4.9 3 1022 (xa = 0.4) Aragonite 0.5 I a 205 xa I c 280 xc 3.3 Calcite 0.3 (5) = 1.912 3 0.999 94 8.7 3 1023 6 3 1022 (xg = 0.4) Gypsum 0.6 Ig 412 xg I a 205 xa 3.0 Aragonite 0.5 (6) = 1.171 3 0.999 08 2.3 3 1022 3.7 3 1022 (xg = 0.4) Gypsum 0.6 I g 412 xg I c 1085 xc 2.0 Calcite 0.1 (7) = 0.721 3 0.999 97 2.2 3 1023 1.4 3 1022 (xg = 0.4) Gypsum 0.05 I g 1006 xg I a 1085 xa 1.7 Aragonite 0.1 (8) = 0.748 3 0.999 99 1.7 3 1023 1.1 3 1022 (xg = 0.4) Gypsum 0.05 I g 1006 xg Fig. 3 Raman spectra of: A, 20 mol% calcite–80 mol% gypsum; B, 80 mol% calcite–20 mol% gypsum; C, sulfated marble sample; D, 20 mol% aragonite–80 mol% gypsum; E, 80 mol% aragonite–20 mol% gypsum; and F, 40 mol% calcite–30 mol% aragonite–30 mol% gypsum.Analyst, January 1997, Vol. 122 35the molar fractions in a sample may be determined from the following relationships: xa = 1.646Ia 205 Ic 280 +1.646Ia 205 +1.927Ig 412 (11) xc = Ic 280 1.646Ia 205 ¥ xa (12) xg = 1.171Ig 412 Ia 205 ¥ xa (13) The validity of these expressions was tested on the spectrum recorded from a powder mixture consisting of 40 mol% calcite, 30 mol% aragonite and 30 mol% gypsum [Fig. 3(F)]. The results were calcite 39.7, aragonite 30.9 and gypsum 29.4 mol%. The deviation of the results obtained was within the experimental error (s in Table 1). It should be noted that the analytical methodology presented here does not depend on the simultaneous existence of the calcium carbonate phases and the gypsum, since there is neither chemical interaction between these species nor overlap of the corresponding bands in the Raman spectra.As a result, the relative intensities used in the analysis are not affected. Moreover, as may be seen from eqn. (1), the intensity of the Raman bands depends on the concentration of the investigated species alone. If an additional compound, besides gypsum, aragonite and/or calcite, is also present and provided that this does not contribute to the RS signal at the proposed frequencies and that there is no chemical interaction among the species present, eqns.(4)–(8) are still valid and the ratio of the gypsum to the calcium carbonate phases can be determined. Theory for Construction of XRD Calibration Curve The XRD spectra of calcite, aragonite and gypsum are shown in Fig. 4. The calcite spectrum exhibits two major peaks associated with the 104 and 113 reflections.Unfortunately, the former coincides with a gypsum peak, so only the intensity of the 113 reflection was used. The peaks attributed to the 121 – and 111 reflections of gypsum and aragonite, respectively, were also used for the quantitative analysis. If the sample is a uniform mixture of two components and extinction and microabsorption effects are neglected, it can be shown that20 IA IB = L ¥ xA xB (14) where L is a proportionality constant which depends on the component, the diffraction line and the mass absorption coefficient of the species present.IA/IB represents the ratio of the intensities of two selected diffraction lines in a mixture of two substances, and xA/xB is the molar fraction ratio of the two substances. A plot of IA/IB versus xA/xB should yield a straight line with an intercept of zero. XRD Calibration Lines Calibration curves for calcite–aragonite, aragonite–gypsum and gypsum–calcite binary mixtures were constructed.For gypsum–aragonite mixtures the plot of Ia 111/Ig 121– versus xa/xg, where the subscripts a and g represent gypsum and aragonite, respectively, and the superscripts the XRD reflections, as shown in Fig. 5. The equation for the calibration line was obtained by linear regression of the experimental data: Ia 111 Ig 121 = 0.18 ¥ xa xg (15) The correlation coefficient (r) was 0.9993 and the s for the slope and the intercept were 3 3 1022 and 2.3 3 1022, respectively.Assuming that xa + xg = 1, the RSD for xg = 0.4 was found to be 12.2%, and DL for gypsum and aragonite were calculated to be of the order of 1–2 and 5%, respectively. Similarly, calibration lines were constructed for calcite– gypsum and aragonite–calcite mixtures using the XRD reflection peaks of 113, 121 – and 111 for calcite, gypsum and aragonite, respectively. Representative spectra are shown in Fig. 6. The cumulative information for the XRD calibration lines is given in Table 2.Calculation of the Molar Fraction of Calcite, Aragonite and Gypsum in a Ternary Sample using the XRD Calibration Curves Assuming that xa + xg + xc = 1 and by employing eqns. (15), (16) and (17) in Table 2, the molar fractions in a sample in which calcite, aragonite and gypsum are present can be determined using the following relationships: xa = 1.13Ia 111 1.13Ia 111 + Ic 113 + 0.20Ig 121 (18) Fig. 4 XRD spectra of the synthetically prepared A, calcite, B, aragonite and C, gypsum.Fig. 5 XRD calibration line for gypsum–aragonite mixtures. 36 Analyst, January 1997, Vol. 122xc = Ic 113 1.13Ia 111 ¥ xa (19) xg = 0.18Ig 121 Ia 111 ¥ xa (20) The validity of the expressions derived for the XRD technique was also tested on a spectrum recorded from the same ternary powder mixture as used for testing the Raman calibration curves, Fig. 6(C). The percentages determined were calcite 41.9, aragonite 30.8 and gypsum 27.3 mol%.The deviation of the results obtained was within the experimental error (s in Table 2). As mentioned for the RS method, the presence of a fourth compound does not affect the validity of the derived equations provided that there is no chemical interaction among the species present or overlap of the XRD peaks used for the analysis with the XRD lines of the additional species. Application to a Sulfated Marble Sample A marble sample taken from Athens National Garden was tested for gypsum using the techniques described here.Raman spectra were excited from several points of the marble surface. Only calcite and gypsum were present [Fig. 3(C)]. Application of eqn. (9) yielded gypsum concentrations between 0 and 8.5 mol%. An external layer from sample surface was removed mechanically and the XRD spectrum was recorded [Fig. 6(D)]. No detectable gypsum was found. This result was expected because only the average percentage from the removed surface layer can be observed with the XRD technique, and the possibility of having material from the inner marble layer, consisting of pure calcite only, in the powder collected from the surface is not trivial.Hence the percentage of gypsum in the XRD-tested material was below the detection limit. Comparison Between RS and XRD Both techniques were used for the determination of the percentage of gypsum on a calcium carbonate surface, and RS exhibited certain advantages over the XRD method: (a) RS was non-destructive for the sample and less time consuming; (b) reliable point-by-point analysis (‘mapping’) of the surface was accomplished using RS, whereas XRD yielded only the average percentage of the bulk, ground powder sample; and (c) from the comparison of the calibration line statistics (Tables 1 and 2) it can be seen that RS exhibited lower SD and lower DL than the XRD calibration curves.Conclusions Methods based on RS and XRD for quantitative determination of the transformation of the surface of monuments into gypsum, known as marble deterioration, were developed.Calibration curves from mixtures of calcite and aragonite (the most stable phases of calcium carbonate), and gypsum were constructed. The much lower detection limits given by RS, the fact that it is non-destructive and the potential use of the technique for chemical mapping of the marble surface are among the major advantages of the RS over powder XRD.The authors are indebted to Professor G.N. Papatheodorou for helpful suggestions and for providing the experimental facilities. They thank Dr. P. Klepetsanis for kindly supplying the gypsum powder. Partial support of this work by the GSRT EPET II Program (contract No. 368/11-1-95) is gratefully acknowledged. References 1 Camuffo, D., Del Monte, M., Sabbioni, C., and Vittori, O., Atmos. Environ., 1982, 16, 2253. 2 Ross, M., McGee, E. S., and Ross, D. R., Am. Mineral., 1989, 74, 367. 3 Skoulikidis, T., and Charalambous, D., Br. Corros. J., 1981, 16, 70. 4 Johanson, L. G., Lindqvist, O., and Mangio, R., Durability Build. Mater., 1988, 5, 439. 5 Gauri, K. L., Chowdhury, A. N., Kulshreshta, N. P., and Punuru, A. R., Stud. Conserv., 1989, 34, 201. 6 Verges-Belmin, V., Atmos. Environ., 1994, 28, 295. 7 Van Houte, G., Rodrique, L., Genet, M., and Delmon, B., Environ. Sci. Technol., 1981, 15, 327. 8 Lipfert, F. W., Atmos. Environ., 1989, 23, 415. 9 Gauri, K. Lal, and Holdren G. C., Jr., Environ. Sci. Technol., 1981, 15, 386. 10 Berner, R. A., Am. J. Sci., 1966, 264, 1. Table 2 Cumulative results for the XRD calibration lines Eqn. s of s of No. Calibration line r slope intercept RSD (%) DL (mol%) Ia 111 xa 12.2 Aragonite 5 (15) = 0.18 3 0.9993 3 3 1022 2.3 3 1022 (xg = 0.4) Gypsum 1–2 Ig 121 – xg Ic 113 xc 13.3 Calcite 4 (16) = 0.18 3 0.9853 1.5 3 1022 2.7 3 1022 (xg = 0.4) Gypsum 1–2 Ig 121 – xg Ic 113 xc 11.2 Calcite 4 (17) = 1.13 3 0.9991 2.4 3 1022 4.2 3 1022 (xa = 0.4) Aragonite 5 Ia 111 xa Fig. 6 XRD spectra of: A, 20 mol% calcite–80 mol% gypsum; B, 80 mol% calcite–20 mol% gypsum; C, 40 mol% calcite–30 mol% aragonite– 30 mol% gypsum; and D, sulfated marble sample. Analyst, January 1997, Vol. 122 3711 Hacker, B. R., Kirby, S. H., and Bohlen, S. R., Science, 1992, 258, 110. 12 Compere, E. L., and Bates, J. M., Limnol. Oceanogr., 1973, 18, 326. 13 Xyla, A., and Koutsoukos, P. G., J. Chem. Soc., Faraday Trans. 1, 1989, 85, 3165. 14 Herman, R. G., Bogdan, C. E., Sommer, A. J., and Simpson, D. R., Appl. Spectrosc., 1987, 41, 437. 15 Silk, S. T., and Lewin, S. Z., Adv. X-ray Anal., 1971, 14, 29. 16 Griffith, P. G., in Spectroscopy of Inorganic-based Materials, eds. Clark, R. J. H., and Hester, R. E., Wiley, Chichester, 1987, pp. 137 and 151. 17 Behrens, G., Kuhn, L. T., Ubic R., and Heuer, A. H., Spectrosc. Lett., 1995, 28, 983. 18 Degen, A., and Newman, G. A., Spectrochim. Acta, Part A, 1993, 49, 859. 19 Strommen, D., and Nakamoto, K., in Laboratory Raman Spectroscopy, Wiley, New York, 1984, pp. 71–75. 20 Whiston, C., in X-ray Methods, Wiley, New York, 1987, p. 113. Paper 6/06167B Received September 9, 1996 Accepted September 24, 1996 38 Analyst, January 1997, Vol. 122

ISSN:0003-2654    

DOI:10.1039/a606167b

出版商:RSC    

年代:1997

数据来源: RSC

摘要:

1 3 5 7 9 11 Cell path length ratio Relative error in the relative SRE level 0.1 0.2 0.3 0.4 0.5 0.00005 0.00005 0.001 0.01 0.05 0.005 0.00001 Error Analysis of the Transmittance Ratio Stray Radiant Energy Test Method in Double-beam Ratio Recording Spectrophotometers Paddy Fleminga and Adrienne Flemingb a Regional Technical College, Sligo, Ireland b Limerick University, Limerick, Ireland The photometric error contributions which the relative errors in determining both the cell pathlength ratio and the transmittance ratio minimum make towards the relative error in determining the relative stray radiant energy (SRE) levels in a ratio recording double-beam spectrophotometer were investigated, thus enabling optimum experimental conditions for SRE determination to be pre-selected. Keywords: Transmittance ratio; stray radiant energy; error analysis; ratio recording spectrophotometer An experimental method for determining an instrumental specification is of uncertain merit unless it has been fully error analysed.The Fleming1,2 and Mielenz et al.3 transmittance ratio spectrophotometric methods for determining the relative stray radiant energy (SRE) level, s, in ratio recording double-beam spectrophotometers involve photometric measurements which are subject to error (cf., Gridgeman).4 Inspection of the equation s = (a-1)(r / a)a/(a-1) 1-r = (a-1)ta 1-ata-1 (1) which relates s to the transmittance ratio minimum, r, to the reference beam’s monochromatic transmittance value, t, where the transmittance ratio is a minimum, and to the cell pathlength ratio, a, suggests that there is a connection between the photometric error and the error in determining s.Remember that r = ata21 and that a, which Burgess and Knowles5 implied may be determined photometrically to a precision of 4%, also contributes to the error in determining the SRE level. This paper investigates the photometric error contributions which the relative errors in determining both the cell pathlength ratio, Da/a, and the transmittance ratio minimum, Dr/r, make towards the relative error in determining the relative SRE levels, Ds/s, in a double-beam spectrophotometer, thus enabling optimum experimental conditions to be pre-selected when using the transmittance ratio SRE determination method.Formulation of Experimental Quantities The dependence of the relative error in determining s on the relative errors in r and a may be derived from eqn.(1) via the equation ln s = ln(a 2 1) + [a/(a 2 1)]ln(r/a) 2 ln(1 2 r) (2) Since r = ata21, eqn. (2) becomes ln s = ln(a 2 1) + aln t 2 ln(1 2 ata21) (3) Differentiating eqn. (3) gives Ds/s = A[B(Da/a) + C(Dt/t)] (4) where A = [(a 2 1)(a21 2 ta21)]21 B = 1 + (a 2 1)ln t 2 ta21 C = (a 2 1)(1 2 ta21) Skoog and Leary6 gave the photometric error in a doublebeam ratio recording spectrophotometer, DT, as k(T + T2)1/2, where k is a constant for a given spectrophotometer.The Perkin- Elmer 551S spectrophotometer specification gives DT = 0.0012 at T = 0.1, yielding k = 0.0036. Therefore, the relative error in t is given by Dt/t = 0.0036(1 + t21)1/2 (5) The relative error function in eqn. (5) has values in the range 0.012 < Dt/t < 0.114 for 0.1 > t > 0.001. If the pathlength ratio is determined spectrophotometrically,5 then its relative error, Da/a, is dependent on the photometric error, DT, for the particular spectrophotometer employed.It may be shown for the Perkin-Elmer 551S instrument that Da/a = (0.0036/ln Tr)[a22(1 + Tr2a)2 + (1 + Tr21)2]1/2 (6) where Tr is the monochromatic transmittance of a dilute sample placed in the reference cell so as to compare its thickness with that of the sample cell and hence to calculate a. The relative error function in eqn. (6) has values in the range 0.049 > Da/a > 0.040 for 1.1 < a < 10 and Tr = 0.8. Note that the dilute solution in question is used solely to establish spectrophotometrically the experimental cell pathlength ratio rather than relying on the nominal pathlengths marked on the cells.Fig. 1 Relative error in the relative stray radiant energy level, Ds/s, plotted against the cell pathlength ratio, a, for seven distinct relative stray radiant energy levels, s = 0.05, 0.01, 0.005, 0.001, 0.0005, 0.0001 and 0.000 05. Analyst, January 1997, Vol. 122 (39–40) 3910–5 10–5 10–4 10–3 10–2 10–1 10–0 10–4 10–3 10–2 10–1 a = 1.25 a =10 Relative stray radiant energy level Transmittance ratio minimum The root mean square relative error, (Ds/s)rms, is given by (Ds/s)rms = A[B2 (Da/a)2 + C(Dt/t)2]1/2 (7) Analysis and Conclusion Eqn.(7) was solved for relative SRE levels in the range 0.05 > s > 0.00005, for cell pathlength ratios in the range 1.1 < a < 10 and for Tr = 0.8. Fig. 1 shows the relative error function as set out in eqn. (7), (Ds/D)rms, plotted against the cell pathlength ratio, a, for seven distinct relative SRE levels.The relative error function increases with increasing cell pathlength ratio for a given relative SRE level, and increases with decreasing relative SRE levels for a given cell pathlength ratio. Sample cells come in standard sizes of 1, 2, 5, 10, 20, 50 and 100 mm. Considering the dependence of the error function on the cell pathlength ratio, we recommend a = 2 as it may be selected in four independent ways from the above cell size set and it also makes eqn.(1) easy to calculate since r = 2t and s = r2/4(1 2 r) = t2/(1 2 2t). Burgess7 has shown graphically the dependence of the transmittance ratio minimum on the reference beam’s monochromatic absorbance ( = 2log10t) in the relative SRE range 0.1 > s > 0.00005 and for a = 1.25. Fig. 2 shows the corresponding graphs for a = 1.25, 1.5, 2, 2.5, 4, 5 and 10, which may be utilized to calculate readily the relative SRE level once the transmittance ratio minimum has been determined.An a priori knowledge of the approximate relative SRE level in a spectrophotometer, e.g., gleaned from its specification sheet, will suggest the monochromatic absorbances of a narrow range of reference beam solutions which need to be prepared for a given a so as to make possible a rapid experimental determination of r. Fig. 2 allows the rapid calculation of the reference beam’s monochromatic absorbance from the transmittance ratio minimum at any given relative SRE level and a value.Spectrophotometers are usually operated in the absorbance mode rather than in the transmittance mode and therefore differential absorbance rather than transmittance ratio is observed in this application. Differential absorbance is a slowly varying function of the monochromatic reference absorbance at and near the differential absorbance maxima, i.e., it is ‘flat-topped.’ References 1 Fleming, P., Analyst, 1990, 115, 375. 2 Fleming, P., Analyst, 1991, 116, 909. 3 Mielenz, K. D., Weidner, V. R., and Burke, R. W., Appl. Opt., 1982, 21, 3354. 4 Gridgeman, N. T., Anal. Chem., 1952, 24, 445. 5 Techniques in Visible and Ultraviolet Spectrometry, Vol. 1, Standards in Absorption Spectrometry, ed. Burgess, C., and Knowles, A., Chapman and Hall, London, 1st Edn., 1981, p. 127. 6 Skoog, D. A., and Leary, J. J., Principles of Instrumental Analysis, Saunders, New York, 4th edn., 1992, p. 132. 7 Burgess, C., in Encyclopaedia of Analytical Science, ed.Townshend, A., Academic Press, London, 1995, pp. 3643–3647. Paper 6/05825F Received August 21, 1996 Accepted October 22, 1996 Fig. 2 Transmittance ratio minimum versus relative SRE level plotted on log–log axes for a = 1.25, 1.5, 2.0, 2.5, 4, 5 and 10. 40 Analyst, January 1997, Vol. 122 1 3 5 7 9 11 Cell path length ratio Relative error in the relative SRE level 0.1 0.2 0.3 0.4 0.5 0.00005 0.00005 0.001 0.01 0.05 0.005 0.00001 Error Analysis of the Transmittance Ratio Stray Radiant Energy Test Method in Double-beam Ratio Recording Spectrophotometers Paddy Fleminga and Adrienne Flemingb a Regional Technical College, Sligo, Ireland b Limerick University, Limerick, Ireland The photometric error contributions which the relative errors in determining both the cell pathlength ratio and the transmittance ratio minimum make towards the relative error in determining the relative stray radiant energy (SRE) levels in a ratio recording double-beam spectrophotometer were investigated, thus enabling optimum experimental conditions for SRE determination to be pre-selected. Keywords: Transmittance ratio; stray radiant energy; error analysis; ratio recording spectrophotometer An experimental method for determining an instrumental specification is of uncertain merit unless it has been fully error analysed. The Fleming1,2 and Mielenz et al.3 transmittance ratio spectrophotometric methods for determining the relative stray radiant energy (SRE) level, s, in ratio recording double-beam spectrophotometers involve photometric measurements which are subject to error (cf., Gridgeman).4 Inspection of the equation s = (a-1)(r / a)a/(a-1) 1-r = (a-1)ta 1-ata-1 (1) which relates s to the transmittance ratio minimum, r, to the reference beam’s monochromatic transmittance value, t, where the transmittance ratio is a minimum, and to the cell pathlength ratio, a, suggests that there is a connection between the photometric error and the error in determining s.Remember that r = ata21 and that a, which Burgess and Knowles5 implied may be determined photometrically to a precision of 4%, also contributes to the error in determining the SRE level. This paper investigates the photometric error contributions which the relative errors in determining both the cell pathlength ratio, Da/a, and the transmittance ratio minimum, Dr/r, make towards the relative error in determining the relative SRE levels, Ds/s, in a double-beam spectrophotometer, thus enabling optimum experimental conditions to be pre-selected when using the transmittance ratio SRE determination method.Formulation of Experimental Quantities The dependence of the relative error in determining s on the relative errors in r and a may be derived from eqn. (1) via the equation ln s = ln(a 2 1) + [a/(a 2 1)]ln(r/a) 2 ln(1 2 r) (2) Since r = ata21, eqn. (2) becomes ln s = ln(a 2 1) + aln t 2 ln(1 2 ata21) (3) Differentiating eqn.(3) gives Ds/s = A[B(Da/a) + C(Dt/t)] (4) where A = [(a 2 1)(a21 2 ta21)]21 B = 1 + (a 2 1)ln t 2 ta21 C = (a 2 1)(1 2 ta21) Skoog and Leary6 gave the photometric error in a doublebeam ratio recording spectrophotometer, DT, as k(T + T2)1/2, where k is a constant for a given spectrophotometer. The Perkin- Elmer 551S spectrophotometer specification gives DT = 0.0012 at T = 0.1, yielding k = 0.0036. Therefore, the relative error in t is given by Dt/t = 0.0036(1 + t21)1/2 (5) The relative error function in eqn.(5) has values in the range 0.012 < Dt/t < 0.114 for 0.1 > t > 0.001. If the pathlength ratio is determined spectrophotometrically,5 then its relative error, Da/a, is dependent on the photometric error, DT, for the particular spectrophotometer employed. It may be shown for the Perkin-Elmer 551S instrument that Da/a = (0.0036/ln Tr)[a22(1 + Tr2a)2 + (1 + Tr21)2]1/2 (6) where Tr is the monochromatic transmittance of a dilute sample placed in the reference cell so as to compare its thickness with that of the sample cell and hence to calculate a.The relative error function in eqn. (6) has values in the range 0.049 > Da/a > 0.040 for 1.1 < a < 10 and Tr = 0.8. Note that the dilute solution in question is used solely to establish spectrophotometrically the experimental cell pathlength ratio rather than relying on the nominal pathlengths marked on the cells.Fig. 1 Relative error in the relative stray radiant energy level, Ds/s, plotted against the cell pathlength ratio, a, for seven distinct relative stray radiant energy levels, s = 0.05, 0.01, 0.005, 0.001, 0.0005, 0.0001 and 0.000 05. Analyst, January 1997, Vol. 122 (39–40) 3910–5 10–5 10–4 10–3 10–2 10–1 10–0 10–4 10–3 10–2 10–1 a = 1.25 a =10 Relative stray radiant energy level Transmittance ratio minimum The root mean square relative error, (Ds/s)rms, is given by (Ds/s)rms = A[B2 (Da/a)2 + C(Dt/t)2]1/2 (7) Analysis and Conclusion Eqn.(7) was solved for relative SRE levels in the range 0.05 > s > 0.00005, for cell pathlength ratios in the range 1.1 < a < 10 and for Tr = 0.8. Fig. 1 shows the relative error function as set out in eqn. (7), (Ds/D)rms, plotted against the cell pathlength ratio, a, for seven distinct relative SRE levels. The relative error function increases with increasing cell pathlength ratio for a given relative SRE level, and increases with decreasing relative SRE levels for a given cell pathlength ratio.Sample cells come in standard sizes of 1, 2, 5, 10, 20, 50 and 100 mm. Considering the dependence of the error function on the cell pathlength ratio, we recommend a = 2 as it may be selected in four independent ways from the above cell size set and it also makes eqn. (1) easy to calculate since r = 2t and s = r2/4(1 2 r) = t2/(1 2 2t). Burgess7 has shown graphically the dependence of the transmittance ratio minimum on the reference beam’s monochromatic absorbance ( = 2log10t) in the relative SRE range 0.1 > s > 0.00005 and for a = 1.25.Fig. 2 shows the corresponding graphs for a = 1.25, 1.5, 2, 2.5, 4, 5 and 10, which may be utilized to calculate readily the relative SRE level once the transmittance ratio minimum has been determined. An a priori knowledge of the approximate relative SRE level in a spectrophotometer, e.g., gleaned from its specification sheet, will suggest the monochromatic absorbances of a narrow range of reference beam solutions which need to be prepared for a given a so as to make possible a rapid experimental determination of r.Fig. 2 allows the rapid calculation of the reference beam’s monochromatic absorbance from the transmittance ratio minimum at any given relative SRE level and a value. Spectrophotometers are usually operated in the absorbance mode rather than in the transmittance mode and therefore differential absorbance rather than transmittance ratio is observed in this application. Differential absorbance is a slowly varying function of the monochromatic reference absorbance at and near the differential absorbance maxima, i.e., it is ‘flat-topped.’ References 1 Fleming, P., Analyst, 1990, 115, 375. 2 Fleming, P., Analyst, 1991, 116, 909. 3 Mielenz, K. D., Weidner, V. R., and Burke, R. W., Appl. Opt., 1982, 21, 3354. 4 Gridgeman, N. T., Anal. Chem., 1952, 24, 445. 5 Techniques in Visible and Ultraviolet Spectrometry, Vol. 1, Standards in Absorption Spectrometry, ed. Burgess, C., and Knowles, A., Chapman and Hall, London, 1st Edn., 1981, p. 127. 6 Skoog, D. A., and Leary, J. J., Principles of Instrumental Analysis, Saunders, New York, 4th edn., 1992, p. 132. 7 Burgess, C., in Encyclopaedia of Analytical Science, ed. Townshend, A., Academic Press, London, 1995, pp. 3643–3647. Paper 6/05825F Received August 21, 1996 Accepted October 22, 1996 Fig. 2 Transmittance ratio minimum versus relative SRE level plotted on log–log axes for a = 1.25, 1.5, 2.0, 2.5, 4, 5 and 10. 40 Analyst, January 1997, Vol. 122

ISSN:0003-2654    

DOI:10.1039/a605825f

出版商:RSC    

年代:1997

数据来源: RSC

摘要:

Simultaneous Determination of Ethinylestradiol and Levonorgestrel in Oral Contraceptives by Derivative Spectrophotometry Juan J. Berzas*, Juana Rodr�ýguez and Gregorio Casta�neda Departamento de Qu�ýmica Anal�ýtica y Tecnolog�ýa de Alimentos, Facultad de Ciencias, Universidad de Castilla-La Mancha, 13071 Ciudad Real, Spain A method for determining ethinylestradiol (ETE) and levonorgestrel (LEV) in mixtures by first-derivative spectrophotometry is described. The procedure does not require any separation step.Measurements are made at the zero-crossing wavelengths and the calibration graphs are linear up to 26 and 33 mg ml21 of ETE and LEV, respectively. The method was applied to the determination of both compounds in five different Spanish commercial low-dose oral contraceptives. Similar results were obtained by an HPLC method. Keywords: Ethinylestradiol; levonorgestrel; derivative spectrophotometry; oral contraceptives At present there are three types of oral contraception available.In the sequential type, estrogen is administered alone for the first week, followed by a lower dosage of the estrogen in conjunction with a progestogen for the remainder of the course. In the second, commonly used, type both an estrogen and a progestogen are present in the tablets (as either a single dose or in three different doses). In the progestogen type, a progestogen alone is administered. Ethinylestradiol (ETE) is a semisynthetic estrogen female sex hormone and levonorgestrel (LEV) is a synthetic steroid with an extremely potent progestational action. The formulation of these steroids in tablets of low dosage, i.e., 30–250 mg per tablet, presented a challenging analytical problem.A sensitive, accurate and rapid procedure is desirable for content uniformity testing of the dosage form. The structure of LEV has a characteristic D4-3-keto group in the A-ring with a different chromophoric power to ETE. The most commonly encountered estrogen is ETE, which is present at a very low dosage level (30–100 mg per tablet) in combination with an orally active synthetic progestin (one of the most commonly used is LEV), which is present at a level of from 5 to 30 times that of the estrogen.Oral contraceptives have had an enormous positive impact on public health for the past three decades and, in the main, there has been a remarkably low incidence of troublesome side-effects. Although estrogens are implicated in an increased incidence of breast and endometrial cancer, epidemiological studies have not provided convincing evidence to support a direct correlation between the use of oral contraceptives and an increase in breast cancer. The modern low-dose oral contraceptives (estrogen–progestogen) require a sensitive analysis method which is unaffected by the small amount of the estrogen and the large excess of progestogen.There have been several reports1–11 of the determination of levonorgestrel or ethinylestradiol, including the use of radioactively labelled derivatives,1,2 dansyl or other fluorescent derivatives,3–5 spectrophotometry or photometry6–9 and gel or column chromatography, but the methods are complicated.10,11 No references were found to the simultaneous determination of ETE and LEV using spectrophotometric methods.The determination of ETE in the presence of noresthisterone by derivative spectrophotometry in methanol or ethanol has been reported,12,13 the recoveries in different tablets being 120–80% for both compounds.The simultaneous determination of ETE and mestranol by derivative spectrophotometry has been proposed, either in a solution of methanol and chloroform14 or in NaOH in methanol.15 In derivative UV/VIS spectrophotometry, the information contained in the spectrum is presented in a potentially more useful form, greatly increasing the versatility of the technique16 –18 and offering a convenient solution to a number of well defined analytical problems, such as the resolution of multi-component systems, removal of sample turbidity, matrix background and enhancement of spectral details.19 Although the use of derivative spectra is not new, it has only become practical in recent years with the development of microcomputer technology, which allows the almost instantaneous generation of derivative spectra.In this paper we demonstrate the ease with which the derivative methods circumvent the problem of overlapping spectral bands and sample turbidity, allowing the simultaneous determination of ETE and LEV without prior separation.The method yielded accurate, rapid and reproducible results for five different commercial products, two of them with three different dosages. The results obtained by the proposed method were compared with those obtained by HPLC with spectrophometric detection, very similar to the method proposed in the US Pharmacopeia. 20 Experimental Apparatus A Beckman (Fullerton, CA, USA) DU-70 spectrophotometer equipped with 1.0 cm quartz cells and connected to an IBM-PS 2 Model 30 computer, fitted with Beckman Data Leader software,21 and an Epson FX-850 printer was used for all absorbance measurements. A Shimadzu (Kyoto, Japan) high-performance liquid chromatograph equipped with a Nova-Pak C18 column (15 3 0.39 cm id, 4 mm), a diode-array detector, a Rheodyne injection valve and connected to a computer fitted with CLASS LC-10 software was used.A Crison (Barcelona, Spain) MicropH 2002 pH meter was used for the pH measurements. Standard Solutions All chemicals and solvents were of analytical-reagent grade. ETE and LEV were obtained from Sigma (St. Louis, MO, USA) and stock standard solutions were prepared in absolute ethanol Analyst, January 1997, Vol. 122 (41–44) 41(100 mg ml21). The purities of the ETE and LEV reported by Sigma were 98.6% and 99.6%, respectively, determined using an HPLC method with spectrophotometric detection at 280 and 242 nm.Procedure Calibration Stock standard solutions of ETE and LEV were placed in 25 ml calibrated flasks to give final concentrations of up to 26 and 33 mg ml21, respectively, adding absolute ethanol to dilute the contents to 25 ml (the resulting final solution was 100% in ethanol). This high percentage of ethanol was necessary to obtain total dissolution of the drugs from the oral contraceptive tablets.The absorption spectra of the samples were recorded against an ethanol blank between 315 and 210 nm at a scan speed of 120 nm min21 and stored in the computer. Firstderivative spectra were obtained with Dl = 8 nm and ETE was determined by measuring the signal of the first derivative spectrum at 293.0 nm (1D293) (zero-crossing point for LEV), and by using an appropriate calibration graph, their concentrations could be determined. These calibrations were performed by varying the concentration of the estrogen, in the absence of the other hormone.The LEV content was also determined by measuring the signal at 249.0 nm (1D249) (zero-crossing point for ETE). These spectra were not treated with a smoothing function because the noise level was low. Assay of pharmaceutical preparations Twenty tablets were finely powdered and an appropriate portion (equivalent to the median mass of two tablets) was dissolved in 8 ml of absolute ethanol by sonication for 15 min, followed by shaking by mechanical means for 20 min.The mixture was filtered, using a Swinnex polypropylene disc filter holder of 13 mm diameter (Millipore, Bedford, MA, USA) with an FH 0.5 mm Fluoropore (PFTE) membrane, into a 10 ml calibrated flask. The residue was washed twice with the same solvent and diluted to volume. The absorption spectra were recorded against an absolute ethanol blank and stored in the IBM-PS computer. For determining ETE and LEV, the absorption spectra were handled as in first-derivative spectrophotometry.Comparison with HPLC reference method Twenty tablets were finely powdered and an appropriate portion (equivalent to the median mass of two tablets) was dissolved in 8 ml of absolute ethanol by sonication for 15 min, followed by shaking by mechanical means for 20 min. The mixture was filtered using a Swinnex polypropylene disc filter holder of 1ter with an FH 0.5 mm Fluoropore (PFTE) membrane into a 10 ml calibrated flask.The residue washed twice with the same solvent and diluted to volume. A 2.5 ml portion of this solution was diluted with water in a 10 ml calibrated flask; the reason for this dilution is to obtain a higher polarity and lower concentration in the sample. HPLC determination was performed on a Nova-Pak C18 60 A column (15 3 0.39 cm id) containing 4 mm packing. The mobile phase was deaerated acetonitrile–methanol–water (3.5 + 1.5 + 4.5) and spectrophotometric detection was performed at 215 nm.The flow rate was about 1 ml min21.20 The differences between the two methods are the dissolution of the tablets (using ethanol–water or acetonitrile–methanol–water) and the column used [C18 (15 3 0.39 cm id) or C8 (15 3 0.46 cm id)]. Results and Discussion Method Development The influence of pH on the absorption spectra of ETE (e280 nm = 2253 l mol21 cm21 in absolute ethanol) and LEV (e240 nm = 17 155 l mol21 cm21 in absolute ethanol) was studied, with a total content of ethanol of 25%.The LEV spectrum showed only a maximum at 246 nm, which remained unchanging in the pH range 1.0–12.4. The ETE spectrum showed a maximum at 280 nm in the pH range 1.0–9.5 and for more alkaline solutions two different bands at 296 and 240 nm. The stability of ETE in very acidic solutions was not very good. For this reason, the best results for analytical purposes were obtained in the pH range 4.0–9.5. The preparation of the samples with 100% of ethanol resulted in spectra very close to those obtained at the optimum pH.Samples were prepared in absolute ethanol solution and the addition of a buffer solution was not necessary. Under these conditions, dilute solutions of ETE and LEV were stable for at least 12 h. The use of absolute ethanol permits the best recovery of the hormones in the oral contraceptive tablets. Derivative Spectrophotometry In Fig. 1 the zero-order spectra of ETE and LEV in the wavelength range 210–315 nm are shown.It can be seen that the absorption spectrum of LEV is very overlapped with the ETE spectrum. The determination of ETE directly could be easy at the start, but the small content of this steroid and the high content of LEV in commercial tablets (the ETE : LEV ratio is normally 1 : 4 or 1 : 5) presumes a large contribution of the spectrum of LEV to the maxima in the spectrum of ETE. The spectra of real samples after dissolution of the hormones and filtration show a very small overlap, not perceptible at the beginning, but the spectrum showed a small y-axis displacement of the absorbance owing to the overlap.This behaviour is revealed when the spectrum of an artificial binary mixture is compared with that of a real contraceptive tablet of similar concentration (Fig. 2). Derivative spectrophotometry is a suitable technique to overcome this problem. The zero-crossing method is the most common procedure for the preparation of analytical calibration graphs.In practice, the measurement selected is that which exhibits the best linear response, gives a zero or near zero intercept on the ordinate of the calibration graphs and is less affected by the concentration of any other component. The shape of the first derivative spectrum is adequate for determining ETE in the presence of LEV and vice versa. Fig. 3 shows the firstderivative absorption spectra of a solution of ETE and a solution of LEV, both solutions in 100% ethanol.It can be seen that owing to the overlapping spectra of these compounds, the zero-crossing method is the most appropriate approach for resolving mixtures of these compounds and it was used in this work with satisfactory results. Fig. 1 Absorption spectra of ethinylestradiol (26.01 mg ml21, broken line), levonorgestrel (7.17 mg ml21, dotted line) and their mixture (continuous line). 42 Analyst, January 1997, Vol. 122Preliminary experiments showed that the signals of the first derivative at 293.0 nm (working zero-crossing wavelength of LEV) are proportional to the ETE concentration and the signals of the first derivative at 249.0 nm (working zero-crossing wavelengths of ETE) are proportional to the LEV concentration.Selection of Optimum Instrumental Conditions The main instrumental parameters that affect the shape of the derivative spectra are the wavelength scanning speed, the wavelength increment over which the derivative is obtained (Dl) and the smoothing. These parameters need to be optimized to give a well resolved large peak, i.e., to give good selectivity and higher sensitivity in the determination. Generally, the noise level decreases with increase in Dl, thus decreasing the fluctuations in the derivative spectrum.However, if the value of Dl is too large, the spectral resolution is very poor. Therefore, the optimum value of Dl should be determined by taking into account the noise level, the resolution of the spectrum and the sample concentration.Some values of Dl were tested and 8.0 nm was selected as the optimum in order to obtain a satisfactory signal-to-noise ratio. In this way, a smoothing function was not necessary. Having established the experimental conditions, the calibration graphs were tested between 2.0 and 26.0 mg ml21 of the ETE in the absence of LEV at 293.0 nm for the first-derivative spectra. The calibration graphs were also tested between 4.0 and 30.0 mg ml21 of LEV in the absence of ETE at 243.0 nm for the first-derivative spectra (Fig. 4). Good linearity was observed in all cases. Statistical Study Tables 1 and 2 summarize the most characteristic statistical data obtained from the different calibration graphs and the reproducibility of the reagent blank and a standard. The reproducibility of particular concentrations of ETE (10.40 mg ml21) and LEV (10.60 mg ml21) were evaluated over 2 d by performing 10 absorption spectrophotometric measurements each day on 10 different samples. The results (Table 1) show that the repeatability for both hormones on each day was satisfactory.The comparison of the average concentrations with the Snedecor test did not show any significant difference at a confidence level of 5%. Determination of ETE and LEV in Synthetic Mixtures Some binary mixtures of ETE and LEV were prepared from the stock standard solutions in the proportions from 1 + 1 to 1 + 4 and were analysed by the proposed derivative spectrophotometric method.Some of these proportions between the two hormones were the same as in commercial contraceptives with the object of checking the relationships of more commercial interest. Table 3 shows the results of the analyses of different mixtures. The recoveries were between 94 and 104% for ETE and between 99 and 100% for LEV for the wavelengths studied. These results show that the method is effective for the simultaneous determination of ETE and LEV by first-derivative spectrophotometry.Fig. 2 (a) Absorption spectra of (A) synthetic mixture of ETE (5.00 mg ml21) and LEV (25.00 mg ml21) and (B) commercial solution of he Ovoplex pills, containing the same concentrations of ETE and LEV. (b) Zoom of the spectra between 270–315 nm. Fig. 3 First-derivative spectra (Dl = 8 nm) of ethinylestradiol (26.01 mg ml21, broken line), levonorgestrel (7.17 mg ml21, dotted line) and their mixture (continuous line).Fig. 4 First-derivative spectra (Dl = 8 nm) for different concentrations of levonorgestrel: a, 2.59; b, 5.18; c, 7.77; d, 10.36; e, 12.95; f, 19.42; g, 25.90; and h, 32.63 mg ml21. Table 1 Precision of the determination of the concentration of ETE (10.40 mg ml21) and LEV (10.60 mg ml21) on different days (n = 10 determinations on each day) Ethinylestradiol Levonorgestrel Average/ s/ RSD Average/ s/ RSD mg ml21 mg ml21 (%) mg ml21 mg ml21 (%) Day A 10.26 0.16 1.55 10.53 0.07 0.65 Day B 10.38 0.08 0.75 10.60 0.08 0.73 Analyst, January 1997, Vol. 122 43Determination of ETE and LEV in Commercial Contraceptives The Spanish pharmacological industry has at present five different low-dose commercial oral contraceptives (Neogynona, Microgynon, Ovoplex, Triciclor and Triagynon) containing ETE and LEV. Two of these contraceptives (Triciclor and Triagynon) have in the formulation three different doses (different proportions of ETE and LEV); in these contraceptives the amount of LEV starts at 0.050 mg per tablet at the beginning of the treatment, later it is 0.075 mg per tablet and at the end is 0.125 mg per tablet.Each oral contraceptive packet contains 21 tablets, six of them corresponding to the lowest dosage of LEV, five to the intermediate dosage and ten to the highest dosage. The amount of ETE starts at 0.03 mg per tablet, becomes 0.04 mg per tablet and at the end is once more 0.03 mg per tablet. The results obtained for the determination of ETE and LEV mixtures in commercial pharmaceuticals are given in Table 4.The relative differences between HPLC and derivative spectrophotometric results were between ±6% for the LEV determination in all oral contraceptives except Triagynon B and C. The relative differences for ETE were ±4% in all the commercial formulations except Triciclor C. The anomalous results found for Triagynon and Triciclor could be due to the presence of some interferences due to the excipients or of dyes that coated these tablets.In all cases the recoveries were calculated with respect to the results obtained by the HPLC method. Good correlations between the two methods were found. The indicated value is the mean of two different analyses of the same commercial batch. Conclusions The proposed derivative method is very suitable for the simultaneous determination of ETE and LEV and can be employed to analyse commercial formulations of low-dose oral contraceptives.The proposed method gave good results when compared with the HPLC method. The authors are grateful to Dr. V. Trigo (Wyeth-Orfi Laboratories) and Dr. C. Barona (Shering Laboratories). Financial support from the DGICYT of the Ministerio de Educaci�on y Ciencia of Spain (Project PB-94-0743) is acknowledged. References 1 Pollow, K., Sinnecker, R., and Pollow, B., J. Chromatogr., 1974, 90, 402. 2 Verma, P., Curry, C., Crocker, C., Titus Dillon, P., and Ahluwalia, B., Clin.Chim. Acta, 1975, 63, 363. 3 Skrivanek, J. A., Ruhlig, M., and Schraer, R., Abstr. Pap. Am. Chem. Soc. 166th Meet. Biol., 1973, 58. 4 Penzes, L. P., and Oertel, G. W., J. Chromatogr., 1972, 74, 359. 5 Short M. P., and Rhodes, C. T., Can. J. Pharm. Sci., 1973, 8, 26. 6 Moreti, G., Cavina, G., Pacioti, P., and Siniscalchi, P., Farmaco, Ed. Prat., 1972, 27, 537. 7 Eldawy, M. A., Tawfik, A. S., and Elshabouri, S. R., J. Pharm. Sci., 1975, 64, 1221. 8 Wu, J.Y. P., J. Assoc. Off. Anal. Chem., 1974, 57, 747. 9 Szepesi, G., and G�or�og, S., Analyst, 1974, 99, 218. 10 Lisboa, B. P., and Strassner, M., J. Chromatogr., 1975, 111, 159. 11 Graham, R. E., and Kenner, C. T., J. Pharm. Sci., 1973, 62, 1845. 12 Korany, M. A., El-Yazbi, F. A., Abdel-Razak, O., and Elsayed, M. A., Pharm. Weekbl., Sci. Ed., 1985, 7(4), 163. 13 Cao, Y., and Zhang, J., Yaowu Fenxi Zazhi, 1984, 4(1), 31. 14 Corti, P., Lencioni, E., and Sciarra, G. F., Boll. Chim.Farm., 1983, 122(6), 281. 15 Corti, P., and Sciarra, G. F., Boll. Chim. Farm., 1981, 120(12), 701. 16 Talsky, G., Mayring, L., and Kreuzer, H., Angew. Chem., 1978, 17, 785. 17 O’ � Haver, T. C., Anal. Chem., 1979, 51, 91A. 18 Fell, A. F., and Smith, G., Anal. Proc., 1982, 19, 28. 19 Cottrell, C. T., Anal. Proc., 1982, 19, 43. 20 United States Pharmacopeia, XXIII Revision, US Pharmacopeial Convention, Rockville, MD, 1995, pp. 881–883. 21 Data Leader Software Package, Beckman, Fullerton, CA, 1987.Paper 6/04558H Received July 1, 1996 Accepted October 15, 1996 Table 2 Calibration data for the determination of ETE and LEV Standard deviation Inter- Range/ cept Slope LOD/ LOQ/ Equation r mg ml21 3105 3105 mg ml21 mg ml21 Ethinylestradiol— 1D293 = 5.8 3 1025 + 72.5 3 1025 C* 0.9999 26 4.2 0.3 0.3 0.9 Levonorgestrel— 1D249 = 1.4 3 1023 + 2.3 3 1023 C* 0.9996 33 40.7 2.3 0.8 2.5 * C = concentration in mg ml21. Table 3 Compositions and recoveries for artificial mixtures Ethinylestradiol Levonorgestrel Sample Actual/ Found*/ Recovery* Actual/ Found*/ Recovery* No.mg ml21 mg ml21 (%) mg ml21 mg ml21 (%) 1 2.06 2.02 98 3.88 3.86 99 2 4.13 4.23 103 7.77 7.77 100 3 6.19 6.12 99 7.77 7.77 100 4 2.06 2.15 104 10.36 10.30 99 5 4.13 3.96 96 10.36 10.35 100 6 2.06 2.01 97 10.36 10.33 100 7 3.10 2.92 94 15.54 15.48 100 8 8.26 8.28 100 20.72 20.58 99 * Average of two determinations. Table 4 Relative differences between HPLC and derivative spectrophotometric methods in the assays of commercial formulations Ethinylestradiol Levonorgestrel Found/mg Found/mg per tablet Relative per tablet Relative Commercial difference difference formulation HPLC 1D293 (%) HPLC 1D249 (%) Ovoplex 0.0460 0.0439 24 0.2300 0.2320 +1 Microginon 0.0295 0.0300 +2 0.1444 0.1505 +4 Neogynona 0.0461 0.0452 22 0.2383 0.2369 21 Triagynon A* 0.0304 0.0314 +3 0.1223 0.1152 26 Triagynon B* 0.0369 0.0361 22 0.0708 0.0775 +9 Triagynon C* 0.0285 0.0292 +2 0.0467 0.0526 +13 Triciclor A* 0.0249 0.0246 21 0.1271 0.1272 0 Triciclor B* 0.0315 0.0325 +3 0.0681 0.0699 +3 Triciclor C* 0.0249 0.0305 +22 0.0488 0.0522 +6 * A = yellow, B = white, C = brown. 44 Analyst, January 1997, Vol. 122 Simultaneous Determination of Ethinylestradiol and Levonorgestrel in Oral Contraceptives by Derivative Spectrophotometry Juan J. Berzas*, Juana Rodr�ýguez and Gregorio Casta�neda Departamento de Qu�ýmica Anal�ýtica y Tecnolog�ýa de Alimentos, Facultad de Ciencias, Universidad de Castilla-La Mancha, 13071 Ciudad Real, Spain A method for determining ethinylestradiol (ETE) and levonorgestrel (LEV) in mixtures by first-derivative spectrophotometry is described.The procedure does not require any separation step. Measurements are made at the zero-crossing wavelengths and the calibration graphs are linear up to 26 and 33 mg ml21 of ETE and LEV, respectively. The method was applied to the determination of both compounds in five different Spanish commercial low-dose oral contraceptives.Similar results were obtained by an HPLC method. Keywords: Ethinylestradiol; levonorgestrel; derivative spectrophotometry; oral contraceptives At present there are three types of oral contraception available. In the sequential type, estrogen is administered alone for the first week, followed by a lower dosage of the estrogen in conjunction with a progestogen for the remainder of the course. In the second, commonly used, type both an estrogen and a progestogen are present in the tablets (as either a single dose or in three different doses).In the progestogen type, a progestogen alone is administered. Ethinylestradiol (ETE) is a semisynthetic estrogen female sex hormone and levonorgestrel (LEV) is a synthetic steroid with an extremely potent progestational action. The formulation of these steroids in tablets of low dosage, i.e., 30–250 mg per tablet, presented a challenging analytical problem.A sensitive, accurate and rapid procedure is desirable for content uniformity testing of the dosage form. The structure of LEV has a characteristic D4-3-keto group in the A-ring with a different chromophoric power to ETE. The most commonly encountered estrogen is ETE, which is present at a very low dosage level (30–100 mg per tablet) in combination with an orally active synthetic progestin (one of the most commonly used is LEV), which is present at a level of from 5 to 30 times that of the estrogen.Oral contraceptives have had an enormous positive impact on public health for the past three decades and, in the main, there has been a remarkably low incidence of troublesome side-effects. Although estrogens are implicated in an increased incidence of breast and endometrial cancer, epidemiological studies have not provided convincing evidence to support a direct correlation between the use of oral contraceptives and an increase in breast cancer.The modern low-dose oral contraceptives (estrogen–progestogen) require a sensitive analysis method which is unaffected by the small amount of the estrogen and the large excess of progestogen. There have been several reports1–11 of the determination of levonorgestrel or ethinylestradiol, including the use of radioactivelyves,1,2 dansyl or other fluorescent derivatives,3–5 spectrophotometry or photometry6–9 and gel or column chromatography, but the methods are complicated.10,11 No references were found to the simultaneous determination of ETE and LEV using spectrophotometric methods.The determination of ETE in the presence of noresthisterone by derivative spectrophotometry in methanol or ethanol has been reported,12,13 the recoveries in different tablets being 120–80% for both compounds. The simultaneous determination of ETE and mestranol by derivative spectrophotometry has been proposed, either in a solution of methanol and chloroform14 or in NaOH in methanol.15 In derivative UV/VIS spectrophotometry, the information contained in the spectrum is presented in a potentially more useful form, greatly increasing the versatility of the technique16 –18 and offering a convenient solution to a number of well defined analytical problems, such as the resolution of multi-component systems, removal of sample turbidity, matrix background and enhancement of spectral details.19 Although the use of derivative spectra is not new, it has only become practical in recent years with the development of microcomputer technology, which allows the almost instantaneous generation of derivative spectra.In this paper we demonstrate the ease with which the derivative methods circumvent the problem of overlapping spectral bands and sample turbidity, allowing the simultaneous determination of ETE and LEV without prior separation. The method yielded accurate, rapid and reproducible results for five different commercial products, two of them with three different dosages.The results obtained by the proposed method were compared with those obtained by HPLC with spectrophometric detection, very similar to the method proposed in the US Pharmacopeia. 20 Experimental Apparatus A Beckman (Fullerton, CA, USA) DU-70 spectrophotometer equipped with 1.0 cm quartz cells and connected to an IBM-PS 2 Model 30 computer, fitted with Beckman Data Leader software,21 and an Epson FX-850 printer was used for all absorbance measurements. A Shimadzu (Kyoto, Japan) high-performance liquid chromatograph equipped with a Nova-Pak C18 column (15 3 0.39 cm id, 4 mm), a diode-array detector, a Rheodyne injection valve and connected to a computer fitted with CLASS LC-10 software was used.A Crison (Barcelona, Spain) MicropH 2002 pH meter was used for the pH measurements. Standard Solutions All chemicals and solvents were of analytical-reagent grade. ETE and LEV were obtained from Sigma (St.Louis, MO, USA) and stock standard solutions were prepared in absolute ethanol Analyst, January 1997, Vol. 122 (41–44) 41(100 mg ml21). The purities of the ETE and LEV reported by Sigma were 98.6% and 99.6%, respectively, determined using an HPLC method with spectrophotometric detection at 280 and 242 nm. Procedure Calibration Stock standard solutions of ETE and LEV were placed in 25 ml calibrated flasks to give final concentrations of up to 26 and 33 mg ml21, respectively, adding absolute ethanol to dilute the contents to 25 ml (the resulting final solution was 100% in ethanol).This high percentage of ethanol was necessary to obtain total dissolution of the drugs from the oral contraceptive tablets. The absorption spectra of the samples were recorded against an ethanol blank between 315 and 210 nm at a scan speed of 120 nm min21 and stored in the computer. Firstderivative spectra were obtained with Dl = 8 nm and ETE was determined by measuring the signal of the first derivative spectrum at 293.0 nm (1D293) (zero-crossing point for LEV), and by using an appropriate calibration graph, their concentrations could be determined.These calibrations were performed by varying the concentration of the estrogen, in the absence of the other hormone. The LEV content was also determined by measuring the signal at 249.0 nm (1D249) (zero-crossing point for ETE). These spectra were not treated with a smoothing function because the noise level was low.Assay of pharmaceutical preparations Twenty tablets were finely powdered and an appropriate portion (equivalent to the median mass of two tablets) was dissolved in 8 ml of absolute ethanol by sonication for 15 min, followed by shaking by mechanical means for 20 min. The mixture was filtered, using a Swinnex polypropylene disc filter holder of 13 mm diameter (Millipore, Bedford, MA, USA) with an FH 0.5 mm Fluoropore (PFTE) membrane, into a 10 ml calibrated flask.The residue was washed twice with the same solvent and diluted to volume. The absorption spectra were recorded against an absolute ethanol blank and stored in the IBM-PS computer. For determining ETE and LEV, the absorption spectra were handled as in first-derivative spectrophotometry. Comparison with HPLC reference method Twenty tablets were finely powdered and an appropriate portion (equivalent to the median mass of two tablets) was dissolved in 8 ml of absolute ethanol by sonication for 15 min, followed by shaking by mechanical means for 20 min.The mixture was filtered using a Swinnex polypropylene disc filter holder of 13 mm diameter with an FH 0.5 mm Fluoropore (PFTE) membrane into a 10 ml calibrated flask. The residue washed twice with the same solvent and diluted to volume. A 2.5 ml portion of this solution was diluted with water in a 10 ml calibrated flask; the reason for this dilution is to obtain a higher polarity and lower concentration in the sample.HPLC determination was performed on a Nova-Pak C18 60 A column (15 3 0.39 cm id) containing 4 mm packing. The mobile phase was deaerated acetonitrile–methanol–water (3.5 + 1.5 + 4.5) and spectrophotometric detection was performed at 215 nm. The flow rate was about 1 ml min21.20 The differences between the two methods are the dissolution of the tablets (using ethanol–water or acetonitrile–methanol–water) and the column used [C18 (15 3 0.39 cm id) or C8 (15 3 0.46 cm id)].Results and Discussion Method Development The influence of pH on the absorption spectra of ETE (e280 nm = 2253 l mol21 cm21 in absolute ethanol) and LEV (e240 nm = 17 155 l mol21 cm21 in absolute ethanol) was studied, with a total content of ethanol of 25%. The LEV spectrum showed only a maximum at 246 nm, which remained unchanging in the pH range 1.0–12.4. The ETE spectrum showed a maximum at 280 nm in the pH range 1.0–9.5 and for more alkaline solutions two different bands at 296 and 240 nm.The stability of ETE in very acidic solutions was not very good. For this reason, the best results for analytical purposes were obtained in the pH range 4.0–9.5. The preparation of the samples with 100% of ethanol resulted in spectra very close to those obtained at the optimum pH. Samples were prepared in absolute ethanol solution and the addition of a buffer solution was not necessary. Under these conditions, dilute solutions of ETE and LEV were stable for at least 12 h.The use of absolute ethanol permits the best recovery of the hormones in the oral contraceptive tablets. Derivative Spectrophotometry In Fig. 1 the zero-order spectra of ETE and LEV in the wavelength range 210–315 nm are shown. It can be seen that the absorption spectrum of LEV is very overlapped with the ETE spectrum. The determination of ETE directly could be easy at the start, but the small content of this steroid and the high content of LEV in commercial tablets (the ETE : LEV ratio is normally 1 : 4 or 1 : 5) presumes a large contribution of the spectrum of LEV to the maxima in the spectrum of ETE.The spectra of real samples after dissolution of the hormones and filtration show a very small overlap, not perceptible at the beginning, but the spectrum showed a small y-axis displacement of the absorbance owing to the overlap. This behaviour is revealed when the spectrum of an artificial binary mixture is compared with that of a real contraceptive tablet of similar concentration (Fig. 2). Derivative spectrophotometry is a suitable technique to overcome this problem. The zero-crossing method is the most common procedure for the preparation of analytical calibration graphs. In practice, the measurement selected is that which exhibits the best linear response, gives a zero or near zero intercept on the ordinate of the calibration graphs and is less affected by the concentration of any other component. The shape of the first derivative spectrum is adequate for determining ETE in the presence of LEV and vice versa.Fig. 3 shows the firstderivative absorption spectra of a solution of ETE and a solution of LEV, both solutions in 100% ethanol. It can be seen that owing to the overlapping spectra of these compounds, the zero-crossing method is the most appropriate approach for resolving mixtures of these compounds and it was used in this work with satisfactory results.Fig. 1 Absorption spectra of ethinylestradiol (26.01 mg ml21, broken line), levonorgestrel (7.17 mg ml21, dotted line) and their mixture (continuous line). 42 Analyst, January 1997, Vol. 122Preliminary experiments showed that the signals of the first derivative at 293.0 nm (working zero-crossing wavelength of LEV) are proportional to the ETE concentration and the signals of the first derivative at 249.0 nm (working zero-crossing wavelengths of ETE) are proportional to the LEV concentration.Selection of Optimum Instrumental Conditions The main instrumental parameters that affect the shape of the derivative spectra are the wavelength scanning speed, the wavelength increment over which the derivative is obtained (Dl) and the smoothing. These parameters need to be optimized to give a well resolved large peak, i.e., to give good selectivity and higher sensitivity in the determination. Generally, the noise level decreases with increase in Dl, thus decreasing the fluctuations in the derivative spectrum.However, if the value of Dl is too large, the spectral resolution is very poor. Therefore, the optimum value of Dl should be determined by taking into account the noise level, the resolution of the spectrum and the sample concentration. Some values of Dl were tested and 8.0 nm was selected as the optimum in order to obtain a satisfactory signal-to-noise ratio. In this way, a smoothing function was not necessary.Having established the experimental conditions, the calibration graphs were tested between 2.0 and 26.0 mg ml21 of the ETE in the absence of LEV at 293.0 nm for the first-derivative spectra. The calibration graphs were also tested between 4.0 and 30.0 mg ml21 of LEV in the absence of ETE at 243.0 nm for the first-derivative spectra (Fig. 4). Good linearity was observed in all cases. Statistical Study Tables 1 and 2 summarize the most characteristic statistical data obtained from the different calibration graphs and the reproducibility of the reagent blank and a standard.The reproducibility of particular concentrations of ETE (10.40 mg ml21) and LEV (10.60 mg ml21) were evaluated over 2 d by performing 10 absorption spectrophotometric measurements each day on 10 different samples. The results (Table 1) show that the repeatability for both hormones on each day was satisfactory. The comparison of the average concentrations with the Snedecor test did not show any significant difference at a confidence level of 5%.Determination of ETE and LEV in Synthetic Mixtures Some binary mixtures of ETE and LEV were prepared from the stock standard solutions in the proportions from 1 + 1 to 1 + 4 and were analysed by the proposed derivative spectrophotometric method. Some of these proportions between the two hormones were the same as in commercial contraceptives with the object of checking the relationships of more commercial interest. Table 3 shows the results of the analyses of different mixtures.The recoveries were between 94 and 104% for ETE and between 99 and 100% for LEV for the wavelengths studied. These results show that the method is effective for the simultaneous determination of ETE and LEV by first-derivative spectrophotometry. Fig. 2 (a) Absorption spectra of (A) synthetic mixture of ETE (5.00 mg ml21) and LEV (25.00 mg ml21) and (B) commercial solution of he Ovoplex pills, containing the same concentrations of ETE and LEV.(b) Zoom of the spectra between 270–315 nm. Fig. 3 First-derivative spectra (Dl = 8 nm) of ethinylestradiol (26.01 mg ml21, broken line), levonorgestrel (7.17 mg ml21, dotted line) and their mixture (continuous line). Fig. 4 First-derivative spectra (Dl = 8 nm) for different concentrations of levonorgestrel: a, 2.59; b, 5.18; c, 7.77; d, 10.36; e, 12.95; f, 19.42; g, 25.90; and h, 32.63 mg ml21. Table 1 Precision of the determination of the concentration of ETE (10.40 mg ml21) and LEV (10.60 mg ml21) on different days (n = 10 determinations on each day) Ethinylestradiol Levonorgestrel Average/ s/ RSD Average/ s/ RSD mg ml21 mg ml21 (%) mg ml21 mg ml21 (%) Day A 10.26 0.16 1.55 10.53 0.07 0.65 Day B 10.38 0.08 0.75 10.60 0.08 0.73 Analyst, January 1997, Vol. 122 43Determination of ETE and LEV in Commercial Contraceptives The Spanish pharmacological industry has at present five different low-dose commercial oral contraceptives (Neogynona, Microgynon, Ovoplex, Triciclor and Triagynon) containing ETE and LEV.Two of these contraceptives (Triciclor and Triagynon) have in the formulation three different doses (different proportions of ETE and LEV); in these contraceptives the amount of LEV starts at 0.050 mg per tablet at the beginning of the treatment, later it is 0.075 mg per tablet and at the end is 0.125 mg per tablet. Each oral contraceptive packet contains 21 tablets, six of them corresponding to the lowest dosage of LEV, five to the intermediate dosage and ten to the highest dosage. The amount of ETE starts at 0.03 mg per tablet, becomes 0.04 mg per tablet and at the end is once more 0.03 mg per tablet.The results obtained for the determination of ETE and LEV mixtures in commercial pharmaceuticals are given in Table 4. The relative differences between HPLC and derivative spectrophotometric results were between ±6% for the LEV determination in all oral contraceptives except Triagynon B and C.The relative differences for ETE were ±4% in all the commercial formulations except Triciclor C. The anomalous results found for Triagynon and Triciclor could be due to the presence of some interferences due to the excipients or of dyes that coated these tablets. In all cases the recoveries were calculated with respect to the results obtained by the HPLC method. Good correlations between the two methods were found. The indicated value is the mean of two different analyses of the same commercial batch.Conclusions The proposed derivative method is very suitable for the simultaneous determination of ETE and LEV and can be employed to analyse commercial formulations of low-dose oral contraceptives. The proposed method gave good results when compared with the HPLC method. The authors are grateful to Dr. V. Trigo (Wyeth-Orfi Laboratories) and Dr. C. Barona (Shering Laboratories). Financial support from the DGICYT of the Ministerio de Educaci�on y Ciencia of Spain (Project PB-94-0743) is acknowledged.References 1 Pollow, K., Sinnecker, R., and Pollow, B., J. Chromatogr., 1974, 90, 402. 2 Verma, P., Curry, C., Crocker, C., Titus Dillon, P., and Ahluwalia, B., Clin. Chim. Acta, 1975, 63, 363. 3 Skrivanek, J. A., Ruhlig, M., and Schraer, R., Abstr. Pap. Am. Chem. Soc. 166th Meet. Biol., 1973, 58. 4 Penzes, L. P., and Oertel, G. W., J. Chromatogr., 1972, 74, 359. 5 Short M.P., and Rhodes, C. T., Can. J. Pharm. Sci., 1973, 8, 26. 6 Moreti, G., Cavina, G., Pacioti, P., and Siniscalchi, P., Farmaco, Ed. Prat., 1972, 27, 537. 7 Eldawy, M. A., Tawfik, A. S., and Elshabouri, S. R., J. Pharm. Sci., 1975, 64, 1221. 8 Wu, J. Y. P., J. Assoc. Off. Anal. Chem., 1974, 57, 747. 9 Szepesi, G., and G�or�og, S., Analyst, 1974, 99, 218. 10 Lisboa, B. P., and Strassner, M., J. Chromatogr., 1975, 111, 159. 11 Graham, R. E., and Kenner, C. T., J. Pharm. Sci., 1973, 62, 1845. 12 Korany, M. A., El-Yazbi, F. A., Abdel-Razak, O., and Elsayed, M. A., Pharm. Weekbl., Sci. Ed., 1985, 7(4), 163. 13 Cao, Y., and Zhang, J., Yaowu Fenxi Zazhi, 1984, 4(1), 31. 14 Corti, P., Lencioni, E., and Sciarra, G. F., Boll. Chim. Farm., 1983, 122(6), 281. 15 Corti, P., and Sciarra, G. F., Chim. Farm., 1981, 120(12), 701. 16 Talsky, G., Mayring, L., and Kreuzer, H., Angew. Chem., 1978, 17, 785. 17 O’ � Haver, T. C., Anal. Chem., 1979, 51, 91A. 18 Fell, A. F., and Smith, G., Anal. Proc., 1982, 19, 28. 19 Cottrell, C. T., Anal. Proc., 1982, 19, 43. 20 United States Pharmacopeia, XXIII Revision, US Pharmacopeial Convention, Rockville, MD, 1995, pp. 881–883. 21 Data Leader Software Package, Beckman, Fullerton, CA, 1987. Paper 6/04558H Received July 1, 1996 Accepted October 15, 1996 Table 2 Calibration data for the determination of ETE and LEV Standard deviation Inter- Range/ cept Slope LOD/ LOQ/ Equation r mg ml21 3105 3105 mg ml21 mg ml21 Ethinylestradiol— 1D293 = 5.8 3 1025 + 72.5 3 1025 C* 0.9999 26 4.2 0.3 0.3 0.9 Levonorgestrel— 1D249 = 1.4 3 1023 + 2.3 3 1023 C* 0.9996 33 40.7 2.3 0.8 2.5 * C = concentration in mg ml21. Table 3 Compositions and recoveries for artificial mixtures Ethinylestradiol Levonorgestrel Sample Actual/ Found*/ Recovery* Actual/ Found*/ Recovery* No. mg ml21 mg ml21 (%) mg ml21 mg ml21 (%) 1 2.06 2.02 98 3.88 3.86 99 2 4.13 4.23 103 7.77 7.77 100 3 6.19 6.12 99 7.77 7.77 100 4 2.06 2.15 104 10.36 10.30 99 5 4.13 3.96 96 10.36 10.35 100 6 2.06 2.01 97 10.36 10.33 100 7 3.10 2.92 94 15.54 15.48 100 8 8.26 8.28 100 20.72 20.58 99 * Average of two determinations. Table 4 Relative differences between HPLC and derivative spectrophotometric methods in the assays of commercial formulations Ethinylestradiol Levonorgestrel Found/mg Found/mg per tablet Relative per tablet Relative Commercial difference difference formulation HPLC 1D293 (%) HPLC 1D249 (%) Ovoplex 0.0460 0.0439 24 0.2300 0.2320 +1 Microginon 0.0295 0.0300 +2 0.1444 0.1505 +4 Neogynona 0.0461 0.0452 22 0.2383 0.2369 21 Triagynon A* 0.0304 0.0314 +3 0.1223 0.1152 26 Triagynon B* 0.0369 0.0361 22 0.0708 0.0775 +9 Triagynon C* 0.0285 0.0292 +2 0.0467 0.0526 +13 Triciclor A* 0.0249 0.0246 21 0.1271 0.1272 0 Triciclor B* 0.0315 0.0325 +3 0.0681 0.0699 +3 Triciclor C* 0.0249 0.0305 +22 0.0488 0.0522 +6 * A = yellow, B = white, C = brown. 44 Analyst, January 1997, Vol. 1

ISSN:0003-2654    

DOI:10.1039/a604558h

出版商:RSC    

年代:1997

数据来源: RSC

摘要:

Spectrofluorimetric Study of the Effects of Cyclodextrins on the Acid–Base Equilibria of Harmine and Harmane† L. Mart�ýn, M. A. Mart�ýn and B. del Castillo* Seccion Departamental de Qu�ýmica Anal�ýtica, Facultad de Farmacia, Universidad Complutense de Madrid, 28040-Madrid, Spain b-Carboline alkaloids are important compounds because they exhibit a variety of pharmacological actions. Their acid–base behaviour can be studied by spectrofluorimetry since these molecules present a remarkable native luminescence.Acid–base equilibria depend on the environment of the molecules and inclusion into cyclodextrin (CD) cavities shifts the acid–base equilibria and alters the apparent pKa values. The influence of CDs on the acid–base equilibria of the model b-carbolines harmine and harmane is described. b-CD and g-CD and the modified b-CDs hydroxypropyl-b-CD (HPb-CD), 2,6-di-O-methyl-b-CD (DMb-CD) and 2,3,6-tri-O-methyl-b-CD (TMb-CD) were used to form the corresponding complexes with harmine and harmane in the pH range 7.8–8.0.In these buffered solutions the complexes with the different CDs exhibit an emission band with resolved peaks at 360 and 380 nm corresponding to the neutral form of harmane and with a remarkable enhancement in the emission intensity compared with aqueous solution. In the case of the complexes with b-CD and g-CD, both the cationic and the neutral emission bands appear. However, for g-CD the cationic band is more intense than the neutral band, the inverse being true for b-CD.In homogeneous aqueous solution at this pH value the cationic band is the only one observed and therefore the presence of the neutral band indicates the formation of inclusion complexes. In the harmane–HPb-CD complexes, the emission bands ascribed to the anionic form are observed after addition of NaOH. This emission is only observed in homogeneous aqueous solution in strongly alkaline media outside the normal pH range.Keywords: Harmane; harmine; cyclodextrin complexes; fluorescence Harmine and harmane (Fig. 1) are b-carboline alkaloids which exhibit a notable native fluorescence and a peculiar acid–base behaviour in the ground and excited states. Thus, the pyridine nitrogen behaves as a base and is easily protonated, and therefore all b-carboline derivatives studied by Bal�on et al.1 present pKa values for this process that vary from 6.2 to 9.5. On the other hand, the pyrrolic nitrogen is acidic and loses its proton in alkaline media, although outside the pH scale (pH > 14).1,2 This behaviour is a typical consequence of the chemical characteristics due to the presence of a p-deficient pyridine ring fused to an electron-excessive indole ring.The acid–base behaviour in the ground state can be easily followed by UV/VIS spectrophotometry, observing the characteristic absorption band which can be attributed to cationic, neutral and anionic species.2,3 However, the acid–base behaviour in the excited state changes remarkably and both the basicity of pyridine and the acidity of pyrrole rings are strongly increased and therefore the pKa* values differ from the pKa.3,4 In the excited state it is also possible to observe another species involved in the acid–base equilibria, which Sakurovs and Ghiggino5 describe as a zwitterion.According to Sakurovs and Ghiggino,5 proton transfer in the excited state is very rapid and excitation of neutral or anionic species formed in the ground state produces the corresponding fluorescent cationic or zwitterionic species in the excited state. Inclusion in cyclodextrin (CD) cavities notably modifies chemical properties such as acid–base or redox behaviour and reactivity.6 A number of groups have studied changes in the acid–base equilibria for several fluorescent molecules included in CDs.Thus, we have shown that modified b-CDs alter proton transfer in carbazole and ellipticine.7 Chattopadhyay8 described differences in the acid–base behaviour of carbazole in b- and g- CD which differ from that observed in homogeneous aqueous solutions.Dissociation processes of 1-naphthol are seriously hampered after inclusion in modified b-CDs,9 a phenomenon which is easily monitored by spectrofluorimetry. The increase or decrease in the observed pKa values after inclusion complex formation depends on the chemical characteristics of the guest molecule and thus the apparent pKa values of nitrophenol derivatives decrease after the inclusion processes.10,11 However, carboxylic acids show the opposite effect and the observed pKa values are increased after inclusion; this is the case for the 1-adamantanecarboxylic acid series,12 cinnamic acid and its analogues13and prostaglandins.14 Considering that b-carbolines are highly fluorescent,15,16 reversed-phase HPLC with fluorimetric detection,17,18 is a very useful technique to determine these compounds in biological fluids.However, problems related to the coexistence of several ionized and neutral species may cause difficulties in the detection. In this present paper we describe the inclusion complexes of harmine and harmane and several CDs as well as the consequences of the complexation processes on the acid– base equilibria of these compounds. We conclude that inclusion into CD allows one to observe in aqueous solution and in the region of neutral pH the emission corresponding to the neutral species of harmine and harmane, an unprecedented and analytically useful observation.Experimental Apparatus and Reagents Uncorrected fluorescence spectra were measured with a Perkin- Elmer (Norwalk, CT, USA) MPF-2A fluorimeter (xenon lamp, † Presented at the VIIth International Symposium on Luminescence Spectrometry in Biomedical Analysis, Sophia Antipolis, Nice, France, April 17–19, 1996. Fig. 1 Structures of harmine and harmane. Analyst, January 1997, Vol. 122 (45–49) 45150 W). All reagents and solvents were of analytical-reagent grade and were used without further purification. Harmine and harmane (free bases) were purchased from Aldrich (Milwaukee, WI, USA), the cyclodextrins (b-CD, HPb-CD, DMb-CD, TMb- CD and g-CD) from Sigma (St. Louis, MO, USA), and ethanol, sodium bromide and chloride from Merck (Darmstadt, Germany). Water was doubly distilled and de-ionized prior to its use. Procedures Freshly prepared ethanolic solutions of harmine and harmane were prepared at a 0.001 m concentration.Aliquots of 10 ml of these solutions were taken and placed in a round-bottomed flask. The solvent was evaporated under reduced pressure at room temperature and then 10 ml of aqueous solutions of the different CDs at a concentration of 0.01 m were added. The solutions of CDs were prepared in water or buffered aqueous solutions and left to stabilize for 24 h prior to their use in order to ensure complete dissolution.The complexes were prepared at different pH values. Besides de-ionized water (pH 5.5), Britton– Welford titrated solutions (0.2 m KH2PO4 with the desired volume of 0.2 m NaOH) were employed to obtain the required pH values. In the case of complexes prepared at pH 7.8, CD solutions were dissolved in the corresponding buffered aqueous solutions. However, for pH 11.0 this was obtained by addition of a suitable amount of NaOH to the aqueous CD solution. The final concentration of b-carboline in the different solutions was 1.0 3 1026 m.The b-carboline–CD solutions were stirred magnetically for 18–48 h in order to obtain the inclusion complexes. Acid–base equilibria in aqueous and CD solutions were studied using the above-mentioned procedure to prepare the inclusion complexes. When the complexes were obtained, successive aliquots of 10 ml of NaOH (10 m) were added in order to study the shifts in the proton transfer processes. Fluorescence quenching of b-carbolines was studied using bromide ion (NaBr) as quencher.NaCl was added to the complex solutions to achieve a constant ionic strength (1.0 m). Aliquots of 10 ml of the NaBr solution were added to study the quenching effect. The concentration of bromide ion varied in the complex solution from 0.001 to 0.1s and Discussion We studied the influence of CD complexation on the acid–base equilibria of harmine and harmane by spectrofluorimetry, believing that this process could seriously affect the determination of these compounds when CDs are employed.Figs. 2 and 3 show the emission spectra of harmane in ethanolic and aqueous solutions. In ethanolic solution, emission bands appear at 360 and 380 nm, which can be attributed to the neutral form. Addition of small amounts of HCl (1 m) caused the appearance of a band at 430 nm, which can be attributed to the cationic form. Addition of NaOH (1 m) induced a decrease in the intensity of the neutral band, together with a very weak increase in the emission at 480 nm which, according to other workers,1,3,5 is due to the formation of a zwitterionic species.The addition of more concentrated NaOH solutions (10 m) or solid NaOH did not cause the appearance of an anionic band. This happened also when harmine and harmane were dissolved in other organic solvents (hexane or propan-1-ol), and also when triethylamine was added as a base. In agreement with observations by Sakurovs and Ghiggino,5 the excitation spectra were the same for cationic, neutral and zwitterionic species.The same tests were performed in aqueous solutions (Fig. 3) and the behaviour was different, since the cationic band only appeared in the aqueous solution at 430 nm, and the addition of acid (1 m HCl) increased slightly this emission corresponding to the cationic band. Addition of NaOH (1 m) produced the zwitterionic band (480 nm) with a notable fluorescence intensity compared with the ethanolic solution and the neutral (370 nm) band.It is important to note that in all solvents studied the emission intensity of the cationic band was considerably higher than that of the corresponding neutral or zwitterionic bands. Addition of more concentrated NaOH (10 m) did not cause the anionic band to appear, because it is necessary to work outside the pH scale.1,2 Fig. 2 Uncorrected excitation and emission spectra of harmane in ethanolic solution (1).Same sample after addition of small amounts of HCl (2) and NaOH (3). Fig. 3 Uncorrected excitation and emission spectra of harmane in aqueous solution (1). Same sample after addition of small amounts of HCl (2) and NaOH (3). The fluorescence intensity for solution 1 is three times lower than that for solution 2 and that for solution 3 is 81 times lower than that for solution 2. 46 Analyst, January 1997, Vol. 122Inclusion complexes with the different CDs were prepared for both b-carbolines studied.Considering the acid–base behaviour of these compounds, the increase in the emission intensity and the shifts in the emission maxima or changes in the fluorescence lifetime or fluorescence quantum yield are proof of the inclusion processes.19 With this in mind, we tried to prepare the inclusion complexes at different pH values (5.5, 7.8, 11.0). Thus, at pH 5.5 only the cationic form was observed, with a weak (5%) increase in the fluorescence intensity of the complex solution with respect to the aqueous solutions of harmine and harmane.Therefore, the formation of inclusion complexes with cationic species is unlikely. This may be due to the fact that cationic species are water soluble and therefore their tendency to be included is low. In the case of HPb-CD, after several days under magnetic stirring the emission of the neutral form together with the predominant cationic emission were observed. A pH of 7.8 was selected because it is very close to the pKa values of harmine and harmane and under such conditions the same concentration of the cationic and neutral species should be present in the ground state.Therefore, inclusion of the neutral species should be favoured, shifting the acid–base equilibria. Under such conditions we obtained the neutral form with a fluorescence intensity higher than that in water. The spectral shape resembles that observed for the neutral form in ethanol; however, in the presence of the CDs it is better resolved and two peaks appear.When complexes were prepared at pH 11.0, the emission corresponding to the neutral form could be observed, but some differences in the spectral shape with respect to those obtained at pH 7.8 can be noted. Thus, the neutral band is not resolved into two peaks and only a peak at 365 nm appears, together with two shoulders at 380 and 420 nm. These changes in the spectral profile may be associated with the existence of non-complexed zwitterionic or anionic species which are present at such pH values in the ground state. However, the additions of NaOH show that this emission corresponds to the cationic form, which is probably produced in the excited state at this pH value.Excitation produces the cationic form owing to very rapid proton transfer in the excited state, as described by Sakurovs and Ghiggino.5 The excited states of b-carbolines are strong bases and deprotonate water.5 From these results, it can be deduced that pH 8.0 is the most suitable for studying the inclusion complexes. HP-b-CD was selected as a model CD to study the time necessary to obtain the inclusion complexes.The complexes with harmine and harmane were prepared following the experimental procedure described, and at pH 7.8–8.0 the b-carboline–HP-bCD complexes showed the neutral band only after 18–24 h under magnetic stirring. A time span of 24 h was chosen to obtain comparable results for the different complexes.At pH 11.0 the results did not improve compared with those obtained at pH 8.0 and, as these conditions can affect the CDs by hydrolysis,20 they were not used in subsequent experiments. Fig. 4 shows the emission spectra for the different harmane–CD complexes using the optimized procedure to obtain the complexes. After 24 h of magnetic stirring the spectra corresponding to the neutral form are present for modified b-CD (HPb-CD, DMb-CD and TMb-CD). For b- CD, a mixture of cationic and neutral forms exists but the equilibrium is shifted towards the neutral form because the intensity of the cationic band is very small.For g-CD, on the other hand, a mixture was also obtained at the same pH values, but the ratio of the species in equilibrium is the opposite to that observed with b-CD. As can be seen in Fig. 4, the shift in the acid–base equilibria depended on the type of CD since they displaced the acid-base equilibria in a different way because of different efficiencies in including the neutral form.In homogeneous aqueous solution, the fluorescence intensity of the cation is about 10 times higher than that of the neutral form, with a considerable difference in quantum yields1 [FF (cation) = 0.76 and FF (neutral) = 0.17], which makes it difficult to observe in aqueous or alkaline solution owing to the protonation of the excited neutral species by rapid proton exchange with the solvent.5 The formation of complexes at pH 7.8–8.0 induced the appearance of emission bands corresponding to the neutral form.This is an important proof of the formation of an inclusion complex because the excited states are protected inside the cavity of CDs and proton transfer is hampered. The fluorescence intensity for the neutral band of harmane in the complexes is only 2–3 times lower than that of the cation. This enhancement in the neutral emission shows the existence of inclusion complexes and also increases the analytical sensitivity.Fig. 5 shows the titration of harmane–HPb-CD complexes with NaOH. The complexes were prepared according to the experimental procedure at pH 7.8. It can be seen that the emission band for the neutral form is resolved into two peaks, and this does not happen in homogeneous aqueous solution. Increasing amounts of NaOH produced a notable decrease in the intensity of the neutral band with the appearance and increase of the corresponding anionic band; the resolution was better than in 14 m KOH1,2 and consequently the alkaloid remained inside the CD.The isoemissive point at 400 nm shows the existence of only two species (neutral and anionic) in the acid–base equilibria. The formation of an anion was not observed in aqueous solution except outside the pH scale and was a consequence of the deprotonation of the pyrrolic nitrogen. We have described previously the anion emission in micellar aqueous solutions of cetyltrimethylammonium bromide.20 Harmane –HPb-CD complexes were the only examples where the formation of an anionic species was observed and this can be attributed to the effects that the CD environment has on the acid–base properties of the guest molecules.Therefore, HPb- CD can isolate harmane from the aqueous environment, hampering proton transfer from the solvent in the excited state. When the neutral form is included in the different CDs, it is not possible to obtain the cationic form after excitation as in homogeneous aqueous solution. Fig. 6 shows the emission spectra of the harmane–b-CD complex at pH 8.0. In the case of b-CD, it was not possible to shift the inclusion equilibrium to obtain only the neutral form, Fig. 4 Uncorrected excitation and emission spectra of the complexes obtained from harmane and the different CDs studied at pH 7.8: (1) b-CD; (2) HPb-CD; (3) DMb-CD; (4) TMb-CD and (5) g-CD. Analyst, January 1997, Vol. 122 47either for harmane or for harmine. In such cases, when NaOH was added an increase in the fluorescence intensity for the neutral band was initially produced. This was followed by the transformation of the neutral form into the zwitterionic species, as shown by the appearance of the emission corresponding to this form. This behaviour can be explained by considering that the cationic form is solubilized in water and that its acid–base equilibrium is similar to that operating in aqueous solution or, alternatively, considering that the indolyl moiety is included with the pyridine end protruding from the CD cavity.This hypothesis is especially valuable in the case of harmine, where the steric hindrance due to the methoxy group on the indolyl moiety makes the inclusion difficult. The effects of NaOH additions on DMb-CD and TMb-CD complexes are very similar. The inclusion complexes produced the emission band corresponding to the neutral form but a tail was observed in the region of the cationic band (Fig. 4). When NaOH was added, a notable decrease was produced in the neutral band with a very weak increase in the emission at 480 nm (zwitterionic species). The absence of an isoemissive point indicates the presence of cationic, neutral and zwitterionic species in the equilibria. The different behaviour of the complexes with increasing amounts of NaOH causes a decrease in the intensity corresponding to the neutral form and an increase in the emission corresponding to the anionic form (in the case of HPb-CD) or the emission of the zwitterionic form for the other CDs.This means an important change in the acid–base behaviour of the b-carbolines as a consequence of the inclusion and the associated selectivity of CDs to include specific species of harmine or harmane. Table 1 summarizes the spectrofluorimetric characteristics of the inclusion complexes studied. While the emission corresponding to the neutral band is present in all cases, for b-CD and g-CD the maximum corresponding to the cationic band in the emission spectra is also present.The presence of this peak is a consequence of the geometrical characteristics of b-CD and g-CD. Thus, b-CD is too small to accommodate b-carboline completely and the molecules are included but the pyridine moiety protrudes from cavity and is protonated by water. The cavity of g-CD is large enough to accommodate harmane and harmine but water molecules may penetrate into the cavity, producing the cationic band when the neutral species are excited.In order to obtain more detailed information concerning the protection afforded by CDs to the excited states, fluorescence quenching of the b-carbolines–CD complexes was tested. This study was carried out for the neutral, cationic and zwitterionic forms obtained after addition of HCl or NaOH to the neutral complexes. In some complexes the correlation coefficient obtained using the Stern–Volmer treatment was not adequate, showing that other quenching mechanisms (static, collisional, energy transfer, etc.) should be present.However, the presence of bromide ion decreased the fluorescence intensity of the complexes. Table 2 shows that neutral forms are effectively protected in the harmane–CD complexes because the slope obtained was lower than that in homogeneous ethanolic solutions where only the neutral band is present. The microenvironments provided by CDs and homogeneous ethanolic solutions are similar.21 Nevertheless, the slope for these complexes was higher than that obtained in water.The emission Fig. 5 Spectrofluorimetric study of the changes in the acid–base equilibrium for the harmane–HPb-CD inclusion complex prepared in buffered aqueous solution (pH 7.8) with addition of 10 ml of 10 m NaOH. Fig. 6 Spectrofluorimetric study of the changes in the acid–base equilibrium for the harmane–b-CD inclusion complex prepared in buffered aqueous solution (pH 7.8) with addition of 10 ml of 10 m NaOH.The dotted line corresponds to the starting solution of the complex. Table 1 Fluorescence characteristics obtained for the CD inclusion complexes with harmane and harmine. The symbols > and < mean the relative fluorescence intensity of one maximum with respect to another Complex Harmane–HPb-CD Harmane–DMb-CD Harmane–TMb-CD Harmane–b-CD Harmane–g-CD lem/nm 362 > 380 362 > 380 362 > 380 362 > 380 > 430 362, 380 < 430 Complex Harmine–HPb-CD Harmine–DMb-CD Harmine–TMb-CD Harmine–b-CD Harmine–g-CD lem/nm 358, 370 358, 370 358, 370 358, 370 > 416 358 < 370 < 416 48 Analyst, January 1997, Vol. 122of the cationic form is also effectively protected against the quencher because the slopes are lower than in water and consequently it is possible that cationic species are partially included. However, in the case of the zwitterionic species a decrease in the fluorescence intensity was produced in the presence of bromide ion, but this phenomenon does not fit the Stern–Volmer relationship and the slopes for the complexes are higher than in aqueous solution.For the harmine–CD complexes, the neutral form presents a weaker quenching effect than in ethanolic solution. The cationic form is also less quenched compared with the corresponding aqueous solution, although the slopes are higher than for harmane complexes. This is certainly associated with the geometrical characteristics of the inclusion complexes.It can also be considered that a small fraction of cationic harmine is included and that the free cationic harmine is quenched as in aqueous solution. It is remarkable that with b-CD the slope is higher than in water, in contrast to the behaviour of g-CD, where the slope is strongly reduced. The differences in the geometries of harmine and harmane produce different inclusion complexes. Thus g-CD can include harmine by the methoxyindole moiety, but the size of the methoxy group does not allow such an inclusion in b-CD or in modified b-CDs.In conclusion, the existence of the inclusion complexes was verified by the changes in the spectrofluorimetric properties, in the acid–base behaviour and by the protection against the effects of the quenching. Nevertheless, the geometry of the inclusion complexes of harmine and harmane can be different because in the case of harmane the indole moiety can be included with the pyridine moiety protuding from the cavity.For harmine this is not easy and a more difficult penetration into the cavity is expected. Considering this selectivity of CDs, their use in chromatography can contribute to enhancing the chromatographic separation of b-carbolines owing to the frequent use of HPLC with fluorimetric detection in b-carboline determination.17,18 References 1 Bal�on, M., Hidalgo, J., Guardado, P., Mu�noz, M.A., and Carmona, C., J. Chem. Soc. Perkin Trans. 2, 1993, 99. 2 Bal�on, M., Mu�noz, M. A., Hidalgo, J., Carmona, M. C., and S�anchez, M., J. Photochem., 1987, 36, 193.F., Zabala, I., and Olba, A., J. Photochem., 1983, 23, 355. 4 Vander Donckt, E., Prog. React. Kinet., 1970, 5, 274. 5 Sakurovs, R., and Ghiggino, K. P., J. Photochem., 1982, 18, 1. 6 Bender, M. L., and Komiyama, M., Cyclodextrin Chemistry, Springer, Berlin, 1978. 7 Sba�ý, M., Ait Lyazidi, S., Lerner, D.A., del Castillo, B., and Mart�ýn, M. A., Anal. Chim. Acta, 1995, 303, 47. 8 Chattopadhyay, N., J. Photochem. Photobiol. A, 1991, 58, 31. 9 Takahashi, K., J. Chem. Soc. Chem. Commun., 1991, 929. 10 Connors, K. A., and Lipari, J. M., J. Pharm. Sci., 1976, 65, 379. 11 Lin, S. F., and Connors, K. A., J. Pharm. Sci., 1983, 72, 1333. 12 Eftink, M. R., Andy, M. L., Bystrom, K., Perlmutter, M. D., and Kristol, D. S., J. Am. Chem. Soc., 1989, 111, 6765 13 Connors, K. A., and Rosanske, T.W., J. Pharm. Sci., 1980, 69, 173. 14 Uekama, K., Hirayama, F., Nasu, S., and Matsuo, N., Chem. Pharm. Bull., 1978, 26, 3477. 15 Abramovitch, R. A., and Spenser, I. D., Adv. Heterocycl. Chem., 1964, 3, 79. 16 Dillon, J., Spector, A., and Nakanishi, K., Nature (London), 1976, 259, 422. 17 Bossin, T. R., and Faull, K. F., J. Chromatogr., 1988, 428, 229. 18 Moncrieff, J., J. Chromatrogr., 1989, 496, 269. 19 Szejtli, J., Cyclodextrins and Their Inclusion Complexes, Akad�emiai Kiad�o, Budapest, 1982. 20 Mart�ýn, L., Mart�ýn, M.A., and del Castillo, B., J. Fluorescence, in the press. 21 Frankewich, R. P., Thimmaiah, K. N., and Hinze, W. L., Anal. Chem., 1991, 63, 2924. Paper 6/02790C Received April 22, 1996 Accepted October 10, 1996 Table 2 Fluorescence quenching study of b-carboline–CD inclusion complexes b-Carboline–CD lem * r† m† b† Harmane–EtOH x 0.947 4.759 1.035 Harmane–H2O x 0.978 0.558 0.958 y 0.984 1.048 0.98 z 0.996 7.7445 0.9552 Harmane–HPb-CD x ‡ ‡ ‡ y ‡ ‡ ‡ z 0.91 5.4 1.03 Harmane–DMb-CD x 0.95 0.858 1.034 y ‡ ‡ ‡ z 0.97 4.454 1.077 Harmane–TMb-CD x 0.949 4.23 1.021 y ‡ ‡ ‡ z 0.95 5.2 1.04 Harmane–g-CD x 0.96 1.037 0.955 y ‡ ‡ ‡ z 0.94 4.17 1.04 Harmane–b-CD x 0.953 0.916 0.999 y 0.985 1.525 1.019 z 0.994 3.547 1.044 Harmine–EtOH x 0.9873 9.63 1.066 Harmine–H2O x 0.969 0.68 0.98 y 0.986 5.109 0.986 z 0.998 15.768 0.986 Harmine–HPb-CD x 0.974 1.83 1.033 y 0.846 1.23 1.082 z 0.965 9.47 1.085 Harmine–DMb-CD x 0.96 1.737 1.04 y 0.98 0.866 1.0007 z 0.994 11.27 1.034 Harmine–TMb-CD x 0.96 2.69 1.07 y 0.966 1.373 1.066 z 0.958 10.71 1.096 Harmine–g-CD x 0.814 0.32 1.045 y 0.97 2.2 1.052 z 0.83 3.7 1.23 Harmine–b-CD x 0.95 3.7 1.02 y 0.92 1.73 1.097 z 0.977 16.02 1.22 * l = Emission wavelength.x: l = 366 nm (harmane); l = 358 nm (harmine). y: l = 480 nm (harmane; l = 476 nm (harmine). z: l = 430 nm (harmane); l = 416 nm (harmine). † r = correlation coefficient; m = slope; b = intercept inordinate.‡ The values obtained could not be adjusted by the Stern–Volmer treatment: Fo/FA = 1 + K(Q). Analyst, January 1997, Vol. 122 49 Spectrofluorimetric Study of the Effects of Cyclodextrins on the Acid–Base Equilibria of Harmine and Harmane† L. Mart�ýn, M. A. Mart�ýn and B. del Castillo* Seccion Departamental de Qu�ýmica Anal�ýtica, Facultad de Farmacia, Universidad Complutense de Madrid, 28040-Madrid, Spain b-Carboline alkaloids are important compounds because they exhibit a variety of pharmacological actions.Their acid–base behaviour can be studied by spectrofluorimetry since these molecules present a remarkable native luminescence. Acid–base equilibria depend on the environment of the molecules and inclusion into cyclodextrin (CD) cavities shifts the acid–base equilibria and alters the apparent pKa values. The influence of CDs on the acid–base equilibria of the model b-carbolines harmine and harmane is described. b-CD and g-CD and the modified b-CDs hydroxypropyl-b-CD (HPb-CD), 2,6-di-O-methyl-b-CD (DMb-CD) and 2,3,6-tri-O-methyl-b-CD (TMb-CD) were used to form the corresponding complexes with harmine and harmane in the pH range 7.8–8.0.In these buffered solutions the complexes with the different CDs exhibit an emission band with resolved peaks at 360 and 380 nm corresponding to the neutral form of harmane and with a remarkable enhancement in the emission intensity compared with aqueous solution. In the case of the complexes with b-CD and g-CD, both the cationic and the neutral emission bands appear.However, for g-CD the cationic band is more intense than the neutral band, the inverse being true for b-CD. In homogeneous aqueous solution at this pH value the cationic band is the only one observed and therefore the presence of the neutral band indicates the formation of inclusion complexes. In the harmane–HPb-CD complexes, the emission bands ascribed to the anionic form are observed after addition of NaOH.This emission is only observed in homogeneous aqueous solution in strongly alkaline media outside the normal pH range. Keywords: Harmane; harmine; cyclodextrin complexes; fluorescence Harmine and harmane (Fig. 1) are b-carboline alkaloids which exhibit a notable native fluorescence and a peculiar acid–base behaviour in the ground and excited states. Thus, the pyridine nitrogen behaves as a base and is easily protonated, and therefore all b-carboline derivatives studied by Bal�on et al.1 present pKa values for this process that vary from 6.2 to 9.5.On the other hand, the pyrrolic nitrogen is acidic and loses its proton in alkaline media, although outside the pH scale (pH > 14).1,2 This behaviour is a typical consequence of the chemical characteristics due to the presence of a p-deficient pyridine ring fused to an electron-excessive indole ring. The acid–base behaviour in the ground state can be easily followed by UV/VIS spectrophotometry, observing the characteristic absorption band which can be attributed to cationic, neutral and anionic species.2,3 However, the acid–base behaviour in the excited state changes remarkably and both the basicity of pyridine and the acidity of pyrrole rings are strongly increased and therefore the pKa* values differ from the pKa.3,4 In the excited state it is also possible to observe another species involved in the acid–base equilibria, which Sakurovs and Ghiggino5 describe as a zwitterion.According to Sakurovs and Ghiggino,5 proton transfer in the excited state is very rapid and excitation of neutral or anionic species formed in the ground state produces the corresponding fluorescent cationic or zwitterionic species in the excited state. Inclusion in cyclodextrin (CD) cavities notably modifies chemical properties such as acid–base or redox behaviour and reactivity.6 A number of groups have studied changes in the acid–base equilibria for several fluorescent molecules included in CDs.Thus, we have shown that modified b-CDs alter proton transfer in carbazole and ellipticine.7 Chattopadhyay8 described differences in the acid–base behaviour of carbazole in b- and g- CD which differ from that observed in homogeneous aqueous solutions. Dissociation processes of 1-naphthol are seriously hampered after inclusion in modified b-CDs,9 a phenomenon which is easily monitored by spectrofluorimetry. The increase or decrease in the observed pKa values after inclusion complex formation depends on the chemical characteristics of the guest molecule and thus the apparent pKa values of nitrophenol derivatives decrease after the inclusion processes.10,11 However, carboxylic acids show the opposite effect and the observed pKa values are increased after inclusion; this is the case for the 1-adamantanecarboxylic acid series,12 cinnamic acid and its analogues13and prostaglandins.14 Considering that b-carbolines are highly fluorescent,15,16 reversed-phase HPLC with fluorimetric detection,17,18 is a very useful technique to determine these compounds in biological fluids.However, problems related to the coexistence of several ionized and neutral species may cause diffion. In this present paper we describe the inclusion complexes of harmine and harmane and several CDs as well as the consequences of the complexation processes on the acid– base equilibria of these compounds.We conclude that inclusion into CD allows one to observe in aqueous solution and in the region of neutral pH the emission corresponding to the neutral species of harmine and harmane, an unprecedented and analytically useful observation. Experimental Apparatus and Reagents Uncorrected fluorescence spectra were measured with a Perkin- Elmer (Norwalk, CT, USA) MPF-2A fluorimeter (xenon lamp, † Presented at the VIIth International Symposium on Luminescence Spectrometry in Biomedical Analysis, Sophia Antipolis, Nice, France, April 17–19, 1996.Fig. 1 Structures of harmine and harmane. Analyst, January 1997, Vol. 122 (45–49) 45150 W). All reagents and solvents were of analytical-reagent grade and were used without further purification. Harmine and harmane (free bases) were purchased from Aldrich (Milwaukee, WI, USA), the cyclodextrins (b-CD, HPb-CD, DMb-CD, TMb- CD and g-CD) from Sigma (St. Louis, MO, USA), and ethanol, sodium bromide and chloride from Merck (Darmstadt, Germany).Water was doubly distilled and de-ionized prior to its use. Procedures Freshly prepared ethanolic solutions of harmine and harmane were prepared at a 0.001 m concentration. Aliquots of 10 ml of these solutions were taken and placed in a round-bottomed flask. The solvent was evaporated under reduced pressure at room temperature and then 10 ml of aqueous solutions of the different CDs at a concentration of 0.01 m were added.The solutions of CDs were prepared in water or buffered aqueous solutions and left to stabilize for 24 h prior to their use in order to ensure complete dissolution. The complexes were prepared at different pH values. Besides de-ionized water (pH 5.5), Britton– Welford titrated solutions (0.2 m KH2PO4 with the desired volume of 0.2 m NaOH) were employed to obtain the required pH values. In the case of complexes prepared at pH 7.8, CD solutions were dissolved in the corresponding buffered aqueous solutions.However, for pH 11.0 this was obtained by addition of a suitable amount of NaOH to the aqueous CD solution. The final concentration of b-carboline in the different solutions was 1.0 3 1026 m. The b-carboline–CD solutions were stirred magnetically for 18–48 h in order to obtain the inclusion complexes. Acid–base equilibria in aqueous and CD solutions were studied using the above-mentioned procedure to prepare the inclusion complexes.When the complexes were obtained, successive aliquots of 10 ml of NaOH (10 m) were added in order to study the shifts in the proton transfer processes. Fluorescence quenching of b-carbolines was studied using bromide ion (NaBr) as quencher. NaCl was added to the complex solutions to achieve a constant ionic strength (1.0 m). Aliquots of 10 ml of the NaBr solution were added to study the quenching effect. The concentration of bromide ion varied in the complex solution from 0.001 to 0.1 m.Results and Discussion We studied the influence of CD complexation on the acid–base equilibria of harmine and harmane by spectrofluorimetry, believing that this process could seriously affect the determination of these compounds when CDs are employed. Figs. 2 and 3 show the emission spectra of harmane in ethanolic and aqueous solutions. In ethanolic solution, emission bands appear at 360 and 380 nm, which can be attributed to the neutral form. Addition of small amounts of HCl (1 m) caused the appearance of a band at 430 nm, which can be attributed to the cationic form. Addition of NaOH (1 m) induced a decrease in the intensity of the neutral band, together with a very weak increase in the emission at 480 nm which, according to other workers,1,3,5 is due to the formation of a zwitterionic species. The addition of more concentrated NaOH solutions (10 m) or solid NaOH did not cause the appearance of an anionic band.This happened also when harmine and harmane were dissolved in other organic solvents (hexane or propan-1-ol), and also when triethylamine was added as a base.In agreement with observations by Sakurovs and Ghiggino,5 the excitation spectra were the same for cationic, neutral and zwitterionic species. The same tests were performed in aqueous solutions (Fig. 3) and the behaviour was different, since the cationic band only appeared in the aqueous solution at 430 nm, and the addition of acid (1 m HCl) increased slightly this emission corresponding to the cationic band.Addition of NaOH (1 m) produced the zwitterionic band (480 nm) with a notable fluorescence intensity compared with the ethanolic solution and the neutral (370 nm) band. It is important to note that in all solvents studied the emission intensity of the cationic band was considerably higher than that of the corresponding neutral or zwitterionic bands. Addition of more concentrated NaOH (10 m) did not cause the anionic band to appear, because it is necessary to work outside the pH scale.1,2 Fig. 2 Uncorrected excitation and emission spectra of harmane in ethanolic solution (1). Same sample after addition of small amounts of HCl (2) and NaOH (3). Fig. 3 Uncorrected excitation and emission spectra of harmane in aqueous solution (1). Same sample after addition of small amounts of HCl (2) and NaOH (3). The fluorescence intensity for solution 1 is three times lower than that for solution 2 and that for solution 3 is 81 times lower than that for solution 2. 46 Analyst, January 1997, Vol. 122Inclusion complexes with the different CDs were prepared for both b-carbolines studied. Considering the acid–base behaviour of these compounds, the increase in the emission intensity and the shifts in the emission maxima or changes in the fluorescence lifetime or fluorescence quantum yield are proof of the inclusion processes.19 With this in mind, we tried to prepare the inclusion complexes at different pH values (5.5, 7.8, 11.0).Thus, at pH 5.5 only the cationic form was observed, with a weak (5%) increase in the fluorescence intensity of the complex solution with respect to the aqueous solutions of harmine and harmane. Therefore, the formation of inclusion complexes with cationic species is unlikely. This may be due to the fact that cationic species are water soluble and therefore their tendency to be included is low. In the case of HPb-CD, after several days under magnetic stirring the emission of the neutral form together with the predominant cationic emission were observed.A pH of 7.8 was selected because it is very close to the pKa values of harmine and harmane and under such conditions the same concentration of the cationic and neutral species should be present in the ground state. Therefore, inclusion of the neutral species should be favoured, shifting the acid–base equilibria. Under such conditions we obtained the neutral form with a fluorescence intensity higher than that in water.The spectral shape resembles that observed for the neutral form in ethanol; however, in the presence of the CDs it is better resolved and two peaks appear. When complexes were prepared at pH 11.0, the emission corresponding to the neutral form could be observed, but some differences in the spectral shape with respect to those obtained at pH 7.8 can be noted. Thus, the neutral band is not resolved into two peaks and only a peak at 365 nm appears, together with two shoulders at 380 and 420 nm.These changes in the spectral profile may be associated with the existence of non-complexed zwitterionic or anionic species which are present at such pH values in the ground state. However, the additions of NaOH show that this emission corresponds to the cationic form, which is probably produced in the excited state at this pH value. Excitation produces the cationic form owing to very rapid proton transfer in the excited state, as described by Sakurovs and Ghiggino.5 The excited states of b-carbolines are strong bases and deprotonate water.5 From these results, it can be deduced that pH 8.0 is the most suitable for studying the inclusion complexes.HP-b-CD was selected as a model CD to study the time necessary to obtain the inclusion complexes. The complexes with harmine and harmane were prepared following the experimental procedure described, and at pH 7.8–8.0 the b-carboline–HP-bCD complexes showed the neutral band only after 18–24 h under magnetic stirring.A time span of 24 h was chosen to obtain comparable results for the different complexes. At pH 11.0 the results did not improve compared with those obtained at pH 8.0 and, as these conditions can affect the CDs by hydrolysis,20 they were not used in subsequent experiments. Fig. 4 shows the emission spectra for the different harmane–CD complexes using the optimized procedure to obtain the complexes.After 24 h of magnetic stirring the spectra corresponding to the neutral form are present for modified b-CD (HPb-CD, DMb-CD and TMb-CD). For b- CD, a mixture of cationic and neutral forms exists but the equilibrium is shifted towards the neutral form because the intensity of the cationic band is very small. For g-CD, on the other hand, a mixture was also obtained at the same pH values, but the ratio of the species in equilibrium is the opposite to that observed with b-CD. As can be seen in Fig. 4, the shift in the acid–base equilibria depended on the type of CD since they displaced the acid-base equilibria in a different way because of different efficiencies in including the neutral form. In homogeneous aqueous solution, the fluorescence intensity of the cation is about 10 times higher than that of the neutral form, with a considerable difference in quantum yields1 [FF (cation) = 0.76 and FF (neutral) = 0.17], which makes it difficult to observe in aqueous or alkaline solution owing to the protonation of the excited neutral species by rapid proton exchange with the solvent.5 The formation of complexes at pH 7.8–8.0 induced the appearance of emission bands corresponding to the neutral form.This is an important proof of the formation of an inclusion complex because the excited states are protected inside the cavity of CDs and proton transfer is hampered. The fluorescence intensity for the neutral band of harmane in the complexes is only 2–3 times lower than that of the cation.This enhancement in the neutral emission shows the existence of inclusion complexes and also increases the analytical sensitivity. Fig. 5 shows the titration of harmane–HPb-CD complexes with NaOH. The complexes were prepared according to the experimental procedure at pH 7.8. It can be seen that the emission band for the neutral form is resolved into two peaks, and this does not happen in homogeneous aqueous solution.Increasing amounts of NaOH produced a notable decrease in the intensity of the neutral band with the appearance and increase of the corresponding anionic band; the resolution was better than in 14 m KOH1,2 and consequently the alkaloid remained inside the CD. The isoemissive point at 400 nm shows the existence of only two species (neutral and anionic) in the acid–base equilibria. The formation of an anion was not observed in aqueous solution except outside the pH scale and was a consequence of the deprotonation of the pyrrolic nitrogen.We have described previously the anion emission in micellar aqueous solutions of cetyltrimethylammonium bromide.20 Harmane –HPb-CD complexes were the only examples where the formation of an anionic species was observed and this can be attributed to the effects that the CD environment has on the acid–base properties of the guest molecules. Therefore, HPb- CD can isolate harmane from the aqueous environment, hampering proton transfer from the solvent in the excited state.When the neutral form is included in the different CDs, it is not possible to obtain the cationic form after excitation as in homogeneous aqueous solution. Fig. 6 shows the emission spectra of the harmane–b-CD complex at pH 8.0. In the case of b-CD, it was not possible to shift the inclusion equilibrium to obtain only the neutral form, Fig. 4 Uncorrected excitation and emission spectra of the complexes obtained from harmane and the different CDs studied at pH 7.8: (1) b-CD; (2) HPb-CD; (3) DMb-CD; (4) TMb-CD and (5) g-CD.Analyst, January 1997, Vol. 122 47either for harmane or for harmine. In such cases, when NaOH was added an increase in the fluorescence intensity for the neutral band was initially produced. This was followed by the transformation of the neutral form into the zwitterionic species, as shown by the appearance of the emission corresponding to this form. This behaviour can be explained by considering that the cationic form is solubilized in water and that its acid–base equilibrium is similar to that operating in aqueous solution or, alternatively, considering that the indolyl moiety is included with the pyridine end protruding from the CD cavity.This hypothesis is especially valuable in the case of harmine, where the steric hindrance due to the methoxy group on the indolyl moiety makes the inclusion difficult. The effects of NaOH additions on DMb-CD and TMb-CD complexes are very similar.The inclusion complexes produced the emission band corresponding to the neutral form but a tail was observed in the region of the cationic band (Fig. 4). When NaOH was added, a notable decrease was produced in the neutral band with a very weak increase in the emission at 480 nm (zwitterionic species). The absence of an isoemissive point indicates the presence of cationic, neutral and zwitterionic species in the equilibria.The different behaviour of the complexes with increasing amounts of NaOH causes a decrease in the intensity corresponding to the neutral form and an increase in the emission corresponding to the anionic form (in the case of HPb-CD) or the emission of the zwitterionic form for the other CDs. This means an important change in the acid–base behaviour of the b-carbolines as a consequence of the inclusion and the associated selectivity of CDs to include specific species of harmine or harmane.Table 1 summarizes the spectrofluorimetric characteristics of the inclusion complexes studied. While the emission corresponding to the neutral band is present in all cases, for b-CD and g-CD the maximum corresponding to the cationic band in the emission spectra is also present. The presence of this peak is a consequence of the geometrical characteristics of b-CD and g-CD. Thus, b-CD is too small to accommodate b-carboline completely and the molecules are included but the pyridine moiety protrudes from cavity and is protonated by water.The cavity of g-CD is large enough to accommodate harmane and harmine but water molecules may penetrate into the cavity, producing the cationic band when the neutral species are excited. In order to obtain more detailed information concerning the protection afforded by CDs to the excited states, fluorescence quenching of the b-carbolines–CD complexes was tested. This study was carried out for the neutral, cationic and zwitterionic forms obtained after addition of HCl or NaOH to the neutral complexes.In some complexes the correlation coefficient obtained using the Stern–Volmer treatment was not adequate, showing that other quenching mechanisms (static, collisional, energy transfer, etc.) should be present. However, the presence of bromide ion decreased the fluorescence intensity of the complexes. Table 2 shows that neutral forms are effectively protected in the harmane–CD complexes because the slope obtained was lower than that in homogeneous ethanolic solutions where only the neutral band is present.The microenvironments provided by CDs and homogeneous ethanolic solutions are similar.21 Nevertheless, the slope for these complexes was higher than that obtained in water. The emission Fig. 5 Spectrofluorimetric study of the changes in the acid–base equilibrium for the harmane–HPb-CD inclusion complex prepared in buffered aqueous solution (pH 7.8) with addition of 10 ml of 10 m NaOH.Fig. 6 Spectrofluorimetric study of the changes in the acid–base equilibrium for the harmane–b-CD inclusion complex prepared in buffered aqueous solution (pH 7.8) with addition of 10 ml of 10 m NaOH. The dotted line corresponds to the starting solution of the complex. Table 1 Fluorescence characteristics obtained for the CD inclusion complexes with harmane and harmine. The symbols > and < mean the relative fluorescence intensity of one maximum with respect to another Complex Harmane–HPb-CD Harmane–DMb-CD Harmane–TMb-CD Harmane–b-CD Harmane–g-CD lem/nm 362 > 380 362 > 380 362 > 380 362 > 380 > 430 362, 380 < 430 Complex Harmine–HPb-CD Harmine–DMb-CD Harmine–TMb-CD Harmine–b-CD Harmine–g-CD lem/nm 358, 370 358, 370 358, 370 358, 370 > 416 358 < 370 < 416 48 Analyst, January 1997, Vol. 122of the cationic form is also effectively protected against the quencher because the slopes are lower than in water and consequently it is possible that cationic species are partially included.However, in the case of the zwitterionic species a decrease in the fluorescence intensity was produced in the presence of bromide ion, but this phenomenon does not fit the Stern–Volmer relationship and the slopes for the complexes are higher than in aqueous solution. For the harmine–CD complexes, the neutral form presents a weaker quenching effect than in ethanolic solution.The cationic form is also less quenched compared with the corresponding aqueous solution, although the slopes are higher than for harmane complexes. This is certainly associated with the geometrical characteristics of the inclusion complexes. It can also be considered that a small fraction of cationic harmine is included and that the free cationic harmine is quenched as in aqueous solution. It is remarkable that with b-CD the slope is higher than in water, in contrast to the behaviour of g-CD, where the slope is strongly reduced.The differences in the geometries of harmine and harmane produce different inclusion complexes. Thus g-CD can include harmine by the methoxyindole moiety, but the size of the methoxy group does not allow such an inclusion in b-CD or in modified b-CDs. In conclusion, the existence of the inclusion complexes was verified by the changes in the spectrofluorimetric properties, in the acid–base behaviour and by the protection against the effects of the quenching.Nevertheless, the geometry of the inclusion complexes of harmine and harmane can be different because in the case of harmane the indole moiety can be included with the pyridine moiety protuding from the cavity. For harmine this is not easy and a more difficult penetration into the cavity is expected. Considering this selectivity of CDs, their use in chromatography can contribute to enhancing the chromatographic separation of b-carbolines owing to the frequent use of HPLC with fluorimetric detection in b-carboline determination.17,18 References 1 Bal�on, M., Hidalgo, J., Guardado, P., Mu�noz, M.A., and Carmona, C., J. Chem. Soc. Perkin Trans. 2, 1993, 99. 2 Bal�on, M., Mu�noz, M. A., Hidalgo, J., Carmona, M. C., and S�anchez, M., J. Photochem., 1987, 36, 193. 3 Tomas, F., Zabala, I., and Olba, A., J. Photochem., 1983, 23, 355. 4 Vander Donckt, E., Prog. React. Kinet., 1970, 5, 274. 5 Sakurovs, R., and Ghiggino, K. P., J. Photochem., 1982, 18, 1. 6 Bender, M. L., and Komiyama, M., Cyclodextrin Chemistry, Springer, Berlin, 1978. 7 Sba�ý, M., Ait Lyazidi, S., Lerner, D. A., del Castillo, B., and Mart�ýn, M. A., Anal. Chim. Acta, 1995, 303, 47. 8 Chattopadhyay, N., J. Photochem. Photobiol. A, 1991, 58, 31. 9 Takahashi, K., J. Chem. Soc. Chem. Commun., 1991, 929. 10 Connors, K. A., and Lipari, J. M., J. Pharm. Sci., 1976, 65, 379. 11 Lin, S. F., and Connors, K. A., J. Pharm. Sci., 1983, 72, 1333. 12 Eftink, M. R., Andy, M. L., Bystrom, K., Perlmutter, M. D., and Kristol, D. S., J. Am. Chem. Soc., 1989, 111, 6765 13 Connors, K. A., and Rosanske, T. W., J. Pharm. Sci., 1980, 69, 173. 14 Uekama, K., Hirayama, F., Nasu, S., and Matsuo, N., Chem. Pharm. Bull., 1978, 26, 3477. 15 Abramovitch, R. A., and Spenser, I. D., Adv. Heterocycl. Chem., 1964, 3, 79. 16 Dillon, J., Spector, A., and Nakanishi, K., Nature (London), 1976, 259, 422. 17 Bossin, T. R., and Faull, K. F., J. Chromatogr., 1988, 428, 229. 18 Moncrieff, J., J. Chromatrogr., 1989, 496, 269. 19 Szejtli, J., Cyclodextrins and Their Inclusion Complexes, Akad�emiai Kiad�o, Budapest, 1982. 20 Mart�ýn, L., Mart�ýn, M.A., and del Castillo, B., J. Fluorescence, in the press. 21 Frankewich, R. P., Thimmaiah, K. N., and Hinze, W. L., Anal. Chem., 1991, 63, 2924. Paper 6/02790C Received April 22, 1996 Accepted October 10, 1996 Table 2 Fluorescence quenching study of b-carboline–CD inclusion complexes b-Carboline–CD lem * r† m† b† Harmane–EtOH x 0.947 4.759 1.035 Harmane–H2O x 0.978 0.558 0.958 y 0.984 1.048 0.98 z 0.996 7.7445 0.9552 Harmane–HPb-CD x ‡ ‡ ‡ y ‡ ‡ ‡ z 0.91 5.4 1.03 Harmane–DMb-CD x 0.95 0.858 1.034 y ‡ ‡ ‡ z 0.97 4.454 1.077 Harmane–TMb-CD x 0.949 4.23 1.021 y ‡ ‡ ‡ z 0.95 5.2 1.04 Harmane–g-CD x 0.96 1.037 0.955 y ‡ ‡ ‡ z 0.94 4.17 1.04 Harmane–b-CD x 0.953 0.916 0.999 y 0.985 1.525 1.019 z 0.994 3.547 1.044 Harmine–EtOH x 0.9873 9.63 1.066 Harmine–H2O x 0.969 0.68 0.98 y 0.986 5.109 0.986 z 0.998 15.768 0.986 Harmine–HPb-CD x 0.974 1.83 1.033 y 0.846 1.23 1.082 z 0.965 9.47 1.085 Harmine–DMb-CD x 0.96 1.737 1.04 y 0.98 0.866 1.0007 z 0.994 11.27 1.034 Harmine–TMb-CD x 0.96 2.69 1.07 y 0.966 1.373 1.066 z 0.958 10.71 1.096 Harmine–g-CD x 0.814 0.32 1.045 y 0.97 2.2 1.052 z 0.83 3.7 1.23 Harmine–b-CD x 0.95 3.7 1.02 y 0.92 1.73 1.097 z 0.977 16.02 1.22 * l = Emission wavelength. x: l = 366 nm (harmane); l = 358 nm (harmine). y: l = 480 nm (harmane; l = 476 nm (harmine). z: l = 430 nm (harmane); l = 416 nm (harmine). † r = correlation coefficient; m = slope; b = intercept inordinate. ‡ The values obtained could not be adjusted by the Stern–Volmer treatment: Fo/FA = 1 + K(Q). Analyst, Jan

ISSN:0003-2654    

DOI:10.1039/a602790c

出版商:RSC    

年代:1997

数据来源: RSC

摘要:

Biomarkers in Hydrolysed Urine, Plasma and Erythrocytes Among Workers Exposed to Thermal Degradation Products From Toluene Diisocyanate Foam Pernilla Lind, Marianne Dalene, Håkan Tinnerberg and Gunnar Skarping* Department of Occupational and Environmental Medicine, University Hospital, S-221 85 Lund, Sweden Blood and urine samples were collected from six workers and two volunteers exposed to thermal degradation products from toluene diisocyanate (TDI)-based polyurethane (PUR) before and during the summer vacation.Air samples were collected on filters impregnated with 9-(N-methylaminomethyl)anthracene. The concentrations of the amines corresponding to 2,4- and 2,6-TDI, i.e., 2,4- and 2,6-toluenediamine (TDA), were determined in urine (U-TDA), plasma (P-TDA) and erythrocytes (E-TDA) after acid hydrolysis as pentafluoropropionic anhydride derivatives by GC–MS. Among the workers urinary elimination phases were seen. The estimated medians of the urinary half-lives were for the slow phase 18 d for 2,4-TDA and 19 d for 2,6-TDA.P-2,4-TDA ranged between 2.5 and 19 ng ml21 and P-2,6-TDA between 4.4 and 30 ng ml21. The estimated median of the half-lives in plasma were 7.8 d for 2,4-TDA and 9.6 d for 2,6-TDA. E-2,4-TDA ranged between 0.5 and 6.6 ng g21 and E-2,6-TDA between 1.2 and 14 ng g21. A significant linear relationship was found between the mean P-TDA and the mean E-TDA. Linear relationships were observed between the mean daily U-TDA and P-TDA and E-TDA.Virtually linear relationships were obtained for P-TDA and E-TDA and the TDI air levels. Proteins from lysed erythrocytes were separated and fractionated by gel filtration. ‘TDI’-modified proteins were found in six out of a total of 80 fractions (fractions 51–56). These co-eluted completely with the haemoglobin (UV, 415 nm). Fractions 51–56 contained 89% of the applied amounts of 2,4-TDA and 81% of 2,6-TDA. Keywords: Biomarkers; gas chromatography–mass spectrometry; toluene diisocyanate; toluenediamine; thermal degradation Occupational exposure to isocyanates is associated with respiratory disease.1–3 The main applications of isocyanates are in the manufacture of elastomers, rigid and flexible polyurethane (PUR) foams, glues and lacquers.Toluene diisocyanate (TDI) dominates in the production of flexible foam and the 2,4- and 2,6-isomers are used typically mixed in a ratio of 80 : 20. Exposure to isocyanates may occur when PUR is thermally degraded.The presence of high concentrations of complex isocyanates, aminoisocyanates and amines has recently been observed in the work atmosphere.4 The methods available for the determination of biomarkers are based on the determination of the amines corresponding to the isocyanates in hydrolysed biological fluids, e.g., 2,4- and 2,6-toluenediamine (TDA).5,6 A detection limit of 0.05 mg l21 of TDA and a precision of 3.5% for 2,4-TDA and 1.6% for 2,6-TDA were found.The amines are liberated under strongly acidic or alkaline hydrolysis conditions, indicating the presence of adducts to proteins and peptides.7 Adducts to plasma albumin among workers exposed to TDI8 and haemoglobin among workers exposed to methylenediphenyl diisocyanate and methylenedianiline have been demonstrated.9 In a study of workers occupationally exposed to TDI, the concentrations of TDA in plasma showed only limited variation during work-days while the concentrations in urine varied considerably.10 The estimated mean half-lives in plasma were 21 (range 14–34) d for 2,4-TDA and 21 (16–26) d for 2,6-TDA, and the urinary half-lives were 5.8–11 d for 2.4-TDA and 6.4–9.3 d for 2,6-TDA.11 Volunteers exposed to TDI showed urinary half-lives of about 1.9 h for 2,4-TDA and 1.6 h for 2,6-TDA in a first phase and of 5 h in a second phase12 and the half-lives in plasma from volunteers were 2–5 h of 2,4- and 2,6-TDA in a first phase and > 6 d in a second phase.13 The aim of this study was to determine the concentrations of TDA in hydrolysed urine (U-TDA), plasma (P-TDA) and erythrocytes (E-TDA) and their relationships in workers exposed to thermal degradation products of TDI-PUR before and during an exposure-free period.A knowledge of the relationships and variations in time between U-TDA, P-TDA and E-TDA is necessary for the sampling strategy and exposure control. Experimental Apparatus The hydrolysed biological samples were analysed with an MD 800 quadrupole mass spectrometer (Fisons Instruments, Altrincham, UK) connected to a GC 8000 gas chromatograph equipped with an A200S autosampler (Fisons Instruments, Milan, Italy).The 2,4- and 2,6-TDA-pentafluoropropionic anhydride (PFPA) derivatives were determined by monitoring the ions at m/z 394 corresponding to the M 2 20 ions and m/z 397 corresponding to the M 2 20 ions of the trideuteriumlabelled internal standards (M = molecular mass).A fusedsilica capillary column with a chemically bonded stationary phase, DB-5 (J & B Scientific, Folsom, CA, USA), 25 m 30.25 mm id with a film thickness of 0.25 mm, was used. Detailed chromatographic and mass spectrometric information was given previously.5 A Waters Model 600 multisolvent delivery system (Millipore –Waters, Milford, MA, USA), a Waters Model 712 WISP and a Waters Model 490 UV detector were used for the determination of isocyanates in air and for the gel separation of haemoglobin.Haemoglobin was fractionated using a Pharmacia –LKB SuperFrac (Pharmacia, Uppsala, Sweden). Chemicals Both 2,4- and 2,6-TDA were obtained from Fluka (Buchs, Switzerland), HCl, NaOH, H2SO4 and K2HPO4 from Merck (Darmstadt, Germany), pentafluoropropionic anhydride (PFPA) from Pierce (Rockford, IL, USA), toluene, isooctane and HPLC-grade acetonitrile from Lab-Scan (Dublin, Ireland), technical-grade TDI (80 + 20 2,4- + 2,6-TDI) and triethylamine (99%) from Janssen (Beerse, Belgium), NH4HCO3 from BDH Analyst, January 1997, Vol. 122 (51–56) 51(Poole, UK), 9-(N-methylaminomethyl)anthracene (MAMA) from Aldrich Chemie (Steinham, Germany), haemoglobin from Sigma (St. Louis, MO, USA) and trideuterated 2,4- and 2,6-TDA [CD3C6H3(NH2)2, TDDA], used as internal standards (IS) from Synthelec (Lund, Sweden). Subjects and Exposure Six workers (A–F) and two volunteers (colleagues) participated in this study. The work took place in a flame lamination factory applying a thin layer of TDI-based flexible foam on to textile fabric.The surface of the PUR foam was partly melted by an open flame immediately prior to the application. The work took place (7 am–4 pm) in two large workshops with an open door in between. Workers A, D, E and F were involved in the lamination process and workers B and C were occupied with cutting the produced laminated fabric. The lamination was performed in both workshops, but not at the same time.Worker D was employed only during the summer and was assumed not to be chronically exposed. The two volunteers were mainly involved in collecting air samples. Procedure Preparation of standard solutions TDI standard solutions were prepared by dissolving TDI in isooctane and further dilutions were made in toluene. Standard solutions of 2,4- and 2,6-TDA were prepared by dissolving accurately weighed amounts in 1 m HCl. The internal standards, 2,4- and 2,6-TDDA, were dissolved in 1 m H2SO4.Air sampling on impregnated glass-fibre filters Three subsequent air samples were collected in the breathing zone per worker during one typical work day of the study during about 8 h. The sampling times of each air sample varied between 100 and 220 min. The average exposure was calculated by taking the total amount of 2,4- and 2,6-TDI on the three filters divided by the total sampling time. Air samples were collected (1 l min21) on 13 mm glass-fibre filters, in a Teflon filter holder, impregnated with a MAMA–glycerol mixture.14 Immediately after sampling, the filters were placed in a test-tube containing 4 ml of toluene. The excess of the MAMA reagent was extracted for 5 min with 2 ml of 0.1 m HCl added to the testtube.A 2 ml volume of the toluene solution was separated and evaporated to dryness. The dry residue was dissolved in 1 ml of the mobile phase added before the LC–UV (254 nm) determinations. The LC column was a Hypersil BDS C18 (150 3 4.6 mm id) with 5 mm particles.The mobile phase (1.5 ml min21) consisted of acetonitrile–water (70 + 30) containing 3% triethylamine in the water and adjusted to pH 3 with phosphoric acid. Peak-height measurements were made and compared with a calibration plot for reagent solutions spiked with TDI to six concentrations in the range 0–1.2 mg ml21. Collection of biological samples Blood and urine samples were collected from six workers (A–F) before and during their summer vacation.Sampling was performed both at the factory and in the workers’ homes. All urine produced during 2 d of exposure and on three exposurefree days were collected. A few urine samples were also collected on different occasions during the 3–4 week vacation. Blood samples were taken once a day during 2 d of exposure and on three exposure-free days. Blood and urine samples were also taken from two volunteers. All urine produced during the day of exposure and the following exposure-free day was collected.Blood samples were taken on three occasions at the day of exposure and two and five times, respectively, during the following eight exposure-free days. Sampling, handling and storing of biological samples All urine samples were collected in polyethylene bottles. The creatinine concentration [determined at the Department of Clinical Chemistry, University Hospital, Lund, by use of Kodak Ektachem clinical chemistry slides (CREA) and Kodak Ektachem 700 XR-C analyser], pH and total volume of each urine sample were determined.Blood was sampled in heparinized tubes (Venoject). Plasma was separated within 8 h and transferred to new tubes. The erythrocytes were washed in 0.9% NaCl two or three times. The urine, plasma and erythrocyte samples were kept frozen at 220 °C until analysis. Gel filtration of haemoglobin Washed erythrocytes from worker A were repeatedly frozen and thawed four times in a double volume of Millipore-purified water (Milli-Q system, Millipore, Bedford, MA, USA).The cell debris was removed by centrifugation at 18 500 g for 15 min. The haemoglobin-containing supernatant was diluted five fold in 50 mm NH4HCO3 and filtered using a 0.2 mm hydrophilic membrane filter (Minisart RC 25; Sartorius, G�ottingen, Germany). A volume of 200 ml of the solution was separated on a Superose 12 (10/30) gel filtration column (Pharmacia Biotech, Uppsala, Sweden) and fraction into 80 fractions of 0.25 ml each.The mobile phase consisted of 50 mm NH4HCO3 at a flow rate of 0.5 ml min21. UV traces were recorded at 415 nm. A total of 50 fractionations were performed and fractions were collected in the same subsequent fraction vials. Hence each fraction vial contained 12.5 ml. The 80 fractions were then evaporated to dryness and the dry residues were dissolved in 1 ml aliquots of Millipore-purified water. Hydrolysis To 1 ml of plasma, 1 ml of urine, 0.5 g of erythrocytes in 0.5 ml of Millipore-purified water or 1 ml of concentrated gel filtration fractions, 100 ml of IS were added together with 1.5 ml of 3 m H2SO4 before hydrolysis at 100 °C for 16 h.In the sample batches the calibration samples were prepared in the range 0–30 ng ml21 TDA with 5.5 ng ml21 IS for plasma, 0–10 ng ml21 TDA with 2.0 ng ml21 IS and 0–60 ng ml21 TDA with 11 ng ml21 IS for urine, 0–10 ng ml21 TDA with 2.0 n ml21 IS for erythrocytes and 0–5 ng ml21 TDA with 1.0 ng ml21 IS for the gel filtration fractions.After hydrolysis the samples were worked-up as described below. Work-up procedure A 5 ml volume of saturated NaOH and 2 ml of toluene were added to the samples. The mixtures were shaken for about 10 min and then centrifuged at 1500 g for 10 min. The organic layers were transferred into new test-tubes and 20 ml of PFPA were added to each. The mixtures were immediately shaken vigorously. The excess of the reagent and acid formed was removed by extraction with 2 ml of 1 m phosphate buffer solution (pH 7.5).The toluene layers containing the amide derivatives and the internal standards were transferred into 1.5 ml autosampler vials with Teflon seals. When analysing the gel filtration samples, about 1 ml of the toluene solution was evaporated to dryness and diluted with 100 ml of toluene. The samples were then ready for injection into the GC–MS system. Statistics To determine if the correlation coefficients are significant, we calculated a t-value: t = ýrý (n 2 2)1 2/(1 2 r2)1 2 (r = correlation 52 Analyst, January 1997, Vol. 122coefficient, n = number of observations). The calculated value of t was compared with the tabulated value at the 95% significance level, using a two-tailed t-test and n 2 2 degrees of freedom. If the calculated value of t was greater than the tabulated value, a significant correlation exists. Results Isocyanates in Air The average air concentrations during the work day of 2,4- and 2,6-TDI at the breathing zone using personal sampling were for worker A 3.5 and 10 mg m23, B 0.9 and 2.6 mg m23, C 1.8 and 6.5 mg m23, D 1.4 and 4.7 mg m23, E 1.6 and 4.1 mg m23 and F 2.1 and 7.4 mg m23, respectively.No representative air samples for the volunteers were collected. Biomarkers in Hydrolysed Urine The U-TDA for workers A–F are shown in Fig. 1. During the work U-TDA was rapidly eliminated and a slower elimination phase were seen during the exposure-free period.For workers B and E one urine sample was considerably above the assumed ideal elimination curve. This was probably due to an unnoticed exposure to TDA dust at the factory after cessation of flame lamination work. The urinary half-lives in the slower phase were estimated by plotting the natural logarithm of the urinary concentrations against time and by dividing ln 2 with the calculated slopes. The half-lives of 2,4- and 2,6-TDA were estimated for worker A to be 32 (r = 0.82, n = 10) and 25 d (r = 0.94, n = 10), B 20 (r = 0.67, n = 23) and 20 d (r = 0.90, n = 23), C 15 (r = 0.93, n = 9) and 18 d (r = 0.91, n = 9), D 4.3 (r = 0.90, n = 8) and 11 d (r = 0.79, n = 8), E 26 (r = 0.36, n = 23) and 22 d (r = 0.88, n = 24) and F 14 (r = 0.93, n = 12) and 16 d (r = 0.97, n = 12), respectively.The medians of the urinary half-lives were for the slow phase 18 d for 2,4-TDA and 19 d for 2,6-TDA. The urinary half-lives for the two volunteers were calculated to 5.3 (r = 0.96, n = 4) and 6.2 h (r = 0.97, n = 4) h for 2,4-TDA and 8.4 (r = 0.96, n = 5) and 7.4 h (r = 0.99, n = 4) for 2,6-TDA (Fig. 2). Biomarkers in Plasma and Erythrocytes The concentrations of 2,4-TDA in plasma were in the range 2.5–19 ng ml21 and those of 2,6-TDA in the range 4.4–30 ng ml21 among the workers. In erythrocytes, the concentrations of 2,4-TDA were in the range 0.5–6.6 ng g21 and those of 2,6-TDA in the range 1.2–14 ng g21 among the workers.Significant linear relationships were seen between the mean concentrations, for all individual samples, in plasma and erythrocytes of 2,4-TDA (r = 0.97, n = 6) and 2,6-TDA (r = 0.97, n = 6) (Fig. 3). P-TDA declined slowly after work cessation. The half-lives in plasma (calculated as for UTDA) were in the range 6.5–12 d for 2,4-TDA (r = 0.76–0.95, n = 4) and 7–11 d for 2,6-TDA (r = 0.76–0.97, n = 4). E-TDA declined too little for the calculation of half-lives.When plotting the mean daily urinary concentrations adjusted for creatinine against P-TDA and E-TDA, relationships were observed. The relationships became more linear with the number of days after cessation of work (Fig. 4). Significant correlations were seen, with the exception of E-2,4-TDA on day 2. Significant linear plots, with the exception of E-2,6-TDA, were obtained on plotting the P-TDA and E-TDA for workers A, B, C, E and F against the TDI air levels measured on day 1 (worker D was only employed during the summer) (Fig. 5). No detectable P-TDA or E-TDA were found in the two volunteers, except for very low concentrations of 2,4-TDA ( < 0.33 ng ml21) in plasma from one of the volunteers. Among a group of 17 unexposed humans, the-TDA and P-TDA were below the detection limit. Conditions for the Hydrolysis of Haemoglobin Volumes of 1 ml of the haemoglobin-containing supernatant (see Gel filtration of haemoglobin) were hydrolysed in 1.5 ml of 3 m H2SO4 at 100 °C for 4, 8, 16, 24 and 48 h.After hydrolysis the samples were subjected to the work-up procedure. No free 2,4- and 2,6-TDA were seen without hydrolysis. The release of 2,4- and 2,6-TDA increased with increasing hydrolysis time (Fig. 6). For the preparation of haemoglobin from the erythrocytes, the recoveries of 2,4- and 2,6-TDA were 77% and 91%, respectively. On analysing the remaining cell debris in the centrifuge tube no TDA was found. Gel Filtration of Proteins in erythrocytes TDA were only found in six of the 80 fractions (fractions 51–56) from the gel filtration separation and co-eluted com- Fig. 1 Concentrations of 2,4-TDA (2) and 2,6-TDA (5) in hydrolysed urine (mg per mmol of creatinine) from workers A–F.All urine produced during 2 work days and on the three days following work cessation (vacation), and on different occasions during the following 3–4 weeks, was collected. Time 0 = 6.00 am on the first day of the study.Analyst, January 1997, Vol. 122 53pletely with the haemoglobin (UV, 415 nm) peak. Fractions 51–56 contained 89% of the applied amounts of 2,4-TDA and 81% of 2,6-TDA (Fig. 7). In the GC–MS traces pure and well separated 2,4- and 2,6-TDA-PFPA peaks were seen. The 2,4-and 2,6-TDA-PFPA peaks in the chromatogram of fraction 54 were found at an unfiltered S/N of 175 and 61, respectively (rms). Only traces of TDA-PFPA were seen in fractions 1–50 and 57–80, due to impurities of TDA in the internal standards.These were also present in the chemical blanks. Discussion Isocyanates occur in many chemical and physical forms in workplace atmospheres. In the TDI foam industry, the monomers dominate in the gas phase. However, when particles are formed, many other compounds may be present in addition to monomers. These can be measured by methods which are based on the reaction of isocyanates with secondary amines containing a chromophore.14 During the thermal decomposition of PUR, in addition to isocyanates, aminoisocyanates and amines are also present.This means that isocyanate exposure measured by ‘conventional’ methods, underestimates the true exposure. It was therefore surprising, but encouraging, to find a relationship between the air data and the P-TDA and E-TDA levels among Fig. 2 Concentrations of 2,4-TDA (2) and 2,6-TDA (5) in hydrolysed urine (mg per mmol of creatinine) from two volunteers. The volunteers were present in the factory from 7 am until 4 pm on day 1.Time 0 = 6.00 am the first day of the study. Fig. 4 Concentrations of 2,4-TDA (5/-) and 2,6-TDA (2/8) in hydrolysed plasma (ng ml21) and erythrocytes (ng g21) and mean urinary concentration (mg per mmol of creatinine) during 1 d of exposure (day 2) and the three following exposure-free days (days 3–5) among six workers (A–F). Day 2: r P- 2,4-TDA = 0.84, r E-2,4-TDA = 0.81, r P-2,6-TDA = 0.86, r E-2,6-TDA = 0.83; Day 3: r P-2,4-TDA = 0.91, r E-2,4-TDA = 0.91, r P-2,6-TDA = 0.94, r E-2,6-TDA = 0.89; Day 4: r P-2,4-TDA = 0.92, r E-2,4-TDA = 0.98, r P-2,6-TDA = 0.97, r E-2,6-TDA = 0.98; Day 5: r P-2,4-TDA = 0.98, r E- 2,4-TDA = 0.97, r P-2,6-TDA = 0.99, r E-2,6-TDA = 0.97. Continuous line = 2,4-TDA; broken line = 2,6-TDA.Fig. 3 Mean concentrations, for all individual samples, of 2,4-TDA (5) and 2,6-TDA (-) in hydrolysed plasma (ng ml21) and erythrocytes (ng g21) among six workers (A–F) during 5 d. Continuous line = 2,4-TDA, r = 0.97; broken line = 2,6-TDA, r = 0.97. 54 Analyst, January 1997, Vol. 122the workers. However, the relationship was different to that for data from a flexible foam factory,11 where about the same levels in plasma were found but three times higher TDI air levels. This can be explained by the insufficiency and the limitations of the air methods used to measure more complex isocyanates. Data from conventional air methods may therefore be used as a relative index of exposure in the absence of methods that take into account all other isocyanates that may influence the industrial hygiene. Further studies of these aspects are in progress.The urine data clearly demonstrate fast elimination phases and one slower phase. It can be assumed that the fast phases reflect the more recent exposure and the slow phase reflects urinary elimination of degradation products of modified proteins, and hence the history of exposure. The urinary elimination half-lives for the slow phase were about the same as the half-life of albumin and the mean half-lives in plasma among TDI foam workers.11 The significant linear relationships between the U-TDA and E-TDA and P-TDA that improved with time after work cessation may also add to the explanation of the presence of modified erythrocytes and albumin metabolites and breakdown products excreted in urine.These products seem to dominate after a few days of work cessation. Worker D, a worker not chronically exposed, had shorter urinary elimination half-lives compared with the chronically exposed workers.The volunteers in this study had about the same urinary halflives as the volunteers in the volunteer study.12 The volunteer study may reflect an interesting difference in uptake, metabolism and excretion between chronically exposed workers and volunteers and the difference in exposure situations. In the workplace studies much more complex isocyanates and different chemical and physical forms are present in the air.Biomarkers of both 2,4- and 2,6-TDA were found in plasma and erythrocytes from all workers. The half-lives in plasma, in Fig. 5 Concentrations of 2,4-TDA (5) and 2,6-TDA (-) among workers A, B, C, E and F and in worker D (2 = 2,4-TDA) and (8 = 2,6-TDA) in: (a), hydrolysed plasma (ng ml21); and (b), hydrolysed erythrocytes (ng g21), plotted against the mean individual air concentrations of 2,4- and 2,6-TDI (mg m23) for the first work day of the study.For the calculation of the linear curves, data from worker D were excluded as he was a seasonal worker and not chronically exposed (r P-2,4-TDA = 0.98, r P- 2,6-TDA = 0.97, r E-2,4-TDA = 0.92, r E-2,6-TDA = 0.85). Fig. 7 Amounts of 2,4-TDA (5) and 2,6-TDA (~) (ng) in the fractions eluted from a gel filtration separation of lysed erythrocytes (cell debris separated) from worker A. Fractions 51–56 contained 89% of the applied amount of 2,4-TDA and 81% of 2,6-TDA and co-eluted with haemoglobin (UV 415 nm, broken line, 100% was saturated detection).Selected ion monitoring chromatogram of fraction 43 represents a fraction not containing TDA, and that of fraction 54 one with ‘TDI’-modified compounds. (m/z = 394.2 is the M 2 20 ions of TDA-PFPA and m/z = 397.2 is the M 2 20 ions of TDDA-PFPA). Fig. 6 Concentrations of 2,4-TDA (5) and 2,6-TDA (-) in lysed erythrocytes (cell debris separated) after hydrolysis for 0, 4, 8, 16, 24 and 48 h at 100 °C.Analyst, January 1997, Vol. 122 55this study, were about 50% shorter than those in the flexible foam worker study11 but about the same as those in the volunteer study.13 The difference may be explained by the presence of a faster phase, but this could not be observed in the study of flexible foam workers as blood samplings were performed at too long intervals. The E-TDA levels were virtually stable during the study and therefore no half-lives could be calculated.They can therefore be assumed to be longer than for plasma. A significant linear relationship between PTDA and E-TDA was observed. No TDA was found in the centrifuged and lysed erythrocyte solution on performing the work-up procedure without hydrolysis. This indicates covalently TDI-modified proteins. The gel filtration demonstrated that the modified protein in fact was haemoglobin. The hydrolysis pattern of haemoglobin was found to be almost the same as for plasma among TDI-exposed workers.7 This is an indication that the nature of the chemical bondings is the same and that the same hydrolysis conditions can be used.There are many aspects of the uptake, metabolism and excretion of isocyanates that are not yet fully understood. However, there are many reasons to believe that modified proteins and peptides are involved in isocyanate-associated disease. It is therefore interesting to know that intracellular in addition to intercellular proteins are modified with TDI among workers.The mechanism by which isocyanates are able to pass cell membranes is not known and further studies on this aspect are in progress. The present study has shown that the history of exposure can be determined by biomarkers in plasma and erythrocytes. Recent exposure does not greatly affect the P-TDA and E-TDA levels. Biomarkers in urine can be used for the same purpose if sampling is performed after the fast elimination phases. This work was supported by the Swedish Work Environment Fund.We thank M. Adamsson, M. Spanne and T. Russin for skilful technical assistance and Professor A. Grubb and Associate Professor J.-O. Jeppson for valuable discussions. References 1 Banks, D. E., Butcher, B. T., and Salviaggio, J. E., Ann. Allergy, 1986, 57, 389. 2 Baur, X., Marek, W., Ammon, J., Czuppon, A. B., Marczynski, B., Raulf-Heimsoth, M., Roemmelt, H., and Fruhman, G., Int. Arch. Occup. Environ. Health, 1994, 7, 310. 3 Vandenplas, O., Malo, J. L., Saetta, M., Mapp, C. E., and Fabbri, L. M., Br. J. Ind. Med., 1993, 50, 213. 4 Tinnerberg, H., Spanne, M., Dalene, M., and Skarping, G., Analyst, 1996, 121, 1101. 5 Skarping, G., Dalene, M., and Lind, P., J. Chromatogr., 1994, 663, 199. 6 Maitre, A., Berode, M., Perdrix, A., Romazini, S., and Savolainen, H., Int. Arch. Occup. Environ. Health, 1993, 65, 97. 7 Lind, P., Skarping, G., and Dalene, M., Anal. Chim. Acta, in the press. 8 Lind, P., Dalene, M., Lindstr�om, V., Grubb, A., and Skarping, G., Analyst, 1997, 122, in the press. 9 Sch�utze, D., Sepai, O., Lewalter, J., Miksche, L., Henschler, D., and Sabbioni, G., Carcinogenesis, 1995, 16, 573. 10 Persson, P., Dalene, M., Skarping, G., Adamsson, M., and Hagmar, L., Br. J. Ind. Med., 1993, 50, 1111. 11 Lind, P., Dalene, M., Skarping, G., and Hagmar, L., Occup. Environ. Med., 1996, 53, 94. 12 Skarping, G., Brorson, T., and Sang�o, C., Int. Arch. Occup. Environ. Health, 1991, 63, 83. 13 Brorson, T., Skarping, G., and Sang�o, C., Int. Arch. Occup. Environ. Health, 1991, 63, 253. 14 Tinnerberg, H., Dalene, M., and Skarping, G., Am. Ind. Hyg. Assoc. J., in the press. Paper 6/06148F Received September 6, 1996 Accepted November 6, 1996 56 Analyst, January 1997, Vol. 122 Biomarkers in Hydrolysed Urine, Plasma and Erythrocytes Among Workers Exposed to Thermal Degradation Products From Toluene Diisocyanate Foam Pernilla Lind, Marianne Dalene, Håkan Tinnerberg and Gunnar Skarping* Department of Occupational and Environmental Medicine, University Hospital, S-221 85 Lund, Sweden Blood and urine samples were collected from six workers and two volunteers exposed to thermal degradation products from toluene diisocyanate (TDI)-based polyurethane (PUR) before and during the summer vacation.Air samples were collected on filters impregnated with 9-(N-methylaminomethyl)anthracene. The concentrations of the amines corresponding to 2,4- and 2,6-TDI, i.e., 2,4- and 2,6-toluenediamine (TDA), were determined in urine (U-TDA), plasma (P-TDA) and erythrocytes (E-TDA) after acid hydrolysis as pentafluoropropionic anhydride derivatives by GC–MS.Among the workers urinary elimination phases were seen. The estimated medians of the urinary half-lives were for the slow phase 18 d for 2,4-TDA and 19 d for 2,6-TDA. P-2,4-TDA ranged between 2.5 and 19 ng ml21 and P-2,6-TDA between 4.4 and 30 ng ml21. The estimated median of the half-lives in plasma were 7.8 d for 2,4-TDA and 9.6 d for 2,6-TDA.E-2,4-TDA ranged between 0.5 and 6.6 ng g21 and E-2,6-TDA between 1.2 and 14 ng g21. A significant linear relationship was found between the mean P-TDA and the mean E-TDA. Linear relationships were observed between the mean daily U-TDA and P-TDA and E-TDA. Virtually linear relationships were obtained for P-TDA and E-TDA and the TDI air levels. Proteins from lysed erythrocytes were separated and fractionated by gel filtration.‘TDI’-modified proteins were found in six out of a total of 80 fractions (fractions 51–56). These co-eluted completely with the haemoglobin (UV, 415 nm). Fractions 51–56 contained 89% of the applied amounts of 2,4-TDA and 81% of 2,6-TDA. Keywords: Biomarkers; gas chromatography–mass spectrometry; toluene diisocyanate; toluenediamine; thermal degradation Occupational exposure to isocyanates is associated with respiratory disease.1–3 The main applications of isocyanates are in the manufacture of elastomers, rigid and flexible polyurethane (PUR) foams, glues and lacquers.Toluene diisocyanate (TDI) dominates in the production of flexible foam and the 2,4- and 2,6-isomers are used typically mixed in a ratio of 80 : 20. Exposure to isocyanates may occur when PUR is thermally degraded. The presence of high concentrations of complex isocyanates, aminoisocyanates and amines has recently been observed in the work atmosphere.4 The methods available for the determination of biomarkers are based on the determination of the amines corresponding to the isocyanates in hydrolysed biological fluids, e.g., 2,4- and 2,6-toluenediamine (TDA).5,6 A detection limit of 0.05 mg l21 of TDA and a precision of 3.5% for 2,4-TDA and 1.6% for 2,6-TDA were found.The amines are liberated under strongly acidic or alkaline hydrolysis conditions, indicating the presence of adducts to proteins and peptides.7 Adducts to plasma albumin among workers exposed to TDI8 and haemoglobin among workers exposed to methylenediphenyl diisocyanate and methylenedianiline have been demonstrated.9 In a study of workers occupationally exposed to TDI, the concentrations of TDA in plasma showed only limited variation during work-days while the concentrations in urine varied considerably.10 The estimated mean half-lives in plasma were 21 (range 14–34) d for 2,4-TDA and 21 (16–26) d for 2,6-TDA, and the urinary half-lives were 5.8–11 d for 2.4-TDA and 6.4–9.3 d for 2,6-TDA.11 Volunteers exposed to TDI showed urinary half-lives of about 1.9 h for 2,4-TDA and 1.6 h for 2,6-TDA in a first phase and of 5 h in a second phase12 and the half-lives in plasma from volunteers were 2–5 h of 2,4- and 2,6-TDA in a first phase and > 6 d in a second phase.13 The aim of this study was to determine the concentrations of TDA in hydrolysed urine (U-TDA), plasma (P-TDA) and erythrocytes (E-TDA) and their relationships in workers exposed to thermal degradation products of TDI-PUR before and during an exposure-free period.A knowledge of the relationships and variations in time between U-TDA, P-TDA and E-TDA is necessary for the sampling strategy and exposure control. Experimental Apparatus The hydrolysed biological samples were analysed with an MD 800 quadrupole mass spectrometer (Fisons Instruments, Altrincham, UK) connected to a GC 8000 gas chromatograph equipped with an A200S autosampler (Fisons Instruments, Milan, Italy). The 2,4- and 2,6-TDA-pentafluoropropionic anhydride (PFPA) derivatives were determined by monitoring the ions at m/z 394 corresponding to the M 2 20 ions and m/z 397 corresponding to the M 2 20 ions of the trideuteriumlabelled internal standards (M = molecular mass).A fusedsilica capillary column with a chemically bonded stationary phase, DB-5 (J & B Scientific, Folsom, CA, USA), 25 m 30.25 mm id with a film thickness of 0.25 mm, was used.Detailed chromatographic and mass spectrometric information was given previously.5 A Waters Model 600 multisolvent delivery system (Millipore –Waters, Milford, MA, USA), a Waters Model 712 WISP and a Waters Model 490 UV detector were used for the determination of isocyanates in air and for the gel separation of haemoglobin. Haemoglobin was fractionated using a Pharmacia –LKB SuperFrac (Pharmacia, Uppsala, Sweden). Chemicals Both 2,4- and 2,6-TDA were obtained from Fluka (Buchs, Switzerland), HCl, NaOH, H2SO4 and K2HPO4 from Merck (Darmstadt, Germany), pentafluoropropionic anhydride (PFPA)erce (Rockford, IL, USA), toluene, isooctane and HPLC-grade acetonitrile from Lab-Scan (Dublin, Ireland), technical-grade TDI (80 + 20 2,4- + 2,6-TDI) and triethylamine (99%) from Janssen (Beerse, Belgium), NH4HCO3 from BDH Analyst, January 1997, Vol. 122 (51–56) 51(Poole, UK), 9-(N-methylaminomethyl)anthracene (MAMA) from Aldrich Chemie (Steinham, Germany), haemoglobin from Sigma (St.Louis, MO, USA) and trideuterated 2,4- and 2,6-TDA [CD3C6H3(NH2)2, TDDA], used as internal standards (IS) from Synthelec (Lund, Sweden). Subjects and Exposure Six workers (A–F) and two volunteers (colleagues) participated in this study. The work took place in a flame lamination factory applying a thin layer of TDI-based flexible foam on to textile fabric. The surface of the PUR foam was partly melted by an open flame immediately prior to the application.The work took place (7 am–4 pm) in two large workshops with an open door in between. Workers A, D, E and F were involved in the lamination process and workers B and C were occupied with cutting the produced laminated fabric. The lamination was performed in both workshops, but not at the same time. Worker D was employed only during the summer and was assumed not to be chronically exposed. The two volunteers were mainly involved in collecting air samples.Procedure Preparation of standard solutions TDI standard solutions were prepared by dissolving TDI in isooctane and further dilutions were made in toluene. Standard solutions of 2,4- and 2,6-TDA were prepared by dissolving accurately weighed amounts in 1 m HCl. The internal standards, 2,4- and 2,6-TDDA, were dissolved in 1 m H2SO4. Air sampling on impregnated glass-fibre filters Three subsequent air samples were collected in the breathing zone per worker during one typical work day of the study during about 8 h. The sampling times of each air sample varied between 100 and 220 min.The average exposure was calculated by taking the total amount of 2,4- and 2,6-TDI on the three filters divided by the total sampling time. Air samples were collected (1 l min21) on 13 mm glass-fibre filters, in a Teflon filter holder, impregnated with a MAMA–glycerol mixture.14 Immediately after sampling, the filters were placed in a test-tube containing 4 ml of toluene.The excess of the MAMA reagent was extracted for 5 min with 2 ml of 0.1 m HCl added to the testtube. A 2 ml volume of the toluene solution was separated and evaporated to dryness. The dry residue was dissolved in 1 ml of the mobile phase added before the LC–UV (254 nm) determinations. The LC column was a Hypersil BDS C18 (150 3 4.6 mm id) with 5 mm particles. The mobile phase (1.5 ml min21) consisted of acetonitrile–water (70 + 30) containing 3% triethylamine in the water and adjusted to pH 3 with phosphoric acid.Peak-height measurements were made and compared with a calibration plot for reagent solutions spiked with TDI to six concentrations in the range 0–1.2 mg ml21. Collection of biological samples Blood and urine samples were collected from six workers (A–F) before and during their summer vacation. Sampling was performed both at the factory and in the workers’ homes. All urine produced during 2 d of exposure and on three exposurefree days were collected.A few urine samples were also collected on different occasions during the 3–4 week vacation. Blood samples were taken once a day during 2 d of exposure and on three exposure-free days. Blood and urine samples were also taken from two volunteers. All urine produced during the day of exposure and the following exposure-free day was collected. Blood samples were taken on three occasions at the day of exposure and two and five times, respectively, during the following eight exposure-free days.Sampling, handling and storing of biological samples All urine samples were collected in polyethylene bottles. The creatinine concentration [determined at the Department of Clinical Chemistry, University Hospital, Lund, by use of Kodak Ektachem clinical chemistry slides (CREA) and Kodak Ektachem 700 XR-C analyser], pH and total volume of each urine sample were determined. Blood was sampled in heparinized tubes (Venoject). Plasma was separated within 8 h and transferred to new tubes.The erythrocytes were washed in 0.9% NaCl two or three times. The urine, plasma and erythrocyte samples were kept frozen at 220 °C until analysis. Gel filtration of haemoglobin Washed erythrocytes from worker A were repeatedly frozen and thawed four times in a double volume of Millipore-purified water (Milli-Q system, Millipore, Bedford, MA, USA). The cell debris was removed by centrifugation at 18 500 g for 15 min. The haemoglobin-containing supernatant was diluted five fold in 50 mm NH4HCO3 and filtered using a 0.2 mm hydrophilic membrane filter (Minisart RC 25; Sartorius, G�ottingen, Germany).A volume of 200 ml of the solution was separated on a Superose 12 (10/30) gel filtration column (Pharmacia Biotech, Uppsala, Sweden) and fraction into 80 fractions of 0.25 ml each. The mobile phase consisted of 50 mm NH4HCO3 at a flow rate of 0.5 ml min21. UV traces were recorded at 415 nm. A total of 50 fractionations were performed and fractions were collected in the same subsequent fraction vials.Hence each fraction vial contained 12.5 ml. The 80 fractions were then evaporated to dryness and the dry residues were dissolved in 1 ml aliquots of Millipore-purified water. Hydrolysis To 1 ml of plasma, 1 ml of urine, 0.5 g of erythrocytes in 0.5 ml of Millipore-purified water or 1 ml of concentrated gel filtration fractions, 100 ml of IS were added together with 1.5 ml of 3 m H2SO4 before hydrolysis at 100 °C for 16 h.In the sample batches the calibration samples were prepared in the range 0–30 ng ml21 TDA with 5.5 ng ml21 IS for plasma, 0–10 ng ml21 TDA with 2.0 ng ml21 IS and 0–60 ng ml21 TDA with 11 ng ml21 IS for urine, 0–10 ng ml21 TDA with 2.0 n ml21 IS for erythrocytes and 0–5 ng ml21 TDA with 1.0 ng ml21 IS for the gel filtration fractions. After hydrolysis the samples were worked-up as described below. Work-up procedure A 5 ml volume of saturated NaOH and 2 ml of toluene were added to the samples.The mixtures were shaken for about 10 min and then centrifuged at 1500 g for 10 min. The organic layers were transferred into new test-tubes and 20 ml of PFPA were added to each. The mixtures were immediately shaken vigorously. The excess of the reagent and acid formed was removed by extraction with 2 ml of 1 m phosphate buffer solution (pH 7.5). The toluene layers containing the amide derivatives and the internal standards were transferred into 1.5 ml autosampler vials with Teflon seals.When analysing the gel filtration samples, about 1 ml of the toluene solution was evaporated to dryness and diluted with 100 ml of toluene. The samples were then ready for injection into the GC–MS system. Statistics To determine if the correlation coefficients are significant, we calculated a t-value: t = ýrý (n 2 2)1 2/(1 2 r2)1 2 (r = correlation 52 Analyst, January 1997, Vol. 122coefficient, n = number of observations).The calculated value of t was compared with the tabulated value at the 95% significance level, using a two-tailed t-test and n 2 2 degrees of freedom. If the calculated value of t was greater than the tabulated value, a significant correlation exists. Results Isocyanates in Air The average air concentrations during the work day of 2,4- and 2,6-TDI at the breathing zone using personal sampling were for worker A 3.5 and 10 mg m23, B 0.9 and 2.6 mg m23, C 1.8 and 6.5 mg m23, D 1.4 and 4.7 mg m23, E 1.6 and 4.1 mg m23 and F 2.1 and 7.4 mg m23, respectively.No representative air samples for the volunteers were collected. Biomarkers in Hydrolysed Urine The U-TDA for workers A–F are shown in Fig. 1. During the work U-TDA was rapidly eliminated and a slower elimination phase were seen during the exposure-free period. For workers B and E one urine sample was considerably above the assumed ideal elimination curve. This was probably due to an unnoticed exposure to TDA dust at the factory after cestion of flame lamination work.The urinary half-lives in the slower phase were estimated by plotting the natural logarithm of the urinary concentrations against time and by dividing ln 2 with the calculated slopes. The half-lives of 2,4- and 2,6-TDA were estimated for worker A to be 32 (r = 0.82, n = 10) and 25 d (r = 0.94, n = 10), B 20 (r = 0.67, n = 23) and 20 d (r = 0.90, n = 23), C 15 (r = 0.93, n = 9) and 18 d (r = 0.91, n = 9), D 4.3 (r = 0.90, n = 8) and 11 d (r = 0.79, n = 8), E 26 (r = 0.36, n = 23) and 22 d (r = 0.88, n = 24) and F 14 (r = 0.93, n = 12) and 16 d (r = 0.97, n = 12), respectively.The medians of the urinary half-lives were for the slow phase 18 d for 2,4-TDA and 19 d for 2,6-TDA. The urinary half-lives for the two volunteers were calculated to 5.3 (r = 0.96, n = 4) and 6.2 h (r = 0.97, n = 4) h for 2,4-TDA and 8.4 (r = 0.96, n = 5) and 7.4 h (r = 0.99, n = 4) for 2,6-TDA (Fig. 2). Biomarkers in Plasma and Erythrocytes The concentrations of 2,4-TDA in plasma were in the range 2.5–19 ng ml21 and those of 2,6-TDA in the range 4.4–30 ng ml21 among the workers. In erythrocytes, the concentrations of 2,4-TDA were in the range 0.5–6.6 ng g21 and those of 2,6-TDA in the range 1.2–14 ng g21 among the workers. Significant linear relationships were seen between the mean concentrations, for all individual samples, in plasma and erythrocytes of 2,4-TDA (r = 0.97, n = 6) and 2,6-TDA (r = 0.97, n = 6) (Fig. 3). P-TDA declined slowly after work cessation. The half-lives in plasma (calculated as for UTDA) were in the range 6.5–12 d for 2,4-TDA (r = 0.76–0.95, n = 4) and 7–11 d for 2,6-TDA (r = 0.76–0.97, n = 4). E-TDA declined too little for the calculation of half-lives. When plotting the mean daily urinary concentrations adjusted for creatinine against P-TDA and E-TDA, relationships were observed. The relationships became more linear with the number of days after cessation of work (Fig. 4). Significant correlations were seen, with the exception of E-2,4-TDA on day 2. Significant linear plots, with the exception of E-2,6-TDA, were obtained on plotting the P-TDA and E-TDA for workers A, B, C, E and F against the TDI air levels measured on day 1 (worker D was only employed during the summer) (Fig. 5). No detectable P-TDA or E-TDA were found in the two volunteers, except for very low concentrations of 2,4-TDA ( < 0.33 ng ml21) in plasma from one of the volunteers.Among a group of 17 unexposed humans, the U-TDA and P-TDA were below the detection limit. Conditions for the Hydrolysis of Haemoglobin Volumes of 1 ml of the haemoglobin-containing supernatant (see Gel filtration of haemoglobin) were hydrolysed in 1.5 ml of 3 m H2SO4 at 100 °C for 4, 8, 16, 24 and 48 h. After hydrolysis the samples were subjected to the work-up procedure. No free 2,4- and 2,6-TDA were seen without hydrolysis.The release of 2,4- and 2,6-TDA increased with increasing hydrolysis time (Fig. 6). For the preparation of haemoglobin from the erythrocytes, the recoveries of 2,4- and 2,6-TDA were 77% and 91%, respectively. On analysing the remaining cell debris in the centrifuge tube no TDA was found. Gel Filtration of Proteins in erythrocytes TDA were only found in six of the 80 fractions (fractions 51–56) from the gel filtration separation and co-eluted com- Fig. 1 Concentrations of 2,4-TDA (2) and 2,6-TDA (5) in hydrolysed urine (mg per mmol of creatinine) from workers A–F.All urine produced during 2 work days and on the three days following work cessation (vacation), and on different occasions during the following 3–4 weeks, was collected. Time 0 = 6.00 am on the first day of the study. Analyst, January 1997, Vol. 122 53pletely with the haemoglobin (UV, 415 nm) peak. Fractions 51–56 contained 89% of the applied amounts of 2,4-TDA and 81% of 2,6-TDA (Fig. 7). In the GC–MS traces pure and well separated 2,4- and 2,6-TDA-PFPA peaks were seen. The 2,4-and 2,6-TDA-PFPA peaks in the chromatogram of fraction 54 were found at an unfiltered S/N of 175 and 61, respectively (rms). Only traces of TDA-PFPA were seen in fractions 1–50 and 57–80, due to impurities of TDA in the internal standards. These were also present in the chemical blanks. Discussion Isocyanates occur in many chemical and physical forms in workplace atmospheres. In the TDI foam industry, the monomers dominate in the gas phase.However, when particles are formed, many other compounds may be present in addition to monomers. These can be measured by methods which are based on the reaction of isocyanates with secondary amines containing a chromophore.14 During the thermal decomposition of PUR, in addition to isocyanates, aminoisocyanates and amines are also present. This means that isocyanate exposure measured by ‘conventional’ methods, underestimates the true exposure.It was therefore surprising, but encouraging, to find a relationship between the air data and the P-TDA and E-TDA levels among Fig. 2 Concentrations of 2,4-TDA (2) and 2,6-TDA (5) in hydrolysed urine (mg per mmol of creatinine) from two volunteers. The volunteers were present in the factory from 7 am until 4 pm on day 1. Time 0 = 6.00 am the first day of the study. Fig. 4 Concentrations of 2,4-TDA (5/-) and 2,6-TDA (2/8) in hydrolysed plasma (ng ml21) and erythrocytes (ng g21) and mean urinary concentration (mg per mmol of creatinine) during 1 d of exposure (day 2) and the three following exposure-free days (days 3–5) among six workers (A–F).Day 2: r P- 2,4-TDA = 0.84, r E-2,4-TDA = 0.81, r P-2,6-TDA = 0.86, r E-2,6-TDA = 0.83; Day 3: r P-2,4-TDA = 0.91, r E-2,4-TDA = 0.91, r P-2,6-TDA = 0.94, r E-2,6-TDA = 0.89; Day 4: r P-2,4-TDA = 0.92, r E-2,4-TDA = 0.98, r P-2,6-TDA = 0.97, r E-2,6-TDA = 0.98; Day 5: r P-2,4-TDA = 0.98, r E- 2,4-TDA = 0.97, r P-2,6-TDA = 0.99, r E-2,6-TDA = 0.97.Continuous line = 2,4-TDA; broken line = 2,6-TDA. Fig. 3 Mean concentrations, for all individual samples, of 2,4-TDA (5) and 2,6-TDA (-) in hydrolysed plasma (ng ml21) and erythrocytes (ng g21) among six workers (A–F) during 5 d. Continuous line = 2,4-TDA, r = 0.97; broken line = 2,6-TDA, r = 0.97. 54 Analyst, January 1997, Vol. 122the workers. However, the relationship was different to that for data from a flexible foam factory,11 where about the same levels in plasma were found but three times higher TDI air levels.This can be explained by the insufficiency and the limitations of the air methods used to measure more complex isocyanates. Data from conventional air methods may therefore be used as a relative index of exposure in the absence of methods that take into account all other isocyanates that may influence the industrial hygiene. Further studies of these aspects are in progress.The urine data clearly demonstrate fast elimination phases and one slower phase. It can be assumed that the fast phases reflect the more recent exposure and the slow phase reflects urinary elimination of degradation products of modified proteins, and hence the history of exposure. The urinary elimination half-lives for the slow phase were about the same as the half-life of albumin and the mean half-lives in plasma among TDI foam workers.11 The significant linear relationships between the U-TDA and E-TDA and P-TDA that improved with time after work cessation may also add to the explanation of the presence of modified erythrocytes and albumin metabolites and breakdown products excreted in urine.These products seem to dominate after a few days of work cessation. Worker D, a worker not chronically exposed, had shorter urinary elimination half-lives compared with the chronically exposed workers. The volunteers in this study had about the same urinary halflives as the volunteers in the volunteer study.12 The volunteer study may reflect an interesting difference in uptake, metabolism and excretion between chronically exposed workers and volunteers and the difference in exposure situations.In the workplace studies much more complex isocyanates and different chemical and physical forms are present in the air. Biomarkers of both 2,4- and 2,6-TDA were found in plasma and erythrocytes from all workers. The half-lives in plasma, in Fig. 5 Concentrations of 2,4-TDA (5) and 2,6-TDA (-) among workers A, B, C, E and F and in worker D (2 = 2,4-TDA) and (8 = 2,6-TDA) in: (a), hydrolysed plasma (ng ml21); and (b), hydrolysed erythrocytes (ng g21), plotted against the mean individual air concentrations of 2,4- and 2,6-TDI (mg m23) for the first work day of the study. For the calculation of the linear curves, data from worker D were excluded as he was a seasonal worker and not chronically exposed (r P-2,4-TDA = 0.98, r P- 2,6-TDA = 0.97, r E-2,4-TDA = 0.92, r E-2,6-TDA = 0.85).Fig. 7 Amounts of 2,4-TDA (5) and 2,6-TDA (~) (ng) in the fractions eluted from a gel filtration separation of lysed erythrocytes (cell debris separated) from worker A. Fractions 51–56 contained 89% of the applied amount of 2,4-TDA and 81% of 2,6-TDA and co-eluted with haemoglobin (UV 415 nm, broken line, 100% was saturated detection). Selected ion monitoring chromatogram of fraction 43 represents a fraction not containing TDA, and that of fraction 54 one with ‘TDI’-modified compounds.(m/z = 394.2 is the M 2 20 ions of TDA-PFPA and m/z = 397.2 is the M 2 20 ions of TDDA-PFPA). Fig. 6 Concentrations of 2,4-TDA (5) and 2,6-TDA (-) in lysed erythrocytes (cell debris separated) after hydrolysis for 0, 4, 8, 16, 24 and 48 h at 100 °C. Analyst, January 1997, Vol. 122 55this study, were about 50% shorter than those in the flexible foam worker study11 but about the same as those in the volunteer study.13 The difference may be explained by the presence of a faster phase, but this could not be observed in the study of flexible foam workers as blood samplings were performed at too long intervals. The E-TDA levels were virtually stable during the study and therefore no half-lives could be calculated.They can therefore be assumed to be longer than for plasma. A significant linear relationship between PTDA and E-TDA was observed. No TDA was found in the centrifuged and lysed erythrocyte solution on performing the work-up procedure without hydrolysis.This indicates covalently TDI-modified proteins. The gel filtration demonstrated that the modified protein in fact was haemoglobin. The hydrolysis pattern of haemoglobin was found to be almost the same as for plasma among TDI-exposed workers.7 This is an indication that the nature of the chemical bondings is the same and that the same hydrolysis conditions can be used. There are many aspects of the uptake, metabolism and excretion of isocyanates that are not yet fully understood. However, there are many reasons to believe that modified proteins and peptides are involved in isocyanate-associated disease. It is therefore interesting to know that intracellular in addition to intercellular proteins are modified with TDI among workers. The mechanism by which isocyanates are able to pass cell membranes is not known and further studies on this aspect are in progress. The present study has shown that the history of exposure can be determined by biomarkers in plasma and erythrocytes. Recent exposure does not greatly affect the P-TDA and E-TDA levels. Biomarkers in urine can be used for the same purpose if sampling is performed after the fast elimination phases. This work was supported by the Swedish Work Environment Fund. We thank M. Adamsson, M. Spanne and T. Russin for skilful technical assistance and Professor A. Grubb and Associate Professor J.-O. Jeppson for valuable discussions. References 1 Banks, D. E., Butcher, B. T., and Salviaggio, J. E., Ann. Allergy, 1986, 57, 389. 2 Baur, X., Marek, W., Ammon, J., Czuppon, A. B., Marczynski, B., Raulf-Heimsoth, M., Roemmelt, H., and Fruhman, G., Int. Arch. Occup. Environ. Health, 1994, 7, 310. 3 Vandenplas, O., Malo, J. L., Saetta, M., Mapp, C. E., and Fabbri, L. M., Br. J. Ind. Med., 1993, 50, 213. 4 Tinnerberg, H., Spanne, M., Dalene, M., and Skarping, G., Analyst, 1996, 121, 1101. 5 Skarping, G., Dalene, M., and Lind, P., J. Chromatogr., 1994, 663, 199. 6 Maitre, A., Berode, M., Perdrix, A., Romazini, S., and Savolainen, H., Int. Arch. Occup. Environ. Health, 1993, 65, 97. 7 Lind, P., Skarping, G., and Dalene, M., Anal. Chim. Acta, in the press. 8 Lind, P., Dalene, M., Lindstr�om, V., Grubb, A., and Skarping, G., Analyst, 1997, 122, in the press. 9 Sch�utze, D., Sepai, O., Lewalter, J., Miksche, L., Henschler, D., and Sabbioni, G., Carcinogenesis, 1995, 16, 573. 10 Persson, P., Dalene, M., Skarping, G., Adamsson, M., and Hagmar, L., Br. J. Ind. Med., 1993, 50, 1111. 11 Lind, P., Dalene, M., Skarping, G., and Hagmar, L., Occup. Environ. Med., 1996, 53, 94. 12 Skarping, G., Brorson, T., and Sang�o, C., Int. Arch. Occup. Environ. Health, 1991, 63, 83. 13 Brorson, T., Skarping, G., and Sang�o, C., Int. Arch. Occup. Environ. Health, 1991, 63, 253. 14 Tinnerberg, H., Dalene, M., and Skarping, G., Am. Ind. Hyg. Assoc. J., in the press. Paper 6/06148F Received September 6, 1996 Accepted November 6, 1996 56 Analyst, January 1997,

ISSN:0003-2654    

DOI:10.1039/a606148f

出版商:RSC    

年代:1997

数据来源: RSC