ISSN:0306-1353    

DOI:10.1039/AA98111FX001

出版商:RSC    

年代:1981

数据来源: RSC

ISSN:0306-1353    

DOI:10.1039/AA98111BX003

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

6 Analytical Atomic Spectroscopy 1.2 PLASMAS Until a few years ago a feature of the reports in this series on plasma atomization and excitation was the relative immaturity of the technique. The somewhat extravagent claims made, e.g., concerning the absence of interferences in the ICP, were a reflection of this. The papers published this year demonstrate a more established situation. The ICP remains the favoured form of plasma and its dominance has been further strengthened. General characteristics of plasma sources, e.g., the effect of easily ionizable elements, identified in previous years, have been confirmed.Indeed, the ICP seems to have matured to the extent that rediscovery is agrowing feature of the literature. It is to be hoped that reports such as this present series will prevent too much duplication of effort, but previous experience in the fields of flames, arcs and sparks suggests that this is a forlorn hope. Keliher (C455) has commented on some aspects of thegrowth of all forms of plasma emission spectroscopy.In a brief but valuable theoretical review, Montfort and Agterdenbos (938) have looked at various models for equilibrium and non-equilibrium plasmas in open and sealed situations, both microwave and radio-frequency excited.The growth of papers reporting the coupling of gas or liquid chromatography to plasma emission spectroscopy has continued to be marked. Chromatographic detection using the ICP, MIP and d.c. plasma has been comprehensively reviewed (2 168).Atomization and Excitation 7 1.2.1 Inductively-coupled Plasmas 1.2.1.I Reviews Ward (429) has presented a tutorial review of the development of ICP instrumentation. The present status of spectrochemical analysis with the ICP in the USSR has been discussed (638). Various workers have reviewed current progress in ICP spectrometry at different conferences (C711, C737, C1330, C1458, C2095, C2102). 1.2.1.2 Plasma Characteristics Several different excitation mechanisms in the ICP have been proposed without, as yet, universal agreement on one theory.As many of the proposed mechanisms involve metastable argon species, studies of these species and the way in which analyte ions might be generated via Penning ionization are of particular value. Both AA and AF measurements have been used to establish the dependence of Ar metastable populations on such parameters as plasma power, electron densities and introduced sample species (C968).An Ar MIP was used as the AA source and a nitrogen-pumped dye laser for fluorescence. The validity of a quasi-thermal relationship between the numerous metastable levels was investigated. Changes of metastable Ar populations in different regions of the plasma as a function of analyte concentration have been used by the same research group as evidence of the importance of metastable Ar in analyte excitation (C 1093).They have further established an equilibration between metastable and resonance levels (C1094). If, as implied, metastable states are overpopulated in a plasma, alternative mechanisms for suprathermal excitation can be postulated.Other useful spatial studies of Ar metastables using the 81 1.52 nm Ar emission line from an MIP to study this S , AA line have been reported (CSOS, C2099). The lower level of this transition is 11.55 eV above the ground state, and the recorded absorbance depended strongly upon power and height of observation. A third independent report (C696) also seems to confirm the above observations, particularly the equilibrium between metastable and ionic levels, and it seems possible that evidence necessary to evaluate new and existing models of excitation mechanisms will soon be available.The careful determination of electron number densities will also aid in mechanistic elucidation. Montaser et al. (2181) have published their method based on observation of the Stark broadened lines of an atomic species. Of a number of elements tested (Al, Ca, H, K, Li and Mg) Al and Mg were considered to be the most suitable.Visser (C2014) reported a value of 8 x 101j electrons cmV3 from determination of the broadening of hydrogen Balmer series lines. By comparison with a calculated Boltzmann excitation temperature it was again shown that the ICP is not in local thermal equilibrium (LTE).Specialist continuous spatial-scanning equipment has been built to make the necessary measurements to investigate deviations from LTE (C2015). Aeschbach (454) has suggested that the electron density distribution exceeds that predicted from LTE calculations, such as those based on the Saha equation, for two reasons. The kinetic energy from the power input of the r.f.field is greater than the kinetic energy of the gas particles; the high gradients in the electron density and temperature lead to the ambipolar diffusion of electrons. Using such a diffusion constant, the electron number density and electronic temperature were calculated to be about an order of magnitude greater than predicted by the Saha equation.The concept of the diffusion of electrons from the discharge region against the Ar flow is probably another useful step towards deriving non- equilibrium models for ICP discharges. A method of calculating energy level populations based on transition rates between states has been proposed by Lovett (C1099). The lower the electron density the greater was the deviation of the calculated energy level populations from those predicted by the Saha and Boltzmann equations due to dominance of radiative recombination.Interestingly, the deviation from LTE began to appear in the region 1 O I s to 10lb electrons cm-3, i.e., the range observed in the ICP. to 4P,8 Analytical Atomic Spectroscopy The importance of defining regions of the discharge in understanding ICP characteristics and the effects of easily ionizable elements was stressed last year (see ARAAS, 1980,10,36 and also Section 1.2.1.5).Koirtyohann et al. (616) have been leaders in this field and have reported vertical intensity distributions using a photodiode array for several atomic and ionic lines. Again it was shown that the zone of interference was shifted, dependent on operating conditions, particularly the injector gas flow and the power level.Detailed profiles of the absorption and emission of atomic and ionic species have shown that Li has little effect on Ca AA, suppresses ionic absorption and enhances both atomic and ionic emission in most regions low in the plasma (C688). Horlick and co-workers (C77, C510, C690, C2098) have also been active studying vertical, lateral and radial profiles using a photodiode array.Neutral atoms have been sub-divided into two basic groups on the basis of vertical spatial emission: those for which the vertical position emission peak correlated positively with the “norm temperature”, and those with a negative correlation. Ionic lines are not species dependent in the same way.Axial profiles of excitation and gas temperatures and different spectral lines have been measured using a vertically movable horizontal slit before the spectrometer (C693). Distinct peaks in the excitation temperature were observed but the gas temperature tended to increase gradually as the observation height increased. Schlieren photography has also been used to study gradients in the ICP (C432); sub-microsecond time resolution was obtained.Variations of emission characteristics with height and their relationship with power have also been studied (CIO). Measured and calculated velocity and temperature profiles have been compared by Barnes and Schleicher (1524). A refined ICP gas flow model was proposed which gave agreement, within experimental error, with observed values.No solution aerosol was assumed in this model but in a later paper Barnes and Genna (1 540) measured temperature and velocity distributions in a low power Ar ICP including an aerosol gas. The channel formed in the discharge by the aerosol gas-flow exhibited temperature and velocity characteristics distinct from the plasma core and wall regions. Spectral line profiles in the plasma are of considerable practical and instrumental importance.Kato et al. (C692) have attempted to measure line profiles at different observation heights with a pressure-scanning Fabry-Perot interferometer. Unfortunately the effect of broadening arising from the use of the semi-empirical apparatus function for deconvolution, seriously influenced the measured profiles.By coupling an interferometer to a conventional monochromator. Edelson and Fassel (2221) were able to measure isotopic spectra of Pb(I1) and U(I1) at 537.2 and 424.4 nm, respectively, from an ICP. The spectral resolution of the system was 0.001 nm at 400 nm and they speculated that commercial monochromators with equivalent resolution could be used for analytical isotopic abundance determinations in the nuclear industry.A study has also been reported of the application of tunable lasers to the study of pressure broadening and related collisional processes in plasmas (1718). The underlying relaxation processes and their time-scales were probed in order to assess limitations on saturable absorption spectroscopy or fluorescence. The use of an Ulbright sphere for spectral calibration purposes has been recommended ( 1 526).The short and long term stability of its radiation at four Balmer wavelengths compared well with a conventional tungsten standard lamp. Tracy and Myers (C7) have successfully used a tungsten ribbon lamp to measure the absolute spectral radiance of ICP background emission. A broad maximum was observed at 450 nm, in a conventional low power Ar ICP, where the spectral radiance reached 1.5 x l o i 2 photons s-’ mm-2 steradian-’ nm-’ .The radiance dropped rapidly below 400 nm reaching 0.025 x 10 ’ at 200 nm. The increasing interest in vacuum ultraviolet wavelengths has been noted previously (see ARAAS, 1980, 10, 35) and this has continued. An experimental configuration for coupling an ICP to a vacuum spectrometer via a He flushed copper tube has been described (315).Atomization and Excitation 9 Several emission lines of Br, C1 and S were observed in the region 120 to 185 nm as were lines due to C, N and 0 impurities.Using an Ar flushed optical path, Omori et al. (C706) have reported analytical applications of the P 178.2 nm and S 180.7 nm lines, in steel and oil analysis.The possible use of the I 183.0 nm line and the Hg 185.0 nm line was also described. Hayakawa et al. (C704) have reported on the use of the following lines in steel and spa-water analysis, noting that He flushing showed no improvement over Ar flushing: As 189.042; B 182.640; I 183.038, P 177.499and 178.287; S 180.730 and Sn 189.989 nm. A thorough study of various P lines for steel and copper analysis led to the conclusion that the best SNR was obtained with the 178.28 nm linecompared to the 213.62or 253.56nm lines, and further, that this V.U.V. line was free of spectral overlaps from Cu and Fe (128).The baseline structure of the Ar ICP particularly at low wavelengths has been studied (C2106). Emission from Ar, CN , NO, 0, and OH species was observed, the latter due to either atmospheric entrainment or trace impurities in the Ar.A quartz-tube torch extension was shown to reduce atmosperhic entrainment greatly. An extended torch has been reported to improve detection limits for several elements, e.g., As and Se, which emit at low wavelengths, particularly when the sample solution contained nitric acid, as nitrogen -oxygen species contributed to the troublesome background (C32).The same extension was also used to enable the determination of N at 174.3 nm in a variety of samples (C705). The use of near infrared wavelengths has been reported previously (see ARAAS, 1980, 10, 36), particularly for the determination of non-metals, and tbis area of study is showing signs of growth. An improved detection limit of 25 ng for 0, at the non-resonance 777.194 nm line, using a gas sampling loop, has been reported (1561; see also ARAAS, 1980, 10, Ref. 858). Oxygen-selective GC - ICP studies were made of several organic compounds but background interferences impaired the selectivity of the application. The spectra of C, H, N and 0 have been qualitatively studied using a near i.r. and i.r.photodiode array spectrometer (1793). The best spectral response was, however, found in the 600-850 nm region. Interesting new studies of C and S emission have been reported by the same research group (1974). The C 909.483 nm and S 921.291 nm lines were recommended and the S line offered detection limits significantly better than S lines in the visible or ‘air-path’ U.V. regions.Detection limits for 20 C lines and 30 S lines were presented with vertical emission intensity profiles. Hughes and Fry (1659) have further reported on 102 Br and 85 C1 lines, all non- resonance lines in the 370 to 990 nm region. Several ionic lines were also observed and detection limits reported. The best SNR were obtained for signals emanating from between the top two turns of the induction coil.Denton (C2097) has commented on the utility of non- metallic detection in the ICP for determining the empirical formulae of nucleotides by direct analysis of the eluant from a LC. Future work will be greatly aided by the appearance of spectral line tables for the ICP. Boumans has further discussed the conversion of copper arc tables into argon ICP tables by computer iteration (1532).Winge et al. (C75) have reported progress on their measurement programme aimed at preparing an atlas to aid in the selection of lines and avoiding interferences. A compilation of prominent lines from these two sources has also been published and will probably be widely used (428). The criteria for wavelength selection in sequential ICP-OES have been discussed, including the use of short wavelength scans to identify interferences (C13). The importance of the optimization of the operating parameters of the ICP is now generally recognized.Mermet (C 1448) has reviewed the influence of different operating parameters. The problems of optimization for several elements simultaneously and the need to identify practical compromise criteria has been stressed (C2115).Simplex optimization is now widely regarded as a suitable multi-parametric technique applicable to the ICP. Cave et al. (393) have further reported on their work using the modified simplex procedure (see also10 A nary t ical A tomic Spectroscopy ARAAS, 1980, 10, 37). De Galan (C1330, C1458) has reported that simplex optimization identified operating conditions close enough together to allow compromise operation for several elements.The sequential simplex procedure also proved highly successful in other laboratories (1539). Boumans et al. (C702) still clearly prefer compromise optimization via representative measurements and rational interpretation based on the concept of ‘norm’ temperatures This less rigorous approach may be most valuable in laboratories with extensive experience of the ICP.Alkali metals tend to offer best SNR at different operating conditions because they are so easily excited and ionized. Joshi (C1158) has suggested that for such elements, low forward power, e.g., 350 W, and reduced gas flows should be used. The noise characteristics of the ICP are now being clarified. Belchamber and Horlick (C80, C502) have shown that below 5 Hz the noise power spectra show a marked dependence on the type of nebulizer used.The spectrally observed noise was correlated with acoustic noise and pressure fluctuations in the spray chamber. Increasing the integration time did not necessarily improve the precision of measurements. Distinct noise peaks were also observed in the region 200 to 400 Hz.This systematic noise arose from rotation of the plasma body (C1120, C2100). Drain arrangements may be a major source of imprecision and the spray chamber should preferably be fitted with a soak away. Since the nebulizing process has been so often identified as a major source of instrumental variability, an attempt has been made to measure time-dependent variations of the sample supply directly using light-scattering ( 1 580 or 13 14).The mist and emission signal fluctuations were found to be strongly correlated, and it was indicated that electronic filtering might enable precision to be improved. Following an analysis of noise and detection limits for the low-power Ar ICP, Boumans et al. (C716) concluded that: high instrumental luminosity is important for full exploitation of the 190 - 250 nm region; the worsening of detection limits by line-rich matrices can be minimized by careful line selection; the ratio of SNR to source flicker represents a better criterion of merit than SNR alone.Ward (470) has proposed a long-term stability test for the ICP. The Ar discharge on all channels of a direct reading spectrometer was monitored without any aerosol gas.The use of internal standards has again been advocated, in this case the use of Y for steel samples, to improve precison (948). Additional reference on preceding topic - C721. Considerable interest is now being shown outside Europe in low-power mixed gas plasmas. Montaser et al. (1615) have published their comparison of Ar and N, cooled plasmas; N, was also introduced into the aerosol gas.Simultaneous measurements of 20 ion and atom lines were made under various operating conditions. Different optimal conditions were found when N 2 was used as the coolant. The reader is referred to previous studies of optimal conditions in Ar and N2 cooled plasmas by Greenfield and Burns (ARAAS, 1980,10, Ref. 477) and Ebdon et al.(ARAAS, 1980,10, Ref. 497). The use of N , instead of Ar as the aerosol gas degraded the detection limits for all elements. Temperature measurements made in Ar and N , cooled Ar plasmas suggested to other workers that while the all Ar ICP was not in LTE the N , cooled Ar ICP may be closer to LTE (C695). While this might suggest more difficulty in identifying compromise conditions for an N /Ar plasma, Ng and Horlick (C29), in contrast, preferred such a plasma as offering wide simultaneous spectral measurement in their correlation studies with a Fourier transform spectrometer system.Spatial profiles have been used to show that the vertical intensity maxima of all elements are highly dependent on the amount of N , coolant in an otherwise Ar ICP (C26). As the N 2 percentage in the coolant increases, the intensity maxima drop until the “pinched” plasma known since Greenfield’s early work (e.g., Proc.SOC. Anal. Chem., 1965, 2 , 111) is observed. Choot (C2101) has reported similar effects with 0,, He and air coolants. Ohls (2091) reported better precision (0.3%) with a higher power N cooled Ar plasma than the 1 To observed with an all Ar plasma. This seems to be because the RSD of the background signal was improved.The impedanceAtomization and Excitation 11 matching characteristics of Ar, N, and air ICP discharges have been measured indirectly (C1113). The plasma impedance for an all N, ICP was almost double that for an all Ar ICP and this has implications for matchbox tuning when switching between discharges.Magnetic flux densities were also measured and found to be related to the power density of the discharge. Using a 40 MHz generator, minimum operating powers were obtained for an air ICP (0.8 kW), a N, ICP (1.3 kW) and an 0, ICP (1.0 kW) (C722). Another preliminary report of a He ICP has been made (C715). As H2 has a thermal conductivity about ten times that of Ar, the influence of H, on the different gas flows of an Ar ICP has been studied (636).A 93.5%/6.5% Ar/H, mixture was used. The net peak area in the low U.V. region was increased and the stability of the plasma improved, thus allowing detection limits to be improved. Above 400 nm no further improvements were observed suggesting that the atoms were already sufficiently excited in this region.Some valuable results are now being reported in the field of plasma source mass spectrometry. A significant event was the publication of a detailed description of the system used at the Ames Laboratory with a list of detection limits (2202). The background spectrum of 1% nitric acid showed few peaks, which were predominantly from argon, water and nitrogen-derived ionic species. The mass spectra of analytes featured largely M + and MO + ions.Linearity over four orders of magnitude, with typical detection limits of 0.002 (Cr) to 0.04 (Y) pg ml-l was reported. Some interference from Na, from the 10 pg ml-I level, was observed but the method clearly shows considerable promise for isotopic measurements. The achievements of the group at the University of Surrey who now claim some detection limits below 1 ng ml-* have been reviewed (C2131, C2132) as have those of the Ames group (C678, C1463).The latter group have examined fundamental properties of charged particles in the ICP with their MS interface (1804 or 2182). Ratios of doubly to singly ionized species were used to determine ionization temperatures. The values obtained agree well with those observed by optical spectroscopy, especially when the influence of the interface is considered, and confirm the evidence for suprathermal ionization in the Ar ICP. 1.2.1.3 Sample Introduction This area of research has clearly been the most active in the past year. Many conference papers have dealt with sample introduction and it will be interesting to note whether these result in full publications in the future.A very useful collection of papers on this topic has appeared (647) introduced by an updating review (648). The performances of different types of nebulizers have been compared (C700). It has been shown that the “washout time” needed to reduce residual memory effects for different nebulizers depends on the type used (C444). The application by Sharp (C682, C923) of the principles of fluid dynamics and thermodynamics to the design of ICP nebulizers has as yet only been presented at conferences.It is to be hoped this work will be published fully as designs for sub-sonic and super-sonic nozzles were indicated and a practical cross-flow nozzle tested. The patterns of flow fields generated by an exhausting jet were also shown to account for the salting up problem, which is particularly severe in concentric nebulizers.Interesting work continues to be published on aerosol characteristics. Various methods for measuring droplet size distributions, including the cascade impactor, microscope and laser diffraction techniques, have been compared (C1406). Browner and co-workers (C681, C2111) continue to contribute to our understanding of observed nebulization phenomena in terms of droplet characterization.Solution composition affects the physico-chemical characteristics of aqueous aerosols and may be reflected, of course, in interference effects (C441). The importance of the effect of spray chamber design on observed aerosol transport phenomena is now being recognized (C1041). The spray chamber essentially acts as an order12 Analytical Atomic Spectroscopy sorter (C2112).Very large sodium droplets in an ICP have been reported to give yellow streaks and ionization and solute vaporization interferences (C440). These can be removed by a double-pass chamber which eliminates larger droplets. Tracy and Myers (C5, C503, C 1 1 14, C1121) have drawn attention to some interesting aerosol behaviour.They have attempted to reduce the nebulizer induced loss of aerosol reported last year (see ARAAS, 1980,10,39) by using an auxillary Ar flow injected directly into the spray chamber (C503). While the effective nebulization efficiency may be doubled any increase in analytical signal is largely negated by the increased total carrier gas flow. Some of the noise associated with aerosol introduction, but not the drift, was removed by filtering the aerosol in a long (up to 15 m) 6.3 mm bore plastic tube between the spray chamber and torch (C5).Improved mixing, elimination of larger droplets and reduced pressure fluctuations were probably the cause of the factor of three drop in noise, but unfortunately the emission signal was also reduced, by 60%. When the injector tip temperature rises, the pressure in the spray chamber was shown to rise sharply (C1114).Elevated tip temperatures appeared to favour laminar flow. Fluctuations in aerosol carrier gas flow-rate cause noise in the ICP (see Section 1.2.1.2), and these may be caused by pressure changes in the spray chamber due to poor drain design (C1121).The drain tubing volume should be maintained constant. Differing success continues to be reported concerning the introduction of organic solvents. The signal enhancement expected when nebulizing volatile organic solvents is not achieved in the low-power Ar ICP due to the lowered excitation temperature occasioned by the introduction of solvent into the discharge (C458, C1122). Signal may however be enhanced if the power is increased.A series of trace elements has been determined in organic solvents (oils diluted with xylene and APDC complexes in MIBK) using a 4 kW nitrogen- cooled plasma (934). Both this report and others (C847, C996) stress the changes in background structure which accompany organic solvent aspiration. When spraying the MIBK the background below 200 nm is highly structured, while a xylene solvent shows little structure (C847).Hence, the preference for xylene as dilution agent when determining S in oils (C1136). Oil in water emulsions have been aspirated into a plasma for determining Pb in gasoline and Ba in lubricating oils (538). Considerable claims were made for the accuracy and reproducibility of the technique even when not using organic standards.Only one new nebulizer has been reported this year. In the ‘jet impact nebulizer’, sample was forced under pneumatic or mechanical pressure through an orifice of 60- 100 pm diameter (C1043, C2096). Directing the ejected high velocity stream of solution against a solid surface generated the aerosol. The distance traversed by the jet before impact was critical.Aerosol production was independent of gas flow and hence the nebulizer has potential for low gas-flow plasmas. A major interest in other nebulizer developments, including concentric nebulizers, has been to increase corrosion resistance particularly to hydrofluoric acid (see also Section 1.2.1.4). A concentric nebulizer with a Pt/Ir capillary and inert fluorocarbon body, was described (C14,651).It was combined with a PTFE spray chamber of minimal dead volume, with an impact rod, and an injector with inserts of PTFE at the base and boron nitride at the tip. The arrangement tolerated 25% V/VHF for 2.5 h. Reduced clean-out times were also reported. Another nebulizer with a Pt/Ir capillary and a PTFE body with a PTFE spray chamber fitted with spoiler has been reported (C1140): In this case the injector tube was fabricated from alumina for HF resistance.A third report described a similar system made of PTFE, Pt/Ir alloy and Zr alloy (C902). Cross-flow nebulizers are clearly becoming more popular. The so-called ‘MAK’ nebulizer has been described in detail (650, 654; see also ARAAS, 1980, 10, Ref. C749 and p.39) and is clearly widely used (C2119).Another “market-leader” is the Babington-type ‘GMK’ nebulizer (see ARAAS, 1980,10, Ref. C852 and p.40), which has now been describedAtomization and Excitation 13 in detail (649). The history of this type of nebulizer has been reviewed with emphasis on recent improvements (652). The resultant recommendation was a rugged cross-flow nebulizer with Pt sample uptake tube, sapphire jewel gas orifice and PTFE body operating at 30 psi for routine use, and a sapphire 'V-groove' pumped nebulizer for high dissolved or suspended solids work (see ARAAS, 1980, 10, 40).A useful, if obvious, comment was made that nebulizer clogging was related more to the percentage saturation of a salt solution than to the percentage total dissolved solids content. Other Babington-type nebulizers have been reported.Thelin (389) has reported a simple glass slot-type nebulizer, capable of spraying 10% rn/Vsteel solutions, coupled to a cyclone spray chamber. This nebulizer was fed by a peristaltic pump. Memory and viscosity effects were eliminated and wash-out time was short. A miniaturized nebulizer, designed for low gas- flow plasmas, was described by Ripson and De Galan (621).The solution was force fed, under Ar pressure or by a pump, to the stainless-steel gas orifice of 100 pm bore. The spray chamber had a volume of 10 ml and 0.05 to 0.2 1 min-' Ar was used. Improvements have been reported in the operation of the glass frit nebulizer (see ARAAS, 1979,9,10). Attempts have been made to optimize features such as the frit porosity, diameter, carrier gas flow-rate and sample delivery rate (C439).Gallimore and Ape1 (C76) have decreased the memory effects observed with their original design by altering the orientation of the frit, which is now inclined at 45" to the vertical, and by rinsing the frit from the carrier-gas side. Typical wash-out times of 1.5 min were reported after high pressure side washing.The efficiency of the new design was 40% at an uptake rate of 0.06 ml min - I . Continued attempts have been made to improve the performance of ultrasonic nebulizers for practical analyses. The construction of a commercial nebulizer has been described (653). Walters (C1407) has reported on the performance of two new vertical designs. These low power nebulizers were tested at low-Ar gas flow-rates. Taylor and Floyd (1836) have published details of their modified commercial system (see ARAAS, 1980, 10, 40).This gave detection limits for very dilute aqueous solutions 5 to 20 times better than pneumatic nebulization for 3 1 elements. Ultrasonic nebulization has consequently been recommended for environmental samples (450). No indication of any desolvation problems was given.A preliminary report of an ultrasonic nebulizer for 300 pl samples has been made (C683). Two approaches are now advocated for the delivery of small samples to the ICP. The first involves discretesample nebulization, (first reported by Greenfield and Smith, ARAAS, 1972, 2, Ref. 189) usually from a PTFE cup. The group at Dortmund have described their approach to samples of limited size or high dissolved solids in some detail (C662, 1536).Samples of 50 - 200 pl were injected into a cup attached to a concentric pneumatic nebulizer. A peristaltic pump was placed between this cup and the nebulizer to prevent entrainment of air which caused poor reproducibility with a low power plasma. Alternatively, by using a high power plasma without the pump, smaller sizes of samples, e.g., 50 pl of serum, could be utilized.Using a 3 kW, N,-cooled Ar ICP, detection limits were improved by factors of between 2 and locompared to continuous nebulization. Total dissolved solids contents up to 50 mg ml-I could be tolerated. Some very promising results for discrete nebulization and simultaneous determination of Ca, Cu, F, K, Mg, Na, P and Zn in serum and whole blood after 10-fold dilution with water have also been reported (356).In this case a PTFE cup, a cross-flow nebulizer, 50 pl samples, a Y internal standard and a 1.4 - 1.8 kW forward power Ar ICP were used. Rare earth elements in carbon steel have been determined by a similar method (606). After dissolving a 1 g sample, Fe was extracted into MIBK and the aqueous phase evaporated to dryness.Using a 40 pl portion of a solution of the residue, detection limits of 0.01 pg ml-I for La, 0.2pg ml -I for Ce, Nd, Pr and Sm, and 0.03 pg ml - I for Eu were obtained.14 Analytical Atomic Spectroscopy The second approach for micro-samples, flow-injection, overcomes problems of air entrainment, and although more expensive, it is more readily automated.The advances in the coupling of flow-injected methods to FAAS are noted elsewhere in this volume (Section 1.3.3.2) and have been used as a basis for devising couplings into the ICP. It would appear that no particular problems are encountered. A demonstration of its applicability to rock and plant analyses has been reported (2169). A flow-rate of 3.2 ml min-I was chosen as a compromise between plasma stability and analytical sensitivity. Greenfield (C680, 1884) has demonstrated that flow-injection techniques allow microlitre sampling with reproducible timing and controlled dispersion which thus improves the speed and precision of analysis.The use of medium dispersion methods to allow a procedure analogous to standard addition, e.g., in the determination of Ca in cement, and large dispersion to generate a calibration curve from one standard solution was described.An example of the speed possible with this approach was the determination of 5 metals (Ca, Fe, K, Mg and Na) simultaneously in lop1 of human blood serum at a rate of 240 samples h-' (C998). The speed of analysis was limited by the digital print-out time.An injection systems using an 8-way motor-driven rotary valve with automated switching and peak detectiodintegration, has been reported to give an RSD of 1.7% for 1 pg ml-' Zn using only a 40 p1 sample (C683). Given the speed at which data can be produced once the sample has been introduced into the plasma, it is no surprise that interest continues to be shown in solid sampling, as sample preparation increasingly becomes the limiting step in many applications.Horlick (C2110) has reviewed approaches such as sample insertion, laser and electrothermal vaporization. Laser vaporization has been reviewed elsewhere in this volume (Section 1.1.2). Different designs of devices which directly introduce samples into the ICP on a graphite probe have been studied.Sommer and Ohls (1 87) have reported work with their sample elevator technique, a graphite cup pneumatically elevated into the plasma where it is induction heated to white heat. Detection limits were improved by about an order of magnitude compared to solution nebulization and the technique was successfully applied to steel and oil analysis. A system has been patented which consists of a quartz insertion cup with pierced cap supported by a carbon pin (1690).Pettit and Horlick (C1056, C1079) have used a system whereby graphite cups were automatically introduced from a 24 cup-carousel. The cups were injected pneumatically to allow analysis of sample types ranging from aqueous solution to powdered coal. Spark source sampling is another approach which is being used in a variety of laboratories. It has been noted that this approach combines the advantages of the spark source, precise sampling with little preparation, and of the ICP, long linear dynamic ranges and minimal matrix effects (C1156).This approach is of major interest in the metallurgical industry, e.g., for aluminium alloys (C2130). Incomplete sample vaporization and inefficient transport presently limit the power of detection of spark sampling to a level inferior to dissolution/aspiration (C686).An interesting approach for powders has been the use of a Danielsson tape-machine (453). The sample on adhesive tape was passed through a spark discharge at a tape speed of 12 m min - I . Eighteen elements in phosphate rock were determined using iron as an internal standard and a high power N,-cooled plasma.Another approach for powders with small particle sizes is slurry nebulization using Babington-type nebulizers (1 293). In this paper, atomization efficiency was shown to be dependent on transport efficiency, particle size, atomization temperature and matrix. Grinding samples to less than 10 pm size was recommended.Ward and Carrara (C107) have also made a preliminary report of slurry nebulization. The introduction of oxide powders from a fluidized bed has been reported (C1057). Under optimized conditions it was found that an air ICP discharge decomposed the samples more efficiently than an Ar ICP. An exponentional dilution chamber was used for single-standard calibration.Atomization and Excitation 15 Interest in electrothermal vaporization seems to be waning, perhaps in the face of competition from the simpler sample insertion devices. A graphite furnace coupled to the ICP has been recommended for volatile elements (C662).The use of a Pt or W filament, connected to the ICP via a small volume (4.5 ml) quartz evaporation chamber, enabled picogram amounts of several elements to be detected following pulse heating (C685).The signals were however transient, occurring between 0.25 and 1 s after discharging the heating capacitor. The potential of the filament-in-furnace vaporization technique for micro- samples has been assessed (C1012). Tungsten wire coils supported on a central filament in a vertical carbon tube furnace were used. The use of hydride generation to introduce samples into the ICP is now an established technique. It is, therefore, unnecessary to review this aspect of sample introduction in detail.The reader is referred to the applications tables to judge the extent to which the technique is being used for “real samples”. The experience of hydride methods for AAS must have aided this ready acceptance.Additionally, for optimal use with the ICP simultaneous hydride generation is required. Ward (C434) has suggested compromise conditions for the determination of As, Bi, Ge, Hg, Sb, Se, Sn and Te in a continuous flow system. The working range was 5 to lo00 ng ml-I and chemical interferences were corrected by masking, standard additions or mathematically by computer. It is noticeable that Pb cannot yet be included in any of these comprehensive schemes.A new nebulization system in which the hydride is formed at the junction of the gas and sample needles of a cross-flow nebulizer has been reported and applied to food analysis (C1053). The gas-phase introduction of volatile metal chelates was briefly mentioned last year (ARAAS, 1980,10,41, Ref. C575). This promising new approach has now been described in detail by Black and Browner (563).Either trifluoroacetylacetone or hexafluoroacetylacetone have been used to complex the metals Co, Cr, Fe, Mn and Zn in aqueous solution, bovine liver and human blood serum. The trifluoro-ligand was more successful as the hexafluoro- ligand often extinguished the plasma. The metal complexes were formed by direct reaction between the sample and chelon in a glass reaction cell, heated at 120 “C for 5 to 10 min.The chelates were then swept by Ar into a conventional quartz torch. Calibration was by standard additions. The avoidance of sample preparation and nebulization is particularly attractive, although the method appears to be relatively complex and time consuming. Improvements in detection limits compared to pneumatic nebulization were reported for all elements except Cr, which also gave low recoveries. An alternative approach was to extract the complexes into xylene and introduce these by pneumatic nebulization (C657,2225).This method was used to determine Al, Cr, Cu, Fe, Mn and Zn in biological and clinical samples. Trifluoroacetylacetone was also chosen as the ligand in another ligand vapour method (C726).Using gas injection a detection limit of 0.65 ng Cu was reported. The interferences of Ni and nitrate on arsine generation are well known. Ho and Tweedy (C446) have reported a novel vapour method to overcome these interferences. The sample was dissolved in 6M HCI and AsCI, distilled using hydrazine sulphate and HBr. The AsCI, was trapped in dilute HNO before conventional ICP analysis. A one minute distillation time was sufficient for complete recovery of As from a 10 ml sample solution.The ICP offers a number of advantages as a gas-chromatographic detector, principally an ability to accept gaseous effluent with minimal peak broadening, high sensitivity, wide linear working range and the power to detect non-metals as well as metals.A GC-ICP instrument has been tested for determining alkyl-lead, alkyltin and ferrocene compounds (C820). A low volume in-line photoionization detector enabled additional monitoring of organic species. The group at Kansas State University have continued to use the ICP for non- metallic detection. The extended plasma torch described above (see Section 1.2.1.2) was used as an 0-selective detector (fixed wavelength 777.194 nm) to detect, e.g., ketones, alcohols16 Analytical A tomic Spectroscopy and hetero-cyclics separated on a Chromosorb P/Carbowax 20M column (1561). Unfortunately some unoxygenated compounds also gave responses.Amines and heterocyclics were detected by the same system using N-selective detection at 821.63 nm (2199).In this case a background correction system was employed and this removed apparent interferences from continuum spectral emission. Volatile hydrides have been determined sequentially by a computer-controlled scanning-monochromator following gas- chromatographic separation (C661, C1148). There are many more potential applications for the ICP as a detector in liquid chromatography. In this case the technique is no longer limited to only volatile compounds but additional problems arise in sample introduction.As yet few workers have addressed themselves to this latter problem preferring to connect the effluent directly to a conventional nebulizer. Smith et al. (C660) have considered the problem that solution acidity and flow-rate affect both column and nebulizer efficiency.Clearly it is not easy to optimize these simultaneously. An interface enabling splitting of an HPLC effluent between a rapid- scanning ICP spectrometer and a non-selective detector has been reported (C1161, C2057). Inorganic metal cations were separated by reverse-phase HPLC using either extraction or paired-ion techniques, as well as ion chromatography.Metal chelates of EDTA and NTA have been separated on an anion-exchange column (Bio-Rad AG 1x4) using 0 . 5 ~ ammonium sulphate as the eluant (C94, 2175; see also ARAAS, 1979, 9, Ref. 1855). The effluent was passed directly to the ICP nebulizer, with detection using a polychromator interfaced to a microprocessor programmed to control chromatographic data acquisition, Linearity and precision data for Cu and Zn complexes of NTA and EDTA were reported.Several elements present in a commercial toluene standard as dialkylbenzenesulphonate salts were separated by size exclusion chromatography on Bio-Beads SX-2 (203 1). A cooled spray- chamber was employed which reduced the amount of toluene reaching the plasma. Consequently a lower power and lower flow-rate were used and sharp peaks were still obtained with detectability comparable to aqueous solutions.This system has been employed for the determination of 15 elements in CHCl 3 , THFand pyridine soluble fractions of solvent refined coal (2032). A synthetic mixture of ferrocenes was also separated. The speciation of As is particularly important in toxicological studies. Morita et al.(1 575) have used a Nagel tertiary amine anion-exchange column and a Nagel Sulphonic acid cation-exchange column to separate As(III), As(V), methylarsonic acid, dimethylarsonic acid and arsenobetaine by HPLC. The use of HPLC - ICP - OES offered a detection limit of 2.6 ng As s - I and it was compared to HPLC - AAS and HPLC - d.c. plasma spectrometry. Speciation and quantitation of orthophosphate, diphosphate and triphosphate has been achieved using an anion-exchange column (p-Bondapak-NH 2 ) , an eluant of 0 . 1 ~ oxalic acid, 0 . 0 1 ~ magnesium sulphate and 0 . 1 ~ NaOH and detection at the 214.9 nm P emission line (2036). The addition of Mg2 + allowed gradient elution without substantial base-line drift. Adenosine mono, di, and tri-phosphates were also separated with a Mg2 + and tartrate eluent.Serum proteins have been separated by gel-permeation chromatography and the P was found to appear at two peaks (C819), one with macro-globulin and one in a small molecule. Major peaks of Cu, Fe and Zn were reported to appear at the albumin position with smaller peaks at the macro-globulin and y-globulin position. These results for Cu and Fe conflict with those of others.An HPLC - ICP combination has also been used to determine condensed phosphates in detergents (C818). The 213.62 nm line was used. 1.2.1.4 Instrumentation Perhaps surprisingly there has been little development in this area this year. After the spate of conference papers on mini-ICP systems a major breakthrough in this area had seemed possible. Meanwhile we have to be content with some useful refinements and one major newAtomization and Excitation 17 instrumental concept.The appearance of the first commercial ICP-AFS system is fully reviewed in the Section on new instrumentation (2.4.1) but is worthy of mention here (C20, 1571, C2109). Some consumer resistance has been noted to rapid-scanning sequential spectrometers until their reliability has been thoroughly proven and, given the history of AFS sources, a similarly cautious approach will probably be adopted to ICP-AFS.The most obvious advantages of the approach are a substantial reduction in spectral interferences, fewer background problems and low cost. Many of the traditional ICP advantages e.g., linear working range and absence of chemical interferences, are maintained and detection limits, while different, are not always inferior to ICP-AES.Potential users will carefully watch the performance of the pulsed hollow-cathode lamps used together with a PMT, interference filter and lenses in independent element modules. Literature reports of several mini-torches previously described at conferences (see ARAAS, 1979, 9, 9 and 1980, 10, 42) have appeared.Lowe (622) has described his demountable torch which consumed 2 I min-' Ar for plasma and coolant gas but which accepted sample Ar flows up to 5 1 min-' . A conventional AA nebulizer could therefore be used. The plasma tube had no flared end and the sample introduction tube extended 4mm above the plasma tube. The ratio of plasma tube 0.d. to outer-tube i.d. was 0.79.With an input power of 3 kW, detection limits for Al, B, Cu, Mo and Ti in the ppb range were reported. Ripson and De Galan (C2128) have reported on the problems of maintaining detectability as the gas flows were decreased on their small water-cooled torches. The problems arose not only from sample introduction but also from cooling of the ICP. The apparent vaporization and ionization interferences observed in a miniature torch have been investigated by Savage and Hieftje (145).By considering the height of observation, applied power and nebulizer gas flow-rate, compromise conditions for simultaneous multi-element analysis were identified which offered minimal interference. Similar factors were also found useful in improving precision in background correction studies with the same torch (359).An extended coolant tube (see below) was recommended to reduce air entrainment and emission from bands such as NO, NH and OH. The same group have postulated that, from consideration of the 'skin-effect', 9 mm is the smallest torch diameter that can be used if aerosol gas is to have an insignificant effect on r.f. coupling (360).To produce the same power density as a conventional torch, 570 W was then required. Such a torch was constructed, operated at 7 1 min-' Ar, and shown to offer detection limits comparable to conventional torches. While easily ionizable elements showed greater effects than in a conventional torch, the new arrangement was successfully used to analyse NBS Orchard Leaves.Three further water-cooled torch designs have been described (160).The most promising required 4 1 min-I plasma gas and 0.8 1 min-' carrier gas. This torch design could be used with a conventional nebulizer and offered competitive detection limits. The cooling water only interacts with the top of the torch which prevents air or steam bubbles being trapped. Some practical modifications to torches have been reported.There has been considerable interest in HF compatible systems (see Section 1.2.1.3). A torch with an alumina central tube capped with either molybdenum and platinum or sapphire has been described for such an application (C100). The sapphire tip did not clog so readily but contributed a 50 ppb A1 blank. Extended torches have also been described above (see Section 1.2.1.2).These are reported to reduce air entrainment and therefore reduce background (e.g., from CN, NO, 0, and OH bands) particularly at low wavelengths and can be purchased as sheaths which fit over the outer tube (1327). Some laboratories may value another suggestion for salvaging torches with damaged or melted outer tubes (2183). The damaged portion is sawn off and replaced by a stepped extension.This extension may also act to prevent air-entrainment if the correct height is selected.18 Analytical Atomic Spectroscopy Capelle and Mermet (C1095) have studied the influence of thegenerator frequency in the range 5-56 MHz. The excitation temperature decreased (from 6600 to 4000 K) with frequency decrease, as did the electron number density.As the Ar continuum intensity decreased faster than the line intensity, SBR increased with frequency. This improvement was confirmed in measured detection limits, although it should be noted that the frequency was the only parameter which was altered. A new 40.68 MHz quartz oscillator r.f. generator employing a Pierce circuit has been described (C717). A novel impedance matching network enabled greater movement of the torch (C2096).Low frequency (30 to 300 Hz) modulation of the r.f. power was shown to have a pronounced effect on the analyte line intensities, especially when concomitants were present (C81). Some “ionization interferences” were reported in “modulated plasmas”. A 256 element linear photodiode array spectrometer was used to generate lateral emission profiles for Mg atomic and ionic lines (968).After direct feeding to a minicomputer for Abel inversion of the data, a three-dimensional picture of the plasma was generated. The emission followed a “hollow bullet” contour low in the plasma which became a solid emitting plume above 12 mm. A critical evaluation of self-scanned photodiode arrays for ICP-OES has been made (C1138). Provided integration times were adjusted properly, powers of detection equivalent to PMT detectors could be achieved but a compromise had to be made between resolution and spectral coverage.Some preliminary results with a Fourier transform spectrometer for ICP-OES have been reported (C1143). Broekaert (2092) has noted that commercial polychromators often squander one order of magnitude in detection limits by sacrificing resolving power in favour of greater thermal stability.In an attempt to improve resolving power more use is being made of echelle spectrometers for ICP-OES. Such a combination has been used to study AE spectral profiles (C710). Another echelle system with wavelength modulation has been used to minimize certain interferences (C720).A new type of 1 m Czerny-Turner vacuum spectrometer with two exit slits at right angles has been developed (C737). This scanning monochromator made measurements at wavelengths as low as 150 nm. An image memory type TV camera was used to measure intensities. Commercial manufacturers continue to extol the advantages of rapid scanning spectrometers (C111, 460, C712, C1141, C1157, C2114).The importance of graphics and software have been stressed (C111, 460, C i 157) and also the possibility of rapid semi- quantitative analysis (C7 12, C2114). A simple and inexpensive slave micro-processor (stepper-motor controller chip) programmable in BASIC has been built to control a 0.85 m double monochromator for ICP-OES (2027). A 64 K bytes memory allowed storage of data from 50 peaks.Considerable flexibility was built into the rapid-scanning facility, and the use of BASIC suggests that operators might be able to use the full power of this system. Full details have now been published of another sophisticated scanning system (1 3 13). All measurements were made in the scanning mode across a line allowing a choice of background correction. A 4 K bytes memory allowed up to 10 elements per run.The smallest programmable step-length was 2.7 x nm. The system of Ferreira et a/. (C1405) offered a step size of 1 x lop3 nm using a stepper motor with a 50:l gearing and angle encoder fitted directly to the grating mount. The resolution reported was impressive (0.0018 nm between 2 signal pulses) and graphics facilities were provided.The ready availability of laboratory minicomputers has led to a remarkable revolution in data acquisition methods in recent years. Pickford (C1331) has reported on the use of such a computer not only to collect the process data for 39 elements from a polychromator but also to provide an interface to a programmable high resolution monochromator. The use of software capable of determining base-line drift in OES, and other factors which may affect the variance of an integrated signal, has been described (368).Spectral evaluation is greatly aided by graphic display, e.g., on a visual display unit with floppy disc storage for later recallAtomization and Excitation 19 and evaluation (C6). Similarly, floppy discs may aid storage of data in archival form for later scrutiny (C858, C1425).De Galan and Kornblum (C687) have speculated that basic data handling and automated instrument control have been mastered. Steps must now be taken to automate the optimization of analysis conditions and the quality control of analytical results. This will be dependent on a fuller understanding of the physical processes in the ICP. Montaser (C 1096) has used a mechanical skimmer located below the usual observation height of an ICP to isolate the central from the outer Ar flows.Improved SNR and detection limits were obtained in the skimmed region, particularly at higher wavelengths. Some evidence of higher levels of ionization-type interferences was found. I .2. I .5 Interferences The recognition and classification of ICP interferences has matured, aiding understanding of how such problems may be overcome. Spectral interferences remain the most troublesome type of interference. Mermet and Trassy (1538) have discussed these interferences particularly drawing attention to the effects of line broadening and the need for careful choice of lines.Such line choices will be aided by the spectral atlases now becoming available (see Section 1.2.1.2).Boumans (462, C708) has discussed the use of his line coincidence tables by means of a computer software package. The need for such tables is illustrated by continuing new reports of interferences, e.g., MoII 213.61 nm on PI 213.62 nm (1831); Al, Ca, Fe and Mn on Si 180.73 nm (1557), although the Ca and Mn interferences was reduced by using a lower observation height; W interferences on 11 elements in steel (2091); Mg 383.2 nm (second order) on K 766.4 nm (first order) (471).Improved resolution reduces some interferences and increasingly echelle spectrometers are been advocated for precisely this application (C84, C456), sometimes as accesories to existing polychromators (656). In this latter instance troublesome spectral interferences on As, Be, Cd, Pb, Sb, Se and TI in coal, shale and fly ash analysis were much reduced.Perhaps the ICP-AFS system referred to above (Section 1.2.1.4) should be regarded as the ultimate in resolution and certainly it is being strongly advocated as a way of circumventing spectral and background interferences (C20, C25, 1571, C2109). Others have also argued for the use of AFS to minimize interferences and the use of this approach, or a flame as a resonance monochromator, has been demonstrated (C727).In a paper which gives an excellent review of spectral interferences, Gautherin et al. (637) have reported results using a 5 mm diameter laser beam which entered the ICP along the axis of symmetry. When viewed at right angles 35 mm above the load coil an A1 detection limit of 3 ng ml - I was obtained.This system enabled the determination of ng ml-’ levels of Al in g 1 -’ levels of Ca, Mo and Zr at 396.2 nm which would be impossible by AES. Any supplier of a scanning system would advocate selecting another line when faced with such a problem, especially if a visual display unit was available to indicate the degree of overlap (346, C437, C449).This may indeed be a viable approach, e.g., the use of B 182.64 or 208.96 nm lines in steel analysis to overcome Fe interference at 249.7 nm (670). A generalized standard addition method, capable of characterizing all analytical wavelengths of an ICP, has been advocated as a method of multi-component analysis for the detection of interference effects, and as allowing the use of the most sensitive wavelengths (2219).The critical importance of background correction in practical analysis methods has yet to be fully recognized. This has been noted in a review and means of correction presented (C1 1). Causes such as stray light, ion -electron recombination, molecular bands and collisional line broadening were also discussed. So-called ‘on-peak’ and ‘off-peak’ background correction methods have been compared (C12, C2108).While the on-peak approach can correct for direct spectral overlaps, it cannot correct for non-specific background shifts. The off-peak, or scanning across the line, approach can correct for such20 A nalytical A tornic Spectroscopy shifts but cannot correct for direct spectral overlaps and is slowe‘r.The use of both approaches has been advocated for marine sediment analysis ((306). Wavelength modulation techniques can be used for real-time background correction. When used with an ICP coupled to HPLC, the frequency of the quartz refractor plate was increased to 200 - 500 Hz so as not to broaden the chromatographic peaks (C719). The use of phosphoric acid is becoming increasingly prevalent in ICP analysis, as traditional phosphate interferences encountered in flames are absent in the plasma.It is probably well known that transport effects may arise from high concenirations of phosphate but the background shifts encountered may not be so well known (C79). Selective spectral-line modulation has been advocated for the detection of resonance radiation and thus for background correction (C718, C1135, C2096).Radiation from the ICP was directed alternately through and around a flame containing the analyte. The detection system was synchronized to the rotation of the chopper. Spectral interferences such as CaII on AII at 396.1 nm were largely eliminated. The advance in our knowledge of the effect of easily ionizable elements in the ICP, consequent upon the contributions from several workers considering spatial plasma characteristics, was highlighted last year (ARAAS, 1980, 10, 44).Koirtyohann et at. (616, C2118) have continued their valuable work in this respect. Severe enhancements, e.g., by Li and K on CaI 422.7 nm and CaII 393.3 nm were observed in the initial radiation zone but these enhancements tended to disappear in the normal analytical zone.In the lower zone, the easily ionizable elements tend to alter the apparent excitation temperature rather than atom formation. Blades and Horlick (C78) have commented on the confusion in the literature about such effects and have given supporting observations for spatial explanations. Ambipolar diffusion appears to be responsible in part for the spatial redistribution of emission.Other contributions have also been made to this subject (C438, C703). 1.2.2 Microwave-excited Plasmas While regular conference reports on microwave plasmas continue to appear, few of these ever seem to become full papers. An exception is probably the use of the MIP as an element specific GC detector. There has been a modest increase in reports about capacitatively- coupled plasmas and perhaps we can detect the first signs of a reawakening interest.The papers in this area are discussed in Section 1.2.2.4. Zander and Hieftje (1794) have reviewed historical, fundamental and practical aspects of MIP’s. This extensive review suggested that the MIP is under-exploited and described in detail suitable instrumentation.Mechanistic aspects werealso covered. Beenakkeret al. (556,2067) have published two reviews of the MIP operated in their TM,, , ,, cavity. The first dealt with mechanistic, design and spectral aspects. The second covered analytical applications of He or Ar MIP’s as GC detectors and in combination with electrothermal vaporizers. Limits of detection and matrix effects were discussed.Haraguchi (C701) has also reviewed the atmospheric He MIP formed in a Beenakker cavity. The existence of non-LTE conditions, available wavelengths for Br, C, CI, F, H, I, N, 0, P and S GC detection and the use of vapour generation sample introduction were outlined. A further review of current MIP studies stressed interferences and how these could be combated (638). 1.2.2. I Fundamental Studies The spectral and mechanistic aspects of He and Ar plasmas in a TM,, I ,, cavity have been studied (C697). In the He plasma, the following temperatures were found rot (OH) I300 K , Tot (N,+) 1400 K, Txc (He) 3300 K, Ton (He) 9500 K, and an electron number density of 4 x 10’ ~ m - ~ . It was suggested that metastable He species played an important role in ionization and excitation, hence the non-LTE conditions observed.The relative atomization efficiences of the halo-carbons in He MIP’s are important in several applications. Nikdel etA tornization and Excitation 21 a/. (C56) were interested in the He MIP as a potential in situ monitor of stratospheric Br and C1 levels. They found that detectability improved at lower pressures, a linear dynamic range of 4 orders, and that halogen signals were independent of molecular speciation.Ail atmospheric pressure N , MIP has been sustained in a 6 mm 0.d. quartz tube in a modified Beenahker cavity (C111). At least 200 W power was required and pneumatic nebulizers could be used to introduce sample into the annular plasma formed. Signal-to-noise ratios, precision and working curve linearity were measured for Ba, Cd, Cu, K , Li, Mg, Na, P, Pb and S and found to be comparable to other MIP sources (C1112). Theexcitation temperature in the N , MIP was lower than in a He or Ar MIP. Atomization in an MIP can be improved by increasing the electron density.When an external magnetic field was applied to an MIP it was pinched (C1097). The resultant increase in electron density gave doubled emission intensity.Sealed lamps, containing Ne and the analyte, were used in a T M , I ,, cavity. Wavelength tab/es for 10 non-metallic elements (Br, C, C1, F, H, I , N, 0, P and S) in the TM,, , He MIP with relative intensities and transitions in the range 190 - 850 nm have been published (1527). Non-resonant emissions from C, N and 0 in the near-infrared (800 to 2000 nm) from a He MIP have been recorded using a Fourier transform i.r.spectrometer (C1125). Douglas et a/. (C55, C1137, 2209) have used a coupled MIP-mass-spectrometer for elemental determination (see also ARAAS, 1980, 10, Ref. C1162). A quadropole mass spectrometer and a 200 W TM,, , ,, Ar MIP were used. Samples were introduced by ultrasonic nebulization and the ions generated in the plasma transferred to the spectrometer by a differentially pumped interface.A wide orifice eliminated boundary layer effects and detection limits in thepg 1 range were reported. Up to 200 ppm Na did not suppress the ion concentrations. This source therefore looks to be a promising competitor to ICP - M S (we Section 1.2.1.2). Analytical application of a variable capacitatively coupled reyonant cavit-v ha5 been reported (C96).The ability to tune the resonance frequency of this cavity to match that of the generator eliminated the need for external tuning stub$. The more efficient coupling resulted in better power transfer. The thickness of the cavity was alyo increased so as to lengthen the placma, hence improving emission intensity. 1.2.2.2 Sample Introduction The MIP is best suited to samples introduced as vapours and this has been reflected in the rriaterial published. Liquid samples can be introduced viaelectrothermal vuporizarion. Alder and Da Cunha (170) have devised a special cell to speed application of samples to a carbon filament vaporizer. A low pressure (5 to 20 Torr) Ar MIP was used and the cell was consequently evacuated.Results similar to those obtained with atmospheric pressure plasmas were reported. A mini-furnace has been coupled to atmospheric VIP's sustained in a 3/4 wave Evenson cavity +or a Beenakker cavity (C1016). The potential for determinin!: trace metals in sea-water was evaluated. The combination of a W wire filament vaporizer and a low power atmospheric MIP i n a modified Beenakker cavity has been reported ( 0 9 ) .A typical two-stage (dry and vaporize) programme minimized plasma perturbation. An intriguing evolved gas analysis/emission spectrometer has been reported for identifying inorganic compounds in environmental particulates (1965). The sample was heated from 25 to 1000 "C at 140 "C min-I in an induction furnace. As the componentsvaporiLed, they were swept into a He MIP i n a TM,, I ,, cavity.Coincidence of metal and non-metal emission indicated the compounds present, but unfortunately problems were encountered with chemical reactions in samples before vaporization. Various vupour genet-ation techniques can be coupled to the MIP. Nitrogen has been evolved from ammonium ions, using sodium hypobromite, and from nitrite and nitrate, (which was first reduced to nitrite) using sulphuric acid (156).The N , was swept by He into a22 Analytical Atomic Spectroscopy Beenakker-type He MIP. Emission of NT at 391.4 nm was monitored, yielding a detection limit of 4 ng N ml - I . Hydrogen chloride, evolved from chloride containing samples by K,SO,/concentrated H,SO,, has been used to determine Cl(536, C922).AgainaHeplasma was used and a detection limit of 6 ng at the 479.5 nm ClII line was reported. 1.2.2.3 Gas-chromatographic Detection Commercial GC - MIP systems have been available for several years, but reports continue to appear of novel combinations from different laboratories (C509, C2057). The MIP is well suited to gaseous samples and simultaneous determination of metals and non-metals can greatly aid identification of chromatographic peaks.A review of element selective detection using the Beenakker plasma has been given by Haraguchi and Fuwa (C1460) with illustrations of trimethylfluorosilane used to determine fluoride and also alkyl mercury determination. The performance of the I /4 wave Evenson and Beenakker cavities has been compared (1 541).No cavity, plasma/eluent gas, or gas pressure combination was found to be best in general even for simple alkanes. An optimum choice had to be made for each application. The group at Amherst have published several papers reporting high-resolution GC using capillary columns and a He MIP in a Beenakker cavity. A chemically deactivated gas switching interface allowed excess solvent to be vented without disrupting the plasma (1762).The inertness of the silanized glass interface was proved by the separation and detection of trialkyl-lead chlorides in tap water. A fused silica wall-coated open-tubular capillary column was also used to separate cyclopentadienyl and carbonylcyclopentadienyl compounds of Co, Cr, Fe, Mn, Ni and V prior to MIP detection (237).Boronate derivatives of various diols were similarly separated and the B emission monitored (2086). The performance of their system for 29 elements has been reported (1574). The use of the fluidic logic gas switching system and a quartz refractor plate for background correction are obviously important in practical analyses. Detection limits were in the range 1 - 150 pg with linear ranges of about 500 - 1OOO.I . 2.2.4 Capacitively-coupled Micro wave Plasma A new microwaveplasma torch for OES has been reported by Feuerbacher (664). The torch is small, only 120 mm long and 50 mm diameter, and does not need tuning or impedance matching. The gold tip of the torch acts as an electrode for coupling 100-700 W power. There is a shielding gas facility and aerosols can be accepted at 0.5 - 3 ml min - I .A modified torch with silicone rubber spacers between the inner and outer walls was reported to produce a more stable vortex flow in the sheath gas (617). The plasma was thereby effectively pinched and stabilized. Higher SNR was reported. A 1 kW, 2450 MHz, CMP Ar plasma and sheath gas was said to melt A1 electrodes (C713).These were replaced by Ag/W alloy and Ti hollow electrodes used to introduce the sample aerosol. Net line intensities increased with power except for B and P . A preliminary report of the optimization of the parameters which affect the analytical performance of a surfatron plasma has been made ((308). Dahmen (665) has reviewed the operation of the CMP. The most sensitive lines, relative sensitivities and detection limits both in aqueous solution and in 0 . 1 ~ Cs+ solution for 46 elements at 242 prominent lines were listed. In the determination of Ni in iron or steel, sensitivity and selectivity were improved by using Mg, or Mg and Fe as spectroscopic buffers (162). The detection limit was 5 pg Ni 1 - I . The Hanamura air plasma (see ARAAS, 1978, 8, Ref. 176) has been applied to the direct measurement of airborne As (C742).The 235.0 nm line was used with a detection limit of 1Opg 1 --I, which is considerably poorer than the 30 ng 1 - I and 20 ng 1-' limits previously reported for Hg and Pb in air (see ARAAS, 1979,9, Ref. 537).Atomization and Excitation 23 1.2.3 D.c. Arc Plasmas A noticeable decrease in the number of papers reporting work with this type of plasma has coincided with the marketing of an ICP source for the echelle spectrometer formerly coupled almost exclusively to the d.c.arc plasma. It will be interesting to see whether this coincidence marks the sign of a sustained decline of interest in the d.c. plasma. Certainly many ICP advocates have been predicting such a decline as the superior detection power and freedom from interferences of the ICP became more widely known, I .2.3.1 Fundamental Studies The design and performance of the commercial three electrode d.c.arc plasma has been reviewed (268). Some of the fundamental processes which contribute to the pinch effects seen with this plasma have been studied (C740). It has been pointed out that the sample aerosol in both the d.c.plasma and ICP are heated from the plasma boundaries; in the former case this boundary is on one side only (C1109). The author uses this to explain the lower background in the d.c. plasma and it is also, perhaps, the cause of lower emission signals. Water-cooled electron probes have been used to compare the relative electron concentrations in the two and three electrode plasma (C50).The three electrode plasma was shown to have the higher concentration by roughly two fold, probably because of the presence of two 7 A current streams. The electron gradient was also greater in this plasma with a 75" electrode angle, particularly when easily ionizable elements were present. More accurate measurements of electron concentrations can be made by measuring the Stark broadening of H lines.Such measurements are tedious if a spatial profile is required. To solve this problem Williams and Coleman (C1098) used a photodiode array and obtained complete 5 mni2 contour maps in one hour using computer control. Again, dramatic changes in electron concentration gradients were observed in the presence of easily ionizable elements known to enhance emission.Attempts to use a d.c. plasma as an atomizer in a Zeeman AA spectrometer were only partially successful (C896). The stray light from the plasma being more intense than that from a graphite furnace, posed additional problems in locking into the lamp signal which was therefore modulated at 6 KHz. Additionally, the short absorption path length, 1 cm, was unfavourable for AA. 1.2.3.2 Sample Introduction Perhaps the most significant contribution in this area has been the publication of the work of Mohamed, Brown and Fry (985) on slurry nebulization. Using a Babington-type V-groove nebulizer (see Section 1.2.1.3), homogenized liver samples were aspirated into the d.c. plasma in an analogous way to work using a flame. The very wide aerosol opening into this torch, 5-fold wider than into a Fassel-tube ICP torch and 10 times wider than an AA slot- burner, eliminated clogging problems and allowed lengthy integration times. It is to be hoped that a more detailed evaluation of this system will soon be reported. Preliminary developments of a rotating d.c. plasma jeA (see ARAAS, 1975, 5, Ref. 1246) including gas- swirling have been reported (C5 1). The aim of the work was to improve aerosol introduction by the application of gas-flow dynamics. Further studies of the application of discrete nebulization have been reported (C735; see also ARAAS, 1980, 10, 47). Nebulization efficiency as high as 27% for a 17 p1 sample at an uptake rate of 0.12 ml min - I was claimed in comparison to the 15% claimed for continuous nebulization. These studies are of interest in the coupling of GC and LC to the d.c. plasma, as is a preliminary report of the use of organic solvents (C 1054). A gas-chromatographic interface system has been designed and implemented for high resolution fused silica column separations (C1139). The interface allowed dual plasma and flame ionization detection with minimal peak broadening. Several applications of this system were described (see also ARAAS, 1980, 10,47).24 A n alyticgl A torn ic Spectroscopy I . 2.3.3. Interferences Emission enhancement by easily ionizable elements remains a major area of interest in d.c. plasmas. On the basis of spatial profiles and temperature measurements, Nygaard and Gilbert (619) have suggested that elements such as Li increase the effective excitation of analyte while decreasing the temperature of the plasma core in the two electrode plasma. Improved penetration of the analyte. combined with higher temperature at the plasma periphery, was given as the explanation of signal enhancements. Similar careful spatial studies of the two electrode plasma for Group IIA element$ showed great \,ariations depending on the element, and concentration (1717). An anodic shift of the ion zone with respect to the atom zone was observed for Be, Ca and Mg, while Sr 5howed the reverse. It was suggested that easily ionized elements interacted early in their passage through the plasma and were ionized by “low energy carriers”. Less easily ionized elements were not excited until they had penetrated further into the plasma where more energetic species were available to excite them. Observed enhancements, measured electron densities and temperatures have been used to produce a mechanistic model of NaCl enhancements (C1110). The data was consistent with an excitation mechanism relying largely on radiative transfer, resonant energy collisions and .4r metastable populations. The degree of ionization in the analytical zone was higher than predicted by LTE. Sodium enhancement cannot be eliminated and proper compensation was advocated. In a study of the effect of NaCl on spectral line intensity and axial particle distributions, it was concluded that the same phenomena govern both the particle density distribution and the intensity optima (1 55 1). While in commercial systems the use of an echelle monochromator minimizes spectral interferences, the scale of potential problems encountered in rare-earth analysis is worth recording. Johnson and Sisneros (C1162) reported 20 OOO emission lines between 200 and 800 nm produced by rare-earth solutions. Between 275 and 475 nm 80 lines per nanometer were reported. Not surprisingly, this presentation emphasized spectral-line selection and background correction.

ISSN:0306-1353    

DOI:10.1039/AA9811100006

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

24 A n a/yticd A torn ic Spectroscopy 1.3 FLAMES 1.3.1 Fundamental Studies Again this year reports of new flame types were conspicuously absent, but a method described for the automatic control of the air supply to an air-fuel oil flame with the object of reducing NOZ, CO and soot formation, has possible application in analytical flames (1 197). The method used the emitted heat to generate an electrical signal which controlled the air Sheuthed flames have been used to increase sensitivity in AAS (1227) and the effects of both N , gas and quartz sheathing on rise velocity profiles in common laminar analytical flames have been shown to be strongly dependent on oxidant/fuel ratios (1535).An apparatus has been described for measuring temperature profiles of flames used as 5ources for atomic and molecular spectioscopy.Atoms introduced a$ a probe were monitored spatially by a laser/vidicon system (1570). The same system was used to map atomic and molecular species in flames using YO and ScO in N ,O/C1 H flames and Ba and CaOH in air/C,HZ flames. Temporal resolution was provided by means of a Bragg cell (1560). Laser Fraunhofer diffraction has been used to study the droplet size distribution produced by nebulizer systems; the measurements were independent of composition, refractive index and velocity (2149). Practical application is rcstricted to particle sizes greater than 2 pm.The conclusions from a theoretical and practical comparison of noire levels for Ca by AAS, AES and AFS in an air/H, flame were in accord with “received” wisdom; viz at the detection limit AES and AFS are background shot- and flicker-noise limited and that AAS CUPPlY.Atomization and Excitation 25 was lamp flicker noise limited (1613).At higher concentrations analyte flicker noise was dominant for all three techniques. Two new noise concepts were identified, “self-absorption noise” and “inner filter” noise, but these are not of major significance. The practical value of these comparisons is limited in view of the flame used, i.e., air/H,.Two-channel instrumentation has been devised to investigate analyte absorption noise by measuring, simultaneously, absorption noise associated with the hydroxyl group and that associated with one of the elements Ag, Cr, Cu or Mg (C893). In a study of the effects of the wavelength modulation frequency, modulation amplitude and modulation waveform on SNR in continuum source AAS, the relationship of frequency and SNR was found to be complex (C1101). At modulation frequencies of 50 - 60 Hz, 3-step square-wave and bi-gaussian-wave modulation gave comparable SNR and both were superior to a sine-wave modulation waveform.The enforced variance technique of determining, by monitoring between measurements, the number of integrations required to provide a predetermined coefficient of variation has been demonstrated with a spectrometer linked to a programmable calculator (C834).Precision (RSD’s) between 0.4% and 0.65% for Al, Ca, Mg and Si were obtained. An expression has been derived for the dependance of instrumental RSD and element concentration in FAES (1400). Criteria for standardization of AA procedures have been laid down (909), and the advantages of a weighted least-squares over a conventional least-squares procedure for calculating calibration curves have been pointed out for the (usual) case where the calibration curve is non-linear or the variance is concentration dependent (152).This method was applied to the AAS determination of Cu using discrete nebulization, of Fe with conventional nebulization using an air/C,H, flame, and of Pb in blood using the Delves cup.Significant improvements in precision were obtained at the extremities of the calibration ranges. Expressions have been devised for correction of AA signals for hyperfine structure of the absorption line and for its shift in relation to the primary emission line (219).The parameters were calculated for Hg 253.7 nm, Mn 403.1 nm and Pb 283.3 nm, at temperatures up to lo00 K for Hg and 3000 K for Mn and Pb. It was claimed that correction for line shift is necessary for accurate results. Other references of interest- Application of organic solvents in AAS: 1727. Burner assembly; explosion hazard: 260. 1.3.2 Atomization and Interference Studies The elucidation of atomization and ionization processes in flames continues to be furthered by application of the single droplet generation approach (see, for example, ARAAS, 1980, 10, Ref. 901 and earlier papers). The mechanisms of vaporization (heat transfer and mass transfer control) and ionization, (pseudo 1st-order collision kinetics) have been incorporated into a model of emission intensity versus time profiles for an air/C2H2 flame (C127, C833, 1533).Vaporization rate constants and first-order ionization rates have been determined for Cs, K and Na for fitting to this 3-equation model. A general theory of the dissociation of particles and molecules and of the diffusion of molecules and atoms in a flame has been presented and this helps to explain the differences in atomization efficiency for different elements and also chemical interference effects.Atom concentrations were shown to be proportional to the dissociation rate constant and to time. The effect of lateral diffusion processes in reducing atomic density in the optical path was explained and the prediction was made that the most efficient atomizer would be a miniature high-temperature flame or plasma in which the sample atoms are introduced immediately below the optical path (C36, 624,2227).Spectroscopic and gas-chromatographic analyses of the flame gases were used to provide air/H, flame composition profiles; it was shown that H atoms play an essential role26 A nalytical A tomic Spectroscopy in Sn atomization but methyl radicals from organic solvents interfere. The optimum position for Sn atomization was found to be just below the tip of the blue primary reaction zone.The effect of flame type and stoicheiometry on the degree of atomization has been studied for several elements. For Bi in AAS, the sensitivity was found to decrease with increase in flame temperature, provided that allowance was made for the different path lengths and different values of aerosol per unit volume of flame gases in the various flames (1354). The effect of air/C, H flame stoicheiometry on the degree of atomization of Cr was studied and absorbance versus height profiles for several Cr compounds investigated (91 3).Compounds found to have low thermal stability by differential thermal analysis and thermogravimetry were found to exhibit higher AAS sensitivity than did thermogravimetrically stable compounds; hence the atomization efficiency was dependent on the Cr compound formed in the flame after desolvation. In a further study, the influence of alkali chloride and nitrate matrices on Cr was found in general to retard Cr atom formation (912).The equilibrium distribution of Ti and Zr for various C/O ratios in C,H, flames was investigated by a free energy minimization approach which took account of solvent uptake rate and of the condensed phase, formed in the flame, which could include oxides, carbides and nitrides (1501). The best fueVoxidant ratios for the AAS determination of Ge were calculated theoretically to beC/O = 0.32and it was concluded that atomization was primarily due to thermal dissociation of GeO (1494); optimum N,O and C,H, flow rates were 6.7 and 5.32 1 min-’, respectively.Of the effects of various concomitants listed, only acetic acid and acetone significantly affected the absorption and they enhanced Ge signals by 27 and 70% respectively. The methods developed were applied to the determination of Ge in magnesium alloys.The importance of small drop size in minimizing interference effects has again been stressed (Cl055). The role of surfactants in reducing the number of large drops was adduced to explain the reduction of interference on Cr by AAS in an air/C,H, flame from matrix components including Al, Co, Fe and Ni (155). In a study of the effect of cationic, anionic and non-ionic surfactants in the AAS determination of Ca, Cr, Fe and Mn in an air/C,H, flame, the conclusions were reached that non-ionic surfactants generally enhanced absorption except for Cr but that for anionic analyte species such as CrO, - or Fe(CN), - anionic surfactants enhanced the absorption (C842).Investigation of interference effects by atomic absorption titration methods has yet further eludicated the mechanisms of phosphate interference on Caand its release by La (365, 1216).The formation of calcium metatitanate as the cause of interference by Ca on Ti by FAAS was also identified (1847). Lanthanum was again used as releasing agent but other major interferences from Al, PO -, SiO, -, SO, -, V and W still occurred.Apparatus for carrying out this titration technique has been described (194) and used to study the change of the absorption signal of Co with changing CN- concentration and that of Mg with changing A1 concentration (255). Many investigations of interferences on specific elements in AAS have been made this year including Ca and Mg in aluminium chloride matrices (1399) and the interferences of SrOH and CaOH on Ca and Li (1221).The interference of PH 3 , present in bottled acetylene, on Ca by a mechanism similar to that of PO, - was removed by the use of La as a releasing agent (C37) (see also “Flame Spectroscopy” R. Mavrodineanu and H. Boiteux, J. Wiley, 1965, p. 62). The interference of iron matrix components on Cr was suppressed by the formation of a Cr/I complex (301, 317).This inhibited the formation of mixed oxides of Cr and the iron matrix elements. The deviation of AAS calibration curves for Hf, Nb, Ta and Zr towards the absorbance axis in HF solution in the N ,O/C H flame was explained as due to oxide formation rather than to self-suppression of ionization (1 163). The explanation was supported by the straightening of the curves and the enhancement produced by adding 2000Atomization and Excitation 27 mg 1 -I Al.This explanation of the effect of A1 on these elements is contradicted by a study of the use of A1 to correct similar calibration curvature for Ti and U in the N20/C2H2 flame (C38). This study showed that A1 enhanced not only the elemental signals but also the T i 0 signal. In the determination of V by AAS in a N 0 / C 2 H flame, the oxidation state of V was shown to be important (1 165).Hydrogen peroxide was used to convert the mixture of V(1V) and V(V), obtained on dissolution, to V(V) before analysis. The formation of polymeric anionic V(V) species gave erratic results which were prevented by the additional use of NH, C1 as a complexing agent. Instrumental methods for combating interference effects included the ingenious use of selective modulation at the analyte wavelength in a FAES method for Ba which overcame interference by CaOH at 554 nm (165).Light from the primary flame was alternately passed through or round a second flame supplied with analyte to achieve the selective modulation. A system previously described for presenting a visual display of absorption spectra over the modulation interval in a continuum source, multi-element, AAS system has been extended to provide a permanent record of the spectra (C125).Its use as a diagnostic tool for spectral interference was advocated. The study and application of Zeeman-effect methods in AAS has again received attention. The results of a study of 210 absorption lines for 38 elements reported their Zeeman splitting sensitivities and for 60 selected transitions, Zeeman-AAS sensitivities were compared with conventional AAS sensitivities (1488). The technique has been applied in FAAS (C42) and for multi-element non-dispersive AFS (C1069).The use of magnetically induced dichroism (MID) with flame atomization has been assessed in terms of detection limits and ease of background correction.It was concluded that MID - AAS was not superior to FAAS in detection limits, but its background correction facility was comparable to Zeeman effect - AAS (C839). Other references of interest- Concentration of cations and reduction of interference by dialysis: 1837. Mechanism of enhancement of Mg by carbon black in AAS: C841.Molecular absorption of Ga compounds; interference on Fe 248.3 nm: 282. 1.3.3 Devices for Sample Introduction (see also section 3.1.4) 1.3.3.1 Nebulizers Amid much general discussion of the importance of understanding the processes of aerosol production (C 1038, C1039), the newer methods of measuring aerosol droplet sizes using laser scattering techniques have again received considerable attention.The contribution from the componenets of the nebulization system, impact bead, chamber and burner bowl have all been assessed (C835, C 1014). Mass transport data from laser scattering measurements have been correlated with data from cascade impactor studies (C 1040). Among several new nebulizer designs an inexpensive (< $200) single droplet generator from the Hieftje group has been described with full constructional details (2174).Greater stability and better noise immunity were obtained from this design. A new concentric nebulizer, which incorporated no revolutionary features, provided modest improvements (2 x ) in AAS detection limits and precision by careful attention to design (C34, 983, 1489). Improved sensitivity by 40 - 50% was also claimed.Maximum precision was attained at sample flow rates lower than those required for optimum signal. The use of peristaltic pumps to supply solution for nebulization has been introduced principally to permit dilutions, additions and standard additions to be made on the continuous flow principle, Up to 20-fold dilutions were possible in the AAS determination of Ca, K, Na and Sr (2160), and up to 160-fold in Ca, Li and Mg determination in serum and Ca28 Analytical Atomic Spectroscopy and Mg in urine with overall RSD's of 1.8 - 6.2% (1569).Additions of releasing agents and standard additions have also been made by means of a triple capillary nebulizer system used for the determination of alkaline earth elements in the presence of Al, PO, -, S O , -and SO, - matrix components (1277).Sample modulation at 8 - 20 Hz has been obtained with a motor driven piston used to apply a periodic pressure wave to the nebulizer in an air/C,H, flame AES system (C1052). The measurement system used a lock-in amplifier and in addition to the reduction in instrument noise (also obtainable by chopping), reduction of analyte flicker noise was obtained.Alternation between sample and blank nebulization at 0.2 Hz has been described (1345) using a motor-driven valve in the aspiration system. Reduced drift and higher precision were achieved at the expense of slower operation and increased sample consumption. A detection limit of 1 pg ml-' for B by AAS was obtained and the apparatus has been used for the determination of major elements in rock standards.An incidental advantage of the system is that aspiration of air between samples can be avoided and this is important for example in the determination of Pb in hydrocarbons. A comparison of the use of pneumatic and ultrasonic nebulizers for AES and AFS has been carried out (986). Despite the large range of nebulization efficiencies, from 3.6 - 68%, the increase in noise at high nebulization efficiences finaily resulted in similar detection limits for the two types for Ca, K, Na and Sr. 1.3.3.2 Microsampling Devices Discrete microsample techniques can usually be considered in two classes - (1) direct injection and (2) dipping capillary aspirators - and these have been discussed in general terms (C837, C1007). Their advantage over conventional nebulization was emphasized for solutions with a high solids content because salting up of the nebulizer and burner are largely obviated.It has also been found that for the direct injection method, used in the determination of Cd, Mg and Rh, the optimum solution volume increases with sample uptake rate (1353). A critical and comprehensive review of discrete microsampling systems for flames and plasmas using pneumatic and ultrasonic nebulizers discussed their advantages as well as applications to diverse sample types (2024).In a comparison of 4 microsampling systems for flames, viz discrete microsample injection, boat and cup techniques, and platinum loop methods, the last method was preferred because of its resistance to oxidants and because it offers control of sample-ashing temperatures (1 302).The discrete nebulization method has been used to follow the kinetics of the absorption of Pt and Ru on alumina being prepared as a catalyst (677). This application makes use of the rapidity of the technique which is assisted by a fast-response chart recorder. Alumina (10 g) was stirred into 50 ml of Pt or Au chloride solutions.To 100 pl aliquots of this slurry, 150 p1 La2S0, solution (2 g 1-') were added and 50 pl samples taken for analysis. Integration has been used with the discrete nebulization method in an air/C H flame for the determination of Cu in biological materials including bovine liver and oyster tissue CRMs (395). Integration was ingeniously triggered by the electrical conductivity of the introduced sample solution.The effects of various parameters were studied and the integrated signal was found to be proportional to analyte mass. The principal applications of discrete sample nebulization have been to the analysis of blood serum, exemplified by a thorough examination and successful application to the determination of Cu, Fe, Mg and Zn by FAAS (C883).The sample treatment for Cu, Fe, and Zn consisted of mixing a l00pl sample with 0.1 Yo Triton-X and 200 pl water. For Mg, a simple 501.11 sample injection into the Teflon cup was used. An RSD of 5% was obtained with a mixed standard solution (all 4 analytes). Results using either calibration curves or a standard addition procedure agreed closely with those obtained by deproteinization with TCA and HCl, with HNO 3 , or after digestion with HNO and H *O 2 .IAtomization and Excitation 29 Many other applications of similar discrete microsample nebulization methods to blood serum analysis have been made which differ mainly in the sample treatment prior to FAAS determination (C884, 1036, 1174, 1191). There appears to be a need for a rationalization, perhaps by interlaboratory comparative studies, of sample preparative techniques for serum analysis.It should be observed that these sample treatments are in general much simpler than many of the deproteinization or digestion techniques commonly used with conventional nebulization and this highlights one of the advantages of this microsample technique. In the AES determination of U in a N O/C H flame, a 3.5-fold enhancement of U emission at 591.5 nm was obtained in 20 mg ml-I Co solution using discrete nebulization (181).Discrete nebulization also avoids the health hazard of continuous nebulization of chlorinated hydrocarbons and this has been exploited in a comparison of solvent extraction using hexamethylene-ammonium-hexamethylenedithocarbamate (HMA - HMDTC) and APDC with MIBK or hydrocarbon solvents for the determination of Cd, Cu and Pb by FAAS (1946) (see also ARAAS, 1980,10, Ref. 91 1). For HMA - HMDTC and APDC, CC1, was the best solvent, MIBK the worst; APDC proved more efficient for Cd and HMA - HMDTC the more efficient for Cu and Pb. The number of reports on microsampling methods usingflow injection techniques (see also Section 2.4.2) have shown a considerable increase this year, and theory as well as practice have been advanced by a simple model (C817, C915, 1297, C2136) (see also Section 3.1.4). This postulated that the dispersion of a sample injected into a carrier stream occurred in a hypothetical, well-stirred, mixing cell whose volume can be found experimentally.The model permits the calculation of dispersion, in terms of injected volume, and also of absorbancekime curves.The use of the model is discussed in terms of the low, medium and high dispersion cases. The first two are analogous, respectively, to discrete nebulization and standard addition, while the last case provides concentration/time profiles using a real mixing cell, and offers a method of standardization. The discussion is illustrated by the determination by FAAS of Ca in iron ore, Cr in steel and Mg in water. The advantages of flow injection methods in AAS appear to be improved speed of analysis, relative freedom from physical (viscosity) interference effects and its ability, like discrete microsample nebulization, to operate with samples of high salt content without clogging.The ability to incorporate with ease, dilutions and additives, both for standardization and for application as releasing agents, has been noted (C836, 1835).Automatic liquid/liquid solvent extraction schemes can easily be incorporated with enhancements relative to the aqueous solution of 5 - 10 times (1 346, C2059). An example of this approach for trace elements by AAS in an air/C H , flame used MIBX or preferably butyl acetate as the carrier into which the aqueous sample was injected (164).A rate of 300 samples/hr with coefficients of variation of < 5% was claimed. Results for Fe in plant digests by FAAS using flow injection were in good agreement with Ar plasma emission spectrometry in the range 4.55 - 11.85 mg 1-' Fe (1467). Diluted chromatographic column eluates have been used with a flow injection procedure in an air/C, H, flame for the determination of Caand Mg in serum albumin (2210).Simultaneous determination of K and Na in a flame photometer and Ca and Mg in an AAS instrument has been reported using a carrier stream incorporating LiNO, and CsCl buffers, and RSD's of about 2Yo were obtained (262). The advantages of quality control of rapid analysis by a computer-managed flow injection system have been emphasized in a scheme which also incorporated two sample loops of different lengths to allow concentration range extension, the preferred result being selected under computer control (C1042).I .3.3.3 Sample Introduction by Volatilization The advantages of minimal sample pretreatment in the Delves Cup technique have again been demonstrated in the analysis of dried plant material introduced as a water slurry (1022).30 Analytical Atomic Spectroscopy Grass, cabbage and the SRMs, orchard and tomato leaves, were analysed for Pb using a standard addition procedure.With 20 pl samples, a precision of about 25vo was attained. Correction for non-specific background absorption was not required since it was time resolved from the analyte signals.The range of Pb contents determined was 0.072 - 2.4pg g-' dry material. Aplatinum filament has been used to volatilize Se deposited on it by controlled potential electrolysis (2141,2167). Electrothermal and electrothermal plus Ar/H , flame volatilization were studied for the determination of Se in complex matrices. A detection limit of 0.5 pg 1 - I was obtained.Volatilization by a ruby laser into an air/C,H, flame has been used in the AAS determination of Ag, Au and Ni for the measurement of layer thickness of electroplate (265). Single-shot laser volatilization gave approximately linear calibration over the ranges 1 - 16 pmol Ag, 0.5 - 8 pmol Au, and 1 - 12 pmol Ni with detection limits of 24, 32 and 73 nmol, respectively.The electrically heated chamber-in-flame method (ARAAS, 1980, 10, Refs. 224, 1001, 1327) has been studied (2052) and adapted for easier use with commercially available equipment (2079). A graphite cup and an air/C,H, flame were used for solution and solid samples. Sensitivities 4 - 10 times better than conventional flame AAS were obtained for the 15 elements studied.Sodium was determined at the ppm level in alumina with an RSD of 9 - 16%. The detection limit for Na was 0.3 pg g-l . Automatic control of a pre-determined temperature has been used to ensure that the absorption signal is in the linear region (1502, 1503). With this control, linear analytical curves were obtained for Cd, Co and Cu with the chamber-in-flame method.Novel methods of sample introduction have been described including one in which the sample was held against a rotating grindstone from which the solid particles were ejected into the gas or air feeds of a flame (1760). It was speculated that it could be used for flame spectroscopy or to feed ions from the flame into a mass spectrometer. A method using volatile metal chelates (Le., dithiocarbamates) formed and trapped on a reverse phase, Chromosorb W-DMCS glass column has been described (1585).The metal chelates were volatilized, by heating the column, into an Ar carrier stream for an AA or ICP measurement. Detection limits in the nanogram range for Cd, Co, Cu, Ni, Pb, Se and Zn were lower than for the conventional nebulization technique. Following sample solution introduction by conventional nebulization, several different methods of aerosolpreconcentration have been investigated.In one of these, the analyte in the aerosol was trapped on a water-cooled silica tube in the flame. The collected species were volatilized in the flame for AAS determination on removal of the water cooling to the tube. Characteristic concentrations obtained were Cd 0.002pg ml - I , Cu 0.002pg ml-', Pb 0.004pg ml - I , Se 0.1 1 pg ml -I and Zn 0.001 pg ml in an air/C, H , flame and As 0.073 pg ml - I in an air/propane flame.Enhancement of Cd and Cu by the addition of V which formed a vanadium oxide coating on the tube was observed (2140) (ARAAS, 1980, 10, Ref. 508 and 1979,9, Ref. 1464). In a second method, aerosol from a conventional nebulizer was dried in a heated chamber and collected on a chromel wire electrode at 14 - 18 kV, on the principle of the electrostatic dust collector (553).The wire was then heated and the volatilized material carried by the oxidant into the flame for measurement. Enhancements of I order of magnitude were obtained for Cd, Pb and Zn. In an attempt to overcome some of the difficulties of this method a water electrode at + 15 kV has been used to replace the chromel wire and this electrode solution nebulized for flame analysis by conventional methods (552).Of 40 elements tested some 36 were effectively preconcentrated and concentrations of Pb in lake water at least as low as 80 ppb could be determined.Atomization and Excitation 31 1.3.3.4 Direct Analysis of Fluid Suspensions A few interesting applications and studies of the use of slurries, suspensions and emulsions for analysis by direct nebulization have been made and perhaps the use of the Babington high- solids nebulizer goes some way to minimizing the practical difficulties of this approach (1471).This type of nebulizer, used in the discrete nebulization mode, has been successfully applied to the analysis by FAAS of liver and beef, sonically homogenized (9 g sample + 4.5 ml H,O), for Cu, Mg and Pb.Quantitative recovery of these elements was obtained when compared with conventional nebulization for sample solutions prepared by wet or dry ashing methods. As this nebulizer was constructed of stainless steel and nickel-plated brass, contamination problems seem inevitable and were admitted.In this connection the wear of this type of nebulizer, constructed of stainless steel or titanium, has been studied experimentally (C105 1). Factors affecting the transport of metal particles in lubricating oils have been investigated for nine nebulizer systems and four spray chambers used for flame, d.c. plasma or ICP spectrometers, by collecting particles of different size at the appropriate height (with the flame or plasma inoperative).For one of the best nebulizers tested, iron particles of 7 pm in di(2-ethylhexy1)azelate base were quantitatively transported but only 62% of 7 - 14 pm size were transported and much less at larger particle sizes (1594). A detailed consideration of the usefulness and experimental parameters of oil/water emulsions for AAS pointed to their advantages in avoiding the flame stoicheiometric problems which occur when organic solvents are used (910).This method was applied to Zn in ointments dissolved in benzene and emulsified, and a precision comparable to that for a standard complexometric procedure achieved. Calcium in lubricating oils has been determined, after emulsification, by FAES in a N,O/C,H, flame (539).The results agreed with those obtained by AAS and a sample dilution procedure. 1.3.3.5 Atomic Absorption1 Chromatographic Techniques These methods have now become so important that they can no longer be simply classed as AA spectrometric detectors for chromatographs and their role, particularly in the field of trace metal speciation, has attracted increasing attention.The technique of gas chromatography -atomic absorption spectrometry has been reviewed (C1332) and the inherent compatibility of GC with flame AAS pointed out. In addition, it has also been shown that with proper attention to maximizing the residence time in the flame, AAS can provide better sensitivity in these applications than can electrothermal methods. In this review tetra-alkyl lead compounds were used to exemplify the operation of the technique and detection limits of 17 pg were obtained for tetraethyl and tetramethyl compounds with a hydrogen diffusion flame from which the atomic species were swept into a flame-heated ceramic tube for determination.The more common combination of liquid chromatography coupled to FAAS has also been applied to tetra-alkyl lead compounds in petrol (1784).In this report acomparison with LC coupled to a U.V. detector showed that the interference from the diluent (xylene) which could occur was absent in the LC - AAS system. With LC - AAS five alkyl-lead compounds were separated and detected whereas only the tetramethyl lead compound was free of hydrocarbon interference with the LC/u.v. system.The detection limit for each of the alkyl- lead compounds in the LC - AAS system was about 10 ng Pb using a 20 pl sample. The most precise method, peak height or area, varied from species to species, but for all five, RSD’s between 4.2 - 6.1 Yo were achieved. A detailed comparison of various separation methods, GC or HPLC, and detection systems, N,O/C,H, flame, furnace pyrolysis, or hydride generation, has been presented for the determination of Sn tetra-alkyls and alkyl-Sn chlorides (1294).A continuous flow system for hydride generation following chromatographic separation was also described. Detection limits ranged from 1 to 20 pg.32 Analytical Atomic Spectroscopy A hybrid gel chromatography-FAAS system has been applied to the study of the binding of metal ions t o linear phosphate (C815).The chromatographic effluent flow, which was lower than the nebulizer uptake rate, was matched by using a 3-way connector in which the third limb aspirated water. In another example, Cd and Zn bound metallothioneins from liver and kidney were determined (1 194). Nebulizer parameters for on-line operation with liquid chromatographs have been investigated and it was found that a non-aspirating setting of the commercial nebulizers gave improved SNR, eliminated the need for post-column sample dilution and reduced gas bubble formation in the chromatograph (144). 1.3.4 Atomic Fluorescence Spectrometry (See also Sections 2.1.4.3 and 2.5.3) Interest in AFS is currently mainly concerned with multi-element systems, but again this year applications papers are few.The potential of AFS has been carefully considered and its characteristics in terms of excitation sources, atomizers, and detection limits reviewed (C1456, C826). The use of hollow-cathode excitation of sputtered elements, pulsed HCI, excitation with an ICP atomizer, and ICP excitation with flame atomization ere all discussed.It was concluded that the manufacture of dye-laser systems was unlikely to be commercially attractive but that multi-element AFS based on more conventional source5 should be able to provide viable systems. This hope has been partly fulfilled with the development of a commercial multi-clement AFS instrument using the ICP as an atomization cell (see Section 1.2.1.4) (C20).The components of a continuum source, multi-element AFS system have been evaluated for their noise contributions with particular reference to methods of modulation and to flame types (628). Of the Ar-shielded flames of air/C,H,, N 2 0 / C z H 2 , and N20/C3H, and an unshielded air/C, H, flame with a liquid fuel component studied, the two air/C, H flame5 were preferred to the N,O/C,H, flame.It was concluded that air/C,H2 should always be used instead of N,O/C, H , unless the atomization efficiency was inadequate. Amplitude modulation was superior for wavelengths below 350 nm and wavelength modulation above. Atomization sources described for continuum source AFS included a miniature (< 0.1 1 min - I ) N , O/H , laminar diffusion flame (629) and an air/C, H or N O/C H flame (582).Both reports made use of an electrothermally heated graphite cup (629) or filament (582) in the flame (see also Section 1.3.3.3). Developments in the techniques of AFS include a novel method of double modulation (see R.C. Elser and J. D. Winefordner, Anal. Chem., 1974,44,698), using a rotating quartz chopper for wavelength modulation and mechanical source modulation (ClO7O).An unuwal AFS instrument makes use of a conventional HCL driven repetitively by an r.f. pulse superimposed on a high current pulse and a stroboscopic (time-resolved) method of signal detection (C828). 1.3.5 Applications of Lasers Although both fundamental and analytical studies of interest to atomic spectroscopists still continue at a high level, no particular development involving lasers appears to offer the breakthrough which would bring about the introduction of lasers to the working analytical laboratory.To many analytical chemists, the papers in this field will therefore appear esoteric and of little interest, which is a marked contrast to the impact of lasers in other fields such as Raman spectroscopy. A useful review by Winefordner (919) compared atomic and molecular spectrometric methods involving lasers and discussed the reasons for the current lack of acceptance of lasers in analytical atomic spectrometry.Perhaps the most important of these is the lack of a real improvement in detection power compared to existing widely used techniques. The potential for improvements in detection limits is made abundantly dear in anAtomization and Excitation 33 excellent review of single-atom detection by Alkemade (982).The review classifies all the various laser-assisted detection techniques potentially useful for single atom measurement, including those involving optical, and electrical signal recovery. His conclusion is that these methods are unlikely to achieve a breakthrough in normal routine analysis but work in this field presents a “thrilling challenge” to atomic spectroscopists. Another relevant article, which represented a personal view of developments in resonance ionization spectroscopy, gave Hurst (2208) the opportunity to describe some recent experiments carried out at the Oak Ridge Laboratory.Although analytical possibilities are mentioned, the emphasis of his article is on fundamental measurements of interest in classical physics. 1.3.5. I Atomic Fluorescence Spectrometry Interesting developments in the measurement of laser excited AF from electrothermal atomizers have appeared this year and have given extremely low detection limits for Pb determinations. Bol’shov et al. (1543) generated a saturated P b vapour in an evacuated quartz cell and could detect as little as 250 atoms cm -3 using a Nd +3 :YAG laser oscillating at 283.3 nm, with the fluorescence measured at 405.8 nm.With a graphite cup atomizer and aqueous solutions, a detection limit of 0.05 pg ml-l was achieved. Detection limits in the lo-’ 5g range have also been reported for P b using a saturated double-resonance emission method (1521).Two lasers, one frequency doubled and operated at 283.3 nm and the other at 600.2 nm, were used to produce saturated atom populations at very high energy levels. Fluorescence emission lines at wavelengths as low as 202.2 nm could then be observed associated with background noise levels reduced by several orders of magnitude compared to longer wavelengths. A commercial electrothermal atomizer was used with the fluorescence emission observed through a sapphire window installed in the side of the atomizer. Falk et al.(C832) discussed the limitations of laser excited flame atomic fluorescence spectrometry and showed that theoretical predictions, based on a flame background limited system, agreed with experimental observations in the visible range (greater than 400 nm).At shorter wavelengths experimental detection limits became substantially higher than predicted by the theory. The effect of background correction, using for example harmonic saturated spectroscopy (244; see also ARAAS, 1980, 10, Ref. 1593) on these theoreticaVpractica1 correlations would be interesting. Horvath et al. (558) have further improved the performance of their laser AFS system described previously (ARAAS, 1978, 8, Ref. 904).Specifically, the present paper compares the performance of a pneumatic nebulizer/capillary burner system with a laboratory built ultrasonic nebulizer/mini burner system. Detection limits for Ca and Sr were similar at 0.01 and 0.1 ng ml - I , respectively, but the latter system gave a five-fold improvement for Mg and Pb (0.002 and 0.2 ng ml-I , respectively) using an O,/C,H,/Ar flame.The resonance lines for Ca and Sr were optically saturated, but those for Mg and Pb were not. In a conference paper, Kosinski et al. (C1045), described some of the advantages of laser AFS using inductively-coupled plasma atomization, which contrasts with the difficulties reported last year (see ARAAS, 1980, 10, Ref. 479). No results are given in the conference abstract but the dependence of detection limit on source intensity is strongly emphasized. Biancifiori et al. (C849) have reported the measurement of laser AFS signals from an atomic vapour produced by cathodic sputtering in a glow discharge chamber. A detection limit of 1 - 5 pg g-I was reported for Na in copper metal.The saturation behaviour of atomic transitions using laser radiation continues to receive much interest, many observations being characterized by the resulting fluorescence signals. Omenetto et a/. (1208) have studied the broadening of atomic absorption lines in H -based flames diluted with Ar and N , , and found that the half-width of the atomic line is proportional to the square root of the logarithm of the laser irradiance (for Ca, In, Na and34 Analytical A fornic Spectroscopy Sr).Further studies of the depletion of excited-state atom populations via chemical reactions with flame species have been reported (C891) (see also ARAAS, 1980, 10, Ref. 1302). The results from a pulsed dye-laser with a short pulse duration were compared with those from a CW laser, and the measurements allowed the determination of a rate constant for the formation of sodium hydroxide molecules of lo7 s--'.Omenetto (C827) has defined a parameter called the saturation irradiance, which is the irradiance producing a steady-state value of the excited-state population which is 50% of the steady-state saturation value, and has shown that this is of significantly different magnitude for 2- and 3-level systems, due to collisional energy exchange with the third level. Zizak et al.(C824,979) have also considered the effects of collisional processes on the populations of energy levels in the region of the saturated excited state level. Levels substantially above the radiatively excited level (thermally assisted lines) follow a Boltzmann distribution, in which the Boltzmann temperature is equal to the temperature of the flame system and can be used for local temperature measurements (979).Levels below the radiatively excited level show a large deviation from the Boltzmann distribution, whilst energy levels close to but above this level are overpopulated with respect to the average population distribution and can be considered in equilibrium with the laser excited level.The use of laser saturated fluorescence measurements for spatialdiagnostic purposesfor flames (C70, 1520) and the inductively-coupled plasma (C1047) has again been reported. Measurements of atoms, ions and molecular species have been described. Besides the above, several other papers have discussed the measurement of flame temperatures based on laser- induced fluorescence. A method based on the molecular fluorescence of OH radicals has again been reported (2178).Alessandretti et al. (2009) have compared the laser AFS method with the spontaneous Raman scattering method in both air/C,H, and air/C,H, flames. Winefordner (1819) has now produced an extensive publication on the five laser AF methods (see ARAAS, 1980, 10, Ref. 1594). Oiher references of interest- Absolute measurements of impurity species in fusion research plasmas by laser AFS: C1049. Extended model for saturation in a 2-level system: 2180. I .3.5.2 Laser-enhanced Ionization (LEI) The principal papers published on this topic in 1981 consist of the full publication of conference papers reviewed in ARAAS, 1980, 10, Section 1.3.5.3.Havrilla and Green (159) have described their work on the use of plate electrodes for laser-enhanced ionization and discuss the importance of measurement parameters on LEI signal recovery. Optimization of sampling height, electrode separation, electrode-flame separation, relative excitation height, fuel - oxidant ratio, and applied voltage all appear to be important.Although a signal from 0.1 ppm In can still be observed in 300 ppm Na in an air/C,H, flame, interference (i.e., change in response) on In occurs at much lower Na concentrations. Subsequently, the same authors (21 88) described an electrode-positioning apparatus and point out that accurate positioning is important to achieve optimum signal recovery. Information on acid effects on the LEI signal of In has now been published (61 8; see also ARAAS, 1980, 10, Ref.C567). These studies showed that neither In nor Ni are plated onto the surface of the cathodes during LEI measurements, and that the In signal, which increased on addition of HNO, due to retention of In in aqueous solutions inside the burner head, was constant for HNO, concentrations from to 1 ~ . The characterization of electrical interferences in LEI has been achieved by comparison of measurements in air/C H and air/H flames (1237).However, the most interesting paper on the subject of interferences in LEI appears to be that on the reduction of matrix effects in an air/C,H2 flame by use of aAtomization and Excitation 35 water-cooled stainless-steel-tube cathode positioned in the flame and just above the primary reaction zone (2029; see ARAAS, 1980, 10, Ref.C1443). No interference on 50 pg ml-' Fe was observed at concentrations up to 300pg ml Na and only a 25% enhancement was found at 3000 pg ml-l Na.. These new working conditions have allowed the determination of traces of Rb in NBS CRM 1575 pine needles (C73). Two-step laser-enhanced ionization of Na in a O,/H,/Ar flame has been reported to give an enhancement factor of 144 compared to single laser - LEI (2033; see ARAAS, 1980, 10, Ref.C1439). A two-step laser process has also been used to obtain sensitive LEI signals for In (1505). A coumarin 47 dye-laser operated at 45 1 . 1 nm and a rhodamine 6G dye-laser operated at 572.8 nm were used for these experiments; measurements of in in lead, tin, and tellurium alloys were also reported.Other reference of interest- Neutral atom depletion by CW laser LEI: C72. I .3.5.3 Laser-induced Impedance Changes in Hollow-cathode Lamps This new technique, which is receiving increased interest, makes use of the optogalvanic effect but with a different experimental configuration to that used for LEI. Several names have been used to describe the process; optogalvanic resonance detection, (C 1 152, 1939); laser-induced impedance change, (LIIC) (2201); intracavity atomic absorption (C15, 1999). These terms individually describe only part of the experimental measurement. A pulsed dye- laser tuned to the absorption transition of the analyte atom (of the element to be determined) is directed through the analyte atomic vapour produced in a flame and on into a commercial hollow-cathode lamp containing the element of interest, which is used as a resonance detector.An optogalvanic signal occurs in the HCL, from a laser-induced impedance change in the HCL discharge, and this is a measure of the incident (i.e., transmitted) radiation from the laser. Thus, as with LEI, no monochromator is required.With a 10 cm air/C,H? flame mounted between the laser and the HCL, detection limits of 5,0.8 and 0.05 pg ml-I have been reported for Cu, Li and Na, respectively (C1152, 1939). Similar measurements have been made with an 0, /H flame positioned inside the cavity of a CW dye-laser (intracavityatomic absorption) and a detection limit for Na below 1 pg I-' was achieved with a flame length of about 1 cm (C15, 1999). Zalewski et a/. (1999) showed that an enhancement of about 4000 is achieved by placing the flame inside rather than outside the laser cavity, and these authors have also investigated the possibility of using non-laser sources, such as a HCI or EDL to induce optogalvanic signals in an HCL (C 15). Intracavity dye-laser measurements of atomic oxygen in a discharge flow tube (1981), and C l in an 0 , / C 2 H 2 flame (1980) have been reported during the past year, but in these cases detection was achieved by normal spectrometric methods. Beenan et a/. (2201) have shown that LIIC signals in HCL's can be used to obtain wavelength calibration of tunable dye-lasers. They have shown that errors as large as 0.0055 nm may result from the distortions in the signal at high operating powers and recommend that sample irradiance should not exceed 10 W cm- 2 . 1.3.5.4 Other Studies Apart from the reviews mentioned above (982, 2208), only one paper has appeared on resonance ionization spectroscopy. Mercury vapour in a vacuum chamber was excited into ionization by three 3 12.8 nm photons via a two photon resonant step, 6 I S , - 7 I S o ( 1 562). Detection of both Hg ions and 253.7 nm Hg emission was reported, with the former being more sensitive. The new class of atomic resonance line lasers based on U . V . photodissociation of metal halides, described in ARAAS last year (ARAAS, 1980, 10,56), were reported (147) to be suitable for Al, Cs, Ga, In, K, Na, Rb and TI, to require no wavelength tuning and to36 Analytical A tomic Spectroscopy provide typical bandwidths of 0.002 - 0.008 nm. Other references of interest- Accurate wavelength tuning of an etalon tuned dye-laser: C 107 1. Laser Fraunhofer diffraction studies of aerosol droplet size distributions: 2 149. (see also Section 1.3.1) New cross-correlation techniques for measuring atomic excited state lifetimes: C694 (see also ARAAS, 1980, 10, Ref. C1440). Power broadening of Na-D lines in a flame irradiated by a pulsed tunable dye- laser: 1719.

ISSN:0306-1353    

DOI:10.1039/AA9811100024

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

36 Analytical A tomic Spectroscopy 1.4 ELECTROTHERMAL ATOMIZATION In the year under review there have been no reports of truly new developments in the design of electrothermal atomizers. Historians of the subject will, no doubt, find it interesting to note how developments have turned full circle. After the pioneering work by L’vov and Woodriff on isothermal atomization this appears, after over a decade of proliferating reports of interference phenomena associated with pulse heated atomizers, to be the optimum mode of atomization for both absorption and emission measurements.One can only speculate as to why these predictions have taken so long to be acknowledged. The importance of more efficient background correction and signal handling are also being stressed.Some workers have investigated various magneto-optical systems for background correction and these show promise for practical analysis. Some interesting reports, attributing previously documented interference effects to inefficient signal response, have added more impetus to the optimization of handling ETA data. 1.4.1 Atomizer Design Several papers which deal with isothermal and quasi-thermal atomization have appeared.Manning and Slavin (C88, C515, 1323), have focused attention on the advantages of using graphite probes for sample introduction into a pre-heated furnace. Improved sensitivity and a freedom from matrix interferences in the atomization of Al were recorded. Rcheulishuili (2048) used a separately heated graphite sample holder from which the sample was volatilized into a hot furnace.The advantages of the quasi-isothermal sampling techniques such as platforms and cups continue to be reported (380, C2135). The obvious limitations of these techniques, in their inherently poor sensitivity for the more refractory elements (C88), surely indicate that these hybrid devices are stop-gaps, albeit useful ones. Atsuya and Itoh ((313, C1459) used a miniature cup, placed within the graphite furnace and heated by conduction, to volatilize solid samples.These workers also noted improved sensitivity for the analysis of solutions. A novel approach was adopted by Holcombe and co-workers (C800, C1061) who adapted a commercial pulse heated atomizer by the incorporation of a condensation site, a re- inforced graphite platform, maintained at a temperature several hundred degrees below the furnace wall.I t was claimed that atomic species could be condensed onto this site while volatile matrix species were preferentially removed. A second atomization was then required for the matrix free analysis. More rigorous tests are required before this imaginative idea proves its worth. The advantages of the rapid heating used in current commercial atomizers was critically reviewed by Riandey and Pinta (C791) who concluded that for refractory elements, such a mode of atomization was still not ideal although it did represent considerable technical progress.Chakrabarti and his colleagues (C501, C1419, C1457)continue to expound the virtues of their capacitively-heated atomizer. The freedom from matrix interferences was said to be mainly attributable to the rapid heating rates (up to 100 K ms- I ) rather than to the isothermalAtomization and Excitation 37 atomization.Such a conclusion may cause controversy in some quarters. The technique was applied to a variety of trace element determinations (133, C790) in NBS reference materials where recoveries, compared to certified values, varied from 98 - 105%.This work is still in its infancy and it may.be some time before its value is realized. Several investigations dealing with techniques for carbon tube coating have been described. Lythgoe (91 1) compared four techniques for tantalum coating; foil lining, pyrolysis from solution, evaporation of the metal and cathodic sputtering.Cathodic sputtering was preferred and resulted in a 3-fold enhancement for the atomization of Si. I t was also the most reproducible and gave the longest tube lifetimes (300 cycles). A soaking technique was used by Takahishi (625) to treat graphite tubes with tantalum (see also ARAAS, 1978, 8, Ref. 950). The tubes were soaked in a 5% m/VTaF, solution under reduced pressure, then baked at 2400 “C.Electron microscope examination showed that the tubes were not coated with TaC but that the small voids in the surface were filled with TaC. Patents have been filed by the Philips company (1691, 1975, 1976) concerning techniques for the production of cuvettes from pyrolytic graphite. A three-stage process was described which included deposition, machining and a second deposition, to fill surface imperfections.A novel atom cell was described by Parker and LaBrecque (C792). The device was operated under high vacuum (10- - 10- ’ Torr) and the sample crucible was heated by bombardment with thermal electrons accelerated by a potential difference of up to 5000 V d.c. at 500 mA. The crucible was positive with respect to the tungsten wire surrounding it.Up to 100 mg samples were charged at one cycle prior to evacuation. The device has not been used for AA measurements and as such is at an early development stage. Torsi et al. (1601) used a custom built carbon furnace to collect atmospheric particulates by electrostatic precipitation. The furnace tube was charged at 1.5 kV and calibration was achieved by normal liquid sampling, assuming 100% collection efficiency.A miniature graphite furnace was used in a novel Zeeman configuration (146). The modulated magnetic field was applied longitudinally and improved background correction was combined with a more stable base line and higher light throughput. The disadvantages were the increased calibration curvature and poorer sensitivity compared to conventional non-Zeeman systems. An atomizer with controlled atmosphere and temperature was described by Cedergren et al.(1028; see ARAAS, 1980,10, Ref. C1170). This was claimed to offer possibilities for controlling and studying interference effects. Examples were presented for the determination of Pb in chloride and sulphate solutions. Good agreement was obtained between experimental results and high temperature equilibrium calculations over large variations in the composition of the gas phase. 1.4.2 Fundamental Studies The use of theoretical models forms an intrinsic part in the investigation of atomization processes occurring in any particular system. Rubeska (C789) reviewed the current state of the art regarding our understanding of these processes. He concluded that the activation energy values obtained greatly depend on the experimental conditions employed and hence, any variations in the values may result in different conclusions regarding the actual mechanistic pathway.The kinetics for atomization in a constant temperature furnace, reported by Fuller (ARAAS, 1974,4,14), was used by Chakrabarti and co-workers (1412; see also ARAAS, 1979, 9, Ref. 1997) to describe the mechanism of atomization in a graphite furnace heated by capacitive discharge. Activation energy values for Cd and Pb were used to develop the model capable of predicting the shape of the absorption/time curves in this particular system. A model has been developed by Tessari (C1429) to describe the “reversible” atomization of metals at a graphite platform.Slavin et al. (C1423) have shown, however, that with such a system small deviations from the steady state may introduce large analytical errors. L’vov and Ryabchuk (2051) made a critical study of atomization38 Analytical A tomic Spectroscopy mechanisms from an evaluation of activation energies. The thermostability of the oxides was said to govern the mechanistic pathway employed.The same authors have continued their work on the atomization of gaseous mono-cyanides (1 187, see also ARAAS, 1979,9,26, and 1980, 10, 59). In an investigation into the atomization of Al (1500), they observed AlCN molecules in a N atmosphere, while in Ar, Al C spectra were recorded. An intense doublet at 193.2 and 193.3 nm could, it was claimed, be used to determine major amounts (> 10- * g) of Al.Models based on high temperature equilibrium calculations (ARAAS, 1980, 10, 59) offer unique possibilities for the understanding of the chemical reactions involved in ETA. The main exponents of the technique, Frech and co-workers (C1081), have shown how optimum conditions for AA measurements can be established from a knowledge of the most important parameters governing the formation of free atoms.As a further example, the reactions involved in the atomization of As (C1422,2166) were described. Surface reactions and surface structure were shown to be important and optimum conditions for the atomization of As were obtained, with the conclusion that the L’vov platform and a stabilizing agent, such as nickel, were required. A parallel study by Kunc and Balkis ( I 199) reported band spectra of As0 and perhaps As *.Sturgeon and Berman (21 5 1) have confirmed their earlier observations on analyte ionization in ETA (ARAAS, 1980, 10,59), concluding that it had a negligible effect on AA signals. Ionization of metal vapours, it was stated, occurred by thermal processes. The microwave attenuation techniques used by these authors have been used again to determine the absolute atomization efficiency of the graphite furnace (C 1062, C1430).The efficiency of the furnace varied with the element under examination, but an average value of lo%, with a range from 1070 (Mo, Sr) to 25% (Na), was reported. Studies of surface reactions have gained some prominence, and a review by Katskov (208) presented the Russian contribution to this work.Styris and Kaye have combined AA and MS techniques to study analyte furnace interactions of Rb in a tantalum furnace (1523) and Ag, Rb and V in a furnace constructed of vitreous carbon (C2063). In the former study, the surface reactions resulted in the formation of the oxide of the analyte, while at higher temperatures, free Rb atoms (which had been chemisorbed onto the tantalum oxide surface) were desorbed as the surface oxide evaporated.The latter study postulated varying oxide states for V as precursors to atomization which occurred ultimately by surface reduction of metal oxide. The analytical implications of chemisorbed oxygen on the surface of graphite atomizers were presented by Holcombe and co-workers (C60, 569). The crystal structure of the graphite surface was said to account for differing activities in graphite and the chemisorption of oxygen onto active sites was capable of increasing the appearance time of Pb by as much as 300 K.The chemistry of the atomizer surface was also investigated by Vickrey et al. (C23, ClO58). Surface treatment with zirconium oxide was observed to affect the rate of atomization from the surface.These metal oxide coatings were said to exhibit a marked surface acidity and the authors, using nitric acid solution, showed that the ash temperature range for elements varied with acidity from 0.1 - 4% in nitric acid. Wahab and Chakrabarti (141 1, 1413) described the factors affecting the atomization of Y from metallic and metal-carbide surfaces.The atom formation process was primarily a thermal decomposition of Y * 0 3 ( s ) above 1800 “C. Better sensitivity for Y wasachieved however using tantalum or tungsten rather than carbon surfaces, due to the absence of side reactions giving rise to the formation of carbides. Additional references on the preceding topic - 213, 214. The transient nature of the processes occurring in ETA makes kinetic studies a useful route for gaining mechanistic information.Many worthwhile studies, on the temporal reactions within electrothermal atomizers are still appearing. L’vov et al. ( 1 5 1 1, 2047) have developed a macrokinetic theory of vaporization based on consideration of the distributionAtomization and Excitation 39 of a sample between the surface and subsurface phases of the atomizer.Two different regions of evaporation were described, one whereby evaporation was controlled by the evaporation rate constant and a second where it was controlled by the diffusion coefficient. This data was used to explain breaks in the Arrhenius curves and to find a connection between activation energies of evaporation in these regions.Clarification of surface effects was also possible. Evaporation rate constants were investigated by Katskov (1398, 1496, 1498) to describe the evaporation of a wide group of elements from a graphite surface. Formation enthalpies were calculated from the falling portion of the absorption/time curve and conclusions as to the route of atomization were drawn. Diffusion coefficients for 3 1 metals in an Ar atmosphere were also calculated. The reaction kinetics involved in the atomization of Si were studied by Muller-Vogt and Wend1 (21 53).The measurements were carried out for conventional graphite tubes and niobium-coated graphite tubes and enhancements in Si sensitivity on the niobium/graphite surface were explained by an increase in the reaction rate of reduction and a decrease in the rate of formation of silicon carbide.Most kinetic calculations are based on studies of absorption/time curves and both Frech et al. (C1063, C1428) and Wassall and Johnson (C19) have explained how the slopes of these curves can be influenced by such factors as sample constituents, heating characteristics, sheath gas, thermal pretreatment and instrument response time.The importance of controlled heating rates and their influence on precision was stressed by two sets of workers. Dymott and Whiteside (C65) explained how such control yvas necessary to improve sensitivity, while Routh et al. (C786, C1064) compared heating rates in two tube designs and discussed the effects of gas flow and matrix on peak shape. Absorption spectra of diatomic molecules: C802 Kinetics of atom formation: C785, 153 1.Spatial and temporal distributions of atoms: C59, 210, C787, C825, C1426. Other references of interest- 1.4.3 Interferences Without doubt, the trend towards a more precisely controlled and isothermal heating environment continues to be the most important contribution towards alleviation of the interference phenomena which have plagued ETA.Slavin (C1117) presented the requirements for interference-free furnace analyses. He discussed the improvements in controlled heating, platform atomization, graphite substrates and background correction, and concluded that the limitations on matrix concentrations were now governed by the efficiency of background correction rather than physio-chemical parameters. Platform atomization continues to be an area of growth and this is reflected in the number of papers dealing with the technique.Koirtyohann er al. (C774) and also the Chakrabarti group (C776, C1420) have produced parallel studies using fast oscilloscope monitoring to assess true peak shapes of atomization from a platform. The conclusions of both teams of workers were that the atomization of relatively volatile elements occurred at higher wall temperatures relative to the platform with a time delay in temperature of approximately 0.8 s between wall and platform.Atomization occurred therefore when the system was close to steady state (isothermal) conditions, resulting in a reduction in interference effects. An interesting commercial system employing a platform microboat and aerosol deposition was described (C64, C795, C1083).Impressive interference reduction, attributable to platform atomization, was complemented by improved drying due to the deposition of microdroplets from the aerosol onto a preheated platform, thus reducing spreading and soaking effects on the graphite. Background absorption from 1070 NaCl on the Pb 217.0 nm resonance line was reduced and negligible chemical interferences up to 10% NaCl were observed.The possibility of aerosol deposition into a furnace at higher temperature has not yet been reported although40 Analytical Atomic Spectroscopy this appears as another possible form of isothermal atomization. Chakrabarti et al. (1559) continue to extol the virtues of their capacitively-heated atomizer for the reduction of interferences.They suggested that the rapid heating rates allowed analyte atoms to exist in a hotter environment with greater dissociation of metal halides. This, they claimed obviated the need for background correction. Examples of the analyses of CRM's were presented. The matrix levels used however (< 1% NaCI) are less than would often be experienced in real analyses. Interference effects attributable to atomizer substrates are not well understood. Slavin and Manning (C61, C797, 1730) have reported how many such interferences were caused by intercalation of the analyte within the graphite structure and have investigated the effects of different substrates.There was clear evidence to suggest that pyrolytic graphite was the most efficient coating, but that the efficiency of interference reduction deteriorated as the protective layer was depleted.Surface interference effects which occurred during the drying stage were reported by Dabeka (C1065, C1443; see also ARAAS, 1980, 10, 61) to be detrimental to analytical precision and were affected by the condition of the graphite surface.Slovak and Docekal (1311) prevented sample diffusion into the graphite during the pretreatment stages, by the use of a hydrophobic film of polystyrene added to the tube. Reports of the multifarious application of matrix modifiers still proliferate in the literature. Routh and McKenzie (C58) suggested a more mechanistic approach to the choice of matrix modifier, which included simultaneous monitoring of the analyte and backgroundhime signals with a view to interpretation of the mechanism of interference reduction.Guevremont and Jamieson (C1060, C1415) produced an interesting investigation into the use of matrix modifiers for the determination of trace metals in sea-water. Using rapid mass spectrometric scans of the pyrolysis products formed in the furnace, data pertaining to the determination of Cd and Zn was obtained.The warning sounded last year (ARAAS, 1980, 10, 57), however, on the ad hoc use of matrix modifiers does not seem to have been generally heeded. The most important reports this year include the use of thiourea (526) to reduce alkali chloride interferences in a molybdenum microtube. Kirkbright eta/. (1) evaluated a variety of compounds for stabilizing Hg and Se.The authors used K , Cr, 0, , with various combinations of AgNO,, KMnO,, Na, S, Cu and Ni, and found at1 to be effective to some degree. However 0.05% K,Cr,O, in 1% HNO, was the recommended reagent for 5tabilizing Hg up to 250 "C and Se to 1200 "C. The fact that most metals could stabilize inorganic and organically bound Te was reported by Langmyhr and co-workers (1026).They recommended Pd and Pt for stabilization up to 1050 "C. The enhancement of Be by the addition of excess calcium at 10- - 10- mole 1- ' has been reported (1384) with the recommendation that H, be added to the sheath gas. Ashing of biological samples in an 0, atmosphere was studied by Holcombe and Eaton (C1082, see also ARAAS, 1974,4, 13 and 1980, 10, Ref.C101). The appearance temperature of Pb was increased and the efficient destruction of the organic matrix was achieved. Ashing at 900 - lo00 "C was possible without loss of Pb. Detrimental effects of ashing in 0, on Cr and Sn however were reported. Several interesting background interferences have come to light. Vajda (1025) listed 81 known line overlaps and recommended the appropriate band widths for background correction.Saeed and Thomassen (2165) recorded spectral interferences from P absorption following atomization of phosphate solutions of As, Sb and Se. Background spectra of the more common organic compounds used in ETA were recorded by Betz et a/. (1 79). A novel multiple wavelength spectrometer was described by Skogerboe et a/. (C63).The device offered the possibility of observing both atomic and molecular events within an electrothermal atomizer. Interference effects may of course be a result of the electronic design of any particular system. The necessity for small time constants in spectrometer electronics was thus stressedAtomization and Excitation 41 by Lundberg and Frech (C853, 1726). A spectrometer system with a time constant of 10 ms was used, together with an on-line data acquisition system.The effects of 0.1'70 NaCl on Pb, which were depressive at a time constant of 270 ms, were enhancing at 10 ms, the peak absorbance occurring at an earlier time. This study was said to invalidate many previously published interference data. Other references of interest- Interference of HClO, : 623. 1.4.4 Developments in Methodology I . 4.4.1 Sample Introduction There have been a considerable number of papers dealing with different aspects of sample introduction in ETA. These can be sub-divided into solid and liquid sampling techniques. An informative study by Fuller and colleagues (1293) compared flame, graphite furnace and ICP for the direct analysis of gelled, stable slurries.Their results indicated that in nebulizer based systems, the relative atomization efficiency was governed by the transport efficiency in the nebulizer, the sample particle size and the atomization temperature. It was found necessary to mill samples to < 10 pm and to match standards both physically and chemically for practical analysis. In ETA however, the particle size only became significant at > 25 pm when sampling error predominated.I t was possible to use aqueous standards with peak height measurement without any loss of accuracy. Slurries of A1 0 were analysed for Ca, Cr, Cu, Fe, Na and Pb by Slovak and Docekal (1848), who employed continuous agitation to maintain sample suspension. Resin suspensions were introduced directly into an atomizer by Isozaki et al.for the determination of Cu in sea-water cencentrate (1665). Shabushnig and Hieftje (375, C 1075) constructed a microdroplet sample applicator for the introduction of liquids into a unheated or heated atomizer. The device comprised a piezoelectric crystal driven glass stylus by which sample solution was drawn from a reservoir by capillary action. An electronic gate allowed the exact number of droplets to be dispensed, and hence resulted in the possibility of single-solution calibration. Good sampling precision (< 1.5%) was obtained and very small sample volumes were required (< 1 .O ml).An aerosol deposition technique was employed by Royal (1481) for the determination of Au. The device lent itself to automation for process control work.A modification of a commercial autosampler was described (385) whereby flowing sample streams could pass through the plastic cups to facilitate sampling. A critical study of solid sampling techniques was undertaken by Frech et a/. (C777, C1084, 2124). They highlighted the limitations, pointing out that aqueous standards often give erroneous results even under isothermal conditions.This was explained by the fact that matrix effects were often related to the solubility of the analyte in the solid matrix. Two devices for simplifying the introduction of solid samples into tube atomizers were described by Takuji and Yamaguchi (C895). Firstly an aluminium cone was inserted into the injection hole of the furnace and weighed resins were thus delivered into the cuvette.About 0.3 mg of sample was used and RSD's of about 10% were obtained. Secondly, a graphite boat, loaded with resin, was introduced via the side entrance furnace. Using Cr as the test element, improved RSD's to about 3% were obtained. Three tools have been described (C66) and patented (1689) for the introduction of solid and semi-solid samples into a graphite furnace, without moving the furnace from the optical axis.Each tool fitted loosely into the sampling orifice. One tool employed a glass tube which was weighed before and after charging the cuvette. A second handled sheet samples by punching them and a third was a graphite rod with a spoon shaped depression for loading granular samples. The devices all allowed for partial mechanization.A vacuum actuated device for delivering solid polymer samples into a42 A naly t ical A tom ic Spectroscopy furnace was described by Danchev et al. (475). Samples weighing 0.02 - 20 mg could be delivered in about 3 s. 1.4.4.2 Magneto-optics Practical applications of ETA have been limited by physico-chemical interferences and background interferences. Although the former becomes less serious due to progress in more efficient heating, the latter still poses one of the most serious limitations in practical ETA.Towards this end, many magneto-optical systems for use in ETA have been reported, (see also Section 2.1) Fernandez et al. (C812, C997, 1326) have evaluated one such magneto- optical effect, the Zeeman effect, for the correction of high background levels in ETA.The system was used in the determination of Se in NBS superalloys (C1087), an application where Co and Fe lines fall within the bandpass of the 196.0 nm Se line. No spectral overlap was observed using Zeeman background correction, whereas conventional deuterium arc background correction overcompensated resulting in negative absorption signals. Kitagawa et al.(565, 572, 595) described the advantages of the coherent forward scattering technique (Faraday effect) and its applicability to ETA. The insensitivity to matrix background scattering, the authors explained, made it a suitable technique for practical analysis. They applied this to thedetermination of Pb in blood, CRM's and volcano ash (595, C823, C1462) and the results showed good agreement with certified values.The application of coherent forward scattering to multi-element analysis using ETA was investigated by Murayama (C821) and also by Masura (150). Using a xenon arc source, the intensity of the forward scattered light was found to be proportional to the square of the number of scattering atoms. Detection limits in the former paper were poorer than by conventional AA (C82 1 ).Additional references on preceding topic- C822. 1.4.4.3 Chromatographic Systems The sensitivity of ETA as a speciation detector for chromatography is now being fully exploited. Tittarelli and Mascherpa (1 761) have separated organophosphorus compounds using reversed phase HPLC. Detection limits of 0.3 mg I - were obtained and the method was applied to P speciation in lubricating oils.Vickrey et al. (4) reported enhanced sensitivity for organo-lead compounds using a zirconium coated tube. The organo-lead compounds were separated by HPLC and the addition of I to the system facilitated the use of aqueous standards for calibration. Various GC techniques were employed by De Jonge et al. ( 1 301) to separate organo-lead compounds. Air was passed through an OV101 column at - 130 "C to trap out the organo-lead compounds. Atmospheric particulates were collected on an impinger in ethylene glycol and heat space techniques were used to introduce the volatile compounds into the chromatograph. Waters were extracted with hexane and gasoline was introduced into the GC without dilution. The eluate from the GC was bled into the inner gas stream of the graphite furnace, maintained at 2000 "C (ARAAS, 1980, 10, Ref. 486). 1.4.4.4 Novel Applications Ottaway et al. (C1131) described the construction and operation of an automatic prompt response monitor for the determination of metals in atmospheric particulates. The instrument was based on ETA-AAS and samples were collected by impaction on the tube wall. Detection limits for Be, 2.1 ng m- 3 , Cd, 5 . 5 ng m- 3 , Mn, 5.d ng m- j , and Pb, 36ng m - were significantly below the threshold values for these elements. A novel detemination of Ni in steels by non-dispersive AFS using ETA was described (604). A detection limit of 10 ng using 10 p1 samples was obtained. iodide was determined indirectly by Nomura andAtomization and Excitation 43 Karasawa (1029) by adding Hg(N0 3 ) 2 to the sample solution and determining Hg. Two Hg signals, one from Hg(NO,), and a second from HgI, were sufficiently time resolved to determine I - in the range 0.13 - 6.4 ppm. Interferences from CN- , S - , SCN- and S , 0, - , were eliminated by adding NaOH and H 0, , and interferences from metals were removed by extraction with 8-quinolinol. Other references of interest: Application of the spectroscopy of diatomic molecules: 275.

ISSN:0306-1353    

DOI:10.1039/AA9811100036

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

Atomization and Excitation 43 1.5 VAPOUR GENERATION Judged by the number of publications appearing in the past year, interest in all the methods of vapour generation has continued at a very high level. This Section of ARAAS deals with the more important developments in instrumentation, while Sections 3.1.1.3 and 3.1.1.4 cover new methods of sample preparation, and also reduction procedures. Inevitably some overlap will be apparent.A new method involving the application of the generation of volatile fluorides appeared late in 1980 and is introduced in Section 1.5.2. The gas-phase introduction of volatile metal chelates into an ICP is also an interesting new development and is discussed in detail in Section 1.2.1.3. 1.5.1 Hydride Generation A useful review of the hydride generation procedures for AAS was published at the end of 1980 (409), and among the aspects covered were state-of-the-art methods of atomization and reactor design.Reamer et al. have now published their detailed study on generation vessel materials using radio-tracer ’jSe (587; see ARAAS, 1980, 10, 63). Their study is of considerable importance to analysis at ultra-trace levels, as it is shown that silanization of glass cells is essential to avoid losses of Se on the cell walls.An alternative is to use PTFE which is less adsorptive as the surface appears to saturate. Reamer et al. (587) and Andreae et al. (1578) used an air/H flarne inside a heated quartz tube atomizer to enhance atomization (see ARAAS, 1980,10, Ref. 1201). Air was introduced into the H2/N, stream just before the quartz tube.This procedure avoided the necessity of removing HCl and H 0 vapours prior to atomization (587) and also reduced gas-phase interferences (C769). An inexpensive hydride generation system has been reported for As, based on the use of boiling tubes for the reaction vessels and introduction of the arsine through the nebulizer system of an H /N diffusion flame (1295). Measured in terms of the number of application studies reported this year, flames and heated quartz cells with AAS measurement appear to represent the most widely used techniques.Automation of both the hydride generation and measurement steps using a flame or electrically heated quartz cell has received much commercial interest in the past year (C121, C765, 965, C 1 15 1, 1265).The availability of automated commercial systems should reduce the tedious nature of this type of analysis, at least for those that can afford them. The introduction of samples by flow-injection (C772, C2060) offers the possibility of a low cost self-constructed approach to automation. In h r o m ’ s method, the sample was injected into a continuously flowing stream of HC1, into which the NaBH, solution was subsequently mixed.The gaseous hydride (of Bi) was removed from the solution in a gas - liquid separator and swept into an electrically heated quartz tube AAS cell. Apparently the flow injection procedure allows interference effects to be minimized but the method by which this was achieved is not clear. Alternatives to the flame and heated quartz cell AAS procedures have been a recurrent theme of this section of ARAAS for several years.The past year, has also seen a number of significant studies in this area, and also some novel commercial developments. The44 A nalyt ical A tomic Spectroscopy combination of hydride generation and conventional graphite furnace atomizers appears to offer the possibility of improved sensitivity (596, C761, 1380, 1578, 2217).Inui et al. (596, C761) used a horizontal glass tube for the generation of hydrides of As (596) and Se (C761), the sample solution (10 - 25 pl) being injected through a side port onto a pellet of NaBH, . The generated hydrides were swept into a conventional electrothermal atomizer already heated to 2400 "C. A significant contribution to this subject was reported by Andreae and Froelich (2217), who reported the determination of Ge in natural waters with a detection limit as low as 0.56 ng 1 - This exceptional performance was achieved using a 250 ml reaction vessel, and collection of the evolved GeH, in a liquid N, cooled trap.When the hydride generation reaction was complete, the condensed vapours were heated rapidly and the germane passed into the graphite furnace of a commercial electrothermal atomizer operated at 2600 "C.The sample vapour was passed from one end of the furnace to the other by using the two normal exit ports. A similar system was adopted for Sb determinations and a detection limit of 0.3 ng 1 - was achieved (1578). These low detection limits are clearly a function of the large sample volume, the hydride collection system, and the increased atomization efficiency at 2400 - 2600 "C.Application of this system to other elements would be interesting. The determination of hydride forming elements by AES using both ICP and d.c. arc plasma detection continues to generate much interest. Goulter et al. (C31, C1149, C1454) have modified the continous flow hydride generation system first described by Thompson et al. (ARAAS, 1978,8, Refs. 524,53 1). A pre-oxidation step has been added which allows the determination of Hg and Pb as well as the normal hydride forming elements. ICP methods for As (401, 2145), P (1480), Pb (1340) and Sn (1388) have also been described. The interesting procedure used to generate PH , (1480) was based on reduction of Ca, (PO,) with A1 powder in a Ta - C tube furnace which produced Ca, P, .This was converted to PH , with 2.7111 HCl. A d.c. arc plasma method has also been described which differentiates between phosphinic and phosphonic acids (962) (see also Section 3.1.1.4). The ICP has also been used as a chromatographic detector for the determination of As, Ge and Sb in ad-mixture (C661, C1148).The hydrides were generated, separated on a column of Chromosorb 102, and each peak detected in sequence with a programmed monochromator. Atomic fluorescence spectrometry is a popular method for the determination of hydride forming elements and the use of non-dispersive atomic fluorescence spectrometry has again been reported by a number of authors (C829, C830, C831, C1002, C1146, 1339,2073).Tsujii et al. (C829, 1339) described a new burner on which an entrained air/H,/Ar flame can be maintained at an H , flow of 0.15 1 min - * . This gave detection limits of 10 pg As and 20 pg Se. A new commercial non-dispersive AFS spectrometer has also been described which incorporated an electrically heated quartz tube atomizer and the usual solar-blind PMT (C831). Electrolytic generation of SnH, has been reported in an AFS method in which the SnH, was preconcentrated in a liquid N, cooled trap before atomization at 750 "C (266).Important developments have taken place in the determination of different forms of As (374, 394, C764, C766, 2150), Sb (C766, 1486, 1578, 2073) and Pb (1808), and these are discussed in detail in Section 3.1.1.4.Two novel instrumental methods of speciation have been described. Arsenic species including arsenite, arsenate, monomethylarsonate, dimethylarsonate and p-aminophenylarsonate have been separated by ion-chromatography using a standard Dionex anion-separator column (2150). The species were determined using two separate eluents but in both cases the column effluent was analysed continuously in a continuous flow hydride generation system.Organo and inorganic Sb compounds have also been separately determined by a chromatographic procedure (1 578). Several detailed studies of inteferences have been reported (C119, C768, C770, C1147, 2139) and are discussed in more detail in Section 3.1.1.4. Almost all relate to interferenceAtomization and Excitation 45 during the hydride generation stage.New releasing agents have been reported for As (C768, 1379), Pb (1321) and Sb (568). The effects of volatile oxides of nitrogen produced by reduction of nitric acid and reported last year (see ARAAS, 1980, 10, Ref. C81) have been described in detail (2192). Instrumentally the most important contribution to removal of interferences has been the demonstration by Dedina (C769) that the flame-in-tube atomizer (see above) reduced gas-phase interferences. The original atomizer of Dedina and Rubeska (see ARAAS, 1980, 10, Ref. 1201) has been redesigned to allow a higher concentration of hydrogen radicals to be produced by the 0, /H2 flame. This was said to reduce gas-phase interferences from other hydride-forming elements by several orders of magnitude. 1.5.2 Fluoride Generation A new method has been described for the determination of Ce, Ta, Th, U and Zr, based on AFS measurement after volatilization of the elements as fluorides (323). The principal application was to the determination of Zr in rare earth metal fluorides. The sample (50 mg) in a PTFE micro-autoclave was heated with ClF,, to convert Zr to ZrF, .The volatile components were evaporated at ambient temperature by means of dry N2. The residue was then heated at 2000 K, in a fused corundum boat and the evaporated ZrF, passed in a stream of He to a tantalum atomizer operated at 2000 K, where it was mixed with atomic Na and atomized. The AFS signal of Zr was generated with a xenon lamp and measured at 360.1 nm and could be used to detect between 0.2 ng and 5 pg Zr in the original sample.The principle of this method looks very interesting but the possibility of applying it to other sample types is not clear. 1.5.3 Mercury Determination The number of novel ideas for cold-vapour Hg determinations is again very few. The principle of collection of the vaporized Hg before transfer to the absorption cell appears to be gaining in popularity, and automation of the procedure is receiving considerable attention.An indirect method for measuring SO, based on cold vapour Hg detection which has been reported this year (1 579) could be of considerable environmental interest (see also Section 3.1.3). Tuncel and Ataman (130) reported an investigation of absorption ceNdesign and showed that the best detection limit was achieved with a cell with a profile which matched that of the hollow-cathode light beam in a commercial AAS instrument.A system in which the reaction vessel and the absorption cell were combined in a single unit, and Hg absorption was measured after 3 min in a stationary mode has been described (1468). This represents a minor adaptation of an earlier procedure (see ARAAS, 1978, 8, Ref. 948). Although AAS continues to be the most widely applied measurement technique, the development and improvement of alternatives is still being reported. A number of authors have reported the use of atomic fluorescence spectrometry (1024, 1296). Ebdon et a/. (1024) have shown that a “windowless” cell obtained by using a sheath of Ar gas allows detection limits down to 20 ng 1 - I to be achieved by AFS (see also ARAAS, 1980,10, Ref. 35). The use of an atmospheric pressure He microwave plasma operated in a TM,,,, Beenakker cavity gave a very low Hg detection limit by AES of 4.1 ng 1 - I (1734). The major problem was from the background emission at 253.7 nm from NO which was overcome by degassing the sample before generation of Hg.The authors believe that introduction of automatic background correction would improve the detection limit to less than 0.5 ng 1 ~ I . Monitoring of Hg in the atmosphere using continuous (257, 1600) and retrospective measurement (383, 1296) methods based on cold-vapour determination has been described by a number of authors. The continuous Zeeman AAS monitor reported in ARAAS last year (see ARAAS, 1980, 10, Ref. 426) has been the subject of two US Government reports (257,46 Analytical Atomic Spectroscopy 1600). Absorption of Hg from the air on 600 mg 25 - 50 mesh Au sponge and subsequent desorption and AAS determination was proposed in one retrospective monitoring procedure (383). Scott et al. (1296) investigated a passive Au wire sampler as a means of monitoring Hg in factory atmospheres.A 1 cm section of the wire was exposed to the atmosphere for the requisite time and levels of Hg between 10 - 120 pg m- could be determined by desorption and AFS detection. The sensitivity of Hg determination is obviously improved by collection of the evolved mercury and subsequent rapid transfer to the measurement cell (see ARAAS, 1979,9,30).The use of Au is now being applied more widely and in addition to the two applications mentioned above, other authors appear to be introducing the procedure into routine applications (634,954,1253). The use of sulfhydryl cotton for collection of both organic and inorganic Hg from sea-water has also been reported (1792). Automation of Hg determinations using continuous-flow apparatus has again been described (C869,1968,2088) and applied to routine analysis of biological samples (2088) and to continuous monitoring of Hg in solution (C869).Automatic AAS (C765, 1265) and ICP (C31) systems for Hg determinations have also been reported. Speciation of Hg by selective reduction procedures .(2223; described more fully in Section 3.1.1.3) and by physical separation procedures continues to generate interest.The separation of inorganic and individual organic compounds of Hg by TLC of the dithizone complexes appears to offer a cheap, sensitive but complicated alternative to other methods of speciation (272, 953, 1810). After complexation and separation, the spots were cut out and heated above 600 "C in a quartz tube, and the Hg determined by cold-vapour AAS.A novel indirect method for the determination of sulphite or SO, has been proposed (1579), in which Hg was generated by the disproportionation of Hg, + on reaction with dissolved SO,, according to: Hg,,+ + 2S03,- ->Hgo + Hg(S03)22- The driving force for the reaction is the complexation of SO, - by Hg(I1); and the optimum pH of 3 was a compromise between that required for efficient complexation/ disproportionation and that required to prevent hydrolysis.The Hg evolved was determined by cold-vapour AAS and yielded an SO , detection limit of 30 pg with a range up to 5 ng. The authors propose to use the method for monitoring SO, in the atmosphere and their procedure may offer an attractive alternative to others used for this determination.Other reference of interest- Generation of Hg vapour by electrically heated graphite capsule: 525. 1.5.4 Methods Based on Molecular Absorption/Emission Two interesting papers have appeared on the determination of NH by gas-phase molecular absorption (171,2216). An improved detection limit of 7 ng ml- was achieved by trapping the NH generated in a liquid N, cooled trap followed by measurement in a 1 m absorption cell (171).The method has also been automated using standard continuous flow analyzer components and a gas - liquid separator to separate the NH which was measured in a T-shaped cell heated to 160 "C (2216). The heated cell is said to remove the base line drift, peak distortion and tailing caused by condensation of water in the inlet tube of the cell.The determination of various species of N by microwave plasma emission spectrometric measurement of N z has been reported (156, 1704). A Beenakker type TM,,, microwave cavity was used to generate an atmospheric pressure He plasma for the determination of NH + , NO - and NO - as N (1 56). Conversiorr of each species to N , was carried out in a reaction cell, NH, + with NaOBr in 0 . 0 5 ~ NaOH, and NO, - using sulphamic acid in 0 . 2 ~ HCl. Nitrate was determined by prior conversion to NOz - with a Cd - Cu reductor column. The N, can be determined using lines for N2 + (391.4 nm), N, (337.1 nm), N (746.8 nm) orAtomization and Excitation 47 NH (336.0 nm), but that for N2 + was most sensitive and yielded a detection limit of 0.004 pg ml- I . For the determination of I5N (1704), NH,Cl was converted to NH3 with a flame, and N, was generated from this with an h.f. discharge. The technique tunable atomic line molecular spectrometry (TALMS) has been reported for the determination of NO, SO,, NO,, and small organic molecules such as formaldehyde and benzene (C807). The source used is one in which the atomic line is split by a magnetic field so that one of the Zeeman components coincides exactly with a molecular absorption line whilst the other concides with an adjacent non-absorbing portion of the spectrum. The difference in the absorption of the two Zeeman components is proportional to the concentration of the molecular species.

ISSN:0306-1353    

DOI:10.1039/AA9811100043

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

CHAPTER 2 lnst rumentat ion 2.1 LIGHT SOURCES This Section describes all sources used in AAS and AFS, with the exception of lasers which are discussed in Sections 1.1.2 and 1.3.5. Hollow-cathode lamps continue to be the most popular source and several workers have exploited their high spectral intensity when they are operated in a pulsed mode. Reported intensity improvements compared with CW operation have been as high as 100-fold (C1418).A 10-fold improvement in the sensitivity of atomic Faraday spectrometry was obtained by pulsing the HCL (572). Walters and co-workers previously reported a preliminary study of r.f. boosted pulsed HCLs (see ARAAS, 1980 10, Ref. 920). They have now used a high resolution spectrometer to study the line-widths (C17) and they have characterized the performance of these lamps with respect to their operating conditions, intensity and line- width data, and cathode vapour propagation times.Although the use of a pulsed HCL with a synchronously pulsed detector was shown previously to give improved SNR (see ARAAS, 1977, 7, Ref. 989), Minami (C1461) has cautioned against pulsed mode operation for AAS since line broadening may adversely affect sensitivity.Broadening is, however, more acceptable in AFS where the increased intensity of the pulsed source is especially advantageous. Demers and Allemand (1 57 1) used pulsed HCL’s, modulated at 500 Hz, for AFS with an ICP atom cell, and r.f. pulsed HCLs have been used with a time resolved AFS instrument (C828). An 8-channel multi-element AF spectrometer (C69) used HCLs and EDLs pulsed up to 16 timedcycle with a cyclic frequency up to 2 KHz. Higher spectral intensity is expected from a HCL if the discharge region is restricted to a small volume and two modified lamps were reported to achieve this. One was an h.f.lamp operated at 144 MHz and powers up to 25 W (C811,274), and the other (1652) had a cathode with a conical rather than a flat base giving a 1.8-fold intensity improvement for Cu.Additional references on the preceding topic - 287, 318. Thermostated air sheathing was used by Walters and Smit (1542) for temperature stabilization of r.f. powered electrodeless discharge lamps (see ARAAS 1980, 10, Ref. 935). The lamps studied (Cd and Zn) gave similar spectral output, line-widths and profiles to microwave EDLs, but were said to be much more stable and reproducible.A high power (500 W) microwave power supply for EDLs has been described (1658). Additional reference on the preceding topic - 1918. Controlled temperature gradient lamps continue to be developed by Cough and Sullivan. Their previous lamps were very suitable for several volatile elements (see ARAAS 1979, 9, Refs. 799, 1678), but they were not successful for P and S because the vapours of these elements reacted with the electrode material. A new design (361) had a resistively heated furnace to vaporize the elements, but the geometry of the lamps ensured that P or S vapours condensed before reaching the electrodes. The power supply consisted of 2 outlets connected in parallel across the lamp, one delivering the excitation current of 750 mA at 120 V, and the other delivering 10 mA at 900 V in order to strike the lamp and provide a continuous current to permit square-wave electronic modulation of the excitation current.Compared with EDLs, these lamps had a faster warm-up time, higher intensity and narrower line-widths without self-reversal. The determination of P (177.5 nm) and S (180.7 nm) by FAAS in a separated N,O/C,H, flame with a N,-purged light path gave detection limits of 10 and 13 ppm, respectively.These lamps have now been fabricated for As, Cd, K, Na, P, Ru, S, Se and Zn. 4950 Instrumentation A glow discharge lamp with a common anode between a thermionic cathode and a cylindrical modulating cathode was used as a light source for AAS (630).Good SNRs and sensitivity were claimed. A demountable electron bombardment volatilization source (C792) contained an element in a sample holder which was positively charged with respect to a W filament. The sample was excited by bombarding it with electrons accelerated at up to 500 V d.c. (500 mA) under vacuum. Other references of interest - Comparison of HCL, EDL and metal vapour discharge lamps: 173 1.Light sources for Zeeman-effect AAS: 1628, 1963. Review of light sources (in Russian): 524. 2.2 OPTICS 2.2.1 Background Correction The use of deuterium background correction when low absorbances are being measured sometimes results in over compensation due to the line absorption of a matrix element. Vajda (1025) illustrated this problem by determining Pb by AAS at 217.0 nm in the presence of Sb which has a resonance line at 217.6 nm.The accuracy of background correction would normally increase when using larger spectral band-widths, but it was shown that the Sb line interfered seriously unless a small band-width was used. The author presented a simple formula for calculating the largest band-width at which the analyte and matrix element are resolved and hence interference will not occur.This led to a very useful table listing 8 1 known pairs of interfering elements and recommended band-widths for use with background correction AAS. A problem of insufficient intensity from the Dz arc in a Perkin - Elmer AA spectrometer was solved (1 164) by removing the least transparent segment of the graded neutral density filter (a spinning partially silvered disc).Woodriff and co-workers have developed a background corrected system for ETA-AAS with the Woodriff furnace (984). Zeeman-effect AAS is becoming increasingly popular as an effective means of background correction in ETA-AAS. Its advantages over conventional AAS with a D, lamp include the effective correction of much higher background levels and background correction at the exact analytical wavelength.Hence, spectral interferences by matrix elements as described in the preceding paragraph do not occur. This was demonstrated by Fernandez et a/. (C1087) who successfully determined Se in nickel-base alloys which could not be analysed by conventional D, correction, since high concentrations of Fe and Co caused spectral interference at the Se 196.0 nm line.Several Zeeman ETA-AAS instruments have been described (146,232) including the Perkin - Elmer Zeeman/5000 AA spectrometer (C86, C812, C997, C1087, 1326) (seealso Section 2.5.2). Two disadvantages of most Zeeman spectrometers are their restricted modulation frequency, giving problems with rapidly changing backgrounds, and sinusoidal modulation allowing only brief measurement at zero magnetic field strength.A modification of the magnet power supply, however (CSOS), overcame these problems to give a zero field measurement of 0.5 ms while retaining a maximum field strength of 10 kG and a modulation frequency of 50Hz. The Zeeman technique has also been used with FAAS (C42) and multi-element non-dispersive AFS (C69) (see Section 2.5.3). Background correction in ICP-OES is easily achieved using programmable monochromators to measure the background on either side of the analytical line and then automatiqally subtract it from the analytical signal (C12, C27,13 13).It can also be effected by wavelength modulation using a quartz refractor plate, and McLaren and Berman (1 888) illustrated the effective background correction obtained with their ICP echelle spectrometerAnalytical Atomic Spectroscopy 51 (see ARAAS, 1980,10, Refs.C1367, C1402) by determining traces of Cd and Pb in biological tissues and sediments. An ICP-echelle spectrometer used as an on-line detector for HPLC (C719) had a quartz plate vibrating at 200 - 500 Hz, and effective background correction was claimed for As (193.7 nm) and P (213.6 nm).Michel et al. (399) described a rotating quartz chopper fabricated from quadrants of differing thicknesses to produce square wave modulation in AFS. Its effectiveness in discriminating against flame and matrix background was demonstrated by the determination of Cu in blood serum. Inductively-coupled plasma atomic fluorescence spectrometry was claimed to avoid many of the background interferences experienced in ICP-OES (C25).Interferences due to radiative recombination continua, molecular emission, stray light and spectral line interferences were all said to be circumvented by this technique when the atomic fluorescence was modulated and the detector electronics were a.c. coupled. Other references of interest - Apparatus for removal of background absorption in AAS: 1856.Atomic magneto-optical rotation spectroscopy for background correction: C1462. Background corrected echelle polychromator system: C456. Circuit for simultaneous monitoring of background corrected AA signals and background only signals: 163. Comparison of methods for background correction in ETA-AAS: C5 16, C 1434. 2.2.2 Optical Systems The usefulness of wavelength modulation in the continuum source AAS system of O’Haver and co-workers (see ARAAS, 1980, 10,70) has been described (C43, C1101). It provided a means of extending the dynamic range to 4 - 6 orders of magnitude by enabling absorbance measurements to be made at various points across the absorption line profile, thus producing several calibration curves with different sensitivity (see Section 2.4.1.2). By modulating at frequencies greater than the source flicker frequency, improved detection limits were also obtained.The modulation waveforms giving the best SNRs were a 3-step square wave and a bi-gaussian wave. The effect of modulation frequency on SNR was investigated and it was found that a 3-step square wave became distorted above 20 Hz, thus degrading the SNR.Hieftjeand co-workers(l65, C718, C1135)describedaninterestingmethod for reducing spectral interferences in ICP-OES by using selective spectral-line modulation. A double- beam optical system was constructed, with one beam traversing a coloured flame which acted as a wavelength-selective absorber, and the other beam acting as a reference. Radiation from the ICP was directed alternatively through and around the flame by means of a mirrored chopper.A synchronous detection system then registered the selectively absorbed analyte wavelength but rejected all others. Thus, there was effectively a very narrow-band highly selective filter which was locked in wavelength to the analytical line. This could be an extremely useful technique, enabling high resolution OES with relatively inexpensive filters or monochromators.The addition of microcomputer controlled stepper motors to both wavelength drives of an echelle monochromator was previously described by Parsons and co-workers (see ARAAS, 1980, 10, Ref. C564). This system has now been described in more detail (2189). Although wavelength increments as small as O.OOO1 nm could be selected during scanning, there were difficulties with wavelength calibration (C8).The prism disperses wavelengths in the vertical plane to act as an order separator and it was found that changing entrance and exit slit heights caused the wavelength to shift in the vertical plane. Thus, it was necessary to include a compensation for slit height when scanning by moving the grating and prism in52 Instrumen tation conjunction with each other.A study of the slit function (plot of intensity emerging from the exit slit vs. wavelength) was made for the system (C16). The scanning echelle monochromator proved especially useful for the investigation of spectral line profiles in the U.V. (C710). A scanning echelle spectrometer was also described by Farnsworth and Walters (C17).Spectrometers incorporating programmable monochromators are now very popular. Human et al. (C854, C1405, C2016) described the conversion of a commercial 1 m Czerny - Turner monochromator to a microcomputer controlled rapid scanning instrument by incorporation of a stepper motor. Wavelengths were scanned in increments of 0.001 nm and an optical encoder attached to the grating axis provided a resolution of 0.0018 nm.The usefulness of computer controlled scanning monochromators for coupled GC-ICP-OES has been illustrated in the determination of volatile hydrides (C661). They were sequentially determined by step scanning to predetermined wavelengths for each chromatographic peak. Additional references on the preceding topic - C97, 195, 306. It has been shown (C18) that the resolution of aberration corrected holographicgratings remains constant with variation in slit height and thus gratings could be useful for spatial studies such as the measurement of emission intensity vs.height profiles in a plasma. A unique opto-electronic signal gating system was used for time dependent observation of laser AFS signals (C71).Scattered radiation could be separated from fluorescence on the basis of time resolution, since scatter is instantaneous while fluorescence has a finite decay time. Other references of interest - Dual polychromator system for ICP-OES: 656. Etalon-spectrograph system for improved spectral resolution: 256. Optical systems for AAS: 1246, 1261.Spectrometer alignment by use of laser diffraction: 2172. 2.3 DETECTOR SYSTEMS Solid state imaging detectors (SSID) continue to be investigated as multi-channel detectors, and though systems providing better SNR characteristics have been constructed in recent years, they are still inferior when compared directly with PMTs. Kubota et al. (C691) mounted an intensified photodiode array in the focal plane of a spectrograph so that a wavelength range of 15.5 nm was covered.Using a number of emission sources, signal-to- dark current ratios were found to be somewhat poorer than for PMTs. Bubert and co- workers previously used a 5-element diode array detector with a GDL for analysing powdered rock samples (see ARAAS, 1979, 9, Refs. 807, 1473). They have now used an array of 20 diodes with the system (1550) and reported a range of detection limits which were slightly poorer than those obtained previously.Further details have been published (1793) of the work of Fry et al. (see ARAAS, 1980,10, Ref. C1040), who used a diode array detector with an ICP for near i.r. studies of non-metals. The difficulties of using SSIDs for simultaneous multi-element AAS have been well documented.Problems of multiplexing line sources and of limited dynamic range have been referred to (see ARAAS, 1974, 4, Ref. 109), but an increased linear dynamic range has been claimed for ETA-AAS (C62). Details were not given, but presumably this was achieved by simultaneously detecting more than one analytical line per element. The same authors multiplexed a line source and a continuum source on their ETA diode-array spectrometer (C63) in order to monitor both atomic and molecular absorption profiles.The molecular absorption measurements allowed a useful study of matrix effects caused by the formation and loss of volatile metal halides. Diode arrays have been found extremely useful for investigating the spacial characteristics of light sources.Horlick and co-workers (C26, C77, C510,968) studied the spacial distribution of atomic and ionic emission lines in an ICP by using a photodiode array spectrometer. Combination ofAnalytical Atomic Spectroscopy 53 vertical and horizontal profiles enabled measurement of the whole ICP emission image. A diode array has also been used to study spacial profiles in a d.c.plasma (C1098). Additional references on the preceding topic - C504, C1138. The use of vidicons for multi-element atomic spectrometry has declined in recent years. A problem of vidicons compared with SSIDs is their capacitive lag which renders them generally unsuitable for monitoring transient signals. This problem was discussed by Staerk etal. (41 1). A SIT has been used (1614) for acomparisonofN,O/C,H, FAES with ICP-OES (see ARAAS, 1980, 10, Refs. 276, 844). Yung and co-workers (1560, 1570) used a silicon target vidicon in conjunction with a laser source in order to study spacially resolved temperature profiles in an air/C H flame. The use of an ICP as a light source for FAFS was reported previously (see ARAAS, 1979, 9, Ref. 1851). Omenetto and co-workers (C727) have now used this instrumentation with the flame acting as a resonance detector for the analyte emission from the ICP.Measurement of the fluorescence with a PMT then gave a very selective means of detection in cases where spectral interferences would be a problem in ICP-OES. In spite of this selectivity, resonance detectors have certain disadvantages. Sensitivity is poor when elements have several atomic lines which all absorb the incident radiation.Also, a high background signal is observed even when measurement of the resonant fluorescence signal is delayed until the discharge pulse has terminated. An improved resonance detector (C751) was claimed to overcome the latter problem, which was said to be due to emission excited by metastable atoms of the Ar carrier gas.Instead of a closed fluorescence cell containing a hollow cathode under reduced pressure, an Ar flow-through cell at 1.7 Torr was used. A substantial improvement in precision was claimed. In optogalvanic resonance detection, instead of measuring resonance radiation from the fluorescence cell, the change in impedance of the discharge is measured. Stephens (1290) designed such a detector for Hg and evaluated its response characteristics when irradiated by a penlight discharge lamp.The output varied linearly with source intensity and had similar quantum efficiency to a vacuum photodiode. An optogalvanic detector has been used in FAAS with a pulsed dye-laser light source (C1152,1939); the detector was a commercial HCL containing the element of interest.A detection limit for Na of 1 ng ml --I was claimed (1999) for an optogalvanic resonance detector when the absorption flame cell was positioned inside the 25 cm cavity of a dye laser. A spectrometer with combined photon counting and dc. amplification (21 57) exploited both the inherent sensitivity and stability of photon counting at low radiational fluxes and the extended linearity and high precision of d.c.amplification at high fluxes. Automatic cross- calibration could be achieved at the intermediate region where both systems were used (see ARAAS, 1979,9, Ref. 536). Additional reference on the preceding topic - 201 8. Other references of interest - Measurement of the gain of a PMT: 2019. Review of imaging detectors for the u.v.: 1519.Use of an ID for spectral emission analyses of gases: 1369. 2.4 INSTRUMENT AUTOMATION With the increasing availability of low cost computers, microcomputers and micro- processors, atomic spectrometers are undergoing revolutionary changes. Computerized data processing is now commonplace, even on many lower priced instruments, and computer- control is a frequent feature of more sophisticated spectrometers.Automated sample54 Instrumen tation introduction systems are becoming increasingly important because they permit the complete automation of computer-controlled spectrometers. In view of their importance in this respect, they are included in this Section under a separate sub-heading. 2.4.1 Computer Control and Data Processing Besides the obvious advantage of faster analysis, computer-controlled instruments have been shown to allow improved accuracy through better analytical quality control, and improved precision by means of more sophisticated data processing.Rapid advances in the study of atomization processes and interference studies are also taking place with the aid of computerized data processing. The current status and future trends in automated spectrometric analysis have been reviewed by De Galan (C687, 1278) who predicted that trends in automation will be increasingly towards the optimization of analysis conditions and the quality control of analytical data.A review of developments in microprocessors and their applications in both control and data processing has been published (2023). Additional references on the preceding topic - C855, C1979. 2.4. I . I . Emission Software improvements continue for instruments equipped with slew-scanning programmable monochromators. The interactive graphics software package for the Perkin - Elmer ICP/5000 system (C6, 346) provides for computer-controlled spectral acquisition over a wavelength range selected by the analyst. Up to 9 spectra may be stored and displayed on a CRT screen for comparison.This facility is useful in methods development when analytical and appropriate background correction wavelengths need to be selected. Further software has also been developed for the Instrumentation Laboratory Plasma 100 instrument. A semi-quantitative program (C28, C712, C2114) is claimed to enable the determination of 27 elements in an unknown sample in about 5 min, with an accuracy within 2 5 - 10%.Calibration of the instrument is with a single standard solution. In order to check for spectral interferences, the instrument automatically scans a small portion of the spectrum and displays it on a CRT. Additional data processing for the Plasma 100 was provided by interfacing a bench-top microcomputer, programmed in BASIC, to the instrument (C858, C1425).This system provided automatic preparation of an analytical report, storage of data in archival form, storage of methods and automation of certain mathematics. It could also provide an intelligent interface between the spectrometer and a larger computer. An ICP spectrometer incorporating a slew-scanning double monochromator was controlled by a microprocessor directly programmed in BASIC (2027), so that the operator could modify the software more easily than with systems dependent on machine language or compiler based language programming.A very important feature of rapid scanning instruments is the ability to select analytical wavelengths accurately and rapidly (see ARAAS, 1980, 10, 69). It was claimed that the continuous measurement of grating position by means of an incremental angle encoder provided wavelength positioning within k0.003 nm ((2111, C2001).A microcomputer automatically corrected the angle position by comparing it with the desired position. It seems unlikely, however, that this method is as reliable as those routines which select the peak by finding the position of maximum emission.A 4K microcomputer was used as part of the interface between a 24K desktop calculator and a scanning monochromator (1313). The calculator was used for operator communication and data handling while the microcomputer performed various control functions including the selection of instrumental parameters and the operation of an autosampler for ICP-OES.Additional references on the preceding topic - 306, C436, C449, C1157, C133 1. A computer coupled to a direct reading emission spectrometer (C 102) controlled certain instrument parameters and collected data from up to 48 channels. A unique A/D conversionAnalytical Atomic Spectroscopy 55 scheme has an individual V-to-F converter and counter for each channel. Counter overflows for each channel could be recorded during a sample burn in order to achieve the required dynamic range.The use of a computer-controlled microphotometer for qualitative emission spectrography was described (C859). Spectral wavelengths were identified from a library in computer memory which contained almost 400 spectral lines covering 68 elements. Their listed intensities were also contained in memory for semiquantitative analysis.The same instrument was used for quantitative analysis (C85 1) with calibration based on a computed relationship between exposure and fractional optical blackening. A computerized micro- photometer has also been used for spectrographic analysis with an ICP (C711). In this case almost 600 lines of 62 elements were stored in memory. Cross-correlation masking has been used in ICP-OES (C 1 120, 1554) and ICP-Fourier transform spectrometry (C29, 1553).In the latter application the technique enabled the analytical information to be extracted from the interferogram without following the time- consuming Fourier transformation step. Correlation was effected by the point-to-point multiplication of the interferogram with the computer-generated mask, which was essentially a synthetic interferogram. A new correlation-based method for determining atomic excited- state lifetimes on a sub-nanosecond time scale has been described by Hieftje and Russo (C694). The method, based on linear response theory, involved cross-correlating the time dependence of the induced atomic fluorescence with the time-dependent characteristics of the light source (a laser).Additional reference on the preceding topic - 1361. A mathematical model has been presented (189) which permitted atomic line shapes to be derived and spectral overlaps to be studied. The model was developed using hyperfine structure, overlap separation values and line broadening calculations. Other references of interest - Calculation of discharge current waveforms in high voltage spark sources: 970.Cluster analysis for the interpretation of geochemical data obtained by OES: 1596. Computer-aided spectral identification of laser-induced plasma emission: 2068. Computer program for OES calibration: 369, 370. Computer program for OES using internal standards: 217. Equation for an electric arc: 1 183.Error propagation in multicomponent analysis: 1563. Estimation of electron number densities in plasmas: 2181. Processing data from a d.c. plasma echelle spectrometer: 1987. Software for determining the precision of integrated signals: 368. Use of Abel inversions for evaluating emission profiles in plasmas: 968, 981. 2.4.1.2 Absorption Computerized data processing provides several useful features in electrothermal atomization-AAS.Rapid recorder response gives high signal resolution enabling peak shapes, appearance times, etc., to be studied. Several workers have studied matrix interferences in this way. In one application (C798), signals were taken directly behind the A/D converter of an AA spectrometer and sent to an external computer. It was possible to collect 50 data points per second in this way and to record the peak onto a graphics plotter.It was suggested by Lundberg and Frech (C853), who used an on-linecomputer AA system, that many previous interference studies are of limited value because slow response recorder systems were used. The use of a Perkin - Elmer Data Station for ETA signal processing has been discussed (C857, C1416j.It was emphasized that the ability to observe absorption peak56 Instrumen tation shape could aid in the optimization of analytical conditions. A microcomputer used for ETA- AAS with a metal microtube (371) was reported to improve detection limits by an order of magnitude due to data enhancement by signal accummulation, scale expansion after background correction and signal smoothing.A fully computer-controlled ETA-AA specrometer has been described (C852) (see Section 2.5.2). Several computer programs in BASIC have been reported for use with ETA-AAS and for the method of additions (1 184, 1188). A 48K microcomputer was used to link two commercial flame atomic absorption spectrometers with autosamplers (400). Six elements were determined in 100 soil extracts or plant digests in about 2% hours.An ingenious means of extending the linear calibration range in simultaneous multi- element atomic absorption continuum source spectrometry (SIMAAC) was described by O’Haver et al. (C1129, 2035). Computerized signal processing enabled intensities to be measured at several wavelengths across each absorption profile during a wavelength modulation cycle.Absorbance values were then calculated at six different positions on each line and six calibration graphs were computed covering a wide sensitivity range (the most sensitive at the line centre). In this way, overlapping linear ranges covering 4 - 6 orders of magnitude in concentration were achieved. A diagnostic FORTRAN program has also been described for this instrument (C 125).By automatically storing the absorption profile of all 16 channels, any spectral interferences could be observed. Other references of interest - Empirical formula for ETA-AAS calibration: 2 146. High temperature equilibrium calculations for optimization of ETA-AAS: C1081. Mathematical corrections for shift and hyperfine structure of spectral lines in AAS: 219. Numerical approach to calibration in OES: C73 1.Simplex techniques for non-linear optimization: 149. Use of enforced variance in AAS: C834, 1177. Variance-weighted least squares regression procedure for AAS: 152. 2.4.1.3 Fluorescence A control and data acquisition system for non-dispersive multi-element atomic fluorescence spectrometry has been described (C69) (see Section 2.5.3). 2.4.2 Automated Sample Introduction Several new automated sample introduction systems have been reported. Besides the obvious advantage of automatic operation they can lead to increased precision, especially when used with micro-atomic spectrometric techniques. A programmable sample changer for FAAS (C35, C837) could be used for conventional continuous nebulization or for the discrete nebulization of micro-sample volumes up to 200 pl.It was claimed that precision for the discrete nebulization of 200 p1 aliquots was similar to continuous sample introduction, and sensitivity was comparable provided that aliquots exceeded 100 pl. The program allowed standard addition as well as normal calibration modes. An in-line dilution system utilizing an 8-channel peristaltic pump was incorporated into a Perkin - Elder AS-50 autosampler ( I 569) used for FAAS.Within batch RSDs of 2% for Ca, Li and Mg in serum and 1 Yo for Ca and Mg in urine were obtained, and 50 samples could be analysed within 20 min. An automatic sampler for ETA-AAS patented by Beckmann Instruments (1896) can beAnalytical Atomic Spectroscopy 57 programmed to deliver multiple sample aliquots with a drying cycle between each aliquot.A modification of the Perkin -Elmer furnace autosampler (385) allowed up to 6 sample streams to flow continuously through the sample cups. A very useful technique for automated sample introduction is flow injection. Olander (C 1042) described a computer-controlled flow injection FAAS system which consisted of an automatic sampler, peristaltic pump, two loop sampling valves and micropressor controlled AA spectrometer coupled to a minicomputer.It was claimed that flow injection peak area measurement provided a wider dynamic range than normal continuous nebulization. It has also been shown (C2059) that solutions with high salt contents could be introduced into a flame without clogging of the burner.The advantages of using flow injection with an ICP were discussed by Greenfield (1884); both speed of analysis and precision were said to be improved. A semi-automated flow injection system was used for determining Bi by hydride generation (C2060). See also Sections 1.3.3.2 and 3.1.4. Additonal references on the preceding topic - 1297, 2169, 2210. A commercial automated hydride generation system for AAS, the Instrumentation Laboratory AVA, has been described (C121, C762, C1447, C1996).It is necessary for the operator to introduce the sample manually (up to 200 ml), but thereafter all steps are performed automatically including the addition of NaBH 4 , stirring, flushing the hydride vapour through a heated absorption cell and triggering the read mechanism of the AA spectrometer.A system capable of automatically generating hydride and which could also be used for cold vapour Hg analysis was said to be compatible with any AA spectrometer (C765). A continuous cold vapour mercury sample introduction system (C869) used a peristaltic pump to introduce both the sample and SnCl, as an air segmented stream int0.a reaction tube. Mercury vapour was automatically removed by means of a gas - liquid separator and passed through a condenser (to remove water vapour) into a quartz flow cell.An RSD of 1.9% at 4 ppb Hg was claimed. Additional reference on the preceding topic - 1265. An automatic device for molecular emission cavity analysis (C112) consisted of an autosampler which dispensed 1 pl volumes into the cavity and a synchronous drive mechanism which moved the cavity into the flame for a predetermined time.Improved reproducibility was obtained with this system. Other reference of interest - Integrator coupled to autosampler for AAS: 271. 2.5 COMPLETE INSTRUMENTS A new feature of Spectrochimica Acta (Part B) is an instrumentation column. The first of these (1 556) included a survey of commercial instrumentation. 2.5.1 Emission Instruments The versatility of sequential-scanning inductively-coupled plasma spectrometers equipped with programmable monochromators is now being demonstrated by applications appearing in the literature. Their ability to select alternative analytical wavelengths rapidly makes method development easier than with direct reading spectrometers. “Wavelength characterization tables” were used with the Perkin - Elmer ICP system to select the most suitable wavelengths for the analysis of geological materials (C9, C507, C1451).The tables provided information on possible spectral overlap and background interferences. The freedom of the selected wavelengths from spectral interferences was then confirmed by using the graphics utility of the instrument, which allows up to 9 spectral scans to be displayed on the CRT.A further application of the system was to the multi-element analysis of soil samples (667). Applications of the Instrumentation Laboratory Plasma 100 system have also been58 Instrumentation described (C449, C2103). Parameters which could be optimized for individual elements included wavelength, plasma viewing height and background correction conditions.The versatility of this instrument was illustrated by method development for metals in oils (C27) and steel (C2008). An Applied Research Laboratories’ 35000 instrument has been applied to the analysis of plant materials (C2120). A new sequential scanning instrument developed by Instruments SA used a 0.64 m monochromator with a holographic grating (C97, C98, C442, C658).A rapid-scanning spectrometer was added to an Applied Research Laboratories’ 33000 direct reader so that both spectrometers viewed an ICP (1313). Several other new systems involving the use of ICP sources with programmable monochromators have been described (195, C737, C1017, C1331, 2027). Additional references on the preceding topic - C1 1 1,460, C1141, 1857, 1882, C2001, 2137.A computer-controlled direct reading spectrometer (C856) enabled versatility of wavelength selection possibly matching that of rapid scanning instruments. It incorporated a 1.5 m Paschen - Runge polychromator equipped with 2 holographic gratings which were irradiated simultaneously by means of a beam splitter. This resulted in 2 separate dispersed spectra at different positions on the Rowiand circle.There were 19 PMTs, each mounted on a separate carriage which could be moved along the perimeter of the circle and positioned anywhere in either of the 2 sectral areas. It was also possible to scan a spectrum continuously with one carriage. A mobile emission spectrometer was designed for in-plant use in the analysis of alloys (C709).Radiation from a d.c. arc was focused via a quartz fibre light guide to a Paschen-Runge type polychromator with a Rowland circle diameter of 750 mm and a reciprocal dispersion of 0.5 nm mm-* . Additional references on the preceding topic - C436, C707, 1614, 1630, C2123. The use of ICP sources with echellespectrometers was previously reported (see ARAAS, 1980, 10, 71), and the system of McLaren and Berman has now been applied to determinations of Cd and Pb in biological tissues and sediments (1888).Wavelength modulation by means of a quartz refractor plate permitted automatic 2-point background correction. A similar system (C720) also enabled derivative spectra to be obtained, and was applied to determinations of Hf in zirconium alloys (C723).An ICP-echelle spectrometer has been used as a detection system for HPLC (C719). Additional reference on the preceding topic - C435. An emission spectrometer equipped with both spark and glow discharge lamp sources was used for determining Al, B, C, P and S in steels (C749). Bubert and Hagenah (1550) applied their spectrometer consisting of a GDL source, Czerny - Turner monochromator and silicon photodiode array detector to the analysis of rock samples (see ARAAS, 1979,9, Ref. 1473). Horlick and Ng described an inductively-coupled plasma Fourier transform spectrometer (C29, C1143). Cross-correlation greatly facilitated the otherwise time- consuming Fourier transformation process (see Section 2.4.1). Hieftje (C2096) coupled his miniature ICP torch (see ARAAS, 1979, 9,9) with a novel detection system termed ‘selective spectral line modulation’ which was said to be capable of isolating specific analytical lines while maintaining high optical throughput (see Section 2.2.2).An automated dual-channel flame atomic emission spectrometer (C74) continuously monitored K and Na in a coal gas stream. Determinations were automatically performed every 2 - 3 min by introducing a fraction of the gas stream into a flame.A 3-channel flame photometer was constructed (1229) by the addition of an extra optical system. The simultaneous determination of Ca, K and Na was then possible.Analytical Atomic Spectroscopy 59 Other references of interest- Metastable transfer emission spectrometer (see also Section 1.1.3.3.): C44 Modified flame photometers for increased sensitivity: 1201, 1973. Spectrospan I11 d.c.plasma echelle spectrometer: 268, 55 1. Scanning ICP spectrometer as a chromatographic detector for volatile hydrides: C1148. Trends in ICP instrument design: C2000. 2.5.2 Absorption Instruments The instrumentation used in simultaneous multi-element atomic absorption continuum sourcespectrometry (SIMAAC) (see ARAAS, 1979,9,41) has been modified slightly (2035).The refractor plate torque motor and driver used for wavelength modulation were replaced by a galvanometer and scanner controller. A computer generated waveform was then used to drive the galvanometer/refractor plate in such a way as to scan across the absorption profile. Extended calibration curves could then be obtained by measuring intensities at several wavelengths across the absorption profile (see Section 2.4.1.2).Applications of SIMAAC to the analysis of fruit juices (C1106) and standard rocks (C1130) were described with particular reference to the compromise conditions needed when determining several elements simultaneously. Additional references on the preceding topic- C780, C1086, C1100, 1189.Several reports have described the Perkin - Elmer Zeeman-effect AA spectrometer (C86, C812, C997, C1087,1326). In this ETA instrument an a.c. magnetic field (up to 8 kG) is applied perpendicular to the optical path through the furnace. It was said that this design improves linear range and sensitivity compared with instruments using a fixed d.c. field, and for most elements over 80% of the sensitivity obtained by conventional ETA-AAS was retained.The efficacy of background correction was demonstrated by determinations in various complex matrices (see Section 2.2.1). A Zeeman-effect AA spectrometer described by Wirz et al. (C896) had a high frequency modulated light source (6 kHz) in a transverse a.c. magnetic field and used a d.c. plasma atom cell.Detection limits were quoted as 100-times poorer than FAAS and this was attributed to the short path length of the plasma. It seems more likely, however, that plasma noise would be the major contributor to the noise of the system and hence to the detection limits. Other Zeeman-effect instruments reported include an on-line Hg analyzer for gas streams (1600) and a multi-channel spectrometer with an array- type detector (1 853).Additional references on the preceding topic- 541, 1629, 1653, 1758, 1855. Kitagawaet al. (565,572) described an ETA instrument operating by the atomic Faraday effect. It was applied to the determination of Pb in blood, volcanic ash and NBS orchard leaves (595). Electrothermal atomization atomic absorption spectrometers benefit greatly from automation.Autosamplers overcome the precision problems otherwise encountered if the analyst lacks practice in sample pipetting, and computerized processing provides rapid recording and characterization of the transient signals. A dedicated ETA-AAS instrument (C852) was equipped with a personal computer operating in BASIC, an automatic sampler and a graphics plotter.The computer, besides processing the ETA signals, controlled the furnace power supply and the sampler. An automated system for determining several trace metals in nickel-base alloys has also been described (C1414). Fast response electronic systems are desirable for ETA-AAS; a modification of a Varian Techtron AA6 spectrometer gave a 10 ms time constant (2176). A dual-channel instrument provided improved sensitivity by double passage of the light beam through the atom cell (1728).Additional reference on the preceding topic- 1698.60 Instrumentation An AA spectrometer based on a Fabry-Perot interferometer (C 141 8) used pulsed HCLs and a photon counting detection system. The interferometer employed rapid piezoceramic scanning of the plates, which allowed the investigation of short-duration pulses.Other references of interest- AA spectrometer with d.c. arc atom cell: 206. Apparatus for safe ignition and extinguishing of N O/C H flames: 1941. GC - AAS instrumentation: 1826. Modified Perkin-Elmer MAS 50 Hg analyzer: 378. Portable AAS for monitoring vacuum coating processes: 1643. Time-resolved AA spectrometer: C1461. Other AA spectrometers: 567, 1167, 1213, 1247, 1269, 1759, 1854, 1922. 2.5.3 Fluorescence Instruments The interest being shown in multi-element atomic fluorescence spectrometry and its current status have been discussed by Winefordner (C826) and Ullman (C1044).Both emphasized inductively-coupled plasma atomic fluorescence spectrometry as an approach to multi- element analysis. Demers and Allemand (C20, C1046, 1571) have developed a non-dispersive multi-element AFS instrument with an ICP atom cell.Up to 12 independent channel modules encircled the ICP, each module comprising a pulsed HCL, optical interference filter, lenses and a PMT. It was claimed that sample throughput was 3 times greater than direct reader ICP-OES systems and spectral selectivity was excellent. With the exception of those for refractory elements, detection limits were often better than with ICP-OES or FAAS, and linear calibration ranges of 4 - 5 '/z orders were obtained.This instrument is now marketed by the Baird Corporation and it is the first multi-element AFS system to be produced commercially since the Technicon AFS-6 in the early 1970s. A time-resolved multi-element AF spectrometer (C828) used stroboscopic photon counting detection in conjunction with pulsed HCLs (200 mA).A digital delay unit controlled an up-down counter which subtracted flame background noise from the fluorescence signal. An 8-channel AFS instrument described by Van Loon and co-workers (C69) used both flames and ETAS, and the light sources (HCLs or EDLs) were pulsed.The pulse frequency could be varied and was controlled by a microcomputer which was also used for signal and data processing. The high spectral intensities produced large scatter signals, but they were overcome by using Zeeman-effect background correction (C1069). Detection limits were said to be as much as 3 orders of magnitude better than AAS. The computer- controlled multi-element AF spectrometer developed by Winefordner et al.(see ARAAS, 1975,5, Ref. 1237 and 1979,9, Refs. 1917,2013) has been further modified (582) by using an electrically heated graphite filament in an Ar sheathed air/C,H, or N,0/C2H, flame (see Section 1.3.4). When using the iiistrument for FAFS (628), noise sources were examined as a function of flame type, modulation approach and element.A commercial non-dispersive atomic fluorescence spectrometer for determining volatile elements as their gaseous hydrides has been produced in China (C831). In this system the generated hydride is swept into a heated silica tube which is irradiated by microwave powered EDLs (2450 MHz). Fluorescent radiation is detected by a solar blind PMT. Linear dynamic ranges of 3 - 4 orders of magnitude were claimed with detection limits of 0.05,O. 1,0.6,0.04, 0. 1, 0.01,0.6 and 0.05 ng for As, Bi, Ge, Hg, Sb, Se, Sn and Ti respectively. A similar system has been described by workers at Perkin - Elmer (C1146), but no indication was given as to whether it also would be available commercially. Michel et al. (C1070) designed a continuum source excited AF spectrometer which incorporated a double modulation technique.The source intensity and wavelengthAnalytical Atomic Spectroscopy 61 modulation had square wave profiles, and their phase and frequency relationships allowed conventional lock-in amplification to be used for signal detection. Other reference of interest- Comparison of nebulizer - burner systems for laser excited AFS: 558. 2.6 NEW COMMERCIAL INSTRUMENTS The accompanying tables describe the current commercially available instrumentation, details having been supplied by the manufacturers or their U.K.agents. The most significant developments are summarized below, and where they have been described at scientific meetings, details will be found in Section 2.5. 2.6.1 Emission Spectrometers New in 1981 are the Spectrumat range of direct readers marketed by Siemens A.G.Optical Emission Services have introduced a spectrometer with optional arc/spark or plasma source. Jobin-Yvon have discontinued the JY 37P and introduced the new direct reading spectrometer model JY 32P. This has 52 photomultiplier positions of which 32 can be used simultaneously. It can be combined with the JY 38 monochromator and detection system and used as a sequential spectrometer.Perkin - Elmer have replaced the ICP 5000 by the ICP 5500. This is a completely automated sequential system which can determine up to 80 elements in an operator-selected multi-element program. The MPD 850 and MPD/ICP 860 of Applied Chromatography Systems are claimed to be the first direct reading spectrometers which can be coupled to either an ICP source for trace metal analysis or a He plasma source for connection to a gas chromatograph. 2.6.2 Absorption Spectrometers The Perkin Elmer model 3030 introduced at the 1982 Pittsburgh Conference is a double- beam instrument with built in CRT and software including all cookbook information. Graphics are available for flame, furnace or mercury/hydride systems, and software is available in 5 languages.Varian have introduced the AA-975, a double-beam instrument which provides capacity for automatic sequential analysis of up to 12 elements by AAS or AES using a compatible flame sample changer or GTA-95 furnace and sample dispenser. Optics are all reflective and hard-dielectric coated. A 12 lamp turret is fitted. All functions are microprocessor controlled.Pye Unicam have withdrawn the SP 2900, and an extremely sophisticated instrument, the PU 9oO0, has been introduced. A novel double-beam system is claimed to eliminate the SNR degradation of conventional double-beam systems. A double- beam reference channel checks and corrects the base line prior to every measurement and the optics are then automatically switched out of the beam during the measurement period.The intelligent system will select and set up wavelength, band pass, lamp current, flame type, gas flow and read time for each element and sample. All this information is contained in the memory system, so the operator is not required to program the instrument. The conditions are then modified (e.g., alternative wavelength selected) automatically in response to the sample signal.Gas control is via a binary flow system with feedback of absorbance signal to automatically set the optimum fuel flow. The instrument detects the presence of the burner or furnace and switches programs accordingly. Up to 8 elements are sequenced automatically for flame or furnace analysis and a universal burner is used for all flames. A flame auto- sampler is built into the instrument.The spectrometer interfaces with the PU 9007 AA data/control system. New from GBC Scientific Equipment is the GBC 901, an improved single-beam spectrometer with increased resolution over the GBC 900. A graphite furnace62 Instrumentation atomizer is also available from this company. A new electrothermal atomizer from Varian, the GTA-95, replaces the CRA 90.This microprocessor controlled instrument has VDU display of information and is fitted for either manual or auto-sampler use. A tungsten strip electrothermal atomizer with Zeeman-effect background correction is now produced by Scintrex Limited. This compact instrument, which does not require water cooling, is easily portable for field use. 2.6.3 Fluorescence Spectrometers The most significant development in commercial instrumentation is the new plasma AF spectrometer marketed by Baird. This was introduced at the 1981 Pittsburgh Conference. It is discussed in Section 2.5.3.Table 2.6A COMMERCIALLY AVAILABLE EMISSION SPECTROMETERS Reciprocal Focal Supplier Model dispersion/ Wavelength length/ Type of Source Special features nmpermm rangelnm m 8 s Applied Research 34000 0.465 170-407 1 .O Low voltage spark Laboratories Ltd., Quantovac 0.69410.347 170-61 0 d.c.arc Wingate Road, 0.93010.4651 170-81 4 Luton, Beds., England 0.310 Applied Research 36000 0.835 240-250 0.5 Low voltage spark Laboratories Ltd., Quantotest ‘En Vallaire CH-1024, Ecu blens/Lusanne, Switzerland Applied Research Laboratories Ltd., 9545 Wentworth Street, P.O.Box 129, California, USA. Socibtb Francaise d’lnstruments Controlbe d’Analyses. B.P. No. 3, F78320, Le Mesnil, St. Denis, France Full computer control to provide concentration printout; full ra e of options including dual casette, floppy and hard disc storage systems, fast printers, V%s, inter-computer links, remote terminals; twin excitation stand facility available including: Ar stand, airstand, GDL, hollow cathode, ICP, Rotrode, d.c.arc. Maximum capacity of 60 channels to be measured simultaneously Transportable Quantometer with Go/NoGo inspection electronics, 10 channels measured simultaneously Baird Corporation, Spectromet 125 Middlesex Turnpike, 1000 Bedford, MA01730, USA. Spectrovac 1000 Baird-Atomic Ltd., Spectromobile Warner Drive, MS 2 Springwood Industrial Estate, FAS 2 Braintree, Essex FAS 2 GT CM77YL, England FAS 2 C 06 or 0.3 210-590 1 .O Arc or spark; Compact, low-cost direct reader with minimum air conditioning requirements; manual master modular monitor tocheck slit alignment; 30channels 0.6 or 0.3 173-767 1 .O Arc or spark; As Spectromet 1000; logarithmic read-out; dual stand option; 50 channels 0.3 or 06 200-1300 1 .O Low voltage Mobile spectrometer for metal sorting or checking; 24channels 0.6 or 0.3 21 0-590 1 .O Rotrode Three fluid analysis spectrometers for SOAP programme.No special environmental facilities modular 3 step d.c. arc and 766.4 required; readout ranges from 0-99.8 and 0-999.9 or 0-999-8 and 0-9998 depending on instrument; manual readout, typewriter or computer interface for most minicomputers Jarrel-Ash Division,‘ 78-090 1.1 or044 420-970 1.5 Fisher Scientific Co., 21 0-485 590 Lincoln Street, 70-130 1.0or0.2 180-3000 3.4 Walthamm MA 02254, depending on 180-1 500 U.S.A.grating 180-750 96-750 0.54 168-500 0.75 9S785 0.54 168-500 0.75 Various available in varisource unit. Includes: spark, low and high vo1taged.c.arcs. Also versatile controlled wave excitation source As above except electronically controlled peak current Wadsworth Spectrograph; 20 in camera; choice of three gratings; N2 purging extends range to 175 nm; optional accessories permit use as direct reader or scanning spectrophotometer Direct reader, computer cdntrolledTable 2.6A COMMERCIALLY AVAILABLE EMISSION SPECTROMETERS - continued Reciprocal Focal nmpermm rangelnm m Supplier Model dispersion/ Wavelength length/ Typeof Source Spacial features (continued) 1500 0.56 or 0.28 200-800 or 1 4 As above 190-400 0.34 or 0.1 7 2ocb510 or Choice of two gratings 190-250 Labtest Equipment Co.’ 31 0 0.56 190-900 1.5 ‘Transource’ high Direct reader; wavelength in first order; CRT; teletype printer or computer read out systems; Ltd., 11828LaGrange V25 0-46 170-450 1 .O voltage trig ered dual airhnert gas and solution excitation stand; V25 vacuum for C and S in ferrous materials Avenue, 2100 0.46 188-455 1 .O discharge.pow Los Angeles, CA 90025, 71 0.52 170-900 2.0 voltage triggered U.S.A. d.c. arc; ICP source for solution analysis Optical Emission Services OES 4500 0.45 or 0.225 170450 1 .O Arc/spark Direct reader.Additional monochromator for extra flexibility, autosampler, built-in VDU, Ltd., 104Tanners Drive, ICP possibility of photoncounting, printer 120 cps Blakelands, MiltonKe nes MK145B&, Eigland Philips lndustrie S.A., PV8020/01 t 0.46 177-41 0 1 .O Monoalternance Computerized emission spectrometric system fitted with 20 preselected lines for steel and Spectrometry Department, 50 Hz spark with iron programmes 131 Boulevard de I’Europe, 6-1301 WAVRE, condition Belgium high energy Philips Analytical PV8350 0.46 177-410 Department, Vacuum Pye-Unicam Ltd., York Street, Cambridge, CB12PX, England PV8250 0.69 190-615 Air 0.59 190-531 0.92 190-820 0.46 190410 PV8210 0.55 190-700 Air 1 .O Monoaltemance spark, d.c.arc, glow-dischar e hollow cathde or ICP 1 .O As for PV8350 1.5 AsforPV8350 Rank-Hilger Ltd..E l 000 0.293-1.1 55 15E-880 Westwood, Margate, Polyvac Kent CT94JL, England 1.5 Various, including high repetition condensed arc, ICP, GUL Integrated spectrometer system with optional dual air/Ar excitation stand; choice of programmable calculator or computer; configurations with dual cassettes or floppy discs; rapid printer; VDU extension options b 0 % b 3 E As for PV8350 3.As for PV8350 Direct reader; solid state electronics; microprocessor control available. Dual gratings give 12 standard systems to select optimum dispersion and wavelength coverage. Special grating if required; dual spark stand 2. m ah a E960 0546or 174.0-447.7 0.75 As El000 Curved entrance and exit slits; microprocessor control available; air or inert gasdischarge 5 0.741 stands s ‘No up-to-date information supplied tNew equipment since publication of Volume 10Spectrametrics Inc., AEZ 0.06 190-800 0.75 Plasmajet Optimized AE system using a high dispersion, high-energy throughout echelle 204 Andover Street, Andover, MA01810, DRlO 0.06 190-800 0.75 Plasmajet U.S.A.spectrophotometer and a high temperature plasma jet excitation source Techmation Ltd., ES9 0.06 190-800 0.75 Plasmajet, flame Built in computer 58 Edgware Way, Edgware, Middlesex, HA8 UP, England RS1 0.06 190-800 0.75 AsES9 One channel, variable wavelength or arc stand Siemens AG, Instrumentation and Control Division, Analytical Systems b t l .Rheinbriicken$tr. 50, D75OO Karlsruhe, West Germany VEB Carl Zeiss Jena, 69 Jena, Carl-Zeiss Str. 1, German Democratic Re ublic t a d Zeiss Scientific Instruments Ltd., P.O. Box43, 2 Elstree Way, Boreham Wood, Herts.WD6 1 NH, England Spectrumatt 0.78 220-750 1 .O Floating anode Direct reader, microcomputer controlled, maximum of 63channels Spectrumatt 0.36 150-450 1.0 Asabove As above, optional special channels for lines above 450 nm Spectrumatt 0.36 110-450 1.0 Asabove As above, special computer and software system for surface analysis (depth profile analysis) lOOOA glow-discharge 1 ooov 1 OOOHV available PGS2 0.74 or 0.37 200-2800 2.075 Arc or spark Atlas for spectra evaluation; wide choice of precision diffraction gratings; high resolving power; dispersion doubling or multiplying as required; automatic transport of cassette, wavelength scale for quick orientation of the user within the spectra; wide range of accessories available including M A 10 laser- microspectral analyser tNew equipment since publication of Volume 10Table 2.6B COMMERCIALLY AVAILABLE PLASMA SPECTROMETERS Generator Supplier Model Reciprocal Focal Output Operating Special features dispersion/ length/ power/ frequency/ nmpermm m kW MHz __ Applied Research$ Quantometer 0.9304465 1 .O 2.5 27.1 2 Paschen-Runge direct reading spectrometer.Full computer control to provide direct Laboratorits Ltd. 34000 ICP concentration print-out; full range of options including dual floppy/hard discs, VDUs, fast printers, remote terminals and computer links, etc. 60 channels measured simultaneously. Useable wavelength range of 17-20 nm.wavelength profile. Electronics provide qualitative analysis on chosen spectrum line. Concentration printout with options for: dual floppylhard discs, fast printers, VDUs, remote terminals. Useable wavelength range of 175-850 nm. Quantometer 0.80 1.0 2.5 27.1 2 Czerny-Turner optical mount with computer controlled scanning of grating for selected 35000 ICP Baird Corporation$ PlasmdAFSt 1 .o 40 Multi-element AFS spectrometer with plasma atom cell.Offers multielement capabilities of ICP-OES with the simplicity and specificity of AAS with almost no spectral or matrix interferences. For more detailed information see Section 2.5 3 Plasma 0.65 1.0 2.5 27.1 2 One metre polychromator with 120 exit slits in a rigid focal curve.Data acquisition is Spectromet 0.33 controlled by a Tektronix 4052 graphic computing system. Optional integral scanning 0.22 monochromator Instrumentations Plasma 100 Laboratory Inc. 2.5 27.1 2 Microcomputer controlled scanning double monochromator for sequential multi-element analysis; instructions for programming appear on video display with single keystroke operation; emission profiles of analytical line appear on video screen for selection of wavelength, for background correction, for study of inter-element effects and observation of spectral interferences: all circuitry for r.f.power generation, monochromator optics, and microcomputer are built into the instrument. Ebert (double) monochromator, crossflow nebulizer Jarrell-Ash Division, * $ 96.975 Fisher Scientific CO.Ltd. 96.988 Jobin-Yvon, Division d'lnstruments, JY38P 16-18RueduCana1, 91 160 Longjumeau, France EDT Research, JY48P 14 Tradin Estate Road London NhlO 7LU, England .- (continued) 'No up-to-date information supplied 0.54 0.75 2.0 27.1 2 Computer controlled direct reader; variable channel; concentration print-out 0.54 0.75 - - Computer control; N + 1 channel scanning attachment, spectrum shifter attachment for automatic background correction; special K and Li channels; data management system Czerny-Turner monochromator, large aperture monochromator (f 5.4, grating size 120 x 140 mrn); manual or computer controlled; constant time integration or in ratio mode; choice of concentric glass nebulizer, adjustable concentric nebu!izer in zirconium, or ultrasonic i? ____ Plasmatherm 1 .o 1.5 2.5 a- ii 27.12 5 nebulizer b 039 1 0 0.46 2 2 56 computer option choice of two standard gratings (or specials if necessa ) 1800 grooves/mm 3' 0 55 4 0 82 $ Durr-Jobin-Yvon Air orvacuum spectrometer, 86 positions of photomultipliers, fully automatic read-out (wavelength range 180-590 nm, reciprocal dispersion 0.55 nm/mm), or%OO grooves/mm (wavelength range 130-41 5 nm, reciprocal dispersion 0 39 nm/mm) scanning entrance slit under computer control for identification of iriterfering spectral lines, background correction, a 2 a $Address as in Table 2 6A tNew equipment since publication of Volume 10 $Address as in Table 2.6C 2(continued) JY32Pt 0.55 0.6 2.2 Kontron GmbH.Postfach ICP 1000/2000 0.6 8057, Eching bei Munchen, Oskar-von-Muller Str. 1, ASS 80 West Germany. ICP 1000/2000 0.25 ES 750 0.6 1.5 4 0.75 1.5 4 and for analysis of elements not installed in the programme, Paschen-Runge monochromator; choice of nebulizers as for JY38P DurrJobin-Yvon Air or Ar purged spectrometer, 50 positions of photomultipliers covering the wavelength range 170-800 nm; full computer controlled system including background correction by scanning primary slit. Gratin is 3600 grooveslmm master holographic.Flying channel monochromator easily incorporatejfor elements not in the basic programme. Can be combined with monochromator and detection system of JY38P; 52 photomultiplier positionsof which any 32 can be measured simultaneously; software based on a Silex microcomputer. Full range of nebulizers and torches including demountable torch as for JY38P Scanning Czerny-Turner monochromator; wavelength range 200-500 nm; Kontron KDT CPU microcomputer; alphanumerical printer, optional printer/plotter Direct reader.Paschen-Runge polychromator; wavelength range 165-500 nm or 165-220 nm; 48 elements 56 27.1 2 27.1 2 Labtest Equipment Co. * * Plasmascan 700 0.35 2 27.1 2 Czerny-Turner monochromator; microprocessor control; enclosed sample pumping system; computer readout system; crossflow, concentric glass or ultrasonic nebulizer Perkin-Elmer Corp.5 ICP 5500 U.V. 0.65 vis. 1.3 0.4 2.5 27.1 2 Completely automated sequential ICP system which can determine up to 80 elements in an operator-selected multi-element programme; analytical parameters including wavelength selection background correction interval and signal handlin are programmable via model 3600 Datb Station included in the system; optical path purgafile permitting analyses to 175 nm; software permits development of analytical methods in Develop mode and operation of system in Analyze mode; five report formats are available at the choice of the operator, including reports with statistics and volume and weight correction if desired.Analytical speed up to 15 elements per minute can be selected by the operator. Maximum precision of 1 % relative is claimed at an analytical speed of 5 elements per minute. Demountable torch usin precision borequartz tubing and alumina injector is standard. Cross-flow nebulizer made o? corrosion-resistant material and corrosion-resistant spray chamber are also standard Philips $ PV8210 Air 0.55 or 0.28 1.5 2 50 Direct reader.Paschen-Runge spectrometer; wavelength rangecovered in 1 st order; remote controlled rovin detector; readout by printer, teletype or digital computer systems. Crossflow nebuf zer PV8250 Air 0.69 or 0.35 0.59 or 0.35 0.92 or 0.46 0.46 or 0.23 PV8350 0.46 Vacuum Rank-Hilger Ltd.* EIOOO 0.293-1.1 55 polyvac E960 0~5464.741 1 .o 2 50 Integrated spectrometer system with built-in source and readout options; spectrometer and nebulizer as for PV8210 1 .o 2 50 Integrated spectrometer system including source and read out options; spectrometer and nebulizer as for PV8210 1.5 27.1 2 Paschen-Rungedirect reading Spectrometer; solid stateelectronics; dual gratingsgive 12 standard systems to select optimum dispersion and wavelength coverage; special grating if required; dual spark stands; microprocessor control available; crossflow or concentric glass nebulizer 0.75 27.1 2 Paschen-Rungedirect reading Spectrometer; curved entrance slits; microprocessor control available; crossflow or concentric glass nebulizer 0 -4 ___.m Table 2.6B COMMERCIALLY AVAILABLE PLASMA SPECTROMETERS - continued 00 Generator Supplier Model Reciprocal Focal Output Operating Special features dispersion1 length/ power/ frequency1 nmpermm m kW MHz Spectramterics 1nc.S Spectraspan 0.06 0.75 Ill6 at 200 mm D.c.Optimized AE system using a high-dispersion, high energy throughput echelle (ICP power and spectrometer. Both d.c. plasma and ICP available; microprocessor controlled, background frequency not corrected modular system capable of expansion for sequential and multi-element reported.) simultaneousoperation Spectraspan As Ill6 As Ill6 As Ill6 Sequential instrument using echelle spectrometer.IV Spex Industries Inc., 1870 3880 ParkAvenue, Metuchen, 1702 NJ 08840, USA. 1704 1269 Glen Creston Instruments Ltd., 16 Dalston Gardens, Stanrnore.Middx. HA7 1 DH, England 1.6 0.5 1.1 0.75 0.8 1.0 0.65 1.26 ‘No up-to-date information supplied tNew equipment since publication of Volume 10 *Address as in Table 2.6A §Address as in Table 2.6C Plasma- Czerny-Turner spectrographhonochromator Therm Slew-scan monochromator 27.12 As1702 - - High Resolution slew-scan monochromatorTable 2.6C COMME.RCIAL,LY AVAILABLE ATOMIC ABSORPTION SPECTROMETERS ________________ Supplier Model Resolution Type of Data (singleldouble) nm output ___ Special features Baird Corporation$ Alpha 1 0.1 Bit parallel (single) BCD (TTL levels) Baird-Atomic Ltd $ Alpha 2 Alpha3 Alpha 4 Alpha computer systems __ _______ - _ _ ~ GBC Scientific Equipment G BC 0.5 IEEE- Pty.Ltd., 7/63 Park Drive, SB900 488 Dandenong, Victoria 31 75, (single) Australia EDT Research, GBC901 0 1 IEEE- 14Trading Estate, (single) 488 London NWlO 7LU, England Single lamp turret; fail-safe gas safety; digital concentration readout As Alpha 1 plus4-position lamp turret As Alpha 1 plus automatic D2 HCL background correction and 2-speed wavelength scanning As Alpha 3 plus 4-position lamp turret Colour or monochrome video display, unlimited curve and report storage on floppy disc; printer option 2 lamp supply; optional background correction; hydride generation and calculator available; length 700 mm, width 200 mm, height 225 mm 2 integration times, all reflective optical system, flame emission, peak height, absorption expansion to 30 times, direct concentration readout, Ti burner, safe burnerigas control, system, 2 lamp supply, 3 point curve control, background correction; graphite furnace and hydride generator available Hitachi Ltd , Nissei Sangyo Co Ltd , Scientific Instruments, International Department, Mori 17th Building, 26-5Toranomon 1 -chome, Minato-ku, Tokyo 105, Japan Nissei Sangyo America Ltd 460E Middlefield Road, Mountain View, California 94043 170-1 0 (single) (single) (double) (single) 170-30 170-50 180-307 0.4 - 0.4 - 0.1 - 1.2 - Also New York, Chicago, Nissei SangyoGMBH, Ross-Strasse 74, 4000 Dusseldorf 30, West Germany 180-60t 1 2 - (single) Nissei Sangyo Co Ltd Sutton Industrial Park, (single) Reading, Berkshire (single) Washington D C 180-50t 1 2 -- (single) London Road, 180-70t 1 2 - RG6 1 AZ, England 18&80t 1 2 - Single lamp mounting, Ndair simultaneously exchanged; concentration readout; continuously variable time constant Concentration read-out, time weighted signal averaging, AAS/AFS measurement, auto zero, N20/air simultaneously exchanged Base-line drift correction; curve corrector, time weighted signal averaging, auto zero 4-lamp turret water cooled premix burner Air/C2H2 or N20/C2H2 selection by synchronized valve system, (C& automatically increased when using N20) AA with D2 background correction and emission measurement direct, (0 5-1 6 s continuously variable) integration, absorbance linear Meter display GA-26 graphite analyser, auto-measuring system, As and Hg analyser available 4-lamp turret water cooled premix burner, Air/C2H2 or N20)/C2H2 selection by synchronised valve system, (C2H2 automatically increased when using N20) AA with D2 background correction and emission measurement direct, integration, peak height, peak area, absorbance and concentration 4-point calibration, auto zero, CRT LED and printer display microcomputer GA3 graphite analyser, automeasuring system, As and Hg analyser available Polarized Zeeman AA and emission Polarized Zeeman AAgraphite furnace and emission pro ramrnable 7 steps and ramp temp range 0-3000 "C measurement direct.peak height, peak area t!onventional flame burner without Zeeman effect Fully equipped instrument incorporating Zeeman AA Graphite Furnace and Zeeman flame system furnace programmable, 7 steps and ramp, temp range 0-3000 'C measurement direct. peak height, peak area Vido terminal on all models, complete data processing computer assures precise control of selected operating parameters Auto process As analyser available, Hg analysers available Q\ \DTable 2.6C COMMERCIALLY AVAILABLE ATOMIC ABSORPTION SPECTROMETERS - continued 4 0 Supplier Model Resolution Type of Data (single/double) nm output Special features Instrumentation Laboratory Inc., 157 0.04 - 68 Jonspin Road, Wilmington (single) MA01887.U.S.A. Microcomputer controlled, calibration curve linearized using two standards. Fully automated gas box is standard feature, optional D2 arc background correction, optional 4-lamp turret and wavelength scan 357 0.04 RS232C Asfor lL457 Instrumentation Laboratory (single) (UK) Ltd., Kelvin Close, Birchwood 451 0.04 RS232C As for IL 551.CRTvideo readout; D2 arc background correction Science Park, Warrington, (single) Cheshire. England (double) 457 0.04 RS232C Microcomputer controlled; calibration curve linearized using up to 5 standards; provides full statistics on results; fully automated gas box is standard feature; optional D2 arc background correction: 4-lamp turret: wavelength scan and alphanumeric printer 55 1 0.04 RS232C (double) 951 CO4 RS232C (double; dual channel) Microcomputer controlled; calibration curve linearized using up to 5 standards; memory will store up to 10 calibration curves simultaneously; VDU displays standard conditions for each element, the working curve and will show transient signals; fully automated fail-safe gas box is standard feature; optional background correction, 4-lamp turret, wavelength scan and alphanumeric printer Microcomputer controlled; calibration curve linearized in both channels using up to 5 standards; CRT video readout, will display 2 elements simultaneously A, B.NBor A-B; internal standard and non-absorbing line background correction; VDU displays standard signals; automatic gas box is standard feature; optional 4-lamp turret; wavelength scan and built-in alphanumeric printer Perkin-Elmer Corporation, 2280 0.2 EIA-RS 232C Main Avenue, Norwalk, (single) CA06856, U.S.A.High energy optical s stem, microprocessor controlled; auto-zero; auto concn.; auto curve with up to 3 standards; peak heigKt; eak area; integration time selectable from 0.2 to 60 s; statistics; flame ignition optional; auto N20 switckng and burner head safety interlocks; optional flame and pressure sensing by microcomputer burner control.D2 arc background correction optional burner head safety interlocks; optional D2arc background correction with automatic intensity control Bodensewerk, Perkin-Elmer 2380 0.2 EIA-RS 232C As model 2280 but all mirror optics: automaticgain control; auto N20 switching; &Co.,GmbH, Postfach 1120, (double) D7770 Uberlinaen. West Germany k 4000 0.3 2wa Semi-automated sequential AA system; automatic gain control; instrument can analyse up to 6 elements with little operator participation; analytical parameters including standardization and signal read-out can beentered and stored internal1y;digital stepper motor wavelength selection; flame ignition, auto N20 switchover, burner head interlocks: optional flame and pressure sensing by microcomputer, burner control; optional double beam background correction for all U.V.and visible wavelengths with automatic intensity (double) EIA-kS 232C 5 control; lamp turret available k 5000 0.3 2way (double) EIA-RS232C (continued) Completely automated sequential AA system; instrument can analyse up to 6 elements with minimal operator participation; all analytical parameters including lamp current, wavelength selection, resolution, double beam background for all U.V.and visible wavelengths; when used in conjunction with HGA 500 it will provide sequential analysis for up to 6 elements with the same analytical ease as flame: when used with ICP-emission accessory it will provide sequential multi-element analysis for up to 20 elements with operator selectable background parameters k fi' gas flows, standardization and signal readout can be entered and stored using magnetic cards; optional 3 3 * Address as in Table 2.6A t New equipment as in Table 2.4A(continued) Zeeman 5000 (double) 3030t (double) 0.3 2wa 0.07 2wa Flame operation both with and without background correction as for model 5000.The instrument can be equipped simultaneously for Zeeman-corrected graphite furnace AA Contains built-in CRT, software including all cook book information; optional graphite display facilitates instrument set up; graphics are available for flame, furnace or MHSsystem and may be printed on an external printer.Exclusive soft keys change function as operator steps through a programme facilitating instrument set-up. Four lamp turret standard, double-beam background correction available; interlocked gas control system available. Automatic calibration with up to 8 standards is built in; integration time selectable from 0.2 to 99 s. Operator may select averaging and RSD if desired. Software resident on floppy disc for easy updating and available in five languages.Quick change mount available for easy change from flame to graphite furnace EIA-hS 232C EIA-hS232C Pye-Unicam Ltd.. SP 9 York Street, Cambridge, (single) CB12PX. England SP 9 Computer PU 9090 PU 9000t PU 9007t AA datdcontrol system Rank-Hilger Ltd., Atomspek Westwood, Margate, Kent H 1580 CT94JL, England (single) 0.2 RS 232 C 0.2 2-way RS 232C 8 modules available with combinations of 4-lamp turret; auto-gas control module with full safety interlocks; scale expansion; 2-standard curvature correction; burner interlock; output for SP 9 Computer and PU 9090 Data Graphics System and 0-10 mV (0-1 A) analogue output as standard Microprocessor data processing and control of flame automatic system for SP 9.Curve correction with 5 standards in fixed and variable ratios. Peak height and/or peak area, full statistics, running mean, error warnings, built-in self-test routines. Integration and peak read times 0.1-1 00 s As SP 9Com uter plus with PU 9095 video dis lay of flame and furnace cook book calibration curves and transient pea! prohles with automatic scaling.&intout of furnace parameters. Intelligent, fully automatic, multi-element microprocessor controlled AA s stem Capable of selecting and then optimisin appropriate instrumental conditions for each element admode of operation. Electronically coded HCLs aiow the instrument to detect element and maximise current. Novel double-beam system eliminates signal-to-noise degradation of conventional double-beam systems.A double-beam reference channel checks and corrects the baseline prior to every measurement. The double-beam optics are then automatically switched out of the beam during the measurement period itself. Master holographic grating. The intelli ent system will select and set-up wavelen th band pass lamp current flame type gas flow and read timegor each element and Sam les All of this inkrmation iscohtained in the memory s .&em and the operator is not required to program tie instrument.The conditions are then modified, e.g., aiernative wavelength selected automaticall in response to the sample signal Gas control is via a binary flow switching system with feedback orabsorbance signal to automaticaily set the optimum fuel flow. The instrument detects the presence of the burner or furnace and switches programs automatically.Up to 8 elements are sequenced automatically for flame or furnace analysis; aflame autosampler is built into the instrument. A universal burner is used for all flames. Built-in printer, 4 controllable formats for multi-element reports, prints out full conditions, programs, error messages and provides hard copy of the video display, including graphics, from the PU 9007AA datdcontrol station Automatic control of furnace programs from the PU 9095 in multi-element furnace analysis.Interfaces with the PU 9000. Microcomputer system with 12"video display, QWERTY keyboard and twin floppy disc drives. High resolution graphics, user programming in all common hi h level languages. Comprehensive software packa e for flame and furnace work.Menu driven selection o?all parameters for multi-element analysis runs. fatal control of the PU 9000. Video display of graphics including transient peaks, calibration curves, ash/atomizecurves. Comprehensive post-run result cornputin andformatting..Floppy disc storage of all programs and data. Any videodisplay, including graphics may ge dumped on the PU 9000 printer 6-lamp turret; autozero and flame ignition; curve correction; integration; programmable calculator.Printer available 0.2 - Shimadzu-Seisakusho Ltd., AA 625 14-5 Uchikanda, 1 Chome, (single) Chiyoda-Ku, Tokyo 101, Japan AA 630 (continued) (single) 0.2 - 0.2 - Quantitative flameemission; flameless capacity, flow lines for air, C&and N 2 0 Quantitative/qualitative flame emission; flameless capacity, flow lines for air, C2H2 and NP; flame monitor; gas pressure monitor: wavelength drive 2Table 2.6C COMMERCIALLY AVAILABLE ATOMIC ABSORPTION SPECTROMETERS - continued 4 N - ~ ~ ~ ~ ~ ~ ~~~~~~~~~~ ~~~ Supplier Model Resolution Type of Data Special features (single/double) nm Output _ _ _ _ _ ~ _ _ _ _ _ _ _______ V.A. Howe & Co. Ltd., AA 640 0.2 - 88 Peterborough Road, (single) London SW63EP, England Automatic background correction; quantitative and qualitative flame emission; flameless capacity: flow lines for air, C2H2 and N20; flame monitor; wavelength drive, integration Varian-Techtron Pty. Ltd., AA 1275 679/701 Springvale Road, (single) Mulgrave, Victoria 31 70, Australia Varian Associates Ltd., 28 Mano Road, Walton-on- Thames, Surrey KT1 2 2QF, England Genesis Centre, (double) Birchwood Science Park South Birchwood, Warrington, Cheshire WA37BM, England Varian Associates Ltd., AA-975t Varian Instruments Division, 61 1 Hansen Way, Palo Alto, CA94303, USA.VEB Carl Zeiss Jena, 69, Jena, Carl Zeiss Str. 1, German Democratic Republic AAS 1 (single) AASlN Carl Zeiss Scientific (single) Instruments Ltd , (N20 P 0 Box43,2 Elstree Way, WD6 1 NH England equipment) 0.2 IEEE-488 RS 232C and parallel BCD Two-lamp turret; overcoated reflective optics; automatic gas control system; compatible with samplers, printer, hydride and furnace atomization systems.Intel 8080 with 1 OK ROM provides signal processing; background correction, absorbance conversion, integration: 3 standard curve fitting; peak height; peak area measurement; lampcurrent control.O2 arc background correction; new integrated high sensitivity atomization system 0.2 IEEE-488 Stores up to 100 sets of operating parameterson in-built floppy-disc. Provides ca ability for automatic sequential anal sisof up to 12 elements by AAor emission using the compatible !SC 55 flame sample changer or the &TA-95 furnace and sample dispenser.Optics are all reflective and hard-dielectric coated; twelve lamp turret; real time signal and result processing by an 8-bit microprocessorwith 24K memory which also controls lamp supplies, turret, photomultiplier supply, monochromator, floppy-disc and the data base. Four modes of integration; operator controlled single reading; accessory controlled multiple reading; automatic continuous reading: operator controlled running mean.Peak detector measures peak height and area in absorbance, concentration and emission. Direct concentration read out using a blank and 3 standards: curve fitting by rational function algorithm. Calibration reslope. Fully programmable gas control 100 mV (600 ohms) for 4-lamp turret; single or triple pass optics; autozero; titanium burner heads; flow lines for air, C2H2, N20; gas potentiometric recorder pressure monitor; gas flow monitor; burner head safety interlock; automatic flame ignition or absorbance converter.- - TEC1 printer or computer, VlDTECl ; signal output 775 mV (5 kohms) for linear recording or absorbance b b No up-to-date information supplied t New equipment since publication of Volume 10 $ Address as in Table 2.6ATable 2.6D COMMERCIALLY AVAILABLE ELECTROTHERMAL ATOMIZERS AND AUTOSAMPLERS Supplier Model Type Control unit Ramp rate range Special features Baird Corporation* A1 70 Graphite rod Programmable, dry, ash (2 stages), atomize; max.temp. 3500°C - Fits most AA spectrometers; air cooled; uses mains power; inert gas shielding; pyrolytic graphite coating for rods in situ; rapid interchange between flame and electrothermal methods GBC Scientific5 GF9OOt Graphite 4 temperature cycles Equipment Pty.Ltd. furnace (Dry, Ash 1, Ash 2, atomize), temperature ramp capabilities on each cycle - Continuousdigital temperature read-out to 30OO0C, water, gas and electrical safety interlocks automatic switch off of inert gas or H2 H2 sheathin available at push of button; p&h button program advance accelerates proper cycfe for cleaning or aborting run Hitachi Ltd.§ GA2Bt Tubeor Cup Programmable, - Manual programme change.Applicable to Hitachi models 180-30 170/10/30/50A. end pointtemp., time. Dry up to 300 "C 0-1 80 s. Ash up to 2000°C 0-1 20 s.Atomize up to 2800°C 0-30 s. max. temp. 3000"C, time0-99s Measuring modes, direct, peak hold GA3t Tube or Cup Programmable, Up to 7 steps Write into CRTscreen via keyboard programme change: measuring modes, direct, peak hold, peak area. Applicable to models 180-50,170-40 Autosampler Up to 31 samples, 10,20,30,40 pI sample injection volume: automatic stop on detection of gas or water pressure drop or abnormal furnace temp.Instrumentations 655 Laboratory Inc. 254 ___ Perkin-Elmer§ HGA 400 Corporation Bodenseewerk, Perkin-Elmer Co., GmbH. A540 HGA 500 Graphite furnace step for auto-zeroing and Autosampler Digital timers for sample for flame or furnace operation spectrometer Programmable, 6 stages; ramp or - auto calibrating the spectrometer boats deposition, triggercircuitry for auto zeroing and auto calibrating the True on temperature read-out; LED display; safety interlock system; automatic cell door; automatic cleaning; cell pressurization; convenient solid sampling capacity using micro Flamelfurnace autosampling technique (FASTAC) with auto calibration; employs an aerosol deposition technique of introducing aerosol into cuvette which is at elevated temperature; the sample volume, which evaporates on contact with graphite surface, is controlled by length of time sample is sprayed into furnace, allowing operator to control sensitivity by varying deposition time 1-99 s - Graphite furnace Microprocessor unit provides up to 8 steps of controlled heating; temp., ramptime, hold time, gas and other furnace and spectrometer control functions are programmed by direct keyboard entry; digital displays provide readout of temp., time and prog.status From 2000 "C s-' High speed temperature accessory permits rapid heating to temperature between 800 to 999 s between and 3000 "C for optimal atomization anytwotemps. Autosampler - - Automatic insertion of up to 35 samples and blank and 35 standards into the furnace; will also perform automatic method of additions; automatic matrix modification; recalibration; automatic triggering of furnace and instrument read cycle for unattended operation Furnace control programs for up to 6 different elements may be stored in 6 program memories; program parameters for more than 6 elements can be stored on magnetic cards and recalled at the touch of one button; the optical temperature sensor and digital gas flow control for 2 different purge gases add to the versatility of the furnace program; when used in combination with the A540 microcomputer furnace autosampler and the Graphite Microprocessor unit provides furnace As for HGA 400 up to 9 stepsofcontrokd heating; temp, ramp time, gas and other furnace and spectrometer control functions are programmed forTable 2.6D COMMERCIALLY AVAILABLE ELECTROTHERMAL ATOMIZERS AND AUTOSAMPLERS Suppller Model Type Control unit Ramp rate range Speclal features ______ each step by direct keyboard ent digital displays provide readout 07' temp, time and program status; up to 6 comp.prog. can be stored and recalled at the touch of 1 key model 5000 AA, up to 35 samples, blank and 3 standards may be analysed for up to 6 elements each without operator attention Pye-Unicam Ltd.§ PU 9095 Graphite Microprocessor control of six video furnace phases and ramps to3000 "C. furnace Voltage or temp. control, no adjustment of photodiode sensor necessary SP 9 Graphite 4-phases, each programmable to furnace furnace 3000°C. Voltage ortemperature control, no adjustment of photodiode sensor SP 9 - Autosampler sampling of 38 f u rnace samples andtwo wash positions. autosampler Selectable number of readingsand vol. for each sample; pos. ident. of blanks, stand. and samples 18 ramp rates, Microprocessor selection and control of all functions includin built-in autosampler 9 linear controls, video displays of parameters and status. Non-volatfe storage of 10 programs. 2-2000°C s-', Gas stop and recorder control on all phases. Video display of peak shapes, calibration and 9 exponential and cookbook conditions when used with PU 9090 Data Graphics System; fits all current Pye-Unicam spectrometers 9 ramp rates, Digital parameter selection, comprehensive status and fault indication. Fits all 2-2000 "C s-l Pye-Unicam spectrometers Rank-Hilger Ltd.$ H 1475 Graphite Programmable; dry, ash, wait, - Water cooled. inert gas shielding furnace atomize; max. temp. 2600°C Scintrex Ltd., AAZ-2t Tungsten Infinitely variable To 2500 "C in Compact instrument which provides high sensitivity comparable to other more 222. Snidercroft Road. StHD drvina-ashina-atomizina cvcle. in rates UD to soDhisticated AA's. Zeeman modulation provides Superior background correction Yet iS Concord, Ontario, Canada LUK 1 B5 Techmation LtdA atomizer with Indicitors foiZeeman mode. Zeeman Arflow, cooling period, PMT background overload, etc., oscilloscope or surveys correction transient recorder output 4000 "C s-' achieved uniquely without moving parts. Does not require watercooling, low power and gas requirement make it suitable under field conditions for geological or environmental Shimadzu-Seisakusho'§ GFAZ Graphite Programmable, do ash, atomize; - Current stabilized to obtain reproducible results furnace max. temp. 3000 Varian-Techtron Pty.5 GTA-95t Graphite Programmable. Temp. Range 0--2000°C s-' Ltd. tube furnace 20-3000 "C, up to 20 temp. steps. Programmable heating rate to Heat injection from 40-1 50 "C 2000 "C s-: . PSC55t sample changer 'No up to date information supplied. tNew equipment since publication of Volume 10 Designed primarily for Varian '75 series spectrometers Graphite tube isolated in enclosed cell VDU gives tables for programming, graphical display of temperatureltime profile with automatic ranging and instantaneous numerical display Operator can select the number of steps displayed Analytical signal superimposed on temperature profile Optional programmable sample dispenser provides blank, 5 standards, 45 samples and chemical modifier Programmable volume from 2-70 &I in 1 &I steps, 4 solutions can be dispensed together Up to 99 multiple injections before atomization, up to 99 replicates on each sample Microprocessor controlled autosampler for flame determinations Sampler has 5 standard and 67 sample positions Keyboard allows programme to be loaded directly into the sample changer Interfaces with all current Varian AA spectrometers ~~ ~ §Address as in Table 2.6C *Address as in Table 2.6C

ISSN:0306-1353    

DOI:10.1039/AA9811100049

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

CHAPTER 3 Methodology 3.1 NEW METHODS This Section describes novel methods of analysis that are considered to be of sufficient general interest to merit discussion here as well as in the appropriate section on specific applications. It also includes consideration of papers by workers who have made a detailed study of experimental parameters of widespread relevance. The aim of this Section is to describe new procedures and methods of analysis that could be of direct relevance to the routine analytical laboratory. 3.1.1 Sample Preparation Techniques 3. I . I . I Sample Preservation, Solubilization, Incomplete and Unusual Dissolution Procedures Considerable effort is often expended in the development of routine ultra-trace metal analysis methods. Unfortunately in many instances, not enough consideration is given to the collection and the preservation of samplesprior to analysis.(see ARAAS, 1979, 9, 59). A common problem is the method of adding the preserving acid to the samples at the time of sampling. Two often conflicting considerations are minimization of sample contamination and minimization of risk to the sampling personnel from the mineral acid used for sample preservation.Guest and Blutstein (2155) solved this problem by placing a clean sealed borosilicate glass ampoule containing redistilled HNO into each sample bottle. The ampoule and inside of each bottle were rinsed several times with the sample, the empty capped bottle was shaken in order to break the ampoule and finally the bottle was filled with the sample and capped.During a study of 13 pre-cleaning procedures for polyethylene bottles (2156), a 48 h soak in 10% WVHN03 was found to be the optimum. After this treatment samples containing 0.5% V/Vhigh purity HNO, could be stored for 28 days with final Cd, Cu, Pb and Zn concentrations of < 0.02,O. 1, < 0.1 and 0.1 ng ml-' , respectively. A number of methods for simplifying sample preparation in order to minimize both preparation time and reagent blanks have been reported.Fuller et al. (1293) compared the use of FAAS, ETA-AAS and ICP-OES for the direct analysis of slurries. Pulse nebulization using a conventional pneumatic nebulizer was recommended for FAAS whilst a slot type nebulizer was essential for use with an ICP. In both instances it was necessary to grind samples to < lOpm, and to match standards and samples both physically and chemically.For ETA work, particle size was only significant at sizes > 25 pm, this being attributed to sampling error. Other workers (1022) have ground dried vegetation to a fine powder which was then slurried with deionized water. An aliquot of the slurry was pipetted into a Delves cup for the determination of Pb.The rate of heating of the cup in the air/C,H, flame was such that the Pb atomic peak was resolved from the scatter peak. Hence, automatic background correction was not required. In a novel slurry-ETA method for Se in nutritional supplements (668), the dried ground sample was suspended in an acidified solution containing 2% m/V Ni(NO,), and mixed non-ionic surfactants.A similar pre-treatment was used to emulsify milk products and was applicable to the determination of As, Hg and Pb. The use of emulsions and aqueous based standards in FAAS analysis continues (see ARAAS, 1980, 10, 87). Copper, Fe and Zn have been determined in milk (175) and milk powder (1473) after formation of stable emulsions using various mixed detergents. Hernandez-Mendez et al.(910) have published their work on the direct FAAS determination 7576 Analytical Atomic Spectroscopy of Zn in ointments using water emulsions. Direct analyses of lubricating oil in water emulsions for Ca (539) and Ba (538) as well as petroleum spirit emulsions for Pb (538) have been reported, aqueous based standards being used for calibration. Enzymatic digestion of biological tissue would appear to have considerable potential.Since only very small quantities of enzyme (typically-0.1070 sample weight) are required to digest a tissue sample. Using this technique Carpenter (1 348) determined Cd, Cu, Pb and T1 in liver and kidney samples. The tissue sample (5 g) was blended with 20 ml of 1~ tris and 10 ml water. Crystalline subtilisin A (5 mg) was then added and the sample was incubated at 55 "C for 1 h with continuous shaking.The solution was then filtered, diluted to 50 ml and the elements determined by FAAS. The filtered tissue digest was stable for at least one month when stored at 5 "C. Sahrawat (1 37) has shown that a simplepartiai digestion of a wide range of finely ground plant tissue with warm 0 . 5 ~ HC1 for 5 min gives quantitative recovery of K thus avoiding the use of the conventional HNO /Hz SO, /HClO, digestion.Other references of interest- Direct chemical pre-treatment of biological samples using CH, , 0, and H, in conjunction with a CRA90 furnace: C1009. Losses of trace metals from seawater on to container surfaces: 2171. Sampling procedure and determination of Pb in canned foods: 1590.Slurry atomization of biological tissue: 985, C105 1, 1471. Slurry atomization of geological samples: C 107. 3. I . I ,2 Dry and Wet Ashing Procedures Papers comparing various dry and wet ashing procedures for biological materials continue to appear at a prodigious rate (248,590, C882, C 1089, C1154'123 1,1282). It would be expected that for a given analysis requirement there would be better agreement concerning the relative merits of the large number of possible ashing procedures.Many papers relating to biological tissue analysis, simply test the proposed method using the NBS standard reference bovine liver and orchard leaves. In the Reviewer's opinion, trace metals are more easily liberated from these dried finely divided samples than from many routine samples.The commonest criticism of the dry ashing technique is that volatile elements such as As, Cd and Se are readily lost. Another, often unappreciated loss, is caused by reaction of the analyte with either the matrix or the digestion vessel to form an acid insoluble form of the analyte (see ARAAS, 1979,9,61). Blanusa and Brevski (936) using radioisotopes have shown that significant amounts of Fe in bovine liver became bound to the porcelain or silica ashing vessel at a temperature as low as 450 "C.Surprisingly, no Cd loss was observed even at 600 "C. Other workers have reported Cd losses at dry ashing temperatures as low as 400 "C (see ARAAS, 1979,9, Ref. 1568). This discrepancy is probably related to the relative amount of involatile CdO reduced to volatile elemental Cd during the dry ashing procedure.Other workers (1788) have found that dry ashing of sewage sludges was less efficient than simple HNO, acid digestion (see ARAAS, 1980, 10, 88). Additional references on the preceding topic- C467, 2038. Analysis of sediment samples for total trace metal content normally involves the use of HF and/or HClO, in the dissolution step.An alternative procedure using a simple 90% HNO, - 10% HC1 (V/V)digestion (153)gavegoodrecoveriesofCd, Co, Cr, Cu, Fe, Mn, Ni, Pb and Zn from NBS SRM 1645 (industrial river sediment). However, when the method was applied to estuarine sediments the recoveries were significantly lower than those obtained using a LiBO, fusion procedure. This again indicates that methods must be tested on routine samples as well as CRMs.A fusion procedure was also recommended for ceramic analysis (1780) and it proved to be more reliable than two alternative wet digestion methods involving HF.Methodololgy 77 The determination of wear metals in oils is very time consuming if conventional wet or dry ashing procedures are employed. Eisentraut et al. (157) have now published details of their method based upon adding a small amount of HF/HNO,/HCl to the sample and heating the mixture at 65 "C for 45 min in an ultrasonic bath.The reaction mixture was then diluted with isopropyl alcohol and MIBK. Quantitative recoveries of Al, Cu, Fe, Mg, Mo, Ni, Sn and Ti were observed (see ARAAS, 1980, 10, Ref. C556). Further use of elemental bromine in dissolution procedures for minerals has been reported (see ARAAS, 1980, 10, 149).Jones and Lee (533) used Br,:HCl (250) for the dissolution of sulphide concentrates in order to determine In and T1. Other workers (1319) used aqua regia/Br, for the determination of Ag and Au in a wide range of rocks and ores, a particular advantage being that the final solution could hold up to 15 pg ml-' Ag without precipitation of AgCl or AgBr . Other references of interest- H,PO, digestion of minerals for ICP-OES: C79.S in coal, plasma ashing: 581. Trace metals in coal/O, bomb: C122. 3.1.1.3 Sample Preparation and Reduction Procedures for Mercury Determination The rate of publication on this topic shows no sign of abating and it is very difficult for the newcomer to determine the optimum analysis procedure considering the bewildering array of published methods and techniques.Improvement in the Hg detection limit has been achieved by various techniques such as adsorption on to Au (634, 1605), sulphydryl cotton fibre (1792), XAD2 resin (2040), Dowex 1-X8 resin (1309), DDC loaded polyurethane foam (2148), and absorption into acidic KMnO, solution (1 890).Redesign of the reaction vessel and absorption cell (1 30, 1468) have also been used to improve the detection limit. It is perhaps surprising that in the above studies the atomicfluorescence technique was not considered. A good comparison of detection limits is afforded by considering the results of Tuncel and Ataman (130) (0.2 ng) using a 250 mm path length absorption tube and Ebdon and co-workers (C917,1024) (0.01 ng) using a simple AFS system. Ottaway and Scott (1296) used an AFS method for the determination of Hg in air using a simple passive gold wire sampler.Additional references on the preceding topic- 383, 397, C868. The following digestion and ashing procedures have been reported for total Hg determination: H SO, /H , 0 , for rocks sediments and soils (1274), low temperature 0 , plasma ashing for high sulphur oilcoke (C870), pyrolysis at 800 "C in the presence of Ag powder for sulphur and pyrites (954), H,SO,/HNO, in a bomb for petroleum (1783), 0, bomb for milk products (1344), NaOH/CdCl, for blood and hair (598) and alkaline KMnO, for biological material (C993).A wide range of biological samples has been combusted in the O,/H, flame of a Wickbold apparatus (1764).Fortunately, simple digestion procedures continue to be developed. The determination of total Hg in sediments using a straightforward aqua regia digestion has been reported (1322). Horvath et al. (167) have confirmed the advantages of the BrCl procedure (see ARAAS, 1978, 8, 62) for the rapid decomposition of a wide range of organomercury compounds in drinking waters.Sodium borohydride (2223) has also been shown to be effective for the breakdown of organomercury compounds. In addition, this study demonstrated that methylmercury compounds are not very stable with respect to storage under acidic conditions. Automation of cold vapour Hg determinations has been reported for urine and blood (2088) and for hair, urine and tissues (2084).Both methods were based on the digestion techniques of Magos (see ARAAS, 1972, 2, Ref. 25). Oda and Ingle (1968) evaluated three78 Analytical Atomic Spectroscopy different designs for continuous flow reduction vessels. The best design was based on stripping the Hg from a thin stream of solution with a countercurrent flow of gas over the solution stream.The detection limit was 0.03 ppb. Other references of interest - Hg in coal, plants and soils by AFS: 2054. Hg in seawater: 1605. TLC for separation of methylmercury: 1810. 3. I . 1.4 Sample Preparation and Reduction Procedures for Hydride Generation This would appear to be one of the biggest growth areas of new method publications. Speciation analysis of environmental samples gives more relevant information than does a total analyte analysis, since the toxicity of the hydride forming elements is often critically dependent upon the actual species present.Howard and Arbab-Zavar (394) used 20% V/V HC1 for dissolution of estuarine samples and generated arsine, monomethyl and dimethyl arsines using NaBH, at pH 5 after adding 0.02M EDTA to minimize interference effects.The liberated arsines were collected in a liquid N, trap and volatilized in order of boiling point. Maher (374) performed a similar study for the analysis of marine organisms and sediments. Samples were refluxed with 0. IM NaOH, evaporated to dryness and the residue dissolved in 8 . 5 ~ HC1. A Dowex 50 AG-X8 resin was used to separate the various species which were collected, converted to As (111) and then individually determined.A further study (2150), which also determined p-aminophenyl arsonate, used ion chromatography to separate five arsenic species. Additional references on the preceding topic - C367, C764, C766, (2867, 1575. Antimony(V) reduction to SbH, by NaBH, was found to be inhibited by the addition of HF (2073) or by maintaining a pH of 8 in the reduction cell (1486), thus allowing selective Sb(II1) determination to be carried out.In addition to Sb(II1) and Sb(V) determinations, methylstiboric and dimethyl stibinic acids have been determined in natural waters (1578) by collecting the hydrides in a liquid N trap after generation from a weakly acidic solution using NaBH, . The collected hydrides were then separated by GLC.Triethyl-lead, diethyl-lead and inorganic Pb have been determined in urine (1808) after adding 0 . 5 ~ malic acid to generate Et,PbH, 0 . 7 5 ~ H2o2/O.04M HClO, for Et,PbH, and 1 . 6 ~ malic acid/O.W5~ K2Cr,0, for total Pb, prior to NaBH, addition. Automation of hydride generation techniques can improve their accuracy, precision and speed of analysis. Dennis and Porter (1984) described a system for the determination of As, Sb and Se in environmental samples that has been successfully used in routine analysis for a number of years.Subramanian (1380) described a similar system for geological analysis. Various other automated systems using ICP atomization of the hydrides have also been reported (C1090, C1091, C1454, 1966, C1996,2145). Additional references on the preceding topic - C772, 1931, 1932, C2060.Caruso et al. (1 589) pursued the technique of directly nebulizing acidic sample solutions and NaBH, through two separate nebulizers into the same spray chamber (see ARAAS, 1980, 10,91). Peristaltic pumps were used to introduce both solutions into their respective nebulizers. The liberated hydrides were passed into an ICP where As, Bi, Sb, Se and Te were determined with sub-ppb detection limits.Additional reference on the preceding topic -C1053. Methods of minimizing interelement effects in As and Se determinations continue to be developed. Gunn (21 95) produced a comprehensive review (76 pp and 184 refs) with emphasis on the minimization of interelement effects. The following methods and masking reagents have been proposed for arsenic determinations: separation of AsCl, prior to hydride generation (C446), use of a disposable plastic syringe as a hydride generation vessel (401,Methodology 79 1295), use of a Zn reductor in place of NaBH, (ClW), addition of thiourea (1295), EDTA (394,2226), KI (2145): picolinaldoxime (1379), Cu2 + to remove Se(1V) interference (2139), and 1, 10-phenanthroline (596).In the last method, 10-25 pl of the digested sample (containing 1,lO-phenanthroline) was directly injected on to a single NaBH, pellet through a side port in a horizontal glass tube directly connected to a graphite tube furnace (596). Phenanthroline has also been used to minimize the effect of Co and Ni on SeH, generation (C761).Studies of the interference of HNO, and Cu on ASH, and SeH, evolution using NaBH, have produced conflicting results. It has now been clearly shown (2192) that nitrate ion does not cause significant inhibition. The inhibition is caused by the presence of nitrite and reduced nitrogen oxides produced by digestion of samples with HNO, . Reduced elemental Cu was shown to catalyse reduction of nitrate to nitrite and hence NO, species.Thus the combined effect of HNO, and Cuz+ ions is more than additive, It was demonstrated that as little as 0.5 pg ml- Cu severely inhibited SeH, evolution in the presence of HNO, . It was also shown that the use of NaBH,CN in place of NaBH, minimized this interference effect (see ARAAS, 1980, 10, 90). It should be noted that NaBH, CN liberates HCN on contact with acids and appropriate safety measures must be taken.Gas-phase interference in flame heated silica tubes has been demonstrated (C769) in the determination of Se in the presence of other hydride forming elements. The effect could be overcome by maintaining an air/H, flame inside the silica tube. In order to minimize the effect of large amounts of Cu on Sb determinations, the addition of KSCN was recommended (676), and thioglycollic acid has been used as an interference suppressor for the determination of Sn in steels and brass (1388).For accurate total arsenic and selenium analysis of many environmental samples, it has been conclusively shown that vigorous digestion procedures are required to ensure quantitative conversion of the various naturally occurring forms of these elements to the appropriate inorganic form.Various procedures which have been recommended for Se include combustion in 0, (602, 1027), HNO,/HClO,/H,SO, (2080) and H,PO,/HNO,/H,O, (2030). Dry ashing with Mg(NO,), at 450 "C was also found to give quantitative recovery of Se (2030). For the determination of As the following procedures have been proposed: dry ashing with Mg(N03), (1474), combustion in 0, (594), HNO,/H,SO,/HF/KMnO, or H,SO,/HNO, in a PTFE bomb at 150 "C for 3 h (966) and autoclave mineralization (1 196).Further work on the generation of lead hydride has been reported using HCl/H,O, (1340), K,Cr,O,/KCN/tartaric acid (1321) and K,Cr,O,/malic acid (1808). Conversion of phosphorus species of phosphine and subsequent introduction into an ICP was used to determine various phosphorus species.Fuwa and Matsumoto (1480) injected the sample containing phosphate into a 'Ta coated graphite tube in the presence of A1 powder. The tube was heated to 1300 "C in order to convert any phosphate to aluminium phosphide; it was then cooled and PH was generated by the addition of 2 . 7 ~ HC1. However, the detection limit was 10 times worse than that observed by direct nebulization of the sample.The differential determination of phosphinic and phosphonic acids was achieved (962) by direct reduction of both acids to PH , using granulated Zn and 1 . 6 ~ HCl containing a catalyst of Sn2 + and Ni2 i- ions. If 1, was added prior to the reduction step, the phosphonic acid was oxidized to H,PO, and was not subsequently reduced by the Zn/HCl.Other references of interest - AFS/hydride generation: 266, 1618, 2073. Determination of Ge: 2217 Effect of acids on generation of ASH, and SeH, : Cll7. Flow injection determination of Bi: C772, C2060. Preconcentration of As, Bi, Sb, Se and Te on La(OH),: 376.80 Analytical Atomic Spectroscopy Reduction cell design: 535, 2191.Reduction of Se(V1) to Se(1V): 406, 1328, 1866. Review -As: 2138. Review - As/Se: 21 95. Review - Sb/Se: 2163. Review - Se: 1329. 3.1.2 Preconcentration Techniques Papers describing new preconcentration procedures continue to proliferate. Problems are envisaged with translating many of these to the analysis of large numbers of samples, since they often require precise attention to detail, which is difficult to achieve for prolonged periods in a routine laboratory. Also, in many cases, it is difficult to discern any marked advantages over previously published methods.With increasing interest in automatic sample introduction (Section 2.4.2) techniques amenable to complete or partial automation have much to recommend them. Two useful reviews on the relative merits of a large number of preconcentration techniques have been published (2022, 2026). 3.1.2.1 Solvent Extraction Multi-element extraction has much to recommend it for routine analysis, a major advantage being reduced sample preparation time.Clark and Viets (529) simultaneously extracted Ag, As, Au, Bi, Cd, Cu, Ga, Hg, In, Pb, Pd, Pt, Sb, Se, Sn, Te, TI and Zn from digested geochemical samples containing ascorbic acid, KBr, KC1, and KI using Aliquat 336 (tricapryl methyl ammonium chloride) and Alamine 336 (tricapryl tertiary amine) dissolved in hexane/MIBK.The resulting extract could be directly nebulized into a flame or ICP (see ARAAS, 1978,8,64). However, when the extracts were used for ETA-AAS analysis, severe interelement effects from high concentrations of co-extracted Cu and Zn were observed.These were minimized by selective back extraction into a number of aqueous reagents (530). Back extraction of CHCl, or Freon extracts has again been recommended (1341, 2144) for both FAAS and ETA-AAS after extraction of Cd, Co, Cu, Fe, Ni, Pb and Zn using either APDC or DDC. This approach overcame the problem of instability of the organic extracts (see ARAAS, 1980, 10, 91).Tsalev and Vasileva (1946) successfully used discrete sample nebulization to determine Cd, Cu and Pb extracted from plant tissue digests using hexamethylene ammonium - hexamethylinedithiocarbamate. Chlorinated solvents (CHCl , , CCl,) gave more efficient extraction than MIBK. A selective extraction scheme (2143) for the determination of Au, Ir, Pd, Pt and Rh in platiniferous materials has been developed.Firstly, gold was extracted from 6~ HC1 into MIBK, secondly Pd and Pt were extracted from 6M HCl/KI into MIBK and finally Rh and Ir were extracted from the remaining aqueous phase by boiling with 2-mercaptobenzothiazole for 1 h and extracting into MIBK. Trace elements in steels (C843) have been selectively extracted from HCl using MIBK/(C,H, 3 ) 4 NI.Nord and Karlberg (1 346) described an automatedextraction system that was not limited by the nebulizer uptake rate. The extraction was performed using an autoanalysis system with a solvent flow of - 1 ml min-I . Part of the resulting organic extract was fed into the sample loop of an injector and when this was periodically activated the contents of the loop were carried by a pumped aqueous carrier stream into the nebulizer.The sensitivity for Cu was increased 5 times compared to direct nebulization into an air/C H flame. Aihara and Kiboku published a series of papers recommending a potassium xanthate/ MIBKextraction system for the determination of Au (956), Fe and Ru (959), In and Te (964) and Pd and Pt (486).Selenium (after conversion to Se(1V)) has been extracted from various matrices. These include geological samples using 2,3-diaminonapthalene/MIBK (577) and HCI/HBr/Fe(II)/Methodology 81 toluene (C90, 908), ores using methyl methacrylate (1476), blood as a chlorocomplex using 6M HCVMIBK (950) and drinking water using APDC/MIBK (1336). Other references of interest - Determination by solvent extraction of - As: 1336, 15 10.-Cd, Pb: 1623, 1740, 1906. -CU: 989, 1623. -Sb: 186,950, 1336, 1623. -V: C898. -Te: 939. 3.1.2.2 Ion-exchange Methods The main advantages of ion-exchange methods are that large concentration factors (up to 100 times) can readily be achieved and rhat the methods are suitable for analyte separation from saline matrices (e.g., seawater). The main disadvatages are that the methods are often limited by reagent contamination, are not suited to the routine analysis of large numbers of samples and that the sample requires some form of digestion pretreatment to ensure quantitative recovery of endogenous trace metals.Seawater analysis has been carried out using Chelex- 100 resin in conjunction with FAAS (C1004) and ICP-OES (1692), concentration factors of 100 being achieved in both cases.During another study (1341) using ETA-AAS, it was found that APDC/DDC/Freon extraction followed by back extraction into dilute HNO, was more reliable. It gave significantly better Cd recovery and significantly lower blanks than the Chelex- 100 resin method. Other workers have recommended complexation with oxine followed by adsorption on to C , -bonded silica gel (2154), complexation with oxine immobilized on silica gel (2222), complexation with 8-hydroxyquinoline-5-sulphonic acid and adsorption on to Biorad 140-1242 AGl-X2 resin (354), complexation with oxine followed by adsorption on XAD2 resin (C864, C865), and adsorption on to XAD resins (1005).Additional references on the preceding topic - 154, 1713. Orpwood (1877) carried out an extensive study on the use of Chelex resins for the determination of trace metals in drinking water.It was found that Ca and Mg compete with trace metals for exchange sites and hence recoveries of Cd, Co, Mn and Zn from hard waters were poor. Additional reference on the preceding topic - 2044. The direct injection of ion-exchange resin slurries into a graphite furnace has been shown to be an effective method of analysis for Cu (593, C803, 1665) and for A1 and Pb (C803).Acid dissolution of ion-exchange resins after adsorption of trace metals and nebulization of the digest appears to be a promising technique (see ARAAS, 1980, 10,92). Barnes and Wan-Young (C1155) discovered a polyacrylamidoxime resin that could be used to preconcentrate 31 elements.Aluminium was not significantly adsorbed by the resin and thus trace impurities in high purity aluminium could readily be determined. A polydithiocarbamate resin (2200) has been examined with respect to uptake of 22 elements including 14 rare earths. For many elements, poor uptake (30 - 40%) was observed. After acid digestion of the resin, poor recovery of Fe(II1) and Th was found and this could lead to unsuspected errors.The separation of trace amounts of Bi from a wide range of other elements was achieved (2193) using an AG 1-X4 quaternary amine anion-exchange resin. Other references of interest - Determination of Ag in ores: 1840. Determination of Pb and Zn in rocks: 926. Determination of trace metals in W: 1672.82 Analytical Atomic Spectroscopy 3.I .2.3 Adsorption Concentration and Other Techniques Adsorption of trace metals on to inert metal wires followed by ohmic heating of the wire to volatilize the adsorbed species appear to have considerable potential as a simple preconcentration technique. The application of Mo, Rh, Ta and W wires for this technique was investigated (562).The integrated analyte absorbance was found to be proportional to both the soaking time and analyte concentration. Satisfactory results were obtained using Mo, Rh and W wires. The technique was successfully applied to the determination of trace levels of Cu in sodium nitrate. Some very impressive detection limits have been obtained by adsorption of ultra-trace metals in polar snow samples on to a W wire loop (1 797).The loop was then ohmically heated using a commercially available ETA-AAS system. Detection limits for Cd, Cu, Pb, and Zn of 0.08, 1.3, 1.5 and 1.2 ng 1- were achieved. A novel co-precipitation method for the determination of Au in copper concentrates was reported by Chaundry and Johnson (C1444). The sample was digested in aqua regia and SnCl, solution was added to the diluted digest.The Au was co-precipitated on to AgCl which was separated and then digested with aqua regia to re-dissolve the Au. A detection limit of 10 ng g- was obtained. Co-precipitation has the disadvantage that the addition of significant quantities of the carrier reagent, as well as the various manipulations required in the isolation and subsequent dissolution of the carrier, can result in significant blank values for many elements.-The following carriers have been reported: Fe(OH), /TiO, for V(949), Fe(0H) for Cr(II1) (540), and DDC for Cd, Cr, Cu, Fe, Hg, Ni, Pb, Se(1V) and Zn (61 1, C2133). Colloidflotation was used to reconcentrate As(V), Ge, Sb(V) and Se(1V) (1617) by using sodium lauryl sulphate and Fe(0H) ,.Bismuth (180) and Sb (613) have also been determined using sodium dodecyl sulphate, sodium oleate and Fe(0H) . Electrochemical preconcentration was employed for ultra trace determination of Cd and Pb in seawater (1337), by direct plating of the metals at the sampling site on to a Hg coated graphite tube. When the tubes were subsequently returned to the laboratory, the metals were determined by direct ETA-AAS analysis (see ARAAS, 1980, 10, Ref. 1309). A German patent (1268) has been granted for a similar type of system. Constant potential electrolysis using a spiral wound Pt wire filament has been used to preconcentrate Se (2141, 2167). The filament was then placed in an Air/Ar/H, flame and the Se was liberated by rapid ohmic heating. Michalik and Stephens (C5 14,553) described a novel preconcentration technique based on electrostatic dustpreconcentration. The aerosol from a conventional pneumatic nebulizer was vaporized in a heating chamber, desolvated and collected on a Chromel wire electrode held at a high potential (+ 15 kV).The wire was then ohmically heated and the liberated analyte transferred to a suitable flame. The system has subsequently been improved (552) by replacing the wire electrode with an aqueous electrode solution (in contact with a Chromel wire).For the 40 elements tested in a lake water matrix, only La, Nd, U, Y and Zr did not respond. Donnan dialysis (2159) has been used to preconcentrate Co and Ni from natural waters after U.V. oxidation to degrade organic matter. Concentration factors up to 22 times could be achieved within one hour, but phosphate interfered (see Anal.Chem. 1977, 49, 1272). Other references of interest - Concentration by evaporation, determination of Se: 414. Determination of Au by co-precipitation on activated charcoal: 203. Determination of lanthanoids by co-precipitation on diantipyrinylmethane: 495 Determination of trace metals in natural waters by co-precipitation on DDC/2-napthol/2,4-dinitroaniline: 303, 304, 305.Determination of Se by volatilization of SeO :602.Methodology 83 3.1.3. Indirect Methods Indirect methods continue to proliferate. Considerable ingenuity has been displayed in the development of some of these methods, though in many instances relatively simple alternative techniques could be readily employed.Halide determinations are a good example, the precipitation of BiF3, PbBrF and CaF, and subsequent acid dissolution of the precipitate having been investigated for the determination of F in waters (1394). The latter (CaF,) method being recommended. Determinations using the fluoride ion selective electrode are considerably more rapid with a detection limit two orders of magnitude better than the proposed indirect precipitation method.Iodide (1029) was determined by adding excess Hg(N03)z to the analyte solution and injecting 10 pl of the resulting solution into a graphite tube furnace. A slow ramp heating rate was employed and two time resolved AA mercury peaks were observed from the decomposition of Hg(NO,), and HgI, . Bromide has been determined by monitoring CuBr molecular emission at 434.1 nm in an 0 , /H , flame (1 806).A method for the determination of total Cl, Br and I (1509), involved adding excess HgZ + to the analyte solution and passing the resulting solution through a cation-exchasge column. Mercury was then determined in the filtrate. A detailed study (C802) has been carried out on the indirect ETA-AAS determination of halides using the absorption of Al, Ga and In monohalide species (see ARAAS, 1980, 10, 93).The results from this study allowed successful analysis of biological samples to be carried out. Additional reference on the preceding topic - 275. A novel method for the indirect determination of sulphite or sulphur dioxide in solution was developed by Marshall and Midgley (1579, 1912).This was based on the reaction - For efficient complexation of Hg(I1) by SO z - and minimization of hydrolysis a pH of 3 was found to be optimum. The sample (5 - 500 pl) was added to a reaction cell containing 5 ml 1 0 - 6 ~ Hg(1) in l @ ’ ~ HN03 and the liberated HgO was continuously stripped from the solution by a gas stream to pass into a standard Hg absorption cell.A remarkable sulphur dioxide detection limit of 30 pg was achieved. The large number of papers describing procedures for the indirect determination of phosphate utilizing molybdo heteropoly acid formation shows little sign of abating (C 123, C801, C878, 920, 941, 951, 1625) (see ARAAS, 1978, 8,65). Anionic detergents (963) were determined by their extraction from alkaline solution as ion pairs with Na+ into MIBK in the presence of large amounts of NaCl.The Na in the MIBK layer was then determined by FAES. Indirect methods have also been proposed for acid anhydrides (278), aliphatic amines (946), alkaloids (C844), alkannins (1650), A1 (1816), amino nitrogen (1249), B (1970), colchicine (1633), enzyme activity (C886), epoxide hydrolase (1 582), fluorine in plasma atmospheres (2232), HCl in air (2089), nitrate (570), sulphate via BaSO, (426,487, C1013, 1410) and Zr (1684).Hgz2+ + 2S032-*Hg(S03)22- + HgO 3.1.4. Nebulization, Vaporization and Atomization Nebulization of slurries is discussed in Section 3.1.1.1. Considerable effort has been expended on the development of improved nebulization systemsfor ICP-OES. Thelin (389) described a slot-type (Babington) nebulizer coupled to a cyclone spray chamber that would tolerate 10% m/Vsteel digest solutions without blockage or significant memory effects.The SBR was 1.5 -2.5 times better than a commercial concentric nebulizer system. A new (commercially available) cross-flow nebulizer, known as the MAK nebulizer, appears to have considerable promise (650, 654).It consists of a heavy-walled glass jet rigidly aligned with a heavy-walled glass solution jet. The heavy rigid tubing and rigid bracing cross member84 A naly tical k tomic Spectroscopy prevent any vibration of the two tips during operation, a common failing of most currently available cross-flow nebulizers. The MAK unit operates at 200 psig, has a gas flow of 0.5 1 min - l , can nebulize 30% m/VNaCl solution and gives typical RSDs of less than 0.5% (see ARAAS, 1980, 10, 94).Goulter (651) has now published details of a corrosion resistant nebulizer fabricated from a fluorocarbon and fitted with a Pt/Ir capillary. Other workers ((2902) have developed a corrosion resistant nebulizer, spray chamber and torch. The nebulizer was constructed from PTFE, Pt/Ir and Zr, the spray chamber from a fluorocarbon and the inner tube of the torch from a-Al 2.0 3 .Ripson and De Galan (621) described a narrow bore stainless steel Babington nebulizer operating on an Ar flow of 50 - 200 ml min- '. The nebulization efficiency for water and organic solvents was comparable to that of a conventional pneumatic nebulizer operating at an Ar flow of 1 1 min- l .Further studies on the glass frit nebulizer (see ARAAS, 1979, 9, 65) have been reported, (C76, C439). A nebulization efficiency of 40% was claimed compared to 3 - 5% for most other pneumatic nebulizers. If the nebulizer was rinsed on the high pressure side of the frit, the rinse time required between samples could be reduced substantially. Ultrasonic nebulization has been found to give large improvements in ICP-OES detection limits when compared to pneumatic nebulization (see ARAAS, 1977,7, Ref. 1691). Floyd and Taylor (1836) have now published their work describing the use of a commercially available ultrasonic nebulizer with desolvation and vapour condensation to determine 3 1 elements in drinking water and sewage samples. Most of the lowest determinable concentrations were below the relevant drinking water statutory limits and were 5 - 20 times better than those obtained using a concentric pneumatic nebulizer.Sample carryover was not a significant problem with either nebulizer. Other ultrasonic nebulization systems have also been reported (649,653, C683, C1407). Discrete sample nebulization into ICPs was used for the simultaneous determination of Ca, Cu, Fe, K, Mg, Na, P and Zn in 50111 of 1 : 10 diluted whole blood and serum (356).Typical RSDs of 2 - 5% were found. Additional references on the preceding topic- C683, C735. Flow injection analysis (see Sections 1.3.3.2 and 2.4.2) has a number of advantages over conventional sample introduction. Sample sizes are small, sample throughput is high, releasing agents can easily be added to the liquid carrier stream, physical interference effects caused by vicosity of the sample can be minimized, and high salt concentrations do not result in clogging of the burner slot.The technique has successfully been applied to the determination of Cu in serum by FAAS (C836). Basson and Van Staden (262) simultaneously determined Ca, K, Mg and Na in natural waters by FAES and FAAS.The rate of analysis was 128 samples h-' . When aqueous samples were injected into a MIBK or butylacetate carrier stream, improved FAAS sensitivity and an analysis rate of 300 samples h-I were claimed (164). Tyson et al. (C817, C915) considered the theoretical aspects of flow injection/FAAS, whilst Greenfield (C680) performed a similar study for ICP-OES. Some Australian workers (C998) described the simultaneous ICP determination of Ca, Fe, Mg, Na and K in 10 pl aliquots of human serum using a flow injection system.The analysis rate of 240 samples h -* was limited by the digital print-out time. It is a sobering thought that even with this limited rate of analysis, it would be possible to screen the serum of the entire Australian population in about 6 years of continuous operation of just one ICP instrument. The vaporization of analyte halides directly from solid samples by heating the sample in the presence of halogen containing compounds may have considerable potential (see ARAAS, 1975,5, Ref. 237). A halide generator that could be interfaced to a flame or an ICP has been developed (C1074). The generator consisted of an induction furnace and a pressurized multi-sample chamber and could be used to analyse a set of samples without disturbing the operating characteristics of the flame or ICP.Spachidis and BaechmannMethodology 85 (1609) published a method for the separation of trace elements at the 20pg g -I level from high purity aluminium after halide formation. It was concluded from theoretical considerations that the separation of most elements except alkali and alkaline-earth elements by selective volatilization of halides should be possible. The determination of NH, in sewage effluent and Kjeldahl digests by gas phase molecular absorption has been automated by Vijan and Wood (2216).The sample was mixed with NaOH using a proportioning pump/mixer system and segmented using N, .The gadliquid stream then passed through a 20 ft glass coil maintained in a heating bath at 98 "C. The nitrogen and NH were separated from the liquid using a gas - liquid separator and passed into a heated absorption cell where the absorption of the 197.2 nm arsenic line was monitored. A detection limit of 0.1 pg ml-' NH, was obtained. Other workers (171) have significantly improved the detection limit of the manual technique as described by Cresser (see ARAAS, 1977, 7, Ref. 131), by collecting the liberated NH, in liquid nitrogen and carrying out the final absorbance measurement in a 1 m path length cell. A characteristic concentration of 20 ng ml-1 and a detection limit of 7 ng ml-' was achieved. Additional reference on the preceding topic- 598.A further report (2232) on the identification of crude oils by monitoring molecular absorption of organic species using an ETA device in conjunction with a very low ramp rate has appeared. (see ARAAS, 1979, 9,29). Considerable improvements in ICP-OES detection limits have been achieved by vaporizing driedsamples from a Pt or Wfilament directly into the plasma (C685).Detection limits for B, P, Pb, Sn and Zn of 4, 100 20, 20 and 30 pg, respectively, were claimed. Ure et al. (2140) have continued their work on atom trapping by mounting a water- cooled silica tube in an air/C,H, flame (see ARAAS, 1979, 9, Ref. 1464). Increases in sensitivity of 18,48 and 80 times for Cd, Pb and Zn respectively were found. Addition of V to the samples enhanced the Cd and Cu sensitivity and this was attributed to a vanadium oxide coating on the collecting tube improving the trapping efficiency.A solid sampling technique for ICP-OES using direct insertion of graphite cups into the plasma was found to give significant improvements in detection limits for steel analysis (187). Pettit and Horlick (C1056, C1079) described an automatic system for insertion of a graphite cup containing solid or dried liquid samples into the plasma.An Ar/O, coolant was more effective in volatilization and atomization of the samples than an Ar coolant. Meyer and Barnes (C1057) have also directly introduced CaO, CaCO , and A1 , 0 powders into an ICP from a fluidized bed. The air-ICP being more effective than the Ar-ICP for this type of analysis.A new tool, that can be partially mechanized, for the introduction of solidsamples into a graphite tube ETA has been described (C66). Other devices for direct analysis of polymer samples (475) and ion-exchange resins (C825) have also been reported. Methods for the minimization of chemical interference effects in ETA continue to be developed. Rapid drying of nebulized samples on to the surface of a graphite microboat (L'vov platform) mounted inside a graphite tube and maintained at 125 "C (C795) has allowed direct analysis of natural waters, brines and biological materials using aqueous calibration standards.The beneficial effect of coating graphite tubes with Zr have been investigated (4, C23). Holcombe et al. (C800, C1061) have adopted the novel approach of heating the dried sample in the graphite tube to a temperature just above the analyte atomization temperature and condensing the analyte on to a secondary cooler surface maintained within thegraphite tube.This allowed the pyrolysis products to diffuse away. The final atomization step then raised both surfaces above the atomization temperature of the analyte.A significant reduction in interferences effects was claimed.86 Analytical Atomic Spectroscopy Other references of interest- Continuum source AFS for Zr determination: 323. Determination of S in air by flash vaporization: 158. ICP-OES in the v.u.v.: 315, C706. Minature N O/H laminar diffusion flames for continuum source AFS: 629. Pneumatic nebulizer design for FAAS: C34. Use of an ETA device as a glow discharge source: C753. 3.2 Further investigations of spectral fine selection have resulted in the publication of new wavelength tables for ICP-OES. Boumans et af. (1532) used detection limit data for more than 800 prominent lines in a computer iteration procedure to establish definitive factors for translating the spectral-line intensities for the Cu arc (US National Bureau of Standards tables) into corresponding tables for the ICP.The latter tables together with a model for quantifying spectral interferences have formed the basis of the two-volume compilation “Line-Coincidence Tables for Inductively Coupled Plasma Atomic Emission Spectrometry” (C708). The earlier wavelength tables of Boumans and Bosveld (see ARAAS, 1979, 9, Ref. 1475) and Winge et af. (see ARAAS, 1979, 9, Ref. 1730) have been combined into a single listing (1428). Some research groups have been concerned with the characterization of noise in ICP- OES. A major noise source is the nebulization process and in the study by Duursma et af. (1 3 14) the aim was to isolate the nebulization process so that its influence on the SNR could be evaluated. A unique feature of their work was the measurement of time-dependent variations in the sample supply to the plasma achieved by monitoring density variations in the sample aerosol using a light scattering technique.The noise associated with the nebulizer and that of the emission signal were characterized by autocovariance functions and power spectral densities, and a strong correlation was found between the two sources of noise.The ability to predict fluctuations in the emission signal from fluctuations in the sample supply has important implications for improving precision in ICP analysis. Belchamber and Horlick (C80, C502) found little improvement in precision with increasing integration time (10 ms - 30 s) for four different ICP nebulizer systems. In order to clarify the situation and indicate the origin of noise in ICP measurements, noise power spectra obtained by Fourier analysis of analyte emission, acoustic emission and spray chamber pressure were recorded.Detection limits for 36 prominent lines between 190 - 320 nm were determined by Boumans et af. (C716) using experimentally measured SNRs and theoretically computed values for the RSD of the background signals.Solutions of elements yielding line-rich spectra were included in the study and an assessment of the separate effects of SNR, source flicker noise and shot noise on detection limits was made. Novel approaches to Calibration based on previous reports in ARAAS have been developed. Harnly and O’Haver (2035) (see ARAAS, 1980, 10, Ref. 458) described a simultaneous multi-element atomic absorption continuum source spectrometer (SIMAAC) which incorporates wavelength modulation. This was used to generate a family of calibration curves having overlapping linear ranges to provide calibration over 4 - 6orders of magnitude for each analyte. The calibration curves were derived from a series of 6 different absorbance measurements made across the analyte absorption profiles.Detection limits corresponding to measurement at the centre of the absorption profile were reported to be comparable to those of conventional FAAS for wavelengths greater than 280 nm. The use of several nonlinear curve-fitting functions for this work has been reported (C1129) as has a study of the effects of wavelength modulation frequency, modulation amplitude and modulation waveform on the SNR (Cl 101). A numerical correction procedure, termed concentration normalization, has been proposed (C731) for use with the separate sampling and excitation DETECTION LIMITS, PRECISION AND ACCURACYMethodology 87 analysis instrument developed by Beatty et al.(see ARAAS, 1979,9, Refs. 74 and 525). The procedure took into account variations in the amount of sample vaporized and also corrected for interelement spectral interference.Advantages claimed for the technique were a reduction in the number of calibration standards and an extension of the working curves for diverse alloy matrices. For the analysis of iron, steel, aluminium, copper and aqueous samples two low-alloy steel RMs were sufficient for the calibration.The generalized standard addition method proposed by Kalivas and Kowalski (see ARAAS, 1979, 9, Ref. 1844) has been adapted to ICP-OES (2219). This multi-element analysis procedure provides a means of wavelength characterization and selection, detection and quantification of interference effects and simultaneous determination of analyte concentrations. The use of Sr as an internalstandard in the determination of Ca in cement by FAAS using the air/C,H, flame has been reported by Dulude et al.(1957) (see ARAAS, 1980,10, Refs. 1052 and 1 158). Internal standardization decreased the variability of atomic absorption under changing flame conditions and the addition of 1% La minimized chemical interference. Calcium was determined in cement RMs with an accuracy of ~ 0 . 3 wt%. Precision was much poorer in the N20/C2H, flame. Yttrium was selected as a suitable internal standard for the determination of Al, As, Mn, P, Si and V in steel by ICP-OES (948). During a study of the effects of HF, ionization suppressants and A1 on the determination of Hf, Nb, Ta and Zr by FAAS (1163), improved linearity, sensitivity, precision and detection limits were realized when sample solutions contained 1070 HF and 0.2% Al.Garden et al. (152) reported improved measurement precision for several A A S determinations through the use of weighted least-squares calibration procedures in preference to the conventional least-squares procedure which assumes that variance is independent of concentration. The theoretical basis of the weighted least-squares method was presented and equations for calculating the regression equation, flagging potential outliers and calculating confidence bands around predicted sample concentrations were included.Applications to the determination of Cu in water by microsampling cup, Fe by conventional FAAS and Pb in blood by Delves cup AAS were demonstrated. Proposals were made by van Dalan and De Galan (909) for the formulation of analytical procedures in FAAS that take into account differences in performance parameters for different instruments.Criteria were presented for the optimum adjustment of instruments and for the definition of a useful concentration range. Other references of interest - Application of enforced variance in AAS: C834. Calibration graphs in Zeeman-effect AAS: C810, 81 1.Experimental and theoretical comparison of precision and noise in FAAS, FAES and FAFS: 1613. Internal standardization in sequential ICP-OES: C1119. Modified Simplex optimization method for ICP: 1539. Survey of A A S instrument performance: C39. Trace determinations at concentrations beyond the linear range by ETA- AAS: 2146. 3.3 STANDARDS AND STANDARDIZATION 3.3.1 Reference Materials Few reports have been specifically concerned with the use of atomic spectroscopic techniques for the characterization of new Reference Materials.The National Research Council of Canada has issued marine sediment RMs certified for matrix, minor and 13 trace elements (MESS-1, BCSS-1) and ICP-OES featured prominently in the analysis programme (C506).88 A nalyt ical A t omic Spectroscopy The preparation, analysis and certification of “Pond Sediment”, the second CRM to be issued by the National Institute for Environmental Studies (NIES), Japan, has been reported (C871, 1616).The material was analysed by 30 laboratories using a wide range of analytical techniques including AAS, ICP-OES, XRF, NAA, photon activation analysis and isotope dilution MS.Elemental composition was considered to be typical of a city pond sediment. A complete account of the preparation, analysis and certification of Pepperbush CRM (see ARAAS, 1979, 9, Ref. 1934) is available from NIES. The Royal Society of Chemistry’s Analytical Division sponsored an international meeting on reference materials in London in February 1982 at which papers were presented on the current standards programmes of the National Physical Laboratory (UK), the National Bureau of Standards (NBS, USA) and the European Community Bureau of Reference (CBR). In addition, the use of RMs in clinical chemistry and geology and the availability of commercial secondary standards were discussed.Summaries of papers are to be published in “Analytical Proceedings”.Other references of interest - Elemental concentrations in NBS biological and environmental CRMs: 241. Metal content of shark muscle powder RM: 2075. Standard samples for general analysis of silicate rocks and minerals: 2072. 3.3.2 Standardization Studies Interlaboratory studies are normally concerned with at least one of the following: establishing accurate determinand values for certification of RMs; developing reference methods of analysis; assessing the performance of participating laboratories. A number of such studies have been reported.An IUPAC reference method for Ni in serum and urine by ETA-AAS (101 8) involves acid digestion of the sample with HNO /HC10, /H SO,, complex formation with APDC and extraction into MIBK before measurement.Analysis of pooled serum and urine by seven laboratories yielded mean Ni values of 12.7 and 8.3 pg 1- with standard deviations of 3.0 and 1.9pg 1 - I , respectively. Brown et al. (18W,1900) reported the findings of an interlaboratory trial for the determination of serum Ca by a CBR reference method based on FAAS. Sources of variability were associated with the preparation and dilution of standard solutions and serum samples, but with appropriate control accuracy within _+2% was achieved.An NBS reference method for Ca in serum by FAAS was modified with minimal loss in precision and accuracy (C1993). Improvements realized were a reduced sample volume (0.1 ml compared to 10 ml), a reduction insample preparation time, increased throughput and an extended calibration range.The NBS has continued its standardization studies in the clinical field with the development of a reference method for the determination of Li in serum by FAAS (1288). Samples of pooled reference sera, with Li concentrations of 0.534 - 2.954 mmol 1 - I determined by isotope dilution MS, were used to establish the accuracy and precision of the method after analysis by 14 laboratories.Over the concentration range studied the standard error for a single laboratory using either manual or semi-automatic pipetting for sample dilution was about 1.5% with a negative bias of about 2%. The results of an interlaboratory study of the NBS reference method for the determination of serum Na by FAES (see ARAAS, 1979,9, Ref. 325) have been published (2215).Selenium results for seven samples of dried powdered foods (cereals, meat, milk products) typical Se content 1 ng g-l, were reported (1475) by eleven laboratories using hydride generation AAS, ETA-AAS, NAA and spectrofluorimetry. Considerable systematic errors were revealed for the study and it was recommended that a RM should be made available for such anlayses.One of the first interlaboratory studies to be specifically concerned with inductively- coupled plasma OES was the analysis of a groundwater sample (C879, C921). The sampleMethodology 89 filtered and acidified to approximately pH 2 was distributed in acid-cleaned polyethylene bottles to 30 laboratories worldwide for analysis. For elements such as Ca, K, Mg and Na present at the 10 - 100 mg 1-l level, the overall RSD of the ICP results was 5 - 7%, whereas for trace metals such as Cr, Cu and Fe present at the 10 pg 1- level, relatively high RSD values of 60 - 80% were typical.It is of interest to note that detection limits for individual elements reported by participating laboratories varied by up to two orders of magnitude. Considerable improvement in analytical capability for the determination of several metals at natural concentration levels was reported by Bewers et al.(1876) for a seawater intercalibration study. Systematic differences were noted between the results for frozen and acidified samples but only small differences were detected for Cd and Cu determinations by Kinard and Gales (480) reported an intrafaboratory comparative study of hydride generation AAS and ETA-AAS for the determination of organic and inorganic As in complex waste waters. The ETA-AAS technique yielded complete recovery of As irrespective of chemical form whereas for hydride generation AAS the mean recovery for organic As was 89%.For consistent results by hydride generation AAS, a closed acid digestion procedure was used before arsine generation.A workshop on the determination of Pb in foods using ETA-AAS demonstrated that reliable intralaboratory determinations over the range 1 - lo00 ng g-l can be obtained in most laboratories provided analysts have been instructed in contamination control (C1092, C1432). For milk analysis with a preliminary acid digestion - solvent extraction technique, seven analysts working independently in the same laboratory reported a Pb concentration of 2.0 M.6 ng g-I (the uncertainty represents one standard deviation).When outlier values were rejected the .uncertainty limit was reduced to a . 3 ng g- l . ETA-AAS a d ASV. Other references of interest - Comparison of methods for Na in fertilizers: 1592. Interlaboratory analysis of sewage sludge: C126.Interlaboratory comparison of nutrient concentrations of plant tissue: 626. Sampling and Analysis for Hg in gaseous process streams: 239. Tooth - Pb interlaboratory analysis: C1306.90 Analytical Atomic Spectroscopy TABLES 3.3A. 1-3.3A.8: CERTIFIED REFERENCE MATERIALS Explanation: The information given in the Tables represents the current availability of RM’s certified for elemental composition.Uncertified RM’s are not included in the listings. The tables have been enlarged compared to previous years and new entries are Table 3.3A.1 Chemicals and Petroleum Products, Table 3.3A.6 Biological, Botanical and Foods, and Table 3.3A.7 Clinical Materials. Categories of RM’s are based on those proposed by the International Organisation for Standardization (ISO).Table 3.3.A. 1 CHEMICALS AND INDUSTRIAL PRODUCTS ~ Supplier Material Bureau National de Metrologies (BNM), 8-10 rue Crillon, 75194 Paris Cedex 04, France Fuel oil Commission of the European Communities, Community Bureau of Reference (BCR), 200, rue de la Loi, B-1049 Brussels, Belgium Organometallic compounds Carbone Lorraine, 45, rue des Acacias, BP164, 75017 Paris, France Commissariat a 1’Energie Atomique, CristalTec, BP no. 85 Centre de tri, 38041 Grenoble Cedex, France Reagents Reagents Industrial Manufacturing Inspection Institute, 6-15-1 Ginza. Chuoku, Tokyo, Japan National Bureau of Standards, Office of Standard Reference Materials, Washington, DC 20234, U.S.A. Primary standards Primary standards, fertilizers, fuel oil, organometallic compounds National Physical Laboratory, Office of Reference Materials, Teddington, Middlesex TW11 OLW, England Produits Chimiques Ugine Kuhlmann, le Rubis Synthetique des Alpes, 38560 Jarrie, France Prolabo, 12 rue Pelee BP200, 75526 - Paris Cedex 11, France Fuel oil, organometallic compounds Reagents ReagentsMethodology 91 Table 3.3A.1 CHEMICALS AND INDUSTRIAL PRODUCTS - continued Supplier Material ~ ~~~ Rhone-Poulenc Chimie Fine, 21, rue JeanGoujon, 75008 Paris, France Reagents Service des Materiaux de Reference (SMR), 1 rue Gaston Boissier, 75015 Paris, France Fuel oil Table 3.3A.2 FERROUS METALS AND ALLOYS Supplier Finely divided form Solid form Amt h r Standardisierung und Warenpriifung 102 Berlin, Wallstrasse 16, D.D.R.Bundesanstalt fur Materialprufung (B AM), 1 Berlin 45, Unter den Eichen 87, Germany Bureau of Analysed Samples Ltd., Newham Hall, Newby, Middlesbrough, Cleveland TS8 9EA, England Bureau National de Metrologie (B.N.M.), 8-10, rue Crillon, 75194 Paris Cedex 04, France Centre Technique des Industries de la Fonderie (C.T.I.F.), 44, Avenue de la Division Leclerc, 92310-Sevre, France Centro Nacional de Investigaciones Metalurgicas, Cuidad Universitaria, Madrid 3, Spain Gosstandart of the USSR, 9 Leninsky Prospekt, 11704, Moscow, U.S.S.R.Institut de Recherches de la Siderurgie Francaise, B.P. 129, 78100-Saint Germain en Laye, France (ASMW) , Unalloyed and alloyed steels, cast irons, slags, ferro alloys Unalloyed and alloyed steels, slags, cast irons, ferro alloys High purity irons, unalloyed and alloyed steels, slags, cast irons, ferro alloys Unalloyed and alloyed steels, cast irons High purity irons C steels C steels Unalloyed and alloyed steels Unalloyed and alloyed steels, cast irons Unalloyed and alloyed steels, ferro alloys, steels cast irons, slags Unalloyed and alloyed92 Analytical Atomic Spectroscopy Table 3.3A.2 FERROUS METALS AND ALLOYS -continued Supplier Finely divided form Solid form Instituto de Pes uisas Tecnol6gicas do Estado de Unalloyed and alloyed SBo Paulo SIA -bT, steels DivisBo de Quimica e Engenharia Quimica, Nutleo de Padr6es Analiticos, Caixa Postal 7141, OlOoO-S~o Paulo - SP , Brazil Iron and Steel Institute of Japan, 9-4,1-Chome, Otemachi Chiyoda-ku, Tokyo, Japan MBH Analytical Limited, Station House, Potters Bar, Herts.EN6 lAL, England Metalimpex, POB 330, H-1376 Budapest, Hungary Unalloyed and alloyed steels, cast irons steels Unalloyed and alloyed Unalloyed and alloyed steels, cast irons Unalloyed and alloyed steels steels, cast irons Unalloyed and alloyed National Bureau of Standards, Unalloyed and alloyed Unalloyed and alloyed Office of Standard Reference Materials, Washington, DC20234, U.S.A.ferro alloys steels, cast irons, steels, cast irons Prolabo, 12, rue PelCe, B.P. 200, 75526-Paris Cedex 11, France S x Industries Inc., 3gO Park Avenue, Metuchen, NJ 08840, U.S.A. Swedish Institute for Metal Research, Drottning Kristinas vag 48, S-11428 Stockholm, Sweden South African Bureau of Standards, Private Bag X191, Pretoria, Transvaal OOO1 , South Africa Steels Unalloyed and alloyed steels, cast irons Unalloyed and alloyed steels, ferro-alloys, slags Ferro alloysMethodology 93 Table 3.3A.3 NON-FERROUS METALS AND ALLOYS Supplier Finely divided form Solid form Air Products Ltd., Special Prods.Dept., Weston Rd., Crewe, Cheshire CW1 lDF, England Aluminium Company of America, Alcon Technical Center, Alcon Center, PA 15069, U.S.A.Aluminium Pechiney , 23 bis, rue Balzac, 75360 Paris Cedex 08, France Amt fiir Standardisierung und Warenpriifung 102 Berlin, Wallstrasse 16, D.D.R. (ASMW), British Aluminium Co. Ltd., Chalfont Park, Gerrards Cross, Bucks. SL90QB, England Bundesanstalt fiir Materialpriifung (BAM), 1 Berlin 45, Unter den Eichen 87, Germany Bureau of Analysed Samples Ltd., Newham Hall, Newby, Middlesbrough, Cleveland TS8 9EA, England BNFMetals Technology Centre, Grove Laboratories, Denchworth Road, Want age, Oxfordshire, England Canada Centre for Mineral and Energy Technology, c/o Coordinator, CANMET, 555 Booth Street, Ottawa, Ontario KlAOGI, Canada Commissariat a 1’Energie Atomique, (C.E.A.) Cristal Tec, B.P.no. 85 Centre de tri, 38041 - Grenoble Cedex, France Cu , Mo , Pb , Ti, Zr base A1 base High-purity metals, Al, Mg base Sn , Al , Mg base Al, Cu Al base Cu, Ni, Al, Mg base High-purity metals, High-purity metals, Al, Mg, Cu, Ni, Sn, Pb base Al, Cu, Ni base Al, Cu, Ni base Cu base A1 , Mg base94 Analytical Atomic Spectroscopy Table 3.3A.3 NON-FERROUS METALS AND ALLOYS - continued Supplier Finely divided form Solid form Centre Technique des Industries de la Fonderie Cubase (C.T.I.F.), 44.Avenue de la Division Leclerc, 92310-Sbvres, France Centre Technique du Zinc, 34, rue Collange, 92300 - Levallois Perret, France Chemicals Inspectionn & Testing Institute, 1-1, CChome, Higashi-Mukojima, Sumida-Ku, Tokyo, Japan Gosstandart of the USSR, 9 Leninsky Prospekt, 117L)4Moscow, U.S.S.R. Inco Europe Limited, European Research and Development Centre, Commercial Development Department, BirminghamB16OAJ, England Instituto de Pesquisas Tecnol6gicas do Estado de SBo Paulo S/A - IFT, DivisBo de Quimica e Engenharia Quimica, NuCleo de Padrdes Analiticos, Caixa Postal 7141, 01OOO-SBo Paulo - SP, Brazil Japan Aluminium Federation, Nihonbashi Meiji Building, 1-3,2-Chome, Nihonbashi, Japan Chuo-Ku, Tokyo, Japan Brass Makers Association, 12-22,l-Chome, Tsukiji, Japan Chuo-Ku, Tokyo, Japan Light Metal Association, 1-3,2-Chome, Nihonbashi, Japan Chuo-Ku, Tokyo, Johnson Matthey Chemicals Ltd., Orchard Road, Royston , Herts.SG8 5HE, England A1 , Cu , Ni base Al, Cu, Mg base High-purity metals, Zn base Al, Cu, Ni base Al, Cu, Ni base Al, Cu, Ni base Cu base A1 , Cu, Mg base Al, Cu, Mg base High-purity metals High-purity metalsMethodology 95 Table 3.3A.3 NON-FERROUS METALS AND ALLOYS -continued Supplier Finely divided form Solid form MBH Analytical Limited, Station House, Potters Bar, Herts.EN6 lAL, England Mercure Jndustrie, 13, rue Saulnier, 92800-Puteawq France A1 , Cu , Ni , Zn , Co base High-purity metals Metalimpex, POB 330, H-1376 Budapest, Hungary National Bureau of Standards, Office of Standard Reference Materials, Washington, DC20234, U.S.A.Planet-Wattohm, 05310-la Roche de Rame, France Prolabo , 12, rue Pelke, B.P. 200, 75.526-Paris Cedex 11, France RhGne- Alpes Mercure , 4, rue des Fauvettes, Mons Vilette D’Authon, 38230 Pont de Cheruy , France Spex Industries Inc., 3880 Park Avenue. Metuchen, NJ 08840, U.S.A. A1 base High-purity metals, Al, Co, Cu, Ni, Pb, Mg, Sn, Ti, Zn, Zr base Al, Cu, Pb, Ni, Ti, Zn, Zr base High-purity metals High-purity metals High-purity metals Cu, Pb, Sn base Table 3.3A.4 GEOLOGICAL MATERIALS Supplier Finely divided form Amt fur Standardisierung und Warenprufung 102 Berlin, Wallstrasse 16, D.D.R.Bundesanstalt fur Materialprufung (BAM), 1 Berlin 4.5, Unter den Eichen 87, Germany Mn, Cr, Sn ores (ASMW) , Fe ores96 Analytical Atomic Spectroscopy Table 3.3A.4 GEOLOGICAL MATERIALS - continued Supplier Finely divided form Bureau of Analysed Samples Ltd., Newham Hall, Newby, Middlesbrough, Cleveland TS8 9EA, England Fe, Mn, Cr, Al, ores fluorspar, sillimanite, Na & Kfeldspar, magnesite, dolomite, limestone Canada Centre for Mineral and Energy Technology, do Coordinator.CANMET, 555 Booth Street, Ottawa, Ontario K1A OG1, Canada Centre National de la Recherche Scientifique, Centre de Recherche Petro raphi ues et Geochimiques (C.N.R.S.It!.R.P.&.), 15, rue Notre Dame des Pauvres, Case Officielle No. 1, 54500 Vandoeuvre-lez-Nancy , France Commission of European Communities, Community Bureau of Reference (BCR), 200 Rue de la h i , B-1049 Brussels, Belgium Geological Survey of Japan, 1-3, Higashi 1-Chome, Yatabemachi, Tsukuba-Gun, Ibaragi, Japan Gosstandart of the USSR, 9 Leninsky Prospekt, 11704 Moscow, U.S.S.R. Instituto de Pesquisas Tecnoi6gicas do Estado de Srio Paul0 S/A-IPT, Divisiio de Quimica & Engenharia Quimica, 0100 SSo Paulo-SP, Brazil International Atomic Energy Agency, Analytical Quality Control Services, Laboratory Seibersdorf , PO Box 590, A-1011 Vienna, Austria Junta de Energia Nuclear, Cuidad Universitaria, Madrid-3, Spain L.R.M., B.P. 3013, 54000 Nancy Cedex, France Sb , Co-Mo , Au , Fe , Mo ores syenite, gabbro, ultramafic rocks, soils Bauxite, granite, iron ores Zn, Sn, Cu, Pb ores; coke Feldspar, clays, granodiorite, basalt U ores, lake sediment and soil Phosphate rocks and clays U ores Lignite RocksMethodology Table 3.3A.4 GEOLOGICAL MATERIALS - continued Supplier Finely divided form Marine Analytical Chemistry Standards Program, Chemistry Division, National Research Council, Montreal Road, Ottawa, KlAORG, Canada Marine sediments National Bureau of Standards, Office of Standard Reference Materials, Washington,DC20234, U.S.A.Fe, Al, Cu, Mo, Li, Zn, W ores, fluorspar, Na & K feldspar, clays National Chemical Lab.for Industry, 1 , Higashi 1-Chome, Yatabemachi, Tsukuba-Gun, Ibaragi, Japan River and estuarine sediments, coal, feldspar, clays, granodiorite , basalt National Institute for Environmental Studies, Division of Chemistry & Physics, Y atabemachi , Tsukuba, Ibaraki, Japan Pond sediment South African Bureau of Standards, Private Bag X191, Pretoria, Transvaal OOO1 , South Africa US Geolo ical Survey, National tenter 972, Reston, Va. 22092, U.S.A. Rocks, Fe, Cr, Pt and Zr ores Diverse Table 3.3A.5 GLASSES, CERAMICS AND REFRACTORIES Supplier Finely divided form Bureau of Analysed Samples Ltd., Newham Hall, Newby, Middlesbrough, Cleveland TS8 9EA, England Centre d'Etudes et de Recherches de L'Industrie des Liants Hydrauliques, 23 , rue de Cronstadt , 75015-Paris, France Silica brick, firebrick, magnesite-chrome, Portland cement , zircon, high purity silica Cement98 Analytical Atomic Spectroscopy Table 3.3A.5 GLASSES, CERAMICS AND REFRACTORIES -continued -~ ~~ Supplier Finely divided form Centre National de la Recherche Scientifique, Centre de Recherche Petrographi ues et Geochimiques (C.N.R.S.lC.R.P.8 .) , 15, rue Notre Dame des Pauvres, Case Officielle No. 1, 54500 Vandoeuvre-lez-Nancy , France Federation Europeenne des Fabricants de Produits Refractaires (P.R.E.), 44, rue Copernic, 75016Paris, France L.R.M., B.P. 3013, 54000 Nancy Cedex, France National Bureau of Standards, Office of Standard Reference Materials, Washington, DC20234, U.S.A.National Chemical Lab. for Industry, 1, Higashi 1-Chome, Yatabemachi, Tsukuba-Gun, Ibaragi, Japan Prolabo, 12 rue Pelee, BP200, 75526-Paris Cedex 11, France She ffield University , Dept. of Ceramics, Glasses & Polymers, Northumberland Road, SheffieId S10 2TZ, England Society of Glass Technology, 20 Hallam Gate Road, Sheffield S10 5BT, England Glasses (2 available) Refractory materials Refractory materials Leadharium, opal, high and low boron, soda lime glasses, silica, aluminosilicate and chrome refractories, Portland cements Sodalime silica, silica, high silicic acid - high boric acid glass Refractory materials G 1 asses Glasses (3 available) Table 3.3A.6 BIOLOGICAL, BOTANICAL AND FOOD MATERIALS Supplier Material Laboratoire National D’Essais (LNE) . 1, rue Gaston Boissier, 75015 Pans, France PlantsMethodology 99 Table 3.3A.6 BIOLOGICAL. BOTANICAL AND FOOD MATERIALS - continued Supplier Material National Bureau of Standards, Office of Standard Reference Materials, Washington, DC 20234, U.S.A. National Institute for Environmental Studies, Division of Chemistry and Physics, Yatabemachi, Tsukuba, Ibarak , Japan Bovine liver, brewers yeast, citrus leaves, orchard leaves, oyster tissues, p qe needles, rice flour, tomato leaves, wheat :lour Pepperbush Table 3.3A.7 CLINICAL MATERIALS Supplier Material Biomerieux, Chemin de L’Orme, Marcy L’Etoile, 69260 Charbonnieres Les Bains, France Biotrol, 1, rue du Foin , 75 140 Paris Cedex 03, France Various liquid preparations Various liquid preparations National Bureau of Standards, Office of Standard Reference Materials, Washington, DC 20234, U.S.A.Freeze-dried urine and serum Table 3.3A.8 ENVIRONMENTAL MATERIALS Supplier Finely divided form Bureau of Analysed Samples Ltd., Newham Hall, Newby, Middlesbrough, Cleveland TS8 9EA, England Furnace dust (LD) Institut de Recherches de la Siderurgie Francaise, B.P. 129, 781WSaint Germain en Laye, France National Bureau of Standards, Office of Standard Reference Materials, Washington, DC20234, U.S.A.Furnace dust (electric) Urban particulate, coal fly ash. filter media, waters100 Analytical Atomic Spectroscopy TABLES 3.3B.l- 3.3B.8 REFERENCE METHODS OF ANALYSIS Explanation - The information given in the Tables is a compilation of reference methods of analysis which have been published and approved by various organizations. In the present context no distinc- tion is made between the terms reference method, standard method, recommended method and official method as used in the literature.Most of the methods are based on atomic absorption spectrometry but flame atomic emission spectrometry, emission spectrography and inductively- coupled plasma optical emission spectrometry also feature in the listings. Each entry in the Table, for the appropriate body, represents either the page number in the standard text or the code number of the reference method. The superscript for each entry signifies the analytical technique as follows: (1) Flame atomic absorption spectrometry (2) (3) (4) (5) Flame atomic emission spectrometry (6) (7) Emission spectrography The names and addresses of the official bodies are given below.Electrothermal atomization - atomic absorption spectrometry Cold vapour - atomic absorption spectrometry Hydride generation - atomic absorption spectrometry Inductively-coupled plasma - optical emission spectrometry 1 AMC: 2. AOAC: 3. APHA: 4. ASTM: 5 . BS: 6 . EPA: Reference Methods of Analysis - Official Bodies Analytical Methods Committee, Analytical Division, The Royal Society of Chemistry, Burlington House, London W 1V OBN, U.K. Association of Official Analytical Chemists, Official Methods of Analysis (13th Ed.), 1111 N 19thStreet-Suite210, Arlington VA 22209, U.S.A.American Public Health Association, Standard Methods for the Examination of Water and Wastewater (14th Ed., 1975), 1015 15th Street NW, Washington DC 20005, U.S.A.American Society for Testing and Materials, 1916 Race Street, Philadelphia PA 19103, U.S.A. British Standards Institution, 2 Park Street, London. W1A 2BS, U.K. US Environmental Protection Agency, Methods for Chemical Analysis of Water and Wastes, Office of Research and Development, Environmental Monitoring Systems Laboratory, Research Triangle Park, NC 2771 1, U.S.A.101 Methodology 7 . IP: 8.ISO: 9. IUPAC 10. NBS: 11. NWC: 12. SABS: 13. USGS: Institute of Petroleum, 61 New Cavendish Street, London. WlM 8AR, U.K. International Organization for Standardization, Case Postale 56, 1211 Geneva20, Switzerland. International Union of Pure and Applied Chemistry. Bank Court Chambers, 2-3 Pound Way, Cowley Centre, Oxford OX4 3YF, U.K.National Bureau of Standards, Office of Standard Reference Materials, Washington DC 20234, U.S.A. National Water Council Standing Committee of Analysts, Dept. of the Environment, Room A416, Romney House, 43 Marsham Street, London SWlP 3PY, U.K. South African Bureau of Standards, Private Bag X191m, Pretoria 0001, South Africa. US Geological Survey, Methods for Determination of Inorganic Substances in Water and Fluvial Sediments Book 5 , Chapter A 1, 12201 Sunrise Valley Drive, Reston, VA 22092, U.S.A.w 0 Table 3.3B. 1 CHEMICALS AND INDUSTRIAL PRODUCTS h) __- Matrix Analyte AOAC ASTM BS 1P IS0 IUPAC SABS Ammonium bicarbonate Baking powders Borates Boric acid Calcium chloride Caustic soda Chlorine (liquid) Cryolite Drugs Fertilizers Fuels, coal coalkoke ash ashes gas turbine gasoline gas (natural) uranium oxide Greases, lubricating Oil, electrical insulating fuel (residual) fuel and crude lubricating (unused) waterborne Pb Al 8.023"' Ca, Mg Ca, Mg 20.043(') Mg, K , Na E449") Hg E538(3) Hg E506(3' Al, F, Fe, Na, Si Na P Ca, K, Na 36*055(" Ca,Cu,Fe, Mg, Mn, Zn 2-109(]) Hg D3684(3) Majorlminor elements D3682(') Trace elements D3683(') Inorganic constituents Trace metals D2788") Trace metals D3605(*) Mn D3831(') Pb D3237(') H2S D2725") U308 E402(') Li, Na D3340(5) c u D3635(') Na D 1318(5) Na, Ni, V Ba, Ca, Mg, Zn Ba, Ca, Mg, Zn Ni, V D3327(') 5050' DP71 lo(') DP69 18(' 122- 1175(')Organic chemicals Paint Paper, boards, pulp Hg (proposed(3) '81) Sb (low conc.) D3717(') Cr (low conc.) D3718(') Pb, Cd, Co (low conc.) D3335(') Hg (low conc.) D3624(3) Pb 5.001 Cd Ca c u Fe Mn Paper, highly opaque pigments Cd, Zn Phosphates (condensed) Ca Phosphoric acid Ca Pigments Cd, Zn Pol yols Na, K Rubber Pb c u Mn Pb, Zn Sodium hydroxide Hg Sodium phosphate Hg Sodium sulphate Ca Timber preservatives As, Cr, Cu Copper naphthenate Sn Urea Biuret D1224(') D4004( ' ) 607511 1(3) DIS3856(') DIS3856/I(') DIS777(') DIS778(') DIS779(') DIS 1830(') DP6101/2(') DP6 10 1/3(') DP6 10 1(') DIS5993(') DIS5994(') D P7 1 02(3) 673-1 976(') 5666/4(') 5 666/7( ' ) DIS4274( ') c.L 0 w1 04 Analytical Atomic Spectroscopy Table 3.3B.2 FERROUS METALS AND ALLOYS Matrix Analyte Cast iron Ferronickel Iron ore Iron/Cr/Ni alloy Steel Silicon steel & ingot/carbon/low- alloy steevwrought iron Tool steel/medium-high alloy steel Pb, Mg c o Al Ca, Mg c u Na, K Pb Bi, Pb Ni c u Pb Pb ASTM E351(') BS i50 K2/20: 64(')* DP7520(') DIS4688(') DIS4692(') DP4693(') DP683 I(') *Australian Standard.Table 3.3B.3 NON-FERROUS METALS AND ALLOYS BS rso Matrix Analyte ASTM Mn 41 40/22(2) DIS3390(') Alumina Zn DIS2071(') Aluminium ores Cd DIS5961(') DIS5666/1-3(3) DIS6061(3) Aluminium & aluminium alloys Cd E34(') 2 Aluminium oxide Cadmium Copper & copper alloys Cr c u Na Pb Zn Ca, Na, V, Zn Ag, Cu, Pb, Zn E396(') A1 Bi Cd Cr Cu, Te Fe Ni Pb, Zn E478") Mg Pb Sb Te Zn Magnesium & magnesium alloys Pb Magnesiudchromium ores Al, Cu, Pb, Zn DIS3982(') 1728/19(') DIS3980(') 1728/23(' ) DIS3256(') 1728/24(') DIS3981(') 1728/20(') 1728/2 1(') 4140/AD2(') DIS4749(') DP7601(') D P7605( * ) D P4740( ' ) DIS58890' 3907/15(')Methodology 105 Table 3.3B.3 NON-FERROUS METALS AND ALLOYS - continued Matrix Analyte ASTM BS i50 Nickel & nickel alloys Ag, Bi, Cd, Co, Cu Fe, Mn, Pb, Zn Al, Si c o Cr c u Fe Mn Trace metals Zn Ca, K, Mg, Na Co, Fe, Mn, Ni Mo, Ti, V Co, Fe, Mn, Mo Ni, Ti, V Cr Al, Cd, Cu, Fe, Pb, Mg E536(') A1 Cd c u Sn Metals Nickel (electronic grade) Mg Pig lead Ag, Bi, Cu, Zn E37(') Powder metals Zinc & Zinc alloys Mg DP635 1(') DP7530/7(') DP75 30/2( ) DP7530/3(') DP7530/4(') DP7530/5(') DP7530/6(') 3727/21(') 3727/20(') 3727/22(') DIS7627/2(') DIS7627/3(') D IS7627/4(') DIS7627/5(') DIS7627/6(') DP4812(') DP4810(') DP4811(') DP7 155(') 3630/15(') 12250 Table 3.3B.4 GEOLOGICAL MATERIALS Matrix Analyte ASTM USGS Gypsum &gypsum products Na Sediment 2 AS Ba Be Ca Cd c o Cr c u Fe Zg Mg Mn Mo Na106 Analytical Atomic Spectroscopy Table 3.3B.4 GEOLOGICAL MATERIALS - continued Matrix Analyte ASTM USGS Ni Pb Sb Se Sn Sr Zn Table 3.3B.5 GLASSES, CERAMICS AND REFRACTORIES Matrix Analyte ASTM IS0 SABS Cement Na, K DP48 13(') 551(1.5) Cement, blended hydraulic Na, K C114('.5) Glazed ceramic surfaces Pb, Cd C738(') Glazed ceramic tile surfaces Pb, Cd C895(') Porcelain enamel surfaces Pb, Cd C872(') Table 3.3B.6 BIOLOGICAL, BOTANICAL AND FOODS Matrix Analyte AMC AOAC IS0 IUPAC MAFF ~~ __ ~~ ~~~ Animal feeds Ca, Cu, Fe 7 49 1 (I) Distilled liquors Cu, Fe 9-029(') Foods c u Ref.G(') 25.083'3) Ref. H(3) 25.044(') Sn 25.136(') Zn 25 - 150(') Fish Pb 25*068(') Mg, Mn, Zn TC/34/SC10(') Hg Pb Fruit & vegetables Zn DIS66362(') Milk Pb 25*063(') Organic matter Cd Ref.A B(') Ni Ref. dl) Se Ref. D(4) Sb Ref. E(4) Zn Ref.Methodology 107 Table 3.3B.6 BIOLOGICAL7 BOTANICAL AND FOODS -continued Matrix Analyte AMC AOAC IS0 IUPAC MAFF Plants Cd Ca co c u Pb Mg Mn Ni K Na Zn Hg Tea Cd, Ni Wines Cu, Fe A. Analyst 1969,94,1153.B . Analyst 1975,100,761. C . Analyst 1979,104,1070. D. Analyst 1979,104,778. E. Analyst 1980, 105,66. F. Analyst 1973,98,458. G. Pure Appl. Chem., 1979,51,385. H. Pure Appl. Chem., 1979,51,2527 Table 3.3B.7 CLINICAL MATERIALS Matrix Analyte BCR IUPAC NBS Blood serum Ca Ref. A(1) 260-36(') K 260-63(') Li 260-69(') Na 260-60(') Blood serum, urine Ni Ref.B(*) A. B. J . Clin. Chem. Clin. Biochem., 1981,19,413 Pure Appl. Chem., 1981 53,773.108 Analytical Atomic Spectroscopy Table 3.3B.8 ENVIRONMENTAL MATERIALS APHAASTM Bs EPA i50 Nwc USGS Matrix Analyte Air particulates Pb Pb Pb Pb Atmosphere, workplace Pb Waters & wastes Ag Al As Au Ba Be Ca Cd c o Cr c u Fe Ir K K, Na Li Ms Mn MO Na Ni 0 s Pb Pd Pt Re Rh Ru Sb SiOz Se Sn Sr Ti TI V Zn Hg 2690Pt 24B) 2690Pt 25(5) D 1428(5) D511(') D3372(') D279 10) 148(') D1886(') D3697(') 159(') D3859 Trace elements D3919(') Majorltrace elements Proposedc6) Sea water, brines K, Li, Na D3561(') Ba D3651(') 40FR46258(') EQL-0380-043"' EQL-0380-044(2) EQL-0380-045(6) 229(') DP6061 159(') Sr D3352(')Methodology 109 Table 3.3C SUPPLIERS OF SPECTROGRAPHIC GRAPHITE ELECTRODES 1 Baird Corporation Inc., 125 Middlesex Turnpike, Bedford, MA 01730, U.S.A. 2 Carbon Products Division, Union Carbide Corp., 270 Park Avenue, New York, NY 10017, U.S.A. (ARL Ltd., Wingate Road, Luton, Beds., England). 3 Labtest Equipment Co., 11828 La Grange Avenue, Los Angeles, CA 90025, U.S.A. 4 Johnson Matthey Chemicals Ltd., Orchard Road, Royston, Herts. SG8 5HE, England. 5 Le Carbone (GB) Ltd., Portslade, Sussex, England. 6 Le Carbone Lorraine, 3741 Rue Jean-Jaures, 92231 Gennevilliers, France. 7 Jarrell-Ash, 590 Lincoln Street, Waltham, MA 02154, U.S.A. 8 Zebac Inc., P.O. Box 345, Bevea, OH 44017, U.S.A. 9 Ringsdorffe-Werke GmbH, 53 Bonn-Bad Godesberg, West Germany (Mining & Chemical Products Ltd., Alperton, Wembley, Middlesex HA0 4PE, England) Spex Industries Inc., 3880 Park Avenue, Metuchen, NJ 08840, U.S.A. (Glen Creston, 16 Dalston Gardens, Stanmore, Middlesex, HA7 lDA, England) Ultra Carbon Corp., P.O. Box 747, Bay City, MI 48706, U.S.A. (Heyden & Son Ltd., Spectrum House, Alderton Crescent, London NW4, England) 10 11 Table 3.3D SUPPLIERS OF STANDARD METAL SOLUTIONS (MS) AND REAGENTS (R) FOR AAS 8 9 10 Aldrich Chemical Co. Inc., 940 W. St. Paul Avenue, Milwaukee, WI 53233, U.S.A. (R) J. T. Baker Chemical Co., 222 Red School Lane, Phillipsburg, NJ 08865, U.S.A. (MS, R) Barnes Engineering Co., 30 Commerce Road, Stamford, CO 06902, U.S.A. (MS) BDH Chemicals Ltd., Poole, Dorset BH12 4NN, England (MS, R) Bio-Rad Laboratories, 2200 Wright Avenue, Richmond, CA 94804, U.S.A. (MS) Carlo Erba, Divisione Chimica Industriale, Via C. Imbonati 24,20159 Milano, Italy (MS) Eastman Organic Chemicals, Eastman Kodak Co., 343 State Street, Rochester, NY 14650, U.S.A. (R) Fisons Scientific Apparatus Ltd., Bishop Meadow Road, Loughborough, Leics. LEll ORG, England (MS, R) Harleco, Div. of American Hospital Supply Corp., 60th and Woodland Avenues, Philadelphia, PA 19143, U.S.A. (MS) Hopkin & Williams Ltd., P.O. Box 1, Romford, Essex RM1 lHA, England (MS, R) 11 V. A. Howe & Co. Ltd., 88 Peterborough Road, London SW6 3EP, England (MS)110 Analytical Atomic Spectroscopy [nstrumentation Laboratory Inc., 113 Hartwell Avenue, Lexington, MA 02173, U.S.A. (MS) Johnson Matthey Chemicals Ltd., Orchard Road, Royston, Herts. SG8 5HE, England (R) Koch-Light Laboratories Ltd., Colnbrook, Bucks., England (R) (Anderman & Co. Ltd., Central Avenue, East Molesey, Surrey KT8 OQZ, England) May & Baker Ltd., Dagenham, Essex RMlO 7XS, England (R) E. Merck, D 61 Darmstadt, West Germany (R) Spex Industries Inc. , 3880 Park Avenue, Metuchen, NJ 08840, U.S.A. (MS) 4LFA Division, Ventron Corp., 152 Andover Street, Danvers, MA 01923, U.S.A. (MS) (Glen Creston, 16 Dalston Gardens, Stanmore, Middlesex AH7 lDA, England) Table 3.3E SUPPLIERS OF ORGANOMETALLIC STANDARDS Angstrom Inc., P.O. Box 248, Belleville, MI 48111, U.S.A. Baird Corporation Inc., 125 Middlesex Turnpike, Bedford, MA 01730, U.S.A. J. T. Baker Chemical Co., 222 Red School Lane, Phillipsburg, NJ 08865, U.S.A. BDH Chemicals Ltd., Poole, Dorset BH12 4NN, England Burt and Harvey Ltd., Brettenham House, Lancaster Place, Strand, London WC2, England Carlo Erba, Divisione Chimica Industriale, Via C. Imbonati 24,20159 Milano, Italy Conostan Div., Continental Oil Co., P.O. Drawer 1267, Ponca City, OK 74601, U.S.A. Durham Raw Materials Ltd., 1 4 Great Tower Street, London EC3R 5AB, England Eastman Organic Chemicals, Eastman Kodak Co., 343 State Street, Rochester, NY 14650, U.S.A. Hopkin and Williams Ltd., P.O. Box 1, Romford, Essex RM1 lHA, England E. Merck, D 61, Darmstadt, West Germany MBH Analytical Ltd., Station House, Potters Bar, Hem. EN6 lAL, England Division of Chemical Standards, National Physical Laboratory, Teddington, Middlesex Twll OLW, England National Spectrographic Laboratories Inc., 19500 South Miles Road, Cleveland, OH 44128, U.S.A. National Bureau of Standards, Office of Standard Reference Materials, Washington, DC 20234, U.S.A. Research Organic/Inorganic Chemical Corp., 11686 Sheldon Street, Sun Valley, CA 91352, U.S.A. ALFA Division, Ventron Corp., 152 Andover Street, Danvers, MA 01923, U.S.A. (Glen Creston, 16 Dalston Gardens, Stanmore, Middlesex HA7 lDA, England)

ISSN:0306-1353    

DOI:10.1039/AA9811100075

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

CHAPTER 4 Applications EXPLANATION OF THE TABLES Each of the Applications Sections, 4.1 to 4.9, is accompanied by a Table which summarizes the principal analytical features of the references from which the corresponding Section is compiled. All relevant references are included in the appropriate Table, while the accom- panying text discusses only the more noteworthy contributions. These Applications Tables form a convenient source of information for analysts interested in particular elements, matrices, sample treatments, or atomization systems. In many cases, sufficient detail is given for the analytical procedure to be followed; absence of such detail usually means that the information was not directly available to the compiler of the table, and the original reference should be consulted.The key to the tables is given below. ELEMENT A/nm MATRIX CONCENTRATION TECH. The elements determined are listed in alphabetical order of chemical symbol, except that, for space economy, multi-element applications (5 elements or more) are given at the end of some tables. The wavelength, in nanometres, at which the analysis was performed. An indication, necessarily brief, of the material analysed.The concentration range or level of the element in the original matrix, expressed as pg g-I for solids and mg 1 -' for liquids. The atomic spectroscopy technique is indicated by A (absorp- tion), E (emission), or F (fluorescence). ATOMIZATION The atomization process is indicated by the abbreviations A (arc), S (spark), F (flame), or P (plasma), usually with some additional descriptive detail, e.g. F, Air/C,H, or P, ICP. ANALYTE The form of the sample, as presented to the instrument, is indicated by S (solid), L (liquid), or G (gas or vapour). SAMPLE TREATMENT A brief indication is given of the sample pre-treatment required to produce the analyte. REF. The number refers to the main Reference section, which gives the title of the paper and the name(s) of the author(s), with address. i l l

ISSN:0306-1353    

DOI:10.1039/AA9811100111

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

112 Analytical Atomic Spectroscopy 4.1 CHEMICALS 4.1.1 Petroleum and Petroleum Products The total number of publications appearing in this field of interest now appears to have stabilized at a relatively low level, perhaps reflecting an appreciation of the real problem areas. The trend towards increased interest in wear metal analysis of lubricating oils continues, but fewer papers are dealing with gasoline analysis. 4. I . 1. I Petroleum A study of the matrix effects encountered in the determination of V in petroleum, by electrothermal atomization-AAS, was undertaken by Buenafama (C794). He concluded that, not only is the nature of the V compound important, but also that of the oil matrix and the other inorganic components,. Sasaki et al. (1 193) described a simple procedure for the analysis of petroleum for Na, Ni and V.Samples were diluted in a three solvent system, iodine added, and the metals determined by ETA-AAS. Equivalent responses from different organometallic species were claimed, together with freedom from interelement interference. Some problems found in the determination of Cu, Ni, and Pb in petroleum products, by ETA-AAS, have been investigated by Karwowska and co-workers (1 376).Solvent extraction procedures were problematical because of variations in sensitivity caused by the nature of the solvent and the chelating agent. Solvent penetration of the atomizer walls was prevented by use of a tantalum boat to contain the samples, Wallace (C1134) and Ediger et al. (C996) considered the problems encountered when the solvents commonly used to dilute oilsamples are aspirated into inductively-coupledplasmas. They investigated changes in plasma background structure caused by various solvents and recommended optimum operating conditions. 4.1.1.2 Lubricating Oils The problem of quantitative determination of suspended wear metals in lubricating oils has again received attention. Saba and Rhine (1594) investigated the efficiency of 9 nebulizers and 4 spray chambers for a suspension of Fe powder in synthetic lubricating oil.A ceramic nebulizer/cylindrical spray chamber combination performed best, quantitatively transporting particles up to 7pn in diameter and transporting 62% of particles in the range 7 - 1 4 ~ . Brown et al, (157) reported further work on the acid dissolution of wear metals prior to nebulization (ARAAS, 1980,10,110).Treatment with HF and aqua regia, dilution with MIBK and isopropyl alcohol and aspiration into a d.c. plasma gave 97 - 103% recovery for a number of elements. A study of the factors influencing wear metal analysis was presented by Marks et a/. (C1102). They considered the type of wear, sampling, sample preparation and the nature of the analytical instrumentation.They concluded that all techniques are useful in showing trends, but that none could be relied upon to give consistently accurate analysis. Winefordner and co-workers (582) described a novel continuum source AF system for the determination of Al, Cu and Mo in jet engine oils. The sample was atomized by an electrothermal atomizer contained within a flame and the elements excited by means of a xenon arc lamp.The application of an emulsion formation technique to the determination of Ca in lubricating oils was described by De La Guardia Cirugeda et al. (539). A commercial emulsifier was used and Sr(NO,), and HClO, added, to control interference from P, prior to anaylsis by FAES in the N,O/C,H, flamc. It is claimed that aqueous standards can be used for calibration. The same authors (538) described a similar procedure for the determination of Ba in lubricating oils using ICP-OES.Applications 113 4.I . I .3 Atomic Spectroscopy/Chromatography In two reports ((2816,1784) Messman and Rains demonstrated the applicability of LC-FAAS to the determination of 5 alkyl-Pb species in leaded gasoline.A reverse phase column with a water/acetonitrile mobile phase was used to separate the Pb species before aspiration into an air/C,H, flame. A major advantage of FAAS over other means of detection is the relative insensitivity to LC effects such as the solvent front. ETA-AAS has been used as an off-line LC detector, by Tittarelli and Mascherpa (1761), for the speciation of organophosphorous compounds.Matrix modification with Lam0 3 ) , was used to obtain acceptable ETA-AAS results and these were then compared with information from the U.V. detector to allow compound identification. 4.1.2 Chemicals and Miscellaneous Applications 4. I .2. I Sample Preparation Co-precipitation was employed by Kamat et al. (173) to separate and concentrate 13 trace elements in phosphorous trichloride and phosphorous oxychloride.The matrix was evaporated in a stream of N2 and the residual H,PO, treated with Bi(NO,), in order to collect the impurities on the precipitated BiPO, . The residue was analysed by d.c. arc-OES. The use of pyrazolone green dyes for the co-precipitation of Pd, Pt and Rh, prior to determination by d.c. arc-OES, was described by Chelnokova et al.(1506). Quantitative co-precipitation was claimed with the iodide salts of the dyesin0.2-0.5~H,SO,. The use of a Pt wirefilament to preconcentrate Se from sulphuric acid was reported by Holen et al. (2141). The Se was deposited by controlled-potential electrolysis and the analysis was subsequently carried out by electrically heating the filament in an Ar/H2 flame for FAAS.Hoschino et al. (562) preconcentrated Ag, Cr, Cu and Pb on various wire fdaments, for ETA- AAS analysis. Impregnation was achieved by soaking, and the effects of immersion time, pH and temperature were investigated. 4.1.2.2 Atomic Absorption Methods The determination of traces of As, Baand Hg inpolymericmaterials was described by Rombach ei al. (545). Samples were decomposed with HNO in a PTFE lined bomb.Hydride generation AAS was employed for As determination and cold vapour AAS for Hg. Three papers have appeared describing the determination of Pb (591), Sn (1782) and Cd and Zn (1707) in poly(viny1 chloride), by FAAS. Lead was determined in the air/C2H, flame after precipitation as PbSO, and dissolution in HNO . For Sn, samples were decomposed with H SO, /H 0 , , prior to aspiration into the N,O/C,H, flame.Decomposition with H,SO,/HNO, was used on samples for the determination of Cd and Zn. Danchev et al. (475) described the construction and operation of a device for the transfer of a weighed polymer sample into a graphite tube furnace. The vacuum actuated device quantitatively transfers mg samples in approximately 3 s and also removes decomposition products.Urbain and Martin (677) described the use of the discreet sample nebulization technique for the analysis of catalyst preparation solutions. The kinetics of impregnation of Pt and Ru onto alumina were followed by sampling 1004 aliquots of solution for analysis, by FAAS, in the N20/C,H, flame. Lanthanum sulphate was added to overcome interference effects.In the FAAS determination of Mo in iron based catalysts (597), addition of K 2 S 0 , was found to suppress interference from Al, Ca, Fe and W in the N,O/C,H, flame. The use of Zrcoatedgraphite tubes for the ETA-AAS determination of Pb (4) and Sn (21 8) in organometallic compounds has been reported. In both cases coating was achieved by overnight soaking in a Zr solution, followed by heat treatment. Increased sensitivity of between 30 and 100 times was claimed.114 Analytical Atomic Spectroscopy 4.1.2.3.Atomic Emhion Methock Heinrich ((282) determined 26 elements in paints, using ICP-OES. Automatic computer correction of interelement interferences was reported to provide results of acceptable accuracy. In contrast, Wong (2037) undertook an investigation into the determination of Ce in paint.After studying a number of effects, including viscosity and interelement interferences, he concluded that matrix matching of samples and reference solutions was required to obtain acceptable results by The use of the ICP in the analysis of nuclear fuels was described by Karnowski and Berg (1833). They reported that, in comparison to AAS, the ICP is particularly suited to the determination of B;Dy, Eu, Gd, Pu, Sm, Ta, Tc, Th, U, Wand Zr.A method for the analysis of solutions used in actinide removal plants was presented by Stone and Olsen (1715). Cerium was determined by ICP-OES in a variety of liquid matrices at between 8 x lod and 3 x 104g 1 - - I . Capdevila and Roca (1484) determined 29 elements in enriched uranium solutions using d.c.arc spectrography. The uranium was first removed from the test solution using tributyl phosphate or trioctylamine. Spectrographic conditions for 18 trace impurities and 14 rare-earth elements in uranium based materials were proposed by Lordello et al. (295). Five different carriers were investigated for the trace impurities and a separation technique, based on adsorption on to alumina, described for the rare earths.A system for the real time determination of K and Na in coal gas and fluidized bed combustor exhausts was described by Fassel and co-workers (C74). A portion of the gas stream was sampled into either a N20/CzHz or O,/H, flame every 2 to 3 min and the FAES result was used for process control or investigation.ICP-OES. 4. I .2.4 Atomic Spectroscopy/Chromatography A study of the most suitable conditions for pneumatic nebulization, for HPLC-AAS combinations, was undertaken by Koropchak and Coleman (144). They showed that adjustment of the FAAS nebulizer to apply a slight back pressure improved signal-to-noise ratio, transport efficiency and reduced gas bubble formation in the LC system.Size exclusion chromatography coupled to ETA-AAS was used by Parks et a/. (1907) to follow the disappearance of low- molecular-weight species during the copolymerization of tributyltin methacrylate and methyl met hacry late. The separation and quantification of condensed phosphates in detergent materials by HPLC-ICP was reported by Ogino et al. (C818). An anion-exchange resin was employed as stationary phase to separate ortho, pyro and tripoly phosphates. An anion-exchange resin was also used by Hirate et al.(C725) as the stationary phase to separate trace elements from acidic uranium solutions prior to ICP analysis. An ICP was used in combination with size exclusion chromatography in an attempt to identify organometallic species in coal (2032). Various solvents were employed to extract the species before evaporation.This was followed by re-dissolution in pyridine and the production of chromatograms for 15 elements. Three papers have appeared from Barnes and co-workers describing combined GC -plasma emission applications. The Friedel -Crafts catalysed redistribution of alkyl groups among Ge, Pb, Si and Sn atoms was followed by separation on both packed and capillary columns and detection by OES using either a d.c. Ar plasma or atmospheric pressure He MIP (225).A fused silica wallcoated capillary column was used in conjunction with an atmospheric pressure He MIP to separate and quantify cyclopentadienyl and carbonylcyclopentadienyl compounds of Co, Cr , Fe, Mn, Ni and V (237). A similar system was used to separate and detect the boronate derivatives of various diols (2086).Addition of H, to the He plasma gas was claimed to improve the plasma characteristics for R determination.Table 4.1A PETROLEUM AND PETROLEUM PRODUCTS k 8 Analyte Form g. Technique, Element Mnm Matrix Concentration Atomization, Sample treatment Ref. 5 Al 394.4 As - Ba 494 Lubricating oils Fuel oil Lubricants Ba Ca Ca - Lubricating oil 422.7 Lubricants 422-7 Lubricating oils - Ca c u 324.7 Li 671 Lubricating oil Lubricating oils Oils Oil - shale gases Petroleum Lubricants F, ETA - Flame (Air/C,H,, N,O~C,H,), L F, ETA - Flame (Air/C,H,, N,O/ A, ETA, L CZH3, L A, ETA, G From 5 nqlg A, cold vapour, G Ash 1 pl for 15 s at 800 OC on graphite filament then atomize into the Ar-sheathed flame.A wavelengthmodulated, continuum source spectrometer was used Study of sample preparation and analytical procedures for ETA and hydride methods Cover 2- 3 g with 2 g La,O, and ash in a platinum crucible with a burner. Calcine at 550 OC for 1 h in a furnace. Dissolve melt in HNO,/H,O and dilute to < 1 mglml La Emulsify and analyse with aqueous standards See Ba, ref. 222 Emulsify 0.1 g with a 10% solution of Emulsogen MS12 in benzene (2 ml).Add 4% aqueous solution of ethoxylated nonylphenol (5 ml), 1 -5% Sr(NO,), (5 ml) and 0 . 4 ~ HCIO, (25 ml), followed by 1 YO KCI (10 ml). Make to 100 ml with H,O and add few drops of ethanol Addition of KOH in 2-ethylhexanoic acid to sample and standard solutions diluted with organic solvent See Al, ref. 582 Use of tantalum boat in carbon tube for atomization of various chelate extractants Zeeman-AAS used to detect Hg in gases heated to 9QO OC in furnace tube HNO,/H,SO, decomposition in closed vessel, followed by heating to fumes on addition of 20 ml HNO, See Ba, ref. 222 c760 222 538 222 539 2164 582 1376 257 1783 222 - c.I m Table 4. I A PETROLEUM AND PETROLEUM PRODUCTS-continued Technique, Analyte Form Element Nnm Matrix Concentration Atomization, Sample treatment Ref.Mo hla Ni Ni Ni Ni Ni P Pb Pb Pb Pb Pb S S se v v v v Lubricating oils Crude oil, petroleurn Oils Heavy oils Oils Crude oil, petroleum Oil Lubricating oils Gasoline Gasoline Oils Gasoline Gasoline Oils Oils Fuel oil Heavy oils Crude oils Oils Crude oil, petroleum F, ETA - Flame (Air/C,H,, N,OIC,H,), L A, ETA, L E, P, ICP, L E, P, ICP, L E, P, ICP, L A, ETA, L A, ETA, L A, ETA, L E, f? L A, - 1 L A, ETA, L A, F, Air/C,H, L A, - I - E, ICP, L E, ICP, L A, ETA, L A, -, G E, P, ICP, L A, ETA, L E, P, ICP, L A, ETA, L See Al, ref. 582 Dilute with a mixture of xylenelMlBKlmethanol (5:4: 1) Scanning double monochramator used Dilute tenfold with MlBK Study of the effect of organic liquids on spectral characteristics See Na, ref. 1193 See Cu, ref. 1376 HPLC- ETA for speciation of organo phosphorus compounds; La(NO,), used for matrix modification Emulsify and analyse with aqueous standards LC - AAS system See Cu, ref. 1376 LC - AAS method using MeCN/H,O mobile phase GC - AAS system See Ni, ref. C996 Comparison of effect of xylene and MlBK diluents on background spectrum See As, ref.C760 See Ni, ref. 575 Study of matrix effects See Ni, ref. c996 See Ni, ref. 1193 582 1193 C27 575 c996 1193 1376 1761 538 C816 1 376 1 784 1821 L - s g 575 2. s c1136 c760 s c m g 1193Various (5) Various (1 1) Various Various (5) Various (1 1) Various Various (13) Various (5) Various Various Lubricating oils - A, F, L E, D.c., P, L - Fuel oil Trace levels - Oil Trace levels - Jet engine oil Trace levels - Motor oil - - Petroleum - - Lubricating oil - - Oils, solutions - - Lubricants - - Oils - Oils Trace levels - A, ETA, L E, P, ICP, L A, F, Air/C,H,, L A, F, N201C,H,, L, E, P, ICP, L 4 - 7 - ME, -, - Digest with HFlaqua regia at 65 OC for 45 min with ultrasonic agitation.Dilute with MlBK and isopropyl alcohol for Al, Cu, Fe, Mg, Mo, Ni, Sn, Ti Comparison of conventional and capacitive discharge ETA techniques involving dry- ashing/dissolution and no preparation, respectively (Cu, Fe, Mn, Ni, Pb) Dilute 10fold with xylene (Al, Ca, Co, Cr, Cu, Fe, Mg, Mn, Pb, Si, Zn) Comparison of techniques involving dilution or acid decomposition of sample (Al, Cr, Fe, Ni, Si) Mix with graphite powder (1 : 4), heat to 220 "C, mix with 6% NaCl and place onto electrodes of double arc in a graphite tube Comparison of sample introduction systems for atomic absorption and emission techniques (Al, Ca, Cr, Cu, Fe, La, Mg, Mn, Ni, Pb, Sc, Ti, v) (Cu, Fe, Pb, Sn, Zn) Dilute with MlBK Review (6 references) b 157 h z g 3.C781 2 934 c1102 1166 1386 1 594 1656 1 822 1893 1951- c. 00 Table 4.1B CHEMICALS AND MISCELLANEOUS MATERIALS Technique, Analyte Form Element Alnm Matrix Concentration Atomization, Sample treatment Ref. 328.1 Ag - containing 0.05 - 2 pg (absolute) E, S, L Dissolve in HNO,, add 0.3 g KBr and 100 pg TI as internal standard (276.8 nrn). Adjust acidity to 0 - 5 ~ , and extract Ag with diantipyrinylmethane in CHCI, and light petroleum. Apply a small volume to the carbon electrode Ag material Al Al As As As As Au B - Phosphoric acid and From lo+% phosphate salts 308.2 Phosphoric acid- - based liquors - Potassium chromate From 193.7 Sulphuric acid pgll level 193- 7 Acrylic fibres 0.04-400pprn 193.7 Plastics Trace levels - Semi-conductor Trace levels 242.8 Silica-based catalysts 0.05-0.5% silicon Carborane - silicone - polymer A, F, Air/C,H,, L E, P, D.c., L E, A, D.c., S F, F, ArlH,, G A, F, Air/C,H,, L E, P, MIP, G Neutralize solution with NH,OH, heat to 60 "C, mix with ZnS suspension, filter, dissolve with 0.1 M HCI, evaporate and dilute with H,O Digest with HNO,IHCI/HCIO,, add Ge as internal standard and dilute with H,O Mix with graphite powder Hydride passed into flameheated quartz tube Dissolve in mineral acids, reduce As(V) to As(lll) with Ti(lll) and separate from Sb with benzene.Generate hydride with KI, SnCI, or Zn powder Decompose 0-4 - 1.1 g polycarbonate, polyethylene, polypropylene or PVC by heating with 65% HNO, (10 mi) at 240 OC for 8 h in a PTFE bomb. Dilute to 100 ml and determine As in a 10 ml aliquot by hydride generation See Sb. ref. 1019 Decompose 0.1 g powder with 40 mi HCU HNO, (3: 1) with heating, cool, and filter.Wash insolubles and combine washings and filtrate. Extract Au with 10 ml isoamyl alcohol shaken with 50 ml solution GC - MIP system for analysis of pyrolysis products of matrixBa Ba Ba Ba Bi Br C, Ca Ca Ca Ca 456.4 - - 230.4 233.5 455.4 306.7 (Cu Br 434.1) - 422.7 317.9 422.7 422.7 Plastics Gunshot residue Trace levels Trace levels El P, ICP, L A, ETA, L See As, ref. 545 Wash subjects hands with dilute HNO,. Add NaCl to sample and standard solutions Analysis of hand swabs Dissolve 5 g in 20 ml H,O, add 1 ml acetate buffer, 2 ml. K,Cr,O,, 2 ml PqNO,), to precipitate Ba Collect, wash, and place precipitate on electrode b 545b 1200 3 g 1767 g' 2053b 209 1806 c1450 258 548 559 571 1182 1 284 1673 144 - L Gunshot residue Alkali halide salts A, - 1 - E, A, A.c., S - 104 - 10-3% Graphite El Grimm: discharge, S El F, O,/H,, L Organic compounds Dissolve in DMF/Me,CO (1 : 1) containing Study of polymer combustion by observation of C, emission from vapour above polymer sheet Dissolve and extract Ca and Mg at pH12 and Cu and In at pH2 with an equal volume of 0.05~ quinolinaol in isoamyl alcohol See Al, ref. 548 CU(N03)2 Polymer combustion products A, F, AirfC,H,, L Potassium chloride Trace levels Phosphoric acid- based liquors Sodi urn hydroxide solutions El P, D.c., L - pg/ml levels A, F, Air/C,H,, L A, F, Air/C,H,, L N * OlC, H 2 I Acidify with HCi; with air/C,H, flame, add LaCI, to suppress interferences Phosphoric acid and phosphates pg/ml level (in solution) Dissolve, adjust to pH13 with NaOH, treat 2 - 6 ml with 10 mi 1 M KCV1 M NaOH (pH13) and 5 ml H20/2-butoxyelthanol (1 : 1) and extract Ca with 10 ml of 3% quinolin-801 in MIBK.Back extract with 8 ml4% EDTA solution, add Sr solution to the aqueous fraction and dilute to 25 ml with H,O. Measure Sr at 460.7 nm as internal standard Dissolve in 0.1 M NaOH - Ca Nit roxi ni I A, F, Air/C2HZ, L A, F, L El A, D.c., S A, F, L - Ca Ca - Insulation cellulose Potassium chromate salts Dilute HNO, 30 Ps/Q From 10 +O/O Mix with graphite powder LC - AAS system Cd 228.8c- Table 4.1 B CHEMICALS AND MISCELLANE.OUS MATERIALS-continued m 0 Technique, Analyte Form Element Nnm Matrix Concentration Atomization, Sample treatment Ref.- Cd cd E;e - ce co 236.4 - - co co co Cr Cr Cr Cr c u c u 240.7 304 401 240.7 324.7 357.9 - 357.9 359.3 324-7 PVC composites - Phosphoric acid and From l O 4 O / ~ phosphate salt Plant st reams - Commercial paints CdMdcatal yst - on alumina 0.2 - 700 pg/I Dyes 5% iron oxide 50-8Oollglg Propylene oxonation - materials Boron oxide pglg level Dyes 3.5% Gallium arsenide Trace levels Catalysts 1 *7'/0 Boron oxide pglg level Potassium chloride Trace levels A, - 9 - A, F, Air/C,H,, L Comparison of acid dissolution procedures See Ag, ref. 1972 E, P, ICP, L - E, P, ICP, L E, P, D.c., L Ash and dissolve residue in HNO,IHCI/H,O, Calcine 1 g of solvent-washed catalyst at 500 OC for 1 h, cool, reweigh and dissolve in 50 mi 1 : 1 sulphuric acid; transfer to 500 ml flask, add LiNO, solution and dilute with H,O; prepare standards containing lo00 pg/rnl Al to match matrix A, F, Air/C,H,, L E, S, D.c., S A, F, Air/C,H,, L A, ETA, L A, F, Air/C,H,, L A, ETA, L 4 - 9 - A, ETA, L A, F, Air/C,H,, L Dissolve sample, or acid digest of sample, in H,O, aq. 50% ethanol or aq. 50 O/O acetone Compress into 6 mm diameter pellet and attach to electrode with methyl methacrylate Comparison of dry ashing and MIBK dilution Dssolve to give a ln/0 B,O, solution in 0.02M %I See Co, ref. 276 M icrocomputer-AAS system Dissolve 0.1 g in H,SO,/H,PO, (1:l)with slow heating of solution. Dilute to 50 ml with addition of Na,SO, to give a 2% concent rat ion See Cr, ref. 176 See Ca, ref. 250 1 707 1972 1715 2037 c106 276 607 1330 1 76 276 b c852 g 3 & 9 $! 1928 -. 176 2. 250 I?. a 4c u c u c u c u c u c u Eu Fe Fe Fe Fe Fe Fe Fe Fe Ge H H Hg - 327 396 324.7 - - - 459.4 248.0 371 -9 248.3 248.3 - L - - 265-1 - - 253.6 Sodium nitrate - A, ETA, S Iron oxide 5-8oPs/s Archaeological - bronzes Pharmaceuticals and - cosmetics Pharmacet ical - tablets Phosphoric acid and From lo4% phosphate salts Gadolinium oxide 0.02% Boron oxide pg/g levels Phosphoric acid- - based liquors Phosphorites H,PO, - Propylene oxonation - material Vitamin-mineral - tablets Potassium chromates From Pharmaceut ical - tablets Copper sulphate - Tet ra-at kyl - organometal I ic compounds Hydrogen isotopes - Helium - Plastics Trace levels E, S, D.c., S A, F, Air/C,H,, L A A, - 7 - A, F, Air/C,H,, L E, A, A.c., S A, ETA, L E, P, D.c., L A, F, Air/C,H,, L A, F, Air/C,H,, L A,-,- E, A, D.c., S A, -, - A, F, Air/C,H,, L E, P, D.c., MIP, G A, cold vapour, L Preconcentration of Cu(lI) ions on Mo, Re, Ta, and W wires See Co, ref. 607 Dissolve in HNO,, aqua regia or HCllHF Determination of alkannins by formation of Cu complex See Ag, ref. 1972 Mix with graphite powder See Cr, ref. 176 See Al, ref. 548 Adjust standard solutions to match Ca : Fe ratio of 70 : 1 for phosphorites Comparison of dry ashing and MlBK dilution Mix with graphite powder - Separate tetra-alkyl compounds of Ge, Pb, Si, Sn by GC Isotopic composition determined with h.f.(1-67 MHZ) discharge at 0.07- 13’3 kPa Image dissector detection with h.f. discharge plasma See As, ref. 545 - L 562 b 2 5 607 $ 932 1650 1962 1972 500 1 76 548 1272 1330 1 593 1673 1962 1971 225 1364 1369 545 c. E3L- Table 4.1 B CHEMICALS AND MISCELLANEOUS MATERIALS-continued h) N Technique, Anaiyte Fonn Element Unm Matdx concentratkm Atomization, Sample tmatment Ref.In In K K K K K K K K Li Li Li Mg Mg Mg 303.9 451.1 303.9 - 766.5 344.67 - 766.5 - 766.5 - 670-8 670.8 - 285.2 - 285.2 Potassium chloride Trace levels Indium 0.007 - 0.4'/0 acetylacetonate Coal gas and ppb levels combustor exhaust Propylene carbonate mg/ml level and tetrahydrofuran Potash Major Insulation cellulose 10 pag Antimony chlorides 0.01 - 0.1 pg/g Ammonium salts - Lead oxide Trace levels Membrane vesicles - Propylene carbonate mg/ml level and tetrahydrofuran Antimony chlorides 0.01 - 0.1 ps/s Membrane vesicles - Potassium chloride Trace levels Anode paste - Phosphoric acid- - based liquors A, F, Air/C,H,, L A, F, L E, A, A.c., L 4 F, L E, F, Air/C,H,, L E, F, L E, F, 1.A, F, Air/C,H,, E, F, Air/C,H,, L E, F, L A, F, Air/C,H,, L 4 - 1 - E, P, D.c., L -- See Ca, ref. 258 Dilute to 100 ml with addition of 5 ml50% HCI Online simultaneous analysis for K and Na Dilute 100- lo00 times with H,O, filter to remove naphthalene precipitate and analyse with aqueous standards Dissolve 20 - 30 g in H,O and apply a few drops to the carbon electrode Evaporate in a stream of argon, add HBr to remove all the Sb, evaporate to dryness and dissolve residue in 1 M HCI.Add isopropanol to final solution Decompose with heating and dissolve residue in 10 ml H,O Dissolve in acetic acid Use of vidicon detector See K, ref. 129 See K, ref. 1359 U s e of vidicon detector See Ca, ref. 258 Sample from an alkaline accumulator See Al, ref. 548 258 1466 c74 129 316 1284 1359 1683 1747 1902 129 k L" 1359 8 1902 $ 258 2. 496 1 $ s 4 m 548Mg Mg Mg Mo Mo N Na Na Na Na Na Na Na Na Na Na Na Nb 285.2 270 280 202.0 313.3 - - - 589.0 330.3 589.0 - 589.0 - 589.0 589.0 - - - Sodium hydroxide solutions Iron oxide Insulation cellulose CdMo catalyst on alumina FebaSed catalysts Helium Coal gas and combustor exhaust Propylene carbonate and tet rahydrof uran Potash Surfactants in water Insulation cellulose Antimony chlorides Ammonium salts Lead oxide Membrane vesicles Catalysts Pol yacrylo nitrile fibres Lithium metaniobate pg/ml levels 0.05- 10% - ppb levels mg/ml level 0 * 18 - 7 * 5% (as Na,CO,) 50-800pg/l Trace levels - 0.59% 0.42% A, F, Air/C,H,, N,O/C,H,, L E, S, D.c., S A, F, L E, P, D.c., L A, F, L E, F, Air/C,H,, L E, F, L See Ca, ref. 559 See Co, ref. 607 - See Co, ref. C106 Add K,SO, to final solution to prevent interference from Al, Ca, Fe and W Image dissector detection with r.f. discharge plasma On-line simultaneous analysis for K and Na See K, ref. 129 See K, ref. 316 Linear alkyl benzene and di.(2-ethylhexyl-)- sulphosuccinate were determined by measurement of Na in MlBK - extracts of the sodium “Ion-pair‘’ complex of each surfactant See K, ref. 1359 - Decompose with heating and dissolve residue in 10 ml H,O Dissolve in acetic acid Use of vidicon detector Treat 0.1 g with 2.5 ml HCI (1 : l), heat on H,O bath, filter, dilute to 50 ml and analyse - See Ti, ref. 322 569 607 1284 c106 597 1369 c74 129 316 963 1284 1359 1683 1747 1902 1928 1950 322 p3 wTable 4.1B CHEMICALS AND MISCELLANEOUS MATERIALS-continued c.E Technique, Anaiyte Fonn Element A/nm Matrix Concentration Atomization, Sample tmatment Ref. Ni Ni Ni Ni P (HPO) P P P P Pb Pb Pb Pb Pb Pb 341.5 231.7 305 582 232.0 - - - 213-6 213.6 217.0 283-3 368.3 217.0 283.3 - - - CdMo catalyst on alumina Boron oxide Iron oxide Propylene oxonation material Detergents and inorganic materials Steel N i-plat i ng bsths Detergents Calcium phosphate Organolead compounds Tet ra-al kyl organometallic compounds PVC Organic materials Gunshot residues Pharmaceutical tablets E, P, D.c., L A, ETA, L E, S, D.c., S A, F, Air/C,H,, L E, F, H,-based, L A, -, - 9 E, P, ICP, - El P, D.c., L E, P, ICP, L E, P, ICP, G A, ETA, L E, P, D.c., MIP, G A, F, Air/C,H,, L See Co, ref.C106 See Cr, ref. 176 See Co, ref. 607 Comparison of dry ashing and MIBK dilution A MECA application Comparison of techniques with the same sequential spectrometer HPLC- ICP system for separation and determination of condensed phosphates Reduce to Ca,P, by heating with A1 powder to 1300 OC in a Ta- C tube atomizer, treat with 2.7M HCI to produce PH, HPLC - ETA speciation using Zr-treated graphite tubes, addition of I, allows use of aqueous standards See Ge, ref. 225 Review of methods with 30 references; best methods- (i) decompose polymer with 95% H,S0,/30% H,O, to precipitate Pbso,, dissolve in dilute HNO,; (ii) remove plasticisers by soxhlet extraction with ethyl ether, then decompose residue with HCIO,/HNO, Comments on sample preparation methods Analysis of hand swabs c106 1 76 607 1330 c112 c643 c738 C818 1480 4 225 591 ' k 5 2 0" 3 5.(378 9 1767 '- Pb Pb Pd 247-6 Pd 342.1 - Pt Pt Pt Rh Ru Ru S S Sb Sb 265.9 266.0 266.0 343.5 349.9 - - 394 259.8 - Copper sulphate - Phosphoric acid and From 10-50/~ phosphate salts Si I ica-based 0-1-1.0~/0 electrolytes Nickel electrolytes - Electrolyte solutions pglml levels Alumina support for WRu catalyst 50 - 20 OOO pglml Nickel electrolytes - Nickel electrolytes - Alumina support for WRu catalyst Anode coatings - 50 - 20 OOO @mi Steel - Organic compounds - Graph it e - Semi-conductor Trace levels silicon A, F, Air/C,H,, L A, F, Air/C,H,, L A, F, Air/C,H,, L E, A, S E, ICP, L E, F, G E, Grimm- discharge, S F, F, Ar/H,, G See Ag, ref. 1972 See Au, ref. 1465 Coprecipitation with iodide salts of bis (4-di methylaminophen yl) (1 -phenyl3met hyl4 chloropyrazolone5il) carbinol and associated compounds; precipitates mixed with GeO, as carrier, Co as internal standard, and graphite Rate of dissolution of PI electrode in ammoniumbased solutions measured Study of adsorption on alumina; add 10 g alumina to 50 mi HCI solution of PtCI,z-/R~C1,2- (2-2.5 911); stir and remove 100 pI aliquots periodically; add 150pl La,SO, solution to each and inject 50 pl volumes into flame See Pd, ref. 1506 See Pd, ref 1506 See Pt, ref. 677 Remove from electrode with molten KOHl KNO, (so0 OC), add metallic Ti to give TilRu ratio of > 24:1, dissolve melt in dilute HCI and analyse GC flame photometric detection Attach Si slice to Cu electrode, produce oxide layer by anodic oxidation and dissolve oxide in HF; add KI and HCI to prepare solution for hydride generatiodnon-dispersive AFS.See Si, ref. 1019, for determination of Si dissolved as oxide 1971 ’ 1972 8 1465 cA 1506 8 8 259 677 1506 1506 677 1 726 c643 21 52 209 1019 a29 c h)I w a? Table 4.1 B CHEMICALS AND MISCELLANEOUS MATERIALS-continued Technique, Analyte F m Element Mnm Matrfx Concentration Atomization, Sample treatment Ref.Sb Sb Sb sc se se Se Si Si Si Sn Sn Sn Sn Sn Gunshot residue Polyester fibre Gunshot residue Phosphates Chemicals and plating baths Sulphuric acid Sulphuric acid Tetra-alkyl organometal I ic compounds Semi-conductor silicon Barium nitrate, sodium chloride Methanol ic solutions of organotins Tetra-al kyl organometallic compounds KOH solution PVC Polymers Trace levels - - - pQll level 3 - 22 nglml 30-60pQll - From pdl level A, ETA, L A, F, filament in ArlH,, L El PI D.c., MIP, G E, P, ICP, L A, ETA, L A, ETA, L E, PI D.c., MIP, G F, heated atomizer, G A, F, NzOICzHz, L A, ETA, - See Ba, ref. 1200 Analysis of hand swabs Optimization study of hydride generation method Extract into toluene with an aromatic ediamine and add NI prior to atomization Dilute Sfold with H,O, electrolyse for 5 min to deposit Se on Pt wire filament work and dry filament. Inject Pt into flame and electrically heat wire See Gel ref. 225 - - See Sb, ref. 1019 Tungstencoated furnace tube used Study of Mo, V and Zr carbide coatings on graphite cup or tube atomizers See Gel ref. 225 Electrolytic reduction of Sn to SnH, H,SO,/H,O, decomposition Size exclusion chromatography used to follow copolymerization process 1200 1210 1767 277 C117 1402 2141 225 1019 1 749 21 8 ?? 225s $ 2 6 6 % 3 1782 8’ @ 1907so:- (Cr 357.9) Acids and sodium - chloride A, F, Air/C,H,, L Sr 460.7 Barium salts 0.ooo5 - 0*002% A, F, Air/C,H,, Ti - Lithium metaniobate 0.1 - 4% E, S, L L V V Yb Zn Zn zn Zr 311.8 CdMo catalyst on - 318.4 Carbon powder 0.001% alumina 369.4 Silica coating 0.01 - 0.05% 213.9 Ointments - - PVC composites - - Pharmaceutical - 360.1 Rareearth metal 0.6 - 85 pg/g tablets fluorides Various - Archival and - (6) library material E, P, D.c., L A, A, A.c., S E, A, s F, ETA, G A, ETA, S Various - Water-based paint Various levels E, P, ICP, L (26) Various - PCI, and 5 - ~ l .J 9 1 9 E, A, D.c., S (1 3) FOCI, - Alkali salts 10-8- 10-3% A, F, L Various - (6) Indirect method involving addition of barium chromate to precipitate Sop and measurement of Cr Polish sample, place in PTFE dish and heat in a glycerol bath; treat surface with HN0,lHF (2 : 1) to etch surface; evaporate the resulting solution, moisten residue with a 0.012 mg/ml Zr solution (internal standard), reevaporate and dissolve residue in 0.03 ml of HN0,lHF See Co, ref. C106 Study of processes in an ac.arc by atomic absorption Laser evaporation of 0.3 pg sample which was deposited on an electrode Dissolve in benzene and emulsify with TWEEN 80 Comparison of acid dissolution procedures Heat 50 mg with CIF, in a PTFE micro autoclave and evaporate volatile components; heat the residue in a corundum boat to 2ooo K, pass evaporated ZrF, into a tantalum atomizer at 2OOO K and mix with Na vapour to give atomic Zr; excite fluorescence with a xenon lamp Optimization study for B, Ca, Cu, Fe, Mg, Na Shake for 1 h and dilute 1Ofold by volume with H,O Evaporate 25 ml in a stream of nitrogen, treat residual phosphoric acid with Bi(NO,), solution to coprecipitate impurities with BiPO, Dissolve in H,O and extract with PMBPl MlBK for Co, Cu, Fe, Mn, Ni, Zn 1748 5. 5 322 c106 205 299 910 1707 1962 323 C22 ca2 1 73 184 c1 N 4L Table 4.1B CHEMICALS AND MISCELLANEOUS MATERIALS-continued Iu bz Technique, Analyte Form Element Alnm Matrlx Concentfa tion Atomization, Sample treatment Ref. Various (8) Various (1 0) (rareearths) Various Various (6) Various (5) Various Various Various (5) Various (9) Various - XRF methyl- - cellulose film standards concent rates - Samarium 0.01 - 1 mg/ml - Organic compounds - - Organometallic pg - ng levels compounds (absolute) - Wool fibres Trace levels - Alkali-metal salts Trace levels - Potassium hydrogen Trace levels carbonate - Lithium carbonate Trace levels - Oxalatoniobic acid Trace levels - Boric acid Trace levels E, P, ICP, L E, S, L A, F, L E, P, MIP, G E, A, S E, A, D.c., S E, A, D.c., S E, A, D.c., S E, A, D.c., S E, A, S Evaluation of JY-38P spectrometer for Be, Cr, Cu, Mn, Ni, Pb, Sr and Zn Porous electrode used for Ce, Dy, Eu, Gd, Ho, La, Nd, Pr, Tb, Y Review of recent literature with 94 refs.Separate by GC on a fused-silica WCOT capillary column connected to a TM,, ,, cavity for Co, Cr, Fe, Mn, Ni, V Ash, add graphite as buffer and O.l0/o In,O, as internal standard, and measure Cu, Fe, Mn, Pb, Sn by standard addition Electrodeposit impurities as electrodes in a 0- 1 @ml solution of the matrix Dissolve 2 g in 15 ml HCI, adjust pH to 5, coprecipitate metal ions with NaDDC, 1-nitro Bnaphthol and Cd2 + in presence of powdered graphite; filter after 12 h, arc in presence of 3 mg NaCl Similar to Various, ref. 280; 8hydroxy- quinoline used, not l-nitroso2-naphthol, for Co, Cr, Mn, Ni, V Dissolve to give a 0.05 g/ml solution of the matrix, electrolyse to deposit impurities and arc carbon electrode in presence of NaCl for Bi, Ca, Cr, Cu, Fe, Mg, Mn, Si, V Dissolve 5 g in 100 ml H,O and preconcentrate on a column of KU2 resin; wash with H,O (3 times), elute cations with dilute H,SO, and evaporate with powdered graphite in the eluate 195 204 224 237 263 279 280 283 5 286 s $ 2 9 0 ; 8 s. 3 8 8Various - (32) Various - (8) Various - (9) Uranium-based - materials Graphite powder - Titanium tetrachloride - - Aluminium Trace levels ammonium sulphate Various - (5) Various - (5) Various - (6) Various - Various - Various - Various - E, A, D.c., S E, A, D.c., S E, A, A.c., S E, A, A.c., S Paper additives - A, ETA, L Ammonium fluoride 3 - 30 pglg E, A, D.c., S (So - 600 pglg for Si) Organic liquids Trace levels Uranium oxide Trace levels E, P, ICP, L E, P, ICP, L High purity graphite Trace levels E, A, D.c.-in- tube, L or S Waveguide materials Trace levels A, ETA, G Art objects - Hydrofluoric acid - solutions E, Laser, S E, ICP, L 321 For rareearth elements, separate from U by sorption on column of AI,O,; arc without carbon or buffer Study of optimum conditions for Ce, Eu, Gd, La, Nd, Pr, Sm, Y Hydrolyse with aq.NH,, ignite precipitate at 850 OC, and form a 9 : 2 : 1 mixture of sample: AgCl: carbon (containing 0.01 O/O Co); determine Al, Cr, Fe, Mn and Sn with Co as internal standard and Mo, Si, V, Zr with Ti as standard Heat 15 g and 0.5 ml H,O in a crystallization concentrator, cool to- 35 OC, dissolve 0.25- 0.5 g of solidified melt in 2 ml H,O and evaporate under i.r.lamp; ignite residue at 400 OC then 1100 OC, and mix residue 2 : 1 with carbon powder for Co, Cr, Cu, Ga, Mg, Mn, Mo, Ni, Pb, Ti and U Decompose 2 g by heating with 8 ml65% HNO, at 160 OC overnight in a PTFE bomb for Cd, Cr, Cu, Pb and Zn 5 ml 1 O/O KCI, evaporate mixture and decompose NH,F at 300 OC.Analyse the resulting concentrate for Cu, Fe, Mn, Pb, Si Mn, Ni, Zn) Dissolve with 7M HNO, and a trace of HF in aTeflon bomb. Separate impurities from U by ion exchange 295 31 0 313 576 Treat 5 g plus 5 g powdered carbon with 605 Dilution with organic solvents (Cd, Cu, Fe, C641 C725 Double d.c.arc arrangement in carbon tube c736 Volatilization of matrix followed by gas c840 phase preconcentration of impurities prior to desorption into furnace atomizer (Co, Cr, Cu, Fe, Mn, Ni) Applications of laser microspectral analysis Use of a Teflon nebulizer and spray chamber C977 C1140 L p3L Table 4.1B CHEMICALS AND MISCELLANEOUS MATERIALS-continued w 0 Technique, Analyte Form Element Nnm Matrix Concentration Atomization, Sample tmatmnl Ref.Various (5) Various Various !16) Various (1 0) Various (29) Various (9) Various (5) Various (7) Various (5) Various (7) Various (6) Various (7) Various (7) - Aluminium chloride - - Insulation cellulose - - Pure graphite powder - - High purity lead salts 0.1 pglg - Nuclear fuel (U) - - Activated charcoal 40 pglg - 20 mglg - Phosphorus - Trace levels containing resins - CaHPO, - - Fe2(S04) 3 - - Neodymium chloride - - Barium nitrate - - Acetylacetonates Trace levels - Carbonate salts Trace levels of Al, In, Zr A, - 7 - A, - 3 - E, A, S E, A, D.c., S E, A.L A, F, L E, A, D.c., S E, A, D.c., S E, A, D.c., S E, A, D.c., S E, A, D.c., S E, A, D.c., S E, A, D.c., S Extract Cr, Cu, Fe, Mn and Ni with NaDDC in iseBuCOMe from acidic solution (pH < 3.5).Oxidize Cr3 + to Cre + by KMnO, and determine separately Review articles 1 O h NaF added as carrier Mix with equal weight of graphite powder and 3% PbF,. Place 0.4 g in electrode (Ag, Bi, Co, Cr, Cu, Fe, In, Mn, Ni) Extract U with tributyl phosphate or trioctylamine Fuse with Na,O, and dissolve in acid (Ag, Au, Ca, Co, Cu, Fe, Ni, Si, Zn) Dry calcination (AJ, Ca, Fe, Mn, Ti) Mix 40 mg powder with equal weigh graphite (Co, Cr, Cu, Fe, Mn, Ni, V) Cu, Mg, Mn, Pb, Zn Mix 10 mg powder with 10 mg graphite and 1 mg NaCl (Co, Cr, Cu, Fe, Mn, Ni, V) Precipitate matrix, filter, evaporate the filtrate with powdered graphite and mix with NaCl (Co, Cr, Cu, Mn, Ni, V) Decompose in presence of powdered graphite at 300 - 500 OC (Co, Cr, Cu, Fe, Mn, Ni, V) Dissolve in H,O/HCI, adjust to pH 4 - 5, add Na DDC, extract with CHCI,, wash extract with H,O and evaporate in presence of graphite powder and NaCl (Co, Cr, Cu, Fe, Mg, Ni, V) 1276 1281 1403 1478 1484 1485 1642 1674 1675 1 676 1677 % z & 1678 $ 3 1679 5.@ 2Various (6) Various (3 Various (6) Various (12) Various (12) Various (1 4) Various (6) Various (5) Various (5) Various (8) Various (1 2) Various Various (6) Various Various (1 9) Various - Zn, Cd nitrates Trace levels - Alkoxides of Al, Ti, Zn Trace levels - Cs, K and Rb Trace levels biphthalates - Organophosphates Trace levels and phosphites - Reagents and high Trace levels purity material - Se I enou rea Trace levels - Lead silicate Trace levels - K and Na nitrates From 0.005°/~ - Sodium fluorosilicate Trace levels - Liquid coke Trace levels - Nuclear fuels - oven effluents - Nonconductor - - Ammonium bifluoride - products - Laboratory chemicals - - Refined coal I - OrganoSi polymers - process solvent E, A, D.c., S E, A, D.c., S E, A, D.c., S E, A, S E, A, D.c., S E, A, D.c., S A, F, N,O/CzHz, L E, A, D.c., S A, -, - AIE, .-, - E, ICP, L Concentration on powdered graphite by evaporation from solution (Bi, In, Mo, Sn, Ti, V) Dissolve in HCI or HF, heat to 60 OC with o-phenanthroline and NH,OH hydrochlorides, adjust to pH 3 - 4, collect on ionexchange resin, decomp.with H,SO,/H,O, and evap. on to graphite (Co, Cr, Cu, Fe, Mn, Ni, V) Decompose 2 g by addition of 2 ml H,SO, and evaporate to dryness. Calcine residue at 600 OC and mix 25 mg with 50 mg graphite (Al, Ca, Cr, Cu, Fe, Mn) Evaporate 2 g with 20 mg powdered graphite and mix with 1 mg NaCl (Al, Ca, Co, Cr, Cu, Fe, Ga, Mg, Mn, Ni, Sn, V) Preconcentrate by filtration and collect on graphite powder(A1, Ca, Co, Cr, Cu, Fe, Mg, Mn, Ni, Pb, Ti, V) Mix with graphite powder and NaCl (Ag, Bi, Ca, Co, Cr, Cu, Fe, In, Mn, Ni, Pb, Sn, Ti, V) Mix 150 mg with equal amount of graphite powder (Cr, Ca, Fe, Mn, Ni, V) Al, Mo, Si, Ti, W Dilute 1 : 4 with powdered graphite (Ca, Cu, Fe, Mn, Pb) Cd, Co, Cr, Cu, Mn, Ni, Pb, Zn Analysis of radioactive material in a glovebox (6, Dy, Eu, Gd, Pu, Sm, Ta, Tc, Th, U, W, Zr) Combination of ETA with plasma A review with one reference Study of parameters for Fe, Mn, Mo, Ni, Pb, and Si Size exciusion chromatography to separate metals, pvridine as eluant Dry solutions on electrode with i.r. lamp (Al, Ca, Cr, Cu, Fe, Mg, Mn, Ni, Sn, Ti) 1680 k % 1682 g 3. s 1 694 1695 1 743 1 744 1746 1750 1754 1828 1833 1871 1892 (21989 2032 2046 L w

ISSN:0306-1353    

DOI:10.1039/AA9811100112

出版商:RSC    

年代:1981

数据来源: RSC