摘要:

Analytical Communications, May 1996, Vol33 (7H-9H) 7H Highlight Determination and Speciation of Mercury in Natural Gases and Gas Condensates Wolfgang Frecha, Douglas C. Baxtera, Berit Bakkeb, James Snell" and Yngvar Thomassed a Department of Analytical Chemistry, Umed University, S-901 87 Umeh, Sweden h National Institute of Occupational Health, P.O. Box 8149 Dep. N-0033 Oslo I , Norway Mercury might be present in metallic (Hgo), organic or inorganic ionic forms which all show unique species-dependent physical, physiological and chemical properties. Species- specific information is thus essential for adequate risk assess- ment, process control, methodological developments and qual- ity assurance of analytical results.At present it is not well known in which chemical forms mercury is present in natural gases and condensates, and, in addition, methods for the determination of total mercury concentrations must be regarded to be of unproven reliability due to the lack of adequate standard reference materials and poor accuracy.' This deficiency of knowledge and methodo- logical experience contrasts the needs from the oil and gas industries for dependable procedures to monitor mercury levels as a result of environmental concern and the requirement to avoid mercury induced corrosion or poisoning of catalysts. For example, mercury induced corrosion on a coil heat exchanger of a liquefaction train resulted in a long term shutdown of the Skikda (Algeria) natural gas processing plant.Problems involved in mercury removal from feedstocks have been discussed.2 In the following, the state of the art for species-specific mercury determinations will be reviewed along with discus- sions on the determination of total concentrations.General Comments on the Determination of Mercury HgO exerts high vapour pressures at ambient temperatures and easily amalgamates with several metals such as, gold and silver, but does not dissolve well in polar solvents such as water.These unique properties facilitate mercury determinations, even at low concentrations, in biological tissues, water and air samples, for which validated methods for total mercury are available. Solids and liquids are normally decomposed to obtain ionic mercury which is subsequently reduced to Hgo by adding, for example, tin(I1) chloride.HgO is then purged from the sample and either directly transported to an atom-specific detector or collected on a noble metal for enrichment. In the latter case, the mercury can then be quantified after heating the collector to free HgO. Air samples are passed either through an oxidizing solution or a solid sorbent to collect mercury. Concentrations are determined in an analogous fashion to solids and liquids. It should be noted that noble metals such as gold will even react with inorganic and organic mercury molecules, which are then released as HgO during heating of the collector.3 Several independent species-specific methods are available, along with a number of certified reference materials, for methyl and inorganic ionic mercury, the most common forms present in biological tissues. Speciation methods utilize chromatographic separation by GC or HPLC, normally coupled to units to produce HgO followed by element selective detectors based on atomic or mass spectrometry.It is also possible to form volatile ethyl derivatives of mercury species directly in aqueous samples. Mercury species can then be selectively volatilized and trapped on a cryogenic GC column.On increasing the column temperature, different mercury forms will elute at characteristic times. Operationally defined methods, utilizing differences in the potential of reducing agents are also frequently used to distinguish between organic and inorganic mercury forms. Determination of Total Mercury Concentrations in Natural Gases, Condensates and Oil Samples From these non-polar sample types, consisting of mainly hydrocarbons, HgO cannot be removed as easily as from aqueous solutions or air.Analytical methods relying on spectrometric detectors therefore include a combustion step followed by methods as outlined above. Treatments with oxidizing solutions or high temperature reactions with air or oxygen are common.Large volumes of reagents are always required in the former, and procedures are complicated and time consuming, increasing the risk for analytical errors and deteriorating detection limits through high and variable blanks. Recently, instrumentation for mercury determinations has been developed consisting of three independently heated furnaces. Oil or condensate samples are introduced on a suitable base material (oxidation catalyst) and combusted in an air stream in the first furnace.Reaction products are directed to the remaining two furnaces to complete combustion followed by selective trapping of mercury vapour on gold-platinum. With this system ng g-1 concentrations of mercury can be relatively rapidly determined. It is also possible to collect mercury directly from natural gases on noble metals.4.5 Quantitative sampling efficiencies could be obtained4 for HgO and dimethylmercury added to dry natural gas when a gold-platinum collector kept at 80°C was used at sampling rates < 2 1 min-1 and volumes below 10 1.For larger volumes, depending on the composition of the natural gas, there was a risk of incomplete mercury collection due to poisoning of the gold-platinum surface.The restrictions on sampled volumes limited the detection of mercury to 30 ng m-3. With an alternative method6 natural gas is passed through a sampling tube filled with I2 adsorbed on silica gel. Mercury is retained as HgI2 and is later recovered by liquid extraction followed by reduction to HgO for determination by atomic spectrometry.The required consumption of reagents for8H Analytical Cowzmunications, May 1996, Vol33 sampling and formation of HgO increases the blank values and thus deteriorates the detection limits of this procedure. measurements, gases are sometimes passed through serially connected quartz tubes, the first of which is filled with finely dispersed silver and the second with gold.The mercury trapped by silver and gold is assigned to inorganic and organic forms, respectively. Usually, mercury is found on both collectors to various proportions indicating the presence of organic and inorganic mercury. Surprisingly, the reliability of this specia- Analytical Performance of Methods Applied in Measurement of Total Mercury in Dry and Wet Natural Gas In order to compare and evaluate the performance of different sampling and analytical methodologies used for measurement of total mercury in gas field exploration and during gas production, controlled test atmospheres of mercury vapour and dimethylmercury simulating dry and wet natural gases, have been generated in a laboratory setting under the auspices of The Norwegian Oil Industry Association.1 The dry gas consisted of a mixture of methane and ethane while wet gas was simulated by saturating argon with n-hexane, n-heptane and n-octane at room temperature. Both HgO and dimethylmercury were introduced to the gas mixtures at realistic levels, ranging from 0.05 to 5 pg m--3. The four industrial, service and research laboratories participating in this collaborative campaign con- ducted measurements using collection systems for total mercury based on solid sorbents (I2),6 acidic liquid scrubbing (Cr2072-) and amalgamation (Au, Au-Pt).Mercury was detected by AAS (n = 3) or atomic fluorescence spectrometry (n = 1) either at the test site or after sampling in the participants own laboratories. Collection of both HgO and dimethylmercury by gold was demonstrated to give accurate and precise measure- ments (relative standard deviation 10%) under dry gas condi- tions.In the course of this study it was, however, clearly demonstrated that, in the presence of higher concentrations of hydrocarbons under wet gas conditions, the collection effi- ciencies of all adsorbents were affected. For some of the methods studied, blank values were larger than the actual amount of mercury introduced into the sampling unit or varied unsystematically.These methods are therefore only suitable for mercury measurements if large gas volumes can be collected (or for higher mercury concentrations). The collection of dimethylmercury from wet natural gas appears to be a greater challenge than that of elemental mercury. From the literature it is known that hydrocarbons may influence the adsorption process4 but the mechanism is not yet known.A reasonable suggestion may be that species from the matrix poison the surface of the collector. Only one laboratory had developed collection procedures where the adsorbent was heated to prevent condensation of the wet gas on the amalgamation surface. Although HgO was collected quantita- tively (100% f 5%) by the heated Au-Pt collector, only 50% of dimethylmercury was recovered from the gas matrix.In order to attempt to understand the tremendous variation in results obtained for total mercury in wet gas, new developments need to further focus on the problem with matrix poisoning of the collector surface. This collaborative campaign showed that there is currently only one collection method with adequate accuracy and precision for measurement of HgO in simulated natural gas.4 The method uses an amalgamation train collector based on two Au-Pt filled quartz tubes heated to 80°C.The relatively low recovery of dimethylmercury suggests, however, an urgent need to improve collection efficiencies and establish further data on the concentration of the different mercury species in natural gas.Mercury Speciation in Natural Gases Total mercury concentrations in natural gas show a great variability, both between different gas wells and with time,4 ranging from a few ng m--3 to mg m--7 levels. For field tion procedure has not yet been verified by independent measurements. In fact, it has been shown elsewhere3 that organic mercury forms, present in air, are collected by both gold and silver, indicating that the speciation procedure outlined above lacks selectivity.It remains to be seen if the hydrocarbon matrix induces a species-specific preference for the collector materials. From results obtained for natural gas condensates, which will be discussed below, more reliable conclusions regarding the presence of mercury species in natural gases can be made.Mercury Speciation in Gas Condensates Several procedures have been developed for characterizing the distribution of mercury species in gas condensates and their fractions.2 It should be mentioned that, since condensates might contain particulate matter-sorbed mercury, filtration is advisa- ble to obtain homogeneous samples and minimize this source of variation in the analytical results.Analytical difficulties in- crease in the presence of hydrogen sulfide due to possible formation of mercury sulfide which has a low solubility. Fig. 1 shows an example of an extraction scheme used to isolate various organic and inorganic mercury species prior to Grignard derivatization of ionic forms (to produce non-polar, butylated derivatives) and species-specific detection by GC microwave induced plasma atomic emission spectrometry.The following species could be tentatively identified in a gas condensate sample: HgO and ionic inorganic mercury, both at the sub-mg 1- level; and monomethyl-, dimethyl-, monoethyl- and diethylmercury, all at the low pg 1-1 level. For the organomercury forms, the efficiencies of the extraction proce- dures were often very poor, and so quantification was not reliable. With this instrumentation it is also possible to determine some mercury species in condensates without extraction.By this method, condensates are treated with Grignard reagent prior to separation and determination of inorganic-, monomethyl- and monoethylmercury compounds.Due to co-elution of some of the mercury forms with carbon compounds present in the condensate, hydrocarbons will cause interference effects, limiting injection volumes and hence giving relatively high detection limits between 2 and 4 pg 1-1 for butylated species. To overcome these interference effects and to improve detection limits, a device has been constructed to combust the compounds eluting from the gas chromatograph and selectively collect individual mercury species on an amalgamation trap.7 The mercury can then be thermally desorbed into the detector for interference free quantification, virtually independently of the chromatographic resolution. This approach permits the direct determination of dimethylmercury in gas condensates down to a detection limit of 0.24 pg 1-l, and of monomethyl- and inorganic mercury, following derivatization, to similar levels.With this system, quality assurance by calibration using gaseous standards directly injected onto the trap after the GC is achievable. Monomethyl-, dimethyl- and inorganic mercury have all been positively identified and quantified in gas condensate samples using the amalgamation trap.An alternative technique to reduce interferences on di- methylmercury determinations was also investigated, namely solid-phase micro-extraction.7 However, due to the lower analyte mass actually introduced into the GC system, detectionAnalytical Communications, May 1996, Vol33 9H I 1 Hg', Eg*+, R-Bg', HJlg in 2. Extract 1 i d snttiplc twice 1. Direct (total inorganic I mercury) I derivatization.I natural grs concfensrte I with I nil watcr, work up 3. Extract 5 ml sample with 1 ml conc. HCl and discard aqueous phase. Neutralize and extract with 5 ml of Lcys teine soh tion. Work up the aqueous phase. ! 4. Extract 5 ml saniplc with 5 ml of Lcysteine solution. Work up the aqueous phase. I[H$ + Hg"], R-Hg' 5. Extract 5 ml sample according to procedure 4, recover condensate and treat RS in 3.Work l i p ttic nqiieoos plimse. Fig. 1 Schematic overview of an operationally defined extraction protocol for the qualitative detection of mercury species in condensate. Mercury forms shown in square brackets are determined as one species due to the procedure employed. Work up of aqueous phases involves neutralization, complexometric extraction of the mercury species into toluene and Grignard derivatization. Determination of the derivatized mercury species is then accomplished by GC with atomic spectrometric detection.limits were two orders of magnitude poorer than obtained with the amalgamation trap. As HPLC does not require volatile, thermally stable analytes, the risk for methodological errors related to derivatization of ionic mercury species, as described above for GC, can be avoided.Recently, Schickling and Broekaert8 developed a method based on HPLC coupled to a continuous eluent oxidation mercury reduction system to introduce HgO to an atomic spectrometric detector. Stability of Hg-forms in Condensates in General Severe baseline disturbances due to the condensate matrix were observed in the region of the diphenylmercury peak, otherwise monomethyl-, monophenyl- and inorganic mercury could be satisfactorily separated.x As the detection limits were at the 10 pg 1-1 level, only inorganic mercury could be quantified in two gas condensate samples.Interestingly, diphenylmercury was found to be unstable in the presence of either monomethyl- or inorganic mercury, and the formation of monophenylmercury was observed.With this method, calibration by standard additions could lead to serious analytical errors because of such transphenylation reactions. Furthermore, the response to in- organic mercury was greater than that for the organic forms, indicating species-dependent, incomplete conversion to HgO. From these result^^.^ there are several clear advantages to using GC for the analysis of condensates at the current state of the art. Firstly, the response to all mercury species is very similar as these are directly introduced to the detector, if required \liu an amalgamation trap, in gaseous form.Secondly, the resolution achieved by GC is much greater, leading to more concentrated analyte bands reaching the detector, thus improv- ing detection limits.Thirdly, if HgO is to be generated from an HPLC eluent, the reagents required will increase blank levels. In GC with an amalgamation trap, the oxygen used to combust the eluent can be efficiently purified on-line. Concluding Remarks Due to the paucity of methods applicable to the speciation of mercury in natural gas and condensates, and the complete lack of reference materials, the determination of total mercury concentrations provides useful, complementary information.Unfortunately, the detection limits of current methods for condensates (at the pg 1-1 level) are often insufficient, bearing in mind that the concentrations of interest in natural gases are at a level of a few ng m--7. This corresponds to concentrations at the ng I-] level in condensate samples, assuming an equimolar distribution of mercury species between gaseous and liquid phases. Consequently, considerable method development for both species-specific and total concentration determinations needs to be carried out in order to fulfil industrial require- ments. References Bakke, B., M.Sc Thesis, University of Oslo and National Institute of Occupational Health, 1995. Sarrazin. P., Cameron, C. J., Barthel, Y., and Morrison, M. E., Oil Gas .I.. 1993, 86. Dumarey, R., Dams, R.. and Hoste, J., Anul. Chem., 1985, 57, 2638. Frech, W., Baxter, D. C., Dyvik, G., and Dybdahl, B., J. Anal. At. S p e c m ~ n . , 1995, 10, 769. Nuturul Gus-Determination of Mercury at High or LOW Pressure, ISOIDIS 6978, International Ot-ganizution for Standardization, British Standards Institution, London, 1988. Bestimn~un~q des Quec.k.silher~~rhaltes, Ref. Nr. E DIN 5 1865: Snell, J. P., Frech, W., and Thomassen, Y., Atzalyst, submitted. Schickling, C.. and Broekaert. J. A. C., Appl. Organornet. Chrm., 1995, 9, 29. 1995-09 Paper 6/01 7471 Received March 12, I996 Accepted Api-il 1 I , 1996

ISSN:1359-7337    

DOI:10.1039/AC996330007H

出版商:RSC    

年代:1996

数据来源: RSC

ISSN:1359-7337    

DOI:10.1039/AC99633FX020

出版商:RSC    

年代:1996

数据来源: RSC

ISSN:1359-7337    

DOI:10.1039/AC99633BX022

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, May 1996, Vol33 (159-162) 159 Analysis of Water-soluble Vitamins by High-performance Liquid Chromatography-Particle Beam-Mass Spectrometry Maria Careria, Romina Cillonia, Maria Teresa Lugarib and Paola Maninia (I Dipartimento di Chimica Generale ed Inorganica, Chimica Analiticu, Chimica Fisica, Universita di Parma, Viale delle Scienze, 43100 Parma, Italy 43100 Parma, Italy Dipartimento Farmaceutico, Universitu di Parma, Viale delle Scienze. Particle beam (PB) LC-MS was investigated for the analysis of eleven water-soluble vitamins.A reversed-phase chromatographic method making use of volatile buffers was set up for the simultaneous separation of this mixture by using narrow-bore columns. After optimization of the PB interface parameters, the effect of the MS source temperature was evaluated at 200 "C and 300 "C in electron impact (EI) and chemical ionization (positive-ion, PCI and negative-ion, NCI) modes.Electron impact mass spectra were library-searchable, but generally sensitivity was better under PCI conditions. Under these conditions and operating in selected-ion monitoring mode, linearity, detection limits and precision of the analysis was explored.Detection limits at the low-ng level were achieved for nicotinamide, nicotinic acid in PCI mode and pyridoxal under NCI conditions; repeatability of ~ 6 % (RSD) was obtained for all the compounds examined except for biotin. HPLC offers an attractive alternative to the more time- consuming chemical and microbiological methods for water- soluble vitamins because of its sensitivity, selectivity and reduced analysis time.However, most of the HPLC methods have been devised for the separation and determination of from two to four vitamins in standard solutions,1.2 simple food materiaW4 or fortified products.5.h Accurate determination of this class of substances is very important in food and pharmaceutical areas. In LC analysis of complex matrices using conventional detection systems, target compounds cannot be unequivocally determined, because of the presence of poten- tially interfering compounds; moreover, changes in retention times can occur owing to matrix effects.In such cases, LC-MS is a useful and powerful technique. Mass spectrometry can be used to obtain the unambiguous characterization and determina- tion of non-volatile and thermally labile substances both for detection in HPLC and as a direct analytical technique such as tandem MS.In the past, LC-MS using direct liquid introduc- tion,' thermospray8.9 and frit-FAB interface 10 have been proposed for the analysis of water-soluble vitamins. Recently, Yamanaka et al. 1 1 applied HPLC coupled with atmospheric pressure chemical ionization (APCI) MS to the assay of thiamine in dried yeast." The objective of the present study was to explore the applicability of particle beam (PB) HPLC-MS for the analysis of various water-soluble vitamins, whose chemical structures are reported in Fig.1; this HPLC-MS technique has previously been used by our research group for the analysis of the fat- soluble vitamins A, D, El2 and K1 13 in complex matrices.Other aims of this work were to investigate the dependence of the ion abundances in the PB mass spectra of these compounds on the source MS temperature and to verify the possibility of overcoming the problem of using water-rich mobile phases with the PB interface by using narrow-bore columns. The com- plementary information gained from electron impact (EI), chemical ionization (CI) in positive-ion (PCI) and negative-ion (NCI) modes was exploited.Experimental Chromatographic separation was carried out on a Hewlett- Packard Model 1050 solvent delivery system (Palo Alto, CA, USA) equipped with a Hewlett-Packard HP 1050 UV detector operating at 254 nm except for pantothenic and dehydroascorbic acids (A = 210 nm) and for biotin (h = 230 nm); the detector signals were monitored using the Maxima Data Acquisition software (Waters, Millipore, Milford, MA, USA).Standard H H CH,OH (CH,),COOH Aswbic acid Biotin CH3 OH o * C O R HOCH,C I I - CCONH -CH,CH,COOH I I CH, H R=OH Nicotinic acid R=NHz Nicotinamidc Pantolhenic acid CH,OH I HOCH I HOFH HOCH I 'iH2 $H,OH I H:Qo r HCI 7 H O C H ~ OH R c,- R-SHO F'yridoxal R=CH2NHz Pyridoxamine Thiamine R=CHzOH Pyridoxine Fig.1 Chemical structures of the water-soluble vitamins investigated.160 Analytical Communications, May 1996, Vol33 solutions were injected with a Rheodyne 7125 injector fitted with a 6 pl sample loop (Rheodyne, Cotati, CA, USA). Chromatographic separation was achieved on an Ultracarb ODS (20) column (250 mm X 2.0 mm, 5 pm), Phenomenex, Torrance, CA, USA.The mobile phase consisted of methanol (solvent A) and 0.02 mol dm-3 ammonium formate buffer, pH 3.75 (solvent B) delivered at 0.15 ml min-1 at room temperature. A binary gradient was formed as follows: B-A 98 + 2 from 0 to 5 min; linear gradient from 98% B at 5 min to 50% B at 15 min; 50% B and 50% A from 15 to 25 min; 50% B from 25 min to 9870 B at 35 min.HPLC-MS was carried out with a Hewlett-Packard HP 1090 chromatograph with the flow rate set at 0.15 ml min-1, equipped with an HP 1050 autosampler and coupled to an HP 59980A PB interface and an HP 5989A MS-Engine quadrupole mass spectrometer (scan range 50-300 u per 1.2 s). Chromato- graphic separations were carried out under the same conditions as for HPLC-UV detection.The volume of sample solution injected varied from 0.1 to 1.0 pl. The particle beam interface desolvation chamber temperature was 70 "C, the nebulizer helium pressure was 0.3 10 MPa, and the capillary position was extended approximately 1.5 mm from the flush position of the nebulizer. It was operated in the EI or CI mode (methane as reagent gas at source pressure for methane CI at approximately 160 Pa); positive CI (PCI) and negative CI (NCI) spectra were obtained using an electron energy of 220 eV.The source temperature was 200 "C unless stated otherwise and the quadrupole temperature was 100 "C. Time-scheduled selected ion monitoring (SIM) experiments were carried out for the analysis of selected water-soluble vitamins using the following schedule: 0-5.8 min, m/z 175; 5.8-8 min, m/z 177, m/z 126, m/z 144; 8-12 min, m/z 124; 12-1 8.5 min, m/z 123; 18.5-23 min, m/z 73, m/z 1 13, m/z 13 1.The mass spectrometer data acquisition was carried out using the HP MS 59940A ChemStation (HP-UX series). Ascorbic acid, biotin, nicotinamide, nicotinic acid, calcium pantothenate, riboflavine and thiamine hydrochloride ( > 99% purity) were purchased from Fluka (Buchs, Switzerland); pyridoxal and pyridoxine hydrochlorides, pyridoxamine dihy- drochloride and dehydroascorbic acid were obtained from Sigma (Milan, Italy).All standards were used without further purification. Stock solutions (0.2 mg ml- I ) were prepared in volumetric flasks by dissolution in and dilution to volume with deionized water, except the standard solution of ascorbic acid, which was made up in 6% m/v aqueous metaphosphoric acid solution. Dilute working standard solutions (0.02 mg ml- I ) were prepared daily.All other chemicals (formic acid, ammo- nium formate) were of analytical-reagent grade and were supplied by Carlo Erba (Milan, Italy). All the solvents were HPLC grade purchased from Lab-Scan (Dublin, Ireland).HPLC-grade water was purified in a Milli-Q system (Millipore, Bedford, MA, USA). Results and Discussion First, with the use of MS as detection system in mind, a reversed-phase chromatographic method making use of volatile buffers is set up for the simultaneous separation of this mixture by using narrow-bore columns and UV detection; generally, the use of ion-pair reagents and of non-volatile buffers is proposed for the separation of water-soluble vitamins.1~~.4.1 1 Fig. 2 shows the full-scan HPLC-PB-PCI-MS trace for a mixture of 10 vitamins obtained using a methanol-0.02 mol dm--3 ammonium formate buffer with gradient elution. The use of a narrow-bore column and thus of quite low flow rates makes the use of a water-rich eluent possible, in spite of the inherent scarce compatibility of the PB interface with aqueous mobile phases.All analytes except dehydroascorbic acid, pyridoxamine and ascorbic acid were well resolved; riboflavine was not detected, even at microgram levels, thus indicating that loss of this highly polar compound occurs during transmission of the analyte molecules through the PB interface. Electron impact, PCI and NCI Mass Spectra of Standard Water-soluble Vitamins The effect of the source temperature was evaluated at 200 and 300 "C; as expected, the abundance of [MI+, [M + HI+, [MI- or [M - HI- under El, PCI and NCI conditions, respectively, rapidly diminished in favour of the fragment ions for all compounds except for nicotinamide, nicotinic acid and thia- mine in all ionization modes and for the vitamers of vitamin B6 (pyridoxal, pyridoxamine and pyridoxine) in EI mode; hence, the source temperature does not affect fragmentation pathways of these vitamins.Electron impact, PCI and NCI techniques were investigated operating at 200 "C. Relative abundances of the ions observed in the PB mass spectra of the substances investigated in the EI and methane PCI and NCI modes are summarized in Table 1.In general, although EI spectra were library matchable, the sensitivity was worse. Using flow injection analyses of the vitamins at a concentration of 0.2 pg pl-1 (10 ~1 injected), PCI provided the highest counts, whereas for dehydroascorbic acid and pyridoxal an increase in response was obtained by performing the NCT experiments; the El and PCT analyses were essentially identical for nicotinic acid (Table 1).For the EI spectra of the vitamers of vitamin Bg, the intensity of the [MI+ ion was high in all cases, particularly for pyridoxal (m/: 167, 76%). The base peaks at m/z 149 and at m/z 151 accounted for the elimination of water and of ammonia from the molecular ions of pyridoxal and of pyridoxamine, respectively.The base peak in the spectrum of pyridoxine was recorded at m / z 94, which corresponds to the loss of a 75 u fragment ([M - CH3-CH20]+). In the NCI spectra of these substances, the deprotonated molecular ions were observed both for pyridox- amine and for pyridoxine, whereas as expected, the [MI-. ion generated by electron attachment was observed for pyridoxal. In addition, in the NCI spectrum of pyridoxamine a chloride adduct of the molecule was detected at m / z 203, the standard being a dihydrochloride.The base peak in the EI mass spectrum of ascorbic acid is the m/z 116 ion, which accounts for the loss of 60 u ([M - CH20H-CHO]+) from the molecular ion m/z 176 (Table 1); similarly, a fragment ion at m/z 114 was visible in the spectrum of dehydroascorbic acid but of lower abundance.The base peak at m/z 55 in the EI spectrum of this compound is characteristic of cyclic ketones corresponding to the resonance-stable ion [CH2=CH-C-O]+ t3 [CH2-CH=C=O]+. In the NCI mode the spectrum of ascorbic acid contained an intense signal at m/z 159 arising from the loss of the OH group; instead, the fragmenta- tion seen in the NCI spectrum of the corresponding oxygenated 5 10 15 20 25 Tim e/m i n Fig.2 Total ion current LC-PB-PCI-MS chromatogram of the water- soluble vitamin standard mixture, where: I, dehydroascorbic acid (2 pg); IT, pyridoxamine ( 5 pg): 111, ascorbic acid (2 pg); IV, thiamine (2 pg); V, nicotinic acid (1 pg); VI, pyridoxal (5 pg); VII, pyridoxine ( 5 pg); VIII, nicotindmide (1 v g ) ; IX, pantothenic acid (2 pg); and X, biotin (2 pg).Analytical Communications, May 1996, Vol33 161 molecule resulted mainly from a release of CH20 and COO forming ions at m/z 144 and m/z 130, respectively. Intense protonated and deprotonated molecular ions were present in the PCI and NCI spectra of the two compounds, respectively.As for nicotinic acid and its amide vitamin PP, the base peak in their EI spectra was the [MI+ ion at m/z 123 and 122, respectively.The elimination of the NH2 and CONH2 moieties from the molecular ion of nicotinamide generated peaks at m/z 106 and 78, followed by the opening of the pyridinic ring (m/z 51); in the same way, losses of 18 u and 45 u fragments accounted for the release of water and the COOH group from the intact molecule of nicotinic acid.For both the substances, PCI and NCI experiments produced [M + HI+ and [M - HI- ions with no other significant amount of fragmentation; this behaviour is similar to that observed in the thermospray spectra of the two compounds.9 In the EI spectrum of biotin, few (but informative) fragment peaks and a weak molecular ion (m/z 244) are present; the base peak at m/z 97 is presumably formed through decyclization of the thiophenic ring and the fragment ion at m/z 85 is attributable to the imidazolic ring.In contrast, PCI and NCI analyses produced the [M + HI+ (m/z 245) and the [M - HI- (m/z 243) ions as the most intense signals in the spectra of this vitamin. A poor EI spectrum was obtained for pantothenic acid; prominent peaks appear at m/z 57 ([CO-NH-CH,]+) and m/z 7 1 ([CO-NH-(CH2)2]+), whereas the molecular ion at m/z 219 was hardly discernible (Table 1).The base peak at m/z 131 in the corresponding PCI spectrum is due to the cleavage of the peptide bond, and the subsequent loss of 18 u from this fragment undergoes the formation of the m/z 113 ion; in addition, an intense [MI+ ion at m/z 219 (20%) is present. A different fragmentation pattern was observed under NCI conditions for this compound; the [M - HI- ion was detected with relative high intensity (59%), the ion [M - H-CH20H]- at m/z 187 being the main fragment; the release of the CH2CH2COOH moiety from the deprotonated molecular ion produced the abundant peak at m/z 145.In contrast to the findings of other authors,lJ good quality mass spectra using both EI and CI ionization modes were obtained for thiamine. The EI spectrum contained mainly.fragments of the thiazole ring at m/z 112 and m/z 143, the molecular ion being scarcely visible. The presence of the most abundant signal at m/z 144 suggests that a decyclization of the thiazole ring occurs under PCI conditions; this spectrum has a close resemblance to that reported using the thermospray HPLC-MS technique.9 In the NCI spectrum an intense [M - 2H]- peak was detected at m/z 263, whereas for the formation of the base peak at m/z 242 a complex mechanism has to be hypothesized, i.e., the elimination of the OH and CH2CH2CH2 moieties followed by the addition of HC1, the standard being a hydrochloride chloride.100 80 0 60 U 2 40 Q, a 20 5 10 15 20 Time/min Fig.3 LC-PB-PCI-MS chromatogram of a water-soluble vitamin stan- dard mixture recorded under time-scheduled SIM conditions: I, dehy- droascorbic acid (100 ng): 11, ascorbic acid (30 ng); 111, thiamine (100 ng); IV, nicotinic acid (10 ng); V, nicotinamide (10 ng); and VI, pantothenic acid (150 ng). Table 1 Direct flow-injection PB EI, PCI and NCI mass spectra of the investigated vitamins (source temperature, 200 "C; scan range, 50-300 u.) Compound Ascorbic acid Biotin Dehydroascorbic acid Nicotinamide Nicotinic acid Pantothenic acid Pyridoxal Pyridoxamine Pyridoxine Thiamine Mr 176 244 174 122 123 219 167 168 169 265 Detection mode EI PCI NCI El PCI NCI EI PCI NCI EI PCI NCI El PCI NCI El PCI NCI EI PCI NCI El PCI NCI El PCI NCI EI PCI NCI Major ions, m/z (relative abundance, %) 116(100), 85(39), 61(29), 176(5) 177(100), 115(26), 159(11) 175(100), 159(90) 97( loo), 85(58), 143(4), 244(3) 245( loo), 201(46), 169(30), 99(20) 243(100), 226( 15), 210(85) 55( loo), 114(55), 84(35) 113( loo), 175(76), 157(37) 174(100), 144(34), 130(30), 114(19), 84(3) 122(100), 78(92), 106(69), 51(46) 123( 100) 121(100), 120(40) 123(100), 105(67), 78(55), 51(42) 124(100), 106(5) 122( 1 OO), 120(20) 71(100), 57(28), 131(2), 219(2) 131(100), 57(29), 219(20), 113(14) 187(100), 145(73), 218(59) 149(100), 167(76), 120(59) 150( loo), 168(54) 167( 100) 151(100), 94(75), 106(33), 123(31), 168(30) 169( loo), 151(67), 136(57), 152(53) 203(100), 167(87), 149(23) 94(100), 106(77), 151(74), 169(50) 152( loo), 170(76), 136( 19) 167(100), 149(74), 168(55), 151(35) 112(100), 85(39), 143(37), 264(1) 144(100), 126(59), 265(4) 242(100), 263(56), 146(48) Base peak abundance 140 000 400 000 25 000 5 000 25 000 30 000 7 000 60 000 115000 38 000 180 000 16 000 150 000 140 000 60 000 8 000 80 000 25 000 70 000 260 000 400 000 6 000 65 000 55 000 70 000 220 000 120 000 120 000 190 000 30 000162 Analytical Communications, May 1996, Vol33 Table 2 Calibration graph results, * linear range, detection limits and precision data for the investigated vitamins using FI-PB-MS Compound Ascorbic acid Biotin Dehydroascorbic acid Nicotinamide Nicotinic acid Pantothenic acid Pyridoxal Pyridoxamine Pyridoxine Thiamine Detection mode, SIMT PCI, mlz 177( 1000) PCI, rnlz 99(200) 169( 150) 201(250) 245(400) PCI, mlz I13(500) 1 75 (500) PCI, mlz 123(1000) PCI, mlz 124( 1000) PCI, SIM, rnlz 73(400) 113(400) 131(200) NCI, SIM, mlz 167(1000) PCI, rnlz 135(200) 151(300) 169(500) PCI, mlz 136(300) 152(500) 170(200) PCI, rnlz 126(400) 144(600) Rangelng 15-150 250-850 150-1 100 20-100 8G270 150-1300 60-250 850-3 100 5 5 0-2000 175-900 a 10-6 3.30 (k0.45) -3.74 (f0.35) - 19.29 (k0.92) -2.65 (f0.12) -3.98 (f0.40) -38.5 1 (k1.52) -11.64 (f1.30) -23.85 (k1.29) -76.27 (f5.62) -38.21 (k1.43) b 10-4 8.79 (k0.43) 1.50 (f0.06) 12.07 (k0.14) 12.53 (k0.21) 6.83 (k0.21) 22.30 (f0.21) 21.51 (f0.84) 2.55 (f0.06) 12.98 (k0.39) 23.20 (f0.24) n 21 20 40 40 18 50 15 30 28 40 * Calibration fitting: y = a + bx.t SIM, ions monitored for each vitamin, dwell time (ms) in parenthesis .. r 0.978 0.984 0.997 0.995 0.992 0.998 0.992 0.992 0.988 0.998 DL+ 15 250 85 5 4 100 6 400 225 90 Amount1 ng 40 100 150 350 500 850 250 500 900 30 50 80 100 165 230 200 500 1000 100 140 200 I400 2200 3 100 680 1350 2000 175 510 820 $ Detection limit, in ng. RSD (%) (n = 5 ) 2.88 1.18 2.68 6.5 1 5.26 3.10 3.48 3.05 0.98 5.98 2.84 1.73 2.14 1.75 2.60 4.58 0.92 2.82 3.66 4.89 1.44 5.69 3.79 3.01 4.58 5.60 4.06 2.03 0.54 1.08 Analytical Characteristics of the PB-MS Determinations of Water-soluble Vitamins The equations relating peak areas obtained in flow injection (FI) mode to amount for each of the vitamins studied are quoted in Table 2.The calculated detection limits (DLs) for a S/N ratio of 3 and precision data at three concentration levels are also given. Linear regression analysis from calibration graphs showed that the correlation coefficients were between 0.992 and 0.998, except for ascorbic acid (0.978) and biotin (0.984).The precision of the analysis gave an RSD of < 6% when low- and high-ng levels were injected (n = 5). For biotin an RSD of > 6% (n = 5,350 ng level) was found; precision of the assay for this compound was hampered by a tailed-shape peak, which was observed also in FI mode.DLs at the low-ng level were achieved for nicotinamide, nicotinic acid and pyridoxal. The minimum detectable quantity for thiamine was 90 ng in the PCI- SIM mode, whereas a detection limit of 2 ng at S/N of 3 using LC-APCI-MS has been claimed. 1 In order to verify the analytical capabilities of the use of the narrow-bore columns with PB-MS detection, time-scheduled SIM (PCI mode) was used; taking into account the detection limits obtained under PCI-SIM conditions, a synthetic mixture of six out of ten vitamins was analysed.Fig. 3 shows the results of a narrow-bore LC-PB-MS analysis of the mixture at 1-15 ng 1.11-1 (10 pl injection volume) of analytes recorded using the peaks listed in Experimental. Promising results from the applicative point of view were obtained for nicotinic acid, nicotinamide and ascorbic acid, which were detectable at low- ng levels. The authors acknowledge financial support from the Minister0 dell’universiti e della Ricerca Scientifica e Tecnologica and from the Consiglio Nazionale delle Ricerche (Italy).References 1 2 3 4 5 6 7 8 9 10 11 12 13 Zapata, S., and Dufour, J. P., J. Food Sci., 1992, 57, 506. Blanco, D., Sanchez, L. A,, and Gutierrez, M. D., J . Liq. Chroma- togr., 1994, 17, 1525. Maeda, Y., Yamamoto, M., Owada, K., Sato, S., Masui, T., and Nakazawa, H., J. AOAC lnt., 1989, 72, 244. Akiyama, S., Nakashima, K., Shirakawa, N., and Yamada, K., Bull. Chem. SOC. Jpn., 1990,63, 280. Chase, G. W., Landen, W. O., Jr., Soliman, A.-G. M., and Eitenmiller, R. R., J. AOACInt., 1993, 76, 390. Kamata, K., Hagiwara, T., Takahashi, M., Uehara, S., Nakayama, K., and Akiyama, K., J. Chromatogr., 1986, 356, 326. Azoulay, M., Desbene, P.-L., Frappier, F., and Georges, Y., J. Chromatogr., 1984, 303, 272. Iida, J., and Murata, T., Anal. Sci., 1990, 6, 269. Iida, J., and Murata, T., Anal. Sci., 1990, 6, 273. Asakawa, N., Ohe, H., Tsuno, M., Nezu, Y., Yoshida, Y., and Sato, T., J. Chromatogr., 1991, 541, 231. Yamanaka, K., Horimoto, S., Matsuoka, M., and Banno, K., Chromatographia, 1994, 39, 91. Careri, M., Lugari, M. T., Mangia, A., Manini, P., and Spagnoli, S., Fresenius’ J. Anal. Chem., 1995, 351, 768. Careri, M., Mangia, A., Manini, P., and Taboni, N., Fresenius’ J. Anal. Chem., 1996, 355,4848. Paper 6100700G Received January 30, I996 Accepted March 20,1996

ISSN:1359-7337    

DOI:10.1039/AC9963300159

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, May 1996, Vo133 ( I 63-165) 163 Enzymic Analysis of L-Lactate Dehydrogenase (LDH) with a Surface Acoustic Wave Impedance Sensor and its Application in the Determination of LDH Activity in Human Serum Ronghui Wang, Qingyun Cai, Wei Wei, Lihua Nie and Shouzhuo Yao* Department of Chemistry and Chemical Engineering, Hunan University, Changsha 41 0082, China A new method is described for the determination of L-lactate dehydrogenase (LDH) using a surface acoustic wave impedance sensor.The assay of this enzyme is based on the change in conductance of the solution caused by enzymic reaction between lactate, NAD+ and LDH. A linear relationship between frequency response and enzyme concentration was obtained and the detection limit was 4.8 U ml-I.The Michaelis constant for lactate and the corresponding maximum initial rate were 0.03 mol 1-1 and 13 kHz min-I, respectively. Influences such as temperature, pH value and interferents were also investigated. The method was successfully applied in the clinical assay of LDH activity in human serum. The results compared well with the reported values and support the clinical diagnosis.L-Lactate dehydrogenase (LDH, EC I . 1.1.27) is an enzyme present in all cells of the body. Determination of serum LDH is useful in the diagnosis of many diseases including hepatitis, acute myocardial infarction and many kinds of carcinoma.' As LDH can catalyse oxidation of lactate to pyruvate in the presence of NAD+: LDH Lactate + NAD+ +--+ Pyruvate + NADH + H+ (1) This reaction is often used to assay the activity of LDH.Several methods such as colorimetry,2 fluorimetry,3 electr~analysis~ and chemiluminescences have been used for the measurement of LDH, most of them using NADH as a coenzyme. However, some of these methods are either time-consuming, require skilful labour or show poor precision and low sensitivity. In this paper, a new sensor system using a surface acoustic wave (SAW) device is described for the determination of LDH activity, using NAD+ as a coenzyme because of its stability and cheapness.SAW devices are small, reliable and very sensitive chemical sensors. Although the development of SAW sensors in liquids is very important for many disciplines, there are still some problems to be solved6 and there are few publications on the use of SAW sensors in the liquid phase.7.8 Recently, a new type of SAW impedance sensor system combining a 61 MHz SAW resonator with a pair of parallel conductive electrodes for liquid investigations has been proposed by our laboratory.9 The frequency of the SAW impedance sensor can be measured within +I Hz.For such a SAW impedance sensor system, there is a linear correlation between the frequency shift (AF) of the sensor and the change of the electrolyte conductance (AK) of * To whom correspondence should be addressed.the liquid, i.e., A F = aAK + b. Here, a and b are constants depending on the SAW device, amplifier circuit and experi- mental conditions. So, the frequency of the SAW impedance sensor shifts according to the conductance change of the substrate solution caused by enzymic reaction between lactate, NAD+ and LDH, and it is possible to correlate the enzymic reaction rate with the frequency change.The proposed SAW impedance sensor is advantageous in its rapid determination, high sensitivity, simplicity of construction and low noise levels. It has been successfully applied in the enzymic analysis of LDH and has also been applied in the determination of LDH activity in human serum.Experimental Materials and Reagents L-Lactate dehydrogenase [LDH, E.C. I . 1.1.27 from rabbit muscle, 230 U mg-* (1 U = 16.67 nkat)] and nicotinamide adenine dinucleotide (NAD) were obtained from Shanghai Biochemical Products Institute (Shanghai, China). L-Lactate was purchased from Hunan Chemical Products Institute (Hunan, China).All other chemical reagents were of analytical- reagent grade. Distilled, de-ionized water was used for the preparation of all solutions. Human serum samples (both healthy persons and patients) were collected freshly from the clinical hospital and were immediately analysed without any treatment or delay. Apparatus The experimental set-up was as described in the previous paper.9 The temperature was thermostated at 26 k 0.2 "C by using air-bath thermostating equipment which was monitored and controlled by a temperature controller.A 5 ml detection cell with a magnetic stirrer was used. Procedures The detection cell was filled with 5 ml of the lactate solution, NAD+ (in 10 mmol 1-1 TRIS-HC1 buffer, pH 9.0) under magnetic stirring and the stable frequency of the SAW impedance sensor was measured (Fo). Then, a small volume of the enzyme solution or the serum sample was transferred into the cell and the enzyme catalysed reaction was started.As a function of time, the resulting frequency shift was recorded as AF = F - Fo, where F was the frequency after adding the enzyme solution or the serum sample. A frequency shift versus time graph was obtained and the initial response rate was calculated from its initial linear section.Then, an initial response rate versus LDH concentration calibration curve was164 Analytical Communications, May 1996, Vol33 constructed and a linear relationship between initial response rate and lactate concentration (-=xK,) was also obtained. Conductimetric measurements were performed for com- parison using the same procedure as above.Here, the detection cell was connected to a HP4129A impedance analyser (Hewlett- Packard, Avondale, PA, USA) by using conductive electrodes with the same cell constant (about 1 cm) and the conductance change versus time graph was constructed. Results and Discussion Determination of Kinetic Enzyme Parameters At a constant enzyme concentration, the initial rate is directly proportional to the substrate concentration when the concen- tration of substrate <<K,.lO Under the applied experimental conditions (50 mmol 1-1 TRIS-HCl buffer, 0.52 mmol 1-1 NAD+, pH 9.0, 26 "C), the concentration of LDH was kept constant at 0.145 mg ml-1 and the lactate concentration could be varied between 0.0909 mmol 1- and 1.45 mmol 1- 1 to obtain a typical Michaelian curve.A linear relationship between the lactate concentration and the initial rate was obtained and the correlation equation of the initial rate versus lactate concentra- tion was described as: VO = 377.5 [lactate] + 24.2 ( r = 0.993, n = 6), where the lactate concentration was expressed in mmoll-l and VO in Hz min-l.V, was determined by extrapolating the quadratic fitting curve of the response curve. K, and V,,, were determined from Lineweaver-Burk graphs. The K , value was estimated to be 0.03 mol 1-1 for lactate and this value is in accordance with the literature. The correspond- ing maximum initial rate (Vmax) was 13 kHz min-1. Using the same procedure as above, a linear relationship between the rate of frequency shift and the NAD+ concentration was also obtained under the present experimental conditions (5 mmol 1-l TRIS-HC1 buffer, 0.01 mol 1-1 lactate, 0.145 mg ml-l LDH, pH 9.0, 26 "C) while the NAD+ concentration ranged from 7.1 to 56.8 pmol l-I.A correlation equation of the initial rate versus NAD+ concentration was obtained as: Vo = 59.81 "AD+] - 7.79 ( r = 0.991, n = 6).This yields a K , value of 0.0089 mol 1-1 for NAD+ and a corresponding maximum initial rate V,,, of 50.3 kHz min-1. Relationship Between Frequency Response and Enzyme Concentration A calibration curve of initial response rate versus enzyme concentration was constructed. In this experiment, the substrate lactate concentration was kept constant at 0.01 mol 1-*, NAD+ at 0.52 mmoll- and the enzyme concentration was in the range up to 42.5 pg ml-I of LDH, at 26 "C and pH 9.0.A correlation of Vo = 18.6 [El + 15.99 ( r = 0.993, n = 7) was obtained, where Vo refers to the initial rate (in Hz min-l) and [El to the concentration of the LDH solution (in pg ml-l). Under these experimental conditions, the detection limit of LDH activity is evaluated to be 4.8 U ml-1 based on three times the S/N.Average recoveries of 95.4 to 104.8% of LDH were obtained, each obtained value being the mean of three measurements; seven separate experiments were carried out (see Table 1). Influences of pH Value, Temperature and Interferents Investigations were carried out to examine the optimum pH and temperature for LDH determination: 0.01 mol 1-1 lactate, 0.52 mmoll- NAD+ and 0.145 mg ml- LDH were selected for this measurement.The LDH has an activity maximum near pH 9.0 and 33 "C. However, these experiments were carried out at 26 "C and pH 9.0 as a compromise between achieving maximum enzyme activity and enzyme stability. Some substances that might possibly affect the LDH activity were studied. Under the same experimental conditions, no obvious change in the sensor response was observed when glycine (1 mol 1-l) and K+ (1 mol 1-l) were present in the reaction system.Activators such as Mg2+ (3 rnol 1-l) can activate LDH by increasing the initial rate by 6%. However, Cu2+ (1 rnol 1-I), Cd2+ (1 mol 1-l) or EDTA (1 mol 1-1) can seriously inhibit the LDH activity, resulting in a decrease in initial enzymic activity of 39.2, 53.9 and 52.4%, respectively.When both EDTA and Ca2+ were present together the enzyme was almost completely inactivated (94.2%). Comparison Study A comparative study was carried out of the SAW impedance sensor and the conductimetric method. The results show a linear relationship between LDH activities measured by the SAW impedance sensor system and the conductimetric method.The regression equation is Y(SAW) = l.O8X(Cond.) - 10.01 with a correlation coefficient of 0.982 (n = 6). Although they both correlate well, the frequency of the SAW impedance sensor system can be easily measured within k 1 Hz or even better and the sensing method avoids the drawbacks of conductimetry which is restricted to low ionic strength media.So, the proposed SAW impedance sensor is much more attractive than the conductimetric one. In Table 2, a comparison of the SAW-sensing method with some other methods previously reported for LDH assay is presented. The immobilized enzyme has a relatively short life and requires skillful labour. The second and the third methods in Table 2 appear to be complicated and require special reagents.The proposed SAW-sensing method is simple and has a lower detection limit than conductimetry and the electrochemical detections 12 and luminescence-based fibre-optic sensor13 methods. Table 1 Recovery determined with the SAW impedance sensor system Enzyme Enzyme added/ VOl found/ Recovery RSD 1.53 44.02 1.51 98.5 1.6 7.00 146.96 7.02 100.3 1.4 14.43 297.16 15.12 104.8 2.5 21.63 412.58 21.32 98.6 1.7 28.83 575.43 30.08 104.3 2.4 36.00 655.05 34.36 95.4 2.8 42.50 804.3 1 42.38 99.7 1 .5 yg ml Hz min-1 yg min-1 (%) (%I Table 2 Comparison of methods for analysing LDH Method Detection Substrate Coenzyme limit/U 1-1 Reference Immobilized enzyme and electrochemical detection Pyruvate NADH 25 12 Flo w-injection analysis combined with luminescence- based fibre-optic sensor Pyruvate NADH 5 13 Bioluminescent flow sensor Pyruvate NADH 0.5 14 Conductimetry Lactate NAD 24 This paper SAW method Lactate NAD 4.8 This paperAnal-ytical Communications, May 1996, Vol33 165 The LDH activity of eight samples of serum was measured by the present method and compared with the results measured by spectrophotometry and conductimetry.The obtained values of activity were the mean of three measurements.The data obtained by the present method agree with the data obtained by the spectrophotometric method, which is being routinely used for clinical analysis. Although the spectrophotometric method is more stable, the SAW method appears to be more sensitive. The results show a linear relationship between the LDH activity measured by the SAW impedance sensor and the spectrophoto- metric method.The regression equation is Y(SAW) = 0.94X(Spect) + 1.53 with a correlation coefficient of 0.998 ( n = 8). Activities of LDH as much as 3.61-6.14 times the usual values for healthy humans have been observed, thereby assisting in the diagnosis of different carcinoma. The SAW- sensing method has an RSD of about 0.8% and the conducti- metric an RSD of about 4.8%.In short, the proposed new method for the determination of LDH has been shown to be an efficient technique for the enzyme assay and is hopeful as a clinical indicator for the detection of carcinoma. The procedure is fast, simple, sensitive and accurate. We are grateful for the financial support of the National Natural Science Foundation and the State Education Commission Foundation of China. References 1 2 3 4 5 6 7 8 9 10 1 1 12 13 14 Scheller, F., Schubert, F., Olson, B., Gorton, L., and Johansson, G..Anal. Lett., 1986, 19, I69 I. Babson, A., and Babson, S., Clin. Ckenr. ( Winston-Salem. N.C.), 1973, 19, 766. Guilbault, G. G., and Zimmerman, R.. And. Chim. Acta.. 1972, 58, 75. Kinoshita, H., Suda, Y., Kawakubo, T., Takayama, K., and Ikeda, T., Mic.rwkeni. J.. 1994, 49, 226. Williams, D. C., and Seitz, W. R., Anal. Ckeni., 1976, 48, 1478. Calabrese, G. S., Wohltjen, H.. and Roy, M. K., Anal. Chern.. 1987, 59, 833. Roederer, J. E., and Bastiaans, G. J., Anal. Chenr.. 1983, 55, 2333. Zhang, D., Crean, G. M., Flaherty, T., and Shallow, A.,Anulysr, 1993, 118. 429. Yao, S., Chen, K., and Nie, L., Anal. Clrini. A ( m . , 1994, 289, 47. Palmer, T., Understancling Enzymes, Ellis Horwood, Chichester, 1981, ch. 9. Barthova, J.. and Janousek, S., Collect. Cxclr. Ckem. Conmiun.. 1978,43,2960. Fonong, T., and Barber, T., Analyst. 1988, 113, 1807. Gautier, S . M., Blum, L. J., and Coulet, P. R., Anal. Ckinr. A m . . 1992,226, 33 I. Girotti, S., Bassoil, C., Loredana Cascione, M., Ghini, S., Carrea, G., Bovara, R., Roda, A., Motta, R., and Petilino, R. J.. Biolumin. Chemilumin., 1989, 3, 41. Paper 6/01 798C Accepted A4ur.c.h 14, 1996

ISSN:1359-7337    

DOI:10.1039/AC9963300163

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, May 1996, Vol33 (167-169) 167 Determination of Some Cephalosporins in Pharmaceutical Formulations by a Fluorescence Quenching Method Jinghe Yang, Guangjun Zhou, Guiling Zhang, Zhikun Si and Jingtian Hu Department of Chemistry, Shandong University, Jinan, 2501 00, China A fluorescence quenching method for the determination of some cephalosporins is described. The method is based on the hydrolysis of the cephalosporins in sodium hydroxide to produce the sulfuret, which can quench the fluorescence intensity of fluorescein mercury.Under optimum conditions a linear relationship was obtained between the concentration of cephalosporins and the fluorescence quenching (AF) of fluorescein mercury. We have studied four cephalosporins: cefadroxil, cephradine, cefotaximum sodium and cefazolinum.Their detection limits are 5.0 x 10-6, 1.2 x 10-5 mg ml-1, respectively. The method has been applied to the determination of some cephalosporins in pharmaceutical formulations. The results are satisfactory. 1.2 x 10-5 and 6.0 X Cephalosporins are a group of (3-lactam antibiotics. In compar- ison with the penicillins, cephalosporins have the characteristic of high antibiosis, low toxicity and allergenicity, which means that they are extensively employed as bacteriostatic drugs.Many methods have been reported for their determination, including spectrophotometry ,1-3 fluorimetry,4 electrochem- istry,' HPLC6.7 and other methods8.9 based on alkaline hydrolysis. However, most of these methods are complicated and lacking in sensitivity. The present method is based on the alkaline hydrolysis of cephalosporins to produce sulfuret, which can quench the fluorescence of fluorescein mercury. On the basis of the above fact, an analytical method for the determination of four cephalosporins such as cefadroxil, cephradine, cefotaximum sodium and cefazolinum was proposed.Tests showed that this method was simple, accurate, sensitive and readily applied to the determination of cephalosporins in injections. Experimental Apparatus All fluorescence measurements were made on an RF-540 spectrofluorimeter (Shimadzu, Japan) with a 150 W Xe lamp as the excitation source. Reagents Fluorescein mercury solution, 1.000 X rnol I-'.Dissolve 0.2239 g of fluorescein mercury [synthesised by fluorescein sodium and mercury(r1) a ~ e t a t e ] ' ~ in 0.1 rnol 1-I NaOH solution and dilute to 250 ml with 0.1 mol 1- NaOH.The stock solution could be kept in a brown vessel for 2 months and the working solution was obtained by diluting the stock solution with 0.1 mol 1-1 NaOH. Cephalospoi-in solutions, 1 .OOO mg ml- 1. Dissolve 0.1 g (accurately weighed) of the cephalosporin (The Chinese Biomedical Institute) in water and dilute to 100 ml with distilled water.The stock solutions could be kept in a refrigerator at 4 "C for 1 week. Working solutions were prepared by diluting with distilled water. All of the other reagents used were of analytical-reagent grade and distilled, deionized water was used. 480 500 550 580 450 500 550 ?Jnm Fig. 1 Fluorescence spectra: ( C I ) excitation (Aem = 520 nm); (h) emission (Aex = 499 nm).1, Fluorescein mercury; 2, fluorescein mercury-alkaline hydrolysate. Conditions: fluorescein mercury, 2.0 X lo-' moll-I; alkaline hydrolysate, 2.0 x 10-4 mg ml-I. 10 8 k 6 4 4 2 0 0.05 0.1 0 15 0.2 0.25 0 3 Concentration/moi I-' Fig. 2 Effect of NaOH concentration: I , cefadroxil; 2, cefotaximum sodium; 3, cephradine; 4, cefazolinum.Conditions: 1, 2, 3, cephalosporins concentration 2.0 X 10-4 mg ml-1; fluorescein mercury concentration, 2.0 x 10-6 rnol I-]; 4, cefazolinum concentration, 2.0 x 10-4 mg ml-1; fluorescein mercury concentration, 1 .0 x 10-6 mol 1-1.168 Analytical Communications, May 1996, Vol33 Procedure Procedure for cefudroxil and cephradine To a 25 ml test-tube, 1 ml of fluorescein mercury (2 X or 2 x 10-6 mol 1-I), 1 ml (for cefadroxil) or 1.5 ml (for cephradine) of 1 moll-' NaOH and an aliquot of cephalospor- ins solution (10-3 or 10-4 mg ml-I) were added in that order.The mixture was diluted to 10 ml with distilled water, shaken and allowed to stand for 15 min. The fluorescence quenching (AF) was measured in a 1-cm quartz cell with excitation and emission wavelengths of 499 and 520 nm, respectively, against a similarly treated reagent blank solution.10 8 z 6 4 2 6 t \ A 0 10 20 30 40 50 60 Time/min Fig. 3 Effect of heating time: A, cefotaximum sodium; B, cefazolinum. Conditions: A, B, cephalosporins, 2.0 X mg ml-I; A, fluorescein mercury, 2.0 x 10-6 mol 1-1; B, fluorescein mercury, 1.0 X 10-6 mol 1-1. Procedure for cefotuximum sodium and cefazolinum To a 25 ml test-tube, aliquots of cephalosporin solutions ( mg ml-1 or 10-4 mg ml-1) and 1 ml of 1 mol 1-1 NaOH solution were added and the mixture was diluted to 10 ml with distilled water. It was then heated in a boiling water-bath for 30 min.After cooling to room temperature, the mixture was diluted to 10 ml with distilled water and mixed well. An appropriate volume of fluorescein mercury solution and 1 ml of the above produced solution were added to a 25 ml test- tube, the mixture was diluted to 10 ml with 0.1 mol 1-1 NaOH solution, mixed well and left to stand for 15 min. The fluorescence quenching (AF) was measured in a 1 cm quartz cell with the excitation and emission wavelengths of 499 and 520 nm against a similarly treated reagent blank solution.Determination of injections The solutions of cephalosporins in injections were made by dissolving accurately weighed samples, equivalent to 0.1 g of Table 1 Calibration graphs for cephalosporins in standard solution Linear range of cephalosporins/ Compound mg ml- Cefadroxil (2-5) x 10-5 Cephradine ( 1 4 ) x 10-5 Cefotaximum (1-10) x 10-9 Cefazolinum (1-7) X 10-4 (1-7) x 10-4 (1-6) x 10-4 sodium (2-9) x 10-4 Concentration of fluorescein mercury/mol 1- 1 2 x 10-6 2 x 10-6 2 x 10-6 1 x 10-6 2 x 10-7 2 x 10-7 2 x 10-7 Regression coefficient 0.9999 0.9988 0.9993 0.9998 0.9997 0.9989 0.9994 Table 2 Structure of cephalosporins and molar yield of sulfuret COOH Compound Cefadroxil Cephradine Cefotaximum R' R2 -H --ti Molar yield of sulfuret (9%) 66.7 74.6 -C-CH, 44.8 II 0Analytical Communications, May 1996, Vol33 169 the cephalosporins, in water and diluting to 100 ml.The determination of injections is identical with the procedure described above. As for cephalosporins in tablets or capsules, since there is interference from the carrier, the tablets and the capsules could be treated according to the method described in references 1 and 8.The present authors did not determine cephalosporins in tablets and capsules. Results and Discussion Fluorescence Spectra The fluorescence spectra of fluorescein mercury ( I ) and fluorescein mercury-alkaline hydrolysate (2) systems are shown in Fig. 1. From this figure, it can be seen that the fluorescence spectra of the two systems are similar, the wavelengths of the excitation and emission being 499 and 520 nm, respectively.However, the fluorescence intensity of system (2) is weaker than that of system (I), owing to the fluorescence quenching caused by the sulfuret in the alkaline hydrolysate. Factors affecting the degradation In order to make cephalosporins produce a maximum yield of the sulfuret, it is important to choose optimum hydrolytic conditions. Experiments showed that the concentration of NaOH and the heating time had a great effect on the fluorescence quenching ( AF).EfSect of NaOH concentration. The effect of NaOH concen- tration on the degradation of cephalosporins was shown in Fig. 2. It can be seen that the fluorescence quenching of the fluorescein mercury caused by the sulfuret in the cephalosporins alkaline hydrolysate was at a maximum when the concentration of NaOH in the solution was 0.10-0.15 mol 1-1.In our experiment, the concentrations of NaOH are 0.1 mol 1-1 for cefadroxil, cefotaximum sodium and cefazolinum and 0.15 mol 1-1 for cephradine. EfSect qf heating time. The experiments showed that the rate of alkaline hydrolysis of cefadroxil and cephradine was very fast.Their fluorescence quenching (AF) reached a maximum when the solutions were stood for 5 min at room temperature. However, the degradation rate of cefotaximum sodium and cefazolinum was very slow at room temperature. In order to speed up the degradation process, it was necessary to heat their sodium hydroxide solutions in a boiling water-bath. The effect of heating time on degradation is shown in Fig.3. It can be seen that the fluorescence quenching (AF) of fluorescein mercury was at a maximum when the solutions were heated for 30 min. Optimum conditions of determination EfSect of fluorescein mercury concentration. The concentra- tion of fluorescein mercury used was determined by the yield of sulfuret in the alkaline hydrolysate of cephalosporins.If the Table 3 Determination of cephalosporins in injections Standard Drug content" Mean It: Cefadroxil 67.2 67.5, 70.0, 6.5.0, 64.0 66.6 It: 2.7 Cephradine 9 1.5 95.0, 90.0, 92.0, 91.0 92.0 f 2.2 Cefazolinum 9 1.9 92.5, 90.0, 92.5, 87.5 90.8 k 2.4 Cefotaximum (injection) (%I Found (%) s (%) sodium - 98.0, 96.0, 97.0, 98.0 97.3 k 0.96 * The standard content of cephalosporins in injections was determined using spectrophotometry by Qilu Pharmaceutical Factory.concentration of fluorescein mercury was too low, the fluores- * cence quenching caused by the sulfuret was found to be so large that the range of the calibration curve became small. If the concentration was too high, the linear relationship would be destroyed, owing to the self-absorption of the fluorescein mercury solution.The appropriate concentrations of fluorescein mercury for different determination ranges of cephalosporins are shown in Table 1. Effect of NaOH concentration. The pH of the solution has a great effect on the fluorescence intensity of fluorescein mercury. Experiments showed that the fluorescence intensity of the system was the strongest and remained constant in the range 0.05-0.8 mol 1-1 of NaOH.Stability of the system Experiments showed that the fluorescence intensity of fluores- cein mercury decreased slowly with time, while for the fluorescein mercury-alkaline hydrolysate system, the fluores- cence intensity reached a maximum after 15 min and remained stable for 1 h, after which it decreased slowly with time.Discussion on degree of the alkaline hydrolysis The structure of the four cephalosporins and their degree of alkaline hydrolysis are shown in Table 2. It can be seen that the structure of the cephalosporin was the primary factor affecting the degree of alkaline hydrolysis, although the concentration of NaOH and the temperature had an effect on it. The more complicated the structure of the cephalosporins, the lower the degree of the alkaline hydrolysis.Calibration curve and detection limit Under the optimum conditions, the calibration values of four cephalosporins were obtained and are shown in Table 1. The detection limits of cefadroxil, cefotaximum sodium, cephradine and cefazolinum are 5.0 X 10-6 1.2 X 10-5, 1.2 X 10-6, and 6.0 X 10-5 mg ml-*, respectively (signal to noise ratio is 2).In comparision with other methods1-9 reported previously, the present method was sensitive for the determina- tion of cephalosporins. Determination of samples The proposed method was applied to the determination of cephalosporins in injections. From Table 3, it can be seen that the results obtained were satisfactory. Therefore, the present method was a simple and highly sensitive technique for the determination of cephalosporins. References 1 2 3 4 5 6 7 8 9 10 Morelli, B., and Peluso, P., Anal. Lett., 198.5, 1S(B9), 1 113. Issopoulos, P. B., Analyst, 1989, 114, 237. Inci Seng, F., and Inci, F., Talanta, 1986, 33, 366. Yu, A. B. C., Nightingale, C. H., and Flanagan, D. R., J . Phurm. Sci., 1977.49, 1002. Mufioz, E., Camacho, L., Avica, J. L., and Garcia-Blanco, F., Analyst, 1989, 114, 1611. Korach, P. M., Lantz R. J., and Brier, G., J. Chromatogr. B , Biomed. Appl., 1991, 105, 129. Zhao, Y. Z., Qian, X. P., and Li, Z. H., Sepu, 1992, 10, 183. Abdalla, M. A., Anal. Lett., 1991, 24, 55. Abdalla, M. A., Fogg, A. G., and Burgess, C., Analyst, 1982, 107, 213. Yang, J., Zhu, G., Si, Z., and Ma, J., Guijinshu, 1990, 11, 34. Paper 6/01 I74H Received February 19, I996 Accepted March 25, I996

ISSN:1359-7337    

DOI:10.1039/AC9963300167

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, May 1996, Vol33 ( I 71-1 74) 171 Elect rogenerated Chem il urn inescence Determination of Some Local Anaesthetics Andrew W. Knight and Gillian M. Greenway School of Chemistry, University of Hull, Hull, North Humberside, UK HU6 7RX This work describes the use of the emerging technique of electrogenerated chemiluminescence (ECL) for the determination of some local anaesthetics using an ECL reaction with tris(2,2’-bipyridine)ruthenium(11). Five local anaesthetic compounds are studied, and experimental conditions optimized to achieve detection limits of 5 X 10-8 moll-’ for bupivacaine and procaine, and 7 X 10-8 mol I-1 for lignocaine.The ECL method was successfully applied to the determination of lignocaine in a commercial product. Chemiluminescence (CL) methodology provides a very sensi- tive and selective means of detection for the analysis of drugs, especially when coupled with the powerful separation technique of HPLC.For the analysis of drugs in complex matrices such as biological fluids, alternative methods such as UV absorption and fluorescence detection are limited by the natural absorbance or fluorescence of numerous compounds present in such samples.However, the problem remains that there are only a limited number of directly chemiluminescent compounds. Hence derivatization procedures are often required. For ex- ample, fluorescein and fluorescamine derivatives of drugs have been determined by their excitation and subsequent light emission with the peroxyoxalate CL reaction. 1~ Much work has focused on the search for CL reactions of underivatized drugs for the additional advantages of simplification in methodology, sample preparation and instrumentation. Examples include methods for the determination of buprenorphine,’ m ~ r p h i n e , ~ paracetamo1,s tetracycline,b and thiamine,7 based on direct CL reactions with a range of oxidizing agents.The electrogenerated chemiluminescence (ECL) reactions of tris(2,2’-bipyridine)ruthenium(r1) [Ru(bpy)”+] provide a means for the direct detection of a range of drugs and pharmaceuticals that contain an aliphatic tertiary amine group.Furthermore, trialkylamines and closely related compounds are difficult to detect by alternative methods, since they do not absorb well in the UV/VIS region, exhibiting low molar absorptivities, and are extremely difficult to derivatize.Electrogenerated chemiluminescence is a technique whereby a CL reaction is initiated from reagents formed in the vicinity of an electrode surface when a potential is app1ied.x Electro- generated chemiluminescence reactions of Ru(bpy)32+ have so far been used to determine a variety of pharmaceuticals containing aliphatic amine groups including, erythromycin,g clindamycin, lo oxaprenolol, 1 1 codeine and related com- pounds,12 amitriptyline,I3 as well as other aminesI4 and aminoacids.1s,16 This paper reports on the use of a Ru(bpy)32+ ECL reaction for the determination of a variety of local anaesthetics. Local anaesthetics are widely used pharmaceutical compounds, which produce a reversible loss of sensation by diminishing the conduction of sensory nerve impulses, near to the site of application or injection. Most useful local anaesthetics have the same general chemical configuration of an amine functional group coupled to an aromatic residue by an ester or amide linkage.The ECL method described herein is based on a CL reaction involving the amine functional group of the local anaesthetic.Five local anaesthetics were chosen for the study whose structures are shown in Fig. 1 . The ECL method was also successfully applied to the determination of lignocaine in a commercial product. Instrumentation The ECL instrumentation used in this work has been described in detail p r e v i ~ u s l y . ~ ~ The ECL system basically consists of an ECL flow cell coupled with a flow injection manifold for sample introduction.The flow cell contains a platinum working electrode over which the reagents flow, and where the CL reaction is initiated in direct view of a facing photomultiplier tube. A platinum counter electrode and a silver pseudo reference electrode are also mounted within the flow cell. The potential of the working electrode is selected and applied by means of a computer controlled three-electrode potentiostat.Reagents Tris(2,2’-bipyridine)ruthenium(11) hexahydrate (Pract-grade, 90-9.5’36) was obtained from Fluka (Gillingham, Dorset, UK). The local anaesthetics, amethocaine, bupivacaine, lignocaine, prilocaine and procaine, were all obtained from Sigma (Poole, Dorset, UK). Anbesol Liquid was purchased from Whitehall Laboratories (Taplow, Berks, UK).The buffer used throughout the study was sodium dihydrogen orthophosphate (AnalaR- grade, 99-100%) obtained from Merck (Poole, Dorset, UK), and the pH was adjusted using solutions of sodium hydroxide (98%) obtained from Rh6ne-Poulenc (Manchester, UK). All solutions were made up using water prepared by reverse osmosis followed by ion exchange (Elgastat UHQ, PS I1 Elga Ltd., High Wycombe, UK).All reagents were used without further purification. Reaction Mechanism The ECL reaction of Ru(bpy)32+ with a tertiary aliphatic amine proceeds as follows. 1 8 A single, appropriate, positive potential is applied to the working electrode to produce the R~(bpy)3~+ active cation by oxidation of the parent complex, and at the same time to oxidize the amine.Oxidation products of the amine react immediately with the water present to form highly reducing intermediates, that can reduce the R~(bpy)~’+ back to Ru(bpy)3*+ in an excited state, which emits light with a maximum at a wavelength of 620 nm. Hence the reaction mechanism proceeds as follows: Ru(bpy)32+ -+ Ru(bpy)s3+ + e- Electro-oxidation (1) R2NCHzR’ -+ R2Nf-CH2R’ + e- Electro-oxidation (2) R*N+*CH2R’ + H20 -+ R2NC.HR’ + H30+ Deprotonation (3)172 Analytical Communications, May 1996, Vol33 R2NC.HR' + R~(bpy)33+ + 2H20 -+ R~(bpy)32+* + R2NH + (4) R'CHO + H30' Ru(bpy)32+* -+ Ru(bpy)32+ + hv Chemiluminescence ( 5 ) Results and Discussion In previous work the applied voltage and solution pH have been shown to be the two most important factors in obtaining an optimum ECL re~ponse.12.1~ The magnitude of the applied positive potential must be at least the oxidation potential of the amine, and the solution pH must be sufficiently high to facilitate the deprotonation step in the ECL reaction.(See reaction 3.) It was expected, however, that there would be some interaction between these two factors, and therefore a multivariate approach was required to optimize the applied voltage and pH simultaneously, and ascertain the significance of their inter- action.This was carried out by preparing a number of sample solutions for each compound, at different pHs covering the range 4.5 to 8.5. Each solution was continuously pumped through the flow cell whilst a simple linear voltage ramp was applied to the working electrode.The ECL response measured for each solution is a peak with a maximum at the optimum voltage for each compound and buffer pH. The peak height also gives the maximum ECL intensity at the optimum voltage for a particular pH (see Fig. 2). By measuring the light output variation with applied voltage, such an experiment is essentially the stationary electrode equivalent of linear voltammetry.moll-1) and Ru(bpy)32+ (1 X 10-3 moll-1) in phosphate buffer (0.05 mol 1-I), and was pumped continuously through the flow cell at a flow rate of 1.2 ml min-I. The concentration of the buffer and the Ru(bpy)32+ reagent for ECL studies were optimized and reported previously. 12,19 The voltage ramp was applied to the working electrode from +0.50 to +2.00 V (versus silver wire), at a scan rate of 20 mV s-1.The variation of ECL intensity with pH is shown in Fig. 3 for each local anaesthetic compound tested. The ECL intensity reached a maximum value at an optimum pH within the range 5.5 to 7.5 for most of the compounds tested. A notable exception was prilocaine, the only secondary amine Each sample solution contained the analyte (1 X in the study, with an optimum pH of approximately 8.5.The ECL intensity observed for lignocaine and procaine tended to increase once again above pH 8-8.5. It is well known that a background reaction producing an ECL emission occurs between OH- and Ru(bpy)32+, which is proportional to the OH- concentration, and hence becomes large and more significant at high pH.20 Hence the pH range above 8.5 is usually avoided if possible in ECL determinations of this type, in order to minimize the blank signal and enhance the limit of detection.Using the S/N as a means to find the optimum pH condition for maximum sensitivity, the optimum conditions determined are shown in Table 1. The S/N given in Table 1 is relative to that of bupivacaine. Table 1 also gives the optimum applied voltage (versus silver wire) for each compound.The optimum voltage varied little between compounds, this was expected as a similar electrochemical reaction is being evoked in each case. The optimum applied voltage also remained constant over the pH range 5.5 to 8.5 for each compound, but fell slightly below pH 5.5 by between 0.04 and 0.20 V.The notable exception was prilocaine, once again, for which the optimum voltage rose slightly at low pH to be +1.32 V at pH 5.5. The differences in observed ECL activity between the compounds tested can be explained by consideration of the structure of the compounds.21 The greater the number and length of the alkyl chains attached to the nitrogen atom involved in the ECL reaction, the greater the stabilization of the electron deficient radical intermediate in the reaction, by an inductive effect.Hence, the production of the radical intermediate is more favourable, and the ECL efficiency is higher. Bupivacaine, with effectively two butyl groups attached to the nitrogen atom, is therefore the most active compound. Lignocaine is a similar compound with shorter ethyl groups attached to the nitrogen, which are less efficient at stabilizing the radical intermediate, and hence the compound exhibits a lower ECL activity compared with bupivacaine.Procaine is similar to lignocaine, but the presence of the additional primary aromatic amine group may lower the signal, since this group may be preferentially oxidized in a competing non-chemiluminescent reaction. This effect is more apparent in amethocaine which contains a more readily oxidized aromatic substituted secondary amine.Finally Amethocaine (Tetracaine) Bupivacaine Lignocaine (Lidocaine) Prilocaine Procaine Fig. 1 Structural formulae of the five local anaesthetics investigated in this study.Analytical Comnzunications, May 1996, Vol33 173 prilocaine was expected to give a lower ECL response since the compound contains a secondary rather than tertiary aliphatic amine group.An in-depth study of the relationship between the structural attributes of potential analyte compounds and their observed ECL activity, including the local anaesthetics studied here, will be the subject of a later publication. A calibration was carried out on the three local anaesthetic compounds which produced the most intense ECL responses, Optimum applied h voltage c v 1 .- 5 404 / \ at the optimum voltage VoltageN Fig.2 ECL response curve for an applied linear voltage ramp. 90 80 70 > 6o z . 50 3 40 30 20 0 10 0 i i .-'-:::::I:/: / /' : I / 4 0 4 5 50 5 5 60 6 5 7 0 7 5 80 85 9 0 PH Fig. 3 Variation of ECL intensity, measured at the optimum applied voltage, with pH. 0, Bupivacaine; H, lignocaine; A , procaine; v, prilocaine; +, amethocaine.Table 1 Optimum experimental conditions for the determination of the local anaesthetics Optimum applied voltage (V) Relative )'PI'SUS signal :blank Compound Optimum pH silver metal ratio Bupivacaine 5.5 1.30 100 Lignocaine 5.5 1.24 63 Procaine 5.5 1.27 44 Prilocaine 7.5 I .26 18 Amethocaine 6.5 1.27 16 namely bupivacaine, lignocaine and procaine.A phosphate buffer carrier stream at pH 5.5 was used, into which 100 1-11 volumes of the standard solutions were injected. Each standard solution contained the local anaesthetic and Ru(bpy)3*+ ( I X 10-3 rnol 1-1) in phosphate buffer solution of pH 5.5. In this way standard solutions were transported into the flow cell and passed over the working electrode, where the ECL reaction was initiated.The working electrode was continuously held at the optimum applied voltage for the particular compound. Hence a response peak was recorded for each injection and a peak height could be measured. Calibrations were carried out in the range 5 X 10-5 to 1 X 10-7 mol 1-1, with five replicate injections per calibration standard.The calibration graphs were linear over the whole concentration range used. The lower and shorter concentration range 0 to 5 X 10-6 mol 1-1 was plotted on a linear scale and used to calculate the limit of detection, as the concentration of the analyte required to give a signal equivalent to the blank plus five times the standard deviation of the blank.The results are shown in Table 2. The method is specific to tertiary aliphatic amines, and hence only these compounds, if present in sufficient quantities, would present positive interference problems. The single exception is oxalate which produces a strong ECL signal under these conditions. Negative interference can be observed in the presence of a high excess concentration of a compound which is more readily oxidized than the tertiary amine, i.e.phenolic compounds. The non-chemiluminescent oxidation reactions of the interfering compound suppresses the ECL oxidation reactions of the analyte. If the ECL technique is used in conjunction with HPLC as a post-column detection system, however, these possible interferences should be eliminated. The proposed ECL method was tested in a preliminary experiment by directly analysing the lignocaine content of a commercial product.Anbesol LiquidTM, used in the treatment of mouth ulcers and denture irritation, was analysed by use of a simple external standard calibration method since no interfer- ences were observed to be produced by components of the product. A small sample of the liquid (approximately 0.3 g) was weighed and diluted to 50 ml with phosphate buffer (pH 5.5).An aliquot of this solution was then diluted by a factor of 10 and Ru(bpy)3*+ was added to a concentration of 1 X mol 1-l. Five replicate 100 pl injections of this sample solution were made and the average peak height was compared to the calibration of lignocaine. The result obtained was a lignocaine content of 0.92 k 0.01%.The manufacturer's claim is a lignocaine content of 0.90%. Hence the ECL method was shown to be suitable for the analysis of lignocaine in a pharmaceutical product with minimum sample preparation. Conclusions The limits of detection for the local anaesthetics achieved with this relatively simple method were comparable with previously reported more complex methodology, for example chromato- graphic22.23 and indirect atomic absorption spectrophoto- metric24 methods, and orders of magnitude better than spectro- photometric methods.25 The determination of local anaesthetics by a conventional CL method has been reported previously by Zhang et a1.2h Their method was based on the CL emission Table 2 Calibration characteristics and limits of detection over the range 0 to 5 X 10-6 mol I-' Correlation Average coefficient % RSD range % RSD LOD/ LOD/ Compound Calibration equation (n = 6) (n = 5 ) (n = 5) moll-' ng ml-1 Bupivacaine j = 3.21 x 106 x+ 1.23 0.998 1.65-3.98 2.52 S X lo-* 16 Lignocaine y = 2.65 x lo6 I + 1.59 0.997 1.3 1-3.43 2.3 1 7 X 10-8 20 Procaine y = 2.49 X lo6 x + 1.49 0.999 0.8 1-3.39 1.79 5 x 10-8 14174 Analytical Communications, May 1996, Vol33 observed on the oxidation of certain local anaesthetic com- pounds with an acidic solution of KMn04.Only the p - aminobenzoate type of local anaesthetics (e.g. amethocaine and procaine), gave measurable responses, and this was thought to be the key functional group responsible for the CL emission. Their limit of detection for procaine was comparable with the present proposed ECL method.The two methods, CL and ECL, are seen as complementary, since those compounds giving the most intense responses in the oxidative CL method (i.e. amethocaine), gave the lowest responses in the ECL study, and vice versa. For example, lignocaine gave no response with the oxidative CL method but was sensitively determined by the ECL method.In conclusion, the proposed ECL method was shown to be suitable for the determination of local anaesthetics, which contain an aliphatic tertiary amine functional group, in aqueous buffered solution. Compounds that also contain an aromatic substituted amine were found to give a lower ECL response, and compounds containing a secondary amine functional group, rather than the tertiary amine group, also gave a markedly low response.After optimization, the ECL method was successfully applied to the determination of lignocaine in a commercial product . References Kobayashi, S., Sekino, J., Honda, K., and Imai, K., Anal. Biochem., 1981, 112, 99. Mahant, V. K., Miller, J. N., and Thakrar, H., Anal. Chim. Acta, 1983, 145, 203. Alwarthan, A.A., and Townshend, A., Anal. Chim. Acta, 1986,185, 329. Abbott, R. W., Townshend, A., and Gill, R., Analyst, 1987, 112, 397. Koukli, I. I., Calokerinos, A. C., and Hadjiioannou, T. P., Analyst, 1989,114, 711. 6 7 8 9 10 I 1 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 Syropoulos, A. B., and Calokerinos, A. C., Anal. Chim. Acta, 1991, 255, 403. Grekas, N., and Calokerinos, A.C., Talanta, 1990, 37, 1043. Knight, A. W., and Greenway, G. M., Analyst, 1994, 119, 879. Danielson, N. D., He, L., Noffsinger, J. B., and Trelli, L., J . Pharm. Biorned. Anal., 1989, 7, 1281. Targrove, M. A., and Danielson, N. D., J . Chromatogr. Sci., 1990,28, 505. Greenway, G. M., and Knight, P. J., Anal. Proc., 1995, 32, 251. Greenway, G. M., Knight, A. W., and Knight, P. J., Analyst, 1995, 120, 2549. Greenway, G. M., and Dolman, S., Anal. Commun., 1996, 33, 139. Downey, T. M., and Nieman, T. A., Anal. Chem., 1992, 64,261. He, L., Cox, K. A., and Danielson, N. D., Anal. Lett., 1990, 23, 195. Jackson, W. A., and Bobbitt, D. R., Anal. Chim. Acta, 1994, 285, 309. Knight, A. W., and Greenway, G. M., Analyst, 1995, 120, 2543. Leland, J. K., and Powell, M. J., J . Electrochem. SOC., 1990, 137, 3127. Knight, A. W., Greenway, G. M., and Chesmore, E. D., Anal. Proc., 1995, 32, 125. Lytle, F. E., and Hercules, D. M., Photochem. Photobiol., 1971, 13, 123. Knight, A. W., Ph.D. Thesis, University of Hull, 1996. Le Guevello, P., Le Corre, P., Chevanne, F., and Le Verge, R., J . Chromatogr. B., Biomed. Appl., 1993, 622, 284. Yu, Z., Abdel-Rehim, M., and Westerlund, D., J . Chromatogr. B., Biomed. Appl., 1994, 654, 221. Valchrcel, M., Gallego, M., and Montero, R., J . Pharm. Biomed. Anal., 1990, 8, 655. Mohamed, A. I., Hassan, H. Y., Mohamed, H. A., and Hussein, S. A., J . Pharrn. Biomed. Anal., 1991, 9, 525. Zhang, X. R., Baeyens, W. R. G., Van der Weken, G., Calokerinos, A. C., and Imai, K., Anal. Chim. Acta, 1995, 303, 137. Paper 6/01 672C Received March 8, 1996 Accepted April 9, 1996

ISSN:1359-7337    

DOI:10.1039/AC9963300171

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, May 1996, Vol33 ( I 75-1 76) 175 Selective Isolation of Anethole from Fructus Anisi Stellati (Star Anise) by Supercritical Fluid Extraction Lin Kwok Liu Department of Chemistry, Hong Kong Baptist University, 224 Waterloo Road, Kowloon, Hong Kong Crude anethole over 90% purity (by GC) was obtained by extracting the ground star anise statically (10 min) and then dynamically (30 min) with supercritical carbon dioxide at a density of 0.35 g ml-l and a temperature of 80 "C.Pure anethole (99% by GC) could be obtained by passing the crude anethole onto a column packed with a mixture of silica and activated charcoal and eluted with 2% ethyl acetate in petroleum ether. Trans-anethole is a naturally occurring food flavouring agent in star anise.The fruit of star anise consists of volatile oil (4-9%) in which trans-anethole is the major component (80-90%) while the rest is pinene, safrole and limonene.1 Typical isolation of anethole from star anise by solvent extraction is tedious, time-consuming, labour-intensive, and causes a considerable contribution to waste production in the laboratory. In contrast, supercritical fluid extraction (SFE) is a practical alternative for the isolation of anethole because this technique can be performed using non-toxic and non-flammable super- critical C02. Although much information has been published to date on SFE,2-7 the use of SFE to selectively isolate anethole from star anise has not been reported.Heikes8 has reported the isolation of safrole (an aromatic oil also present in anise) and related alkylbenzenes in sassafras teas by SFE.The isolation of semivolatile flavour compounds from cinnamons using SFE has recently been reported.9 Experimental Standard and Reagents Trans-anethole standard, analytical-reagent grade methanol, ethyl acetate, petroleum ether, dichloromethane and Celite were obtained from Aldrich (Milwaukee, WI, USA).High- and low- grade carbon dioxide were obtained from Hong Kong Oxygen (Hong Kong). Instrumentation A Hewlett-Packard 7680T (Wilmington, DE, USA) super- critical fluid extractor with a Hewlett-Packard 1050 series modifier pump was used for the SFE extractions. All analyses were performed on an HP 5890 gas chromatograph equipped with an HP 5972 mass selective detector fitted with an HP-5 (cross linked 5 % phenylmethyl silicone, film thickness 0.25 pm) capillary column (300 X 0.35 mm id).The initial oven temperature was 80 "C; this was ramped to 300 "C at 10 "C min-1 and held for 5 min at the final temperature. Preparation of Samples Star anise samples were purchased locally and ground to powder with a blender before use. Two layers of filter paper discs cut to the internal diameter of the extraction thimble were placed on the bottom thimble cap and then 0.2 g of Celite was added to the thimble.A mixture of 0.5 g of the sample and 1.5 g of Celite was prepared in a pestle and mortar and then transferred quantitatively to the extraction thimble. Approxi- mately 0.2 g of Celite and two more filter paper discs were placed on top of the sample in the thimble so that the volume of the extraction cell was just filled.The procedures for the spike extraction experiments were similar to those described above for real samples experiments except that 1.5 g of Celite was spiked with 0.02 g of standard anethole. SFE Extraction Conditions A 7 ml extraction thimble (1 1 mm id X 95 mm) and SFE-grade carbon dioxide (99.90% purity) were used in the extraction.The sample was placed in the thimble as mentioned previously. Before extraction, a modifier was added to the extraction cell by pipetting 0.5 ml of methanol onto the sample in the extraction cell. The thimble was then placed into the extractor at a chamber temperature of 80 "C followed by a 10 min static extraction period and 30 min dynamic extraction with pure supercritical carbon dioxide at a flow rate of 0.5 ml min-1.After the static extraction, the supercritical fluid was depressurized onto a sorbent trap packed with diol and cooled to 5 "C. Once the SFE step was finished, the analytes were recovered by rinsing with 1 ml of dichloromethane at a flow rate of 1 ml min-' and collected in a 1 .5 ml vial. The rinse trap and nozzle temperatures were maintained at 30 "C and 25 "C, respectively. The procedure for extracting spiked samples was exactly the same as above except that 0.1 ml of methanol was added in the extraction cell to wet the sample.Results and Discussion Selectivity The selectivity of the method was assessed by extracting real samples using the optimum SFE conditions. Anethole over 90% purity (by GC) was consistently obtained after SFE for 30 min.Under the conditions used, the average amount of anethole extracted was 6.78% and other components were not extracted. Table 1 shows the results of spike recoveries on samples under dry and wet conditions. Method Validation on Real Samples Since good spike recoveries do not necessarily indicate good recovery of real samples because the analytes are not neces- sarily in the same chemical or physical locations in or on the sample matrix, additional validation methods are required to establish the reliability of the proposed SFE method.The extraction efficiency of the SFE method was performed by comparing the quantities of anethole extracted from real176 Analytical Communications, May 1996, Vol33 samples using solvent extraction, and SFE for 30 min with supercritical CO2. The average amount of anethole extracted from 10 replicate real samples by SFE, and solvent extractions were 6.78% (3.2% RSD) and 5.93% (8.2% RSD), respectively. The amount of anethole found in the samples by both methods was within the range quoted in the literature (3-8%).' The SFE results correlated with the solvent extraction data for all the real and spiked samples.The relatively low RSD value obtained by the SFE method reflected greater selectivity and precision than the solvent extraction method. It showed that SFE for 30 min with supercritical C02 was superior to dichloromethane solvent extraction (12% more extracted anethole than with dichloro- methane extraction).The reliability of the proposed SFE method was assessed by multiple sequential extractions of a single sample using SFE and solvent extractions. Star anise was extracted with dichlo- romethane (1 5 h reflux) after SFE for 30 min with supercritical CO2 to see whether or not any additional anethole was recovered. In contrast, the dichloromethane extract performed after the supercritical CO2 extraction yielded a minimal amount (about 1 %) of additional anethole, indicating that supercritical CO2 yielded quantitative recovery ( > 90%) of the native anethole. Table 1 Recovery of anethole from 10 replicates spiked with 0.02 g anethole Matrix Supercritical fluid condition* C02 wet C02 + 5% methanol wet C02 + 10% methanol wet co2 dry C02 + 5% methanol dry C02 + 10% methanol dry Average o/c recovery+ RSD (%) 80 2 95 2 65 3 71 3 45 3 52 3 * Celite 1.5 spiked with 0.02 g anethole.Wet condition means 0.1 ml methanol added to the sample before the extraction. + Experimental conditions: 10 min static followed by 30 min dynamic extraction at 80 "C and 1916 psi. Conclusion The developed method is especially suitable for the deter- mination of anethole in food, natural products and/or biological samples.The first fraction collected after the SFE step can be analysed directly by GC or HPLC without further purification. The optimized SFE method is fast, efficient and selective as compared with the solvent extraction method. It consumes far less organic solvents, by an order of 10 to 100 times, than solvent extraction.This substantially reduces the cost per sample, both for purchase of solvents as well as the disposal of the used toxic wastes. SFE is a significantly more 'environmen- tally friendly' method than traditional methods of sample preparation. The author thanks Hong Kong Research Grant Council for funding this project; Grant number REC/94-95/04. References Pharmacopoeia of the People's Republic of China, ed.Tu Guoshi, The People's Medical Publishing House, Beijing, English edn., 1988. Sueprcritical Fluid Extraction and its use in Chromatographic Sample Preparation, ed. Westwood, S. A., Chapman & Hall, New York, 1993. Camel, V., Tambute, A., and Gaude, M. J., J . Chromatogr., 1993, 642, 263. Janda, V., Bartle, K. D., and Clifford, A. A., J . Chromatogr., 1993, 642, 283. Analysis with Supercritical Fluids: Extraction and Chromatography, ed. Wenclawiak, B., Springer-Verlag, New York, 1992. Analytical Supercritical Fluid Extraction, ed. Luque de Castro, M. D., Valcarcel, M., and Tena, M. T., Springer-Verlag, New York, 1994. Supercritical Fluid Technology: Theoretical and Applied Approaches in Analytical Chemistry, ed. Bright, F. V., and McNally, M. E. P., American Chemical Society, Washington, DC, 1992. Heikes, D. L., J. Chromatogr. Sci., 1994, 32, 253. Miller, K. G., Poole, C. F., and Chichila, T. M. P., J . High Resolut. Chromatogr., 1995, 18, 416. Paper 6101472K Received March 1 , 1996 Accepted March 28, 1996

ISSN:1359-7337    

DOI:10.1039/AC9963300175

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, May 1996, Vol33 ( 1 77-1 81) Use of Haematoxylin in the Spectrophotometric Determination of Alkaloids in Pharmaceuticals, Galenicals and Powdered Plants Enaam Y. Backheeta, Hassan F. Askalb and Gamal A. Salehb LI Department of Pharmacognosy, Faculty of Pharmacy, Assiut University, 715I6-Assiut, Egypt f7 Department of Pharmaceutical Analytical Clzemistiy, Faculty of Pharmacy, Assiut University, 7151 6-Assiut, Egypt A simple and sensitive method for the determination of about fifteen alkaloids is described.The procedure is based on the reaction of the alkaloids with haematoxylin reagent in aqueous medium to give a red-violet colour (A,,, = 555 nm). Several variables affecting the colour development were studied and optimized. The method was used to determine 2-40 pg ml-1 of the final measured alkaloidal base solution.There was a considerable increase in the absorptivity, which was dependent on both the basicity (pK,) and molar concentration of the alkaloid present. The simplicity of the method permits rapid analysis, suitable for routine quality control. The interference of about 28 substances which are commonly prescribed with alkaloids was studied.No interference due to additives commonly present in the pharmaceutical preparations was observed. The method could be applied for the analysis of alkaloids in pure and pharmaceutical dosage forms, galenicals and powdered plants. Alkaloids occupy a prominent historical position among extensively employed medicaments. Accordingly, various ap- proaches to their colorimetric analysis were reported.They include extraction by ion pairing with a dye anion, 1-4 quantita- tive precipitation as reineckate and the subsequent dissolution in acetone"6 and charge transfer complex formation with iodine and ~~-acceptors.7-~0 The British Pharmacopoeia ( 1993) speci- fies the extraction of individual alkaloids followed by acid-base titration and TLC check on foreign bases.1 1 Some individual alkaloids were determined by measuring the developed colour after oxidation,I2 condensation13 or nitration.14 Most plant alkaloids are derivatives of tertiary amines, while others contain primary, secondary or quaternary nitrogen. Their basicity therefore varies greatly, depending on which of the four types is represented.The pK, values are 0-4 for very weak bases (purines), 4-7 for weak bases (papaverine, pilocarpine and reserpine) and 7-1 1 for medium strength bases (tropine, atropine, nicotine and emetine). l 5 As most alkaloids are weakly UV-absorbing and the reported methods for their analysis are time-consuming, the development of a simple, rapid and sensitive procedure is required. Haematoxylin has been used mainly for the analysis of metals,I6 in manufacturing of ink and chiefly as a stain in microscopy. l 7 Oxidized haematoxylin has been used for the determination of some penicillins and cephalosporins.18 The aim of the present work is to develop a method suitable for the quantitative analysis of some of these alkaloidal bases, depending upon their pK, values, using haematoxylin reagent.The method could be utilized for the determination of many alkaloids in pure and pharmaceutical formulations as some plants and liquid extracts containing alkaloids. 177 well as in Experimental Apparatus Uvidec-320 (Tokyo, Japan) and Perkin-Elmer 3B UV/VIS (Norwalk, CT, USA) spectrophotometers with I cm quartz cells were used. All volumetric measurements were made with standard glassware.Reagents and Materials All solvents and reagents were of analytical-reagent grade. Doubly distilled water was used throughout. Haematoxylin. Obtained from Aldrich, Poole, Dorset, UK; 0.2% (m/v) is prepared fresh daily by dissolving 100 mg in 2 ml of 0.5% (m/v) boric acid and diluting to 50 ml with doubly distilled water,. Boric acid. Solutions, 0.5 and 0.05% (m/v) were prepared in distilled water.Alkaloids. Tropine, quinine, ephedrine, nicotine and ajmaline bases; brucine trihydrate; emetine hydrochloride; sparteine, (L)- lobeline and strychnine sulfates; (m)-homatropine and hyo- scine hydrobromides; codeine phosphate and physostigmine salicylate were obtained from different manufacturers and were used as working standards.Formulations The following available commercial preparations were analy- sed: (1) Ephedrine HCl tablets (Kahira Pharmaceuticals and Chemicals, Cairo, Egypt), containing 0.5 grain of ephedrine HC1 per tablet; (2) Efanol tablets (Memphis, Cairo, Egypt), containing ephedrine HCI (20 mg), chlorpheniramine maleate (2 mg), dihydroxypropyl theophylline (250 mg) and pheno- barbitone (10 mg) per tablet; (3) Ephedrine sulfate ampoules (Misr, Cairo, Egypt), containing ephedrine sulfate (50 mg) per ampoule; (4) Asthmolase tablets (Memphis), containing ephed- rine HCI (20 mg), papaverine HC1 (10 mg), homatropine (0.5 mg), caffeine (20 mg) and phenobarbitone (20 mg); ( 5 ) Tepedrine tablets (Misr), containing ephedrine HCl (25 mg), theophylline (120 mg) and phenobarbitone (8 mg); (6) Atropine sulfate ampoules (Nile, Cairo, Egypt), containing atropine sulfate ( 1 mg ml-I); (7) Isoptoatropine eye drops (Alcon, Puurs, Belgium), contain atropine sulfate ( 1 %), benzalkonium chloride (0.01 5%) and hydroxypropyl methyl cellulose (0.5%); (8) Bellacid tablets (CID, Cairo, Egypt), containing extract of Belladona siccum (0.01 g) and phenobarbitone (0.02 g); (9) Supergine ampoules (Memphis), containing homatropine178 Analytical Communications, May 1996, Vol33 methylbromide (2.5 mg), dipyrone (1 g) and papaverine (50 mg); (10) Codacetine tablets (Kahira Pharmaceuticals and Chemicals), containing salicylamide (300 mg), paracetamol (250 mg) and codeine phosphate (10 mg); (11) Vegaskine tablets (Alex, Alexandria, Egypt), containing acetyl salicylic acid (300 mg), paracetamol (200 mg) and codeine phosphate (10 mg); (12) Buscopan ampoules (CID), containing hyoscine butyl bromide (20 mg); and (13) Butacid tablets (CID), containing hyoscine butyl bromide (10 mg).Liquid extracts Nux vomica (CID), Belladonna (Paul Mugenburg, Germany), Hyoscyamus (Memphis), Ipecacuanha (Arab Drug, Egypt) were used. Powdered plants Cinchona bark consists of the dried bark of Cinchona succirubra Pavon, family Rubiaceae; Rauwolfia consists of the dried rhizome and roots of Rauwolfia serpentina, family Apocynaceae; and cigarette samples were obtained from Eastern Company, Egypt. Procedure Standard solutions Dissolve a calculated amount of alkaloidal base in absolute ethanol and dilute quantitatively to 0.02-0.4 mg ml-1.For alkaloidal salts, dissolve calculated amounts in about 20 ml of distilled water and transfer to a 100 ml separating funnel. Add 2.0 ml of 33% ammonia solution and extract with three 25 ml portions of chloroform, passing the separated organic layers through 5 g of anhydrous sodium sulfate supported in a small funnel. Evaporate chloroform, dissolve the residue in 10 ml of ethanol in a 100 ml calibrated flask and make up to the mark with the same solvent.Sample solutions Liquid preparations (ampoules and eye drops). Transfer aliquot portions equivalent to 50 mg of alkaloidal base into 50 ml of distilled water in a 100 ml separating funnel and continue as described above for standard solutions. Tablets of alkaloidal salts. Place an accurately weighed amount equivalent to 50 mg alkaloidal base from composite of 20 powdered tablets in 50 ml of distilled water in a 100 ml separating funnel and continue as described above for standard solutions. Galenical preparations Evaporate 10 ml of the liquid extract to dryness. Extract the residue with water acidified with sulfuric acid (1 + 100).Transfer the aqueous extract into a 100 ml separating funnel through a filter paper and wash with 2 ml of dilute sulfuric acid (1 + loo), add the aqueous washings to the mother liquor and render it alkaline to litmus with 33% ammonium hydroxide solution ( = 2 ml).Shake with successive portions of chloroform until the alkaloids are completely extracted (3 x 25 ml). Evaporate the combined chloroformic extract on a water-bath to dryness and extract with ethanol in a 100 ml calibrated flask.Dilute quantitatively 1.0 ml of this solution to 100 ml with ethanol. Powdered plants A 20 g sample of cinchona bark or rauwolfia root powders, dried and ground to 40 mesh was heated under reflux with 100 ml of 70% ethanol for 2 h. The extract was concentrated to 25 ml. Extract with water acidified with sulfuric acid (1 + 100) and continue as described above under Galenical preparations beginning from ‘Transfer the aqueous extract .. . ’. Cigarette Fillers Remove sufficient filler from tobacco product. Weigh 2 g of the filler into a tared 250 ml Erlenmeyer flask, add 100 ml of 0.05 mol 1-1 sulfuric acid and place on a shaker for about 15 min. Transfer into a separating funnel through a filter paper.Render alkaline to litmus with 33% ammonium hydroxide solution and continue as described above under Galenical preparations. General procedure Pipette 1.0 ml of the standard or sample solutions into a dry 10 ml calibrated flask, add 1.0 ml of haematoxylin reagent. Allow to stand for 30 min at 25 f 5 OC, add 1 ml of 0.05% boric acid solution. Complete to the mark with distilled water and measure the absorbance at 555 nm against a reagent blank.Results and Discussion Addition of haematoxylin aqueous solution to an aqueous alcoholic solution of any of the investigated alkaloidal bases 0.8 - I Wavelengthhm Fig. 1 reaction with atropine (16.5 yg ml-l). Absorption spectra of A, 0.2 mg ml-1 haematoxylin and B, its Table 1 Some spectral characteristics of the investigated alkaloids 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Alkaloidal base Tropine Atropine Homatropine Ephedrine Sparteine Quinine Emetine Ajmaline Codeine Brucine Strychnine Nicotine Phy sostigmine Hyoscine Lobeline Linear rangel yg ml-I 2-12 2-20 5-25 2-15 2-20 5-25 5-40 5-35 2-30 5-40 5-40 2-15 5-30 5-30 5-25 Intercept (a) -0.0173 -0.0064 -0.0180 0.0060 -0.0060 - 0.0090 -0.0040 -0.005 1 0.0020 -0.0050 0.0047 -0.0103 0.0150 0.0020 -0.0064 Slope (h) 0.07 16 0.0346 0.0370 0.0540 0.0400 0.0273 0.0 184 0.0263 0.0283 0.0200 0.0253 0.0525 0.0300 0.0223 0.0375 Y 0.9996 0.9952 0.9996 0.9999 0.9992 0.9957 0.9970 0.996 1 0.9996 0.9985 0.9985 0.9986 0.9997 0.9990 0.9994Analytical Communications, May 1996, Vol33 179 yields an intense red-violet colour.The absorption spectrum of the resulting chromogen is characterized by the formation of a new band (A, = 555 nm) which was not found in the spectrum of either the reagent or the alkaloids (Fig. 1). A fresh solution of haematoxylin possesses an absorption band at 290 nm. The stability of the reagent was investigated. Haematoxylin was found to be unstable in water as its solution darkens on standing. Trials were made for the stabilization of its aqueous solution and it was found that 2.0 ml of 0.5% (m/v) boric acid solution per 50 ml reagent solution is a suitable medium for this purpose (haematoxylin borate complex).The effect of diluting solvents was studied. As assay solvent, water afforded maximum sensitivity and stability. Methanol, ethanol, propan- 1-01, propan-2-01 and acetone were not suitable as diluting solvents because they gave lower absorbance readings.Halogenated and other water immiscible solvents were found to be unsuitable as the formed chromogen is not extractable into these solvents. The stability of the colour was investigated. Maximum absorbance readings at 555 nm were achieved after 30 min duration followed by quenching the reaction with 1.0 ml of 0.05% boric acid solution.Stability was then manifested for more than 30 min at 25 k 5 "C. The effect of reagent concentration was investigated. The colour develops strongly as the amount of haematoxylin reagent in the reaction mixture was increased from 0.5 to 2.0 ml of 0.2% solution. Therefore, 1.0 ml was used throughout the experi- mental work.r 1 8.0 I I I I I I A 7.5 8.0 8.5 9.0 9.5 10.0 10.5 pK, of alkaloid bases Fig. 2 Molar absorptivity t'ersus pK, values of haemotoxylin reaction products with alkaloids. (pK, values from ref. 19 except for compound 1 from ref. 20 and 5 from ref. 17). Compound numbers refer to Table 1. Under the specified reaction conditions, Beer's law was obeyed for 2-40 pg ml-1 of the final assayed solution.Slopes, intercepts and correlation coefficients are given in Table 1. The mean of six replicate analyses of the different investi- gated alkaloids gave s values of 1.20-1.90%. This level of precision is adequate for the quality control analysis of pharmaceutical preparations and natural products. The sensitivity of the assay, expressed as &-values varies with different alkaloids in a regular pattern which was found dependent upon the pK, values of these alkaloids, Fig.2. Regression analysis of the above correlation by the method of least squares afforded r = 0.9801 and a regression equation of E = 3063.8 + 679.9 pK, From this equation, E for the coloured Table 2 Analysis of certain alkaloids in the presence of some coformulated ingredients Ingredients* Phenobarbitone"ac,eJ1 DipyroneU.< .If Salic ylamidec-e Cetrimidee Hydroxypropylmethylcellulosea~h Acetylsalicylic acid< Paracetamol' Dexamethasonee Sodium benzoatee Sodium citratea Sodium salicylatee Sodium penicillin Ge Sulfadimidineh Sulfagaunidineh Sulfanilamide" Vitamin BI" Vitamin BZh Vitamin Bhl' Nicotinamideh Vitamin Cy Naphazoline nitrate' N i ke thami dee Piperazine hydrate" Menthole Diiodohydrox yquinolineh Bismuth carbonate" Kaolin'z Potassium bromide" Amount addedlmg 400 250 300 2 500 300 250 1 15 10 5 10 150 1500 400 3 1 1 10 10 125 250 1800 50 13 150 1500 400 Recovery of alkaloid (%) k s+ 98.44 f 1.46 100.58 f 1.57 99.27 k 1.85 104.17 f 0.98 99.03 f 1.39 98.59 k 1.41 100.35 f 1.06 103.55 f 1.60 101.10f 1.11 101.31 rf: 1.61 98.78 f 1.23 99.27 f 0.98 99.17 f I .53 139.89 k 1.67 98.34 k 1.25 96.95 f 1.8 1 100.42 f 0.83 97.65f 1.11 99.03 f 1.39 100.49 f 1.48 149.50 f 1.85 100.49 k 0.86 300.29 _+ 1.46 98.78 f 1.35 101.11 f 1.39 99.17f 1.67 100.83k 1.11 102.49 f 1.61 *u,r'.e.h are alkaloids atropine (0.5 mg), codeine (15 mg), ephedrine (SO mg), homatropine (2000 mg), respectively, added to the above in- gredients.? n = 3 Table 3 Analysis of some pharmaceutical dosage forms containing alkaloids Dosage forms" Atropine sulfate ampoules Isoptoatropine eye drops Bellacid tablets Ephedrine tablets Efanol tablets Ephedrine sulphate ampoules Asthmolase tablets Tepedrine tablets Supergine ampoules Codacetine tablets Vegaskine tablets Buscopan ampoules Butacid tablets Label Found Added/ claimlmg (%) f s i mg I 1% 1 0* 32.5 20 50 20 25 25 10 10 20 10 99.91 f 1.72 1 101.93f 1.78 1% 100.75 f 2.01 1 98.62 f 1.58 32.5 99.91 f2.27 20 98.90 k 1.80 50 100.74 f 1.85 20 99.27 f 1.23 25 99.03 f 1.67 25 98.60f 1.41 10 99.12 f 0.72 10 100.44k 1.18 20 98.96 f 1.63 10 Recovery (%) f S+ 100.80 f 1.46 100.98 f 1.90 99.20 f 1.70 98.67 f 1.85 99.30 _+ 1.72 100.50 f 1.72 99.27 f 1.48 100.29 2 0.98 100.83rt 1.11 100.70f 1.58 101.23f 1.41 99.1 1 k 1.78 l00.89f 1.04 * See text for detailed composition.+ n = 4. * As extract Belladona siccum and calculated as atropine.180 Analytical Communications, May 1996, Vol33 product of an alkaloid with the reagent could be predicted from its corresponding pK, value. Specificity and Interference The proposed procedure has the advantage that the assay is performed at 555 nm in the visible region away from UV- absorbing interferents that might be co-extracted from dosage forms, liquid extracts or powdered plants.The failure of papaverine, reserpine, pilocarpine, colchicine and ergotamine to interact with the reagent is certainly related to their weak basicity (pK, 6.4,6.6,6.9, 1.7 and 6.3, respectively).Purine bases: caffeine, theobromine and theophylline gave a negative response under the specified reaction conditions: these compounds have very low pK, values and actually behave as weak acids.2l Table 2 shows the results obtained from the analysis of some of the investigated alkaloids in the presence of ingredients commonly co-formulated with them.Interference from acidic substances could be eliminated by extraction at alkaline pH but highly basic lipophilic compounds, if present, would interfere with the analysis. Analysis of Pharmaceutical Preparations The strongly red-shifted band combined with the high intensity of absorption and the very low reagent background makes this method suitable for routine alkaloid analysis with minimum interference.The proposed procedure was applied to assay a number of commercial preparations containing alkaloids. Common ingredients of formulations such as tablet excipients were found not to interfere in the method when following the proper extraction procedure. The results in Table 3 indicate high accuracy and confirm the suitability of the proposed procedure for the analysis of the investigated alkaloids in the micro range. In tablets containing phenobarbitone (Efanol, Asthmolase and Table 4 Assay of alkaloids in some galenicals Table 5 Assay of alkaloids in some powdered plants Amountlg Recovery Liquid extract Found Reported* Added (%j k S+ Nux Vomica 2.40; 2.45-2.55 1.0 98.80 k 1 .50 Ipecacuanha 2.039 1.95-2.05 2.0 99.20 k 1.20 Belladonna 0.357 0.27433 0.4 97.60 5 1.70 Hyoscyamus 0.901 0.95-1 .05 1 .0 97.50 ? 1.90 * Ref.21 and 22. f n = 3, t Calculated as strychnine. !i Calculated as emetine. 1 Calculated as atropine. Amountlg Recovery Plant Found Reported Added (%j. k s* Cinchona bark 6.36; 5-10t 6.00 97.20 -t 2.30 Rauwolfia root 0.85s 0.7-2.4t 0.85 97.80 f I .70 Cigarette fillers p.221 0.1711 0.20 98.10 * 2.00 * n = 3.t Calculated as quinine. Ref. 22. a Calculated as ajmaline (non reserpine alkaloids). 1 Calculated as nicotine. 11 Ref. 23. OH HO HO 1 H - 0 0 t I 0 0 ' H J 5 OH - 0 OH +I Alkaloid OH H6 2 Alkaloidal salt OH + H' 4 Scheme 1 3Analytical Communications, May 1996, Vol33 181 Tepedrine tablets), the extraction at the alkaline pH was sufficient to separate the acidic phenobarbitone from the basic alkaloids.Chloropheniramine maleate, in spite of its consider- able basicity, was found not to interfere with the analysis of Efanol tablets. This could be attributed to its low content relative to ephedrine content as well as the higher solubility of chloropheniramine base in aqueous phase. The data suggest good recoveries of added standards. Analysis of Galenical Preparations and Powdered Plants Table 4 gives the alkaloid contents of several liquid extracts analysed by the proposed method.The values were found to be consistent with the reported concentration ranges of these alkaloids.21.22 In addition, analysis of alkaloids in some powdered plants could be also carried out. Table 5 shows the total alkaloid contents in cinchona bark, rauwolfia root and in cigarette fillers.No interference was observed from the presence of the very complex matrix of the plants. The accuracy of the method was confirmed by the good recovery of added alkaloids to the respective liquid extract. Reaction Mechanism Haematoxylin (1) has been previously oxidized to give haematein (2).*4,25 Since the formation of haematein is greatly facilitated by alkaline conditions, it seems likely that the removal of hydrogen is involved and it is suggested that alkaloids are the essential species responsible for the hydrogen acceptance.In this respect, the hydroxy group at position 9 of haematoxylin was oxidized to the corresponding keto deriva- tive. The second stage is the ionization of the produced haematein which corresponds to the appearance of the 555 nm peak.The ionized haematein will possess two possible resonance forms (3,4) and these account for the intense colour of the ionic solution. The actual structure of the haematein ion will approximate to something between the two structures (3) and (4) and will be similar to (5). Scheme 1 shows the possible react ion path way. Conclusion The proposed method is simpler, less time-consuming and more sensitive than the official titrimetric method, it could be applied for the routine quality control analysis of the investigated alkaloids in pure, pharmaceutical and galenical forms as well as in some powdered plants after adopting a suitable extraction procedure.The method is particularly recommended for assaying the tropine alkaloids and other weak UV-absorbing alkaloids such as ephedrine and sparteine.The proposed procedure could be considered specific for alkaloids having pK, values greater than 7.6. However, it must be considered non- specific with regard to differentiation between them. These shortcomings do not affect the utility of the method in routine analysis and content uniformity determination of singly pre- scribed alkaloids. The disadvantages may also be overcome by coupling the proposed method to a suitable separation proce- dure.The enhanced sensitivity of the method allows for additional handling without risk of increasing errors. References 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Malat, M., Anal. Chim. Acta, 1979, 109, 191.Moriyasu, M., Ichimaru, M., Nishiyama, Y., and Kato, A., Bunseki Kagaku, 1993, 42, 659. Huang, T., and Jin, H., Zhongguo Yaoxue Zazhi, 1991, 26,733. Popov, D. M., Krasnokutskay, I. S., Litvnenko, T. N., and Safronova, T. O., Farmatsiya (Moscow), 1987, 36, 55. Weyers, J., and Skora, M., Dim. Pharm., 1962, 14, 201. Bandelin, F. J., J . Am. Pharm. Assoc. (Sci. Ed.), 1967, 39, 130.Krishnamurthy, M., and Muralikrishna, U., Ind. Drugs, 1985, 22, 171. Taha, A., and Rucker, G., Arch. Pharm. (Weinheim) Ger., 1977,310, 485. Gomaa, C., and Taha, A., J . Pharm. Sci., 1975, 64, 1398. Taha, A., and Gomaa, C., J . Pharm. Sci., 1976, 65, 986. British Pharrnacopeia 1993, HM Stationery Office, London, 1993, vol. 1. Rai, M., Raniachandrand, K. N., and Gupta, V. K., Analyst, 1994, 119, 1833. Smith, C. L., and Cooke, M., Analyst, 1987, 112, 1515. Nir-Grosfeld, I., and Weissenberg, E., Drug Stand., 1957, 25, 180. Wagner, H., Bladt, S., and Zgainski, E. M., Plant Drug Analysis, Springer-Verlag, Berlin, 1984, p. 5 I . Bishop, E., Indicators, Pergamon Press, Oxford, 1972, p. 383. The Merck Index, ed. Budavari, S., Merck, Rahway, NJ, USA, 1989, 1 1 th edn. Sastry, C. S. P., Satyanarayana, P., Rao, A. R., and Singh, N. R. P., Mikrochim. Acta, 1989, I, 17. Reynolds, J. E. F., Martindale, The Extra Pharnzacopeia, Royal Pharmaceutical Society of Great Britain, The Pharmaceutical Press, London, 1993, 30th edn. Martin, A. N., Swarbrick, J., and Cammarat, A., Physical Pharmacy, London, Febiger, Philadelphia, 1969, 2nd edn. Egyptian Pharmacopeia, Arabic Edition, El-Amiria Press, Cairo, Egypt, 1972. Evans, W. C., Trease and Ei>ans, Pharmacognosy, Great Britain University Press, Cambridge, 1994, 13th edn. Gottsho, A. M., Lin, J. A., Duck, W. N., and Losty, T. A,, J . Assoc. Off. Anal. Chem., 1988, 71, 11 10. Masoud, M. S., and Hagaag, S. S., Ind. J . Chem., 1982, 21A, 323. Saleh, G. A., and Askal, H. F., Anal. Lett., 1995, 28, 2663. Paper 6/01 230B Received February 20, I996 Accepted April 3, 1996

ISSN:1359-7337    

DOI:10.1039/AC9963300177

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Analytical Communications, May 1996, Vol33 183 Traceability at Pittcon 96 R. F. Walker Laboratory of the Government Chemist, Queens Road, Teddington, Middlesex, ~- UK 7’WI I-OLY VALID ANALYTICAL MEASUREMENT The traceability of analytical measure- ments was a major theme at this year’s Pittcon meeting. CJTAC, the relatively recently formed working group ‘Co-oper- ation on International Traceability in Ana- lytical Chemistry’, continued its close association with the Pittsburgh Conference by organizing two major international meetings on the traceability of reference materials and of food analysis.Reference Materials, Traceability and Accreditation The meeting on ‘Reference Materials, Traceability and Accreditation’ looked at recent developments aimed at providing traceable and transparent quality in the field of reference material production.Delegates also discussed the issues in- volved in the supply and use of reference materials, as well as the possibility of developing an internationally recognized chemical measurement system linking high-level reference measurements to working-level test measurements. Dr. Bernard King (LGC, UK) gave a keynote presentation which adressed the question of how can the user be sure that the certified reference materials (CRMs) they buy are suitable for their require- ments. Dr.King noted that international organizations like CITAC can work with others to; provide an adequate range of reference materials which are marketed on a world-wide basis; supply information to users about the materials available and best practice in their use; establish mechanisms for identifying priority needs and for international collaboration on reference material production and marketing; ensure that reference materials are fit for purpose and that users are provided with an assur- ance of quality; and develop a structured chemical measurement system which pro- vides traceability across national bounda- ries and provides links between high level reference measurements and working level analysis.Dr. Larry Keith (Radian Corporation) and Dr. Paul Bouis (J. T. Baker Inc.) noted that there are now so many analyte/matrix combinations involved in environmental measurements that they far exceed what few CRMs are currently available or even what are capable of being produced by the national CRM producers.This naturally creates a problem of traceability to appro- priate national or international standards when producing commercial reference ma- terials. To try and overcome this situation, the American Chemical Society (ACS) Committee on Analytical Reagents and a committee on environmental monitoring are working together to provide an ACS perspective on the production and use of reference materials for which there is no such traceability.The approach that is gaining favour is to produce a series of written procedures with agreed acceptance criteria which, when met, qualify a com- mercially-produced environmental refer- ence material to be certified as complying with a specified grade. To try and provide appropriate trace- ability to national CRMs, the US National Institute of Standards and Technology (NIST) are working with secondary refer- ence material producers to provide specific linkages, matrices and protocols (including statistics), along with procedures for ver- ification.Mr. Bill Reed (NIST) described the work ongoing at NIST and noted that, in order to achieve this objective, it was necessary to devise suitable descriptions of traceability, not only for the measurement procedure, but also for sampling, sample handling, the reference materials used to validate the linkages, limitations in the composition of the matrix and the statistics used to describe the uncertainty of the certified value.Mr. Peter Unger (American Association for Laboratory Accreditation, A2LA) made the case for the need to ensure that all reference material producers have an ade- quate quality system in place which is in compliance with I S 0 9001/2, together with adequate testing competence (compliance to IS0 Grade 25).Finally, Dr. Ron Walker (LGC) described the soon-to-be-published IS0 Guide 34 document which provides specific guidance on the quality system requirements for reference material pro- duction.The Guide includes details on all aspects of production including sample preparation, homogeneity and stability testing, traceability, characterization, measurement uncertainty and also wider issues, such as storage, packaging and customer service. Following on from this, Ron Walker concluded the formal presen- tations by summarizing the possible ways forward for ensuring that CRM producers comply with minimum quality require- ments.These included the formal accredi- tation and certification requirements, together with the proposal that an inter- national register of producers could be compiled. Registration would require a producer to meet certain criteria including, for example, accreditation to I S 0 Guide 25, registration to IS0 9000 series stan- dards and compliance with the IS0 Guides concerned with reference material production (i.e., Guide 31, contents of certificates; Guide 34, quality system requirements; Guide 35, certification procedures).Food Analysis The second meeting organized by CITAC was concerned with ‘Traceability of Food Analysis’. Issues discussed included the different perspectives on traceability be- tween food analysts, who rely extensively on method validation, and chemical met- rologists, who employ a hierarchy of reference methods and materials ideally linked to the International System of Units (SI).Bernard King took the lead in setting the picture and his presentation explored some of the common misunderstandings that are in danger of creating a gulf between analysts and metrologists.Dr. Wayne Wolf (US Department of Agriculture) noted that the Association of Official Analytical Chemists International (AOACI) had recently expanded its mis- sion statement to include the promotion of quality measurements, arising from its members’ needs to obtain accuracy state- ments in application of AOAC Official Methods of Analysis.The Association has also established a Technical Division on Reference Materials, which is promoting the use of CRMs and check samples for method validation, collaborative studies and individual measurement assessment. Bill Reed (NIST) described how refer- ence materials are certified by various national CRM producers to exacting metrological levels and how these qualities are then transferred down the measurement chain.In this context NIST are developing a number of NIST Traceable Reference Materials (NTRMs). The European Union is working towards establishing a single market without trade barriers, sound technical regulation and economic growth. In order to achieve these objectives it will be necessary to improve184 Analytical Communications, May 1996, Vol33 the reliability and comparability of chemical measurements.As a result there should be fewer repeat analyses and fewer disputes arising from different analytical results. Dr. Maire Walsh (Irish State Laboratory) described how the European Commission is working with a variety of independent organizations such as EURACHEM to facilitate the mutual acceptance of test data, and how in the context of food analysis the concepts of reference laboratories and specifications for method performance are being devel- oped.Finally, Dr. Robert Kaarls (Netherlands Measurement Institute, NMI) noted that 20 years after the mole has been accepted as the SI unit for the amount of (chemical) substance, a new programme of work has been initiated by the Cornit6 International des Poids et Mesures (CIPM) to help establish the traceability of chemical meas- urements at the highest metrological level.Two studies are already well advanced, covering gas analysis and trace elemental analysis by isotope dilution mass spec- trometry. Additional work is also under consideration on traditional techniques, such as gravimetry and titrimetry, and on the difficult area of developing primary methods for the analysis of trace amounts of organic compounds. Future CITAC Plans CITAC has plans to host a joint workshop with EURACHEM on the traceability and comparability of amount of substance measurements to be held in the Nether- lands in September, 1996. Further infor- mation about this and other CITAC activ- ities is contained in CITAC News. Copies and further information are available from the CITAC Secretariat, Laboratory of the Government Chemist, Queens Road, Ted- dington, Middlesex TWll OLY (Tel. 0181-943-7612; Fax 0181-943-7565; in- ternet: citac@ 1gc.co.uk). Paper 6102367C

ISSN:1359-7337    

DOI:10.1039/AC9963300183

出版商:RSC    

年代:1996

数据来源: RSC