ISSN:0003-2654    

DOI:10.1039/AN98712FX001

出版商:RSC    

年代:1987

数据来源: RSC

摘要:

ANALAO 112(1) 1-112 (1987)The AnalystJanuary 1987The Analytical Journal of The Royal Society of ChemistryCONTENTS117Directly Coupled Chromatography - Atomic Spectroscopy. Part 2. Directly Coupled Liquid Chromatography - AtomicSpectroscopy. A Review-Les Ebdon, Steve Hill, Robert W. WardDi- and Tributyltin Species in Marine and Estuarine Waters. Inter-laboratory Comparison of Two Ultratrace AnalyticalMethods Employing Hydride Generation and Atomic Absorption or Flame Photometric Detection-Aldis 0.Valkirs, Peter F. Seligman, Gregory J. Olson, Frederick E. Brinckman, Cheryl L. Matthias, Jon M. BellamaDetermination of Scandium in Coal Fly Ash and Geological Materials by Graphite Furnace Atomic AbsorptionSpectrometry and Inductively Coupled Plasma Atomic Emission Spectrometry-M.Bettinelli, U. Baroni,N. Pastorelli27 Multi-element Pre-concentration by Solvent Extraction Compatible with an Aqua Regia Digestion for GeochemicalExploration Samples-Ivan Rubeska, Benilda Ebarvia, Edita Macalalad, Dahlia Davis, Nenita Roque31 Analytical Reference Materials. Part VI. Development and Certification of a Sediment Reference Material for SelectedPolynuclear Aromatic Hydrocarbons-Hing-Biu Lee, Geeta Dookhran, Alfred S. Y. Chau37 Analytical Reference Materials. Part VII. Development and Certification of a Sediment Reference Material for TotalPolychlorinated Biphenyls-Hing-Biu Lee, Alfred S. Y. Chau41 Determination of Uranium(V1) in Process Liquors by Ion Chromatography-John J. Byerley, Jeno M. Scharer, George F.Atki nson45 Gas Chromatographic Determination of Acrolein in Rain Water Using Bromination of 0-MethyloximeHarumitsuNishikawa, Tomokuni Hayakawa, Tadao Sakai49 Simple Gas Chromatographic Determination of the Distribution of Normal Alkanes i n the Kerosene Fraction ofPetroleum-Suresh C.Vishnoi, Shiv D. Bhagat, Vidya B. Kapoor, Sneh K. Chopra, Rajamani Krishna53 Determination of Polychlorinated Biphenyls in Waste Oil by Gas - Liquid Chromatography-Richard E. Lawn, Shaun A.ToffeI57 Simultaneous Measurement of Phenobarbital, Carbamazepine, Phenytoin and 5-(4-Hydroxyphenyl)-5-phenylhydan-toin in Serum by High-performance Liquid Chromatography-Siraj A. Mira, Yousry M. El-Sayed, Samira I. Islam61 Lithium Ion-selective Electrodes Containing TOPO: Determination of Serum Lithium by Flow Injection Analysis-Robert Y.Xie, Gary D. Christian65 Chemically lmmobilised Bi-enzyme Electrodes in the Redox Mediated Mode for the Flow Injection Analysis of Glucoseand HypoxanthineG. J. Moody, G. S. Sanghera, J. D. R. Thomas71 Determination of Free Hydrofluoric and Nitric Acids in Pickling Bath Liquors Using a Fluoride-selective Electrode andAlkalimetric Titration-Kaj Lindroos75 Spectrophotometric Determination of Carbaryl and Propoxur Using Aminophenols and PhenylenediamineC. S. P.Sastry, D. Vijaya, D. S. Mangala79 Determination of iorazepam by Fluorimetric and Photochemical - Fluorimetric Methods-Jesus Rodriquez Procopio,Pedro Hernandez Hernandez, Lucas Hernandez Hernandez83 Selective Photometric Titration of Calcium or Magnesium with EDTA Using Various Thiols as Masking Agents-ColinG. Halliday, Michael A. Leonard87 Turbidimetric Determination of Chlorhexidine Using Flow Injection Analysis-J. Martinez Calatayud, P. Campins Falc6,A. Sanchez Sampedrc91 A New Type of Argon lonisation Detector-S. A. Beres, C. D. Halfmann, E. D. Katz, R . P. W. ScottSHORT PAPERS97 Diazotised 4-Nitroaniline as a Chromogenic Reagent for the Determination of Trace Amounts of Pyrrole in Aqueous101 Spectrophotometric Determination of Dobutamine Hydrochloride Using 3-Methylbenzothiazolin-2-one H y d r a z o n e105 SOFTWARE REVIEW106 BOOK REVIEWS109 INSTRUCTIONS TO AUTHORS23Solution-Ahmad K. Ahmad, Younis I. Hassan, Wadala A. BashirMichael E. El-KommosTypeset and printed by Heffers Printers Ltd, Cambridge, Englan

ISSN:0003-2654    

DOI:10.1039/AN98712BX003

出版商:RSC    

年代:1987

数据来源: RSC

摘要:

ANALYST, JANUARY 1987, VOL. 112 17 Di- and Tributvltin SDecies in Marine and Estuarine Waters. Inter4aborato;y Comparison of Two Ultratrace Analytical Methods Employing Hydride Generation and Atomic Absorption or Flame Photometric Detection Aldis 0. Valkirs and Peter F. Seligman Marine Environment Branch, Naval Ocean Systems Center, San Diego, CA 92152, USA Gregory J. Olson and Frederick E. Brinckman Surface Chemistry and Bioprocesses Group, National Bureau of Standards, Gaithersburg, MD 20899, USA and Cheryl L. Matthias and Jon M. Bellama Department of Chemistry and Biochemistry, University of Maryland, College Park, MD 20742, USA Di- and tributyltin compounds present in marine and estuarine waters at sub-parts per billion (<pg 1-1) levels were determined using two different chemical speciation procedures.Generally, good analytical agree- ment was obtained from split samples independently analysed by a simultaneous hydride generation - dichloromethane extraction procedure followed by gas chromatographic separation and flame photometric detection (GC - FPD, performed at the National Bureau of Standards) and by a hydride generation procedure followed by purge and trap collection with boiling-point separation and atomic absorption detection (HG-AA, performed at the Naval Ocean Systems Center). Sea water samples containing tributyltin at sub-p.p.b. levels can be stored frozen (-20 "C) in polycarbonate containers for up to 2-3 months without any serious loss of analyte. Keywords: Antifouling marine biocides; atomic absorption and flame photometric detection; h ydride generation; inter-laboratory comparisons; organotin speciation Organotin compounds are increasingly used industrially as catalysts, plastics stabilisers and biocides.1 Tributyltin species are among the most effective organotin biocides and their worldwide use as active agents in antifouling coatings, especially for ships,2 is rapidly expanding. The increasing use of tributyltin-based antifouling coatings has raised concern about their environmental fate and the effects on non-target organisms of the toxic tributyltin species released from the coatings. Bioassays with algae,3 oyster ,4 crabs and mussel larvae,6 mysid shrimp,' copepods8 and fish9 have shown sub-lethal and lethal effects of tributyltin at parts per billion (pg 1-1) and lower levels in water.Consequently, some nations have issued regulations (France) or proposed regulations (UK) to control the use of tributyltin-based antifouling coatings on small craft in an attempt to protect marine life (generally shellfish) near harbours and marinas. The US Environmental Protection Agency recently initiated a special review on the environmental use of organotin com- pounds. 10 In order to evaluate effectively the environmental risks associated with tributyltin biocide usage and to develop monitoring strategies, analytical methods capable of the detection and speciation of key diagnostic butyltin species in environmental waters at concentrations as low as parts per trillion (ng 1-1) levels must be developed and employed. The determination of the toxic tributyltin species and its less toXic3J1J2 primary degradation product, dibutyltin, is of paramount importance.The relative abundances of these species offer clues to their persistence and toxic impacts at ambient levels in environmental systems. However, the chemical determinations of these species in natural waters are difficult and cannot be achieved by conventional total tin analysis. Only recently have methods been described for the requisite sub-pg 1-1 speciation of butyltins in aquatic environ- ments. These include chromatographiclsl6 or boiling- point17-20 separation of butyltin species, followed by atomic absorption,16-18320 mass spectrometric,l4 or flame photo- metricl3715>19 detection. Sample derivatisation by Grignard reaction14-16 or hydride generation13J7-20 is usually employed.Unfortunately, there have been no standards on which to base assessments of the relative accuracy of these analytical methods. Consequently, an inaugural inter-laboratory com- parison of methods for the detection of tributyltin in de- ionised water was organised and recently performed by the National Bureau of Standards (NBS), employing a stable, chromatographically purified aqueous tributyltin research material.21 In general, the 35 participant laboratories per- formed well in determining total tin (as tributyltin) at the parts per million (mg 1-1) level, but speciation of the tin at sub-pg 1-1 levels is an entirely different problem. A new intercom- parison is planned using a more dilute mixed butyltin species research material.It is also necessary to analyse butyltin species in natural water samples to determine if it is possible to compare data from different laboratories at low, ambient levels (ng 1-1) in marine and estuarine waters. Similarly, sample preservation procedures such as freezing must also be evaluated in order to determine their applicability for the storage and exchange of environmental samples. Such procedures were effective in the storage of sea water samples containing organoarsenic species, with respect to distribution and stability, for a period of three months at -20 OC.22 For these reasons, and for reporting preliminary results on concentrations of butyltin species in a variety of coastal marine and estuarine waters, the Naval Ocean Systems Center (NOSC) and the NBS undertook a joint intercomparison to measure tributyltin and its primary degradation product, dibutyltin, in shared marine and estuarine water samples collected on the east and west coasts of the USA and in England.18 ANALYST, JANUARY 1987, VOL.112 Experimental Glassware or plasticware (polycarbonate) was used directly if new, or else it was leached with 10% nitric acid solutions for 8-12 h and rinsed repeatedly with de-ionised water. Samples were collected from ships or piers in new polycarbonate (8-20 1) containers. Sampling sites included the west (San Diego Bay, CA, nine sites) and east (Chesapeake Bay, MD, three sites) coasts of the USA and the east coast of England (two sites). The sampling containers were submerged to 0.5-1.0 m depth, the caps were removed by hand, the bottles filled and the caps replaced underwater prior to removal from the water.In this manner, we avoided collecting surface microlayer films, which can contain butyltin compounds in relatively high concentrations. 1 3 ~ 5 On return to the laboratory, samples were either imme- diately frozen (-20 "C) and shipped, or aliquots were transferred into 1-1 polycarbonate containers, frozen and shipped. Prior to analysis, the samples were thawed at room temperature or by gentle warming (40 "C), but were not allowed to exceed room temperature and were analysed while still cool. Frozen Sample Storage Evaluation An effective sample storage procedure was considered essen- tial to allow successful comparisons of the determination of butyltin concentrations measured in split samples, and for the development of monitoring procedures.Frozen sea water samples containing tributyltin at sub-pg 1-1 concentrations were evaluated in order to document their long-term stability. A large volume of sea water, circulated over panels painted with antifouling paint containing tributyltin, was collected in a 20-1 polycarbonate bottle. This unfiltered source solution was poured into individual 500-ml polycarbonate bottles and frozen at -20 "C. Individual bottles were then removed at various times and analysed by hydride generation - atomic absorption (HG - AA) for tributyltin. Very little, if any, dibutyltin or monobutyltin was initially present. Subsequent analysis did not, therefore, address these species, as concen- trations were very near or below the detection limits (5 ng 1-1).Hydride Generation - Atomic Absorption Method At NOSC, the HG - AA methodl7J8 for producing volatile tin species with detection by modified hydrogen flame atomic absorption spectrometry was an adaptation of methods described by Braman and Tompkinslg and Hodge et a1.20 Inorganic tin and organotin compounds are derivatised to stannane and the respective alkyltin hydrides by sodium borohydride before detection. Briefly, a sample was placed into a 500-ml modified gas washing bottle and acidified to pH 5.0-5.5 with 2 M acetic acid. Hydride derivatives were formed by the addition of 4% mlV sodium tetrahydroborate(II1) prepared in a 1% mlV sodium hydroxide solution in distilled water.A ratio of 1 ml of sodium tetrahydroborate(II1) solution to 100 ml of sample was used to generate hydride species, which were purged from solution with helium carrier gas and trapped in a glass U-tube (2 mm i d . ) packed with 0.01-0.015 g of 3% OV-1 on Chromosorb W HP (80-100 mesh) and immersed in liquid nitrogen. Inlet and outlet lines to the U-trap and detector were made of FEP Teflon. The solution was purged for a total of 5 min after the addition of sodium tetrahydroborate(II1) to ensure the maximum removal of tin hydrides from solution. The trap was then removed from the liquid nitrogen bath and the tin species were separated and detected sequentially according to their boiling-points as they distilled from the trap. The tin species were carried into a quartz tube, atomised in a hydrogen - air flame and detected by an atomic absorption spectrometer at 286.3 nm.Gas flow-rates with respect to hydrogen, air and helium were 220, 140 and 40 ml min-1, respectively. The volatilisation of tributyltin hydride (Bu3SnH) required heating the trap in an oil-bath (180 "C). Standardisation was accomplished by the addition of an appropriate alkyltin standard (in ethanol) to the unknown, or by calibration graphs with the values calculated by peak integration. Concentrations of butyltin species are reported as ng 1-1 of tributyltin chloride or dibutyltin dichloride. The detection limit for dibutyltin and tributyltin hydrides was 5 ng 1-1. We demonstrated a relative standard deviation of 6.3% (standard deviation of a single measurement divided by the mean) for the analysis of five replicate determinations prepared in sea water at a concentration of 10 ng 1-1 of tributyltin chloride.The redistribution of butyltin groups resulting from the sodium tetrahydroborate(II1) derivatisa- tion process has not been detected in butyltin standards analysed at NOSC. The recent analysis of a tributyltin reference material prepared in distilled water at the NBS, containing tin only as the tributyl species,21 has shown no evidence of re-distribution. Gas Chromatographic Separation - Flame Photometric Detec- tion Method At the NBS, 100-200-ml aliquots of sea water samples were analysed by a simultaneous extraction (dichloromethane) and hydride generation method followed by gas chromatography - flame photometric detection (GC - FPD).13 Briefly, for a typical analysis of saline water with a butyltin concentration in the sub-pg 1-1 range, the following procedure was used. To 100 ml of sample in a 125-ml glass separating funnel, equipped with a Teflon stopcock and Teflon-lined screw-top, were added 2.8 ml of dichloromethane and 2.0 ml of 4% mlV aqueous NaBH4.In addition, a 10-p1 spike of a 0.5 p.p.m. aqueous solution of dipropyltin dichloride was added as an internal standard. The funnel was capped and shaken by hand for 1 min, vented and then shaken (240 strokes min-1) on a wrist-action shaker for 10 min. After a 5-min settling period, the lower organic layer was removed. An additional 1.4 ml of dichloromethane was added and the extraction procedure repeated. The organic layers were combined (approximately 2 ml) in polypropylene centrifuge tubes and evaporated to 100-200 p1 or less under a gentle stream of air.Appropriate reagent blanks were carried through the entire procedure. A gas chromatograph equipped with a flame photometric detector was used for the determinations. Chromatographic separations were carried out on a 1.83 m (6 ft) x 2 mm i.d. glass column packed with 1.5% OV-101 (liquid methylsili- cone) on Chromosorb G HP (100-120 mesh). A hydrogen-rich flame was employed, supported by H2 flowing at the measured rate of 110 ml min-1, air at 70 ml min-1 and N2 (zero grade) carrier gas at 20 ml min-1. The FPD was equipped with a 600-nm cut-on interference filter (band pass 600-2000 nm) to monitor the SnH molecular emission.The output signal from the FPD was recorded simultaneously on a strip-chart recorder and an integrator - plotter. For all runs reported here, the column temperature was programmed at 23 "C for 2 min and then heated to 170 "C at 32 "C min-1. The detector temperature was maintained at 200 "C and the injection port at 150 "C. The determination of butyltin species was performed either by the method of standard additions or from calibration graphs using sea water or estuarine samples containing little or no measurable levels of butyltin species. Butyltin and di- propyltin chlorides were used as received for the preparation of standard solutions for determination. Concentrations of butyltin species are reported as nanograms of tributyltin chloride or dibutyltin dichloride per litre.The GC - FPD method gives detection limits for di- and tributyltin species ofANALYST, JANUARY 1987, VOL. 112 L SnH4 HG-AA T 19 I Pr2SnH2 GC - FPD approximately 5 ng 1-1, with a relative standard deviation of 1&15% at 10 ng l-V3 I Results Typical chromatograms for the HG - AA and GC - FPD methods are shown in Fig. 1. Di- and tributyltin species were detected in the 14 coastal marine and estuarine surface water samples in concentrations ranging from trace to several hundred ng 1-1. Tables 1 and 2 show the means ( X ) for 2-5 replicate analyses (depending on the sites), standard devia- tions of a single measurement (S,) and the percentage relative standard deviations (RSD) of single measurements. With a few exceptions, the concentrations of the dibutyltin species were lower than those of the tributyltin species.In general, agreement was good between the two analytical methods employed. In all but five samples the NBS and NOSC I SnH4 Blank L Blank Y I ' I I 1 Time/m i n 0 4 8 Fig. 1. Chromatograms from analysis of environmental samples by HG - AA and GC - FPD methods. For HG - AA, Bu2SnH2 = 16 ng 1-1 and Bu3SnH = 9 ng 1-l. For GC - FPD, Bu,SnH2 = 28 ng 1-l and Bu3SnH = 90 ng 1-1 values were within 20% of their mean concentration for tributyltin. With dibutyltin, only four of the values varied by more than 20% of the means. The RSDs for replicate analyses were also similar for the two methods, being in the range 11-15% for the two tin species. There was no consistent trend for one method yielding a higher concentration of butyltins than the other.In 5 of 14 instances (36%), the GC - FPD values were greater than the HG - AA values for tributyltin. Also, 5 of 14 analyses showed greater values of dibutyltin with GC - FPD. Only two samples showed serious discrepancies (more than a two-fold difference) for tributyltin concentrations: the San Diego Bay-5 (four-fold) and the Bradwell Ck. marina (which barely exceeded two-fold) sites. Three sites showed such discrepancies with dibutyltin: the San Diego sites 8 and 9 (two- and three-fold, respectively) and the MAFF Laboratory (five-fold) site. Plots of HG - AA values compared with GC - FPD values for dibutyltin and for tributyltin showed good agreements with the theoretical line of slope 1.0 (Figs. 2 and 3). The analytical results (using the HG - AA method) for three frozen sample storage sets measured at different times and containing tributyltin at sub-pg 1-1 concentrations are presented in Table 3.Individual analytical values represent single determinations performed by one of three analysts on one of two instruments. Data from set I1 show that a slow loss of tributyltin occurs during the frozen storage of sea water samples containing tributyltin at sub-pg 1-1 levels. The regression line for these data has a slope of -0.00149 pg 1-1 per week, with a standard deviation of 0.00027. This corre- sponds to a tributyltin loss of about 1.4% per week of the fitted time zero value (0.112 pg 1-1). After 41 weeks of storage, the tributyltin level was less than 50% of the initial concentration (Table 3).In this study, replicate analyses of sea water samples containing di- and tributyltin gave average RSDs of 11-15%. Given this level of uncertainty, the frozen storage of samples for up to 2-3 months should not cause serious analytical problems. A significant negative slope was not obtained from tributyltin values regressed with storage time using data from sets I and 111. This is probably due to the relatively short storage times of sets I and 111, which were only one-third and one-sixth of the time of set I1 storage. Discussion This study evaluated the determination of dibutyl- and tributyltin species at ng 1-1 concentrations in natural waters Table 1. Concentrations of tributyltin detected in marine and estuarine waters GC - FPD (NBS) HG - AA (NOSC) X S, R.S.D., '/o X S, R.S.D.,% SanDiegoBay-6 .. . . 5* -t - 9 2.0 22 BayBridge,MD . . . . 6 1.2 20 NMS Baltimore Harbor, MD . . 11 0.8 7 12 --$ SanDiegoBay-7 . . . . 22 1.1 5 21 1.5 7 SanDiegoBay-4 . . . . 23 2.5 11 18 2.1 12 MAFF Lab., England . . 37 11 30 68 0 0 SanDiegoBay-8 . . . . 79 8.2 10 95 10 11 SanDiegoBay-5 . . . . 90 33 37 19 7.1 37 Annapolis,MD . . . . 97 5.1 5 103 5.2 5 SanDiegoBay-1 . . . . 184 30 16 270 26 10 SanDiegoBay-2 . . . . 338 60 18 369 38 10 - - - SanDiegoBay-9 . . . . 50 8.1 16 93 24 26 SanDiegoBay-3 . . . . 162 2.7 2 209 13 6 Bradwell Ck., England . . 732 --t - 332 43 13 Average . . . . . . . . 15 13 * Tributyltin chloride, ng 1-1. t Single analysis. 3 Not measurable.20 ANALYST, JANUARY 1987, VOL. 112 Table 2. Concentrations of dibutyltin detected in marine and estuarine waters GC - FPD (NBS) HG - AA (NOSC) MAFF Lab., England Bay Bridge, MD .. Baltimore Harbor, MD San Diego Bay-6 . . San Diego Bay-7 . . San Diego Bay4 . . Annapolis,MD . . San Diego Bay-4 . . SanDiegoBay-9 . . Bradwell Ck., England San Diego Bay-5 . . San Diego Bay-3 . . SanDiegoBay-2 . . SanDiegoBay-1 . . Average . . . . . . * Dibutyltin dichloride, ng 1-1. t Single analysis. $ Not measurable. X . . 2* . . 5 . . 11 . . 11 . . 13 . . 17 . . 19 . . 20 . . 20 . . 23 . . 28 . . 60 . . 263 . . 270 . . S , R.S.D., "/o --t 0 0 1.6 15 1.5 14 0.3 2 2.8 16 1 .o 5 5.1 26 2.7 14 7.9 28 7.7 13 18 7 48 18 13 - - - X 11 NM$ 11 16 12 52 29 23 49 19 25 75 21 1 270 S, R.S.D., Oh 1.1 10 - - - - 1.2 8 0.6 5 9.5 18 4.1 14 4.6 20 4.2 9 4.0 21 2.1 8 0 0 14 7 21 8 11 Table 3.Frozen storage of sea water samples containing tributyltin at sub-pg I-' concentrations Set I: Set 11: Set 111: April-August 1984 April-September 1985 September-November 1985 Weeks 0 0.3 0.9 1.1 2.1 3.3 4.1 4.9 6.0 6.9 7.0 7.9 8.9 9.9 11.7 13.7 13.9 14.0 Concentration* 0.13 0.13 0.14 0.11 0.13 0.09 0.13 0.10 0.11 0.10 0.10 0.10 0.11 0.09 0.13 0.12 0.15 0.10 * Tributyltin chloride, pg I-'. Weeks 0 0.1 1 .o 2.1 2.1 2.3 2.3 2.4 2.9 3.0 3.0 3.1 3.3 4.1 4.1 4.9 4.9 5.0 5.0 5.1 Concentration * 0.112 0.070 0.093 0.086 0.130 0.120 0.128 0.102 0.102 0.103 0.133 0.110 0.121 0.080 0.090 0.111 0.090 0.090 0.087 0.083 Weeks 5.3 5.4 5.6 5.6 5.7 5.7 7.6 7.6 11.4 16.6 16.7 16.9 17.9 18.4 18.4 18.7 18.9 20.9 41.4 41.6 Concentration* 0.105 0.085 0.090 0.080 0.071 0.082 0.097 0.097 0.073 0.057 0.060 0.075 0.089 0.073 0.103 0.054 0.071 0.099 0.043 0.046 Weeks 0 0.3 4.1 4.3 4.3 4.4 4.7 5.3 5.4 5.6 5.7 6.4 Concentration* 0.063 0.071 0.058 0.061 0.068 0.068 0.088 0.069 0.110 0.076 0.092 0.086 300 r I m - 8 2oo (3 I 0 100 200 300 400 GC - FPD, ng I-' Fig.2. Graph of HG - AA values versus GC - FPD values for tributyltin species from environmental sea water samples. The line is the theoretical slope, 1 .O 200 7 I 0 c - 8 100 GC - FPD, ng I-' Fig. 3. Graph of HG - AA values versus GC - FPD values for dibutyltin species from environmental sea water samples. The line is the theoretical slope, 1.0ANALYST, JANUARY 1987, VOL. 112 using two different measurement methods. The HG - AA method employs the direct hydridisation of the bulk sample, purging by an inert gas, trapping on a chromatographic substrate at liquid nitrogen temperatures and butyltin detec- tion by highly element-specific atomic absorption spec- trometry following boiling-point elution from the trap.The GC - FPD method employs hydride generation coupled with simultaneous extraction into an organic solvent (CH2C12). It is clear from this study that both methods give similar analytical results for di- and tributyltin species in marine and estuarine waters. It is possible that the few instances of disagreement (showing a 2-fold or greater difference in results) resulted from factors related to the chemical or physical composition of samples that affected one method more than the other. However, no obvious anomalies (oil pollution, excessive particulates, algal blooms, etc.) were noted in these samples.Frozen Sample Storage The data presented in Table 3 indicate that frozen storage of sea water samples containing tributyltin is a reasonably effective method of sample preservation for a period of approximately 2-3 months. Data from set 11, which was obtained over a 10-month period, showed that a slow loss of tributyltin occurred during frozen storage. The mechanism of this loss may not have involved de-alkylation as increases in degradation products (mono- and dibutyltin) were not noted. About half of the initial tributyltin concentration was present after a 10-month storage period. Data from sets I and 111, representing 6.4-14.0 weeks of frozen storage, did not show statistically significant negative slopes, and also confirm that frozen storage for several weeks does not result in significant losses of tributyltin.Frozen storage in polycarbonate bottles has proved to be effective in preserving sea water samples. Previous work has shown that storage of sea water samples containing tributyltin in polyethylene plastic containers resulted in substantial (62%) adsorptive losses from initial values after a one-week period at 4 OC.23 Samples stored in polycarbonate plastic, Pyrex glass and Teflon containers exhibited adsorptive losses of 3, 4 and 7%, respectively.23 Efforts are continuing to evaluate the stability of frozen sea water samples containing tributyltin for a period of approximately 1 year. Chemical speciation of butyltins in marine and estuarine waters at ambient (ng 1-1) levels is difficult and as a result there is little data available for toxicologists and environmen- tal agencies to evaluate. As legislation restricting the use of organotin antifouling paints is growing worldwide, there is a rapidly increasing need for monitoring data for the toxic tributyltin species and its less toxic degradation product, dibutyltin. This study is the first effort to make inter-labora- tory comparisons of methods for butyltin speciation in natural waters.Efforts of this type are important to establish intercomparability of methods and data as environmental monitoring programmes are undertaken. A related, parallel effort at NBS21 has chromatographically generated a stable aqueous tributyltin research material, used in conducting the first worldwide methods intercomparison for organotin measurements. The next phase of this effort is underway and will generate a mixed butyltin species research material for use in a inter-laboratory comparison involving speciation of the material diluted to sub-pg 1-1 levels.21 Development Center under program element 63724N. Ana- lytical support provided by Giti Vafa and Peter Stang is gratefully acknowledged. The NBS research was supported in part by the Office of Naval Research and the David Taylor Naval Ship Research and Devlopment Center. We gratefully acknowledge shiptime on the RN Ridgeley Warfield made available to us by the University of Maryland. We also thank Dr. Robert Paule of NBS for his helpful advice on data presentation and statistical considerations.Portions of this work will be included in the dissertation of C. L. M. to be submitted as a requirement for the PhD degree from the University of Maryland. The technical assistance provided by the advanced technol- ogy division of Computer Sciences Corporation, San Diego, CA, USA is gratefully acknowledged. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. References Blunden, S. J., Hobbs, L. A., and Smith, P. J., in Bowen, H . J. M., Editor, “Environmental Chemistry,” Royal Society of Chemistry, London, 1984, p. 51. Gitlitz, M. H., J. Coatings Technol., 1981, 53, 46. Walsh, G. E., McLaughlan, L. L., Lores, E. M., Louie, M. K., and Deans, C. H . , Chemosphere, 1985, 14, 383.Waldock, M. J., andThain, J. E., Mar. Pollut. Bull., 1983, 14, 411. Laughlin, R . B., French, W., Johannesen, R . B., Guard, H. E., and Brinckman, F. E., Chemosphere, 1984, 13, 575. Beaumont, A. R., and Budd, M. D . , Mar. Pollut. Bull., 1984, 15, 402. Salazar, M. H., and Salazar, S. M., in “Proceedings of the 11 th US - Japan Experts Meeting on Management of Sediments Containing Toxic Substances,” November 4-6, 1985, Seattle, WA. U’ren, S. C., Mar. Pollut. Bull., 1983, 14, 303. Seinen, W., Helder, T., Vernig, H., Penninks, A . , and Leeuwangh, P., Sci. Total Environ., 1981, 19, 155. Fed Regist., January 8th, 1986, 51, No. 5. Wong, P. T. S., Chau, Y . K., Kramer, O., andBengert, G. A . , Can. J. Fish Aquat. Sci., 1982, 39, 483. Laughlin, R. B., Jr., Johannesen, R . B., French, W., Guard, H. E., and Brinckman, F. E., Environ. Toxicol. Chem., 1985, 4, 343. Matthias, C. L., Olson, G. J., Brinckman, F. E., andBellama, J. M., Environ. Sci. Technol., 1986, 20, 609. Mueller, M. D., Fresenius Z . Anal. Chem., 1984, 317, 32. Maguire, R . J., Chau, Y . K., Bengert, G. A . , Hale, E . J., Wong, P. T. S., and Kramer, O., Environ. Sci. Technol., 1982, 16, 698. Maguire, R. J., Tkacz, R. J., J. Chromatogr., 1983, 268, 99. Valkirs, A. O., Seligman, P. F., Vafa, G., Stang, P. M., Homer, V., and Lieberman, S. H., Technical Report No. 1037, Naval Ocean Systems Center, San Diego, CA, 1985. Valkirs, A. O., Seligman, P. F., Stang, P. M., Homer, V., Lieberman, S. H., Vafa, G., and Dooley, C. A , , Mar. Pollut. Bull., 1986, 17, 319. Braman, R. S., and Tompkins, M. A., Anal. Chem., 1979,51, 12. Hodge, V. F., Seidel, S. L., and Goldberg, E . D., Anal. Chem., 1979, 51, 1256. Blair, W. R., Olson, G. J., and Brinckman, F. E., NBSIR 86-3321, National Bureau of Standards, Gaithersburg, MD, 1986. Yamamoto, M., Fujishige, K., Tsubota, H., and Yamamoto, Y . , Anal. Sci., 1985, 1, 47. Dooley, C. A., and Homer, V., Technical Report No. 918, Naval Ocean Center, San Diego, CA, 1983. Research performed at NOSC was sponsored by the Office of the Chief of Naval Research, Energy Research and Develop- ment Program and the David Taylor Naval Ship Research and Paper A61135 Received May 6th, 1986 Accepted July 7th, 1986

ISSN:0003-2654    

DOI:10.1039/AN9871200017

出版商:RSC    

年代:1987

数据来源: RSC

摘要:

ANALYST, JANUARY 1987, VOL. 112 23 Determination of Scandium in Coal Fly Ash and Geological Materials by Graphite Furnace Atomic Absorption Spectrometry and Inductively Coupled Plasma Atomic Emission Spectrometry M. Bettinelli, U. Baroni and N. Pastorelli Central Laboratory, ENEL-DCO, Via Nino Bixio 39, 29700 Piacenza, Italy Two procedures for the determination of scandium in coal fly ash and other geological materials have been developed. The first procedure consists in dissolution of the samples with a mixture of nitric, perchloric and hydrofluoric acids in a sealed PTFE vessel followed by determination by graphite furnace atomic absorption spectrometry. In the second procedure, the sample is decomposed by fusion with lithium tetraborate, the melt is dissolved in 5% nitric acid and scandium is determined by inductively coupled plasma atomic emission spectrometry using an argon plasma.The limits of determination are 0.80 and 0.63 pg g-1, respectively, for the two methods. The scandium concentration in eight coal ash samples obtained by GFAAS and ICP-AES compare well with those determined previously by NAA. Keywords: Scafidium determination; coal fly ash; geological materials; graphite furnace atomic absorption spectrometry; inductively coupled plasma atomic emission spectrometry Studies of the particle size dependence of element concentra- tions in fly ash can be classified into two categories.1 The first category consists of those studies that relate the elemental concentrations to the particle size of size-classified material (sufficient mass of size-classified material must be collected in order to allow gravimetric determination prior to elemental analysis).The second category consists of the many studies that have employed inertial cascade impactor systems for aerodynamic size classification. Because only small masses of material may be collected on the stages of the cascade impactor, the specific elemental masses of the particles deposited on each stage are often ratioed to the mass of an element that does not demonstrate a marked dependence of concentration on particle size. One of the elements that has been frequently used for this purpose is scandium. This element was selected because it is present at very low levels in the natural environment, it is essentially non-volatile at furnace temperatures , and also because most fly ash analyses are carried out by neutron activation analysis (NAA) , which gives a detection limit of about 0.002 pg of Sc.2 Gladney et aZ.,3 for example, reported the concentration of Sc in several geological environmental NBS standard materials, but referred only to NAA and flame atomic absorption spectrometric (FAAS) techniques. More recently, ICP-AES and DCP-AES techniques have been successfully employed to determine trace amounts of elements in geological and related materials, including Sc and rare earth elements.4-13 In contrast, very few papers have been published on the determination of Sc by ETA-AAS.1620 Sen Guptal6 determined Sc in silicate rocks after coprecipi- tation with calcium oxalate and hydrated iron(II1) oxide. Working in the peak-height mode (Tatom.= 2500 "C) using pyrolytically coated graphite tubes, the sensitivity for Sc was 37 pg per 0.0044 A, superior to that found (50 pg per 0.0044 A) in a tantalum-lined furnace.17 Sen Gupta later showed18 that, at 2000 "C, greater sensitivity can be achieved by using a tantalum foil-lined graphite furnace (1.2 pg per 0.0044 A) instead of a pyrolytically coated furnace (13 pg per 0.0044 A). Wu and Ma19 reported the direct determination of Sc in soils by graphite furnace AAS (GFAAS) with a pyrolytically coated graphite tube lined with both tungsten and tantalum foil. Sample aliquots (10 ~ 1 ) were atomised at 2730 "C for 10 s in a 0.5 1 min-l argon gas flow. The absolute limit of determination was 4.6 pg of Sc.Atnashev et aZ.20 reported the determination of Sc by ETA-AAS using a tungsten coil atomiser, carrying out the pulsed atomisation in a laminar flow of Ar - N2 (10 + 1). However, no characteristic amounts data for Sc were reported by Atnashev et aZ.,20 but it was reported that the results compare well with corresponding values obtained with a L'vov graphite cell. In this work we compared two different procedures for sample dissolution: acid attack with nitric - perchloric - hydrofluoric acids and lithium tetraborate fusion. Scandium was determined in the solution derived from the mixed acid digestion procedure by GFAAS, whereas in the fusion solution we used the ICP-AES technique. Experimental Reagents Standard solutions of Sc were prepared from a 1000 mg 1-1 stock standard solution for atomic absorption spectrometry (Aldrich-Chemie , Steinheim, FRG) by dilution with de-ion- ised water or lithium tetraborate.Perchloric acid (70% mlv), nitric acid (65% m/V) and hydrofluoric acid (40% mlv) were all Suprapur reagents (E. Merck, Darmstadt, FRG). Lithium tetraborate solution (5 g 1-1) was prepared from Baker Analyzed reagent flux grade material (J. T. Baker Chemicals, Denveter, The Netherlands). Water was purified in a Milli-Q system (Millipore, S.P.Q. Milan, Italy). Apparatus A Perkin-Elmer 5000 atomic absorption spectrometer equipped with a D2 arc background corrector, an AS-40 autosampler and an HGA-500 graphite furnace was used for the Sc determinations. A 7500 data station was used for the display and storage of the fast atomisation signals.Peak absorbance and integrated absorbance signals were calculated using the HGA Graphics I1 software and PR 210 printer - plotters were used for printing out the analytical information and the high-resolution peak profiles. The optimum instrumental parameters for Sc determinations (Table 1) were established after extensive investigations. Pyrolytically coated graphite tubes were used in all determinations. A Perkin-Elmer ICP/6000 inductively coupled plasma atomic emission spectrometer was used for all Sc determina-24 ANALYST, JANUARY 1987, VOL. 112 Table 1. AAS instrumental operating conditions Model 5000 spectrophotometer: Wavelength . . . . . . . . . . . . 391.2 nm Calibration mode . . . . . . . . . . Peak area Integration time .. . . . . . . . . 6 s Background corrector . . . . . . . . Deuterium arc Spectral slit width . . . . . . . . . . 0.2 nm HGA-500 graphite furnace: Step, n Temperature/"C Ramp time/s Hold time/s 1 80 1 4 2 120 10 10 3 500 10 10 4 1700 20 10 5 2700 O* 6 6 2800 1 3 Purge gas . . . . . . . . . . . . Argon, interrupted Sample volume . . . . . . . . . . 20 pl Alternative volume . . . . . . . . . . 20 pl * Maximum power heating mode; read activated at -2 s. Table 2. ICP-AES instrumental operating conditions . . . . . . . . Incident RF power 1250 W Reflected RF power . . . . . . . . <5 w Plasma gas flow-rate . . . . . . . . 15 1 min-1 Auxiliary gas flow-rate . . . . . . . . 0.3 1 min-1 Nebuliser gas pressure . . . . . . . . 26 lb in-2 Viewing height . . . . . . . . .. 14 mm above load coil tions in lithium tetraborate solutions; the operating conditions used are given in Table 2. For sample decomposition, a Perkin-Elmer Autoclave 3 was used. Neutron activation analyses were performed with the TRIGA Mark I1 reactor at the University of Pavia. Standards and samples were irradiated for 50 h at a flux of 1.2 X 1012 neutrons cm-2 s-1; the radionuclide used was 46Sc with a half-life of 83.80 d and a characteristic peak at 889 keV. Acid Dissolution Procedure About 0.25 g of sample was taken with 4.0 ml of 65% nitric acid, 2.0 ml of 70% perchloric acid and 4.0 ml of 40% hydrofluoric acid in the PTFE beaker of the Autoclave-3 and heated in a drying oven for 3-5 h at 150 "C. After cooling the contents of the PTFE beaker were slowly evaporated to dryness on a low-temperature (200 "C) alumi- nium block.After the addition of 1 ml of nitric acid and 10 ml of water, the solution was transferred into a polypropylene bottle and diluted to 100 ml with de-ionised water. If any undissolved residue was visible at this stage, the solution was filtered through a 0.5 pm Fluoropore membrane filter and the residue solubilised with 2.0 ml of 70% perchloric acid and 5.0 ml of 65% nitric acid. The mixture was heated on the plate nearly to dryness and, after washing with 5.0 ml of water, was then heated until the evolution of dense white fumes. The PTFE beaker was then cooled, 1.0 ml of nitric acid plus 10 ml of water were added to dissolve the salts and the solution was combined with that in the polypropylene bottle.This final solution was then diluted to 100 ml. Fusion Procedure Approximately 0.25 g of sample and 0.5 g of lithium tetra- borate were placed in a 50-ml platinum crucible and thor- oughly mixed. The sample - flux mixture was fused in a muffle furnace at 1000 k 50 "C for 45 min. When the fusion was complete, the cooled crucible was placed in a 100-ml beaker, a small PTFE-coated stirrer was inserted and 25 ml of 5% nitric acid were added. The solution was heated for 15-20 min at 50-60 "C on a magnetic stirrer and then transferred into a 100-ml poly- propylene flask. The same procedure was repeated with a second aliquot of nitric acid and, when dissolution was complete, the contents of the flask were adjusted to volume with de-ionised water. Blanks containing only lithium tetraborate were also fused in the manner described above.These solutions are stable over a period of several months and may be used for the determination of ten major elements, plus some trace elements. Results and Discussion Optimisation of HGA Operating Parameters In order to select the optimum ashing temperature, a study of the effect of this parameter on the absorption signal of Sc was carried out. Aliquots of 20 pl of a 50 pg 1-1 Sc solution were injected into the pyrolytically coated graphite tubes and drying and atomisation cycles of 120 "C for 10 s and 2700 "C for 6 s, respectively, were employed. The integrated absorption signal produced during the atomisation step was detected at 391.2 nm and recorded by the HGA Graphics I1 software. On varying the ashing temperature between 1000 and 2000 "C, we observed that the signal remained constant at temperatures up to 1800 "C (Fig.1). Hence, an ashing temperature of 1700 "C was selected for all further determina- tions. Interferences Sen Guptalg reported that there are no inter-element interfer- ences in the determination of lanthanides in GFAAS and that interferences from associated common elements can be eliminated by heating the sample at about 1800-2000 "C before atomisation. The final solution in work of the Sen Guptals is, however, relatively free from other elements because it is first submitted to a double calcium oxalate and a single hydrous iron(II1) oxide co-precipitation step. Our final solution, in contrast (250 mg of sample diluted to 100 ml), contains several milligrams of matrix elements such as Si, Al, Fe, Mg, Ca, Na and K. For this reason we verified the interference effects of a synthetic solution containing 100 mg 1-1 of Fe, Mg, Na, K and P, 500 mg 1-1 of A1 and 1000 mg 1-l of Ca and Si on the absorbance signal of 20 pg 1-1 of Sc.Fig. 2 shows that under the experimental conditions reported in Table 1 the interfer- ences due to the principal elements in our matrices are negligible. GFAAS Determinations According to the concept of "characteristic mass" (rn,) developed by Slavin and Carnrick,21 we calculated this quantity by injection of the analyte in aqueous solution (1% HN03) and by the method of standard additions. The instrumental parameters and the experimental con- ditions are given in Table 1 and the results for rn, are reported in Table 3.The characteristic mass (m,) averaged on ten scandium determinations in aqueous solution YO HN03) was 31.2 rt 4.0 pg per 0.0044 A s (CV = 12.7%) but with two different values, 34.6 f 1.9 pg per 0.0044 A s (n = 5, CV = 5.6%) and 27.9 k 1.2 pg per 0.0044 A s (n = 5, CV = 4.4%) for two different lots of pyrolytically coated graphite tubes. The characteristic amounts found by the standard additions method for NBS 1633, 1633a, 1645, 1646, 278 and 688 standards compare very well with those in water. The precisions for the reported characteristic mass data appear to be better than 10-20%.ANALYST, JANUARY 1987, VOL. 112 25 0.050 a 0.040 Table 3. Characteristic mass data (pg per 0.0044 A s) for Sc in aqueous solution and in different NBS materials A ( a ) - - Sample Water (1% HN03) NBS 1633 NBS 1633a NBS 1645 NBS 1646 NBS 278 NBS 688 Number of samples 5 3 3 3 3 3 3 (n) Characteristic mass (mo) Lot A* 34.6 2 1.9 32.4 k 1.7 33.3 f 1.2 33.9 2 1.5 32.5 f 2.5 34.3 f 2.0 32.8 k 1.9 * Lots A and B are two different lots of pyrolytically coated graphite tubes.Lot B* 27.9 k 1.2 25.3 k 0.8 25.3 f 1.8 24.7 k 1.7 24.8 f 2.6 29.2 f 1.1 25.1 f 2.6 Average 31.2 k 4.0 30.0 f 5.9 29.2 f 4.5 29.3 f 5.2 28.7 k 4.8 31.8 f 3.1 28.9 k 4.6 c v , Yo 12.7 19.7 15.4 17.9 16.7 9.8 16.0 0.21 v) a 0.09 1 1 I I I I I I 1 I I I 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 Ashing temperaturePC Fig. 1. Typical ashing curve for 1 ng o f Sc in 1% HN03 0 1 2 3 4 Time/s Fig. 2. of Sc and (b) 20 pg 1-1 of Sc plus Fe, Mg, Na, K and P (100 mg l-l), A1 (500 mg 1-l) and Ca and Si (loo0 mg 1-1) High-resolution peak profiles for ( a ) 20 pg 1 Fig.3. Graphic spectral scans of Sc I1 at 361.384 nm. Conditions: 100 pg 1-1 of Sc in A, lithium tetraborate; B, NBS 1633a solution; C, NBS 688 solution; and D, NBS 1646 solution. Background positions: B1, -0.126 nm and B2, +0.140 nm After about 50-70 firings of a new pyrolytically coated graphite tube at the recommended temperatures, the inte- grated absorbances decrease because of the destruction or loss of some of the pyrolytic coating and the tube should be discharged. The detection limit, DL (calculated as the concentration corresponding to three times the standard deviation of the blank), was 2.0 pg 1-1, whereas the lowest quantitatively determinable concentration, LQD (defined as the concentration corresponding to ten times the standard Table 4.Results for NBS reference materials (pg g-l) This work NBS Certified standard GFAAS ICP-AES values Coal fly ash: 1633 25.1 23.5 28.7k2.0 - 1633a 40.62 5.1 41.3 k 4.8 (40)* River sediment: Estuarine sediment: Obsidian rock: Basalt rock: 1645 (1.8) (1.6) (2) 1646 9.4 f 3.1 10.9 k 1.9 (10.8) 278 6.1 f 1.3 3.5 k 0.1 (5.1) 688 39.1 k 2.4 40.4 2 0.7 (38.1)* * Values reported but not certified by NBS. Other values3 26.6 2 1.7 38 2 3 2.6 10.4 4.9 f 0.5 36.2 Table 5. Comparative concentrations of Sc (pg g-l) in different coal ash samples This work Sample GFAAS ICP-AES NAA BMAl . . . . . . 2 6 f 1.7 29f0.6 27f1.4 BMA2 . . . . . . 20k1.1 20f0.4 GMAl .. . . . . 3352.0 36f0.6 3 4 f 1 . 7 GMA2 . . . . . . 3.550.08 2.8k0.03 3 k 0 . 2 BCLl . . . . . . 8k0.09 9k0.07 10f0.5 11 f 0.5 BCL2 . . . . . . 951.0 1OkO.4 BMIl . . . . . . 25k1.3 28k0.6 25k1.3 BM12 . . . . . . 2420.7 25k0.3 2921.5 19 5 1.0 deviation of the blank and a dilution factor of 400), was 2.7 pg g-1 of Sc. ICP Determinations The choice of the Sc I1 line at 361.384 nm was based on a systematic consideration of the detection limit, the expected elemental concentration in the sample solution and spectral interferences. Fig. 3 shows a scan from 0.5 nm below to 0.5 nm above the scandium 361.384 nm line, overlaid with scans of three NBS reference materials. These, and other scans not reported in the figure, indicate that, for these and similar types of samples, the Sc emission line is sufficiently free from background interferences to be utilised in a reasonably straightforward manner.Background correction must be carried out accurately because the background under the analyte line is extremely structured and could cause spectral interference problems, particularly if there were any broadening as the concentrations increased.26 ANALYST, JANUARY 1987, VOL. 112 The detection limit was 1.6 pg 1-1, whereas the lowest quantitatively determinable concentration was 2.1 pg g-1 of sc. Comparison of GFAAS and ICP Results Results for scandium determinations obtained in this work compare well with NBS reference values (reported but not certified) and literature values. The accuracy data are given only for those samples for which the Sc concentration exceeded the LQD value, as quantitative determinations can usually be made with satisfactory accuracy and precision only above this concentration level.The scandium concentration determined by ICP-AES for the NBS 1645 sample was found to be less than the estimated DL, whereas the value for NBS 278 appears to be biased low. The poor recoveries also obtained for this element using the acid dissolution - HGA technique indicate that incomplete solubilisation (dissolution or fusion) for the NBS 1645 sample was responsible for the low scandium results. It is evident from these data that accurate determinations can be carried out by both the methods; the fusion - ICP-AES method gives a better precision then the acid digestion - GFAAS procedure.Similar conclusions can be drawn from the data in Table 5 relative to other coal ash samples analysed for their Sc content either by mixed acid digestion - GFAAS and fusion - ICP-AES procedures or by neutron activation analysis. Conclusions Both the fusion - ICP-AES and acid digestion - GFAAS procedures are capable of the successful routine determina- tion of Sc in coal fly ashes, provided that the samples contain relatively high concentrations of this element (i.e., 1C20 times the LQD). The major source of error in the experimen- tal reproducibility lies (mostly for acid solubilisation) in the preparation of the sample powder solution. In general, the data obtained for fusion - ICP-AES or acid digestion - GFAAS determinations are as good as those obtained for NAA determinations.4. 5. 6. 7 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. References Fischer, G. L., and Natusch, D. F. S., in Karr, C., Editor, “Analytical Methods for Coal and Coal Products,” Volume 111, Academic Press, New York, 1979, Chapter 54, p. 489. Weaver, J. N., in Karr, C., Editor, “Analytical Methods for Coal and Coal Products,” Volume I, Academic Press, New York, 1979, Chapter 12, p. 377. Gladney, E. S., Burns, C. E., Perrin, D. R., Roelandts, I., and Gills, T. E., “1982 Compilation of Elemental Concentration Data for NBS Biological, Geological and Environmental Standard Reference Materials,” NBS Special Publication 260-88, National Bureau of Standards, Washington, DC, 1984. Brenner, I. B., Watson, A. E., Steele, T. E., Jones, E. A., and Goncalves, M., Spectrochim. Acta, Part B, 1981,36,785. Crock, J. G., and Lichte, F. E., Anal. Chem., 1982,54, 1329. Uchida, U., Uchida, T., and Iida, C., Anal. Chim. Acta, 1980, 116, 433. Floyd, M. A., Fassel, V. A., and D’Silva, A. P., Anal. Chem., 1980, 52, 2168. McLaren, J. W., Berman, S. S., Boyko, V. J., and Russel, D. S., Anal. Chem., 1981,53, 1802. Fries, T., Lamothe, P. J., and Pesek, J. J., Anal. Chim. Acta, 1984, 159,329. Nadkarni, R. A., Anal. Chem., 1980,52,929. Cantillo, A. Y., Sinex, S. A., and Helz, G. R., Anal. Chem., 1984,56, 33. Bolton, A., Hwang, J., and Vander Voet, A., Spectrochim. Acta, Part B, 1983,38, 165. Barnes, R. M., and Mahanti, H. S., Spectrochim. Acta, Part B, 1983,38, 193. Sen Gupta, J. G., Talanta, 1981, 28, 31. Sen Gupta, J. G., Geostand. Newsl., 1982, 6, 241. Sen Gupta, J. G., Anal. Chim. Acta, 1982, 138, 295. L’vov, B. V., and Pelieva, L. A., Can. J. Spectrosc., 1978,23, 1. Sen Gupta, J. G., Talanta, 1985, 32, 1. Wu, Z,, and Ma, Y., Fenxi Huaxue, 1983, 11, 423; Anal. Abstr., 1984, 46, 6G10. Atnashev, V. B., Muzgin, V. N., and Atnashev, Yu. B., Zh. Anal. Khim., 1982, 37, 1590; Anal. Abstr., 1983,44,6B16. Slavin, W,, and Carnrick, G. R., Spectrochim. Acta, Part B, 1984,39,271. Paper A615 Received January 6th, 1986 Accepted August 26th, 1986

ISSN:0003-2654    

DOI:10.1039/AN9871200023

出版商:RSC    

年代:1987

数据来源: RSC

摘要:

ANALYST, JANUARY 1987, VOL. 112 27 Multi-element Pre-concentration by Solvent Extraction Compatible with an Aqua Regia Digestion for Geochemical Exploration Samples Ivan Rubeska UNDP, P.O. Box 7285 ADC, Mia Road, Pasay City, Metro Manila, Philippines and Benilda Ebarvia, Edita Macalalad, Dahlia Ravis and Nenita Roque Bureau of Mines and Geosciences, North Avenue, Dilirnan, Quezon City, Philippines The interference of nitric acid in the organic solvent extraction of metal iodide complexes as ion associates with the trioctylmethyl ammonium ion was investigated. The results show that the interference is caused by nitric oxide extracted into the organic solvent in the form of nitrosyl compounds and that this can be eliminated by the addition of sulphamic acid or urea. A multi-element extraction procedure was applied to geochemical exploration samples digested with aqua regia.Ag, Cd, Se, Te and TI were determined by AAS in the extracts down to 100 p.p.b. levels. Keywords : Multi-element extraction; nitric acid interference; trace element determination; geochemical samples; trioctylmeth yl ammonium chloride Geochemical exploration often involves the determination of a number of trace elements, many of which are well below the detection limits of AAS or ICP-AES. For such elements a multi-element organic solvent extraction would be an attrac- tive solution1 if it could be made sufficiently simple and could cover all the elements of interest. Most of the trace elements of interest in geochemical exploration are easily solubilised by digesting the samples with boiling aqua regia and this is probably the most widely used decomposition procedure when looking for primary mineralisation.It may be performed on a large scale with very simple equipment, i.e., test tubes, an aluminium block and a hot-plate, and even fairly resistant minerals such as pyrite and cinnabar are attacked. For samples with a high iron content, one of the most promising multi-element extraction systems is the extraction of iodide or bromide complexes into 4-methylpentan-2-one (IBMK), possibly as ion associates with long chain aliphatic polyamines or polyphosphines. Two reagents of this kind have been extensively applied, namely, trioctylmethylammonium (TOMA) chloride2-5 and trioctylphosphine oxide (TOPO) .cg Unfortunately, the extraction of iodide or bromide com- plexes is incompatible with an aqua regia digestion of the samples unless the solutions are dried and then re-dissolved in, for example, HC1.This would make the procedure prohibi- tively time and labour consuming for geochemical explora- tion. In the course of establishing analytical procedures for the geochemical exploration of epithermal gold deposits, 16 trace elements had to be determined, possibly from a single sample decomposition. Seven of these (Cu, Pb, Zn, Co, Ni, Mn and Mo) had sufficient ratios of abundances to detection limits to provide meaningful analyses by direct flame AAS of the sample digest and four after hydride (As, Bi and Sb) or cold vapour (Hg) generation. For the five remaining (Ag, Cd, Se, Te and Tl), the possibility of applying the extraction of iodide complexes as ion associates with TOMA or TOPO were explored.Reactions causing interference were identified as being due to nitric oxide and an extraction procedure appropriate for the geochemical analyses of soil, stream sediment and rock-chip samples, where some precision may be sacrificed for greater speed, was established. Experimental Sample solutions were prepared by boiling 2 g samples in a test-tube with 8 ml of aqua regia for 30 min. After cooling, the volume was adjusted to 20 ml with distilled water. The final concentration was not critical and varied between 3 and 4 M HCl. After settling, 10 ml of this solution were drawn into a polypropylene syringe fitted with a 10 cm X 1.2 mm i.d. plastic capillary tube extension to the inlet, followed by 2 ml of ascorbic acid (30%), 2 ml of a 2 M KI solution and 3 ml of IBMK containing 5% V/V TOMA or 5% m/V TOPO.After shaking for about 1 min the syringes were left to stand, and after the separation of the phases the organic layer was transferred into glass vials, which were then capped. This solution was then used for measuring several different trace elements by AAS. A Varian-Techtron Model 1475 AA spectrometer was used, either with a microsampling flame attachement or a GTA 95 graphite furnace. For flame AAS, 50 pl of the extract were delivered to the nebuliser with a micropipette. A 3 s integration time was used, giving the operator enough time to keep the whole absorbance signal within the integrating period. Ag and Cd were determined down to 0.1 p.p.m.in the sample. Se, Te and T1, which have insufficient detection limits by flame AAS, were determined in a graphite furnace using pyrolytic graphite coated tubes for Se and Te and a pyrolytic graphite platform for T1. Te and T1 were determined down to 0.1 p.p.m. and Se down to 0.2 p.p.m. from integrated absorbance readings. A vessel containing IBMK was kept in the carousel of the automatic sampler to slow down the evaporation of the organic solvent from the sample extracts loaded in open vials. The carousel was always kept covered. Instrumental conditions are given in Table 1.10 Results and Discussion Preliminary experiments using extraction from 3 M HCl with the addition of up to 1 ml of nitric acid to the 10 ml of extracted Table 1.Conditions of measurement in graphite furnace AAS Element . . . . Se Te Tl Analytical line/nm . . 196.0 214.3 276.8 Atomisation surface Wall Wall Platform Samplevolume/pl . . 10 10 5 Modifiervolume/pl . . 3 3 2 Temperature programme/("C; sramP + shold) Drying . . . . 110;5 + 10 110;5 + 10 500;5 Pyrolysis . . . . 500;8 500; 8 600; 10 + 5 Atomisation . . 2800; 1 + 1.2 2600; 1 + 1.2 2600; 1 + 2 1200; 1 + 3 Modifier for Se and Te . . Modifier form . . . . 1% H2S04 + 0.5% Mg 1200; 1 + 3 10% HN03 + 0.5% Cu + 0.5% Mg28 ANALYST, JANUARY 1987, VOL. 112 solution showed that extraction with TOMA is less affected by the nitric acid than extraction with TOPO and the former was therefore used for further experiments. It soon became evident that the presence of nitric acid in the limited amount used did not affect the extraction itself, but rather the stability of the extracts.If determined immediately after separation, the correct amount of extracted metals was usually found. Metals in the extract showed different stabili- ties. Tellurium was very rapidly lost and was therefore used for checking the stability of the extracts. In sample extracts with a high metal content a yellow precipitate was formed within a few hours and the original brown colour became orange. Iodine, Pb, Cu, Zn and Ga were identified in the precipitate by XRF. Extracts with a lower metal content turned bright yellow, and even if no precipitate was visible the Te content fell markedly. If the extract was stirred just before introduction into the graphite furnace with an automatic sampler, highly scattered readings for Te were observed.As the determinations were repeated the higher values eventually disappeared. This behaviour indicated that Te was still present but was in the form of particles that were gradually settling out after stirring. If after separation the organic phase was stored undisturbed in a glass vial, the colour change from brown to yellow always began from the top surface. Discolouration was particularly frequent with samples high in iron. The extracts also emitted vapours, highly irritant to the eye. From these observations and some basic chemistry the following sequence of events may be seen: 1, Nitric acid oxidises iodide in the aqueous phase with the production of nitrous acid and nitric oxide.The latter forms nitrosyl compounds that are extracted into IBMK. 2, In the IBMK, nitric oxide catalyses the oxidation of iodide to iodine by atmospheric oxygen. 3, As iodide is lost, soluble metal iodide complexes turn into less soluble forms and precipitate. 4, Iodine reacts with IBMK in the enolic form, giving an iodinated ketone. Halogenated ketones are well known tear gas compounds. If this sequence of reactions is a correct description of the processes involved, the instability of the extracts is not due to the nitric acid itself but rather to the products of its reduction, i.e., nitrous acid and nitric oxide. These may be eliminated by the addition of urea, sulphamic acid or ammonia in general. The presence of the ammonium ion in TOMA explains why extracts with this reagent were more stable than with TOPO. Both urea and sulphamic acid were investigated.The addition of 1 ml of a 1% solution of these reagents was found to be sufficient to secure the stability of the extracts. A correlation between the instability of extracts and a high iron content in the samples pointed to an iron nitrosyl halide as the NO-containing species extracted. The amount of NO extracted, and presumably the stability of extracts, could therefore be assessed from the amount of iron in the organic phase. Iron extracted under different conditions, i . e . , 1 .o v) Q 0.5 A with and Fig. 1. Signal of iron extracted into IBMK - TOMA in presence of increasing amounts of NO2- from 2 M HCI, 0.266 M KI and: A, ascorbic acid (4%) + sulphamic acid (0.066%); B, sulphamic acid only; C, ascorbic acid only; and D, neither without the addition of ascorbic and/or sulphamic acids and with increasing concentrations of nitric oxide, was determined by microsampling flame AAS using the Fe line at 372 nm.Nitric oxide was added as sodium nitrite, which is reduced to nitric oxide by the excess iodide. The solutions contained 25 mg of iron added as FeC13. Other conditions of the extraction were identical with those used for the sample solutions, i.e., the final concentrations were 2 M HC1, 0.266 M KI and 4% ascorbic acid. The results show that in the presence of nitric oxide iron is extracted both in the absence and the presence of ascorbic acid. The addition of ascorbic acid reduces the amount extracted but less so than sulphamic acid.With an increase in the amount of nitrite, the amount of Fe extracted in the absence of ascorbic acid levels off at about 0.003 M NO*-, but increases further in its presence (Fig. 1). The addition of ascorbic acid alone thus cannot prevent the extraction of iron and nitric oxide into the organic phase. Only a combination of ascorbic and sulphamic acids reduces iron extraction to a negligible level. Whether nitric oxide is also extracted in some other form is difficult to assess. However, if so, the amount must be limited as no instability of extracts unconnected with a high iron content in the samples was ever observed. It is also difficult to assign any valence to the iron in the nitrosyl compounds extracted. Iron nitrosyl halides with metal valencies 1-111 are known.It is plausible that the nitrosyl compound extracted is a negatively-charged ferrous complex that forms an ion asso- ciate with TOMA or protonated IBMK. This is suggested by the observation that when carrying out the extraction under identical conditions but without the addition of sulphamic acid and using KBr instead of KI, sample solutions high in iron turned green on addition of ascorbic acid. This green compound is partially extracted into IBMK, as seen by the intense green of the organic layer. On standing, the green gradually fades and a red - brown precipitate forms at the contact with the aqueous layer. It is known that negatively charged nitrosyl halides of Fell are green, neutral halides are red and cationic halides are brown.11 A less intensive green colour is observed when adding ascorbic acid to the iodide solutions.It is, however, obscured by the brown of the extracts. The elements Ag, Cd, Se, Te and T1 were determined in the same extracts in order to check the expected correlation between the amount of iron extracted and the stability of the extract. Ag and Cd showed no systematic variation with the amount of nitrite added; the scatter of values was within 5%. I I 1 1 I I / I I I I 1 I 0 2 4 6 8 1 0 2 4 6 8 1 0 “ O Z ~ ] / ~ M Fig. 2. Signals of Se, Te and T1 measured from same solutions as in Fig. 1. (a) Solution A; (b) solution B; ( c ) solution C; and (d) solution D of Fig. 1. A, Se; B; Te; and C, Tl. Amounts extracted: Te and T1,l FLg; Se, 2.5 pgANALYST, JANUARY 1987, VOL.112 29 Signals of Se, Te and T1 determined in the graphite furnace are plotted in Fig. 2. Thallium, in a similar manner to Ag and Cd, does not show any systematic variation. Se and Te, at the highest N02- concentration and both with the addition of ascorbic acid only and without any addition, show significantly lower values indicating the instability of the extracts. The same two solutions also have the highest Fe content. The signal of Te in the extract without the addition of ascorbic acid, determined 30 min after extraction, was 0.166 A s and fell to 0.02 A s within 1 h. If the two readings for the unstable extracts are excluded there is no systematic variation. The scatter of readings has a relative standard deviation of about lo%, which is not statistically significant.Although 1 ml of 1% sulphamic acid removes all instability up to the concentration level of NO investigated, a 2% solution was used in the determinations. Out of more than 2000 samples analysed by this procedure, only about 10 showed any instability by decolourising. These were repeated using a smaller aliquot of the sample solution and dilution with 3 M HC1. The relative standard deviation of the determinations is about 10% for Ag and Cd, 15% for Se and Te and 20% for TI, which is adequate for geochemical exploration purposes. Conclusions The results reported in this paper indicate that it is possible to apply the extraction of iodide and/or bromide complexes of metals as ion associates with TOMA to the analysis of geochemical exploration samples, even in the presence of up to 10% V/V FINO3 if certain provisions are made.In particular, the time during which the nitric acid and iodide are in intimate contact must be short in order to minimise the amount of nitrous acid formed. The sample solution extracted should, therefore, always be clear as the presence of clay colloids slows down the separation of phases. If the sample solutions are clear the separation is very fast because of the high density of the aqueous layer. As the instability of the extracts is due to nitric oxide extracted in the form of nitrosyl compounds, it may be eliminated by reducing both nitrous acid and nitric oxide (which are in equilibrium) to nitrogen by the addition of sulphamic acid or urea to the extracted solutions.Ascorbic acid must be added to counteract the oxidation of iodide to iodine, which would make the separation of phases difficult. It is also needed to reduce some of the extracted elements to their lower valencies that form the iodide complexes. Extracts prepared under these conditions with the addition of sulphamic acid and kept in a refrigerator have been found to be stable for many days. As many elements of interest in geochemical exploration form relatively stable iodide and/or bromide complexes (e.g., Ag, Au, Bi, Cd, Cu, Ga, Hg, In, Pb, Sb, Se, Sn, Te, T1 and Zn) this extraction system has potentially very broad applications.3.10 The enrichment factor attainable is evidently limited by the presence of common base metals (Cu, Pb and Zn), which use up the reagents and may saturate the organic phase.This is not generally a limiting factor for geochemical exploration samples. The main advantage of the method is that the extraction may be applied to samples digested with aqua regia, which is a much simpler, and for geochemical exploration more widely used, oxidative decomposition procedure than HCl + KC103,2 HC1 + H2025 or fusion with potassium pyrosul- phate.12 All these have been used in order to make the extraction of iodide complexes with TOMA applicable to geochemical exploration samples. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. References Thompson, M., Analyst, 1985, 110, 443. Viets, J . G.. Anal. Chem., 1978, 50, 1097. Clark, J. R., and Viets. J. G., Anal. Chem., 1981, 53, 61. Motooka, T. M., Mosier, E. L., Sutley, E. T.. andViets, J . R., Appl. Spectrosc., 1979, 33, 456. O’Leary, R. M., and Viets, J. G., At. Spectrosc., 1986, 7, 4. Burke, K. E., Analyst, 1972, 97, 19. Burke, K. E., Tulanta, 1974, 21, 497. Bedrossian, M., Anal. Chem., 1978, 50, 1898. Janousek, I . , Coll. Czech. Chem. Commun., 1978, 43, 2136. Rubeska, I., Ebarvia, B., Macalalad, E., Ravis, D., and Roque, N., “Multielernent Extraction System for the Deter- mination of Trace Elements in Geochemical Exploration Samples,” PHI 85/001, Internal Technical Report GCR/86/3, 1986. Moeller, T., “Inorganic Chemistry,” Wiley, New York, 1952, Viets, J. G., O’Leary, R. M., and Clark, J. R., Analyst, 1984, 109, 1589. Paper A61229 Received July I7th, I986 Accepted August lath, 1986 p. 603.

ISSN:0003-2654    

DOI:10.1039/AN9871200027

出版商:RSC    

年代:1987

数据来源: RSC

摘要:

ANALYST, JANUARY 1987, VOL. 112 31 Analytical Reference Materials Part Vl.* Development and Certification of a Sediment Reference Material for Selected Polynuclear Aromatic Hydrocarbonst Hing-Biu Lee, Geeta Dookhran and Alfred S. Y. Chau Quality Assurance and Methods Section, Analytical Methods Division, National Water Research Institute, Environment Canada, Burlington, Ontario L7R 4A6, Canada A naturally contaminated environmental sediment reference material (EC-1) was developed and analysed for selected polynuclear aromatic hydrocarbons (PAH). Freeze-dried and homogenised subsamples of EC-1 were Soxhlet extracted and the extracts were cleaned-up on activated silica gel and alumina columns. The levels of PAH in this material were determined by using the following three independent analytical methods: (1) GC - FID, (2) GC - MS and (3) reversed-phase HPLC with fluorescence and UV detectors.Up to a total of 72 replicate determinations were performed and the results obtained by each method were in good agreement with each other. lnterlaboratory PAH results for EC-1 obtained in a round-robin study also confirmed the in-house data. The results for ten PAH of which the between-method difference was less than k 10% were pooled t o generate the certified values. Keywords: Pol ynuclear aromatic hydrocarbon determination; certified reference material; quality assurance; sediment samples Polynuclear aromatic hydrocarbons (PAH) are ubiquitous environmental pollutants as they are naturally occurring and are also formed by the pyrolysis of carbonaceous materials at high temperatures.The routine determination and monitoring of PAH in environmental samples are essential because of their persistence and carcinogenic and mutagenic properties. Although PAH levels in open lake or marine surface waters are low, they are readily adsorbed and accumulated by sediments and particulate matter and pg g-1-ng 8-1 levels of PAH have been reported in many sediment samples.2-5 Methods for the determination of PAH in environmental samples have been reviewed previously.6 The most popular detection techniques involve either gas chromatography (GC) or high-performance liquid chromatography (HPLC) . The determination of PAH by HPLC with a fluorescence detector has been used by many workers7-9 as many PAH are highly sensitive to this detector.The judicious selection of excitation and emission wavelengths also makes the detector highly specific, which minimises the interferences from unresolved co-eluting PAH resulting from highly complicated mix- tures.7-10 In many instances the fluorescence detector is complemented by a variable-wavelength UV detector for the detection of those PAH with a low sensitivity to the former. 11-13 GC determinations of PAH are usually carried out with high-resolution capillary columns.2-5J4J5 The quality of these chromatograms is much higher than those obtained by packed columns and therefore capillary columns are considered essential for complicated environmental samples. Although PAH are generally detected by a flame ionisation detector (FID), electron-capture detectorsl6J7 can also be used under special circumstances for these compounds.The determina- tion of PAH by a sensitive and selective mass spectrometer interfaced to a high resolution capillary column is currently the most powerful approach.lgJ9 In this instance, mass spectra of a sample provide positive identification of known PAH or structural information for unknown PAH. By using the selected ion monitoring (SIM) technique, sub-nanogram amounts of PAH can easily be detected. * For Part V of this series, see reference 26. t This material is currently not for sale and not available for general distribution. Two interlaboratory studies on the determination of PAH in sediment samples have been reported.20.21 The results of these studies and that organised by our section22 indicated that widespread results were obtained from different laboratories.For many parameters, interlaboratory relative standard devia- tions (after the rejection of outliers) ranged from 30 to 60%, suggesting that there was a need to improve the accuracy of data obtained from PAH determination in environmental samples. Although several certified reference materials (CRM) for PAH determination have been prepared by the US National Bureau of Standards,23 only one of them, a sample of urban particulate matter, was in an environmental matrix. In order to fulfil quality assurance and method research require- ments, the development of sediment CRMs for PAH is therefore necessary. In this paper, we describe the development and certification of the first in a series of sediment reference materials for method evaluation and other in-house and interlaboratory quality assurance applications.Certified or reference values for the following 14 PAH are reported: phenanthrene (phen), anthracene (anth), fluoranthene (F), pyrene (Py), benzo- [alanthracene (B[a]A) , chrysene (chry) , benzo[b]fluoran- thene (B[b]F), benzoCk1fluoranthene (B[k]F), benzo[e]py- rene (B[e]P), benzo[a]pyrene (B[a]P), perylene (pery), inde- no[l23cd]pyrene (I[cd]P), dibenz[ah]anthracene (D[ah]A) and benzo[ghi]perylene (B[ghi]P). All of the above, except B[e]P and perylene, are listed as US EPA priority pollu- tan ts .24 Experimental Preparation of Sediment Reference Material Approximately 450 kg of wet sediment were collected from a landfill site in Hamilton Bay, Ontario, Canada. The sample, coded EC-1, was freeze-dried, crushed, sieved, blended and subsampled. Further details of this procedure have been published elsewhere.25.26 Extraction of PAH in Sediments A 10.00 g aliquot of EC-1 was extracted in a Soxhlet apparatus with 350 ml of 59 + 41 V/V acetone - hexane for 8 h at a rate of 8 cycles h-1.For the comparison of recoveries, Soxhlet32 ANALYST, JANUARY 1987, VOL. 112 extraction with other solvent systems and extraction using a soni~ator25~26 were also evaluated. The results of this compari- son are given under Results and Discussion. Clean-up of Sediment Extracts The combined organic extract was shaken with 400 ml of 2% KHC03 in a 1-1 separating funnel for 1 min with frequent venting.After the layers had separated, the aqueous layer was drained into a 500-ml separating funnel and then discarded. The organic layers in the two separating funnels were combined and passed through 100 g of anhydrous Na2S04 in a column. The funnels were washed with 2 x 10 ml of hexane and the washings were again applied to the column. After the last trace of solvent was removed from the Na2S04 column by vacuum suction, the dry extract was evaporated down to ca. 5 ml using a rotary evaporator with a 35 "C water-bath. A 400 x 10 mm i.d. glass clean-up column with either a coarse porosity fritted disc or a glass-wool plug was filled with a freshly prepared slurry of 10.0 g of silica gel (Davison grade 923, 100-200 mesh activated at 130 "C for 18 h before use) in hexane with 1 cm of anhydrous Na2S04 at the top.The concentrated sediment extract in hexane was quantitatively transferred on to the column and drained just into the Na2S04 layer. The sample flask was rinsed with 2 ml of hexane and the rinsing again applied to the column. This process was repeated twice. The column was then eluted with 50 ml of hexane and the eluate was discarded. This fraction contains chloroben- zenes, PCBs and several chlorinated insecticides if they are also present in the sample. The column was further eluted with 60 ml of 40 + 60 V/V dichloromethane - hexane. This fraction was collected in a 250-ml round-bottomed flask and was evaporated down to ca. 5 ml using a rotary evaporator as described above. After the addition of 20 rnl of hexane and 3 ml of isooctane, the evaporation was repeated until the volume was ca.3 ml. A second clean-up column was prepared by filling a 230 X 5 mm i.d. disposable Pasteur pipette having a glass-wool plug at the bottom with 5 cm of activated neutral alumina (Woelm, Brockmann activity 1 , 100-200 mesh) and 5 mm of anhydrous sodium sulphate at the top. This column was eluted with 5 ml of hexane and the eluate was discarded. The concentrated extract, after silica gel column clean-up, was applied to the column, rinsing through with 3 x 1 ml of hexane. The column was further eluted with hexane until a total of 10.0 ml of hexane was collected. This fraction contained aliphatic hydrocarbons and other non-polar co-extractives that had not been removed by the silica gel column.The PAH were removed from the alumina column by elution with toluene until a volume of 10.0 ml was collected. Gas Chromatography with Flame Ionisation Detection (GC - FID) A Hewlett-Packard 5880A gas chromatography equipped with a Grob-type split - splitless capillary injection port, a flame ionisation detector, a 7671A autosampler and Level IV terminals was used. A 30 m X 0.25 mm i.d. DB-5 fused-silica capillary column of 0.1 pm film thickness (J and W Scientific) operating under the following conditions was used for PAH analysis. Temperatures: injection port and detector, 275 "C; oven initial temperature 70 "C, hold 1.5 min at 70 "C, programming rate 1, 30 "C min-1 (from 70 to 160 "C), rate 2, 2 "C min-1 (from 160 " to 260 "C), hold 10 min at 260 "C. Flow-rates: hydrogen, 30 ml min-1; air, 240 ml min-I; detector make-up gas (helium), 25 ml min-1.Carrier gas, helium; column head pressure, 15 lb in -2. Splitless valve on for 90 s. A 2-p1 aliquot of the final extract was injected in the splitless mode without dilution. Gas Chromatography with Mass Spectrometry (GC - MS) The system consisted of a Hewlett-Packard 5880A gas chromatograph, as described above, a 5970B mass-selective detector (MSD) a 9816s computer and a 9133XV disc drive. The DB-5 capillary column was directly interfaced with the electron-impact ion source (70 eV) for maximum sensitivity. The GC operating conditions were identical to those used in the FID determination, except that the detector gases were not needed and the column head pressure was 4 lb in-2.A 2-p1 aliquot of a 20-fold diluted sample extract was analysed. The data were acquired by the following two modes: (a), linear scanning from rnlz 50 to 300 in order to obtain abundance data of major fragments for compound identification purposes; and (b), selected ion monitoring (SIM) for quantitative analysis. In the latter instance, the following molecular ions characteris- tic of PAH were monitored: 1, rnlz 178 for phen and anth; 2, rnlz 202 for F and Py; 3, rnlz 228 for B[a]A and chry; 4, mlz 252 for B[b]F, B[k]F, B[e]P, B[a]P and pery; 5, mlz 276 for I[cd]P and B[ghi]P; and 6, rnlz 278 for D[ah]A. The dwell time for each ion was 100 ms. Three labelled internal standards, i.e., phen-dlo, chry-d12 and B[ghi]P-Wl2 were used for the calibration of response factors.High-performance Liquid Chromatography (HPLC) A system including a Waters Model 510 pump, a Rheodyne 7125 loop injector and 20- j.d loop, a 4.6 mm i.d. x 25 cm long Zorbax ODS column (DuPont, 5-6 ym particle size) and a Schoffel Model FS 970 fluorescence detector were used. Mobile phase (isocratic), 85 + 15 acetonitrile - water; flow-rate, 1.0 ml min-1. The detector wavelength was set at 280 nm (excitation) and 389 nm (emission). A 20-pl aliquot of a 100-fold diluted sample extract was injected. Standards and Standard Solutions Most PAH standards are available from Aldrich Chemical or Eastman Kodak. Certified reference materials of PAH are also available from the Commission of European Communi- ties, BCR, Brussels. Individual stock solutions were prepared by dissolving 50.0 mg of each PAH in a 100-ml low actinic calibrated flask with toluene; some PAH required gentle heating or sonification to dissolve.Appropriate amounts of the 14 PAH stock solutions in proportions similar to those found in EC-1 were pipetted into a low actinic calibrated flask and diluted to volume with toluene. This solution, which contained PAH at yg ml-1 levels, was used in the GC analyses as an external standard. A standard for the HPLC analysis of EC-1 was prepared by diluting the above solution with the HPLC mobile phase. Results and Discussion The sediment reference material EC-1 was originally prepared and analysed for PCBs.25 It is a fine (200-325 mesh), silty clay sediment naturally contaminated with many toxic organics and metals. Although homogeneity tests were not performed for PAH before subsampling, subsequent determinations on various lots of sediment subsamples did not reveal any inhomogeneity for PAH (Table 1).Therefore, EC-1 is considered sufficiently homogeneous for use as a PAH reference material. Soxhlet extraction has been used by many workers for the determination of organics, including PAH, in sediments. Although most of the extraction in this work was carried out with 41 + 59 hexane - acetone, comparative extraction was also investigated using solvent systems such as 1 + 1 benzene - methanol, cyclohexane and hexane. The results for theANALYST, JANUARY 1987, VOL. 112 33 determination of PAH in EC-1 under various extraction conditions are given in Table 2. No difference in PAH recoveries from EC-1 subsamples was observed between the four solvent systems, although the non-polar solvents (cyclohexane and hexane) gave extracts much lighter in colour.Further experiments by Soxhlet extraction were all carried out with the 41 + 59 hexane - acetone because this solvent was easy to evaporate and was also used in our multi-residue extraction procedure. A longer extraction time (24 vs. 8 h) did not produce higher results (Table 2). Ultrasonic extractions25 of EC-1 with 1 + 1 hexane - acetone were also carried out, and again recoveries of PAH by this technique were identical to those obtained by the Soxhlet method (Table 2). As different solvent systems and different extraction methods gave the same results, it was therefore concluded that the extraction recoveries of PAH from EC-1 were quantitative in these experiments.Table 1. Homogeneity test of various lots of EC-1 subsamples for PAH. Concentrations in pg g-1 Bottling sequence . . Start Middle End Bottleno. . . 144 434 1014 2029 3769 4784 F . . . . 19.6 21.9 20.5 20.9 21.2 20.6 B[u]A. . . . 8.9 7.5 8.3 7.8 8.0 8.5 B[a]P . . . . 5.2 4.6 4.5 4.9 4.7 4.4 I[cd]P. . . . 5.0 5.0 4.3 5.6 4.5 4.7 A rotary evaporator was used to evaporate organic extracts containing PAH. At a water-bath temperature of 40 "C, this technique was found to be satisfactory for the 14 PAH determined in this work. Quantitative recoveries of the hydrocarbons were obtained unless the solution was evapor- ated to dryness. However, if the determination of the more volatile PAH, such as naphthalene, is required, a Kuderna - Danish evaporator equipped with a three-stage Snyder column should be used to minimise evaporative losses of the volatiles.Silica gel, neutral alumina and Florisil have been commonly used for the clean-up of sediment extracts containing PAH.2,11>27,28 Fully activated silica gel and neutral alumina both gave 295% recoveries of all PAH when microgram amounts of the hydrocarbons were spiked directly on to the columns. Activated Florisil also worked well for most PAH, however, as reported earlier, the satisfactory recovery of B[a]P could not be obtained on this column.11 Gel permeation chromatography with Sephadex LH-20 is also a popular approach to the clean-up of sediment extracts for PAH determination,2.20>28?29 especially for the separation of ali- phatic and aromatic hydrocarbons.In this work, clean-up of sediment extracts was carried out using a 20-g Sephadex LH-20 column with a 1 + 1 benzene - methanol elution system according to the method of Giger and Schaffner.2 However, it was found that in our work the Sephadex column did not further improve the clean-up of the EC-1 extracts after they were subjected to the silica gel and alumina columns. Table 2. Concentrations of PAH (pg g-1) in EC-1 obtained under various extraction conditions. Average of three analyses Solventsystem* . . . . A B C D A E Replicates . . . . . . 3 3 3 3 3 3 Extraction method . . Soxhlet Soxhlet Soxhlet Soxhlet Soxhlet Ultrasonic Extractiontime . . . . 8 h 8 h 8 h 8 h 24h 3x3min Phen . . . . .. . . 15.9 Anth . . . . . . . . 1.3 F . . . . . . . . 22.2 Py . . . . . . . . 17.2 B[a]A . . . . . . . . 8.5 Ghry . . . . . . . . 8.4 B[b]F . . . . . . . . 9.0 B[k]F . . . . . . . . 4.2 B[e]P . . . . . . . , 5.3 B[a]P . . . . . . . . 5.0 Pery . . . . . . . . 0.8 D[ah]A . . . . . . 1.4 B[ghi]P . . . . . . 4.3 I[cd]P . . . . . . . . 5.2 16.0 1.7 21.5 15.1 8.3 8.3 7.9 4.4 4.9 4.8 0.7 5.2 0.8 2.6 14.7 1.2 23.3 15.8 8.2 8.2 8.3 5.2 5.1 5.1 1 .o 5.3 1.5 4.8 15.9 1.3 22.7 16.8 8.0 7.5 7.8 4.0 4.6 4.9 0.8 2.9 1.3 2.9 15.2 1.5 22.4 16.2 8.2 8.7 7.8 4.6 5 .0 5.1 0.8 5.3 1.5 4.4 14.2 1 .o 21.4 16.8 10.2 7.5 8.2 3.6 4.4 4.9 0.8 5.4 1.2 4.4 * A = 41 + 59 hexane - acetone; B = 1 + 1 benzene - methanol; C = cyclohexane; D = hexane; and E = 1 + 1 hexane - acetone. Ti meimin Fig.1. gel and alumina column clean-up Total ion current chromatogram of EC-1 extract after silica loo 80 r 10 4 20 30 40 Time/m in34 ANALYST, JANUARY 1987, VOL. 112 Table 3. Mean concentrations of selected PAH (pg g-1, dry mass) in reference material EC-1. Uncertainty is one standard deviation Table 5. Certified concentration of selected PAH in reference material EC-1. Uncertainty is one standard deviation Phen . . Anth F . . Py . . BbIA Chry . . B[bIF B[kIF BkIP BbIP I[CdlP Pery . . D[ah]A B[ghi]P GC-FID GC-MSD HPLC* No. of analyses 30 12 30 . . . . . . 15.8f 1.2 16.2+ 1.5 14.9+0.4$ . . . . . . 1.3k0.3 1 . 2 f 0 . 3 0.8+0.1$ . . . . . . 22.5k2.0 23.6f 1.9 23.8k2.1 . . . . . . 16.8k 1.9 17.8-t 1.5 16.2f2.1 . . . . . . 8.5fO.9 8.7k1.0 8 . 8 f 0 . 6 . . . . .. 9.2f0.9-t 9.6f l.lt 7.920.84 . . . . . . 7.6k 1.2 8.5f0.9 8.0k0.5 . . . . . . 4.5f0.6 4.4k0.6 4.4k0.3 . . . . . . 5 . 2 f 0 . 6 5.7k0.8 5.4f0.3$ . . . . . . 5.4kO.7 5.8k0.7 5.0f0.6 . . . . . . 0.8fO.2 1.7k0.2 1.7k0.1$ . . . . . . 5.8fO.5 6.1k1.0 5.5k0.4 . . . . . . 1.3 20.2 1.5 k0.2 1.1 20.2 . . . . . . 4.6kO.7 5.4f1.0 4.9k0.4 * Results obtained by fluorescence detection except phen, anth, t Results include triphenylene. $ Results obtained by UV detection and no. of replicates was five. chry, B[e]P and pery. ~~~~ ~ Table 4. Interlaboratory results on selected PAH (pg g-l) in EC-1 No. of Parameter results* Range Median Mean f s.d. Phen . . Anth F . . Py . . B [ 4 4 BPIF WIF B[eIP BbIP I[CdlP Chry . . Pery . . D[ah]A B[ghi]P . . . . 11 9.9-24.35 . . . . 9 0.35-13.18 .. . . 15 14.87-45.3 . . , . 13 9.58-26.0 . . . . 11 4.6-15.6 . . . . 10 6.7-44 .O . . . . 11 3.68-15.2 . . . . 11 2.8-16.61 . . . . 8 3.12-7.76 . . . . 15 2.61-30.0 . . . . 7 1.05-2.19 . . . . 11 3.12-7.6 . . . . 10 1.44-11.0 . . . . 13 0.45-20.31 16.80 1 S O 21.81 18.50 7.60 8.80 6.75 3.63 5.36 4.50 1.16 4.90 2.35 4.73 16.57 f 4.59 3.92 f 4.70 23.45 f 7.39 18.42 f 5.21 8.41 f 3.04 13.70 f 11.86 8.08 k 3.64 5.58 f 4.17 5.55 f 1.50 6.58 f 6.77 1.49 f 0.48 5.10 k 1.36 3.62 f 2.35 7.32 f 6.48 * Some laboratories did not provide all the results obtained. Therefore, Sephadex and Florisil column clean-ups were not employed. At the early stage of PAH determination, a 12-m OV-1 capillary column was used. This column successfully resolved all 14 PAH of interest, including the following isomeric pairs: phen and anth, B[a]A and chry, and B[b]F and B[k]F. However, even better resolution of the PAH could be obtained by using a 30-m, thin-film DB-5 column.In the latter instance, base-line resolution was observed for all 14 PAH except B[b]F and B[k]F. This efficient column provided adequate resolution of the PAH and co-extractives in this complex EC-1 sample (Fig. 1). Initially, the PAH in EC-1 were identified by their retention times as they were chromatographed on the OV-1 and DB-5 columns. Their identities were positively confirmed by operat- ing the MSD in the scanning mode and comparing the mass spectra with authentic standards. For all 14 PAH, the match quality were better than 9800 (best match = 10000). As chrysene and triphenylene were not resolved on our GC columns and their mass spectra were very similar, we were not able to tell which one of the two, or whether a mixture of both PAH was present in EC-1.Nevertheless, this peak was quantitated against a chrysene standard in subsequent deter- minations. By monitoring only the characteristic molecular ions of the PAH, the MSD was extremely selective, as indicated by the reconstructed multi-ion current profile of an EC-1 extract (Fig. 2) versus the total ion current chromato- Concentration/ Parameter CLg g-l Phen . . . . . . 15.8 k 1.2 F . . . . . . . . 23.2 f 2.0 Py . . . . . . 16.7f2.0 B[a]A . . . . . . 8.7 f 0.8 B[b]F . . . . . . 7.9 k 0.9 B[k]F . . . . . . 4.4 f 0.5 B[e]P . . . . . . 5.3 f 0.6 B[a]P . . . . . . 5.3 k 0.7 I[cd]P .. . . . . 5.7 f 0.6 B[ghi]P.. . . . . 4.9 f 0.7 gram (Fig. 1). As the molecular ion is the most abundant ion for each PAH, SIM quantitation of these PAH was also highly sensitive. Under the conditions used the MSD was approxi- mately 100 times more sensitive than the FID in these PAH determinations. Because of availability of equipment, only nine of the 14 PAH in EC-1 were determined by HPLC. The samples were chromatographed on a reversed-phase C18 column and the nine PAH were detected by a filter fluorimeter (Table 3). Under isocratic conditions, several PAH eluted closely with each other and were not separated by modification of mobile phase composition. For example, B[a]A and chry, and B[b]F, pery and B[e]P were the two groups of co-eluting PAH. However, by operating the fluorescence detector at he, 280 and he, 389 nm, B[a]A was selectively detected in the presence of chry as the latter had very little fluorescence sensitivity at such wavelengths .7JO For similar reasons, the determination of B[b]F was not significantly interfered with by the presence of B[e]P and perylene in the same sample. Extracts of EC-1 were also determined by an independent laboratory for the 14 PAH using the gradient elution HPLC technique detailed in the US EPA Method 610.24 Under these conditions, all 14 PAH were resolved and they were quanti- tated by the fluorescence and UV detectors connected in series.The GC - FID, GC - MSD and HPLC results of the PAH in EC-1 are summarised in Table 3. Recently, we organised an interlaboratory study on the determination of PAH in sediment samples, including EC-1.22 The interlaboratory results submitted by 14 Canadian partici- pants were generated by a variety of extraction, clean-up and detection methods.Interlaboratory round-robin results alone are generally considered insufficient to certify environmental reference materials as the precision and accuracy of these results are obtained in an uncontrolled manner. Nevertheless, based on what we have learned in the 100 or more interlabora- tory studies organised by us, the interlaboratory mean or median results are usually good estimates of the true values in unknown samples. As shown in Table 4, the interlaboratory results further confirmed our in-house results as the two were in good agreement with each other.Levels for ten PAH (phen, F, Py, B[a]A, B[b]F, B[k]F, B[e]P, B[a]P, I[cd]P and B[ghi]P) in EC-1 were certified and their values are listed in Table 5. In these instances, the agreement between the in-house results using three indepen- dent detection techniques (Table 3 ) were all better than k 10%. The certified values were obtained by calculating the weighted averages of the pooled in-house results26 and the uncertainty was one standard deviation. Values for chrysene could not be ascertained because both GC methods were unable to resolve chrysene and triphenylene. The other three PAH are relatively minor components in EC-1 and the results listed in Table 3 are for information only. Currently, this sediment reference material is kept at -20 "C in the dark. No degradation of PAH in the sample hasANALYST, JANUARY 1987, VOL.112 been detected since the work was initiated approximately two years ago. In conclusion, a naturally contaminated lake sediment certified reference material was developed and certified for ten PAH at the yg 8-1 level. The certified values were derived by repetitive in-house analysis using three different methodol- ogies, i.e., GC - FID, GC - MS and HPLC techniques. These values were further confirmed by an interlaboratory study. This material is a valuable tool in the development and evaluation of analytical methods for PAH, and for the generation of accuracy statements in in-house and interlabora- tory quality assurance activities in such analysis. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. References Dipple, A., in Searle, C .E., Editor, “Chemical Carcinogens,” American Chemical Society, Washington, DC, 1976, ACS Monograph 173, pp. 245-314. Griger, W., and Schaffner, C . , Anal. Chem., 1978, 50, 243. Laflamme, R. E., and Hites, R. A., Geochim. Cosmochim. Acta, 1978, 42, 289. Bjarseth, A., Knutzen, J., and Skei, J., Sci. Total Environ., 1979, 13, 71. Bieri, R. H., Cueman, M. K., Sith, C. L., and Su, C . W., Int. J. Environ. Anal. Chem., 1978, 5 , 293. Lee, M. L., Novotny, M., and Bartle, K. D., “Analytical Chemistry of Polycyclic Aromatic Compounds ,” Academic Press, New York, 1981. Das, B. S., and Thomas, G. H., Anal. Chem., 1978, 50, 967. Ogan, K., Katz, E., and Slavin, W., Anal. Chem., 1979, 51, 1315. Konash, P. L., Wise, S. A., and May, W. E . , J. Liq.Chromatogr., 1981,4, 1339. Das, B. S., and Thomas, G. H., “Proceedings of the 9th Materials Research Symposium, April 10-13, 1978, Gaithers- burg, MD ,” National Bureau of Standards Special Publication 519, April 1979. Dunn, B. P., and Armour, R. J., Anal. Chem., 1980,52,2027. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 35 Joe, F. L., Jr., Salemme, J., andFazio, T., J. Assoc. OF. Anal. Chem., 1982, 65, 1395. Black, J. J., Dymerski, P. P., and Zapisek, W. F., Bull. Environ. Contam. Toxicol., 1979, 22, 278. Lee, M. L., and Wright, B. W., J. Chromatogr. Sci., 1980,18, 345. Lee, M. L., Vassilaros, D. L., and Later, D. W., Znt. J. Environ. Anal. Chem., 1982, 11, 251. BjGrseth, A., and Eklund, G., J. High Resol. Chromatogr. Chromatogr. Commun., 1979, 2 , 22. Miller, D. A., Skogerbock, K., and Grimsrud, E. P., Anal. Chem., 1981, 53, 464. Konig, J., Balfanz, E., Funcke, W., and Romanowski, T., Anal. Chem., 1983, 55, 599. Howard, A. G., and Mills, G. A., Int. J . Environ. Anal. Chem., 1983, 14, 43. MacLeod, W. D., Jr., Prohaska, P. G., Gennero, D. D., and Brown, D. W., Anal. Chem., 1982, 54,386. Hilpert, L. R., May, W. E . , Wise, S. A., Chesler, S . N., and Hertz, H. S., Anal. Chem., 1978, 50, 458. Lee, H. B., and Chau, A. S. Y., 1985, unpublished results. May, W. E., Chesler, S. N., Hertz, H. S . , and Wise, S. A., Int. J. Environ. Anal. Chem., 1982, 12, 259. US Environmental Protection Agency, “Guidelines Establish- ing Test Procedures for the Analysis of Pollutants,” Federal Register, Volume 49, No. 209, Washington, DC, Oct. 26,1984, Chau, A. S . Y., and Lee, H. B., J. Assoc. 08. Anal. Chem., 1980,63,947. Lee, H. B., Hong-You, R. L., and Chau, A. S. Y., Analyst, 1986, 111, 81. Tan, Y. L., J. Chromatogr., 1979, 176, 319. Giger, W., and Blumer, M., Anal. Chem., 1974,46, 1663. Ramos, L. S., and Prohaska, P. G., J. Chromatogr., 1981,211, 284. pp. 112-120. NOTE-Reference 25 is to Part V of this series. Paper A61208 Received June 27th, 1986 Accepted August 8th, 1986

ISSN:0003-2654    

DOI:10.1039/AN9871200031

出版商:RSC    

年代:1987

数据来源: RSC

摘要:

ANALYST, JANUARY 1987, VOL. 112 37 Analytical Reference Materials Part VII.* Development and Certification of a Sediment Reference Material for Total Polychlorinated Biphenylst Hing-Biu Lee and Alfred S. Y. Chau Quality Assurance and Methods Section, Analytical Methods Division, National Water Research Institute, Environment Canada, Burlington, Ontario, 17R 4A6, Canada A lake sediment reference material naturally contaminated with PCBs was prepared. Subsamples of this material were subjected t o repetitive in-house analyses for total PCBs. The quantitative recovery of PCBs was demonstrated by performing ultrasonic and Soxhlet extractions under various conditions. Sample extracts were usually cleaned-up with Florisil and the cleaned extracts were shown to be free from other major interferences, except for sulphur, which was later removed by mercury. The presence of PCBs in the samples was confirmed by the perchlorination of the sample extract and by GC - MS techniques.Most of the sample extracts were quantitated against a 1 + 1 + I mixture of Aroclors 1242, 1254 and 1260 on a 3% OV-I packed column using the technique described by Webb and McCall. These results were further confirmed by the quantitative GC - M S analyses of EC-1 and by interlaboratory results provided by independent laboratories. The reference material showed no signs of degradation of its PCB content over a six-year storage period a t -20 "C in the dark. Keywords: Polychlorinated biphenyl determination; certified reference materials; quality assurance; lake sediment samples Aroclors or mixtures of polychlorinated biphenyls (PCBs) with various percentages of chlorine by weight were manufac- tured in the United States by Monsanto Chemical Company.Because of their general inertness, PCBs were widely used as transformer dielectric fluids, plasticisers and flame retardants, etc., in industry. Although the production of Aroclors has been curtailed in recent years owing to government regula- tions, the disposal, dump leaching and accidental spillage of the millions of pounds of these persistent Aroclors manufac- tured in the preceding years still cause environmental pollu- tion problems in air, water and biota samples. Aroclors are classified as priority pollutants by the US Environmental Protection Agency and environmental moni- toring of the compounds was begun in the 1960s because of their toxicity and persistence. Although PCB determinations are considered routine, the results from many naturally contaminated samples are often unsatisfactory, as indicated by the many interlaboratory round-robin studies organised by our section192 and by other parties.3.4 Typically, the interlabor- atory relative standard deviation of PCB results in naturally contaminated sediment samples at low pg g-1 levels is between 25 and 50%.Hence, there is a need to develop a real-life reference material in order to monitor the perfor- mance of laboratories involved in PCB determinations. Several years ago, our section initiated the research and development of a number of sediment certified reference materials (CRMs) for toxic organics such as PCBs,5 chloro- benzenes6 and polynuclear aromatic hydrocarbons,7 in order to fulfil the requirements of our on-going quality assurance programmes. Two marine sediment CRMs were recently available commercially, however, no details were given on how these CRMs were characterised.In this paper, we report our rigorous approach to analyse and certify the total PCB contents in our sediment CRM coded EC-1. * For Part VI of this series, see p. 31. t This material is currently not for sale and not available for general distribution. Experimental Preparation of Sediment Reference Material Detailed procedures for sample collection, preliminary dry- ing, freeze-drying, blending, bottling and homogenity testing before and after subsampling have been described in previous publications .5,6 Extraction and Clean-up of Sediment Samples The Soxhlet and ultrasonic extractions of PCBs in sediment samples and the partitioning and evaporation steps were identical to those employed for the chlorobenzene determina- tions.6 The clean-up of sediment extracts was carried out in a 500 x 19 mm i.d.glass column filled with 20.0 g of activated Florisil and 10 mm of anhydrous sodium sulphate at the top. The PCBs in the extract were eluted by 200 ml of hexane. After the addition of 3 ml of iso-octane and a few boiling chips, this hexane fraction was evaporated down to ca. 3-5 ml using a three-stage macro Snyder column and a heating mantle. The concentrated extract was diluted to 10.0 ml with iso-octane. Sulphur and sulphur compounds in the cleaned-up extract were removed by vigorous agitation with mercury until the metal remained shiny.Gas Chromatography with Electron-capture Detection (GC - ECD) A Hewlett Packard 5700 series gas chromatograph equipped with a Ni-63 electron-capture detector, a Model 7671A autosampler and a Model 3390 reporting integrator was used. The column was a 1.8 m x 2 mm i.d. glass column packed with 3% OV-1 on 100-120 mesh Gas Chrom Q. The temperatures were: injection port, 250 "C; detector, 300 "C; and column, 185 "C. The flow-rate was 25 mi min-1 and the carrier gas argon - methane (95 + 5). Aroclors 1242,1254 and 1260 were supplied by the US Environmental Protection Agency. The working standard was a mixture of 1 + 1 + 1 Aroclors 1242, 1254 and 1260 in iso-octane with a total concentration of 600 pg p1-1.The quantitation of PCBs was carried out by the peak matching technique described by Webb and McCall.8ANALYST, JANUARY 1987, VOL. 112 38 Table 1. Ions, congeners and concentrations of calibration standards used in the capillary column GC - MSD analysis of EC-1 mtt Calibration congener Concentration/ Homologue Quantitate ion Confirm ion Congener no. * pg 4-1 c11 . . . . . . . . Cl;! . . . . . . . . c13 . . . . . . . . Cl'$ . . . . . . . . c15 . . . . . . . . C16 . . . . . . . . c1, . . . . . . . . C18 . . . . . . . . c19 . . . . . . . . CllO . . . . . . . . 188 222 256 292 326 360 394 430 464 498 190 224 258 290 328 362 396 432 466 500 1 5 29 50 87 154 181 200 209 209 50 50 50 100 100 100 150 150 250 250 250 - - Chrysene-d,, .. . . 240 * Adopted from reference 9. Table 2. Summary of results (pg g-1 total PCBs) from EC-1 sediment reference material by packed column Webb - McCall quantitation method Extraction Ultrasonic Soxhlet No. ofanalyses . . . . 97 72 R a n g e k g-1 . . . . 1.85-2.15 1.88-2.17 Mean k SD . . . . . . 2.02 k 0.07 1.97 k 0.08 Gas Chromatography with Mass-selective Detection (GC - MSD) A Hewlett Packard 5880A gas chromatograph equipped with a split - splitless injection port, a Level I1 terminal, a Model 7671A autosampler, a Model 5970B mass-selective detector (MSD), a Model 9816s computer and a Model 9133XV 15 megabytesdiscdrivewereused, togetherwitha30m x 0.25mm i.d. DB-5 fused-silica capillary column, which was directly interfaced to the electron impact ion source for maximum sensitivity. The GC conditions were: injection port, 275 "C; interface, 280 "C; column initial temperature, 70 "C (held for 1.5 min); programming rate, 30 "C min-1 (70- 170 "C), 2.5 "C min-1 (170-260 "C); and oven temperature held at 260 "C for 15 min.The splitless valve was on for 1.5 min and the column head pressure was 4 lb in-2. Fully automated sample injection, data acquisition, data editing and report generation was made possible with the existing "Sequencing" software and a keystroke program on the GC terminal. The detector was operated in the selected ion monitoring (SIM) mode for both confirmation and quantitation. In both instances, two characteristic ions (one for quantitation and the other for confirmation) were monitored for each PCB homologous series (Table 1).For quantitative work, the procedure described by Budde and co-workers10J1 was used, except that the 2,2' ,3,4,4' ,5,6-heptachlorobiphenyl (congener 181) instead of 2,2',3,4',5,6,6'-heptachlorobiphenyl (congener 188) was used owing to availability. Chrysene-d12 was used as an internal standard and the dwell time for each ion was 100 ms. Sediment extracts were quantitated against an iso-octane mixture of nine congeners of various concentrations (Table 1). At each level of chlorination, one PCB congener in the calibration mixture was used as the concentration standard for all isomers in that group, e.g., congener 29 was used to quantitate all trichlorobiphenyls in the sample. The only exception was that decachlorobiphenyl was used as the concentration calibration standard for both nona- and deca- chlorobiphenyls in the sample.The total PCB concentration was obtained by the summation of all concentrations in each homologous series. All PCB congeners were obtained from Ultra Scientific, Hope, RI, USA. Chrysene-d12 was obtained from Aldrich Chemical, Milwaukee, WI, USA. Results and Discussion Efforts were made to ensure that the PCBs in EC-1 were quantitatively recovered. The ultrasonic and Soxhlet extrac- tion techniques that are routinely used for the extraction of organics in sediments were employed and compared. In order to achieve a valid comparison between the extraction methods, identical clean-up (Florisil column) and quantitation (packed column Webb - McCall) procedures (see later discussions) were used in these samples.A total of 97 PCB determinations were carried out on EC-1 by the ultrasonic extraction technique using a 1 + 1 mixture of acetone and hexane. A total of 72 determinations were also carried out on the same certified reference material by Soxhlet extraction using 59 + 41 acetone - hexane. All of these results are summarised in Table 2. It is obvious from these data that nearly identical results were obtained from both techniques. As the presence of moisture has been reported to provide better recoveries of some organochlorines in soil samples,12 EC-1 samples with 0 or 30% moisture content were Soxhlet extracted simultaneously for comparison. A t-test was applied to the means of these results and no difference was found in these means at the 95% significance level, indicating that a moisture content of 0 or 30% in the sediment samples had no effect on the recovery of PCBs.Another t-test was applied to the sample results obtained by ultrasonic extraction under similar conditions and again no difference in PCB recovery was observed. As similar PCB results were obtained from samples Soxhlet extracted for 3,7, 20 and 72 h, it was implied that PCBs were exhaustively removed from this sediment after 3 h of Soxhlet extraction. Different solvents were also employed to see if the recovery of PCBs was solvent dependent. The results indi- cated that both acetone and methylene chloride gave similar recoveries of PCBs to 59+41 acetone - hexane, whereas non-polar solvents such as hexane and benzene gave slightly lower (ca.90%) recovery. Florisil has been used by many workers13 to remove co-extractives in sediment samples before organochlorine and PCB determinations. Although some workers have preferred to use Florisil deactivated with a few percent. of water for the column clean-up7 activated Florisil was used in this study because it gave a better separation of some organochlorines and polynuclear aromatic hydrocarbons from PCBs. In order to ensure that the Florisil clean-up was effective, some of the samples were subjected to additional clean-up steps. In triplicate runs, the PCB fraction of the Florisil- cleaned EC-1 extract was further cleaned-up on a Celite and 3% deactivated silica-gel column14 and, in another instance, on an activated neutral alumina coIumn,15 according to published methods.In both instances, no change in the PCB profile and amount could be observed in the EC-1 extractsANALYST, JANUARY 1987, VOL. 112 39 0 10 20 30 40 50 Tim eim i n Fig. 1. GC - ECD chromatogram of the PCB fraction in EC-1. A 1.8 m X 2 mm i.d. 3% OV-1 column was used Table 3. Composition of PCBs in EC-1 as determined by quantitative GC - MS analyses (six replicates). Uncertainty is one standard deviation EC-1 Homologue c11 . . . . . . . . Clz . . . . . . . . c13 . . . . . . . . Clq . . . . . . . . c1s . . . . . . . . C16 . . . . . . . . c1, . . . . . . . . Clfj . . . . . . . . c19 . . . . . . . . CllO . . . . . . . . Concentration/ ni3g-l Nd * Nd 184 k 12 (9.9)t 453 k 24 (24.3) 688 k 33 (36.9) 280 f 21 (15.0) 161 k 6.7 (8.6) 98 k 7.6 (5.3) Nd Nd Total concentration .. 1864 * Nd = None detected. t Figures in parentheses show % m/m of each chlorobiphenyl in EC-1. after additional clean-up. The Florisil-cleaned EC-1 extract was also subjected to an ethanolic KOH treatment at 80-90 "C for 30 min. Again, no change in the PCB components could be observed before and after the additional clean-up. These experiments indicated that Florisil-cleaned EC-1 extracts were free from any major interference from other organochlorines. It should be noted that a few chlorobenzenes, p,p'-DDE and Mirex present in EC-1 could not be separated from the PCBs by column chromatography. Their presence, however, would not affect the PCB results as their concentrations were relatively low compared to those of the PCBs.Most of the EC-1 extracts were chromatographed on a 6-ft 3% OV-1 column operated at 185 "C. The PCBs were quantitated by the established peak-matching technique developed by Webb and McCall.8 This method was used as it has been demonstrated to be better than other packed column techniques in a collaborative study.16 It is still the official method for the quantitation of total PCBs in our Water Quality laboratories and is also approved by the US EPA (Method 608).17 Samples were quantitated against a 1 + 1 + 1 mixture of Aroclors 1242, 1254 and 1260; preliminary runs of sample extracts had indicated that the PCB components in EC-1 were very similar to this mixture (Fig. 1). Extracts of EC-1 were subjected to the perchlorination procedure described by Armour18 using antimony penta- chloride.The formation of decachlorobiphenyl in these reactions confirmed the presence of PCBs in the EC-1 extracts. The perchlorination results were not used quantita- Table 4. Summary of interlaboratory results for total PCBs in EC-1 Studynumber . . . . N-27 DQC-3 No. of laboratories . . 15 14 No. of results used* . . 25 12 Range of results/pg g-1 0.96-3.41 1.11-3.26 Mediadpgg-1 . . . . 1.96 1.75 Mean f s.d./vgg-1 . . 2.05 k 0.61 1.98 2 0.69 * After rejection of outliers. tively to determine the PCB concentration in EC-1 as the latter was a complicated mixture of several Aroclors. A concentrated extract of EC-1 (containing approximately 3.0 pg ml-1 of total PCBs) was analysed on a 30-m DB-5 column interfaced to a mass-selective detector.Data were acquired in the selected ion monitoring mode for the detection of the ten chlorobiphenyl homologous series, i. e., from mono- to decachlorobiphenyl. Two characteristic ions were used for each homologous series: one for quantitation and the other for confirmation, as shown in Table 1. Although the mono-, di-, nona- and decachlorobiphenyls were not present in EC-1 in detectable amounts, the presence of tri-, tetra-, penta-, hexa-, hepta- and octachlorobiphenyls in this reference material was confirmed by the presence of both characteristic ions at the right retention times and in the expected ratios for each of the six homologous series listed above. PCBs in EC-1 extracts were also quantitatively determined by GC - MSD using the method described by Budde and co-workers, 10911 As PCB homologues have overlapping reten- tion time windows, special precautions were taken to avoid interferences by PCB congeners containing more chlorines. Under the electron ionisation mode, a PCB molecule under- goes fragm5ntation by the loss of two chlorines, and to a lesser extent by the loss of HC1 and Cl,19,2O thus causing interference in the determination of PCBs with one or two less chlorine atoms.In this work, the level of chlorination in each PCB peak was previously determined by a full scan run of a concentrated Aroclor mixture. The level of chlorination in a sample PCB peak was first assigned by the observed relative abundance of the two corresponding characteristic ions. This, together with the information obtained in the full scan run, was generally sufficient to eliminate interference generated by fragmenta- tion ions produced by co-eluting PCBs with more chlorines.The results of the quantitative GC - MSD determination of PCBs in EC-1 are shown in Table 3. In this instance, the concentration of each PCB homologous series and the total PCB concentration were obtained. As indicated, the GC - MS results further confirmed the GC - ECD results as the total PCB concentrations obtained by these two different quantita- tive methods (Tables 2 and 3) varied by less than 10%. The over-all lower sensitivity of the MSD to most PCBs, especially the hepta- and higher chlorobiphenyls, rendered some PCB peaks undetected by this detector at low concentrations. This could be the reason why, in the example of EC-1, that the total PCB results obtained by mass-selective detection were slightly lower than those obtained by electron-capture detection.Reference material EC-1 was used in two interlaboratory round-robin studies in two different years. In both instances, the participants were requested to analyse the material for total PCBs by using their own in-house methods and stan- dards. The interlaboratory results (Table 4) were diversified because of the different extraction and clean-up method- ologies and the calibration standards and quantitation tech- niques employed by various participants. 1~ Despite all these variations, the interlaboratory medians and means of the PCB results in both studies were in excellent agreement with the in-house results as summarised in Tables 2 and 4.The stability of PCBs in EC-1 under cold storage conditions (-20 "C in the dark) was monitored twice annually. The40 ANALYST, JANUARY 1987, VOL. 112 results give no evidence of degradation during storage over the last six years.21 In conclusion, we have successfully prepared and certified a naturally contaminated lake sediment reference material (EC-1) for total PCB contents on the basis of 169 in-house determinations. The reference value, 2.00 k 0.05 pg g-1 (uncertainty is one standard deviation), generated by com- bining all the GC - ECD results (Table 2), was further supported by GC - MS results and interlaboratory results in two round-robin studies. This certified reference material is currently being used in many of our intralaboratory and interlaboratory quality assurance programs for PCB deter- minations.1. 2. 3. 4. 5. 6. References Lee, H. B., and Chau, A. S. Y., “National Interlaboratory Quality Control Study No. 27-PCBs in Naturally Contami- nated Dry Sediments,” Inland Waters Directorate Report Series No. 72, Environment Canada, 1981. Lee, H. B., Dookhran, G., and Chau, A. S . Y., “Dredging Quality Control Study No. 3 (DQC3)-Analysis of PCBs in Naturally Contaminated Dry Sediments and Standard Solu- tions,” National Water Research Institute Contribution, Burl- ington, Ontario, 1985 Alford-Stevens, A. L., Budde, W. L., and Bellar, T. A., Anal. Chem., 1985, 57, 2452. Musial, C. J., and Uthe, J. F., J . Assoc. Off. Anal. Chem., 1983, 66, 22. Chau, A. S. Y., and Lee, H. B., J. Assoc. Off. Anal. Chem., 1980,63,947. Lee, H. B., Hong-You, R. L., and Chau, A. S. Y., Analyst, 1986, 111, 81. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. Lee, H. B., Dookhran, G., and Chau, A. S. Y., Analyst, 1987, 112, 31. Webb, R. G., and McCall, A. C., J . Chromatogr. Sci., 1973, 11, 366. Ballschmiter, K . , and Zell, M., Fresenius 2. Anal. Chem., 1980,302, 20. Gebhart, J. E., Hayes, T. L., Alford-Stevens, A. L., and Budde, W. L., Anal. Chem., 1985,57, 2456. Silvon, L. E., Gebhart, J. E., Hayes, T. L., Alford-Stevens, A. L., and Budde, W. L., Anal. Chem., 1985,57,2464. Williams, I. H., J. Assoc. Off. Anal. Chem., 1968, 51, 715. Lee, H. B., Chau, A. S. Y., and Kawahara, F. K., in Chau, A. S. Y., and Afghan, B. K., Editors “Analysis of Pesticides in Water,” Volume 11, CRC Press, Boca Raton, FL, 1982, pp. 1-40. Armour, J. A., and Burke, J. A., J. Assoc. Off. Anal. Chem., 1970,53,761. Telling, G. M., Sissions, D. J., and Brinkman, H. W., J. Chromatogr., 1977, 137, 405. Sawyer, L. D., J . Assoc. Off. Anal. Chem., 1978, 61, 282. U.S. Environmental Protection Agency, Fed. Regist., 1984,49, No. 209, 89. Armour, J. A., J. Assoc. Off. Anal. Chem., 1973, 56, 987. Safe, S., and Hutzinger, O., J. Chem. SOC. Perkin Trans. I . 1972, 686. Tindall, G. W., and Wininger, P. E., J. Chromatogr., 1980, 196, 109. Lee, H. B., and Chau, A. S. Y.. unpublished results. No~~-Reference~7 is to Part VI of this sehes. Paper A61220 Received July 9th, 1986 Accepted August 26th, 1986

ISSN:0003-2654    

DOI:10.1039/AN9871200037

出版商:RSC    

年代:1987

数据来源: RSC

摘要:

ANALYST, JANUARY 1987, VOL. 112 41 Determination of Uranium(V1) in Process Liquors by Ion Chromatography John J. Byerley and Jeno M. Scharer Department of Chemical Engineering, University of Waterloo, Waterloo, Ontario N2L 3G 7, Canada and George F. Atkinson Department of Chemistry, University of Waterloo, Waterloo, Ontario N2L 3G I , Canada Two methods are described for the determination of Uvl in process liquors using ion chromatography based on cation separation and cation - anion separation with ammonium sulphate - sulphuric acid as the eluent. The UV1 species is detected spectrophotometrically at 520 nm after post-column reaction with 4-(2- pyridy1azo)resorcinol. Chromatographic and detector variables, such as eluent composition and concentra- tion, metallochromic indicator concentration and eluent and indicator flow-rates, are discussed.The method is linear for peak heights up to 15 pg ml-1 and has a quantitation limit of 0.04 pg ml-1 using direct injection. Keywords: Uranium(V1) determination; ion chromatography; 4-(2-pyridylazo)resorcinoI; process liquors; mixed-mode column An important requirement of the hydrometallurgical industry is the rapid, inexpensive and reliable determination of metallic species in process liquors and mill effluents. These analytical requirements have generally been met using a number of instrumental spectroscopic techniques, including flame emis- sion spectrometry (FES) , atomic absorption spectrometry (AAS) and atomic emission spectrometry (AES). When these and other related methods are combined with various excita- tion sources and aqueous solution handling procedures, at least 75 elements can be determined.It is to be expected that with this broad range of applicability particular examples will arise where an alternative technique should be considered. Uranium and certain other metals that form refractory oxides in a flame are difficult to determine by atomic absorption spectrometry. This problem is not serious when atomic emission spectrometry is applied in conjunction with direct and inductively coupled plasma excitation sources. Satisfac- tory determination of aqueous uranium in the pg ml-1 range is possible. Lynch et al. 1 have reported a flow injection method involving solvent extraction and adsorptiometric determina- tion of uranium in leachates and effluents which gave a detection limit of 0.1 pg ml-1.Neutron activation analysis is a popular non-spectroscopic method for the determination of uranium in aqueous solution; however, the availability of a neutron source and the need for a preliminary de-watering step are important considerations. The detection limit claimed by commercial suppliers for this method using standard procedures is 0.01 pg ml-1. Our research has been directed towards studying the absorption - desorption characteristics of aqueous metallic species in contact with various microorganisms.2 Of particular interest has been the behaviour of the uranium(V1) present in acid process leach liquor when in contact with biomass. The process liquor contains uranium(V1) at levels of the order of 200 pg ml-1 and substantial levels of other species, particularly iron(II1).Metals such as thorium, zinc, copper, nickel and cobalt are also present in the liquor, which contains about 2.5 g 1-1 of sulphate. In the absorption - desorption studies of uranium(V1) in contact with biomass, both equilibrium and kinetic experi- ments are required. The large number of samples taken in these experiments were initially analysed for uranium(V1) using neutron activation analysis. However, reproducibility problems and the absence of a convenient neutron source prompted a search for an alternative method. Recently a number of reportss5 have indicated that ion chromatography can be applied to the determination of many transition metal species in aqueous solution. The method relies on the use of low-capacity cation- and anion-exchange materials.Cation- exchange separations are obtained using a surface-sulpho- nated microporous polystyrene - divinylbenzene resin. Anion- exchange separations are accomplished using the surface- sulphonated resin coated with aminated latex particles. This exchanger does exhibit a residual cation-exchange capacity. Both exchangers are reported to offer a high efficiency and excellent pH stability. A wide range of selectivity for transition metals can be achieved by a variation in particle size, functional groups, degree of latex cross-linking and, most importantly, by the use of both neutral and anionic complex- ing agents in the eluent. The determination of UVI as U022+ by ion chromatography was reported by Riviello using a sulphate eluent .6,7 This paper describes the application of ion chromatography, post-column derivatisation and spectropho- tometric detections.9 for the determination of UVI in process leach liquors.Experimental All ion chromatographic analyses reported were carried out with a Dionex Series 2010i ion chromatograph equipped with a Dionex Ionpac membrane reactor for post-column derivatisa- tion. Detection was at 520 nm using a Cary 219 UV-visible spectrophotometer fitted with a 10 mm (8 p1) flow cell. Two chromatographic columns were employed for the Um deter- minations, an HPIC-CS2 cation separator and HPIC-CSS cation - anion separator (Dionex). Both separators were operated with the appropriate guard columns. Eluent reagents, ammonium sulphate and sulphuric acid were of analytical-reagent grade and were obtained from J.T.Baker Chemical Company. The metallochromic indicator, 4-(2- pyridy1azo)resorcinol (PAR) , was supplied as the mono- sodium salt monohydrate by Aldrich Chemical, Milwaukee, WI, USA. The indicator was supplied to the membrane reactor from a nitrogen pressurised reservoir. Other com- ponents of the indicator solution, ammonia solution and acetic acid were of analytical-reagent grade (J. T. Baker Chemical Company). Synthetic uranium solutions used for this study were prepared from analytical-reagent grade uranyl nitrate (BDH, Canada). A large carboy of biologically produced uranium process liquor was obtained from Dennison Mines, Elliott Lake, Ontario, Canada. The analysis of this bulk sample for UVI was carried out by Dennison Mines.The reported Uw concentration was confirmed by a commercial laboratory using neutron activation. The addition of other42 ANALYST, JANUARY 1987, VOL. 112 metal species to synthetic uranium solutions was carried out using atomic absorption standards. The metal content of process liquors and synthetic solutions was determined using atomic absorption spectrometry. Results and Discussion Typical chromatograms of synthetic solutions containing 2.0 and 1.0 pg ml-1 of UVI as U022+ are shown in Fig. 1. These chromatograms were obtained using an anion - cation mixed- mode separator (Dionex HPIC-CS5) and a cation separator (Dionex HPIC-CS2) followed by post-column derivatisation and spectrophotometric detection.The chromatogram for the CS5 column exhibits an early peak (above and below the absorption base line), which results from water in the sample and unretained ionic species eluting in the void volume. This early dip has been eliminated in the chromatogram for the CS2 column by matching the sample matrix with the eluent. This was found to be good practice whenever possible, but is especially significant when separating ions at low levels. In comparing the two columns, it was observed that the cation separator produced a UO22+ response that was 30% higher than the anion-cation separator under the same chromatographic conditions. Table 1 summarises the con- ditions used for operating both columns and the post-column detection system. These conditions are similar to those reported by Riviello6.7 for the cation separator and were routinely used for U V I determination.Alternative conditions were often used for certain analytical requirements; the effects of these variations on column performance serve to charac- terise the columns. 300 200 100 0300 200 100 0 Timeis I 1 I I I I I I I 200 100 0200 100 0 Time/s Fig. 1. Typical chromatograms for synthetic UVI solutions, 1.0 and 2.0 pg ml-I. (a) CS5 column and ( b ) CS2 column. Conditions as in Table 1 Table 1. Operating conditions for cation (Dionex HPIC-CS2) and cation - anion (Dionex HPIC-CS5) columns Eluent . . . . . . . . Eluentflow-rate . . . . Column pressure drop . . Metallochromic indicator Indicatorflow-rate . . . . Samplematrix . . . . Samplevolume . . . . Absorptionscale .. . . Wavelength . . . . . . 0.02 M (NH4)2S04 and 0.20 M H2SO4 1 .O ml min-1 720-740 lb in-2 = 49&5100 kPa (CS5) 48&510 lb in-2 = 3310-3520 kPa (CS2) 4 x 10-4 M 4-(2-pyridylazo)resorcinol 3.0 M NH3 solution 1 .O M CH,COOH 0.4 ml min-1 Variable 50 p1 0.20 absorbance units full scale (a.u. f.s.) 520 nm Table 2 summarises the column performance for the determination of UvI employing both cation and cation - anion separation modes. Peak heights were normally used as an indication of UVI concentration. Fig. 2 shows chromatograms obtained by direct sample injections of 5.0, 2.0 and 1.0 pg ml-1 UVI standards using the cation - anion separator column (CS5) under standard operating conditions. Similar performance was obtained using the cation separator column (CS2).Both columns produced chromatograms that verify the linearity of the chromato- graphy up to 10 pg ml-1. An extension of the linear range was possible with an adjustment in operating conditions. An increase in the metallochromic indicator flow-rate was effective in increasing the range, but a much higher base-line variation was observed. A decrease in the eluent flow-rate at constant indicator flow-rate extended the linear range to 15 pg ml-1 with no adverse effect on the base-line stability, but the peak heights were slightly reduced. As noted in Table 2 the retention times observed for the two separation modes were somewhat different. For the cation column separation, which shows a shorter retention time (125 s), it is possible that the complex formed with U022+ in the eluent stream dissociates on the column and moves through as a simple aquated species.If the cation - anion column is used, it is possible that the anionic uranyl sulphate species present undergo anion exchange and that the corre- sponding aquated species are retained on the substrate resin by cation exchange. The dual functionality of this column would be expected to provide a superior selectivity for transition metals and may be the reason for the longer retention times. The performance of the system was investigated using routine techniques and minor alterations in the column and detection conditions. As noted in Table 2, the quantitation limit is estimated to be 0.090-0.120 pg ml-1 using modest spectrophotometric sensitivity. The limit of detection is lower than the quantitation limit.An increase in injection volume from 50 to 100 pl reduced the detection limit proportionally. Larger injection volumes may be used in some instances, but when process samples containing high concentrations of other ionic species that are readily retained by the column are injected, the intervals between sample injection must be greatly increased to avoid interference from these slow eluting species. An improvement in the quantitation limit is more readily achieved by increasing the sensitivity of the detecting instru- ment and reducing the base-line variation. In the case of the cation separator, using a full-scale absorbance range of 0.02 and a reduced metallochromic indicator concentration (1 x 10-4 M), with all other conditions remaining the same, the direct injection of a standard containing 0.10 pg ml-1 of UVI yielded a peak height of about 11.5% of the full scale and a base-line variation of 2%.Fig. 3 shows the chromatograms obtained for three standard UVI solutions containing 0.20, ~ ~ Table 2. Column performance. Operating conditions as in Table 1 c s 2 Retentiontime* . . . . 125s Linearrange . . . . 610pgml 1 Peak height . . . . 13% full scale at 1 .O pg ml- (0.20 a.u.f.s.) Quantitation limit (direct injection) . . 0.090 pg ml-1 Base-line variation . . 0.4% of full scale (0.20 a.u.f.s.) Signaltonoiseratio . . 3 : 1 c s 5 225 s <10pgml-l 1Ooh full scale at 1 .O pg ml-1 (0.20 a.u.f.s.) 0.120 pg ml- 0.4% of full scale (0.20 a.u.f.s.) 3 : 1 * Retention time reported as the time from sample injection to elution peak.ANALYST, JANUARY 1987, VOL.112 43 300 200 100 0300200 100 0 200 100 0 Timeis Fig. 2. Linearity verification, CS5 column. Synthetic Uvl solutions, 5.0, 2.0 and 1.0 pg ml-l; conditions as in Table 1 200 100 0 100 0 100 0 200 100 0 Time/s Fig. 4. Evidence of reproducibility of CS2 column. Synthetic UVI solution, 1.10 pg ml-l; conditions as in Table 1 I To.01 AU I I I I Y I Timeis Fig. 5. conditions as in Table 1 Uranium process liquor chromatogram using CS5 column; 200 100 0200 100 0 200 100 0 Ti m e/s Fig. 3. Low-level chromatograms using CS2 column. Synthetic Um solutions, 0.20, 0.10 and 0.05 pg ml-1; reagent, 1 x 10-4 M; other conditions as in Table 1 0.10 and 0.05 pg ml-1 of UVI using the above conditions. The quantitation limit if a signal to noise ratio of 3 : 1 is assumed is about 0.05 pg ml-1.It was observed that if the (NH&S04 concentration in the eluent was increased to 0.08 M the peak height was enhanced by about 25% with no effect on the base-line variation but with the retention time reduced to about 100 s. Under these conditions the quantitation limit may be decreased to 0.4 pg ml-1. The performance of the system using the cation - anion separator was found to be similar to this, but the reduced peak height response for this column resulted in a proportionately higher quantitation limit. The quantitation and detection limits of both columns could be greatly improved by the substitution of a concentrator column in place of the direct injection sample loop.When analysing samples containing UVI concentrations below about 0.50 pg ml-1 it was necessary to be particularly careful in flushing the entire system, otherwise sample to sample contamination resulted in memory effects. It was generally observed that the optimum chromatographic perfor- mance was achieved when the sample matrix was matched to the eluent. This was particularly true in the determination of low level samples. If the sample matrix is perfectly matched (which is not always possible) to the eluent, the water dip is eliminated. Both the cation and cation - anion separator modes showed excellent reproducibility. Fig. 4 shows the chromatograms of four successive injections of a standard 1.1 pg ml-1 Uvl sample using the cation separator column. Taking the peak heights of the four peaks shown and a further four injections, the digitally recorded data gave a mean peak height absor- bance above base line of 0.0283 and a standard deviation of 0.0005. The matrix of the sample used in this series was matched to the eluent, although minor mismatch was found to have little influence on reproducibility.The delivery of a sufficient and constant supply of derivati- sation reagent (PAR) to the post-column reactor was an important factor in achieving a linear response, acceptable reproducibility and minimum base-line variation. This did not normally present a problem. Some consideration was given to the question of the kinetics of colour development. Experi- ments conducted independently of the flow system showed that complete derivatisation occurred in less than 2 s.In our flow system, the time interval from the post-column contact of reagent and eluted UVI to spectrophotometric detection was about 2 s. Increasing this time interval by 50% had no detectable effect on the analytical results if the tendency for peak broadening was ignored. It should be noted that this behaviour cannot be assumed for all metal species where post-column derivatisation and downstream spectrophoto- metric detection is employed. Work being carried out by the authors indicates that the detection of ThIV by spectro- photometry using PAR requires a residence time of more than 40 s for full colour development.10 Synthetic samples containing UVI (2.0 pg ml-1) and Ni", Co", CuII, ZnII, FeII (10.0 pg ml-I), FeIII (20.0 pg ml-1) and ThIV (2.0 pg ml-I), both individually and in combination, were prepared in sulphate solution.Passing through the cation anion separator the bivalent metals appeared as one peak with a retention time of about 550 s. The elution of FeI" did not begin until 800 s and resulted in a very broad peak with an estimated retention time of 1130 s. Chromatograms of samples containing UVI and each metal in turn verified the makeup of the bivalent metal peak. Retention times varied from 530 s for Zn" to 565 s for CoII. As expected, FeIII was strongly retained and was eluted at the same time as observed when present in the composite sample. The presence of ThIV was not observed on any of the chromatograms. This is probably because it was not suf- ficiently retained under the chromatographic conditions, or because the conditions for derivatisation were not appropriate for its detection.The reproducibility and linearity of the Uvl peak heights after many injections of composite samples was comparable to the performance observed for pure UVI samples, indicating minimal impurity accumulation and column capacity loss. An acid leach uranium process liquor (pH 2.3) was reported by Dennison Mines to contain 160 pg ml-1 of UVI, 1120 pg ml-1 of FeIII, 9.0 pg ml-1 of Zn", 2.6 pg ml-1 of Cu", 2.3 pg ml-1 of CO" and 2.0 pg ml-1 of Ni", together with unreported amounts of aluminium, calcium, magnesium and thorium. The UVI concentration was confirmed using neutron activation analysis by NAS, Hamilton, Ontario. This liquor was diluted 100-fold with the eluent and injected repeatedly into the system. Fig.5 shows the chromatogram obtained for the sample of process liquor. This chromatogram was similar to that obtained for the synthetic composite sample. All the peaks were reproducible and there was no indication of ionic accumulation. The presence of large amounts of Fe"1 in uranium process liquors requires an interval at least 25 min between injections44 ANALYST, JANUARY 1987, VOL. 112 700 600 500 400 300 200 I 0 Ti meis Fig. 6. Uranium process liquor chromatogram using CS5 column. Eluent, 0.08 M (NH4)+S04, 0.2 M H,SO,; other conditions as in Table 1 to ensure the complete elution of Fe"'. The Fe"' could be rapidly eluted by using oxalate, although re-establishing equilibrium with the original eluent would then need to precede a further determination.As a reasonable compro- mise, if the (NH4)*S04 level was increased to 0.08 M, the elution of Fe"1 was complete within 12 min. Fig. 6 shows the chromatogram obtained. The use of ion chromatography for the determination of Uvl in a wide range of aqueous solutions, both synthetic and industrial, is now routine practice in our laboratory. Over 500 samples have been analysed using the system described with no apparent loss of column response. Biological processes related to the sample create no observable problems in the procedure. The method deserves consideration for the deter- mination of Uvl in the low pg ml-1 range and could be used successfully in the ng ml-1 range with pre-concentration. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. The support of the National Sciences and Engineering Research Council of Canada and the Department of Supply and Services of Canada (Contract OSQ84-00277) is gratefully acknowledged. References Lynch, T. P., Taylor, A. F., and Wilson, J. N., Analyst, 1983, 108,470. Byerley, J. J., Scharer, J. M., and Charles, A. M., "Evaluation of Biomass for UVI Recovery from Process Streams," Govern- ment of Canada Department of Supply and Services, Ottawa, No. OSQ84-00277, May 1985 pp. 1-178. Pohl, C. A., and Riviello, J. M., Paper No. 108, 24th Rocky Mountain Conference, Denver, CO, August 1982. Herberling, S. S., and Riviello, J . M., 27th Rocky Mountain Conference, Denver, CO, July 1985. Riviello, J. M., in Naden, D., and Streat, M., Editors, "Ion Exchange Technology," Society for Chemical Industry, Lon- don; Ellis Horwood, Chichester, 1984, pp. 584-594. Riviello, J. M., personal communication, 1985. Application Note No. 48, Dionex, Sunnyvale, CA, August 1983. Fritz, J . S., and Story, J. N., Anal. Chem., 1974,46, 825. Fritz, J. S . , and Story, J . N., Talanta, 1974, 21, 892. Byerley, J. J., Atkinson, G. F., andTrang, C. V., unpublished work. Paper A6190 Received March 17th, 1986 Accepted July 31st, 1986

ISSN:0003-2654    

DOI:10.1039/AN9871200041

出版商:RSC    

年代:1987

数据来源: RSC

摘要:

ANALYST, JANUARY 1987, VOL. 112 45 Gas Chromatographic Determination of Acrolein in Rain Water Using Bromination of 0-Methyloxime Harumitsu Nishikawa and Tomokuni Hayakawa Gifu Prefectural Research Institute for Environmental Pollution, 58-8, Yabuta, Gifu-shi, 500 Japan and Tadao Sakai Department of Chemistry, Asahi University, 185 7 Hozumi, Hozumi-cho, Gifu, 501-02 Japan A gas chromatographic method using an electron-capture detector was developed for the determination of acrolein based on the bromination of 0-methyloxime. Acrolein was determined by gas chromatography with a 3% silicone Ge XE-60 packed column and the calibration graph showed good linearity in the range 0-0.06 yg ml-1 of acrolein in aqueous solution. The detection limit was 0.4 ng ml-1 of acrolein (signal t o noise ratio = 2) and the relative standard deviation from five determinations of 0.04 yg ml-1 of acrolein in aqueous solution was 4.5%.This method is satisfactory for the selective and reproducible determination of trace amounts of acrolein in rain water. Keywords: Acrolein determination; gas chromatography; acrolein bromination; 0 -methyloxime; rain water Aldehydes are present in vehicle exhaust gases and are formed by photochemical reactions with hydrocarbons in air. As aldehydes are therefore related to photochemical oxidant concentrations, it is very important to be able to determine them in air, as an indication of air pollution levels. Acrolein is a particularly important aldehyde and spectrophotometric and fluorimetric methods'-5 are generally used for its deter- mination.Several methods using gas and liquid chromato- graphy also have been reported for the determination of aldehydes, including acrolein,6-s but these methods are usually limited in either sensitivity or selectivity. Oxime derivatives used in the gas chromatographic deter- mination of carbonyls are rnethoximes,9JO benzyloximes,llJ2 p-nitrobenzyloximesll and pentafluorobenzyloximes. 13-15 For these methods, except for pentafluorobenzyloximes, flame- ionisation or nitrogen-specific detector systems are used. Although a gas chromatographic method based on the bro- mination of acrolein has been reported in earlier papers,l6,17 the brominated product of acrolein is unstable and the reproducibility of the determination is poor. This paper reports a method for the sensitive and selective determination of acrolein involving the bromination of acrolein O-methylox- ime and gas chromatography with electron-capture detection. Experimental Reagents and Materials Methoxylamine hydrochloride (MOA.HC1) (Wako Pure Chemical Industries, Osaka, Japan) was dried under reduced pressure. Other reagents used were of analytical-reagent grade.A standard solution of acrolein was prepared by dissolving 100 mg of the purest grade of acrolein available in distilled water and diluting to 100 ml. The brominated derivative of acrolein 0-methyloxime was supplied by Tokyo Kasei Kogyo (Tokyo, Japan), and the Sep-PAK CI8 (SP) cartridge was from Waters Associates (Milford, MA, USA). Apparatus and Conditions A Hitachi 073 gas chromatograph (GC) with a 63Ni electron- capture detector (ECD) and a Hitachi 663 GC with a flame- ionisation detector (FID) and a flame thermionic detector (FTD) were used.The following conditions were used for the GC with ECD: a 2 m glass column packed with 3% silicone GE XE-60 on 60-80 mesh Chromosorb W AW DMCS; column temperature, 90 "C; injection and detector tempera- ture, 170 "C; and carrier gas (nitrogen) flow-rate, 40 ml min-1. The GC with FID conditions were as follows: 1, a 2 m glass column packed with 20% TCP on 60-80 mesh Chromosorb W AW DMCS; a column temperature of 80 "C; an injection and detector temperature of 120 "C; and a carrier gas (nitrogen) flow-rate of 40 ml min-1; 2, a 2 m glass column packed with 10% DEGS on 60-80 mesh Chromosorb W AW DMCS; a column temperature of 140 "C; an injection and detector temperature of 170 "C; and a carrier gas (nitrogen) flow-rate of 40 ml min-1.The GC with FTD conditions were the same as the GC-FID conditions except for the carrier gas, which was helium, also with a flow-rate of 40 ml min-1. A Hitachi M52S GC - MS was used with a 10-20 eV ionisation energy. General Procedure A 1 ml aliquot of 2 M sodium acetate and 1 ml of MOA.HC1 (5 mg ml-1) are added to 5 ml of sample solution containing about 0.3 pg of acrolein. The mixture is allowed to stand for 10 min at room temperature and then 1 ml of 1.5 M sulphuric acid, 0.2 ml of 0.2 M potassium bromate and 2 g of potassium bromide are added and dissolved with stirring. After standing for 15 min at room temperature, the excess of bromine is reduced with 0.05 M sodium thiosulphate.The solution is forced through an SP cartridge and the derivative in the cartridge is eluted with 1.5 ml of diethyl ether. A 4 pl portion of the eluate is measured by GC with ECD and the peak-height method is used for the determination of acrolein. Results and Discussion Derivatisation Reaction Aldehydes are known to react with MOA to form O-methyl- oxime.10 In this work, we investigated a method based on the bromination of acrolein 0-methyloxime, followed by GC with ECD to determine acrolein with a high sensitivity and selectivity. It is assumed that the reaction proceeded as follows: CH,--CH-CHO + CH3-O-NH2 ---* CHFCH-CH=N-O-CH~ + H20 CH&H-CH=N-O-CH3 + Br2 + CH2-CH-CH=N-O-CH3 I 1 Br Br46 ANALYST, JANUARY 1987, VOL.112 !L 1 0 5 1 0 0 5 1 0 tR/min Fig. 1. Chromatograms of acrolein and its derivatives by GC-FID. 20% TCP: column temperature, 80 "C; carrier gas (N2), 40 ml min-1. 10% DEGS: column temperature, 140 "C; carrier gas (NJ, 40 ml min-1. (a Pre-reaction, A = acrolein; (b) 0-methyloxime derivative, A = acetic acid, B = brominated derivative of acrolein - MO A = acro r' ein - MO; and (c) brominated derivative of 0-methyloxime, r L DEGS L I I I I 0 5 0 5 rR/min Fig. 2. Chromatograms of acrolein and its derivatives by GC-FTD. 20% TCP: column temperature, 80 "C; carrier gas (He), 40 ml min-l. 10% DEGS: column tem erature, 140 "C; carrier gas (He), 40 ml min-l. (a) Pre-reaction;hby 0-methyloxime derivative, A = acrolein - MO; and (c) brominate derivative of 0-methyloxime, A = bromi- nated derivative of acrolein - MO In order to identify the reaction procedure described above, the solutions extracted with diethyl ether ( i e ., the pre- reaction solution, the solution containing 0-methyloxime and that containing the brominated derivative of 0-methyloxime) were measured by GC-FID and GC-FTD (Figs. 1 and 2). (b) 164 50 100 150 m/z 177 245 200 250 Fig. 3. Mass spectrum of acrolein - MO.Br derivative. (a) Bromi- nated derivative of acrolein 0-methyloxime and ( b ) synthesised 2,3-dibromopropionaldehyde 0-methyloxime I 1 I I 0 2 4 6 Amount of MOA.HCl/mg Fig. 4. Effect of amount of MOA on the formation of acrolein - MO. Acrolein, 1.0 pg; sodium acetate (2 M), 1.0 ml 0.5 1.0 1.5 2.0 Amount of 1.5 M H2SOdjml Fig. 5. Effect of volume of 1.5 M H2S04 on bromination of acrolein - MO. Acrolein, 1.0 pg; MOA.HC1, 5 mg On the chromatograms obtained by GC-FID, the acrolein peak [peak A in Fig.1 (a)] disappeared and two new peaks [Fig. 1 (b)] appeared with the formation of 0-methyloxime. Furthermore, in Fig. 1 (c), two of the peaks decreased and a new peak (peak B) appeared. This new peak is assumed to be caused by the bromination of O-methyloxime. Peak A in Fig. 1 (c) was identified as acetic acid by GC - MS measurement. On the chromatograms obtained by GC-FTD (Fig. 2), the acrolein peak did not appear, but two large peaks appeared as the O-methyloxime derivative was formed. Then, as can be seen in Fig. 2 (c), these peaks disappeared and a new large peak appeared as bromination took place.From these results, it is assumed that the nitrogenous compound was formed and converted to different nitrogenous compounds by bromination. The two peaks of 0-methyloxime may be due to syn- and anti-isomers. The peaks resulting from the brominated derivative of 0-methyloxime could not be separated under the proposed conditions. The extent of the reaction, determined by GC-FID from those compounds remaining after each reaction period, was about 92% for 1 mg of acrolein in 5 ml of aqueous solution. The mass of the brominated derivative of acrolein O-methyl- oxirne is shown in Fig. 3. The highest peak (mlz 245) of theANALYST, JANUARY 1987, VOL. 112 47 molecular ion peaks (mlz 243,245 and 247) appeared with an ionisation energy of 10 eV. The fragment peaks are assumed to be (M - Br) at mlr 164 and 166, and (M - 2Br) at mlz 85. The mass spectrum agreed with that of the standard 2,3-dibromo- propionaldehyde 0-methyloxime obtained from Tokyo Kasei Kogyo, Tokyo, Japan.As a result, the brominated derivative of acrolein O-methyl- oxime is assumed to be 2,3-dibromopropionaldehyde 0-methyloxime. 0-Me thy loxime For ma tion Levine et al. have reported the formation of 0-methyloxime in methanol solution.10 In this work, the formation of acrolein 0-methyloxime in aqueous solution was studied in order to be able to dissolve potassium bromide adequately in the second step. Fig. 4 shows the effect of MOA.HC1 in sodium acetate buffer solution. The yield of the product was maximum and constant in the range 4-6 mg of MOA.HC1. The temperature of the reaction had no influence, even at room temperature (15-20 "C), 40 and 60 "C.The minimum time for the completion of the reaction was found to be 10 min. Bromination of 0-Methyloxime The effect of acidity on the bromination of acrolein O-methyl- oxime is shown in Fig. 5. The yield was maximum in the range 0.8-1.0 ml of 1.5 M sulphuric acid, corresponding to pH 1.3-1.7. The peak height was constant in the range 2 4 g of KBr, and the yield was constant in the range 0.1-2.0 ml of 0.2 M potassium bromate. The minimum reaction time for complete bromination was found to be 10 min after the addition of the reagents. The derivative was stable for at least 5 d. Use of Sep-PAK CI8 Cartridge The brominated derivative of O-methyloxime was completely extracted with one 5 ml aliquot of diethyl ether.An SP 0 10 tRlmin Fig. 6. Typical chromatogram of rain water by GC-ECD. A = Acrolein derivative Table 1. Retention time of derivatives Retention time/ Relative Parent compound min retention time Acrolein . . . . 7.39 1 .oo Methacrolein . , 5.39* 0.73 6.951. 0.94 Crotonaldehyde . . 9.07 1.23 * Peak of brominated derivative of methacrolein. t Peak of brominated derivative of methacrolein 0-methyloxime. cartridge was used to enrich and clean up the acrolein derivative, and the following procedure was established. Pump the sample solution after reaction through the cartridge with a 10 ml syringe. Remove the cartridge from one syringe to another and pump 1.5 ml of diethyl ether through to elute the acrolein derivative. The recovery of the derivative obtained from the cartridge using 0.2 pg of acrolein was 98%.Selection of GC Column In order to separate the acrolein peak from those compounds with ethylenic bonds, such as methacrolein and crotonal- dehyde, five different columns were tested, namely, 1% PEG-HT, 10% TCEP, 10% DEGS, 3% silicone GE XE-60 and 5% silicone OV-225. As a result, it was found that the 3% silicone GE XE-60 column was the most efficient and effective for the separation of the peaks. The retention times are shown in Table 1. It was found by GC - MS that two peaks (5.39 and 6.95 min) of methacrolein were of the brominated derivative of methacrolein and the brominated derivative of meth- acrolein 0-methyloxime, respectively. This result shows that the 0-methyloxime formation from methacrolein had not proceeded to completion.Calibration Graph and Precision A calibration graph was prepared with an acrolein standard solution under the optimum experimental conditions. The relationship between peak height and the concentration of acrolein in aqueous solution was linear over the range 0-0.06 yg ml-1. The relative standard deviation from five replicates was estimated to be 4.5% for 0.04 yg ml-1 of acrolein. The detection limit (signal to noise ratio = 2) was 0.4 ng ml-1 of acrolein. Recovery of Acrolein from Rain Water In order to assess the proposed method, recovery experiments were carried out on mixtures of rain water and acrolein standard solutions. The results obtained are shown in Table 2. The proposed method can be satisfactorily applied to the determination of acrolein in rain water, as the recovery obtained was in the range 90-101%.Table 2. Recovery of acrolein from rain water Acrolein Rain water sample A . . . . . . . . B . . . . . . . . * ND = Not detected. Added PH Pg - 4.5 0.050 0.200 0.050 0.200 - 4.8 Found Recovery, Pg Yo ND* - 0.045 90 0.202 101 ND - 0.047 94 0.192 96 Table 3. Results obtained for the determination of acrolein in rain water Acrolein*/ng ml-1 Range Rain water sample Mean A-1 . . . . . . 1.8 1.5-2.0 A-2 . . . . . . NDT A-3 . . . . . . 2.8 2.5-3.1 A-4 . . . . . . 2.2 2.1-2.2 B-1 . . . . . . ND * Mean of three determinations. t ND = Not detected.48 ANALYST, JANUARY 1987, VOL. 112 Determination of Acrolein in Rain Water Acrolein was determined in rain water samples by the proposed method. Table 3 shows the results obtained, and Fig.6 shows a typical chromatogram of acrolein in rain water. Conclusion An improved method for the determination of acrolein in aqueous solutions, based on the bromination of 0-methylox- ime followed by GC with ECD, has been established. This method is highly sensitive and selective for the determination of acrolein in rain water compared with previous methods. We thank Dr. H. Tsukube, Okayama University, for his instructive advice and are grateful to Dr. K. Hojo, Tokyo Kasei Kogyo, for supply of standard materials. 1. 2. 3. 4. References Cohen, I. R., and Altshuller, A. P., Anal. Chem., 1961, 33, 726. Japanese Industrial Standard, JIS K 0089, 1983. Alarcon, R. A., Anal. Chem., 1968, 40, 1704. Suzuki, Y., Imai, S . , and Hamaguchi, A., Bunseki Kagaku, 1979, 28,445. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. Suzuki, Y., and Imai, S . , Anal. Chirn. Acta, 1982, 136, 155. Saito, T., Takashina, T., Yanagisawa, S . , and Shirai, T., Bunseki Kagaku, 1983, 32, 33. Swarin, S. J., and Kipari, F., J. Liq. Chromatogr., 1983,6,425. Kuwata, K., Uebori, M., Yamasaki, H., Kuge, Y., and Kiso, Y., Anal. Chem., 1983, 55, 2013. Fales, H. M., and Luukkainen, T., Anal. Chem., 1965,37,955. Levine, S . P., Harvey, T. M., Waeghe, T. J., and Shapiro, R. H., Anal. Chem., 1981, 53, 805. Magin, D. F., J. Chrornatogr., 1979, 178, 219. Magin, D. F., J. Chromatogr., 1980, 202, 255. Nambara, T., Kigasawa, K., Iwata, T., and Ibuki, M., J. Chromatogr., 1975, 114, 81. Kobayashi, K., Tanaka, M., and Kawai, S . , J . Chrornatogr., 1980, 187, 413. Nishikawa, H., Takahara, Y., Mori, H., and Hayakawa, T., J. Jpn. SOC. Air Pollut., 1984, 19, 387. Nishikawa, H., Hayakawa, T., and Ikeda, S., J. Chromatogr., 1986, 351,566. Nishikawa, H. , and Hayakawa, T., Bunseki Kagaku, 1985,34, 729. Paper A61185 Received June 6th, 1986 Accepted August 5th, 1986

ISSN:0003-2654    

DOI:10.1039/AN9871200045

出版商:RSC    

年代:1987

数据来源: RSC

摘要:

ANALYST, JANUARY 1987, VOL. 112 49 Simple Gas Chromatographic Determination of the Distribution of Normal Alkanes in the Kerosene Fraction of Petroleum Suresh C. Vishnoi, Shiv D. Bhagat, Vidya B. Kapoor, Sneh K. Chopra and Rajamani Krishna Indian Institute of Petroleum, Dehra Dun 248005, India An internal standard technique has been applied to the determination of the distribution of normal alkanes in the kerosene fraction of petroleum using a capillary column. The applicability of packed columns for such a determination has also been studied and compared with the existing Universal Oil Products (UOP) method. Keywords: Normal alkanes determination; subtractive gas chromatography; internal standards technique; molecular sieves The determination of the concentration and concentration distribution of normal alkanes in hydrocarbon mixtures is of considerable importance in the petroleum and petrochemicals industries.Normal alkanes from petroleum sources are an important feed stock for the petrochemical industries; the long chain alkanes can be converted to lubricant and fuel additives, plasticisers, industrial surfactants, flotation agents, solvents and raw materials for protein synthesis by means of oxidation, halogenation, esterification, fermentation, etc. Such wide applications have generated a new interest in the refinery processes for recovering long chain alkanes from petroleum. Flow properties, such as viscosity, viscosity index, fluidity, pour point, etc., of heavy petroleum fractions largely depend on the n-alkane content.The distribution of n-C11-n-C14 alkanes obtained from the kerosene fraction has immense potential in the manufacture of biodegradable detergents. In view of this importance, various methods and techniques have been proposed to determine the distribution of n-alkanes in the kerosene fraction of petroleum. The determination of n-alkanes in complex hydrocarbon systems by their selective adsorption on molecular sieve 5A was suggested by Brenner and Coats1 as early as 1958. Since then the molecular adsorption technique has been invariably used by several workers2.3 in spite of its limitations. The mechanism of selective adsorption and the structure and properties of molecular sieves have been discussed in detail by many workers. Nelson4 determined the n-alkane content of petroleum distillates by calculating the difference in mass of the zeolite before and after adsorption.O'Connor and co-workers596 suggested recovering adsorbed n-alkanes from the sieve by a diffusion-controlled process for quantitative determination. Larson and Becker7 used volumetric tech- niques for the determination of n-alkanes in olefin-free petroleum distillates. Whithams-9 used a subtractive method using a conventional GLC column with and without a molecular sieve and the n-alkanes were determined by the difference between the two chromatograms. These methods, however, were inadequate for low concentrations of n-alkanes. Eggertsen and Groen- ningsl" and later Blytas and Peterson11 modified this method so that n-alkanes were desorbed from the molecular sieve by heating and were then determined on a GLC column.Hydrofluoric acid followed by isooctane extraction12.13 was used for the recovery of adsorbed n-alkanes by the destruction of the molecular sieve. Petrovic and Vitorovicl3 reported the direct gas chromatographic determination of C9-Ct4 n-alkanes in the kerosene fraction using an open tubular column of Apiezon L. Hine" used an open tubular column for the determination of the total n-alkane content in petroleum fractions. Johanson et al. 16 described the possibility of determining hydrocarbons by structural group types in gaso- line and distillates. There is no standard analytical method for the determina- tion of n-alkanes in the kerosene fraction, except for the Universal Oil Products (UOP) method.17 This method is based on a subtraction technique using two gas chromato- graphs in series separated by a molecular sieve column, but has certain inherent limitations.In this paper we propose a simple and straightforward capillary gas chromatographic method for the determination of the n-alkane distribution in straight-run kerosene fractions. The method makes use of the high resolution capability of an open tubular column (WCOT) to separate the n-alkanes from branched components and an internal standard technique18 for fast, reliable and accurate determinations. A simplified procedure is also discussed utilising the applicability of packed columns for such determinations and is suggested as an alternative to the UOP method.17 Experimental Two gas chromatographs with flame-ionisation detectors were employed, one for the capillary and one for the packed- column studies.The former was a Varian gas chromatograph (Model 3700) with a chromatographic data system (CDS-111) and potentiometric recorder (Model 9176). The provision to install the capillary column was used in order to achieve the separation of individual n-alkanes from branched peaks. A fused-silica open tubular column of 50 m X 0.2 mm with methylsilicone phase of 0.2 pm film thickness was used. The injector and detector blocks were set at 300 and 320 "C, respectively, and the column was programmed from 85 to 250 "C at 4 "C min-1 with 4 min of initial hold-up time. Nitrogen was used as a carrier gas at an average linear velocity of 18.5 cm s-1, corresponding to a flow-rate (uncorrected) of 1.5 ml min-1.A 0.1-pl sample was injected with a split ratio of 70 : 1. The second gas chromatograph (Perkin-Elmer Model 3920 B) was installed with a 3 m x 2 mm i.d. packed column of 5% SE-30 (Methyl E-301) on Chromosorb P, 80-100 mesh. The injector and detector temperatures were kept at 300 and 320 "C, respectively. The initial column temperature was 45 "C and it was temperature-programmed at a rate of 4 "C min-1, with an initial hold-up time of 8 min, to a final temperature of 220 "C. Nitrogen was used as the carrier gas with flow-rate of 30 ml min-1, and a 0.2-p1 sample was used for the determination. A Spectra-Physics minigrator and recorder were used for computing the data. High purity n-alkanes were used to prepare reference blends and internal standard samples.A de-normalised reference stock was prepared from kerosene samples in two stages for making the calibration blends. The kerosene sample was subjected to urea adduction and the last traces of n-alkanes were removed by molecular sieve adsorp-50 ANALYST, JANUARY 1987, VOL. 112 tion.19 The calibration blend was prepared by mixing a known amount of de-normalised reference stock with a pure n-alkane reference blend. Table 1. n-alkane concentration of 29.976%. The reproducibility of individual n-alkane concentrations from run to run is shown in The peak subtraction technique using a molecular sieve column was applied in order to investigate whether the n-alkanes were masked by branched alkanes. The technique Results and Discussion can quantify the extent of contribution of branched to the The Proposed method is based on a wall-coated open tubular n-alkanes and the concentration of individual n-alkanes can column of SE-30 (methylsilicone), which has the best solvent therefore be determined with high precision.The subtraction Characteristics of a non-Polar Phase for separating complex was achieved using a Linde molecular sieve 5A packed in the hydrocarbon mixtures according to boiling Points. Some quartz liner of the split injector of the gas chromatograph. The properties of this capillary column were determined in order molecular sieve was activated in the injector port itself by to show the efficiency of the stationary phase. The number of heating at 300-350 "c. A section of the subtracted and theoretical Plates of the capillary COhmn used was 254 X lo3 unsubtracted chromatograms obtained using the capillary with a coating efficiency of 69.8% and a capacity ratio of 5.1 column is shown in Fig.2; it was noticed that the number of for n-tridecane- The separation mmber (Trennzahl) for branched alkanes obscured by n-alkanes was negligible owing consecutive pairs of n-CI3 and n-C14 was found to be 49.3. This to the high resolution of the capillary column. has been used as a ~ X W N ~ of Column efficiency under Packed-column investigations using a single gas chromato- temperature-programmed conditions and gives the maximum graph with a flame-ionisation detector appear to be promising number of peaks that can be separated between two sequential as an alternative to the uop method.Data from individual homologues. n-alkanes in the same kerosene sample were obtained using a Fig. 1 shows a typical chromatogram of a kerosene sample packed column with internal standards. The total concentra- obtained under the operating conditions outlined above. Each tion of n-alkanes in the sample (Table 2) was 31.372y0, which n-alkane has been well separated from the neighbouring was about 1.396% higher than the value reported when using branched components by utilising the high resolving power of the fused-silica capillary column. This higher value was the capillary column. TWO different chromatograms of the expected, owing to the limitations of the packed column in sample were recorded; one as above and the other with resolving n-alkane peaks from the branched-chain corn- n-hexadecane added as an internal standard.Area and ponents. The concentration of branched components base-line sensitivity parameters were taken into account for obscured by n-alkanes was calculated using chromatograms of the accurate peak detection using the CDS-111 data system. the sample obtained with and without molecular sieve The values obtained for individual n-alkane concentrations subtraction (Fig. 3). The deviation in the values obtained for (obtained by area normalisation and internal standard tech- total n-alkane concentration in the kerosene sample using niques) were found to be in good agreement. The COncentra- packed and capillary columns does not exceed the error tion of n-CI6 in the original kerosene sample was determined expected in routine GC determinations.by the area normalisation technique for calculating the individual concentration of n-hexadecane. The precision of the above proposed internal standard method was examined by determining five replicate injections of the sample and the standard deviation and coefficient of variation were found to be 0.4726 and 1.5766, respectively, with an average total c17 Y C9 c18 1 I I I I I I I I 36 32 28 24 20 16 12 8 4 Ti me/m i n Fig. 2. column (a) without molecular sieve and (b) with molecular sieve Section of chromatogram of kerosene sample on capillary Fig. 1. Kerosene sample on fused-silica capillary column Table 1. Repeatability and precision in capillary-column method Concentration, 70 m/m n- Alkane components 1 2 3 4 5 C-8 .. . . . . . . 0.28 0.26 0.29 0.29 0.30 C-9 . . . . . . . . 0.84 0.85 0.88 0.88 0.91 C-10 . . . . . . . . 2.42 2.45 2.52 2.52 2.57 C-11 . . . . . . . . 5.09 5.25 5.14 5.15 5.00 C-12 . . . . . . . . 6.31 6.20 6.37 6.39 6.51 C-13 . . . . . . . . 6.00 5.89 6.06 6.10 6.21 C-14 . . . . . . . . 4.53 4.45 4.58 4.66 4.68 C-15 . . . . . . . . 2.28 2.31 2.37 2.40 2.44 C-16 . . . . . . , , 0.78 0.79 0.81 0.82 0.84 C-17 . . . . . . . . 0.36 0.35 0.37 0.37 0.38 C-18 . . . . . . . . 0.32 0.31 0.33 0.33 0.34 C-19 . . . . . . . . 0.19 0.19 0.20 0.20 0.21 C-20 . . . . . . . . 0.15 0.14 0.15 0.16 0.16 Total . . . . . . . . 29.55 29.44 30.07 30.27 30.55 Average, 70 mlrn 0.2840 0.8720 2.4960 5.1260 6.3560 6.0520 4.5800 2.3600 0.8080 0.3660 0.3260 0.1980 0.1520 29.976 Standard deviation, % mlm 0.0152 0.0277 0.0602 0.0913 0.1135 0.1186 0.0946 0.0652 0.0239 0.0114 0.0114 0.0084 0.0084 0.4726 Coefficient of variation, % 5.3401 3.1822 2.4138 1.7805 1.7856 1.9600 2.0655 2.7624 2.9548 3.1152 3.4975 4.2256 5.5043 1.5766ANALYST, JANUARY 1987, VOL.112 51 Table 2. Repeatability and precision in packed-column method Concentration, YO m/m n- Alkane components 1 2 3 4 5 C-8 . . . . . . . . 0.27 0.29 0.31 0.30 0.33 C-9 . . . . . . . . 0.89 0.93 0.98 1.01 1.05 c-10 . . . . . . . . 2.49 2.56 2.63 2.67 2.70 C-11 . . . . . . . . 5.13 5.22 5.24 5.32 5.40 C-12 . . . . . . . . 6.38 6.31 6.53 6.44 6.60 C-13 . . . . . . . . 6.08 6.2 6.19 6.38 6.38 C-15 . . . . . . . . 2.59 2.61 2.60 2.72 2.75 C-16 . . . . . . . . 0.87 0.89 0.90 0.83 0.86 C-17 .. . . . . . . 0.54 0.51 0.52 0.49 0.51 C-18 . . . . . . . . 0.29 0.28 0.31 0.30 0.31 C-19 . . . . . . . . 0.24 0.21 0.23 0.23 0.25 C-20 . . . . . . . . 0.13 0.14 0.16 0.14 0.15 Total . . . . . . . . 30.58 30.90 31.44 31.72 32.22 C-14 . . . . . . . . 4.68 4.75 4.84 4.89 4.93 Average, Yo mim 0.3000 0.9720 2.6100 5.2620 6.4520 6.246 4.818 2.654 0.87 0.514 0.2980 0.2320 0.1440 31.372 Standard deviation, O/O mlm 0.0224 0.0634 0.0851 0.1026 0.1157 0.1310 0.1022 0.0750 0.0274 0.0181 0.0130 0.0148 0.0114 0.65093 Coefficient of variation, YO 7.4536 6.5230 3.2623 1.9492 1.7917 2.098 2.1217 2.8272 3.1478 3.5300 4.3753 6.3933 7.9179 2.075 Table 3. Accuracy and coefficient of variation in packed-column method Actual Observed concentration, YO m/m Coefficient n- Alkane concentration, of variation, components % mtm 1 2 3 4 5 YO Accuracy, YO C-11 .. . . . . . . 4.18 4.25 4.35 4.4 4.1 4.3 2.69 2.39 C-12 . . . . . . . . 4.80 4.90 4.94 5.0 4.75 4.88 1.89 1.87 C-13 . . . . . . . . 3.87 4.0 3.95 3.85 4.15 4.10 2.98 3.61 C-14 . . . . . . . . 2.08 2.10 2.15 2.2 2.3 2.22 3.44 5.48 c-15 . . . . . . . . 0.75 0.72 0.77 0.70 0.67 0.75 5.40 4.50 Total . . . . . . . . 15.68 15.97 16.16 16.15 15.97 16.25 0.78 2.64 a ) I “L, Fig. 3. sieve and ( b ) with molecular sieve Kerosene sample on packed column (a) without molecular In order to check the precision and accuracy of the packed- column method, a high purity calibration blend was prepared by weighing portions of a de-normalised kerosene reference stock and a blend of pure n-alkanes from n-undecane to n-pentadecane.The calibration blend was determined on the packed column and the concentration of each n-alkane was calculated and compared with the actual values. The agree- ment between the actual and observed concentrations was found to be between 1.875 and 5.48%, as shown in Table 3, and the coefficient of variation was 0.78%. The proposed packed-column method has many advantages over the UOP method. 17 The flame-ionisation detector that was used in place of the thermal-conductivity detector, apart from being more sensitive, has the distinct advantage of having the same quantitative response to equal masses of any hydrocarbon, thus avoiding having to account for the response factors of individual components. Peak broadening, which may be caused by the use of a transfer line in the UOP method,” is eliminated by using a single gas chromatograph.Also, the optimisation of the detector current in both gas chromatographs in order to match the detector signals for the relative distribution of an isoalkane blend is not required in the proposed method. The total n-alkane concentration depends on the factor used for the conversion of area percent. to mass percent. in the non-distributive mode. Any error in preparing the calibration blend will affect the total concentra- tion of n-alkanes obtained for the sample. As the sample is injected twice in the proposed method (with and without the molecular sieve) the accuracy in a sample injection of 0.2 pl is well within acceptable limits. Conclusion The proposed internal standard method €or the determination of normal alkanes using a fused-silica capillary column is both fast and reliable.The proposed method using a packed column is simpler and more sensitive than the existing UOP method17 for the determination of normal alkanes and gives reasonable accuracy compared with an open tubular column. References 1. 2. 3. Brenner, N., and Coates, V. J., Nature, (London), 1958, 181, 1401. Barall, E. M., and Bauman, F. J . , J . Gas Chromatogr., 1964,2, 256. Beroza, M., and Insco, M. N., in Ettre, L. S . , and McFadden, W. H., Editors, “Ancillary Technique of Gas Chromato- graphy,” Wiley-Interscience, New York, 1969, pp. 89-114 and Nelson, K. H., Anal. Chern., 1957,29, 1026. O’Connor, J. G., and Norris, M. S . , Anal. Chem., 1960, 30, 701. O’Connor, J. G., Burrow, F. H., and Norris, M. S . , Anal. Chem., 1962,34, 83. 127-130. 4. 5. 6.52 ANALYST, JANUARY 1987, VOL. 112 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. Larson, L. P., and Becker, H. C., Anal. Chem., 1960,32,1215. Whitham, B. T., Nature (London), 1958, 182, 392. Whitham, B. T., Nature (London), 1961, 192, 966. Eggertsen, F. T., and Groennings, S . , Anal. Chem., 1961,33, 1147. Avondale, PA, 1966. Blytas, G. C., and Peterson, D. L., Anal. Chem., 1967, 39, 1434. Knight, H. S . , Anal. Chem., 1967, 39, 1452. Brunnock, J. V., Anal. Chem., 1966, 8, 1648. Petrovic, K., and Vitorovic, D., J . Chromatogr., 1972,65, 155. Hine, B. T., Chromatographia, 1984, 18, 679. Johanson, N. G., Ettre, L. S . , and Miller, R. L., J. Chromat- ogr., 1983, 256, 393. 17. “Normal Paraffins by Subtractive Gas Chromatography,” UOP VO8-411-75, Universal Oil Products Inc., Des Plaines, IL, 1975. “Application Laboratory Report,” No. 1005, Hewlett Packard, Chen, N. Y., and Lucki, S. J., Anal. Chem., 1970, 42, 508. 18. 19. Paper A612 76 Received August 13th, 1986 Accepted August 20th, 1986

ISSN:0003-2654    

DOI:10.1039/AN9871200049

出版商:RSC    

年代:1987

数据来源: RSC