摘要:

Determination of Trace and Ultra-trace Elements in Saline Waters by Inductively Coupled Plasma Mass Spectrometry After Off-line Chromatographic Separation and Preconcentration Journal of Analytical Atomic Spectrometry KYM E. JARVIS AND JOHN G. WILLIAMS NERC ICP-MS Facility Centre for Analytical Research in the Environment Imperial College Silwood Park Ascot Berkshire UK SL5 7TE ELENA ALCANTARA Department of Civil Engineering Imperial College South Kensingron London S W7 UK JULTAN D. WILLS Groupe de Geochimie URA-CNRS Dl 767 Place Eugene Bataillon F-34090 Montpellier CEDEX 5 France A method for the determination of trace and ultra-trace elements in sea-water by ICP-MS after preconcentration and matrix removal using a commercially available ion-chelation system was developed and optimized.Cellulose-immobiliz,ed ethylenediaminetriacetic acid was used as the column material and elution profiles were determined with an on-line configuration. Recovery experiments were carried out firstly in a simple matrix (spiked de-ionized water) and secondly in a complex matrix (synthetic sea-water) employing a preconcentration factor of either 10 or 100. In both types of matrix recoveries of between 90 and 110% were obtained for Cd Co Cu Ni Pb U Y and the 14 REEs. Excess recoveries were obtained for Zn and are thought to result from reagent contamination. Reproducibility was in the range 0.20-7.25% coefficient of variation. Optimized conditions were used to analyse National Research Council of Canada CRMs Estuarine Water SLEW-1 Open Ocean Sea-water NASS-4 and Nearshore Sea-water CASS-2.Accuracy and precision are related to concentration. Cd Co Cu and Ni show good accuracy for concentrations greater than 1 ng m1-I. Both short term instrumental precision and between sample reproducibility are typically better than 10% RSD. Although the rare earth elements are not certified in any of the reference materials a relatively smooth sea-water normalized plot is seen for SLEW-1. Keywords Inductively coupled plasma mass spectrometry; preconcentration techniques; saline waters The accurate determination of trace and ultra-trace elements in sea-water is one of the most important and challenging tasks in analytical chemistry. Heavy metals Cd Cu Ni Pb and Zn and actinide elements Th and U are often required for routine monitoring of marine environmental pollution and for ocean modelling programmes.In particular the REEs are used as tracers in the study of submarine volcanism and also to determine the degree to which riverine water influences the sea-water-estuarine mixing zone.' In reviewing the analytical techniques currently used for inorganic trace analysis ICP-MS initially appears to be one of the most appropriate for ultra-trace determination in saline solutions. The technique offers excellent sensitivity and very low detection limits (ng 1-' and pg 1-I). It is a very rapid multi-element technique (in principle 70 elements can be determined in < 2 ml of sample solution in < 2 min) and offers linear calibration ranges over at least five orders of magnitude. During the last ten years it has become one of the most widely used techniques for the determination of REEs and other trace and ultra-trace elements in a wide variety of matrices.2 However the high salt concentration of sea-water makes it a difficult matrix both physically and chemically for analysis by ICP-MS.The high dissolved-solids loading of sea-water ( M 1-3%) results in matrix deposition on the torch injector spectrometer interface and ion lenses often leading to severe degradation of sen~itivity.~ In addition the analyte response is suppressed by major components of the ample.^-^ A simple remedy is to dilute the sample; however this decreases the already extremely low analyte levels present in naturally occur- ring saline waters in some cases to an undetectable level.The nature of the sea-water matrix also causes a number of polyatomic ion interferences (e.g. 40Ar23Na+ on 63Cu) which can considerably degrade the analytical capabilities of the instrument. Clearly separation of the matrix before analysis by ICP-MS is desirable. In some cases a matrix removal step prior to analysis is sufficient to allow quantitative deter- mination of the analytes. However in others cases such as the analysis of open ocean sea-water the levels of most analytes are below ICP-MS limits of detection. Hence a preconcentration step together with matrix removal is essential. This study deals with the optimization of a method for the simultaneous determination of REEs and transition elements in sea-water by ICP-MS after off-line automated chelation preconcentration.Cellulose-immobilized EDTrA (ethylenedia- minetriacetic acid) was investigated as a chelating agent. The preconcentration method developed is simple and straightfor- ward to carry out and combined with ICP-MS can provide considerable information about the trace and ultra-trace composition of saline solutions. INSTRUMENTATION A commercially available chromatography system (Tracecon Knapp Logistik Automation A-8042 Graz Austria) was used. Fig. 1 shows a schematic of the unit and the relationship between the analytical column pumps and valves switched during the separation and preconcentration cycle. Each of the pinch valves and peristaltic pumps can be individually manipu- lated by the system control software. The chromatography unit was supplied with a solid phase of 0.25mg EDTrA on fibrous cellulose held in a disposable syringe.This functional Journal of Analytical Atomic Spectrometry October 1996 Vol. 11 (91 7-922) 91 7Table 2 ICP-MS instrument operating conditions i M eluent waste Eluent in Sample Fig. 1 Schematic of separation and pre-concentration unit P peristal- tic pump; V valve; B bubble sensor; and C analytical column containing cellulose-immobilized ethylenediaminetriacetic acid ( EDTrA) group has been found to give excellent preconcentration characteristics for the REEs as well as for a range of transition metals.' The control software allowed the daily calibration of all three pumps in order to determine the sample and reagent volumes accurately. Capacitive bubble sensors were employed to prevent the analytical column from running dry.Chelating agent in an immobilized form was packed into a column conditioned and the sample solution loaded. The analytes were complexed by the chelating agent while the sample passed through the solid phase. To make this possible the sample and the chelating agent were buffered to a pH suitable for the formation of complexes between the analytes (divalent or trivalent metal cations e.g. REEs transition elements U and Th) and the chelating functional group (EDTrA). The effect of sample solution pH on the enrichment was studied by Schramel et aL8 and pH 4 was found to be optimum for the quantitative recovery of trace metals. The remaining matrix (monovalent and some divalent ions e.g. Na Mg Ca C1) was either not retained in the column and was sent directly to waste or was weakly chelated and subsequently removed.by a buffer wash. By doing this the necessary matrix removal step was achieved. A small volume of eluent acid was then passed through the column releasing the metal cations from the solid phase and was either sent for direct analysis to the ICP-MS unit (on-line operation) or was collected for later analysis (off-line oper- ation). The preconcentration factor achieved was the ratio of initial sample volume to final eluent volume collected. The time taken to complete each step of the procedure is shown in Table 1. The major time-consuming step was the loading of the sample solution onto the column particularly if a large preconcentration factor was required.Separated samples were analysed using a PlasmaQuad PQ2 STE (VG Elemental Winsford Cheshire UK). Operating conditions for routine analysis were employed and are shown in Table 2. Residual Na Ca and Mg were determined by ICP- AES (JY 24 monochromator system; Instruments SA UK Stanmore Middlesex UK) using the Mg 285.213nm Ca 422.673 nm and Na 589.592 nm spectral lines. Plasma conditions- Instrument Forward power/W Reflected power/W Coolant gas flow/l min-' Nebulizer gas flow/l min-' Auxiliary gas flow11 min-' Znte$ace- Sampling cone Skimmer cone Nebulizer- Type Solution up-take rate/ml min-' Data acquisition parameters- Dwell time per channelfps Number of channels Total measurement time/s Scan range/m/z Skipped mass regions/m/z PlasmaQuad STE 1350 < 5 14 0.80 0.5 Ni 1 mm orifice Ni mini-skimmer 0.7 m de Galan V-groove 0.5 160 20 per mass unit 60 23.6-239.4 27.4-42.6 79.4-80.6 EXPERIMENTAL In order to check the validity of the method two dim of experiments were carried out.First elution prof plotted to determine the necessary volume of eluent to separate the elements of interest and secondly experiments were performed to assess the capabilii column to preconcentrate the trace elements quans The accuracy of the method was tested by analysis National Research Council Canada (NRCC Ottawa CRMs Estuarine Water SLEW-1 Open Ocean 2 NASS-4 and Nearshore Sea-water CASS-2. Reagents Standards and Test Solutions Nitric acid acetic acid and ammonia solution 'Aristar' grade (Merck Lutterworth Leicestershire 1 moll-' ammonium acetate solution was used to samples while a 0.1 moll-' ammonium acetate buffel was used to condition the column prior to sample Solutions were adjusted to pH 4.0k0.1 and were before use by passing them through an EDTrA colu exclusively for reagent clean-up).A 1 moll-' HN03 was used as the eluent. All buffers eluents and test were prepared in high-purity deionized water (18 MC were stored in pre-cleaned (leached with 10% HNO polyethylene bottles. A series of test solutions were prepared from 10 stocks (SPEX Industries Edison NJ USA) to contai 1.0 and 10 ng ml-I of Cd Co Cu Ni Pb Th U Y the 14 REEs. Approximately 1 ml of 1 moll-' an Table 1 Chromatographic unit operating conditions employed for recovery experiments showing the pumps valves time periods an volumes used in each step Step 1 2 3 4 5 6 7 8 Pump P-2 P-2 P- 1 P- 1 P-2 P-3 P-3 P-3 Valve v-2 v-3 v-2 v-3 v-3 v-3 v- 1 v-2 Volume/ ml 2 3 2 300* 3 0.5 3 15 Time/ 18 27 24 3673 27 22 133 667 S Event Buffer solution purges the tubing Column conditioning Sample solution purges the tubing Sample solution is loaded onto the Buffer solution removes the matrix Eluent solution purges the tubing Elution of metals Column cleaning * Volume loaded determines the pre-concentration factor achieved. 91 8 Journal of Analytical Atomic Spectrometry October 1996 Vol.11acetate buffer was added and the pH was adjusted to 4.0k0.1 with 2 mol 1-' HNO and ammonia solution. A stock solution of synthetic sea-water (similar major element content to SLEW-1) was prepared by dissolving 268.82 g of NaC1 101.65 g of MgC12-6H20 and 21.908 g of CaClz.6H20 (AnalaR) in 11 of de-ionized water. The solution was purified prior to spiking by passing it twice through the separation-preconcentration system.The stock solution was diluted by a factor of 10 prior to separation and preccrncen- tration. A working solution was prepared by transferring 100ml of stock solution to a 500ml calibrated flask. Trace multi-element spikes were added followed by 15 ml of 0.1 moll-' ammonium acetate buffer and dilution to volume with deionized water. The pH was adjusted to 4 with 2 11-101 1-' HNO and ammonia solution. BLANK LEVELS AND CONTAMINATION Initial studies indicated a contamination problem for some elements (Co Cu Ni Pb and Zn) as high recoveries were obtained whereas for the less naturally abundant elements such as REEs Th and U good recoveries (typically 100 t 5%) were recorded.The most likely sources of contamination were reagents column material and laboratory environment. Each of the reagents used (buffer eluent and deionized water) was analysed directly by ICP-MS (see Table 3). The concentration of most elements was relatively low and was similar in all the reagents used. However Cu Ni Pb and Zn displayed a significant degree of contamination from the column itself with a Zn concentration of over 50 ng ml-' in the procedural blank solution. Repeated washing of the column did not reduce these levels of contamination. It was also observed that new columns contained appreciable concentrations of these elements. A rigorous regime was implemented in order to minimize possible sources of contamination.Solutions were prepared in a laminar flow hood. When a new column was installed it was thoroughly washed by passing 200ml of eluent then 30ml of de-ionized water 15 ml of eluent and finally 15 ml of de-ionized water through it. Between different recovery experiments the column was washed following the same procedure as above except that only 100ml of eluent were used in the first step. Tubes were washed twice firstly with 2 mol I-' HNO and secondly with de-ionized water. The minimum quantity of reagents was also employed (i.e. the pH was carefully adjusted trying not to add excessive amounts of reagents to the samples and blanks). PERFORMANCE EVALUATION A number of criteria were assessed to evaluate the overall effectiveness of the method.These included the shape of the elution profiles the percentage recovery of each analyte in both simple and complex matrices after passage through the chromatography unit repeatability effectiveness of matrix removal and accuracy of the overall technique for natural samples. Recovery experiments are a vital measure of the efficiency of any separation technique. To assess recovery the unit was operated off-line. The analytical procedure used is shown in Table 1 by loading different sample volumes in step 4 different preconcentration factors were obtained (e.g. when loading 300 ml the preconcentration factor obtained was 100; for 30 ml the pre-concentration factor obtained was 10). The flow rates were maintained at 4.9 6.7 and 1.4 ml min-' respectively for the sample (P-1) buffer (P-2) and elution (P-3) pump.To test the overall recovery of the separation-preconcentration system three 300 ml aliquots of spiked de-ionized water containing 0.1 0.5 and 1.0ngml-' test elements were loaded onto the column. They were stripped off with 3 ml of acid eluent and collected giving a preconcentration factor of 100. To test the effectiveness of the chromatography system to not only preconcentrate the elements of interest but also as a method of matrix removal a complex synthetic matrix was prepared. A 300 ml portion of spiked synthetic sea-water containing 1 ng ml-' of mixed REEs and trace metals was loaded onto the column. Trace elements were stripped off with 3 ml of acid eluent and collected giving a preconcentration factor of 100. Repeatability was assessed using three 30ml Table 3 Concentration (ng ml-') of trace element contaminants in reagents used during preparation and elution of samples.The most significant levels of contamination are seen for Zn in a procedural blank which has been passed through the chromatography column Element c o Ni c u Zn Y Cd La Ce Pr Nd Sm Eu Gd Tb DY Ho Er Tm Yb Lu P Th U De-ionized water 0.183 0.776 0.204 0.326 0.551 0.288 0.192 0.284 0.137 0.196 0.159 0.208 0.189 0.275 0.287 0.23 1 0.288 0.290 0.09 1 0.193 0.072 0.187 0.065 1 moll-' HNO 0.191 0.9 18 0.469 2.13 0.584 0.319 0.195 0.293 0.136 0.198 0.156 0.208 0.184 0.274 0.289 0.23 1 0.288 0.293 0.090 0.193 0.160 0.189 0.08 1 Procedural blank* pre-column 0.178 0.766 0.294 1.12 0.552 0.354 0.196 0.285 0.137 0.198 0.155 0.209 0.190 0.274 0.295 0.232 0.29 1 0.289 0.088 0.193 0.093 0.190 0.069 Procedural blank* post-column 0.261 2.42 4.79 0.617 0.417 0.512 0.317 0.138 0.212 0.158 0.208 0.189 0.275 0.292 0.232 0.300 0.289 0.090 0.193 1.46 0.197 0.371 55.8 0.1 mol 1-' buffer? 0.195 0.823 0.855 1.13 0.556 0.361 0.194 0.287 0.136 0.201 0.160 0.209 0.182 0.274 0.290 0.232 0.290 0.291 0.09 1 0.193 0.164 0.190 0.065 1 mol 1-l buffer? 0.190 0.732 0.334 0.861 0.560 0.398 0.207 0.289 0.142 0.198 0.146 0.209 0.186 0.275 0.293 0.234 0.294 0.294 0.090 0.197 0.158 0.208 0.065 Detection limit1 0.020 0.225 0.404 2.7 1 0.056 0.099 0.006 0.015 0.002 0.003 0.006 0.035 0.010 0.002 0.0 12 0.002 0.005 0.006 0.005 0.005 0.140 0.005 0.025 * De-ionized water-buffer-NH and HNO to adjust to pH 3.? Ammonium acetate solution. 1 Replicate determinations of procedural blank (n = 5 ) 30. Journal of Analytical Atomic Spectrometry October 1996 Vol. 11 91 9aliquots of 10ngml-' mixed REEs and trace metal solution. These were eluted from the column with 3ml of acid eluent and collected giving a preconcentration . factor of 10. The effectiveness of the matrix removal process was evaluated using a 30 ml sample of 10 ng ml-' mixed REEs and trace metal solution. This was eluted with 3 ml of acid eluent and collected (giving a preconcentration factor of 10). A 300ml portion of spiked synthetic sea-water containing 1 ng ml-' of mixed REEs and trace metals was also treated in the same manner. Both collected eluents were analysed by ICP-AES to determine the concentration of residual Na Mg and Ca after the matrix removal step.The accuracy of the procedure was determined by the replicate analysis of three certified reference materials from the NRCC. A 200 ml aliquot of each was taken and 1% v/v of buffer solution was added. The solution was then adjusted to pH4 with 2 mol I-' nitric acid and ammonia solution. The sample was loaded onto the column and then 2ml of eluent acid was used to strip the column and collect the ions. RESULTS AND DISCUSSION The percentage recovery for all elements of interest was assessed in a simple matrix (see Table 4). Percentage recoveries at an initial concentration of 1.0 or 0.5 ng ml-' were better than 100+10% for all elements except Cd Cu Ni and Zn (1 ngml-') and Ni U and Zn (0.5 ngml-').At 0.1 ng ml-' initial concentration recoveries were better than 100 5% with the exception of Cd Ni Th and most notably Zn. Thorium recovery was poor (60+8%) although at higher concentrations Th appeared to be quantitatively recovered. Zinc contamination clearly results in an apparently high recovery. Good precision (n=3) was obtained for all meas- urements with values between 0.20 and 7.25% RSD. Recovery values obtained from a synthetic sea-water matrix are shown in Table 5 and display a slight overall improvement over those from the simple matrix. Excellent recoveries were obtained at a preconcentration factor of 100 at 1 ngml-' for Co REEs Pb Y and U. Thorium recovery tended to be Table 4 Percentage recovery for some transition elements actinides and REEs in spiked deionized water after a preconcentration factor of 100 Percentage recovery Element Cd Ce c o c u DY Er Eu Gd Ho La Lu Nd Ni Pb Pr Sm Tb Th Tm U Y Yb Zn 0.1 ng ml-'* (n=4) 110f6 103 f 3 103 f 5 96f 13 100f2 102f 1 102f 1 103f2 101 &2 98f3 97f 1 100f4 120f7 97f2 102f3 101 f 3 102f 1 60f8 100f2 95f 1 100f2 99*2 669 f 134 0.5 ng ml-'* (n= 1) 108 107 101 102 110 104 108 108 109 106 105 104 89 104 106 102 107 100 108 112 106 106 117 1.0 ng ml-'* (n= 1) 115 101 105 128 99 97 99 98 99 102 95 103 115 101 103 103 97 90 98 91 110 98 111 * Concentration prior to preconcentration. Table 5 Percentage recovery for some transition elements actinides and REEs in synthetic sea-water after x 10 and x 100 preconcen- tration factor Percentage recovery Element Cd Ce c o c u DY Er Eu Gd Ho La Lu Nd Ni Pb Pr Sm Tb Th Tm U Y Yb Zn Preconcentration factor x loo* 1.0 ng ml-'* 94 100 99 105 99 102 99 99 99 97 99 99 83 97 98 98 98 89 99 95 105 98 119 Preconcentration factor x lot 10.0 ng ml-'* 95+ 1 102 f 1 117+3 104+ 8 99f 1 101 f 1 101 f 1 102f0 99f1 96+ 1 96f 1 99fl 112f5 96f2 103 + 1 102f 1 99f0 91 +2 98+ 1 96.5 f 1 109+2 97+ 1 184 + 73 * Starting concentration 1.0 ng m1-I.Starting concentration 10 ng ml-'. slightly low at 89%. Recovery for Ni was also a little low at 83% in the x 100 experiment. Zinc the element most prone to contamination showed an anomalously high value (120%). Data for a times 10 preconcentration factor at 10 ng ml-' was a little more variable. Excellent precision (n=3) was achieved in this experiment .with an RSD of 7-26%.In general recovery values were in the same range and were independent of the preconcentration factor (10 or loo) the test solutions (spiked water or synthetic sea-water) and the concentration of the starting solution (0.1 0.5 1.0 or 10 ng ml-I). For both types of matrices good recoveries were obtained for Ni Pb U Y and the 14 REEs while recoveries for Cu and Zn tended to be high due to contamination from reagents. The effectiveness of matrix removal is illustrated in Table6. In both experiments the amount of total dissolved solids (TDS) in the final solution was well below the limit tolerated by the ICP-MS instrument used in this work i.e. <0.1% TDS. An assessment of accuracy was made by analysing three certified reference materials (Tables 7-9).Although reference values are only available for a limited suite of elements (Cd Co Cu Mo Ni Pb U and Zn) they serve to demonstrate the application of the method. Measurements for SLEW-1 (Table 7) were made over a period of 12 months. Co Cu and Ni displayed good repeatability over the analysis period. The certified Cd concentration is low (0.018 ng ml-') and poor repeatability is expected at this level while Pb and Zn are Table 6 Concentrations (pg ml-') of some matrix elements determined by ICP-AES before and after matrix separation Solution Na Mi? Ca Total Before matrix removal 27 880 10650 2910 41 440 After matrix removal* 105 17 11 133 After matrix removalt 203 41 22 266 * Solution preconcentrated by a factor of 10. t Solution preconcentrated by a factor of 100.920 Journal of Analytical Atomic Spectrometry October 1996 Vol. 11Table7 Analysis of Estuarine Water SLEW-1 after separation and preconcentration by a factor of 100. Precision is expressed at the 95% confidence interval Measured value/ ng ml-' Element Cd c o c u Ni Pb Zn Experiment 1 * (n= 1) 0.083 0.084 2.04 0.831 0.148 1.07 Experiment 2 t (n= 1) 0.026 0.057 1.60 0.695 0.070 1.91 Experiment 3$ (n=4) 0.01 6 f 0.001 0.053 f 0.007 1.47 f 0.16 0.562 f 0.058 0.109+0.006 1.42 f 0.5 1 Reference value/ ng ml- ' 0.01 8 f 0.003 0.046 f 0.007 1.76 f 0.09 0.743 f 0.078 0.028 f 0.007 0.860 & 0.150 * Single aliquot month 1. t Single aliquot month 8. $ Four separate aliquots triplicate scans month 12. Table 8 Analysis of Nearshore Seawater CASS-2 after separation and preconcentration by a factor of times 100.Precision is expressed at the 95% confidence interval Measured value*/ ng ml-' Element (n=4) Cd 1.83f0.10 c o 2.7 f 0.3 c u 62.0 f 4.6 Ni 24.3 f 4.6 Pb 20.2 f 0.7 Zn 161 f 67 Reference value/ ng ml'-' 1.9 f 0.4 2.5 f 0.6 67.3 f 3.9 29.8 f 3.6 1.9 f 0.6 197f :I2 ~~ ~~~ * Four separate aliquots triplicate scans. Table 9 Analysis of Open Ocean Seawater NASS-4 after separation and preconcentration by a factor of 100. Precision is expressed at the 95% confidence interval Measured value*/ ng ml-' Reference value/ Element (n=4) ng ml'-' Mo 8.84f0.60 8.88 f 1.37 U 2.68 & 0.12 2.68 f (1.08 * Four separate aliquots; Cd Co Cu Ni Pb and Zn were not determined. clearly prone to contamination.Method reproducibility (Table7 experiment 3) was better than 15% RSD with the exception of Zn where reproducibility was poor at the lppb level. CASS-2 (Table 8) contains generally higher concen- trations of trace elements than SLEW-1 and accordingly both accuracy and intra-sample precision are improved. Lead dis- played a significant degree of contamination and inaccurate results. At x200 ng rnl-' Zn contamination was probably masked by the naturally elevated concentrations of this element in SLEW-1. Open Ocean Sea-water NASS-4 contains very low concentrations of most trace elements with the exception of Mo and U. Agreement between measured and reference values was excellent (< 0.5%) with an intra-sample precision of < 6% RSD where n=4 (Table 9). Unfortunately there are no saline water reference materials available which are certified for the REEs.An assessment of the internal consistency of the data set may be made by normalizing the REE data to chondritic abundance? and plotting the results against atomic number. Due to the coherent behaviour of the group a smooth curve should result for all elements except Ce and Eu which because of their difrerent oxidation states may display anomalous behaviour.'" The REE data for both CASS-4 and SLEW-1 (Table 10) displayed smooth chondrite-normalized curves with the exception of Sm which had anomously high results for both reference materials Table 10 Measured REE concentrations (ng ml-') in SLEW-1 and CASS-4 after separation and preconcentration by a factor of 100 SLEW-1 Element La Ce Pr Nd Sm Eu Gd Tb DY Ho Er Tm Yb Lu Batch 1 0.05 1 48 0.052 13 0.003 19 0.01 1 58 0.01 1 62 0.000 45 0.002 59 0.000 33 0.002 11 0.000 47 0.001 41 0.OOO 24 0.001 26 0.00021 Batch 2 0.014 08 0.016 01 0.002 86 0.011 66 0.012 69 O.OO0 41 0.002 41 O.Oo0 37 0.002 22 0.000 55 0.001 55 0.OOO 27 0.001 62 O.OO0 28 CASS-4 0.005 10 0.001 59 O.OO0 52 0.002 22 0.001 91 O.OO0 07 O.OO0 46 O.OO0 06 o.oO0 50 0.OOO 15 o.OO0 40 O.OO0 06 O.OO0 48 0.000 07 and between batches of SLEW-1.When REE concentrations in SLEW-1 were normalized to those of CASS-4 (i.e. a sample prepared using the same method) the anomalous result was eliminated (Fig. 2). This suggests that Sm contamination has occured at some point in the preparation procedure. P v) 0 I 52100 .E I I La Ce Pr Nd Sm Eu Gd T b Dy Ho Er Trn Yb Lu !! 8 Fig.2 Measured concentrations of the REEs in SLEW-1 normalized to measured concentrations in CASS-4. Very close agreement is seen between A Batch 1 and B Batch 2 of SLEW-1 from Pr to Lu. Poor agreement for La and Ce may be due to batch differences or recovery problems. Normalization to a second sample prepared using the same method eliminates the anomalously high Sm value (easily identified if concentrations are normalized to chondrite values) in CASS-4 and SLEW-1 suggesting that Sm contamination has occurred Journal of Analytical Atomic Spectrometry October 1996 Vol. 11 921CONCLUSIONS The separation procedure described here is rather time- consuming with a single column in operation; it does offer the potential for simultaneous determination of 23 elements includ- ing the REEs and U. This is in contrast to some chelation procedures which tend to be limited to a relatively restricted range of elements." The NERC ICP-MS Facility is in part supported by the Natural Environment Research Council and this support is gratefully acknowledged. One of us (E.A.) was in receipt of a EC COMMET bursary during the period of this work. Technical advice and assistance was provided by Bridget C.H. Gibson. REFERENCES 1 Elderfield H. Upstill-Goddard R. and Sholkovitz E. R. Geochim. Cosmochim. Acta 1990 54 971. 2 Jarvis I. and Jarvis K. E. Chem. Geol. 1992 95 1. 3 4 5 6 7 8 9 10 11 Williams J. G. and Gray A. L. Anal. Proc. 1988 25 385. Beauchemin D. Mclaren J. W. and Berman S . S. Spectrochim. Acta Part B 1987 42 467. Hutton R. C. and Eaton A. N. J. Anal. At. Spectrom. 1988,3,547. Gregoire D. C. Spectrochim. Acta Part B 1987 42 895. Garbe-Schonberg C.-D. Bruhn R. and Michaelis M. presented at the Conference on Plasma Spectrochemistry Granada Spain 1993. Schramel P. Xu L.-Q. Knapp G. and Michaelis M. Mikrochim. Acta 1992 106 191. Nakamura N. Geochim. Cosmochim. Acta 1974 38 575. Jarvis K. E. Chem. Geol. 1988 68 31. McLaren J. W. Lam J. W. H. Berman S. S. Akatsuka K. and Azeredo M. A. J. Anal. At. Spectrom. 1993 8 279. Paper 6/03822K Received May 3 I 1996 Accepted August 12 1996 922 Journal of Analytical Atomic Spectrometry October 1996 Vol. 1 1

ISSN:0267-9477    

DOI:10.1039/JA9961100917

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Determination of Long-lived Radioisotopes Using Electrothermal Vaporization-Inductively Coupled Plasma Mass Spectrometry* JORGE S. ALVARADO AND MITCHELL D. EKICKSON Environmental Research Division Argonne National Laboratory 9700 South Cass Avenue Argonne IL60439 USA A general method for the determination of long-lived radioisotopes by integrating electrothermal vaporization and inductively coupled plasma-mass spectrometry (ETV- ICP-MS) to vaporize environmental samples with complex inorganic matrices is described. The method required no sample pretreatment and minimized sample size. The rationale was to use chemical modifiers such as CHF to form metal fluorides with much lower boiling-points than other metal compounds (such as oxides and carbides). Given sufficiently high temperatures and long reaction times samples in other chemical forms are converted into elemental halides and vaporized.The characterization and application of ETV- ICP-MS for the determination of radioisotopes is described. The detection limits for 99Tc 238U u6U U2Th 230Th and 226Ra were similar to those obtained with ultrasonic nebulization ( USN-ICP-MS). Absolute detection limits ranged from 0.6 fg for 226Ra to 5 fg for 238U. Analytical calibration plots were linear over a range of 2-3 orders of magnitude. Matrix effects caused by Group IA and IIA elements were minimized by changing the nature of the sample and by using temporal-thermal programming without affecting analytical performance. Comparison studies between ETV- ICP-MS and classical radiometric techniques were performed for various environmental samples.Keywords Electrothermal vaporization; sample introduction; inductively coupled plasma mass spectrometry; long-lived radioisotope; chemical modifier ICP-MS has proved to be a cost-effective trace analytical technique for many transition elements but has only recently been applied to the radiochemical field. This technique offers advantages such as low detection limits (typically n,g 1-I) mass-selective detection and multicomponent detection. In determining long-lived radionuclides ICP-MS is rapidly sur- passing other techniques such as differential-pulse polarogra- phy,' radiochemical neutron activation,2 ion chromatography3 and classical ph~tometry.~ The main advantage of ICP-MS is its ability to determine (1) long-lived radionuclides with low- intensity radiation and (2) alpha-emitting radionuclides that require tedious radiochemical separations.Despite the fact that ICP-MS itself was a major improve- ment techniques of sample introduction are critical for the performance of the plasma source. Browner and Boorn5v6 define the goals of sample introduction as 'the reproducible transfer of a representative portion of sample material to the atomizer cell with high efficiency and with no adverse inter- ference effects.' Owing to its simplicity and high reproducibility * The submitted manuscript has been authorised by a contractor of the US Government under contract No. W-31-109-ENG-38. Accordingly the US Government retains a non-exclusive royalty-free license to publish or reproduce the published form of this contribution or allow others to do so for US Government purposes.Journal of Analytical Atomic Spectrometry pneumatic nebulization (PN) is the most common method of introducing aqueous samples; however the efficiency is poor with reported typical efficiencies for PN of 1% with ICP spe~trometry.~ Ultrasonic nebulization (USN) has been used as a replacement for PN. Although USN has a more efficient production of droplets ranging up to 30% the instrument has longer clean-out times and tends to enhance matrix effects. Since its introduction electrothermal vaporization (ETV) has also been used for the introduction of samples into the plasma. Some of the advantages over traditional methods of sample introduction are high analyte transport efficiency producing higher sensitivity; reduction of polyatomic ion inter- ferences; reduction of oxide formation of the analyte plasma gas and matrix components; small volumes of sample required; ability to analyse organic liquids strong acids liquid high in solids and slurries; reduction of non-spectroscopic interferences through selective volatilization; and improved performance using chemical modification techniques.However the poor reproducibility and the formation of refractory molecular species involving the furnace (or cup) material are considered major disadvantages. Physical carriers such as the halogens and halocarbons have been used as aids to volatilization in order to reduce losses and improve sample transport. Ng and Caruso7 showed that pyrolytic carbon cups typically used during the ETV process exhibit memory effects multiple peaks and broad bands that can be minimized by using a tantalum coating procedure.8 Studies by Barnes and Fodorg and by Gun et a1.I' reported that when nickel was added elements such as selenium and arsenic owing to their volatility were retained during the ashing process increasing memory effects." Ediger and Beres12 performed an interesting experiment with sodium chloride magnesium nitrate palladium nitrate and tellurium nitrate as chemical modifiers.This study was performed on 20 elements and the amount of modifier present was more important than the physical characteristic of the modifier itself. The major exception here was uranium which seemed to show large suppressions for all of the solution-based modifiers evaluated except for palladium in moderate amounts.Fonseca and Miller- Ihli13 used palladium as a physical carrier to reduce the difference in response between samples and aqueous standards in slurry analysis. Nickel and Zadg~rska,'~ using an ETV device showed that the total evaporation of impurities in ceramic powders was achievable by mixing the sample with BaO and CoF,. Although many of these experiments showed promising results an increase in memory effects and the matrix compo- nents make the efficacy of chemical modification techniques very specific to the determination of specific analytes in specific matrices. In order to create a general procedure for the determination of any element in any type of sample and to be able to compare with standards a method that eliminates the matrix matrix interferences and memory effects should be Journal of Analytical Atomic Spectrometry October 1996 Vol.11 (923-928) 923studied. Ren and Salin" described the determination of trans- ition elements by using Freon 12 for the analysis of sediments and coal fly ash. Experimental results showed that by using a gas halogenation reagent nearly 100% vaporization was achieved for samples with a high content of aluminium silica zirconium and tantalum. Matrix effects were reduced and the formation of oxides and carbides was minimized. The objective of this work was to develop a general method that can be used to introduce any sample for the determination of radionuclides in environmental samples by ETV-ICP-MS.Our approach builds on that described previously by Ren and Salin," where samples are vaporized in the presence of halogen- ated gas by using high temperatures to decompose the halogen- ated compounds and create fluoride compounds that have lower vaporization temperatures and to eliminate interferences from the formation of oxides and carbides. For these experi- ments CHF3 was used because of its high purity and longer reaction times. Minimal modification makes the instrument easy to use and the process convenient for automation. EXPERIMENTAL Instrumentation The ICP mass spectrometer used in these experiments was a PlasmaQuad I1 (VG Elemental Winsford Cheshire UK). Instrument operating conditions are listed in Table 1. Two types of sample introduction technique were employed a U-5OOOAT ultrasonic nebulizer (Cetac Technologies Omaha NE USA) and a VG Microtherm Mark I11 electrothermal vaporizer (also manufactured by Fisons Instruments).The electrothermal vaporizer consists of a graphite tube open on both sides to permit the passage of argon used as the carrier gas. Pyrolytic coated carbon tubes from Fisons Instruments were used. Some of the carbon tubes were soaked in a tantalum solution as described by Zatka16 and by Ng and Caruso' to form the tantalum carbide layer. The development of a gold colour on the surface of the cup indicated the formation of the tantalum carbide.8 In ETV the sample is introduced through a small inlet on top of the graphite tube which is closed by a graphite rod when vaporization is taking place.On either side of the graphite cup are two graphite electrodes that heat the graphite tube when high voltages are passed through them. The sample undergoes three processes drying ashing and atomization. Operating conditions are described in Fig. 1. Reagents The standard solutions of the radionuclides of interest were quantified by isotope dilution employing radionuclides trace- Table 1 Instrument operating conditions Instrument and parameter Operating value ICP-MS Fisons PlasmaQuad I1 ( VG Elementa1)- Incident power 1350 W Reflected power ow Coolant gas flow rate Auxiliary gas flow rate Carrier gas flow rate Ultrasonic nebulizer U-5000AT (Cetac Technologies)- Condenser temperature 0 "C Heater temperature 150 "C Sample uptake rate Argon carrier gas flow rate Trifluoromethane gas flow rate Sample injection volume 25 p1 13.8 1 min-' 0.94 1 min - ' 0.8-1.0 1 min-' 1.2 ml min-' 0.8-0.9 1 min-' 1.0 ml min-' Electrothermal vaporizer Microtherm Mark III ( VG Elemental)- "Flash" Program "Ramp" Program Timels TemperaturePC Timds TemperaturePC Drying 0-60 0-80 60- 120 80-110 Ashin 120-190 130-160 Va orizarion 169-160 160-168 ~~ ~~.168- 168 168-180 110-400 400-600 600-2600 2600-2600 2600-25 25 -25 Drying 0-40 0-100 40-50 100-100 Ashing 50-70 70-100 100-100 Va orizarion 115-118 106-1 15 118-1 18 118-128 100-750 750-750 750- 1 100 1100-2400 2400-2400 2400-25 25-25 Fig. 1 Electrothermal vaporization operating conditions able to NIST. Standard solutions of radium-226 (SRM 4958) technetium-99 uranium-238 uranium-236 and thorium-232 were used.Standard solutions were prepared by volumetric dilution of each radionuclide with 5% v/v nitric acid (Ultrex 11-Ultra Pure Reagent J. T. Baker Phillipsburg NJ USA) and distilled de-ionized water (18 mR cm Nanopure System Barnstead Thermolyte Dubuque IA USA). Matrix effects were studied by using nitrate salts of sodium calcium potassium and magnesium from J. T. Baker. Solutions containing 1-1000 mg 1-' of each element were prepared. Tantalum metal powder hydrofluoric acid (1 + 1) and oxalic acid dihydrate from Aldrich (Milwaukee WI USA) were used to prepare the soaking solution necessary to produce the tantalum carbide coatings. This soaking solution contained 6% m/v tantalum. The gases used in ICP-MS and ETV were liquid argon and 99.995 electronic-grade Halogen-23 (CHF3) from AGA Gas (Cleveland OH USA).Samples The following samples were used tap water from Chicago Illinois and Lemont Illinois; river water from the Fox River and Kankakee River in Illinois; well water from Lemont Illinois and Borden Indiana; lake water from Herrick Lake in Illinois; spring water; and groundwater from the Gaseous Diffusion Plant in Paducah Kentucky were used. The samples were collected in plastic bottles and acidified to a pH of about 2 with nitric acid. Samples were divided into two fractions and one of the fractions was analysed for uranium,I7 thorium,I7 radium,18 and technetiumlg by using radiochemical techniques for confirmation. RESULTS AND DISCUSSION Instrument Optimization ICP mass spectrometers are used more often for the analysis of liquids via continuous solution nebulization which produces a stable ion signal.In a traditional procedure for optimization instrument parameters such as gas flow rates and ion lens stack settings are tuned to obtain a maximum signal. Owing to the transient gaseous and 'dry' nature of the ETV signal and the changes in the physical and chemical conditions during each step in the vaporization process the optimization param- 924 Journal of Analytical Atomic Spectrometry October 1996 Vol. 11eters were shown to be different from the typical conditions for optimization. Different tuning parameters have been studied to obtain optimum conditions during the ETV process. Gray et aL20 used the 38Ar2 dimer to optimize ion lens voltages torch position and nebulizer flow rate.Further optimization was performed by using single-ion monitoring of 115i[n to compensate for the vaporization pulse effect. For these studies USN was used to optimize the instrument using a 0.5 pg 1-' solution of beryllium magnesium cobalt bismuth indium lead and uranium. The signal was maximized for '151n. The standard deviation was less than 2% with ten consecutive 60 s integrations. USN was used for optimization because the extra desolvation provided by this instrument makes conditions that resemble (to a first approximation) the dry plasma conditions in ETV. After the gross calibration the ETV device was calibrated for each individual element. The typical response is shown in Fig. 2. When the auxiliary gas flow rate was increased from 0.80 to 1.06 1 min-l a plateau was reached with no changes in the signal until high flow rates were used.A different response was obtained when the nebulizer gas flow rate was changed from 0.86 to 1.06 1 niin-' as shown in Fig.2. The signal proved to be optimum for a 0.05 ppb solution of 238U when a carrier gas flow r'rte of 0.96 1 min-' was used. This signal showed a maximum that is very sensitive to any changes in flow rate. A small change can produce a decrease in the signal of up to 25% of the optimum signal. Chemical Modifiers ETV has been shown to exhibit memory effects multiple peaks and broad bands when the pyrolytically coated tubes interact with the element under study. Initial observations showed a response similar to that shown in Fig.3(u) where a reproduc- ible broad band was obtained for 232Th. An explanation for these effects is the possible formation of carbide or oxide compounds (with much higher boiling-points) that compete in the vaporization-atomization process. The formation of tantalum carbide (mp = 3983 "C) is neces- sary to restrict carbide formation with the element under study. Atomization and carbide formation are competing processes. More specifically the tantalum carbide formed is more stable than the carbides of elements such as thorium arsenic uranium or radium. When atomization occurs the tantalum carbide layer prevents the formation of any other carbide species and as a consequence single and sharp peaks of the element under study are obtained. Unfortunately no noticeable change in the 2m i Argon flow/l min" Fig.2 Variation of uranium-238 (0.05 pg I-') signal as a function of nebulizer and auxiliary gas flow rates 40 36 32 28 24 20 16 12 8.0 4.0 $2 = o z Fig. 3 Vaporization behaviour of thorium-232 (a) argon as carrier gas; and (b) trifluoromethane as carrier gas signal occurred when the tantalum-coated graphite tube was used in place of the uncoated tube. The formation of carbide was not the source of the broad signal. Chemical modifiers are known to increase sample transport and to minimize the formation of other interfering compounds. Elements such as thorium and uranium have a tendency to form stable oxides with boiling-points as high as 2350 and 3390 "C for UOz and Tho2 respectively. Fluoride compounds are usually characterized by having lower boiling-points than most oxides ( UF6 = 64.8 "C; UF = 1400 "C).When chemical modifiers such as CHF3 were used a single sharp signal [Fig. 3(b)] was observed. When CHF3 is heated it decomposes to form free fluorine radicals that react with the element under study producing fluoride complexes with lower boiling-points. These complexes have a much higher volatility than the oxide compounds. In addition the CHFJ helps in the elimination of matrix components the reduction of memory effects and the reduction of the background levels. The fluorine radicals in addition to reacting with the analyte react with other parts of the matrix such as silicates carbides and oxides; for example the Si02 (bp = 2230 "C) present as particles in liquid samples and present as a major constituent in soils can be eliminated Journal of Analytical Atomic Spectrometry October 1996 Vol.11 925by the formation of SiF (bp = - 86 "C). In addition to improv- ing the signal for uranium and thorium a reduction in the background levels and in memory effects was observed. No signal improvement was observed for radium and technetium. Linearity and Detection Limits Detection limits were determined and the results are listed in Table 2. Detection limits were defined as the concentration that yields a signal-to-noise ratio of three. The noise was defined as the standard deviation of ten non-consecutive blank determi- nations. These detection limits were calculated three times on three non-consecutive days to test for reproducibility.Table 2 shows the detection limits in femtograms when no chemical modifier was used and when CHF was used as a chemical modifier. An improvement of two or three orders of magnitude was observed. Trifluoromethane did not improve the signal of radium or technetium but improvement in the background caused a slight improvement in detection limits. Differences were observed in the detection limits of similar isotopes. Uranium-238 showed detection limits five times higher than 236U. Because 236U is not a natural isotope in the environment lower back- grounds were observed and thus better detection limits. This behaviour and the detection limits obtained are comparable to the results obtained by USN-ICP-MS2' and with the exception of radium meet the minimum requirements for the detection of radioisotopes in the environment.Ebdon and Gooda1122 described the thermochemical effects in hexafluoroethane-modified argon and concluded that the introduction of halocarbons resulted in an apparently unusual plasma spectrochemistry which yields non-linear calibration graphs. This effect was attributed to classical mass action buffering involving highly stable metal fluoride compounds. In contrast to this earlier work,22 we found that when 25 pl aliquots of 226Ra 236U 238U "Tc 230Th and 232Th were introduced into the ETV device the system showed a linear response over two orders of magnitude for each element. The log-log slopes were 0.99 for radium 1.00 for thorium 0.99 for uranium and 1.07 for technetium. No 'roll-over' was observed at higher concentrations.Fig. 4 shows the calibration plot for each element. Matrix Effects ETV offers distinct advantages over conventional methods of sample introduction. The gaseous nature of the sample reduces the effects of solvent and sample matrix interferences. In addition the ability to use temporal-thermal programming allows selective removal of the sample matrix constituents. Matrix effects can increase signal intensities in ICP-MS by facilitating ionization processes; conversely matrix effects can quench signals. These competing results as well as secondary effects produced by changing plasma temperature such as analyte desolvation solute volatilization compound dis- Table 2 Detection limits for long-lived radioisotopes. *Detection limit is defined as three times the pooled standard deviation of ten non- consecutive blank analyses ETV unmodified j Isotope fg Uranium-238 180 Uranium-236 48 Thorium-232 1600 Thorium-230 -$ Radium-226 1 Technetium-99 -1 ETV trifluorome t hane/ fg 5 0.9 2 1.4 0.6 1.5 USNt/ fg 18 25 0.5 0.3 0.3 0.8 * Sample size 25 pl.t From ref. 21. -$ Not determined. 5.61 5.22 - 4.83- 7 I w 2 4.44 - $ 4.06- c *g 3.67 - cn ,O 3-28 - 2.89 - 2.50 ' -3150 -2193 -2136 -159 -dl -0k4 -0107 log concentration added (ppm) Fig. 4 Calibration plots for technetium (H) radium (+) thorium (0) and uranium (+ ) sociation ionization and excitation can lead to complex relationships between the analyte and various matrix constitu- ents. Tanner et aL2 described space-charge effects that take place in ICP-MS due to a perturbation of the electrostatic field in an ion flow system.If the space-charge field is sufficiently strong a self-repulsion develops which acts to spread the ion radial distribution and thereby reduces the transmission of other ions. Matrix effects have been observed not only in the ICP-MS system but also in dc plasma24 and microwave-induced plasma atomic emission ~pectrornetry.~'.~~ Following the classical Le Chatelier's principle when easily ionizable elements (EIEs) are added to the sample in high concentrations the classical description states that the ioniz- ation of EIEs produces electrons that shift the ion-atom analyte equilibrium in the direction of neutral analyte atoms decreasing the signal observed on the mass spectrometer. Although analytical intensities-to-background ratios can be improved two or three times by deliberate EIE doping enhancement is generally regarded as an interference because of the different EIE content of samples and the unpredictable behaviour of the plasma.Matrix effects on uranium and thorium were studied by monitoring the mass of 236U and 232Th respectively. Matrix effects for uranium are shown in Fig. 5. For uranium the typically described EIE effect was observed when elements such as potassium sodium and magnesium were introduced. A comparison of the ionization potentials of sodium (5.138 eV) and potassium (4.339 eV) can be used to explain why potassium 120 I I 0 ' 0.b 0.50 1.b 1.k 2.b 230 3. log concentration added (ppm) Fig. 5 Effect of sodium (+) potassium (0 ) calcium (m) and mag- nesium (+) on the uranium-236 signal using flash vaporization 926 Journal of Analytical Atomic Spectrometry October 1996 Vol.1 1produces a larger quenching than sodium on the uranium signal. When calcium was introduced a maximum signal was observed at concentrations close to 10 ppm. Depressions at high concentrations of added EIEs were probably due to overloading of the plasma. Effects of EIEs on the thorium signal are shown in Fig.6. Sodium and potassium produced a decrease in the signal; however in this instance sodium produced a decrease in the signal of about 50% when a concentration of approxirnately 5 ppm was introduced and potassium produced a 5 -10% decrease in the signal. This behaviour is not consisteni with the EIE effect and is a good example of combination cffects taking place in the plasma and at the plasma interface.'When calcium and magnesium were introduced as matrix inter- ferences an increase in the thorium signal was observed. Magnesium and calcium produced an increase in the th+)rium signal of 100 and 8O% respectively when concentrations of 5 ppm of each element were introduced. In general when elements from Groups IA and IIA were introduced into the plasma the effects on analyte ion and atom ratios could be described by classical EIE effect3 in a few instances; however much of the behaviour was not consist- ent and depended on both the EIE and the analyte. The behaviour is not yet fully understood but may be the re>ult of a combination of classical equilibrium and space-charge effects.For this reason the presence of EIEs and other elements in the sample matrix can be considered more as an interference than as a signal improvement. In an attempt to minimize the matrix effects we used the unique characteristics of the ETV of temporal-thermal pro- gramming to vaporize and remove matrix interferences selec- tively. Figs. 7 and 8 show the effects of sodium and calcium respectively on the uranium signal when flash vaporization and ramp vaporization as described in Fig. 2 were used. When sodium was introduced no effects were observed for concen- trations up to 100 ppm. Calcium showed no effects at concen- trations close to 80 ppm. By performing thermal programming it was possible to minimize matrix effects when samples exhibited concentrations of approximately two orders of mag- nitude of each interfering element.In addition temporal- -ther- mal programming can be used to vaporize compounds with similar chemical and different physical compositions (e.g. U308 or UO,) selectively. A decrease in the signal at concentrations higher than these concentrations was observed when both elements calcium and sodium were introduced. This effect can be explained in the same way as before due to overloading of the plasma. Thermal programming allows the matrix effects in environmental samples to be minimized usually with concen- trations of EIEs lower than those used in this experiment. 120.00 m.00 $ 3 80.00 u 60.00 h v c ul v) .- .- A - a 0 z 40.00 20.00 0.00 log concentration added (ppm) Fig. 7 Effect of sodium on the uranium-236 signal using flash vaporiz- ation and thermal programming (ramp) 120.00 ' 100.00 h 8 Y 80.00 0) v) .- 60.00 .- 8 - a 0 z 40.00 20.00 0.00 ' 0.h 0.h tbo 1.h 2.b 230 3.log concentration added (ppm) Fig. 8 Effect of calcium on the uranium-236 signal using flash vaporiz- ation and thermal programming (ramp) Precision Precision measurements are illustrated in Fig. 9 for 99Tc. The data represent the worst-case scenario during this analysis. For ten different samples the blank and the standard with a concentration of 99Tc of 1.21 ng l-' were introduced into the ETV device; the average signal for the sample was 232 counts s - l with a relative standard deviation of 6.6%. For the blank the average signal was 24 counts s-' with a relative standard deviation of 9.2%.Other elements such as radium uranium and thorium presented relative standard deviations lower than 7%. 2Ml 1 500 Acid blank Te-sample 2.2 15.4 Mean 24 232.2 v) RSD I 9 .2% 6.6% . log concentration added (ppm) Fig.6 nesium (+ ) on the thorium-232 signal using flash vaporization Effect of sodium (+) potassium (0) calcium (m) and mag- mm 0 Sample number 1 Fig. 9 Precision measurement of technetium-99 and blank samples Journal of Analytical Atomic Spectrometry October 1996 Vol. 1 1 927Table3 Comparison of the results for the determination of uranium-238 in water Sample ETV-ICP-MS/pg 1- a-Spectrometry/pg 1-' Tap water- Chicago IL Lemont IL River water- Fox River IL Kankakee River IL Well water- Lemont IL Borden IN Herrick Lake Spring water Others- 0.001 5 f 0.0004 ND* 0.006 f 0.002 0.0043 f 0.0007 ND* 0.012 f0.003 0.003 f 0.001 0.001 5 f 0.0004 < 0.09 < 0.09 3.27 0.12 < 0.06 < 0.09 < 0.06 < 0.09 < 0.06 * Not detected Table4 Comparison of the results for the determination of thorium-232 in water ~~ Sample ETV-ICP-MS/pg 1-' a-Spectrometry/pg 1- ' Tap water- Chicago IL 0.17 & 0.04 0.27 & 0.05 Lemont IL 0.27 & 0.08 0.28 f 0.06 Fox River IL 0.8 f 0.2 0.92 & 0.09 Kankakee River IL 1.4 f 0.3 1.5f0.1 Lemont IL 0.24 f 0.07 0.30 f 0.07 Borden IN 0.07 f 0.03 0.06 f 0.02 Herrick Lake 0.19 f 0.06 0.16 & 0.05 Spring water 0.15 f 0.05 0.13 f 0.04 River water- Well water- Others- Table5 Comparison of the results for the determination of technetium-99 in water ETV-ICP-MS/ Membranelfi-counter*/ Sample ng 1-' ng I-' Paducah-5920 1.4+0.2 Paducah-6275 26f2 1.2 _+ 0.1 25f3 * Low-background proportional counter Results of Environmental Sample Analysis Samples of tap water river water well water lake water and spring water from the Illinois and Indiana areas were collected as described previously and analysed for 226Ra 238U 235U 232Th 230Th and "Tc.Owing to the very low concentrations in natural samples and the detection limits of the system no results were obtained for 226Ra 235U and 23@Th. Results obtained by using external calibrations for 238U 232Th and "Tc are reported in Tables 3-5 which also compare each isotope measurement obtained in this work with the results obtained by using isotope dilution alpha-spectrometry and low-background proportional counters.There was good agree- ment between the values obtained in this work and the values obtained by the radioanalytical measurements. CONCLUSIONS ETV-ICP-MS shows several advantages the system is easy to use is capable of handling microlitre sample volumes and is highly sensitive. Also plasma excitation processes are more efficient with ETV owing to the removal of solvent before the sample is introduced. The detection limits for long-lived isotopes such as "Tc 238U 236U 232Th and 226Ra were sim- ilar to those obtained with USN-ICP-MS. No sample pre-treatment was required. ETV can be used for the analysis of real samples owing to the elimination of matrix interferences. The use of a halogen- ated gas such as CHF3 as a carrier/chemical reactor gas proved to be very effective as a sample digester in the ETV- ICP-MS analysis.This gas minimized the effect of carbides and oxides by creating compounds with lower boiling-points. In addition the gas can be used to eliminate matrices such as Si02 particles or as a major component in a soil sample by forming fluorinated compounds such as SiF4 (gas). The use of temporal-thermal programming in addition to halogenation selectively eliminates matrix constituents without adversely affecting analytical performance. The authors thank Lesa L. Smith and Kent A. Orlandini for their work in preparing and characterizing the water samples used in this study. This work was funded by the Laboratory Management Division of the Office for Environmental Restoration and Waste Management US Department of Energy. Argonne National Laboratory is operated by the University of Chicago for the US Department of Energy under contract number W-3 1-109-ENG-38.REFERENCES 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 Daes S. K. Kulkarni A. V. and Dhaneshwar R. G. Analyst 1993 118 1153. Franek M. and Krivan V. Anal. Chim. Acta 1993 274 317. Jackson P. E. Carnevale J. Fuping H. and Haddad P. R. J. Chromatogr. A 1994 67 181. Rohr U. Meckel L. and Ortner H. M. Fresenius' J. Anal. Chem. 1994 348 356. Browner R. F. and Boorn A. W. Anal. Chem. 1984 56 787A. Browner R. F. and Boorn A. W. Anal. Chem. 1984 56 875A. Ng K. C. and Caruso J. A. Anal. Chim. Acta 1982 143 209. Toth L E. Transition Metal Carbides and Nitrates (Refractory Material Vol.7) Academic Press New York 1971. Barnes R. M. and Fodor P. Spectrochim. Acta Part B 1983 38 1191. Gunn A. M. Millard D. L. and Kirkbright G. F. Analyst 1978 103 1066. Alvarado J. S. PhD Thesis Northern Illinois University 1990. Ediger R. D. and Beres S. A. Spectrochim. Acta Part B 1992 47 907. Fonseca R. W. and Miller-Ihli N. J. Appl Spectrosc 1995 49 1403. Nickel H. and Zadgorska Z. Fresenius' J. Anal. Chem. 1995 351 158. Ren J. M. and Salin E. D. Spectrochim. Acta Part B 1994 49 555. Zatka J. V. Anal. Chem. 1978 50 538. Smith L. S. Crain J. S. Yaeger J. S. Horwitz E. P. Diamond H. and Chiarizia R. J Radioanal. Nucl. Chem. 1995 194 151. Alvarado J. S. Orlandini K. A. and Erickson M. D. J. Radioanal. Nucl. Chem. 1995 194 163. Orlandini K. A. King J. and Erickson M. D. Methods for Evaluating Environmental and Waste Management Samples DOE/EM-O089T Rev-2 US Department of Energy Richland Washington September 1994. Gray D. J. Wang S. and Brown R. Appl. Spectrosc. 1994 48 1316 Crain J. S. Smith L. L. Yaeger J. S. and Alvarado J. S. J. Radioanal. Nucl. Chem. 1995 194 133. Ebdon L. and Goodall P. Spectrochim. Acta Part B 1992 47 1247. Tanner S. D. Cousins L. M. and Douglas D. J. Appl. Spectrosc. 1994,48 1367. Miller M. Keating E. Eastwood D. and Hendrick M. S. Spectrochim. Acta Part B 1985 40 593. Alvarado J. S. and Carnahan J. W. Appl. Spectrosc 1993 47 2036. Wu M. and Carnahan J. W. Appl. Spectrosc. 1992 46 163. Paper 6/02394K Received April 9 1996 Accepted July 10 1996 928 Journal of Analytical Atomic Spectrometry October 1996 Vol. 11

ISSN:0267-9477    

DOI:10.1039/JA9961100923

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Journal of Analytical Atomic Spectrometry Characterization of Spent Nuclear Fuels by Ion Chromatog raphy-Inductively Coupled Plasma Mass Spectrometry* JOSEFA M. BARRERO MORENO J. IGNACIO GARCIA ALONSOt PHILIPPE ARBORE GEORGOS NICOLAOU AND LOTAR KOCH European Commission JRC Institute for Transuranium Elements Postfach 2340 761 25 Karlsruhe Germany Ion chromatography (IC) coupled to ICP-MS was applied to the characterization of two different spent fuel samples [uranium oxide ( U02) and mixed uranium-plutonium oxide (MOX)] and the results were compared with those obtained by other techniques. Isotope dilution analysis with ion chromatographic separation was applied to the determination of the fission products (Rb Sr Cs Ce Nd Sm Eu and Gd). The standard additions method was employed for the determination of monoisotopic fission products and actinides (Y La Pr 147Pm 237Np "'Am 243Am and 244Cm).Total U and Pu were determined only by ID-TIMS. Nd Am and Cm isotope concentrations were determined also by ID-TIMS. gamma spectrometry. Based on the ID-TIMS results for U 9 134Cs and I3'Cs were determined in parallel by 144Ce 154EU Pu and 148Nd fuel burn-up was calculated and the value introduced into the computer code KORIGEN in order to calculate the complete fuel inventory based on the known irradiation parameters for the two fuels. The agreement between the experimental IC-ICP-MS results and the theoretical calculations was within 15% for most isotopes. Keywords Inductively coupled plasma mass spectrometry; chromatography; isotope dilution; spent nuclear fuel ion The inventory of fission products and actinides in spent nuclear fuel is often required when new types of fuels and/or reactor operating conditions are being investigated.Fuels containing minor actinides (Np Am Cm) for transmutation purposes and fuels subjected to very high burn-ups are examples where post-irradiation examination of the spent fuel has to be performed. Current techniques used in the Institute for Transuranium Elements for the determination of fuel param- eters include mainly ID-TIMS (for U Pu Am Cm and Nd) and gamma spectrometry (for 134Cs 137Cs '44Ce ls4Eu and lo6Ru). Based on the heavy metal and Nd isotope concen- trations determined the fuel burn-up can be calculated; this is then applied in the computer code KORIGEN,' together with the known irradiation history of the fuel and the reactor operating conditions. The complete inventory of the fuel is then calculated and the results are compared for the other measured isotopes.However the determination of Am. Cm and Nd by TIMS is very time-consuming because of the difficult chemical separations that have to be performed. The determination of gamma-emitting nuclides is simpler but suffers from poor accuracy and precision in spent fuel samples for some nuclides. In order to obtain a larger amount of data and simplify the handling of the samples we have applied ICP-MS to the characterization of spent fuel ICP-MS is a sensitive multi-elemental technique that offers very low detec- * Presented at the 1995 European Winter Conference on Plasma Spectrochemistry Cambridge UK January 1995.7 To whom correspondence should be addressed. Present address Department of Physical and Analytical Chemistry Faculty of Chemistry Julian Claveria 8 33006 Oviedo Spain. tion limits and broad elemental coverage with the additional capabilities of providing isotopic abundance information and absolute concentrations by applying the technique of isotope dilution. However the complete determination of fission prod- ucts and actinide isotopes in spent nuclear fuels by ICP-MS is hindered by the presence of some isobaric interferences which cannot be corrected f ~ r . ~ ~ In order to overcome these interferences chemical separation is needed. The coupling of liquid chromatography to ICP-MS is well documented in the literature for trace metal speciation and environmental analy- S ~ S .~ Also the separation of the lanthanide elements in order to eliminate isobaric interferences has been described." We have used also ion chromatography (IC) coupled to ICP-MS for the elimination of isobaric interferences in the analysis of spent fuel samples." Fission products that can be determined in spent fuel solutions by IC-ICP-MS include Rb Sr Y Cs Ba La Ce Pr Nd Pm Sm Eu and Gd. Other fission products including Zr Mo Tc Ru Rh Pd and Te are not completely dissolved in 7 mol 1-1 nitric acid and remain partially as insoluble residues.' For this reason we have selected only the first series of elements together with the actinides Np Am and Cm for a comparison with the standard techniques and with the theoretical calculations. EXPERIMENTAL Instrumentation An ELAN 5000 ICP-MS instrument (Perkin-Elmer SCIEX Thornhill Ontario Canada) was modified to handle radioac- tive samples in a glove box and coupled with a 45001 high- pressure chromatographic pump (Dionex Sunnyvale CA USA).Details of the instrumental set-up have been published previously." The Dionex high-pressure pump was located outside the glove-box while the injection valve (Dionex Part No. 42766 200 pl loop) guard column and analytical column were placed inside the glove-box. The effluent from the chroma- tographic column was directed to the cross-flow nebulizer. A peristaltic pump was used to carry the waste to a vessel inside the glove-box. Standard plasma and ion lens operating con- ditions are summarized in Table 1.They were optimized as described previously." Reagents and Materials Lanthanide separations were performed using Dionex CG5 and CS5 mixed bed columns using a 0.1 mol 1-' oxalic acid (Suprapur Merck Darmstadt Germany)-0.19 moll - ' lithium hydroxide (analytical-reagent grade Merck) eluent. For the separation of the actinides and Rb Sr Y Cs and Ba Dionex CGlO and CSlO cation-xchange columns were used with a 1 mol 1-' nitric acid (Suprapur Merck) eluent. Natural element standards (1000 mg ml-') of Rb Sr Y Cs Ba La Ce Pr Nd Sm Eu and Gd were obtained from SPEX (Grasbrunn Germany). The 237Np standard was obtained from Los Alamos National Laboratory (Los Alamos NM USA) Journal of Analytical Atomic Spectrometry October 1996 Vol. 11 (929-935) 929Table 1 Operating conditions Rf power/W Argon outer gas flow rate/ 1 min-' Argon intermediate gas flow rate/l min-' Argon nebulizer gas flow rate/ 1 min-') Nebulizer type Spray chamber Load coil-sampler distancelmm Quadrupole working pressure/Pa Sampler and skimmer cones Ion lens 1050 15 0.8 1 .o Cross-flow Scott-type double-pass Ryton 15 (fixed) 2.13 10-3 Platinum Setting (% of range) Ions lens Fission products Actinides Bessel box B 45 55 Bessel box P 50 50 Einzel lens El 30 30 Photon stop S2 40 35 as the metal and dissolved in 7 moll-' nitric acid.Its concen- tration was standardized by titration.12 All standard solutions spikes and samples were prepared by dilution by mass in polyethylene bottles. Nitric acid (Merck Suprapur) and Milli-Q water (Millipore Eschborn Germany) were used for all dilutions.Experimental Procedures Sample preparation One pellet (about 30 g) of irradiated U 0 2 fuel and another of mixed uranium-plutonium oxide (MOX) fuel were dissolved (in duplicate) in 7 mol I-' nitric acid in a hot cell facility by reflux and diluted with 4 moll-' nitric acid. A second dilution by mass was performed using 1 moll-' nitric acid and 5 ml of these solutions containing about 100 pg g-' of fuel solution these conditions and Rb Sr and Y could also be eluted with 1 moll-' nitric acid from the CSlO column. Additionally Rb eluted separated from Sr (possible interference between "Rb and 87Sr in the natural elements) and "Sr was separated from 90Zr which was also present in the samples. Quantification procedures The sample spiked sample and spike solution were injected into the chromatograph consecutively.Mass ranges used were 85-90 133-138 234-237 and 239-244 for the separations performed with the CSlO column and 139-160 for the separa- tions using the CS5 column. Detection was carried out by scanning at 1 point per u with a 0.1 s integration time per point with approximately 2.5 s per scan. The data were trans- ferred to a personal computer for further treatment. Integration of the chromatographic peaks was performed using the software GRAMS/386 (Galactic Industries Salem NH USA). The element concentrations in the measured solutions were adjusted so that no detector dead-time correction needed to be applied. The mass discrimination factor was determined using the natural elements spike solution.' The computer code KORIGEN KORIGEN is a dimensionless simulation program i.e.it does not take into account the position of the fuel element in the reactor which calculates the fuel inventory after irradiation based on different data sets which include fuel burn-up (U Pu I4'Nd data) original fuel inventory (235U enrichment initial Pu concentration and isotopic composition) reactor type (neutron flux and energy) and irradiation history (power generated versus time). Also databases on decay characteristics of different nuclides and fission and neutron capture cross- sections are used in the calculations. The accuracy of the predictions is considered to be between 1 and 3% for major actinide isotopes (U Pu) and up to 10% for fission products and minor actinides.' RESULTS Semi-quantitativeData were transferred into a glove-box for further dilution and spiking.A solution containing natural Rb Sr Y Cs Ba La Ce Pr Nd Sm Eu and Gd was prepared by taking aliquots by mass in the glove-box in order t o spike the sample for analysis. For the determination of Np Am and Cm only a Np standard was used. The mass spectrum obtained during the semi-quantitative evaluation of the MOX sample for the analysis of fission products is shown in Fig. l(a) and that for the actinides in Optimization of the isotope dilution procedure In order to optimize the spike addition the concentrations of the isotopes present in the fuels were determined semi- quantitatively based on the response curve of the in~trument.~ For this purpose solutions containing In Tb and Th were added at concentration levels of about 50 ng ml- '.Mass spectra were measured for all samples in the range 80-160 u for fission products and 230-25Ou for actinides using the standard operating conditions. Separation ofjission products and actinides by ion chromatography The separation of the lanthanides was performed using the experimental conditions previously described'' with 0.1 mol 1-1 oxalic acid-0.19 moll-' lithium hydroxide eluent. Under these conditions the lanthanides elute in order of increasing atomic number. The separation of the actinides was carried out using 1 moll-' nitric acid eluent as described previously." The samples were treated with solid silver@) oxide in order to oxidize all plutonium valencies to Puv' and after 5 min the solutions were injected into the chromatograph.It was observed that Cs could be separated from Ba under Fig. l(b). The peaks corresponding to added In and Tb in Fig. l ( a ) and Th in Fig. l(b) were used as internal standards for the evaluation of the concentrations. Indium was used for Rb Sr and Y terbium for Cs Ba La Ce Pr Nd Pm Sm Eu and Gd and thorium for the actinides. The assignment of concentrations to the different fission and actinide isotopes was carried out based on the main fission isotope for the various fission chains.' Similar experiments were performed for the UOz fuel. No matrix interferences were expected at about 100 pg g-' levels." Based on the results obtained the concentration of the multielemental spike solution was adjusted so as to have the same total elemental concentration as the sample for the isotope dilution analysis and standard additions procedures.Determination of the Lanthanides Lanthanides (La to Gd) are eluted in about 25 min from the CS5 column. The separation obtained for the uranium oxide fuel is shown in Fig. 2 for all the isotopes monitored (139- 160u). As can be observed the order of elution was La Ce Pr Nd Pm Sm Eu and Gd as expected. Maximum peak intensities were always lower than 250 000 counts s-'; hence no dead-time correction procedures needed to be applied. All isobaric interferences present in the original sample are elimin- 930 Journal of Analytical Atomic Spectrometry October 1996 Vol. 11v) 80 90 100 110 B 120 130 140 150 160 1 0 230 235 240 245 Mass Fig.1 Semi-quantitative analysis of fission products and actinides in spent nuclear fuel. (a) Mass spectrum of fission products in a MOX fuel with In and Tb (50 ng g-I) added as internal standards. (b! Mass spectrum of the actinides in the same fuel using Th (50 ng g -') as internal standard 0 La I 0 200 400 600 800 lo00 1200 1400 1600 Time/s Fig. 2 IC-ICP-MS trace obtained for the lanthanide elements present in UO fuel using an oxalic acid-LiOH eluent. Mass range monitored was 139-160 ated. The oxide polyatomic ions from '39La 14'Ce 14'Pr and '42Ce which would interfere with most Gd isotopes measured are observed in the chromatogram at the retention time corresponding to the metals. Direct isobaric interferences between Ce and Nd Nd and Sm Pm and Sm and Sm Eu and Gd are also resolved.Chromatograms obtained at selected masses where some isobaric interferences occur are shown in Fig. 3. The separation between 144Ce (half-life 285 d) and 144Nd (decay product) is shown in Fig. 3 ( 4 and similarly Fig. 3(b) shows the separation between 14'Nd and 14%m. This separation is important as it allows the fuel burn-up to be determined by measuring the 148Nd concentration (the fission yield of this isotope is similar for both 235U and 239Pu fission thus reducing uncertainties arising from estimating the relative number of fissions from these two nuclides). Isotope dilution analysis for Ce Nd Sm Eu and Gd was performed as described previously'' using natural elements as spikes. The isotopes 14'Ce 142Nd lS2Sm '"Eu and ls8Gd were used as reference isotopes in the spike while '42Ce 144Nd "'Sm lS3Eu and lS6Gd were the reference isotopes in the sample.The mass discrimination factor',' was determined using the natural elements for the range 140-160 u and it was found to show no long-term variations. Applying linear regression to four independent determinations of the mass 2500 2000 I500 lo00 500 -. I m o IuNd L 0 500 lo00 1500 2 .d . 0 500 lo00 1500 Time/s Fig.3 Single-ion monitoring of the lanthanides in the U02 fuel (cooling time about 3 years) at selected masses where isobaric inter- ferences occur due to neutron capture or parent-daughter pairs. (a) Separation between '44Ce (half-life 285 d) and 144Nd. (b) Separation between 14'Nd (burn-up indicator) and 148Sm Journal of Analytical Atomic Spectrometry October 1996 Vol.11 931discrimination factor on four different days (76 data points) gave a slope the mass discrimination factor per unit mass of -0.0066 (a=0.0015). This value was applied to correct all isotope ratios measured. The method of standard additions was applied to the determination of monoisotopic La and Pr. The concentration of '47Pm was determined assuming an equal response to '47Sm. The results obtained for the uranium oxide fuel are summarized in Table 2 and those for the MOX fuel in Table 3. The results are compared with those obtained by ID-TIMS (some Nd isotopes) gamma spectrometry (154E~ and 144Ce) and with the theoretical calculations using the computer code KORIGEN. Determination of Rb Sr Y Cs Np Am and Cm The separation was performed using the CSlO column and 1 moll-' nitric acid as eluent as described previously.'' Under these conditions the elution order was Rb Cs Sr and Y for the fission products and Np Pu U Am and Cm for the actinides.The resolution between Rb and Sr and Cs and Ba and the differences in isotope abundances from the natural elements allowed the use of natural elements as spikes for the isotope dilution procedure. The intensity at mass 237 during elution for the MOX fuel solutions is shown in Fig.4(a). As can be observed the peak corresponding to 237Np can be resolved from the tailing of the 238U peak although in the mass spectrum it only appears as a shoulder on the 238U peak [see Fig. l(b)]. The separation between Pu and Am for the same fuel is demonstrated in Fig.4(b) showing the response at mass 241. Separation between Np and Pu was not necessary as there is no mass spectral overlap. The separation between U and Pu was observed to degrade with increasing number of injections which was assumed to be due to unspecific adsorp- tion and reduction of Pu on the column as the retention time for Np and U did not change. This could be restored by cleaning the column with 6 moll-' nitric acid for 30 min. Rb Sr and Cs were determined by the method of isotope dilution analysis using the natural elements as spikes as described above for the lanthanides whereas the determination of Y and Np which have only one stable isotope each was performed by the method of standard additions. Am and Cm isotopes were determined assuming the same response as for 237Np.The results obtained for these elements in both samples are summarized in Tables 4 and 5 and compared with those obtained using the standard procedures. DISCUSSION Precision of Duplicate Analyses Based on the data presented in Tables 2-5 the differences between IC-ICP-MS values and those from the other methods observed when analysing the spent fuel samples by ICP-MS can be assessed. For this purpose the mean values were calculated and corrected for the small differences in the concen- tration of the dissolver solutions. The values of the relative deviation for the duplicate analyses showed typical precisions under 10% except for isotopes at very low concentrations and for La and Pr values which were determined by the method of standard additions.The poor results obtained for monoiso- topic lanthanides could not be explained. In general terms the precision seems to be better for the MOX fuel than for the UOz fuel. Typical precision values for the MOX fuel were under 5% which we believe could be a practical value for these types of analyses using chromatographic separation and ICP-MS detection. The precision of ID-TIMS for the determi- nation of Nd isotopes was better than 1% in all instances (duplicate analyses) which is adequate for burn-up measure- ments using 148Nd. Accuracy of the Concentrations Measured It is difficult to check the accuracy of a new procedure when there is no reference material available for the type of analysis Table 2 Inventory of the lanthanide elements in uranium oxide spent fuel dissolver solutions (atoms x 10I6 g-l) and comparison with TIMS gamma spectrometry and KORIGEN Dissolver solution 1 Dissolver solution 2 Element La Ce Pr Nd Pm Sm Eu Gd Mass 139 140 142 144 141 142 143 144 145 146 148 150 147 147 148 149 150 151 152 154 153 154 155 154 155 156 158 160 Gamma ICP-MS TIMS spectrometry KORIGEN 4.82 - - 6.27 - 6.66 6.77 - 5.71 5.92 0.04 - 0.06 0.05 - 5.63 5.81 - 0.26 0.20 - 2.46 2.64 7.54 - - 8.05 2.87 2.92 - 2.89 3.76 3.77 - 4.01 1.81 1.87 - 1.84 0.92 0.95 - 0.93 - - 0.18 0.20 0.61 - - 0.51 0.99 - - 0.87 0.002 - - 0.01 1.44 - - 1.48 - 0.05 0.03 - 0.50 0.41 - 0.21 0.19 - 0.44 0.54 0.09 - 0.16 0.20 - 0.06 0.03 - 0.to 0.05 0.02 - - 0.04 1.03 0.97 - 0.22 0.13 0.01 - - 0.007 - - - - - - - - - - - - - - Gamma ICP-MS TIMS spectrometry KORIGEN 6.24 4.69 6.62 6.77 5.69 6.72 0.04 - 0.05 0.05 - 5.61 4.45 - 0.26 0.24 2.45 3.04 8.01 8.53 3.30 2.88 - 2.87 4.34 3.71 - 3.99 2.08 1.86 - 1.83 1.05 0.96 - 0.93 - 0.18 0.16 - 0.5 1 0.76 - 0.87 1.16 - - 0.01 0.007 - - 1.48 1.64 - - 0.05 0.04 - 0.49 - 0.21 0.48 0.22 0.44 0.52 0.09 - 0.07 0.20 - 0.06 0.02 - 0.10 0.05 0.04 0.02 1.03 0.99 - 0.22 0.14 - - 0.007 0.01 - - - - - - - - - - - - - - - - - - - - - - - - - - 932 Journal of Analytical Atomic Spectrometry October 1996 Vol.11Table 3 Inventory of the lanthanide elements in mixed oxide spent fuel dissolver solutions (atoms x 10l6 g-I) and comparison with TIMS gamma spectrometry and KORIGEN Dissolver solution 1 Dissolver solution 2 Gamma spectrometry Gamma spect rome t ry - - - 0.04 - - Element La Ce Mass 139 140 142 144 141 142 143 144 145 146 148 150 147 147 148 149 150 151 152 154 153 154 155 154 155 156 158 160 ICP-MS 6.79 8.27 7.41 0.02 6.80 0.12 5.62 7.70 4.30 4.58 2.59 1.51 0.28 1.31 1.22 0.02 2.14 0.13 0.81 0.39 1.01 0.19 0.01 0.12 0.03 0.86 0.20 0.006 TIMS KORIGEN 7.98 7.93 6.86 0.04 7.12 0.11 5.15 6.79 3.93 4.16 2.41 1.40 0.33 1.14 0.96 0.03 2.01 0.15 0.88 0.38 0.77 0.33 0.11 0.17 0.09 0.67 0.22 0.02 ICP-MS 8.43 8.24 7.41 0.03 8.54 0.14 5.86 7.85 4.47 4.77 2.74 1.59 0.3 1 1.27 1.17 0.03 2.1 1 0.12 0.78 0.39 1.08 0.22 0.04 0.13 0.04 0.93 0.22 0.006 TIMS KORIGEN 7.96 7.91 6.84 0.04 7.11 0.11 5.14 6.77 3.92 4.14 2.40 1.40 0.33 1.14 0.96 0.03 2.00 0.15 0.88 0.38 0.77 0.33 0.1 1 0.17 0.09 0.66 0.22 0.02 - 0.03 Pr Nd - 3.94 4.23 2.48 1.43 - 3.89 4.18 2.42 0.32 Pm Sm Eu Gd - 0.52 - Table 4 Inventory of Rb Sr Y Cs Np Am and Cm isotopes in uranium oxide spent fuel dissolver solutions (atoms x loi6 g-I) and comparison with TIMS gamma spectrometry and KORIGEN Dissolver solution 1 Dissolver solution 2 Gamma spectrometry Gamma TIMS spectrometry Element Rb Sr Mass 85 87 86 88 90 89 133 134 135 137 237 24 1 243 244 ICP-MS 1.18 2.45 0.03 3.16 4.01 3.84 6.08 0.34 2.17 6.77 1.16 0.37 0.57 0.37 TIMS KORIGEN 0.83 1.87 0.007 2.53 3.50 3.34 5.35 0.32 1.64 6.05 1.47 0.55 0.74 0.40 ICP-MS 0.92 2.05 0.03 3.25 4.15 3.61 5.56 0.36 2.07 6.48 1.43 0.38 0.65 0.41 KORIGEN 0.82 1.86 0.007 2.52 3.48 3.32 5.33 0.32 1.63 6.02 1.46 0.55 0.73 0.40 Y c s - 0.30 6.25 - - NP Am - 0.49 0.71 0.48 Cm performed.However in this instance the results can be com- pared with those obtained by two independent methods for some isotopes and with the computer predictions for 21.11 the isotopes measured which can be considered to be accurate to about 10% for well-characterized reactors.' It must be borne in mind that the computer predictions are based on measure- ments of the total U total Pu total Am and Cm and I4*Nd concentrations which in our case were taken from the ID-TIMS results.Figs. 5 and 6 show the comparison of ICP-MS measurement data and KORIGEN predictions for the UOz and MOX fuels respectively. The data for the duplicate analyses are included in the graphs. As cxi be observed for both fuels there is a systematic difference between the obtained and calculated concentrations of about 10 -15% for virtually all elements. Hence the ICP-MS determined fuel burn-ups are about 10-15% higher than the values introduced into the computer code in order to calculate the fuel inventories (the fission product inventory increases linearly with the fuel burn-up). If we compare the results presented for Nd isotopes in Tables 2 and 3 by ICP-MS and ID-TIMS it can be seen that except for one solution the results are always 5-1570 higher by ICP-MS.Because the 14*Nd concentrations measured by ID-TIMS were used to determine the fuel burn-up the differences between ICP-MS and ID-TIMS for Nd will be reflected throughout the comparison of ICP-MS measurements with KORIGEN values. The results obtained by gamma spectrometry for 144Ce and IS4Eu are typically higher than those obtained by ICP-MS and for 144Ce are in closer agreement with the KORIGEN predictions.However for ls4Eu there is a large difference between ICP-MS and gamma spectrometry with the computer predictions far from both results. The results obtained for 134Cs and '37Cs presented in Journal of Analytical Atomic Spectrometry October 1996 Vo!. 11 933v) 4 20000 B 15000 iij 10000 SO00 - I G o B 0 100 200 300 400 500 600 16000 14000 12000 10000 8000 6000 4000 2000 0 0 200 400 600 800 1000 1200 Time/s Fig.4 Single-ion monitoring of the actinides in the U02 fuel at selected masses where isobaric interferences occur. (a) Separation between 237Np and 238U detected at mass 237. (b) Separation between 2 4 r P ~ and 241Am Tables 4 and 5 show a much better agreement.For the UOz fuel (Table 4) the 13'Cs values obtained by ICP-MS are 4 and 8% higher than by gamma spectrometry. However for the MOX fuel in Table 5 the results are 5% higher in one instance and 11% lower in the other. For both techniques the 13'Cs t a 35 30 25 20 15 10 5 0 0 10 20 30 40 KORIGEN data Fig. 5 Comparison between the computer predictions and the exper- imental values for the U02 fuel. Duplicate analyses are shown with open and filled circles. The line corresponds to a slope of 1 30 25 20 E l5 h 10 5 0 10 I5 20 25 30 0 5 KORIGEN data Fig. 6 Comparison between the computer predictions and the exper- imental values for the MOX fuel. Duplicate analyses are shown with open and filled circles. The line corresponds to a slope of 1 Table 5 Inventory of Rb Sr Y Cs Np Am and Cm isotopes in mixed oxide spent fuel dissolver solutions (atoms x 10l6 g-') and comparison with TIMS gamma spectrometry and KORIGEN Dissolver solution 1 Dissolver solution 2 Element Mass Rb 85 87 Sr 86 88 90 Y 89 c s 133 134 135 137 NP 23 7 Am 24 1 243 Cm 244 ICP-MS 0.84 1 .a0 0.03 2.77 3.35 2.80 8.81 0.28 5.25 9.10 0.87 3.86 3.30 2.00 Gamma TIMS spectrometry KORIGEN 0.72 1.54 0.003 2.03 2.76 2.63 8.20 0.23 4.40 7.94 0.58 4.56 3.88 1.43 ICP-MS 0.79 1.67 n.d.2.78 3.40 2.94 8.90 0.3 1 5.12 8.98 0.85 4.2 1 3.26 2.01 Gamma TIMS spectrometry KORIGEN 0.72 1.53 0.003 2.02 2.75 2.62 8.18 0.23 4.39 7.92 0.57 4.53 3.87 1.42 934 Journal of Analytical Atomic Spectrometry October 1996 Vol. 1 1results are always higher than the computer predictions.On the other hand the data for 134Cs obtained by gamma spec- trometry show better agreement with the computer predictions than those obtained by ICP-MS. From the data presented in Tables 4 and 5 for Np Am and Cm it can be observed that the values obtained by ID-TIMS for Am and Cm are about 10-30% higher than those otbtained by ICP-MS in contrast to Nd which showed the opposite effect for three of the solutions analysed. The agreement with the predicted concentrations for Am isotopes is better for TIMS in the UO fuel but worse for the MOX fuel. For 244Cm ICP-MS values are closer to the predicted concentrations. The Np concentrations found are lower than predicted for the U 0 2 fuel but higher for the MOX fuel. From the data shown in Tables 2-5 it is difficult t o draw clear conclusions about the accuracy of the proposed rnethod for the determination of fission products in spent fuel s:,mples.The data obtained by gamma spectrometry are not conclusive but seem to indicate a higher burn-up than that calculated based on 14*Nd by TIMS. In summary the accuracj of the proposed method seems to be on average better thaln 15% (assuming that the TIMS results for Nd are correct) which for the purposes of high burn-up studies or new fuel dcvelop- ments for transmutation can be considered to be acccptable considering the large amount of data that can be plovided and the reduced radiation hazard for the operator in ccimpari- son with existing techniques. Accuracy of the Isotope Ratios For most elements the isotope ratios measured by I(:P-MS and those predicted by KORIGEN are in close agreement (data can be extracted from Tables 2-5).Minor differences can be explained by small changes in the neutron capture cross-sections of the fission nuclides with neutron entsrgy or by experimental errors. However significant differences were found for Eu and Gd isotopes in all samples which carinot be explained by measurement errors. These differences [can be attributed to errors in the computer-assigned neutron apture cross-section for lS3Eu. If the neutron capture cross-sect ion for lS3Eu is lower than assumed the formation of 154E~. which does not occur by direct fission and lssEu will be rt'aduced which was observed experimentally. Similarly the c oncen- trations of 154Gd and "'Gd which are decay productil of the Eu isotopes will be lower than predicted thus explainjmg the differences observed also for Gd.It is clear that the direct measurement of isotope ratios of fission elements by I(:P-MS can help in the refining of neutron capture cross-sectioiis. CONCLUSIONS ICP-MS has been applied to the characterization 01' spent nuclear fuels with IC separation and the method of isotope dilution analysis o r that of standard additions. The results obtained for duplicate measurements of two different fut:l types showed that the precision attainable was better than I.()% for most nuclides measured. In comparison with other techniques ICP-MS is capable of providing data on a larger number of isotopes both short medium long-lived and stable which will help in the extrapolation of the current prediction techniques based on KORIGEN or similar computer codes to higher burn-ups or to experimental fuels for transmutation purposes.ICP-MS yields more precise measurements than gamma spec- trometry and the ICP-MS data fit better to the general trend observed for other isotopes. However the precision of ICP-MS measurements after IC is worse than that attainable by TIMS. Also discrepancies of up to 15% between ICP-MS and TIMS for Nd isotopes could not be explained. These discrepancies resulted in differences between the observed and computer- predicted fuel inventory of a similar magnitude. Similar discrep- ancies were observed between TCP-MS and TIMS for Am and Cm isotopes but in opposite directions which rules out the existence of dilution errors in the ICP-MS methodology.The isotope ratios determined by ICP-MS showed good agreement with the predicted values. This allows the correction of estimated neutron capture cross-sections for those nuclides where the discrepancies between both techniques were too large to be due only to experimental errors. This could help in establishing better databases for burn-up-dependent neutron capture cross-sections and to extrapolate current data for different reactor operating conditions or for non-commercial nuclear fuels. REFERENCES 1 2 3 4 5 6 7 8 9 10 11 12 Fischer U. and Wiese H. W. Verbesserte konsistente Berechnung des nuklearen Inventars abgehrannter D WR-Brennstofe auf der Basis von Zell-Abbrand- Verfahren mit KORIGEN Kernfors- chungszentrum Karlsruhe Report 3014 1983.Garcia Alonso J. I. Babelot J.-F. Glatz J.-P. Cromboom O. and Koch L. Radiochim. Acta 1993 62 71. Garcia Alonso J. I. Thoby-Schultzendorff D. Giovanonne B. Koch L. and Wiesmann H. J . Anal. At. Spectrom. 1993 8 673. Garcia Alonso J. I. Thoby-Schultzendorff D. and Koch L. Proceedings of the 15th Annual Symposium on Safeguards and Nuclear Material Management European Safeguards Research and Development Association (ESARDA) Report 26 EUR 15214 EN Ispra 1993 pp. 485-489. Garcia Alonso J. I. Thoby-Schultzendorff D. Giovanonne B. Glatz J.-P. Pagliosa G. and Koch L. J. Anal. At. Spectrom. 1994 9 1209. Garcia Alonso J. I. Garcia Serrano J. Babelot J.-F. Closset J.-C. Nicolaou G. and Koch L. in Applications of Plasma Source Mass Spectrometry II eds. Holland G. and Eaton A. N. Royal Society of Chemistry Cambridge 1993 p. 193. Betti M. Garcia Alonso J. I. Arbore Ph. and Koch L. in Applications of Plasma Source Mass Spectrometry I I ed. Holland G. and Eaton A. N. Royal Society of Chemistry Cambridge 1993 p. 205. Garcia Alonso J. I. Anal. Chim. Acta 1995 312 57. Byrdy F. A. and Caruso J. A. Environ. Sci. Technol. 1994 28 528A. Braverman D. S. J. Anal. At. Spectrom. 1992 7 43. Garcia Alonso J. I. Sena F. Arbore Ph. Betti M. and Koch L. J . Anal. At. Spectrom. 1595,10 381. Cromboom O. Garcia Alonso J. I. Koch L. Goerten J. Roesgen E. Wagner H. G. Ottmar H. and Eberle H. Proceedings of the 4th International Conference on Facility Operations-Safeguards Interface American Nuclear Society IL 1992 pp. 431. Paper 6/01 290F Received February 22 1996 Accepted June 13 1996 Journal of Analytical Atomic Spectrometry October 1996 Vol. 11 935

ISSN:0267-9477    

DOI:10.1039/JA9961100929

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Optimization of Secondary Cathode Thickness for Direct Current Glow Discharge Mass Spectrometric Analysis of Glass Journal of Analytical Atomic Spectrometry WIM SCHELLES STEFAN DE GENDT AND RENE E. VAN GRIEKEN University of Antwerp (UIA) Department of Chemistry Univerciteitsplein 1 B-2610 Antwerpen Belgium Direct current glow discharge mass spectrometry can be used for the analysis of solid non-conducting samples by application of the secondary cathode technique. In the work reported the thickness of this secondary cathode a conducting diaphragm placed in front of the sample was evaluated. Variation of the secondary cathode thickness results in a variation of the electrode configuration and of the discharge voltage required for stable atomization of the non-conductor. Both parameters were evaluated separately using conducting samples.Thickness and voltage were found to have an opposite effect on the sample signal intensity obtained and on the crater shape created. Generally it could be concluded that for the determination of elements at low concentrations in glass plates a thin secondary cathode is preferable but that the use of a thick secondary cathode results in a flatter crater profile which is important for depth profiling. Keywords Glow discharge mass spectrometry; non-conductor analysis secondary cathode Glow discharge mass spectrometry (GDMS) has already proved to be an interesting tool for the direct (ultra) trace analysis of bulk solid samples.'-3 Because the sample is !,putter- consumed in a controlled way by the plasma the technique also allows depth ~ r o f i l i n g .~ - ~ However because the sample acts as a cathode in a low pressure discharge the application of GDMS seems a priori to be restricted to conducting and semi-conducting materials. Nevertheless non-conducting solid samples can be directly analysed by using either an rf powered source' or a dc plasma using the secondary cathode technique.' Mixing with a conducting host is another possibility for the analysis of non-conducting samples with a dc device hut this requires a grinding step which can cause serious contaminati~n.~-" The use of the rf mode is very appealing because solid non- conductors such as glass plates can be directly analysed without exposing an auxiliary conductor to the discharge. The application of rf GDMS has already led to promising and impressive res~lts.'~*'~ However no commercial rf GDMS equipment is yet available although coupling with commer- cial mass spectrometers has already been successfully perf~rmed.'~-'~ As an alternative it is also possible to analyse solid non- conducting materials with dc GDMS by means of a secondary cathode (or mask).The sample under examination is exposed to the discharge through the aperture of this conducting diaphragm. Owing to the continuous in situ .;putter- redeposition of the secondary cathode material a thin con- ducting layer is formed on the non-conducting sample. Bombarding particles can penetrate this layer and consequently atomize the non-conducting sample. The successful use of this method relies on the appropriate thickness of the conducting redeposited coating; a too thin or incomplete layer will cause discharge instabilities whereas a too thick layer prevents the bombarding particles from reaching the non-conducting sample.The main disadvantages of this method are therefore the critical choice of the discharge conditions (determining the building of the conducting layer) and the introduction of possible blank values caused by the exposure of the secondary cathode to the glow di~charge.'~*~' There are on the other hand also certain advantages; for example for the analysis of flat non-conducting samples the sample thickness does not influence the sample signal intensity obtained as is the case for rf GDMS.I2 Moreover the secondary cathode technique has been shown to offer good precision (c 10% RSD for both internal and external precision) and sub-ppm limits of detection for most elements and matrices.It has already been successfully applied to the bulk analysis of various sample types for example glass plates polycarbonate Teflon aluminium oxide and Ma~or.'~-,~ To date the thickness of the secondary cathode mostly 0.25 mm has always been chosen in an arbitrary way. However the cathode-anode distance in an analytical glow discharge is mostly of the order of only 0.5 mm. It is therefore not unreasonable to assume that changing from for example a 0.25 to a 1 mm thick secondary cathode will significantly affect the discharge conditions and the results obtained. In this work the thickness of the secondary cathode was evaluated for the analysis of glass. Attention was focused on the sample s i g i ~ l intensity obtained and the sample crater profiles created.EXPERIMENTAL SECTION Glow Discharge Mass Spectrometer A VG9000 double focusing glow discharge mass spectrometer (VG Elemental Winsford Cheshire UK) was used; it has been described in detail elsewhere.24 The working resolution (5% peak height) was about 4000. A dual detection system con- sisting of a Daly detector and a Faraday cup can provide a dynamic range of about 9 orders of magnitude (1 x 10-'*-1 x lo-' A). The 'new' flat cell was used; this has been described elsewhere.,' The sample holder front plate on anode potential had a 7.5 mm diameter hole. The cell was cryo-cooled to reduce the background caused by residual gases.The glow discharge was supported by high-purity argon (Air Liquide Lyon France; 99.9997%). Materials The non-conducting sample used was a common Si0,-based glass plate. Because of the methodological scope of this work knowledge of the exact composition was not required. The secondary cathodes were all made of 99.9% pure tantalum (Sigma-Aldrich Bornem Belgium). The conducting samples used were either made of the same tantalum or of 99.5% aluminium (Advent Research Materials Halesworth UK). Journal ojAnalytica1 Atomic Spectrometry October 1996 Vol. 11 (937-941) 937RESULTS AND DISCUSSION Influence of the Secondary Cathode Thickness on the Sample Signal Intensity First identical glass samples were measured using secondary cathodes with different thicknesses 0.12 0.25 0.5 1.0 and 2 mm.All the tantalum diaphragms had a 4mm diameter aperture and the anode hole size was 7.5 mm as determined by the choice of the sample holder front plate. As demonstrated previously the possible discharge conditions for effective atom- ization of the glass are restricted.” The discharge current was set at 3mA and the discharge voltage was chosen according to the optimum and stable glass signal intensity. The results are shown in Table I. Secondary cathodes with different thick- nesses were used immediately after each other to restrict instrumental influences. For each measurement a thorough optimization of the ion beam was performed in order to make a comparison of the absolute signal intensities useful. Afterwards the experiment was repeated to check the reproduc- ibility.Both series revealed clearly comparable trends but the absolute signal intensities were slightly different for both due to instrumental parameters. For this reason the results reported in Tables 1-3 are those obtained with only one set of consecutive measurements. It is important to note that the use of the 2mm thick secondary cathode always (i-e under all discharge conditions and anode diameters used) caused an unstable discharge after approximately 45 min; also a small central spot of redeposited tantalum on the glass could be observed after removal of the sample. Hence a 2mm thick tantalum secondary cathode appears to be excluded for practi- cal use. Nevertheless the values for this thickness (obtained before the instabilities were observed) are reported because they confirm the observed trends.The effect of the secondary cathode thickness on the optimum discharge voltage the absolute sample signal intensity and the ‘weight’ of the blank values” will now be discussed based on the results obtained listed in Table 1. It is clear that the required discharge voltage for stable atomization of the glass increases with increasing thickness of the secondary cathode. This can be understood as a larger amount of tantalum is exposed to the discharge when a thicker secondary cathode is used. Hence there is more sputtered tantalum in the discharge whereas the non-conducting area to be covered remains the same. An excess of redeposition would result in a thick tantalum coating on the glass and would consequently prevent its atomization.A decrease in relative redeposition (expressed as the fraction of redeposited atoms of the total amount of sputtered atoms) can be achieved by decreasing the discharge pressure thereby increasing the discharge voltage when working in a constant-current mode. This simple explanation is indeed confirmed by the results obtained. In this context it is also worth mentioning that not only is the area of the secondary cathode that is exposed to the discharge important but also the distance to the non- conducting surface to be covered. This is indicated by the following outcomes. If a 0.25 mm thick tantalum mask in combination with a 10 mm anode diameter is used an area of 69.1 mm2 of tantalum is exposed to the discharge. It has previously been shown that with this configuration and appro- priate discharge conditions (3 mA 0.9 kV) it is possible to atomize a glass sample in a stable way.” On the other hand the present work shows that the use of 2mm thick tantalum secondary cathode in combination with for example a 5 mm anode diameter (only 32.2mm2 of tantalum exposed to the discharge) always results in an excess of redeposited tantalum regardless of the discharge conditions used. The only possible explanation for this result is that a remote area of the sputtered secondary cathode has a much lower contribution or none at all to the necessary redeposition on the non-conducting sample.This is not surprising as it is known that most sputtered atoms are thermalized very close ( < 1 mm) to the cathode,26 and thus that redeposition of atoms on the sample occurs close to the place where they are sputtered.Despite the increased optimum discharge voltage for thicker Table 1 Measured influence of the secondary cathode thickness on the sample signal intensity obtained Secondary cathode Sample-anode Discharge voltage/ Signal intensity Relative intensity Signal intensity ratio 0.12 0.62 0.65 4.6 1.07 22 0.25 0.75 0.75 4.3 1 .oo 29 0.50 1 .o 0.85 3.4 0.79 48 1 .o 1.5 1.05 3.0 0.70 200 2.0 2.5 1.20 0.13 0.03 3500 thickness/mm distance/mm kV of 28Si/10- “A of 28Si 18 1 ~ ~ 2 8 s i Table 2 discharge voltage and the sample-anode distance. The reference secondary cathode thickness is 0.25 mm. Results for tantalum Calculated influence of the secondary cathode thickness on the sample signal intensity obtained by combination of the influence of the Secondary cathode Sample-anode Relative intensity Relative intensity t hickness/mm distancejmm (voltage determination) (distance determination) Combination 0.12 0.62 0.62 1.18 0.73 0.25 0.75 1 .oo 1 .oo 1 .oo 0.50 1 .o 1.38 0.73 1 .oo 1 .o 1.5 2.14 0.39 0.83 2.0 2.5 2.’7 1 0.1 1 0.30 Table 3 Calculated influence of the secondary cathode thickness on the sample signal intensity obtained by combination of the influence of the discharge voltage and the sample-anode distance.The reference secondary cathode thickness is 0.25 mm. Results for aluminium Secondary cathode Sample-anode Relative intensity Relative intensity t hickness/mm distance/mm (voltage determination) (distance determination) Com bination 0.12 0.62 0.57 1.13 0.64 0.25 0.75 1 .oo 1 .oo 1 .oo 0.50 1 .o 1.43 0.79 1.13 1 .o 1.5 2.30 0.49 1.13 2.0 2.5 2.94 0.19 0.56 938 Journal of Analytical Atomic Spectrometry October 1996 Vol.11secondary cathodes the absolute signal intensity of 28Si rep- resentative of the glass sample is seen to decrease slightly (see Table 1). Moreover the ratio between the signal intensity of '"Ta relative to that of 28Si increases for thicker secondary cathodes. This ratio is representative for the 'weight' of the blank values in the tantalum; it determines the 'apparent' blank va1~es.l~ For a better understanding of the results the influences of the discharge voltage and of the sample-anode distance on the sample signal intensity were evaluated separ- ately.These measurements were performed with Conducting samples to avoid the inevitable link between both parameters when a secondary cathode is used. The purpose of the first set of measurements was tct deter- mine the relative signal intensities as a function of the cathode- to-anode distance. These measurements were performcd with tantalum and aluminium plates. The cathode-anode distance was varied by changing the thickness of the isolating Teflon spacer between the two electrodes. Constant discharge con- ditions were used for all measurements 3 mA 0.75 bV. The sample signal intensity in this instance represented by the '"Ta or "Al signal was measured for four different cathode- anode distances. Based on these data an exponential plot was fitted as represented in Fig.1. The signal intensities for other cathode-anode distances could be calculated by mcans of correlation equations (where I is the signal intensity iri A and D is the distance between the sample and the anode) For Ta 1=(1.01 x 10-9)-1.260 ( 1 ) 1 = ( 3 . 8 9 1 0 - 9 - 0 . 9 4 0 (2) For Al An increased cathode-anode distance clearly results in a decreased sample signal intensity. Although it is difficult to establish the exact cause of this result (it is probably a combination of differences in sputtering redeposition diffusion and ionization processes) some qualitative reasoning 1s given here. First the increased distance between the sample and the anode also implies a longer path for sputtered sample particles to reach the exit slit where the ions are extracted to the mass spectrometer. The longer this path the greater the chawe that ions and atoms travelling through the cell are lost because of radial diffusion towards the cell body. Although the results obtained meet this theoretical explanation it has to be con- sidered that the relative variation of the sample-slit distance is fairly small.Another probably more plausible cause can be found in variations of the kinetic energy of the particles bombarding the glass sample. Because a constant discharge 6.OE- 1 0 5.OE-10 \ x = V) 4.OE-10 C C - 3.OE-10 0 0) c c iz 2.OE-10 1.OE-10 !\ Ta \ \ \ \ \ \ \ * \ \ ,A1 \ \ \ \ * \ \ \ \ \ .-a \ O.OE+O Cathode-anode distance / rnm Fig. 1 Influence of the cathode-anode distance on the sample signal intensity measured for tantalum and aluminium plates voltage is used for the whole set of measurements the maximum kinetic energy with which the ions (and indirectly also the fast neutrals) impinge on the sample surface is also constant.However an increased cathode-anode distance results in an increased chance of collisions for ions and neutrals on their path towards the sample surface. Increasing the cathode-anode distance can therefore be considered qualitatively as loss of bombarding energy. This can probably also be considered as one of the main reasons for the increasing '8'Ta:28Si signal intensity ratio with increasing secondary cathode distance as shown in Table 1. The tantalum area exposed to the discharge remains the same but owing to the longer path and conse- quently the larger number of collisions the ions impinging on the glass surface will have lost more kinetic energy than those bombarding the secondary cathode.Summarizing it can be concluded that a thin secondary cathode is preferable to reach low detection limits as a consequence of both the higher sample signal intensity and the lower blank contribution. In a second set of measurements with conducting samples (again with tantalum and aluminium plates as samples under investigation) the relative influence of the discharge voltage on the sample signal intensity obtained was evaluated. The cathode-anode distance ( 1 mm) and the discharge current ( 3 mA) were kept constant whereas the discharge voltage was varied from 0.65 to 1.20kV. In this instance a linear plot could be fitted through the experimentally obtained points in the measured range as illustrated in Fig.2. The correlation equations were found to be as follows (where I is the signal intensity in A and V is the discharge voltage in kV used) For Ta I = 1.39 x lO-'V- 6.8 x lo-'' (3) I=7.46 x 1 0 - 9 V - 3 . 9 x lo-'' (4) The general outcome is not surprising since it is known that an increased voltage results in an increased sample signal intensity at least when dealing with conducting samples in a restricted range. It is however obvious that the above- mentioned linear equations should not be extrapolated especially not towards lower voltages as threshold effects may occur. After evaluation of the effect of the sample-anode distance and the discharge voltage on the observed sample signal intensity both parameters were combined.This was performed by calculation and multiplication of relative signal intensities. The data for a 0.25 mm thick secondary cathode were used as for Al 1.OE-9 i 4 0) 6.OE-10 c L I / - - 0 &4.OE-10 i7i 2.OE- 1 0 / / / Ta / / / / f / / / / f Al ./' Discharge voltage / kV Fig. 2 Influence of the discharge voltage (in the constant-current mode) on the sample signal intensity measured for tantalum and aluminium plates Journal of Analytical Atomic Spectrometry October 1996 Vol. 11 939reference values because 0.25 mm is the standard thickness a secondary cathode are restricted. The previously suggested used to date. The results are presented in Tables 2 and 3 for both tantalum and aluminium. For the sake of clarity one example based on the tantalum data is illustrated here.If only the effect of the increased distance is considered [using eqn. (l)] the use of a 1 mm thick mask would result in a signal intensity that is only 39% (0.39 in Table 2) of the signal intensity obtained with a 0.25 mm mask. On the other hand considering the corresponding discharge voltage used (see Table 1 column 3) a signal intensity that is a factor of 2.14 higher than the reference value should be expected for the 1 mm thick mask as determined from eqn. (3). Combining both effects (0.39 x 2.14) results in a calculated sample signal intensity of 0.83 times the reference intensity. The combined data calculated in this way are more or less in agreement with the experimentally obtained data for the glass analysis for a secondary cathode thickness in the range 0.12-1 mm both the calculated and the experimental data reveal only slight variations of the sample signal intensity obtained.Influence of the Secondary Cathode Thickness on the Sample Crater Profile Owing to the continuous sputter-consumption of the sample GDMS also allows depth profiling. In order to obtain a good depth resolution a flat crater bottom is de~irable.~-~ The crater shapes obtained for the glass samples under a 3 mA discharge current and the discharge voltages mentioned above (see Table l) were measured using a mechanical profilometer (Dektak 3030 Veeco Plainview NY USA). The results are shown in Fig. 3. A significant effect can be seen with increasing secondary cathode thickness and thus with increasing dis- charge voltage the concave character decreases. However it is clear that the best crater shape (for the 1 mm thick secondary cathode) is still not really flat.For conducting samples it has been demonstrated that variation of the discharge conditions allows variation of the crater With increasing dis- charge voltage in the constant current mode the concave shape of the crater bottom disappears and possibly changes into a convex shape. However the discharge conditions to achieve stable atomization of a non-conducting sample using 0 2 6 8mm Fig.3 Influence of the secondary cathode thickness on the crater shapes created after 1 h of sputtering (note the difference in vertical and horizontal scales). (a) 0.12 mm thick 0.65 kV/3 mA (b) 0.25 mm thick 0.75 kV/3 mA (c) 1 mm thick 1.05 kV/ 3 mA idea that a completely flat crater bottom- cannot be obtained for certain samples with .this method,Ig is thus confirmed.The crater shape is known to depend not only on the discharge conditions but also on the electrode geometry In order to evaluate this effect simulations of the ion trajector- ies were performed using the SIMION program.28 The results are represented in Fig. 4; a complete overview of the configur- ation used has been described elsewhere.* SIMION allows one to calculate the electric field strength between two electrodes with a different potential and the resulting ion trajectories. Bombarding ions focused towards the sample surface create a convex crater bottom whereas defocused ions cause a concave crater bottom.Only a homogeneous current distri- bution perpendicular to the sample surface will result in a flat crater bottom and thus in a good depth res~lution.~ These ion trajectories are simulated for a thin and a thick secondary cathode; the schematic representations are depicted in Fig. 4. The curved line on anode potential (lOOOV) represents the cathode dark space-negative glow interface region; this virtual electrode is drawn because the so-called cathode fall almost equals the complete discharge voltage drop. The shape and position of this interface from where the ions are accelerated have been calculated previously for the VG9000 flat cell geometry by means of a three-dimensional mathematical Although these calculations were performed for a 3 mA/1.0 kV discharge the results are used in this methodologi- cal study because it aims more at trends rather than at absolute results.The surface of the non-conducting sample exposed to the discharge is also represented by an electrode on cathode potential (OV) because of the thin conducting layer on the glass sample. The simulation indicates that the thickness of the secondary cathode is an important factor. The thicker the secondary cathode the larger the deviation of the electrode configuration from a set of two infinite parallel electrodes. The use of two parallel electrodes implies completely perpendicular ion trajectories and would thus result in an ideal rectangular crater shape. Therefore from the electric field point of view a thin secondary cathode seems to be preferable to obtain a flat crater bottom.A thick secondary cathode results in a defocus- ing of the bombarding ions and thus in principle in a concave crater profile. Both the theoretical effect of the discharge conditions and the electrode configuration can now be combined and com- pared with the real crater shapes obtained with different secondary cathode thicknesses. According to the results of the SIMION simulations a thicker secondary cathode causes a concave crater bottom. An increasing discharge voltage (in the Fig.4 Ion trajectories simulated with SIMION for a thin and a thick secondary cathode. 1 Glass sample; 2 tantalum secondary cathode; 3 anode cell body; 4 virtual electrode on anode potential (to simulate the cathode dark space-negative glow interface); and 5 ion trajectories 940 Journal of Analytical Atomic Spectrometry October 1996 Vol.1 1constant-current mode) results however in a more convex crater profile as demonstrated for conducting samples as well as for glass plates.4* l9 The use of a thick secondary cathode implies a higher discharge voltage thus opposite effects (elec- trode configuration uersus discharge voltage) are involved. In practice it is seen that the effect of the discharge voltage plays a major role since the concave character decreases with increasing discharge voltage (see Fig. 3). Summarizing. one could conclude that the different discharge conditions inherent to a varying secondary cathode thickness rather than the thickness itself have a pronounced effect on the crater t;,hape created.CONCLUSION The thickness of the secondary cathode has in the obst:rved range from 0.12 to 1 mm only a minor effect on the sample signal intensity obtained. However the ‘weight’ of the blank values due to the sputtering of the secondary cathode is more pronounced for thick secondary cathodes. Therefore the use of a thin secondary cathode (0.12-0.25 mm thick) is preferable to reach low limits of detection and should therefori be chosen for bulk analysis of non-conducting materials. The thickness of the secondary cathode also influences indirectly the crater shape obtained. For glass and for the reported discharge conditions a thick secondary cathode (1 mml was found to favour a flat crater. However at present rf GDMS seems to be more appealing for depth analysis because 4 if the restricted discharge conditions when using a secondary cathode.This work was partly funded by the Belgian ‘Nationaal Fonds voor Wetenschappelijk Onderzoek’ (NFWO) to whom one of the authors (S.D.G.) also acknowledges financial support. W.S. is financially supported by the ‘Vlaams Instituut voor de bevordering van het Wetenschappelijk-technolcigisch Onderzoek in de Industrie’ (IWT). The authors thank 8:. De Cauwsemaecker and R. Saelens for technical support. REFERENCES 1 Harrison W. W. Barshick C. M. Klinger J. A. Ratliff P. H and Mei Y Anal. Chem. l990,62,943A. 2 Harrison W. W. J. Anal. At. Spectrom. 1992 7 75. 3 King F. L. and Harrison W. W. in Glow Disi,harge Spectroscopies ed. Marcus R. K. Plenum Press New York 1993 ch.5 pp. 175-214. 4 Raith R. Hutton R. C. and Huneke J. C. J. Anal. At. Spectrom. 1993 8 867. 5 Bengtson A. Spectrochim. Acta Part B 1994 49 411. 6 Jakubowski N. and Stuewer D. J. Anal. At. Spectrom. 1992 7 951. 7 Coburn J. W. and Kay E. Appl. Phys. Lett. 1971 19 350. 8 Milton D. M. P. and Hutton R. C. Spectrochim. Acta Part B 1993 48 39. 9 Tong S. L. and Harrison W. W. Spectrochim. Acta Part B 1993 48 1237. 10 Woo J. C. Jakubowski N. and Stuewer D. J. Anal. At. Spectrom. 1993 8 881. 11 De Gendt S. Schelles W. Van Grieken R. and Muller V. J. Anal. At. Spectrom. 1995 10 681. 12 Marcus R. K. Harville T. R. Mei Y. and Shick C. R. Jr. Anal. Chem. 1994,66 902A. 13 Winchester M. R. Duckworth D. C. and Marcus R. K. in Glow Discharge Spectroscopies ed. Marcus R.K. Plenum Press New York 1993 ch. 7 pp. 291-328. 14 Duckworth D. C. Donohue D. L. Smith D. H. Lewis T. A. and Marcus R. K. Anal. Chem. 1993 65 2478. 15 Shick C. R. Jr. Raith A. and Marcus R. K. J. Anal. At. Spectrom. 1993 8 1043. 16 Shick C. R. Jr. Raith A. and Marcus R. K. J. Anal. At. Spectrom. 1994 9 1045. 17 Saprykin A. I. Becker J. S. and Dietze H.-J. J. Anal. At. Spectrom. 1995 10 897. 18 Shick C. R. Jr. and Marcus R. K. Appl. Spectrosc. 1996,50,454. 19 Schelles W. De Gendt S. Muller V. and Van Grieken R. E. Appl. Spectrosc. 1995 49 939. 20 Schelles W. De Gendt S. Maes K. and Van Grieken R. E. Fresenius’ J. Anal. Chem. 1996 355 858. 21 Schelles W. Maes K. De Gendt S. and Van Grieken R. E. Anal. Chem. 1996 68 1136. 22 Betti M. Rasmussen G. and Koch L. paper presented at the 1994 Winter Conference on Plasma Spectrochemistry San Diego CA USA paper S12. 23 Schelles W. and Van Grieken R. E. Anal. Chem. in the press. 24 Robinson K. and Nayler R. Eur. Spectrosc. News 1986 68 18. 25 van Straaten M. Gijbels R. and Vertes A. Anal. Chem. 1992 64 1855. 26 Bogaerts A. van Straaten M. and Gijbels R. J. Appl. Phys. 1995 77 1868. 27 Demeny D. Sziics L. and Adamik M. J. Anal. At. Spectrom. 1992 7 707. 28 SIMION Software Version 4.0 Idaho National Engineering Laboratory Idaho Falls ID. 29 Bogaerts A. Gijbels R. and Goedheer W. Anal. Chem. 1996 68 2296. Paper 6103821 B Received May 31 1996 Accepted July 5 1996 Journal of Analytical Atomic Spectrometry October 1996 Vol. 11 941

ISSN:0267-9477    

DOI:10.1039/JA9961100937

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Comparison of Systems for Eliminating Interferences in the Determination of Arsenic and Antimony by Hydride Generation Inductively Coupled Plasma Atomic Emission Spectrometry ANNA RISNES AND WALTER LUND" Department of Chemistry University of Oslo P.O. Box 1033 N-031 i Oslo Norway The determination of arsenic and antimony by hydride generation inductively coupled plasma atomic emission spectrometry is discussed. A comparison is made of different procedures for eliminating redox (HNO,) and transition metal interferences; two pre-reduction agents three continuous-flow systems and two gas-liquid separators were studied. The pxe- reduction agents were KI and thiourea which were added in a continuous-flow mode either before or after the introduction of tetrahydroborate. The spray chamber served as a gas-liquid separator; alternatively a U-tube separator was used.In the recommended system 10% m/v thiourea is added to the sample stream before 1.0% m/v tetrahydroborate is introduced using the spray chamber as a gas-liquid separator. Thiourea eliminated the oxidation interference from HNO I and also the interference from 50 mg 1-' of Ag+ Cu2+ hi2+ and Fe3+. Keywords Hydride generation; inductively coupled plasma atomic emission spectrometry; arsenic and antimony determination; on-line pre-reduction; thiourea; potassium iodide Sample introduction by hydride generation is used frequently when As Sb and Se are determined by atomic spectrom:try; the technique provides very low detection limits and in add1 tion removal of interferences. Hydride generation is particu! arly well established in the field of AAS,' but it is even riore suitable for ICP-AES because this technique eliminates the gas-phase interferences which may be encountered in AA ?i.2-4 For ICP-AES a continuous-flow hydride generation system is more suitable than a batch system because it ensures plasma stability and multi-element capability In the determination of As and Sb a pre-reduction agent is normally used to reduce any pentavalent As and Sb ta the trivalent state because a low sensitivity is obtained when these elements are present in the higher oxidation state.For samples containing HNO As and Sb may be present in the pentavalent state since trivalent As and Sb are easily oxidized by HNO,. The pre-reduction of the elements to the trivalent state can be achieved by adding K15-8 or t h i ~ u r e a ~ .~ to the sample solution prior to the hydride generation process. However the pre- reduction will interfere with the determination of Se becmse Se" is partly reduced to the elemental state.'-'' Schramel and Xu" found that this interference was eliminated when tht. KI addition was carried out in a flow system. The addition of KI or thiourea also serves to minimize the interference froin a number of transition metals.' The hydrides can be separated from the liquid phase by different devices; the most common is probably a U-tube Journal of Analytical Atomic Spectrometry separator." From the U-tube the hydrides can be introduced directly at the bottom of the torch without disconnecting the spray chamber using a special interface.' However the normal spray chamber can also be used as a gas-liquid separator.In this work different continuous-flow procedures for elimin- ating redox (HNO,) and transition metal interferences are compared. The pre-reduction agents KI and thiourea were added either before or after the introduction of the tetrahydro- borate. The spray chamber or a U-tube served as a gas-liquid separator. EXPERIMENTAL Apparatus and Operating Conditions A Perkin-Elmer (Norwalk CT USA) ICP 5500B sequential inductively coupled argon plasma atomic emission spec- trometer was used. The plasma generator (27.12 MHz) and torch box were identical with those of the Perkin-Elmer ICP 6500 whereas the data system was a Model 3600 computer with a PR-100 printer.An all-quartz torch was used; the outer diameter was 20 mm and the inner diameter of the injector tip was 1.4 mm. For the hydride generation Gilson ( Villiers-le-Bel France) Minipuls 3 (4 channels) peristaltic pumps PVC/PTFE tubing and poly (propylene) T-junctions were used. The acidified sample sodium tetrahydroborate solution and pre-reduction agent were continuously pumped in separate streams to the T- junctions where the reagents were mixed and the hydrides generated. The three different flow systems used are shown in Fig. 1; the flow rates are given in Table 1. In Flow System 1 the pre-reduction agent is added ofS-line to the tetrahydrobo- rate solution; in Flow System 2 the pre-reduction agent is Flow System 1 Sample1 blank Sample/ Flow System 2 Flow System 3 * To whom correspondence should be addressed.Fig. 1 Flow diagrams for the three flow systems Journal of Andytical Atomic Spectrometry October 1996 Vol. 11 (943-948) 943Table 1 Flow rates of sample and reagents Flow rate/ml min-' Gas-liquid Pre-reduction separator Sample NaBH agent Waste U-type 9.2 4.2 9.2* Spray chamber 3.5 1.5 3.5* 13.5t * Only in Flow Systems 2 and 3. 7 6.9 ml min-' for Flow System 1. introduced into the sample stream after the hydride generation whereas in Flow System 3 the pre-reduction agent is introduced before the hydride generation takes place. When the spray chamber served as a gas-liquid separator the reaction mixture passed through the solution inlet of the cross-flow nebulizer while argon at a rate of 1.0 1 min-' passed through the gas inlet.When a U-type gas-liquid phase separ- ator12 was used the hydrides were transported by an argon flow of 1.0 1 min-' to the bottom of the quartz torch through a specially designed ball-joint adapter with a separate hydride inlet. The adaptor which is shown in Fig. 2 constitutes the interface between the spray chamber and the quartz torch with its ball-joint. The adaptor allowed the introduction of hydrides without disconnecting the spray chamber. The operating conditions for the ICP-AES measurements and the hydride generation are given in Table 2. The emission signals were corrected for the blank which was determined using 5 moll-' HCl as the sample. The blanks were only a few per cent. of the analyte signals. 1 cm Fig. 2 Ball-joint adaptor with separate hydride inlet.The adaptor constitutes the interface between the quartz torch and the spray chamber Table 2 Operating conditions for the ICP-AES instrument and the hydride generation ICP-AES- Rf-effect Flow rate of plasma gas Flow rate of auxiliary gas Observation height Integration time Wavelength As Sb Se Background correction Hydride generation- Reducing agent Acid concentration in sample Flow rate of carrier gas (Ar) 1.05 kW 14.0 1 min-' 0.6 1 min- 18 mm 1.0 s 193.70 nm 217.58 nm 196.03 nm kO.08 nm 1.0% m/v NaBH in 0.1 moll-' NaOH 2.0 2.4 or 5.0 mol 1- ' HCl 1.0 mi min-' Reagents and Samples All reagents were of analytical-reagent grade and de-ionized water was used throughout. The 1% m/v sodium tetrahydro- borate solution was prepared by dissolving NaBH (Fluka Buchs Switzerland) in 0.1 moll-' NaOH. The solution was filtered before use and stored in a polyethylene flask at 4°C.Solutions containing 1-10% m/v of KI and thiourea respect- ively were prepared daily by dissolving the reagents in 5 moll-' HCl. Single-element 1 g 1-' stock standard solutions were pre- pared from the salts listed in Table 3. Working standard solu- tions of concentrations lower than 1 g 1-' were prepared fresh daily by dilution. The sample solutions contained 2.0 2.4 or 5.0moll-' HCl and in some experiments also l.lrnoll-' HNO,. The analyte concentration was usually 50 pg 1-' . Recommended System and Procedure The recommended system for hydride generation is shown in Fig. 3. Flow System 3 is used; the sample (in 2.0 moll-' HCl) is first mixed with the pre-reduction agent 10% m/v thiourea (in 5.0mol1-' HC1).The flow rate of both streams is 3.5 ml min- '. The hydrides are then generated by introducing 1.0% m/v NaHB4 (in 0.1 moll-' NaOH) at a rate of 1.5 mlmin-'. The spray chamber is used as a gas-liquid separator. RESULTS AND DISCUSSION Different pre-reduction agents can be used for the selective reduction of As and Sb from the pentavalent to the trivalent state. Potassium iodide is probably the most commonly used reagent but KBr,I3 ~-cysteine'~*'~ and thiourea6*' have also been recommended. Preliminary experiments showed that L-cysteine worked well for Sb but the intensity of the As signal decreased with increasing concentration of L-cysteine. For 4-6% m/v L-cysteine the emission intensity for As was equal to the blank value.In this work the effects of KI and thiourea were studied. Flow Systems The three flow systems shown in Fig. 1 were tested by analysing 50pgl-I solutions of pentavalent and trivalent As and Sb. For each system the effect of pre-reduction agent and gas- liquid separation was studied by calculating the signal intensity ratios AsV :As"' and SbV Sb"'; the ratios should be 1.0 when the pre-reduction is complete. The following concentrations of KI and thiourea were used 0 2 4 6 8 and 10% m/v. In Flow System 1 the pre-reduction agent is added ofl-line to the tetrahydroborate solution; Schramel and Xu" used the same approach. In Flow System 2 the pre-reduction agent is introduced into the flowing stream after the introduction of tetrahydroborate; this approach was also used by Nygaard Table 3 Salts used for preparation of 1 g 1-' stock standard solutions of analytes and interferents; the solutions also contained 0.05 mol 1-' HCl Analyte As"' AsV Sb"' SbV Se" Interfering metal ions Fe3 + cu2 + Ni2 - Ag + Salt Na AsO Na,HAsO * 7H20 Spectrascan (Teknolab Norway) Na2Se03 5H20 KSbOC4H404 '0.5H20 FeC13 -6H,O CuCl 2H20 NiSO 6H,O Ag2S04 944 Journal of Analytical Atomic Spectrometry October 1996 Vol.1 1I I Argon 1.0 I min-1 Sample 3.5 ml min-1 in 2.0 mol 1-1 HCI r I u 13.5 ml min-1 10 9% thiourea 3.5 ml min-1 in 5.0 mol F1 BCI Waste Fig. 3 The Recommended System for hydride generation based on Flow System 3 with 10% m/v thiourea as pre-reduction agent and the spray chamber as gas-liquid separator. All experimental parameters ai.e given in the figure and Lowry' and Pretorius et In Flow System 3 the pre- reduction agent is mixed with the sample before tetrahydrobo- rate is added. The results obtained for the three flow systems are shown in Table 4. For all three systems the signals obtained for trivalent As and Sb were unaffected by the presence of the pre-reduction agent. From Table4 it can be seen that low values are obtained for the As" As"' and Sb" Sb"' ratios in the absence of a pre- reduction agent; the As ratio is in the range 0.2-0.5 whereas the Sb ratio is 0.05-0.2. The As and Sb ratios are lower for Flow Systems 2 and 3 than for Flow System 1. The use of an extra channel in Flow Systems 2 and 3 (in the absence of pre- reduction agent 2.0moll-' HCl was introduced) leads to a dilution which affects the two oxidation states differently. When the pre-reduction agent was introduced by means of Flow Systems 1 and 2 complete pre-reduction was obtained for SbV with 10% m/v KI and the spray chamber as a gas- liquid separator as shown in Table4.However the pre- reduction of AsV was far from complete. Experiments were also carried out with reagent concentrations higher Lhan 2.0 moll-' HCl 1 % tetrahydroborate and 10% pre-reduction agent and with a higher flow rate of the pre-reduction agent but the As ratio remained below 1.0. The best results were obtained when Flow System 3 was used as shown in Table 4. Here complete pre-reduction was achieved for both Asv and SbV by 10% thiourea (and for Sb" also by 10% KI) although only when the spray chamber was used as a gas-liquid separator.The results indicate that it is preferable to carry out the pre-reduction before the hydrides are generated by NaBH4. For all three flow systems the pre-reduction by KI was more effective (faster) for SbV than for As" as has also been observed by other worker^.^*'^ From Table4 it can be seen that it is best to use the spray chamber as a gas-liquid separator. This may partly be due to the low flow rates which were used when the reaction mixture was introduced through the solution inlet of the cross-flow nebulizer because this resulted in a prolonged reaction time. From the experiments with the three flow systems the following conditions are recommended to achieve complete pre-reduction of both As" and Sb" (hereafter called the Recommended System) 10% m/v thiourea is used as pre- reduction agent Flow System 3 is used for mixing the different reagent streams and the spray chamber is used as a gas-liquid separator.The Recommended System is shown in Fig. 3. The concentration of thiourea used is much higher than that required when thiourea is added directly to the sample solution prior to the hydride generationg The detection limits obtained with the Recommended System were 0.3 pg 1-' As and 0.5 pg 1-' Sb; the values correspond to twice the standard deviation for a solution of 1/5 the back- ground equivalent concentration of each element.'8w19 The detection limit was also calculated for Se; a value of 0.5 pg I-' Se was obtained. The detection limits are higher than those obtained using the U-type separator; in the latter instance the detection limits were 0.1 pg 1-' As and 0.2 pg 1-' Sb.The Table 4 Effect of pre-reduction agent and gas-liquid separator on the signal intensity ratios Asv As"' and Sbv Sb"' for Flow Systems 1 2 and 3. The concentration of each species was 50 pg 1-' in 2.0mol1-' HCl. The standard deviations were 0.01-0.05 (n=3) Pre-reduction agent Flow system 1 1 1 1 1 2 2 2 2 2 3 3 3 3 3 3 KI KI Thiourea Thiourea KI KI Thiourea Thiourea KI KI Thiourea Thiourea Thiourea Concentration (YO m/v) 0 2 10 2 10 0 2 10 2 10 0 2 10 2 6 10 Gas-liquid separator U-type Spray chamber AsV:As"' 0.46 0.54 0.60 0.60 0.64 0.29 0.32 0.45 0.33 0.64 0.22 0.26 0.5 1 0.34 0.70 0.89 SbV Sb'" 0.20 0.58 0.99 0.28 0.36 0.05 0.3 1 0.58 0.2 1 0.46 0.05 0.37 0.76 0.21 0.41 0.54 AsV:Asl" 0.54 0.57 0.55 0.59 0.65 0.43 0.42 0.46 0.47 0.77 0.32 0.33 0.43 0.38 0.82 0.99 SbV Sb"' 0.19 0.67 0.97 0.25 0.47 0.17 0.69 0.99 0.39 0.77 0.08 0.93 1.03 0.50 0.86 0.98 Journal of Analytical Atomic Spectrometry October 1996 Vol.1 1 945inferior detection limits for the Recommended System are due to the low sample introduction rate and a high background caused by a less complete gas-liquid separation with droplets entering the p l a ~ m a . ~ Unfortunately the presence of thiourea depresses the Se" signal,' because SeIV is partly reduced to the elemental state. The effect of the Recommended System on this process was studied; it was found that the Se" signal was depressed by 14%. Effect of HNO in Sample Solution Nitric acid is often present in solutions of real samples after the sample decomposition step.The effect of HNO is of particular interest because the acid can oxidize trivalent As and Sb to the pentavalent state and in addition it will lower the effective concentration of the pre-reduction agent. The effect of 1.1 mol I-' HN03 on the signal intensities of As"' Sb"' and Se" is shown in Table 5 using the Recommended System. As can be seen the signal intensities for As and Sb in the presence of HNO were only 0.33 and 0.07 respectively relative to the intensities obtained in the absence of HNO,. These values are almost the same as those given in Table 4 for the AsV As"' and SbV Sb"' ratios using Flow System 3/spray chamber without a pre-reduction agent.Thus the results in Table 5 show that HNO oxidizes trivalent As and Sb to the pentavalent state even at room temperature. For our study this explanation is more likely than inhibition of hydride evolution by nitrogen oxides.20 The positive effect of thiourea is also demonstrated in Table 5; 10% m/v thiourea eliminates the depression of the As and Sb signals caused by 1.1 moll-' HNO,. For Se the presence of HNO has no effect since Se" is not oxidized by HNO,. However thiourea depresses the Se" signal because SeIV is partly reduced to the elemental state. Interference From Transition Metals In high concentrations metals such as Fe Cu Ni Co Ag Au Pt Pd and W can interfere with the generation of the h y d r i d e ~ . ' - ~ - ~ ' - ~ ~ The predominating reaction is probably the reduction of the interfering metal ion by tetrahydroborate which results in the formation of metal particles or metal b o r i d e ~ ; ~ ~ ~ ' the finely dispersed precipitate may adsorb and decompose the hydrides.The interference can be minimized in different ways.26 A simple approach is to increase the concen- tration of HCl.27*28 A decrease in the tetrahydroborate concen- tration may also have a beneficial effect.29 Alternatively a masking agent can be used; a variety of such agents have been evaluated.,' More recently the use of EDTA diethylenetriami- nepentaacetic acid (DTPA) and tartrate,2-4*31 ~ - c y s t e i n e ' ~ * ' ~ ' ~ ~ and thiourea' was described. Thiourea has been used to mask the interference from transition metals in the determination of As9* and Sb.6,9734 Thiourea was found to be a better masking agent than KI.6 Thiourea is one of a few ligands that forms complexes with metal ions in strongly acidic solution.In this work the interference from Ag' Cu2+ Ni2+ and Fe3+ was studied; these are among the elements that often interfere the most.' The interfering metals were studied at concentrations of 0 0.05 0.5 5.0 and 50 mg 1-'; the results for 5.0 and 50 mg 1-' concentrations are shown in Tables 6 7 and 8 for As Sb and Se respectively. When the concentration of Ag and Cu was above 5 mg 1 - ' a precipitate was formed when NaBH was introduced in the absence of thiourea. The precipi- tation was prevented by thiourea and thus a time-consuming acid cleaning of the system was avoided. From Table 6 it can be seen that the As signal is depressed by all of the transition metals at a concentration of 50 mg 1-'.However in the presence of thiourea the interference from the transition metals is eliminated. It can also be seen that HNO alone depresses the As signal more than the transition metals do owing to the oxidation of As"' to AsV; the signal is not further depressed by 50 mg 1-' of Ag Ni Cu or Fe. Thiourea eliminates the combined interference from HNO and trans- ition metals. The effects observed for Sb are similar to those described for As but the signal depression by the transition metals and HNO is more severe as shown in Table 7. The depression by HNO is ascribed to the oxidation of Sb"' to SbV; the signal is not further depressed by 50mg 1-1 of Ag Ni Cu or Fe.Again thiourea eliminates the interference from the transition metals and the combined effect of HNO and these metals. Table 5 signal intensities; 1.00 means no depression effect. The Recommended System was used for hydride generation; n = 8 Effect of HN03 and thiourea on the signal intensity of 50 pg l-.' of AS"' Sb"' and Se" respectively. The results are given as relative 2.4 mol I-' HCl 1.1 mol 1-' HN03+2.4mol I-' HCl Without thiourea 10% thiourea Without thiourea 10% thiourea Analyte Mean S Mean S S Mean S Mean As"' 1 .oo 0.022 1.01 0.013 0.33 0.01 1 0.97 0.008 Sb"' 1 .oo 0.0 12 1 .oo 0.013 0.07 0.007 0.97 0.012 Se" 1 .oo 0.022 0.84 0.019 0.99 0.01 5 0.72 0.065 Table 6 Effect of transition metals HN03 and thiourea on the signal intensity of 50 pg 1-' As"'.The results are given as relative signal intensities; 1.00 means no depression effect. The Recommended System was used for hydride generation. The standard deviations were 0.01-0.06 (n = 2) Interferent 2.4 mol 1 - ' HCI 1.1 mol I-' HN03+2.4mol 1-' HCI Metal ion Concentration/mg 1- 0 5.0 50 Ni2 + 5.0 50 c u 2 + 5.0 50 Fe3 + 5.0 50 Ag+ Without thiourea 1 .00 1.01 0.59 0.98 0.85 0.95 0.67 0.64 0.60 10% thiourea 1.01 1.02 0.98 0.98 1.04 0.99 1.03 0.96 1.03 Without thiourea 0.33 0.33 0.32 0.32 0.30 0.3 1 0.30 0.3 1 0.32 10% thiourea 0.97 0.96 0.95 0.93 0.96 0.97 0.97 0.94 0.95 946 Journal of Analytical Atomic Spectrometry October 1996 Vol. 11Table7 Effect of transition metals HNO and thiourea on the signal intensity of 50pg I-' Sb"'. The results are given as relative signal intensities; 1.00 means no depression effect.The Recommended System was used for hydride generation. The standard deviations were 0.01-0.04 (n=2) In te rfe ren t 2.4 rnol I - ' HCI 1.1 mol 1-' HN03+2.4 mol 1-' HCI Metal ion Concentration/mg I-' Ag+ Ni2 + c u 2 + Fe3 + 0 5.0 50 5.0 50 5.0 50 5.0 50 Without thi1:urea 1 .oo 0.80 0.20 1 .oo 0.81 1.03 0.07 0.37 0.08 10 % thiourea 1 .oo 1.01 0.97 0.99 1 .oo 1 .oo 0.98 0.96 0.95 Without thiourea 0.07 0.06 0.06 0.10 0.05 0.07 0.07 0.10 0.1 1 10% thiourea 0.97 0.98 0.98 0.94 0.99 0.95 0.97 0.99 0.91 Table 8 Effect of transition metals HNO and thiourea on i.he signal intensity of 50 pg I-' SetV. The results are given as relative signal intensities; 1.00 means no depression effect. The Recommended System was used for hydride generation.The standard deviations were 0.01 -0.07 (n=2) Interferent 2.4 rnol 1-' HCI 1.1 rnol I - ' HN03+2.4mol I - ' HCI Metal ion Concentration/mg 1- ' 0 5.0 50 5.0 50 5.0 50 5.0 50 Without thiourea 1 .oo 0.13 0.98 1.01 1.04 0.72 0.97 0.93 - 10% thiourea 0.84 0.85 0.82 0.8 1 0.83 0.77 0.15 0.82 0.57 Without thiourea 0.99 0.1 1 0.96 0.94 0.96 0.41 0.97 0.98 - 10% thiourea 0.72 0.80 0.62 0.67 0.67 0.59 0.09 0.70 0.49 Table9 Effect of a mixture of Ag+ Ni2+ Cu2+ and Fe3+ in equal amounts HN03 and thiourea on the signal intensity of As"' Sb"' and Se" for a mixture of the analytes (50pg I-' of each). The results are given as relative signal intensities; 1.00 means no depression effect. The Recommended System was used for hydride generation.The standard deviations were 0.01-0.06 (n = 2 ) Analyte As'" As"' A s'" Sb"' Sb"' Sb"' SeiV SetV Interferent concentration/ mg 1-'* 0 5 .O 50 0 5.0 50 0 5.0 2.4 m01 1- ' HCl Without thiourea 10% thiourea 1 .oo 0.95 0.78 1.00 0.60 0.40 1.00 0.18 1.03 0.98 0.99 1.03 0.97 0.96 0.80 0.62 1.1 rnol 1-1 HNO,+2.4 rnol 1-' HCI Without thiourea 10% thiourea 0.37 0.35 0.35 0.07 0.08 0.05 0.94 0.25 0.96 0.95 0.96 0.95 0.93 0.92 0.73 0.5 1 * Concentration of each of the interfering metals Ag+ Ni2+ a h 2 + and Fe3+. For Se the signal depression by the transition metals is very marked for Ag and less so for Cu while Ni and Fe do not interfere significantly at 50 mg l-' as shown in Table 8. Nitric acid does not depress the Se signal because HNO cannot oxidize Set" to Se".However thiourea depresses the Se" signal particularly when 50 mg 1 - l Cu is present and has a positive effect only on the interference from Ag. In Table 9 the relative signal intensities of As"' Sb"' and Se" are given for solutions containing a mixture of As Stl and Se the transition metals Ag Ni Cu and Fe and HNO For As and Sb it can be seen that thiourea eliminates the inter- ference from HN03 and a mixture of 50 mg 1-' of each of the transition metals. Even for Se thiourea partly eliminate#*; the interference from the transition metals probably becaus.;e of the positive effect on the interference from Ag (Table 8). CONCLUSION Trivalent As and Sb are oxidized to the less sensitive ptmta- valent state when HNO is present in the sample solution. Pentavalent As and Sb are fully reduced to the trivalent state by 10% m/v thiourea in a continuous-flow system using the spray chamber as a gas-liquid separator.Thiourea is more effective than KI for the pre-reduction of AsV. Thiourea also eliminates the interference from 50mg1-' of each of the transition metal ions Ag' Cu2+ Ni2+ and Fe3+. However the use of a continuous-flow system for the addition of the pre-reduction agent does not eliminate the interference from thiourea on the SeIV signal. REFERENCES 1 Dedina J. and Tsalev D. L. Hydride Generation Atomic Absorption Spectrometry Wiley Chichester 1995. 2 Wickstrram T. Lund W. and Bye R. J . And. At. Spectrom. 1995 10 809. 3 Wickstr~m T. Lund W. and Bye R. Analyst 1995 120 2695. 4 Wickstrarm T. Lund W. and Bye R. Analyst 1996 121 201.5 Nakahara T. Anal. Chim. Acta 1981 131 73. 6 Nakahara T. and Kikui N. Anal. Chim. A d a 1985 172 127. Journal of Analytical Atomic Spectrometry October 1996 Vol. 11 9477 Sinemus H. W. Melcher M. and Welz B. At. Spectrosc. 1981 2 81. 8 Nygaard D. D. and Lowry J. H. Anal. Chem. 1982 54 803. 9 Uggerud H. and Lund W. J. Anal. At. Spectrom. 1995 10 405. 10 Thompson M. Pahlavanpour B. Walton S. J. and Kirkbright G. F. Analyst 1978 103 705. 11 Schramel P. and Xu L.-Q. Fresenius’ J. Anal. Chem. 1991 340,41. 12 Thompson M. Pahlavanpour B. Walton S. J. and Kirkbright G. F. Analyst 1978 103 568. 13 Thompson M. Pahlavanpour B. and Thorne L. T. Water Res. 1981 15 407. 14 Chen H. Brindle I. D. and Le X.-C. Anal. Chem. 1992 64 667. 15 Chen H. Brindle I. D.and Zheng S. Analyst 1992 117 1603. 16 Pretorius L. Kempster P. L. van Vliet H. R. and van Staden J. F. Fresenius’ J. Anal. Chem. 1992 342 391. 17 Nadkarni R. A Anal. Chim. Acta. 1982 135 363. 18 Inductively Coupled Plasmas in Analytical Atomic Spectrometry eds. Montaser A. and Golightly D. W. VCH New York 2nd edn. 1992 p. 262. 19 Method Development for ZCP Spectrometry User Manual Perkin- Elmer Norwalk CT. 20 Brown R. M. Fry R. C. Moyers J. L. Northway S. J. Denton M. B. and Wilson G. S. Anal. Chem. 1981 53 1560. 21 22 23 24 25 26 27 28 29 30 31 32 33 34 Nakahara T. and Kikui N. Spectrochim. Acta Part B 1985 40,21. Smith A. E. Analyst 1975 100 300. Henden E. Analyst 1982 107 872. Bye R. Talanta 1986 33 705. Bax D. Agterdenbos J. Worrell E. and Kolmer J. B. Spectrochim. Acta Part B 1988 43 1349. Nakahara T. Spectrochim. Acta Rev. 1991 14 95. Welz B. and Melcher M. Spectrochim. Acta Part B 1981,36,439. Hershey J. W. and Keliher P. N. Spectrochim. Acta Part B !986 41 713. Astrsm O. Anal. Chem. 1982 54 190. Nakahara T. Prog. Anal. At. Spectrosc. 1983 6 163. Wickstrsm T. Lund W. and Bye R. J. Anal. At. Spectrom. 1995 10 803. Le X.-C. Cullen W. R. Reimer K. J. and Brindle I. D. Anal. Chim. Acta 1992 258 307. Peacock C. J. and Singh S. C. Analyst 1981 106 931. Narsito Agterdenbos J. and Bax D. Anal. Chim. Acta 1991 244 129. Paper 61033066 Received May 13 1996 Accepted July 5 1996 948 Journal of Analytical Atomic Spectrometry October 1996 Vol. 11

ISSN:0267-9477    

DOI:10.1039/JA9961100943

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Behaviour of a Single-bore High-pressure Pneumatic Nebulizer Operating With Alcohols in Inductively Coupled Plasma Atomic Emission Spectrometry 1 Journal of I Analytical Atomic Spectrometry JOSE L. TODOLI ANTONIO CANALS AND J;ICENTE HERNANDIS Departamento de Quimica Analitica Universidad de Alicante 03071 Alicante Spain The behaviour of a single-bore high-pressure pneumatic nebulizer (SBHPPN) with alcohols in (ICP-AES) was investigated. A standard Meinhard nebulizer was used for comparison. To this end the drop size distribution of the primary aerosols the analyte and solvent transport rates W and Stor the fraction of solvent transported in liquid and vapour forms SKq and Sap the excitation temperature of the plasma T, and the molecular emission intensity of the C band were determined for solvents of different physical properties.The effect of the physical properties of the solvent on the nebulization process was also studied. The results show that the SBHPPN gives rise for all the solvents studied to primary aerosols that have smaller mean diameters than those produced by the Meinhard nebulizer for the same gas and liquid flows. The relative decrease in volume median diameter (I)"*) when switching from the Meinhard nebulizer to the SBHPPN is more noticeable for alcohols than for water. W is significantly higher for the SBHPPN than for the Meinhard nebulizer particularly at high liquid flows. However the differences between their S values are less pronounced. Under similar conditions the SBHPPN gives rise for all the solvents studied to higher emission intensities than the Meinhard nebulizer.However the relative signal enhancements achieved by changing from the SBHPPN to the Meinhard nebulizer are lower than the corresponding analyte transport enhancements. The relative signal enhancements achieved by switching from water to alcohols are lower for the SBHPPN than for the Meinhard nebulizer. This behaviour can be explained in terms of the higher S values associated with the SBHPPN which lower T in comparison with the Meinhard nebulizer. Keywords Inductively coupled plasma; organic solvent; alcohols; pneumatic nebulization; single-bore high-pressu 1.e pneumatic nebulizer One of the most important applications of organic sol~ents in atomic spectrometry is as a means to increase the senisitivity of the determination.',* However in low-medium power (ICP- AES) the organic solvent load has a strong influence on the physical and chemical properties of the plasma and iii most instances the sensitivity improvements achieved have nc it been as significant as expected.Therefore much effort has been devoted in trying to understand fully the interactions between organic aerosols and analytical plasmas that could account for the observed behaviour. It has been observed that under the same conditions (ie. rf power viewing height and Glolvent plasma load) organic solvents tend to decrease the excitation temperat~re~-~ and the electronic density of the pla~ma.~ Boumans and Lux-Steiner' attributed these effects to the enhancement of the thermal conductivity of the plasma due to the presence of the pyrolysis products of the organic clolvent (ie.Cz CN CO CH C I etc). Blades and Caughlin7 suggested that such an enhancement in the thermal conductivity of the plasma was mainly due to the energy absorbed by [.he C2 molecules to dissociate and ionize since in a water-loaded plasma the predominant molecules i.e. OH require much less energy to dissociate. In addition solvent pyrolysis products cause a large increase in the background emission of the plasma. Kreuning and Maessen3 measured the spatial distribution of the species C2 CN and C I in chloroform- and toluene-loaded plasmas and observed that the emission maxima of these species appeared in different zones of the plasma. Thus whereas C2 appears at low observation heights in the inner zone of the plasma CN appears higher in the outer zone since its formation results from the interaction of C atoms with the surrounding air.Recent studies conducted by Weir and have shown that methanol and chloroform behave in a similar fashion. These distributions of the solvent pyrolysis products which depend on the experimental conditions (i.e. rf power nebulizer/carrier gas flow liquid flow solvent nature) cause the effect of the organic solvents on the emission intensity to have a marked spatial character in ICP-AES. As regards the nature of the solvent employed Botto'l concluded that the energy consumption from the plasma was higher for those solvents with the highest H:C ratio since they have higher atomization energies. Maessen et al.12 classi- fied the solvents into two groups one for solvents of low vapour pressure (e.g.water xylene or IBMK) that do not affect the stability of a plasma of 1.75 kW and another for solvents with high vapour pressure (e.g. chloroform methanol and ethanol) that clearly deteriorate the behaviour of the plasma. Later Kreuning and Maessen3 concluded that for a given solvent load to the plasma the most significant variables are the load of carbon and the relative molecular mass of the solvent. Recent studies conducted by Weir and Bladesg reveal that the C:O ratio markedly affects the appearance of the plasma. These workers make a distinction between solvents with a C 0 ratio close to 1 (e.g. methanol) and those with a C:O ratio higher than 1 (e.g. chloroform). When working with water-ethanol mixtures McCrindle and Rademeyer6 have reported that the excitation temperature increases with the percentage of ethanol up to 15% (v/v) and then decreases.All these detrimental effects associated with the use of organic solvents become more noticeable on increasing the solvent The amount of solvent that reaches the plasma (S,,,) when using pneumatic nebulization depends on the following factors (1) The nebulizing gas flow Q,. Increasing gas flow gives rise to a decrease in the mean drop size of the primary aeros01'~ and to a larger amount of evaporated solvent thus causing Stot to in~rease.'~ (2) The liquid flow Q1. Increasing Q1 causes S, to increase even though the mean drop size of the primary aerosol increase^.'^ The increase in S, is less pronounced at high Q values.I2 (3) Solvent nature.As regards the influence of the solvent nature on St, two factors have to be taken into consideration. Firstly the fraction of solvent transported in liquid form (Sliq) is a function of the mean drop size of the primary aerosol i.e. the smaller the mean drop size of the primary aerosol the Journal of .Analytical Atomic Spectrometry October 1996 Vol. 1 1 (949-956) 949higher the value of Sliq. Secondly the fraction of solvent transported in gaseous form (Svap) depends on the relative volatility of the solvent. These two factors are not completely unrelated. On the one hand the rate of solvent vaporization is proportional to the aerosol surface. In addition the vapour pressure of a droplet increases on decreasing its curvature radius.16 Hence a reduction in the aerosol mean drop size facilitates solvent evaporation.On the other hand solvent evaporation would cause Sliq to decrease; however this hypo- thetical decrease is compensated for at least partially by the reduction in the mean drop size of the aerosol caused by the solvent evaporation which in turn facilitates the transport of the analyte. Since in general organic solvents show lower surface tension and higher volatility than water organic aero- sols usually have smaller mean drop sizes than aqueous aerosols and Sto values are usually higher for the f~rmer.".''.'~ Particular attention has been paid to the rate of solvent transported in vapour form. Long and Browner17 compared the effects on the plasma of introducing water as a vapour and as a liquid aerosol.They noticed that the introduction of vapour causes the electronic density to increase compared with the introduction of the same amount of water as droplets. These droplets diffuse very little from the axis of the central channel and hence most of the power consumption corre- sponds to this zone of the plasma." With organic solvents most of the solvent load reaches the plasma as vapour its side diffusion rate then being much higher and therefore it occupies a greater zone of the plasma. Hence when using organic solvents in ICP-AES it is advisable to take into account the physical state in which they will reach the plasma." (4) Introduction system. For a given spray chamber and specific nebulization conditions the characteristics of the pri- mary aerosol and hence the values of Sliq and Stat depend on the nebulizer ernpl~yed.'~-~' The effects of the organic solvents described above have been observed when using c~ncentric,'*~*'~ cross-flowg-" or V- groove3 pneumatic nebulizers.However other nebulizers have shown a different behaviour. Nisamaneepong et a1." reported that the use of a glass-frit nebulizer enhanced the tolerance of the plasma to the organic solvent load as compared with a concentric nebulizer. They explained their observations in terms of the small mean drop size and monodispersion of the aerosols generated. Later Brotherton et ~ 1 . ~ ' observed the same behaviour with a double grid nebulizer. Other more efficient nebulizers such as ultrasonic21.22 or therm~spray'~ nebulizers need to be used in conjunction with a desolvating system to reduce the load of organic solvent which otherwise would prevent any analysis from being performed.These papers highlight the importance of nebulizer design on the interfering effect of organic solvents. Some suggestions have been proposed to alleviate the detri- mental effects of organic solvents on the characteristics of the plasma (1) Working at high rf power. Many studies have shown that increasing rf power gives rise to higher excitation temperatures7 and hence to higher emission intensity."." (2) Using low gas flow.14 This leads to a lower solvent load to the plasma and to a longer residence time of the sample in the plasma. However lowering Q reduces the amount of analyte transported to the plasma.Therefore Q should not be too low. (3) Using refrigerated spray chambers. Maessen et ~ 1 . ' ~ found that when using toluene chloroform or 1,2-dichloro- ethane as solvents the signal-to-noise ratio for many elements increased significantly when the spray chamber was thermostated between -5 and -10°C. With ethanol and methanol cooling of the spray chamber was indispensable for the stability of the plasma. Pan et al.' obtained similar results. Bottoll reported that the power consumption decreased when spray chamber cooling was used. (4) Working with a narrower injector tube. Boumans and Lux-Steiner' and Nobile et ~ 1 . ' ~ designed a torch with a narrower injector tube which was particularly well suited to organic solvents. Recently a single-bore high-pressure pneumatic nebulizer (SBHPPN) has been developed in our laboratories which is suitable for FAAS,26,27 ICP-AES" and ICP-MS.29 In compari- son with the standard Meinhard nebulizer the SBHPPN provides higher sensitivities and lower limits of detection (LODs) when working with aqueous solutions of metals. In addition it does not show any tendency to become clogged because of high salt content in the solutions.As was indicated above the type (i.e. positive or negative) and extent of the effects of organic solvents on the plasma is a function of the nebulizer employed among other variables. Therefore when a new nebulizer is designed and evaluated one should test its applicability for the introduction of organic solutions. Hence the aim of this work was to evaluate with the SBHPPN the type and extent of the effect of alcohols in ICP-AES. To this end the drop size distribution of the primary aerosol the analyte and solvent transport rates the fraction of solvent transported in liquid and vapour forms the exci- tation temperature of the plasma T,, and the molecular emission intensity of the C band were determined for four alcohols with different physical properties.The effect of the physical properties of the solvent on the nebulization process was also studied. EXPERIMENTAL The design and working principle of the SBHPPN has been described in previous The glass nozzle employed in this work has an outlet bore whose section is 5.92 x mm2. A Meinhard nebulizer (TR-30-A3) was employed for comparison (gas outlet section 28.3 x mm2; liquid outlet section 0.82 mm').The liquid flow was varied between 0.6 and 2.0ml min-' by means of an Iso-Chrom HPLC pump (Spectra-Physics San Jose CA USA). Table 1 lists the significant physical properties of the five solvents employed. The gas flow was varied by means of a Model FC260 mass flow controller (Tylan Torrance CA USA). Air was employed for the measurement of the aerosol drop size distributions (DSDs) and argon for the remainder of the experiments. Although the DSDs obtained with air and argon are different their behaviour is similar as regards the variables considered in this work.I3 The DSDs of the primary aerosols were measured at a distance of 11 mm from the nebulizer tip in all instances except when the variation of the mean drop size along the radial position was measured; in this instance the distance was 15 mm to allow the aerosol cone to broaden.The drop size characteriz- ation was performed by means of a Model 2600c Fraunhofer laser diffraction system (Malvern Instruments Malvern Worcestershire UK) equipped with a lens with a focal length of 63 mm that allowed for the measurement of drop diameters from 1.2 to 118 pm. The instrument was previously calibrated with a reticule.31 The DSDs were calculated using a model- Table 1 Physical properties of the solvents* Solvent a/dyn cm - t P/CPS 4 Water 70.4 1 .oo 0.09 Methanol 20.3 0.50 1.00 Ethanol 21.4 1.13 0.69 Propan-2-01 19.8 2.08 0.27 Butan-1-01 22.8 2.38 0.11 *Measured at 20 "C. Surface tension. $Viscosity. §Relative volatility defined as the relationship between the liquid volume of solvent i and the volume of methanol required to saturate a given empty pace.'^^^^ 950 Journal of Analytical Atomic Spectrometry October 1996 Vol.11independent algorithm which does not presume any distri- bution function. The following parameters were employed for characterizing the primary aerosols volume median diameter (Dv.so) Sauter mean diameter ( 0 3 . 2 ) and volume concentration (VC) which is the ratio of the liquid volume of aerosol contained in the measurement zone of the laser beam to the total volume of this measurement zone in the particle ~ i z e r . ~ ~ . ~ ~ The solvent transport rate (Stat) was measured by a continu- ous indirect m e t h ~ d . ~ . ' ' ~ . ~ ~ The analyte transport rate (wet) was measured by a direct method i.e.by capturing the aerosol on a glass-fibre filter (Type A/E 47 mm diameter) of 0.3 pm size ~ o r e ' ~ . ~ ~ (Gelman Sciences Ann Arbor MI USA). Solutions of Mn (100 pg ml-') in each of the solvents Gdudied were used for this purpose. The Mn retained was extracted with hot 1.0% (m/m) HNO and determined by ICP-AES. The emission intensity was measured with a Modcl 2070 ICP-AES instrument (Baird Bedford MA USA). Solutions of Mn (1 pg ml-') in each of the solvents studied were used for this purpose. Table 2 lists the instrumental conditions employed. The emission intensity of the C2 molecular band (h= 516.468 nm) and the plasma excitation temperature T,,,. were also measured so as to assess the effect of organic solvcnts on the characteristics of the plasma.For calculating Z, the emission intensity of five ionic Fe lines was measured using solutions of this element (50 pg ml-') in each of the s1)lvents employed with the assumption that the population of the species of a given element obeys the Boltzmann distri- bUtion.3.5.7.18.34.35 Table 3 lists the characteristics of these lines. A straight line is obtained by plotting ln(~qp~qp/gqAqp I versus E (see Table 3 for definition of the symbols) the slope of which is - 1/KT,,c,3,s-7,'8,34,36 where K is the Boltzmann constant. The standard deviation of this method is abtlut 200 K,3 which is not too large for qualitative studies as in o iir case. RESULTS AND DISCUSSION Nebulization Process The peculiarities of the SBHPPN (i.e. working at high press- ures and having just one bore for both streams gas and liquid) Table 2 Experimental conditions for the measurement of the emis- sion intensity Outer gas flow rate Intermediate gas flow rate Nebulizer gas flow rate Liquid flow rate Spray chamber Rf power Slit-width Wavelength Integration time Viewing height PMT gain Torch 16 1 min-' 1.7 1 min-' Variable Variable Double-pass Sccj tt type 1000 w 0.1 nm 257.610 nm 0.2 s 7 mm ALC 4 For use with orpanic (V= 100 cm3) solvent^^*^^ Table 3 Iron ionic lines employed to measure the plasma excitation temperature* Wavelength/nm g '4,,/10-8 s-' E 'eV 259.940 22.1 4 17 275.574 21.1 5 .48 274.648 11.7 5 59 256.253 12.8 5 82 266.664 24.1 8 07 make it advisable to carry out a preliminary study of the fundamentals of the nebulization process with this nebulizer.Fig. 1 shows the effective section of the gas outlet A oersus Qi at a constant gas pressure for water and the four alcohols. In order to calculate A an equation that expresses this parameter as a function of Q and the gas pressure required to reach this gas flow was used.37 In all instances A decreases when Q1 is increased since the width of the liquid film adhering to the inner walls of the nozzle increases. Fig. 1 also gives some information about the influence of the physical properties of the solvent on A,. Thus for instance when comparing water and ethanol i.e. solvents of similar viscosity but different surface tension (Table l) it appears that surface tension hardly affects A which is about the same for both solvents.As regards viscosity Fig. 1 shows that the higher the solvent viscosity the smaller the effective section. Therefore when the solvent viscosity is increased a higher gas pressure should be applied if one seeks to keep the gas flow constant. In pneumatic nebulization the efficiency of the energy transfer from the gas stream to the liquid stream increases with the turbulence of the gas stream,37 and hence the mean drop size of the aerosol decreases. For the flow values used in this work the Reynolds number (Re) for the SBHPPN is up to 20 times higher than for the Meinhard nebulizer. Hence aerosols with a smaller mean drop size are to be expected for the former. DSD of the primary aerosols Influence of gas and liquidjows Fig. 2(a) shows that for both nebulizers and for all the solvents studied Dv.50 values decrease when Q is increased as expected since the energy available for nebulization is also increa~ed.'~-'' As regards the effect of Qi Fig.2(b) shows a different behaviour for both nebulizers. With the Meinhard nebulizer Dv,50 increases when Qi is increased for all the solvents in agreement with previously reported results." With the SBHPPN Dv,so for alcohols behaves similarly but for water Dv,so decreases slightly when Q1 is increased. This fact can be explained by taking into account that the surface tension values for alcohols are usually very low and hence the viscosity becomes a more important factor than with water.37 Thus the liquid vein remains unbroken for a longer time before being disintegrated by the gas stream.Therefore when Qi is increased the rupture of the liquid vein takes place in positions increasing downstream from the nebulizer tip where the energy of the gas flow becomes lower. This effect seems to predominate over the reduction in A caused by the increase in Q (Fig. 1). Obviously the intensity of this effect will be greater for solvents of high viscosity. Thus for solvents of low viscosity such as methanol the increase in Dv,50 is less pro- nounced than for solvents of high viscosity such as butan- 1 0.5 1.0 1.5 2.0 Q /ml min-' *p fundamental state; q upper state; g, orbital degenerancy of the upper state; Aqp Einstein coefficient for spontaneous emissior (8 A values from ref. 3); E, excitation potential. Fig. 1 Effect of the liquid flow rate on the 'effective' outlet section A for the SBHPPN.A Water; B methanol; C ethanol; D propan-2-01; and E butan-1-01. Gas pressure = 14 atm Journal of Analytical Atomic Spectrometry October 1996 Vol. 1 1 95120.0 1 (a) I \ 3 0 a 5*0* 0.50 0.60 0.70 0.80 0.90 6 Qg /1 min-’ 0.50 0.70 0.90 1.10 1.30 Q /ml min-’ Fig.2 Variation of Dv,50 for the solvents and nebulizers employed. (a) Dvis0 uersus Qg; Q1= 1.2 ml min-’. (b) Dv,50 uersus Q1; Qg=0.57 1 min- . (dotted line) Meinhard nebulizer; (solid line) SBHPPN. A Water; B methanol; C ethanol; D propan-2-01; and E butan-1-01 1-01. For solvents with a high surface tension such as water viscosity plays a secondary role. Comparison of SBHPPN and Meinhard nebulizer Fig. 2 shows that the SBHPPN generates organic aerosols the Dv,50 of which are 1.5-3.0 times smaller than those generated by the Meinhard nebulizer under the same conditions.The decrease in Dv,50 when changing from the Meinhard nebulizer to the SBHPPN is more important for alcohols than for water. The excess of kinetic energy of the gas stream associated with this change gives rise when applied to liquids with a surface tension much lower than that of water to an excess of surface which is much more significant for alcohols than for water. Efects of surface tension and viscosity of the solvent The effect of the surface tension on Dv,50 can be seen by comparing the curves of butan-1-01 and water (Fig.2). For any given set of conditions butan-1-01 gives rise to aerosols with a smaller Dv,50 than those produced by water as expected from the surface tension values,14*32*38 although the viscosity of the former is twice that of the latter (Table 1).As regards the viscosity its effect can be seen by comparing the curves of methanol ethanol propan-2-01 and butan-1-01 (Fig. 2). These solvents have similar surface tension values but different viscosities (Table 1). When working with the Meinhard nebulizer the values of Dv,50 follow the same order as the solvent viscosities whatever the conditions employed. This behaviour can be explained in terms of the ability of the viscosity to dampen the instabilities of the liquid vein and to lengthen it.37 However with the SBHPPN the behaviour is not the same. Fig. 2 shows firstly that the differences between the Dv,50 values are smaller than with the Meinhard nebulizer and secondly that the Dv,50 values do not follow the order of the solvent viscosities exactly.This divergence can be attributed to the fact that as already mentioned for the SBHPPN but not for the Meinhard nebulizer the effective gas outlet section (As) decreases when changing from one solvent to another of higher viscosity. Thus a higher pressure is required to keep the gas flow constant which in turn means more kinetic energy for surface generation. The final result with these high viscosity solvents is that the aerosols obtained with the SBHPPN have smaller Dv,50 values than expected from their surface tension and viscosity alone. Eflect of the volatility of the solvent Although the mean drop size depends to some extent on the volatility of the ~olvent,~’ the total liquid volume is the aerosol property that is most influenced by the solvent volatility. This influence was monitored by means of the so-called ‘volume concentration’ VC.Table4 summarizes the VC values for the four alcohols employed under several Qg and Q1 conditions. It appears that for the same Q1 Q and nebulizer VC values follow the rev- erse order of the solvent volatilities i.e. VC( butan-1-01)> VC(propan-2-01) > VC(ethano1) > VC(methano1). Hence meth- anol is the solvent that has lost the most liquid volume due to volatilization as expected. Comparison between nebulizers (SBHPPN versus Meinhard) is not possible because VC is a function of the gas (aerosol) velocity and this velocity is higher for the SBHPPN than for the Meinhard nebulizer under the same Q and Q1 conditions.Variation of the mean drop size along the radial position From a visual inspection pneumatically generated aerosols have the shape of a cone the angle of which decreases as Q is increased. In order to optimize the design from the point of view of aerosol transport it is important to know how droplets of different sizes are distributed within the cone. Fig. 3 shows the variation of D3,2 in relation to the transversal distance from the cone axis for both nebulizers. In this instance D3,2 was plotted instead of Dv,50 because the variations correspond- ing to Dv,50 for the SBHPPN appeared to be masked by the large values of Dv,50 for the Meinhard nebulizer. Some differ- ences between the two nebulizers are apparent.With the Meinhard nebulizer the largest drops are usually found in the outer zone of the cone whatever the solvent employed whereas the aerosol contained in the inner zone has a higher proportion of small drops. Similar observations have been reported pre- viously for technical atomizers (i.e. devices designed to generate aerosols in non-spectroscopic applications such as mechanical Table 4 Volume concentration values of the primary aerosols obtained for the alcohols using the SBHPPN and the Meinhard nebulizer vc (Yo) Solvent Methanol* Ethanol* :Propan-2-01* Butan-1-ol* Methanol? Ethanol7 Propan-2-01? Butan-1-01? Methanol$ Ethanol$ Propan-2-011 Butan-1 -011 Meinhard nebulizer 0.0620 0.07 17 0.0756 0.0882 0.1380 0.141 1 0.1532 0.1651 0.0893 0.1031 0.1152 0.1228 SBHPPN 0.0301 0.0353 0.04 16 0.0469 0.0676 0.08 13 0.0946 0.0966 0.0482 0.0559 0.06 13 0.0634 *Qg=0.57 1 min-’; Ql=0.6 ml min-’.tQg=0.57 1 min-‘; Ql=1.2 rnl min-l. $&,=0.87 1 min-’; Q,= 1.2 ml min-’. 952 Journal of Analytical Atomic Spectrometry October 1996 Vol. 1 1,;*A Fig. 3 Variation of D3+2 in relation to the transversal distance from the cone axis for both nebulizers. A Water; B methanol; C:' butan- 1-01. (dotted line) Meinhard nebulizer; (solid line) SBHPPN. Q,=0.68 1 min-'; Q,=2 ml min-'; distance from nebulizer tip to laser beam= 15 mm engineering and combustion p r o c e s s e ~ ) ~ ~ - ~ ~ the working prin- ciples of which are the same as for the Meinhard nebulizer and for other pneumatic nebulizer^.^^-^^ The SBHPPN pro- vides a more homogeneous distribution of drop sizes particu- larly when working with water.The slight increase in observed with alcohols on the outside of the cone which does not appear with water might be the result of the preferential evaporation of small droplets circulating in this zonc. From Fig. 3 it is also apparent that under the same set of Q and Q1 the cone angle is larger for the Meinhard nebuliier than for the SBHPPN. These observations clearly show that the geometriec of the interaction between gas and liquid streams are different in the two nebulizers. In the Meinhard nebulizer the aerosol is generated from the surface of the cylindrical liquid vein since the gas stream flows externally to it,37 whereas in the SHHPPN the liquid vein flows as a film externally to the gas stream.These results support the coupling of Meinhard nebulizers to double-pass spray chambers so as to remove most of the large drops by impaction against the inner walls of the inner tube which would not contribute effectively to the final emission signal and would also contribute to the inciease in signal fluctuation^,^^ thus deteriorating the LOD. As regards the SBHPPN this coupling does not seem to be so beneficial since small droplets as well as large droplets are likely to be lost that otherwise would contribute to the signal. For the same reason the spray chamber should be lengthened for the SBHPPN since otherwise the high velocity of the aerosol jet causes many potentially useful droplets to be lost through impaction against the bottom walls of the spray chamber.In spite of all these considerations a conventional double-pass Scott chamber was used in this work for both nebulizers so as not to change more than one factor at a time. Transport Efficiency The transport rates W, and St, finally obtained as well as the DSD of the tertiary aerosol are the result of cornbining the characteristics of the primary aerosol and the processes that take place in the spray chamber. The interaction of the tertiary aerosol with the plasma determines the analytical signal i.e. the emission signal and the LOD. Under the same conditions organic solvents proviide Wot and S, values higher than those provided by water whereas organic tertiary aerosols have smaller average drop sizes than aqueous tertiary aerosols. These facts are expected lo have opposite effects on the plasma the net result on the signal being unpredictable.Solvent transport eflciency Under any set of conditions S, is expected to increase when Q is increased owing to two parallel factors. Firstly the mean drop size of the primary aerosols decreases thus facilitating the transport and secondly evaporation is favoured since there is more gas and a larger surface. Stot is plotted against Q in Fig. 4(a). The results confirm the expected trends. Several reports on pneumatic nebuli~ation'~*'~~'~ state that S, increases again when Q1 is increased. Since Stat= Q1 x E where E is the solvent transport effi~iency,'~ it is clear that the increase in the amount of aerosol predominates over the reduction of the transport efficiency caused by the increase in the mean drop size of the primary aerosol.Fig. 4(b) shows the same tendency as increasing Q in our case. When S, values are compared in terms of the solvent employed it appears that volatility (or vapour pressure) is the most significant physical property of the ~ o l v e n t . ~ * ' ~ * ' ~ ~ ' ~ Fig. 4 shows that the arrangement of the solvents according to their Stot values coincides with the arrangement according to their volatility (Table 1). Finally when Stot values are compared in terms of the nebulizer employed it is clear that the SBHPPN provides higher values than the Meinhard nebulizer as expected since the aerosols produced by the SBHPPN have smaller D,,50 and D3,2 values than those produced by the Meinhard nebulizer.Analyte transport eficiency Fig. 5 shows the influence of several variables on KO,. The trends shown here are very similar to those obtained for S,,,. As mentioned before the transport values are higher for the SBHPPN than for the Meinhard nebulizer. Table 5 gives the relative transport values (SBHPPN/Meinhard) for solvent (Stot)rel and for analyte (Wot)rel. It can be seen that ( ~ o t ) r e l values are significantly higher than (Stot)rel values which indi- cates that part of the solvent is transported in vapour form. The amount of solvent that reaches the plasma in vapour form Svap and in liquid form SIiq can be estimated from mass balance considerations for both solvent and ana1~te.I~ Table 5 shows that Sli is much higher for the SBHPPN than for the 4.0 -I 1 - @ 0.40 0.45 0.50 0.55 0.60 0.65 - 'Y) - 4.0 9 Y % 3.0 2.0 1 .o 0 Q,/I min-' 0.50 0.70 0.90 1.10 1.30 Q /ml min-I Fig.4 Variation of S, for the solvents and nebulizers employed. (a) S, versus Q,; QI = 1.0 ml min- ' . (b) S, versus Q1; Q =OSO 1 min- ' . (dotted line) Meinhard nebulizer; (solid line) SBHPPN. A Water; B methanol; C ethanol; D propan-2-01; and E butan-1-01 Journal of Analytical Atomic Spectrometry October 1996 Vol. 11 953m 0 100- 80 - 60 - 40 - 20 - d x o 4 v) Q g / 1 rnin-' I - 0.40 0.45 0.50 0.55 0.60 0.65 I 3 1401 1 2 1201 (b) I 601 40 2ol * ..._._._........ * ..... rl *... ........... .* A O L I 0.5 0.7 0.9 1.1 1.3 Q /ml min-' Fig. 5 Variation of KO for the solvents and nebulizers employed. (a) Tot uersus Qg; QI = 1.0 ml min-'. (b) To versus Q,; Qg=0.50 1 min-'.(dotted line) Meinhard nebulizer; (solid line) SBHPPN. A Water; B methanol; C ethanol; D propan-2-01; and E butan-1-01 Meinhard nebulizer as expected from their DSDs whereas Svap is similar for both nebulizers. Obviously the solvent with the highest volatility i.e. methanol gives rise to the highest Svap values. Kreuning and M a e ~ s e n ~ ~ state that the solvent evaporation factor4 is the parameter that determines the frac- tion of solvent in vapour and in liquid form. According to these workers methanol should be the solvent with the highest Svap value whereas water should show the lowest value as is in fact the case (Table 5 ) . Sliq in turn depends on the DSD of the primary aerosol (the higher the proportion of small drops in the primary aerosol the higher the Sliq should be) and on solvent volatility (the higher the solvent volatility the lower the Sliq should be).Thus for instance Sliq for methanol is about the same as for ethanol. The smaller mean drop size of the primary aerosol for methanol is probably compensated for by the lower volatility of ethanol. Emission Intensity Fig. 6(u) shows the variation of the net emission intensity Znet uersus Qg for the nebulizers and solvents employed. For the SBHPPN [Fig. 6(u)(ii)] when Q is increased the emission intensity increases up to a maximum value and then decreases whatever the solvent. For the Meinhard nebulizer the behav- iour is similar but no maximum is found for water and butan- l-ol [Fig. 6(u)(i)]. These results can be explained on the basis of opposite effects that take place when Qg is increased.Since the mean drop size of the primary aerosol becomes lower and there is an increasing volume of dry gas the following occurs (i) W, increases; (ii) S, increases; (iii) the gas load increases; and (iu) the sample residence time in the plasma decreases. Factor (i) will contribute to a higher emission signal whereas factors (ii) (iii) and (iu) will act in the opposite direction. Fig. 6 also shows that the SBHPPN gives rise to higher emission signals than the Meinhard nebulizer since their W, values are higher (Fig. 5). However the relative emission enhancement on switching from the SBHPPN to the Meinhard nebulizer is lower than the corresponding analyte transport enhancement. Also the signal maxima are more noticeable for the SBHPPN than for the Meinhard nebulizer probably because the S, values are also higher for the former (Fig.4). Hence the plasma excitation temperature seems to be lowered when the SBHPPN is used. As regards the effect of the solvents in addition to the above-mentioned factors (i) and (ii) another factor should be borne in mind the nature of the solvent. Fig.6(u) shows that the position of the maxima as well as the intensity val- ues are dependent on the solvent employed. For the com- binations methanol-SBHPPN and propan-2-ol-SBHPPN [Fig. 6(u)(ii)] the I, drop after their maxima is more pro- nounced than for the remaining solvents. Moreover for propan-2-01 maxima are located at a lower Q value ( i e . 0.39 1 min-I). On comparing the net intensity values obtained for the different alcohols it can be observed that the increasing order of net intensity does not coincide with the increasing order of W,,,.Thus for both nebulizers ethanol is the solvent that affords the highest net intensity values. For the remainder of the solvents tested the increasing order of net intensity depends on the gas flow rate employed. For example at high Q values propan-2-01 gives net intensity signals even lower than those of water whereas at low Qg values water gives the lowest I, values. In order to explain these results it is necessary to take into account the interaction between the plasma and the solvent. When organic solvents are introduced into the plasma a factor to be considered is the concentration and nature of the Table 5 Relative values (SBHPPN/Meinhard) of the analyte and solvent transport rates and liquid and gas fractions of the solvent transport rate Solvent Water Methanol Ethanol Propan-2-01 Butan- l-ol Water Methanol Ethanol Propan-2-01 Butan-1-01 Q J 1 min-' 0.43 0.43 0.43 0.43 0.43 0.50 0.50 0.50 0.50 0.50 Qi/ ml min-' 1 .o 1 .o 1 .o 1 .o 1 .o 0.6 0.6 0.6 0.6 0.6 (Stot )re,* 1.44 1.32 1.27 1.42 2.10 ( Kot )re1 t 4.00 2.58 2.66 2.27 3.00 SBHPPN 220 181 1 926 595 373 Meinhard nebulizer 248 1716 1125 613 349 SBHPPN 213 910 926 709 610 Meinhard nebulizer 52 351 336 306 21 1 1.33 1.35 1.16 1.31 1.74 2.12 1.60 1.81 1.95 2.46 157 1796 938 579 317 168 1432 1094 599 280 176 819 83 1 740 649 82 509 428 363 27 5 *(Stot)rcl =(Stat )SBHPPN/(Stot)Meinhard- t(W tot ) re] =(W tot )SBHPPN/(W tot )Meinhard.$See ref. 15. 954 Journal of Analytical Atomic Spectrometry October 1996 Vol. 1 1120 100 m 80 2 x 60 - 0) %" 40 20 0 m I3 K Y s u 0.34 0.37 0.40 0.43 0.46 0.49 600 -1 I / 400 0.34 0.37 0.40 0.43 0.46 0.49 800 m 600 2 Ky 400 s u 200 0 0.34 0.37 0.40 0.43 046 0.49 0.34 0.37 0.40 0.43 0.46 0.49 8000 M C 6000 i5001 5000 .I 0.34 0.37 0.40 0.43 0 46 0.49 M h c Q,' 8000 7500 7000 6500 6000 5500 5000 (d 0) 0.34 0.37 0.40 0.43 0.46 0.49 1 min-' Fig.6 ICP emission parameters uersus Q for the different .,olvents and nebulizers employed. (a) Net emission intensity I,, for ionic line of Mn; (b) net emission intensity of C molecular band lcPc; (c) excitation temperatures. (i) Meinhard nebulizer; (ii) SBHPPN. A Water; B methanol; C ethanol; D propan-2-01; and E butan-1-01.Q,= 1.0 ml min-' solvent pyrolysis products C being the most significant 3*9*10*12 Fig. 6(b) shows C2 emission intensity I C X versus Q It can be observed that for a given nebulizer Ic-C increases with Qg and that for a given solvent the IC-c values obtained v ith the SBHPPN are higher than those for the Meinhard nebulizer. These results are in agreement with the behaviour of Stot (Fig. 4). On comparing the behaviour of I, and I uersus Qg it appears that both intensities increase with Q until the maximum value of I, is reached. From this point I, decreases whereas I increases even faster t lian at lower Q values thus indicating that the plasma exhitation temperature is lowered. As regards the nature of the -,olvent Fig. 6(b) shows that I decreases in the following order propan-2-01 > ethanol > butan-1-01 >methanol. This is the result of the amount of carbon transported to the plasma which is related to Stat and the number of C-C bonds that exist in each solvent molecule.For methaiiol the C species comes from recombination reaction^.^ Fig. 6(c) shows the influence of Qg on T, for all the :.olvents and nebulizers used. It can be seen that T, decreases when Q is increased and also when the SBHPPN is used These results can be attributed to the higher Stot values obtained (Fig. 4). As regards the nature of the solvent propan-2-01 shows the lowest T, values. In agreement with this it also provides the highest I values and the sharpest deciease in I, when Q is increased. Methanol also shows a 1( w T,, which probably highlights the negative effect of a solvrnt load that is too high (Fig.4). It can also be observed that ihe T, values for alcohols are lower than that for water. Thi; might explain why the net emission enhancement obtaintad with alcohols with respect to water is lower than that of the analyte transport rate as shown in Table 6. An additional result that can be derived from Table6 is that the enhancements in transport rate and emission signal achieved by the SBHPPN when switching from water to alcohols are less important than those observed for the Meinhard nebulizer. Table6 includes the LOD for Mn in each of the solvents employed showing that propan-2-01 and water provide the highest LOD values for both nebulizers and that the values obtained for the SBHPPN are lower than those for the Meinhard nebulizer by a factor of about two.The effect of Q1 on the emission intensity is similar to that of Qg although less detrimental since gas load and residence time do not change on varying Q1. CONCLUSIONS (1) The SBHPPN gives rise for all the solvents studied to primary aerosols with smaller Dv,50 values than those produced by the Meinhard nebulizer under the same Q and Q1 con- ditions. The relative decrease in Dv,50 when switching from the Meinhard nebulizer to the SBHPPN is more noticeable for alcohols than for water. In addition for the SBHPPN D3,2 and Dv,50 show little variation along the radial (transversal) position in the aerosol cone whereas for the Meinhard nebul- izer the largest drops are usually found in the outer zone of the aerosol cone.(2) The analyte transport rates Wtot are significantly higher for the SBHPPN than for the Meinhard nebulizer particularly at high liquid flows. However the differences between their S, values are less pronounced. Journal of Analytical Atomic Spectrometry October 1996 Vol. 11 955Table 6 Relative values (solvent i/water) of analyte transport rate and net emission intensity and limits of detection (LOD) for the solvents and nebulizers employed* ( Wot )Ire t (LOD/ng ml- ’)§ Meinhard Solvent 1 min-’ nebulizer SBHPPN Qd Water 0.43 1 .o 1 .o Methanol 0.43 7.6 4.9 Ethanol 0.43 7.0 4.6 Propan-2-01 0.39 6.0 3.4 Butan-1 -01 0.43 4.2 3.1 Meinhard nebulizer SBHPPN 1 .oo 1 .oo 1.64 1.52 3.40 3.14 2.42 2.16 1.64 1.40 Meinhard nebulizer SBHPPN 11.9 5.5 2.8 1.5 2.9 1.2 11.0 6.2 5.4 2.3 *Q1 = 1.0 ml min-’.t(kl(ot)’rel = ( ~ o t ) s o ~ v c n t i/(kl(ot)water. f(Znct)lrel =(Inet)solvent i/(Znet)water. §LOD calculated according to the 3sb criterion where s is the standard deviation from ten replicates of the blank. The experimental conditions were optimized for each solvent. (3) The SBHPPN gives rise for all the solvents studied to higher emission intensities than the Meinhard nebulizer under the same Q and Q conditions. However the relative signal enhancements achieved on changing from the SBHPPN to the Meinhard nebulizer are lower than the corresponding analyte transport enhancements. (4) The relative signal enhancements achieved on switching from water to alcohols are lower for the SBHPPN than for the Meinhard nebulizer. ( 5 ) Conclusions (3) and (4) can be explained in terms of the higher Stot values associated with the SBHPPN which lower the excitation temperature of the plasma in comparison with the Meinhard nebulizer.(6) Solvent viscosity has a small effect on the characteristics of the primary aerosol generated with the SBHPPN at least up to 2.3 cP. This makes the SBHPPN particularly promising for the analysis of high viscosity samples (e.g. wear metals in lubrication oils). This aspect is currently under study in our laboratories. The authors thank the DGICYT (Spain) for financial support of this work (Projects PTRI91-0029 and PB92-0336). REFERENCES 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Cresser M. S. in Solvent Extraction in Flame Spectroscopy Analysis Butterworths London 1978. Boorn A.W. and Browner R. F. in Inductively Coupled Plasma Emission Spectroscopy ed. Boumans P. W. J. M. Wiley New York 1987 p.151. Kreuning G. and Maessen F. J. M. J. Spectrochim. Acta Part B 1989 44 367. Boorn A. W. and Browner R. F. Anal. Chem. 1982 54 1402. Pan C. Zhu G. and Browner R. F. J. Anal. At. Spectrom. 1990 5 537. McCrindle R. I. and Rademeyer C. J. J. Anal. At. Spectrom. 1995 10 399. Blades M. W. and Caughlin B. L. Spectrochim. Acta Part B 1985 40 579. Boumans P. W. J. M. and Lux-Steiner M. Ch. Spectrochim. Acta Part B 1982 37 97. Weir D. G. and Blades M. W. J. Anal. At. Spectrom. 1994 9 1311. Weir D. G. and Blades M. W. J. Anal. At. Spectrom. 1994 9 1323. Botto R. I. Spectrochim. Acta Part B 1987 42 181.Maessen F. J. M. J. Kreunning G. and Balke J. Spectrochim. Acta Part B 1986 41 3. Canals A. Wagner J. Browner R. F. and Hernandis V. Spectrochim. Acta Part B 1988 43 1321. Browner R. F. Canals A. and Hernandis V. Spectrochim. Acta Part B 1992 41,659. Canals A. Hernandis V. and Browner R. F. Spectrochim. Acta Part B 1990 45 591. Cresser M. S. Prog. Anal. At. Spectrosc. 1982 5 35. Long S. E. and Browner R. F. Spectrochim. Acta Part B 1988 43 1461. 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 Pan C. Zhu G. and Browner R. F. J. Anal. At. Spectrom. 1992 7 1231. Nisamaneepong W. Haas D. L. and Caruso J. A. Spectrochim. Acta Part B 1985 40 3. Brotherton T. Barnes B. Vela N. and Caruso J. J. Anal. At. Spectrom. 1987 2 389.Wiederin D. R. Houk R. S. Royce K. W. and D’Silva A. P. Anal. Chem. 1990 62 1155. Botto R. I. and Zhu J. J. J.Ana1. At. Spectrom. 1994 9 905. Laborda F. de Loos-Vollebregt M. T. C. and de Galan L. Spectrochim. Acta Part B 1991 46 1089. Maessen F. J. M. J. Seeverens P. J. H. and Kreuning G. Spectrochim. Acta Part B 1984 39 1171. Nobile A. Jr. Shuler R. G. and Smith J. E. Jr. At. Spectrom. 1982 3 73. Todoli J. L. Canals A. and Hernandis V. Spectrochim. Acta Part B 1993 48 373. Todoli J. L. Canals A. and Hernandis V. Spectrochim. Acta Part B 1993 48 1461. Todoli J. L. Mufioz M. Valiente M. Hernandis V. and Canals A. Appl. Spectrosc. 1994 48 573. Hernandis V. Todoli J. L. Canals A. and Sala J. V. Spectrochim. Acta Part B 1995 50 985. Mora J. Hernandis V. and Canals A. J. Anal. At. Spectrom. 1991 6 573. Hirleman E. D. in Liquid Particle Size Measurement Techniques eds. Tishkoff J. M. Ingebo R. D. and Kennedy J. B. ASTM STP848 Philadelphia 1984 p. 35. Malvern Instruments 2600 Laser Diffraction User Manual Malvern 1987. Tarr M. A. Zhu G. and Browner R. F. J. Anal. At. Spectrom. 1992 7 813. Alder J. F. Bombelka R. M. and Kirkbright G. F. Spectrochim. Acta Part B 1980 35 163. Nakamura S. J. Anal. At. Spectrom. 1995 10 467. Walker A. L. Curry D. L. and Fannin H. B. Appl. Spectrosc. 1994,48 333. Sharp B. L. J. Anal. At. Spectrom. 1988 3 613. Farino J. and Browner R. F. Anal. Chem. 1984 56 2709. Boorn A. W. Cresser M. S. and Browner R. F. Spectrochim. Acta Part B 1980 35 823. Tishkoff J. M. Hammond D. C. Jr. and Chraplyvy A. R. J. Fluids Eng. 1982 104 313. Schulte G. Dannehl M. and Bauckhage K. Presented at Partec 89 Nurnberg 1989. Dodge L. G. Rhodes D. J. and Reitz R. D. Appl. Opt. 1987 26,2144. Faske A. J. PhD Thesis Georgia Institute of Technology USA 1986. Shum S. C. K. Johnson S. K. Pang H. and Houk R. S. Appl. Spectrosc. 1993 47 575. Liu H. and Montaser A. Anal. Chem. 1994 66 3233. Olesik J. W. and Fister J. C. 111 Spectrochim. Acta Part B 1991 46 851. Kreuning G. and Maessen F. J. M. J. Spectrochim. Acta Part B 1987 42 677. Paper 610321 5J Received May 8 1996 Accepted July 11 1996 956 Journal of Analytical Atomic Spectrometry October 1996 Vol. 11

ISSN:0267-9477    

DOI:10.1039/JA9961100949

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Classification of Emission Lines of the Group IllB Elements Aluminium Gallium and Indium Excited by Grimm Glow Discharge Plasmas Using Several Different Plasma Gases KAZUAKI WAGATSUMA Institute for Materials Research Tohoku University Katahira 2-! - 1 Sendai 980 Japan In order to clarify the spectroscopic characteristics of the Group IIIB elements in Grimm glow discharge plasmas the relative intensities of their emission lines in the wavelength range 160-800 nm were investigated and tabulated when using different plasma gases oiz. argon neon argowhelium and neowhelium. It was characteristic of the glow discharge light source that the emission spectra observed were strongly dependent on the type of plasma gas employed. Extremely intense ionic lines were observed when neon rather than argon was employed as the plasma gas.Furthermore different groups of ionic lines could be observed only in the helium- containing plasmas. The excitation of these ionic lines caii be principally attributed to charge-transfer collisions between the analyte atoms and the plasma gas ions. Keywords Grimm glow discharge plasma; Group I I I B element; emission line; plasma gas The hollow anode glow discharge tube as first developed by Grimm,'V2 has been employed extensively in atomic emission spectrometry. This analytical technique has been appl led to surface analysis and to the elemental analysis of solid because cathode sputtering occurring on the sample surface results in the ejection of analyte atoms.8 Unlike the numerous reported applications little fundamental rexearch on the emission mechanisms has been performed.Several papers have pointed out that the emission lines excited hy the glow discharge depend strongly on the plasma gas 25 20 15 2 10 5 0 A12+ 20 - ..... ...._.... ,. ,...... - 3p4s 7s-6d- -7P =-7f - 4f 6s-M- -6p -\ 6f . 5s =4d= -5p -\5f t 4 N e' ~ . 5 ~ Arm 15.76 eV Fig. 1 Energy level diagram of singly ionized aluminiuni Journal of Analytical Atomic Spectrometry employed;g-" however a complete wavelength table of the glow discharge spectrum has not been published. It is therefore worthwhile classifying the emission lines for analytical use so that the analytical line and the operating conditions of the excitation source can be properly determined. It has been reported that in Grimm glow discharge particular ionic lines are selectively excited through (quasi-)resonance charge-transfer ~ollisions'~ between a ground state ion of the plasma gas (G) and a ground state atom of the sample (M) G + + Mg-+Gg+(M')*.This reaction produces a highly populated excited state of the sample (M+)* which subsequently leads to the emission of an ionic line. It should be noted that such excitations occurring in the glow discharge plasma are primarily dependent on the properties of the plasma gas such as the ionization energy which implies that selectively excited ionic lines might be found for various elements. h cn C 3 - .- 2 2 e .- c 4- .- (d v cn Q c c 0 fn fn W c. .- .- .- E n 357 358 359 360 361 Wavelengthhm Fig. 2 Spectral scans of A1 I1 emission lines in the wavelength range 357-361 nm excited by an argon (a) an argon-helium mixed gas (b) and a neon plasma (c). Discharge conditions Ar 4.0 x lo2 Pa; 500 V (a) Ar 4.0 x lo2 Pa plus He 5.3 x lo2 Pa; 500 V (b) and Ne 5.3 x lo2 Pa; 600 V (c) Journal of Analytical Atomic Spectrometry October 1996 Vol.1 1 (957-966) 957Table 1 Observed emission lines of aluminium and their assignment Wavelength/nm Assignment Reduced intensity* in vacuo I1 162.562 I1 167.079 I1 171.944 I1 172.124 I1 172.127 I1 172.495 I1 172.498 I1 175.062 I1 176.010 I1 176.197 I1 176.395 I1 176.581 I1 176.773 I1 182.857 I1 183.283 I1 183.481 I1 185.592 I1 185.802 TI 185.999 I1 186.231 I1 192.996 I1 193.237 I1 193.449 I1 193.470 I1 193.691 I1 193.925 I1 196.146 I1 196.265 I1 196.269 I1 196.273 I1 199.053 I1 207.466 I1 209.492 I1 209.540 I1 209.545 I1 209.571 IT 209.576 I1 209.579 I 214.623 1215.138 I 215.140 I 216.948 I 217.475 I 217.477 I1 219.329 I1 219.488 I1 219.493 I1 219.608 I1 219.614 I1 219.618 I 220.535 I 221.075 I 221.082 I 226.416 I 226.979 I 226.992 I1 232.490 I1 232.615 I1 232.620 I1 232.712 I 236.778 I 237.385 I 237.408 I 238.080 in air 207.400 209.425 209.473 209.478 209.504 209.509 209.5 12 214.555 21 5.070 21 5.072 2 16.880 2 17.407 2 17.409 2 19.260 2 19.4 19 219.424 2 19.539 2 19.545 219.549 220.466 22 1.006 22 1.01 3 226.346 226.909 226.922 232.419 232.543 232.549 232.641 236.706 237.3 13 237.336 238.007 Upper (eV) 5 s 'So (15.047) 3p 'P (7.4205) 3d 3D (11.847) 3d 3D (11.847) 3d 3D2 ( 11.846) 3d 3D (11.846) 3d 3D3 (11.846) 7f IF3 (17.680) 3p 3P2 (11.688) 3p 3P1 (1 1.672) 3p 3P (1 1.688) 3P 3P (1 1.665) 3p 3P (1 1.672) 3p4s jP (18.097) 3p4s 3P (18.081) 3p4s 3P0 (18.073) 4s 3S (11.316) 4s 3S (11.316) 6f 'F3 (17.264) 4s 3S (11.316) 3p4s 'P ( 18.097) 3p4s 3P1 (18.081) 3p4s 3P (18.097) 3p4s 3P (18.081) 3p4s 3P0 (18.073) 3p4s 3P (18.081) 8f 3F (17.993) 9f 3F3 (18.163) 9f 3F3 (18.163) 9f 3F2 (18.163) 3d ID (13.649) 5f 'F3 (16.574) 7f 3F (17.764) 7f 3F3 (17.763) 7f 3F3 (17.763) 7f 3F (17.762) 7f 3F2 (17.762) 7f 3F (17.762) 9d 2D3/2 (5.7767) 9d 'D,/ (5.7768) 9d 2D3/ (5.7767) 8d 'D3/ (5.7148) 8d 2Ds,2 (5.7148) 3p3d 3F (17.499) 3p3d 3F3 (17.495) 3p3d 3F3 (17.495) 3p3d 3F2 (17.492) 3p3d 3F (17.492) 3p3d 3F2 (17.492) 7d 'Ds12 (5.6220) 8d 2D3,2 (5.7148) 7d 'D3/2 (5.6218) 7d 'D3/2 (5.6218) 6d 2D3/ (5.4758) 6d 2D,I (5.4761) 6f 3F (17.179) 6f 3F3 (17.176) 6f 3F3 (17.176) 6f 3F (17.174) 5d 'D5/ (5.2367) 6s 'SlI2 (5.2215) 6d 'D312 (5.4758) 5d 'D3/ (5.2362) 5d 'D3/2 (5.2362) Lower (eV) 3p P (7.4205) 3p 3P0 (4.6360) 3p 3P (4.6436) 3p 3P (4.6436) ] 3p 3P2 (4.6590) 3p 3P2 (4.6590) ] 3p ID2 (10.598) 3p 3P (4.6436) 3p 3P (4.6360) 3p 3P (4.6590) 3p 3P (4.6436) 3p 3P2 (4.6589) 4s 3S (11.316) 4s 3S (11.316) 4s 'S1 (11.316) 3p 3P0 (4.6360) 3p 3P (4.6435) 3p 'D (10.598) 3p 3P (4.6589) 3p 3P (11.673) 3p 3P0 (11.665) 3p 3P (1 1.688) 3p 3P (1 1.672) 3p 3P (11.672) 3p 3P2 ( 11.687) 3p 3P (1 1.672) 3d 3D3 ( 11.846) 3cl 3D2 ( 11.846) 3cl 3D2 ( 11.846 j 31) 'PI (7.4205) 'D (10.598) 3D3 (11.846) 3d 3D3 ( 11.846) 3d 3D2 (1 1.846) ] 3d 3D3 (11.846) 3d 3D2 ( 11.846) 3d 3D ( 1 1.847) 3s 'so (0.0000) 1 313 2P,/ (0.0000) 31) 2P,/ (0.0000) 1 1 31) 'P3/ (0.01 39) 31) 2P3/2 (0.0139) 3p 'P3/2 (0.0139) 31) 'P312 (0.0139) 3d 3D3 (1 1.846) 3d 3D3 ( 11.846) 3d 3D2 ( 11.846) ] 3d 3D3 ( 11.846) 3d 3D ( 11.846) 3d 3D ( 11 347) 3d 2P,/ (0.0000) 3p ZP,/ (0.0000) 1 1 1 31) 'P3/2 (0.0139) 3p 'P3/ (0.0139 j 31) 'P3/ (0.0139) 3p 2P3/2 (0.0139) 3d 3D3 (11.846) 3d 3D3 (1 1.846) 3d 3D ( 11.846) 3tl 3D3 ( 11.846) 313 2P112 (0.0000) 31) 'P3/2 (0.0139) 3p 'P3/ (0.0139) 313 'P3/2 (0.01 39) Art < 1 200 20 25 45 < 1 5 5 20 10 5 < 1 < 1 < 1 6 10 < 1 15 < 1 < 1 < 1 < 1 < 1 <1 < 1 ArT[ 10 < 1 < 1 < 1 20 40 30 60 <1 < 1 ArT[ 50 100 80 130 < 1 < 1 < 1 70 y 8 Ar-Hef 30 520 170 5 10 850 15 60 50 210 60 65 160 120 50 100 3 10 45 5 10 45 (40) 130 55 65 30 20 340 130 80 40 30 60 65 140 110 90 (80) 140 250 220 370 8 8 8 220 330 55 25 Ne§ 310 1500 540 1600 2800 <1 120 80 380 70 95 < I < 1 < l 330 970 < I 1600 100 < 1 < 1 < 1 < 1 < 1 < 1 <1 1100 < 1 <1 < 1 5 8 9 15 < 1 < 1 < 1 15 30 25 45 < 1 < 1 < 1 25 40 10 5 * Normalized per unit amount of the sputtered sample.t Discharge conditions Ar 4.0 x 10' Pa; 400 V; 42.2 mA. $ Discharge conditions Ar 4.0 x 10' Pa plus He 5.3 x 10' Pa; 450 V; 38.6 mA. 0 Discharge conditions Ne 5.3 x 10' Pa; 700 V; 20.8 mA. 7 Not estimated due to overlapping with gas emission lines. 11 Standard for estimating the reduced intensities. 958 Journal of Analytical Atomic Spectrometry October 1996 Vol. 11Table 2 Observed emission lines of aluminium and their assignment Wavelenrrthl Assignment Reduced intensity* (nm) (in iir)' 1236.706 1237.314 I1 247.526 I 236.798 1257.509 I1 263.152 I1 263.769 I1 263.818 I1 263.825 I1 263.853 I1 263.860 I1 263.867 1265.248 I 266.039 I1 266.916 I1 281.618 I1 299.547 I1 299.817 I 308.215 I 309.270 I 309.284 I1 358.660 I1 358.699 I1 358.713 I1 358.725 I1 358.751 I1 364.926 I1 365.099 I1 365.500 I1 370.334 I1 373.195 I1 373.377 I1 386.618 I 394.401 I 396.153 I1 466.296 I1 559.348 I1 585.387 I1 586.176 I1 618.143 II 618.220 I1 618.335 I1 620.175 I1 622.648 I1 623.153 I1 624.320 I1 633.579 IT 681.708 I1 682.313 I1 683.712 I1 692.042 I1 704.205 Upper (ev) 5d ,D3/2 (5.2362) 5d ,D5! (5.2367) 5p 'P (15.605) 4f 'F3 (15.308) 5f 3F (16.545) 5f 3F3 (16.544) 5f 3F3 (16.544) 5f 3F (16.544) 5f 3F (16.544) Sf 3F (16.544) 5s 'S1/ (4.6728) 5s 2S112 (4.6728) 3p 3P1 (4.6436) 4s 'So (11.822) 6d 3D, (17.210) 6d 3D,,2 (17.210) 3d 'D,,' (4.0214) 4d 'D3/2 (4.8265) 4d 'D5,2 (4.8271) 3d ,D5/2 (4.0216) 3d ,D3/2 (4.0214) 4f 3F4 (15.302) 4f 3F3 (15.302) 4f 3F3 (15.302) 4f 3F (15.301) 4f 3F (15.301) 5d 3D1 (16.468) 5d 3D1,2 (16.468) 5d 3D2 (16.468) 5d 'D2 (16.603) 6s 3S1 (16.392) 6s 3S1 (16.392) 6s 'So (16.462) 4s 'Sl/ (3.1426) 4s 2Sl/z (3.1426) 'P (13.256) 3 'D (15.472) 6f 3F4 (17.179) 6f 3F3 (17.176) 6g 3G3 (17.307) 6g 3G3,4 (17.307) 6g 3G3.4,5 (17.307 6g 'G (17.307) 4d 3D (15.062) 4d 'D1.(15.062) 4d 3D1,2,3 (15.062 5p 'P (15.605) 5s 3S1 (14.889) 5s 3S1 (14.889) 5s 3S1 (14.889) 5s 'So (15.047) 4p 3P (13.076) Lower (eV) 3p (0.0000) 3p 2P1 (0.0000) 3p 'P3/2 (0.0139) 3p 'D (10.598) 3p 'P, (0.0139) 3p 'D (10.598) 3d 3D3 (11.846) 3d 'D3 (1 1.846) 3d 3D2 (1 1.846) 3d 3D3 ( 1 1.846) 36 3D (1 1.846) 3d 3D1 (1 1.847) 3p 2P3/2 (0.0139) 3p 'P (7.4205) 4p 3P1 (13.073) 4p 'P (13.076) 3P ,p1/2 ( 0 .~ 0 ) 3s 'so (0.0000) 3p 2P1 (0.0000) 3p ,P3/2 (0.0139) 3p ,P3/2 (0.0139) 3d 3D3 ( 11.846) 3d 3D3 (1 1.846) 3d 3D2 (1 1.846) 3d 3D3 (1 1.846) 3d 'D (1 1.847) 4p 3P (13.071) 4p 3P1 (13.073) 4p 3P2 ( 13.076) 4p 'P1 (13.256) 4p 3P0 (13.071) 4p 3P1 (13.073) 4p 'PI (13.256) 3p 'D (0.0000) 4p 'P1 (13.256) 4d 3D3 ( 15.062) 4d 3D2 (15.062) 4f 3F2 (15.301) 4f 3F3 (15.302) 4f 3F4 (15.302) 4f 'F3 (15.308) 4p 3P0 (13.071) 4p 3P (13.073) 4p 3P (13.076) 3d 'D (13.649) 4p 3P0 (13.071) 4p 3P1 (13.073) 4p 3P2 (13.076) 4p 'P (13.256) 4s 3S1 (11.316) 3p 2P1/2 (0.0000) 3p 'P3/2 (0.0139) Art 60 100 II <1 140 230 < 1 <1 1 1 <1 <1 <1 70 140 < 1 30 <1 < 1 2700 4100 <1 < 1 < 1 <1 <1 < 1 < 1 <1 < 1 8700 9900 < 1 <1 <1 <1 < 1 <1 <1 < I <1 <1 <1 < 1 <1 <1 < 1 <1 ArB Arll Ar-He$ 190 320 15 540 890 280 45 30 30 30 230 480 2 250 4 4 7300 14 000 1800 1300 1100 45 160 90 20 30 45 12 OOO 18 000 60 55 8 5 8 8 15 8 40 120 (230) 3 10 20 30 35 Ne§ 35 55 200 80 150 1700 <1 <1 <1 <1 40 75 40 180 <1 < 1 1800 3100 14000 10 OOO 7900 1 3 3 3 <1 <1 <1 4100 5300 2200 840 < 1 < 1 < 1 <1 <1 1100 3300 5800 140 50 160 320 700 840 Nen * Normalized per unit amount of the sputtered sample.t Discharge conditions Ar 4.0 x 10' Pa; 400 V; 47.2 mA. 1 Discharge conditions Ar 4.0 x lo2 Pa plus He 5.3 x 10 Pa; 450 V; 42.0 mA. 6 Discharge conditions Ne 5.3 x 10 Pa; 700 V; 23.6 mA. f Not estimated due to overlapping with gas emission lines. // Standard for estimating the reduced intensifier. In this paper the emission lines of the Group IIIB elements aluminium gallium and indium emitted by Grimm glow discharge plasmas were investigated. The emission lines were classified according to their transition schemes and the relative intensities were compared for several different plasma gases uiz. argon neon argon-helium and neon-helium mixed gases. A common feature of the Group IIIB elements is that the second ionization energy is relatively high whereas the first ionization from the neutral atom takes place easily.Hence there are distinct excited levels belonging to the various electron configurations leading to spectra rich in the ionic emission lines. It has been reported that the hollow cathode laser based on Al I1 emission lines can be pumped by energy transfer from neon ions. Steers and Leis15 have indicated that the intense A1 I1 lines assigned to the 3s4f-3s3d transition can be excited principally through charge-transfer collisions in which ground state ions of neon are involved. Previous studies have also shown that for the Group IIIB elements the spectra obtained with the glow discharge might be different from those obtained with other sources and might depend on the plasma gas used. Hence complete wavelength tables are required not only for analytical applications but also for investigations into the excitation mechanisms.EXPERIMENTAL The structure of the glow discharge tube employed here has been described el~ewhere.'~ It was constructed according to the original model reported by Grimm.' The inner diameter Journal of Analytical Atomic Spectrometry October 1996 Vol. 11 959a 1 182 183 184 185 186 187 Wavelengthhm Fig. 3 Spectral scans of A1 I1 emission lines in the wavelength range 182-187 nm excited by an argon (a) an argon-helium mixed gas (b) and a neon plasma (c). Discharge conditions Ar 4.0 x 10’ Pa; 550 V (a) Ar 4.0 x lo2 Pa plus He 5.3 x 10’ Pa; 550 V (b) and Ne 5.3 x 10’ Pa; 700 v (c) A 0- 200 400 600 80 2 0 0 4 0 0 6 0 0 8 0 0 of the hollow anode was 8.0mm and the distance between the anode and cathode was adjusted to be 0.2-0.3mm.The discharge tube was evacuated down to about 1.3 Pa (1 x lo- Torr) and then the plasma gas was introduced. High- purity argon (99.9995%) neon (99.99%) and helium (99.9999%) were used. Flow control of the plasma gas was carried out with a ball (on/off) valve and a needle valve which were inserted in each gas line. The partial pressure of each gas was regulated with the needle valve and read on a Pirani gauge at the gas inlet of the tube when the ball valves of the other gas lines were closed. The scales of the gauge had been calibrated for each pure gas. The discharge power was supplied with a dc power supply device (Model PAD lK-0.2L Kikusui Electronics Tokyo Japan).All measurements were conducted in constant-voltage mode. A Fastie-Ebert mounting spectrograph (Model GE-340S Shimadzu Kyoto Japan) equipped with a photomultiplier tube (Model R-955 Hamamatsu Photonics Tokyo Japan) was employed to measure the spectra in the wavelength region 230-800 nm. The focal length is 3.4 m. The grating has 1200 grooves mm-’ and a blaze wavelength of 300 nm. The emission spectra in the 160-245 nm wavelength range were recorded on an Eagle-mounting vacuum spectrometer (Model EGV-200 Shimadzu) equipped with a CaF window and a photomultipl- ier tube (Model R-l66UH Hamamatsu Photonics). The focal length is 2.0 m. The concave grating has 1200 grooves mm-l and a blaze wavelength of 170nm. The spectrometer was evacuated to about 1.3 Pa with a rotary vacuum pump.The intensity data were recorded at scan rates of 0.008 nm s-l (GE-340s) and 0.010 nm s-l (EGV-200) with an analogue chart recorder. (b) I (8) Al I1 624.32 nm I ! ’ 0 L o -0 - -0 _I L.-8.-:-;; 200400600800 2 0 0 4 0 0 6 0 0 8 0 Discharge voltageN Fig. 4 Discharge voltage dependence of the discharge current (a) the sputtering rate (b) and of the emission intensities of A1 I1 281.618 nm (c) A1 I1 358.660 nm ( d ) A1 I1 624.320 nm (e) and A1 I 396.153 nm (f). Plasma gases Ar 4.0 x lo2 Pa (circles) and Ne 5.3 x 10’ Pa (squares) 960 Journal of Analytical Atomic Spectrometry October 1996 Vol. 1115 t 3 - 2 ' 1 ' 0 A (e) Al II 209.49 nm A / A/ ,/ A' A A - o-mo--@ -I . 0 400 600 800 1000 i .1" i i ,m' 11' ,i i I I d 60 40 20 0 d . 6 8 - 400 600 800 1000 100 t 0 v) .- .% 20 i c Q Z O 400 600 8CiO loo0 Discharge voltageN Fig.5 Discharge voltage dependence of the discharge current (id) and of the emission intensities of A1 I1 162.562 nm (b) A1 I1 167.079 nm (c) A1 I1 172.492 nm (d) and A1 I1 209.492 nm (e). Plasma gases Ar 4.0 x 102 Pa (circles) Ar 4.0 x lo2 Pa plus He 5.3 x ld Pa (triangles) and Ne 5.3 x lo2 Pa (squares) Table 3 Observed emission lines of gallium and their assignmeni Wavelength/nm Asbignment Reduced intensity* in vacuo I1 153.551 I1 153.631 I1 166.937 I1 169.581 I1 179.944 I1 181.399 I1 184.526 I1 209.135 1219.603 1223.661 I 225.989 1229.487 1229.855 1233.894 1233.923 1237.201 I 241.942 I 245.082 I 250.094 in air 209.069 219.534 223.592 225.989 229.416 229.784 233.822 233.851 237.129 241.869 245.008 250.019 Upper (eV) 4d 3D3 (14.118) 4d 'D2 (14.1 14) 4d ID2 (13.355) 4d 'D (13.355) 5s (12.763) 5s ?St (12.763) 5s 3S1 (12.763) 4p 3P1 (5.9283) 9s 2s1/2 (5.6457) 9s 2s1/2 (5.6457) 7d 2D5/2 (5.5886) 6d 'D312 (5.4025) 6d 'D5/2 (5.4032) 6d 'D3p (5.4025) 8s 2S,12 (5.4963) 7s 2Sl/2 (5.2269) 7s 2S112 (5.2269) 5d 'DU2 (5.0598) 5d 'D3/2 (5.0588) Lower (eV) 4p 3P2 (6.0441) 4p 3P2 (6.0441) 4p 3P (5.9283) 4p 3P2 (6.0441) 4p 'Po (5.8730) 4p 3P1 (5.9283) 4p 3P2 (6.0441) 4s 'So (O.oo00) 4P 2p1/2 (O.oo00) 4P 2pl (O.oo00) 4P %/2 (O*ooOo) 4P (O.oo00) 4p 'P3/2 (0.1024) 4p 'P3/2 (0.1024) 4p 2P3/2 (0.1024) 4p 2P3/2 (0.1024) 4p 'P3/2 (0.1024) 4p 'P3/2 (0.1024) 4p 'P312 (0.1024) Art <1 < 1 < 1 < 1 < 1 <1 <1 780 10 10 10 30 15 45 6 10 15 80 loon Ar-He$ 15 < 1 < 1 30 100 400 640 1500 70 50 30 120 40 190 25 30 65 360 460 N4 160 40 30 220 1400 4200 6800 6OOo <1 < 1 < 1 5 5 20 1 5 10 45 50 * Normalized per unit amount of the sputtered sample.t Discharge conditions Ar 4.0 x 102 Pa; 450 V; 28.0 mA. $ Discharge conditions Ar 4.0 x I d Pa plus He 5.3 x lo2 Pa; 501 V; 26.8 mA. Q Discharge conditions Ne 5.3 x lo2 Pa; 800 V; 12.3 mA. Standard for estimating the reduced intensities. Aluminium plates (99.99% purity) tin-gallium alloy blocks (containing about 20% m/m Ga) and tin-indium alloy blocks (containing about 20% m/m In) were prepared as the cathode samples. It is particularly difficult to handle gallium metal as the sample because of its low melting-point. However pro- longed stable discharges could be attained for the alloy samples.The spectral analysis of gallium or indium emission lines was not significantly affected by the appearance of emission lines Journal of Analytical Atomic Spectrometry October 1996 VoL 11 961100 \ ‘. \ \ \ a\ - ‘h 7’ \A/o ’b Helium partial pressure/l O2 Pa - I - - . - (h) A1 I1 209.49 nrn 7 - 4 - / / m 2 - L/ 17’’ // // . / / A * . . ‘ - Fig. 6 Variations in the discharge current (a) the sputtering rate (b) and in the emission intensities of A1 I1 162.562 nm (c) A1 I1 167.079 nm ( d ) Al I1 172.492 nm (e) A1 I1 186.231 nm (f) A1 I1 207.466 nm (8) and A1 I1 209.492 nm (h) as a function of helium partial pressures added to an argon (circles) or a neon (squares) plasma. Discharge conditions Ar 4.0 x 10’ Pa; 700 V and Ne 5.3 x 10’ Pa; 800 V of tin because only a few such lines were observed in the spectrum. of the sputtered sample (the sputtering rate) so that they can be compared for different plasma gases.From the viewpoint of the optical transition the observed A1 I1 lines can be RESULTS AND DISCUSSION Aluminium Table 1 summarizes the A1 I and A1 I1 emission lines measured with the vacuum spectrometer and Table 2 the aluminium lines measured with the air-path spectrometer. The first and second columns give the wavelength values which were calcu- lated from the energy level data in ref. 20 and their assignments respectively. The remaining columns provide the emission intensities observed with argon gas an argon-helium mixture and neon gas. The intensities are normalized per unit amount classified -into the following groups 3s3p + 3s2 (resonance transition) 3sns (n =4 5) -+ 3s3p 3P2 + 3s3p 3snd (n = 3 4) 4 3s3p 3snf (n =4 5 6 .. .) + 3 ~ 3 d 3p4s 4 3s4s or 3p2 3p3d + 3s3d Fig. 1 shows a schematic energy diagram of singly ionized aluminium together with the ionization potentials of argon neon and helium. 962 Journal of Analytical Atomic Spectrometry October 1996 Vol. 1 1I 25 - 20 2 15 10 5 Ar Ne He In2+ - 61 -51 r 7s -6d- -7P - 41 - as -7d- - 5p2=5d= - ==6P 6s '-" Ar + T - IErn 21 .! 5P - 15 76 eV I In+ .. . . 7 5.78 ev 0 - i ' A- Fig. 8 Energy level diagram of singly ionized indium Fig. 7 Energy level diagram of singly ionized gallium Table 4 Observed emission lines of gallium and their assignme!it Wavelength/ (nm) (in air) I 241.868 I 245.008 I 250.019 I 250.072 I1 251.351 I1 270.052 I 271.966 I1 278.036 I 287.424 I 294.364 I 294.417 I1 296.940 I1 297.081 I1 297.160 I1 297.461 I1 337.467 I1 337.581 I1 347.039 I1 370.581 I1 374.488 I1 392.462 1403.297 I 417.203 I1 425.104 I1 425.376 I1 425.557 I1 426.174 I1 633.418 I1 641.960 I1 645.650 Assignment Reduced intensity* Upper (eV) 7s 2Sl (5.2269) 5d 2Dsiz (5.0598) 5f 'F3 (18.286) 4d 'D2 (13.355) 6s 'Sli2 (4.6598) 5s 'So (13.233) 4d 2DS,2 (4.3131) 5f 3F2 (18.285) 5f 'F3 (18.286) 5f 3F2 ( 18.285) 5f 3F2,3 (18.285) 4f 'F3 ( 17.028) 4f 3F2 (17.027) 6d 3D3 (18.291) 7s 3S1 (18.038) 7s 3S1 (18.038) 7s 'So (18.103) 5s 2S1/2 (3.0733) 5s 2 S l i 2 (3.0733) 4f 3F2 (17.027) 4f 'F3 ( 17.028) 4f 3F2.3 (17.027) 4f 3F2,3,4 (17.027) 5d 2D3i2 (5.0588) 5d 'D3/2 (5.0588) 4d 'D3,2 (4.3 123) 4d 2D3/2 (4.3123) 5p 3P2 (14.720) 5p 3P1 (14.694) 5p 'Po (14.683) ~ Lo..ver (eV) 4p 2P /2 (0 1024) 4p 'P (0.1024) 4p 2P ,2 (0.1024) 4d 'D I (13.355) 4p 'P (8.7654) 4p 'P ,2 (0.1024) 4p 'P (8.7655) 4p 'P (0.1024) 4p 'P ,2 (0.1024) 4d 3D (14.1 11) 4d 3D (14.114) 4d 3D (14.1 14) 4d 3D (14.118) 4d 'D (13.355) 4d 'D (13.355) 5p 3P (14.720) 5p 3P (14.694) 5p 3P (14.720) 5p 'P (14.945) 4p 'P (0.1024) 4d 3D (14.111) 4d 3D (14.114) 4d 3D (14.114) 4d 3D (14.1 18) 5s 3S1 (12.763) 5s 3S1 (12.763) 5s 3S1 (12.763) 4p 2P ,' (O.OOO0) 4p 2P 2 (0.0000) 4p 'P 2 (0.0000) Art 3 45 15 < 1 < I 110 < 1 1500 2300 480 <1 < I < l < 1 < 1 < 1 < 1 < 1 < 1 < I 6400 9500 < 1 < 1 <1 < 1 <1 < 1 < 1 loon Ar-He 20 220 480 65 30 35 420 40 4300 7500 1400 95 25 120 220 200 55 55 30 60 200 12 000 18 OOO 500 150 500 lo00 80 20 3 NeS 3 30 70 7 2 400 8 1800 900 1600 280 < 1 < 1 < 1 < 1 < 1 < 1 < 1 < 1 < 1 < 1 2800 4800 < 1 < I < 1 < 1 3200 300 60 * Normalized per unit amount of the sputtered sample.t Discharge conditions Ar 4.0 x lo2 Pa; 450 V; 30.2 mA. $ Discharge conditions Ar 4.0 x lo2 Pa plus He 5.3 x 10' Pa; .;OO V; 29.2 mA. 9 Discharge conditions Ne 5.3 x lo2 Pa; 800 V; 13.8 mA. Standard for estimating the reduced intensities. Fig. 2 indicates spectral scans in the wavelength range 357-361 nm measured with argon gas alone (a) with an argon- helium mixed gas (b) and with neon gas (c). In the argon- excited plasma no A1 11 lines appear in the spectrum consisting of several emission lines of argon ions.However in addition to neon lines at least three A1 I1 lines uiz. 358.660 358.713 and 358.751 nm lines are observed when using neon gas. The A1 I1 lines result from the 3s4f-3s3d transition (see Tables 1 and 2). The addition of helium to the argon plasma enables these A1 I1 lines to be emitted although the intensities are weaker than those in the neon plasma. Fig. 3 shows spectral scans in the wavelength range 182-187 nm. Intense triplet A1 I1 lines which are attributed to transitions from the 3s4s excited level (1 1.316 eV) appear in the neon-excited spectrum. Journal of Analytical Atomic Spectrometry October 1996 Vol. 1 1 963Table 5 Observed emission lines of indium and their assignment Wavelength/nm Assignment Reduced intensity* in vucuo in air I1 157.153 I1 158.645 I1 165.743 I1 167.199 I1 167.412 I1 170.007 I1 170.256 I1 171.662 I1 177.066 I1 177.484 I1 184.248 I1 193.063 I1 193.633 I1 196.672 I1 197.746 11 207.935 207.869 I1 230.686 230.6 15 Upper (eV) 6d 'D (15.707) 5p 'P (7.8150) 7s ' S o (15.295) 5d 'D (12.656) 5p2 3P2 (13.087) 5d 3D2 (12.667) 5d 'D (12.656) 5p2 3P0 (12.567) 5d 3D3 ( 12.684) 5d 3D2 (12.667) 5p2 'D (12.103) 5p2 'D (12.103) 6s 3S1 (11.644) 5d 'D ( 14.1 19) 6s 3S1 ( 1 1.644) 6s 'S (11.644) 5p 3P (5.3745) Lower (eV) 5p 'P (7.8150) 5p 'P (7.8150) 5p 'Po (5.2413) 5p 3P (5.6817) 5p 'P (5.3745) 5p 3P1 (5.3745) 5p 'P (5.3745) 5p 3P (5.68 17) 5p 3P (5.6817) 5p 'P (5.3745) 5p 'P (5.6817) 5p 3P0 (5.2413) 5p 'P (7.8150) 5p 'P (5.3745) 5p 'P (5.6817) 5s ' S o (O.oo00) 5s ' S o (O.oo00) Art < 1 4 < 1 < 1 < 1 < 1 < 1 < 1 2 < 1 < 1 < 1 < 1 < 1 1 2 loql Ar-He$ 1 7 < 1 3 1 7 2 < I 20 3 < I < 1 1 5 4 8 130 Ne$ 8 50 25 15 3 40 10 5 110 15 4 15 6 20 15 35 190 * Normalized per unit amount of the sputtered sample.t Discharge conditions Ar 4.0 x 10 Pa; 450 V; 31.6 mA. $ Discharge conditions Ar 4.0 x 10 Pa plus He 5.3 x 10' Pa; 500 V; 26.8 mA. 9 Discharge conditions Ne 5.3 x 10 Pa; 800 V; 11.0 mA. 1 Standard for estimating the reduced intensities. Such triplet A1 I1 lines are also observed with argon gas although the intensities are much lowered compared with those in the neon plasma. It is further noted that in the argon- helium spectrum there is another group of A1 I1 emission lines uiz.182.857 183.283 and 183.481 nm lines which are not emitted by the neon or argon plasma. These A1 I1 lines result from the 3p4s-3~4~ transition and the upper levels have excitation energies of more than 18 eV (see Tables 1 and 2). It is possible to observe these A1 I1 lines also in neon-helium mixed gas plasmas which implies that helium is principally responsible for the excitations. Fig. 4 shows the voltage dependence of the discharge current (a) the sputtering rate (b) and the emission intensities' of some aluminium lines (c)-(j') when argon (circles) and neon (squares) are employed as the plasma gases. It was found that both the sputtering rate and the discharge current were raised with increasing discharge voltage and that at the same voltage both were greater in the argon-excited plasma than in the neon-excited plasma.As indicated in Fig. 4 ( n the intensity of the A1 I 396.153 nm line emitted by the argon plasma is always larger than that emitted by the neon plasma which roughly corresponds to changes in the sputtering rate. However the observed intensity variations of the A1 I1 lines are unlikely to be explained by differences in the discharge conditions between argon and neon gas. The intensities of the A1 I1 358.660nm [Fig. 4(d)] and A1 I1 624.320 nm [Fig. 4(e)] lines in the argon plasma are almost independent of the applied voltage whereas those in the neon plasma increase with increasing voltage. The intensity of the A1 I1 281.618 nm [Fig.4(c)] line exhibits a similar voltage-dependence for both plasmas but the intensity emitted by the argon plasma is weaker.Fig. 5 shows intensity variations of some A1 I1 lines measured in the vacuum ultraviolet wavelength region. The results obtained with an argon-helium mixed gas are also plotted. The voltage-current characteristics are different from those in Fig. 4 although the same discharge conditions were employed; this is principally because the structure of the glow lamp is altered in order for it to be set in the vacuum spectrometer. Regardless of the discharge voltages the A1 I1 209.492 nm line [Fig. 5(e)J which results from the 3s7f-3s3d transition is not significantly excited either in the argon or the neon plasma indicating that this line is observable only with the argon- helium mixed gas.On the other hand the intensity of the A1 I1 964 Journal of Analytical Atomic Spectrometry October 1996 167.079 nm line [Fig. 5(c)] which is attributed to a resonance transition exhibits a similar voltage dependence for all the plasma gases employed even though the intensity in the neon plasma is generally greater than in the other plasmas. The intensity variation of the A1 I1 162.562 nm line [Fig. 5(b)] is similar to that of the A1 I1 358.660nm or A1 I1 624.320nm line [Fig. 4(d) and (e)]. Fig. 6 shows variations of the discharge current (a) the sputtering rate (b) and the emission intensities of the vacuum A1 I1 lines (c)-(h) as a function of helium partial pressures added in an argon (circles) and a neon (squares) plasma. It was found that both the sputtering rate and the discharge current decrease with increasing amounts of helium in the plasmas.Such a decline in the discharge conditions might lead to a decrease in the emission intensities. Nevertheless the addition of helium leads to complicated changes in the A1 I1 intensities depending on the nature of the lines. The intensities of the A1 I1 207.466nm line [Fig. 6(g)] and the A1 I1 209.492 nm line [Fig. 6(h)] correspond to the amount of helium added in a straightforward manner which is hardly dependent on the type of matrix gas (argon or neon). Such results imply that helium species in the plasmas mainly contribute to ioniz- ation and excitation processes for the corresponding excited states of aluminium ions. The A1 I1 182.857 nm line A1 I1 193.449 nm line etc.exhibit similar behaviour regarding the addition of helium. On the other hand the addition of helium to the neon plasma causes a significant decrease in the intensity of the A1 I1 162.562 nm line [Fig. 6(c)] whereas the intensity gradually increases in the argon plasma. Similarly the intensity of the A1 I1 172.495 nm line [Fig. 6(e)] or the A1 I1 186.231 nm line [Fig. 6 0 1 decreases with the neon-helium plasmas. When the neon-helium plasma includes helium partial pressures of 5.6 x lo2 Pa the intensity of the A1 I1 162.562 nm line is reduced by a factor of 8 and that of the A1 I1 186.231 nm line by a factor of 2 thus differing for each A1 I1 line analysed. The addition of helium to argon plasmas generally enhances the A1 I1 intensities regardless of the reduction of the sputtering rate and many more A1 I1 lines are observable compared with excitation with argon gas alone.It can be concluded from these results that additional excitations caused by helium species principally determine the spectral pattern of A1 I1 emission lines. However in a neon plasma the addition of helium results in intensity decreases of the A1 I1 lines which Vol. 11Table 6 Observed emission lines of indium and their assignment Wavelength/ nm (in air) I1 230.615 1252.137 I 252.299 I 256.015 I 260.175 I1 268.312 1271.027 1271.394 I1 274.974 1275.388 1277.535 1283.691 1285.812 I1 289.018 1293.263 I1 294,104 I 295.700 I 303.936 I1 313.861 I1 314.671 I1 315.835 I 325.609 I 325.856 I1 333.848 I1 379.528 I1 383.468 I1 384.220 I1 384.293 I1 390.194 1410.176 1451.130 I1 463.821 I1 464.468 I1 465.571 I1 465.680 I1 467.355 I1 468.100 I1 468.473 I1 585.318 I1 590.329 I1 591.517 I1 591.867 I1 609.624 11 611.601 Assignment Reduced intensity* Upper (eV) 5p ,P1 (5.3745) 7d 2D5/2 (5.1901) 7d 2D3/2 (5.1869) 6d 2D3,z (4.8413) 8s 2S1/2 (5.0382) 6f ,F4 (17.303) 6d ,D5/2 (4.8475) 6d 2D3/2 (4.8413) 5f 'F (16.611) 5p 4P,/ (4.4659) 5p 4P,i (4.6433) 5p 4P,/2 (4.3366) 5p ID2 (12.103) 6s ' S o (12.029) 5p 4P,/ (4.4659) 5f 'F2 (16.606) 5f ,F3 (16.606) 5f ,F4 (16.608) 7s 2s1 (4.5007) 7s 2s1/ (4.5007) 5d 2D3/2 (4.0780) 5d ,D5/2 (4.0809) 5d 'D3/2 (4.0780) 7p 'P (15.816) 7d ,D (16.709) 4f 'F (15.336) 4f ,F (15.329) 4f ,F2 (15.329) 7d 'D2 (16.787) 6s 'S1/ (3.0218) 6s 'Slj2 (3.0218) 4f ,F2 (15.329) 4f IF3 (15.336) 4f ,F3 (15.329) 4f ,F2 (15.329) 4f 'F (15.336) 4f ,F4 (15.331) 4f 'F3 (15.329) 6d ,Dl (15.465) 6d ,D (15.469) 6d ,Dl (15.465) 6d 'D (15.704) 6d ,D ( 15.476) 6d ,D (15.469) ~- Lower (eV) 5s ' S o (0.oO00) 5P ,PI/ (O.oo00) 5p ,P3/2 (0.2743) 5p 2P3/2 (0.2743) 5p ,P3/2 (0.2743) 5p 2P3/2 (0.2743) 5p ,P3/2 (0.2743) 5d ,D (12.684) 5p2 'D (12.104) 5p 2P1/ (0.0000) 5P ,p1/2 ( 0 .~ 0 ) 5P 2P1/2 ( 0 . ~ 0 ) 5p ,P3/2 (0.2743) 5p 'P (7.8150) 5p 'PI (7.8150) 5d 3D1 (12.656) 5d 3D (12.667) 5d ,D ( 12.684) 5p 'P3/2 (0.2743) 5p 2P3/2 (0.2743) 5P 2p1/2 ( 0 . 0 ~ ) 5p ,P3/2 (0.2743) 5p ,P3/2 (0.2743) 5p2 ID (12.103) 6p ,P2 (13.443) 5p2 'D (12.103) 5p2 'D (12.103) 5p2 ID (12.103) 6p 'P1 (13.610) 5d ,Dl (12.656) 5d ,D2 (12.667) 5d ,D2 (12.667) 5d 3D (12.667) 5d 3D3 (12.684) 5d ,D3 (12.684) 5d ,D (12.684) 6p 3P0 (13.348) 6p ,P1 (13.370) 6p ,P1 (13.370) 6p 'P (13.610) 6p ,P (13.443) 6p ,P2 (13.443) 5P ,p1/* ( O .O ( W 5p 'P3/2 (0.2743) Art 1w <1 <1 8 1 <1 25 3 <1 10 1 6 2 <1 25 2 3 190 <1 < 1 <1 330 160 <1 < 1 <1 <1 <1 <1 1200 1600 <1 <1 <1 <1 <1 <1 <1 <1 <1 <1 < 1 < I <1 Ar-He$. 150 8 1 50 7 1 170 25 5 45 4 30 8 10 110 20 4 780 2 2 5 2100 590 <1 3 25 <1 <3 1 1900 2800 50 6 45 4 <1 120 3 <1 1 <1 2 10 < I Ne§ 220 <1 <1 10 2 <1 20 3 1 10 1 8 4 130 15 85 4 130 <1 <1 1 280 90 15 <1 670 5 60 <1 280 500 330 50 280 30 4 740 15 10 20 3 35 Ne II 8 * Normalized per unit amount of the sputtered sample. t Discharge conditions Ar 4.0 x lo2 Pa; 450 V; 33.4 mA.$ Discharge conditions Ar 4.0 x 10 Pa plus He 5.3 x 10 Pa; 500 V; 28.4 mA. Q Discharge conditions Ne 5.3 x 10 Pa; 800 V; 11.0 mA. 7 Standard for estimating the reduced intensities. 11 Not estimated due to overlapping with gas emission lines. can be excited with the neon gas although some A1 I1 lines are first excited in the presence of helium. This is due to the reduction in the sputtering rate as indicated in Fig. 6(4. In addition the addition of helium may have an influence on the relative populations of neon species through neon-helium collisions and therefore on excitation by neon itself. As can be seen from Fig. 1 singly ionized aluminium h.as several excited states favourable for a charge-transfer collision between a ground state ion of neon and a ground state atom of aluminium.These are the 3s5s the 3s4d the 3s5p and the 3s4f excited states. It should be noted that the sensitive A1 I1 lines emitted by the neon plasma results from transitions from these excited states. The small difference in the total energies may contribute to (quasi-) resonance energy transfer thus resulting in the increased population of the corresponding excited levels. Also this reaction follows Wigner's spin rule." A characteristic of helium-containing plasmas is that A1 I1 lines with relatively high excitation energies can be observed whereas they cannot be emitted by neon and argon plasmas. The corresponding upper energy levels are deduced from the 3p4s the 3p3d or the 3s7f electron configuration. It is expected from Fig.1 that charge-transfer collisions in which helium ions are involved will be beneficial in yielding these states of the aluminium ion. It was also found in the neon-excited plasma that the A1 I1 lines originating from the states having lower energies such as the 3s34 state generally give larger emission intensities. The A1 I1 172.495 nm line is a typical example. Possible channels for their excitations are stepwise de-excitations from the excited states having higher energies. The reactions are expressed as 3s4f- 3s3d4 3s3p 3s5p+3s4s43s3p 3s5s (3s4d)+3s4p+3s4s In the neon plasma energy transfer from neon ions results in the increased population of the 3s4f state etc. The subsequent stepwise de-excitations may induce emissions of the corre- sponding A1 I1 lines. Journal of Analytical Atomic Spectrometry October 1996 Vol.1 1 965Gallium Table 3 summarizes the wavelengths of Ga I and Ga I1 emis- sion lines measured with the vacuum spectrometer and Table 4 the gallium lines measured with the air-path spectrometer. The format of Tables 3 and 4 is the same as that of Tables 1 and 2. Fig. 7 shows a schematic energy diagram of singly ionized gaIIium together with the ionization energies of argon neon and helium. Sensitive Ga I1 lines emitted by the neon plasma result from a 4s5p-4~5~ or 4s4d-4~4~ transition. The Ga I1 633.418nm line (5p 3P,) is a typical line belonging to the transition schemes. As with the excitation mechanism of the A1 I1 lines the excitation to the 4s5p and 4s4d states may be explained mainly by charge-transfer collisions between ground state ions of neon and gallium atoms.The energy surplus for instance 0.84eV in the 5p3P2 level is larger compared with those in the A1 I1 lines. This is theoretically too large for charge transfer to occur effectively; however charge transfer reactions might still be considered to explain the intensity enhancement in the neon plasma. Furthermore there are certain types of Ga I1 lines such as the Ga I1 426.174 nm line which are emitted only by the argon-helium plasma. These Ga I1 lines which are attributed to the 4s4f (or 4s5f)-4s4d transition are also observed in a neon-helium plasma. It is possible that the 4s4f and 4s5f excited states are populated through charge-transfer collisions in which ground state ions of helium are involved and sub- sequent step-wise de-excitations.CONCLUSIONS In Grimm glow discharge spectrometry the relative intensities of the emission lines of the Group IIIB elements presented in this work demonstrate that their excitations are principally determined by the nature of the plasma gas. It was found that some ionic lines are selectively emitted by a neon plasma and that some ionic lines appear only when a plasma gas containing helium is employed. These phenomena can be explained by the assumption that charge-transfer collisions between analyte atoms and plasma gas ions is the major mechanism for determination of the spectrum. Indium Table 5 summarizes the wavelengths of In I and In I1 emission lines measured with the vacuum spectrometer and Table 6 the indium lines measured with the air-path spectrometer. The format of Tables 5 and 6 is the same as that of Tables 1 and 2. Fig. 8 shows a schematic energy diagram of singly ionized indium together with the ionization energies of argon neon and helium. Sensitive In I1 lines appear in the neon-excited spectrum. These In I1 lines for example the In I1 468.100 nm or In I1 591.867 nm line are attributed to the 4s4f-5s5d or the 5s6d-5~6~ transition. As can be seen from Fig. 8 selective excitations to the 5s4f and the 5s6d states can take place through charge-transfer collisions between ground state ions of neon and indium atoms. REFERENCES 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 Grimm W. Naturwissenschaften 1967 54 586. Grimm W. Spectrochim. Acta Part B 1968 23 443. Belle C. L. and Johnson J. D. Appl. Spectrosc. 1973 27 118. Takadoum J. Pivin J. C. Pons-Corbeau J. Berneron R. and Charbonnier J. C. Surf. Interface Anal. 1984 6 175. Teo W. B. and Hirokawa K. Surf. Interface Anal. 1988 11,421. Bengtson A. Eklund A. Lundholm M. and Saric A. J. Anal. At. Spectrom. 1990 5 563. Payling R. Jones D. G. and Gower S. A. Surf. Interface Anal. 1993 20 959. Boumans P. W. J. M. Anal. Chem. 1972 44 1219. Johansson S. Phys. Scr. 1978 18 217. Wagatsuma K. and Hirokawa K. Anal. Chem. 1985 57 2901. Steers E. B. M. and Fielding R. J. J. Anal. At. Spectrom. 1987 2 239. Wagatsuma K. and Hirokawa K. Anal. Chem. 1988,60 702. Wagatsuma K. and Hirokawa K. J. Anal. At. Spectrom. 1989 4 525. Wagatsuma K. and Hirokawa K. Spectrochim. Acta Part B 1991 46 269. Steers E. B. M. and Leis F. Spectrochim. Acta Part B 1991 46 527. Steers E. B. M. and Thorne A. P. J. Anal. At. Spectrom. 1993 8 309. Duffendack 0. S. and Black J. G. Phys. Rev. 1929 34 35. Rozsa K. Janossy M. Csillag L. and Bergou J. Phys. Lett. A 1977 63 23 1. Wagatsuma K. and Hirokawa K. Surf. Interface Anal. 1984 6 167. Moore C. E. Atomic Energy Levels NBS Stand. Circular 467 US Government Printing Office Washington DC 1958 vols. 1-111. von Engel A. Ionized Gases Clarendon Press Oxford 1965. Paper 6J039.588 Received June 5 1996 966 Journal of Analytical Atomic Spectrometry October 1996 Vol. 11

ISSN:0267-9477    

DOI:10.1039/JA9961100957

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Comparison of the Analytical Performance of Flame Atomic Magneto- optic Rotation Spectrometry in the Faraday Configuration With That of Flame Atomic Absorption Spectrometry Journal of Analytical Atomic Spectrometry AHMET T. INCE,* JOHN B. DAWSON AND RICHARD D. SNOOK Department of Instrumentation and Analytical Science (DIAS) University of Manchester Institute of Science & Technology (UMIST) P. 0. Box 88 Manchester U K M60 1 QD Atomic magneto-optic rotation spectrometry (AMORS) is an analytical technique for the determination of elements and has much in common with AAS. This paper compares the analytical performance of the two techniques for several elements using flame atomization. Using the same instrumentation for all measurements Mg and Ag were studied in depth by AAS and AMORS.The best AMORS results were obtained when the angle between the planes of polarization of the polarizing and analysing prisms was +45". Under these conditions the detection limits and linear ranges were slightly better than those obtained by AAS. Keywords Atomic magneto-optic rotation; atomic absorption; frame atomization Faraday configuration; magnesium and silver determination In 1924 Hanle' observed the rotation of the plane of polarized light when it passed through atoms located within a magnetic field. The theory of magneto-optic rotation was introduced by Corney et a1.' on a semi-classical basis. The determination of elements by magneto-optic rotation was first proposed by Church and Hadeish? in 1974; they determined mercury in air as did Stephens4 in 1978.Since that time magneto-optic rotation has been shown to be potentially useful as an analyt- ical technique for trace element analyses.'-12 AAS is a well established technique and has been successfully utilized since the 1 9 4 0 ~ ' ~ and commercialized since 1955.14 A general prob- lem in AAS is background absorption by molecular species and the sample matrix. Atomic magneto-optic rotation spec- trometry (AMORS) is a more recent technique for trace element analysis and has not yet been commercialized. However by the very nature of its physical measurement it is a technique which discriminates against background absorp- tion. AMORS depends on the rotation of the plane of polariz- ation of resonance radiation by birefringence and dichroism generated in a magnetized atomic vapour at the wavelength of the resonance line of the analyte element. The measurement of the rotation of the incident plane polarized light also leads to AMORS having a multi-element spectroscopic capability when a continuum light s o ~ r c e ~ ~ * ' ~ is employed.This arises because only rotated light at the particular wavelength under investigation is subsequently detected. In its simplest form the AMOR spectrometer consists of a magnetized atomic vapour situated between a pair of crossed polarizers a light source and a detector. The magnetic field can be applied either in the transverse (Voigt) or longitudinal (Faraday) configuration. If the atomic vapour is not magnetized ~~~ * Present address National Metrology Institute (UME) Scientific and Technical Research Council of Turkey P.0. Box 21 41470 Gebze-Kocaeli Turkey. or is absent light from the source is blocked by the crossed polarizer arrangement so that no light reaches the detector. This configuration is illustrated in Fig. 1 where only light leaking from the polarizers can be detected. This crossed polarizer system produces signals with a quadratic dependence on concentration; hence detection limits are poor and means of linearizing the calibration graph have been sought. THEORY Generalized theoretical treatments have been presented by Dawson and c~-workers~'-~~ and Kankare and StephenP for crossed and partially crossed polarizers in the Voigt and Faraday configurations. The conclusion of these treatments as applicable to a Faraday-configured instrument will be summa- rized here as a basis for examination of the experimental results. Polarizer-Analyser Axes Orthogonal ('Crossed') The intensity of the transmitted radiation I is given by the expression ctL2 K2 I,= ~ [( 1 - p)2 + 4psin28] 8 where P=aR/aL (dichroic component); ctL and aR are the attenuation coefficients of the left- and right-handed circularly polarized components of the electric vector (amplitude 6) of the incident linearly polarized radiation; intensity = I = 1/02 28 is the phase difference between the components induced by birefringence of the magnetized atomic vapour and is proportional to the number of atoms N in the optical path.When p-1 and 8 is small eqn. (1) reduces to i.e. IT cc N 2 (3) For low concentrations of analyte eqn.(3) predicts a quadratic response to changes in analyte atom numbers. Such a response curve may be linearized by taking the square root of the transmitted intensity as the analyte-dependent function. It should be noted that the attenuation coefficients ctL and ctR are also a function of N . When N is small aL and d R approach unity. Polarizer-Analyser Axes Offset (Uncrossed) by a Small Angle The general expression for the intensity of radiation transmitted by a system consisting of a magnetized atomic vapour between Journal of Analytical Atomic Spectrometry October 1996 Vol. 1 1 (967-972) 967Magnet solenoids Monochromator Fig. 1 Block diagram showing the AMOR spectrometer arrangement. HCL hollow cathode lamp; PSU power supply unit; L1 L2 L3 L4 lenses F flame; PMT photomultiplier tube polarizers uncrossed by a small angle A is (4) CtL2 Vo2 I T = - [( 1 - p)2 + 48 sin2(A + 8)] 8 - [(I -p)2+4p(sin2 8 cos2 A+cos2 o sin2 A (5) 8 1 2 1 + - sin 28 sin 2A) When 8=0 the transmitted intensity is a L 2 L IT = - [( 1 -p)2 + 4psin2 A] 8 and represents the background against which the analytical signal is measured. The analytical signal is however linearized and enhanced when 8 is close to 0 and the term 1/2 (sin 28 sin 2A) of eqn.( 5 ) becomes dominant. In these circum- stances if the offset angle is set negative i.e. against the direction of optical rotation the initial gradient of the cali- bration graph will be negative. The sensitivity is increased and background signal cancelled if the difference signal between the positive and negative offset angles (+A) is used as the analytical signal I D .From eqn. ( 5 ) ID = IT( + A) -IT( - A) Thus (sin 2A sin 28) M L ~ R I ~ I D = - 2 When N is small and aL and aR x 1 IDKN (7) Polarizer-Analyser Axes at 45" This condition is a special case of eqn. ( 5 ) with A = +45" and can be used to calculate the optical rotation by the magnetized atomic vapour from the measured transmitted intensities 1.45" and 1-45" of the two orientations. Substitution of sin45" = 1/$; cos45" = 1/$; sin( - 45") = - 1/$ and cos(- 45") = 1/$ in eqn. (5) gives I+450= 8 ( 1 + p 2 + 2p sin 28) (9) and (1 +D2-2p sin 28) - 8 aR210 Combining eqns. (9) and (10) leads to (11) e= -sin-' 1 (a) 1+p2 ( + 4 5 " 4 - 4 5 " ) K N 2 +45" + I - 45" Eqn. (1 1) predicts a linear relationship between the number of atoms in the optical path and a simple function of the measured transmitted intensities modified by a dichroic term ( 1 + p2)/2p.Linear Range In order to facilitate comparison between the analytical per- formance of the various modes of operation of the apparatus the upper limit of linearity ( U ) will be defined as the concen- tration at which the gradient of the calibration graph is half that at the origin or in the event of a concave curve twice that concentration. The relative linear range (R) will be expressed as the ratio of the upper limit of linearity down to five times the detection limit D i.e. (R = U/5D). The detection limit is calculated as the analyte concentration corresponding to twice the standard deviation of a determination at near zero analyte concentration.The relative linear range is an indication of the analyte concentration ratio over which analy- ses can be effected with an RSD of 10% or better. EXPERIMENTAL Apparatus Atomic magneto-optic rotation generated by several analyte elements was studied using atomization in an air-C2H2 flame with the magnetic field in the Faraday configuration. The apparatus used is shown schematically in Fig. 1 and its specifi- cation summarized in Table 1. A particular feature of the apparatus was the use of a Rochon prism as the analysing polarizer. This prism was chosen because it generates two orthogonally polarized beams of radiation with a small angle of separation from an incident radiation beam. The prism was mounted on the optical axis such that the two beams passed through the spectrometer entrance slit one above the other.On emerging from the spectrometer the beams were directed to two separate photomultipliers. By means of this arrange- ment in addition to single channel measurement of IT for 968 Journal of Analytical Atomic Spectrometry October 1996 VoE. 1 1Table 1 Specification of instrumentation Monochromator RLD*/A O mm-' Grating spacing/lines mm- ' Entrance slit-width/mm Exit slit-width/mm Polarizer Anal yser Electromagnet maximum strength/T Atomizer Beam area on flame/cm2 Pathlength/mm HCL current/mA PMT voltage/V Nebulizer uptake/ml min Coil current/A Rank Hilger monospek. 8 1200 0.5 0.5 Glan-air prism Rochon prism 0.7 30 Air-acetylene flame 0.4 20 8 lo00 8 * RLD is the reciprocal linear dispersion.various configurations of the polarizers simultaneous channel measurement of I + 450 and I - 450 could be made. The polarizing component was a Glan-air prism which was attached to a rotatable mount so that the angle of polarization of the incident radiation relative to the polarization axis of the fixed analyser could be varied. Four principal orientations of the polarizer were used crossed polarizers A =O; small angle offset A= & 6 O ; large angle offset A = +45" and for atomic absorption measurements A = 90". Test solutions were pre- pared from analytical-reagent grade chemicals and doubly distilled water. Preliminary Experiments Several experiments were conducted with a view to selecting the two elements with the strongest analytical signal for further investigation.The crossed polarizer configuration was used and the response of the AMOR spectrometer to Ag Ca Cu Mg Ni and Pb was determined. The atomic transitions and wavelengths of the analytical lines are given in Table 2; the splitting components were found from the All of these analytes gave an AMORS signal but those of Ca and Ni were poor and so were not studied further. For Ca this lack of intensity was attributed to non-optimum flame con- ditions for the element. The dependence of the signal on the magnetic field strength was examined for Ag Cu Mg and Pb. The results are shown in Fig. 2. As the magnetic field strength increases the rotating power of the atomic vapour is initially enhanced as the components of the Zeeman-split analyte spectral line move apart and their optical interaction with the emission line from the light source increases.However at high magnetic fields the components may move beyond the profile of the emission line and thus the AMORS signal will begin to decrease. All the response curves in Fig. 2 show a maximum. Multiple maxima arise from the fine structure and splitting patterns of the spectral line. As Ag and Mg appeared to be the most sensitive elements they were chosen for further study using magnetic field strengths corresponding to the maximum Table 2 Atomic transitions for analytical lines used in this work & 6 component splitting Element Term symbols l/nm pm T-' (refs. 20-22) Ag+ :sl/2 -'2p3/2 328.1 k 5.0 Ca so JPI 422.6 k 8.2 cu* 2s1,2 d2P3,2 324.7 +_ 5.0 Mg 'so +'PI 285.2 & 3.8 Ni* 'F4 +3Gj 232.0 + 3.0 Pb 3P0 +3P1 283.3 5.6 * Zeeman splitting component has more than one sigma component on each side of the analytical wavelength.1.4 1.2 ' 1 E 0 T3 3 C .- - 8 0.0 - a K 0 a 0.6 U 0 .- 0.4 0.2 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Applied magnetic field strengthn Fig. 2 Optimization of magnetic field strength in crossed polarizer configuration for A Mg (40 ppm); B Ag (80 pprn); C Cu (100 ppm); and D Pb (1000 ppm) responses of the two elements (for Ag 0.27T and for Mg 0.55 T). RESULTS AND DISCUSSION AMORS for Ag and Mg With Crossed Polarizers In the conventional AMOR configuration i.e. with crossed polarizers the calibration graphs obtained for Mg and Ag and are shown in Figs. 3A and 4A respectively. As predicted by 3 T I 2.5 0 10 20 30 40 50 60 Concentration (ppm) Fig.3 AMORS calibration graphs for Mg.A Crossed polarizer; B square root of curve A amplitude; C small positive offset polarizer (+6"); D small negative offset polarizer (-6"); and E difference between curves C and D Journal of Analytical Atomic Spectrometry October 1996 Vol. 11 9690 10 20 30 40 50 oa Concentration (ppm) Fig. 4 AMORS calibration graphs for Ag. A Crossed polarizer; B square root of curve A amplitude; C small positive offset polarizer (+ 6"); D small negative offset polarizer (- 6"); and E difference between curves C and D various workers"*'* [eqn. (3)] the calibration graphs are non- linear and appear to be quadratic in nature; this can be demonstrated by taking the square root of the transmitted intensity and plotting this parameter against concentration to obtain a more linear calibration graph (Figs.3B and 4B). AMORS for Ag and Mg With Small Angle Offset Polarizers When the polarizers are uncrossed radiation is transmitted by the system and all analytical measurements are made against this background signal. With a small positive offset (A = + 6") increasing slightly concave response curves were observed (Figs. 3C and 4C). When the offset was negative (A= -6") the transmitted intensity initially decreased with increasing analyte concentration to produce minima in the response curve (Fig. 3D and 4D) at 45 pg ml-' Ag and 17 pg ml-' Mg. These minima arise because the plane of the radiation incident on the analyser is rotated by the AMOR effect through an orientation which is orthogonal to the transmission axis of the analyser.The initial gradients of the curves for both positive and negative offsets are almost linear. When the difference signal between positive and negative offsets is plotted (Figs. 3E and 4E) the effect of the background signal is eliminated and the curve begins at the origin. AMORS for Ag and Mg With 45" Offset Polarizers Another method of obtaining a linear calibration graph is to use the linear relationship between the angle of rotation of the linear plane polarized light and sample concentration c pre- dicted by eqn. (1 1). When dichroism is small p~ 1 and the dichroic term I( 1 + P2))/2p3 z 1. Hence d= - 1 s i n - l ( ; + 4 5 - ~ - 4 5 0 ) ~ N 2 + 45" + 1-450 In order to measure 8 the axes of the polarizer and analyser prisms were set at 45" relative to each other.The latter prism (Rochon) then generated the two signals and I-45.= from which 8 was calculated using eqn. (12). Calibration graphs obtained at several magnetic field strengths are shown in Figs. 5 and 6 for Mg and Ag respectively. The graphs are initially linear but then begin to curve and 'roll-over'. Theoretically [eqn. (ll)] the function reaches its maximum value corre- sponding to an optical rotation of 45" i.e. the axes of the polarized radiation and the analysing prism are parallel in one detector channel and orthogonal in the other. In practice the combined effect of the dichroic term [( 1 +P2)/2P] and stray light becomes significant and leads to maxima of less than 45".Although the general form of the response curves is as predicted by eqn. ( l l ) the change in shape with change in magnetic field strength demonstrates the crucial role of the interaction of the plane of the emission and absorption line profiles in determining the precise shape of the curves. As the magnetic field changes the dichroic effect and the proportion of non- interactive radiation (stray light) change as the overlap between the emission and absorption line profiles changes. Thus eqn. (12) should be modified to eqn. (13) to include stray light terms I An example of this situation is seen in Fig. 6C where at high concentrations of Ag there is little change in the apparent signal because the peak has been broadened by the complex splitting peaks of the lines and much of the optical rotatable light has been absorbed and stray light has become significant.This effect may also be attributed to the second 'peak' seen in Fig. 2. AAS Studies With the AMOR Spectrometer Atomic absorption curves were obtained using the AMOR apparatus with the polarizers in the 45" setting. The absorbance (A) of the atomic vapour was calculated as I + 45" + I - 45")o ( I + 45" + I - 45% A =log where the subscript '0' indicates measurements when a blank solution is aspirated and 'c' when analyte is present. Calibration graphs without and with an applied magnetic field are pre- sented in Figs. 7 and 8 respectively. The curves are initially linear but as stray light becomes a greater proportion of the measured intensity curvature begins to arise.Fig. 8 includes two AMOR calibration graphs to facilitate visual comparison of the linearities of AMORS and AAS. Until the 'roll-over' in the AMOR curves occurs the relative curvatures of all the curves are similar. The reduction in AAS sensitivity when the magnetic field is applied is relatively small (25% for Mg and 33% for Ag) and demonstrates that the optimum field strength for AMOR measurements is much less than that required to generate the maximum Zeeman AAS (ZAAS) signal. This difference is due to the fact that in AMORS the signal is generated by interaction of the overlapping emission and absorption line profiles whereas in ZAAS the maximum signal is generated by separating the emission and absorption line profiles as completely as possible in order that the intensity (field 0n):intensity (field off) ratio is as large as possible.970 Journal of Analytical Atomic Spectrometry October 1996 Vol. 1 10 20 40 60 80 100 120 140 (::,oncentration (ppm) Fig. 5 Calibration graphs for Mg with 45" offset polarizer cc' ?figuration. Magnetic field strengths used A 0.225; B 0.35; C 0.5; and D 0.65 T 35 1 C T 0 40 80 120 160 20C 240 280 320 360 400 440 480 520 560 600 Concentration (ppm) Fig. 6 Calibration graphs for Ag with 45" offset polarizer configuration. Magnetic field strengths used A 0.225; B 0.275; and C 0.37 T 0.35 0.30 0.25 a 2 0.20 2 0.15 a 0.10 0.05 0 10 20 30 40 50 t Concentration (ppm) Detection Limits Upper Limit of Linearity and Linear Range A summary of the experimental results is presented in Table 3.The best results for the AMORS system were obtained when the polarizer axes were offset by 45" and the worst when the axes were orthogonal. The latter is a consequence of the poor sensitivity of a quadratic response system at low concentrations and failure of the crossed polarizers to exclude non-interactive radiation completely. Although the upper limit of linearity was comparable to that with other configurations the relative linear range was poor. When the polarizer axes were offset from orthogonal by 6" the detection limits and relative linear range were improved 10-fold. However the greatest improve- ment was obtained when offset angles of 45" were employed. Under these conditions detection limits were further reduced and the upper limits of linearity extended.These improvements in performance are attributable to the increase in light trans- mitted by the system in this configuration which in turn leads to greater accuracy in measurements. As AAS is a well established technique for the determination of elements and in our situation can be carried out with the Fig. 7 Absorption calibration graph using AMOR apparai US. No magnetic field. A Mg; and B Ag same instrumentation as AMORS it provides an effective means for evaluating the merits of AMORS. The sensitivities 971 Journal of Analytical Atomic Spectrometry October 1996 Vol. 11linear ranges of the AAS measurements were worse (z 50%) than the best of those obtained by AMORS using 45" offset polarizer axes but were better than those obtained with all the other AMORS configurations.When AAS measurements were made with a magnetic field applied to the atomic vapour there was a reduction in sensitivity of approximately 50% and a doubling of the detection limits. The studies of Ag and Mg by both AMORS and AAS did not reveal any differences in their analytical behaviour which might lead to the preference of one technique over the other. CONCLUSION In the conventional crossed polarizer configuration AMORS 0" 50 100 150 2b 250 300 350 - o Concentration (ppm) Fig. 8 Comparison of atomic absorption calibration graphs with magnetic field and AMOR calibration graphs for Mg and Ag obtained Magnetic field strengths for Mg and Ag are 0.55 and 0.27 T respectively has an inherent quadratic dependence on analyte concen- with the 45" POlariZer Configuration.* degrees; + 5 absorbance. tration. This characteristic can be linearized by offsetting the polarizer prism from orthogonality by a small angle (x6"). Table 3 Detection limits and linear ranges in AMORS and AAS Configuration AMORS Field strength/T Crossed polarizers (linearized by taking square root of IT) Detection limit (ppm) Upper limit of linearity (ppm) Relative linear range (ppm) Offset polarizers A=++" Detection limit (ppm) Upper limit of linearity (ppm) Relative linear range (ppm) Detection limit (ppm) Upper limit of linearity (ppm) Relative linear range (ppm) A = -6" Difference [+6"-(-6")] Detection limit (ppm) Upper limit of linearity (ppm) Relative linear range (ppm) 45" Offset polarizers Detection limit (ppm) Upper limit of linearity (ppm) Relative linear range (ppm) AAS No field on Sensitivity (ppm per 0.0044 A ) Detection limit (ppm) Upper limit of linearity (ppm) Relative linear range (ppm) With field Sensitivity (ppm per 0.0044 A ) Detection limit (ppm) Upper limit of linearity (ppm) Relative linear range (ppm) Analyte 0.27 10 60 1.2 0.6 > 40 > 13.3 0.6 7.3 22 0.4 40 20 0.4 220 110 1.2 0.6 > 50 > 20 1.8 1.1 200 36 0.55 3 55 3.7 0.4 > 50 > 25 0.4 6.5 13 0.3 15.3 23 0.1 50 100 0.4 0.15 45 60 0.5 0.4 > 120 > 60 This modification improved the detection limit and relative linear range by an order of magnitude.The use of +45" offset angles produced the lowest detection limits with the greatest linear ranges. From our studies it appears that this arrangement is the most suitable for analytical applications of AMORS.The detection limits and relative linear ranges obtained were comparable to or better than those obtained by AAS using the same apparatus. AMORS however has the potential to discriminate against background absorption. The poor detec- tion limits resulting from the experimental configuration could be improved by increasing the optical path through the flame and more efficient utilization of the atomic vapour. Moreover as the magnetic field strength required to generate an AMORS signal is less than that for ZAAS an AMORS system can offer greater flexibility in the design of magnet/atomizer systems using either flames or furnaces. REFERENCES 1 Hanle W. Z . Phys. 1924 30 93. 2 Corney A. Kibble B. P.and Series G. N. Proc. R. SOC. London Ser. A 1966 293 70. 3 Church D. A. and Hadeishi T. Appl. Phys. Lett. 1974 24 185. 4 Stephens R. Anal. Chim. Acta 1978 98 291. 5 Davis L. A. Krupa R. J. and Winefordner J. D. Spectrochim. Acta Part B 1986 41 1167. 6 Hermann G. CRC Crit. Rev. Anal. Chem. 1988 19 323. 7 Davis L. A. and Winefordner J. D. Anal. Chem. 1987 59 309. 8 Debus H. Hanle W. Scharman A. and Wirz P. Spectrochim. Acta Part B 1981 36 1015. 9 Jolly G. S. and Stephens R. Spectrochim. Acta Part B 1984 39 335. 10 Monnig C. A. and Hieftje G. M. J. Anal. At. Spectrom. 1988 3 679. 11 Kitagawa K. Shigeyasu T. and Takeuchi T. Analyst 1978 103 1021. 12 Yamamoto M. and Murayama S. J. Opt. SOC. Am. 1979,69,781. 13 Woodson T. T. Rev. Sci. Instrum. 1939 10 308. 14 Walsh A. Spectrochim. Acta 1955 7 108. 15 Yamamoto M. Murayama S. Ho M. and Yasuda M. Spectrochim. Acta Part B 1980 35 43. 16 Kankare J. and Stephens R. Spectrochim. Acta Part B 1983 38 1301. 17 Kersey A. D. Dawson J. B. and Ellis D. J. Spectrochim. Acta Part B 1980 35 865. 18 Kersey A. D. and Dawson J. B. Anal. Proc. 1981 18 187. 19 Dawson J. B. King P. R. Duffield R. J. and Ellis D. J. J. Anal. and detection limits of AAS measurements made in this work are up to times than those achieved using tional equipment under optimum conditions. This reduced sensitivity is attributable to a short optical Path through the flame ( z 20 mm) inefficient flame geometry and non-optimum flame gas composition. However the lack of sensitivity does not invalidate compari- sons between AAS and AMORS as this limitation will affect both techniques similarly. The detection limits and relative At. Spectrom. 1989 4 245. 20 Wagenaar H. C. PhD Thesis Delft University 1976. 21 Mavrodineanu R. and Boiteaux H. Flame Spectroscopy Wiley 1965 New York p. 31. 22 Preli F. R. Jr. Dougherty J. P. and Michel R. G. Spectrochim. Acta Part B 1988 43 501. Paper 6/01 71 2F Received March 1 I 1996 Accepted June 7 1996 972 Journal of Analytical Atomic Spectrometry October 1996 Vol. 11

ISSN:0267-9477    

DOI:10.1039/JA9961100967

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Short Multifactorial Plan for the Determination of Trace Metals in Complex Matrices by Flame Atomic Abso rpt i o n Spectrometry* SVJETLANA LUTEROTTI Department of Analytical Chemistry Faculty of Pharmacy and Biochemistry University of Zagreb 10000 Zagreb Republic of Croatia Youden and Steiner's short multifactorial experimental plan was successfully applied to the determination of Gun Zn" and Mnn by FAAS in synthetic samples simulating whole liver homogenates of 1 5 and 1 29 mass ratios. An alternative method of treating the data is described which makes the modified short multifactorial concept a useful means for the rapid and reliable recognition of those working parameters that are most likely to jeopardize the accuracy of the analytical results. Keywords Copper(zz) zinc(zz) and manganese(zz) determination whole liver homogenate; flame atomic absorption spectrometry statistical treatment; ruggedness testing Trace metals in human tissues are of interest in medical investi- gations and their levels can be related to certain biochemical abnormalities or nutritional deficiencies. Generally when trace metals in biological materials are concerned of the variety of techniques proposed FAAS is still widely used because it is precise simple sensitive and relatively free from interferences.' From the standpoint of simplicity and accuracy the determi- nation of trace metal ions in tissue homogenates seems to be very However the complexity of whole tissue homogenates which originates not only from the const I tuents of the tissue material at the actual dilution level but also from the presence of agents used for the preparation of the final sample (e.g.medium for homogenization detergent etc.) may seriously affect the accuracy of the method. Incorrect analytical results may arise from interferences that are not removed or accommodated by chemical modification,6 standard additions matrix-matching or correction of non-specific background absorption. If disturbances produced by the matrix components are not significant or are mutually compensated calibration may be based on simple or even aqueous standard solutions. A short multifactorial plan which was first introduced by Plackett and Burman,' was applied by Youden and Steiner' and Fisherg for ruggedness testing and recently by Doerffel'' for evaluating the selectivity of analytical procedures.Our concern has been focused on demonstrating the usefulness of this concept for rapid and simple recognition of those working parameters that have the greatest effect on the accuracy of the determination of Cu'' Zn" and Mn" by FAAS in complex matrices matching whole liver homogenates. EXPERIMENTAL Materials All the chemicals used except for human albumin lyophilysate were of analytical-reagent grade. Water that had been distilled after first being de-ionized in an all-glass apparatus was used throughout. * Presented at the XXIX Colloquium Spectroscopicum Internationale (CSI) Leipzig Germany August 27-September 1,1995. TRIS-acetate buflers (pH 8.6) The buffer used as a simple Journal of Analytical Atomic Spectrometry I I Ante KovaEiLa 1 medium at a 1 5 mass ratio (hereafter referred to as TAB) was prepared as a mixture of 0.25 mol 1-' sucrose and 0.05 moll-' tris( hydroxymethy1)me- thylamine in water.The pH was adjusted to 8.6 by dropwise addition of glacial acetic acid. The mixture was stored at 4°C overnight and the pH regulated if necessary. A 5-fold dilution of TAB with water was performed to obtain a medium for solutions of 1:29 mass ratio (hereafter referred to as TAB/5). Hydrochloric acid Solutions of 1.0 and 0.2 mol 1-' HC1 were used as simple media for 1 5 and 1 29 mass ratios respectively. Human albumin An opalescent suspension of human albumin in water (mass fraction 20%) was prepared from lyophilized human albumin (ImunoloSki Zavod Zagreb Croatia) and stored at 4 "C.In order to ascertain the influence of NaCl on human albumin the content of sodium in human albumin was determined as described previously." A concentration of 1.54 mg g-' of sodium was found in human albumin lyophilysate. Analyte standard solutions A stock solution of Cu" (514.6mg 1-') was prepared by dissolving CuC1,*2H20 in 0.1 mol 1-' HCl. It was stan- dardized complexometrically. A 10-fold dilution (51.5 mg 1-' Cu 0.01 moll-' HC1) was used for preparing synthetic samples (1 5 and 1 29 mass ratios) of the following concentrations 2.1 1.5 1.0 0.5 0.2 and 0.1 mg 1-' Cu". A Zn" stock solution (500.0 mg 1-') was prepared by dissolving Zn metal in concentrated HCl. The solution was 0.1 moll-' with respect to HCl. From an intermediate standard solution (50.0 mg 1-' Zn 0.01 mol 1-' HCl) the following working standard solutions were prepared 1.5 1.0,0.7,0.4,0.2 and 0.1 mg I-' Zn".By using a Titrisol solution (Merck Darmstadt Germany) a stock solution of Mn" was prepared (1000mg 1-' Mn 0.2 moll-' HC1). Dilution of a solution containing 50.0 mg 1-' Mn and 0.01 mol 1-' HC1 served for the preparation of working standard solutions containing 2.0 1.5 1.0,0.5,0.2 and 0.1 mg I-' Mn". Metal ion working standard solutions were prepared in diluents that simulated the matrices at dilution level@) required for determination of the respective analyte. The dilution levels applied corresponded to a 1:5 and/or 1:29 mass ratio of whole rat/bovine liver' in water TRIS-acetate buffer and HCl. The pH values of the Cull Zn" and Mn" solutions in water were 4.6-6.1 3.7-5.0 and 3.6-5.6 respectively.The albumin- containing solutions were slightly opalescent (pH 3.9-4.9) with slow flocculation taking place. An increase in the opalescence was observed with decreasing metal ion concentration i.e. by Journal of Analytical Atomic Spectrometry October 1996 Vol. 11 (973-978) 973Table 1 Experimental scheme for seven variables (ref. 8)* System 1 2 3 4 5 c91 TAB[HCl] - - H2O H2O H2O H2O T T - - - HA - HA - - NaCl NaCl NaCl - KH2P04 - KH2P04 KH2P0 - DBC DBC _. - ~ ~~~~ * Variable parameters (mass ratios of simulants indicated in parentheses) (1) basic medium-A H 2 0 versus a TRIS-acetate buffer. pH 8.6 TAB (1 5) and TAB15 (1 29) or HC1 1.0 mol 1-' (1 5) and 0.2 mol 1 -' ( 1 29); (2) Triton X-100 (T)-B none versus b volume fractions of 0.5% (1:5) and 0.1% (1:29); (3) human albumin ( H A F C none versus c mass fractions of 1.1% (0.78 mmol 1-' Na') (1:5) and 0.22% (0.16mmol 1-' Na') (1:29); (4) NaCl-D none uersus d 6.34mmol I-' (1:5) and 1.27mmol 1-' (1:29); ( 5 ) KH2P04-E none versus e 12.55 mmol 1-' (1 5) and 2.51 mmol 1-' (1 29); (6) deuterium-arc background correction (DBCFF none uersus f DBC; and (7) sample ageing- G to = 0 h versus g t24 = 24 h.approaching the isoelectric point at pH 4.9. For Zn" and Mn" opalescence was reduced in the presence of Triton X-100. The pH values of TRIS-acetate buffered systems of Cu" were 8.0-8.3 and of Zn" and Mn" 8.2-8.4. Solutions containing human albumin were clear at a mass ratio of 1 29 and slightly opalescent at a mass ratio of 1 5.Triton X-100 stabilized the systems with Zn" and Mn". The pH values recorded in 0.2 and 1.0 mol 1-' HC1 systems were 0.7-0.8 and 0.1-0.2 respectively. Denaturation of albumin was evident from a slight opalescence in the 1 29 mass ratio systems or from immediate high turbidity with extensive sedimentation in the 1 5 mass ratio systems. Triton X-100 reduced these effects. 'Field' blanks were prepared so as to contain all the appro- priate matrix components at the respective dilution levels but with analyte omitted. The matrix composition and experimental scheme are given in Table 1. Apparatus FAAS measurements were performed using a Perkin-Elmer (Norwalk CT USA) Model 305B flame atomic absorption spectrometer equipped with a deuterium-arc power supply and a Perkin-Elmer Model 56 recorder.For pH control a pHM85 Precision pH-meter (Radiometer Copenhagen Denmark) with a combined pH electrode (Radiometer) was used. Analytical Procedures Short multifactorial experimental concept An experimental scheme for seven variables after Youden and Steiner* was applied to systems with Cu" Zn" and Mn" and is presented in Table 1. By deliberately varying seven factors simultaneously the necessity of carrying out a large number of experiments in order to vary each factor independently is avoided and only eight experiments are required. Each param- eter was tested at two alternative levels (r and y namely A/a through G/g). In our case two comparisons of simple media (water-TRIS-acetate buffer and water-HC1) resulted in a total of 12 experiments which were designed to determine how the calibration sensitivity is affected by the following (1) The use of media of markedly different acidities for liver homogeniz- ation.For this purpose water (level A systems 1-4) as a medium for homogenization was replaced with either TRIS- acetate buffer (level a TAB or TAB/$ systems 5-8) or HC1 (level a 1.0 or 0.2 rnol 1-' HCl systems [9]-[12]). (2) The presence of the non-ionic surfactant Triton X-100 (T) (level b). (3) The presence of human albumin (HA) (level c). (4) The presence of NaCl (level d ) . (5) The presence of KH2P04 (level e). (6) The use of deuterium-arc background correction (DBC) (level f). (7) Ageing of synthetic samples over a period of 24 h (t24) (level g ) in comparison with fresh samples (to) (level G).Therefore synthetic samples were prepared using three types of simple media i.e. water TRIS-acetate buffer and HC1 for each analyte. Weakly acidic i.e. aqueous systems were com- pared with either weakly alkaline or highly acidic systems. The remaining six variables were tested within both comparisons of simple media; uiz. water-TRIS-acetate buffer and water-HC1. Matrix-matched standard solutions simulated whole liver homogenates of mass ratios 1 5 and 1 29. Accordingly con- centrations of matrix components namely TRIS-acetate buffer HC1 Triton X-100 human albumin NaCl and KH2P04 were adjusted to the dilution level required while the analyte level was fixed at its linear concentration range. The above dilution levels correspond to both rat and bovine whole liver.The respective concentrations of major inorganic ions in both rat and bovine whole liver are comparable. FA AS measurements Experiments were performed according to the experimental scheme of the short multifactorial concept given in Table 1. Instrumental parameters are given in Table 2. All measure- ments were performed under standard pressure and flow rate conditions for both air (2.1 x lo5 Pa 22.5 1 min-') and acety- lene (5.5 x lo4 Pa 3.9 1 min-'). A single-slot burner head 10.2cm in length was used. DBC was used according to the experimental scheme in Table 1. In order to avoid the deterior- ation of measurements with DBC no scale expansion was applied. The solids that were formed as a consequence of human albumin denaturation in 1.0 mol I-' HCl media were allowed to settle and the supernatant liquid was aspirated into the nebulizer system of the instrument.Table 2 Instrumental conditions for FAAS measurements Element c u Zn Mn Lamp type HCL* EDLt HCL* Lamp current/mA 15 Lamp power/W 6 Wavelengthlnm 324.8 213.9 279.5 Scale expansion None None None 20 - - - Slit setting (bandpasslnm) 4 (1.0) 4 (1.0) 4 (1.0 * Hollow cathode lamp. t Electrodeless discharge lamp. 974 Journal of Analytical Atomic Spectrometry October 1996 Vol. 11Evaluation of Analytical Results The analytical signals were processed in the peak-height mode with an integration time of 2 s. Analytical signals i.e. absorp- tion values (YO) were converted into analytical information values i.e. absorbance. The mean of five readings was taken.These were corrected for the blank values. Single calibration lines (Table 3) and weighted averages of calibration lines at levels r and 7 (collective lines Tables 4-7) were obtained by the method of least squares using the computer package LSQANAL.12 Depending on whether the intercepts of the calibration lines were found to be statistically different from zero at the 99% probability level or not. ideal or non-ideal calibration lines were established. An analysis of parallelism of the collective calibration lines13 was carried out using the computer programs TWOLINE and TWOLINE1.l' RESULTS AND DISCUSSION The calibration sensitivity (CS) is defined as the slope of the analytical calibration line.14 It is a critical performance charac- teristic of an analytical method and a highly reliable indicator of the bias of a method.It was therefore chosen as a basis of obtaining an insight into the effects of seven variable param- eters on the accuracy of analytical results. One of the difficulties in interference studies is the possibility of complex interactions between the major constituents pro- ducing effects on the apparent concentration of the trace analyte that are not predictable from results obtained with each major constituent separately." To date the computational approaches that have been proposed to detect such interactions have been fairly c ~ m p l e x . ' ~ - ' ~ This makes the application of a simple short multifactorial plan even more attractive. The results for the 12 analytical systems for each analyte obtained with the short multifactorial concept are presented in Table 3.Tables 4-7 display two methods of treating the data one proposed by Youden and Steiner' including calcu- lation of the differences DCSr- CS for each variable parameter based on the results given in Table 3 and our approach. According to Youden and Steiner,' the absolute differences DCSr-CS should be calculated and related to the 21/2 of criterion (Tables 4-7 2nd column). The standard deviation of eight individual results of is equal to 7 of= [2/7C (DCSr- CS )2]1'2 1 Any value of DCSr-CS can be considered significant (p<0.05) if lDCSr-CS,I >2'l2a'. If attention is paid to the corresponding relative values (Tables 4-7 3rd column) obtained from DCSr-CS,(%)= [(DCSr-CS,)/CS,] x 100 Table 3 CS values for systems 1-8 [9]-[ 123 with Cu" Zn" and Mn" (ref.8) CU" * cu" t Zn" i Mn" * System 1 2 3 4 5 6 7 8 [ 101 1111 c 121 c91 CSS f a 0.0191 fO.OOO4 0.0181 f0.0004 0.02 10 f 0.0003 0.0195+0.0004 0.01 62 f 0.0002 0.01 66 f 0.000 1 0.01 8 1 f 0.0002 0.01 63 & 0.0002 [ 0.0 192 f 0.00021 [ 0.01 72 f 0.00081 [ 0.0204 f 0.00021 [ 0.0 179 f 0.00021 r 0.9990 0.999 1 0.9994 0.9990 0.9996 0.9998 0.9996 0.9994 [ 0.99983 [ 0.99493 [ 0.99971 [ 0.99961 CSS f a 0.0191 f 0.0004 0.0198 fO.0003 0.01 77 f 0.0003 0.0 19 1 0.0002 0.0 193 f O.OOO3 0.0188 fO.OOO2 0.01 99 f O.OOO4 0.01 86 f 0.0002 C0.0199 f0.0004] [ 0.01 89 f 0.0003] [0.0190 +0.0003] [ 0.0 192 & 0.00021 r 0.9990 0.9996 0.9993 0.9996 0.9993 0.9997 0.9988 0.9998 [ 0.99891 [ 0.99931 [ 0.99931 [ 0.99961 CSS f a 0.1210 f 0.001 7 0.0984 f 0.0025 0.0994 f 0.0032 0.1253 f 0.0026 0.092 1 f 0.00 12 0.1037 f 0.001 5 0.09 12 f 0.0027 0.1051 f0.0016 [ 0.1 142 f 0.00233 [ 0.1003 f 0.00203 [0.1053f0.0019] [ 0.1 1 18 f 0.001 81 r 0.9995 0.9984 0.9974 0.9989 0.9996 0.9995 0.9978 0.9994 :0.9990] 10.99901 0.99921 :0.9994] CSS f a 0.0 190 f 0.0002 0.01 97 -t O.OOO1 0.02 16 f 0.0003 0.02 1 1 f 0.0002 0.01 74 f 0.0001 0.0 168 f 0.0002 0.01 88 f 0.0002 0.0 190 f 0.0002 [ 0.01 88 f 0.00021 [ 0.01 56 k 0.0001 3 [ 0.0205 f 0.0003] [ 0.0202 f 0.0002) r 0.9997 0.9999 0.9994 0.9998 0.9999 0.9997 0.9997 0.9998 [ 0.99981 [ 0.99981 [ 0.99961 [ 0.99981 * Mass ratio 1 5.t Mass ratio 1 29. Line intercepts are not statistically different from zero at a probability level of 99%.Number of points n=6. Table 4 Treatment of data for Cu" systems at a 1 5 mass ratio (1.0 mol l-.' HCI systems in square brackets) DCSr- CS C&*' t f fJ 0.0194 f 0.0003 A bsol u tea + 0.0026 [ + O.O008] -0.0012 [-0.0013] + 0.0009 [+0.0017] + 0.0003 - 0.0004 [ - O.O004] [ - O.O005] -0.0007 [ - O.OOO31 [ - O.ooOo] 2l'*a' = O.OO24 [ 0.00 181 + 0.0004 Relative (%) + 13.5 [ + 3.91 + 5.0 [ + 8.51 + 1.4 - 7.0 [ - 7.01 - 2.3 [- 1.93 [ - 2.73 - 4.0 [- 1.31 [-0.11 + 2.4 0.0175 +0.0003 [0.0184f0.0003] 0.01 86 f 0.0004 [ 0.0199 f 0.00021 0.01 79 f 0.0003 [ 0.01 89 f 0.00023 0.0182f0.0004 C0.0188 f 0.00041 0.0175f0.0003 [ 0.01 88 f 0.00023 0.0182 & O.OOO3 [0.0190f0.0004] Proposed treatment t-test for parallelism of lines and y TI. 0.9976 0.9972 [ 0.99741 0.995 1 [ 0.99873 0.9977 [ 0.998 1 3 0.9940 [ 0.99561 c1.9957 [ 0.99891 0.9976 [ 0.99623 CS,**t fa 0.0 168 f 0.0002 [0.0187f0.0003] 0.01 86 f 0.0004 C0.0197 f0.0003] 0.01 74 f 0.0003 [ 0.01 80 f 0.00031 0.0181 f0.0005 C0.0192 +0.0004] 0.0 178 f 0.0003 C0.0193 f0.00021 0.01 85 f O.OOO4 [0.0192fO.OOO4] 0.0179 k O.OOO4 [0.0191f0.0003] rY 0.9985 [ 0.99663 0.9946 [ 0.99751 0.9972 [ 0.99743 0.9930 [ 0.99563 0.9974 [ 0.99853 0.9959 [ 0.99531 0.9935 [ 0.99751 t-Value Inference1 7.6 18 S§ C1.6971 [NS] 2.259 Sll [ 3.2261 2.527 C5.4191 0.457 C0.7021 0.942 [ 1 .080] 1.977 C0.7741 0.507 [ 0.1941 FCSrJCS 1.15 0.94 [0.93] 1.07 [1.11] 0.99 [ 0.983 1.03 [0.98] 0.95 [ 0.983 1.01 c 1.041 c 1.001 ~~ ~ * Weighted average for N=4 lines total number of points n=24.t Line intercepts not statistically different from zero at a probability level of 99%.$ SINS = significant/no significant difference between CS values. 9 p < 0.001.~ p < 0.05. 11 p < 0.005. ** p < 0.025. tt p z 0.05. Journal of Aiialytical Atomic Spectrometry October 1996 Vol. 11 975Table 5 Treatment of data for Cu” systems at a 1 29 mass ratio (0.2 mol I-’ HCI systems in square brackets) ~~ Proposed treatment DCSr- CS t-test for parallelism of lines r and y Rela ti ve W) - 1.3 t-Value 0.852 [ 1.3533 1.583 [2.558] 0.25 1 [ 1.3153 2.412 [ 1.4821 4.121 C2.7011 0.161 [ 1.6391 1.337 [ 0.4343 FCSr/CSy 0.99 Absolute’ - 0.0002 [ - 0.00041 + 0.0004 [ + 0.00071 -0.0001 [ - 0.00043 - 0.0006 [ + 0.00041 -0.0010 [ - 0.00O71 -0.0001 [ + 0.00041 + 0.0004 [ - 0.000 13 2lW =0.0010 [0.0010] CSr* f u 0.01 89 f 0.0002 c s y * f 0 0.0 19 1 f 0.0002 C0.0193 +0.0002] 0.0188 f 0.0002 [0.0187+0.0002] 0.01 90 f 0.0001 [ 0.0193 f 0.0001] 0.0187 f 0.0002 [ 0.0189 f 0.00021 0.01 95 f 0.0002 [ 0.0 194 f 0.00023 0.0190 f 0.0002 [ 0.0 189 f 0.00021 0.0188 f 0.0002 [ 0.01 91 f 0.00021 rr 0.9985 rY 0.9990 [0.9991] 0.9984 [ 0.99891 0.9993 [ 0.99941 0.9989 [ 0.99841 0.9992 [ 0.999 1 3 0.9982 [ 0.99861 0.9986 [ 0.99831 [ 0.983 1.02 [ - 2.01 + 2.3 0.0 192 f 0.0002 [ 0.0 194 f 0.00021 0.0190 f 0.0002 [ 0.0 189 f 0.00023 0.0193 f0.0002 C0.0193 f0.00021 0.01 85 f 0.0002 C0.0187 + 0.00021 0.0 190 f 0.0002 C0.0193 +0.0002] 0.0192 f 0.0002 [0.0190f0.0002] 0.9992 [ 0.99901 0.9982 [ 0.99821 0.9989 0.9990 [ 0.99871 0.9993 [ 0.999 13 0.9990 [ 0.99931 co.99921 c1.041 1 .00 [0.98] 1.03 [ 1.021 0.95 [0.96] 1 .00 [ 1.021 1.02 [ 0.991 [ + 3.41 - 0.4 [- 1.91 + 3.3 - 5.4 [+2.1] Did E le [- 3.73 - 0.2 [ + 2.31 + 1.9 [- 0.61 -~~ ~ *Weighted average for N = 4 lines total number of points n = 24.tLine intercepts not statistically different from zero at a probability level of 99%. $S/NS =significant/not significant difference between CS values. §p < 0.025. Yp < 0.001 JJp < 0.010. Table 6 Treatment of data for Zn” (0.2 mol I-’ HCI systems in square brackets). Dilution level 1 29 Proposed treatment DCSr - CS t-test for parallelism of lines r and y Relative (”/I + 12.4 [ + 2.93 - 1.4 t-Value 4.020 [ 0.9921 0.391 C0.6281 1.994 C0.3261 0.323 [ 0.2091 1.516 C0.8391 1.305 [ 3.7051 3.453 [ 2.3 171 Absol u tea + 0.01 30 [ + 0.00323 - 0.001 5 [ - 0.00203 - 0.0072 [+0.0010] -0.0012 [ - 0.00071 + 0.0056 CSr**t f o 0.1 1 10 f 0.0028 CS,*.t&U 0.0980 & 0.0016 0.10531 f 0.0029 [0.1105$+ 0.00233 0.108 1 1 _+ 0.0024 [0.1089$f 0.00251 0.10513 & 0.0028 [0.1098$ & 0.00251 0.1018Sf 0.0031 [0.1108$ f 0.00243 0.103911*** f 0.0021 [0.106511*** f 0.00161 0.0988 & 0.0014 [0.1059$ & 0.00191 co.i079$+ o.oois] riY Ala FCSrICS 1.13 [ 1.033 0.99 [ 0.981 0.93 0.99 [ 1.013 rr 0.9927 rY 0.9970 [ 0.99791 0.99 15 [ 0.99501 0.9946 [ 0.9941 3 0.9920 [ 0.9940) 0.9896 [ 0.99481 0.9955 [ 0.99761 0.9976 [0.9964] 0.1038 f0.0024 [0.1085 f0.00223 0.1010~f0.0027 0.1039 f0.0025 [0.1091$ f0.00201 0.10731 f0.0020 [ 0.108 1 $ f 0.00213 0.1 109 11 *** & O.OO49 [0.117611*** $0.0026) 0.1103$+0.0030 [ 0.1 130 & 0.00243 [0.1100$fO.0020] 0.9940 C0.99531 0.99 17 C0.99611 0.9934 [ 0.99631 0.9962 [ 0.99563 0.9790 [ 0.99491 0.99 15 [ 0.99491 [- 1.81 - 6.9 CIC [+ 1.03 - 1.2 [ - 0.61 + 5.3 “SI NS [ 0.991 1.05 Ele [ - 0.00271 + 0.0127 [ - 2.43 + 12.1 “sl NS [ 0.983 1.07 [ 1.101 1.12 [ 1.071 [+ 15.73 + 11.1 [ + 6.41 [ + 0.01721 +0.0116 21‘2~’ =0.0177 [ + O.OO701 [ 0.01441 *Weighted average for N =4 lines total number of points n = 24.tS/NS = significant/not significant difference between CS values. SLine intercepts not statistically different from zero at a probability level of 99%. §p <0.001. qp x0.05. IINon-ideal calibration lines. **NO significant difference between line intercepts.ttSignificant difference between lines intercepts (t = 2.532 p < 0.02). S:p < 0.005. @p x0.025.it is evident that Youden and Steiner’s’ criterion tolerates changes in CS even higher than 10%. Therefore in order to obtain more reliable information on the influence of variable parameters on the value of CS t-testing is proposed for parallelism of weighted averages of calibration lines obtained at levels be accounted for by the physical interferences of the protein namely the change in viscosity and surface tension but also by the chemical interferences of the acidic medium itself. In contrast in alkaline TRIS-acetate samples a soluble anionic form of albumin was favoured. The effects observed could be assigned to the physical interference of the human albumin. These assumptions are supported by the behaviour of the 1 5 and y (Tables 4-7 columns 4 6 8 and 9).Behaviour of Investigated Systems The physico-chemical characteristics and behaviour of the systems examined were mainly governed by the acidity of the sample media particularly at a mass ratio of 1 5. Accordingly slow precipitation of human albumin was observed in slightly acidic aqueous systems at the isoelectric point (PI= 4.9) and extensive denaturation in highly acidic solutions. The latter effect could be responsible for the possible occlusion of metal ions within the precipitate and a significant decrease in the value of CS. It follows that the effects observed could not only mass ratio simulant solutions containing CU”. the albumin-containing systems. The stabilizing activity of Triton X-100 was evident for all Copper(I1) systems When the determination of Cu“ in whole animal liver homo- genates is examined the systems of practical importance are those of 1 5 or 1 29 mass ratios.’ Although the first dilution level is appropriate for whole rat liver homogenates with typical concentrations of Cu” of 0.8 mg 1-’ the latter is useful 976 Journal of Analytical Atomic Spectrometry October 1996 Vol.11Table 7 Treatment of data for Mn" (1.0 mol I-' HCl systems in square brackets). Mass ratio 1 5 DCSr- CS CSr*'t 10 0.0203 f 0.0002 Dld Absolu te8 + 0.0024 [ + 0.00161 -0.0019 [ - 0.00261 - 0 . m [ + O.OoO81 +0.0001 [ + O.OoO61 +0.0002 [- 0.0010] -0.o001 [ + 0.00041 -0.0005 [ - O.oolO] 21'2d = 0.0023 [ 0.0025] ~~ Relative W) +11.6 [+ 7.91 - 10.4 [- 14.01 + 0.2 [ + 4.01 - 0.4 [ + 2.91 - 1.0 [ - 5.01 - 0.4 [ + 2.21 - 2.7 [ - 5.41 0.0182f0.0003 C0.0183 +0.0003] 0.0192 & 0.0003 C0.0199 +_0.0003] 0.0191 fO.OOO1 C0.0198 f0.00021 0.0191 _f0.0004 [0.0191 f0.00051 0.0191 f0.0003 C0.0198 f0.00021 0.0189 f 0,0003 [ 0.0190 f 0.0005] Proposed treatment t-test for parallelism of lines r and y rr 0.9984 0.9977 [ 0.99603 0.9966 [ 0.99801 0.9996 [ 0.99941 0.9957 [ 0.99301 0.9974 C0.99861 0.9964 [ 0.99353 CS,*.t & u 0.01 80 f 0.0002 C0.0187 f0.00041 0.0201 f 0.0003 [ 0.0208 f 0.00023 0.0192 f0.0003 [0.0191+_0.0005] 0.0192 f 0.0005 C0.0193 f O.OoO51 0.0193 & 0.0003 0.0192f0.0004 C0.0193 1 0.0005] 0.0194 & 0.0003 [ 0.0200 k 0.0002) [0.0201 f 0.00021 rY 0.9984 [ 0.99453 0.9978 0.9964 [ 0.99371 0.9935 [ 0.99221 0.9973 [ 0.99881 0.9957 0.9968 [ 0.99851 [ 0.99931 [ 0.99291 t-Value 7.392 [ 3.3771 4.976 [ 6.7923 0.060 0.179 [ 1.0931 0.387 [ 1.8543 0.162 CO.8221 1.090 [ 1.9973 [ 1.5371 F CSr/CS 1.13 [ 1.091 0.91 [0.88] 1 .00 1 .oo [ 1.033 0.99 [0.95] 1 .oo 0.97 C0.951 c1.041 [ 1.023 * Weighted average for N=4 lines total number of points n =24.Line intercepts not statistically different from zero at a probability level of 99%. 2 S/NS = significantlno significant difference between CS values. 9 p ~0.001. p <0.005. 11 p c0.05. for both whole rat and NET SRM 1577a Bovine Liver homogenates with corresponding Cu concentrations of (1.2 and 1.4 mg 1-'. All these concentrations are covered by the linear concentration range for Cu" which was established to be 0.1-2.1 mg I-'. Low-level Cu samples may call for a lower dilution level e.g.a 1 5 mass ratio. Higher dilution levels are preferred whenever possible to reduce the effects of clogging ion interferences memory effects and viscosity differences encountered in FAAS. The linearity range for Cu" was justified by the favuurable correlation coefficients (r) and relative standard deviations of CS( RSDCS) which were 2 0.999 and 0.8-2.4% respectively for single calibration lines and > 0.998 and 0.7-2.6% respect- ively for collective calibration lines. The fact that the intercepts of either single (Table 3) or collective (Tables 4 and 5) calibration lines were found not to be statistically different from zero at a probability level of 99% enabled the slopes of the ideal calibration lines to be used as the basis for evaluation of the effects of variable parameters.As expected the slopes of the calibration lines namely [he CS values are more strongly affected at lower dilution levels than at higher dilution levels (Tables 4 and 5 ) . Analysis of the data for the 1 5 mass ratio systems (Table 4) shows that according to Youden and Steiner's' criterion only the influence of TRIS-acetate buffer could be recognized as significant. A marked decrease in CS in TAB indicates that this medium is not interchangeable with water. The opposite is true for 1.0mol I-' HCl medium. t-Testing for parallelism of lines also shows that the influence of Triton X-100 and human albumin is significant. Owing to the opposite directions of these effects they are expected to cancel each other lout. Data for the 1:29 mass ratio systems are displayed in Table 5.The data show that there is no interference from either alkaline or acidic media. It is known that Cu" is susceptible to alkaline media;" hence the absence of interference from TAB/5 is surprising although this phenomenon has been observed previo~sly.'~ However Cu" suffered a marked decrease in CS (of approximately 15%) in TAB i.e. in alkaline medium at a mass ratio of 1 5 (Table 4). The unexpected behaviour of Cu" systems at a 1 29 mass ratio has still to be elucidated. As can be seen from Table 5 Youden and Steiner's' criterion failed to isolate any working parameter as an uncertain one. The influence of both NaCl and KH2P04 seems to be signifi- cant when water is replaced with TAB/5. Owing to the opposite directions of these effects they should cancel each other out.The replacement of water with 0.2mol I-' HCl however highlighted the physical interference of Triton X-100 which led to a decrease in CS; a significant increase in CS was observed in the presence of KH2P04. Again oppositely directed effects are expected to cancel each other out. Zinc@) systems Owing to the high sensitivity of FAAS nearly all Zn determi- nations in biological materials can be performed rapidly and easily using fairly high dilution levels.' The systems of practical importance for the measurement of Zn in whole liver homogen- ates are those at a mass ratio of 1:29 with typical Zn concentrations of 1.1-1.2 mg 1-' for both whole rat liver and NIST SRM 1577a Bovine Liver.' The calibration procedure is therefore performed within the linear concentration range of Zn of 0.1-1.5 mg 1-'.Favourable r values (> 0.997 for single calibration lines and 3 0.99 for collective calibration lines) and respective RSDCS values of 1.3-3.2 and 1.4-4.4% justify the linear regression in this concentration range. According to Youden and Steiner's' criterion (see Tables 3 and 6) none of the seven variable factors influences the analytical results within analytical systems 1-8 significantly. When considering systems 1-4 through [9]-[ 121 only the influence of DBC can be recognized as significant. t-Testing for parallelism of lines r and y and the FCS,/CS ratio indicate that a simple alkaline medium (TAB/5) and ageing of simulated samples for 24 h may significantly decrease the value of CS for systems 1-8. This is in accordance with the well known fact that an alkaline medium leads to precipitation of zinc hydrox- ide'' and to losses of Zn ions due to sorption on glass surfaces.2o It follows that TRIS-acetate buffer pH 8.6 and water are not interchangeable media.The physical interference caused by human albumin (mass fraction 0.22%) is evident from an increase in the value of CS. A similar effect has been reported previously." When an aqueous medium is replaced with a highly acidic medium (0.2mol I-' HCl) significant influences of DBC and solution ageing are observed. The latter is evident from a decrease in the value of CS and is probably due to occlusion of Zn within the precipitate of denatured albumin. Nevertheless water and HC1 matrices seem to be Journal of Analytical Atomic Spectrometry October 1996 Vol.11 977interchangeable. These facts are in accordance with the pH-dependent behaviour of 2n.l' A pronounced influence of DBC was observed for systems 1-8 as well as for systems 1-4 through [9]-[ 121. The use of DBC markedly changes the characteristics of the calibration lines by introducing persistent line intercepts which may differ significantly from one another. It was also observed that t- testing for parallelism of non-ideal lines (even with not signifi- cantly differing intercepts) in comparison with ideal lines made the recognition of significant differences between CS values more difficult (e.g. a bias of 7% caused by DBC could not be claimed as significant). This makes ideal calibration lines even more favourable. Hence they were obtained and compared throughout the study of Zn" systems except when the influence of DBC was evaluated.Since DBC and solution ageing seem to be highly uncertain factors the use of fresh standard solutions and background correction throughout is recommended. Manganese(11) systems When determining Mn the more sensitive ETAAS methods seem to be more appropriate than FAAS.'12' When Mn is present at high concentrations FAAS still seems to be usefuL5 The linear concentration range for Mn" was established to be 0.1-2.0 mg I-' covering the typical concentrations for Mn of 0.4-0.5 mg 1-' in 1 5 mass ratio homogenates of both whole rat liver and NIST SRM 1577a Bovine Liver.' Values of r for single (Table 3) and collective (Table 7) calibration lines were >0.999 and >0.99 respectively while RSDCS values were found to be 0.6-1.5 and O.6-2.6% respectively.Ideal calibration lines were established throughout the study. t-Testing of parallelism of lines showed a significant decrease in CS in both TRIS-acetate buffer and 1.0mol 1-' HC1. The former is probably due to hydrolysis'' and sorption phen- omena,20 whereas the latter could probably be attributed to chemical interference encountered with FAAS.22 In contrast significant increases in the value of CS caused by Triton X-100 were observed in both media. Hence neither Tris-acetate buffer nor 1.0 mol 1-' HC1 are directly interchangeable with water; however they might be interchangeable with water if they contained Triton X-100. Applying Youden and Steiner's' criterion only the influence of TAB within systems 1-8 and of Triton X-100 within systems 1-4 through [9]-[ 121 was recognized as significant.Both methods of data treatment indicate that replacing water with 1.0 mol 1-' HC1 results in more pronounced effects of variable parameters than replacing water with buffer. CONCLUSION In general the effects observed by varying several working parameters did not affect either the range of linear response or the intercepts of the calibration lines but strongly affected the calibration line slopes. This justified the use of CS values as the basis for drawing the conclusions presented here. t-Testing of parallelism of collective calibration lines at two alternative levels of the chosen variable parameters (r and y ) proved to be more selective in evaluating the effect of each parameter on the value of CS than Youden and Steiner's' criterion.The former approach extracts more uncertain factors than the latter which tolerates errors even higher than 10%. However Youden and Steiner's' differences may still be useful in indicating the direction of the analytical bias. Therefore combining both approaches enables the net effect of more parameters to be anticipated. Chemical interferences from the simple media used were not only observed as such but were also reflected in the behaviour of acidic systems containing protein. The disturbances and physical interferences from non-ionic surfactant strongly affec- ted the determination of the analytes examined. Together with the influence of DBC and ageing for Zn" they should be taken as the main sources of inaccuracy in the results.From a practical point of view our results make it possible not only to compare the behaviour of either highly acidic or alkaline whole liver homogenates with those prepared in water but also to suggest the use of aqueous standards for calibration instead of matrix-matched standards. It therefore follows that reliable calibration with aqueous standards is possible if Cu" at a 1:29 mass ratio is to be determined from a TABS homogenate. For Mn'' homogenates containing Triton X-100 should be used whereas for Cu" at a mass ratio of 1:5 and for Zn" the use of an alkaline medium is not appropriate. Matrix-matched standards might also be replaced with aque- ous standards during the determination of Cu" at both mass ratios from HC1 homogenates containing Triton X-100.The same is true for Mn" whereas the determination of Zn" can successfully be performed from simple HCl homogenates. Although interactions between the parameters varied simul- taneously and cannot be excluded from the effects observed agreement of the results obtained here with those recently obtained by systematic investigations of Zn" systems," shows that the short plan concept can successfully replace comprehen- sive systematic investigations. This justifies the use of the short plan concept in conjunction with t-testing of parallelism of calibration lines for recognizing experimental factors as poss- ible sources of systematic error. This approach should also be applicable to other analytical systems.The author thanks Professor Dr. A. Bezjak for providing the computer packages LSQANAL TWOLINE and TWOLINEl. REFERENCES 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 Delves H. T. Prog. Anal. At. Spectrosc. 1981 4 1. Jackson K. W. and Mitchell D. G. Anal. Chim. Acta 1975,80,39. Hinners T. A. Fresenius' Z . Anal. Chem. 1975 277 377. Mohamed N.,vand Fry R. C. Anal. Chem. 1981 53 450. Luterotti S. Zanic-GrubigiC T. and JuretiC D. Analyst 1992 117 141. Carnrick G. Schlemmer G. and Slavin W. Am. Lab. 1991 23 120. Plackett R. L. and Burman J. P. Biometrika (London) 1946 33 305. Youden W. J. and Steiner E. H. Statistical Manual of the Association of OfJicial Analytical Chemists Association of Official Analytical Chemists Washington DC 1975. Fisher B. V. Anal. Proc. 1984 21 443. Doerffel K. Pharmazie 1994 49 216. Luterotti S. Analyst 1995 120 925. Bezjak A. personal communication. Tallarida R. J. and Murray R. B. Manual of Pharmacologic Calculations with Computer Programs Springer New York 2nd edn. 1987 pp. 19-22. IUPAC Analytical Chemistry Division Commission on Spectrochemical and other Optical Procedures for Analysis Spectrochim. Acta Part B 1978 33 242. Pszonicki L. Lechotycki A. and Krupinski M. Talanta 1988 35 465. Thompson M. Walton S. J. and Wood S . J. Analyst 1979 104 299. Waughman G. J. and Brett T. Environ. Res. 1980 21 385. Kragten J. Atlas of Metal-Ligand Equilibria in Aqueous Solution Ellis Horwood Chichester 1978. Luterotti S. unpublished work. Smith A. E. Analyst 1973 98 65. Paynter D. I. Anal. Chem. 1979 51 2086. Ramirez-Mufioz J. Atomic-absorption Spectroscopy and Analysis by Atomic-Absorption Flame Photometry Elsevier Amsterdam London New York 1968 p. 268. Paper 6/01 306F Received February 23 1996 Accepted June 7 1996 978 Journal of Analytical Atomic Spectrometry October 1996 Vol. 11

ISSN:0267-9477    

DOI:10.1039/JA9961100973

出版商:RSC    

年代:1996

数据来源: RSC

摘要:

Thermally Stabilized Iridium on an Integrated Carbide-coated Platform as a Permanent Modifier for Hydride-forming Elements in Electrothermal Atomic Absorption Spectrometry Part 2.* Hydride Generation and Collection and Behaviour of Some Organoelement Species DIMITER L. TSALEV? ALESSANDRO D'ULIVO AND LEONARD0 LAMPUGNANI CNR Istituto di Chimica Analitica Strumentale Via Risorgimenh 35 56100 Pisa Italy MARCO D I MARCO AND ROBERTO ZAMBONI Universita di Pisa Dipartimento di Chimica e Chimica Industriale Via Risorgimento 35 561 00 Pisa Italy The in situ collection of volatile hydrides in an electrothermal atomizer with an integrated platform pre-treated with 110 pg of Zr or 240 pg of W and 2 pg of Ir for permanent modification was studied. An optimization study of the performance characteristics of an automated FI-HG-ET.4AS system based on an FI hydride generator interfaced with a transverse-heated graphite atomizer and longitudinal Zeeman- effect background correction was elaborated.The HG step for Asu' AsV BiII' Sb"' SbV SeIV Sn'" and Te" as well as for several alkylated species of As and Sn was optimized by means of a full factorial 3' design the factors being the concentrations of acid and tetrahydroborate (or their supply rates in pmol s- '). The corresponding regression equations are tabulated and representative response surfaces and contour diagrams are plotted. All inorganic hydrides except for SIIH are generated and collected with high efficiency at tetrahydroborate concentrations of 0.25-0.4% m/v sample acidity of 1.5-3 moll-' HCl trapping temperatures of 400 "C and a purge gas flow of argon of 100-130 ml min-'.The optimum conditions for stannane and alkyltin hydrides are pH 1-4 tetrahydroborate concentrations of 0.2-0.4% m/v trapping temperatures between 400 and 600°C and argon flow rates of 60-120 ml min-'. Arsine monomethylarsine and dimethylarsine are effectively collected on both coatings at temperatures between 400 and 500°C and purge gas flow rates of 70-120 ml min-'. Optimum HG conditions differ strongly for As"' AsV monomethylarsonate and dimethylarsinate species with this FI system unless L-cysteine is added. Organoelement species of As Sn and Se are thermally stabilized in a similar manner on both Ir-Zr- and Ir-W- treated platforms the least stable species being selenornethionine and trimethylselenonium.The best levelling- off effect on the integrated absorbance for different analyfe species (isoformation) is observed for As and the worst for organotins particularly for trialkylated species such as tributyltin trimethyltin and trimethylselenonium. Relatively better isoformation is achieved for organotins on Ir-W- and for organoselenium on Ir-Zr-treated platforms. The long-term stability of the Ir-Zr and Ir-W modifier coatings during at least 600-700 thermal cycles is demonstrated. The Ir-Zr treatment is preferred to Ir-W for hydride trapping owing to * For Part 1 of this series see ref. 1. -f On leave from Faculty of Chemistry University of Sofia. Sofia 1 126 Bulgaria. Journal of Analytical Atomic Spectrometry lower atomization temperatures longer lifetime of the atomizer and an absence of double peaks.Such peaks persist for Bi and Te on Ir-W-treated platforms. The best characteristic masses in integrated absorbance measurements with Ir-Zr-treated platforms are close to those for the direct injection mode &. 35 107 83 43 104 48 31 32 153 146 148 145 and 152 pg for AS"' Bim SbI'I Se" Snw Te" monomethylarsonate dimethylarsinate monomethyltin dimethyltin trimethyltin diethyltin and monobutyltin respectively. Analytical results for As Sb and Se in certified reference materials (water and autoclave-decomposed sediments) are in good agreement with the certified contents. Keywords Electrothermal atomic absorption spectrometry hydride generation and collection in situ enrichment; chemical modiJication; iridium- and zirconium-treated platform; iridium- and tungsten-treated platform; sediment; water; methylarsenic; organotin; organoselenium In Part 1 of this series,' a permanent Ir modifier stabilized on a Zr- or W-treated platform was proposed with prospective application in direct ETAAS for the determination of volatile analytes (As Bi Cd Pb Sb Se Sn Te T1) and more promisingly in HG-ETAAS for in situ trapping/enrichment of volatile hydride-forming elements ( HFEs).~ The scant earlier reports on permanent modification in HG-ETAAS3-' have since been supplemented by several papers on HG-ETAAS,6-8 HG-FANES9 and HG-ETV- ICP-MS," as well as on direct ETAAS with permanent modi- fication.l1.l2 In addition to the classical techniques for injection of more sophisticated and efficient approaches for applying modifier coatings have been reported uiz.electro- plating'*'' and sputtering.l'*l2 Bulska and Jedrall' electro- plated Pd or Rh onto the inner surface of the graphite tube thus extending the analytical lifetime for the direct ETAAS determinations of As and Se up to 80 and 160 firings in the presence of Pd and Rh respectively. Rademeyer et deposited Ir on the inner surface of the graphite tube by cathodic sputtering of Ir in a low-pressure Ar discharge; the integrated absorbance for Se was then effectively constant for about 750 firings while the peak height signal gradually decreased after about 350 firings in direct ETAAS. Several workers have studied trapping of hydrides of AS,^*^.^.^^ Bi,'v9 Ge,8 Pb,4 Sb,9 Se,',',' Sn346 and Te9 on tubeP9 or platforms6Tl0 pre-coated for permanent modification with Pd,8*10 Ir,6*9 Journal of Analytical Atomic Spectrometry October 1996 Vol.11 (979-988) 979Pd-Ir,5,6 Nb,6 Ta 9 9 W Zr3,477*8 and Pd-Zr.' Attempts have been made to extend the useful lifetime of modifier coatings which range from 80 firings (Pd," Zr7) to 300' or 40O6 complete trapping and atomization cycles. Among the prob- lems faced with these coatings are partial" and non-uniform surface coverage10q'2 and re-distribution of modifier within the atomizer,12 peak shape distortion6*' and even appearance of double peaks,1*6*12 gradual impairment of sensitivity the peak height signals being more vulnerable,6*12 carry-over,6 overstabilization of certain analytes,'*'2 etc.Studies on the behaviour of some organoelement species with known biological and environmental occurrence and importan~e~~'~*'~ in vapour generation and in situ trapping are interesting with a view to at least two possible application fields (i) species-independent quantification by HG-ETAAS or ETAAS which relies on simplified sample pre-treatment thus eliminating the need for time-consuming digestions and species isoformation; and (ii) speciation in which certain analyte species could be generated selectively and determined by HG-ETAAS or be separated by another technique (chroma- tography extraction cryogenic trapping etc2) and then be quantified by ETAAS or HG-ETAAS. While the in situ trapping of inorganic hydrides in graphite atomizers (HG-ETAAS) is well recognized,2 very little work has been published so far on the possibility for trapping of organo- element species of HFES.~~"-'~ Thus Sturgeon and c o - w ~ r k e r s ~ ' ~ ~ ~ proposed a fractionation/quantification scheme for As species which employed selective HG cation-exchange preconcentration HG after UV oxidative digestion of organoe- lement species and simultaneous HG of arsine monomethylar- sine and dimethylarsine in a batch mode system in 0.5 mol I-' HCl with collection of hydrides in a heated graphite tube at 600"C in which the pyrolytic layer had been removed by abrasion.16 Zhang et al." collected dimethylselenide and diethylselenide in Pd-coated graphite tubes at temperatures between 600 and 900°C.Ni et ~ 1 . ~ trapped stannane (SnH,) and tributyltin hydride [(C4H9)3SnH] at temperatures in the ranges 400-600 and 100-700 "C respectively in Zr-treated graphite tubes down to characteristic masses for peak height measurements (m,) of 14 and 20 pg and limits of detection of 58 and 78pg respectively.Sturgeon et a1." generated and trapped tetraethyllead [Pb(C,H,),] with an over-all efficiency of 58-t 3% and mp=28 pg; the trapping temperatures were within 200 and 600°C with pyrolytic graphite coated tubes and 100-800 "C with Pd-treated pyrocoated tubes. The aim of this study was to evaluate the performance characteristics of Ir-Zr- and Ir-W-treated platforms in an automated FI-HG-ETAAS system as well as in direct ETAAS determinations uiz. effects of HG parameters purge gas flow collection (trapping) temperature long-term stability of measurements behaviour of some organoelement species and applicability to real samples/digests.EXPERIMENTAL Apparatus A Perkin-Elmer (Uberlingen Germany) Model 4100 ZL atomic absorption spectrometer with a transverse-heated graphite atomizer (THGA) and longitudinal Zeeman-effect background correction and an AS-71 autosampler were employed in ETAAS studies. Standard THGA graphite tubes with inte- grated platforms (Part No. B300-0643) were pre-treated with about 1.2-1.3 pmol of a carbide-forming element (1 10 pg of Zr or 240pg of W) and then with 2pg of Ir,' and are referred t o in this paper as 'Ir-Zr-treated' or 'Ir-W-treated'. In FI-HG-ETAAS studies the spectrometer was interfaced with a Perkin-Elmer FIAS-400 flow injection system as shown in Fig.1. The system was operated under the Software Version C L ' TO HG R - -w Fig. 1 Schematic diagram of the manifold and the instrumental set- up for HG and in situ collection. P and P peristaltic pumps of the FIAS-400 system; L sample coil; L and L reaction coils; S sample; C carrier; R reductant; W waste; GLS gas-liquid separator; and F filter (for details see text and Table 1) 7.21 (Part No. B0509524). The PTFE capillary of the AS-71 was disconnected from the autosampler arm and was replaced by a quartz capillary which was connected to the PTFE tubing carrying hydrides from the gas-liquid separator. The quartz capillary was carefully adjusted so as to inject hydrides over the heated platform of the THGA without touching the platform or the tube walls the optimum distance between the quartz tip and the heated platform surface being about 1.3-1.5 mm.The optimized parameters for the FIAS-400 mani- fold are given in Table 1. In Table 2 is given the FIAS-400 programme which is synchronized with the THGA electrother- mal atomizer programme (Table 3). These programmes are slightly modified compared with those recommended by the manufact~rer,'~*~~ in order to reduce carry-over in experiments with broadly varied compositions of reagents and carrier and to ensure the long-term stability of coatings during numerous cycles/firings. Table 1 FIAS-400 FI manifold coils tubings and flow rates (see also Fig. 1 and text) Sample coil (L,) Reaction coil ( L,) Reaction coil ( L3) Filter (F) Sample conduit Carrier conduit Reductant conduit Waste from gas-liquid PTFE 500 pl 640 x 1 mm id or Teflon PFTE 100 x 1 mm id PFTE 300 x 1 mm id PFTE membrane Part No.B050-8567 Tygon 1.52 mm id 'yellow/blue' 6.4 ml min-' 80 rev min-' of P Tygon 1.52 mm id 'yellow/blue' 6.4 ml min-l 80 rev min-' of P Tygon 1.14 mm id 'red/red' 4.0 ml min-' 80 rev min-' of P Tygon 3.18 mm id 'white/black' 10.5 ml min-' 80 rev min-' of P PFA 2 ml 640 x 2 mm id separator Table 2 FIAS-400 programme for HG-ETAAS Step No. Pre-fill 1 2 3 4 5 Pump 1/ Time/s rev min-' 10/15* 100 10/40t 100 8 0 30/80t 0 8 0 5 0 Pump 2/ rev min-' 0/50* 80 0 80 0 80 Valve position Fill Fill Inject Inject Inject Fill * The higher setting is preferred for reduced carry-over in experi- t For 0.5 and 2 ml sample coils respectively. ments with varied compositions of acid and reductant.980 Journal of Analytical Atomic Spectrometry October 1996 VoE. 11Table 3 THGA electrothermal atomizer programme. Compromise trapping temperature ('injection temperature') 400 "C (see Fig. 4 for the intervals studied). Autosampler arm position 'out' during steps 1-4 Step Temperature/ Ar flow rate/ No. "C Ramp/s Hold/s ml min-' Read 1 Variable* 1 12 250 - 2 600 4 4 250 - 3 20007 0 5 0 Yes 4 2100$ 1 3 250 - * Varied in the range 30-600 "C during optimization studie:,. t Compromise setting (see text); must not exceed 2050 "C. Compromise setting (see text); must not exceed 2150 "C. Solutions of other reagents were prepared from L-( +)-ascorbic acid (No. 1018 Baker Analyzed Reagent; J. T. Baker Phillipsburgh NJ USA) potassium iodide (pro analysi; Carlo Erba Milan Italy) and tartaric acid (Analyticals Carlo Erba).The water used in all operations was prepared by means of a Milli-Q water-purification system (Millipore Milford MA USA). Certified and Standard Reference Materials The CRMs and SRMs were obtained from the National Research Council of Canada Ottawa Canada uiz. Riverine Water Reference Material (RM) for Trace Metals (SLRS-3) Nearshore Seawater RM for Trace Metals (CASS-3) Marine Sediment RM for Trace Elements and Other Constituents (BCSS-1 and MESS-1); the Institute for Reference Materials and Measurements Commission of the European Communities Geel Belgium viz. Estuarine Sediment (CRM 277) and River Sediment (CRM 320); and NIST Washington DC USA viz. SRM 1572 Citrus Leaves. The THGA temperature programme for direct ETAAS determinations with Ir-W-treated platforms is given in 'Table 4 . For other temperature programmes and experimental details see ref.1. Reagents The preparation of stock standard solutions has been described in Part -1.' Other reagents for preparation of stock solutions of some organoelement species their formulae abbreviations Sample Preparation and supplier are given in Table 5. Working standard solutions for direct ETAAS were prepared before use in 0.2% v/v HNO (2 ng of As or Se and 4 ng of Sn). A stock solution of 5% m/v NaBH in 0.3% m/v NaOH was prepared by dissolving sodium tetrahydroborate pellets (for AAS Spectrosol; BDH Poole Dorset UK) in an aqueous solution of NaOH (30% solution Suprapur; Merck Darmstadt Germany) and filtration through a 0.45 pm mem- brane filter.This solution was stored refrigerated for up to 2 weeks and was diluted before use (daily) as required. Typically a 0.2% m/v NaBH,-0.05% m/v NaOH solution was used unless stated otherwise. A 10% m/v L-cysteine stock solution was prepared by dissolving the solid reagent (No. 30089; Fluka Biochemika Buchs Switzerland) in 0.05 moll-' HCl (Suprapur; hlerck). Table 4 THGA temperature programme for direct ETAAS determi- nations with Ir-W-treated platform Step Temperature/ Ar flow rate/ No. "C Ramp/s Hold/s ml min-' Read 1 110 1 20 250 2 130 5 20 250 3 800/1400* 10 10 250 - 4 1600/2100* 0 5 0 Yes 5 2000/2100* 1 3 250 - - - * For Pb and As respectively. Table 5 Organoelement compounds and their abbreviations Formula Name and supplier Abbreviation CH3AsO(ONa),.6H20 Sodium methylarsonate* M MA (CH3),02AsNa.3H,0 Sodium cacodylate* DMA CH,SnCl Methyltin trichloridet M MT (CH3)2SnC12 Dimethyltin dichloridet DMT (CH ),SnCI Trimethyltin chloridet.ThlT Water samples acidified to pH 1.6 with HNO were diluted with an appropriate acid or acid plus pre-reductant solution Procedure A 9 + 1 dilution with HC1 for SetV determination; Procedure B 19+1 dilution with 20% m/v Kl-20% m/v ascorbic acid-10 moll- ' HC1; Procedure C 9 + 1 dilution with 10% m/v L-cysteine solution. Blank and standard solutions were diluted in the same manner thus providing reagent/acid- matched solutions for the calibration graph or standard additions technique in FI-HG-ETAAS. Pre-reduction times were not less than 1 and 8 h with L-cysteine and KI-ascorbic acid respectively. Sediment and plant samples (0.2 g) were acid-leached or completely decomposed in Parr PTFE bombs (Parr Instrument Company Moline IL USA; Model 4782) by microwave digestion according to DUlivo et uZ.,~ with some modifi- cations as given below.Microwave heating was applied for 1 min at 850 W and for 1 min at 0 W; the cycle was repeated ten times. Procedure D sediment (0.1 g) leached with 2 ml of HNO then diluted to 50ml with water. Procedure E plant soaked with 2ml of HNO for 24 h then autoclave- decomposed in the microwave oven as above; diluted to 50 ml with water. Procedure F sediment leached with 3 ml of HNO and 1 ml of HCl then diluted to 50 ml with 3 mol I-' HCl. Procedure G sediment digested with 3 ml of HNO 1 ml of HCl and 0.5 ml of HF then diluted to 50 ml with 1 mol I-' HC1; an aliquot of digest pre-reduced with 1% m/v L-cysteine at 60 "C for 1 h in 0.025 moll-' HCl for As or 1 moll-' HC1 for Sb determination; 0.5% m/v H3B03 added to complex Procedure H (for Pb by direct ETAAS) as Procedure G but without pre-reduction step.Procedure I (for Sn) as Procedure G but digests diluted to a final 0.15 mol 1-' HC1 concentration. Experimental Design for Studying Hydride Generation and Collection Diethyltin dichloridet DI'T The effect of concentrations (or supply rates in pmol s-') of (C,H,),SnCI C4H9SnCl Monobutyltin trichloridet M BT (C4H,),SnOOCCH3 Tributyltin acetate? TUT the reagents NaBH and HCl was studied by varying these C,HllN02Se DL-Selenomethionine$ SE M concentrations within the practical ranges for the present HG (CH313Sel Trimethylselenonium iodide$ TMS system viz.0.05-0.4% m/v NaBH in 0.05% m/v NaOH. . 01-3 mol 1-' HCl and pH'1-7 (for in). A full factorial experi ment of the 32 type was applied according to the plan suggested by the program of the Statistical Graphics System Version 6 (ManUgistics Rockvik MD USA). Randomized experiments with three replicate measurements were performed the highest * Carlo Erba (Milan Italy). t Alfa Johnson-Matthey (Karlsruhe Germany). $1 Janssen (Geel Belgium). 5 Synthesized according to the method described by Palmer et a/.21 and tested as detailed by D'Ulivo et Journal of Analytical Atomic Spectrometry October 1996 Vol. 11 981mean value of integrated absorbance (QA) for each analyte species being assigned a value of 100% recovery (Y%). In optimization experiments of hydride trapping the Ar purge gas flow rate was varied between 60 and 210 ml min-' at a fixed trapping temperature of 400 "C and then the trapping temperature was varied between 30 and 600°C at a fixed Ar flow rate of 125mlmin-'.The chemical parameters of HG were fixed in these experiments as follows 0.2% m/v NaBH,-0.05% NaOH and 0.5 mol 1-' HC1 except for Se" and Te" (1.2moll-' HC1) and for SnIV (2% m/v H,BO,-0.1 moll-' HC1) and organotins (1% m/v tartaric acid pH 1). The analyte levels were kept at 2.5 ng except for SnIV and organotins (5 ng) and AsV (10 ng). RESULTS AND DISCUSSION Optimization of Hydride Generation The results for numerous hydride-forming elements/species derived from the 32 full factorial experiment are summarized in Table 6.The three-dimensional response surfaces are described by regression equations of the type Y% = k + k [H '3 + k2 [NaBH,] + k [H' ][NaBH,] + k4 [ H f I 2 + k [NaBH,l2 The normalized integrated absorbance signal (Y%) does not necessarily represent the absolute chemical yield but rather the relative over-all effect of all relevant factors (HG and transport trapping efficiency and effects on ETAAS signal) on analyte response. Nevertheless the over-all efficiency of this particular FI-HG-ETAAS system appears to be very high for most analytes except for the pentavalent species AsV and SbV which give lower signals by a factor of about eight and four compared with the trivalent forms As"' and Sb"' respectively.For the remainder of the elements the best characteristic masses for integrated absorbance measurements m are AS"' 43 pg; BiI'I 107 pg; SbI" 87 pg; Se" 43 pg; Te" 48 pg; and Sn" 156 pg which are comparable to the corresponding figures from Part 1' for direct ETAAS measurements with an Ir-Zr-treated plat- form viz. 30 176 92 43 50 and 71 pg respectively. Further {experiments on FI-HG-ETAAS with more favourable HG media helped to improve these figures e.g. in the presence of 1% m/v L-cysteine down to m of 35 110 83 and 104 pg for AS'" Bi"' Sb"' and Sn" respectively. Stannane (SnH,) generation is obviously critically dependent on pH in a dilute HC1-2% m/v H,BO medium (Fig. 2) which renders this element outside the scheme for multi-element optimization and unified conditions as is possible for As Bi Sb Se and Te (Fig.3). Further improvement of the character- istic mass for the inorganic Sn" (i-SnIv) can be achieved viz. from 156 pg in the above medium to 104 pg in the presence of 1 YO m/v ~-cysteine-0.15 mol 1-' HC1 or 1% m/v tartaric acid (pH 1). Pentavalent species of As and Sb are also practically outside the working field in Fig. 3 because of lower sensitivity coupled with higher blanks arising from more concentrated 1 2 3 4 5 6 7 PH Fig.2 Overlay of the contour plots at 90% response (normalized integrated absorbance signal) for six Sn species versus pH and reduct- ant concentration. The species symbol is placed on the side of ~ 9 0 % response. Hydride collection on Ir-Zr-treated platform.The 10-80% contours for i-SnIv are also shown Table 6 reagent supply rates (pmol s-') derived from the 3' full factorial experiment Model fitting results for the regression equations Y% versus reagent concentrations [HCl/mol 1 - I and NaBH (% m/v)] or Y% versus Coefficients Analyte ko Reagent concentrations- As"' 42.3 AsV 32.0 MMA 28.0 DMA 51.9 Bill1 90.8 Sb"' 79.7 SbV 11.2 Se" 85.2 TeIV 67.9 Reagent supply rates- As"' 42.3 AsV 32.0 MMA 28.0 DMA 51.9 Bit11 90.8 Sb"' 79.7 SbV 11.2 Se" 85.2 Te" 67.9 SnIV 2.09 MMT 3.20 DMT 2.66 TMT - 7.23 DET -3.51 MBT 6.62 kl 24.5 26.2 - - 17.6 - 65.7 -0.327 6.97 -21.5 - 1.66 15.8 0.230 0.245 -0.165 -0.616 - 0.003 1 -0.201 - 0.01 56 0.148 0.0654 9097 9159 9214 9059 9311 9037 k2 95.8 100 196 178.1 27.2 72.0 21.6 57.1 244 5.44 - 5.70 11.1 10.1 1.542 4.08 1.57 3.23 13.9 - 0.322 - 1.54 - 1.20 3.17 1.78 - 2.76 k3 - 10.4 35.4 - 37.5 -46.3 - 0.287 - 6.06 - 63.5 4.29 - 1.31 - 0.0056 0.0188 -0.0120 - 0.0246 - 0.0001 - 0.0032 -0.0338 0.002 28 - 0.0007 - 0.401 0.127 0.128 0.382 0.0148 0.0660 k4 - 4.65 3.34 0.0742 6.51 0.206 - 3.34 - 0.0004 - 0.0009 - 10.1 16.3 - 1.81 0.000 29 0.001 43 6 x - 0.0001 0.000 57 2 x 10-5 -3 x 10-4 - 852.3 - 857.9 - 863.0 - 846.0 - 872.1 - 846.4 k5 0.035 1 444 - 14.8 -95.1 -9.51 - 74.6 - 39.4 14.2 -42.5 0.000 081 1.43 - 0.0478 - 0.306 - 0.0306 - 0.240 -0.127 0.002 28 -0.137 -0.0319 0.144 0.105 - 0.283 0.1 a4 0.226 982 Journal of Analytical Atomic Spectrometry October 1996 Vol.11AsV 9( 0 0.5 1.0 115 2:O 2:5 3:O HCI I mol I-' Fig. 3 Overlay of the contour plots at 90% response (nornialized integrated absorbance signal) for nine hydride-forming species uersus acid and reductant concentrations. The element symbol is placed on the side of >90% response.Trapping on Ir-Zr-treated platform NaBH solutions (> 0.35% m/v). Pre-reduction to their tri- valent forms preferably with 1 YO m/v L-cysteine-dilute HCl was therefore practised in subsequent work with real sample solutions and digests. It is worth noting that interesting and analytically useful improvements in the hydride generation pattern of As Sb Bi Sn and alkylated species of As and Sn are observed in the presence of L-cysteine which will be discussed in Part 3 of this series. Position of the Quartz Pipette Tip The manufacturer recommends adjusting the depth position so that the quartz pipette tip stops approximately 1 5 mm above the L'vov platform.20 Chaudhry et aL2' observed that the Pd deposition and hydride injection modes had a critical effect on the results obtained (for GeH,); hence the eflect of the distance between the pipette tip and platform surface was tested for two typical analytes As"' and BiIII at three tip positions-low normal and high i.e.1.0 1.5 and 2.0 mm taking as a guide the outer diameter of the tip (about 1.3 mm). Since no effect was observed and moreover the precision of such variable settings was poor this parameter was fixed at approximately 1.4-1.5 mm in all subsequent experiments. Effect of Trapping Temperature The effect of trapping temperature for six inorganic hydrides is shown in Fig. 4. As expected by analogy with Pd-treated atomizers,2 the trapping efficiency for Ir-Zr- and Ir-W-treated platforms is very high the optimum temperatures being gener- ally within the range 300-500 "C.Slightly higher temperatures (40&500"C) are needed for ASH SnH and H,Se as com- pared with Pd- or Pd-Ir-treated s ~ r f a c e s ~ ' ~ ~ which could be due to the lower reactivity of the stabilized Ir compared with Pd at lower temperatures and particularly on tungsten carbide (WC)-coated platforms. Therefore the optimum trapping tem- peratures for Sb Se and Sn are higher by about 1WcC for Ir-W compared with Ir-Zr coatings. On the other hand the precision of measurements is impaired at both extremes of the temperature range uiz. at 30°C for As on both coatings and at 600°C for Sb Se and Te on Ir-W and for Sn on It-Zr.Peak height signals are higher but less reproducible than integrated absorbance measurements. A good compromise temperature for the simultaneous collection of all six hydrides would be 400°C which entails only a minor sensitivity loss compared with the optimum single-element conditions f i x As on Ir-Zr (<5%) Sb on Ir-W (<5%) and Sn on Ir-W (10%). No damage to the quartz probe was experienced at this temperature during several thousand injections. In Fig. 4 the effect of trapping temperature on pyrocoated and W-treated platforms for two typical hydrides with different thermal stabilities uiz. ASH and BiH is also shown. Bismuthine (BiH,) is trapped readily on both surfaces even at room temperature (30 "C) and is thermally stable until about 300 "C.The surface treatment was found to have a minor effect on characteristic masses for integrated absorbance measure- ments c$ mo= 102 112 and 122 pg for Bi trapped on pyro- coated W-coated and Ir-W-coated platforms respectively. Arsine (ASH,) is very inefficiently trapped on pyrocoated platforms even at 500°C (m =254 pg and very poor pre- cision) whereas trapping on W-treated (m = 72 pg) and Ir-W- treated platforms (m = 79 pg) is almost quantitative. Recently Haug and Liao6 collected SnH on WC-coated tubes/platforms at 600 "C with apparent success. Effect of Argon Flow Rate The purge gas flow rate could be expected to affect not only the trapping efficiency but also the stripping of hydrides out of solution. Nevertheless its effect is relatively small within a fairly broad range from 60 to 210 ml min-'.A slight decline in sensitivity is observed at higher Ar flow rates (Ir-W > Ir-Zr e.g. down to -22% for As and -15% for Se with Ir-W- coated platforms). It is worth noting that flow rates above 3 ml s-' are very high in comparison with the volume of this type of atomizer ( ~ 0 . 5 ml). The precision of measurements is not adversely affected by higher flow rates and a good compromise condition for all six elements would be an Ar flow rate of 100-130 ml min-'. Build-up of Residues After prolonged work particularly in the continuous-flow mode with 10ml samples (e.g. at levels of Se and Sb below 0.1 pg 1-I) a gradual build-up of black deposits on the lower (heated) part of the quartz capillary tip was observed.Presumably these are residues of reduced As products because the As levels exceeded those of Sb and Se by one or two orders of magnitude. Although these residues appeared to have no detrimental effect they could be considered as possible sources of sensitivity drift in view of the observations by Brovko,26 who found good trapping of ASH and H,Se on graphite surfaces modified with a thin Sb layer deposited by pyrolysis of SbH,. Therefore the residues were removed by briefly soaking the quartz tip in a small plastic cup containing HN0,-HC1 and then rinsing with water. Residues were thus readily dissolved without disassembling the probe (Caution acid mist formed!). Double Peaks Persistent double peaks with relatively good reproducibility were observed for Bi and Te with Ir-W-treated platforms in direct ETAAS measurements and (less so) in HG-ETAAS.The appearance of double or distorted peaks has also been reported by other ~ o r k e r s . ~ * * ' ~ ? ~ ~ While such peaks corrupt measure- ments of peak height (A,) signals their incidence could well be tolerated in the integrated absorbance mode. One may expect that part of the hydride is trapped on the tube walls and is subsequently atomized earlier giving a fast pre-peak; this effect should be most pronounced with the least stable hydrides viz. BiH and H2Te. However double peaks are also observed for the injected samples in direct ETA AS,' wherein no hydride trapping/atomization is taking place. Another explanation could be that the surface coverage of the treated Journal of Analytical Atomic Spectrometry October 1996 Vol.1 1 983100- 80 - 60 - 40 - 20 - 60J 40 - 20 - 4 0 ! - 1 - I - 1 - I - 4 - 1 1 0 100 200 300 400 500 600 Te # # # 0'' d 60 - ' 0 40 - 20 - * Se o r . I . . I - 0 100 200 300 400 500 600 6o 1 20 401 Sb 0 4 . I ~ 1 . I - I - I - I 0 100 200 300 400 500 600 0 100 200 300 400 500 600 1 Trapping temperaturePC Fig. 4 Effect of trapping temperature for different hydrides on the normalized integrated absorbance signal. Trapping on platforms treated with Ir-Zr (filled circles) Ir-W (open circles) pyrolytic graphite (cross symbols and dotted line) and W (star symbols and dashed line). Ar flow rate fixed at 125 ml min-' platform with modifier components is incomplete and non- uniform. Such imperfect coatings have been experimentally observed for Pd,l0*l1 Rhl' and IrI2 on graphite1°-12 or metallic platforms,1° particularly when coatings are formed by injec- tion12 rather than by electroplating'O~" or sputtering.lo9l2 Supposedly three types of surface site are present on an Ir-W-treated platform pyrolytic graphite WC and WC with Ir deposits.These sites can contribute in a different manner to the processes of thermal stabilization of analytes during pyrol- ysis and subsequent atomization. In Fig. 5 are shown the pyrolysis-atomization curves for Bi and Te on pyrocoated W-treated and Ir-W-treated platforms. As can be seen both analytes are thermally unstable on pyrocoated and W-treated platforms i.e. above 300°C for Te and 3W5OO"C for Bi. Atomization signals for both elements start at very low tem- peratures (< 1000 "C) in the absence of Ir.On the basis of these observations which unfortunately have not yet been confirmed by studies on the distribution of coatings on the platform surface these double peaks might originate from two surface sites uiz. those not coated with Ir (pyrographite and WC-coated) and Ir-WC-coated. In the hydride-trapping mode the first peak is lower presumably because in this instance the hydrides should be preferentially trapped on the noble metal sites and even if trapped on an uncoated surface might then be redistributed during pyrolysis by migrating to the more 'sticky' Ir-W-coated sites. Thus the production of more uni- form and efficient coatings would be a good challenge for manufacturers of instruments and graphite parts.Behaviour of Some Organoelement Species Thermal stabilization and isoformation in direct E TAAS Efficient thermal stabilization of As"' AsV monomethylarson- ate (MMA) and dimethylarsinate (DMA) in 0.2% v/v HNO on Ir-Zr- and Ir-W-treated platforms up to pyrolysis tempera- tures (Gyr) of 1300 "C was observed with no marked differences between the various As species. The optimum atomization temperatures (TJ for all species are 2000 and 2100°C with Ir-Zr- and Ir-W-treated platforms respectively whereas reproducibility of measurements is impaired at 2100 versus 2000 "C for both coatings. DMA exhibits slightly better sensi- tivity for both integrated absorbance and peak height measure- ments as well as with both coatings; the reasons for this behaviour are not yet known. Representative mean values for characteristic masses for QA and A measurements (m and mp 984 Journal of Analytical Atomic Spectrometry October 1996 Vol. 1 16o 1 -.-Pyre I Bi - I- Ir-W -g 100- 2 N .- - 80- 60 - 40 - 20 - - N- Ir-W Te 0 500 1000 1500 2000 Temperaturd'C 0 I I - - I I I Fig.5 Pyrolysis-atomization curves for Bi and Te deposited on different platform surfaces pyrolytic graphite coated (solid line); W-treated (dotted line); and Ir-W-treated (dashed line) respectively) with Ir-Zr-treated platforms at T = 2000' C are m = 31 30 and 26 pg and mp = 30 31 32 and 26 pg and with Ir-W-treated platforms at T = 2100 "C are m = 38 35 33 and 28 pg and mp= 20 20 19 and 17 pg for AS"' AsV MMA and DMA respectively.Therefore both permanent coatings provide an acceptable degree of levelling-off of the signals for various species (isoformation) as well as thermal stabilization for these As species.Preference could be given to the Ir-Zr coating owing to the lower Tt slightly lower m and better reproducibility of measurements at 2000 "C (RSD = 1-3% for 3-5 successive runs). For Sn species in 0.2% v/v HNO the Ir-W-coated plat- forms provide better (albeit not complete) isoformation than the Ir-Zr coating. Pyrolysis temperatures up to 1400 C are tolerated. Similar sensitivities are observed with both perma- nent coatings for example with Ir-Zr-treated platforms at T,,=2000°C m,=110 104 113 141 95 112 and 196pg and mp=72 70 73 89 63 72 and 133 pg; and with Ir-W-treated platforms at T,,=21OO0C m,=92 90 88 108 91 92 and 134 pg and mp = 38 38 38 47 39 38 and 57 pg for 1-Sn" for monomethyltin (MMT) dimethyltin (DMT) trimet hyltin (TMT) diethyltin (DET) monobutyltin (MBT) and tributyltin (TBT) respectively.The mp figures with Ir-Zr-treated plat- forms can be improved by approximately 25-30% at T,,= 2100 "C; however no further improvement in isoformation results. Thus TMT and TBT appear to persist during pyrolysis up to 1400°C and are then volatilized/atomized in a different manner than the remainder of the Sn species. Pyrolysis-atomization curves for Se" SeV' selenomethion- ine (SEM) and trimethylselenonium (TMS) in 0.2% v/v HNO with Ir-Zr- and Ir-W-treated platforms (not shown) reveal small analyte losses above 1100-1200°C for all species. Organoselenium species are stabilized in a similar manner on both coatings but persistently give lower signals than the inorganic SeIV and Se". The corresponding characteristic masses (m,) at T,,=2100 "C are 43 42 47 and 50 pg with Ir-Zr- and 49 54 72 and 48 pg with Ir-W-treated platforms for Se" SeV1 SEM and TMS respectively.It follows from the above studies that the best isoformation is achieved for As and the worst for Sn species and that lower signals are obtained for some di- and trialkylated species such as (CH3),Sel (CH3),SnCl and (C4H9)3SnOOCCH3. These observations are generally in agreement with the literature on thermal stabilization and isoformation of organoelement species Se,31-35 Sn,36-39 e t ~ . ~ ' . ~ ~ ) namely that (i) chem- ical modifiers are often better thermal stabilizers than isofor- mers of various specie^,'^,'^,^' and (ii) coatings of carbides (reviewed in ref.42)13*36,43 and solid modifier residues36 are generally less effective than are the modifier additions to each injected sample aliquot. Trapping of ursine and methylarsines In Fig. 6 is shown the effect of trapping temperature for ASH CH,AsH2 and (CH,),AsH on the normalized integrated absorbance signal. All three hydrides are efficiently trapped on Ir-Zr- and Ir-W-treated platforms at optimum temperatures in the range 400-500°C. Trapping of methylated hydrides (DMA>MMA) starts even at room temperature on Ir-Zr- and less so on Ir-W-treated platforms. Dimethylarsine is even trapped at 30 "C on a pyrolytic platform (m,=59 pg) and on a W-treated platform (m =46 pg) but is thermally unstable on these surfaces and readily lost at pyrolysis temperatures in the range 100-200 "C and above 200-300 "C respectively.The best characteristic masses (m,) for the over-all procedure of HG and trapping are 43 77 and 56 pg (HG from dilute HC1) and 35 31 and 32 pg (HG from 1% m/v L-cysteine- dilute HCl) for inorganic As"' (i-As"') MMA and DMA respectively. These figures are favourable compared with the characteristic masses (m,) for 'injected arsenic' viz. 31 30 and 26 pg for i-As"' MMA and DMA respectively. The effect of Ar flow rate is not critical with Ir-Zr-treated platforms and the optimum settings are fairly low viz. 70-120 ml min-'. Higher flow rates of the purge gas could be expected to improve stripping from solution particularly for the less volatile alkyl hydrides; cfi the boiling-points of -55 2 and 35.6 "C for ASH CH,AsH and (CH,),AsH respectively.On the other hand the surface of the Ir-W-coated platform appears to be less reactive than Ir-Zr for trapping hydrides. Hence the resulting effect is a slight species-dependent decline in sensitivity at higher Ar flow rates e.g. by 4 6 and 11% with Ir-Zr and 1 2 and 17% with Ir-W coatings for ASH CH,AsH2 and (CH,),AsH respectively. Trapping of stannane and alkyltin hydrides The best results for trapping of all six tin hydrides on an Ir-Zr-treated platform are obtained at higher trapping tem- peratures in the range 400-600°C and at Ar flow rates of 60-120 ml min-'. A slight sensitivity decrease (10%) is observed for C4H9SnH at 600°C and for all species at Ar flow rates above 130mlmin-'.Under the optimized con- ditions the best characteristic mass (m,) values for the over- all HG/trapping procedure are 158 153 146 148 145 and 152 pg for i-SnIv MMT DMT TMT DET and MBT respect- ively which are higher than the corresponding m values for 'injected tin' viz. 110 104 113 141,95 and 112 pg respectively thus indicating chemical yields between 65 and 95%. Tin appears to be one of the analytes that is generated as hydrides with chemical yields lower than 100%; see e.g. previous estimates in the literature uiz. 75?40,~~ 45-80%45 and 85%.46 Optimization of hydride generation for organoelement species In Table 6 are given the regression equations for the response surfaces describing the HG process for inorganic and organo- element hydrides.Selected cross-sections at 90% response for .Journal of Analytical Atomic Spectrometry October 1996 Vol. 11 985Fig. 6 Effect of trapping temperature for different arsenic hydrides [ASH CH,AsH2 and (CH,),AsH] on the normalized integrated absorbance signal. Hydride collection on four types of platform pyrocoated WC- Ir-Zr- and Ir-W-treated. Ar flow rate fixed at 125 ml min-' each species are overlayed for Sn (Fig. 2) and As species (Fig. 3). The contour plot for As (Fig. 3) shows that unfortu- nately there are no suitable conditions for the simultaneous generation of all three hydrides nor for their fractional determi- nation (speciation). There are probably very serious kinetic limitations to arsine generation from inorganic AsV (i-AsV) with this ~ y s t e m ~ ~ .~ ~ and therefore even the best characteristic mass (m,) for these species (see the upper right corner of Fig. 3) is approximately 8-fold higher than the m for i-As"' (275 versus 45 pg respectively). These differences between i-AslI1 i-AsV MMA and DMA in 0.1-0.5 moll-' HC1 medium are known to be much less pronounced in batch mode HG.2,16929 Analytically useful shifts of the 90% response contours are observed in the presence of L-cysteine pre-reductant which will be discussed in Part 3. The optimized conditions for the generation of tin hydrides are fairly uniform (Fig. 2) all six hydrides are generated with high efficiency (relative response > goo/,) at pH < 4 and reduct- ant concentrations above 0.20-0.25% m/v NaBH,.However the sensitivity for TMT is significantly lower than for the other species despite the high apparent yield of (CH3)3SnH gener- ation (%YO) as estimated from the ratio of the characteristic masses viz. 141 and 148 pg for 'injected' and generated/ trapped TMT respectively. Thus TMT and MBT are outside the scope of simultaneous multi-species trapping and quantification. performed. Part of these results are presented in Fig. 7 as the stability of rn uersus the number of firings. The characteristic mass values appear to be consistent considering that the effect of the day-to-day reproducibility is also superimposed on these results i.e. the effects of ageing of tubes radiation sources and FI manifold tubing precision of settings of numerous instrumental parameters alignment etc. The graphite tubes eventually cracked after more than 700 firings due to corrosion and thinning in the upper part of the tube around the injection hole probably because of the effect of entrained air during prolonged gas-stop steps of hydride trapping and atomization.The performance of two of the tested tubes coated with Ir-W was impaired after working in the direct ETAAS mode with digests containing acidic mixtures 120 1 4 0 i 100 80 < 60 40 \ OSb-3 AS-5 1 Long-term Stability Our experience is based on observations during the lifetime of 20 individual THGA graphite tubes with integrated platforms treated with Ir-Zr or Ir-W. Although no particular experi- ments had been devoted to the study of long-term performance it was possible to extract information from the archived data for several tubes during 1 year when these studies were 2o i- 0" f .I * ' " * I ' I * 0 100 200 300 400 500 600 700 No. of firings Fig. 7 Long-term stability of characteristic mass (m,) for As Sb Se and Sn as derived from archived data for six different graphite tubes in direct ETAAS (filled circles) and HG-ETAAS (open circles). Platforms treated with Ir-Zr (Nos. 1-5) or Ir-W (No. 6 ) 986 Journal of Analytical Atomic Spectrometry October 1996 Vol. 11of HN0,-HC10 or HN0,-HCI-HF resulting in sensitivity drift and background absorption. Therefore the use of such aggressive acids in sample preparation for direct ETAAS is discouraged. Applications Some applications to CRMs and SRMs of environmental waters and digested sediments and plant tissue were performed. The results given in Table 7 are in acceptable agreement with the certified values.As expected the sample preparation step was critical in these assays which were also supplemented by parallel analyses of the same digests by continuous-flow hydride generation non-dispersive atomic fluorescence spec- trometry (CF-HG-NDAFS).23*24 Discussion on the sample digestion is outside the scope of this paper; therefore. only a few comments will be given here. Lower results were obtained in direct ETAAS for Pb by acid leaching with HNO only (Procedure D) e.g. recoveries of 66 83 and 95% for CRM 320 MESS-1 and BCSS-1 respectively and for As in hIESS-1 and BCSS-1 respectively and for As in MESS-1 (77%). Small additions of HF (0.5 ml Procedure G ) are needed for complete recovery of Sb from BCSS-1 and CRM 320 sediments.Nevertheless the results for Sb were persistently lower for the CRM 277 Estuarine Sediment viz. recoveries of 8249% by FI-HG-ETAAS and 82-90% by CF-HG-NDAFS. The L-cysteine pre-reductant was given preference to KI-ascorbic acid in these assays because of faster pre-reduction and lower blanks; however the effect of sample acidity is Table 7 with Ir-Zr-treated platform* Analysis of environmental CRMs and SRMs by HG -ETAAS Certified Analyte Sample valuet Environmental waters (pg l - ' t As SLRS-3 0.72 k 0.05 CASS-3 1.09 & 0.07 Sb SLRS-3 0.12 f 0.01 Se" CASS-3 0.020 f 0.005 Sediments (pg g- 'b As BCSS- 1 11.1 f 1.4 CRM 277 47.3+ 1.6 CRM 320 76.7k3.4 Pb BCSS-1 22.7 It 3.4 MESS-1 34.0f6.1 CRM 320 42.3_+ 1.6 Sb BCSS-1 0.59 f 0.06 CRM 320 (0.6nI MESS-1 0.73 kO.08 Se MESS-1 0.34 f 0.06 BCSS- 1 0.43 f 0.06 Sn CRM 320 (6.0nl Plant (pg g-'& Pb SRM 1572 13.3f2.4 Found value$ 0.70 & 0.03 1.08 f 0.04 0.11 kO.08 0.1 3 -t 0.07 0.016 f 0.008 11.2+0.3* 11.8 f 0.03 46f 1 48.6 f 0.8 77 f 3* 7 9 f 2 2 1 .O f 0.9* 31.4* 39.6 -t 0.01 * 0.57 f 0.01 0.57 2 0.02 0.63 f 0.04 0.53 f 0.05 0.73 i- 0.09 0.74 0.02 0.31 f 0.04 0.43 f 0.04 6.25 f 0.05 13.1 &0.2* Notes C .Cys,a C .Cys,a B KI,a C .Cys,a A.a D,b Ci,Cys,a Ci ,Cys,a Ci,Cys,a rb,Cys,a Ci ,Cys,a €i,a €1 ,a € I ,a 0 ,Cys,a (r,Cys,b Cr,Cys,a,b Ca,Cys,a OI,Cys,a C,,Cys,a F .a F .a I Cys,b F .a * Results marked with an asterisk were obtained by direc! ETAAS t Mean _+ 95% confidence limits.1 Mean k s for three analyses.tj For code of sample preparation procedure see Experimental. a Calibration graph; b standard additions; KI pre-reductant 7 Information value only. with Ir-W platform. KI + ascorbic acid; Cys pre-reductant L-cysteine. critical in the presence of L-cysteine As being particularly vulnerable to acidities above 0.05 mol 1-' HCl. The beneficial effects of L-cysteine will be discussed in Part 3. Sample Throughput Rates In the FI-HG-ETAAS mode the throughput rates are approxi- mately 17 and 28 samples h - l with a 2000 and 500 pl sample coil respectively. In the continuous-flow mode with the maxi- mum possible collection time of 99 s," this figure is 16-20 samples h-l. In direct ETAAS the throughput rates are 30 and 10 samples h - l with injection volumes of 10 and 60 p1 (4 x 15 pl multiple injection mode) respectively. CONCLUSIONS Ir-Zr-treated platforms are more suitable than those treated with Ir-W for trapping hydrides of As Bi Sb Se Sn and Te in an automated FI-HG-ETAAS system owing to lower atomization temperatures longer lifetime of the atomizer (> 600 firings) less critical hydride trapping parameters and absence of double peaks.All inorganic hydrides except for SnH are generated and collected with high efficiency (relative response of more than 90%) at tetrahydroborate concentrations of 0.25-0.4% m/v sample acidity of 1.5-3 moll-' HCl trapping temperatures of 400 "C and Ar flow rates of 100-130 ml min-l. These con- ditions may be considered as a prospect for future simultaneous multi-element enrichment and quantification.Organoelement species of As Sn and Se are thermally stabilized in a similar manner on Ir-Zr- and Ir-W-treated platforms the least stable species being SEM and TMS. The best isoformation is observed for As and the worst for Sn species. Lower integrated absorbance signals are obtained for some di- and trialkylated species such as TBT TMT and TMS. Relatively better isoform- ation is observed for organotins on Ir-W- and for organo- selenium on Ir-Zr-treated platforms. Arsine monomethylarsine and dimethylarsine are effectively trapped on both coatings at temperatures in the range 400- 500°C. Stannane and alkyltin hydrides are trapped on Ir-Zr- treated platforms at temperatures between 400 and 600 "C the optimum collection temperature for all six species being 500 "C.Optimum HG conditions for AS'" As" MMA and DMA differ strongly within the range over which the chemical parameters were varied viz. 0.1-3 moll - HCl and 0.05-0.4% m/v NaBH,. Therefore there are no suitable conditions for the simultaneous generation and collection of all arsenic hydrides nor for their speciation. Stannane and the examined alkyltin hydrides are generated and collected with high efficiency within a fairly uniform field of parameters at pH values below 4 and reductant concentrations above 0.20-0.25% m/v NaBH,. The Ir-W-treated platforms are more suitable for direct ETAAS assays owing to better thermal stabilization and lower reagent blanks. However Bi and Te persistently produce double peaks probably originating from different deposition/ atomization sites because of non-uniform surface coverage with the modifier.Applications to As and Pb determinations in digests are possible provided that aggressive acid mixtures such as HC1-HClO and HC1-HN0,-HF are not employed in the sample preparation schemes. L-Cysteine is preferable to the KI-ascorbic acid pre- reductant for AsV and SbV although the optimum acidities for HG are substantially shifted to lower H + concentrations which is not compatible with acidic sample digests. These shifts however also affect the HG pattern of other analytes (Bi Sn methylated organoarsenicals and organotins) and can be utilized analytically as will be discussed in Part 3. Journal of Analytical Atomic Spectrometry October 1996 Vol. 11 987Financial support to D. L.T. by the Consiglio Nazionale delle Ricerche Istituto di Chimica Analitica Strumentale Pisa is gratefully acknowledged. This work was supported in part by the Ministry of University Scientific Research and Technology ( Progetto Speciale Sistema Lagunare Veneziano) Italy. One of the authors (R. Z.) thanks the CNR National Committee for the Environment and Habitat Technologies Italy for financial support. REFERENCES 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Tsalev D. L. D’Ulivo A. Lampugnani L. Di Marco M. and Zamboni R. J. Anal. At. Spectrom. 1995 10 1003. DEdina J. and Tsalev D. L. Hydride Generation Atomic Absorption Spectrometry Wiley Chichester 1995. Ni Z.-m. Hang H.-b. Li A. He B. and Xu F.-z. J. Anal. At. Spectrom. 1991 6 385.Yan X.-p. and Ni Z.-m. J. Anal. At. Spectrom. 1991 6 483. Shuttler I. L. Feuerstein M. and Schlemmer G. J . Anal. At. Spectrom. 1992 7 1299. Haug H. O. and Liao Y. Spectrochim. Acta Part B 1995 50 1311. Gabros S. Walcerz M. Bulska E. and Hulanicki A. Spectrochim. Acta Part B 1995 50 1669. Ni Z.-m. and Zhang D.-q. Spectrochim. Acta Part B 1995 50 1779. Dittrich K. Franz T. and Wennrich R. Spectrochim. Acta Part B 1995 50 1655. Marawi I. Olson L. K. Wang J. and Caruso J. A. J. Anal. At. Spectrom. 1995 10 7. Bulska E. and Jedral W. J. Anal. At. Spectrom. 1995 10 49. Rademeyer C. J. Radziuk B. Romanova N. Skaugset N. P. Skogstad A. and Thomassen Y. J. Anal. At. Spectrom. 1995 10 739. Tsalev D. L. Atomic Absorption Spectrometry in Occupational and Environmental Health Practice CRC Press Boca Raton FL 1984 vol.11. Tsalev D. L. Atomic Absorption Spectrometry in Occupational and Environmental Health Practice CRC Press Boca Raton FL 1995 vol. 111. Sturgeon R. E. Willie S. N. and Berman S . S. J. Anal. At. Spectrom. 1986 1 115. Sturgeon R. E. Siu K. W. M. Willie S. N. and Berman S. S. Analyst 1989 114 1393. Zhang L. Ni Z.-m. and Shan X.-q. Can. J. Appl. Spectrosc. 1991 36 47. Sturgeon R. E. Willie S. N. and Berman S . S. Anal. Chem. 1989,61 1867. The THGA Graphite Furnace Techniques and Recommended Conditions Publication B3210.20 Part No. B050-5538 Bodenseewerk Perkin-Elmer Uberlingen 1992. The FZAS-Furnace Technique Recommended Analytical Conditions 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 4 1 42 4.3 44 45 46 47 and General Information Publication B3212.20 Part No.0993-5204 Bodenseewerk Perkin-Elmer Uberlingen 1993. Palmer I. S. Fischer D. D. Halverson A. V. and Olson E. Biochim. Biophys. Acta 1969 177 336. DUlivo A. Lampugnani L. Sfetsios I. Zamboni R. and Forte C. Analyst 1994 119 633. DUlivo A. Lampugnani L. Pellegrini G. and Zamboni R. J. Anal. At. Spectrom. 1995 10 969. D’Ulivo A. Lampugnani L. Faraci D. and Zamboni R. unpublished work. Chaudhry M. M. Ure A. M. Cooksey B. G. Littlejohn D. and Halls D. J. Anal. Proc. 1991 28 44. Brovko I. A. Zh. Anal. Khim. 1987 42 1627. Sturgeon R. E. and Gregoire D. C. Spectrochim. Acta Part B 1994 49 1335. Odanaka Y. Matano O. and Goto S. Bunseki Kagaku 1979 28 517. Tsalev D. L. Mandjukov P. B. and Stratis J. A. J. Anal. At. Spectrom. 1987 2 135. Larsen E. H. J. Anal. At. Spectrom. 1991 6 375. Welz B. Schlemmer G. and Voellkopf U. Spectrochim. Acta Part B 1984,39 501. DoEekalova H. DoEekal B. Komarek J. and Novotny I. J. Anal. At. Spectrom. 1991 6 661. Krivan V. and Kueckenwaitz M. Fresenius’ J. Anal. Chem. 1992,342 692. Laborda F. Viiiuales J. Mir J. M. and Castillo J. R. J. Anal. At. Spectrom. 1993 8 737. Deaker M. and Maher W. J. Anal. At. Spectrom. 1995 10 423. Vickrey T. M. Harrison G. V. Ramelow G. J. and Carver J. C. Anal. Lett. 1980 13 781. Parks E. J. Blair W. R. and Brinckman F. E. Talanta 1985 32 633. Katsura T. Kato F. and Matsumoto K. Anal. Chim. Acta 1991 252 77. Matsumoto K. Katsura T. and Kato F. Anal. Sci. 1991 7 (Suppl.) 451. Tsalev D. L. Slaveykova V. I. and Mandjukov P. B. Spectrochim. Acta Rev. 1990 13 225. Tsalev D. L. At. Spectrosc. 1991 12 169. Volynskii A. B. Zh. Anal. Khim. 1987 42 1541. Tsalev D. L. Dimitrov T. A. and Mandjukov P. B. J. Anal. At. Spectrom. 1990 5 189. Sturgeon R. E. Willie S . N. and Berman S . S. Anal. Chem. 1987 59 2441. Mandjukov P. B. Djarkova V. and Tsalev D. L. in 6. Colloquium Atomspektrometrische Spurenanalytik ed. Welz B. Bodenseewerk Perkin-Elmer Uberlingen 1991 pp. 597-608. Frolov I. A. Kulakov S. I. and Yakish G. M. Zh. Prikl. Khim. 1977 50 2561. Tsalev D. L. Sperling M. and Welz B. Analyst 1992 117 1735. Paper 61045661 Received July 1 1996 988 Journal of Analytical Atomic Spectrometry October 1996 Vol. 11

ISSN:0267-9477    

DOI:10.1039/JA9961100979

出版商:RSC    

年代:1996

数据来源: RSC