摘要:

DECEMBER, 1968 THE ANALYST Vol. 93, No. I I13 The Effect of Filtration and Centrifugation on Raw Sugar Polarisation Analysis BY R. A. M. WILSON, C. G. SMITH, R. H. JAMES AND R. R. WALLACE (The Colonial Sugar Refining Co. Ltd., Central Laboratory, Sydney, Australia) In raw sugar analysis, the apparent sucrose content is determined as its polarisation value, a measure of the rotation of plane-polarised light passing through a solution of the sugar, which is directly proportional to sucrose concentration when the impurity level is low. Before taking a reading with a polarimeter, a “normal” raw sugar solution (26.000 g per 100 ml) is clarified by addition of basic lead acetate solution to precipitate impurities, which are removed by simple filtration ; about 50 ml of filtrate are needed for analysis.When the solution is filtered, the sucrose concentration increases because of preferential absorption of water by the filter-paper. Initially in the current study, the magnitude of the absorption has been examined, and it has been determined that the first 10 ml of filtrate should be discarded to minimise this effect. Differences of 1 part in 40,000 in the polarisation value of sugar solutions, however, can now be detected by using modem photo-electric polarimeters and the refined analytical techniques described in this paper. This compares with 1 part in 2000 when using older visual instruments and standard tech- niques. Therefore, an insight into evaporation and humidity effects has also been gained, and a comparison of filtration and centrifugation, an alternative method of clarification, made.Centrifugation has practical advantages over filtration in laboratories in which large numbers of analyses are performed. In addition, it is not subject to preferential absorption and evaporation errors, and currently appears to give the best estimate of polarisation value, as defined and specified by the International Commission for Uniform Methods of Sugar Analysis (ICUMSA) . THE polarisation value of a sugar is a measure, in units called International Sugar Degrees (” ISS), of the rotation of plane-polarised light passing through a solution of the sugar. The basis of the 100” point on the International Sugar Scale is defined by the International Commission for Uniform Methods of Sugar Analysis (1CUMSA)l as the optical rotation of the ‘normal” (26.000 g per lo0 ml weighed in air) solution of pure sucrose in a 200*000-mm tube at 20*00° C at the wavelength of the green line of the mercury-198 isotope (A = 546.227 nm in vacuo).The “normal” solution of a raw sugar is usually not clear enough to enable a reading to be taken with a polarimeter, and is, therefore, first clarified by the addition of basic lead acetate solution. This forms a flocculent precipitate that settles at the bottom of the flask, carrying with it suspended particles. The solution is separated from the precipitate by filtration, for which a 3-inch diameter stemless filter funnel (covered to minimise evaporation) resting in a 200-ml filter glass is used. The polarisation value of the filtrate is then determined as the optical rotation in O ISS measured in a sugar polarimeter.Hardin and Zerban2 have reported that the first portions of sugar solution passing through a filter-paper undergo an increase in polarisation, which they attributed to preferential absorption of water from the solution by the filter-paper. To minimise this effect, they recommended that at least 25 ml of initial filtrate be discarded. The latest official ICUMSA method for raw sugar polari~ation,~ however, specifies that 1 O m l be discarded. 0 SAC and the authors, 773774 WILSON et al. : EFFECT OF FILTRATION AND [Analyst, Vol. 93 At the time of the 14th Session of ICUMSA, in 1966, it was not known whether an air-dried filter-paper would reach equilibrium with the solution after only 10ml had been filtered, or whether the concentration of the subsequent filtrate, and hence the polarisation value, would still be affected.It was, therefore, recommended by this Session4 that "the optimum amount of filtrate discarded be investigated." This amount needs to be sufficient to ensure that the filtrate is unaffected by preferential absorption, but as small as possible to reduce the sample volume requirements, the time needed to complete the analysis and the consequent risk of concentration changes from solution evaporation. The primary aim of the work presented in this paper, therefore, was to determine the optimum amount of filtrate to be discarded. During the course of the work, however, the development of improved analytical techniques, in conjunction with the use of a high-precision photo-electric polarimeter, enabled a detailed study of evaporation and humidity effects, and a comparison of filtration and centrifugation methods, to be undertaken.In the usual method of centrifuging, the leaded raw sugar solution is sealed in a 60-ml capacity stainless-steel tube of about 2.7-cm diameter and centrifuged with either an MSE High Speed 17 or a Sorvall SS-3 Automatic Superspeed centrifuge. The speed is increased to 14,000 r.p.m. and then decreased to zero, under automatic braking. The cycle time is about 9 minutes for the MSE centrifuge and 5 minutes for the Sorvall machine. Under these conditions, the clarity of the final solution, measured as optical density at both 646 and 589 nm, is as good as, or better than, that of filtered solutions. If a raw sugar solution is clarified by centrifugation there is a significant difference in its polarisation value, but centrifugation has practical advantages over filtration in a labora- tory in which large numbers of samples are being analysed.In addition, centrifuged samples should not suffer from the effects of evaporation, preferential absorption or other possible sources of error occurring with filtration; therefore, centrifugation should give a better estimate of the polarisation value, as defined and specified by ICUMSA, than filtration. The 14th Session of ICUMSA,* in 1966, also recommended "that the optimum moisture content in filter-paper . . . be investigated.'' The current study has been extended to include a determination of the equilibrium moisture content of filter-paper and a limited attempt to gauge the effect of filter-paper moisture content on the polarisation result.EXPERIMENTAL TECHNIQUE USE OF HIGH-PRECISION POLARIMETER- Before 1960, the polarisation value of a clarified raw sugar solution was determined with a visual polarimeter, calibrated in O ISS, by measuring the rotation of plane-polarised light passing through the solution contained in a cell of 200-mm length. For a single, clear solution, two consecutive readings may differ by up to O*lOo ISS (1 part in 10oO). The usual practice is to determine the average of five readings to the nearest 0.01" ISS, but the standard error of the mean is about 0-02" to 0-04" ISS, depending on the type and condition of the instrument, and the experience of the analyst. In the early 1960's, the Bendix-NPL, Model 143A, automatic polarimeter was introduced into this Laboratory for raw sugar analysis.This instrument combines such features as short cell length (10 mm) and, therefore, small angular rotation, with the Faraday magnetic - optic effect for modulation of the plane of polarisation and compensation for the angle of rotation. These features made possible a considerable increase in the precision of results, so that consecutive readings seldom differ by more than 0.01" ISS. The use of a Solartron 1420.2 digital voltmeter to monitor the output enables each reading to be made to the nearest 0.0025" ISS (1 part in 40,000). The Bendix polarimeter can, therefore, be used to examine effects producing much smaller polarisation errors than could previously be detected with visual p olarimet ers .COLLECTION OF SMALL ALIQUOTS OF FILTRATE- To take full advantage of the high instrument precision, special analytical technique5 required to be developed to minimise experimental errors in the preliminary solution prepara- tion steps. In particular, special apparatus was designed to facilitate collection of filtrate in small aliquots without introducing sources of error, while at the same time preserving procedural and environmental consistency with normal filtration.December, 19681 CENTRIFUGATION ON RAW SUGAR POLARISATION ANALYSIS 775 A thistle funnel of 20-ml capacity was attached with wax to the underside of a 3-inch diameter, stemless, glass filter funnel.To keep evaporation at an absolute minimum, 10 ml of filtrate were first collected in the thistle funnel and then, by loosening a spring-clip, released quickly through a short piece of small-bore plastic tubing into a 10-ml calibrated Quickfit test-tube (Fig. 1). The test-tubes were stoppered immediately after filling; the filter funnel was not kept full during filtration, and was, of course, always kept covered. Cover Fig. 1. Filtration apparatus A “normal” solution (250 ml) of raw sugar (sucrose content between 96 and 100 per cent.) was prepared, defecated with basic lead acetate solution and filtered. For each such solution, 80 ml of filtrate were collected in 10-ml portions, as described above. Simultaneously, the usual filtration procedure was followed as a control, the first 10 ml of filtrate being discarded and about 70ml collected.Carlson Ford, BIC quality, U4512 filter-papers were used throughout this work. Other papers of similar grade may be used, provided that they are quick filtering and yield a sparkling clear filtrate ; preferential absorption characteristics may then be different. POLARISATION OF SMALL ALIQUOTS ( ‘‘MINI-POLS”)- At least 50ml of solution are required to flush and fill a 200-mm visual polarimeter tube, and 30ml to flush and fill the 10-mm long, 1-ml capacity, Bendix polarirneter flow- through cell by the usual gravity siphoning method. By using a vacuum, however, it was found that a minimum volume of 8 ml of solution was required with a Bendix polarimeter cell when two consecutive solutions differed by no more than 4” ISS; 8 ml of solution, with a small air bubble between samples to reduce sample mixing, was found to flush and fill the cell with no detectable contamination, that is, to within 1 part in 40,000.Without the small capacity of the Bendix cell, a detailed study of the effect of filtrate volume on polarisation would not have been possible. 1” C. In addition, the polarimeter cell is provided with a water jacket, and strict temperature control was maintained at 20.0” 4 0.1” C . RESULTS TRIAL 1- Polarisations of twenty-eight raw sugars were examined by the “mini-pol” procedure described above. By using the average polarisation value of the control over all twenty-eight samples as a base-line, the graph of average polarisation value against the progress of the altration was plotted (Fig.2). Results show a decrease in polarisation as filtration progresses, until about 40ml have been filtered. This presumably results from preferential absorption of water from the solution during the early stages of filtration. At this point the curve is about 0~001” ISS lower than All of the trials were carried out in a temperature-controlled laboratory at 20”776 WILSON et a2. : EFFECT OF FILTRATION AND the control. The polarisation then increases steadily until, after 80 ml, it is about 0@09" ISS higher than the control. The increase is, presumably, caused by evaporation of water from the solution. Filter-papers used in the trial were in equilibrium with ambient conditions and contained about 7 per cent. of moisture, determined by both Karl Fischer and oven-drylng methods.The ambient relative humidity during the trial was 70 per cent. TRIAL 2- The alternative procedure of centrifugation was compared with filtration. Trials with thirty-eight raw sugar samples showed a highly significant (more than 99.9 per cent. signifi- cance) difference of 0-013" ISS between the polarisation values of solutions clarified by filtration and the same solutions clarified by centrifugation, the results for the former being higher than those for the latter. A horizontal line representing the result by centrifugation as 0*013" ISS lower than the filtered control is drawn in Fig. 2. [Afinalyst, Vol. 93 s - 0.005 - 0.0 13" ISS Fig. 2. Effect of filtrate volume on polarisa- tion at 70 per cent. relative humidity (air-dry filter-papers used with about 7 per cent.of moisture). A, Experimental curve; and B, curve corrected for evaporation 0-85 x evapora- tion at 60 per cent. relative humidity subtracted (see Fig. 5) 80 0-10 10-20 20-30 30-40 40-50 50-60 60-70 71 Progress of filtration - (10-ml portions of filtrate) Fig. 3. Effect of filtrate volume on polarisation at 60 per cent. relative humidity (air-dry filter-papers used with about 7 per cent. of moisture). A, Experimental curve; and B, curve corrected for evaporation (see Fig. 6) Earlier trials with 151 samples had shown a difference of 0.025" ISS, but 0.013O ISS of this was later attributed to insufficient rinsing of the centrifuge tubes and lids with the test solution. Tubes and lids are normally washed with water and, in the early trials, were rinsed only once with the test solution; three rinsings were subsequently found to be necessary, and this practice was adopted for the thirty-eight sugars mentioned above, and for later trials TRIAL 3- The polarisations of seven raw sugars were examined by the "mini-pol" procedure as in Trial 1, except that the control on each sample was centrifuged. Filter-papers again con tained about 7 per cent.of moisture, but the atmospheric humidity was lower (60 per cent.) The average polarisation value compared with the control is plotted against the progress of filtration in Fig. 3. The form of the curve is similar to that for Trial 1, except that the right-hand portion (50 to 80 ml) is about 20 per cent.steeper than for Trial 1 ; this reflects a higher rate of evaporation at the lower prevailing ambient humidity.December, 19681 CENTRIFUGATION ON RAW SUGAR POLARISATION ANALYSIS 777 The lowest value of the curve at about 30 ml is higher by about 0*015O ISS, with respect to the centrifuged control, than that in Trial 1, with respect to the filtered control. This agrees well with the result of Trial 2 for the difference in polarisation between filtered and centrifuged samples. TRIAL A To examine further the filter-paper preferential absorption effect, Trial 3 was repeated on eight raw sugars, except that the filter-papers were pre-conditioned to contain about 12 per cent. of moisture. The average polarisation value compared with the centrifuged control is plotted against the progress of filtration in Fig.4. Fig. 4. Effect of filtrate volume on polarisa- tion at 60 per cent. relative humidity (moisture- conditioned filter-papers used with about 12 per cent. of moisture). A, Experimental curve; and B, curve corrected for evaporation (see Fig. 6) In the early stages, the polarisation increased as filtration progressed until the value of the curve at 40 ml was about the same as the control. The moisture content of the paper was presumably greater than the equilibrium value for the paper in contact with the “normal” sugar solution and, therefore, sucrose instead of water was preferentially absorbed from the solution during the early stages of filtration. From 50 to 80 ml, the curve is similar to that for Trial 3, and suggests that evaporation has become the dominant effect.TRIAL 5- To obtain further insight into the evaporation effect, the “mini-pol” procedure was carried out with six previously clarified raw sugar solutions. However, no filter-papers were used. To be consistent with the other trials, about 90 ml of clarified solution were added to the filter funnel (without a filter-paper) and 80 ml collected in 10-ml portions at the same rate as in Trials 1 and 3. The rate of flow was controlled by replacing the thistle funnel with a short length of small-bore plastic tubing and a screw-clip. In this way the preferential absorption effect by the filter-paper was eliminated. Average polarisation results are plotted against the progress of filtration in Fig. 5. The polarisation of the 70 to 80-ml portion is 0.027” ISS higher than that of the 0 to 10-ml portion, indicating that evaporation causes a significant increase in polarisation during filtra- tion, even when elaborate precautions are taken to prevent it.The prevailing relative humidity during Trial 6 was 60 per cent. DISCUSSION In earlier polarisation work, an accuracy of better than 0-01” ISS could not be expected. However, the use of modern photo-electric polarimeters and refined analytical techniques have enabled smooth filtration graphs to be drawn, in which the points are calculated and778 WILSON et al. : EFFECT OF FILTRATION AND plotted with a precision as high as 0~001" ISS, polarisation differences consequently detected with a range of precision far beyond that previously possible and factors causing small but consistent errors studied in detail for the first time.The factor most likely to affect the polarisation of a sugar solution during filtration is preferential absorption of water or sugar by the filter-paper. The true polarisation value of the solution would then be approached as the volume of filtrate increases. This effect for the first 40 ml of filtrate is shown in Fig. 2. However, the sharp rise in polarisation over the last 30ml of filtrate cannot be explained by preferential absorption, and is almost certainly caused by evaporation. An alternative method for the removal of precipitate from a "leaded" raw sugar solution is centrifugation. Trial 2 shows that filtration gives a higher polarisation value than centri- fugation, presumably as a result of preferential absorption and evaporation, and a comparison of Trials 1 and 3 (Figs.2 and 3) confirms this finding. When the lead precipitate in a raw sugar solution is removed by centrifugation, the effects of preferential absorption and evaporation are eliminated. [Analyst, VOl. 93 Two questions remain to be answered- (i) how much initial filtrate should be discarded to minimise the effect of preferential absorption and to obtain a filtered polarisation result as close as possible to the correct polarisation value ; and (ii) what is the correct value, and whether a centrifuged solution gives, on average, a more accurate estimate than the corresponding filter solution? It is best first to consider the two effects, preferential absorption and evaporation, separately .PREFEREXTIAL ABSORPTION- Karl Fischer moisture determinations have shown that filter-papers used for polarisation analysis have a moisture content of 6 to 8 per cent. when in equilibrium with ambient condi- tions of 20" C and 60 to 70 per cent. relative humidity. Oven-drying methods, although not precise, have confirmed a value of 7 per cent. for moisture. Papers with this moisture content preferentially absorb water from sugar solutions, as shown in Figs. 2 and 3. If the moisture content were increased, we would expect less preferential absorption in the early stages of filtration and, if sufficiently high, a preferential desorption, that is, the equivalent of an absorption of sucrose from the solution and a consequent underestimate of polarisation from the initial filtrate runnings.However, we would not expect the increase in polarisation from 40 to 80ml of filtrate, presumably from evaporation effects, to be more than slightly affected by different initial filter-paper moisture contents. These conclusions are confirmed by the results of Trial 4, shown in Fig. 4, in which papers pre-conditioned to 12 per cent. of moisture were used. Fig. 4 shows a reversed preferential absorption effect, followed by the characteristic increase in polarisation over the last 40 ml of filtrate collected. Therefore, to minimise the effect of preferential absorption, filter-papers should apparently have a moisture content of 9 to 10 per cent., when it should not be necessary, in theory, to discard any of the first runnings of filtrate.This is contrary to some beliefs that an optimum moisture content of 20 per cent. is required. AMOUNT OF FILTRATE TO BE DISCARDED- It is difficult to condition filter-papers to a pre-determined moisture content, and the first filtrate runnings need to be discarded, in any event, as the first few millilitres are some- times slightly cloudy. It is, therefore, most convenient to use papers with the moisture content in equilibrium with ambient conditions (about 7 per cent.). Under these conditions, sufficient of the first runnings must be discarded to ensure a subsequent sparkling, clear filtrate and that the effect of preferential absorption has become small enough to be insignificant. Figs. 1 and 3 indicate that at least 10 ml need to be dis- carded.However, the amount discarded should be kept to a minimum so that the time required for the analysis will be reduced to a minimum; there will be no need to replenish the solution in the filter funnel; and risk of evaporation will be minimised. For example, we see (Fig. 2) that if a 20 to 80-ml portion is collected, rather than a 10 to 70-ml portion, preferential absorption errors will be reduced, but evaporation errors will be increased with respect to the control.December, 19681 CENTRIFUGATION ON RAW SUGAR POLARISATION ANALYSIS 779 At least 50ml of filtrate need to be collected for flushing and filling a 200-mm visual polarimeter tube. As both the preferential absorption and evaporation effects result in an overestimate of polarisation, the 50, 60 or 70-ml portion that gives the lowest polarisation should be collected.In Table I, the average results for individual 10-ml portions over spans of 50, 60,70 and 80 ml are compared for Trial 1 with the filtered control, and for Trial 3 with the centrifuged control $Zus 0.013" ISS. (0*013" ISS is the difference between filtered and centrifuged polarisations determined in Trial 2.) TABLE I COMPARISON OF AVERAGE POLARISATION RESULTS OF SEVERAL 10-ml Polarisation difference (" ISS) ; solution average minus control FRACTIONS WITH CONTROL SAMPLE RESULTS Trial 1 Trial 3 (Fig. 2) (Fig. 3) Filtered control (10 to 80 ml) . . . . o*ooo - Centrifuged control +0.013" ISS . . - 0-000 0 to 60 ml 0.027 0.028 0 to 60 ml 0.023 0,025 0 to 70 ml 0.021 0.023 0 to 80 ml 0.019 0.023 10 to 60 ml 0.002 0.004 10 to 70 ml 0-003 0.006 10 to 80 ml 0.003 0.008 20 to 70 ml 0.001 0.007 20 to 80 ml 0-003 0.009 30 to 80 ml 0-003 0.01 1 The agreement between the 10 to 80-ml portion in Trial 1 and the filtered control is excellent and, in Trial 3 with the centrifuged control, satisfactory.Clearly, if the first 10 ml are not discarded, the polarisation result will be seriously in error (by about 0.02" ISS). However, if additional filtrate is discarded, results will not differ significantly for most applications. We therefore recommend that the first 10 ml be discarded and the filtration continued until 50 to 60 ml have been collected. I I I I I I 0-10 10-20 20-30 30-40 40-50 W-60 60-70 70- 0.0ooI: 1 Progress of filtration - (10-ml portions of filtrate) Fig.5. Evaporation during filtration at 60 per cent. relative humidity (no filter- paper used) EVAPORATION- To study the evaporation effect independently of the preferential absorption effect, rrial5 was conducted with solutions that were "filtered" under specially controlled conditions without filter-papers. There is a steady rise in polarisation throughout which, in the late780 WILSON et al. : EFFECT OF FILTRATION AND stages from 50 to 80 ml, is approximately linear in form, with a slope (Fig. 5) of 0.0060" ISS per ml. The slope of the 50 to 80-ml portion of the curve in Fig. 3 is 0*0045" ISS per ml; this is less steep than the evaporation curve of Fig. 5, presumably because of a small, gradually diminishing effect of preferential absorption.The slope of the 50 to 80-ml portion of the curve in Fig. 4 is, however, 0.0060" ISS per ml, the same as for the evaporation curve in Fig. 5. If the evaporation curve is now subtracted from the experimental curves in Figs. 3 and 4, the resultant curve, corrected for evaporation, represents a plot of preferential absorption alone. The best estimate of polarisation should now be that value which is approached asymptotically as preferential absorption diminishes. For Trial 3 (Fig. 3), the corrected curve is relatively flat and approaches closer to the centrifuged control result in the final stages of the filtration. After 80 ml have been filtered, the corrected curve is only 0.005" ISS higher than the centrifuged control. For Trial 4 (Fig. a), there is a similar effect but, because the preferential absorption effect is reversed, the approach to the centrifuged control is from the low polarisation side; after 80 ml have been filtered, the corrected curve is 0.005" ISS lower than the centrifuged control.This indicates that clarification by centrifugation causes no significant errors in the determination of raw sugar solution polarisations. Centrifugation is, therefore, a more accurate method than filtration and gives a better estimate of the polarisation, as defined and specified by ICUMSA. As mentioned earlier, centrifugation is preferred for large numbers of raw sugar analyses because it is less susceptible to the effects inherent in the filtration procedure; it is also rapid in comparison with filtration and requires less bench space.[Analyst, VOl. 93 HUMIDITY EFFECTS- Apart from the different controls, the curves of Figs. 2 and 3 represent similar trials differing only in humidity conditions; Trial 1 was carried out at 70 per cent. and Trial 3 at 60 per cent. relative humidity. Examination of the curves over the last 30ml of filtrate shows that the experimental curve in Fig. 2 has a slope of 0.0038" ISS per ml compared with 0*0025" ISS per ml for that with 60 per cent. humidity in Fig. 3. Therefore, the evapora- tion effect seems to be only 0.85 times as large at 70 per cent. as at 60 per cent. relative humidity. If the evaporation curve in Fig. 5, with all values reduced by a factor of 0-85, is now subtracted from the experimental curve in Fig. 2, again the preferential absorption effect alone is illustrated.Again the corrected curve approaches the centrifuged result obtained in Trial 2, and plotted in Fig. 2, but here it seems that the effect of evaporation based on the above assumptions has been slightly overestimated. Evaporation at 70 per cent. relative humidity is probably only about 0-8 times that at 60 per cent. The amount of water required to saturate air of 70 per cent. humidity at 20" C is calculated to be about 0.8 times the amount required to saturate air of 60 per cent. humidity. These trials were conducted in a laboratory at 20" C, considerable precautions being taken to prevent evaporation. The over-all effect in Trial 5 for 60 per cent. relative humidity was a polarisation increase of 0.027" ISS from the first 10-ml portion to the last.Under routine conditions, for lower ambient humidities and higher ambient temperatures, the effect is likely to be several times greater. The average effect over a 50 or 60-ml portion in a laboratory in the tropics could easily be 0.05" ISS for a raw sugar, or 0.05 per cent. of the polarisation value for cane juices and sugar products. CONCLUSIONS When a raw sugar solution is defecated with basic lead acetate and filtered through a single filter-paper that has a moisture content in equilibrium with the atmosphere, polarisation is significantly increased as a result of preferential absorption of water by the filter-paper. Although this effect continues until at least 50 ml of filtrate are collected, the effect is negligible for most practical purposes after the first 10 ml are filtered.In accordance, therefore, with the official method of the International Commission for Uniform Methods of Sugar Analysis for raw sugar polarisation, the initial 10ml should be discarded. The optimum moisture content of filter-papers to minimise the preferential absorption effect is 9 to 10 per cent. However, for convenience, papers in equilibrium with the atmosphere (containing 6 to 8 per cent. of moisture) should be used.December, 19681 CENTRIFUGATION ON RAW SUGAR POLARISATION ANALYSIS 781 Even when great care is taken to prevent evaporation, there is a steady increase in the polarisation of small consecutive aliquots of filtrate, apart from the preferential absorption effect. The effect is about 20 per cent. higher for 60 per cent. relative humidity ambient conditions than for 70 per cent., with both at 20” C. For routine analysis under tropical conditions, errors as large as 0.05” ISS could be expected. To minimise this effect, not more than 10 ml of initial filtrate should be discarded, and only as much as is required for the polarisation measurement, usually about 50 to 60 ml, should be collected thereafter. A raw sugar defecated with basic lead acetate solution and clarified by centrifugation, rather than by filtration, is not subject to preferential absorption and evaporation effects. Centrifuged solutions give the better estimate of polarisation value (as defined and specified by ICUMSA) , whereas filtered solutions have polarisation values 0.013” ISS higher. The authors thank the Colonial Sugar Refining Company Limited for permission to publish this paper. REFERENCES 1. 2. 3. 4. Saunier, R., Editor, “Proceedings of the XIVth Session of the International Commission for Uni- Hardin, G. H., and Zerban, F. W., La Plr Sug. Mfv, 1924,73,388; Ind. Engng Chem., 1924,16, 1175. Gross, D., Editor, “Proceedings of the XIth Session of the International Commission for Uniform Saunier, R., Editor, “Proceedings of the XIVth Session of the International Commission for Uni- Received July llth, 1968 form Methods of Sugar Analysis, 1966,” ICUMSA, Subject 6, p. 16. Methods of Sugar Analysis, 1958,” ICUMSA, Subject 21, p. 86. form Methods of Sugar Analysis, 1966,” ICUMSA, Subject 11, p. 61.

ISSN:0003-2654    

DOI:10.1039/AN9689300773

出版商:RSC    

年代:1968

数据来源: RSC

摘要:

782 ABaZyst, December, 1968, Vol. 93, pp. 782-787 Recent Methods for Determining Traces of Nitrogen in (Mineral Oils BY P. GOUVERNEUR AND F. VAN DE CWATS (KoninklijkelSheEl-Laboratorium, A msterdam (Shell Research N . V.)) Three methods for the determination of traces of nitrogen in oil are discussed, viz., those based on extractive percolation, oxy-hydrogen combus- tion and hydrogenation - coulometry. Their scopes are compared with respect to their lower detection limit, the range of products to which they can be applied and their speed. ALTHOUGH organic nitrogen compounds are only minor constituents of mineral oils, they citn play a significant rdle, both during the manufacture and in the performance of oil products. The nitrogen concentrations involved are usually small, most often in the parts per million range, and sometimes below that level. Fortunately, there is a choice of analytical methods available today, thus enabling successful results to be achieved in this concentration range.Three of them are discussed in some detail and their scopes compared with respect to their lower detection limit, the range of products to which they can be applied and their speed. In each of the following methods ammonia is the final reaction product, but the des- tructive treatment and the method of ammonia formation are different. The first method involves extractive percolation of the oil sample through sulphuric acid on a carrier, followed by Kjeldahl digestion of the nitrogen-containing concentrate; the second, oxy-hydrogen com- bustion, whereby the nitrogen compounds of the oil are initially converted into nitrogen oxides, and later into ammonia by Devarda reduction; and the third, catalytic hydrogenation, followed by microcoulometric titration of the ammonia formed. Dumas’ combustion method is not included in the present discussion because of the interference encountered in this method from dissolved molecular nitrogen, the amount of which often exceeds the combined nitrogen fraction by several times. EXTRACTIVE PERCOLATION THROUGH SULPHURIC ACID ON A CARRIER- This is essentially a pre-concentration method and has been found a successful tech- nique for concentrating all of the nitrogen compounds (basic and non-basic) contained in oil fractions.It is based on percolation of the sample through concentrated sulphuric acid distributed on an inert carrier, e.g., fine-grade pumice, in a small column.After passage of the oil, the contents of the column are subjected to Kjeldahl treatment and the ammonia is determined titrimetricallyl or, with small samples, spectrophotometricaUy.2 Since its introduction in 1962, the technique has proved useful for a wide variety of petroleum distillate fractions ranging from low-boiling fractions to lubricating oils with nitrogen levels from 0 to 500 p.p.m. TABLE I EXTRACTIVE PERCOLATION METHOD : NITROGEN CONTENT OF REFINED KEROSENES 2-litre samples; percolation rate 500 ml per hour Nitrogen content, parts per thousand million Sample Natural Added Total present Total found A 19* 1137 132 122; 131 B - 8; 14 C - 31; 28 D - 194; 199 A I \ * Average from multiple determinations by same method.t Nitrogen-containing kerosene. Paper presented at the Joint Symposium on Limits of Detection in Analysis, April 17th and 18th. 0 SAC and the authors. 1968, Enschede.GOUVERNEUR AND VAN DE CRAATS 783 It has been found more recently that this technique can be easily extended to the parts per thousand million range for highly refined oil fractions. Table I shows parts per thousand million amounts of nitrogen in kerosene samples, detected by percolation of 2-litre samples at a rate of 500ml per hour. 0 /El I00 90 Y c 8 80- 5 L Q) Q 70 5 6 0 - f! 2 $ 50- M E 40- Z 30 - - - 0 I I I 20' Qo 70 80 90 IC Sulphuric acid strength, per cent. Fig. 1. Strength of sulphuric acid distributed on pumice versus nitrogen recovery for the extractive percolation method: 0 straight-run heavy gas oil, 200 p.p.m. of nitrogen; 0 catalytically cracked light gas oil, 122 p.p.m.of nitrogen A = Reducing valves B = Flame arrester C = Precision regulating valves D = Control valve and rotameter for primary E = Control valve and rotameter for hydrogen F = Control valve and rotameter for secondary G = Metal pressure safety valves H = Flame arrester J = Burner oxygen oxygen K = Sample feed unit L =Condenser M = Adaptor N = Flow indicator 0 = Scrubber P = Condenser Q = Absorber R = Splash bulb S = Control valve and rotameter for suction T = Water flow meter Fig. 2. Diagram of apparatus for nitrogen determination by oxy-hydrogen combustion784 [Analyst, Vol. 93 The recommended concentration of the sulphuric acid is 98 per cent., although in later work this proved to be less critical than originally believed.Fig. 1 indicates that results are not greatly affected unless the concentration of the sulphuric acid drops below 94 per cent. OXY-HYDROGEN COMBUSTION- Oxy-hydrogen combustion is a well known method for decomposing organic material quickly and effectively, and the Wickbold apparatus developed for this purpose is widely used nowadays for determining traces of sulphur and halogen.5 In this method the sample is fed into an oxy-hydrogen pilot flame and burned with a hot flame in an excess of oxygen. The combustion products are then drawn through a scrubber containing a suitable absorbent for trapping the component to be determined.Combustion rates can be high, e.g., from 1 to 5 ml per minute, depending on the nature of the sample. The determination of nitrogen in this way is a more recent development4 and com- prises the following steps: conversion of the organic nitrogen compounds in the hot flame, producing nitrogen oxides (mainly nitric oxide) ; trapping of the oxides formed on sodium chlorite on an alumina carrier, thereby producing a nitrite - nitrate mixture; and wet reduction of this mixture with Devarda alloy producing ammonia, which can be determined in the usual way. The combustion apparatus used is shown in Fig. 2. A satisfactory sample feed-rate is about 2 ml per minute, and a useful upper limit for the amount of nitrogen to be convertedin one run is 5 mg, preferably contained in 5 to 50 ml of sample.In this way any overloading of the sodium chlorite reagent is prevented. The reduction of the nitrite - nitrate mixture to ammonia and the isolation of the latter is a one-step operation that proceeds simply and conveniently, so that the whole method is relatively quick in comparison with traditional methods for the detection of trace nitrogen. Fifteen determinations can be made in a single working day. GOUVERNEUR AND VAN DE CRAATS: RECENT METHODS FOR TABLE I1 OXY-HYDROGEN METHOD : CONVERSION OF NITROGEN COMPOUNDS All compounds dissolved in benzene - kerosene Compound Nitrobenzene . . .. Aniline . . .. .. Azobenzene . . .. Pyridine . . a . .. Octylamine . . .. Heptyl cyanide . . .. Quinoline .. .. .. Indole . . .. .. Nitrogen theory, p.p.m.23 1 1016 296 246 204 171 203 215 Nitrogen average found, p.p.m. 222 1028 281 99 204 169 200 207 Recovered as ammonia, per cent. 96 101 96 40 100 99 99 96 In model experiments with blends of pure nitrogen compounds in nitrogen-free fuel, the conversions into ammonia were between 95 and 101 per cent., as Table I1 shows. Azobenzene, however, is an exception. This compound evidently splits off an appreciable fraction of its nitrogen as elemental nitrogen, which is only sparingly oxidised in the flame. However, this is of little consequence for the present purpose because tke occurrence of compounds of this type in petroleum has not been reported. TABLE I11 OXY-HYDROGEN METHOD : NITROGEN CONTENT OF OIL FRACTIONS Nitrogen by Nitrogen by extractive percolation, oxy-hydrogen combustion, Sample p.p.m.p.p.m. Naphtha .. .. .. .. .. 97 94; 95 Light gas oil . . .. .. .. 15 16; 13; 16 Heavy gas oil* . . .. .. .. 267 261; 254 Gas oil concentrate* . . .. .. 577 542; 540 Luboil* . . .. .. .. .. 616 618; 613 Dark steam cylinder lubricant* .. 478 470; 496 Shale oil* .. .. .. . . 1.59 per cent. 1.57; 1.59 per cent. * Diluted.December, 19681 DETERMINING TRACES OF NITROGEN IN MINERAL OILS 785 Some results obtained on actual oil samples are presented in Table 111, the oils having nitrogen contents from 15 p.p.m. upwards. In general, satisfactory agreement was found with results obtained by the extractive percolation method, both for low-boiling and for residual oil fractions, and also in the exceptional case of the shale oil sample.It may be concluded, therefore, that the nitrogen compounds of oils are generally converted into nitrogen oxides during the oxy-hydrogen combustion treatment. The conversion of molecular nitrogen, on the other hand, is poor and amounts to only a few per cent. under the flame conditions. In consequence, the effect of the portion of dissolved nitrogen present in the oils is not sig- nificant. Unfortunately the situation is quite different for the gases used for the combustion. Their molecular nitrogen impurity level often ranges from 100 to 1000 p.p.m. v/v and, as the gas consumption is relatively high, the blank contribution is significant. For this reason the method is no longer attractive for oils containing less than 15 p.p.m. of nitrogen. CATALYTIC HYDROGENATION - COULOMETRIC METHOD- This method was published by Martin6 in 1966 and comprises catalytic hydrogenation of the oil sample with nickel catalyst, followed by microcoulometric acid - base titration, in a special cell, of the ammonia formed.An outstanding feature of this method is that it is capable of detecting nanogram amounts of nitrogen in milligram amounts of oil in a short time. Moreover, the use of such small samples and a relatively large excess of nickel catalyst appreciably reduces the effect of nickel de-activation by poisoning and coke deposits (a tradi- tional drawback in Ter Meulen type hydrogenations). Hydrogen r Insert scrubber t Hydrogenation furnace injection Fig. 3. Hydrogenation - coulometric nitrogen method A schematic arrangement is given in Fig. 3.Fig. 4 shows the equipment, which is available from Dohrmann Instruments Company (Mountain View, California). The pro- cedure is as follows: the 1 to 10-pl oil sample is injected into a 400" C zone and subjected to hydrogenation in a current of ultra-pure hydrogen (24 litres per hour) through nickel catalyst. The catalyst supplied by the manufacturer, viz., granular nickel of high purity, is used at 900°C in contrast with Martin's catalyst, which was nickel on magnesium oxide used at 440" C. An insert scrubber tube containing, e.g., magnesium oxide serves to remove inter- fering acidic products that originate from accompanying sulphur and halogens. The am- monia leaving the tube is fed into a titration cell designed to operate at a pre-determined constant hydrogen-ion concentration in the pH range 5 to 6.The cell contains 0.04 per cent. sodium sulphate solution as the electrolyte and has four electrodes, a sensing pair (hydrogen electrode v e m w lead - lead sulphate reference electrode) and a generating pair (two platinum electrodes). Any change in concentration (incoming ammonia) is detected by the sensing electrode pair as a potential difference, which leads through the coulometer amplifier to the generation of hydrogen ions at the generator electrode. The required current is recorded via a precision series resistance on a potentiometric recorder, the peak area representing the current integral with the chart speed as the time basis. The number of coulombs required for neutralisation is then known. Initially, the application of this method to oil samples failed, although ammonia produced by heating weighed sub-micro amounts of ammonium chloride was completely recovered.This suggested that the trouble was in the catalytic hydrogenation step, and the experiencesFig. 4. Hydrogenation - coulometric apparatus To fuce page 7851786 GOUVERNEUR AND VAN DE CRAATS: RECENT METHODS FOR [Arta&St, VOl. 93 of colleagues (E. M. Fredericks and E. D. Peters in a private communication) have confinned it. They suggested a pre-treatment of the nickel catalyst and also ultra-purification of the hydrogen used. Fresh catalyst should be subjected in sit% to an oxidative treatment (pure oxygen at 900" C) and subsequently to a reducing treatment (pure hydrogen at 400" C), with pure helium at the intermediate stages for safety reasons.This treatment was repeated as soon as the catalyst began to show signs of de-activation. The hydrogen used was purified by palladium diffusion with an A.5 Diffusion Unit manufactured by Johnson, Matthey and Co. Ltd. HYDROGENATION - Nitrogen theory, p.p.m. 1000 500 250 125 50 TABLE IV COULOMETRY: OCTYLAMINE I N KEROSENE Nitrogen average Injected, found, Recovery] P1 p.p.m. per cent. 0-5 to 1 667 67 0-5 t o 2 374 75 1 237 95 1 130 104 2 48 96 Table IV illustrates the conversion of octylamine when blended with nitrogen-free kerosene. Best results were obtained for the more dilute blends, and for practical reasons the 50 p.p.m. nitrogen level was selected for further work. Results relating to the conversion of other nitrogen compounds at this level are shown in Table V.Satisfactory results were obtained for the compounds tested, except for azobenzene. Here again, the tendency of TABLE V HYDROGENATION - COULOMETRY : CONVERSION OF NITROGEN COMPOUNDS All compounds dissolved in kerosene; 2 4 samples injected, inlet temperature 400" C Compound Nitrobenzene . . .. Aniline . . .. .. Azobenzene . . .. Pyridine . . .. .. Octylamine . . .. Heptyl cyanide . . .. Quinoline .. .. .. Indole . . .. .. Nitrogen theory, p.p.m. 65 50 50 50 49 55 60 60 Nitrogen average found, p.p.m. Gl 48 50 40 47 55 53 53 Recovered as ammonia, per cent. 94 96 100 80 96 100 106 106 this compound to produce elemental nitrogen when decomposed obviously prevents complete conversion, as in the oxy-hydrogen method.Results for oil samples are shown in Table VI. In general they are in reasonably good agreement with known values, even those for the shale oil sample. The high-boiling dark steam cylinder lubricant gave a low value, however. TABLE VI Inlet temperature 400" C, magnesium oxide scrubber 450" C HYDROGENATION - COULOMETRY: NITROGEN CONTENT OF OIL FRACTIONS Nitrogen by extractive percolation, Sample p.p.m. Naphtha .. .. .. .. 97 Kerosene . . .. .. .. 3.2 Light gas oil* . . .. .. .. 187 Heavy gas oil* .. .. .. 267 Gas oil concentrate* . . .. .. 577 Dark steam cylinder lubricant* . . 478 Shale oil* - . .. .. .. 1.59 per cent. * Diluted in kerosene to a 50 p.p.m. level. Nitrogen by hydrogenation- coulometry, p.p.m. 94; 95 3; 4 165; 180 251; 283 588; 623 334; 392 1.43; 1.69 per cent.December, 19681 DETERMINING TRACES OF NITROGEN IN MINERAL OILS 787 These experiments led us to the conclusion that the hydrogenation - coulometric method is most effective for products that are sufficiently volatile below 400” C, and that it is rapid, the signal being available almost instantly and the calculated result well within 15 minutes, including the necessary dilution.The lower detection limit at present appears to be I p.p.m. Further work will be necessary to extend the method to products boiling well above 400” C. CONCLUSION In Table VII the salient features of the methods discussed have been summarised. TABLE VII SUMMARY OF CHARACTERISTICS OF THREE METHODS FOR TRACE NITROGEN DETERMINATION IN OILS Extractive percolation Useful product range : low boiling ..* . .. + highboiling .. .. .. + - residual . . .. .. .. Useful nitrogen range, p.p.m. . . 0.01 to 600 Sample requirements . . .. . . 10 ml to 2 litres Repeatability, usual . . .. . . 6 per cent. amount Repeatability, best . . .. . . 0.01 p.p.m. Analysis time: Number of analyses per person per present 1 test elapsed . . .. .. 6 hours 8-hour day .. .. .. 3 Oxy-h ydrogen combustion + + + 16 to 1000 6 to 40 ml 6 per cent. amount present 6 p.p.m. 46 minutes 16 Hydrogenation - coulometry + (+I 1 to 60 6 per cent. amount present 1 p.p.m. 16 minutes I 2 to 10 pl 30 The extractive percolation method, although of limited speed, is capable of determining parts per thousand million concentrations of nitrogen in certain refined oil fractions. The oxy-hydrogen combustion method is more rapid, and is characterised by its general applicability to oil products with nitrogen contents of over 15 p.p.m. The “Dohrmann” technique (hydrogenation - coulometry) appears to be the most rapid, and is useful for products that are sufficiently volatile below 400” C and have nitrogen contents of over 1 p.p.m. REFERENCES 1. 2. 3. Gouverneur, P., Analytica Chim. Acta, 1962, 26, 212. Smith, A. J., Cooper, F. F., jun., Rice, J. O., and Shaner, W. C., jun., Ibid., 1968, 40, 341. “Proceedings of the Colloquium on Experiences with the Wickbold Combustion Apparatus for Organic Elementary Analysis, Hanau, February, 1966,’’ Heraeus-Schott Quarzschmelze GmbH, Hanau, Germany. Gouverneur, P., Snoek, 0. I., and Heeringa-Kommer, M., Artalytica Chim. Acla, 1967, 39, 413. Martin, R. L., Analyt. Chem., 1966, 38, 1209. 4. 6. Received May 7th, 1968

ISSN:0003-2654    

DOI:10.1039/AN9689300782

出版商:RSC    

年代:1968

数据来源: RSC

摘要:

788 Arcalyst, December, 1968, Vol. 93, $q5. 788-791 Determination of Chloride in Chloridecontaining Materials with a Chloride Membrane Electrode BY J. C. VAN LOON (Department of Geology, University of Toronto, Toronto 5, Ontario) The determination of chloride by using the membrane-type chloride electrode is described. Procedures are given for the analysis of chloride materials, including silver halides, both in the presence and absence of bromide and iodide. RECENTLY a solid-state silver chloride type membrane electrode has been developed com- mercially for the measurement of chloride activities. This electrode, together with an expanded-scale pH meter, can be used for the rapid determination of chloride concentration in solutions of constant ionic strength. Only the presence of OH-, Br-, I-, S2-, CN-, NH, and S,0,2- in the solutions to be measured interfere directly with the results.1 The avail- ability of this system for rapid studies in the field and laboratory when large numbers of analyses are required makes it potentially useful.Before the commercial development of the silver chloride membrane electrode the use of silver - silver chloride electrodes for the determination of chloride has been recorded for boiler waters,2 chloride in ground water,5 free chloride in the presence of many cations,4 also sweat, urine and miscellaneous solutions.6 This type of electrode developed redox potential errors in the presence of strong oxidising substances, a limitation not shared by the new silver chloride membrane electrode. In this paper procedures are recorded for the rapid determination of chloride in soluble substances and insoluble silver halide materials, both when bromide and iodide are present and in their absence.EXPERIMENTAL APPARATUS- An Orion Specific Ion Chloride Electrode Model 94-17 was used, in conjunction with a Beckman Expandomatic pH meter. A scale expansion of 200-mV full scale was sufficient for these measurements. REAGENTS- All the water used was distilled and checked for the absence of chloride in amounts that would cause interference. Chromic acid sohtiolz-Dissolve 40 g of analytical-reagent grade chromic oxide in 400 ml of sulphuric acid (1 + 2). Standard sodium chloride solution, 0.1 M. Dilute working standards for analysis of sample solutions were prepared by combining all the ingredients used to prepare the sample but substituting an aliquot of standard solution for the chloride sample in the graduated flask.All reagents were tested for the presence of chloride contamination and found to contain insignificant amounts. Solutions were stored in polythene bottles. PROCEDURES FOR SOLUBLE MATERIALS- (a) No bromide or iodide present-Weigh an appropriate amount of sample (final chloride concentration 10-2 to 1 0 - 4 ~ ) into a 250-ml beaker and dissolve it in water. Rinse it com- pletely into a 1000-ml graduated flask and dilute to volume with water. Millivolt readings for these slowly stirred solutions were checked against the appropriate standards (meter was set to read +lorn0 mV for M chloride standard solutions), with the nearest standards being read before and after each sample.0 SAC and the author.VAN LOON 789 (b) Bromide and iodide @resent-Weigh an appropriate amount of sample (final chloride concentration 10-2 to l o - 4 ~ ) into a 250-ml Erlenmeyer flask. Wash the sample into the bottom of the flask with 5 ml of water and tilt it to allow the sample to run into one edge. Add 20 ml of the chromic acid solution and bubble nitrogen into the mixture with a fritted bubbler for 15 minutes. Rinse the bubbler inside and out and remove it from the flask. Wash the solution in the flask carefully into a 500-ml graduated flask and dilute to volume with water. Millivolt readings for these slowly stirred samples were checked against the appropriate standards (meter was set to read +lO.O mV for M chloride solutions), with the nearest standards being read before and after each sample.PROCEDURES FOR SILVER HALIDE- to 10-4 M) gently in 0.7 g of potassium carbonate and a small amount of spectrographic carbon in a platinum crucible until all of the silver, as the metal, has formed a coherent coating on the bottom of the crucible. Place the crucible upright in a 250-ml beaker and cover with 200 ml of water. Place a small stirring rod in the crucible and stir gently to complete dissolu- tion of the melt. Remove and rinse the crucible and stirring rod, then wash the solution from the beaker into a 500-ml graduated flask and dilute to volume with water. With standards prepared as indicated above and the meter set to read +lO.O mV for the M chloride standard solution, the mV readings for the samples were obtained on slowly stirred solutions.The nearest standards are read before and after each sample. (d) Bromide and iodide present-Fuse the sample gently as above, then cool the crucible and add 5 ml of water to the upright crucible on a magnetic stirrer. Stir with a small stirring rod until the dissolution is complete, then rinse the contents of the crucible with a minimum of water (4 to 5 ml) into a 250-ml Erlenmeyer flask. Add 30 ml of the chromic acid solution and support the flask in a tilted position so that the contents run to one side of the flask. Place a bubbler in the solution and bubble nitrogen through the solution for 15 minutes. Rinse the bubbler inside and out and wash the solution from the Erlenmeyer flask into a 500-ml graduated flask. Dilute to volume and measure the millivolt readings of the slowly stirred samples against appropriate standards with the meter set to read +lO.O mV for the M chloride standards.It is important not to allow any metallic silver to contact the concentrated chromic acid solution as silver is easily oxidised to Ag+ in this medium, with the resultant consumption of chloride. The most common causes of inconsistent values when the above methods are used can be summarised as follows: failure to allow sufficient time for the electrode reaction to reach a steady state; the presence of small particles of silver metal in the fusion, which sub- sequently are oxidised by the chromic acid (can be observed because of a resulting cloudy final solution); and failure to check higher and lower standards before and after each sample.RESULTS AND DISCUSSION METER NOISE AND DRIFT- Meter readings were found to fluctuate by * 5 mV, because of the movement of the hands around the chloride electrode. This was corrected by wrapping the electrode and two thirds of its body in three layers of aluminium foil and grounding the foil, somewhere on the lower body, to an outlet. The electrode response to changes in chIoride concentration was not instantaneous and often drifted towards the correct value over an interval of 60 seconds. Response times became even longer after long exposure of the electrode to many samples containing chromic acid solution. This problem was easily rectified by a light buffing of the membrane with fine emery paper.Slow response in the concentration range below 1 0 - 3 ~ could be, to some degree, overcome by stirring slowly with a thin stirring rod. REFERENCE ELECTRODE- It was, of course, impossible to use the normal standard calomel electrode directly in contact with chloride sample solutions, hence a small diameter calomel electrode was inserted in a larger diameter glass casing from a broken reference electrode. The casing with its fibre removed was plugged with a small amount of agar-agar - 4 ~ ammonium nitrate and then filled with 4 M ammonium nitrate. This electrode was clamped to the side of the chloride electrode with the Orion Specific Ion Electrode Holder. (c) No bromide or iodide +resent-Fuse a sample (final chloride concentration The nearest standards are read before and after each sample.790 VAN LOON: DETERMINATION OF CHLORIDE I N CHLORIDE- [Arta&St, VOl. 93 INTERFERENCES- Serious interferences with the use of this electrode would be caused by OH-, S2-, CN-, NH,, S,OS2-, Br- and I-.Of these, Br- and I- are the only ions often encountered in solutions resulting from chloride minerals. In addition, complexing cations, which remove free chloride from solution, could cause serious errors if not kept at an insignificantly low concentration by the use of dilute solutions. An agreement, within the level required by the accuracy of the method, of ionic strengths between samples and standards must be maintained in order to determine chloride concentration directly. The bromide and iodide interference can be eliminated by anion exchange,6 but this method was found to be too slow to be practical in this instance.Evans’ recommended the use of chromic acid in 8 to 9 N sulphuric acid for the separation of bromide from chloride in wide ranges of bromide-to-chloride ratios. It was necessary to prove that the chromic acid system worked for the expected removal of iodide and that it did not cause serious problems with the operation of the chloride electrode. Aliquots containing chloride (so that the final chloride content will be ~ O - * M ) and (or) bromide and (or) iodide, in 1 + 1 + 1 ratios, were treated according to the proposed pro- cedure, to determine firstly if chloride was stable during the oxidation, secondly whether bromide and iodide were eliminated quantitatively by the proposed procedure, and thirdly if chromic acid solution interfered with the operation of the chloride electrode.Tests were also carried out to see if the resultant C9+, produced during the oxidation, interfered in these solutions. Results for these experiments are listed in Table I, and from these it is obvious that the proposed procedure can be used without encountering an error of greater than 10 per cent. Although CI3+ was shown to interfere, its molar concentration is usually less than chloride and at this level does not cause serious error. Test TABLE I TESTS WITH CHROMIC ACID OXIDATION Maximum deviation Number of times from standard, performed mV Stability of chloride in presence of chromic acid mixture (no Br- or I- added) .. .... .. 4 Removal of I- by oxidation . . 3 oxidation . . .. .. . . 5 Removal of Br- and I- together by Possible interference by CP+ : MC1- - MCra+ (1 + 3) . . * . .. 1 MCl- - MCra+ (1 + 10) .. . . 1 MC1- - MCrS+ (1 + 29) .. .. 1 - 0.7 f0-8 & 1.0 + 0-7 +3-1 + 5.0 Maximum error (approximate), per cent. 3 6 10 3 15 23 TEST OF PROPOSED PROCEDURE- Many weighed samples of potassium chloride and sodium chloride, analytical-reagent grade salts, were subjected to the procedures (a) and (b), with aliquots of bromide and iodide solution being added to the sample to test the procedure (b). These results are given in Table 11. Several naturally occurring chloride minerals were analysed by using procedures (a) TABLE I1 ANALYSIS OF “KNOWN” SALTS Weight of samples Procedure taken, Sample used mg KCl .... a 100 NaCl .. .. a 100 b 50.0 b 50.0 d 11 to 22 AgCl .. .. c 18 to 21 Theoretical chloride, per cent. 47.7 47.7 60.7 60.7 24-7 24-7 Mean of chloride found, per cent. 46.5 46.6 60-6 60.1 24.0 23.8 Number of separate determinations 6 4 6 3 4 5 Standard deviation 1.18 1.71 0.56 1.27 0.22 ‘0.66December, 19681 CONTAINING MATERIALS WITH A CHLORIDE MEMBRANE ELECTRODE TABLE I11 ANALYSIS OF CHLORIDE MINERALS 791 Sample Halite (NaC1) . . .. Sylvite (KC1) . . .. Procedure used a b b a b b b U U Carnallite (KMgC1,.6H20) U b b Weight of sample, mg 49-0 47-0 49.0 48.5 50.4 51.8 49.1 48.7 51.3 50.7 50.6 49.0 Theoretical amount of chloride, m€! 29.7 28.5 29.7 29.4 24.0 24.6 23.4 23.2 24.5 19.4 19.2 18.8 Amount of chloride found, Deviation, mg mg 30-5 + 0-8 29-8 + 1.3 28.7 - 1.0 28.4 - 1.0 23-0 - 1.0 23.2 - 1.4 24.1 + 0.7 21.6 - 1.6 25-7 + 1.2 18.4 - 1.0 18.8 - 0.4 18.6 - 0.2 and (b) and the results are listed in Table 111.Several analyses with precipitated silver chloride and silver chloride jh?zls equal amounts of precipitated silver bromide or silver iodide, or both, were subjected to procedures (c) and (d) and the results are listed in Table 11. From these results it is obvious that the proposed procedures can be used for the rapid deter- mination of chloride in the presence of bromide and iodide when the greatest accuracy is not required. The time required for a single analysis is about 20 to 30 minutes with pro- cedures (b) and (a), but this can be lowered to about 8 minutes for each when many samples are analysed by integrating the steps. Procedures (a) and (c) required 2 to 5 minutes for completion. The author thanks the Department of University Affairs of the Province of Ontario for financial assistance for this work, and Dr. J. Mandarin0 of the Royal Ontario Museum for the samples. REFERENCES 1. 2. 3. 4. 6. 6. 7. Instruction Manual, Halide Electrodes, Orion Research Corp., Cambridge, Mass. Seidel, R. L.. U.S. Atomic Energy Corp. KAPL-M-3MS-93, 1958, p. 35. Black, W., J . Amer. Wal. Wks Ass., 1960, 52, 923. Onashi, H., and Morozumi, I., Hokkaido Doigaku Kogat ubu Kenkyu Hokoku, 1967,42, 131. Stem, M., Shwachman, H., Licht, T. S., and deBethune, A. J., Analyt. Chem., 1958, 30, 1506. De Geiso, R. C., Rieman. W., and Lindenbaum, S., Ibid., 1954, 26, 1840. Evans, B. S., Analyst, 1931, 56, 590. For mode 94-17. Received May 7th. 1968

ISSN:0003-2654    

DOI:10.1039/AN9689300788

出版商:RSC    

年代:1968

数据来源: RSC

摘要:

792 Analyst, December, 1968, Vol. 93, $9. 792-796 Oxidimetric Determination of Thiocarbonate Sulphur with Chloramine-T, Potassium Ferricyanide and Potassium Permanganate BY K. N. JOHRI AND N. K. KAUSHIK (Department of Chemistry, University of Delhi, Delhti 7, India) Oxidimetric methods for determining the concentration of aqueous solutions of potassium thiocarbonate by using chloramine-T, potassium ferricyanide and potassium permanganate are discussed. The chloramine-T method is based on the reaction of potassium thiocarbonate with a known excess of chloramine-T in alkaline medium at 6OoC and back-titration of the unreacted chloramine-T against a standardised solution of sodium thiosulphate, with starch as indicator. TwentyLfour equivalents of the oxidant per mole of potassium thiocarbonate are consumed, showing that the three sulphur atoms of thiocarbonate are oxidised to sulphate.When ferri- cyanide is used twenty-four equivalents of the oxidant per mole of potassium thiocarbonate are also consumed in an alkaline medium at 60" C. However, in acidic medium, potassium permanganate is found to oxidise the three sulphur atoms of thiocarbonate to elemental sulphur. The following molar relationships are established- 12 moles of chloramine-T = 1 mole of K&S, = 5 moles of K,CS, 24 moles of K,Fe(CN), = 1 mole of K,CS, 6 moles of KMnO, BECAUSE of the important analytical applications of potassium thiocarbonate reported lately in the literature,ls2Js4s6 an investigation of efficient and rapid chemical procedures for the determination of thiocarbonate sulphur present in aqueous samples of the reagent was considered necessary.A gravimetric methode in which thallium(1) nitrate is used for the quantitative precipitation of thiocarbonate sulphur and titrimetric procedures involving iodimetric, as well as iodatometric, oxidation of the thiocarbonate contents of samples have been rep~rted.~ Iodic acid alone was found to react with the three sulphur atoms of the thiocarbonate; iodine and potassium iodate each oxidised one of the sulphur atoms, the other two forming carbon disulphide. The iodimetric and gravimetric method~8s~~~O available relate only to the determination of sulphide sulphur and not to thiocarbonate sulphur. The results obtained by the present method are not only quantitative, but have confirmed that three moles of hydrogen sulphide per mole of potassium thiocarbonate are available in reacting solutions under optimum conditions.As chloramine-T is able to break the C-S, N-S and S-S bonds in a variety of sulphur compounds,ll to 17 oxidising all of the sulphur quantitatively to the sulphate form, it was of interest to investigate the reaction between chloramine-T and potassium thiocarbonate. The active constituent of chloramine-T is the hypochlorite ion, which is obtained by hydrolysis of chloramine-T. Chloramine-T is preferred to hypochlorite because of its relatively high stability. The oxidation of thiocarbonate sulphur was studied in both acidic and alkaline media, and a simple titrimetric procedure for the determination of potassium thiocarbonate has been developed by making use of the oxidation in alkaline medium.Potassium ferricyanide has also been studied as an oxidant for thiocarbonate sulphur in hot alkaline medium. An investigation of the use of this oxidant was necessary because, being a weak oxidant, it is more selective than others. Moreover, oxidation of various organic and inorganic sulphur compounds in alkaline medium has been reviewed by Sant,l* but no reference to thiocarbonate sulphur in this respect is found in the literature. However, recently Deshmukhlg has standardised potassium thiocarbonate amperometrically by using ferricyanide as oxidant, with osmium tetroxide as catalyst, and has reported that only one third of the thiocarbonate sulphur reacted at room temperature. 0 SAC and the authors.JOHRI AND KAUSHIK 793 Furthermore, evaluation of thiocarbonate sulphur in acidic medium has been carried out with potassium permanganate, a strong oxidant that reacts quantitatively with the hydrogen sulphide liberated by the thiocarbonic acid produced in acidic solutions of potassium t hiocarbonate.CHLORAMINE-T METHOD REAGENTS- Chloramine-T, 0.1 N-This solution was kept in amber-coloured bottles and standardised iodimet rically.20 Sodium thiosulphate, 0.1 N-This was prepared from analytical-reagent grade material and standardised against potassium iodate. Starch solution, 1 $er cent., aqueous. Potassium thiocarbonate, 2 M-An aqueous solution was prepared by the direct method, and, after standardisation,, was used to prepare suitable dilutions.Other reagents used were of analytical-reagent grade. PROCEDURE- Transfer 10 ml of 0.02 M potassium thiocarbonate into a 250-ml conical flask containing 50 ml of standard chloramine-T solution, made alkaline with 5 ml of M sodium hydroxide. Heat to 60" C for half an hour by immersing the flask in a hot water bath. Cool to room temperature, acidify with 20 ml of 5 N sulphuric acid and add 25 ml of 10 per cent. potassium iodide solution. Titrate the liberated iodine with standard sodium thiosulphate solution. The amount of chloramine-T consumed by potassium thiocarbonate is thus obtained from the titre value. Calculate the thiocarbonate sulphur content of the sample from the relationship- Blank corrections were not necessary in these experiments. The titre values were found to be reproducible, and the results of a few representative experiments are given in Table I.1 ml of N chloramine-T = 7-75 x 10-3 g of K2CS3 = 4.008 x g of s. TABLE I RESULTS OF THE DETERMINATION OF SULPHUR IN POTASSIUM THIOCARBONATE WITH CHLORAMINE-T IN ALKALINE MEDIUM AT 60°C Volume of 0.02 M potassium thiocarbonate added, ml 10 10 5 5 3 3 Number of equivalents of oxidant per mole chloramine-T of potassium consumed, thiocarbonate ml used 47-98 23.99 47-84 23-92 24.02 24-02 24-00 24.00 14-38 23.96 14-40 24.00 Volume of 0.1 N Sulphur present, mg 19-20 19.20 9.60 9.60 5.76 5.76 Sulphur found, * g 19-23 19-18 9.63 9.62 5-76 6-77 Difference, *g + 0.03 - 0.02 + 0.03 + 0.02 0.0 + 0.01 DISCUSSION It is evident from the results shown in Table I that the reaction of chloramine-T with alkaline potassium thiocarbonate at 60°C is such that the three sulphur atoms undergo oxidation. For the complete oxidation of sulphur in potassium thiocarbonate to the sulphate ion twenty-four equivalents of oxidant per mole of potassium thiocarbonate would be needed according to the following equations- (;) CH3.C,H4.S02NC1 - Na+ + H20 -+ CH3.C,H4.S02NH2 + Na+ + Ocl-.(ii) CSS2- + 120Cl- + 40H- -+ 3S0,2- + 2HzO + 12C1- + C02. Thus, 12 moles of chloramine-T = 1 mole of K2CS3. Experiments carried out in acidic medium revealed incomplete oxidation, even at a higher temperature. This was caused by the partial oxidation of potassium thiocarbonate to elemental sulphur which, once formed, resists further oxidation by chloramine-T. The794 JOHRI AND KAUSHIK : OXIDIMETRIC DETERMINATION OF [Analyst, Vol.93 sulphur precipitated during these experiments in acidic media was observed to be suspended in solution. Oxidation in alkaline medium at room temperature (20" C) also did not proceed to completion, and separation of elemental sulphur was clearly observed as a white turbidity. However, when the temperature was raised to 60" C this turbidity vanished and the oxidation of the entire sulphur was found to be quantitative. The finely divided sulphur reacts with hot alkali,21 forming sulphide, sulphite and thiosulphate, all of which can be oxidised to sulphate by chloramine-T. FERRICYANIDE METHOD REAGENTS- Potassium ferricyanide, 0.1 this was prepared by dissolving analytical-reagent grade material in redistilled water and standardised by titrating against standard sodium thio- sulphate solution.Sodium thiosulphate, 0.1 N. Starch solution, 1 per cent., aqueous. Sulphuuric acid, 5 N. Zinc sulphate, 0.5 M-This was prepared by dissolving an analytical-reagent grade sample in redistilled water. Potassium thiocarbonate, 2 M. PROCEDURE- In a series of experiments to determine the optimum conditions for the complete oxida- tion of thiocarbonate sulphur with ferricyanide, carried out at room temperature, the time of reaction was varied and the amount of potassium thiocarbonate used kept constant. The excess of ferricyanide was determined by back-titrating against standardised thio- sulphate solution. The reaction was slow at room temperature, the number of equivalents of oxidant per mole of potassium thiocarbonate increasing from 2.60 to 8-06 in 45 minutes at &minute intervals.However, in the second set of experiments the reactants were heated to 60" C and the reaction was found to be completed within 15 to 20 minutes. RECOMMENDED PROCEDURE- Introduce a measured aliquot of the test solution containing not more than 17 mg of sulphur into a measured excess volume of standard 0.1 N ferricyanide, previously made alkaline with 5 N sodium hydroxide so that its alkalinity is about 3 N. Heat to 60" C for 15 to 20 minutes by immersing the flask in a hot water bath. Cool to room temperature, and titrate the excess of ferricyanide against standardised thiosulphate after acidifying with 5 N sulphuric acid, adding 25 ml of 10 per cent.potassium iodide solution and an excess of zinc sulphate solution so that all of the resulting ferrocyanide can be precipitated as zinc ferrocyanide. Calculate the thiocarbonate sulphur content of the sample from the relationship- Use starch as the indicator. 1 ml of N ferricyanide = 7.75 x g of K, CS, = 4.008 x g of s. The results of a few representative experiments are given in Table 11. TABLE I1 RESULTS OF THE DETERMINATION OF SULPHUR IN POTASSIUM THIOCARBONATE WITH FERRICYANIDE IN ALKALINE MEDIUM AT 60°C Volume of 0.06 M potassium Volume of 0.1 N added, consumed, 1 14.38 1 14.40 2 28.76 2 28-78 3 43.20 3 43-22 thiocarbonate f emcyanide ml ml Number of equivalents of oxidant per mole of potassium t hiocarbona te used 23-96 24-00 23-96 23.98 24.00 24-01 Sulphur present, mg 5-76 5-76 11-52 11.52 17.28 17-28 Sulphur found, mg 5-76 5-77 11-52 11.53 17.32 17-33 Difference, mg 0.0 +o*or 0.0 + 0.01 + 0.04 + 0-06December, 19681 THIOCARBONATE SULPHUR WITH CHLORAMINE-T 795 DISCUSSION It was seen from the results that the reaction between ferricyanide and potassium thiocarbonate is time consuming.It takes about 45 minutes to oxidise only one of the three sulphur atoms to sulphate at room temperature, after which the titre value remains constant. However, the reaction at 60°C is rapid, and the three sulphur atoms of the thiocarbonate are oxidised to sulphate, as seen in Table 11. The reaction of ferricyanide with potassium thiocarbonate in alkaline medium at 60°C can be expressed- CS,2- + 24Fe(CN),3- + 260H- -+ 24Fe(CN),4- + 3so,2- + 13H20 + CO Thus, 24 moles of ferricyanide = 1 mole of K2CS,.Before titrating the excess of ferricyanide, sufficient zinc sulphate must be added to precipitate all the ferrocyanide as zinc ferrocyanide. Otherwise, the end-point would not be sharp because of the formation of Prussian blue, the presence of which has been verified by making the solution alkaline (colour fades). POTASSIUM PERMANGANATE METHOD PROCEDURE- Transfer 10ml of 0.1 N potassium permanganate solution to a 250-ml conical fiask and add 10 ml of 0.1 N sulphuric acid. Introduce, gradually, freshly prepared potassium thio- carbonate solution from a microburette with a bent nozzle, keeping the tip of the nozzle beneath the liquid surface. Continue adding until the colour of the potassium permanganate is discharged.Repeat the observations with different amounts of potassium permanganate. Calculate the potassium thiocarbonate sulphur content of the sample solution from the relationship- 1 ml of N KMnO, = 31.006 x lO-,g of K,CS, = 16.03 x 10-3g of s. DISCUSSION The results in Table I11 show that the reaction of potassium thiocarbonate with acidified potassium permanganate is such that the three sulphur atoms of thiocarbonate undergo oxidation to elemental sulphur according to the following equations- MnO,- + 8Hf + 5e- -+ Mn2+ + 4H20 CS,2- + 3(0) -+ C032- + 3s + 6e- k, 6KMn0, + 9H2S04 + 5K2CS, --3 3K2S0, + GMnSO, + 9H20 + 5K,C03 + 15 S Thus, 6 moles of KMnO, = 5 moles of K,CS3. The results of titrimetric evaluation with permanganate are accurate for 0.006 to 0.06 M concentrations of potassium thiocarbonate.The volume of permanganate taken should be such that not more than 8 mg of sulphur are precipitated out after the complete reaction. TABLE I11 RESULTS OF THE DETERMINATION OF THIOCARBONATE SULPHUR WITH POTASSIUM PERMANGANATE IN THE PRESENCE OF 0.05 TO 0.1 N SULPHURIC ACID Volume of 0-05 N potassium permangana te taken, ml 5 5 10 10 15 15 Titre of 0-03 M potassium thiocarbonate. ml 1-38 1.40 2.80 2-82 4.46 4.44 Number of equivalents of oxidant per mole of potassium thiocarbonate used 6.03 5.95 5.95 5.91 5.60 5.63 Sulphur present, mg 4.00 4.00 8-12 8.12 12.18 12-18 Sulphur found, mg 3.97 4.03 8-06 8.12 12-84 12-78 Difference, mg - 0.03 + 0-03 - 0.06 0.0 + 0-66 + 0.60 The fact that the three sulphur atoms of thiocarbonate are oxidised by permanganate and no trace of carbon disulphide separated was confirmed by the negative result of a colori- metric test.22 In this test a drop of the solution should produce a stable, pink colour if acetone796 JOHRI AND KAUSHIK and elemental sulphur are present together with free carbon disulphide. The reaction of permanganate with potassium thiocarbonate in alkaline medium at room temperature was also studied, and it was found that only one sulphur atom of thiocarbonate had undergone oxidation to sulphate, while the other two formed carbon disulphide.Furthermore, results at 60” C are not reproducible. To ensure that the liberated hydrogen sulphide undergoes oxidation, potassium thio- carbonate is added to the permanganate, dropwise, from a burette.CONCLUSIONS The results are quantitative, and suitable dilution of more concentrated solutions of potassium thiocarbonate is necessary to obtain accurate results. Under optimum conditions the three sulphur atoms of the thiocarbonate are oxidised, thus giving a rapid method for determining and identifying thiocarbonate from other sulphur compounds. The authors thank Professor R. P. Mitra, F.N.I., Head of the Department of Chemistry, University of Delhi, for the facilities provided, and one of us (N.K.K.) is also grateful to the Council of Scientific and Industrial Research, New Delhi, India, for the award of a Research Fellowship. REFERENCES 1. 2. 3. 4. 6. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. Johri, K. N., J .Scient. Ind. Res., 1969, 18B, 430. -, J . Indian Chem. Soc., 1961, 38, 935. -, Analyst, 1961, 86, 487. -, Indaan J . Appl. Chem., 1963, 26, 114. -, “Chemical Analysis without Hydrogen Sulphide using Potassium Trithiocarbonate,” Asia Johri, K. N., and Singh, K., Indian J . Chem., 1965, 3, 158. -- , Analyst, 1965, 90, 746. Jacobs, M. B., “The Analytical Chemistry of Industrial Poisons, Hazards and Solvents,” Second Edition, Interscience Publishers Inc., New York and London, 1949, p. 323. Scott, W. W., “Standard Methods of Chemical Analysis,” Fifth Edition, D. Van Nostrand, New York, 1939, Volume 1, pp. 911 and 923. Duval, C., “Inorganic Thermogravimetric Analysis,” Elsevier Publishing Co. Ltd., Amsterdam, London and New York, 1953, p. 134. Murthy, A. R. V., Curr. Sci., 1953, 22, 342. Nair, C. G. R., and Murthy, A. R. V., J . Scient. Ind. Res., 1962, 21B, 146. 3 , Chem. & Ind., 1962, 1539. -- , Mh. Chem., 1963, 94, 134. Rao,’V. R. S., and Murthy, A. R. V., Talanta, 1960, 4, 206. Aravamudan, G., and Rao, V. R. S., Ibid., 1964, 11, 55. Jose Jacob, T., and Nair, C. G. R., Ibid., 1966, 13, 154. Sant, B. R., and Sant, S. B., Ibid., 1960, 3, 261. Deshmukh, G. S., and Garde, P., Bull. Chem. SOC. Japan, 1967, 40, 1643. Bishop, E., and Jennings, V. J., Talanta, 1968, 1, 197. Nair, C. G. R., and Murthy, A. R. V., Proc. Indian Acad. Sci., 1962, 55A, 168. Publishing House, Bombay 1, 1963. -- Urbanski, T., Talanta, 1962, 9, 799. k.K Received March W, 1968

ISSN:0003-2654    

DOI:10.1039/AN9689300792

出版商:RSC    

年代:1968

数据来源: RSC

摘要:

AnaZyst, December, 1968, Vol. 93, f@. 797-798 797 The Determination of Molybdenum in Mixtures Containing Molybdenum Disulphide by Atomic-absorption Spectrophotometry BY R. J. JULIETTI AND J. A. E. WILKINSON (Morganite Research and Development L t d . , Battersea Church Road, Battersea, London, S. W. 11) A rapid method is described for determining the total molybdenum content of mixtures containing molybdenum disulphide, graphite and a resin. The novel feature of the technique is the decomposition of the mixture by fusion with sodium hydroxide. Dissolution of the melt in sulphuric acid permits the determination of molybdenum by atomic-absorption spectro- photometry without interference. Results at the 8 per cent. molybdenum level are accurate to within &0.1 per cent. MOLYBDENUM disulphide is finding increasing use in industry as a solid lubricant. It is widely used as an additive to oils and greases, particularly for high temperature lubrication, in the form of a dispersion for “dry-film” impregnation, and in bearing components and electrical contacts.Rapid methods for its determination at the percentage level are, there- fore, of value. EXPERIMENTAL Molybdenum disulphide is particularly difficult to decompose ; the usual methods involve dissolution in strong acids, e.g., as in Defence Specification DEF-2304,l in which fuming perchloric acid is used. In this laboratory we have also used mixtures of nitric and hydro- bromic acids and of nitric and sulphuric acids for dissolution. A combustion methodof decomposition has also been reported.2 The samples normally received for analysis contain molybdenum disulphide, graphite and a resin, and the determination of molybdenum is sufficient to define the molybdenum disulphide content.In earlier methods if the sample contained a resin, it was decomposed by heating at 450” C, a t which temperature there is no loss of molybdenum. The material was then heated with a mixture of acids, usually containing nitric acid to accelerate the attack. After filtration and removal of all nitrate by fuming with concentrated sulphuric acid, the solution was passed through a Jones’ reductor and titrated in the usual way. This method, although sufficiently accurate for process control analysis (k0.1 per cent.), was time con- suming. The rapid finish afforded by atomic-absorption spectrophotometry suggested a means of shortening the procedure.If an acid decomposition were used, the interference from sulphate and nitrate could be overcome as suggested by David.3 However, we found that decomposition by fusion with sodium hydroxide is much more rapid than by any other method. The molybdenum disulphide dissolves at just above the fusion temperature of the alkali, and further, by heating for a little longer at a higher tempera- ture, the bulk of the graphite can also be decomposed. The fusion was carried out in a nickel crucible and the melt dissolved in a standard amount of dilute sulphuric acid (1 + 1). No interference was found from either sodium or nickel. Interference from the variable amount of sulphate resulting from the oxidation of molybdenum disulphide is swamped by the presence of the large excess of sulphuric acid.The following procedure has been found satisfactory. PROCEDURE- 50mm diameter) and fuse over a bunsen burner. walls with a layer of flux. Weigh 5 g of sodium hydroxide pellets into a nickel crucible (about 40mm tall and On cooling, swirl the crucible to coat the 0 SAC and the authors.798 JULIETTI AND WILKINSON Weigh 0.5 g of sample, containing about 8 per cent. of molybdenum, and spread it over the top of the melt by tapping. Fuse gently, without a lid, over a small bunsen flame until the effervescence subsides (about 15 minutes). Do not swirl. Transfer the crucible to a muffle furnace at 650” C and leave for 30 minutes. Allow to cool, place the crucible upright in a 250-ml beaker containing 50 ml of distilled water and cover the beaker with a watch-glass. Cautiously add 20 ml of dilute sulphuric acid (1 + 1)* to the crucible, with the watch-glass slightly displaced.Heat, if necessary, to complete the dissolution of the melt. Remove and wash the crucible, collecting the washings in the beaker, and boil the solution for 5 minutes to remove any traces of hydrogen sulphide. Allow the solution to cool and filter it through a No. 40 filter-paper into a 250-ml cali- brated flask. Dilute to the mark and compare the absorbance with that of the standard solution by atomic-absorption spectrophotometry. STANDARD SOLUTION OF MOLYBDENUM- Weigh accurately about 0.6 g of analytical-reagent grade molybdenum trioxide, previously dried for 1 hour at 150” C.Dissolve it in 50 ml of 10 per cent. w/v sodium hydroxide solution. Add 20 ml of dilute sulphuric acid (1 + 1). Cool and make up to 250 ml with distilled water. This solution is stable for at least 1 month. ATOMIC-ABSORPTION MEASUREMENTS- A Techtron A.A.4 atomic-absorption spectrophotometer was used, and the conditions were as follows: wavelength, 313.3 nm; slit width, 50 pm; flame, air - acetylene; burner, 50-mm slot A.B.40, high temperature burner, which was found to give much less variation in absorbance readings for molybdenum than the 100-mm, A.B.41, burner supplied as a standard fitting with the instrument; air pressure, 1 bar (15 p s i . ) ; acetylene flow, approximately on position 4 of the flow meter, but adjusted to give maximum absorbance when a solution containing molybdenum is aspirated; height of light path above burner, about lOmm, but also adjusted for maximum absorbance after the flame conditions have been set ; and aspiration rate, about 2 ml per minute.RESULTS Two samples were analysed for molybdenum by the above procedure. Fourteen results on the first sample, which contained a resin, had a mean value of 7-71 per cent. of molyb- denum, compared with values of 7-79 and 7.84 per cent. by our previous chemical method. The mean value of eight results on the second sample (not containing a resin) was 9.11 per cent. Values of 9.08 and 9.13 per cent. of molybdenum were obtained on this sample by the gravimetric benzoin a-oxime method. The coefficient of variation on the two sets of results by the atomic-absorption procedure described was less than 1 per cent. in each instance. CONCLUSION This method has been used successfully in this laboratory and, compared with our previous method, has the following advantages. (i) There is no need for a preliminary decomposition of the resin. (ii) The dissolution of molybdenum disulphide is more rapid. (iii) No nitrate ions are introduced. (iv) The total working time is much shorter. The authors thank the directors of Morganite Research and Development Limited for permission to publish this paper. REFERENCES 1. 2. 3. Received Jzrly 4th, 1968 * This reagent should be prepared and added as accurately as possible with a measuring cylinder, as Defence Specification DEF-2304, January 1966, “Molybdenum Disulphide (Powdered) ” ZX-35, H.M. Stationery Office, London, 1966. Kalnin, I. L., Analyf. Chem., 1964, 36, 886. David, D. J.. Analyst, 1968, 93, 79. the amount of sulphuric acid in the standard and sample solutions must be the same.

ISSN:0003-2654    

DOI:10.1039/AN9689300797

出版商:RSC    

年代:1968

数据来源: RSC

摘要:

Analyst, December, 1968, Vol. 93, pp. 799-801 799 Determination of Antimony in Titanium Dioxide by Atomic-absorption Spectrophotometry BY J, C. MERANGER AND E. SOMERS (Research Laboratories, Food and Drug Directorate, Department of National Health and Welfare, Ottawa, Canada) A method is described for the determination of the antimony content of titanium dioxide by atomic-absorption spectrophotometry, based on the extraction of antimony with isobutyl methyl ketone. This method is more sensitive than determination by emission-spectrographic and as sensitive as the colorimetric technique, and can detect 25 p.p.m. of antimony in a 0-1-g sample of titanium dioxide. THE titanium dioxide used in pharmaceutical preparations, and as a food colour, may be contaminated with antimony, and there are several official specifications that give limits for the total, or acid-extractable, antimony.Spectrophotometric determination of the antimony in titanium dioxide with Brilliant green has been described by Galliford and Yardley,l and Ratcliffe and Stevens2; the latter workers showed that many of the pharmaceutical samples of titanium dioxide examined exceeded the British Pharmaceutical Codex limits for antimony. Atomic-absorption spectrophotometry was used by Mostyn and Cunningham3 for the determination of antimony in non-ferrous alloys and, in view of the convenience and sensitivity of this technique, we have applied it to the determination of antimony in titanium dioxide. The atomic-absorption analysis has been compared with colorimetric and spectro- graphic methods, and we have determined the effectiveness of dilute hydrochloric acid in extracting antimony from titanium dioxide.EXPERIMENTAL SPECTROGRAPHIC AKALYSIS- A Jarrell-Ash, Model 15000 Lpl, 1*5-m, wide-angle, grating emission spectrograph was used. Titanium dioxide samples (10 mg) were mixed with 30 mg of spectrographic-grade graphite and introduced into 3.2-mm electrodes with 2 x 15-mm craters, with a 3.2-mm counter electrode. The samples were arced for 25 seconds at 4 and 6A to completion; the resulting spectra were recorded on Kodak SA-1 film. The intensity of the antimony 259.8 nm line in the unknown samples was compared with that obtained from mixtures of spectro- scopically pure antimony tetroxide and titanium dioxide. FUSION OF TITANIUM DIOXIDE- The method followed was that of the Food and Agricultural Organisation Specifications for Identity and Purity of Food Additives4 Titanium dioxide (0.1 g) was fused with 5 g of potassium hydrogen sulphate and 10mg of glucose in 30-ml Kjeldahl flasks.The melt was dissolved by heating with 10 ml of sulphuric acid (96 to 98 per cent. w/v) then, when cool, made up to 100 ml with 30 per cent. w/v hydrochloric acid. COLORIMETRIC DETERMINATION- Aliquots of the above solution, containing from 1 to 1Opg of antimony, were oxidised with solid cerium(1V) sulphate, extracted with di-isopropyl ether, and the optical density of the complex formed with Rhodamine B measured at 555 nm.4 0 SAC and the authors.800 MERANGER AND SOMERS: DETERMINATION OF ANTIMONY IN TITANIUM [Artahst, VOl. 93 ATOMIC ABSORPTION- A Perkin-Elmer, Model 303, atomic-absorption spectrophotometer, equipped with a Boling burner and a null recorder read-out accessory coupled to a Westronics, Model S11A/U, 11-inch strip-chart recorder, was used. The operating parameters were : acetylene flow- rate, 6; compressed air flow-rate, 13; solution uptake, 3.2 ml per minute; slit, position 4; wavelength, 217.6 nm; scale expansion, x 3 ; meter response, 4; and antimony hollow- cathode lamp, current 30mA.STANDARD SOLUTIONS- Standards of 1, 10 and 100 pg of antimony per ml were freshly prepared by dissolving analytical-reagent grade potassium antimony tartrate, KSbC,H,O,. 1/2H,O, in hydrochloric acid (30 per cent. w/v). EXTRACTION WITH HYDROCHLORIC ACID- The extraction method is that given in the U.S.Food and Drug Administration Color Additive regulation^.^ Titanium dioxide (10 g) was boiled for 15 minutes in 50 ml of 0.5 N hydrochloric acid, then samples of the solution were filtered (Whatman No. 42) and made up to 100ml. RE s ULTS SOLVENT EXTRACTION OF ANTIMONY- When the acidic solutions from the potassium hydrogen sulphate fusion of titanium dioxide were nebulised in the atomic-absorption spectrophotometer high results were obtained, probably because of light-scattering by salt particles. In addition, this method was not sensitive enough to determine the low levels of antimony present in food-grade titanium dioxide. To overcome these inadequacies, an extraction procedure was devised. The extrac- tion of the ammonium antimony pyrrolidinedithiocarbamate complex* was attempted with isobutyl methyl ketone.However, the chelate was not formed under the conditions of high acidity used in this work. Chelation was subsequently found to be unnecessary as isobutyl methyl ketone quantitatively extracted the antimony( 111) or (V) directly from the aqueous acidic solutions. Consistent extraction of a standard antimony solution was obtained from solutions of hydrochloric and sulphuric acids in the acid concentration range of that of the dissolved, fused titanium dioxide (Table I). TABLE I WITH ISOBUTYL METHYL KETONE (10 ml), BY ATOMIC ABSORPTION EFFECT OF ACID CONCENTRATION ON THE EXTRACTION OF 50p.g OF ANTIMONY Hydrochloric acid, Sulphuric acid, per cent. w/v per cent. w/v & - 10 12 14 6 13 Optical density* .. . . 0.042 0.041 0.041 0.041 0.042 0.043 * Average of duplicate determinations. ATOMIC-ABSORPTION DETERMINATION OF EXTRACTED ANTIMONY- Aliquots (10 ml) of the acidic solution from fused titanium dioxide were transferred, by pipette, into a 125-ml separating funnel, 5 ml of water and 1 O m l of isobutyl methyl ketone added and the funnel shaken vigorously for 1 minute. After the layers had separated, the aqueous layer was drawn off and the organic solvent directly aspirated in the atomic- absorption spectrophotometer. For samples of titanium dioxide with less than 100 pg of antimony per g, the 10-ml aliquot can be increased to 40m1, but 20ml of water, instead of 5m1, must then be added. Two calibration graphs were prepared by following the above procedure with antimony standards over the range 10 to 100pg and 0 to 1Opg.One millilitre of sulphuric acid (98 per cent. w/v) was added to the standards and they were diluted to 10 ml with 30 per cent. w/v hydrochloric acid. Blank values were obtained from the acidic solutions alone. A linear relationship between optical density and antimony concentration was found over the range 0.1 to 10 pg per ml, with a lower detection limit of 0.1 pg of antimony per ml.December, 19681 DIOXIDE BY ATOMIC-ABSORPTION SPECTROPHOTOMETRY 801 When several samples are analysed it is desirable to have a stock solution of antimony (5 pg per ml) extracted into isobutyl methyl ketone. This solution, appropriately diluted, can be nebulised in the spectrophotometer at fixed intervals, and so slight variations in sensitivity can be normalised.The analysis of a series of synthetic mixtures of titanium dioxide and standard antimony solutions (Table 11) showed good recoveries of antimony over the range 250 to 10,OOO p.p.m. TABLE I1 ANTIMONY RECOVERIES FROM 0.1 g OF TITANIUM DIOXIDE Antimony added, pg Antimony recovered, pg* Percentage recovery 25 27 108 60 62 104 100 105 105 400 438 110 1000 905 91 * Average of duplicate determinations. COMPARISON OF ANALYTICAL METHODS- The antimony content of three commercial samples of titanium dioxide (anatase form) was detennined by emission-spectrographic, colorimetric and solvent-extraction atomic- absorption methods. The samples were chosen to represent a range of antimony content, sample A being a high quality, food-additive grade.Table I11 shows that atomic-absorption, combined with extraction, provides a method of analysis that is both consistent with, and as sensitive as, the other two methods. The lower limit of detection of antimony in a 0-1-g sample of titanium dioxide is 25 p.p.m. Extraction with hydrochloric acid removed only a very small proportion of the total antimony. TABLE I11 COMPARISON OF METHODS OF ANALYSIS FOR THE DETERMINATION OF THE ANTIMONY CONTENT OF TITANIUM DIOXIDE, AS P.P.M. Extraction with 0.6 N hydrochloric acid r Atomic Atomii Sample Spectrographic Colorimetric absorption Colorimetric absorption* A < 100 < 25 < 25 (0.1 <Om7 B loot 100 120 <om1 (0.7 C 12,600t 11,300 11,000 4.9 4.6 All values are averages of duplicate determinations.* Direct aspiration of aqueous solution, i.e., not extracted with isobutyl methyl ketone. t &lo per cent. CONCLUSIONS Until recently, the methods available for the determination of low concentrations of antimony in titanium dioxide were erratic and imprecise.lS2 Atomic-absorption spectroscopy is a rapid, accurate and interference-free analytical technique, and the isobutyl methyl ketone extraction method we have adopted combines the advantages of atomic absorption, together with sensitivity equal to the existing colorimetric methods. Acid extraction of antimony is used as the basis for official specifications of titanium dioxide.4~6 Whatever the toxicological significance of the antimony extracted by hydro- chloric acid may be, it is clear that this technique gives no useful information as to the total antimony content of the sample. We are grateful to Mr. R. E. Horton and Mr. C. C. Durham of the Department of Energy, Mines and Resources, Ottawa, for the analyses by emission spectrography. REFERENCES 1. Galliford, D. J. B., and Yardley, J. T., Analyst, 1963, 88, 653. 2. Ratcliffe, R. J. M., and Stevens, S. G. E., Mfg Ckenz. Aerosol News, 1964, 35, (9), 83. 3. Mostyn, R. A., and Cunningham, A. F., Amzlyt. Chem.. 1967, 39, 433. 4. “Specifications for Identity and Purity of Food Additives,” Food and Agricultural Organisation, 5. Fed1 Register, 1966, 31, 1065. Received July 5th, 1968 Rome, 1963, Volume 2, p. 26.

ISSN:0003-2654    

DOI:10.1039/AN9689300799

出版商:RSC    

年代:1968

数据来源: RSC

摘要:

802 Afialyst, December, 1968, Vol. 93, $9. 802-804 The Determination of Zirconium in Mineral Rutile with Alizarin Red S BY N. B. STANTON ( A ssociated Minerals Consolidated Limited, Southport, Queensland, A ustralia) A colorimetric procedure has been developed for the determination of the zirconium oxide content of rutile. A fusion with potassium hydrogen difluoride engbles the rutile to be brought into solution rapidly and eliminates the task of filtering off silica. The zirconium is separated from the titanium by a phosphate precipitation, but the interference of tin inherent in the method involving phosphate precipitation followed by ignition to oxide, is eliminated by dissolving the zirconium phosphate and applying a colorimetric finish, with Alizarin red S. Calibration is effected by making additions of zircon to titanium oxide and carrying each through the entire procedure, thus eliminating reagent-blank difficulties. It has been used in the range of 0 to 1 per cent.of zirconium oxide. Good agreement with independent determinations by X-ray fluorescence has been obtained. THE main constituents of the mineral sands of Australian east coast beaches are zircon, rutile and ilmenite. The guarantees under which these products are sold include the maximum amounts of each present as impurities in the others. A need, therefore, exists for a method for determining the zirconium oxide content of rutile. From this content the percentage of zircon, which is accepted as the sole contributor to the zirconium oxide content, can be inferred.The guarantee at the present time requires a maximum of 1 per cent. of zirconium oxide. Methods previously used for this purpose were variants of two main procedures. The first was the precipitation of zirconium with diammonium hydrogen orthophosphate. As pointed out by Wood and McKenna,l tin interferes in this method. Rutile commonly contains about 0.1 per cent. of tin oxide, and it has been shown in this laboratory that tin oxide added to rutile can be recovered by phosphate precipitation. A cassiterite concentrate that had previously been analysed for tin oxide was used for additions to a rutile sample, which was then fused with potassium hydrogen difluoride and subjected to a phosphate precipitation technique to be described, followed by ignition of the phosphate precipitate. The weight of contained tin oxide added was 6.2 mg.The difference in weights between the oxide obtained from the rutile and from the “spiked” rutile was 5.2 mg. The method is, therefore, not satisfactory for this purpose. The second method involves precipitation with mandelic acid. This approach is tedious and time consuming and, further, is not suited to the determination of amounts at the lower end of the range encountered, e.g., 0.3 per cent. of zirconium oxide. It appears that a certain amount must be present before the precipitate can be collected. Mayer and Bradshaw2 first put forward the use of Alizarin red S as a reagent for the determination of zirconium, with reference to magnesium alloys. Snell3 states that titanium interferes in the determination of zirconium with Alizarin red S, and Wood and McKennal correct for a small amount of interference by their method of calibration.An attempt by the author to determine zirconium in rutile directly was unsuccessful because of the difficulty of controlling the hydrochloric acid concentration of the dissolved melt, and the added disadvantage of the presence of sulphate ions. The approach adopted was that suggested by Vinogradov and Ryabchikov.* The titanium was first separated by precipitation of the zirconium with phosphate, the precipitate re-dissolved and a colorimetric finish applied. However, the technique used by these authors for re-dissolving the precipitate on the paper with 5 per cent. oxalic acid was not successful under the conditions experienced, and this approach was modified.The method used for colour formation was that recommended by Wood and McKennal and the U.S. Atomic Energy Commission.6 0 SAC and the author.STANTON 803 EXPERIMENTAL Calibration was carried out by weighing about 1-5,4*5, 7-5,lO and 15 mg of finely ground zircon consecutively into five platinum crucibles, each containing 6 g of potassium hydrogen difluoride and 1 g of Johnson, Matthey “Specpure” titanium dioxide. These additions correspond to 0.10, 0.30, 0.50, 0.675 and 1.0 per cent. of zirconium dioxide, respectively. Each was then taken through the full procedure to be described. This approach was adopted for two reasons, Firstly, Wood and McKennal did not use zirconium nitrate solution because the reagent contains more than the theoretical amount of zirconium, resulting from gradual decomposition of the zirconium nitrate; instead, they used metallic zirconium.It was, however, possible to obtain a pure grade of zircon of known zirconium content for this work. Secondly, the problem of adjusting for the reagent blank is overcome as it is automatically incorporated in the calibration. Any approach in which an attempt is made to subtract a blank reading, obtained by taking the reagents alone through the procedure, from an observed reading is not considered valid. Anomalous enhancement and depression effects on colour development can occur. Wood and McKennal refer to the enhancement effect of titanium on zirconium solutions, as compared with the effect on the reagent blank.The author has made a comparison in some instances of the colour development obtained by adding a solution of the element, and the colour reagent, to distilled water with that obtained by taking the element through the entire procedure and subtracting a reagent blank. An enhancement effect occurs with the former. For these reasons it is desirable to simulate the actual method in its entirety when carrying out the calibration. COLOUR DEVELOPMENT- Wood and McKenna heated the solution for 5 minutes a t 70” to 80°C. Vinogradov and Ryabchikov advocate bringing it to the boil. It was found that neither approach, as applied to our problem, guaranteed full colour development, and prolonged boiling, as recommended by the U.S. Atomic Energy Commission,5 was adopted successfully.Double filtration of the Alizarin red S solution, as recommended by Wood and McKenna, is considered essential. REAGENTS- Alizarin red S solution, 0-1 per cent.l-Dissolve 1.0 g of Alizarin red S in about 300 ml of hot water, boil and then filter the solution through a pad of paper pulp. Dilute the solution to 1 litre and again filter through a pad of paper pulp. PROCEDURE- Fuse 1 g of previously ground sample with 6 g of potassium hydrogen difluoride in a platinum crucible. Cool the melt, transfer it to a platinum basin and add 60 ml of sulphuric acid (1 + 1). Heat gently until copious fumes of sulphuric acid are evolved. This removes the silica, and all of the titanium and zirconium is brought into solution. Cool the solution, dilute to 450 ml, and transfer it to a 600-ml beaker.Add 25 ml of 30 per cent. hydrogen peroxide and 25ml of 20 per cent. dibasic ammonium hydrogen orthophosphate solution; allow to stand overnight in a warm place. Filter the solution through a Whatman No. 31 paper and wash with 5 per cent. sulphuric acid, to which a little hydrogen peroxide has been added, until no colour remains on the paper. Continue washing (6 or 7 times) with 5 per cent. ammonium nitrate solution. Wash the precipitate from the paper into a 250-ml beaker with hot water, then add 10 ml of 10 per cent. sodium hydroxide solution. Bring the mixture to the boil and allow the precipitate to coagulate for 20 minutes in a steam-bath. Filter on a Whatman No. 41 filter-paper and wash the precipitate with 5 per cent. ammonium nitrate solution.Wash the precipitate from the paper into the same 250-ml beaker with hot distilled water and add 25 ml of 5 N hydrochloric acid; the volume should now be about 100 ml. Boil until the volume is reduced to 40 ml, cool, make up to 100 ml and filter through a Whatman No. 41 filter-paper. To a 10-ml aliquot of the filtrate, add 2.5 ml of 5 N hydrochloric acid. Add 5 ml of 0.1 per cent. Alizarin red S and make the solution up to 50ml. After allowing it to stand for 1 hour, measure the absorption a t 560nm. The instrument used was a Bausch and Lomb Spec- tronic 20. METHOD804 STANTON COMPARISON WITH X-RAY FLUORESCENCE DETERMINATION- The accuracy of the method has been demonstrated by comparison of the results with those obtained by X-ray fluorescence determination, in which the method of additions, as described by Birks,6 was used, background correction being allowed for.Results are shown in Table I. TABLE I DETERMINATION OF ZIRCONIUM OXIDE IN RUTILE BY CHEMICAL AND X-RAY FLUORESCENCE METHODS Zirconium oxide (ZrO,), per cent., by- Sample No. Chimica1 method X-ray fluorescence mithod 1 0.27 0-28 2 0-36 0-37 3 0.38 0-40 4 0-46 0.43 5 0.5 1 0.50 6 1.10 1.06 CONCLUSION With this method a single determination can be carried out in 24 hours. It requires a slightly longer time than the phosphate precipitation method, but eliminates the error caused by tin oxide inherent in the latter. The calibration method initially is time consuming, but no further calibration is required for a given set of reagents. The assistance of Miss Rose Thomas in carrying out some of the chemical determinations is acknowledged with thanks. Acknowledgment is also accorded to the Directors and Management of Associated Minerals Consolidated Limited for permission to publish this paper. REFERENCES 1. 2. 3. 4. 5. 6. Wood, D. F., and McKenna, R. H., Analyst, 1962, 87, 880. Mayer, A., and Bradshaw, G., Ibid., 1952, 77, 476. Snell, F. D., and Snell, C. T., “Colorimetric Methods of Analysis,” Volume IV, Organic, 11, Third Edition, D. Van Nostrand Co. Inc., New York; Macmillan & Co. Ltd., London, 1954, p. 447. Vinogradov, A. P., and Ryabchikov, D. I., Edztors, “Detection and Analysis of Rare Elements,” National Science Foundation, Washington, 1962, p. 366. “Manual of Special Materials Analytical Laboratory Procedures,” U. S. Atomic Energy Commission Report ANL-5410, 1955, p. 49. Birks, L. S., “X-Ray Spectrochemical Analysis, ” Interscience Publishers Inc., New York, 1959, p. 67. Received March 25th, 1965 Amended J d y loth, 1967

ISSN:0003-2654    

DOI:10.1039/AN9689300802

出版商:RSC    

年代:1968

数据来源: RSC

摘要:

AIzaZyst, December, 1968, Vol. 93, $9. 805-809 805 The Effect of Ethanol on the Colorimetric Determination of Formaldehyde and Glycollic Acid BY R. H. STILL, K. WILSON AND B. W. J. LYNCH (Department of Chemistry and Biology, Hatfield College of Technology, Hatfield, Hertfordshire) Contrary to previous reports ethanol has been found to interfere with the colorimetric determination of formaldehyde and glycollic acid with 1,8-dihydroxynaphthalene-3,6-disulphonic acid (chromotropic acid) or 2,7-di- hydroxynaphthalene in concentrated sulphuric acid. The effect is greater for glycollic acid than for formaldehyde. The mechanism for these observations is given. DURING the study of the rheological properties of some sodium carboxymethylcelluloses a routine method for the determination of the sodium glycollate impurity was required.The standard procedure, developed by Eastenvood,l involves the preliminary removal of the salt impurities by washing with hot (50" to 60" C), 80 per cent. aqueous ethanol until the wash liquor no longer gives a positive test for chloride ions. The glycollic acid content of these ethanolic wash liquors is then determined by a colorimetric method based on the conversion of glycollic acid into formaldehyde, and its determination with chromotropic acid. Optical density measurements are made at 570nm, and are related to concentration by using a calibration graph prepared for a series of standard aqueous glycollic acid solutions. The use of standard aqueous solutions of glycollic acid is based on previous observations by Bricker and Johnson2 that ethanol and methanol do not interfere with the colour reaction in the determination of formaldehyde.In a later paper, Bricker and Vai13 modified their procedure to remove volatile organic impurities after adding chromotropic acid but before adding concentrated sulphuric acid. Even so, they were still of the opinion that these two specific alcohols did not, in fact, interfere with the colour development of the formaldehyde. This view is also supported by the later work of Ekberg and S i l ~ e r , ~ and B e r o ~ a . ~ A conductimetric method more suitable for the routine determination of sodium glycol- late was developed,6 but when it was applied to the ethanolic wash liquors from sodium carboxymethylcellulose it was found to give consistently higher values for the glycollate content than those obtained by the colorimetric method of Eastenvood.Subsequent investi- gation revealed the discrepancies to be caused by the interference by ethanol in the colour development with chromotropic acid. This led to a re-appraisal of Easterwood's method and that of the standard colorimetric determination of formaldehyde in ethanolic solution with chromotropic acid. The chemistry of the reaction ofchromotropic acid with form aldehyde is not known with certainty. Feigl' postulates that, as aromatic hydroxy compounds condense with formaldehyde to yield colourless hydroxyphenylmethanes, it is probable that the initial step consists of a condensation of the phenolic chromotropic acid with formaldehyde, followed by oxidation to a 9-quinonoidal compound of the type shown below- SO,H SO,H 'A' CHJ<> / \ H O < d <JPH 'S0,H SOsH / Concentrated sulphuric acid participates in the reaction both as a dehydrant and an oxidising agent.0 SAC and the authors.806 STILL et al. : EFFECT OF ETHANOL ON THE COLORIMETRIC [Alzalyst, Vol. 93 Eyler, Klug and Diephus* have used 2,7-dihydroxynaphthalene in concentrated sulphuric acid to detect and determine formaldehyde colorimetrically. Several colour reagents have been investigated by West and Sen,9 who concluded that chromotropic acid and 2,7-dihydroxy- naphthalene were outstanding for determining formaldehyde. EXPERIMENTAL REAGENTS- Glycollic acid-(Koch-Light crystalline), recrystallised from water, m.p. 79.5" to 80" C.A standard aqueous or ethanolic solution was prepared containing 1.32 x moles per litre. Chromotropic acid-Obtainable from British Drug Houses Ltd. for formaldehyde deter- mination. A 5 per cent. aqueous solution of the disodium salt was prepared immediately prior to use. 2,7-Dihydroxynaphthalene-A solution in 95 per cent. sulphuric acid containing 0.1 g per litre was prepared. Sulphuric acid, sp.gr. 1.84. Ethautod-The procedure for the preparation of the various aqueous ethanol samples is similar to that for 80 per cent. aqueous ethanol. An 80-ml aliquot of absolute ethanol was introduced into a 100-ml calibrated flask and made up to the mark with distilled water. Formaldehyde-An aqueous solution was obtained by the dry distillation of para- formaldehyde, and passage of the formaldehyde vapour formed (after rejection of the first 30 per cent.of the distillate to remove polymer) into ice-cold, continuously stirred, distilled water. This solution was standardised by using both sodium sulphite and iodimetric met hods. lo y1l Standard aqueous and ethanolic solutions containing 1.66 x moles per litre of formaldehyde were prepared by dilution. Ethyl glycollate-Standard aqueous and ethanolic solutions containing 1 -34 x moles per litre were prepared. Diethyl formad-This was prepared by the method of Vogel12 as a colourless liquid which, when fractionally distilled, yielded a fraction b.p. 87" to 89" C and nL0 1.3730. Standard aqueous or ethanolic solutions containing 1.32 x moles per litre were prepared. APPARATUS- photometer and a Beckmann DB recording spectrophotometer. Optical density measurements were made with 1-cm cells on a Unicam SPW spectro- PROCEDURE BEER - LAMBERT PLOTS OBTAINED BY USING AQUEOUS AND 80 PER CENT.AQUEOUS ETHANOLIC SOLUTIONS OF THE SUBSTRATES- Chromotropic acid determinations-An aliquot of the standard solution of the substrate (glycollic acid, formaldehyde, diethyl formal or ethyl glycollate) dissolved in water or 80 per cent. aqueous ethanol was added, from a 2-ml microburette, to a 50-ml calibrated flask. Distilled water or 80 per cent. aqueous ethanol was then added to adjust the total volume to 2 ml, then 1 ml of chromotropic acid was added, followed by concentrated sulphuric acid to bring the total volume to about 40ml. The reactants were mixed by shaking and the flask was immersed in a boiling water bath for exactly 30 minutes.The flask was removed and allowed to cool to room temperature and the volume adjusted to the mark with concen- trated sulphuric acid so that, for the 80 per cent. aqueous ethanolic solution, the final concentration of ethanol in the reaction mixture was therefore 3.2 per cent. v/v. A blank determination with either water or 2 ml of 80 per cent. aqueous ethanol was made by using the same procedure. When the blank solution was used as reference the optical density at 580 nm (the Amax. was found to be 580 nm, its position remaining constant, independent of the presence of ethanol) was measured. 2,7-Dihydroxynaphthalere determinationsa-An aliquot of glycollic acid dissolved in water or 80 per cent.aqueous ethanol was added from a 2-ml microburette to a 50-ml flask. Dis- tilled water or 80 per cent. aqueous ethanol was then added to adjust the volume to 2 ml. A 20-ml aliquot of 2,7-dihydroxynaphthalene was added, the contents of the flask wellDecember, 19681 DETERMINATION OF FORMALDEHYDE AND GLYCOLLIC ACID 807 mixed and the flask immersed in a boiling water bath for exactly 30 minutes. The flask was removed and allowed to cool to room temperature and the volume adjusted to the mark with distilled water. EFFECT OF ETHANOL CONCENTRATION ON THE MOLAR EXTINCTION COEFFICIENTS WITH Solutions of the four substrates in water, ethanol or aqueous ethanol were prepared, giving a range of ethanol concentration in the final reaction mixture of 0 to 4.0 per cent.and a substrate concentration of 3.2 x 10-5 moles per litre. The optical density was measured as previously by using an appropriate aqueous, ethanolic or aqueous ethanolic blank. The molar extinction coefficients were then calculated. CHROMOTROPIC ACID- RESULTS AND DISCUSSION TABLE I Colour Substrate reagent Glycollic acid Glycollic acid Glycollic acid Glycollic acid Formaldehyde Formaldehyde Ethyl glycollate Ethyl glycollate Diethyl formal Diethyl formal .. .. A * . .. A .. .. B .. . . B .. .. A .. .. h .. .. A .. .. A .. .. A .. . . A Solution Aqueous 80 per cent. aqueous ethanol Aqueous 80 per cent. aqueous ethanol Aqueous 80 per cent. aqueous ethanol Aqueous 80 per cent. aqueous ethanol Aqueous 80 per cent. aqueous ethanol Molar extinction coefficient, litres moles-' cm-I x 10-0 1-85 0.64 2.32 0-3 1 1.84 1-45 1-68 0.63 1-55 1-19 Colour reagent A is chromotropic acid.Colour reagent B is 2,7'-dihydroxynaphthalene. Beer - Lambert plots for all the substrates studied in both aqueous and ethanolic solution, were straight lines passing through the origin. The molar extinction coefficients calculated from the slopes of these lines are summarised in Table I. Percentage ethanol in the reaction mixture Effect of ethanol concentra- tion of molar extinction coefficient E : [7 formaldehyde; 0 glycollic acid; 4 diethyl formal; x ethyl glycollate all with chromotropic acid Fig. 1.808 STILL et al. : EFFECT OF ETHANOL ON THE COLORIMETRIC [Analyst, Vol. 93 The influence of ethanol concentration on the molar extinction coefficients for the four substrates with chromotropic acid is shown in Fig. 1.These results show quite clearly that ethanol does interfere with the colorimetric determination of all four substrates with chromo- tropic acid and that the effect is greater for glycollic acid and ethyl glycollate than for formaldehyde and diethyl formal. The range of ethanol concentrations used in this work is considerably greater than that used by Ekberg and Silver4 when they reported that ethanol had no effect. Our results confirm that with the ethanol concentration used by these workers, the effect would be far less than the experimental error. Significantly, the molar extinction coefficients of these compounds in the absence of ethanol are not the same. In all of these reactions the species responsible for the production of the coloured complex with chromotropic acid is the protonated form of formaldehyde (I).Glycollic acid Ethyl glycollate Diethyl formal H0.CH2C<zH II fH+ II f H + OH2 The generation of this reactive intermediate species from glycollic acid involves the intermediate formation of the acyl carbonium ion (11) and the subsequent loss from it of carbon monoxide. In the absence of ethanol these equilibrium reactions are displaced, in favour of the production of (I), as indicated by glycollic acid and formaldehyde which have the same molar extinction coefficients (1-85 and 1-84 x lo4 litres moles-l cm-l). For ethyl glycollate and diethyl formal, however, the production of (I) involves the release of one and two molecules of ethanol, respectively.Ethanol, being a stronger nucleophile than water, will attack the intermediate carbonium ions with the resultant incomplete production of (I) and a lowering of their molar extinction coefficients (1.68 and 1.55 x lo4 litres moles-1 cm-1, respectively, for ethyl glycollate and diethyl formal). The production of (I) from all four compounds in ethanolic solution involves equilibrium reactions that will be progressively reversed by ethanol, resulting in the observed lowering of the molar extinction coefficient. The influence of ethanol on the molar extinction coefficients for ethyl glycollate and glycollic acid is greater than that observed for formaldehyde and &ethyl formal, as both reactions involve the acyl carbonium ion intermediate (11).The attack on (11) by ethanol under these reaction conditions, in which ethanol is in considerable molar excess, will favour the formation of the ester. In addition there is the reaction of (I) with ethanol which is common to all four substrates. It is evident from these results that for the determination of glycollic acid in ethanolic solution, Eastenvood’s method must be amended to include an appropriate ethanolic blank. Alternatively, as the presence of ethanol greatly decreases the sensitivity of the colour reaction with chromotropic acid, the preliminary removal of it by evaporation to dryness is recommended. The authors thank Mr. M. E. Shreeve and Mr. S. C. Elliston for some preliminary observations made in this study.December, 19681 DETERMINATION OF FORMALDEHYDE AND GLYCOLLIC ACID REFERENCES 809 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. Easterwood, M., Analyt. Chem., 1967, 29, 981. Bricker, C. E., and Johnson, H. R., Ind. Engng Chem. Anulyt. Edn, 1945, 17, 400. Bricker, C. E., and Vail, W. A., Anulyt. Chem., 1950, 22, 720. Ekberg, D. R., and Silver, E. C., Ibid., 1966, 38, 1421. Beroza, M., Ibid., 1964, 26, 1970. Still, R. H., Wilson, K., Shreeve, M. E. , and Boardman, W., in preparation. Feigl, F. , “Spot Tests in Organic Analysis,” Seventh Edition, Elsevier Publishing Co., Amsterdam, Eyler, R. W., mug, E. D., and Diephus, F., Analyt. Chem.. 1947, 19, 24. West, P. W., and Sen, B., 2. analyt. Chem., 1956, 153, 177. Walker, J. F., “Formaldehyde,” American Chemical Society Monograph, No. 159, Third Edition, Reinhold Publishing Corporation, New York, Amsterdam and London, 1964. C u d n g , W. M., Hopper, I. V., and Wheeler, T. S., “Systematic Organic Chemistry,” Fourth Edition, Constable and Co., London, 1950, p. 492. Vogel, A. I., “A Textbook of Practical Organic Chemistry, including Qualitative Organic Analysis,” Second Edition, Longmans, Green & Co. Ltd., London, New York and Toronto, 1964, p. 325. Received April 8th, 1968 London and New York, 1966, p. 434.

ISSN:0003-2654    

DOI:10.1039/AN9689300805

出版商:RSC    

年代:1968

数据来源: RSC

摘要:

810 Artalyst, December, 1968, Vol. 93, fie. 810-816 Simple, Rapid Quantitative Determination of Amino-acids by Thin-layer Chromatography BY MARY E. CLARK (Department of Organismic Biology, University of California, Irvine, California 92664, U.S.A .) A simple, rapid method for the quantitative determination of complex mixtures of amino-acids is described. After separation on thin layers of cellulose mounted on flexible plastic sheets, the chromatograms are sprayed with ninhydrin and developed under controlled conditions. The spots are cut out and eluted with 2 ml of 50 per cent. propyl alcohol and the optical density at 570 nm determined with a microspectrophotometer. From 4 to 6 chromato- grams can be eluted and read in 1 day. An accuracy of about 5 nmoles is obtained over a range of 12.5 to 50 nmoles.A t higher concentrations, accuracy is within A10 per cent. Standard graphs are reproducible for at least 5 months. DURING an investigation of the composition of the free amino-acid pools of marine inverte- brates, a quick and reliable method for quantitative analysis was sought, Although mixtures of several amino-acids can be separated rapidly by thin-layer chromatography, its application has been severely limited because of the difficulties involved in making the method quantita- tive. Previously described methods involve scraping the coloured spots from glass plates, and either their subsequent elution for absorption spectrophotometryl or their analysis by reflectance spectroph~tometry.~~~ Both methods require considerable time in the handling of individual spots.The present paper describes a method in which commercially available thin layers of cellulose supported on a flexible plastic backing are used. After separation of the amino-acids and development of the coloured ninhydrin reaction products, the spots are cut out with scissors and eluted directly, in a few minutes, with a small volume of solvent. The optical density is then determined with a microspectrophotometer, the sensitivity and reproducibility of the method being of the same order as those previously described. EXPERIMENTAL STANDARD AMINO-ACID MIXTURE- Throughout the development of this method a single batch of a standard mixture of 17 amino-acids, obtained from CalBiochem (Kit No. AA-5), was used. This contained 2.5 +_ 0.05 pmoles per ml of each amino-acid.The graphs obtained by using this mixture were later compared with a known mixture of amino-acids, made up to resemble the free amino-acid pool of an experimental animal and with a concentration factor of 30 between the least and most concentrated amino-acids. As shown in Table 11, amino-acids of the known mixture gave values within the error appropriate to their concentration levels. It would thus appear that the CalBiochem mixture provided a suitable standard. SEPARATION OF THE AMINO-ACIDS- Samples were spotted on to 20 x 20-cm sheets of MN-Polygram Cell 300 (MN 300), which is a cellulose powder produced by Machery-Nagel and Co. (Duren), and supported on inert sheets of poly(ethy1ene terephthalate). To increase the reproducibility of separation, spots were always placed at the same corner of each chromatogram, in relation to its position in the package as supplied.Spots were applied with Drummond “Microcap” capillary pipettes, which were rinsed once with distilled water to ensure complete transfer of the contents. * Present address : W. M. Keck Engineering Laboratories, California Institute of Technology, Pasadena, California 91109, U.S.A. 0 SAC and the author. A hot air stream was directed on to the spot during application.CLARK 81 1 Chromatograms were run in Kontes Chromaflex developing tanks, with the solvent system described by Jones and Heath~ote.~ To increase further the separation of the larger spots, the times were slightly increased; thus, solvent I, which was isopropyl alcohol - formic acid - water (40 + 2 + lo), was run for 3Q hours and solvent 11, t-butyl alcohol - ethyl methyl ketone - ammonia solution - water (25 + 15 + 5 + 5 ) ) was run for 3 to 35 hours.Even so, with amounts of each amino-acid greater than about 25 nmoles, arginine and lysine overlap considerably. Between solvents I and I1 the chromatograms were dried for at least 5 minutes in a stream of cold air, and a line was then scored just below the zone of yellow impurities to separate them from the remainder of the chr~matogram.~ On removal from solvent I1 the chromatograms were subjected to a cold air stream for at least 20 minutes, and then stored between sheets of clean paper. COLOUR DEVELOPMENT- Before treating the chromatograms with ninhydrin they were once again subjected to a stream of cold air for 20 minutes to remove any ammonia absorbed from the atmosphere during storage.In our laboratory, where developing chromatograms could not be isolated from other chemical work, this proved the best insurance against high background colour. Chromatograms were sprayed with 7 to 8 ml of freshly prepared 2 per cent. ninhydrin in absolute ethanol,5 as this amount permits saturation without dripping. As soon as the ethanol was evaporated in a cold air stream, the chromatograms were moved to a dark cupboard, under conditions of constant temperature and humidity. The importance of the physical conditions for reproducible colour development of ninhydrin-reaction products in paper chromatography has been demonstrated by Wellington, who found that, for most amino-acids, 20" C and 35 to 40 per cent.relative humidity gave maximum colour produc- tion.696 Reproducible, though less intense, colour was produced under other conditions, provided they were constant. The effects of temperature and humidity on intensity of colour production on thin layers was not investigated here, reliance being placed on the air conditioning in the laboratory to maintain constant conditions. The temperature in the cupboard in which the chromatograms were developed remained between 23.3" and 26.1" C and the relative humidity between 60 and 64 per cent. over a period of several weeks. The chromatograms were developed for 24 to 30 hours, as this length of time was found to yield a maximum reaction on paper.5 ELUTION OF SPOTS- As elution entails destruction of the chromatogram, it is useful to have a record to indicate which spots were overlapping.Each spot was outlined with a blunt pencil and identified. In addition, background spots were drawn in at this time. The background varies across the chromatogram, especially in the direction of the second solvent. This may be caused either by impurities in the cellulose layer or by impurities in the solvents (or by both sources). It is convenient to divide the amino-acids into four groups, with a single background spot for each group. This spot was drawn the same size as the largest spot in the group, and when the other spots in this group were cut out, sufficient background was added so that all the spots were the same size as the background spot.The chromatograph was recorded with a Polaroid camera and Polaroid Land Projection film, Type 146-L. Fine-tipped dissecting scissors were used to cut out individual spots, care being taken to avoid finger-prints. To prevent chipping of the cellulose layer and loss of coloured material it was necessary to cut in straight lines. Each spot was placed in a 1.2 x 10-cm test-tube. If the spots were large, they were cut in half so that they were small enough to be completely covered by 2 ml of eluting fluid, and the pieces placed in the tubes with the cellulose layers facing outwards for efficient elution. Two millilitres of 50 per cent. propyl alcohol in distilled water were transferred by pipette into each tube. Elution required 20 minutes, after which each tube was shaken vigorously and left for 20 minutes to allow the cellulose particles to settle.DETERMINATION OF OPTICAL DENSITY- The optical density at 570 nm was determined for each spot, including the blanks, after setting the instrument at 100 per cent. transmission for 50 per cent. propyl alcohol. Micro- cuvettes and a conventional spectrophotometer can be used, or, for more rapid analysis,812 CLARK : SIMPLE, RAPID QUANTITATIVE DETERMINATION [Arta&st, Vol. 93 a Gilford Microspectrophotometer 300 is convenient. Both have been used in obtaining the standard graphs in the present study. In the latter method, about 1.5 ml of the coloured eluate is carefully decanted into a small vial, avoiding transfer of cellulose particles, prior to taking a reading.The optical density of the appropriate blank was then subtracted from that of each spot. CALBIOCHEM STANDARD MIXTURE- As is well known, equimolar amounts of the various amino-acids do not produce the same amount of coloured reaction product with ninhydrin. Although proline appears to give a more purplish colour on thin layers of cellulose than on paper, the amount of colour is insufficiently reproducible to yield quantitative results, and no proline values are included here. The results of analyses for three different amounts of the standard mixture are shown in Table I. The number of replicates (.n) at each concentration is given at the top of each column, and for all three amounts, replicates were made over a period of 5 months. As is to be expected, the percentage error decreases with increasing amounts of amino- acids. This is primarily caused by the uncertainties introduced by variable amounts of background colour.Thus, those amino-acids which produce less colour with ninhydrin, such as aspartic acid, glycine, cystine, histidine and tyrosine, tend to give a higher error than more chromogenic amino-acids. The variability of leucine, isoleucine and phenylalanine is also caused by background effects. These amino-acids lie in the upper right-hand quadrant of the chromatogram where the background is always deeper and more variable. It will be seen that, at concentrations ranging from 12.5 to 50 nmoles, the absolute error for most amino-acids remains at about the same level, between 3 and 5 nmoles. At the highest concentration, the total load on the chromatogram was 0.85 pmoles of amino-acid. When the mean optical density values for the individual amino-acids are plotted against concentration, rectilinearity (indicating agreement with Beer's law) is obtained for all amino- acids, except threonine, isoleucine and leucine (Fig.1). Even for these three amino-acids, reproducible graphs are obtained, and thus such graphs can be used as standards for the determination of amino-acids in unknown samples. RESULTS 0.500 0.400 r).zci? 03909 0.100 10 20 30 40 50 0.500 C 1 0,400 0.300 0.200 0.100 10 20 30 40 50 Nanomoles Fig. 1. Optical density veisus concentration graphs of several amino-acids. Values a t 12-5, 25 and 50 nmoles are taken from Table I. Values at 5 nmoles are based on 1 determination; those a t 37.5 are the mean of 2 deter- minations.Threonine and leucine do not give straight-line graphs. (The graph for isoleucine resembles that of leucine.) All other amino-acids follow Beer's lawTABLE . I MEAN OPTICAL DENSITIES, STANDARD DEVIATIONS, PERCENTAGE ERROR AND ABSOLUTE ERROR FOR VARYING CONCENTRATIONS OF CALBIOCHEM AMINO-ACID MIXTURE 1/2 Cystine Aspartic acid Glutamic acid Arginine Lysine . . Glycine . . Serine . . Histidine Alanine Tyrosine.. Valine . . Methionine Threonine Isoleucine Leucine Phen ylalanine 7 x . . 0.054 . . 0.051 * . 0.110 . . 0.093 . . 0.099 . . 0.053 . . 0.102 . . 0.050 . . 0.121 . . 0.071 . . 0.111 . . 0.106 . . 0.113 . . 0.083 . . 0.058 . . 0.070 12.5 nmoles n = 5 Standard deviation f 0.034 f 0.007 f 0.015 f 0.025 f 0.024 f 0.015 f 0.022 f 0.026 f 0.013 f 0.024 f 0,022 f 0.012 & 0-040 f 0-025 f 0-018 f 0.008 Error, per cent.63 14 14 27 24 28 22 44 11 34 20 11 35 30 31 11 - X = Mean optical density. n = Number of determinations. 3: nmoles f 7.9 f 1.8 f 1.8 f 3.4 f 3.0 f 3.5 f 2.8 f 5.5 f 1.4 f4.3 f 2.5 1.4 f 4.4 f3.8 f 3.9 f 1.4 25.0 nmoles n = 6 r A Standard Error, deviation per cent. 0.120 f 0.039 33 0.106 f 0.028 26 0.201 f 0-050 25 0.231 f 0.039 17 0.111 f 0*028 25 0-197 f 0.046 23 0.146 f 0.025 17 0.228 f 0.009 4 0.148 f 0.034 23 0.211 f 0.034 16 0.204 f 0.042 21 0.204 f 0,017 13 0.183 f 0.029 16 0.148 f 0.032 22 0.140 & 0.026 19 0-200 f 0.023 12 I f nmoles f 8.3 f 6.5 f 6.3 f 4.3 f 3.0 f 6.3 f 5-8 k4.2 f 1.0 f5.8 f 4.0 f 5.3 f 3.3 f 4.0 f 5.5 f4.8 50.0 nmoles n = 4 A I \ Standard Error, U deviation per cent.f nmoles rn 0.212 f 0.017 8 -fs 4.0 0.456 f 0.053 12 f 6.0 0.477 f 0.076 16 f 8.0 4 0.418 f 0.022 5 f 2.5 M 0.263 & 0.014 5 3: 2-5 0.512 & 0.019 4 5 2.0 0 0.286 f 0.018 6 f 3.0 9 0 0.479 f 0.038 8 f 4.0 0.455 f 0-017 4 f 2.0 0.388 f 0.029 7 0.177 f 0.036 20 flO*O c3 0.421 f 0.039 9 f 4.5 5 0.184 & 0.021 11 f 5.5 5 0.421 f 0.063 15 f 7.5 gj 0-396 f 0.066 17 f 8-5 0.288 & 0.009 3 f 1.5 $ 4814 CLARK : SIMPLE, RAPID QUANTITATIVE DETERMINATION [Autahyd, VOl. 93 ARTIFICIAL MIXTURE RESEMBLING THE FREE AMINO-ACIDS OF A POLYCHAETE WORM- In preliminary investigations, the composition of the free amino-acid pool of the poly- chaete worm Stauronereis rudolphi (Delle Chiaje) was determined on de-proteinised and de- salted aliquots of whole-body homogenates.(Details of the preparative procedures will be described elsewhere.) Nineteen amino-acids were identified (excluding the amino-acid derivative taurine, which does not give reproducible colour reactions by the present method and must be determined by alternative procedures.) These ranged in concentration from 0.5 mmoles per kg of body water (histidine) to 15 mmoles per kg of body water (aspartic acid). An artificial mixture was then prepared (Table 11) and analysed by the method described above. A chromatogram of this mixture is shown in Fig. 2. The results of two individual chromatograms of this mixture are shown in Table 11. The difference between the amount found and that expected falls, in most instances, near or within the range of error expected, as determined from the percentage error of the nearest amount used in preparing the standard graphs (Table I).Arginine and lysine spots overlap with each other and with asparagine, which gives a weak, brown colour with ninhydrin, but this latter error can be obviated by hydrolysing the mixture before it is used for a chromato- gram. There were no standard graphs for asparagine nor tryptophane, and these were read on the lysine and threonine graphs, respectively. For tryptophane this resulted in a systematic error, as the colour for tryptophane under these conditions was about half that of threonine. It should also be noted that the CalBiochem standard mixture contained cystine (1-25 pmoles per ml), whereas the artificial mixture contained cysteine, which yields a much weaker colour with ninhydrin.Aspartic acid consistently seems to give slightly high values, and alanine consistently slightly low ones. This may be a result of an interaction during migration of the amino- acids, when those present in disproportionately high concentration exert a salt-like effect on the movement of some of the other amino-acids. The total recovery of amino-acids was within 10 per cent. of that expected. DISCUSSION The method described here provides a rapid means of quantitatively determining small amounts of amino-acids. It requires about 3 days to carry out the whole procedure, of which only the third day, when the spots are eluted and read, requires continuous attention from the researcher.It was found possible to elute and read as many as six chromatograms in 1 day. It should be emphasised that during the development of this technique, every effort was made to treat replicate chromatograms on different days at each stage of the procedure, so that maximum variability of the conditions would be observed. In this way, it was possible to establish the true variation in reproducibility in our laboratory. In addition, all results were obtained on complex mixtures of at least 17 amino-acids; thus the range of error reported is conservative in all respects. It is probable that some of the error, especially at low concentrations, could be reduced by carrying out all the steps following the evaporation of the second solvent in a separate room kept scrupulously free from ammonia and other chemical fumes.There appear to be only two earlier methods for the quantitative determination of amino-acids with thin-layer chromatography. Frodyma and Frei,2s3 who use reflectance spectrophotometry, found that when silica gel on glass plates was used scanning did not give sufficiently reproducible results, and that it was therefore necessary to scrape the spots off the glass surface before determinations were made. By running simultaneous standards they achieved a 5 per cent. reproducibility, within a range of about 5 to 200 nmoles for various amino-acids. However, as the percentage reflectance is not linear with concentration nor, for several amino-acids, is it linear even with the square root of the concentration, analysis of complex mixtures necessitates running a complete set of standard graphs with each unknown.The other method is that recently published by Bondivenne and Busch,l who used cellulose 300 MN spread on glass plates and after colour development scraped the individual spots off the glass and eluted the colour through sintered-glass filters. As this required 4 ml of eluant, the sensitivity of their method is about one half that of the present method. The smallest amount of the four amino-acids analysed by these workers was 50 nmoles, at which level the error ranged from 3.7 to 8.5 per cent. When a chromatogram was made of theFig. 2. The chromatogram of an artificial mixture of amino-acids resembling the composition of the free amino-acid content of the pool of the polychaete worm, Stauronereis rudolphi.Solvent I was run in the vertical direction, solvent I1 in the horizontal To face page 8 14)TABLE I1 Cysteine . . Aspartic acid Glutamic acid Arginine . . Asparagine Lysine . . Glycine . . Serine .. Histidine . . Alanine . . Tyrosine . . Valine . . Methionine Tryptophane Threonine Isoleucine Leucine . . Phenylalanine Total . . .. .. ,. . . .. .. .. .. .. .. .. ,. .. .. . . .. .. .. . . RECOVERY OF KNOWN MIXTURE OF AMINO-ACIDS FROM TWO THIN-LAYER CHROMATOGRAMS Artificial mixture of amino-acids, mmoles per litre . . 2.0 . . 15.0 . . 10.0 * . 2.0 . . 5.0 . . 5.0 . . 10.0 . . 2.0 . . 0.5 . . 10.0 . . 1.0 . . 1.0 . . 1.0 . . 1.0 * , 1.5 * . 1-0 . . 1.5 . . 1.6 . . 71.0 nmoles found 1.0 71.6 43.6 13.8 5.6 12.8 40.2 9.4 1.8 37.0 6.4 4.4 4.6 1.8 4-4 4.4 7-6 5.8 276.2 nmoles expected 8.0 60.0 40.0 8.0 20.0 20.0 40.0 8.0 2.0 40.0 4.0 4.0 4.0 4.0 6.0 4.0 6-0 6-0 284.0 Difference found, nmoles + 11.6 + 3.6 - 7*0* 4- 5*8* - 14.6 - 7*2* + 0.2 + 1.4 - 0.2 - 3.0 $.2.4 + 0.4 + 0.6 - 2.2* - 1.6 + 0.4 + 1.6 - 0.2 Recovery, per cent. . . 97.3 * Spots where large errors were expected-see text for explanations. Approximate difference expected, nmoles f 5.0 f 6.0 f 2.4 f 2.2 -f: 2.4 & 4.4 f 1.8 0.9 4 1.6 f 1.4 f 0.8 f 0 - 4 - I f2.1 f 1.2 4 1.9 f 0.7 nmoles found 10.0 168.2 102.1 24.8 15.2 35.0 87.2 18.2 1-4 89.7 13.2 7.6 9.6 4.8 11.4 12.2 16.6 10.4 637.6 89.8 nmoles expected 20.0 150.0 100.0 20.0 50.0 50.0 100.0 20.0 5.0 100.0 10.0 10.0 10.0 10.0 15.0 10.0 15.0 15.0 710.0 Difference found, nmoles - 10*0* + 18.2 + 2.1 + 4.8* - 34.8* - 15*0* - 12.8 - 1.8 - 3.6 - 10.3 + 3.2 - 2.4 - 0.4 - 5*2* - 3.6 + 2.2 + 1-6 - 4.6 Approximate 's difference expected, 2 nmoles & 6.6 2 f15.0 $j f 6.0 f 3.4 2 f 4.6 f 2.2 ;p 4 f 4.0 M f 3.4 f 2.0 X f 1.1 & 5.3 & 3.0 f 4.7 f 11.7 cd 0 - 2 I?816 CLARK four amino-acids together, in amounts exceeding 100 nmoles of each, the error increased, ranging from 7.8 to 13.5 per cent.for five replicate determinations. It would thus appear that the present method offers a more rapid analysis, with somewhat improved accuracy and sensitivity. Recently Heathcote and Washington' have published a sensitive method based on paper chromatography, in which they state that, for most amino-acids, determination of amounts of from 1 to 250nmoles is possible with an over-all accuracy of _+lo per cent.It is not clear if this high degree of accuracy at lower concentrations could be obtained in complex mixtures, but it would appear that their method is more reproducible at lower concentrations than the present one. It entails, however, a rather elaborate, prolonged procedure for eluting the spots. The staining reagent used by these workers was chosen for its increased sensitivity, which is about five times that of ninhydrin alone. However, as the elution procedure required 10 ml for each spot, optical densities were equivalent to those obtained here. In addition, as with all methods in which paper chromatography is used, the time required in the solvent systems is about three times that necessary for separation on thin layers. When sufficient material is available to carry out several chromatograms on the same sample, better accuracy can be expected with the present method and this may prove less time-consuming than the more elaborate method of Heathcote and Washington, which would be the method of choice when sample size is limited. This work was supported by USPHS Grant 12889, awarded to Dr. Grover C . Stephens, in whose laboratory the work was done. I am most grateful to Dr. Stevens for making this study possible, and for his encouragement throughout. REFERENCES 1. 2. 3. 4. 5. 6. 7. Bondivenne, R., and Busch, N., J . Chromat., 1967, 29, 349. Frodyma, M. M,, and Frei, R. W., Ibid., 1964, 15, 501. - - , Ibbid., 1965, 17, 131. Jonis, K., and Heathcote, J. G., Ibid., 1966,24, 106. Wellington, E. F., Can. J . Chem., 1952,30, 581. - , Ibid., 1953,31, 484. Heathcote, J. G,, and Washington, R. J., AnaZyst, 1967, 92, 627. Received Muy 6th. 1968

ISSN:0003-2654    

DOI:10.1039/AN9689300810

出版商:RSC    

年代:1968

数据来源: RSC

摘要:

Aszalyst, December, 1968, Vol. 93, $9. 817-820 81 7 The Continuous Polarographic Determination of Small Amounts of Nitroglycerine in Plant Effluent BY A. R. HOLLAND AND A. G. S. BENHAM (Imperial Metal Industries Limited, Summerfeld Research Station , Kidderminster) A fully automatic polarographic method is described for the continuous monitoring of nitroglycerine plant effluent, for which an arbitrary maximum level of 100 p.p.m. of nitroglycerine has been set. Conventional polarographic principles are involved, but the polarograph used has been specially designed, and incorporates some novel features that make it admirably suitable for this and similar work. Constructional details are given of the polarograph, and of an alarm system that operates when the permitted level of nitroglycerine in the effluent is exceeded.IN the manufacture of nitroglycerine by the injection nitration process (Nitroglycerine Aktiebolaget, Gyttorp), an aqueous solution is produced, the nitroglycerine content of which, after neutralisation, is about 0-5 per cent. This solution also contains about 2.5 per cent. of sodium nitrate, 1.5 per cent. of sodium carbonate and 0.1 per cent. of sodium sulphate. At this stage the nitroglycerine is in suspension, and most of it is recovered by a “settling out” process, the remainder (about 0.1 per cent.) being diluted with a 20-fold excess of water before it is discharged to waste. An arbitrary maximum level of nitroglycerine in this effluent has been set at 100 p.p.m., and the problem was how to provide a reliable analytical procedure so that in the event of this level being exceeded any failure at the dilution stage could be detected immediately and rectified.It was therefore considered necessary to have an automatic sampling and analysis system for the continuous determination of levels of nitroglycerine normally present in this effluent, and some form of warning (alarm) that would alert plant operators when the amount of nitroglycerine in the effluent was in excess of the permitted level. Small amounts of nitroglycerine can be determined by a spectrophotometric procedure in which N-1-naphthylethylenediamine is used, but the use of any colorimetric reagent is excluded because of the turbid nature of the effluent, especially during periods of heavy rainfall. Nitroglycerine produces three distinct polarographic waves (-0.25, -0.45 and -0-76 volt versus mercury-pool anode) in a quaternary ammonium halide solution, and a procedure involving the use of this electrolyte, by Williams and Kenyon,l was used in the following experimental work.EXPERIMENTAL CALIBRATION GRAPH- In preliminary experiments with a Cambridge photographic-recording polarograph, a linear calibration graph was obtained from the peak heights recorded at -0.75 volt when simulated effluent samples containing up to 150 p.p.m. of nitroglycerine were analysed; nitrogen was used to de-oxygenate the samples. In further experiments, an alternative (cheaper) base electrolyte (ammonium chloride - potassium chloride) was used; sodium sulphite was added to the electrolyte to replace the use of nitrogen, because it was envisaged that a gaseous de-oxidant would lead to complications in any fully automatic system.Satisfactory results were obtained, and this electrolyte was used in subsequent experiments. 0 SAC and the authors.818 HOLLAND AND BENHAM : CONTINUOUS POLAROGRAPHIC DETERMINATION [AflabSt, VOl. 93 DESIGN OF THE ANALYTICAL SYSTEM The following automatic stages were planned. (i) Sampling of effluent every 15 minutes. (ii) Addition and mixing of base electrolyte. (iii) Transfer of prepared sample to a polarographic cell. (iv) Analysis of prepared sample. (v) Removal of analysed sample from polarographic cell. The sampled eflluent is pumped through a filter into a header tank provided with an overflow. A week's supply of base electrolyte is also stored in another header tank, and this tank is provided with a floating lid to minimise evaporation losses.Supplies of base electrolyte and sample enter the polarographic cell by means of a gravity feed via the solenoid-controlled valves, S1 and S2 (Fig. l), and the chokes are adjusted so that equal volumes of both solutions can pass into the mixing chamber. In this way, the use of low-capacity pumps, which we have not found reliable in operation over long periods, is eliminated. A period of 30 seconds is allowed for the delivery of a 100-ml sample and also for the same volume of electrolyte; the system is self-purging. Electrolyte storage -ve - Sample '-1 Valve Valve 4 Mixer Filter Drain Fig. 1. Schematic layout ELECTRICAL CIRCUITS (FIG 2)- The two mercury cells are connected in series to provide the polarographic cell current via an adjustable resistor, R1, a potential divider, R3 (or R5), R4 and an adjustable resistor, R2. The recorder is connected across R2, and the oscillating potential produced by the dropping-mercury electrode is damped by a condenser connected across R2, so that a constant current of 10pA flows at full-scale deflection of the recorder. The polarographic current flows through the cell via a platinum wire, which is in electrical contact with the (negative) mercury reservoir, and a second platinum wire in similar contact with the (positive) mercury pool in the neck of the flask (Fig.1). The positive current is taken from the centre noint of the notential divider via the microswitch.MS.3. which makes OPERATION- The position of the valve voltmeter, V2, is shown in Fig. 2; the meter reading is adjusted to -1.00 volt by means of R1. The reading on V1 is noted on the moving-coil meter, then V2 is disconnected. Any variation in voltage can be corrected by restoring V2 to the original reading.December, 19681 OF SMALL AMOUNTS OF NITROGLYCERINE IN PLANT EFFLUENT 819 The sampling sequence is programmed by a 4 r.p.h. mains timer, which operates the microswitches, MS1, MS2 and MS3, by a camshaft. At zero time, the mains-operated solenoid valves, S1 and S2, are energised through MS1, thus permitting 1OOml of the sample and the same volume of the base electrolyte to pass into the mixing chamber. After a lapse of 9& minutes to allow the solution to de-oxygenate, the microswitch, MS3, makes contact and allows a potential of -1.00 volt to be applied instantaneously to the cell.The concen- tration of nitroglycerine is shown on the recorder (Fig. 2). After 5 minutes, MS3 breaks to stop the cell current and the sequence is repeated. Switches are provided to enable manual operations to be made, and a second mains timer (12 r.p.h.) is operated by MS2 to permit the slide-wire, R5, to be used when required. The nitroglycerine peak is displayed on a self-balancing potentiometer strip-chart recorder, which is fitted with a mercury switch that operates an alarm system when the nitroglycerine content of the sample exceeds 100 p.p.m. Fig. 1 shows a schematic layout of the unit, and Fig. 2 the electrical circuit.230V Alarm Potent iomet r ic R,, R, = 100-k i2 variable resistors R3, R, = 100-k I2 fixed resistors R2 = 500-k variable resistor SW = Single pole change over Fig. 2. Electrical circuits EFFECT OF VARIABLES- Sodizcm carbonate-At the 50 p.p.m. nitroglycerine level, the presence of 4 per cent. of sodium carbonate gave an apparent nitroglycerine level of 47.5 p.p.m. Under plant operating conditions, this level of sodium carbonate is most unlikely to occur, so that a normal variation in sodium carbonate content is not significant. Glycerol and sodiztm nitrate-The presence of 1 per cent. of glycerol in the effluent would indicate serious malfunctioning of the plant, but even at this level of glycerol the effect on the determined nitroglycerine content of the effluent is insignificant.The same observation was made in the presence of 1 per cent. of sodium nitrate. Calcizcm-Because it is possible that calcium could enter the effluent stream between the plant and the sampling point, the effect of 2 per cent. of calcium oxide was investigated. This produced a lowering of the recorded nitroglycerine content by about 2 per cent. TemfJeratwe-Wide fluctuations in the temperature of the effluent do not occur, and the small structure in which the polarograph is housed is controlled at the same temperature as that used to calibrate the instrument, i.e., 20" C.820 HOLLAND AND BENHAM It was, however, established that a 1” C rise above this temperature increased the peak height of the polarographic wave by about 1.5 per cent. Temperature compensation could be arranged in the electrical circuits, if necessary.METHOD REAGENTS- Base electrolyte-Dissolve 48-1 g of ammonium chloride, 7.5 g of potassium chloride, 0 4 g of sodium sulphite, Na2S0,.7H20, and 0 6 g of gelatine in warm distilled water and cool. Dilute the solution to 1 litre with distilled water and mix. Standard nitroglycerine solzdiovt-Dissolve 100 mg of nitroglycerine in 15 ml of methanol, transfer the solution to a l-litre calibrated flask, dilute to the calibration mark with distilled water and mix. This solution contains 100 p.p.m. of nitroglycerine. CALIBRATION- Establish that 100 ml of sample and 100 ml of electrolyte are discharged into the mixing chamber during each 15-minute cycle. Variable chokes are fitted in each delivery pipe to the mixing chamber, which should be adjusted, if necessary.Mix the base electrolyte and the standard nitroglycerine solution, adjust the temperature to 20” C, transfer the mixture to the polarographic cell via the mixing chamber and record the polarogram at a fixed potential of -1.00 volt. Dilute the standard nitroglycerine solution with distilled water to provide standard solutions containing 20, 50 and 75 p.p.m. of nitroglycerine, and similarly record the polaro- grams obtained from these solutions ; include a blank determination. PROCEDURE- sample by reference to the calibration graph. glycol dinitrate and is usually known as “A-type nitroglycerine.” Proceed as outlined under Calibration, and calculate the nitroglycerine content of the The nitroglycerine referred to throughout this paper does not contain any ethylene CONCLUSIONS The proposed procedure can be satisfactorily used for the continuous determination of nitroglycerine in a typical effluent at the levels of nitroglycerine normally present in these samples, i.e., less than 100 p.p.m., with an accuracy of 1 p.p.m. The polarogram can be recorded and interpreted visually, and the output of the recorder can be fitted with a mercury switch to operate an alarm system when the nitroglycerine content of the effluent exceeds 100 p.p.m. There is no obvious reason why a similar unit should not be used for the continuous determination of certain other constituents in flowing streams, e.g., metallic impurities. The unit described has been in continuous operation for 12 months. No serious instru- mental difficulties have been encountered, and periodic checks with solutions of known nitroglycerine content have confirmed the reliability of the method. The authors acknowledge the assistance given by Mr. D. Facer of this laboratory, who carried out most of the preliminary experiments. REFERENCE 1. Williams, A. F., and Kenyon, D., Talanta, 1959, 3, 160. Received August 16th, 1968

ISSN:0003-2654    

DOI:10.1039/AN9689300817

出版商:RSC    

年代:1968

数据来源: RSC