摘要:

SEPTEMBER, 1967 THE ANALYST Vol. 92, No. 1098 Radioactive Tracer Methods in Inorganic Trace Analysis: Recent Advances A Review"' BY J. W. McMILLAN (Analytical Sciences Division, A .E.R.E., Harwell, Didcot, Berks.) SUMI\IARY OF CONTENTS Introduction Isotopic methods Single isotope dilution Substoicheiometric isotope dilution Isotopic-exchange methods Concentration-dependent distribution methods Isotope-derivative and derivative-dilution methods Isotope-displacement methods Radio-release methods Nomisotopic methods Radiometric titration Assessment of the practical utility of radiometric methods INTEREST in the production of highly pure materials for the electronics and atomic energy industries has provided the main stimulus for the development of extremely sensitive methods of analysis for the determination of trace elements.Recently, several radiometric techniques have been developed for the determination of sub-microgram traces in materials. These techniques and their applications are reviewed, and an attempt is made to assess their practical utility. For convenience, in reviewing this topic, named radiometric techniques are classified in groups. Radiometric methods may be broadly divided into two types, involving the use of isotopic labelling, and non-isotopic labelling. Most recognised techniques fall mainly within one of these groups. Thus, isotope dilution, substoicheiometric isotope dilution, isotope exchange and saturation analysis are all isotopic methods ; whereas, radio release, radio displacement, and derivative dilution are of the non-isotopic variety.The one notable exception to this classification is radiometric titration, which spans both groups. Therefore, for the purpose of this review radiometric methods are classified under three main headings, Isotopic methods, Non-isotopic methods and Radiometric titration. ISOTOPIC METHODS All isotopic methods are dependent on the inverse proportionality of specific activity and inactive isotope concentration, i.e. , 1 Specific activity cc The division of methods within this group is solely dependent on how this inverse proportionality is utilised. Only methods that have been applied at the sub-microgram level are described in detail. A general appreciation of isotope-dilution methods can be obtained from published reviews.l s 2 The accuracy of isotope-dilution methods has been dealt with by Weiler.3 * Reprints of this paper will be available shortly.For details see Summaries in advertisement pages. [inactive isotope] ' 539540 MCMILLAN : RADIOACTIVE TRACER METHODS I N [Analyst, Vol. 92 SINGLE ISOTOPE DILUTION- The principle of isotope-dilution analysis is simple and well known. When an active tracer is mixed with its inactive isotopes the over-all specific activity If the specific activity of the tracer is So, will be reduced. where A , is the activity of a weight W, of tracer plus carrier. Then, if activity A , is added to a material containing a weight W, of inactive isotopes of the tracer, the specific activity will become S , where . - (2). A , .. .. .. s, =- W,+WX - - S, can be measured by recovering a pure fraction W , of (W, + Wx) and determining its activity A , .Wx is given by solution of equations (1) and (2): If W , is small compared with TV, then .. A0 w, = w, *- A2 The application of this principle to trace analysis is limited by the difficulty of measuring W2. Although this may be done at the microgram level, for instance by c~lorimetry,~ at lower levels this becomes increasingly difficult. The introduction of substoicheiometric isotope-dilution analysis5 has provided a solution to this problem. SUBSTOICHEIOMETRIC ISOTOPE DILUTION- In substoicheiometric isotope-dilution analysis, equal amounts of the element being determined are removed from the tracer solution and the solution formed by isotope dilution. The activities in these amounts, namely a, and a,, will be directly proportional to the specific activities of the tracer solution, and the solution formed by isotope dilution.Therefore, equation (3) can be re-written as w x = WO& 1 ) .. .. .. . . ( 5 ) . The essential pre-requisite for the use of this technique is the establishment of a means of removing equal amounts of the element being determined from two solutions, possibly of different volume, containing different total amounts of that element. This problem has been dealt with in great detail by Riiiieka and Sta1-9~9~,7 who established that equal amounts may be recovered by substoicheiometry. This technique involves the addition of equal substoicheiometric amounts of a reagent to the two solutions, and its quantitative stoicheiometric reaction, in each, with the element being determined.The product of the reaction must then be separated from the unreacted element. RiiiiEka and Starq5 conclude that only two types of reagent are suitable for substoicheio- metric isotope-dilution analyses at the sub-microgram level. These are organic reagents forming complexes that are easily separated from the unreacted element by solvent extraction or by ion exchange. Theory shows5y6 that in the metal chelate-extraction method, only reagents that are stable at high dilution and form complexes with high extraction constants are suitable for use at the sub-microgram level. Dithizone and cupferron meet these requirements. In the ion-exchange method it has been shown5s7 that only ligands forming complexes with high stability constants are satisfactory.Complexones, such as EDTA, meet this requirement, and also that of stability at low concentration. These methods have been used successfully for the substoicheiometric analysis of a number of elements. Dithizone extraction has been used for the determination of copper,s merc~ry,~JO silver11J2J3 and zinc,l* and cupferron has been used as a reagent for iron.15September, 19671 INORGANIC TRACE ANALYSIS : RECENT ADVANCES 541 Iron,' 9 1 3 indium16 and individual rare earths1' have been determined by the complexone ion-exchange method. Model experiments have demonstrated that iron,l5 indium16 and copper8 can be determined down to 10-log, and zinc14 and mercuryg to 10-9g, by these techniques. When dealing with practical situations the question of interferences inevitably arises.Selectivity can be obtained in several ways. For the mercury - dithizone complex and the iron - cupferron complex the extremely high values of their extraction constants, compared with those of other metals, suffice to give the necessary selectivity. For those metal complexes with intermediate or low values of extraction constant, masking or pre-separation can be used to increase selectivity. For instance, iodide has been used as a masking agent in the determination of copper,s EDTA in that of ~ i l v e r , ~ ~ ~ ~ ~ and diethyl dithiocarbamate in that of zinc.14 In the complexone ion-exchange method, selectivity is obtained in a similar way. For example, pre-separation of iron with acetonyl acetone,' and of indium with cupferron,lG is used in their determination by the complexone ion-exchange method.Another important feature is the value of the blank. For a satisfactory analysis the value of the blank should be reduced to less than a tenth of the sample value.5 This necessitates the use of specially purified reagents and ultra-clean apparatus. The problem of the blank value restricts the use of substoicheiometric isotope-dilution analysis to solutions, and easily soluble materials.5JS In a recent attempt to determine zinc in germanium oxide, sensitivity was limited by the 1 pg of zinc introduced in an initial fusion.19 The sensitivity of these methods is ultimately limited by the specific activity of the tracer solution used. Minor variants of substoicheiometric isotope-dilution analysis exist.These are multiple radioactive isotope dilution20 and double substoicheiometric isotope di1uti0n.l~ 9 l 3 In the first, different amounts (activities A and A,) of a tracer of known specific activity, So, are added to equal amounts of material containing Wx, the amount of inactive isotopes of the element being determined. The specific activities of the two isotope-dilution solutions formed are - SOW, A s ---1- - w, + w, w, + w,' ' . .. and .. .. sow, .. s - A2 - Y - W , + W x w,+wx ' * If equal amounts, W,, are removed from each, then their respective activities are a,= SxW - sow~x'w, .. .. .. . . ,-w1+ and Solution of these equations gives W, .. .. w, = Wl w2 (a37 - ax) w, ax - w, a y The optimal conditions in this type of analysis are given by making This method has been used for the determination of antimony in lead at w, > WX.l ... . W , = W , and the microgram level.20 The antimony is extracted into toluene from hydrochloric acid as its methyl-Giolet ion-association complex. In double substoicheiometric isotope dilution, identical amounts, A o, of a tracer are added to equal amounts of material containing Wx, the amount of inactive isotopes of the element being determined. To one of these is added a known amount of inactive isotopes, W,, and to the other a second amount, W,. The weight of element contributed by the tracer, W,, must be negligible compared with Wx, W, and W,.542 MCMILLAN RADIOACTIVE TRACER METHODS IN [Analyst, Vol. 92 The specific activities of the two isotope-dilution solutions are ... . .. . . s - A0 - w, + wx * . and If equal amounts, W,, are now removed from each solution their activities will be a , = S , W - ow, . . .. a . . . (13) - w x + w, and a 2 = S 2 W - *w, .. . . * . . . - w x + w, Solution of these equations for W , gives .. .. . . . . (15). a2 wz - a , Wl w, = a1 - a2 Optimal conditions are again given when W , = W , and W2> W,. This method has been used for the determination of silver by dithizone extraction, and iron by the complexone ion-exchange technique.13 An attempt has been made to use this principle combined with differential constant-voltage cou10metry.l~ $21 Model experiments have demonstrated that cadmium in the range 2 to 400 pg may be determined by this method. Extension of this technique to the sub-microgram level is being investigated.A brief comparison of these variants with normal substoicheiometric isotope-dilution analysis is of interest. Under optimum conditions the multiple isotope-dilution equation (10) simplifies to Wx=W1 3 - 1 .. . . . . (16) as W , a y < W , a,. In these a, and a y are, in effect, both measures of the specific activity of the tracer solution. In multiple isotope dilution the activity of the tracer solution a y is determined in the presence of the sample environment, the effect of W , being swamped by the large excess of tracer added. It is debatable that this is an advantage over measuring the tracer activity alone, ao. How- ever, as the condition W2> W , == W , applies, this advantage is partially lost as the environment will affect the measurement of a, far more than ay.A possible disadvantage of the multiple isotope-dilution method is that the substoicheiometric principle must hold over a wider metal-ion concentration range for one reagent concentration than does the normal method. When comparing double substoicheiometric isotope-dilution analysis with the normal method similar arguments can be put forward regarding the effect of sample environment. The fact that W,, the weight of isotopes in the tracer, must be negligibly small compared with W , and W , , means that the normal method is potentially at least 100 times as sensitive as the double-dilution method. An advantage of double dilution is that a tracer of unknown specific activity can be used providing that the condition mentioned above is fulfilled.ISOTOPIC-EXCHANGE METHODS- ( a x ) * * Equations (5) and (10) are therefore of the same form. In these methods exchange reactions of the following type are utilised- M*X + MY + (M,M*)X + (M,M*)Y (* indicates active isotope labelling). MX and MY must be readily separable by some technique, such as solvent extraction, and the most strongly complexing ligand, be it X or Y, must be associated stoicheiometrically with one portion of M before exchange. If the amount of M associated with Y, [MIy, is unknown, and that with X, [MI,, is known. then .. .. .. where D is the distribution of M* between organic solvent and aqueous phases, assuming MX is the extractable species.September, 19671 INORGANIC TRACE ANALYSIS : RECENT ADVANCES 543 One example of this type of method is the determination of mercury by isotopic exchange with mercury( 11) dibutyl phosphorothioate.22 Mercury-203 has been used to label the mercury(I1) dibutyl phosphorothioate or the ionic mercury. Mercury has been determined down to 10-7g by this method.Iodide in natural waters has been determined by an isotopic-exchange method.23 Iodide tracer is added to the water and, after conversion to iodine, is extracted into benzene. Up to this stage quantitative recovery is unnecessary as the yield can be determined by means of the tracer. The iodine in the benzene is exchanged with a known amount of iodide and the amount of iodine is evaluated from the distribution of the activity and the amount of iodide used.The method can be used to measure down to 1 pg of iodine per litre of natural water. Oxby and Dawson have developed a method for iodine based on the same exchange react ion. 24 The remarks made in the section on substoicheiometric isotope-dilution analysis regarding sensitivity, tracer specific activity, selectivity, reagent purity and stability, apparatus cleanliness, and blanks are of equal importance in isotopic-exchange methods. This is not unexpected as these two types of methods are closely related. CONCENTRATION-DEPENDENT DISTRIBUTION METHODS- Concentration-dependent distribution methods (C.D.D.) involve the use of calibration graphs; the latter relate the initial concentration of the determined species either to its distribution in a 2-phase system, or to that of a substance reacting with it.25 Substoicheio- metric radiometric methods is an alternative name for methods of this type.26 Kyrs has discussed the theoretical aspects of C.D.D.for four 2-phase systems25; they are heterogeneous sorption conforming to the Langmuir or Freundlich isotherms, extraction with a constant amount of an extracting agent, and sorption or extraction in the presence of a constant amount of a chelating agent. A theoretical treatment of the C.D.D. method has also been reported by De Voe.21 Comparisons of C.D.D. and substoicheiometric isotope dilution lead to several general conclusions concerning their relative merit and applicability.21 925 Firstly, substoicheiometric isotope dilution is preferred, providing a suitable reagent is available for the concentration level required, because it is the more reproducible method.Secondly, C.D.D. can be used to increase sensitivity when the limit of applicability of substoicheiometric isotope dilution has been reached, or when no suitable reagent exists for the use of the latter technique. Several C.D.D. methods have been developed to date. Filter-paper impregnated with Prussian blue absorbs caesium from low concentration solutions. This has been used for the determination of the specific activity of carrier-free, caesium-137 solutions.27 Model experi- ments have shown that sub-microgram amounts of cobalt, nickel, zinc, thallium, silver, cadmium and iron may be determined by absorption on to manganese dioxide impregnated filter-paper.13 This absorbent has been applied to the determination of cobalt in nickel oxide in the range 0.3 to 0.02 per cent.13 Cobalt may also be determined by ion-exchange separation in the presence of a limiting amount of EDTA.26 This technique can also be used for the substoicheiometric isotope-dilution determination of cobalt ; comparison with the C.D.D.method illustrates the superior sensitivity of the latter. Caesium may be deter- mined by extraction with calcium dipicrylaminate in nitrobenzene28 ; in certain circumstances preliminary separation of caesium with polyiodide is necessary. A further method, involving the use of a calibration graph, that might be classified as belonging to the C.D.D. type is a combination of isotope dilution and the ring-oven method.29 In this method an active tracer, containing a weight, W,, of the element being determined, is added to material containing an unknown amount of that element, Wx.A constant amount, W,, of (W, + W,) is precipitated and is separated from the remainder by the ring-oven method. I** + I- + 1.1" + I*-, The specific activity, S , of these 2 fractions is identical, so that . . .. - A1 W , + W x - W , W , -- Ax S = . . (18) where A, and A, are the activities of the portions containing (W, + W , - W,) and W,, respectively. Solution for W , gives Wx=W1($+1) -w() .. .. .. . . (19).544 MCMILLAN RADIOACTIVE TRACER METHODS IN [Analyst, Vol. 92 A A , A calibration graph can be obtained of -2 against Wx, for constant values of W , and W,. This should be a straight line. However, deviations from this occur, particularly at low values of W,.These deviations can be explained on the basis of a decrease in W , as W , becomes small, and is possibly a function of the solubility product of the precipitate. Calcium, phosphate and iron have been determined by this method with sensitivities ranging between tens and tenths of micrograms. From the definition of C.D.D. methods, it can be seen that non-isotopic methods could be devised in which this principle is used. Some isotope-derivative methods that do not fall strictly within the definition of the method could be regarded as non-isotopic C.D.D. met hods. NON-ISOTOPIC METHODS The principle of these methods is that the amount of activity derived from a labelled reagent when it reacts with the species being determined is related to the amount of that species.An alternative name for this group of methods is radio-reagent technique^.^^ Included in it are isotope-derivative, derivative-dilution, radio-release and isotope-displace- ment methods. ISOTOPE-DERIVATIVE AND DERIVATIVE-DILUTION METHODS- These two methods are considered together as the basis of each is identical. In these methods the species being determined is made to react quantitatively with a radioactive reagent of known specific activity to form a radioactive compound. The com- pound is separated and purified, and its activity is measured. Assuming that the recovery is quantitative, the amount of the species being determined, Wx, is calculated from . . (20) where A , is the activity of the compound formed, S , is the specific activity of the radioactive reagent, and E is the ratio of equivalent weights of the species being determined and the radioactive The two critical requirements in this procedure2 are to ensure quantitative reaction by correct selection of the reagent and reaction conditions, and to recover the pure radioactive compound.The latter can be achieved by direct quantitative separation of the reaction product ; by adding macro amounts of inactive product as carrier and correcting the recovery on the basis of the chemical yield; and by adding a small amount of reaction product labelled with a different tracer and correcting the recovery radiometrically. The distinction between isotope-derivative and derivative-dilution methods is associated with the recovery procedure adopted.It appears that isotope-derivative methods make use of the first procedure, and derivative dilution all three. In addition, it must be said that derivative dilution is a method introduced into organic analysis and, apparently, used exclusively in that fie1d.l s2 Paper-chromatographic procedures exist for the isotope-derivative determination of traces at and below the microgram leve1.31$32 Lead can be determined in the presence of 100-fold excesses of calcium, strontium, barium, iron, copper, zinc, manganese, molybdenum, cobalt, aluminium, mercury, thallium and magnesium by paper ~hromatography.~~ After separation, lead is determined by spraying the chromatogram with phosphate labelled with phosphorus-32, eluting the excess, and measuring the activity associated with the lead spot.By this procedure lead can be deter- mined down to 1 pg. In a modification of this technique duplicate chromatograms are produced for each sample.32 One is used for the identification of component positions by means of tracers. The other is used for the quantitative determination of those separated com- ponents by means of a radioactive precipitant. Sulphate labelled with sulphur-35 has been used as a precipitant for the determination of calcium down to 0.002 pg, strontium to 0.005 pg, barium to 0.008 pg and lead to 0.01 pg. Labelled phosphate has been used to determine beryllium and zirconium at the 10-4 pg level. Radioactive chromate, titanate, fluoro- zirconate, carbonate, sulphide and molybdophosphate may also be applied as precipitants. reagent.September, 19671 INORGANIC TRACE ANALYSIS : RECENT ADVANCES 545 Although it has been stated that quantitative reaction of the reagent and determined species is a pre-requisite of the isotope-derivative technique, methods based on partial reaction can be used, providing the extent of this is known or can be measured.Thus methods with rigid reaction conditions and calibration are of the first type,21y33 and those involving the use of isotopic labelling of the determined species are of the second.34 Cali- bration and isotopic labelling can be used in combination.21 A method involving the use of calibration is the combined gas-chromatographic separa- tion and radiometric determination of metal chlorides.33 In this method metal chlorides are passed through a column of broken brick impregnated with chlorine-36-labelled hydrogen chloride.During their passage through the column, the metal chlorides are separated and simultaneously labelled with chlorine-36 by exchange. On elution from the column the metal chlorides are determined radiometrically. The feasibility of the technique has been demon- strated by separating and determining iron, arsenic, germanium and tin chlorides. Application to a wide range of metals appears possible, and sensitivity is claimed to be of the order of Microgram amounts of steel components, nickel oxide impurities and copper - beryllium alloy have been separated by paper chromatography and determined by several methods of this type which rely on the use of calibration graphs.21 Iron, copper and man- ganese were determined by precipitation with phosphorus-32-labelled phosphate. Nickel, copper, cobalt, manganese, cadmium, zinc and mercury were determined by precipitation of their ferrocyanides followed by exchange with iron-59. 3M,[Fe(CN),] + 4Fe*3+ -+ Fe,*[Fe(CN),], + 6M2+ (* indicates active isotope labelling.) In addition, manganese was determined by oxidation to manganese dioxide, which was used as an absorbent for cobalt-60, the amount of activity absorbed being a measure of the amount of manganese dioxide present.The determination of cobalt by extraction with zinc - diethyl dithiocarbamate labelled with ~ulphur-35~~ is an example of a method in which partial reaction is measured by isotopic labelling. After extraction, the excess of reagent is removed, and the cobalt determined on the basis of the sulphur-35 activity associated with it.As the initial cobalt extraction is not quantitative it is carried out in the presence of cobalt-60 tracer, and the yield is corrected radiometrically. Isotopic labelling can be used not only to correct for non-quantitative reaction at the determination stage, but also for losses in preliminary separations. This technique has been used for certain paper-chromatographic separations that ultimately rely on isotope-derivative methods of the calibrated type for the final determination.21 g. ISOTOPE-DISPLACEMENT METHODS- The order of stability,35 or relative order of extraction,22 or extraction constants, for a series of metals with an organic reagent is utilised in this method.Thus an unknown amount of a metal forming a complex of high stability will release a proportionate amount of activity from a labelled metal complex of lower stability. If the released and complexed activities can be separated then the amount of one or both present can be used as a measure of the amount of the displacing metal. This principle is used in the determination of mercury by displacing silver, labelled with silver-llOm, from silver dibutyl phosphorothioate.22 Separation of the released and complexed silver is achieved by solvent extraction. The method has been used for the determination of as little as lO-'g of mercury. Another application is the determination of lead by displacement of thallium, labelled with thallium-204, from its diethyl dithio- carbamate complex.35. Extraction is again used for separation of the released and complexed metal.As the displacement is non-quantitative in this instance, lead-210 is used as an isotopic tracer so that a radiometric correction may be made. The method is sensitive to 0.05 pg of lead. Methods of this type are unlikely to be used if comparable isotopic-exchange methods are feasible. One reason is that correction for the effect of blanks for both the reagent and determined metals will be necessary. In addition selectivity through masking will be more difficult as both the reagent and determined metal must be unaffected by the masking agent.546 MCMILLAN : RADIOACTIVE TRACER METHODS I N [Analyst, Vol. 92 RADIO-RELEASE METHODS- Radio-release methods are dependent on the release of a radioactive species from a labelled carrier when it reacts chemically with the trace material concerned.The species released is not combined with the trace material and may,36~37~38~39 or may not,40,41742943 be isotopic with the carrier. The trace material may react directly36 3 3 7 9 3 8 9 3 9 y 4 0 9 4 1 9 4 3 or indirectly42 with the carrier; normally the reaction is stoicheiometric but need not be ~ 0 . ~ 0 The reaction must be highly selective, at least within the context of a particular application, and sensitivity is invariably attained by labelling the carrier with an isotope of high specific activity. This technique has found two main applications, namely, trace analysis in waters and in gases. Both vanadate and dichromate are used as non-active species for tracing water. In acid conditions silver labelled with silver-1 10m reacts with vanadate36 or dichromate39 to release silver ions, the reaction is stoicheiometric and the released activity is readily measured.The quoted sensitivities are 0.1 p.p.m. and 0.01 p.p.m. for vanadate and dichromate, respec- tively. Iron and chloride, normally present in natural waters, cause interference, but this can be eliminated by the use of masking agents. Dissolved oxygen in water can be determined by its reaction with thallium labelled with thallium-204 and measurement of the activity released3' 4T1+ 0, + 2H20 + 4Tlf + 40H-. The sensitivity attained was 0-2 p.p.m., although it is theoretically possible to measure 1 part in 1012 by this method if nitrate, which interferes, is absent.Several gas-analysis methods are based on the release of krypton-85 from krypton ~lathrates,~, 943 or materials whose surfaces have been impregnated with krypton-85.40941 Ozone down to 1 part in 10lo in air can be measured by the release of krypton-85 from its quinoline clathrate.43 Water will interfere, if not removed, but sulphur dioxide and hydrogen sulphide will not. The same reagent may be used for the indirect determination of sulphur This reacts with sodium chlorite ; the chlorine dioxide produced oxidises the clathrate, thus releasing krypton-85. The method is sensitive to 0.05 p.p.m. of sulphur dioxide ; it does, however, suffer appreciable interference from nitrogen dioxide and ozone. Krypton-85 can be incorporated into solids by bombardment with ionised krypton, or by diffusion of krypton into the surface at high temperatures and pressures.40 The krypton-85 is confined to a surface layer about lo3 to lo4 deep.This gas will then be released only by chemical attack of the surface, providing the material is not heated to temperatures exceeding 200" C below its melting-point. Kryptonated pyrolytic graphite and copper have been examined as materials for the analysis of oxygen.40 Copper is the more satisfactory material, the reaction rate being dependent on the reciprocal of the absolute temperature for incompletely oxidised surfaces. By adjusting the temperature, oxygen levels between lo5 and p.p.m. can be measured. A correction must be made for the interference by ozone.The converse of this reaction, namely the reduction of kryptonated platinum oxide, has been used for the determination of hydrogen.41 To determine hydrogen in air, it is neces- sary to dilute the platinum oxide with aluminium oxide. Under these conditions 10 p.p.m. of hydrogen can be measured. Normal constituents of air cause no interference. Sulphur dioxide in air will react with labelled potassium iodate releasing iodine-131 .38 By this means sulphur dioxide at the p.p.m. level may be measured in a 10-litre sample; however, hydrogen sulphide interferes significantly if present. The selectivity they possess is largely a result of the well defined, simple systems to which they are applied. Because the analysis is often performed on the sample directly, and because of the simple form of the labelled carrier in many instances, reagent blank problems are minimal.One of their main advantages is their ready adaptation to automatic operation. Present radio-release methods are highly sensitive. RADIOMETRIC TITRATION Radiometric titrations are those in which the course of the titration is followed by the change of radioactivity within the system. The activity can be introduced into the system in the form of a labelled titrand, titrant or indicator. In addition, one or more species can be titrated, and one or more titrants can be used. The number of radioactive tracers required is governed by the complexity of the system. As in conventional titrimetry direct or backSeptember, 19671 INORGANIC TRACE ANALYSIS : RECENT ADVANCES 547 titration may be used.Measurement of the change of activity within the system invariably requires a phase separation that occurs naturally in precipitation titrimetry; in all other systems auxiliary methods of separation must be used. The features of radiometric titration are dealt with thoroughly in reviews by Braun and T01gyessy~~ 9 4 5 and radiometric titration procedures involving precipitations, complex formation, solid indicators, redox reactions and constant-current coulometry are all described. It is evident that few of these methods can at present be used in trace analysis, although the chelate-ion exchange technique and the complex-extraction technique can both be used at or just below the microgram level. The lower limit attained by precipitation methods is normally of the order of a milligram, but zirconium44 and neodymium46 have been deter- mined at about the 10 pg level.All other methods in their present state of development are of little interest for trace analysis because of their lack of sensitivity. Radiometric titration was first extended to the microgram level by the introduction of the complex-extraction technique.47 This method is based on the formation of metal chelates that can be separated from unreacted metal ions by solvent extraction. The course of the titration is followed by measuring the activity in both phases. Cobalt has been determined down to the 0.2 pg level by using dithizone as titrant and cobalt-60 as an isotopic tracer.47 Only metals having a higher extraction constant than cobalt interfere, and then only if they are present in greater than 10-fold excess.Zinc has been determined similarly at the 0.3 pg leveL48 Non-isotopic titration of a metal ion can be carried out, providing its extraction constant is higher than that of the indicator, for example, zinc has been determined down to 0-4pg in this way with cobalt-60 as indicator.47 The most recent advance in the use of radiometric titration for trace analysis has been the introduction of the chelate-ion exchange technique.49 This method is based on the reaction of metal ions with EDTA to form a chelate that may?be separated from unreacted metal ions by the use of ion exchange. The practicability of the method has been demon- strated at the microgram level by using model systems.Indium-114m has been used as an isotopic tracer for the determination of indium, and as a non-isotopic tracer for the determination of cobalt. For a metal to be of use as a non-isotopic tracer the stability of its chelate must be lower than that of the metal being determined. Radiometric titration methods have been little used compared with many of the methods described earlier in this review. This is most probably because radiometric titration procedures are often, in effect, multiple versions of the other methods. However, the presence of unexpected interferences will probably be revealed more readily by multiple, rather than single, measurement methods. ASSESSMENT OF THE PRACTICAL UTILITY OF RADIOMETRIC METHODS When attempting to evaluate the worth of radiometric methods of trace analysis a comparison with activation analysis inevitably ensues.Activation analysis is preferred because of the absence of reagent blank problems. However, reasons for using radiometric methods become apparent when the limitations of activation analysis are enumerated. The sensitivity of activation analysis is limited by the nuclear properties of the element being determined, together with the flux and energy of the activating species. When the limit set by these parameters is reached, extra sensitivity may be available through radio- metric methods in certain circumstances. Isotopic methods will be potentially capable of increasing sensitivity if a high specific activity tracer can be obtained, perhaps by using irradiation times exceeding those practicable for activation analysis, or if a cm-ier-free tracer can be obtained, for instance through fission. In addition, it is feasible that sersitive methods could be developed for elements having unfavourable activation cross-section s through the use of non-isotopic radiometric methods.Invalidation of activation analpis because of interference through secondary nuclear reactions is a further circumstance in which radiometric analysis might be substituted with advantage. For routine analysis of simple, well defined materials radiometric analysis possesses certain intrinsic advantages. With the availability of high specific activity tracers these methods are sensitive, often rapid and simple, and are potentially amenable, in many instances, to automation.5 48 MCMILLAN The ultimate test of any analytical method is whether or not it is applied as a working Of the methods reviewed certain substoicheiometric isotope- The future method to a practical situation.dilution and radio-release procedures have already qualified on this basis. should see the emergence of many more. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. REFERENCES Alimarin, I. P., and Bilimovitch, G. N., I n t . J . Appl. Radiat. Isotopes, 1960, 7 , 169. “Radioactive Isotope Dilution Analysis,” R.C.C. Review, 2, U.K. Atomic Energy Authority, The Radiochemical Centre, Amersham, Bucks, April, 1964.Weiler, H., I n t . J . Appl. Radiat. Isotopes, 1961, 12, 49. Ralph, W. D., Sweet, T. R., and Mencis, I., Analyt. Chem., 1962, 34, 92. RBiiCka, J., and Star?, J., Atom. Energy Rev., 1964, 2, 3. Star$, J., and RbiiEka, J., Ibid., 1961, 8, 775. RbiiCka, J., and Star$, J., Ibid., 1962, 9, 617. -- , Ibid., 1961, 8, 535. Suzuki, N., Scient. Rep. Tohoku Univ., Ser. 1, 1959, 43, 161. -, J . Chem. SOC. Japan, 1959, 80, 370. De Voe, J. R., Editor, “N.B.S. Technical Note, 248,” U.S. Department of Commerce, National -, Editor, “N.B.S. Technical Note, 276,” U.S. Department of Commerce, National Bureau of Starf, J., and RbiiCka, J., Talanta, 1961, 8, 296. Starg, J., RBiiCka, J., and Salamon, M., Ibid., 1963, 10, 375. RBiiCka, J., and Starf, J., Ibid., 1964, 11, 691. Prasilova, J., Ibid., 1966, 13, 1567.Star$, J., and RfiiiEka, J., Ibid., 1964, 11, 697. Ballaux, C., Dame, R., and Hoste, J., Analytica Chim. Acta, 1966, 35, 141. Zimakov, I. E., and Rozhavsky, G. S., Zav. Lab., 1958,24, 922; Ind. Lab., 1958,24, 1030. De Voe, J. R., Editor, “N.B.S. Technical Note, 404,” U.S. Department of Commerce, National Handley, T. H., Analyt. Chem., 1964, 36, 153. Richter, H. G., Ibid., 1966, 38, 772. Oxby, C. B., and Dawson, J. B., in “Proceedings of the Symposium on Radiochemical Methods, KyrS, M., Analytica Chim. Acta, 1965, 33, 245. Landgrebe, A. R., McClendon, L. T., and De Voe, J. R., in “Proceedings of the Symposium on Kyrg, M., and Kadlecovk, L., Analytica Chim. Acta, 1965, 33, 481. -~ , Ibid., 1966, 36, 215. Weis;, H., and Klockow, D., Mikrochim. Acta, 1963, 1082. Reynolds, S. A., and Leddicotte, G. W., Nucleonics, 1963, 21 (S), 128. van Erkelens, P. C., Analytica Chim. Acta, 1961, 25, 570. Welford, G. A., Choitis, E. L., and Morse, R. S., Analyt. Chem., 1964, 36, 2350. Tadmor, J., Ibid., 1964, 36, 1565. van Erkelens, P. C., Analytica Chim. Acta, 1962, 26, 46. -, Ibid., 1962, 26, 32. Gillespie, A. S., and Richter, H. G., Analyt. Chem., 1964, 36, 2473. Richter, H. G., and Gillespie, A. S., Ibid., 1962, 34, 1116. Ross, H. H., and Lyon, W. S., in “Proceedings of the Symposium on Radiochemical Methods, Salzburg, 1964,” Volume 11, I.A.E.A., Vienna, 1965, p. 285. Richter, H. G., and Gillespie, A. S., Analyt. Chem., 1965, 1146. Chleck, D., Maehl, R., Cucchiara, O., and Carnevale, E., I n t . J . Appl. Radiat. Isotopes, 1963, 14, 581, 593 and 599. Chleck, D., in “Proceedings of the Symposium on Radiochemical Methods, Salzburg, 1964,” Volume 11, I.A.E.A., Vienna, 1965, p. 273. Hommel, C. O., Bersin, R. L., Filipov, A. M., and Brousaides, F. J., U.S. Report No. NYO-2767/ Tracer Lab. Inc., Waltham, Massachusetts Clearing House for Federal Scientific and Technical Information, August, 1962. Baker, P. S., Lafferty, R. H., Gerrard, M., and Rupp, A. F., Editors, “Isotopes and Radiation Technology,” U.S.A.E.C., 1963, 1, 53. Braun, T., and Tolgyessy, J., Talanta, 1964, 11, 1277. Tolgyessy, J., Jesenak, V., and Braun, T., in “Proceedings of the Symposium on Radiochemical Methods, Salzburg, 1964,” Volume 11, I.A.E.A., Vienna, 1965, p. 199. Maryanov, B. M., and Serebrennikov, V. J., Zh. Analit. Khim., 1963, 18, 58. Duncan, J. F., and Thomas, F. G., J . Inorg. Nucl. Chem., 1957, 4, 376. Spitzy, H., Mikrochim. Acta, 1960, 789. Starg, J., RbiiEka, J., and Zeman, A., Talanta, 1964, 11, 481. > , Talanta, 1961, 8, 228. -- Bureau of Standards, Washington, D.C., 1964, p. 10. Standards, Washington, D.C., 1966, p. 111. Bureau of Standards, Washington, D.C., 1966, p. 141. Salzburg, 1964,” Volume 11, I.A.E.A., Vienna, 1965, p. 229. Radiochemical Methods, Salzburg, 1964,” Volume 11, I.A.E.A., Vienna, 1965, p. 321. Received March 3rd, 1967

ISSN:0003-2654    

DOI:10.1039/AN9679200539

出版商:RSC    

年代:1967

数据来源: RSC

摘要:

Analyst, September, 1967, Vol. 92, $fi. 549-552 549 Rapid Determination of Arsenic in Copper and Brass Neutron-activation Analysis BY A. P. GRIMANIS AND A. G. SOULIOTIS (Chemistry Department, Nuclear Research Center, “Democritos,” Athens, Greece) A rapid neutron-activation analysis method is presented for the deter- mination of microgram amounts of arsenic in copper and brass. Neutron- irradiated samples are dissolved in hot perchloric acid in the presence of an arsenic carrier. The resulting solution is made 7.4 M with respect to perchloric acid and 1.2 M with respect to hydrobromic acid. The arsenic-76 and the arsenic carrier are extracted quantitatively into benzene. After washing the benzene layer with 7.4 M perchloric acid - 1.2 M hydrobromic acid mixture, arsenic-76 is counted directly in the organic phase.The radio- chemical purity of the arsenic-76 extract was demonstrated by using a multi-channel analyser, together with decay studies. After the irradiation, the time required to complete the analysis is less than 15 minutes, which is at least four times shorter than that previously reported for activation analysis methods. The proposed method was successfully applied to the determination of arsenic in copper and brass primer cups and primer pockets (arsenic content 1 to 50 p.p.m.). THE usual methods for the separation of arsenic are based either on the distillation of arsenic(II1) chloride or on the precipitation of arsenic as sulphide with hydrogen sulphide from hydro- chloric acid s01ution.l~~ The distillation method is time consuming, while the precipitation of arsenic with hydrogen sulphide is interfered with by many elements, giving difficultly soluble sulphide salts.Recently, several extraction methods for separating arsenic have appeared in the l i t e r a t ~ r e . ~ , ~ , ~ , ~ , ~ y8 Several workers combined extraction procedures and spectrophotometric methods to determine arsenic in copper and copper alloys.1° s11Y12 The determination of arsenic by activation analysis in different materials is also des- ~ r i b e d . ~ ~ ~ ~ ~ J ~ ~ ~ ~ ~ ~ ~ Non-destructive neutron-activation analysis, in conjunction with y-scintil- lation spectrometry, cannot be used successfully to determine small amounts of arsenic in copper or copper alloys. The neutron-induced copper-64 of the copper or copper-alloy samples has an annihilation energy peak at 0.51 MeV, which is very close to that radiation characteristic of the neutron-induced arsenic-76, and constitutes a serious interference in this type of analysis.Thus to determine arsenic in such samples by neutron-activation analysis it is necessary to separate arsenic-76 radiochemically from copper-64 produced during the irradiation of a copper sample. Yajimals and co-workers applied extraction techniques and neutron-activation analysis to determine gold, arsenic and antimony in copper. Grimanis and Leddicottel9 developed a rapid extraction procedure for separating arsenic from other elements by using the system benzene - 7.0 M perchloric acid - 1-0 M hydrobromic acid, and applied this separation technique in an activation-analysis method for determining arsenic in copper, copper alloys and organic materials.The arsenic, after quantitative extraction into benzene, was re-extracted into the aqueous phase and reduced to the metal for counting. However, the time required for this and other methods when using precipitation techniques as final steps for the radio- chemical separation of arsenic is about 1 hour (for the formation of the precipitate, filtration, washing, drying and weighing). In this paper a modification of the Grimanis and Leddicotte method is described, in which the time required for the last step is eliminated as arsenic-76 is counted directly in the benzene phase. The total time taken for the analysis is reduced to 15 minutes.550 GRIMANIS AND SOULIOTIS : RAPID DETERMINATION OF ARSENIC [Anahst, Vol.92 EXPERIMENTAL Preliminary experiments were carried out to establish the modified radiochemical pro- cedure for the separation of arsenic. For the extractions an arsenic carrier and an arsenic radiotracer, prepared as described below under “Reagents,” were added to the aqueous phase. In one series of experiments we studied the effect on the extractability of arsenic into benzene, of copper, zinc, brass or iron dissolved in the 7-4 M perchloric acid - 1.2 M hydrobromic acid mixture. It was found that the presence of up to 80 mg of these metals and metal alloy in the acid mixture does not interfere with the quantitative transference of arsenic into the organic phase (Table I). The benzene layer containing the arsenic was washed twice with the same volume of 7.4 M perchloric acid - 1.2 M hvdrobromic acid. I t was found that less than 0.2 per cent.of the akenic present in the organic phase was removed by this washing (Table I). TABLE I EXTRACTION OF ARSENIC INTO BENZENE FROM 7.4 M PERCHLORIC ACID - 1.2 M HYDROBROMIC ACID MIXTURE CONTAINING METAL OR ALLOY Aqueous phase, 25 ml of 7.4 M perchloric acid - 1.2 M hydrobromic acid con- taining metals or alloy as shown and 0.4 mg of arsenic per ml; organic phase, benzene; phase ratio, 1 : 1 (v/v) ; extraction time, 2 minutes; equilibration time, 1 minute Arsenic remaining in thc organic phase after washing twice with the same volume of 7.4 M HC10, - 1-2 M HBr, per cent.* Metals or alloy dissolved in the aqueous phase, 80 mg Arsenic extracted in the organic phase, per cent.* None Copper Zinc Brass Iron 99.9 99.9 99.8 99.8 99.8 > 99.6 > 99.6 > 99.6 > 99.6 > 99.6 * All figures are mean values of duplicate analyses.Another series of experiments was carried out to determine the procedure for the control of the proposed neutron-activation analysis method. Samples of copper and brass were analysed for arsenic by this modified neutron-analysis method and by the method described by Grimanis and Leddicotte. The results found for arsenic by the two methods are shown in Table I1 and it can be seen that they give comparable results. TABLE I1 DETERMINATION OF ARSENIC IN COPPER AND BRASS BY THE TWO NEUTRON-ACTIVATION ANALYSIS METHODS Concentration of arsenic, p.p.m., found by- Grimanis and Sample Leddicotte’s methodla Proposed method Brass .. . . . . 15.6, 15.7 15.3, 15.3 Copper . . . . . . 4-8, 5-1 4-6, 5.0 Brass . . .. . . 45.8, 47- 1 46-5, 47.7 Copper . . . . .. 1.2, 1.3 1.2, 1.3 METHODS REAGENTS- The materials used were of analytical-reagent grade. Arsenic carrier solation-Prepare a solution to contain 10 mg of arsenic per ml by dissolving 1.774 g of arsenic pentoxide in distilled water and making up to 100 ml in a calibrated flask. Standard arsenic solation-Prepare a solution to contain 50 pg of arsenic per ml by trans- ferring, with a micropipette, 0.5 ml of the arsenic carrier solution into a 100-ml calibrated flask and diluting to volume with distilled water. Arsenic radiotracer-Weigh 177 mg of arsenic pentoxide (As20,.2H,O) into aluminium foil and irradiate it by using the pneumatic transfer system at a neutron flux of 2-2 x 1012September, 19671 IN COPPER AND BRASS BY NEUTRON-ACTIVATION ANALYSIS 551 neutrons per cm2 per second for 15 minutes.After 1 hour, dissolve the irradiated arsenic pentoxide in distilled water, transfer the resulting solution to a 100-ml calibrated flask and dilute to volume with water. Use 1 ml of radiotracer for each preliminary experiment containing a total y-activity of 7.5 x lo6 counts per minute. Perchloric acid, 70 per cent. w/w. Nitric acid, sp.gr. 1.40. Hydrobromic acid, 48 per cent. w/w. Benzene. Rinsing solution (7-4 M perchloric acid - 1-2 M hydrobromic acid)-Make a mixture consisting of 190 ml of 70 per cent. perchloric acid, 40 ml of 48 per cent.hydrobromic acid and 70 ml of water. IRRADIATION- Weigh about 80 mg of each copper or brass sample into polythene tubes (10 x 20 mm). Transfer with a pipette 0 5 m l of standard arsenic solution into another polythene tube of the same size. Heat-seal the tubes and pack them in plastic sheets. Irradiate the targets for 35 minutes in a neutron flux of 2.2 x 10l2 neutrons per cm2 per second. (We used the pneumatic transfer system of the “Democritos” swimming-pool reactor.) PROCEDURE- After irradiation, open the samples and standards behind a shield. Transfer each copper sample into a 50-ml beaker containing 1 ml of arsenic carrier solution and 3 to 4 drops of concentrated nitric acid. Add 5 m l of 70 per cent. perchloric acid and heat the beaker on an electric plate until white vapour of perchloric acid is evolved.After dissolution, add 11 ml of 70 per cent. perchloric acid and 5-5 ml of water, and cool the beaker in an ice-bath. When the solution is at room temperature, add 3-5 ml of 48 per cent. hydrobromic acid solution and decant the solution into a 100-ml separating funnel containing 25 ml of benzene. Rinse the beaker with 5 ml of the 7.4 M perchloric acid - 1.2 M hydrobromic acid mixture, with the aid of a wash-bottle. Shake the funnel for 2 minutes. Leave the two phases to equilibrate for 1 minute and discard the aqueous phase. Wash the organic phase with an equal volume of 7-4 M perchloric acid - 1.2 M hydrobromic acid mixture and, after equilibration, discard the washing. Repeat the back-washing of the organic phase with the acid mixture and discard the washing.Transfer by pipette 5 ml of the organic phase into a 5-ml calibrated flask. Transfer with a micropipette 1 0 0 ~ 1 of standard arsenic solution into a 50-ml beaker containing 1 ml of arsenic carrier solution. Add 16 ml of 70 per cent. perchloric acid solution and 4.5 ml of water. Cool the solution to room temperature in an ice-bath and add 3.5 ml of 48 per cent. hydrobromic acid solution. Follow the procedure used for the copper sample. DETERMINATION OF RADIOACTIVITY- Take a count of the calibrated flasks containing the arsenic-76 from the irradiated copper sample or from the arsenic standard with a sodium iodide (thallium-activated) well-type y-scintillation counter (2 x 2% inches). IDENTIFICATION AND CONTROL OF RADIOCHEMICAL PURITY OF ARSENIC-76- To identify arsenic-76 and to be assured of its radiochemical purity a y-ray spectrometric examination and half-life determination were made of the radioisotope contained in the organic phase. The y-ray spectrometric examination was carried out to determine whether or not there was any y-ray contaminant in the benzene phase from the analysed sample.Thus, 15 minutes after withdrawal from the reactor, a portion of the organic phase containing the separated arsenic-76 was quantitatively examined by scintillation counting with a 400-channel pulse height analyser. The energy spectrum examined showed no indication of any radionuclide emitting y-rays, except arsenic-76 at 0.56 MeV.20y21 The half-life of the isolated radioelement was determined by plotting a decay curve for the arsenic-76 contained in the organic phase on “semilog” graph paper and was obtained from the slope of the straight line calculated by the method of least squares.A value of 26.25 hours, with a standard error of k0.23 hours was found. This served as an additional indication of the radiochemical purity of the isolated radionuclide.552 GRIMANIS AND SOULIOTIS RESULTS AND DISCUSSION A neutron-activation analysis method has been developed for determining microgram amounts of arsenic in copper and brass. The method is characterised by its rapidity and simplicity, being convenient for a series of analyses. At least 1 pg of arsenic can be deter- mined by this method. However, the limits of quantitative measurement could be extended downward if the duration of irradiation or the neutron flux, or both, were increased.TABLE I11 DETERMINATION OF ARSENIC IN PRIMER CUPS AND PRIMER POCKETS BY NEUTRON-ACTIVATION ANALYSIS Primer cups made of copper- British primer cups .. Italian primer cups. . .. Greek primer cups . . .. Austrian primer cups . . Primer cups made of brass- German primer cups, 6.45 mm Italian primer cups, 6-45 mm Greek primer cups, 6-45 mm Greek primer cups, 5.45 mm Arsenic, p.p.m. . . 1.2, 1.3 .. .. . . 1.9, 2.0 .. ,. . . 4-9, 5.2 , . . . 18.3, 18.4 .. . . . . . . . . 3.8, 4.3 . . . . .. 21.9, 23.7 . . . . .. 32.6, 33.8 .. .. .. 46.5, 47.7 Primer pockets made of byass- German primer pockets, 6.45 mm . . . . . . 2.6, 2.9 Italian primer pockets, 6-45 mm . . ,.. . 15.3, 15.3 Greek primer pockets, 6.45 mm . . . . .. 32.5, 33.4 Greek primer pockets, 5.45 mm . . . . .. 17.6, 18-0 The proposed neutron-activation analysis method was applied successfully for the deter- mination of arsenic in copper and brass primer cups and primer pockets, and results for duplicate analyses are given in Table 111. The concentration of arsenic found ranged from 1.3 to 47.1 p.p.m., depending on the quality of copper and brass used in each analysed sample. I t seems to us that the method may also be applied successfully for the determination of arsenic in zinc or iron-based alloys. The authors express their thanks to Maria Vassilaki-Grimani and to Basil Papadopoulos for their capable assistance in developing this method. 1. 2. 3. 4. 5. 6. 7.8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. REFERENCES Hillebrand, W. F., and Lundel, G. E. F., “Applied Inorganic Analysis,” Second Edition, John Wiley & Sons Inc., New York, 1953, p. 260. Sandell, E. B., “Colorimetric Determination of Traces of Metals,” Third Edition, Interscience Publishers, New York and London, 1959, p. 175. Irving, H. M., Q. Rev. Chem. Soc., 1951, 5, 200. Morrison, G. H., and Freiser, H., “Solvent Extraction in Analytical Chemistry,” John Wiley & Sons Inc., New York, 1957, p. 193. Morrison, G. H., and Freiser, H., Analyt. Chem., 1958, 30, 632. , Ibid., 1960, 32, 37R. -- , Ibid., 1962, 34, 64R. Morrison, G. H., Ibid., 1964, 36, 93R. Freiser, H., Ibid., 1966, 38, 131R. Scholes, I. R., and Waterman, W. R., Analyst, 1963, 88, 374. Tanaka, K., Japan Analyst, 1961, 10, 612. Karanov, R. A., and Karolev, A. N., Zh. Analit. Khiun., 1965, 20, 639. Bock-Werthmann, W., and Schulze, W., “Actiwierungs Analyse,” A tomkernenergie-Documentation, Beim Gmelin Institute, Report AED-C-14-1, Hahn-Meitner Institut fur Kernforschung, Berlin, 1961. Bock-Werthmann, W., Ibid., Report AED-C-14-2, Hahn-Meitner Institut fur Kernforschung, Berlin, 1963. -, Ibid., Report AED-C-14-3, Hahn-Meitner Institut fur Kernforschung, Berlin, 1964. Bowen, H. J. M., and Gibbons, D., “Radioactivation Analysis,” Clarendon Press, Oxford University Lyon, W. S., “Guide to Activation Analysis,” D. Van Nostrand Co. Inc., Toronto, New York Yajima, S., Kamemoto, Y., Shiba, K., and Onoda, Y., J . Chem. SOC. Japan, Pure Chem. Sect., Grimanis, A. P., and Leddicotte, W. G., Trans. Amer. Nucl. Soc., 1962, 5, 195. Grouthamel, C. E., Editor, “Applied Gamma-ray Spectrometry,” Pergamon Press, New York, Heath, R. L., “Scintillation Spectrometry, Gamma-ray Spectrum Catalogue, ’’ U S . Atomic Energy Commission Report IDO-16880, Second Edition, 1964. Received November 23rd, 1966 , -- Press, 1963. and London, 1964. 1961, 82, 38; Chem. Abstr., 1961, 55, 8160e. Oxford, London and Paris, 1960, p. 194.

ISSN:0003-2654    

DOI:10.1039/AN9679200549

出版商:RSC    

年代:1967

数据来源: RSC

摘要:

Analyst, September, 1967, Vol. 92, pp. 553-557 553 The Determination by Atomic-absorption Spectroscopy of Several Elements, including Silicon, Aluminium and Titanium, in Cement BY LUIS CAPACHO-DELGADO AND DAVID C. MANNING (The Perkin-Elmer Cor$oration, Norwalk, Connecticut) A method is described for the rapid determination of silicon, aluminium] titanium, magnesium, iron, manganese, sodium, potassium, lithium and strontium in cement. The sample preparation consists of a hydrochloric acid digestion, followed by filtration] sodium carbonate fusion of the residue and aqueous dissolution of the fusion. National Bureau of Standards cements are treated as unknowns to establish the accuracy of the method. The nitrous oxide - acetylene flame and high brightness hollow-cathode lamps are used to determine several of the refractory oxide-forming elements. THE currently accepted methods of cement analysis for the metal constituents involve relatively long and tedious procedures.Gravimetric, titrimetric and colorimetric methods require several weighings, separations or titrations, or combinations of these. Emission flame- photometric methods1 12 33 y4 have been proposed for several of the elements (sodium, potassium, magnesium and manganese). These methods generally specify lengthy separation steps to eliminate interferences. Atomic absorption has also been proposed for determining sodium, potassium, calcium, magnesium and manganese in cement .5 j 6 Silicon, aluminium and titanium in siliceous materials have not been reported previously as being determined by atomic absorption.EXPERIMENTAL INSTRUMENTATION- This work was carried out with a Perkin-Elmer atomic-absorption spectrophotometer, model 303, equipped with a recorder read-out. A nitrous oxide - acetylene burner was used for the determination of aluminium, silicon, titanium and strontium, and an air - acetylene burner for the other elements. The nitrous oxide - acetylene burner flame slot was 0.019 inch wide by 3 inches long.* The air - acetylene burner flame slot was 0.015 inch wide by 4 inches long. High-brightness lamps' j 8 were used to improve the signal-to-noise ratio for silicon and titanium. Scale-expansion (3 times) was used for the silicon and titanium determinations. The instrument settings are summarised in Table I. TABLE I INSTRUMENT SETTINGS FOR THE PERKIN-ELMER ATOMIC-ABSORPTION SPECTROPHOTOMETER, MODEL 303 Wavelength] Lamp current, Element A Slitt mA Aluminium Titanium1 Silicon: .. Magnesium Iron . . Manganese Sodium . . Potassium Lithium Strontium . . . . .. . . .. .. .. .. .. . . . . . . . . .. . . .. .. .. .. .. 3092.6 3642.7 2516.1 2852.1 2483.3 2794.8 5890.0 7664.9 6707.8 4607.3 30 40 30 6 40 20 5003 5003 20 10 t Slit settings are given, followed by the approximate spectral slit-width 1 Scale-expansion 3 times used. 5 Osram discharge lamp used. provided by the setting. * Perkin-Elmer now recommends the use of a slot with dimensions 0.019 inch wide by 2 inches long.554 CAPACHO-DELGADO AND MANNING : DETERMINATION BY [Anahst, VOl. 92 STANDARDS PREPARATION- Silicon standards consisted of aqueous solutions of sodium silicate.Titanium standards were made with titanium sulphate and iron, magnesium, manganese, sodium, potassium, aluminium, strontium and lithium standards with the chloride or nitrate of the element. It was found necessary to add about the same amount of calcium chloride and hydrochloric acid to the aluminium standards as was present in the sample solutions. The titanium standards likewise contained calcium chloride and hydrochloric acid, as well as about the same amount of aluminium chloride as was present in the sample solutions. Strontium standards were prepared with additions of 1 mg of potassium per ml (as the chloride). Reasons for these additions will be given later. To evaluate the accuracy of the technique, National Bureau of Standards standard cement samplesg were considered as unknowns, and were analysed with aqueous solutions of the elements as standards.In the subsequent discussion, the N.B.S. cement standards will be called “samples.” SAMPLE PREPARATION- The sample preparation procedure is essentially the first part of the A.S.T.M. referee method for silica determination,1° together with a sodium carbonate fusion to bring the silica into solution. Transfer 1.00 g of the sample into a 200-ml evaporating dish and moisten it with 10 ml of cold water, swirling the dish. Add 10 ml of concentrated hydrochloric acid and allow it to digest with gentle heat and agitation until dissolution is complete. Evaporate the solution to dryness under an infrared heat lamp. Without further heating, add 10ml of dilute hydrochloric acid (1 + l ) , cover the dish and allow it to digest for 10 minutes. Dilute the solution with 10 ml of hot water, filter with a Whatman No.541 filter-paper, or equivalent, and wash the separated silica with hot hydrochloric acid (1 + 99). Transfer the filtrate into a 100-ml calibrated flask, dilute to the mark with water and use this solution for the determination of aluminium, iron, titanium, magnesium, manganese, sodium, potassium, strontium and lithium. Transfer the paper containing the residue into a platinum crucible; dry and carefully ignite the paper. Weigh 1 g of anhydrous sodium carbonate on to a watch-glass and transfer about three-quarters of it into the crucible and mix carefully; then cover the mixture with the remaining sodium carbonate.Cover the crucible and carefully heat the mixture until fusion is complete (10 to 15 minutes). After cooling, dissolve the melt in water, transfer the solution to a 100-ml calibrated flask and dilute to the mark with water. This will be designated the “fusion solution.’’ Titanium, manganese and lithium are determined in the undiluted acid-soluble solution. The aluminium, iron, strontium, sodium and potassium are determined by using a dilution (10 times with water) of an aliquot of the acid-soluble solution in order to bring their con- centrations within a convenient analytical range. Silica is determined by using a dilution (40 times with water) of an aliquot of the fusion solution. This will be designated the “acid-soluble solution.” TABLE I1 DETERMINATION OF SILICON DIOXIDE, ALUMINIUM OXIDE, MANGANESE ( 111) OXIDE AND IRON(III) OXIDE IN CEMENTS Silicon Aluminium Manganese(II1) Iron( 111) dioxide by- oxide by- oxide by- oxide by- - - - Atomic N.B.S., absorption, N.B.S., absorption, N.B.S., absorption, N.B.S., absorption, per per cent.per per ccnt. per per cent. per per cent. Sample cent. f-A-, cent. cent. f-A-, cent. r----A-----, 177 21.92 21.6 21.2 - - - - 0.05 0.046 0.048 2.39 2-58 2.63 2-38 1011 21.03 20.9 20.9 5-38 5.44 5.35 5.35 0.03 0.039 0-035 2.07 2.12 2-38 2-10 1013 24.17 23.5 24-1 3.31 3-40 3.34 3.34 0.05 0.053 0.049 3.07 3.15 3-19 3.01 1014 19-49 18.9 19-1 6.39 6.41 6.21 6.25 0.07 0.071 0.064 2.50 2.60 2.72 2.55 1015 20-65 20.6 19.8 5.04 5-04 4-94 4.94 0.06 0.061 0.056 3.27 3.38 3.35 3.22 1016 21.05 20.0 20.0 4.97 4.53 4.48 4.58 0.04 0.043 0.040 3-71 3.81 3.71 3.61September, 19671 ATOMIC-ABSORPTION SPECTROSCOPY OF SEVERAL ELEMENTS 555 About 75 to 90 per cent.of the titanium is in the acid-soluble solution, the remainder being in the fusion solution. With this exception the separations are sufficiently complete, so that silicon can be determined in the fusion solution and the other elements in the acid- soluble solution. Titanium is determined in each solution and the results totalled. If incom- plete separation of any element is suspected, it is possible to check each solution for the element concerned, except for sodium, as this is added as the carbonate in the fusion procedure. ANALYTICAL RESULTS The results for silica are given in Table 11.The numbers in the first column identify the N.B.S. cement samples. Certificate values are also given, followed by atomic-absorption results obtained from individually weighed and prepared samples. O+ 10 I00 I 000 I0,OOO I I I TABLE I11 DETERMINATION OF MAGNESIUM OXIDE, POTASSIUM OXIDE, SODIUM OXIDE AND LITHIUM OXIDE I N CEMENTS Sample Magnesium oxide by- --7Gz7 N.B.S., absorption, Per per cent. Sample cent. (-*----, 177 2.45 2.35 2.38 1011 1-12 1.08 1.06 1013 1.39 1.37 1.39 1014 2.80 2.71 2-75 1015 4-25 4-15 4.28 1016 0-42 0.39 0.42 Potassium oxide by- Atomic N.B.S., absorption, Per per cent. cent. (-*---, 0.57 0.58 0.58 0.26 0.29 0.29 0.32 0.38 0.36 0.99 0.99 0.97 0.87 0.88 0.86 0.04 0.048 0.042 Sodium oxide by- - N.B.S., absorption, Per per cent.cent. (--A-, 0.14 0.14 0.14 0.08 0.08 0.08 0-20 0.21 0.22 0.24 0.24 0.26 0-16 0.16 0.16 0.55 0-57 0.61 Lithium oxide by- N.B.S., absorp- per tion, cent per cent. - 0.0037 0.002 0.0037 0.001 0.0027 0.005 0.0041 0-004 0.0028 0.012 0.0132 rAtomlc Results for iron, manganese, magnesium, potassium, lithium and sodium are given in Tables I1 and 111. Their determination by using simple aqueous standards of the element was straightforward. Magnesium concentration varies quite widely among cements, and this element is extremely sensitive to atomic absorption. It was convenient to make higher dilutions of the acid-soluble solution (from 30 to 300 times) so as to bring the concentrations between 0-2 and 2 p g per ml of magnesium. Preliminary results for aluminium with aluminium chloride solution standards were about 7 per cent.high, probably as a result of suppression of aluminium ionisation by the calcium in the cement. Therefore, calcium chloride and hydrochloric acid were added to the standards in about the same amounts as were present in the samples. The results for aluminium are shown in Table 11. Fig. 1. Effect of A, sodium silicate and B, aluminium chloride on the optical density of a solution of 20 pg of titanium per ml556 CAPACHO-DELGADO AND MANNING : DETERMINATION BY [Analyst, Vol. 92 In attempting to use simple standards of titanium (as the sulphate), the totalled titanium results of the samples were high by 10 to 15 per cent., which indicated that there was inter- ference. A study was made of the effects on the absorption signal from titanium of the major constituents of cement, as well as those of several acids, by adding these in varying concentrations.Hydrochloric acid, up to 10 per cent. v/v, had no effect on the absorption of 20pg per ml of titanium. With 5 per cent. v/v hydrochloric acid and 0.5 per cent. of calcium, added as the chloride, titanium absorption was increased by 4 per cent. Additions of sodium silicate and aluminium chloride increased the absorption of 20 pg per ml of titanium, as shown in Fig. 1. The concentration of aluminium in cement is about 15 to 25 times that of the titanium. The increase in the titanium absorption caused by this amount of aluminium is analytically significant, as shown in Fig. 1. Titanium standards were prepared containing about the same concentrations of calcium chloride, aluminium chloride and hydrochloric acid as were present in the cement samples, and this series of standards was also used to determine the titanium in the fusion solution that contained principally sodium silicate.The error introduced by not using titanium standards with added sodium silicate is small because only about 10 to 15 per cent. of the titanium is in the fusion solution. Results for titanium are given in Table IV. TABLE IV DETERMINATION OF TITANIUM DIOXIDE IN CEMENTS Titanium dioxide by- Sample 177 1011 1013 1014 1015 1016 N.B.S., per cent. 0.26 0.25 0.20 0.25 0.26 0.34 Atomic absorption, per cent. r----&----- 7 Acidic solution Fusion solution Total 0.200 0.053 0.25 0.210 0-032 0.24 0.165 0.033 0.20 0.207 0.030 0.24 0.2 17 0,043 0.26 0.290 0.062 0.35 It has been shown that in using the nitrous oxide - acetylene flame for strontium, greater sensitivity and greater freedom from chemical interferences is achieved, compared with the cooler air - acetylene flame.ll When using the nitrous oxide - acetylene flame for the deter- mination of strontium it is necessary to suppress strontium ionisation by the addition of a relatively large amount of an easily ionised metal.12 Additions of 1 mg of potassium per ml (as the chloride) were made to samples and standards and the results are given in Table V.Additions of I per cent. of lanthanum worked equally well as an ionisation suppressant, and enabled the same sample aliquot to be used for the potassium determination.TABLE V DETERMINATION OF STRONTIUM OXIDE IN CEMENTS Strontium oxide by- > Atomic absorption, per cent. 7- m N.B.S., Sample per cent. ~ Z U S lanthanum plus potassium 0.05 0.052, 0.050 0.052, 0.055 177 1011 0.11 0.107 0.118 1013 0.08 0.083 0.092 1014 0-26 0-273 0.270 1015 0.11 0-107 0.105 1016 0.25 0.256 0.266 To determine the precision of the method, several repeat determinations of aluminium were made with a single cement sample, with the results shown in Table VI. Three samples were individually prepared and the aluminium determined on each of eight days. The mean of all determinations is 6.56 per cent. of alumina and the standard deviation 0-096, or 1-5 per cent. relative standard deviation. As the analytical procedure for the other elements is similar, the precision for them would be about the same.September, 19671 ATOMIC-ABSORPTION SPECTROSCOPY OF SEVERAL ELEMENTS 557 About 34 hours were required to prepare four cement samples for analysis, including dilutions, and the determination of one element in the four samples, including the time required for calculation, took 15 minutes. This time would be less if an automatic concen- tration read-out device13 were used and the total time taken for analysis would then depend on the number of elements.The determination of silicon, magnesium, aluminium, iron, titanium, sodium and potassium in 4 samples of cement by one analyst required about 5 hours, including sample preparation. A larger number of samples requires proportionately less time for each because of greater efficiency.TABLE VI ANALYTICAL PRECISION FOR ALUMINIUM ON EIGHT SUCCESSIVE DAYS BY USING THREE INDIVIDUALLY PREPARED SOLUTIONS Aluminium oxide, per cent. A B C 6-54 6.35 6.35 6.50 6.54 6.48 6-55 6.50 6.55 6.60 6.60 6.60 6-59 6-62 6.59 6-60 6.68 6.68 6-81 6-59 6.60 6.59 6.57 6.45 Mean 6-59 Mean 6-55 Mean 6.54 Mean 6.56 per cent. Standard deviation 0.096. A I \ 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. REFERENCES Diamond, J. J., Analyt. Chem., 1956, 28, 328. Ford, C. L., Bull. Amer. SOC. Test. Muter., December, 1960, No. 250, p. 25. U.S. Army Corps of Engineers, Waterways Experiment Station, Vicksburg, Mississippi, Miscel- Wilson, T. C., and Krotinger, N. J., Bull. Amer. SOC. Test. Muter., 1953, 189, 56. Takeuchi, T., and Suzuki, M., Talanta, 1964, 11, 1391. Leithe, W., and Hofer, A., Mikrochim. Acta, 1961, 2, 268. Sullivan, J. V., and Walsh, A., Spectrochim. Acta, 1965, 21, 721. Cartwright, J., Sebens, C., and Manning, D. C., Atomic Absorption Newsletter, Perkin-Elmer U.S. Department of Commerce, Washington, D.C., National Bureau of Standards Miscellaneous American Society for Testing Materials, A .S. T.M. Stand C114-61, Philadelphia, Pennsylvania, Amos, M., and Willis, J., Spectrochiun. Acta, 1966, 22, 1325. Manning, D. C., and Capacho-Delgado, L., Analytica Chim. Acta, 1966, 36, 312. Keats, G. H., Atomic Absorption Newsletter, Perkin-Elmer Corporation,. 1965, 4, 319. laneous Paper No. 6-111, November, 1954. Corporation, 1966, 5, 91. Publication 260, 1964, “Catalog and Price List of Standard Materials.” 1961, Sections 53 to 58. Received September 23rd, 1966

ISSN:0003-2654    

DOI:10.1039/AN9679200553

出版商:RSC    

年代:1967

数据来源: RSC

摘要:

558 Analyst, September, 1967, Vol. 92, jq5. 558-564 Direct Complexometric Determination of Zirconium(1V) in Relation to Polymerisation BY B. C. SINHA AND S. DAS GUPTA (Central Glass & Ceramic Research Institute, Calcutta, 32) Polymerisation and de-polymerisation of zirconium on boiling and ageing in various concentrations of nitric, hydrochloric and sulphuric acids have been systematically studied and the optimum experimental conditions for complete de-polymerisation and titration of zirconium ions worked out. A simple and accurate method is described for the direct titration of zirconium ions with EDTA. Most of the elements, other than phosphorus and fluorine, do not interfere. The method has been successfully used for determining zirconium in zircon and commercial zirconium dioxide.ZIRCONIUM is known to form a complex with ethylenediaminetetra-acetic acid (EDTA) in acidic solutions. A study of the literature reveals a confusing picture concerning the suitability of acids and their strengths for direct titrations with EDTA. Musil and Theisl reported a method in which the titration of zirconium was carried out in N hydrochloric acid solution. Budevsky, Pencheva, Russinova and Russeva2 preferred to titrate in about 0-5 N sulphuric acid. Banerjee3 and Fritz and Fulda4 selected the pH ranges of 1.5 to 2.5 and 1.3 to 1-5, respectively. Pribil and Vesely5*6 observed that comparatively dilute solutions of hydro- chloric, sulphuric and perchloric acids (below 0.6 N) were not suitable for direct EDTA titrations of zirconium because the results obtained were always low.They recommended a method in which zirconium in 0-3 to 0.6 N nitric acid solution was heated just to boiling and immediately titrated with EDTA. On longer boiling, however, low results were obtained. It was also reported that titrations in freshly prepared zirconium solutions always gave low values and reached the theoretical values only after 15 hours in N nitric acid and 8 days in N perchloric acid.' This unsatisfactory state with EDTA titrations of zirconium is probably the result of incomplete understanding of the polymerisation of the zirconium ion and its complex forma- tion with the anions present in the solution. Recent studies5969s have revealed that both ZF+ and Zr02+ are likely to be present, even in strongly acidic solutions of zirconium salts.Lower concentrations of hydrogen ions and relatively higher concentrations of zirconium ions favour polymerisation of Zr02+ to polycations, probably of the type [ Zr <$zr14+ . These polymerised zirconyl ions are likely to be responsible for the low titration results. Complete de-polymerisation, as well as control over the procedure, will therefore be the pre-requisites for direct titration of zirconium with EDTA. The present investigation of the polymerisation and de-polymerisation of zirconium in several acids at various concentrations was undertaken with a view to ascertaining the most suitable acid, and the concentration, in which it may be completely de-polymerised and titrated. It was also the purpose to utilise the results of this investigation to develop a simple, satisfactory and rapid method for the direct determination of zirconium with EDTA.EXPERIMENTAL REAGENTS- All reagents were of analytical-reagent quality. EDTA solution, 0-05 M-Prepare by dissolving 18-61 g of the disodium salt of EDTA in distilled water and diluting to 1 litre. Standardise by titrating the solution with a standard zinc solution in acetate buffer (pH 5.2), with xylenol orange as indicator. EDTA solution, 0-02 M-Prepare from the 0.05 M EDTA solution by appropriate dilution. Zirconium nitrate solution, 0-05 M---Prepare a stock solution by dissolving 13.5 g of zirconyl nitrate, ZrO(N03!,.2H,0, in 140 ml of nitric acid (8 N). Then boil the solution for 5 minutes and dilute to 1 litre. The final acid strength was N.The molarity of the zirconiumSINHA AND DAS GUPTA 559 solution was checked by an indirect complexometric method,6 based on the addition of a known excess of EDTA and back-titrating the excess with bismuth nitrate at pH 1 to 2, with xylenol orange as indicator. Zirconium chloride and sulphate solutions, 0-05 M-Prepare by precipitating zirconium hydroxide and dissolving it by boiling with the appropriate acids. Standardise the solutions as described for the zirconium nitrate solution. Bismuth nitrate solution, 0.05 M-Prepare by dissolving 12-12 g of bismuth nitrate in 30 ml of concentrated nitric acid and diluting to 500ml. Check the molarity by direct titration at pH 1 to 2 with standard EDTA, with xylenol orange as indicator. Iron solution-Dissolve 12.0 g of ammonium iron(II1) sulphate, Fe,(SO,),.(NH,),S04.- 24H20, in 200 ml of water containing 5 ml of sulphuric acid (sp.gr. 1-84), and boil to obtain a clear solution. Precipitate iron( 111) hydroxide with ammonia solution. Filter the precipitate, wash and dissolve it in 125 ml of warm nitric acid (4 N) and dilute to 500ml. The acid strength of the solution was N, and 1 ml of the solution contained 2-79 mg of iron. Titanium solution-Dissolve 4.60 g of potassium titanyl oxalate, K,TiO(C,O,) ,.2H,O, in 200 ml of water containing 10 ml of hydrochloric acid (1 + 1). Then precipitate titanium hydroxide by adding ammonia solution, wash the precipitate, dissolve it in 125 ml of warm nitric acid (4 N) and dilute to 250 ml. The acid strength of the solution was 2 N, and 1 ml of the solution contained 2.40 mg of titanium.Xylenol orange solution, 0.2 per cent., aqueous. Nitric acid, 4 and 10 N. Hydrochloric acid, 4 and 10 N. Sulphuric acid, 4 N. Sodium hydroxide, 2 N. EFFECT OF BOILING ON POLYMERISATION- The effect of the time of boiling on polymerisation of zirconium ions in various concen- trations of nitric, hydrochloric and sulphuric acids was studied. Solutions were prepared by diluting, in conical flasks, aliquots of 5 ml of 0-05 M zirconium chloride, sulphate and nitrate solutions to 100 ml in each instance with the requisite amounts of acid to bring the acid strength to the desired values. The solutions were refluxed from the commencement of boiling for varying periods of time up to 20 minutes and titrated while hot (above 90" C) with standard EDTA, with xylenol orange as indicator.The titre values corresponding to the first change of the indicator from pink - red to lemon yellow, the EDTA solution being added dropwise (1 drop per second), were noted. 1.0 t- Time. minutes Fig. 1. Effect of boiling on the poly- merisation of zirconium ions in 0.5, 0.75 and N hydrochloric acid, curves A, B and F, respectively; in 0.25 N sulphuric acid, curve C; and in 0.5, 0.75 and N nitric acid, curves D, E and G, respectively560 SINHA AND DAS GUPTA : DIRECT COMPLEXOMETRIC DETERMINATION [Analyst, VOl. 92 Polymerisation was indicated by a low titre value, and also by reversion of the end-point after some time (a few seconds to a few minutes), depending on the extent of polymerisation.The results are presented graphically in Fig. 1, and the following observations made. Zirconyl ions polymerised on boiling in up to 0.75 N nitric and hydrochloric acids and 0.25 N sulphuric acid, and polymerisation increased with the time of boiling up to 5 minutes in nitric and sulphuric acids, and 10 to 15 minutes in hydrochloric acid. Beyond these time limits there was little further polymerisation. Polymerisation was found to be at its maximum in hydrochloric acid. Zirconyl ions did not polymerise on boiling, even for 20 minutes, in N nitric and hydro- chloric acids and in 0-5 N sulphuric acid, or at higher concentrations. The sharpness of the end-point of xylenol orange decreased with the increase in acid strength above N with nitric and hydrochloric acids and 0.5 N with sulphuric acid, and the indicator was found unsuitable above 1.5 N with nitric and hydrochloric acids and 0.75 N with sulphuric acid.The maximum titre values, corresponding to the actual amount of zirconium present, were obtained in nitric acid medium. EFFECT OF BOILING ON DE-POLYMERISATION- From the study of polymerisation it was clear that zirconyl ions did not polymerise on prolonged boiling in N nitric or hydrochloric acids or 0.5 N sulphuric acid, or at higher concentrations, but it was not known whether de-polymerisation could take place on boiling at these acid strengths. Therefore, the effect of boiling on de-polymerisation in 1, 2 and 3 N nitric and hydrochloric acids and 0.5, 1 and 3 N sulphuric acid was studied on solutions containing an appreciable amount of polymerised zirconyl ions.Aliquots of 5 ml of the 0.05 M zirconium solutions, in conical flasks, were each diluted to 25 ml with water to bring the acid strength to nearly 0.2 N, and refluxed for about 5 minutes to polymerise zirconyl ions. Each solution was then diluted to 40 ml with an adequate amount of 1 0 ~ acid and water to bring the acid strength to the desired values, and refluxed for varying periods of time, starting from boiling, up to 15 minutes. The solutions were then diluted to 200 ml, with acid if required, to render them N with respect to nitric and hydro- chloric acids, and 0.5 N with respect to sulphuric acid. The solutions were then heated to boiling and titrated against standard EDTA solutions, with xylenol orange as indicator.L O O 15 Time, minutes Fig. 2. Effect of boiling on the de-poly- merisation of zirconium ions in N, 2 N and 3 N hydrochloric acid, curves A, C and F, respectively; in N, 2 N and 3 N nitric acid, curves B, E and G, respectively; and in 0.5 N sulphuric acid, curve D The results of these experiments are presented in Fig. 2, and the following observations made. De-polymerisation depended on the nature of the acid, its strength and the time of boiling. Polymerised zirconium ions de-polymerised on boiling in solutions of N nitric and hydrochloricSeptember, 19671 OF ZIRCONIUM(IV) IN RELATION TO POLYMERISATION 561 acids and 0.5 N sulphuric acid, and at higher concentrations. De-polymerisation increased with increase in acid strength and time of boiling. Maximum de-polymerisation could be effected by boiling a zirconium solution for 5 minutes in 3 N nitric, hydrochloric or sulphuric acids.The titre value corresponding to the maximum de-polymerisation in nitric acid (Fig. 2) was equivalent to the actual amount of zirconium taken, indicating complete de-polymerisation, while the maximum titre values corresponding to maximum de-polymerisation in hydrochloric and sulphuric acids were always slightly lower than those in nitric acid. It is interesting to note that during the preparation of the stock solution of 0.05 M zirconium nitrate, the solution was boiled for 5 minutes in 8 N nitric acid and it did not, therefore, contain any polymerised zirconyl ions. EFFECT OF AGEING ON DE-POLYMERISATION- From the results of our present study of polymerisation on boiling, it can be remarked that zirconium(1V) will polymerise on ageing in acid strengths below N with nitric and hydrochloric acids and 0.5 N with sulphuric acid.However, the extent of polymerisation will depend on the concentration of zirconium(1V) and the temperature of ageing. At higher concentrations of acids, de-polymerisation is also to be expected but it is not very clear under what conditions complete de-polymerisation will occur. Therefore, the effect of progressive ageing on polymerised zirconyl solutions at various acid strengths was studied at room temperature (30" rt: 2" C). Polymerised zirconyl solutions (0.05 M) in 1, 2 and 3 N nitric and hydrochloric acids and in 0 6 , l and 3 N sulphuric acid were prepared from 0.05 M zirconium nitrate stock solution.Aliquots of 100ml of this solution, after appropriate dilution to bring the acid strength to 0.2 N, were refluxed for 5 minutes to polymerise zirconyl ions. Zirconium hydroxide was then precipitated from each aliquot with sodium hydroxide solution in the presence of triethanolamine, and the precipitate dissolved in the required amounts of the various acids to maintain the desired acid strengths after diluting each solution to 100 ml. Five millilitres of each of these solutions were then diluted to 100 rnl with the requisite amounts of the respective acids to bring the final strengths to N with respect to nitric and hydrochloric acids and 0.5 N with respect to sulphuric acid. Each solution was then heated to boiling and titrated against EDTA, with xylenol orange as indicator.The titrations were carried out daily for the first few days, and later at longer intervals. De-polymerisation was indicated by the increase in titre values. The results are presented in graphical form in Fig. 3, from which the following observations are made, 5.5 5.08- - 5.0 4.5 - S 4.0 3.5 3.0 2.5 l00%Zr Time, days Effect of ageing on the de-polymerisation of zirconium ions in N, 2 N and 3 N hydrochloric acid, curves A, F and G, respectively; in N and 2 N nitric acid, curves B and H, respectively; and in 0.5 N, N and 3 N sulphuric acid, curves C, D and E, respectively Fig. 3.562 SINHA AND DAS GUPTA : DIRECT COMPLEXOMETRIC DETERMINATION [Analyst, VOl. 92 De-polymerisation increased on standing with the increase in acidity from 1 to 3 N with nitric and hydrochloric acids and 0.5 to 3 N with sulphuric acid.The extent of de-polymerisation depended on the nature of the acid, being greatest in nitric acid and smallest in sulphuric acid. The rate of de-polymerisation was quite fast at the beginning of the experiments and then gradually decreased to zero, indicating no further de-polymerisation, i.e., the maximum de-polymerisation possible at that degree of acidity, which was attained within 2 days in 3 N nitric acid and required 4 and 15 days in 2 N and N nitric acid, respectively. The titre values corresponding to the maximum de-polymerisation in 2 and 3 N nitric acid were found to be the same, and almost equivalent to the actual amount of zirconium taken, indicating complete de-polymerisation of zirconyl ions.Therefore, if complete de-polymerisation is to be affected by ageing, the zirconyl solution must be at least 2 N in nitric acid. Hydrochloric and sulphuric acids were found to be unsuitable. PROCEDURE FOR THE DETERMINATION OF ZIRCONIUM WITH EDTA- From the present studies on polymerisation and de-polymerisation, it is evident that for direct titration of zirconium with EDTA, nitric acid is the best medium. It has also become clear that de-polymerisation of any polymerised zirconyl ion can easily be effected by boiling it for 5 minutes in 3 N nitric acid, and the titration can best be carried out in the hot solution after diluting to N. The following procedure was accordingly developed for EDTA titration of zirconium.To an aliquot of sample solution containing up to 91 mg of zirconium, in a conical flask, add the calculated amount of 8 N nitric acid and water to make the volume up to 40 to 60 ml, and its acid strength 3 N. Boil the solution for 5 minutes and then dilute it to between 100 and 200ml (for higher concentrations of zirconium the volume should preferably be diluted to 200 ml) with the calculated amount of 4 N nitric acid and water to maintain the acid strength at N. Heat the solution to boiling, add 0-4ml (8 drops) of xylenol orange indicator and titrate while hot with dropwise addition (1 drop per second) of 0.05 M EDTA. Swirl the solution during the titration and heat it, if required, to maintain the temperature above 90" C. Continue the titration slowly until the end-point is indicated by a sharp change of colour from pink to lemon yellow.DISCUSSION The present studies on polymerisation and de-polymerisation of zirconium ions in various concentrations of acids reveal that, for the direct titration of zirconium with EDTA, the following basic steps are necessary. (1) Zirconium ions should be completely de-polymerised. This can easily be effected by boiling the solution in 3 N acid (preferably nitric acid) for 5 minutes. (2) To avoid polymerisation of zirconyl ions on heating or boiling, the EDTA titration should be carried out in N hydrochloric acid or 0-5 N sulphuric acid, or preferably in N nitric acid at, or above, 90" C. TABLE I DETERMINATION OF ZIRCOKIUM Zirconium present, mg 2-27 4.56 11-41 22-80 45.48 63-37 91.16 Zirconium found, mg 2.18 4.40 11.05 22.91 45.81 63.22 91.47 Difference, mg - 0.09 -0-16 - 0.36 s 0 .1 1 + 0.23 +0*31 -0.15 The analytical procedure described is based on these observations and the results obtained are presented in Table I. They compare very favourably with the actual amounts of zir- conium taken, indicating the accuracy of the method. The maximum amount of zirconium that could be determined was 91 mg in 200 ml of solution, compared to only 35 mg in 300 ml of solution claimed by Pribil and Vesely6 by their method.September, 19671 OF ZIRCONIUM(IV) IN RELATION TO POLYMERISATION 563 As the titrations were carried out in N nitric acid solutions, most elements did not inter- fere. Interfering elements reported so far are fluorine, phosphorus, iron, bismuth and titanium (in amounts greater than 60mg).Fluoride and phosphate, even in traces, were also found to interfere with the present method, but iron and titanium up to 181.51 and 155.68 mg, respectively, had no effect (Table 11). Bismuth when present in excess (62.70 mg or more) interfered. TABLE I1 DETERMINATION OF ZIRCONIUM IN THE PRESENCE OF DIFFERENT AMOUNTS OF IRON, BISMUTH AND TITANIUM 7 Iron 55-95 181.51 55.95 1.8 1 -5 1 - Taken, mg p---h------- Bismuth Titanium - - - 71.61 - 155.68 - 71.61 - 155.68 41.80 - 62-70 - 1 Zirconium 23.81 23.81 9.67 9.67 9-67 23.81 23.81 9-67 21.89 21.89 Zirconium found, mg 23.90 23-72 9-76 9-58 9.58 23.63 23.81 9.58 21-80 22.58" Zirconium difference, mg + 0.09 - 0.09 + 0.09 - 0-09 - 0.09 -0.18 - 0.09 + 0.69 - - * End-point was not sharp.The procedure was applied for the determination of zirconium in zircon and zirconium dioxide as follows. Fuse 2 g of sodium carbonate in a nickel crucible so that an inside lining of sodium carbonate occurs after cooling, and add to it 0.5 g of well ground (in an agate mortar) and dried sample and 5 g of sodium peroxide. Heat the crucible over a low flame to dull redness. Swirl the crucible occasionally with tongs and continue heating for 10 minutes. Cool the crucible and extract the melt quantitatively with water in a beaker. Heat the beaker containing the extracted mass for 10 minutes on a steam-bath and filter the solution through a Whatman No. 42 filter-paper or equivalent. Wash the residue containing zirconium hydroxide two or three times with a 5 per cent.solution of sodium hydroxide and then twice with hot water. Cool the solution, dilute to 250 ml and proceed as described in the Procedure for the determination of zirconium with EDTA. Dissolve the residue in 75 ml of hot nitric acid (4 N). TABLE I11 DETERMINATION OF ZIRCONIUM I N ZIRCON AND COMMERCIAL ZIRCONIUM DIOXIDE Sample Zirconium dioxide Zirconium dioxide by the present Mean by gravimetric Difference, method, value, method, per cent. per cent. per cent. per cent. Zircon 66-21 66.28 66-15 65.91 0.24 65.97 Commercial zirconium dioxide 96-79 95.84 95.98 -0.14 95.87 95.85 * Standard deviation was 0-102. The determination in triplicate of zirconium in one zircon and one commercial zirconium dioxide sample, together with the results obtained by the classical m e t h ~ d , ~ are presented in Table 111. The results compare favourably with each other. It is interesting to note that the time taken to determine zirconium in zircon by the present method is only 2 hours compared with 12 to 14 hours by the classical method.564 SINHA AND DAS GUPTA The authors thank Mr. K. D. Sharma, Scientist-in-Charge of the Central Glass & Ceramic Research Institute, for permission to publish this paper, and Dr. S. Kumar for his keen interest in the work. 1. 2. 3. 4. 5. 6. 7. 8. 9. REFERENCES M u d , A., and Theis, M., Z . analyt. Chem., 1955, 144, 427. Budevsky, O., Pencheva, L., Russinova, K., and Russeva, E., Talanta, 1964, 11, 1225. Banerjee, G., 2. analyt. Chem., 1955, 147, 105. Fritz, J. S., and Fulda, M. O., Amlyt. Chem., 1954, 26, 1206. Pribil, R., and Vcsely, V., Z . analyt. Chem., 1964, 200, 332. ~~ , Talanta, 1964, 11, 1197. Prib;l, R., Ibid., 1965, 12, 925. Larsen, E. M., and Wang, P., J . Amer. Chem. SOC., 1954, 76, 6223. Scott, W. W., and Furman, N. H., “Standard Methods of Chemical Analysis,” Fifth Edition, Received November 4th, 1966 D. Van Nostrand Company Inc., New York, 1939, p. 1103.

ISSN:0003-2654    

DOI:10.1039/AN9679200558

出版商:RSC    

年代:1967

数据来源: RSC

摘要:

Analyst, September, 1967, Vol. 92, $I$. 565-566 565* Determination of Copper in Trace-element Superphosphate by A.C. Polarography BY G. CURTHOYS AND J. R. SIMPSON (University of Newcastle, New South Wales, AustraZia) Copper in trace-element superphosphate is determined by an a.c. polarographic technique, in which the copper is polarographed in a M solution of hydrochloric acid. A METHOD, which is both rapid and accurate, is described for determining copper in trace- element superphosphate (from 0.01 to 5 per cent, of copper) by an a.c. polarographic technique. The copper is polarographed in a M solution of hydrochloric acid. The presence of phosphate, sulphate, iron and aluminium, usually found in superphosphate, does not interfere with the curve for copper. The results are in excellent agreement with those obtained by the official sulphide separation procedure.METHOD The reagents, samples and apparatus were essentially the same as those described in our previous publication,l except that a standard copper solution was used in place of the standard zinc solution. PROCEDURE- Prepare four standard solutions by weighing 2.0000 g of the normal superphosphate into each of four 250-ml beakers. Introduce 2.0 ml of the standard copper sulphate into the first beaker, 4-0 ml into the second, 6-0 ml into the third and 12.0 ml into the fourth. Add to each beaker 5 ml of concentrated nitric acid and 5 ml of 10 N hydrochloric acid, and evaporate the solution to dryness. Dissolve the residue in 10 ml of 10 N hydrochloric acid and 70 ml of hot distilled water.Boil the solution, then filter it through a Whatman No. 41 filter-paper and wash the residue with six small washings of hot distilled water. Cool the filtrate, dilute it to exactly 100 ml in a calibrated flask and mix it well. Transfer a portion of the solution to a 100-ml squat beaker and determine it polarographically between 0 to 0.45 volt with respect to the mercury-pool anode. Add to the sample 5 ml of 15 N nitric acid and 5 ml of 10 N hydrochloric acid, and evaporate the solution to dryness. Dissolve the residue in 10 ml of 10 TU’ hydrochloric acid, then boil, filter, wash and dilute the solution to 100 ml and determine it polarographically as above. From the standard solution draw a calibration graph and determine from it the amount of copper present in the trace-element superphosphate.Place 2.0000g of the trace-element superphosphate in a 250-ml beaker. RESULTS The reduction of copper ions in M hydrochloric acid base electrolyte gave a well defined peak occurring at a half-wave potential of -0-26 volt with respect to the mercury-pool anode. The results of the peak heights obtained from the addition of different amounts of the standard copper sulphate solution to normal superphosphate are shown in Table I. When plotted these results give a straight line passing through the origin. TABLE I PEAK HEIGHTS AT DIFFERENT COPPER CONCENTRATIONS Concentration of copper, Current, g per litre PA 2-00 x 10-3 7 4.00 x 10-3 15 6-00 x 10-3 22 12-00 x 10-3 44566 CURTHOYS AND SIMPSON The polarogram for the normal superphosphate without addition of standard copper solution is shown in Fie.1 (a), and that for the copper in the trace-element superphosphate v in Fig. 1 ( b ) . 3c - -0.1 - 0.3 Potential, volts Fig. 1. Polarogram for (a), normal superphosphate without standard copper solution and ( b ) , the copper in the trace-element superphosphate Comparison of the results obtained by the official sulphide separation followed by the iodimetric determination of copper2 and by the a.c. polarographic procedure are shown in Table 11. Sample TABLE I1 COMPARISON OF RESULTS Sulphide separation; copper, per cent. Copper superphosphate . . .. 0.054 0.053 0.053 0.054 0.053 0.054 Mean . . . . . . 0.0535 Standard deviation . . 0.00061 A.C. polarography; copper, per cent. 0.054 0.053 0.054 0-054 0.054 0.054 0.0538 0-00047 The a.c. polarographic procedure for the determination of copper in trace-element siinernhosDhate is accurate and rapid. and the analvsis takes about 1 hour. REFERENCES 1. 2. Curthoys, G., and Simpson, J. R., Analyst, 1966, 91, 195. Horwitz, W., Editor, “Official Methods of Analysis of the Association of Official Agricultural Chemists,” Eighth Edition, Association of Official Agricultural Chemists, Washington, D.C., 1955, p. 24. Received January 13th, 1967

ISSN:0003-2654    

DOI:10.1039/AN9679200565

出版商:RSC    

年代:1967

数据来源: RSC

摘要:

Analyst, September, 1967, Vol. 92, pp. 567-568 567 Voltammetric Studies with Different Electrode Systems Part II.* Tungsten as Reference Electrode in Polarography BY V. T. ATHAVALE, M. R. DHANESHWAR AND R. G. DHANESHWAR (Analytical Division, Bhabha Atomic Research Centre, Trombay, Bombay-74, India) The tungsten electrode has properties similar to the molybdenum electrode, which has been used successfully as a reference electrode in polarography. The study was carried out on a d.c. polarograph with a dropping-mercury electrode and a tungsten or molybdenum reference electrode system for silver and zinc reductions, as well as on a cathode-ray polarograph with silver - tungsten, silver - molybdenum, silver amalgam - tungsten and silver amalgam - molybdenum electrode systems for the reduction of various cations.For zinc reduction, tungsten and molybdenum electrodes behave in a similar way, but generally the performance of tungsten electrode does not compare favourably with that of the molybdenum electrode, especially for silver reduction. IN a search for alternative reference electrodes to the S.C.E. and mercury-pool electrodes, the molybdenum wire electrode has already been examined under widely varying experi- mental c ~ n d i t i o n s . l ~ ~ J ~ ~ The performance of the tungsten wire electrode as a reference electrode has also been studied. The chemical properties of tungsten and molybdenum are very ~ i m i l a r , ~ and both show similar behaviour as pH electrode^.^^^ In media of constant pH both electrodes should behave as reference electrodes whether they are used as such in polarography or potentiometry. A recent communication reported a study of the bi-metallic electrode systems silver - molybdenum and silver amalgam - molybdenum in relation to the silver-ion reduction on a cathode-ray p~larograph.~ It has been shown that silver or silver amalgam electrodes respond preferentially to silver ion when it is mixed with several other ions.The comparison of molybdenum and tungsten electrodes was made mainly by using these electrode systems so that some additional information regarding the behaviour of silver and silver amalgam electrodes was also obtained. For reduction of other ions a d.c. polarograph and the dropping- mercury electrode were used for the comparison. EXPERIMENTAL A KlOOO cathode-ray polarograph (Southern Instruments Ltd.) was used for examination of silver or silver amalgam cathodes, and a manual recording Du-Bellay polarograph was used with the dropping-mercury electrode.Potentials were measured on a direct-reading Vibron Electrometer, model 33B (Electronics Instruments Ltd.). Other experimental conditions have been reported.* RESULTS AND DISCUSSION Two concentrations of silver ion were examined on the cathode-ray polarograph with silver - molybdenum, silver - tungsten, silver amalgam - molybdenum and silver amalgam - tungsten electrode systems. Linear current - concentration relationships were obtained only for the molybdenum reference electrode, and no wave at all was obtained for the lower silver concentration with the silver amalgam - tungsten system.Further, to investigate this difference in behaviour, potential - concentration relation- ships were examined for silver ion and also for nickel, zinc, cadmium, iron and copper-ion solutions. Polarographic current - concentration relationships ultimately depend upon whether the particular electrode responds to the ion under consideration in a Nernstian sense. The results for potential - concentration relationships will therefore also indicate whether any current - concentration linearity can be expected in polarography. The potentials were measured in unstirred solutions so as to simulate the polarographic conditions, and may therefore differ slightly from those obtained in stirred solutions. However, the results will usefully define the nature of the response of the particular electrode system in polarography and potentiometry.Fig. 1 shows the curves obtained for a silver indicator and a molybdenum or tungsten reference electrode. * For details of Part I of this series, see reference list, p. 568.568 ATHAVALE, DHANESHWAR AND DHANESHWAR 400 - - 300 - - > E 2oQ- - Id r: .- - V 5 100 - IL - 0- - too - Ni” I Log metal-ion concentration, pM 2 3 4 5 Log rnetal-ion Log metal-ion concentration, pM concentration, pM Fig. 1. Potential - log metal-ion concentration curves with silver indicator, 28 s.w.g. and 7 mm long, and molybdenum or tungsten reference electrodes, 22 s.w.g. and 15 mm long, for Zn2+, Cd2++, Ni2+, Cu2+, Fe3+ and Ag+ ions. Curves 0, with molybdenum reference electrode ; curves 0, with tungsten reference electrode With the silver indicator - molybdenum reference system, a nearly straight line of zero slope is shown for nickel, iron, copper, cadmium and zinc, with deviations at concentrations above 1 0 - 3 ~ for iron and zinc.For silver ions a straight line with a slope of 45 mV is obtained. With a silver indicator - tungsten reference system a near-zero slope is shown for cadmium ion, and a greater but still deficient slope is obtained for zinc, nickel and copper ion. Therefore, it may be assumed that for these elements tungsten behaves in some measure as a reference electrode in a similar way to molybdenum. Serious deviations are, however, obtained in the curves for iron and silver, and the function of the tungsten electrode as a reference electrode is impaired.Silver polarograms were also taken on a d.c. polarograph with the dropping-mercury electrode as the cathode and molybdenum or tungsten as the reference electrode in a supporting electrolyte of 0.1 M sulphosalicylic acid and 1 per cent. sulphuric acid.3 The background current obtained with a tungsten anode is considerably higher than that for molybdenum. Moreover, in the presence of silver ion the silver reduction current is actually smaller than the residual current for the tungsten electrode. In silver voltammetry, therefore, the molybdenum electrode is to be preferred to the tungsten electrode. Zinc waves were recorded on a d.c. polarograph by using a dropping-mercury electrode and either of these reference electrodes. The current - concentration proportionality was excellent in both instances, and the values of the currents were comparable for both reference electrodes.The half-wave potential values were similar and remained constant (-0.82 volt) over the concentration range. Despite the close similarity between their chemical properties and pH responses molybdenum is a better reference electrode than tungsten. REFERENCES 1. 2. 3. 4. 5 . 6. 7. Athavale, V. T., Burangey, S. V., and Dhaneshwar, R. G., J . Electvoanal. Chem., 1965, 9, 169. --- , in Shallis, P. W., Editor, “Proceedings of the SAC Conference, Nottingham, Dhaneshwar, R. G., and Kulkarni, A. V., Indian J . Chem., 1966, 4, 533. Athavale, V. T., Dhaneshwar, R. G., andDhaneshwar, M. R., J . Electroanal. Chem., 1967, 14, 31. Mellor, J. W., “A Comprehensive Treatise on Inorganic and Theoretical Chemistry,” First Edition, Volume 11, Longmans, Green and Co., London, New York and Toronto, 1948, pp. 512 and 729. Issa, I. M., and Khalifa, H., Analytica Chirn. Acta, 1954, 10, 567. El Wakkad, S. E. S., Rizk, H. A., and Ebaid, I. G., J . Phys. Chem., 1955, 59, 1004. NOTE-Reference 2 is to Part 1 of this series. 1i65,” k. Heffer & Sons Ltd., Cambridge, 1965, p. 446. Received August 22nd, 1966

ISSN:0003-2654    

DOI:10.1039/AN9679200567

出版商:RSC    

年代:1967

数据来源: RSC

摘要:

Analyst, September, 1967, Vol. 92, $9. 569-574 669 Determination of Dialkyltin Stabilisers in Aqueous Extracts from PVC and other Plastics BY R. SAWYER (Ministry of Technology, Laboratory of the Governwent Chemist, Cornwall House, Stamford Street, London, S.E. 1) A method is described for the determination of dialkyltin salts leached from PVC into aqueous extracting media. The reagent used, 4-(2-pyridylazo) resorcinol, is stable and water soluble, but forms a chloroform-soluble complex with dialkyltin compounds. Indication of the screening efficiency of EDTA for possible trace metal contaminants is also given. DIALKYLTIN compounds have been used as stabilisers in compounded PVC formulations for several years and they are generally present to the extent of 1 to 2 per cent. in the finished polymer.The increased use of PVC in the food packaging, and disposable medical articles fields, has directed attention to the need for analytical methods for determining traces of polymer additives that may be extracted by materials in contact with the plastic. Several methods have been published for the determination of alkyl tins in various contexts. The reagent most used for this purpose appears to be dithizone; reports of the use of dithizone for determination of diethyl tins in fungicides and medicaments,l total alkyl tin in PVC2 and as a spray reagent for detection of the various alkyl tins following separation on thin-layer plates (S. Greenfield, private communication) have appeared. Diphenyl carbazone has also been reported as a reagent for dibutyl tin dichloride in chloroform solutions derived from bactericidal agent^.^ Major objections to the use of these two reagents stem from their inherent instability in solution and from the fact that complexes and reagents are both chloroform soluble, thus necessitating the use of “reagent loss” methods of measurement or the use of highly absorbing reference solutions.Catechol violet has been proposed as a reagent for dialkyl tin compounds in fats and The formation of alkyl tin complexes with various nitrogen and oxygen donor ligands has been r e p ~ r t e d , ~ ~ ~ ~ ~ ~ ~ and it is likely that any one of several such ligands could be used as the basis of a spectrophotometric method. A report of the use of one such reagent, 4-(2-pyridylazo) resorcinolg (PAR), for determination of diethyl lead in aqueous solutions also proposed the use of the reagent for diethyl tins.A study of the use of this reagent in the plastics extract field was undertaken, and attention was paid to the elimination of interferences by the metal ions commonly found in plastic formulations or occurring in ancillary equipment used as components of composite fabricated articles. Metal salts commonlp used in PVC formulations include those of calcium, magnesium, cadmium, zinc and tin, These may be present either as metal soaps or as stabilisers. PAR has been described as a reagent for lead,lO vanadium,ll cobalt, uranium,12 copper, nickell3; coloured complexes with iron(II), iron(III), zinc(II), cadmium(II), tin(II), titanium(1V) and rare earths were also reported.12 A system has been devised in which the PAR - alkyl tin complex is selectively extracted from solutions containing possible interfering ions.REAGENTS- Alkyl tin salts-The various alkyl tin salts used were obtained from Pure Chemicals 1,td. Standard solutions were prepared, as and when required, in chloroform at concentrations of 1 x 10-3 M. Dilutions to 6 and 12 x M were generally used in the preliminary work. Twice the calculated amount of acetic acid was added to dibutyl and di-octyl tin oxides to bring these into solution in chloroform. 4-(2-PyridyZaxo) resorcinol (PAR), 2 x M-Dissolve 0.474 g of monosodium salt of PAR in distilled water in a 1-litre flask. Dilute to volume. (Stock solution.) Ethylenediavniizetetra-acetate, 0.1 M-Dissolve 37.2 g of ethylenediaminetetra-acetic acid and 8 g of sodium hydroxide in distilled water, then dilute to 1 litre.(Stock solution.) Chloroform-Analytical-reagent grade. EXPERIMENTAL570 SAWYER : DETERMINATION OF DIALKYL TIN STABILISERS [A nabst, VOl. 92 2 co 0.5 Jd - aJ 6 C; 0.4 >; .- v) S aJ -0 - 5 0.3 .- U 0" APPARATUS- S$ ec trot h otameters-Unicam S P500 and S P 8 00. Meter-Pye Dynacap. - - - I I I I 1 8 9 10 I I 12 O6 6 Wavelength, mp Fig. 1. Spectra over pH range 8.6 to 12 by extrac- tion of PAR-dibutyl tin complex a t 12 x 1 0 - 6 ~ in chloroform: 10 ml of dibutyl tin maleate, 20 ml of EDTA and 10 ml of PAR PROPERTIES OF THE ALKYL TIN COMPLEXES- Chloroform-soluble pink complexes are formed when aqueous solutions of PAR are shaken, either with aqueous suspensions of dialkyl tin salts or with solutions in chloroform of the dialkyl tin compounds.The pink chloroform-soluble complex of dibutyl and di-octyl tin salts, Em,,. at 518 mp, is readily formed from aqueous PAR at pH in the range 8 to 12. Addition of 0.1 N sodium hydroxide to the separated chloroform layer will decolorise the complex. Fig. 1 shows a typical set of spectra produced over a pH range by extraction of PAR - dibutyl tin complex at 12 x M in chloroform. The molar extinction coefficient at pH 9 to 11 is 4.15 x 104 for dibutyl and di-octyl tin. Fig. 2 shows a plot of the variation of E,,, (absorbance at 518 mp in a 2-cm cell) with pH over the range 8-5 to 12.5 at the same concentrations.September, 19671 IN AQUEOUS EXTRACT FROM PVC AND OTHER PLASTICS 57 1 Wavelength Fig.3. maleate present Extraction of PAR into chloroform, no dibutyl tin The spectra obtained at pH below 9.0 show evidence of a second peak of Em,,. at1384 mp due to PAR solubility in chloroform, and distortion of the visible spectrum is apparent from spectra shown in Fig. 1. Extraction of PAR into chloroform from aqueous solutions of varying pH without dibutyl tin, is illustrated in Fig. 3. Negligible transfer to chloroform is obtained for aqueous solutions at pH 10 and above. None of the PAR - metal complexes listed in the introduction is soluble in chloroform when the aqueous phase is raised to a pH above 9, but all of tlie metals indicated form complexes in competition with dibutyl tins and consume the reagent preferentially.Addition of EDTA to tlie aqueous phase at pH above 9 screens a metal - tin alkyl system to allow formation of the chloroform-soluble dialkyl tin complex with PAR. Table I indicates the freedom from interference by common TABLE I REACTION OF SELECTED METAL IONS WITH DIBUTYL TIN - PAR COMPLEX 25ml of a chloroform solution of dibutyl tin dilaurate at 6 x 1 0 - 6 ~ was used Apparent concentration of dibutyl tin in CHCl, Present in Metal 100-ml aqueous phase, mg M X Fe3+ 3.0 6- 1 6.2 Zn2+ 3.0 6.0 30* 5.9 6.1 Pb2+ 3-0 6.2 30" 5.8 6.1 Cd2+ 3.0 6.1 30* 5-8 6.0 cu2+ 3.0 6.0 30* 6.1 5.8 Ni2+ 3.0 6-0 6-0 Sn4+ 3.0 6-0 30* 6.1 6-1 Sn2+ reduces the complex. Colour formation equivalent t o Sn4+ if 1 ml of H,O, is added to mixed reagents. * EDTA used a t 4 times the concentration indicated in procedure.572 SAWYER : DETERMINATION OF DIALKYL TIN STABILISERS [ArtabSt, VOl.92 V I I I I I 1 r Molarity of dibutyl tin maleate x lo6, 20 ml of chloroform Molarity of PAR x106, 20 ml of 0 . 1 ~ EDTA a t pH 10.5 0 2.5 5 7.5 10 12.5 15 17.5 20 20 17.5 15 12.5 10 7.5 5 2.5 0 Fig. 4. Job curve metal ions. No detectable colour reaction is obtained with mono-alkyl and tetra-alkyl tins at the concentration indicated above ; tributyl and triphenyl tin salts give reactions equivalent to approximately one half of 1 per cent. of that of the dibutyl tin salts at the same concen- tration. A mole ratio plot obtained for dibutyl tin - PAR is illustrated in Fig. 4 and indicates a 1 : 1 complex. A plot at constant concentration of dibutyl tin against varying concentrations of PAR is indicated in Fig.5, a constant optical density being obtained when the ratio of PAR - dibutyl tin exceeds 100 : 1. 1 2 4 10 20 40 100 201 Log mole ratio, PAR - dibutyl tin maleate Fig. 5. Reagent required to give maximum optical density of complex in chloroform ANALYTICAL PROCEDURE FOR THE DETERMINATION OF DIALKYL TIN SALTS IN EXTRACTS FROM Pvc FORMULATIONS EXTRACTION PROCEDURE- Those commonly encountered in the medical appliance and in the food and drug container fields involve filling the article with hot distilled water or 0.9 per cent. saline solution and main- taining it at a fixed temperature of 60" to 85" C according to specification, for periods varying Aqueous extracts from PVC articles may be prepared in a variety of ways.September, 19671 IN AQUEOUS EXTRACT FROM PVC AND OTHER PLASTICS 573 from one hour to several days.Typical sets of conditions are set out in B.S. 2463,14 and B.S. 3505.15 Alternatively, the use of autoclave techniques has been recommended for preparation of extracts from samples of film and tubing. In these methods a fixed ratio of total surface area to liquid volume is used; typical conditions require the exposure of 3 cm2 of surface per ml of extracting medium at temperatures of 110" C16 or 121" C17 (autoclave) for periods of 1 hour. In these experiments extracts were prepared with distilled water at 85" C for 1 hour in whole containers or autoclave at 110" C on tubing. Comparison determinations were carried out on the extracts by using the dithizone method of Aldridge and Cremer.1 REAGENTS- PAR - EDTA-Mix together 100 ml of 0.1 M EDTA stock (loc.cit.) and 50mlof PARstock, adjust the pH to between 10 and 11, shake with 50 nil of analytical-reagent grade chloroform. Allow to settle and discard the chloroform layer. Chloroform-Anal yt ical-reagen t grade. Dibutjd tin maleate standard, 1 x M-Dissolve 0.0867 g of dibutyl tin maleate in Dilute suitable aliquots to to 1 x M. chloroform, dilute to 250ml in a standard calibrated flask. give a range of standard solutions in chloroform from 1 x METHOD- Place in a separator a suitable volume (usually 25 ml) of the extract, obtained as detailed above, add 30 ml of the combined EDTA - PAR reagent and 10 ml of chloroform. Shake the mixture for 1 minute and allow to settle, run off the chloroform layer through a small cotton- wool plug into a l-cm stoppered cell. Read the optical density of the solutions against a chloroform reference at 518 mp, record the optical density and compare it with that of a series of standards prepared from 25ml of water, 30ml of PAR-EDTA and 10ml of standard dibutyl tin in chloroform.RESULTS AND DISCUSSION Standard calibration curves prepared with di-octyl and dibutyl tin salts are almost indistinguishable from one another. Information relating to the nature of the alkyl tin stabiliser is often not available and for convenience the resalts are expressed as dibutyl tin maleate. Comparison results obtained by the method described and that of Aldridge and TABLE I1 COMPARISON OF THE RESULTS OBTAINED BY THE METHOD DESCRIBED WITH THOSE OF ALDRIDGE AND CREMER Dibutyl tin maleate extracted per cm2 of surface exposed, p g Type of plastic ml PAR Dithizone Extract volume, f A -l ::: PVC tubing 1 .. .. 200 2.22 2.05 2 .. . . 200 0.99 0.91 3 .. . . 200 0.61 0.66 4 .. .. 200 0.64 0.70 5 * . . . 200 0.61 0.67 6 .. .. 200 0.58 0-51 8 .. . . 200 3.79 3-88 77- . . . . 200 0.0 0.0 Dibutyl tin maleate in total extract, pg Type of Volume, f A \ blood transfusion set ml PAR Dithizone $PVCAl .. . . . . 47 20-7 22-6 A2 . . . . , . 46 20.8 21.6 B .. .. . . 66 130 125 c . . . . .. 41 0 0 * Autoclave a t 110" C, 1 hour, 3 cm2 surface per ml in distilled water. t This formulation stated to contain no alkyl tin. $ Set fillcd with distillcd water and maintained at 85" C for 1 hour.574 SAWYER Cremer are shown in Table 11.The results obtained by the two methods are in good agree- ment and in this respect one method is no better than the other; however, the reagent stability and convenience of handling of aqueous PAR solutions is far superior to that of the dithizone solutions and in this respect the proposed technique offers a significant advantage. The author wishes to acknowledge the assistance of Misses B. G. Cox and L. M. Grisley in the experimental work on the method, and thanks the Government Chemist for his per- mission to publish this paper. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. REFERENCES Aldridge, W. N., and Cremer, J. E., Analyst, 1957, 82, 37. Chapman, A. H., Duckworth, M. W., and Price, J. W., Brit. Plast., 1959, 78. Skeel, R. T., and Bricker, C. E., Analyt. Chem., 1961, 33, 428. Adamson, J. H., Analyst, 1962, 87, 597. Yasuda, M., and Tobias, R. S., Inovg. Chem., 1963, 2, 207. Tanaka, T., Komura, M., Kawaski, Y., and Okawara, R., J , Organometallic Chew., 1964, 1, 484. Barbieri, R., Faraglia, G., Giustiniani, M., and Roncucci, K., J . Inorg. Nucl. Chem., 1964, 26, 203. Roncucci, L., Faraglia, G., and Barbieri, R., J . Organometallic Chew., 1964, 1, 427. Pilloni, G., and Piazzogna, G., Analytica Chim. Acta, 1966, 35, 325. Dagnall, R. M., West, T. S., and Young, P., Talanta, 1965, 12, 583. Gagliardi, E., and Ilmaier, B., Mikrochim. Acta, 1967, 180. Pollard, F. H., Hanson, P., and Geary, W. J., Analytica Chim. Acta, 1959, 20, 26. Geary, W. J., Nickless, G., and Pollard, F. H., Ibid., 1962, 26, 575. “Specification for Transfusion Equipment for Medical Use,” British Standard 2463 : 1962, Appen- “Unplasticized PVC pipe (type 1420) for cold water supply,” British Standard 3505 : 1962, Gullbring, R., Vox Sang., 1964, 9, 513. U.S. Pharmacopoeia XVII, 1965, p. 904. dix F, 29. Appendix H, 1966 amendment. Received April 12th, 1967

ISSN:0003-2654    

DOI:10.1039/AN9679200569

出版商:RSC    

年代:1967

数据来源: RSC

摘要:

Analyst, September, 1967, Vol. 92, $@. 575-577 575 Quantitative Determination of Trace Amounts of Some Dialkyl Phthalates by Gas - Liquid Chromatography BY W. BUNTING AND E. A. WALKER (Ministry of Technology, Laboratory of the Govevnment Chemist, Cornwall House, Stamford Street, London, S.E. 1) Dialkyl phthalates a t concentrations of 1 to 10 p.p.m. in aqueous solution have been separated and determined by gas - liquid chromatography after extraction into hexane. The column comprised Chromosorb G coated first with 1 per cent. polyvinylpyrrolidone and then with 1 per cent. Carbowax 20M, and the effluent concentrations were monitored with an electron-capture detector. IN an investigation of the effects of prolonged contact of water with the plastics used in medical equipment, a method was required for the determination of dialkyl phthalates in aqueous solutions in the range of 1 to 10 p.p.m.Several examples are given in the literature for the separation of dialkyl phthalates by gas - liquid chromatography. Both polar and non-polar columns at low loading are used in conjunction with flame-ionisation or thermal conductivity detector~.1~2,3,4 These were found to be unsatisfactory from the quantitative point of view when working at the levels required. All the chromatograms exhibited severe tailing when using a DMCS-treated support such as Chromosorb G; furthermore, the large aqueous injections required to obtain peaks of measurable size resulted in the production of ghost peaks, thus adding to the difficulties of quantitative measurement. Other workers5 have used solvent extraction techniques to concentrate small amounts of plasticisers, but in our own experience, extraction into hexane eliminated the ghost peaks but had no effect upon the quantitative aspects of the method, whereas concentration of the plasticisers by solvent evaporation appeared to result in loss of dialkyl phthalate.Recently the qualitative detection of low concentrations of dialkyl phthalates in hexane solution by using an electron-capture detector has been reported.6 This detector was, there- fore, adopted for the investigation, as it was then possible to use moderately sized injections (2 pl). Concentration of the hexane solution was unnecessary although the problem of eliminating absorption effects still remained.In a critical study of silanisation of solid supports Kirkland' suggested that the low surface energy associated with the de-activated support made it difficult to coat with polar phases because of poor wetting of the surface; coating the stationary phase, however, with polyvinylpyrrolidone has been found to overcome this.8 Coating was therefore carried out on the DMCS support before addition of the stationary phase. The phases selected for detailed study were SE30, a non-polar phase, and Carbowax 20M, a polar phase, both in our experience well suited to prolonged use at low loadings and relatively high temperatures, As polyvinylpyrrolidone deteriorates at temperatures above 220" to 230" C, the column was first conditioned at 230" C for 24 hours, and then used at 190" C.At this stage some improve- ment was evident in quantitative work when using the Carbowax column. The SE30 column showed improvement, but to a lesser degree. This improvement was found to continue daily in the polar column when it was maintained at the working temperature until, after 10 days, peak asymmetry was reduced to a minimum and the column was found to give satisfactory reproducible quantitative results. As, however, this improvement might have been brought about through saturation of the active surface by repeated injection, two new columns were prepared and conditioned as before, omitting sample injection. After about 10 days, both columns were found to be satisfactory. A series of standard solutions of di-(3,5,5'-trimethylhexyl)phthalate in hexane was prepared and chromatographed. A plot of peak area vcvsus concentration over the range576 BUNTING AND WALKER : QUANTITATIVE DETERMINATION OF TRACE [AIza,!ySt, VOl.92 1 to 10 p.p.m. was prepared and found to be linear. It was also found that this linear relationship persisted up to the 100 p.p.m. level. No attempt has been made to extend it further. The lowest limit of detection has not been determined, but it seems reasonable to suppose that considerably lower concentrations could be detected by increasing the sample volume. EXPERIMENTAL The instrument used was an Aerograph 1520 fitted with a 1 : l stream splitter. The two equal effluent gas streams were monitored by flame-ionisation and electron-capture detectors to demonstrate the difference in response to both sample and solvent. Nitrogen used as carrier gas was dried by passing through molecular sieve 5A. The conditions for chromatography are given in Fig.1, which shows the separation obtained when chromato- graphing four dialkyl phthalates. I I I 20 15 10 5 0 Time, minutes Fig. 1. Chromatogram of 4 pl of a mixture of A, diethyl phthalate; B, dibutyl phthalate; C, di-(2-ethylhexyl) phthalate ; and D, di-(3,5,5'-trimethylhexyl) phthalate (each a t a concentra- tion of 1 p.p.m.) showing relative response to electron-capture (----) and flame-ionisation (- - - - ) detectors. Column, 1 per cent. Carbowax 20 M on DMCS Chromosorb G, 80 to 100 mesh, coated with 1 per cent. polyvinylpyrrolidone (5 feet x tinch, 0.d.) ; temperature, 190" C; flow-rate, 20 ml per minute; instrument, Aerograph 1520 ; sensitivity of electron-capture and flame- ionisation detectors 1 x 8 and 0.1 x 16, respectively COLUMN PREPARATION- Fifty millilitres of an ethanol solution containing 0.15 g of polyvinylpyrrolidone (Applied Science Laboratories) were added to 15 g of Johns Manville 80 to 100 DMCS Chromosorb G, i.e., polyvinylpyrrolidone equivalent to 1 per cent.of the support, then dried overnight in an oven at 105" C. The prepared support was then coated in a similar way with 1 per cent. of stationary phase by using a solution of Carbowax 20M in methylene chloride. After removal of the solvent, the column packing was dried at 105" C for 3 hours. The columnSeptember, 19671 AMOUNTS OF DIALKYL PHTHALATES BY GAS - LIQUID CHROMATOGRAPHY 577 (5 feet x &-inch 0.d.stainless steel) was then packed and conditioned overnight at 230” C in a stream of nitrogen, followed by 10 days at 190” C, after which it was ready for use. A column so prepared has been in continuous operation for 6 months without evident deterioration. Complete recovery of phthalates from aqueous solution was effected by extracting a 50-ml aqueous aliquot successively with 10, 5 and 5 ml of hexane that had previously been checked, and found free of phthalates.6 The hexane extracts were then bulked and shaken with about 0.5 g of anhydrous sodium sulphate to eliminate moisture. Within the limits of experimental error recovery was complete. Y 1 15 10 Time, minutes Fig. 2. Typical example of the type of chromatogram obtained after extraction of a poly(viny1 chloride) -type plastic : A, diethyl phthalate ; B, dibutyl phthalate ; C, di-(2-ethylhexyl) phthalate; and D, di-(3,5,5’-trimethylhexyl) phthalate Fig.2 is included to illustrate a typical example of the type of chromatogram obtained from an actual sample, by using only the electron-capture detector, from which it may be seen that di- (3,5,5’-trimethylhexyl) and di (2-ethylhexyl) phthalates are easily determined. If required, a better separation of the lower boiling dialkyl phthalates can be obtained by carrying out the separation at a lower column temperature, but for the present purpose only determination of di-(3,5,5’-trimethylhexyl) phthalate was required. The procedure for obtaining an aqueous extract from the plastic has been discussed by Sawyer.9 REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. Komers, R., Colln Czech. Chem. Commun., Engl. Edn., 1963, 28, 1549. Zulaica, J., and Guichon, G., Analyt. Chem., 1963, 35, 1724. Esposito, G., Ibid., 1963, 35, 1439. Leebrand, R. J., Facts Meth. Scient. Res., 1965, 6, 4. Diemair, W., and Pfeilsticker, K., Z . analyt. Chem., 1965, 212, 53. Lee, D. R., Britton, J., Jeffcoat, B., and Mitchell, R. F., Nature, 1966, 211, 521. Kirkland, J. J., “4th International Symposium on Gas Chromatography, 1963,” Academic Press, Vanden Heuval, W. J. A., Gardiner, W. L., and Horning, E. C., Analyt. Chem., 1963, 35, 1745. Sawyer, R., Analyst, 1967, 92, 569. Received April llth, 1967 New York, p. 77.

ISSN:0003-2654    

DOI:10.1039/AN9679200575

出版商:RSC    

年代:1967

数据来源: RSC

摘要:

578 Analyst, September, 1967, Vol. 92, $9. 578-580 Mobile Laboratory Methods for the Determination of Pesticides in Air Part III.* Mevinphos BY G. A. LLOYD AND G. J. BELL (Plant Pathology Laboratory, Hatching Green, Harflenden, Herts.) The collection of airborne droplets or vapour of mevinphos and the mobile laboratory method for its determination are described. PUBLISHED procedures for the determination of mevinphos? (Z-methoxycarbonyl-l-methyl- vinyl dimethyl phosphate) involve infrared spectrophotometry, gas - liquid chromatography, bioassay or cholinesterase inhibiti0n.l Colorimetric procedures are more practicable for use in a mobile laboratory for the reasons given in the introductory paper to this series.2 In order to measure concentrations of mevinphos in air in the region of the tentative threshold limit (0.1 mg per m3)3 a method, sensitive to about 5 pg, is required when air is sampled at normal breathing rates (10 to 20 litres per minute) for periods possibly as short as 5 minutes.A suitable procedure is described whereby mevinphos is hydrolysed in cold aqueous alkali - isopropyl alcohol to afford principally the keto-isomer of methyl acetoacetate, which slowly enolises in the solvent mixture. Addition of bromine to the double bond of the enol yields a bromo-ketone that reacts readily with cyanide ions to liberate cyanogen bromide. This is determined colorirnetrically by Aldridge's procedure4 with minor modifications,5 the major change being the substitution of p-phenylenediamine6 for benzidine in the chromogenic reagent because the latter substance is a potent carcinogen.The maximum colour is developed in 2 hours at 40" C, during which time the absorption peak shifts from 480 to 520 mp. The presumed reaction is- NaOH NaOH P.O.C=CH.COCH, CH,.C.CH,.COCH, + CH,.C=CH.COCH, , I It 0 CH,.C.CH.COCH, OBr 0 II I II The colour development stage of the procedure is precise but, for exact work, carefully controlled conditions of hydrolysis of mevinphos and enolisation of the resultant acetoacetate ester are necessary. The optimum conditions of time and temperature for the conversion of mevinphos, or an equivalent amount of methyl acetoacetate, to the enol form of the ester are similar (20 hours at 20" C), thereby indicating that mevinphos is rapidly hydrolysed. The yield of enol is, however, only 74 5 2 per cent.of that found by reacting the ester with alkali and no significant difference occurs in yield or stability of the enol, whether sodium or lithium hydroxide is used. Complex intermediate reactions occur and not all the pesticide is hydrolysed to the ester, acetone being the principal by-product.1 As acetone does not give the cyanogen bromide reaction under the conditions described, the reaction with alkali can be studied over a range of temperature or with time. The graphs show the importance of including standards for reference with a series of unknowns when, in particular, the optimum temperature cannot be maintained over the required period. * For details of Part I of this series, see reference list, page 580. t British Standards Institution, Recommended Common Names for Pesticides, B.S.1831 : 1965.LLOYD AND BELL 579 0 I 1 1 I 10 20 30 40 Time, hours, a t 20°C Temperature, "C Fig. 1. Optimum time, in alkali - isopropyl Fig. 2. Optimum temperature for 20 alcohol hours, in alkali - isopropyl alcohol METHOD APPARATUS- for air sampling.2 Dimpled bubblers-Whatman, GF/A, glass fibre filters, diameter 5.5 cm and filter holders Glass-stoppered tzibes-These are graduated at 20 ml, capacity 40-ml to stoppers. Water-bath, 40" C. REAGENTS- All reagents are of analytical-reagent grade. Trapping solzdion-Dilute 1 volume of 0.2 N aqueous sodium hydroxide to twice its Hydrochloric acid, 2.7 2 0.05 N-Mix 1 volume of hydrochloric acid (about 11 N) with Saturated bromine water-Store over bromine. Phenol - bromide-Dissolve 2 g of phenol and 5 g of potassium bromide in 100 ml of water.Sodium cyanide, 0.5 per cent. w/v in water. p-Pheizylenediamine hydrochloride-Dissolve 5 g in 100 ml of 2 per cent. v/v hydrochloric Pyridine hydrochloride-Mix 60 volumes of pyridine, 40 volumes of water and 10 volumes AZdridge's reagent-Mix 11 volumes of the pyridine reagent and 2 volumes of the Standard mevinphos-Stock solution, 1 mg per ml in isopropyl alcohol. Set the sampling devices in the breathing zones of operators in the manner described Vnpours-Charge each bubbler with 20 ml of trapping reagent and add more isopropyl volume with isopropyl alcohol. 3 volumes of water and adjust the normality against standard alkali. acid. of hydrochloric acid. j5-phenylenediamine reagent immediately before use.AIR SAMPLING- elsewhere. alcohol according to the intended period of sampling (Table I). Renew after 48 hours. TABLE I ADJUSTMENT OF BUBBLER CHARGINGS ACCORDING TO SAMPLING TIMES Intended sampling time, minutes . . 5 5 to 10 11 to 15 16 to 21 22 to 26 27 to 31 32 to 36 37 to 40 Added isopropyl alcohol, ml . . 2 4 5 7 8 9 10 11 At the end of the sampling period (5 to 30 minutes), transfer the contents of each bubbler to a graduated tube, adjust the volume to 20ml with isopropyl alcohol, stopper and stand it for 18 to 24 hours at room temperature (15" to 20" C). At the same time as the samples are taken, dilute an aliquot (about 1 mg) from the standard solution of mevinphos to 200 ml with the trapping solution. Draw air through each bubbler at 10 litres per minute.580 LLOYD AND BELL DroPZets OY dast-Draw air at 10 litres per minute through glass-fibre filters.Transfer the contaminated filters to tubes containing 25 ml of the trapping reagent. Stopper the tubes and stand them for 18 to 24 hours at 15” to 20” C. Filter if necessary and transfer aliquots (normally 20 ml) to graduated tubes, and adjust the volume, as required, to 20 ml with the trapping reagent. ANALYSIS- Add 1.0 ml of 2.7 N hydrochloric acid to 20-ml portions of the hydrolysed extracts and mix well. Then to the solution add 1.0 ml of bromine water, stopper, mix and allow to stand for 1 minute, then add 1.0 ml of phenol - bromide solution, stopper, mix well for 10 seconds. Ensure that no free bromine remains around the stopper.Add 1.0ml of sodium cyanide solution, stopper, mix, and shake for 30 seconds, then add 4.0 ml of Aldridge’s reagent; mix, and stand the stoppered tube in a water-bath a t 40” C for 2 hours. Measure the optical density of the solution at 520 mp through a l-cm light path. Deduct the appropriate “air’, or reagent blank and refer to a calibration graph prepared from 0 to 200-pg amounts of mevinphos hydrolysed under identical conditions of time and tem- perature. The reagent blank is normally equivalent to 4 & 1 pg of mevinphos. Make the appropriate corrections when aliquots have been taken from the original extract and multiply the amount of mevinphos found in a bubbler and on a glass-fibre filter by 10/9 and 10/9-5, respectively, to allow for the “slip” through the bubbler (10 per cent.) and a minimum recovery of 95 per cent.from the glass-fibre filters. When high accuracy is not important, and a + l o to 20 per cent. error is acceptable, reference may be made to a calibration graph prepared in a similar manner from 0 to 100-pg amounts of methyl acetoacetate. The results obtained from the graph are multiplied by 1.93 to convert to micrograms of mevinphos and again by 10/7 to correct for a possible minimum conversion of mevinphos to the ester (normally about 74 per cent.). The calibration graph is linear over the range 0 to 200 pg of mevinphos. REFERENCES 1. Porter, P. E., Yun-Pei Sun, and Archer, T. E., in Zweig, G., “Analytical Methods for Pesticides Plant Growth Regulators, and Food Additives,” Academic Press, New York and London, 1964, Volume 2, p. 351. Lloyd, G. A., and Bell, G. J., Analyst, 1966, 91, 806. American Conference of Governmental Hygienists, Threshold Limit Values for 1964, Archs Envir. Hlth, 1964, 9 (4), 545. Aldridge, W. N., Analyst, 1945, 70, 474. Bakes, J. M., and Jeffery, P. G., Talanta, 1961, 8, 641. Bark, L. S., and Higson, E. G., Ibid., 1964, 11, 471. 2. 3. 4. 5. 6. NOTE-Reference 2 is to Part I of this series. Received Januavy 23rd, 1967

ISSN:0003-2654    

DOI:10.1039/AN9679200578

出版商:RSC    

年代:1967

数据来源: RSC

摘要:

Analyst, September, 1967, Vol. 92, $9. 581-583 581 A Micro Method for the Determination of Capsaicine in Capsicum Fruits BY M. S. KARAWYA,* S. I. BALBAA," A. N. GIRGIS AND N. 2. YOUSSEF (The Research and Control Centre, The Egyptian Organisation for Pharmaceuticals, Chemicals and 1Vedical Appliances, Giza, U.A . R.) A micro method for the determination of capsaicine in capsicum fruits is described. The fruits are extracted with 70 per cent. ethanol and chromato- graphed on kieselguhr plates. The capsaicine spots are eluted, coupled with diazobenzenesulphonic acid and the resulting red colour measured in a suitable spectrophotometer. The method is sensitive to amounts of cap- saicine ranging between 10 and 200 (ug. The preparation and purification of capsaicine on an alumina column is also reported.THE Panel of the Joint Committee of the Pharmaceutical Society and the Society for Analytical Chemi~tryl?~ reviewed the different methods of assay of capsicum fruits. The Panel suggested a method involving the use of diazobenzenesulphonic acid and sodium nitrite, together with sodium iodide, sodium hydroxide and hydrochloric acid to produce a measurable red colour with capsaicine. The sensitivity range of the method is 350 to 700pg of capsaicine. The present work made use of the thin-layer chromatographic techniques for the separa- tion of capsaicine from extracts of capsicum fruits before determining it colorimetrically. EXPERIMENTAL MATERIAL AND REAGENTS- Pure capsaicine-This is prepared by extracting the powdered dry capsicum fruits with light petroleum, then concentrating and cooling the extract.The deposited crude crystals, after being washed with cold light petroleum, are dissolved in ether and mixed with a few grams of alumina. The ether is allowed to evaporate, and the powder inserted on the top of a column of alumina. The impurities are eliminated by washing with isopropyl alcohol - light petroleum - water (23 + 75 + 2) and then capsaicine is eluted with isopropyl alcohol - light petroleum (94 + 6) and repeatedly crystallised from light petroleum - ether (9 + 1). Tests of purity of the prepared capsaicine-Melting-point 65" C; mixed melting-point with authentic capsaicine, 64.5" to 65" C; on chromatographing a mixture of prepared and authentic capsaicine on a kieselguhr plate, only one spot is obtained when developed with light petroleum - ethanol (99 + 1).Ethanol, absolute-B.P. quality. Ethanol, 70 per cent. v/v. Light petroleum, 40" to 60" C. Kieselguhr, acid washed. Colour reagent-Prepare a fresh solution of 0.1 per cent. diazobenzenesulphonic acid3 in 8 per cent. aqueous solution of sodium carbonate. The diazonium salt is prepared by dis- solving 25 g of sulphanilic acid in 125 ml of 10 per cent. potassium hydroxide solution, cooling and adding 100ml of 10 per cent. sodium nitrite solution. Add the solution dropwise to an ice-cold mixture of 40ml of hydrochloric acid and 20ml of water, with stirring. The temperature of the reaction is not allowed to exceed 8" C. Filter the diazonium salt, wash it successively with ice - water, ethanol and ether, then dry it in air.* Permanent address : Pharmacognosy Department, Faculty of Pharmacy, Cairo University, Cairo, U.A.R.582 KARAWYA et al. : A MICRO METHOD FOR THE [Analyst, Vol. 92 PROCEDURE- In a spectrophotometric study of the red colour resulting from coupling capsaicine with diazobenzenesulphonic acid reagent, a Unicam SP700 spectrophotometer was used. The optimum wavelength for the measurement of the colour is found to be 505 mp, at which wavelength the colour of the excess of reagent does not interfere. The colour is found to develop well at pH 11.5; the intensity remains stable for about 6 hours and is not affected by light and a temperature range of 12" to 32" C. The colour is found to obey Beer's law in a concentration range of 10 to 200 pg per 5 ml.DEVELOPMENT AND MEASUREMENT OF THE COLOUR- Subject to the above precautions, the following method is recommended. Add 0.5 ml of 0.1 per cent. diazobenzenesulphonic acid in 8 per cent. aqueous solution of sodium carbonate to 4ml of capsaicine (10 to ZOOpg) solution in 70 per cent. ethanol and make up to 5 ml with 70 per cent. ethanol. Measure the developed colour by means of a IJnicam SP600 spectrophotometer at 505 mp with a No. 2 filter and the blue photocell against a blank of 4-5 ml of 70 per cent. ethanol and 0.5 ml of the colour reagent. SEPARATION OF CAPSAICINE FROM EXTRACTS OF CAPSICUM FRUITS BY THIN-LAYER CHROMATO- A slurry was prepared by mixing 5 parts of kieselguhr and 0.1 part of starch with 18 parts of water, and heating at 80" C for 10 minutes.Plates (10 x 18 cm) were coated with this slurry, adjusted to 0.3 mm thickness, air dried, activated at 105" C for 30 minutes and allowed to cool in a desiccator. Chromatographic runs of capsaicine and 70 per cent. ethanolic extracts of capsicum fruits were made with light petroleum (40" to 60" C) - absolute ethanol (99 + 1) as a developer and the colour reagent for location of spots. A good separation of capsaicine (R, 0.58) from the pigments of the extract (R, 0.87) was obtained. QUANTITATIVE RECOVERY OF CAPSAICINE FROM KIESELGUHR PLATES- The plates were loaded with volumes corresponding to amounts of capsaicine ranging from 10 to 200 pg, with an Agla micrometer syringe. Two peripheral pilot spots were made on each plate. After develop- ment, the central part of the plate was covered with a sheet of glass and the two peripheral pilot spots were located by spraying with the colour reagent.The non-sprayed developed spot areas, which were at the level of the pilot spots, were separately scraped with a thin spatula. The powder obtained from each spot area was collected on a glazed paper and transferred quantitatively to a micro column (0.4 x 10 cm). Complete elution of capsaicine was achieved with 10 ml of 70 per cent. ethanol. The volume was concentrated on a water-bath to about 4m1, 0 5 m l of the colour reagent was added, the volume adjusted to 5 ml with 70 per cent. ethanol and the developed colour measured at 505 mp. From the results shown in Table I it is clear that capsaicine can be recovered almost quantitatively.GRAPHY- Serial dilutions of capsaicine in ethanol were prepared. TABLE I RECOVERY OF CAYSAICINE FROM KIESELGUHR PLATES Capsaicine added, pg . . . . . . . . 10 20 50 100 150 200 Capsaicine recovered, p g . . . . . . 10 19.8 48 101 146 195 DETERMINATION OF CAPSAICINE IN CAPSICUM FRUITS- The above method was applied for the quantitative determination of capsaicine in the fruits of different species of capsicum, vix., C. mi~~imum, C. frutescens and a commercial sample of capsicum fruits obtained from the local market. One gram of a representative sample of the powdered fruits was extracted with 70 per cent. ethanol in a small continuous extrac- tion apparatus until it was exhausted (tested by adding the colour reagent to drops of the extract) and the extract was concentrated on a water-bath to 10 ml. An amount of the extract containing 50 to 150pg of capsaicine was applied with an Agla micrometer syringe on the starting line of a kieselguhr plate which was developed and eluted, the eluate being treated with the colour reagent and the developed colour measured as described above.September, 19671 DETERMINATION OF CAPSAICINE IN CAPSICUM FRUITS 583 The results, as shown in Table 11, are then deduced from an absorbance - concentration graph of the capsaicine recovered from kieselguhr plates. TABLE I1 PERCENTAGE OF CAPSAICINE IN THREE SAMPLES OF CAPSICUM FRUITS Percentage of capsaicine I Experiment C. minimum C. frutescens Commercial sample 1 0.746 0-417 0.596 2 0.747 0.421 0.593 3 0-748 0.419 0.596 4 0-750 0-420 0.592 5 0.749 0.423 0.598 Mean 0.748 0.420 0.595 REFERENCES 1. 2. __ , Ibid., 1964, 89, 377. 3. Report of the Joint Committee of the Pharmaceutical Society and the Society for Analytical Chemistry, Analyst, 1959, 84, 603. Waldi, D., “Chromatography,” E. Merck A.G., Darmstadt, Second Edition, p. 68. Received September 5th, 1966

ISSN:0003-2654    

DOI:10.1039/AN9679200581

出版商:RSC    

年代:1967

数据来源: RSC