attributed in part to: the change of the residual blank absorbance of a brown yellow dimeric iron(II1) phenanthroline species with increasing concentration of sulfur dioxide; incompleteness of the reduction of iron(II1); and the evaporation of solution when passing air. The evaporation error at 50 "C amounts to approximately 4% after passing 50 liters of air through the sample. If the standardization is carried out using sulfur dioxide permeation tube, this error is reduced to less than 1 %. The calculated absorbance for the reduction of iron(II1) with sulfur dioxide from aqueous bisulfite at room temperature and the formation of tris-1 ,lo-phenanthroline iron(I1) is in good agreement with the following reaction:
+
+ +
+
2 Fe(III)Ac,pher~,~-~ 2S02 2H20 (6 - 2 y ) phen $ 2 Fe(II)phen32+ 2 x AcS2OB2- 4 H+
+
Thus, the slow change of absorbance after 15 minutes can be in part attributed to the secondary reaction. In the gas flow system at 40 "C the calculated absorbance corresponds to the oxidation of sulfur dioxide to sulfate by the reaction:
+ SO2 + 2 H 2 0 + (6 - 2y)phen e 2 F e ( I I ) ~ h e n ~+ ~ +2xAc- + 4 H+ + S042-
2 Fe(III)pheqAcZ3-'
Amounts of formaldehyde, nitric, nitrous oxide, and mercaptans comparable to sulfur dioxide concentration had no apparent effect on the absorbance. Hydrogen sulfide interferes seriously and must be removed; chloride interferes if present in excess of 100 ppm.
+
In this reaction sulfur dioxide is oxidized to dithionate which is quite inert and can be oxidized only rather slowly to sulfate.
RECEIVED for review March 9, 1970. Accepted June 19, 1970. Work partially supported by the Institute of Gas Technology.
Radiopolarograhy of Thallium R. A. Culp' and A. F. Findeis2 Department of Chemistry, University of Alabama, University, Ala. 35486
SINCE1958, SEVERAL PAPERS have appeared in the literature concerning the application of the polarographic method to solutions containing trace quantities of radioactive isotopes (1-4). If the radioactive isotope is polarographically reduced to an oxidation state which associates with the mercury drops, the drops can be separated from solution and the radioactivity associated with each drop can be measured. The purpose of this note is to show, first, that diffusion coefficients for thallium calculated from radiopolarographic data differ considerably from those calculated from polarographic data. Second, by calculating the number of atoms associated with the drops from both radiopolarographic and polarographic data, it was possible to measure a quantity apparently related to the residual charge on the drop, Finally, in an effort to rapidly convert the measured radioactivity to absolute activity, a new sample preparation technique was developed for counting @-radiationusing a cylindrical geometry foil flow counter. In previous radiopolarographic work, relative measurements have been made, but there has been no attempt to relate the radioactivity, on an absolute basis, to the number of coulombs determined from polarographic results. The comparisons of the radiopolarographic and polarographic data show that the radioactivity associated with the drops is a combination of the diffusion controlled process and specific adsorption or postelectrolysis that occurs after the drop has detached from the capillary.
In the measurements of @-radiation, care must be taken to correct for the nature of the sample and the characteristics of the counting system. These have been identified in some detail by Libby (5). Self absorption by the sample is of primary concern. In order to eliminate this, several techniques have been used. Love ( I ) suggested that the reduced radioactive isotope be dissolved from the mercury drop with a small amount of acid followed by mounting the sample on a thin plastic film and evaporating to dryness prior to counting with an end-window or a 4-.lr counter. Wildman and Schaap (2) placed the radioactive drop in the center ring of a ringed planchet and evaporated the mercury at low heat on a hot plate sealed in a vacuum desiccator. These techniques are very time consuming and are of little value with short lived isotopes. Ferenczy (3), neglecting any absorption of the @-radiation by the mercury drop, recommended the use of Plexiglas planchets that are conical shaped in order to have reproducible geometry for counting. Blazek and Wagnerova (4) investigated the effect of the time of collection of the mercury drops on the measured radioactivity. They concluded that by collecting the drops for short periods the absorption could be neglected for those isotopes studied and chose to use a technique similar to that used by Ferenczy. Even though this technique is used for those isotopes with relatively high penergy, it is undesirable since the conversion to absolute activities is somewhat speculative.
Present address, Department of Chemistry, Kent State University, Kent, Ohio 44240 a Present address, Chemistry Section, National Science Foundation, Washington, D. C. 20550
EXPERIMENTAL
~
(1) D. L. Love, Anal. Chim. Acta, 18,72 (1958). (2) E. Wildman and W. B. Schaap, Abstracts, 142nd National Meeting, American Chemical Society, Atlantic City, N. J., Sept. 1962, p 18B. (3) Z. Ferenczy, Acta. Chim. Hung. Tomus, 26, 229 (1961). (4) J. Blazek and D. M. Wagnerova, Collect. Czech. Chem. Commun., 29,915 (1964).
Apparatus. The polarograph used was a Sargent Model XV Polarograph. The polarographic cells were made from borosilicate glass tubing with one sidearm for degassing and another to pass nitrogen over the solution. The mercury pool was used as the reference electrode. The polarographic cells were immersed in a Sargent Thermonitor constant temperature water bath. All radioactive counting data were ( 5 ) W. F. Libby, ANAL.CHEM., 29, 1566 (1957).
ANALYTICAL CHEMISTRY, VOL. 42, NO. 11, SEPTEMBER 1970
0
1285
4600
-
- 2.0
0
2 4 6 8 [THALLIUM], M/L
X
1 IO‘
0
possible with a hypodermic syringe followed by removal of the remaining solution with filter paper. The drops were then mounted on copper planchets. Preparation of Sample Mounting Planchets. In initial studies rectangular (1 3/4-in~h by 1-inch) planchets were prepared from copper sheet. The planchets were cleaned with nitric acid and amalgamated, in the center, with a thin film of mercury. All mercury and copper salts were then washed from the surface of the planchet and it was bent to fit the cylindrical brass sample holder of the Geiger counter. The radioactive mercury drops were placed on the amalgamated planchet and spread over a controlled area to provide a constant area per drop. After it was observed that the collection of five drops of mercury was adequate for the measurement of reproducible radiopolarograms, commercial (15/18inch diameter) copper planchets were used. Polarograms from the collection of only one drop were satisfactory, but the reproducibility of the measurements was less than desired. This was in a large part associated with the manipulation of the dipper during collection of the single drop.
Figure 1. Comparison of radiopolarography and diffusion current measurements for a 204TI solution on addition of inactive thallium
obtained with a decade scaler, Model N-221, supplied by Hamner Electronics Co., Inc. The Geiger counter used in this work was a foil flow counter which has been described by Williams and Findeis (6) and is similar to a counter described by Sugihara, Wolfgang, and Libby (7). The dropping mercury electrode was constructed of a 15-cm length of marine barometer tubing and had a drop time in 0.1M KC1 of 4.76 sec and a mass flow rate of 1.77 mg per sec at a mercury column height of 73 cm and an applied potential of -0.60 V. Reagents. The zo4Tl samples were purchased from the Nuclear Research Chemicals Division of Mallinckrodt Chemical Works. The samples contained 5 mCi of zo4Tl per 0.125 ml of solution and had a specific activity of 1390 mCi per gram of TI which indicated that the zo4Tl was 0.278x of the total thallium in the sample. Thallium nitrate was used for addition of nonradioactive thallium to the solutions. Reagent grade KCl was used as the supporting electrolyte and distilled water was used for making all dilutions. No maximum suppressor was used. Procedure. The zo4Tl sample was transferred to a 25-ml volumetric flask and diluted to volume with distilled water. Further dilutions were made from this stock solution by placing the desired amount of the radioactive solution into a 25-ml volumetric flask to which 10 ml of 0.250M KC1 had been added. Ordinary thallium was added when desired followed by dilution to volume. This solution was added to the polarographic cell and placed in the constant temperature water bath until an equilibrium temperature of 25.0 ==! 0.1 “ C was reached. The solutions were not degassed after preliminary experiments revealed only a slight decrease in activity deposited in the drop on the plateau of the radiopolarograms with the removal of dissolved oxygen. Drops were collected at potentials corresponding to the polarographic wave of thallium by placing a small glass dipper in the solution under the dropping mercury electrode and catching the drops as they fell. The dipper consisted of a small cup constructed from 4-mm glass tubing and was 4 mm deep. A small curved handle of 1-mm glass rod was attached to the tubing so that the dipper could be manually manipulated in the cell. The potential was applied to the cell only during the time that the drops of mercury were collected. The radioactive solution was removed from the dipper by first removing as much as
(9 F. W. Williams and A. F. Findeis, ANAL.CHEM., 37,857 (1965). (7) T. T. Sugihara, R. L. Wolfgang, and W. F. Libby, Rev. Sei. Insfrum., 24, 511 (1953). 1286
RESULTS AND DISCUSSION
Solutions were prepared which contained 3.8 x 10-8M *04Tl(1.4 X l W 5 MT1) and drops were collected at potentials corresponding to the plateau of the radiopolarographic wave. The radioactivity associated with the drops was measured and corrected to absolute activity. A plot of the corrected radioactivity us. the number of mercury drops collected yields a straight line which passes through the origin, for the collection of over 25 drops or a collection time of over two minutes, thus indicating that there was uniform absorption of the /3-radiation by the thin film of mercury. This differs from observations by Blazek and Wagnerova ( 4 ) in which they reported that the measured radioactivity was not linear with the number of drops collected but had a negative deviation with increasing collection time. Their results were due to an increasing selfabsorption by the mass of mercury in the sample holder. This self-absorption has been made uniform by the sample mounting technique used in this work. The radioactive samples are very tenaciously held by the amalgamated planchet and even if dropped, very little, if any, of the radioactive sample will be lost. The fact that the radiopolarograms have the same shape as polarograms obtained by electrochemical methods is in itself misleading. Log A/(Amax, A ) us. E plots do not yield the same slope as the usual log i/(id - i) us. Eplots usually providing a value smaller than unity for the number of electrons transferred in the reduction of Tl(1). Straight line plots of the measured radioactivity us. the concentration of *o4Tl in solution are obtained only if the ratio of radioactive to nonradioactive thallium is a constant. Using the appropriate correction factors for conversion to absolute activity (8) and calculating the diffusion coefficient from the Ilkovic equation, we cm2 sec-’ for the solution find a value for D of 8.4 X obtained from the supplier, which after dilution was 1.4 x lO-5M Tl(1). Upon addition of nonradioactive thallium, keeping the Zo4Tlconcentration constant, the radioactivity associated with each drop of mercury decreases so that the calculated value of the diffusion coefficient based on a radioactivity measurement decreases to 2.6 X 10-5 cm2 sec-1 for a solution that is 1.0 X 10-aM in total thallium concentration. The values of the diffusion coefficient for T1 in 0.1M KC1 which are reported in the literature range from 1.78 X 10-5 to
-
(8) E. Bleuler and G. J. Goldsmith, “Experimental Nucleonics,” Reinhart and Co., Inc., New York, N. Y., 1952.
ANALYTICAL CHEMISTRY, VOL. 42, NO. 11, SEPTEMBER 1970
cm2 sec-1 (9-11). A value of 2.05 X 10-5 cm2 sec-1 was obtained by conventional polarography in this work. It should be noted that the calculation of absolute activity in the drops amalgamated on the planchet requires substantial correction for self absorption. This correction was 1.79 times the measured activity for a sample containing 5 drops of mercury from the electrode uniformly spread over 1 cm2. Figure 1 shows a comparison of the radioactivity associated with each drop of mercury along the plateau of the radiopolarograms with the limiting current from the electrochemical polarograms at various concentrations of thallium. Since our results for the calculation of the diffusion coefficient from radiochemical data are initially high and upon addition of nonradioactive thallium approach the value determined from electrochemical data, it is clear that some process in addition to diffusion is required to explain the level of activity associated with the drops. Wildman and Schaap ( 2 ) suggested that post-electrolysis may occur after the drop has detached from the capillary. Delahay and coworkers (12-14) have studied thallium compounds with respect to the electrical double layer with cation as well as anion specific adsorption, and have measured surface charges corresponding
2.0 X
(9) J. J. Lingane and J. M. Kolthoff, J . Amer. Chem. SOC.,61, 825 (1939). (10) G. W. Smith and F. Nelson, ibid., 76, 4714 (1954). (11) Von H. Strehlow, 0. Madrich, and M. V. Stackelberg, Z. Elecrrochem., 55, 250 (1951). (12) P. Delahay and G. G. Sushielles, J . Phys. Chem., 70, 647 (1966). (13) G . G . Sushielles, P. Delahay, and E. Solon, ibid., p 2601. (14) B. Baron, P. Delahay, and D. J. Kelsh, J. Electroanal. Chem., 18, 187 (1968).
to 11 pcoulombs/cm2. In either case it is assumed that the increased amount of radioactivity associated with the mercury drops at the various thallium concentrations is due to the residual charge on the drop. A value related to this charge can be estimated by calculating the number of coulombs required to account for this difference. The charge on the drop that is effective in either post-electrolysis of specific adsorption of thallium is 24.6 ycoulombs/cm2, which is not in bad agreement with that reported by Delahay (although in the work reported by Delahay the concentration ranges were 2 to 3 orders of magnitude larger than those reported here). This indicates that the use of diffusion coefficients from conventional polarography may lead to erroneous results if applied to calculations of mass transport into the mercury drops in radiopolarography. The difference between diffusion coefficients from conventional polarography and radiopolarography, even when the large excess of inactive thallium is present, is outside the error inherent in the conversion of measured activity and represents a true differencewhich will be discussed at a later date. This technique provides a useful experimental tool for the study of electrochemical processes which may not be studied by conventional means. It has been used in obtaining radiopolarograms from solutions prepared from exempt quantities of radioactivity (less than 1 pCi) as well as for the licensable quantities reported in this work. A detailed report of the effect of other cations and anions in buffered and unbuffered systems on the radiopolarography of thallium is in preparation.
RECEIVED for review July 7, 1969. Resubmitted July 6, 1970. Accepted July 6,1970.
Determination of Water in Ion-Exchange Resins: Anion Exchange Resins H. D. Sharma and N. Subramanian Department of Chemistry, University of Waterloo, Waterloo, Ontario, Canada
DURING THE COURSE of study of proton magnetic resonance (PMR) and of kinetics of ion-exchange in ion-exchange resins ( I , 2), we have investigated various methods of determination of water in the resin matrix. The methods are: drying a resin sample at a suitable temperature and pressure in the presence or in the absence of a dehydrating agent ( 3 , 4 ) ; titrating the sorbed water against the Karl Fischer reagent (5-8); and recording the proton magnetic resonance spectra (1) H. D. Sharrna and N. Subramanian, Can. J. Chem., 48, 917 (1970). (2) H. D. Sharma, R. E. Jervis, and L. W. McMillan, J. Phys. Chem., 74, 969 (1970). (3) K. W. Pepper, D. Reichenberg, and D. K . Hale, J. Chem. Soc., 1952, 3129. (4) W. R. Heumann and F. S . Rochon, Can. J. Chem., 43, 3483 (1965). ( 5 ) A. Dickel and J. W. Hartrnann, 2.Phys. Chem. (Frankfurt am Main), 23, 1 (1960). (6) F. X. Pollio, ANAL.CHEM.,35, 2164 (1963). (7) W. R. Heurnann and F. D. Rochon, ibid., 38, 638 (1966). (8) E. Blasius and R. Schmitt, Z. Anal. Chem., 241, 4 (1968).
of a suspension of resin beads (9). A comparative study of the applicability of these methods for cation exchange resins was reported by us earlier (10). For anion exchange resins, the drying method has been followed extensively. In our study, a white organic residue was observed in a liquid nitrogen trap when the resins in various ionic forms were dried at 50-55 OC under 1-mm pressure. The Karl Fischer method gives accurate values of water content for cation exchange resins of the sulfonic acid type (Dowex AG 50W), but its applicability has not been examined for anion exchange resins. The PMR method relies on the measurement of the chemical shift and the areas of the “external” and “internal” water signals (Figure 1). This method can be expected to yield reliable results if the two signals are well separated. In this note we present a comparative study of water determination in Dowex (9) J. P. de Villiers and J. R. Parrish, J. Polym. Sci., Part A , 2, 1331 (1964). (10) H. D. Sharrna and N. Subrarnanian, ANAL.CHEM., 41, 2063 (1969).
ANALYTICAL CHEMISTRY, VOL. 42, NO. 11, SEPTEMBER 1970
1287
