Historical Articles

September, 1952 issue of Plating

 

Radioactive Isotope Dilution Method for
Determining Sulfate Concentration in
Chromium Plating Baths

By Stanley L. Eisler

---

ABSTRACT
A procedure employing the isotope dilution technique is described. The method has been found to give results which are accurate to within 1 to 3 per cent of the true values of sulfate concentration. The reproducibility of the method was found to be very good as evidenced by a 2 percent average deviation from the mean in a series of seven samples run on two different occasions. As had been expected comparative tests by the gravimetric procedure consistently gave higher results than the radiometric method owing to the errors caused by co-precipitation.

INTRODUCTION
The accuracy of the sulfate ion determination in chromium plating solutions is very important because of the low concentration of sulfate ion in the bath and the necessity for maintaining as near as possible an optimum ratio of chromium trioxide to sulfate ion.

As pointed out in another article¹ the presently used gravimetric method of analysis by which the sulfate ion is precipitated as barium sulfate has proven unreliable because of co-precipitation of other anions and cations and the possibility of dirt or other impurities being collected in the precipitate. According to Karaoglanov² the amounts of ions contaminating the precipitate are related to the solubility of the correspond Mg sulfate and the solubility of the corresponding barium salt, in the case of anions. He claims that this cannot be explained on the basis of adsorption, but on the assumption that with the formation of barium sulfate, a secondary chemical precipitation process takes place in which the cations and anions in question participate. In another article³ Karaoglanov states that many procedures give good results only because of compensation of errors and he theorizes that such products as (BaCl)₂SO₄ are formed and affect the determination proportionately to their insolubility.

This investigation was inaugurated to eliminate the shortcomings noted above and, if possible, to reduce the time required for the analysis.

THE ISOTOPE DILUTION METHOD
The basic principle of the isotope dilution procedure as given by Friedlander and Kennedy⁴ involves the addition to the unknown mixture of a known weight of the compound to be determined, the tracer containing a known amount (activity) of radio-actively tagged molecules. After the pure compound has been isolated from the mixture its specific activity (activity per unit weight), is determined, and compared with that of the added compound the extent of the dilution of the tracer shows the amount of inactive compound present in the original unknown. The formula for calculating the weight of the unknown from the specific activities is presented later.

A very important advantage of the isotope dilution method, according to Schweitzer and Whitney⁵, is that a quantitative isolation of the final compound is not at all necessary. For this reason, it was thought possible to use in place of the conventional barium chloride a precipitating agent which produces a less insoluble product. If this were possible, many of the shortcomings of the barium sulfate precipitation method would be eliminated.

Fig. 1. Filtration appartatus

PRELIMINARY EXPERIMENTS
The radiosulfur used in this investigation was purchased from the Oak Ridge National Laboratory by authorization of the Atomic Energy Commission, Isotopes Division. It was received in the form of dilute H₂S³⁵O₄ and added to a standardized H₂SO₄ solution to give an activity level of approximately 0.05 microcurie per milliliter.

A number of tests were conducted using benzidine dihydrochloride as the precipitating agent in place of barium chloride. Tests on dilute sulfuric acid alone provide very successful. However, when precipitation of- the benzidine sulfate from a reduced chromic acid solution (approximately of plating solution concentration) was attempted, it, proved unsuccessful. This result was attributed to the very high acidity of the solution, caused by the acids used in the reduction of the hexavalent chromium.

The use of benzidine dihydrochloride having proved unsuccessful, several other precipitating agents, such as strontium chloride, lead nitrate, and antimony chloride were tried. They also failed because the solubility products of the sulfates were higher than the concentrations found in the plating solutions.

The next step, therefore, was to develop a method using barium chloride which would eliminate the effects of co-precipitation that cause the inaccuracies in the gravimetric procedure. The new method is based on the assumption that precipitates of controls and samples alike would contain the same amount of impurities when prepared from similar solutions. The amount of impurities cannot be considered negligible but, at least should be comparable and for that-reason their effect on the true result should be minimized.

PREPARATION OF PRECIPITATE
During the preliminary tests using benzidine dihydrochloride and sulfuric acid solutions, the benzidine sulfate precipitate was prepared by adding the precipitant to the acid, centrifuging, decanting, recentrifuging, decanting and then transferring the precipitate to a stainless steel cup. The precipitate was then dried and counted. It was believed that when this method was applied to the tank solutions, using an aliquot portion, the method would be much shorter than the lengthy filtration procedure. Transfer of the precipitate to the cups is not a critical step since a quantitative separation is not necessary when utilizing the isotope dilution method. However, counting results obtained by this method indicated that the non-uniform thickness of the precipitate over the entire cup area produced errors.

To produce a precipitate of uniform thickness over the entire counting area, it was decided to use a filtration apparatus as shown in Fig. 1.

This apparatus utilizes a plastic funnel-with a .871 inch diameter top surface. The top surface has a slight spiral groove and cross lines cut into the surface to facilitate flow of the liquid to the four holes spaced at 90° intervals near the outer rim of the funnel. These four holes are connected by diagonal holes through the thickness of the funnel to a center larger hole which drains into the suction flask. The holes in the top surface were spaced so as to avoid build up of a precipitate at the center as was encountered when only one central hole was employed. The filter paper disc is mounted on the top surface of the funnel and the a tight fitting plastic cylinder with an inside diameter of .702 inch is pressed over the disc. This results in a uniform precipitate with a diameter of .702 inch and an area of 2.5 cm². Use of this apparatus also produces a preformed filter paper disc which may be easily mounted on an aluminum disc and then held in place by an aluminum ring which slips over the disc and paper.

SELF ADSORPTION CONSIDERATIONS
According to Aten6 there are three factors which can influence the efficiency of the counting measurement of a sample, namely, back-scattering, size, and self adsorption. In the work done, the first two are made essentially constant by standardization of the material of the disc and the size of the precipitate, respectively. However, the problem of self adsorption was more difficult as the precipitate weights and correspondingly the weights per unit area (or sample thickness as generally defined in radiochemistry) varied considerably from sample to sample, due to the varying amounts of sulfate in the unknown. Schweitzer and Stein7 state that the true activity is diminished by an amount dependent on sample thickness. However, they also state that this effect may be overcome by the use of infinitely thick samples where the measured activity is proportional to the specific activity.

Since the specific activity is the value required for calculation of the extent of dilution in the isotope dilution method, it was decided to increase the sample size so that in all cases a sample of infinite thickness would be obtained. By doing this it was found that all weighings could be eliminated. This was possible as a value proportional to specific activity is obtained directly, while normally it is necessary to divide the true activity by the weight to get the true specific activity which is then used in the isotope dilution formula to determine the weight of the unknown.

The amount of sample required was based on work previously done, which indicated that there was no increase in counting rate above a sample thickness of approximately 25 mg/cm² in the case of BaSO₄ precipitates. This is shown graphically in Fig. 2.

Fig. 2. Sample thickness of BaSO4 versus counting rate.

FINAL APPROVED TEST PROCEDURE
The final approved procedure for sulfate analysis in chromium plating tank solutions is as follows:

1. Pipette a 15 ml sample into a 250 ml beaker.
2. Add 1.5 ml of approximately N/10 H₂SO containing about .05 microcurie of H₂S³⁵O₄ per ml.
3. Add 50 ml of water, 10 ml concentrated hydrochloric acid, 15 ml glacial acetic acid and 20 ml ethyl alcohol.
4. Boil for 15 minutes.
5. Add 15 ml of 10 per cent BaCl₂ solution. Boil one minute and then allow to stand at least one hour.
6. Filter each sample, using a 1-1/8 inch disc of S & S No. 597 filter paper mounted on the filtration apparatus shown in Fig. 1.
7. Mount the filter paper disc on an aluminum disc and ring assembly, dry under an infra-red lamp, and count.

All counting was done by placing the sample on the first shelf of a lucite mount located in a lead shield with 1/2 inches of lead shielding, so that the sample was 10 mm from the window of the counter tube. The counter tubes were Tracerlab Model TGC-2 tubes with mica window thicknesses of 1.8 and 1.9 mg/cm², respectively.
The control samples are prepared by placing 15 ml . of a 25 percent CrO₃ solution in a 250 ml beaker and adding 7.5 ml of the N/10 H₂SO₄ solution containing the radioactive sulfuric acid. Then the procedure previously given is followed from steps 3 to 7 inclusive.

The formula for calculating the weight of unknown by use of the isotope dilution method is as follows:

where W = Weight of unknown
Wa = Weight of known added
Sa = Specific activity Of control
Sb = Specific activity of mixture

As previously stated, the specific activity is proportional to the counting rate of infinitely thick samples, so the counting rates may be substituted directly in the above equation for the values Sa and Sb.

Expanding this formula to cover the conditions of the proposed test procedure and to have the result in oz gal of SO⁼₄, as usually expressed in plating circles, the following steps are involved.

Normality of acid used for control x 49 = mg of H₂SO₄/ml.

Let


Ratio x mg of H₂SO₄/ml x 1.5 = mg of H₂SO₄ in unknown sample.
Mg of H₂SO₄ in unknown sample divided by 15 = mg H₂SO₄/ml of unknown sample.
Mg H₂SO₄/ml x 0.1335 = oz H₂SO₄/gal.
oz H₂SO₄/gal x .98 = oz SO⁼₄/gal.



= oz SO⁼₄/gal or Ratio x mg H₂SO₄/ml in control x 0.013083 = oz SO⁼₄/gal

TEST RESULTS
Two series of tests were planned to check the accuracy and reproducibility of the new method. The first ,series of tests involved the use of known samples by which it was hoped to prove the accuracy of the method. The second series of tests was conducted using actual tank samples to prove the reproducibility of the method.

For the first series of tests two H₂SO₄ solutions of like normality were prepared. To one of these solutions sufficient H₂SO³⁵O₄ was added to produce an , activity level of .05 microcurie per milliliter. Samples were prepared containing varying amounts of the two solutions. To these solutions there was added 15 ml of the 25 percent CrO3 solution to simulate the actual conditions of tank samples. The final approved procedure was then followed for preparation of counting samples and counting rates were determined. Based on these counting rates, the dilution ratios were calculated and compared with the true ratios based on the quantities of H₂SO₄ and radioactive H₂SO₄ used. Control samples were counted for one minute and all others to a total of 6400 counts. All counting rates were corrected for background and coincidence loss. The results of two identical tests are presented in Tables I and II.

It will be noted the average per cent errors of the two tests were 1.1 and 3.3 respectively, giving an overall average of 2.2 per cent. This degree of error is considered above average for work involving radioactivity which is influenced by the randomness of atomic disintegration and counter efficiency among other factors. It is believed that this margin of error is a big improvement over that of the gravimetric procedure which may show reproducibility but the accuracy of which is questionable.

The second series of tests consisted of running actual tank samples from the chromium plating department at this Arsenal. Samples from seven different tanks were analyzed. The final approved test procedure was followed in its entirety. Duplicate samples were run on each of two days and are designated Tests L-18 and L-18a. Each sample was counted twice on each of two scalers. The average values of the eight counting determinations run on each day are presented in Table III.

The sulfuric acid containing radioactivity used for these tests was .1014 N. This gave a constant value for K = 0.065004 where K = mg H₂SO₄/ml in control x 0.013083.

Referring to Table III, it will be noted that there was exceptionally good agreement between the two tests in all cases except tanks 4 and S. This degree of reproducibility combined with the degree of accuracy shown in the first series of tests indicates the reliability of this method.

Table IV presents the comparative data between the isotope dilution method and the gravimetric method. The per cent deviation from the mean of the two tests run by the isotope dilution method are also presented in column 3. It will be noted that the average deviation is only 1.9 per cent, which is considered to be very good for the determination of sulfate ion concentration. The difference between values obtained by the two methods expressed as percentages of the values determined gravimetrically are presented in column 5. It will be noted that the gravimetric value was consistently higher and the differences ranged from 4 to 12 per cent with the majority between 10 and 12 per cent.

Calculations have also been made to show that when counting rates taken on only one scaler are used, that the final results will be comparable to those calculated from the counting rates taken on two scalers. This will facilitate the counting procedure so that automatic counting equipment may be used to advantage.

SUMMARY
Since the accuracy of the gravimetric method-has long been questioned and the indicated better accuracy for the isotope dilution method is detailed above, there is a distinct advantage to be gained by the use of the latter method. Therefore, considering the precision and accuracy of the method, the use of the isotope dilution technique is strongly recommended. Other considerations such as quicker results, no tedious weighing operations, and automatic counting further favor the use of the new procedure.

The use of this method is, of course, limited to laboratories having the necessary facilities and equipment for radiochemical work. However, the number of industrial and government laboratories having these facilities is increasing each year. It is believed that a larger number of laboratories will consider installing the necessary equipment, since major remodeling of chemical laboratories is not required to do work such as has been discussed in this article. Shortage of trained personnel, another major deterrent to the expansion of radiohemical work is being alleviated, somewhat, at the present time.

The author wishes to express his appreciation to his co-workers at the Rock Island Arsenal Laboratory for their assistance and to the Ordnance Corps, Research and Development Division of the Department of the Army and supervisory staff of the Laboratory for permission to publish the information in this paper.

LITERATURE CITED
1. S. L. Eisler, “Determination of Sulfate in Chromium Baths IJsing Radiabarium”, Metal Finishing 50, 71-74 (January, 1952).
2. Z. Karaoglanov, Z. physik. Chem. A178, 143-156 (1937.
3. Z. Karaoglanov, Z. anal. Chem. 106, 129-146 (1936).
4. G. Friedlander and J. W. Kennedy, “Introduction to Radiochemistry”, John Wiley & Sons, New York (1949).
5. G. K. Schweitzer and I. B. Whitney, “Radioactive Trace Techniques”, D. Van Nostrand, New York (1949).
6. A. H. W. Aten, Jr., Nucleonics 6, No. 1, 68 (Jan. 1950).
7. G. K. Schweitzer and B. R. Stein, Nucleonics 7, No. 3, 65 (Sept. 1950).

---