Historical Articles

December, 1952 issue of Plating

 

Notes on the Electrodeposition of Thick Gold Deposits

CHARLES L. BAUER
Chemist, National Bureau of Standards

---

I. INTRODUCTION
In connection with a special project, it was necessary to produce in a continuous electroforming operation smooth gold deposits up to a thickness of approximately 0.01 inch (0.25 mm). Efforts to produce this type of deposit by means of the conventional cyanide gold plating baths were unsuccessful. It was, therefore, necessary to investigate different types of baths and operating conditions. Because time did not permit a systematic study of all the variables that may affect a gold deposit, this brief report is devoted principally to the definition of conditions found to be favorable to the aforesaid objective. A more exhaustive study would be required to define and explain fully the conditions that affect the production of thick gold deposits.

A survey of the literature revealed very little published work on the deposition of thick gold deposits. For ornamental purposes very thin coatings are used, ranging from a few millionths to a few hundred-thousandths of an inch (of the order of 0.1 - 1 µ). In special engineering applications thick deposits are used, e. g., during the war as much as 0.001 inch (0.025 mm) of gold was used on radar equipment. There has been little, if any, demand for much thicker coatings or for electroformed gold objects, and consequently, suitable baths have not been developed. An exception is the gold chloride-hydrochloric acid bath, used principally in gold refining, from which thick, smooth deposits are produced. However, for the electroforming application involved in the present study, the very corrosive action of the chloroauric acid on the stainless steel mandrel and the poor throwing power of the bath eliminated the use of such a bath. It was decided, therefore, that the cyanide bath was the most practical type to be studied.

II. COMPOSITION OF PLATING BATH AND CONDITIONS OF PLATING
On the basis of exploratory studies, a high-concentration gold cyanide bath and conditions of operation were developed, from which deposits 0.000.009 inch (0.15-0.23 mm) thick were electroformed in a continuous operation. The composition of the bath and the operating conditions were:

Pure gold anodes were used. Vigorous agitation obtained from a combination of mandrel rotation and air agitation, was beneficial to the deposit cathode efficiency of nearly 100 per cent was obtained. The conditions and composition were carefully controlled during the plating operation.

At its peak of performance the bath gave smooth deposits up to 0.015 inch (0.38 mn) thick, which did not have the coarsely crystalline appearance that was characteristic of thick gold deposits obtained from many other combinations of composition and operating conditions tried. The life of the bath was limited, for with continued operation the bath produced coarsely crystalline deposits, and “spot-plating” occurred. (The term “spot-plating” is applied to the deposition of the metal only at random places on the cathode surface.) The main reason for the failure of the bath to continue to produce the desired deposits was the increase in the carbonate content of the bath, caused by the decomposition of the free cyanide. Efforts to control the cyanide decomposition were not very successful. One method that was promising was the use of lithium cyanide as a source of cyanide, inasmuch as lithium carbonate precipitates as it forms. The present high cost of lithium salts is a disadvantage of this method.

The effects of changes in concentration of the bath constituents, especially the concentration of gold and cyanide, were studied. In addition, the rate of decomposition of cyanide and the resulting increase in the carbonate content were investigated.

(a) Gold Concentration
The various concentrations of gold that were tried gave reproducible results. With very low concentrations, gassing occurred at the cathode except at very low current densities. Concentrated gold baths were selected for reason of the following advantages: (1) a longer life, because the gold content is depleted during operation under the conditions given, and (2) a higher deposition rate, because the limiting current density may be increased by increasing the gold concentration. The gold was introduced as commercial sodium or potassium gold cyanide. Commercial potassium gold cyanide, containing 67.5 per cent of gold, was generally used, but sodium gold cyanide, which contains 46 per cent of gold, is just as satisfactory. The limiting gold concentration was determined by the solubility of the potassium gold cyanide complex, KAu(CN)₂. At the concentration adopted, the gold cyanide complex precipitated when a bath that had been in operation for some time was cooled. This precipitation was probably caused by the formation and accumulation of potassium carbonate in the bath. However, the limited solubility of the gold cyanide complex did not shorten the life of the bath, for the increasing carbonate content made the bath inoperable before the cyanide complex precipitated at the high operating temperature.

(b) Cyanide Concentration
The effect of changing the free cyanide concentration was investigated by varying the ratio of cyanide to gold. The results are expressed as the ratio between the molar concentrations of free cyanide and of gold at various gold concentrations. The potassium cyanide concentration was increased from 1.3 to 391 g/l (0.17 to 52.4 oz/gal), and the mole ratios of potassium cyanide to gold ranged from 2:1 to 60:1. Runs of six hours at a current density of 5.4 amp/ft² (0.55 amp/ dm²) gave deposits about 0.005 inch (0.13 mm) thick. With low ratios of potassium cyanide to gold, the deposits were coarsely crystalline and treed at the edges, and had a duller appearance than those produced from baths with higher ratios. As the ratio was increased, the deposits became smoother and lost their crystalline appearance. At still higher ratios the deposits were again very coarsely crystalline, and in many cases spot-plating occurred. The best results were obtained with a mole ratio of about 20:1, regardless of the gold concentration used. For any bath the deposits improved as the bath was used, until a peak of performance was reached. As the bath was used further, the crystallinity of the deposits increased and spot-plating occurred. The bath with a ratio of 20:1 and with a concentration of 18 g/l (2.2 troy oz/gal) of gold gave superior deposits over the longest period of tine. Under its best operating conditions this bath gave a deposit with a matte finish and no marked crystallinity.
Higher current densities, up to 18 amp/ft² (2 amp/ dm²), produced smooth, fine-grained deposits with the 20:1 ratio of potassium cyanide to gold, but whether or not this ratio is optimum at the higher current densities was not determined.

(c) Cyanide Decomposition and Carbonate Control
Free cyanide decomposes in hot solutions, particularly at high current densities, and frequent additions of potassium cyanide have to be made to maintain a constant concentration. As the cyanide is decomposed the carbonate content of the bath increases. The decomposition reactions may form: hydrogen cyanide, ammonia, urea, carbonate, formate, and cyanate1. Hydroxide, formate, and cyanate ions are intermediate in the formation of carbonate, which is the principal decomposition product.
Because carbonate is the principal decomposition product, a study was made of the effect of increasing the carbonate content in a gold cyanide bath. Potassium carbonate was tried in initial concentrations from 0 to 70 g/l (0 to 9.4 oz/gal) in new baths. A mole ratio of about 2:1 with respect to gold, that is, a concentration of 25 g/l of potassium carbonate in the bath containing 18 g/l (2.2 troy oz/gal) of gold gave the best deposits. With no or very little carbonate, the deposits had a dull appearance. With high concentrations of carbonate, the deposits were very crystalline, similar to those obtained with high concentrations of cyanide. Since carbonate rapidly accumulates in a bath, no initial addition was made in the selected bath.

Methods for controlling the carbonate content of a cyanide bath were studied. The most promising method which could be used on a continuously operating bath and which would introduce no harmful ions was the use of lithium cyanide as a source of cyanide. Lithium cyanide is extremely soluble, whereas lithium carbonate has a low solubility. As the cyanide is decomposed the carbonate formed is precipitated as lithium carbonate. Continuous filtration would remove the precipitated carbonate. The solubility of lithium carbonate in water decreases with the temperature as follows:

The solubilities of lithium carbonate in the plating bath would probably be of the same order of magnitude. As stated previously, the desired mole ratio of carbonate to gold is about 2:1. Therefore, in the selected bath, the concentration of carbonate, as lithium carbonate, should be about 13 g/1 (1.8 oz/gal). Several determinations of carbonate in operating baths, saturated with respect to lithium carbonate, yielded concentrations of lithium carbonate in the range of 15 to 18 g/l. Gold deposits from these baths were satisfactory except for roughness, which was probably caused by suspended matter.

(d) Addition Agents
Of about thirty addition agents tried, only two showed any improvement in the deposits. These two were vanillin and potassium sulfite. Vanillin in a concentration of 0.3 g/l (0.04 oz/gal) was beneficial. Deposits about 0.005 inch (0.13 mm) thick were much finer grained from a bath to which vanillin had been added than from one in which there was no vanillin. The simultaneous use of both vanillin and potassium sulfite appeared to be advantageous.

III. ANALYSIS OF GOLD SOLUTIONS
In the determination of gold it was found that evaporation of the bath sample with either hydrochloric or nitric acid, as recommended in the literature, did not completely destroy the gold cyanide complex. A yellow precipitate, probably aurous cyanide, was never completely destroyed, even with subsequent evaporation. It was necessary to fume the sample with concentrated sulfuric acid, whereby the cyanide was destroyed and the gold was precipitated as metal. If no impurity such as silica or iron was present, the gold was precipitated quantitatively and in a compact form, which could be filtered out, washed, ignited, and weighed.

Since it was usually necessary to remove traces of silica and iron, the longer procedure outlined below was used. Details for the final precipitation of the gold are given by Gilchrist³.

1. A sample from the bath was evaporated with sulfuric acid to the appearance of dense white fumes.
2. The residue was treated with aqua regia, diluted, and filtered to remove silica.
3. The filtrate was evaporated almost to dryness, the residue was dissolved in water, and the pH was adjusted to 1.5 by adding sodium hydroxide, using thymol blue as an indicator.
4. The solution from 3 was heated and potassium nitrite was added to precipitate the gold.
5. The gold was filtered, washed with dilute hydrochloric acid to remove iron, washed with water, ignited, and weighed.
Standard methods were used to determine cyanide and carbonate.

ACKNOWLEDGMENT
Acknowledgment is made to Vernon A. Lamb for advice and assistance in this investigation.

3. Raleigh Gilchrist, “New Procedure for the Analysis of Dental Gold Alloys” J. Research, National Bureau of Standards, 20, 745 (1938).