Experimental Details

The basic electrolyte used was as follows:

The standard plating procedure was 5 amp/sq. ft for thirty minutes at 40°C unless otherwise stated.

A plating jig which completely eliminates the use of stopping-off lacquers was designed and is shown in exploded view in Fig. 1. The beryllium-copper strip is clamped against a stainless steel backing plate, using P.T.F.E. as an insulator.

The strips were made from one batch of Mallory 73 beryllium-copper strip, 0.375 × 0.005 inch. The strip was stretch-straightened and then cropped into 3.5 inch lengths. At this stage the metal was in the cold rolled condition and it was then given the standard heat treatment in air at 350°C for two hours. This treatment was carried out in a clamp which held the strips flat.

The strips were measured for the zero curvature in a modified Soderburg and Graham apparatus shown in Fig. 2 and weighed before mounting in the plating jig.

Degreasing of the strip in cold trichorethylene which was ultrasonically agitated was followed by cathodic cleaning for ninety seconds. After rinsing the strips were transferred to a 1.5 per cent phosphoric acid solution and anodically etched for forty seconds at 7 amp/sq. ft. After a thorough rinse the strip was ready for plating.

After plating, the jig was rinsed in a drag-out tank and dismantled. The strip was dried and weighed. Three readings of the curvature were taken, the strip being moved along each time so that an average value of the chord height was obtained.

Internal stress was calculated by use of the equation:

S =internal stress

d =deposit thickness, calculated

from the increase in weight

EB-Young’s modulus for the

basis metal

b =width of the strip

t =thickness of basis strip

h = height of chord measured by Soderburg and Graham apparatus

I=moment of inertia about the neutral axis, given by

For all experiments the vaIues of E_B, b, t and R were regarded as constant.

In the case where a paint was being studied, a copper panel 2×1 inches was first coated with the paint and air dried for twenty-four hours. This panel was immersed in 200 ml of the rhodium solution at the temperature under consideration for thirty minutes, and then removed. The solution was thoroughly stirred and the stress measurement made.

Where a paint solvent or other water insoluble substance was being studied, 1 g of the substance was introduced into 200 ml of rhodium solution, at the temperature under consideration, and stirred for thirty minutes. The solution was filtered and the stress measurement made.

Results

Curve A of Fig. 3 shows the results obtained by varying the temperature of the standard rhodium solution. Curve B shows the stress against temperature when a commonly used stopping-off paint, Detel CRP, had been in contact with the solution. It will be noted that the rate of contamination rises sharply with temperature. Electrographic tests revealed that the deposits produced at 30°C from both solutions were cracked, while all other deposits were crack free.

The organic constituents of a paint of this type can be divided into three groups,

The resin used in this paint is a chlorinated rubber, and a rhodium solution treated with this substance gave a value of stress of 111,000 lb/sq. in. showing no change compared with the pure rhodium solution.

The plasticisers used in commercial paints vary considerably, but tests carried out on two typical examples showed that rhodium solution saturated with dibutyl phthallate and with Aroclor 1248 (a chlorinated diphenyl from Monsanto Chemicals Limited) gave values of 114,000 and 110,000 lb/sq. in. respectively, again very similar to those from the pure solution. For this reason no further tests were carried out on other plasticisers.

The solvents used are often commercial grades of solvent naphtha and stress measurements made on rhodium solutions treated with various samples of these solvents showed considerable variation. These were shown to agree closely with variations in the sulphur content of the solvent as seen in Table I.

A study of the various solvent samples by vapour phase chromotography showed that the sulphur compound present was essentially thiophen. The results of adding this compound directly to the rhodium solution are shown in Table II.

As a result of these findings a paint was formulated in co-operation with Detel Products Limited using solvents of low sulphur content. The stress/temperature results for this paint, designated Detel RHP, are shown as curve C in Fig. 3. It will be seen that the increase of stress compared to the paint-free solution is virtually constant with changes of temperature.

Discussion

The rhodium sulphate electrolyte will normally produce highly stressed deposits. It has been shown that small additions of organic sulphur compounds will markedly increase this stress level.

It appears that previous determination of internal stress in rhodium deposits using a method relying on paints for stopping-off one side of the cathode must be suspect, and only an apparatus completely avoiding the use of paint can be used with confidence.

The contamination from Detel CRP paint has been shown to vary with temperature, becoming considerably greater at the higher temperatures tried. The Detel RHP paint, containing sulphur-free solvents, did not exhibit this effect.

The introduction of this new paint to the plating shop has improved operations on two counts; first, the stress cracking originating at the paint/plating interface is almost eliminated and secondly, the frequency of treatment of the solution with active carbon in order to remove organic contaminants is much reduced. Whereas with standard paints the plating solution was being treated after every ten days’ operation, the interval has now been increased to as long as eight to ten weeks.

The author is indebted to Mr H. A. Slade, of Detel Products Limited, for advice and assistance in the provision of paint samples.