An electron microscopic investigation has revealed that the electroless deposition of copper occurs by repeated three‐dimensional nucleation at catalytic sites on a substrate. The nature of the catalytic sites formed by immersion in the standard stannous and palladium chloride solutions was observed and found to influence the structure of the subsequently deposited metal. Initially, copper nuclei about 25Aå in diameter form aggregates an order of magnitude larger. As the autocatalytic reduction reaction continues the aggregates increase in size until they become energetically unstable; then recrystallization occurs. The resulting continuous copper films exhibit a heterogeneous microstructure with grain size variations of more than an order of magnitude and also numerous twin faults.

A new technique has been developed for producing laminated composites of cyclic multilayered alloy (CMA) electrodeposits. The thickness and composition of the individual layers of the CMA deposits are altered precisely and conveniently by cyclic modulation of the cathodic current or potential during electrodeposition. It is thus possible to modify the structure to obtain laminated composite coatings which may have desirable engineering properties. Wear and corrosion resistances, mechanical hardness and strength, as well as certain magnetic, optic, and electronic properties of the plated composite alloys should be, in principle, straightforward to design and fabricate. The CMA technique was demonstrated by plating a wide variety ofAg-Pd CMA structures. Fine structures, with repeat distances as small as 0.05 ~m, were obtained. Both square and triangular waveforms were applied to produce sharp or gradual composition variations, respectively. The resulting deposits * Electrochemical Society Active Member. 1 Present address: MPI, Santa Clara, California 950,51. -~Present address: Oxy Metal Industries, Warren, Michigan 48089.

The electroless gold plating process using potassium borohydride as the reducing agent has been investigated for impurity effects, material compatibility, bath agitation effects, and thickness uniformity and line resolution in selective plating of patterned substrates. Impurities may cause a decrease in plating rate [Ni(II) ], bath instability [Ni(II), Co(II), Fe(II) ], thickness nonuniformity (polyethylene, organics in deionized water), and nodule formation (some surfactants). Bath agitation is beneficial: it increases plating rate, minimizes porosity of thin deposits, and eliminates nodule formation. Edge build-up generally occurs in selective pattern plating but, with proper selection of bath compositions and agitation conditions, it can be maintained below 10% in the thickness range of 1-12 ~m. The rate of lateral growth of electroless gold deposits is about 60% of that of perpendicular growth under optimum plating conditions. Also considered in this paper are certain aspects of the scale-up and waste disposal problems associated with electroless gold plating.Electroless gold plating has been found to be useful in a variety of applications, especially for selective plating on patterned substrates for electronics applications. Such applications generally require pure soft gold with a thickness in the range of 1-15 ~m. A bath developed in this laboratory (1) has been found to be quite suitable for forming such deposits. Previous papers described the general bath characteristics (1), physical properties of deposits (2), bath operation with replenishment (3), reaction mechanism (4), and the nucleation and growth of deposits (5). The purpose of this paper is to describe several other aspects of the process which are important from the practical viewpoint and which have hitherto not been discussed. The topics covered include impurity effects, material compatibility, bath agitation effects, deposit thickness uniformity, and line resolution in selective plating of fine line patterned substrates. General recommendations are made as a guide for users of this process. Specific applications will be described in separate communications. Solution Preparation and Plating ProcedureCompositions of three electroless gold plating baths used are listed in Table I. Bath A was used often in our earlier studies (1-3) including those of impurity effects and porosity described in this paper. More recently, baths B and C have been used exclusively. These two baths contain less KCN and KBH4 and, therefore, are more preferable than bath A for practical reasons. Bath B gives the highest deposition rate (5-7 ~m/hr at 70~176 with vigorous agitation), but deposits with acceptable physical properties can be obtained only when plated with agitation. Bath C is slower plating (2 ~m/hr at 70~ with agitation) but gives better thickness uniformity on thin deposits in fine line plating. Details will be described in subsequent sections. It is convenient to prepare the baths by dilution of 5• concentrated stock solutions. These solutions can...

The electrochemical and structural aspects of gold deposition from dilute cyanide, citrate, and phosphate‐buffered plating solutions were studied using rotating disk electrodes. Current‐potential curves were recorded galvano‐statically for 0.005M Au solutions at 60°C, and the morphology of deposits approximately 1µ thick, which corresponded to different regions of the i–V curves, was determined by scanning electron microscopy. Transmission electron microscopy was used to elucidate the early stages of growth (∼0.1µ). The cyanide system, which has an exchange current density of 0.82 mA/cm2 and a transfer coefficient of 0.7, showed the greatest tendency to form an out‐ward growth morphology consisting of fine features termed spikes. These features are responsible for the characteristic brown appearance of deposits plated under certain conditions. Increasing current density and decreasing rotation speed, in general, both favor the formation of spikes which occurs over a given range of potential false(150 normalmV<|η|<300 normalmVfalse) and not at a particular concentration of gold at the interface. The spikes, which are typically 0.1– 1.0µ in size and >108/cm2 in density, were also observed for deposits from the citrate system and to a much lesser extent in the phosphate system. At low overpotentials, a smooth, layer type of growth is formed. The formation of the various morphological features is governed by the electrode potential below the limiting current density and by mass transport near or above limiting current.

The as-plated surfaces of cobalt electrodeposits obtained using various combinations of the common plating variables as well as some thinned foils of these samples were studied. The basal plane was found to be predominantly perpendicular to the substrate leading to <1120> and <1010> fiber axes. Structural features associated with each fiber axis are explained in terms of crystal symmetry and twinning. Using the concepts of free and inhibited outgrowth it was not possible to predict the structure resulting from plating under a given set of variables. The cubic phase of cobalt was observed after deposition at room temperature and low pH. ) unless CC License in place (see abstract). ecsdl.org/site/terms_use address. Redistribution subject to ECS terms of use (see 169.230.243.252 Downloaded on 2015-02-02 to IP Vol. 113, No. 5 DEPOSITION MODES OF COBALT ) unless CC License in place (see abstract). ecsdl.org/site/terms_use address. Redistribution subject to ECS terms of use (see 169.230.243.252 Downloaded on 2015-02-02 to IP