The electronics and telecommunications industries require high reliability in electrical connections. Thus, gold and gold alloys have long served as the standard for the contact or surface material in applications such as separable connectors, relays and lead frame metallization. Their low resistivity for electrical conduction, high chemical passivity for resistance to film formation, high bondability for ease of thermocompression and other forms of joining and their Base of application by standard manufacturing processes such as electroplating, sputtering, welding or cladding are ideally suited to the requirements of these devices.Historically, gold was used in relatively thick layers deposited over the entire contact region, hence there was little concern over the substrate metal and its interaction with the gold or the environment? The surface region would remain as pure gold with its favourable properties through the device lifetime. However, the significant increases in the cost of gold and the concurrent economie pressures of this decade have stimulated appreciable efforts toward replacement of or reduction in gold usage per device. The more conservative use of gold has been the approach usually pursued. In this, a reduction from classically specified thicknesses had to be accomplished without any sacrifice in reliability. Thus, careful consideration had to be given to the behaviour of gold in thin film systems involving more than one metal.Several potential mechanisms which can degrade the performance of a thin gold over base metal system are possible when the gold is thinned significantly. Thin gold films are more likely to be porous (1) and invite corrosion of the substrate metal despite the nobility of gold. They are subject to more rapid removal by wear (2), which exposes the base metal substrate with its inherent unreliable performance. Thin gold films are also subject to penetration by the substrate materials by diffusion (3), which can place oxidizable base metals at the contacting surface and degrade performance due to increased electrical resistance. All of these potential hazards involving the interaction of gold and base metals in thin film systems have been the object of considerable attention over the last five to ten year period. This article will concentrate on a review of the diffusion behaviour of gold/substrate systems, in particular those where nickel and/or copper are the substrate materials.
Electrodeposited hard gold with 0.6 at.% cobalt has a hardness about four times that of annealed bulk gold and this high hardness cannot be reproduced by standard metallurgical methods. By measuring the hardness for gold and gold alloys subjected to various quenching, annealing, and deformation processes, all common hardening mechanisms such as solution hardening, precipitation hardening, strain hardening, and ’’voids’’ hardening were eliminated as possible major hardening contributors with the exception of the grain size effect. Pure gold with grain sizes ranging from 200 Å to 3 μm were prepared using sputtering deposition by varying the substrate temperature during deposition from 55 to 310 °C. Larger grain sizes from 5 to 200 μm were prepared by annealing cold-drawn gold wire at 300–750 °C. The hardness versus grain diameter followed the Hall-Petch relation up to a grain size of 0.1 μm. Beyond that, the hardness increased less rapidly. At the grain size of electrodeposited hard gold of 250–300 Å, the sputtered pure gold gave the same hardness value also. Therefore, the grain size effect accounts for the observed high hardness of electrodeposited hard gold, with other mechanisms accounting for only small alterations.
Low-temperature (75–150 °C) diffusion of copper through gold is studied with Auger electron spectroscopy. Grain-boundary or pipe diffusion measurements and calculations are discussed with boundary-value-problem assumptions. Electroplated cobalt-hardened gold on electroplated copper printed-wiring-board fingers is used. In addition, contact-resistance data is presented for different heating times and temperatures and a plot of CR values versus amount of surface film is shown. Chlorine, which evolves from the board material upon heating, reacts with copper which has diffused to the surface. This reaction provides a much better sink for the copper than oxidation. Diffusion coefficients are estimated which are several orders of magnitude larger than those previously reported for the Au-Cu system and reasons are discussed. Because of the chlorine emanating from the board, data from accelerated tests, in which the printed wiring board is also heated to elevated temperatures, may give erroneous predictions when extrapolated using an Arrhenius plot. Results demonstrate that diffusion via the defect structure is not the rate-controlling step in the accumulation of copper and/or copper compounds on the gold surface when exposed to normal atmospheres. The surface accumulation is limited by the growth kinetics of the surface film.
The low-temperature diffusion of nickel through gold has been evaluated using Auger electron spectroscopy. Samples used were printed-wiring-board fingers having a nickel ’’diffusion barrier’’ plated between the copper substrate and the cobalt-hardened gold contact surface. A coefficient was determined for diffusion via the gold defect structure at 150°C using both first-arrival and surface-accumulation approaches. Both methods yield comparable results with a coefficient D′ of (2–7) ×10−15 cm2/sec. This is between one and two orders of magnitude smaller than that determined for copper under nearly identical conditions. Results of other investigations are compared and contrasted to the present findings. The surface reaction of nickel with air and chlorine-containing environments is sufficiently rapid that diffusion is the rate-limiting step in the accumulation of nickel at the gold surface.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.