US 3821692 A
Abstract available in
Claims available in
Description (OCR text may contain errors)
United States Patent [191 Barnard [111 3,821,692 June 28, 1974 1 SLOTTED ELECTRICAL CONNECTOR OF COPPER-BASED ALLOY SEPARATED FROM AN INDIUM COATING BY A BARRIER LAYER  Inventor: Ralston White Barnard, Dunwoody,
 Assignee: Bell Telephone Laboratories Incorporated, Murray Hill, NJ.
 References Cited UNITED STATES PATENTS 1,904,241 4/1933 Kammercr 339/278 C 2,417,967 3/1947 B001 r 339/278 C 3,175,181 3/1965 Grabbe l .1 339/278 C 3,511,921 5/1970 Pasternak 339/98 Primary Examiner-Richard E. Moore Attorney, Agent, or Firm-C. E. Graves; G. S. lndig  ABSTRACT Means are disclosed to' maintain the long-life resistance characteristics of contacts used in solderless connectors and fabricated from indium plated, copper based alloy. A barrier layer is plated between the alloy of the contact and the indium layer beneficially used with aluminum conductor wire.
4 Claims, 3 Drawing Figures PMENIEDJUNZG m4 (PRIOR ART) X, X\ i BACKGROUND OF THE INVENTION This invention relates to connectors for establishing permanent electrical connections to insulated wires, and particularly to such connectors as are intended for use with wires having aluminum conductors.
Connectors for interconnecting a plurality of insulated wires, without the necessity of stripping the .insulation from the wires prior to insertion-into the connector, are widely used and are known as solderless connectors." Typical of such connectors is that disclosed by E. .I. Levin et al. in U.S. Pat. No. 3,0l2,2l9, dated Dec. 5, l96l. These connectors include a metal contact, such as phosphor bronze or brass, formed in a U-shape with slots in the open legs of the U and extending inward from the free edge. When the wires are inserted into a base and mated with a cover, the insulated wires are forced into the slots of the contact. Since the slots are narrower than the diameter of the conductors of the wires, the insulation is pierced or ripped back from the conductor, exposing a portion thereof. Contact is then made between the exposed conductors and the contact.
Although these connectors worked extremely well with copper'conductor wire, problems were encountered in making connections to aluminum conductor wire. The formation of a coating of aluminum oxide on the conductor surface acted as an electrical insulator. Since the oxide was not electrically conducting, it was necessary to insure that the oxide layer was removed during the connection process. However, even when sound connections were established. it was found that the contact resistance would rise with age. This was due, in substantial measure, to the appearance of oxide at the conductor/contact interface and the reduction of contact pressure due to creep in the aluminum.
To prevent this time-related deterioration in the connection, J. P. Pasternak proposed, in U.S. Pat. No. 3.5l 1,921 dated May 12, 1970, that the contact member be plated with indium. Indium is nonoxidizing, solid but readily flowable and conductive. It is self-annealing and does not work-harden under compression. As a result, the indium was able to penetrate small cracks in the oxide layer and make contact with the underlying aluminum conductor. At the same time, a gas tight seal was formed preventing further oxidation at the interface. Although indium plated contacts make excellent connections to aluminum conductors, indium is relatively expensive and it has been found that such contacts have a surprisingly short shelf life.
The short shelf life has been found to result from the formation of a hard, brittle alloy of the copper in the contact and the indium at the contact/plating interface. The interdiffusion and alloying of the copper of the contact with the inditfm has been shown to be quite fast, even at room temperature. By way of illustration, it has been found that the time to alloy 0.2 mils of indium is only years at room temperature. If compliant or elemental indium is not present when the connection is first made, metal-to-metal contact may not be established and the connection may not be reliable. Since elemental indium must be present when the connection is first made to ensure good electrical contact, it can be seen that the effective shelf life is probably somewhat less than the 5-year period.
It is therefore an object of my invention to significantly prolong the shelf life of indium plated contacts of copper-based material used in solderless connectors.
It is a further object of .my invention to prolong the shelf life of such contacts by preventing interdiffusion and alloying between the indium and the copper of the contact.
It is yet another object of my invention to reduce the thickness of the indium plating on such contacts.
SUMMARY OF THE INVENTION The electrical resistance instability and degradation of an indium plated contact of copper-base material, resulting from formation of a hard, brittle alloy of copper-indium at the contact/plating interface, is effectively prevented by placing a barrier layer of metal between the contact and the indium layer. Although the mechanisms differ, aluminum, iron, lead and nickel have been found to be effective for the metal of the blocking layer.
In a specific embodiment of the invention, a phosphor-bronze contact is plated with a 0.1 mil thickness of nickel followed by 0.2 mil thickness of indium.
DESCRIPTION OF THE DRAWING FIG. 1 is a perspective view of a portion of a connector contact embodying the barrier layer of my invention;
FIG. 2 is a cross-section view of the conductor/contact interface for a prior art connector; and
FIG. 3 is a cross-section view of the conductor/contact interface for the contact shown in FIG. 1.
DESCRIPTION OF THE ILLUSTRATIVE EMBODIMENT An embodiment of my invention is shown in FIG. 1 where a contact 14 is formed in a U-shape with wire receiving slots 13 extending inward from the open end of the U. Contact is established between contact 14 and conductor 12 of wire 11, which has been forced into slot 13. Although aluminum, iron lead and nickel have all been found suitable for the blocking layer, as will be discussed later, the combination of economy and ease of manufacture make nickel the: preferred material with todays technology. Accordingly, the contact 14 shown in FIG. 1 has a 0.1 mil plating of nickel over a formed phosphor bronze part followed by 0.2 mils of plated indium.
Although the plating thicknesses specified herein are I applicable over substantially all of the surfaces of contact 14, the specific region referred to is the sidewall of slot 13. Since wire 11 is forced into slot 13 to pierce the insulation and make contact with conductor 12, the region of particular significance for connection purposes is the area of contact between the sidewall and the conductor.
A prior art contact is shown in FIG. 2. The underlying contact 20 is a formed phosphor bronze piece. Contact 20 is then plated with a 0.4 mil thickness of indium. As can be seen in FIG. 2, the region of elemental indium 22 has been severely diminished by the formation of alloy 21 by diffusion. The remaining region of elemental indium 22 is only marginally usable to establish a stable connection to aluminum conductor 12. In
fact, in spots the alloy 21 has completely penetrated the indium layer 21 to contact conductor 12.
Using my invention, the improved contact structure 14 of FIG. 3 is obtained. Contact 14 comprises an underlying phosphor bronze piece 25 plated with 0.1 mil thickness of nickel 26 followed by 0.2 mil thickness plating of indium 27. The indium layer 27 makes contact with conductor 12. The reduced thickness of the indium is made possible due to the barrier layer 26 which prevents alloying in indium layer 27. As a result, substantially the entire plated thickness of layer 27 is useful, without the necessity of overplating to accommodate deterioration due to aging and alloy formation.
The indicated thicknesses of layers 26 and 27 are nominal only. The nickel layer 26 could be as thin as approximately 0.07 mil before the porosity of the layer would become prohibitive. Conversely, a thickness exceeding about 0.12 mils would unduly stiffen the contact and deleteriously affect the connectors performance by excessively notching the aluminum conductors as they are forced into the slots in the contact. Although the thickness of layer 27 should be at least 0.2 mil to be effective, there is no real upper limit to the thickness. Since the excess indium is essentially wiped of the contact by the inserted wire, thicker layers of indium merely increase the excess.
As was indicated earlier, nickel is not the only suitable material for the blocking layer. In searching for suitable materials for this-layer, certain criteria were first established.
To block the formation of the undesired alloy, a blocking layer of metal is used to prevent an indium/- copper interface from being established. The perfect metal for such a blocking layer would interdiffuse very slowly (if at all) with copper and indium, and would form no intermediate phases with either.
To prevent the formation of an alloy between the metal of the blocking layer and the indium, the two should be either completely soluble or completely insoluble in the solid state. Phase diagrams of indium containing systems reveal that indium is completely soluble in very few metals, and in general forms many intermetallic phases due to partial solid solubility. The metals in which indium is soluble are generally those in which extremely rapid interstitial diffusion of copper occurs. Thus, little is gained by these metals in terms of blocking capabilities.
Metals which form intermetallic phases with indium are unsuitable if the rate of interdiffusion is so fast that the barrier layer/indium intermetallic phase forms at the same rate as the copper/indium intermetallic phase. A metal which is insoluble in indium would be satisfactory, but many are difficult to plate. For example, plating the phosphor bronze contact with a blocking layer of aluminum prior to the plating of the indium would be effective. However, the plating of aluminum on bronze is both expensive and dangerous since it must be done under airless conditions in an ether-based solutron.
Although iron and indium are almost mutually insoluble in each other and form no intermetallic phases, the plating of iron is also quite difficult. The plating bath must be maintained at a high temperature and impurities must be critically controlled. Even once a satisfactory iron plating is obtained, it must be heat treated to relieve the brittleness of the iron.
The use of tin was found to be unsuccessful due to the formation of intermetallic phases of both coppertin and and tin-indium. Although the formation of the intermetallic phases would not of itself be a serious problem, it was found that the diffusion was quite rapid at both interfaces, allowing the intermetallic phases to form quickly.
Cadmium was also found to be unsuccessful as a blocking layer. Originally it was thought that the slower rate of copper-cadmium interdiffusion and the lack of cadmium-indium intermetallic phases would produce an effective blocking layer. However, extensive alloys of copper-cadmium-indium were found to form. Zinc also proved to be unsuitable. Although no alloy forms at the bronze-zinc interface, considerable interdiffusion occurs. Zinc and indium also interdiffuse to some extent at their interface. Because of this an intermetallic alloy could form.
Since copper diffuses rapidly through most metals around indium on the periodic table, an attempt was made to find a metal suitable for the blocking layer which would not alloy with copper and which would separate the copper and indium to delay their eventual alloying. Lead was the best choice from this group. Although lead forms an intermediate phase with indium, it is not a hard intermetallic. In addition, lead and copper are almost insoluble in each other in the solid state.
Copper diffuses through lead by an interstitial mechanism and although rapid, is not as fast as through indium. The interdiffusion of lead and indium causes the two metals to rapidly lose their identities. As a result, an alloy does form at the copper/barrier layer interface, but at a much slower rate than at the copper-indium interface. Interestingly enough, although an intermediate phase of lead-indium forms, it retains many properties of the constituent metals. It is quite safe and has a low tensile strength. Electrical resistance tests reveal an excellent resistance aging characteristic.
Predictably, a very thin layer of nickel will prevent the diffusion of copper. Although nickel and indium interdiffuse rapidly at elevated temperatures near the melting point of indium, it was found, quite surprisingly, that nickel and indium interdiffuse very slowly at temperatures below that melting point. For example, it
would take at least 40 years for 0.2 mils of indium to convert to nickel-indium alloy at a temperature of 1 10C. At room temperature, it would take much longer.
Even beyond obviating the alloying problem, the composite plating represents improvements over the heavier indium plating. For example, nickel is relatively cheap and easy to plate. Since indium plates readily over fresh nickel, the thinner plating of indium is easier to apply and handle than the thicker plating of indium over phosphor bronze as taught by the prior art. Cost comparisons indicate that the described composite plating will be no more expensive than the prior art plating of FIG. 2. I
Although nickel has been indicated as the preferred material, it should be apparent that this assumes particular available manufacturing facilities and use environment. For example, where use of contact 14 to make a number of reterminations is planned, the utility of lead as the barrier should be seriously considered. The lead-indium intermediate phase is harder than pure indium. This may be a particular advantage in this use environment by preventing the removed wires from carrying away too much plating from the contact.
It is to be understood that the embodiments described herein are merely illustrative of the principles of my invention. Various modifications may be effected by those skilled in the art without departing from the spirit and scope of my invention.
What is claimed is: t
l. A connector for making electrical connection to at least one wire comprising means for positioning said wire to be connected;
a contact of copper-based material having at least one receiving slot for wire to be connected;
a layer of indium at the sidewall of each slot;
means for forcing said positioned wire into a said slot to establish both mechanical and electrical contact between the conductor of the wire and the indium layer; and
means for separating the indium layer from the Contact to prevent interdiffusion and alloying therebetween and for establishing electrical continuity between the contact and the indium wherein the separating means comprises a layer of metal and wherein the metal of the layer is selected from the group consisting of aluminum, iron, lead and nickel.
2. A connector in accordance with claim I wherein the layer of metal comprises nickel between 0.00007 and 0.00012 inches in thickness.
3. A connector for making electrical connection to at least one bare or insulated wire, the connector being of the type including means for positioning wire to be connected, an indium plated contact of copper-based material having at least one wire receiving slot for wire to be connected, and means for mating with the positioning means to force said wire into said slot so that electrical connection is established to the conductors of the wire, wherein the improvement comprises a barrier layer of metal positioned between the contact and the indium plating, thereby maintaining electrical continuity between the contact and the indium while preventing interdiffusion and alloying between the copper of the contact and the indium wherein the layer of metal is selected from the group consisting of aluminum, iron, lead and nickel.
4. A connector in accordance with claim 3 wherein and 0.00012 inches in thickness.