DE1924845B2 - Semiconductor arrangement with metallic contacts and / or metallic connecting layers - Google Patents

Description

30th

The invention relates to a semiconductor arrangement with metallic contacts and / or metallic connecting layers on the semiconductor surface and / or on an insulating layer over the semiconductor surface; in particular a non-hermetically encapsulated integrated semiconductor device.

When encapsulating semiconductor assemblies with plastics, their gas permeability often results in corrosion phenomena on the metallic contacts and connection layers, which can * o lead to failure of the semiconductor assembly, especially if it is an integrated semiconductor circuit with numerous contacts and connection layers. In this case, the extent of corrosion phenomena, in particular ⁴ S also depends on the metal used systems, which may include, for example, gold and aluminum or molybdenum and gold, wherein the system tungsten gold relatively corrosion resistant, but also is in need of improvement.

Detailed information on the problems of non-hermetic encapsulation can be found, for example, in of OE-PS 2 59 014, from which a semiconductor arrangement with a semiconductor body of a certain conductivity type is known, which has a flat main surface and a zone adjoining part of this main surface has opposite conductivity type, this zone forming a pn junction, the edge of which on the Main surface is and wherein a layer of semiconductor oxide is formed on the main surface, which the The edge of the pn junction is covered and has an opening which forms part of the surface of the zone opposite conduction type exposed and which is further designed so that in contact with this *> 5 exposed surface of the zone of opposite conductivity type within the opening of the layer Semiconductor oxide and covering the adjacent parts covering the edge of the pn junction the same a layer made of an active metal, for example titanium, tantalum, zirconium, niobium, chromium, Vanadium or hafnium, is arranged and that over this layer of active metal one or more There are layers of contact metal, for example platinum, silver, nickel, palladium, rhodium or gold, it is assumed that the interface between a layer of an active metal such as e.g. B. Titanium or tantalum and a dielectric oxide such as e.g. B. silicon dioxide a practically insurmountable Barrier to the penetration of harmful gases from the environment of the semiconductor device forms. The cited patent thus deals with the problem with non-hermetically encapsulated semiconductor arrangements pn junctions reaching to the surface of the substrate against the disruptive diffusion of Protect contaminants from the atmosphere.

The present invention, however, is based on the problem of corrosion phenomena on the metallic contacts and / or metallic connecting layers in the case of not completely hermetically encapsulated To avoid semiconductor arrangements.

This object is achieved in a semiconductor device of the type described in that the Contacts or connecting layers made from a mixture of molybdenum or tungsten and one Modifier metal or metalloid are made, which have a greater resistance to corrosion than molybdenum or tungsten, which is able to form passivating oxide layers and its oxide is more stable than the oxides of molybdenum and tungsten, respectively.

This embodiment of a semiconductor arrangement proves to be particularly advantageous where the Semiconductor arrangement works in an atmosphere with high humidity, because then it will also occur not hermetically sealed encapsulation does not show any signs of corrosion. In addition, the specified Material combination also good on silicon and silicon oxide. Another advantage is that such a barrier-forming metal for the improved electrical contacts and electrical Connection layers with gold do not enter into any intermetallic compounds.

The invention is explained below using preferred exemplary embodiments in conjunction with a drawing explained in more detail. It shows

F i g. 1 is a plan view of a semiconductor die in FIG which a planar transistor has been formed and which has an insulating layer on its surface, the has openings for attaching contacts,

F i g. 2 shows a section along line 2-2 in FIG. 1,

3 shows a schematic section through a device for high-frequency sputtering, as used for Attaching the contacts to the semiconductor die can be used,

F i g. FIG. 4 is a top plan view of the semiconductor die of FIG. 1 after attaching contacts and Connection points,

F i g. 5a and 5b show a section and a plan view of the semiconductor chip shown in FIG Attaching the supply lines and embedding them in plastic, the cutting direction of FIG. 5a through the line 5a-5a in FIG. 5b is indicated; however, the upper part of the plastic capsule has been cut off in Figure 5b, in order to be able to show the arrangement more clearly,

F i g. 6 shows a section through an integrated circuit

with connections only in a single plane,

7 is a plan view showing the arrangement of circuit elements in one of the functional units of a semiconductor wafer, which is shown in Pig.8 is and interconnections are required on several levels,

F i g. 8 is a plan view of this semiconductor die; which contains several functional units,

F i g. 9 is a circuit diagram of one of the functional units the F i g. 7 and

10 to 12 several sections to explain the production of the integrated shown in FIG Circuit according to line 10-10 F i g. 8th.

The invention provides contacts and interconnections for semiconductor devices which are corrosion resistant and which can be used to advantage on either unencapsulated or non-hermetically encapsulated semiconductor devices. A homogeneous mixture of predetermined amounts of molybdenum or tungsten and a modifier metal is used, which has a greater corrosion resistance than molybdenum or tungsten, in order to increase the corrosion resistance of the latter. The advantages that can be achieved by using molybdenum are described in US Pat. No. 3,341,753. It is worth noting that molybdenum and tungsten adhere extremely well to silicon and silicon oxide, that they do not form intermetallic compounds with gold, and that they act as a barrier between gold and the semiconductor material, which is important when gold is used as a clad metal where there is both one Protective layer as well as a conductor with low resistivity.

One of the requirements that must be made of a modifier metal is that it must be more corrosion-resistant than molybdenum or tungsten, with which it is to be used. This means that the modifier metal must only be slightly soluble in aqueous acids - with the exception of hydrofluoric acid; Furthermore, the oxides of the modifier must be more stable than the oxides of molybdenum or tungsten, and they must only be slightly soluble in a wide variety of acids. The modifier metal must also be able to form a passivating oxide layers, namely in the "" ^r most varied Oxydationspotentialen; Finally, the modifier must be highly resistant to corrosion. But it should also have a relatively high melting point and low self-diffusion.

However, none of the modifier metals are the The requirements outlined above are met, with gold metallurgically stable enough to use it as a barrier layer to be used with a metal-gold contact.

The ideal modifier metal is insoluble in molybdenum or tungsten and does not form any with it Links. However, any metal that meets the above requirements is either soluble in molybdenum or tungsten or forms a compound with it. The requirement that that Modifier metal must not form compounds with molybdenum or tungsten and must not dissolve in them should, with regard to the etchability of the mixture of molybdenum or tungsten and the Modifier metal set up, because the etchability changes with the percentage of an added Element in a mixture. Etching becomes more and more difficult, the greater the percent solubility, and compounds that form interconnections are the most difficult to etch when the desired circuit pattern is achieved should be formed. Examples of modifier metals that have all of the desired properties with the exception that they do not form a compound with or are soluble in molybdenum or tungsten, are the following elements: titanium, tantalum, chromium, zirconium, hafnium and silicon.

ίο The preferred modifier metal is titanium, which is soluble in molybdenum in any ratio, but does not form any compounds with it or with tungsten; in addition, tungsten is practically insoluble ^{below 700 ° C. in titanium.} Titanium, for its part, is only soluble up to about 4% by weight in tungsten at 600 ° C. The mutual diffusion in tungsten-titanium mixtures in which the titanium content exceeds 4% is limited by the concentration and is therefore quite low. The other metals mentioned above are also suitable as modifier metals, but they are less preferred in their order for the following reasons: Tantalum and chromium do not form compounds with molybdenum or tungsten, but they are in molybdenum and tungsten in any ratio soluble.

The higher oxides of chromium are water-soluble and consequently not as resistant to corrosion when one applies a high positive bias. Zircon forms compounds with molybdenum or tungsten, as well as this is the case with hafnium and silicon.

A homogeneously mixed layer of molybdenum or tungsten and a modifier metal, which is preferably titanium, can easily be applied by high-frequency or triode spraying using cathodes manufactured using the usual powder metallurgy technology. Titanium concentrations above about 4% in tungsten-titanium mixtures result in pseudo-alloys in which the excess of 4% Although the titanium content is not alloyed or dissolved in tungsten, the increased titanium content is still the desired one Causes corrosion resistance in tungsten.

The lower limit of the proportion of the modifier metal in the mixture with molybdenum or tungsten is determined by the fact that this proportion of modifier metal the corrosion resistance of the molybdenum or tungsten increases to a sufficient extent. Random Small amount impurities from the aforementioned modifier metals, such as them Inadvertently into which molybdenum or tungsten can get into, have no influence on the corrosion resistance of molybdenum or tungsten, from which the contacts are formed, for example. It must So a noticeable amount of modifier metal was intentionally added to the molybdenum or tungsten will. In practice, the lower limit for the proportion of the modifier metal is approximately 3% by weight. The passivation or corrosion resistance of the contact decreases with the proportion of the modifier metal in the mixture with molybdenum or tungsten. However, if the proportion of the modifier metal exceeds about 20% - this value shows excellent corrosion resistance of the contacts - so the modifier metal also brings disadvantageous properties of the mixture with it.

The upper limit for the amount of modifier detail in the mixture with molybdenum or tungsten is determined by two things: Firstly by that Proportion of modifier metal that the molybdenum or tungsten matrix can absorb without that

Modifier metal reacts with the gold layer and consequently increases its specific resistance, and secondly by the etchability of the layer, which increases with the increase in the proportion of modifier metal decreases. Furthermore, it becomes difficult to make the metallic contact of molybdenum or tungsten and modifier metal to be precisely defined when the modifier metal reacts with the adjacent gold layer. the Mixture of molybdenum or tungsten and modifier metal is extremely difficult to etch if the modifier metal content exceeds about 35% and the mixture becomes metallurgically unstable and leads to a reaction of the modifier metal with the gold, if the proportion of the modifier metal is above about 60%. The use of contacts that are more than approximately 35% or 60% Containing modifier metal is possible, but more careful work must be done and the preparation takes more time as it becomes more difficult to correctly locate the individual contacts and interconnects To make shape. A modifier metal content of approximately 20% is most expedient. The specified Percentages are averages and do not exactly apply to every modifier metal, because these differ slightly in their properties, so that the specified Move percentages.

For the sake of simplicity, only the use of molybdenum and titanium is described below. where molybdenum can always be replaced by tungsten, unless otherwise stated will.

The F i g. 1 and 2 show a semiconductor die 10 in FIG which a transistor has been formed which has a base 11 and an emitter 12 and wherein the remaining plate forms the collector 17. He was in the conventional planar technology created by successive diffusion of donors and acceptors and the intermediate formation of silicon oxide masks.

In the planar process, an oxide layer 13 is formed on the surface of the lamina 10, namely the layer thickness over the collector 17 must be thicker than over the base, so that the usual stepped design results. For high frequencies, the active parts of the transistor should be extremely narrow, so that the elongated emitter 12 is, for example, 0.05 mm wide and less than 0.25 mm long. The base 11 then has an area of approximately 10 ~ ² mm ² , for example. Two openings 14 and 15 are provided for the base contacts, while one opening 16 has been arranged for the emitter contact; the latter is the same that was also used for diffusing in the emitter. Because of the extremely small size of the base and emitter contact areas, the contacts themselves must be extended beyond the silicon oxide layer so that lead wires can be attached. The half-liner itself has, for example, an edge length of 0.75 mm and is 0.1 mm thick.

To put a layer of a mixture of molybdenum and a modifier metal on the wafer 10 to attach - this layer forms an emitter contact 18 and a base contact 19 (see FIG. 4) - is die 10 which is part of a larger semiconductor wafer along with a number of others Disks mounted on a carrier plate 20 in a conventional HF spray device 21 (see FIG. 3). Of the For the sake of simplicity, only the formation of the molybdenum-titanium contacts is described. Contacts off other compilations can be made in the same way. The carrier plate 20 serves as an anode in the high-frequency spray circuit, while an atomizer plate 22 is provided as the cathode, which the metal which is to be deposited on the semiconductor die 10. Of course there are several Atomizer plates available if different metals are to be deposited. In the present In this case, the atomizer plate 22 consists of the desired mixture of molybdenum and titanium. It can be manufactured in the usual way by powder metallurgical processes. The atomizer plate 22 is carried by a holding plate 23 which is electrically operated via a switch with a not shown HF generator is connected.

In order to deposit the gold, a gold atomizer plate 24 is attached to a holding plate 25 which in turn is also connected to the HF generator via a switch. Of course it provides the F i g. 3 is only a schematic drawing, so that from it the arrangement of the various elements cannot be taken exactly. To have uniform metal layers on all semiconductor wafers To achieve this, each atomizer plate must be at least as large as each semiconductor wafer to be coated and also should the clearances between each atomizer plate and the surface of each Be the same semiconductor wafer. As a result, each sprayer has a device within its chamber 26, with the help of which the selected atomizer plate can be adjusted for the respective precipitation process can.

It is now, for example, argon under a pressure of approximately 6 to 20 μbar through an inlet 27 'into the Sprayer 21, whereupon the high-frequency energy between carrier plate 20 and holding plate 22 creates; A frequency of approximately 15 MHz has proven to be expedient, and the process of the Precipitation should be maintained long enough for a molybdenum-titanium layer to form on the semiconductor wafer 10 arises, which is approximately 250 nm thick. After that, the HF energy is switched off and between the carrier plate 20 and the gold atomizer plate 24 are applied. It then remains switched on until the molybdenum-titanium layer a gold layer, for example 1000 nm thick, has arisen. After that, will HF energy is switched off, the flow of argon is interrupted and the semiconductor wafer 10 is connected to the device 21 taken. Of course, the molybdenum-titanium layer can also be applied in another way, for example by triode atomization. The same applies to the gold layer. But you can also go to the usual Use vapor deposition if this seems cheaper.

Then the excess parts of the

Gold and molybdenum-titanium layers are removed by selectively etching the semiconductor wafers, in particular using photoresist masks. Suitable for this

mi is particularly a photoresist material, which in certain Way is exposed to ultraviolet light through a mask at the places where the gold and the molybdenum-titanium layer should remain. The unexposed parts of the photoresist layer are then

(i> Removed by a developer liquid. The photoresist mask is then only over those parts of the gold and Moiybdän-titanium layer that the emitter and Basic contacts as well as the extended lines

should form as they are the F i g. 4 shows.

The semiconductor wafer is then etched to remove the parts of the not covered by the mask Gold layer 27 to remove. A suitable etchant for gold is an alcoholic solution of iodine and Potassium iodide. The now exposed parts of the molybdenum-titanium layer 28 are now also etched away, for example by means of a basic solution of potassium ferricyanide, so that the contacts 18 and 19 arise, which consist of the firmly adhering layers 27 and 28; the molybdenum-titanium layer 28 also adheres extremely well to the silicon oxide layer 13 and the surface of the semiconductor die 10. The latter is now ready to be assembled and packaged. An example a non-hermetically sealed encapsulation is a plastic casing 5, as shown in FIG. 5a and 5b show. After the semiconductor die 10 has been attached to a collector lead 9, is a gold wire 6 connected both to the base contact 19 and to a base lead 8, for example with the help of so-called "ball bonding", and another gold wire 7 connects between the emitter contact 18 and an emitter lead 10. The plastic casing 5 is then placed over the semiconductor wafer 10 educated.

In order to create a good ohmic contact of low resistance between the silicon and the molybdenum-titanium layer, it is necessary that the surface area of the silicon in which the contact is to be formed has a high density of impurities and is either N or P conductive ; As is well known, this applies to all contacts made from difficult-to-melt metals. If either boron or phosphorus is used as the doping material, the surface concentration should be greater than 2 ¹⁹ atoms / cm ³ , in particular greater than 10 ²¹ atoms / cm ³ . Of course, electrical contacts to silicon surfaces with a lower density of impurities can also be produced, but the contact resistance at the contact increases as the concentration of the impurities decreases.

In typical transistors, the N-conducting emitter usually has a very high concentration of impurities on, especially in the area of its surface, since the emitter zone by a second diffusion has been created. The base zone usually has a lower density of impurities, but it has at least on the surface also a high degree of doping, so that a contact is low Resistance can be produced. If this is not the case, before the contact material is deposited In a further diffusion process, a flat P-conductive diffusion zone was additionally created. That Diffusion can occur through openings of approximately the same size and approximately the same place as the openings 14 and 15 which have been formed for the production of the base contacts, and preferably these are the same openings. In the case of integrated circuits, special diffusion steps are necessary It is even more necessary to generate high surface concentrations in the area of the contacts. This is coming this is because the collector contact is made on top of the die in an area where it is around an epitaxial layer of low impurity density or around the first diffusion zone in a threefold manner diffused area can act usually has a relatively low Störstcllenkonzentration to the two following diffusion stepsc to be able to perform. Furthermore, the base zone of the transistor or one is also integrated Circuit usually created simultaneously with the formation of resistors created by diffusion. Tuesday the specific resistance of the material that forms these "resistances" should be relatively high the impurity density in the base zone must be quite low. As a result, the concentration should be ar Dopants in the area of the collector, the base;

ίο and the resistances are increased where contacts must be attached.

The F i g. 6 shows an integrated circuit in section with interconnectors in only one plane needed. It is in a P-conductive silicon wafer 30 and comprises a transistor at the left end "with a collector 31 generated by diffusion, base 3i and emitter 33; it is an NPN transistor. At the right end there is a resistance in an insulating zone 34, the resistance itself being formed by a P-conducting diffusion zone, unc is denoted by 35. Before the second diffusion step for introducing N-conductive impurities is carried out will, d. H. before the formation of the emitter 33, wire in an insulating layer 36, which, for example, au; Silicon oxide can exist, an opening made at the point where the collector contact should be formed; consequently, at the same time, mil a diffusion region 37 with a high density of N + impurities is formed in the emitter 33. Diffusion areas 38.3u and 42 with a high P + junction density are subsequently generated by selective diffusion of boron, the insulating layer 36 being used as a mask Can be found. Then in the insulating layer in the usual way by means of a photoresist layer and a Etching process openings created, namely ar those places where contacts to the Transistoi and to which resistance are to be produced. Finally, in the manner described, a Molybdenum-titanium layer 40 is applied and this is covered in front of a gold layer 41, whereupon this layer can be selectively removed, creating the desired line and contact pattern. The Kollektoi 31 is connected to one end 38 of the resistor by an interconnection 39 which extends across the Insulating layer runs. This interconnection as well as the other contacts and interconnections of the integrated circuit a layer 40 of: a mixture of molybdenum and titanium, which is attached to the Insulating layer 36 and the exposed surface portions of the semiconductor die 30 adhered, as well as over it a gold layer 41, which in turn is firmly adhered to the layer 4C.

A special diffusion of P ⁺ impurities to produce a low resistance at the contact can be unnecessary if an extremely thin film of aluminum or platinum is applied to the silicon surface and, in the case of platinum, is sintered into it, so that platinum silicide is formed before one the molybdenum-titanium layer applies.

Fig. 7 shows an integrated circuit which requires interconnections in more than one level, A semiconductor wafer 70 made of silicon, for example, has a number of types formed therein functional elements. In the picture there are 16 such elements are shown, but there could be many more. Each of these functional elements 7t through 86 includes a number of transistors, resistors, capacitors and the like connected to each other

result in the desired circuit. For example, the functional element 73 can comprise the circuit shown schematically in FIG. 9 and in the top view in FIG. 8. In the illustrated embodiment, these are PNP transistors 90 to 93 and NPN transistors 94 to 100; in addition, three input connections A, B and X and one output connection G are provided. These connections as well as a voltage supply connection V correspond to the connections denoted by the same letters in FIG. 7th

If the four functional elements 73, 76, 81 and 86 of the total of 16 elements 71 to 86 are to be connected to one another in a certain way, so that a special electrical circuit is created, as shown in FIG. 7 is shown, the connections B, D, / and OR-functional elements 73, 76, 81 and 86 are connected to one another with the aid of an intermediate connection 87. In the same way, the connections V, F, L and R are connected to one another by means of an interconnection 88 and the connections X, H, and Q are connected to one another by means of an interconnection 89. When one considers that there are already a large number of electrical interconnections in a first level connecting the various transistors to one another and to other circuit elements and connections, it can be seen that the interconnects 87 to 89 must necessarily cross some of the interconnections of the first level, as shown in FIG. 8 reveals. For this reason, and since the interconnections for the functional elements are expediently made separately from those for the individual circuit elements, they are formed in a second level according to FIG. 7 and an insulating layer is placed between them and the interconnections of the first level.

The transistors and other switching elements can be formed in or on the semiconductor wafers 70 in a known manner, for example using epitaxial technology or by diffusing in dopants. Thus, FIG. 10 shows in section a part of the integrated circuit of FIG. 8 before the application of any metallic interconnections. An NPN transistor 94 comprises an N-conducting collector formed by the semiconductor wafer itself, as well as a diffusion-created P-conducting base 110 and an N-conducting emitter 111. A resistor R \ is formed by a P-conducting and through Diffusion-created zone 112 formed, which was produced simultaneously with the base 110 of the transistor 94. An insulating layer 113, which can consist of silicon oxide, for example, lies on the surface of the semiconductor wafer and is graduated, as can be seen from the drawing, which results from the successive diffusion steps. Openings are then formed in the insulating layer in order to subsequently be able to ohmically connect the metallic interconnections lying in the first level to the semiconductor components.

In a next step, a thin molybdenum film, for example approximately 120 nm thick, is made 116 deposited on the surface of the insulating layer 113, which is in ohmic contact with the semiconductor material at the openings in the insulating layer. The film 116 can also be composed of a mixture of molybdenum and titanium, but is corrosion resistance this mixture is not yet required for this film, as it is covered by a further insulating layer 119 (see Fig. 12) is covered and protected. To deposit the molybdenum film you can a wide variety of processes can be used, as is also the case with the molybdenum-titanium layers has already been explained. A is then applied to the molybdenum film by any of the known methods Gold layer 117 deposited, which can be, for example, 750 nm thick. In their place, however, can also other highly conductive metals, such as copper, are provided

ίο that is metallurgically stable with molybdenum. A second molybdenum layer 118, for example approximately 120 nm thick, is formed on the gold layer, whereupon the three layers of molybdenum, gold and molybdenum are successively etched in the manner described above, so that the interconnections of the first level are produced; the interconnection 104 is formed, for example, which connects the base 114 of the transistor 94 to one end 112 of the resistor R \ ; an interconnect 101 is ohmically connected to the emitter 111 of the transistor 94, while an interconnect 102 ohmically connects the collector 70 of the transistor 94 to the voltage supply terminal V (see Fig. 11). Each interconnection thus has a lower layer of molybdenum (116), an intermediate layer of gold (117) and an upper layer of molybdenum (118).

An insulating layer 119, for example made of silicon oxide, is then applied to the molybdenum layer 118 by any of the known methods, for example by vapor deposition or spraying and then selectively etched in order to expose the surface of the molybdenum layer 118 only where the voltage supply connection V is located (see FIG. 12 ). The purpose of the insulating layer 119 is to electrically isolate the interconnections between the two planes.

The insulating layer can also consist of silicon nitride, aluminum oxide or other inorganic or consist of organic insulating materials. With the preferred

In the th embodiment, the insulating layer 119 is formed from silicon oxide which is sprayed on by means of HF sputtering in a layer thickness of approximately 2000 nm. In order to create a better ohmic contact between the interconnections of the two levels, the uppermost molybdenum layer 118 is selectively etched away in the area of the voltage supply connection V , in particular with the aid of a photoresist mask, so that the contact with the gold layer 117 can be made directly.

A molybdenum-titanium layer 120, for example approximately 120 nm thick, is formed on the insulating layer 119, for example by high-frequency sputtering, on which a gold layer 121, in particular, is vapor-deposited, the thickness of which is expediently approximately 750 nm. Then the gold layer and molybdenum-titanium layer 121 and 120, respectively, are selectively etched to create the desired pattern for the interconnects 122 of the second level; a contact between the two levels is created at connection V between the gold layer 117 and the molybdenum-titanium layer 120. As already mentioned, the gold can easily be etched in an alcoholic solution of iodine and potassium iodide, while the molybdenum-titanium layer is etched using another etching solution is removed, namely expediently with potassium ferricyanide, which can also be used for pure molybdenum. The top gold layer 121 adheres well to the molybdenum-titanium layer 120.

Finally, an outwardly leading gold wire (not shown) can be connected to the gold layer 12t , for example by applying pressure and heat.

One of the many advantages of the system shown in FIG. 12 is the extremely good adhesion of the molybdenum and the mixture of molybdenum and titanium (layers 118 and 120, respectively) to the insulating layer 119. An improved adhesion is observed which can be attributed to the titanium, when using mixtures of molybdenum and titanium with up to 3% titanium. The latter increases the adhesion of molybdenum to silicon oxide or to glass surfaces as a result of the extremely strong Ti-O bonds that arise, and improves the adhesion to gold through Ti-Au bonds that occur at the interface between gold and the molybdenum system. Titan emerges. If more levels of interconnection and contact are required, all but the topmost level can be made of pure molybdenum, while this should be made of a combination of niolybdenum-titanium with gold; the uppermost gold layer has the advantage that connecting wires leading to the outside can easily be attached. If you want to lead a lead directly to the lower molybdenum layer, it can be useful to create a gold layer only on the molybdenum where it is exposed through an opening in the insulating layer above, so that the gold wire can be connected to the relatively small gold layer .

To the great advantages achieved by the invention To demonstrate the corrosion resistance, the following are a number of examples are recorded:

In a conventional water drop test at a voltage of 6 volts at a number of thin lines from the to metal under test and a drop of a 10- ³ -prozentigen phosphoric acid on the lines to complete the circuit, a three-layer metal strip fails molybdenum-gold-molybdenum within 20 seconds since the anodic molybdenum layers were etched away. A three-layer strip with 80% molybdenum, 20% titanium-gold-80% molybdenum and 20% titanium lasted at least 20 minutes before the hard-to-melt alloy and gold on the anode were etched away. The dissolution of both metals proceeded with about the same straightforwardness.

In examining the metallurgical stability of a metal strip of a mixture of 19% titanium and 81% molybdenum was of about 10- ⁴ mm thick, was provided with an about 70 μιτι thick layer of gold and then heated to 600 ⁰ C, during a Hour in a nitrogen atmosphere. The sheet resistance of the strip was 0.044 ohms at the beginning, but it increased to 0.098 ohms, which shows how little the titanium reacts with the gold. In another test with a similar material, the sheet resistance, which was originally 0.064 Ohm, did not increase at all at 500 ° C. over a period of 5 minutes. So both experiments show which

the system has great metallurgical stability.

The following examples show the advantages of using a mixture of tungsten with titanium can be achieved: In a conventional water drop test with a voltage of 6 volts on a number of thinner

Lines because the top tungsten layer etched away from the to metal under test and a drop of a 10- ³ -prozentigen phosphoric acid on the lines to complete the circuit fails, a three-layer metal strips made of tungsten-gold tungsten within 5 min. In the vicinity of the cathode was; the evolution of hydrogen leads to a strongly alkaline reacting zone around the cathode. A three-layer strip with 90% tungsten and 10% titanium-gold, 90% tungsten and 10% titanium lasted at least 30 minutes before the gold layer on the anode was etched away.

When testing the metallurgical stability, a metal strip made of a mixture of 18% titanium and 82% tungsten, which was approximately 10 ~ ⁴ mm thick, was provided with an approximately 7 χ 10 " ³ mm thick layer of gold and then heated to 600 ° C, The sheet resistance of the strip was initially 0.073 ohms and dropped to 0.049 ohms after 10 minutes, due to the fact that the gold was annealed and degassed in the process, and the sheet resistance was 0.049 even after 30 minutes This test shows that the titanium of the tungsten-titanium mixture does not penetrate into the gold layer, because if this were the case, the sheet resistance of the metal strip should have increased because of the increase in the resistance of the gold layer.

In an accelerated transistor life test, a number of NPN transistors were used with different contact and interconnection systems brought into a room, which at 85 ° C 85% relative humidity. A voltage of +2 volts was applied to the emitters, the base of each The transistor was grounded and a voltage of +6 volts was applied to the collectors. Transistors, whose contacts and interconnections consisted of a tungsten and a gold layer, were after 64% failure for 96 hours and 95% failure after 250 hours due to broken contacts or interconnection lines. Transistors with contacts and interconnections from 80% Tungsten and 20% titanium as well as a gold layer were only about 5% after 96 hours and only closed after 250 hours 12.5% failed. In the case of the transistors, their metal layers Consisting of 97% tungsten and 3% chromium as well as a gold layer, only resulted after 96 hours a failure rate of 2% and after 250 hours a failure rate of 13%.

In addition 6 sheets of drawings

Claims (4)

Palent claims: 1. Semiconductor arrangement with metallic contacts and / or metallic connecting layers the semiconductor surface and / or on an insulating layer over the semiconductor surface, thereby characterized in that the contacts or connecting layers are made from a mixture Molybdenum or tungsten and a modifier metal or metalloid are made up of a larger Has corrosion resistance than molybdenum or tungsten, which is capable of passivating To form oxide layers and the oxide of which is more stable than the oxides of molybdenum or tungsten. '5 2. Semiconductor arrangement according to claim 1, characterized in that the modifier metal or -Metalloid is titanium, tantalum, chromium, zirconium, hafnium and / or silicon. 3. Semiconductor arrangement according to claim 1, characterized in that on at least part of the metallic connecting layer there is a metal layer of higher electrical conductivity. 4. Semiconductor arrangement according to claim 3, characterized in that the metal layer is higher Conductivity consists of gold.