Diamond component with rear side contact and a method for the production thereof

The invention relates to a component with a rear side contact and to a method for manufacturing same. Components of this type are used for example as sensors, actuators, active components, transistors, diodes, capacitors, resistors and the like, in particular for sensors which are resistant to high temperatures or to chemicals, radiation or magnetic fields. Actuators or sensors of this kind can be used in particular in the field of medical technology.

Conventionally, components of this type are contacted on the rear side by means of electrical through connection technology (via-hole technology). This technology is described for example in Heinz Beneking, Halbleiter Technologie (=semiconductor technology) B. G. Teubner-Verlag Stuttgart, 1991 and is used mainly in MMIC (Monolithic Microwave Integrated Circuits) technology. These are high-frequency ICs in which the rear side of the substrate represents an earth surface, i.e. the rear side of the IC is completely metallised. To the front side of the substrate are attached the components (resistors, transistors and the like), and additional non-sheathed cables. In order to realise an electrical earth connection of the components from the front to the rear side, through-holes are introduced into the substrate, the flanks of which holes are covered with metal. Thus the metallisation of the hole is accessible from the front side and from the rear side and can be connected on the front side to the components and on the rear side to the earth. What is disadvantageous about this is that not only is the rear side of the substrate partially coated with metal but also the function element itself consists of chemically reactive material. As a rule, the function element consists of an electronic or optoelectronic structure formed from semiconductor material such as GaAs. This semiconductor material is largely chemically reactive (and corrodes for example) and must be passivated (e.g. with SiO 2 or Si 3 N 4 ), which in turn makes very high demands of passivation technology.

Via-hole technology is therefore substantially limited to electronic/optoelectronic components.

The object of the present invention is to make available a component and a method for manufacturing same, which renders possible an electrical rear side contact whilst simultaneously preserving an inert component surface.

According to the invention, the component is produced by an electrically conductive layer being deposited on a substrate on the front side in the region of a future contact region and thereafter possibly additional layers being applied to form the function element. Then in the area of the contact region, the substrate is removed, starting from the rear side of the substrate, for example by a physical or chemical etching method, at least down to the electrically conductive layer and the recess so formed is metallised. Thus a component is produced which has no through-hole at the contact location and is consequently only metallised on the rear side. This renders possible a high radiological, chemical and physical resistance of the surface of the component since the metallic contact on the rear side is completely covered on the front side. Noble metal, base metal or even an alloy can therefore be used without taking into account the requirements made of the upper side of the component. Consequently an electrical rear side contact is produced whilst preserving the component surface. The physical, chemical and/or mechanical properties of the surface can therefore be completely exploited since the latter is freely available and not limited by metallised through-holes,. Thus additional diamond layers can also be applied to the surface of the component. With a complete diamond coating of the front side of the component, the excellent chemical, physical, electrical thermal and/or mechanical surface properties of the diamond can be exploited.

The electrically conductive layer can also be applied already as a conductive doped diamond layer, for example with a boron doping. The doping can here be designed in a variable manner in both vertical and horizontal directions, according to which component with the diamond coating is to be constructed on the front side of the substrate.

As the substrate can be used, for example, silicon, quartz, glass or even free-standing diamond. Non-doped diamond here has the advantages that it is completely non-conductive, such that it is possible to dispense with insulation between the diamond substrate and electrically conductive layers of the function element or of the contact metallisation. This saves further steps in the manufacturing process such that production is simplified overall. In this case the diamond substrate can also be produced as one piece with the function element, which can also be constructed from diamond layers.

Where silicon is used as the substrate it is necessary to insert an electrically insulating layer between the substrate and the electrically conductive layer of the function element and also between the substrate and the metallic rear side contact. This can be deposited for example as silicon oxide or silicon nitride or can be produced by oxidation of the silicon substrate to form silicon oxide.

Advantages about the use of diamond for the function element or for the complete coating of the front side of the component is in particular the fact that the diamond is resistant to high temperatures, completely non-conductive and chemically inert and has great hardness. Furthermore, a component with a diamond coating on the front side is suitable as a detector material for corpuscular radiation or high-energy X-rays since it is completely resistant to radiology. In particular it has no self-heating. The metal layers which are sensitive to radiation are then located protected merely on the rear side of the component.

The components according to the invention can be used as sensors, actuators, active components, transistors, diodes, capacitors, resistors and the like. As a result of their high stability they are particularly suitable as sensors, for example for pressure, impedance, temperature, in particular as sensors which are resistant to chemicals, corpuscular radiation, X-rays, high temperatures and/or magnetic fields. They can therefore be used in the field of medical technology, such as for example NMR tomography and also for instruments of micro-invasive surgery. There, for example, they can be used in catheters, which have to be completely sealed with respect to the human organism, in order to achieve high biomedical compatibility. This requires also the here guaranteed complete absence of chemically reactive elements such as e.g. metals on the catheter surface which comes into contact with the patient's body.

By providing contact on both sides of the diamond layers by means of metallisation, in addition to the above-mentioned applications vertical structures and components such as e.g. back-gate transistors, photodiodes, photoconductors etc. can be realised. In addition, pn-diodes can be produced by epitaxial diamond layers with differing doping.

Diamond layers can be deposited and doped for example on silicon substrates over the entire surface in layer thicknesses of between several μm and several mm. The deposition is usually carried out through the use of plasma-supported or hot wire methods. Selective depositions is here possible with the aid of SiO 2 as the mask material. The layers so produced have, by comparison with natural diamonds, roughly identical mechanical, chemical and physical properties.

The removal of the substrate in the contact region can be effected by means of conventional etching methods. For example, dry or wet chemical etching or even physical etching is available for this purpose.

The definition of all the structures is achieved by means of known lithographic methods such as are usual in microsystem technology.

Some components are described by way of example in what follows.

The figure shows:

FIG. 1 is a component according to the invention;

FIG. 2 is a component according to the invention; and

FIG. 3 is a further component according to the invention.

FIG. 1 shows a component according to the invention in which a function element 2 is applied to a silicon substrate 4 . Between the silicon substrate 4 and the function element there is an electrically insulating diamond layer 3 which serves as the seeding layer during the production of the component.

The function element 2 has an electrically active, doped layer, for example a p ++ -layer (having quasi-metallic behaviour). This function element 2 is electrically contacted on its underside by a metallisation layer 6 formed from a contact metal such as, for example, WSi, WSi:N, Ti, Au. This metallisation layer 6 extends through an opening in the substrate 4 to the rear side of the function element 2 . It is in turn electrically insulated from the silicon substrate 4 via a silicon dioxide layer 5 . To produce this component, first of all the seeding layer 3 of diamond is deposited on a silicon substrate 4 . Then the electrically active p ++ -layer 2 , doped with boron, is deposited on the seeding layer in the structure shown. The layer thicknesses are typically approx. 500 nm for the electrically active doped layer 2 and approx. 100 nm for the seeding layer 3 . Thereafter, the opening is introduced in the rear side of the substrate 4 . For this structuring of the rear side, first of all the opening in the silicon substrate 4 is etched by wet chemical anisotropic means in KOH. An Si 3 N 4 mask was here used to structure this opening. Then the non-doped seeding layer 3 was etched down to a defined depth within the electrically active doped layer 2 . This dry chemical etching took place in an Ar/O 2 plasma with an etching rate of approx. 350 nm/min and a depth of approx. 350 nm. Here, too, the structure was defined by means of a standard lithographic method.

As the next step, the silicon substrate 4 is insulated by oxidation or by sputter deposition of SiO 2 as the insulating layer 5 . WSi, WSi:N, Ti and/or Au is then applied as a contact metal to this insulating layer 5 and to the doped, electrically active layer 2 , from the rear side of the substrate 4 , by sputter deposition and vapour deposition to form the contact metal layer.

The deposition of the diamond layers 3 and 2 took place in an MPECVD (Microwave Plasma Enhanced Chemical Vapour Deposition) system. The doping of the electrically active layer 2 takes place by means of a solid boron source. This is directly introduced in plasma for doping in the form of a wire for a period of approx. 2 hours during the deposition of layer 2 . The doping substance concentration achieved is approx. 10 20 cm −3 . In principle, however, the doping can also take place by implantation after the deposition of the diamond layer 2 .

The component so applied can serve as a heating element for example in conjunction with a further rear side contact of this type. The function element 2 can, however, directly serve at the end of a catheter inserted into tissue to introduce currents of high intensity or high voltages locally into the tissue directly adjacent to layer 2 .

FIG. 2 shows a further component according to the invention. The same constituent parts of the component are identified with the same reference numerals as in FIG. 1 and there is therefore no description of them in what follows. In addition to the component shown in FIG. 1, the electrically active doped layer of the function element is coated with a further diamond layer 1 , which is nominally non-doped. This nominally non-doped layer has a thickness of up to 5 μm. Because of this diamond layer 1 , the component becomes extremely resistant on its surface to chemical, physical or other influences. Since the component with diamond layer 1 only has a diamond layer with respect to the outer side, in particular the resistance to high temperature, the chemical stability and the lack of sensitivity to corpuscular radiation or hard X-rays can be exploited.

FIG. 3 shows a further component according to the invention in which the same reference numerals identify the same constituent parts as in FIG. 2 . By contrast with FIG. 2, the substrate 4 does not consist of silicon but also of a diamond layer (diamond-quasi-substrate). Since diamond is completely non-conductive, the insulating layers 3 and 5 can be dispensed with. Nevertheless complete occlusion of all currents between the contact metal layer 6 and the electrically active doped layer 2 is guaranteed.

In all the figures, the electrically conductive layer 2 and the metal layer 6 produce the rear side contact for additional layers and function elements, or constituent parts of the function element, arranged on the front side of the component, to which further track conductors, lines or electrically conductive layers can be connected. The metal layer 6 is suitable as a soldering point on the rear side of the component.