WO2001052331A1

WO2001052331A1 - Method for producing luminous structures on silicon substrate

Info

Publication number: WO2001052331A1
Application number: PCT/EP2001/000034
Authority: WO
Inventors: Heiner Röcken
Original assignee: RUBITEC Gesellschaft für Innovation und Technologie der Ruhr-Universität Bochum mbH
Priority date: 2000-01-10
Filing date: 2001-01-04
Publication date: 2001-07-19
Also published as: DE10000707B4; DE10000707A1; AU2001239219A1

Abstract

The invention relates to a method for producing light-emitting structures on silicon substrate. A defined area of the pre-doped silicon substrate is contradoped by means of focussed ion implantation. A lateral npn or pnp junction region is produced by over-compensating the original charge carrier density. Light is emitted in the reverse-biased pn or np junction region when said region is flown through by a current. Masking steps can be totally avoided by using focussed ion implantation for producing light-emitting structures.

Description

Manufacturing process for luminous structures on silicon substrate

The invention relates to a method for producing light-emitting structures in silicon.

LEDs and other technically usable, light-emitting semiconductor components usually consist of compound semiconductors such as, for. B. GaAs, GaP or GaAsP. This is due to the fact that the electroluminescence properties are favored by the electronic band structures of these so-called Ill / V materials.

With the usual large-scale integration technologies, the problem arises that the most widespread material in chip manufacture is silicon. Since silicon and the compound semiconductors mentioned have different crystal structures, the integration of light-emitting components in microelectronic systems on silicon substrates is complex and difficult. In particular, conventional epitaxial methods are out of the question for the reason mentioned.

It has been known since 1955 that electroluminescence occurs in silicon at reverse pn junctions. A so-called breakdown current occurs, which is accompanied by a lighting phenomenon when the electrical voltage applied to the pn junction in the reverse direction exceeds a certain value (see, for example, R. Newman, Physical Review, Volume 100, 1955, pages 700- 703). This immediately gives rise to the possibility of integrating luminous components on a silicon substrate. To produce the pn over Masking and subsequent doping by means of diffusion or ion implantation required. The process is therefore complex and uneconomical.

Furthermore, it is known that electroluminescence occurs on specially treated, porous surfaces of a silicon substrate (see P.M. Fauchet, Journal of Luminescence, Volume 80, 1999, pages 53-64). The production of the porous surface is very complex and the steps involved are difficult to control. Due to the porosity, the surface of the material is particularly large and therefore extremely reactive and sensitive. A successful use of this technology for the production of light-emitting components is not yet in sight.

The object of the invention is to provide a method for producing light-emitting components on silicon substrate, which does not require additional masking steps. It is desired that the luminous structures are long-term stable and insensitive to environmental influences (atmosphere, temperature).

This object is achieved in a method of the type mentioned at the outset in that spatially limited areas of a predoped silicon substrate are counter-doped by focused ion implantation.

The use of focused ion implantation enables a precisely structured doping of the semiconductor surface. There is no need for complex masking steps. The counter-doping according to the invention can be incorporated into existing silicon technology for producing microelectronic circuits without problems. It is particularly advantageous that optoelectronic circuits with integrated lighting elements can be implemented in this way. Conceivable applications arise in displays, light guide amplifiers and other micro-optical components.

It is advisable to choose the one for focused ion implantation

Doping dose such that the charge carrier density of the predoped silicon substrate is overcompensated. On pre-doped n-type silicon result laterally limited p-type areas. Lateral npn junctions are generated by the overcompensation. In the same way, alternatively, pnp junctions can be produced by creating structured n-type regions on p-type silicon. When a sufficiently high voltage is applied, the pn or np transition operated in the blocking direction is lit.

Experiments have shown that light-emitting structures produced in accordance with the invention on a silicon substrate have a broad, white-appearing emission spectrum. The light emission is limited to a narrowly limited area of the pn junction area of less than a micrometer width.

To produce a usable light-emitting component, it is expedient to carry out the focused ion implantation in a laterally delimited region on the semiconductor surface. This can be prepared, for example, by a masking step followed by wet or dry chemical etching. The result is a pre-doped silicon island that is electrically isolated from the environment, which can accommodate additional optical or electronic components.

To contact the light-emitting structure, metal-semiconductor junctions are expediently provided in the periphery of the laterally delimited region. The two contact points must be separated from one another by the area doped according to the invention. If, for example, a p-doped structure was generated by the focused ion implantation, the contacting takes place in the n-doped region. When a voltage is applied, one of the pn junctions is in the forward direction and the other in the reverse direction. A breakthrough current flows through this, resulting in light emission.

Experiments have shown that is possible to produce the light emitting structures according to the inventive method is advantageous when the pre-doping of the silicon substrate 10 ¹⁶ - 10 ^19, preferably 5 χ 10 ¹⁷ dopant atoms per cc is. The counter-doping should then 10 ¹⁸ - 10 ²¹ carried out, preferably with 5 x 10 ¹⁹ dopant atoms per cc. If possible, the dose for counter-doping should be about two orders of magnitude larger than in the predoped substrate.

The structuring of the light-emitting component expediently takes place in that, in the case of focused ion implantation, a predetermined (hole) mask, which is located in the ion beam, is imaged on the semiconductor surface by means of ion optics. The imaging process allows the structures specified by the mask to be reproduced in a greatly reduced form. It is ion-optically easily possible to focus the ion beam in at least one direction perpendicular to the beam direction to 0.01 microns - 10.0 microns. This advantageously results in dimensions for the luminous structures produced according to the invention which are compatible with the structures of modern large-scale integration technologies.

An embodiment of the manufacturing method according to the invention is explained below with reference to the drawings. Show it:

1 shows a schematic cross section of a luminous component produced according to the invention;

Figure 2: Semiconductor surface in different

Stages in the production of a light-emitting structure according to the invention.

FIG. 1 shows a cross section through an n-doped silicon crystal 1, in which a p-doped zone 2 was produced by the method according to the invention. The npn junction is composed of an n-region 3, a p-region 2 and an n-region 4. The figure shows stationary donors 5 and acceptors 6, as well as mobile charge carriers in the form of holes 7 and electrons 8. In the Transition areas of the npn structure result in zones 9 and 10, which are depleted of charge carriers. If you lay over the n-areas 3 and 4 a voltage is present, depending on the polarity, one of the transitions 9 and 10 is in the blocking direction and the other in the forward direction. Light is emitted when a breakdown current flows through the reverse junction.

FIG. 2 illustrates the production of a light-emitting component using the method according to the invention. The starting material, which is shown in FIG. 2A, is commercially available, p-doped silicon, which is present as a macroscopic, single-crystalline disk 11 (wafer).

The first process step (FIG. 2B) consists in electrically separating a layer 12 close to the surface of approximately 50 to 500 nanometers in thickness from the substrate. This is done either by an insulator layer or - as shown here - by a pn junction. So there is a doping of the near-surface silicon layer by implantation with an n-type dopant, e.g. B. with 10 ¹⁷ atoms of arsenic per ccm.

In the next step (FIG. 2C), a laterally delimited area (mesa) with a size of a few micrometers to a few millimeters is generated. This is done in that a mask 13 is applied photolithographically to the n-doped layer 12. The n-type silicon not covered by the mask is removed by plasma etching.

In FIG. 2D, the masking layer 13 was removed and a new mask 14 was applied, which leaves the end regions of the previously generated mesa free. A high-dose implantation of arsenic, e.g. B. with 5 x 10 ²⁰ atoms per ccm. Then a suitable metal (e.g. cobalt) is allowed to diffuse in to form the silicide. In this way, a metal-semiconductor junction suitable for contacting is created, via which the light-emitting component can later be supplied with current.

FIG. 2E shows the component after the masking layer 14 has been removed. An n-doped region 16 can be seen, which has contact regions 17 and 18 at its ends. In FIG. 2F, a p-doped, line-like structure 20 is now generated according to the invention by means of a focused ion beam 19, which extends so deep into the prepared near-surface silicon layer that the two adjacent n-conducting regions 21 and 22 are completely separated from one another. The implantation is carried out, for example, with 10 ¹⁹ boron atoms per ccm. The ion beam 19 is focused on about 0.01 microns to 10 microns. The line-like structure is created either by deflecting the ion beam or by ion-optical imaging of a corresponding mask. When a voltage is applied to the contact points 17 and 18, line-like light emission also occurs along the pn junction 20, 21 or 20, 22.

Claims

claims

1. A process for the production of light-emitting structures in silicon, characterized in that spatially limited areas of a predoped silicon substrate are counter-doped by focused ion implantation.

2. The method according to claim 1, characterized in that the charge carrier density of the predoped silicon substrate is overcompensated by the counter-doping.

3. The method according to at least one of claims 1 and 2, characterized in that the ion beam is focused in at least one direction perpendicular to the beam direction to 0.01 microns to 10.0 microns.

4. The method according to at least one of claims 1 to 3, characterized in that the focused ion implantation takes place in a laterally limited area on the semiconductor surface, which is prepared on the predoped silicon substrate by a masking step with subsequent wet or dry chemical etching.

5. The method according to claim 4, characterized in that in the periphery of the laterally delimited area metal-semiconductor junctions

Contacting are generated.

6. The method according to at least one of claims 1 to 5, characterized in that the silicon substrate is predoped with 10 ¹⁶ to 10 ¹⁹ , preferably with 5 x 10 ¹⁷ dopant atoms per ccm.

7. The method according to at least one of claims 1 to 6, characterized in that the counter-doping with 10 ¹⁸ to 10 ²¹ , preferably with 5 x 10 ¹⁹ dopant atoms per ccm.

8. The method according to at least one of claims 1 to 7, characterized in that in the focused ion implantation, a (hole) mask is imaged on the semiconductor surface by means of ion optics.