KR960006696B1 - Metal plug forming method - Google Patents

Abstract

The method is to deposit the tungsten after the contact area is divided into N+ contact area(4) and P+ contact area(5) because the tungsten deposition rate differs from that of the substrate. The method comprises the steps : (a) opening the N+ contact area(4), (b) depositing of 500A tungsten(7) at 230-250deg.C, (c) opening the P+ contact area(5) and polysilicon contact area(6), (d) depositing tungsten(9) at 300deg.C, and (e) depositing aluminium(10).

Description

Method of forming a two-stage selection metal plug for semiconductor devices

1 is a cross-sectional view of a semiconductor device showing a step of depositing tungsten after opening a predetermined portion of an N ⁺ contact-side insulating film on a silicon substrate.

2 is a cross-sectional view of a semiconductor device showing a step of opening a predetermined portion on a P ⁺ contact portion and a polysilicon after forming a photoresist film over the entire structure.

3 is a cross-sectional view of a semiconductor device showing a step of depositing aluminum alloy after depositing tungsten on a P ⁺ contact and a polysilicon contact;

* Explanation of symbols for main parts of the drawings

1: silicon substrate, 2: device isolation oxide film,

3: oxide film, 4: N ⁺ contact portion,

5: P ⁺ contact portion, 6: polysilicon,

7: tungsten optional, 8: photosensitive film,

9: tungsten optional, 10: aluminum alloy.

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a two-stage selective metal plug process of semiconductor devices, and in particular, in order to keep the tungsten metal penetrating into the silicon substrate and the difference in tungsten growth thickness constant, the contact portion may be N ⁺ contact portion and P ^+. After divided into contact portions, the present invention relates to a two-stage selective metal plug process for depositing tungsten. In general, tungsten selected in the semiconductor device manufacturing process was deposited using a low pressure chemical vapor deposition reactor. At this time, the reactor gas is SiH ₄ and WF ₆ and the transport gas is H ₂ . In this case, tungsten deposition is a reaction in which tungsten penetrates into silicon in the WF ₆ reaction by silicon (Si). This reaction is N ^+, depends on the type of the P ⁺ contact, particularly this reaction takes place much more than the N ⁺ region P ⁺ region. This is fundamentally due to the difference in the electronegativity of the dopant in the contacts of N ⁺ and P ⁺ . During the deposition of tungsten, the seed layer is initially formed through the Si reduction reaction of WF ₆ and the seed layer formation rate is affected by the Si surface state. That is, as the donor type dopant is present rather than the acceptor, seed layer formation is initially performed. Thus, the final growth thickness of tungsten also differs. In general, the silicon reduction reaction is a function of temperature, and at 300 ° C., the N ⁺ region is 350 kV and the P ⁺ region is 1000 kV, so the encroachment is about 250 kPa. This phenomenon is more severe when natural oxides are present. Because of this, the failure caused by more junction leakage current in N ⁺ / P ⁺ junction, the failure due to the final thickness difference in the N ⁺ , P ⁺ region is as follows. In other words, when tungsten is deposited to a target of 5000 mW at 300 ° C., it grows normally at about 5000 mW in the N ⁺ contact, but may not grow at only about 1000 to 2000 mW in the P ⁺ contact. For this reason, there is a problem of causing electrical property deterioration in combination with a later aluminum sputtering process.

In order to solve the above-mentioned problems, the contact region is formed at 250 ° C. to form a N ⁺ region and a P ⁺ contact region in order to maintain a constant tungsten silicon penetration reaction and a growth thickness difference of tungsten depending on the contact type of the substrate. Tungsten was deposited at the following low temperature and maintained silicon consumption of less than 100 Pa. That is, after deposition of tungsten on the N ⁺ region after forming the contact of the P ⁺ contact region and the polysilicon region, by the deposition of tungsten at a temperature of about 300 ℃ N ⁺ region that has already been deposited is no longer deposited tungsten, and P ⁺ Tungsten grows only on the site and the polysilicon contact area, so that tungsten penetrates into the silicon and the thickness difference between the N ⁺ and P ⁺ contact areas caused by tungsten deposition when the N ⁺ and P ⁺ contacts are exposed simultaneously. It aims to overcome.

Hereinafter, the present invention will be described in detail with the accompanying drawings. 1 shows an N-well region and a P-well region, a device isolation oxide film 2 and a transistor in a silicon substrate 1 by a conventional method, and an oxide film 3 formed thereon, followed by N ^+. A sectional view of a semiconductor device showing a state in which a predetermined portion of the oxide film 3 on the contact portion 4 is etched to open N ⁺ contacts, and then tungsten is deposited about 500 kV. In this case, tungsten is deposited using WF ₆ and SiH ₄ gas at a low temperature of about 230 to 250 ° C. using an LPCVD reactor.

FIG. 2 shows that after the process of FIG. 1, the photoresist film 8 is deposited on the entire structure, the P ⁺ contact portion 5 and the polysilicon 6 are etched by a metal contact mask process to form a contact. It is sectional drawing of the semiconductor element which shows a state.

FIG. 3 shows the step of depositing tungsten on the P ⁺ contact portion 5 and the polysilicon portion 6 to a predetermined thickness after the process of FIG. 2, and then depositing the aluminum alloy 10 on the entire structure. A cross section of a semiconductor device. At this time, as shown in FIG. 1, tungsten is not grown on the deposited N ⁺ contact portion 4, and a problem of penetration of tungsten from the N ⁺ contact portion 4 into the silicon substrate 1 is caused. The reliability of the device is also improved. As described above, by dividing the active region of the semiconductor device into an N ⁺ region and a P ⁺ contact region, it is possible to suppress the phenomenon that the selective metal penetrates into the silicon substrate region during the selective metal deposition by the CVD method. Therefore, the reliability of the metal floc process is increased by reducing the failure of junction leakage current of weak N ⁺ / P junctions compared to the P ⁺ / N region, and the difference in final tungsten growth thickness in N ⁺ and P ⁺ contacts is also controlled. Can be. This increases the reliability of the later aluminum sputtering process.

Claims (3)

In a two-step selective metal plug process of a semiconductor device in which an N-well region and a P-well region are formed on a silicon substrate, and an element isolation oxide film and a transistor are formed, the oxide film 3 on the N ⁺ contact portion 4 is formed. After etching a predetermined portion to form a contact, depositing tungsten (7), depositing a photoresist film (8) on the entire structure, and then an oxide film on the P ⁺ contact portion 5 and the polysilicon 6 (3) etching only a predetermined portion, and depositing tungsten (9) and then depositing an aluminum alloy (10). The method of claim 1, wherein the N ⁺ contact portion (4) is deposited at a low temperature of about 230 to 250 ℃. The method of claim 1, wherein the P ⁺ contact portion is deposited at a high temperature of 300 ° C. or higher.