Process for producing nitrogen-doped semiconductor wafers

Details Process for producing nitrogen-doped semiconductor wafers Process for producing nitrogen-doped semiconductor wafers

2000-08-31

2002-02-26

Hiteshew, Felisa (Department: 1765)

Single-crystal, oriented-crystal, and epitaxy growth processes;

Processes of growth from liquid or supercritical state

Having pulling during growth

C117S066000, C117S073000, C117S076000, C117S916000, C117S930000

Reexamination Certificate

active

06350314

ABSTRACT:

BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a process for producing nitrogen-doped semiconductor wafers, with the nitrogen being derived from a dopant gas which contains NH
3
. The process comprises the steps of pulling the single crystal from a melt of molten semiconductor material and cutting the nitrogen-doped semiconductor wafers off the pulled single crystal.
2. The Prior Art
It is known to cut semiconductor wafers off a single crystal. The single crystal is obtained by zone refining (floating zone method, or FZ method) or by pulling from a molten material which is contained in a crucible (Czochralski method, or CZ method). U.S. Pat. No. 4,591,409 has described a process which achieves a uniform distribution of nitrogen in a single crystal which has been pulled using the CZ method. According to this process, it is necessary to ensure the presence of a gas, such as dinitrogen oxide, during the pulling operation.
A scientific article reports experiments involving reacting nitrogen gas or a gas mixture of helium and NH
3
with molten silicon (W. Kaiser, C. D. Thurmond,
J.Appl.Phys.
30, No. 3, 427-431 (1959)). However, the article does not contain any teaching as to how semiconductor wafers can be reproduced which contain nitrogen in a predetermined concentration and with a uniform distribution. Due to the thermal instability of NH
3
, reproducible doping cannot be achieved without suitable measures.
SUMMARY OF THE INVENTION
The present invention relates to a process for producing nitrogen-doped semiconductor wafers, comprising the steps of pulling a single crystal from a melt of molten semiconductor material; feeding a dopant gas which contains NH
3
to the semiconductor material whereby nitrogen is derived from said dopant gas; feeding the dopant gas to the semiconductor material at most until pulling begins for that part of the single crystal from which the semiconductor wafers are cut; and cutting the nitrogen-doped semiconductor wafers off the pulled single crystal.
Compared to the process described in U.S. Pat. No. 4,591,409, this process of the present invention has the advantage that the axial distribution of nitrogen in the single crystal is more uniform. That is the axial gradient of the nitrogen concentration in the single crystal is less pronounced. Furthermore, the process prevents oxygen derived from the dopant gas from corroding graphite-containing internal surfaces or structures in the pulling apparatus, forming CO and, consequently, contaminating the semiconductor material with carbon.
According to the invention, a dopant gas is fed to the semiconductor material during the pulling of the single crystal in a pulling apparatus at most until that part of the single crystal is pulled which is intended to be processed further to form semiconductor wafers. This part of the single crystal is the entire, or at least virtually the entire, part of the single crystal which is of cylindrical shape. The conical parts which adjoin this cylindrical part are preferably not processed further to form semiconductor wafers. After the feed of dopant gas to the semiconductor material has been terminated, the pulling of the single crystal is completed in the usual way, for example in a pure inert gas atmosphere.
If pulling is carried out using the CZ method, the feed of dopant gas to the semiconductor material should preferably be commenced at the earliest when the semiconductor material contained in a crucible has melted completely. If pulling is carried out using the FZ method, the feed of dopant gas to the semiconductor material should be commenced at the earliest when the pulling of a so-called thin neck has already commenced. Preferably the feed of dopant gas should begin when the pulling of the thin neck has ended and an initial cone is being pulled using the FZ method. In this case, the feed of dopant gas to the semiconductor material is preferably ended before the pulling of the cylindrical section of the single crystal has commenced.
The dopant gas is passed through the pulling apparatus at a defined flow rate and with a defined concentration of NH
3
for a defined time. Preferably, the dopant gas is fed to the semiconductor material in the cooled state. It is also preferable for the flow of dopant gas to be passed to the open surface of the molten semiconductor material. For example, the flow of gas can be fed through a cooled tube (in particular in the case of the FZ method) or through a heat shield surrounding the single crystal which is to be pulled (in particular in the case of the CZ method), close to the open surface of the molten material. Furthermore, in the case of the CZ method it is preferable, in order to achieve an axially uniform distribution of nitrogen in the single crystal, if the single crystal is pulled in a magnetic field, the field lines of which are axially oriented.
It has been established that the presence of an excessively high concentration of nitrogen in the molten material can prevent a single crystal from growing and that nitrogen which is dissolved in the molten material in practice can no longer escape from the molten material. Therefore, it is particularly preferable for the concentration of nitrogen in the single crystal not to be allowed to increase beyond a limit amount of 5*10
15
atoms/cm
3
, preferably not beyond 3*10
15
atoms/cm
3
. Accordingly, the flow rate of dopant gas through the pulling chamber, and the concentration of NH
3
in the dopant gas, and the time for which the dopant gas is fed to the semiconductor material are to be selected in such a way that wherever possible the limit amount is not exceeded. Typically, for example, to pull a single crystal with a weight of from 30 to 120 kg using the CZ method, a total quantity of from 0.01 to 20 liters (s.t.p.), preferably from 0.1 to 3 liters (s.t.p.) of NH
3
should be sufficient. This is because given an effective feed of gas directed toward the surface of the molten material, nitrogen is absorbed by the semiconductor material from approximately 25% to 50% by volume of the NH
3
supplied. If the gas is fed less effectively, the amount of dopant gas required increases accordingly.
The dopant gas provided is an NH
3
-containing gas, preferably a mixture of NH
3
and an inert gas, particularly preferably an NH
3
/argon mixture. The semiconductor material can be either silicon or germanium, and preferably is silicon.
A single crystal which is produced in the manner described above is divided into semiconductor wafers of the desired thickness in a known way, for example using a wire saw or an annular saw. Usually, only that part of the single crystal which is substantially cylindrical is completely or partially divided into semiconductor wafers.
The present invention will now be further illustrated by reference to the following Example which is not to be deemed limitative of the present invention in any manner thereof.
DETAILED DESCRIPTION OF PREFERRED EMBODIMENT


REFERENCES:
patent: 4591409 (1986-05-01), Ziem et al.
patent: 5935320 (1999-08-01), Graef et al.
patent: 6191009 (2001-02-01), Tamatsuka et al.
patent: 0 170 788 (1986-02-01), None
patent: 0284107 (1988-09-01), None
patent: 0 829 559 (1998-03-01), None
W. Kaiser, C.D. Thurmond, J. Appl. Phys. 30, No. 3, 427-431 (1959).
W. Kaiser, Nitrogen in Silicon, Journal of Applied Physics, vol. 30, No. 3, p. 427-431.

Affiliated with

Also associated with

Rating

Process for producing nitrogen-doped semiconductor wafers does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Process for producing nitrogen-doped semiconductor wafers, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Process for producing nitrogen-doped semiconductor wafers will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2963343