Method for producing silicon monocrystal and silicon...

Single-crystal – oriented-crystal – and epitaxy growth processes; – Processes of growth from liquid or supercritical state – Having pulling during growth

Reexamination Certificate

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

C117S013000

Reexamination Certificate

active

06174364

ABSTRACT:

BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a method for producing a silicon monocrystal used for fabrication of a semiconductor integrated circuit device and the like, and a silicon monocrystal wafer.
2. Description of the Related Art
A silicon monocrystal wafer used as a substrate for a semiconductor integrated circuit device is manufactured mostly by the Czochralski method (CZ method). CZ method is a method wherein a seed crystal of a silicon monocrystal is immersed in a silicon melt molten at high temperature of 1420° C. or more in a quartz crucible, and then gradually pulled with rotating the quartz crucible and the seed crystal to grow a silicon monocrystal in a columnar shape. Generally, a larger diameter of a crystal to be grown results in higher solidification latent heat which is radiated on solidification of a melt, and requires a lower pulling rate. For example, the pulling rate for a crystal having a diameter of 200 mm is generally 0.40 to 1.5 mm/min. When a wafer made of the silicon monocrystal thus manufactured is subjected to a high temperature thermal oxidation treatment at 1000° C. or more, there is sometimes caused in the wafer an oxidation-induced stacking faults in a ring shape distribution (hereinafter referred to as a ring OSF).
However, in a silicon monocrystal wafer manufactured with a relatively high pulling rate, the ring OSF gets out of the wafer, or is present at the peripheral portion of the wafer, and at the inner part of the wafer, vacancies which are lattice points having no silicon atoms are incorporated excessively at a solid-liquid interface and are aggregated to grow into a observable defect on cooling of the crystal, which is referred to as a grown-in defect. Fusegawa et al. have first disclosed that the grown-in defect can be observed by using a Secco etching solution which can reveal the defect selectively (Japanese Patent Application Laid-open (kokai) No 4-192345). The defect is referred to as FPD (Flow Pattern Defect). Afterward, other methods to detect the defect have been studied to find defects such as those referred to as COP (Crystal Originated Particle), LSTD (Laser Scattering Tomograph Defect). However, recent studies have revealed that they are the same. Namely, observation and analysis using an electron microscope revealed that they are voids in a shape of an octahedron formed by aggregation of vacancies (occasionally referred to as Negative crystal).
The size of the grown-in defect is 0.2 &mgr;m at the largest, and therefore such a defect had almost no effect on yield of the device, when the degree of integration of the device was small, and the design rule was one &mgr;m or more. However, it has been revealed that the defect has adverse effect on the device when the design rule is one &mgr;m or less. When the grown-in defect exists in or near the device active layer, junction leak failure is caused. When the grown-in defect exists on the surface of the wafer, oxide dielectric breakdown voltage failure, junction leak failure or the like is caused. Accordingly it is necessary to decrease density and size of grown-in defects, or eliminate them, or prevent formation of the defects in order to cope with increase of a degree of integration in a device.
In order to prevent formation of the grown-in defect due to vacancies, there have been developed and manufactured by way of trial in about 1990 so-called low pulling speed crystal wherein a ring OSF at the peripheral portion of the wafer is constringed to a center portion of the wafer (crystal). It is well known by the manufacturers of crystal that a lower pulling rate results in a smaller diameter of the ring OSF, and the ring OSF is constringed to a center portion of the crystal at a pulling rate not higher than a certain rate. However, such lowering of a pulling rate in manufacture of a wafer has been averted, since OSF formed on the surface becomes the largest, which has adverse effect on the device formed thereon, and productivity is lowered because of low pulling speed.
In such a circumstance, Shinoyama et al. disclosed that a lower pulling rate results in constriction and elimination of the ring OSF at the center portion of the crystal (Ouyoubutsuri (Applied Physics), 60, p.766, 1991). Higetsu et al. presented that an oxide dielectric breakdown voltage failure was caused on the inside of the ring OSF of the wafer, but it was not caused on the outside thereof (at the 7th crystallography symposium of the crystallography subcommittee of Japan Society of Applied Physics, p.27, 1990). The presentation triggered development and manufacture by way of trial of low pulling speed crystal. W. V. Ammon et al. made experiments and revealed that a pulling rate at which a ring OSF is constringed at the center portion of the crystal, Pcrit (mm/min) is proportional to a temperature gradient G in the center of crystal (° C./mm) along the pulling direction, and can be given by using the following formula: Pcrit/G=0.13 mm
2
/° C. min, and they published it (Japanese Patent Application Laid-open (kokai) No. 7-257991, and Journal of Crystal Growth vol. 151, p. 273-277, 1995). This is the first work which experimentally shows the theory proposed by Voronkov that type and density of excessive point defect depend on P/G (V. V. Voronkov: Journal of Crystal Growth, vol. 59, p.625, 1982).
However, manufacturers of crystal have recognized that preferentially etched pits, which are completely different in a size and a shape from those of FPD, that is a grown-in defect due to vacancies, are observed on the outside of the ring OSF, or on the wafer wherein the ring OSF is constringed to be eliminated (hereafter generically referred to as outside of the ring OSF, since it is the same as the wafer wherein the region outside of the ring OSF extends all over the surface). They have not come into question at an early stage, since they have been considered as having no effect on an oxide dielectric breakdown voltage. However, it was revealed that a failure due to leakage occurred in the yield of the device. Accordingly, there arose necessity of a wafer wherein there exists no grown-in defect (hereafter referred to as LEP) which leads to the large preferentially etched pit (herein referred to as Large Etch Pit, and abbreviated to as LEP; occasionally referred to as interstitial dislocation loop, dislocation cluster, large dislocation).
It has thus revealed that completely different grown-in defects generate on the inside and the outside of the ring OSF region. As mentioned above, it is now evident that FPD which is the grown-in defect on the inside of the ring OSF is a void resulting from aggregation of vacancies. However, LEP existing at low density on the outside thereof has not yet been identified. From a comparison with the results of the studies relating to grown-in defects with a floating zone method (FZ method), it is predicted that LEP is an aggregate of interstitial silicon atoms, and is a dislocation loop and a cluster thereof. They are also grown-in defects as they are formed during cooling of crystal.
As described above, the development of the wafer having none of FPD, LEP and the ring OSF has become important and necessary for the manufacturers of crystal.
Hourai et al. disclosed data implying probability of a wafer wherein none of FPD, LEP and the ring OSF exists (M. Hourai et al. :Progress in Semiconductor Fabrication, SEMICON/Europe, 1993 Technical Conference, Geneva, March/April, 1993). The data are shown in
FIG. 1
, which is a sketch showing approximately one fourth part of the wafer which is taken with X-ray topography after decorating grown-in defects by thermal diffusion of copper. As shown in the figure, there is a region having no grown-in defect between the ring OSF region and LEP (dislocation loop and cluster thereof) region. Namely, it was implied to form the region there exists neither FPD nor LEP (dislocation cluster) outside of the ring OSF, and to enlarge the region by controlling the crystal growing condition.
Then, Hourai et al. inv

LandOfFree

Say what you really think

Search LandOfFree.com for the USA inventors and patents. Rate them and share your experience with other people.

Rating

Method for producing silicon monocrystal and silicon... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Method for producing silicon monocrystal and silicon..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for producing silicon monocrystal and silicon... will most certainly appreciate the feedback.

Rate now

     

Profile ID: LFUS-PAI-O-2483515

  Search
All data on this website is collected from public sources. Our data reflects the most accurate information available at the time of publication.