Method and apparatus for manufacturing a silicon single...

Single-crystal – oriented-crystal – and epitaxy growth processes; – Apparatus – For crystallization from liquid or supercritical state

Reexamination Certificate

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

C117S217000, C117S220000, C117S222000

Reexamination Certificate

active

06364947

ABSTRACT:

BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a method and an apparatus for manufacturing a silicon single crystal having few crystal defects and a uniform oxygen concentration distribution, as well as to a silicon single crystal and silicon wafers manufactured by the same.
2. Description of the Related Art
Along with a decrease in size of semiconductor elements for coping with the increased degree of integration of semiconductor circuits, quality requirements are recently becoming severer on silicon single crystals which are grown by the Czochralski method (hereinafter referred to as the CZ method) for use as materials for substrates of semiconductor circuits. Particularly, there has been required a reduction in density and size of grown-in defects such as flow pattern defects (FPD), laser scattering tomography defects (LSTD), and crystal originated particles (COP).
In connection with the description of the above-mentioned defects incorporated into a silicon single crystal, first are described factors which determine the concentration of a point defect called a vacancy (hereinafter may be referred to as V) and the concentration of a point defect called interstitial (hereinafter may be referred to as I) silicon.
In a silicon single crystal, a V region refers to a region which contains a relatively large number of vacancies, i.e., depressions, pits, or the like caused by missing silicon atoms; and an I region refers to a region which contains a relatively large number of dislocations caused by excess silicon atoms or a relatively large number of clusters of excess silicon atoms. Further, between the V region and the I region there exists a neutral (hereinafter may be referred to as N) region which contains no or few missing or excess silicon atoms. Recent studies have revealed that the above-mentioned grown-in defects such as FPD, LSTD, and COP are generated only when vacancies and/or interstitials are present in a supersaturated state and that even when some atoms deviate from their ideal positions, they do not appear as a defect so long as vacancies and/or interstitials do not exceed the saturation level.
According to a popular view, in the CZ method, the concentration of vacancies and/or interstitials depends on the relation between the pulling rate of crystal and the temperature gradient G in the vicinity of a solid-liquid interface of a growing crystal (see FIG.
4
), and another type of defect called oxidation-induced stacking fault (OSF) is present in the vicinity of the boundary between the V region and the I region (Erich Dornberger and Wilfred von Ammon, J. Electrochem. Soc., Vol. 143, No. 5, May 1996; T. Abe, H. Harada, J. Chikawa, Paper presented at ICDS-12 Amsterdam, Aug. 31-Sep. 3, 1982).
According to a conventional pulling method, in view of cost of growth and assuming that OSF does not exist in a V-rich region, pulling is mostly performed in the V-rich region, in which a crystal can be grown at a relatively high growth rate. Also, a thermal history or the like during pulling has been controlled so as to reduce crystal defects generated in the V-rich region. For example, according to control practiced in the conventional pulling method, a transit time across a temperature zone of 1150-1080° C. is made relatively long so as to reduce the density of defects, each of which is conceivably a cluster of vacancies such as FPD, so that there can be improved the dielectric breakdown strength of oxide film, which is a factor for evaluating device characteristics. However, recent studies have revealed that such a method as controlling a thermal history (a transit time across a certain temperature zone) during pulling can reduce the density of defects, but the size of defects rather increases with a resultant failure to reduce a total volume of defects.
Thus, there has recently been made an attempt to reduce the pulling rate for quality improvement in spite of an increase in manufacturing cost or to increase a temperature gradient in the vicinity of the solid-liquid interface of a crystal as much as possible, to thereby manufacture a crystal that partially or entirely has an I-rich region, in which FPD, LSTD, COP, and like defects are observed less often. However, recent studies have revealed that even in the I-rich region, relatively large-sized Secco etch pit defects (hereinafter referred to as L-SEPD) are present at a portion located away from the boundary region between the V-rich region and the I-rich region. L-SEPD is conceivably a dislocation loop formed of a cluster of excess interstitial silicons. L-SEPD may be more likely to have an adverse effect on device characteristics than are FPD, LSTD, COP, and like defects generated in the V-rich region.
A recent tendency to increase the degree of integration of semiconductor devices requires uniformity of properties over a silicon wafer surface. Particularly, the oxygen concentration distribution is desired to be uniformly distributed over the surface of a wafer since the distribution directly influences the yield of devices.
Conceivable factors responsible for an impairment in the oxygen concentration distribution over the surface of a silicon wafer obtained from a crystal grown by the CZ method include convection of silicon melt, conditions of a gas atmosphere, rotation of a crystal, and rotation of a crucible. Particularly, since a cooling rate differs between an outer peripheral portion and an inner central portion of a growing crystal ingot, a crystal growth interface (a solid-liquid interface) does not become flat, thus having an adverse effect on the oxygen concentration distribution over the surface of a wafer.
Specifically, in the CZ method, at an inner portion of a crystal growth interface, crystal growth is relatively slow because of relatively slow cooling. As a result, the crystal growth interface becomes upwardly convex. A wafer obtained by slicing the thus-grown silicon bar has growth striations on the surface derived from different times of growth. Consequently, an oxygen concentration is distributed over the wafer surface in accordance with variations in oxygen concentration in the direction of crystal growth.
Conventionally, such variations and distribution of oxygen concentration derived from the profile of a solid-liquid interface are considered unavoidable in growing a single-crystal ingot by the CZ method.
Accordingly, taking for granted that variations and distribution of oxygen concentration derived from the profile of a solid-liquid interface are present to some extent, an attempt to improve the oxygen concentration distribution over a wafer surface has been carried out through the control of the above-mentioned factors such as rotation of a crystal.
SUMMARY OF THE INVENTION
The present invention has been accomplished to solve the above-mentioned problems, and an object of the invention is to obtain at high productivity a silicon single crystal and a silicon wafer by the CZ method such that neither a V-rich region nor an I-rich region is present and a defect density is very low over the entire crystal cross section, as well as to improve the oxygen concentration distribution over the surface of a silicon wafer.
According to a first aspect of the present invention, there is provided a method for manufacturing a silicon single crystal in accordance with a Czochralski method, wherein during the growth of a silicon single crystal, pulling is performed such that a solid-liquid interface in the crystal, excluding a peripheral 5 mm-width portion, exists within a range of an average vertical position of the solid-liquid interface ±5 mm.
As a result of pulling a crystal such that the crystal growth interface (solid-liquid interface) in the crystal, excluding a peripheral 5 mm-width portion, exits within a range of an average vertical position of the solid-liquid interface ±5 mm, the crystal has only a neutral region (hereinafter referred to as the N region) and has neither a V-rich region nor an I-rich region, which contain many defects. Also, the oxyg

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 and apparatus for manufacturing a silicon single... 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 and apparatus for manufacturing a silicon single..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for manufacturing a silicon single... will most certainly appreciate the feedback.

Rate now

     

Profile ID: LFUS-PAI-O-2833073

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