Crystal section shape measuring method

Details Crystal section shape measuring method Crystal section shape measuring method

1999-10-28

2002-06-25

Rosenberger, Richard A. (Department: 2877)

Optics: measuring and testing

Shape or surface configuration

Triangulation

C356S030000, C356S635000, C117S014000

Reexamination Certificate

active

06411391

ABSTRACT:

CROSS-REFERENCE TO RELATED APPLICATIONS:
This application claims priority from Japanese Application No. JP 10-316248 filed Nov. 6, 1998, the entire content of which is hereby incorporated by reference.
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a crystal section shape measuring method which optically measures the sectional shape of a single crystal pulled by the Czochralski method (CZ method).
2. Description of the Related Art
The CZ method is one of the methods for producing single crystal as the raw material for a semiconductor. In the CZ method, as shown in
FIG. 7
, a crucible
2
provided in a furnace body
1
of the CZ furnace is filled with a crystal melt liquid
3
, from which the single crystal
4
is pulled by a pulling apparatus
5
while being rotated by a rotating apparatus
6
. Upward motion of the crucible
2
is controlled in such a way that the heating center of a heater
7
keeps a constant relative position with the liquid level, in order to evenly heat the crystal melt liquid
3
by the heater
7
.
It is preferable that the single crystal
4
takes a target shape at its upper and lower ends while being pulled. It is also preferable that the crystal body and seed crystal have a uniform diameter equal to the target value during the same process. Moreover, it is preferable to keep its deformation factor [(maximum diameter -minimum diameter)/minimum diameter], representing distortion from the roundness of the sectional shape of the single crystal
4
, at an allowable level.
As for the product quality, it is preferable to control the density of oxidation-induced stacking faults (hereinafter referred to as OSF) to a low level. OSF, as one of the crystal evaluation criteria, is a stacking fault caused by the phenomenon wherein oxygen, dissolved in the crystal to form a solid solution, separates out as an oxide while the crystal is thermally treated for oxidation. The OSF density decreases as pulling speed increases because of the accelerated quenching of the crystal. It is therefore preferable to increase the pulling speed. This also advantageously increases production efficiency.
However, increasing the pulling speed increases the deformation factor, possibly beyond the allowable limit, thus decreasing product yield. It is therefore preferable to set the optimum pulling speed at which the crystal is pulled while keeping the deformation factor within an allowable range, for improved single crystal yield and production efficiency and securing product quality. It is therefore important to accurately measure the sectional shape of the single crystal while it is pulled and thereby to accurately determine its deformation factor.
The known methods for measuring the sectional shape of the single crystal being pulled by the CZ method falls into two general categories. One is the weight method, which tries to determine the crystal diameter from its weight, and the other is the optical method, which tries to determine the crystal diameter using an optical apparatus, such as a CCD camera.
Pulling the crystal by the CZ method, however, is accompanied by the formation of projections
4
a
, referred to as crystal habit lines, regularly formed in the peripheral direction on the outer peripheral face of the single crystal
4
, as shown in FIG.
9
. The projections
4
a
extend in the crystallographic axis direction, and are formed at peripheral positions characteristic of the crystal orientation of the single crystal
4
. In order to accurately determine the deformation factor of the crystal, it is preferable to measure the shape of the crystal habit portions.
The weight method, which tries to determine the crystal diameter from the weight and length of the single crystal pulled, covers only the average diameter and is incapable of measuring the detailed sectional shape involving the crystal habit lines. The optical method, on the other hand, measures the shape more accurately than the weight method, because it reads the diameter of the fusion ring high in brilliance, formed at the interface between the crystal melt liquid and the single crystal, as the diameter of the crystal.
The optical method, as shown in
FIG. 7
, takes an image (measures light) of the base of the single crystal
4
by an optical apparatus, e.g., a one-dimensional CCD camera
8
, set at an upper oblique position over the crystal
4
, through a window
9
provided at the top of the furnace body
1
. The points C and C, at which the fusion ring A formed around the base of the single crystal
4
intersects the light measuring line B—B, are located from the brilliance change at these points C and C, in order to measure sectional shape of the crystal
4
, as shown in FIG.
8
.
More precisely, the intersection points C and C are continuously located, while the single crystal
4
makes one rotation, to find the interval W(&agr;) between the points C and C by the following equation:
W
(&agr;)=
L
(&agr;)−
R
(&agr;)
wherein,
L(&agr;) and R(&agr;) are the detected positions of the intersection points C and C, and &agr; is the angle of rotation of the single crystal. The diameter across the entire periphery of the single crystal
4
is thus measured.
However, when the one-dimensional CCD camera
8
is set in such a way that its light measuring line B—B passes through the crystal center O, the fusion ring A will stand in the single crystal
4
's light when the diameter of the crystal diminishes, causing a measuring error and perhaps making the measurement impossible.
In the actual pulling process, therefore, the one-dimensional CCD camera
8
is set in such a way as to take a photograph (measures light) of the crystal center O of the single crystal
4
's base on the side of the camera. As a result, the light measuring line B—B intersects the fusion ring A on the camera side from the crystal center O. In this case, the crystal diameter is determined by the following equation from the interval W between the intersection points C and C, measured by the one-dimensional CCD camera
8
:
D
=(
W
2
+4
r
2
)
½
wherein,
D is the crystal diameter,
W is the interval between intersection points C and C, and
r is the distance from the crystal center O to the light measuring line B—B.
However, when the one-dimensional CCD camera
8
is set in such a way as to have the light measuring line B—B on the camera side (this side) from the crystal center O, two crystal habit lines
4
a
and
4
a
, opposite each other about the crystal center O, cannot pass the light measuring line B—B simultaneously, the one following the other to pass the line. Therefore, accuracy of diameter measurement decreases significantly, when the diameter is measured in the vicinity of the crystal habit line
4
a
by the conventional optical method, which tries to determine the distance W between the intersection points C and C from the difference between the detected C positions L(&agr;) and R(&agr;).
Furthermore, none of the presently known methods can accurately sense the liquid level for controlling upward motion of the crucible, which means that the measured liquid level invariably involves an error. As a result, the light measuring line B—B of the one-dimensional CCD camera
8
will deviate from the initially set position, causing the distance (r) between the crystal center O and the light measuring line B—B to change. The measured diameter D therefore. involves an error.
In order to solve these problems, Japanese Patent Laid-open No. 63-256594 discloses a method which moves the light measuring line B—B of the. one-dimensional CCD camera
8
in the direction perpendicular to the line, trying to find the true diameter from the crystal diameters determined before and after the movement and from the distance of the movement. However, even this method cannot avoid the decreased accuracy of the diameter measurement in the vicinity of the crystal habit line, because the light measuring line is apart from the crystal center.
Under these circumstances, the inventors of the pr

Affiliated with

Also associated with

Rating

Crystal section shape measuring method does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Crystal section shape measuring method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Crystal section shape measuring method will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2950490