JP3210489B2 - Silicon wafer and method for evaluating withstand voltage of oxide film - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an oxide film withstand voltage of a silicon wafer.

[0002]

2. Description of the Related Art Various devices are incorporated in a surface layer of a silicon wafer. In order for each device to operate electrically well, it is necessary that the withstand voltage of an oxide film exceeds a reference value. However, the factors that affect the oxide film breakdown voltage of devices have not always been sufficiently elucidated. Therefore, conventionally, a MOS capacitor for actually measuring the oxide film breakdown voltage was actually created, and the oxide film breakdown voltage was actually measured. Was.

[0003]

However, the above-mentioned conventional method for evaluating the withstand voltage of an oxide film is actually a method for measuring the withstand voltage of an oxide film.
Since an S capacitor is prepared and the breakdown voltage of an oxide film is measured, there is a problem that it takes a long time. That is, it takes about 20 hours even if the time required from the device preparation to the measurement of the oxide film breakdown voltage is simply added, and in practice, it takes about one week due to various preparations and the like. Therefore, the present invention provides an evaluation method of an oxide film withstand voltage of a silicon wafer, which can easily and quickly evaluate an oxide film withstand voltage, and therefore the cost required for the evaluation is remarkably low. It is intended to provide a simple silicon wafer.

[0004]

Means for Solving the Problems The present inventor has repeated studies to achieve the above-mentioned object, paying attention to scattered light when a silicon wafer is irradiated with a laser, and the density of a laser scatterer that generates the scattered light. Have a strong causal relationship with the oxide film breakdown voltage, that is, it has been found that the oxide film breakdown voltage is degraded when a large number of laser scatterers are present in the surface layer of the silicon wafer, thus completing the present invention. That is, the present invention is a method for evaluating the oxide film breakdown voltage of a silicon wafer, comprising evaluating the oxide film breakdown voltage non-defective product rate or defective product rate of the silicon wafer based on the laser scatterer density of the silicon wafer surface layer, The silicon wafer is characterized in that the surface has a laser scatterer density of 5 × 10 ⁵ / cm ³ or less.

[0005]

Embodiments of the present invention will be described below. A single crystal silicon ingot is manufactured by the pulling method, that is, the Czochralski method, and sliced, wrapped, chamfered,
Each step of chemical polishing was performed to obtain a silicon wafer sample. The specifications of the sample were 6 inches in diameter, crystal axis <100>,
P-type, boron-doped, resistivity 10 to ²⁰ Ωcm, oxygen concentration 12 to 15 × 10 ¹⁷ atoms / cm ³ . The sample was subjected to various heat treatments to change the density of the laser scatterers, and thereafter, the density of the laser scatterers on the wafer surface layer was measured. FIG. 1 shows a laser scatterer measuring device. A laser beam having a wavelength of 1.3 μm is vertically emitted from the laser emitting device 2 toward the surface of the silicon wafer 1, and this beam is focused on the surface of the silicon wafer to arbitrarily define a plurality of points on the wafer surface. , The wafer is scanned by sliding. When the beam hits a defect, a slight phase shift occurs. By detecting this shift, the defect is detected. That is, as shown in FIG.
In this device, the laser
The phase of the applied laser beam differs depending on the prism 2
Split into two polarized beams and incident from one side of the wafer.
Let And when one of them crosses the defect
A phase shift occurs, causing a phase difference with another beam
This phase difference after passing through the wafer.
Detecting defects by detecting with an analyzer
It is.

FIG. 2 shows the vicinity (0) of the surface of the silicon wafer 1.
(3 μm) and the oxide withstand voltage is 3
The relationship with the B mode defective product ratio of MV / cm or more and 8 MV / cm or less is shown. As is clear from the figure, there is a remarkable correlation between the laser scatterer density and the B-mode reject rate, that is, it is well understood that the B-mode reject rate increases as the laser scatterer density increases. You. FIG. 3 shows that the density of the laser scatterer and the withstand voltage of the oxide film are 8 MV / cm.
The relationship with the above C-mode non-defective rate is shown, and as is apparent from the figure, it is understood that the C-mode non-defective rate decreases as the laser scatterer density increases.

In FIGS. 2 and 3, the density of the laser scatterer is measured by the laser scatterer measuring apparatus of FIG. 1, and on the other hand, a MOS capacitor for actually measuring the oxide film withstand voltage is formed to reduce the oxide film withstand voltage. These figures are obtained as a result of measurement, but once these figures are obtained, the oxide film breakdown voltage of the silicon wafer can be quickly evaluated. That is, in the measurement of the laser scatterer density by the laser scatterer measuring apparatus of FIG. 1, for example, the area that can be measured per point by the present apparatus is 128 × 450 μm, so that at least 3
You need more than 0 points, but it still takes about an hour. There are two types of laser scanning directions: the depth direction of the wafer and the horizontal direction. In this embodiment, the scanning is performed in the horizontal direction to take a value near the surface (depth of 0 to 3 μm). At each position, laser scatterers were measured at 6 points, respectively, for a total of 30 points, and were set as representative values in the wafer plane. In this way, after that, for example, by using FIG. 3, the oxide film breakdown voltage non-defective rate of the silicon wafer can be immediately evaluated,
As a result, the oxide film breakdown voltage can be remarkably quickly evaluated as compared with the case where the oxide film breakdown voltage is obtained by the conventional direct measurement. Further, when the specification of the oxide film breakdown voltage non-defective rate is required to be, for example, 95% or more, it is understood from FIG. 3 that the laser scatterer density needs to be about 5 × 10 ⁵ / cm ³ or less. The laser scatterer density of the wafer surface layer was measured, and the measured value was 5 × 10 ⁵
/ Cm ³ or less, it is possible to easily and quickly determine whether or not to use the silicon wafer.

[0008]

The method of the present invention evaluates the oxide film breakdown voltage non-defective product rate or defective product rate of a silicon wafer based on the laser scatterer density of the surface layer of the silicon wafer.
The oxide film breakdown voltage of each silicon wafer can be easily and quickly evaluated. Further, the present invention is a silicon wafer having a surface layer having a laser scatterer density of 5 × 10 ⁵ / cm ³ or less, and thus has a remarkably good silicon film withstand voltage ratio of about 95% or more.

[Brief description of the drawings]

FIG. 1 is a diagram showing a method for measuring a laser scatterer density.

FIG. 2 is a diagram showing a relationship between laser scatterer density and B-mode defective product rate.

FIG. 3 is a diagram showing the relationship with the non-defective rate in the C mode.

[Explanation of symbols]

 1. Silicon wafer 2. Laser emitting device

────────────────────────────────────────────────── ─── Continued on the front page (72) Inventor Mitsuo Kono 2612 Yonomiya, Hiratsuka-shi, Kanagawa Prefecture Komatsu Electronic Metals Co., Ltd. (56) References JP-A-6-242036 (JP, A) JP-A-6-112292 ( JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) H01L 21/66 G01N 21/00

Claims (3)

(57) [Claims] 1. The small amount of light that occurs when a beam hits a defect.
By detecting defects by detecting the phase shift
And measures the laser scatterer density on the surface of the silicon wafer
And evaluating the oxide film breakdown voltage non-defective product rate or defective product rate of the silicon wafer based on the evaluation result. Wherein whether crab surface of the laser scattering body density by the method of claim 1, wherein is 5 × 10 5 ^/ cm ³ or less
Therefore, a method of determining whether to adopt a silicon wafer . 3. The small amount of light produced when the beam hits a defect.
By detecting defects by detecting the phase shift
And measures the laser scatterer density on the surface of the silicon wafer
When the scatterer density is 5 × 10 ⁵ / cm ³ or less
Silicon wafer.