JP2945557B2 - Method for measuring oxide film coverage on silicon wafer surface - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for measuring a coating rate of a silicon wafer surface for determining a coating rate of an oxide film on a silicon wafer surface.

[0002]

2. Description of the Related Art Conventionally, the coverage of an oxide film formed on the surface of a silicon wafer has been measured by X-ray photoelectron spectroscopy (XPS) or ellipsometry (polarization analysis).

[0003] The coating ratio is generally considered to be the ratio of the area of the oxide film on the silicon wafer surface to the total area of the silicon wafer surface.

However, instead of directly measuring such an area ratio, the surface analysis is performed by the above-mentioned X-ray photoelectron spectroscopy to indirectly determine the coating ratio. Alternatively, the film ratio is indirectly determined by measuring the polarization state of the light reflected from the silicon wafer by the above-mentioned ellipsometry method.

[0005]

In the above-mentioned X-ray photoelectron spectroscopy, since the measurement is performed in a high vacuum, a process for obtaining such a high vacuum is required. Therefore, there is a disadvantage that the measurement cannot be performed easily. Further, since the measuring device for performing this measurement is expensive, it cannot be performed at low cost. Therefore, there is a disadvantage that the economy is poor.

[0006] On the other hand, the above-mentioned ellipsometry method can easily carry out the measurement and can use an inexpensive measuring device, as compared with the X-ray photoelectron spectroscopy method. Further, there is a disadvantage that the coating ratio cannot be obtained with sufficient accuracy.

According to the present invention, measurement can be performed easily,
In addition, it is an object of the present invention to provide a method for measuring the coating rate on the surface of a silicon wafer, which can accurately determine the coating rate.

[0008]

Generally, the coverage of an oxide film on a wafer surface is expressed as a ratio of the area of the oxide film to the area of the wafer surface. The inventor of the present invention has found that the peak intensity of the longitudinal wave optical mode (LO) of the long-wavelength polar lattice vibration of the oxide film on the wafer surface, which will be described in detail later, almost matches the growth state of the oxide film on the wafer surface, that is, the coating rate. Therefore, the present inventors have found that this peak intensity can be used as a coating rate, and have accomplished the present invention. Longitudinal wave optical mode (LO) of long wavelength polar lattice vibration when observing an oxide film on the wafer surface
Frequency, peak intensity, and shear wave optical mode (T
The inventors have found that the split width between the frequency O) and the longitudinal wave optical mode (LO) frequency has a correlation with the growth state of the oxide film, and have accomplished the present invention.

In order to solve the above-mentioned problems, the first of the present application
The present invention includes a step of measuring a peak intensity of a longitudinal wave optical mode (LO) of a long-wavelength polar lattice vibration of an oxide film when the surface of a reference wafer is saturated with an oxide film; Measuring the peak intensity of the longitudinal wave optical mode (LO) of the long-wavelength polar lattice vibration of the oxide film, and obtaining the ratio of the peak intensity of the measured wafer to the peak intensity at the time of the saturation as the coating ratio. The gist is a method for measuring the coating rate of an oxide film on a wafer surface.

The second invention of the present application is directed to a longitudinal wave optical mode of a long-wavelength polar lattice vibration of the oxide film and a coverage ratio of the oxide film from the time when no oxide film is present on the surface of the reference wafer to the time when the oxide film is saturated. LO) determining a relative relationship of peak intensities in advance;
Measuring the peak intensity of the longitudinal wave optical mode (LO) of the long-wavelength polar lattice vibration of the oxide film of the measurement wafer at a predetermined time; The gist of the present invention is a method for measuring the coating rate on the surface of a silicon wafer including the step of obtaining the rate.

[0011] The third invention of the present application is directed to a transverse wave optical mode (TO) of a long-wavelength polar lattice vibration of an oxide film from the time when no oxide film is present on the surface of the reference wafer to the time when the oxide film is saturated. ) And the relative frequency of the longitudinal wave mode (LO) with respect to the splitting width, and the transverse wave mode (TO) and the longitudinal wave of the long-wavelength polar lattice vibration of the oxide film of the measurement wafer at a predetermined time. A method for measuring the coating rate of a silicon wafer surface, which includes a step of measuring a split width of a frequency of an optical mode (LO) and a step of obtaining a coating rate at a measurement time point of a measurement wafer based on the measured value based on the measured value. Make a summary.

The fourth invention of the present application is directed to a longitudinal wave optical mode of a long-wavelength polar lattice vibration of the oxide film and a coverage ratio of the oxide film from the time when no oxide film is present on the surface of the reference wafer to the time when the oxide film is saturated. LO), a step of determining a relative relationship with the frequency in advance, a step of measuring the longitudinal wave optical mode (LO) frequency of the long-wavelength polar lattice vibration of the oxide film of the measurement wafer at a predetermined time, and the measured value Therefore, the gist of the present invention is a method of measuring the coating rate on the surface of a silicon wafer including the step of obtaining the coating rate at the measurement point of the measurement wafer based on the above-described relative relationship.

Hereinafter, in the present specification, the longitudinal wave optical mode of the long-wavelength polar lattice vibration is simply called the longitudinal wave optical mode, and similarly, the transverse wave optical mode of the long-wavelength polar lattice vibration is simply called the transverse wave optical mode.

The oxide film on the wafer surface is said to be amorphous rather than single crystal, and in this sense, the lattice vibration of the oxide film does not exist correctly. However, since the vibration modes measured by the parallel polarization oblique incidence transmission method and the normal incidence method are the same as the polarization characteristics in the case of a single crystal thin film,
In the present invention, these are described as lattice vibrations.

[0015]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The measurement of the longitudinal wave mode and the transverse wave optical mode of an oxide film used in the present invention will be described below.

An oxide film grown on the surface of a wafer is first formed in an island shape as time passes if the oxide film is left in an atmosphere containing oxygen from a point where no oxide film is present, for example, immediately after cleaning with hydrofluoric acid. Furthermore, as the standing time elapses, it grows from an island shape to a network shape (net shape), and so-called long-distance ordering in this network proceeds. Further, as the time elapses, the wafer grows so as to cover the entire surface of the wafer and eventually saturates. That is, the natural oxide film covers the entire surface of the wafer.

In the growth process of the oxide film, the observed transverse-wave optical mode and longitudinal-wave optical mode of the long-wavelength polar lattice vibration of the oxide film change according to the growth stage, and the change is quantified. The coating rate of the oxide film grown on the wafer surface can be measured.

The modes of the long-wavelength polar lattice vibration of the oxide film are, for example, parallel polarization Brewster angle incidence infrared transmission spectroscopy (hereinafter simply referred to as Brewster angle incidence method) and normal incidence infrared transmission spectroscopy (hereinafter referred to as "Brewster angle incidence method"). , Simply referred to as the normal incidence method).

Referring to FIG. 1, the measurement by the Brewster angle incidence method will be described below.

The first reference wafer 1 is washed with about 4.8% hydrofluoric acid (HF) and then with water to remove an oxide film on the surface of the first reference wafer 1. The concentration of hydrofluoric acid may be any concentration as long as the oxide film on the surface of the wafer can be removed.

The first reference wafer 1 is a floating zone silicon wafer (plane orientation (100), specific resistance of 1 kΩcm or more, thickness of about 700 μm), and both sides are mirror-processed.
In this state, there is almost no oxide film on the surface of the first reference wafer. The first reference wafer is left in a clean room and naturally oxidized. That is, the first reference wafer 1 is naturally oxidized by the atmosphere, so that a natural oxide film is gradually grown on the surface thereof.

The floating zone silicon wafer is a silicon wafer formed from a silicon single crystal manufactured by a floating zone melting method.

The incident light I is incident on the first reference wafer 1 at an incident angle B along the arrow and is transmitted therethrough. The incident angle B is the Brewster angle of silicon (73.7 degrees). A natural oxide film 10 is formed on the surface of the first reference wafer 1.

The incident light I is parallel polarized light.

[0025] measuring the incident parallel polarization intensity I _O of the incident light I transmitted light intensity I _A. These measurements can be performed using a conventional Fourier transform infrared spectrophotometer (FTIR) or the like.

A second reference wafer 2 is prepared separately from the first reference wafer 1. The second reference wafer 2 is the same silicon wafer as the first reference wafer 1. However, this second reference wafer 2 is immediately after being cleaned with hydrofluoric acid,
The natural oxide film on the surface is completely peeled off.

The first Brewster angle incidence method described above
In the same manner as in the measurement of the transmitted light intensity I _A of the reference wafer 1 to measure the transmitted light intensity I _B of the second reference wafer 2.

First reference wafer 1 and second reference wafer 2
Is based on the fact that parallel polarized light is incident on the second reference wafer 2 and the first reference wafer 1 at the Brewster angle B, respectively. It is to prevent multiple reflections from occurring inside the reference wafer 1 and the second reference wafer 2.

Based on Equation 1, the difference absorbance spectrum A
_Ask for _B. Here, d _A is the thickness of the first reference wafer 1, and d _B is the thickness of the second reference wafer 2. Note that the measurement parameters I _O , I _A , and I _B are spectra and are functions of wave numbers.

[0030]

(Equation 1) Differential absorbance spectrum A _B corresponds to the absorbance spectrum of the oxide film 10 of the first reference wafer 1 surface. An example of the difference absorbance spectrum A _B determined at each time point in the growth process of the natural oxide film of the aforementioned (after washing hydrofluoric acid, when standing for 8 days) are shown in 5a in the graph of FIG.

Next, the measurement by the vertical incidence method will be described with reference to FIG.

The vertical incidence method is the same as the above-described Brewster angle incidence method except that the incident light I is incident on the first reference wafer 1 and the second reference wafer 2 at a right angle R. The spectrum _AR is determined. An example of the difference absorbance spectrum A _R obtained at each point in the growth process of the natural oxide film is shown at 5b in the graph of FIG.

In the vertical incidence method, multiple reflections occur inside the first reference wafer 1 and the second reference wafer 2.

[0034] The differential absorbance spectrum A _B obtained by Brewster angle incidence method, transverse optical (Transv
error optical mode (hereinafter simply referred to as TO mode) and longitudinal wave optics (longitudina).
An optical mode (hereinafter, simply referred to as an LO mode) is observed. In the difference absorbance spectrum A _R obtained by the normal incidence method, only the TO mode is observed. That is, by performing the measurement on both the Brewster angle incidence method and the vertical incidence method for the first reference wafer 1, the above-described TO mode and LO mode can be measured separately.

FIG. 3 shows the LO mode and TO mode frequencies of the oxide film on the first reference wafer and the LO peak intensity. The difference absorbance spectrum 5a measured by the Brewster angle incidence method has the LO mode frequency (L
O) and TO mode frequencies (TO) are observed.
Further, the frequency (TO) of the TO mode is observed in the difference absorbance spectrum 5b measured by the vertical incidence method.

Using these measuring methods, the LO mode and the TO mode in the above-described natural oxide film growth process were measured, and the change with time was examined. Naturally, it is preferable to perform these measurements continuously.However, due to operational constraints, appropriate measurements at each point in time are performed over time, and the results of continuous measurement are used using those measurement results. Can be substituted for After removing the oxide film from the wafer by hydrofluoric acid cleaning, leave the wafer in a clean room, and then 0 hours (that is, no oxide film), 8 hours, 20 hours, and 2.8.
Days, 8.8 days, 16.8 days,
At each time point after the elapse of 33 days, measurement was performed by the Brewster angle incidence method, and the above-mentioned difference absorbance spectrum A _B
I got The results are shown in the graph of FIG.

When 0 hour has elapsed, 7 hours have elapsed, and 21 hours have elapsed.
Time elapsed, 2.8 days passed, 8.9 days passed, 1
At 6.8 days and 33.1 days, measurements were made by the vertical incidence method to obtain the above-mentioned difference absorbance spectrum A _R. The results are shown in the graph of FIG.

[0038] Since in the Brewster angle incidence method as described above so as to prevent multiple reflection, the peak intensity of the LO mode from the difference absorbance spectrum A _B obtained by this injection method (hereinafter, referred to as LO peak intensity) It can be obtained with high accuracy.

As can be understood from FIG. 4A, the LO peak intensity is substantially zero when there is no oxide film on the wafer.
Therefore, as the time of leaving the wafer elapses, the LO
The peak intensity increases. This LO peak intensity directly reflects the process until the wafer surface is covered with the natural oxide film. Moreover, the LO peak intensity is theoretically well proportional to the area of the oxide film formed on the wafer surface. Therefore, by determining the ratio of the peak intensity at the measurement time to the LO peak intensity at the saturation time of the wafer, the coating rate of the oxide film on the wafer can be defined.

In the first invention of the present application, specifically,
L at the time when the surface of the reference wafer is saturated with the oxide film
The O peak intensity is measured, and the LO peak intensity at the measurement time point of the measurement wafer having substantially the same physical properties as the reference wafer is measured. Then, the ratio of the LO peak intensity of the measurement wafer to the LO peak intensity at the saturation point of the reference wafer (LO peak intensity of the measurement wafer / LO peak intensity at the saturation point of the reference wafer)
To determine the coating ratio.

To be precise, the LO peak intensity at the saturation point of the measurement wafer is used as a denominator. For this purpose, the measurement wafer must be left until the oxide film is saturated.
However, it is not possible to determine the coating rate of the measurement wafer at the time of measurement. Therefore, by using a wafer of the same physical properties as the measurement wafer as the reference wafer,
The LO peak intensity when the oxide film on the reference wafer is saturated can be regarded as the LO peak intensity when the oxide film on the measurement wafer is saturated.

Also, as can be understood from FIGS. 4A and 4B, after the hydrofluoric acid cleaning (stripping of the oxide film), the frequency of the LO mode is increased to about 1110 c as the standing time elapses.
It has shifted from m ^-1 to about 1212 cm ^-1 . On the other hand, the frequency of the TO mode is almost 1050 cm ⁻¹ and hardly changes. Therefore, the difference between the frequency in the LO mode and the frequency in the TO mode (hereinafter referred to as the TO-LO split width) also increases with time. Figure 3 shows the TO-LO split width
As shown.

This LO mode frequency shift and TO
It can be said that the increase in the LO division width over time directly reflects the process of forming a network of silicon oxide (oxide film) due to oxidation of the wafer surface.

The reason is that the frequency of the TO mode is determined by the local force field without being affected by the macroscopic electric field induced by the LO mode polar lattice vibration, so that the natural oxide film grows from an island to a network. Even hardly changes.
On the other hand, since the LO mode frequency is affected by the macroscopic electric field induced by the LO mode polar lattice vibration, it is considered that the natural oxide film grows and shifts to a higher wavenumber side as long-range ordering progresses. .

Therefore, the LO mode frequency and TO
The -LO split width corresponds to the formation and growth of the native oxide film network, and the native oxide film coverage of the wafer left for an arbitrary time is determined by the LO mode frequency and the TO-L
TO-L of the first reference wafer 1 with respect to the value of the O division width
It has a close relationship with the O division width. Therefore, by measuring the LO mode frequency or the TO-LO split width, the coverage of the natural oxide film on the wafer surface can be obtained indirectly. As described above, since the frequency of the TO mode does not substantially change with time, the frequency of the LO mode and the TO-LO split width have substantially the same meaning.

The second to fourth inventions of the present application focus on the fact that the above-mentioned LO peak intensity, LO mode frequency, and TO-LO split width are closely related to the growth process of the oxide film on the wafer surface. However, the inventors have found that it is possible to obtain the coating ratio of the oxide film on the wafer surface using these.

The outline of the second to fourth inventions of the present application will be described.

The second to fourth inventions of the present application include the following first to third steps.

In the first step, a reference wafer is prepared, the LO peak intensity and / or the LO mode frequency and / or the TO-LO split width of the reference wafer are measured with time, and each measurement at each measurement time point is performed. This is a step of previously obtaining a relative relationship between the value and the coating ratio. At the time of measurement, the coating ratio may be determined by the X-ray electron spectroscopy or ellipsometry described above, or may be determined by the LO peak intensity ratio determined by the first invention of the present application. The latter is preferable because it can be automatically obtained as a result by performing the above-described measurement over time without performing an additional measurement operation.

The relative relationship can be obtained as appropriate by displaying a graph.

The second step is a step of preparing a measurement wafer and measuring the LO peak intensity and / or the LO mode frequency and / or the TO-LO split width of the measurement wafer.

A measurement wafer having substantially the same physical properties as the reference wafer is used so that the above-mentioned correlation can be applied as accurately as possible. Conversely, it is necessary to perform the first step using a wafer having substantially the same physical properties as the measurement wafer to be measured as a reference wafer.

The third step is a step of obtaining a film ratio from the measurement result measured in the second step based on the relative relationship obtained in the first step.

In the second invention of the present application, the coverage of the natural oxide film on the wafer surface is obtained by using the LO peak intensity.

In the third aspect of the present invention, the coverage of the natural oxide film on the wafer surface is determined by using the TO-LO split width.

In the fourth invention of the present application, the coverage of the natural oxide film on the wafer surface is obtained by using the frequency of the LO mode.

[0057]Measurement example  (First step) Double-sided mirror polished as the first reference wafer
Floating zone silicon wafer (plane orientation (100), resistivity
(Anti-resistance of 1 kΩcm or more, thickness of about 700 μm) was prepared. Next
The second reference wafer has substantially the same floating zone silicon wafer.
I prepared Eha. About 4.8% of the first reference wafer
After washing with hydrofluoric acid, it was washed with water to remove the oxide film on the surface.
Release the first reference wafer from the oxide film into the clean room.
At a certain point,
Peak intensity, TO-
The LO division width D and the frequency of the LO mode were determined. Table of results
It is shown in FIG. In addition, measured LO peak intensity, TO mode
Frequency and LO mode frequency are linked by a curve, without oxide film
The change with time from the time point to the saturation point is shown in the graph of FIG.
Roughly shown. In FIG. 5, the peak intensity of LO mode and
And the frequency are plotted with ○, and the frequency in TO mode is plotted as △
It is plotted with.

Further, the time when the wafer is left for 10.0 days is defined as the saturation time when the oxide film covers the entire surface of the wafer,
The ratio (b / a) of the LO peak intensity (refer to FIG. 5: indicated by b) at the measurement time to the O peak intensity (37) (refer to FIG. 5: indicated by a) was determined as the coating ratio at the measurement time, and the table was obtained. 1

FIG. 6 shows a graph of the correlation between the LO peak intensity and the coating ratio (b / a) of the first reference wafer. Similarly, TO-LO split width D
FIG. 7 shows the correlation between the coating mode and the coating rate, and FIG. 8 (8a) shows the correlation between the LO mode frequency and the coating rate. These correlations may not be a straight line as shown by the curve (8b) in the graph of FIG. 8, but the correlation may be obtained anyway, and the shape of the line of the graph does not matter.

(Second Step) Next, a floating zone wafer substantially identical to the first reference wafer is prepared and used as a measurement wafer.
Using the second reference wafer used in the process, measurement was performed using the above-mentioned Brewster angle incidence method and normal incidence method, and the LO peak intensity, TO-LO split width D, and the frequency of the LO mode were obtained. The results shown were obtained.

TO mode frequency = 1050 cm ⁻¹ LO mode frequency = 1180 cm ⁻¹ TO-LO split width D = 130 cm ⁻¹ LO peak intensity = 30 (arbitrary unit) (Third step) Next, obtained in the second step Based on the results, the coating ratio was determined.

Since the LO peak intensity of this measurement wafer is 30, the coating ratio is 30/37 = 81%. In this measurement example, the coating ratio was calculated as (LO peak intensity at the time of measurement (b) /
Since it is defined by the LO peak intensity (a) at the time of saturation, the coating ratio could be obtained only by calculation without using the graph of FIG.

Next, when the film ratio is determined from the TO-LO split width, the TO-LO split width of the measurement wafer is 130 cm.
^Since it is ^-1 , the film coverage is calculated from FIG. 7 to be about 74%.

Similarly, when the film rate is determined from the LO mode frequency, the LO mode frequency of the measured wafer is 1180.
cm ⁻¹ , and the film ratio was calculated from FIG.
Becomes

As a reference, FIG. 9 shows a graph in which the horizontal axis represents the standing time of the first reference wafer, and the vertical axis represents the above-mentioned coating ratio (b / a). Since the coating ratio obtained from the LO peak intensity of the measurement wafer is 81%, it can be seen from the graph of FIG. 9 that the standing time of the measurement wafer is about 4.3 days.

For reference, the first reference wafer was not washed with hydrofluoric acid but with alkali cleaning (SC-1 cleaning, NH ₄ cleaning).
OH: H ₂ O ₂ : H ₂ O = 1: 1: 5 (vol%), 1
FIG. 5 also shows the measurement results for the case where the sample was left for 60 minutes at 60 to 70 ° C.). In this case, the peak intensity and the frequency in the LO mode are plotted with ●, and the frequency in the TO mode is plotted with ▲.

As can be understood from FIG. 5, in the case of alkali cleaning, both the vibration in the TO mode and the vibration in the LO mode have little change over time in the frequency and intensity, and the TO
The mode frequency and the LO mode frequency are almost equal to the frequency at the stage when the wafer surface is almost entirely covered with the oxide film after the hydrofluoric acid cleaning. Therefore, in the case of alkali cleaning, it is considered that the oxide film is formed on almost the entire surface of the wafer even immediately after the cleaning.

The present invention is not limited to the embodiment described above. For example, the first reference wafer may be a floating silicon wafer instead of a floating zone silicon wafer. The pulled silicon wafer is a silicon wafer processed by performing a series of processing steps on a wafer cut from a silicon single crystal manufactured by a pulling method (so-called Czochralski method).

The finish state of the surface of the first reference wafer may be a state in which etching processing or mirror processing has been performed.

The second reference wafer does not need to be the same silicon wafer as the first reference wafer, but may be any silicon wafer satisfying the following conditions.

The second reference wafer has the same bulk characteristics as the first reference wafer. That is, when the first reference wafer is a raised silicon wafer, the second reference wafer is a raised silicon wafer having substantially the same dissolved oxygen concentration, and when the first reference wafer is a floating zone silicon wafer, The second reference wafer is also a floating zone silicon wafer. In addition, the surface of the second reference wafer is in the same finished state as the surface of the first reference wafer, or has been subjected to double-sided mirror processing.

In the above-mentioned Brewster angle incidence method, the incident angle of the parallel polarized light is set to the Brewster angle. However, the incident angle may be an arbitrary angle other than the Brewster angle.
In this case, since multiple reflections occur on the surface and inside of the wafer, an error may occur in the measured LO peak intensity.

Further, the thicknesses of the first reference wafer and the second reference wafer may be different. In this case, the thickness is corrected by the above-described equation (1).

Further, in the above-described embodiment, the second reference wafer is prepared and measured each time the Brewster angle incidence method and the vertical incidence method are measured in the first step, but the present invention is not limited to this. Not something. For example, as storing the transmitted light intensity I at a time that measurement After measuring the _B of the second reference wafer, omit the measurement of the transmitted light intensity of the second and subsequent second reference wafer, the instead The stored measurement values of a two-reference wafer may be used. Or by measuring the transmitted light intensity I _A of the first reference wafer immediately after washing hydrofluoric acid is stored, may be used the measured value instead of the transmitted light intensity I _B of the second reference wafer. In such a case, only the first reference wafer is required, and the second reference wafer is not required.

When only the measurement of the LO mode polar lattice vibration of the oxide film is required, as in the second or fourth invention of the present application, the measurement by the perpendicular incidence method can be omitted. Further, the present invention is not limited to the measurement of the coverage of a natural oxide film by natural oxidation, but can be applied to the measurement of the coverage of an oxide film by oxidation other than natural oxidation. However, it is preferable that the thickness of the oxide film in this case is substantially the same as the thickness of the natural oxide film (extremely thin).

[0076]

As described above, according to the present invention, the coating ratio of the oxide film on the wafer surface can be obtained by using the LO peak intensity, the TO-LO split width, and the LO mode frequency of the oxide film. In addition, the coating ratio can be measured simply and accurately.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram for explaining parallel polarization Brewster angle incident infrared transmission spectroscopy in a coating ratio measurement according to a preferred embodiment of the present invention.

FIG. 2 is an explanatory diagram for explaining normal incidence infrared transmission spectroscopy.

FIG. 3 is a graph showing a difference absorbance spectrum of an oxide film.

FIG. 4 is a graph showing a change over time in a difference absorbance spectrum of an oxide film on a surface of a wafer.

FIG. 5 is a graph showing the change over time in the frequency of the TO mode and the LO mode of the wafer and the LO peak intensity.

FIG. 6 is a graph showing the correlation between the LO peak intensity and the coating ratio.

FIG. 7 is a graph showing the correlation between the TO-LO split width and the coating ratio.

FIG. 8 is a graph showing the correlation between the LO mode frequency and the coating ratio.

FIG. 9 is a graph showing the change over time in the coating ratio.

[Explanation of symbols]

 B incident angle I incident light 1 first reference wafer 2 second reference wafer 10 oxide film

[Table 1]

Claims (4)

(57) [Claims] A step of measuring a peak intensity of a longitudinal wave optical mode (LO) of a long-wavelength polar lattice vibration of the oxide film when the surface of the reference wafer is saturated with the oxide film; Measuring the peak intensity of the longitudinal-wave optical mode (LO) of the long-wavelength polar lattice vibration of the oxide film on the surface of the sample, and obtaining the ratio of the peak intensity of the measured wafer to the peak intensity at the time of the saturation as the coating ratio. The method for measuring the coating rate of an oxide film on the surface of a silicon wafer containing silicon. 2. The oxide film coverage and the peak intensity of the longitudinal wave optical mode (LO) of the long-wavelength polar lattice vibration of the oxide film between the time when no oxide film is present on the surface of the reference wafer and the time when the oxide film is saturated. And the step of measuring the peak intensity of the longitudinal-wave optical mode (LO) of the long-wavelength polar lattice vibration of the oxide film of the measurement wafer at a predetermined point in time. A method for measuring a coating rate on a silicon wafer surface, comprising a step of obtaining a coating rate at a measurement time point of a measurement wafer based on the measurement. 3. The ratio of oxide film coverage from the time when no oxide film is present on the surface of the reference wafer to the time when the oxide film is saturated, and the transverse-wave optical mode (TO) and longitudinal-wave optical mode of long-wavelength polar lattice vibration of the oxide film. A step of previously determining the relative relationship between the frequency of the mode (LO) and the splitting width, and a transverse optical mode (TO) and a longitudinal optical mode (LO) of the long-wavelength polar lattice vibration of the oxide film of the measurement wafer at a predetermined time. A step of measuring a split width of the frequency of the silicon wafer, and a step of obtaining a coating ratio of the measurement wafer at the time of measurement based on the above-mentioned relative relationship from the measured value. 4. A coating rate of an oxide film from a time when no oxide film is present on a surface of a reference wafer to a time when the oxide film is saturated, and a longitudinal wave optical mode (LO) frequency of a long-wavelength polar lattice vibration of the oxide film. And a step of measuring the longitudinal wave optical mode (LO) frequency of the long-wavelength polar lattice vibration of the oxide film on the measurement wafer at a predetermined time, and the above-described relative relationship from the measured value. A method for measuring the coating rate on the surface of a silicon wafer, comprising the step of obtaining the coating rate at the time of measurement of the measurement wafer based on the above.