JP5630426B2 - Cleanliness evaluation method for vapor phase growth apparatus - Google Patents

Description

  The present invention relates to a method for evaluating the cleanliness of a vapor phase growth apparatus used for producing an epitaxial wafer.

  As a method for detecting metal impurities in a silicon wafer, there is a wafer lifetime (hereinafter sometimes referred to as WLT for short) method (for example, see Non-Patent Document 1). As a typical method of the WLT method, there is a microwave photoconductive decay method minority carrier lifetime method (hereinafter, abbreviated as μPCD method). In this method, for example, light is applied to a sample (substrate), and the lifetime of the minority carriers generated is detected by a change in the reflectance of the microwave, thereby evaluating the metal impurities in the sample.

  When the metal is taken into the wafer, this WLT value becomes small. Therefore, the metal contamination in the heat treatment furnace can be managed by measuring and evaluating the WLT value of the wafer subjected to heat treatment or vapor phase growth. . That is, a wafer for contamination control is prepared and heat-treated in a heat treatment furnace used in an actual process, and the WLT value of the wafer after the heat treatment is measured to determine whether or not the heat treatment furnace is contaminated with metal impurities. be able to.

"Issues with Silicon Crystal / Wafer Technology" (Realize Inc., issued on January 31, 1994) pp. 265-269

Since this WLT method is simple but can detect even a small amount of contamination, it is widely used for contamination management of heat treatment furnaces and contamination management of a vapor phase growth apparatus (epi deposition apparatus). In particular, in the case of a vapor phase growth apparatus, a silicon wafer having a P ⁻ or N ⁻ conductivity type is prepared, and a silicon epitaxial layer is formed on the silicon wafer using the vapor phase growth apparatus to be evaluated. The cleanliness of the vapor phase growth apparatus can be evaluated by measuring the WLT value of the epitaxial wafer by the above-described μPCD method.

  However, even if the cleanliness of the vapor phase growth apparatus is evaluated by the WLT method using an epitaxial wafer for lifetime monitoring (hereinafter sometimes referred to as a monitor wafer), the actual cleanliness is not reflected. There is a problem that the cleanliness cannot be evaluated with high sensitivity.

  This invention is made | formed in view of the said problem, Comprising: It aims at providing the cleanliness evaluation method which can evaluate the cleanliness of a vapor phase growth apparatus with high sensitivity.

  The present invention has been made to solve the above problems, and is a method for evaluating the cleanliness of a vapor phase growth apparatus. A silicon wafer having a thickness of 1000 μm or more and 2000 μm or less is prepared, and the vapor phase growth apparatus is used. And producing a monitor wafer having a silicon epitaxial layer grown on the silicon wafer, measuring the lifetime value of the monitor wafer, and determining the cleanliness of the vapor phase growth apparatus from the measured lifetime value of the monitor wafer. Provided is a method for evaluating the cleanliness of a vapor phase growth apparatus characterized by evaluating.

  In this way, by using a silicon wafer having a thickness of 1000 μm or more and 2000 μm or less as the substrate of the monitor wafer, the ratio of surface recombination among the recombination of minority carriers in the monitor wafer can be reduced. As a result, the detection upper limit value of the wafer lifetime can be increased, so that the cleanliness of the vapor phase growth apparatus can be evaluated with high sensitivity.

  The present invention also provides a method for producing a silicon epitaxial wafer, characterized in that a silicon epitaxial wafer is produced using the vapor phase growth apparatus evaluated by the method for evaluating the cleanliness of the vapor phase growth apparatus.

  If a silicon epitaxial wafer is manufactured in this way, a high-quality epitaxial wafer with little contamination can be manufactured with a high yield.

  According to the method for evaluating the cleanliness of a vapor phase growth apparatus according to the present invention, the upper limit of detection of the wafer lifetime can be increased, and the cleanliness of the vapor phase growth apparatus can be evaluated with high sensitivity. Therefore, even if the impurity contamination level is low, it can be accurately evaluated with high sensitivity. In addition, by manufacturing a silicon epitaxial wafer with a vapor phase growth apparatus evaluated that there is no contamination by the method according to the present invention, a high-quality silicon epitaxial wafer with little contamination can be manufactured with a high yield.

It is a flowchart which shows the outline of the cleanliness evaluation method of the vapor phase growth apparatus of this invention. It is a cross-sectional schematic diagram of the epitaxial wafer (monitor wafer) which has a two-layer structure. It is a graph which shows the wafer lifetime value of the monitor wafer of the Example to which this invention is applied, and a comparative example.

  Hereinafter, the present invention will be described in detail, but the present invention is not limited thereto.

  As described above, there has been a problem that the cleanliness cannot be evaluated with high sensitivity even if the cleanliness of the vapor phase growth apparatus is evaluated by the WLT method using a monitor wafer as in the past.

  The present inventor has repeatedly studied the above problems. In the monitor wafer, when minority carriers diffuse and reach the surface, surface recombination occurs there. If the ratio of surface recombination increases, an accurate wafer lifetime cannot be measured. Then, in order to reduce the ratio of surface recombination, it was thought that the thickness of the wafer should be increased, and the present invention has been achieved.

  Hereinafter, embodiments of the present invention will be specifically described with reference to the drawings, but the present invention is not limited thereto.

  FIG. 1 shows an outline of a method for evaluating the cleanliness of a vapor phase growth apparatus according to the present invention.

  In the method for evaluating the cleanliness of the vapor phase growth apparatus according to the present invention, first, as shown in FIG. 1A, a silicon wafer having a thickness of 1000 μm or more and 2000 μm or less is prepared. The detailed reason for setting the thickness in such a range will be described later. Further, the diameter, surface orientation, conductivity type, resistivity, etc. of the silicon wafer prepared here are not particularly limited.

  Next, as shown in FIG. 1B, a monitor wafer in which a silicon epitaxial layer is grown on the prepared silicon wafer is manufactured using a vapor phase growth apparatus to be evaluated. FIG. 2 shows the structure of a two-layer monitor wafer (epitaxial wafer). The monitor wafer 10 includes a silicon wafer 11 and a silicon epitaxial layer 12. The diameter, plane orientation, conductivity type, resistivity and the like of the silicon epitaxial layer 12 are not particularly limited.

  Next, as shown in FIG. 1C, the lifetime value of the produced monitor wafer is measured. The method for measuring the wafer lifetime value can be a known method, and is not particularly limited, but is preferably performed by the μPCD method, which allows easy measurement.

  Next, as shown in FIG. 1D, the cleanliness of the vapor phase growth apparatus to be evaluated is evaluated from the measured lifetime value of the monitor wafer. When impurities, particularly metal, are taken into the silicon epitaxial layer of the monitor wafer, the lifetime value is reduced. Therefore, if the monitor wafer is manufactured using the vapor phase growth apparatus to be evaluated and the monitor wafer is contaminated and the lifetime value is small, it is evaluated that the cleanness of the vapor phase growth apparatus is low. it can. Conversely, if the decrease in lifetime value is small, it can be evaluated that the contamination of the monitor wafer originating from the vapor phase growth apparatus is small, and the cleanliness of the vapor phase growth apparatus can be evaluated to be high.

  Hereinafter, the reason why the thickness of the thick silicon wafer prepared in the present invention is set to 1000 μm or more and 2000 μm or less will be specifically described.

  Wafer lifetime is strongly influenced by the thickness of the wafer being measured. The reason is due to carrier recombination at the wafer surface. As described above, in the μPCD method, minority carriers are excited by light. The minority carriers diffuse in the wafer and eventually recombine with the lifetime corresponding to the amount of contamination of the wafer and disappear. However, if the thickness of the wafer is smaller than the minority carrier diffusion length, many of the excited minority carriers reach the surface before recombination inside the wafer. In this case, carrier recombination occurs at the surface according to a certain surface recombination rate regardless of the contamination level of the wafer.

  The structure of a conventional lifetime monitoring epitaxial wafer (monitor wafer) is a 200 mm or 300 mm diameter wafer having a thickness of about 700 μm, and a silicon epitaxial layer of about 10 μm is formed on a silicon wafer as a substrate. Is. In this case, since the thickness of the wafer is smaller than the diffusion length of minority carriers, the excited minority carriers easily reach the surface of the wafer. Therefore, a lot of carrier recombination components on the wafer surface are contained, and the required lifetime information inside the monitor wafer (bulk portion) is diluted. This problem is particularly remarkable for metal impurity elements (for example, Fe and Cu) having a high diffusion rate in silicon.

The diffusion length of minority carriers in a silicon wafer is expressed by the following formula.
L = √ (D × τ)
L: minority carrier diffusion length D: minority carrier diffusion coefficient τ: lifetime

The diffusion coefficient of minority carriers in silicon is approximately 12.5 cm ² / sec for holes and approximately 35.5 cm ² / sec for electrons. When the lifetime is 1000 μsec, the diffusion length is 1000 to 2000 μm. If the thickness of the monitor wafer is equal to or longer than this diffusion length, the minority carrier component diffused on the wafer surface is reduced and the number of carriers recombining with the surface is also reduced. As a result, the proportion of carriers recombined inside the monitor wafer increases, and contamination can be detected with high sensitivity.

  Moreover, the thickness of the silicon wafer used as a board | substrate shall be 2000 micrometers or less. The thicker the silicon wafer that is the substrate, the smaller the rate of surface recombination and the greater the information component regarding contamination. However, it is economically wasteful if the thickness is increased. Further, the wafer transfer apparatus used in the vapor phase growth apparatus cannot be transferred if the wafer is too thick. For these reasons, the thickness of the silicon wafer serving as the substrate is set to 2000 μm or less, which is the same as the carrier diffusion length.

  For the reasons described above, in the present invention, in order to reduce the contribution due to surface recombination in the wafer lifetime of the monitor wafer, the thickness of the silicon wafer serving as the substrate is increased, specifically, 1000 μm or more and 2000 μm or less.

  On the other hand, the thickness of the silicon epitaxial layer formed on the silicon wafer to be the substrate is not particularly limited and can be set as necessary. For example, even if it is a thin silicon epitaxial layer of about 1 to 20 μm, typically about 10 μm as in the prior art, the thickness of the silicon wafer serving as the substrate is sufficiently thick, such as 1000 μm or more and 2000 μm or less. Can be sufficient to reduce the effect of carrier surface recombination.

  Furthermore, a silicon epitaxial wafer can be manufactured using the vapor phase growth apparatus evaluated by the cleanliness evaluation method of the vapor phase growth apparatus of the present invention as described above. By manufacturing a silicon epitaxial wafer with a vapor phase growth apparatus that has been evaluated to be free of contamination by the cleanliness evaluation method according to the present invention, a high-quality silicon epitaxial wafer with little contamination can be manufactured with a high yield. In particular, a silicon epitaxial wafer can be manufactured based on an evaluation that sufficiently reflects the contamination of a metal having a high diffusion rate in silicon.

  EXAMPLES Hereinafter, although an Example and a comparative example are given and this invention is demonstrated further more concretely, these do not limit this invention.

(Example)
First, a large number of P-type silicon wafers having a diameter of 200 mm, a resistivity of 10 Ωcm, and a thickness of 1500 μm were prepared as silicon wafers.

  Next, the vapor phase growth apparatus to be evaluated was opened to the atmosphere, and so-called maintenance work was performed. In general, when maintenance work is performed, the inside of the reactor of the vapor phase growth apparatus is slightly contaminated, and it takes several days after the maintenance until the contamination source is almost removed and the influence of the contamination is almost eliminated (contamination is withered).

  In this way, a vapor phase growth apparatus that has undergone maintenance work is prepared. Using this apparatus, a P-type silicon epitaxial layer having a resistivity of 10 Ωcm is deposited on the silicon wafer prepared as described above by 10 μm. Monitor wafers were produced once a day. While the monitor wafer was not manufactured, the vapor phase growth apparatus was heated in the same sequence as the growth of the silicon epitaxial layer to continue the process of removing the contamination source.

  The monitor wafer thus produced was subjected to a surface treatment by chemical passivation, and a wafer lifetime value was measured using a wafer lifetime measuring apparatus by a μPCD method. FIG. 3 shows the daily transition of the wafer lifetime value in the example. In the monitor wafer produced immediately after the maintenance, the wafer lifetime value is about 800 μsec, but the wafer lifetime value increased with the passage of days thereafter, and increased to about 4500 μsec on the 10th day.

(Comparative example)
Next, as a comparative example, an example is shown in which a monitor wafer having a thin silicon wafer thickness (less than 1000 μm) as in the prior art is manufactured.

  First, a large number of P-type silicon wafers having a diameter of 200 mm, a resistivity of 10 Ωcm, and a thickness of 725 μm were prepared as silicon wafers.

  A vapor phase growth apparatus in which the same maintenance work as that of the example was performed was prepared, and a monitor wafer of a comparative example was produced when the elapsed time after the maintenance was the same as the production of the monitor wafer of the example. The condition for producing the monitor wafer is that a P-type silicon epitaxial layer having a resistivity of 10 Ωcm is deposited on the silicon wafer by 10 μm.

  The monitor wafer thus produced was subjected to surface treatment by chemical passivation, and the wafer lifetime value was measured using a wafer lifetime apparatus by the μPCD method. In FIG. 3, the daily transition of the wafer lifetime value in the comparative example is shown. In the monitor wafer manufactured immediately after the maintenance, the wafer lifetime value is about 800 μsec. However, the wafer lifetime increased with the passage of days thereafter, and settled to about 1800 μsec on the second day. After that, there was no big change until the 10th day.

  Compared to the examples, it appears that the contamination level does not change after the second day in the comparative example. This means that the comparative example has not been able to detect a reduction in the contamination level after the second day. That is, it was confirmed that the example has higher impurity detection capability.

  The present invention is not limited to the above embodiment. The above-described embodiment is an exemplification, and the present invention has substantially the same configuration as the technical idea described in the claims of the present invention, and any device that exhibits the same function and effect is the present invention. It is included in the technical scope of the invention.

10: Monitor wafer (epitaxial wafer),
11 ... silicon wafer, 12 ... silicon epitaxial layer.

Claims (2)

A method for evaluating the cleanliness of a vapor phase growth apparatus,
Prepare a silicon wafer having a thickness of 1000 μm to 2000 μm (excluding 1000 μm)
Using the vapor phase growth apparatus, producing a monitor wafer having a silicon epitaxial layer grown on the silicon wafer,
Measure the lifetime value of the monitor wafer,
A cleanliness evaluation method for a vapor phase growth apparatus, wherein the cleanliness of the vapor phase growth apparatus is evaluated from the measured lifetime value of the monitor wafer.   A method for producing a silicon epitaxial wafer, comprising producing a silicon epitaxial wafer using the vapor phase growth apparatus evaluated by the method for evaluating the cleanliness of the vapor phase growth apparatus according to claim 1.

Images