Method for judging source material allowance of MBE source furnace based on epitaxial wafer surface

Abstract

The invention discloses a method for judging the source material allowance of an MBE source furnace based on the surface of an epitaxial wafer, and the method comprises the steps: selecting proper points on the surface of the epitaxial wafer, measuring the cycle, the half-peak width and the mismatch value of the surface of the epitaxial wafer through XRD, and respectively calculating the cycle, the half-peak width and the mismatch value of the epitaxial wafer through standard deviation; whether the source materials of the source furnace are insufficient or not is judged by designing a control group, an experience control group and an evidence group and comparing standard deviation ratios of the groups, and the allowance of the source materials of the molecular beam epitaxy source furnace is judged more rigorously. The method allows the sufficiency of the source material of the molecular beam epitaxy source furnace to be monitored in real time, and stable epitaxial wafer growth is ensured through XRD and data analysis. Compared with a traditional statistical method, the method is more accurate and scientific, resource waste is avoided, and the quality of the epitaxial material is improved.

Description

Technical Field

[0001] The present application relates to the field of semiconductor technology, and in particular to a method for judging the remaining amount of source material in an MBE source furnace based on the surface of an epitaxial wafer. Background Art

[0002] Molecular beam epitaxy (MBE) is a high-precision thin film growth technology that is often used to prepare semiconductor materials and other nanostructures. The source furnace is the core component of the molecular beam epitaxy equipment, where the source material is heated to a high temperature and evaporated from the source furnace to produce a molecular beam that is deposited on the surface of the epitaxial wafer and gradually grows into a thin film or crystal.

[0003] The requirements for precision in the molecular beam epitaxy growth process are extremely high. When the source furnace source material is sufficient, the temperature of the source furnace is controlled to control a stable evaporation rate to ensure high precision requirements; when the source furnace source material is insufficient, it will seriously affect the surface uniformity, half-peak and mismatch size of the epitaxial material, thereby reducing the quality of the epitaxial material. Therefore, it is necessary to accurately judge the source furnace source material surplus, open the cavity for maintenance in time, and replenish the source material. The existing source furnace source material surplus judgment method usually uses the statistical growth days to compare with the previous data that meets the growth conditions to roughly estimate the surplus. This method is not systematic and is relatively single. It is easily affected by uncertain factors, so it is often inaccurate and will affect normal growth. Therefore, it is urgent to find a method that can accurately judge whether the source furnace source material is insufficient. Summary of the invention

[0004] The present invention provides a method for judging the remaining amount of source material in an MBE source furnace based on the surface of an epitaxial wafer, comprising the following steps:

[0005] (1) Grow an epitaxial wafer under the condition of sufficient source material in the source furnace, which is recorded as the control group; when the growth thickness is close to 90-95% of the total growth thickness corresponding to the estimated source material weight, grow an epitaxial wafer under the same growth conditions as the control group as the empirical control group;

[0006] (2) uniformly selecting n points on the surface of the epitaxial wafer, using X-ray diffraction technology to measure and record the corresponding values ​​of the period, half-peak width, and mismatch on the surface of the epitaxial wafer;

[0007] (3) Using the standard deviation formula, calculate the standard deviation values ​​of the period, half-peak width and mismatch of the control group and the empirical control group, respectively, record them as the control group period standard deviation A, half-peak width standard deviation B, mismatch standard deviation C and the empirical control group period standard deviation A', half-peak width standard deviation B', mismatch standard deviation C', and calculate the ratio Denoted as n A '、n B '、n C ', record the standard deviation ratio and keep it on file;

[0008] The standard deviation formula is as follows: Among them, σ is the standard deviation, which represents the degree of dispersion of the data and the degree of dispersion of the data distribution; X i : represents the specific measurement value of the period, half-peak width or mismatch of each measurement point of the data; μ: represents the average value of the period, half-peak width or mismatch of all measurement points; n: represents the number of data measurement points, that is, X i The number of

[0009] For specific production, whether it is adding materials or growing, the operation on the same growth material is repeatable, and the standard deviation data of the control group and the empirical control group are highly consistent, so their standard deviation data can be reused;

[0010] (4) using the methods of steps (2) and (3) to obtain the standard deviations of the period, half-peak width and mismatch of the epitaxial wafer in the subsequent growth process, which are respectively recorded as the period standard deviation A1, the half-peak width standard deviation B1, and the mismatch standard deviation C1; and calculating the ratio of the period standard deviation A1, the half-peak width standard deviation B1, and the mismatch standard deviation C1 to the control group period standard deviation A, the half-peak width standard deviation B, and the mismatch standard deviation C Denoted as n A1 、n B1 、n C1 ;

[0011] (5) When the ratio n A1 、n B1 、n C1 Less than n A '、n B '、n C 'When the supply is 90% of the supply, it is judged that the source furnace source material is sufficient.

[0012] Furthermore, when the ratio n A1 、n B1 、n C1 Greater than n A '、n B '、n C When the growth rate is 90%, the following steps are performed to conduct growth confirmation experiments:

[0013] (6) Stop the sample holder from rotating, and ensure that the remaining growth conditions are consistent with those of the control group. Grow an epitaxial wafer thickened on the basis of the control group as the growth support group. Use the method in step (2) to measure the period, half-peak width and mismatch of the epitaxial wafer surface of the growth support group, and use the standard deviation formula in step (3) to calculate the period standard deviation A1', half-peak width standard deviation B1' and mismatch standard deviation C1' of the growth support group. Calculate the ratio of the period standard deviation A1', half-peak width standard deviation B1' and mismatch standard deviation C1' of the growth support group to those of the control groups A, B and C. Denoted as n A1 '、n B1 '、n C1 ';

[0014] When n A1 '、n B1 '、n C1 ' is greater than or equal to n A '、n B '、n C ', it is judged that the current state is that the source furnace is insufficient in source material, and the cavity is opened for maintenance to replenish the source material; otherwise, the above steps (4)-(6) are repeated.

[0015] Specifically, in step (2), n≧20.

[0016] In the subsequent growth process, when the source material is sufficient, the prepared epitaxial wafers can be randomly sampled and measured; when the remaining amount of the source material is close to the remaining amount of the empirical control group, each prepared epitaxial wafer is measured.

[0017] The empirical control group can also be one of the multiple epitaxial wafers grown under the same growth conditions as the control group when the growth thickness is close to 90-95% of the estimated total growth thickness corresponding to the source material weight. The epitaxial wafer whose slope of the curve of the period standard deviation A', the half-width standard deviation B' and the mismatch standard deviation C' changing with the number of grown wafers shows an inflection point (or suddenly increases) is selected from the multiple epitaxial wafers as the empirical control group.

[0018] The thickness of the growth material on the surface of the epitaxial wafer in the growth support group is 1.5-3 times the thickness of the growth material on the surface of the epitaxial wafer in the control group.

[0019] The beneficial effects of the present invention include:

[0020] The present invention provides a method for judging the surplus of MBE source furnace source material based on the surface of an epitaxial wafer. The appropriate point on the surface of the epitaxial wafer is selected to measure the period, half-peak width and mismatch value of the surface of the epitaxial wafer by XRD, and the period, half-peak width and mismatch value of the epitaxial wafer are calculated respectively by standard deviation. By designing a control group, an empirical control group and a supporting group and comparing the standard deviation ratio of each group, it is judged whether the source furnace source material is insufficient, and the surplus of the molecular beam epitaxy source furnace source material is judged more rigorously. This method allows real-time monitoring of the adequacy of the molecular beam epitaxy source furnace source material, and ensures stable epitaxial wafer growth by means of XRD and data analysis. Compared with traditional statistical methods, this method is more accurate and scientific, helps to avoid waste of resources and improve the quality of epitaxial materials. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 The present invention is a flow chart for judging the remaining amount of source material in the MBE source furnace for growing superlattice materials based on the surface of the epitaxial material. DETAILED DESCRIPTION

[0022] The present invention is further illustrated and described below in conjunction with embodiments, but the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the present invention and embodiments, all other inventions and embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] Unless otherwise specified, the experimental methods used in the following examples are conventional methods.

[0024] Unless otherwise specified, the materials and reagents used in the following examples can be obtained from commercial sources.

[0025] Embodiment 1: A method for judging the residual amount of source material in an MBE source furnace based on the surface of an epitaxial wafer. The method of this embodiment can be used to judge the residual amount of Ga and Al in an MBE source furnace, and specifically comprises the following steps:

[0026] (1) When the source material has just been replenished and the source furnace has sufficient source material, a 2000nm GaAs / AlGaAs epitaxial wafer is grown on a GaAs substrate for the first time, which is recorded as the control group;

[0027] In this step, the sample holder preparation, temperature, flow rate and other growth parameters are precisely set according to routine operations. The Ga rate is 0.5 ML / s, the Al rate is 0.5 ML / s, the As beam current is 1.5E-6 Torr, and the substrate temperature is 580°C.

[0028] The total growth thickness T corresponding to the weight of the source material is estimated. When the growth thickness is close to 90-95% of T, another 2000nm GaAs / AlGaAs epitaxial wafer is grown under the same growth conditions as the control group, which is recorded as the empirical control group.

[0029] (2) After the growth is completed, the period, half-peak width and mismatch of the epitaxial wafer surface are measured using X-ray diffraction (XRD) technology. 20 measurement points are selected, and the measurement points are evenly distributed in different areas of the epitaxial wafer to ensure accuracy.

[0030] (3) Use the standard deviation formula to calculate the standard deviation of the period, half-peak width and mismatch. The standard deviation of the period, half-peak width and mismatch of the control group when the source furnace material is sufficient are recorded as A (standard deviation of the period), B (standard deviation of the half-peak width) and C (standard deviation of the mismatch); the standard deviation of the period, half-peak width and mismatch of the empirical control group under the critical state when the source furnace material is insufficient are recorded as A' (standard deviation of the period), B' (standard deviation of the half-peak width) and C' (standard deviation of the mismatch). The standard deviation formula is as follows:

[0031] Among them, σ is the standard deviation, which represents the degree of dispersion of the data and the degree of dispersion of the data distribution; X i : represents the specific measured value of the period, half-peak width or mismatch of each measurement point; μ: represents the average value of the period, half-peak width or mismatch of all measurement points; n: represents the number of measurement points, i.e. X i The number of .

[0032] (4) Calculate the ratio of the standard deviation A', standard deviation B' of half-peak width and standard deviation C' of the period of the empirical control group when the source furnace is insufficient to the control group A, B and C when the source furnace is sufficient. Denoted as n A '、n B '、n C ', record the standard deviation ratio and keep it on file.

[0033] (5) The period, half-peak width and mismatch of the epitaxial wafer surface during the subsequent growth process are measured using the method of step (2), and the standard deviation formula of step (3) is used to calculate the values ​​of the period standard deviation A1, the half-peak width standard deviation B1, and the mismatch standard deviation C1. The ratio of the period standard deviation A1, the half-peak width standard deviation B1, and the mismatch standard deviation C1 to the control groups A, B, and C is Denoted as n A1 、n B1 、n C1 In the subsequent growth process, when the source material is sufficient, the prepared epitaxial wafers can be randomly sampled and measured; when the remaining amount of the source material is close to the remaining amount of the empirical control group, each prepared epitaxial wafer is measured.

[0034] (6) Analyze the ratio. When the ratio n A1 、n B1 、n C1 Less than n A '、n B '、n C ', it is judged that the source furnace source material is sufficient. A1 、n B1 、n C1 Greater than n A '、n B '、n C When the growth rate reaches 90%, the growth confirmation experiment is finally verified.

[0035] (7) Stop the sample holder from rotating, and keep the other growth conditions the same as those of the control group, and grow a 4000nm GaAs / AlGaAs epitaxial wafer as the growth evidence group. Use the method of step (2) to measure the period, half-peak width and mismatch of the epitaxial wafer surface of the growth evidence group, and use the standard deviation formula of step (3) to calculate the standard deviation A1' of the period, the standard deviation B1' of the half-peak width and the standard deviation C1' of the mismatch of the growth evidence group. The ratios of the standard deviation A1' of the period, the standard deviation B1' of the half-peak width and the standard deviation C1' of the mismatch to those of the control groups A, B and C are: Denoted as n A1 '、n B1 '、n C1 ';

[0036] When n A1 '、n B1 '、n C1 ' is greater than or equal to n A '、n B '、n C ', it is determined that the current state is that the source furnace is insufficient in source material, and the cavity is opened for maintenance to replenish the source material; otherwise, repeat steps (5)-(7).

[0037] Embodiment 2: A method for judging the remaining amount of MBE source furnace source material based on the surface of the epitaxial wafer

[0038] The method of this embodiment can be used to determine the residual amount of Ga and Al in the MBE source furnace, and specifically includes the following steps:

[0039] (1) Under the initial condition that the source furnace source material is sufficient, a 2000nm GaAs / AlGaAs epitaxial wafer is grown on a GaAs substrate for the first time, which is recorded as the control group;

[0040] In this step, the sample holder preparation, temperature, flow rate and other growth parameters are precisely set according to routine operations: Ga beam rate is 0.5 ML / s, Al beam rate is 0.5 ML / s, As beam is 1.5E-6 Torr, and substrate temperature is 580°C.

[0041] The total growth thickness T corresponding to the weight of the source material is estimated. When the growth thickness is close to 90-95% of T, several more 2000nm GaAs / AlGaAs epitaxial wafers are grown under the same growth conditions as the control group to screen the empirical control group.

[0042] (2) After the growth is completed, the period, half-peak width and mismatch of the epitaxial wafer surface are measured using X-ray diffraction (XRD) technology. At least 20 measurement points are selected, and the measurement points are evenly distributed in different areas of the epitaxial wafer to ensure accuracy.

[0043] (3) Use the standard deviation formula to calculate the standard deviation of the period, half-peak width and mismatch. The standard deviation of the period, half-peak width and mismatch of the control group when the source furnace material is sufficient are recorded as A (standard deviation of the period), B (standard deviation of the half-peak width) and C (standard deviation of the mismatch); the standard deviation of the period, half-peak width and mismatch of the empirical control group under the critical state when the source furnace material is insufficient are recorded as A' (standard deviation of the period), B' (standard deviation of the half-peak width) and C' (standard deviation of the mismatch). The standard deviation formula is as follows:

[0044] Among them, σ is the standard deviation, which represents the degree of dispersion of the data and the degree of dispersion of the data distribution; X i : represents the specific measured value of the period, half-peak width or mismatch of each measurement point; μ: represents the average value of the period, half-peak width or mismatch of all measurement points; n: represents the number of measurement points, i.e. X i The number of .

[0045] The epitaxial wafer where the slope of the curves of the period standard deviation A', the half-peak width standard deviation B' and the mismatch standard deviation C' with the number of grown wafers has an inflection point (or suddenly increases) is selected as the final empirical control group.

[0046] (4) Calculate the ratio of the standard deviation A', standard deviation B' of half-peak width and standard deviation C' of the period of the empirical control group when the source furnace is insufficient to the control group A, B and C when the source furnace is sufficient. Denoted as n A '、n B '、n C ', record the standard deviation ratio and keep it on file.

[0047] (5) The period, half-peak width and mismatch of the epitaxial wafer surface during the subsequent growth process are measured using the method of step (2), and the standard deviation formula of step (3) is used to calculate the values ​​of the period standard deviation A1, the half-peak width standard deviation B1, and the mismatch standard deviation C1. The ratio of the period standard deviation A1, the half-peak width standard deviation B1, and the mismatch standard deviation C1 to the control groups A, B, and C is Denoted as n A1 、n B1 、n C1 In the subsequent growth process, when the source material is sufficient, the prepared epitaxial wafers can be randomly sampled and measured; when the remaining amount of the source material is close to the remaining amount of the empirical control group, each prepared epitaxial wafer is measured.

[0048] (6) Analyze the ratio. When the ratio n A1 、n B1 、n C1 Less than n A '、n B '、n C', it is judged that the source furnace source material is sufficient. A1 、n B1 、n C1 Greater than n A '、n B '、n C When the growth rate reaches 90%, the growth confirmation experiment is finally verified.

[0049] (7) Stop the sample holder from rotating, and keep the other growth conditions the same as those of the control group, and grow a 4000nm GaAs / AlGaAs epitaxial wafer as the growth evidence group. Use the method of step (2) to measure the period, half-peak width and mismatch of the epitaxial wafer surface of the growth evidence group, and use the standard deviation formula of step (3) to calculate the standard deviation A1' of the period, the standard deviation B1' of the half-peak width and the standard deviation C1' of the mismatch of the growth evidence group. The ratios of the standard deviation A1' of the period, the standard deviation B1' of the half-peak width and the standard deviation C1' of the mismatch to those of the control groups A, B and C are: Denoted as n A1 '、n B1 '、n C1 ';

[0050] When n A1 '、n B1 '、n C1 ' is greater than or equal to n A '、n B '、n C ', it is determined that the current state is that the source furnace is insufficient in source material, and the cavity is opened for maintenance to replenish the source material; otherwise, repeat steps (5)-(7).

Claims

1. A method for judging the remaining amount of MBE source furnace source material based on the surface of an epitaxial wafer, characterized in that: The steps include: (1) Grow an epitaxial wafer under the condition of sufficient source material in the source furnace, which is recorded as the control group; when the growth thickness is close to 90-95% of the total growth thickness corresponding to the estimated source material weight, grow an epitaxial wafer under the same growth conditions as the control group as the empirical control group; (2) uniformly selecting n points on the surface of the epitaxial wafer, using X-ray diffraction technology to measure and record the corresponding values ​​of the period, half-peak width, and mismatch on the surface of the epitaxial wafer; (3) Using the standard deviation formula, calculate the standard deviation values ​​of the period, half-peak width and mismatch of the control group and the empirical control group, respectively, record them as the control group period standard deviation A, half-peak width standard deviation B, mismatch standard deviation C and the empirical control group period standard deviation A', half-peak width standard deviation B', mismatch standard deviation C', and calculate the ratio Denoted as n A '、n B '、n C ', record the standard deviation ratio and keep it on file; The standard deviation formula is as follows: Among them, σ is the standard deviation, which represents the degree of dispersion of the data and the degree of dispersion of the data distribution; X i : represents the specific measurement value of the period, half-peak width or mismatch of each measurement point of the data; μ: represents the average value of the period, half-peak width or mismatch of all measurement points; n: represents the number of data measurement points, that is, X i The number of (4) using the methods of steps (2) and (3) to obtain the standard deviations of the period, half-peak width and mismatch of the epitaxial wafer in the subsequent growth process, which are respectively recorded as the period standard deviation A1, the half-peak width standard deviation B1, and the mismatch standard deviation C1; and calculating the ratio of the period standard deviation A1, the half-peak width standard deviation B1, and the mismatch standard deviation C1 to the control group period standard deviation A, the half-peak width standard deviation B, and the mismatch standard deviation C Denoted as n A1 、n B1 、n C1 ; (5) When the ratio n A1 、n B1 、n C1 Less than n A '、n B '、n C 'When the supply is 90% of the supply, it is judged that the source furnace source material is sufficient.

2. The method according to claim 1, characterized in that When the ratio n A1 、n B1 、n C1 Greater than n A '、n B '、n C When the growth rate is 90%, the following steps are performed to conduct growth confirmation experiments: (6) Stop the sample holder from rotating, and ensure that the remaining growth conditions are consistent with those of the control group. Grow an epitaxial wafer thickened on the basis of the control group as the growth support group. Use the method in step (2) to measure the period, half-peak width and mismatch of the epitaxial wafer surface of the growth support group, and use the standard deviation formula in step (3) to calculate the period standard deviation A1', half-peak width standard deviation B1' and mismatch standard deviation C1' of the growth support group. Calculate the ratio of the period standard deviation A1', half-peak width standard deviation B1' and mismatch standard deviation C1' of the growth support group to those of the control groups A, B and C. Denoted as n A1 '、n B1 '、n C1 '; When n A1 '、n B1 '、n C1 ' is greater than or equal to n A '、n B '、n C ', it is judged that the current state is that the source furnace is insufficient in source material, and the cavity is opened for maintenance to replenish the source material; otherwise, the above steps (4)-(6) are repeated to grow.

3. The method according to claim 1 or 2, characterized in that: During the subsequent growth process, when the source material is sufficient, spot checks are performed on the prepared epitaxial wafers; when the remaining amount of the source material is close to the remaining amount of the empirical control group, each prepared epitaxial wafer is measured.

4. The method according to claim 1 or 2, characterized in that: In the step (2), n≧20.

5. The method according to claim 1 or 2, characterized in that: The empirical control group is replaced by one of the multiple epitaxial wafers grown under the same growth conditions as the control group when the growth thickness is close to 90-95% of the total growth thickness corresponding to the estimated source material weight, and the epitaxial wafer at which the slope of the curve of the period standard deviation A', the half-peak width standard deviation B' and the mismatch standard deviation C' changing with the number of grown wafers shows an inflection point is selected from the multiple epitaxial wafers as the empirical control group.

6. The method according to claim 1 or 2, characterized in that: The thickness of the growth material on the surface of the epitaxial wafer in the growth support group is 1.5-3 times the thickness of the growth material on the surface of the epitaxial wafer in the control group.

Images