Method for judging film growth accuracy

Abstract

The invention provides a method for judging film growth accuracy, which comprises the following steps of: S01, providing a reference sheet, and respectively acquiring a first dark state threshold voltage of the reference sheet in a dark state and a first light state threshold voltage of the reference sheet in a light state to obtain a reference threshold voltage of the reference sheet; s02, measuring and obtaining a first tape-out threshold voltage of the reference sheet; s03, setting the ratio of the reference threshold voltage to the first tape-out threshold voltage as a deviation coefficient K; s04, providing an epitaxial wafer, respectively obtaining a second dark state threshold voltage of the epitaxial wafer in a dark state and a second light state threshold voltage of the epitaxial wafer in a light state, obtaining an actually measured threshold voltage of the epitaxial wafer, and obtaining a second tape-out threshold voltage according to a ratio of the actually measured threshold voltage to the deviation coefficient; and S05, comparing the second tape-out threshold voltage with the first tape-out threshold voltage, and judging the film growth accuracy according to a comparison result. According to the method, the threshold voltage deviation of the epitaxial wafer caused by process fluctuation can be obtained, and the accuracy and consistency of threshold voltage judgment between batches can be improved.

Description

Technical Field

[0001] This invention relates to the field of semiconductor manufacturing technology, and in particular to a method for determining the accuracy of thin film growth. Background Technology

[0002] Molecular beam epitaxy (MBE) works by placing different high-purity materials in a specific crucible furnace, where the temperature of the furnace bottom and opening can be precisely controlled by electric heating. Different types of gaseous beams are simultaneously sprayed onto a substrate at a specific temperature in different proportions to form corresponding crystalline thin films on the substrate.

[0003] Molecular beam epitaxy (MBE) controls the epitaxial layer thickness by controlling time and beam current. The order and proportion of different material spraying methods are altered by adjusting the shutter sequence and needle valve. The amount of raw material in the crucible decreases with usage rate and time; at the same furnace temperature, the beam current gradually decreases, and the film growth rate also gradually slows down. However, fluctuations in film growth and the natural consumption of raw material in the crucible can lead to inaccuracies in the epitaxial structure, resulting in significant material and economic losses during the mass production stage of MBE.

[0004] Therefore, for those skilled in the art, designing a method to determine the accuracy of thin film growth is a pressing technical problem that needs to be solved. Summary of the Invention

[0005] This invention provides a method for judging the accuracy of thin film growth. This method can obtain the threshold voltage deviation of epitaxial wafers caused by epitaxial process fluctuations, improve the accuracy and consistency of threshold voltage judgment between batches, and improve the accuracy of process calibration in the mass production stage of epitaxial wafers.

[0006] To achieve the above objectives, the present invention provides a method for judging the accuracy of thin film growth, comprising:

[0007] S01: Provide a reference wafer of a high electron mobility transistor, wherein the reference wafer is a standard sample with an epitaxially grown thin film, and obtain the first dark state threshold voltage of the reference wafer in the dark state and the first light state threshold voltage of the reference wafer in the light state, and obtain the reference threshold voltage of the reference wafer based on the first dark state threshold voltage and the first light state threshold voltage.

[0008] S02: Measure and obtain the first wafer threshold voltage of the reference wafer;

[0009] S03: Set the ratio of the reference threshold voltage to the first wafer threshold voltage as the deviation coefficient K;

[0010] S04: Provide an epitaxial wafer of a high electron mobility transistor, wherein the epitaxial wafer is a test sample with an epitaxially grown thin film, and obtain the second dark-state threshold voltage of the epitaxial wafer in the dark state and the second light-state threshold voltage of the epitaxial wafer in the light state, respectively, obtain the measured threshold voltage of the epitaxial wafer based on the second dark-state threshold voltage and the second light-state threshold voltage, and obtain the second wafer fracturing threshold voltage based on the ratio of the measured threshold voltage to the deviation coefficient;

[0011] S05: Compare the second wafer threshold voltage with the first wafer threshold voltage, and determine the accuracy of the thin film growth of the epitaxial wafer based on the comparison result.

[0012] Optionally, in steps S01 and S04, the first dark-state threshold voltage and the first light-state threshold voltage of the reference wafer, and the second dark-state threshold voltage and the second light-state threshold voltage of the epitaxial wafer are obtained by the current-voltage curves in the electrochemical capacitance-voltage method.

[0013] Optionally, step S05 may further include:

[0014] If the difference between the second wafer threshold voltage and the first wafer threshold voltage is less than or equal to the preset voltage value, it is determined that the thin film growth accuracy of the epitaxial wafer meets the requirements.

[0015] If the difference between the second wafer threshold voltage and the first wafer threshold voltage is greater than the preset voltage value, it is determined that the film growth accuracy of the epitaxial wafer does not meet the requirements.

[0016] Optionally, in step S05, if it is determined that the accuracy of the thin film growth of the epitaxial wafer does not meet the requirements, the process parameters for growing the epitaxial layer of the epitaxial wafer are adjusted based on the second wafer threshold voltage and the first wafer threshold voltage.

[0017] Optionally, in step S01, the step of obtaining the reference threshold voltage of the reference chip based on the first dark state threshold voltage and the first light state threshold voltage includes:

[0018] A first scaling factor is set between 0 and 1, and the difference between 1 and the first scaling factor is set as a second scaling factor. The reference threshold voltage is the sum of the product of the first dark state threshold voltage and the first scaling factor and the product of the first light state threshold voltage and the second scaling factor.

[0019] Optionally, in step S04, the step of obtaining the measured threshold voltage of the epitaxial wafer based on the second dark-state threshold voltage and the second light-state threshold voltage includes:

[0020] The measured threshold voltage is the sum of the product of the second dark-state threshold voltage and the first proportionality coefficient and the product of the second light-state threshold voltage and the second proportionality coefficient.

[0021] Optionally, the equivalent beam pressure of the source furnace before and after epitaxial growth of the reference wafer and the epitaxial wafer can be obtained by using an ion gauge.

[0022] Optionally, the diffraction peak value of the epitaxial wafer is obtained by X-ray diffraction, and the diffraction peak value is compared with the simulation test value to determine whether the component ratio and the film thickness of the epitaxial wafer meet the requirements.

[0023] Optionally, the center wavelength of the epitaxial wafer can be obtained by photoluminescence spectroscopy, and the compositional uniformity of the epitaxial wafer can be obtained based on the center wavelength.

[0024] Optionally, the sheet resistance value of each test point on the epitaxial wafer is obtained through sheet resistance testing to obtain the sheet resistance uniformity of each test point; the sheet resistance uniformity of each test point is compared with the sheet resistance threshold range to determine whether the doping uniformity of the epitaxial wafer meets the requirements. This application provides a method for judging the accuracy of thin film growth. This method can fuse the dark state threshold voltage and light state threshold voltage of the epitaxial wafer and the reference wafer respectively to obtain the second wafer flotation threshold voltage of the epitaxial wafer, and then compare the second wafer flotation threshold voltage of the epitaxial wafer with the first wafer flotation threshold voltage of the reference wafer to obtain the threshold voltage deviation of the epitaxial wafer caused by epitaxial process fluctuations. This setting can avoid the fluctuations of reference wafer testing and process, improve the accuracy and consistency of threshold voltage judgment between batches, realize batch-to-batch consistency control of GaAs pHEMT wafer fabrication, and improve the accuracy of process calibration in the mass production stage of epitaxial wafers.

[0025] Furthermore, by analyzing the epitaxial wafer testing stage, this method can quickly calculate the deviation of epitaxial growth, determine the optimal epitaxial growth conditions and uniformity, and greatly save time and material costs. Attached Figure Description

[0026] Figure 1 This represents the different barrier layer thicknesses d in the prior art. i The corresponding IV characteristic curve, where the horizontal axis represents the gate voltage of the device and the vertical axis represents the drain-source current.

[0027] Figure 2 This is a flowchart of a method for judging the accuracy of thin film growth in a preferred embodiment of the present invention.

[0028] Figure 3 The above are IV characteristic curves of an n-type epitaxial wafer in dark and light states in a preferred embodiment of the present invention, wherein C1 is the IV curve in dark state and C2 is the IV curve in light state.

[0029] Figure 4 This is a schematic diagram of the structure of an ion gauge in a preferred embodiment of the present invention.

[0030] Figure 5 This is a schematic diagram illustrating the application scenario of the ion gauge in a preferred embodiment of the present invention.

[0031] Figure 6 This is a schematic diagram of the beam test results of Ga or In in a preferred embodiment of the present invention, where the horizontal axis represents time and the vertical axis represents the equivalent pressure of the beam.

[0032] Figure 7 This is a comparison chart of XRD simulation data and actual epitaxial growth test data in a preferred embodiment of the present invention. The horizontal axis represents the X-ray diffraction angle (2θ), and the vertical axis represents the intensity of the diffraction signal. The red curve in the chart represents the simulation data of the epitaxially grown thin film, and the blue curve represents the test data of the epitaxially grown thin film.

[0033] Figure 8 This is a photoluminescence spectrum of an epitaxial wafer in a preferred embodiment of the present invention.

[0034] Figure 9 This is a sheet resistance mapping pattern of an epitaxial wafer in a preferred embodiment of the present invention. Detailed Implementation

[0035] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The advantages and features of the present invention will become clearer from the following description. It should be noted that the drawings are all in a very simplified form and use non-precise proportions, and are only used to facilitate and clarify the illustration of the embodiments of the present invention.

[0036] The terms “center,” “longitudinal,” “lateral,” “length,” “width,” “thickness,” “upper,” “lower,” “front,” “rear,” “left,” “right,” “vertical,” “horizontal,” “top,” “bottom,” “inner,” “outer,” “clockwise,” “counterclockwise,” “axial,” “radial,” and “circumferential” indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the present invention and simplifying the description, and are not intended to indicate or imply that the mechanism or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the present invention.

[0037] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "fixation," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection, an electrical connection, or a connection that allows communication between them; they can refer to a direct connection or a connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0038] In the prior art, the tape-out threshold voltage Vto is defined as the critical voltage at which the gate voltage depletes the two-dimensional electron gas (2DEG) in the channel. Taking a pseudo-high electron mobility transistor (PHEMT) as an example, the formula for calculating the critical voltage is as follows:

[0039] Where ϕB represents the Schottky barrier height (eV) between the metal and AlGaAs; ΔE C Indicates the conduction band offset (eV) of the AlGaAs / GaAs heterojunction; n D Indicates AlGaAs doping concentration (cm⁻³); d i ε is the barrier layer thickness (nm); i Σ represents the dielectric constant of AlGaAs (≈12.9ε0); Σ represents the interfacial polarization charge density (C / m²), which is related to spontaneous polarization and piezoelectric polarization.

[0040] The quadratic term of di represents The linear term of di represents the contribution of the depletion layer charge to the electric field; increasing the thickness significantly increases Vto. Reflecting the effect of polarization charge, it is positively correlated with thickness. The negative adjustment terms represent ϕB and ΔE. C Provides a fixed value as the reference offset for Vto.

[0041] Figure 1 d represents different barrier layer thicknesses i The corresponding IV characteristic curve. For example... Figure 1 As shown, the barrier layer thickness d i Increasing the critical voltage can improve the device's performance, so the thickness of the epitaxial layer on the substrate surface can be obtained through the critical voltage, so that the device can meet the required performance.

[0042] Furthermore, those skilled in the art can combine and integrate the different embodiments or examples described herein, as well as the features of the different embodiments or examples, without contradiction.

[0043] Reference Figure 2 and Figure 3As shown, a preferred embodiment of this application provides a method for determining the accuracy of thin film growth, including:

[0044] S01: A reference wafer of a high electron mobility transistor is provided. The reference wafer is a standard sample with an epitaxially grown thin film. The first dark-state threshold voltage V-dark1 (i.e., the threshold voltage test value in the dark state) and the first light-state threshold voltage V-light1 (i.e., the threshold voltage test value in the light state) of the reference wafer are obtained. The reference threshold voltage VS0 of the reference wafer is obtained based on the first dark-state threshold voltage V-dark1 and the first light-state threshold voltage V-light1.

[0045] S02: Measure and obtain the first fabrication threshold voltage Vt0 of the reference wafer.

[0046] S03: Set the ratio of the reference threshold voltage VS0 to the first fabrication threshold voltage Vt0 as the deviation coefficient K, i.e., K = VS0 / Vt0. The deviation coefficient K is the core coefficient used to quantify the actual deviation between the design value (reference threshold voltage VS0) and the test value (first fabrication threshold voltage Vt0) of the reference wafer.

[0047] S04: Provide an epitaxial wafer of a high electron mobility transistor, the epitaxial wafer being a test sample with an epitaxially grown thin film. Obtain the second dark-state threshold voltage V-dark2 of the epitaxial wafer in the dark state and the second light-state threshold voltage V-light2 of the epitaxial wafer in the light state. Based on the second dark-state threshold voltage V-dark2 and the second light-state threshold voltage V-light2, obtain the measured threshold voltage V1 of the epitaxial wafer. Based on the ratio of the measured threshold voltage V1 to the deviation coefficient K, obtain the second wafer fabrication threshold voltage Vt1 (i.e., the calibration value of the wafer fabrication threshold voltage), Vt1 = V1 / K = V1 * Vt0 / VS0.

[0048] S05: Compare the second wafer threshold voltage Vt1 with the first wafer threshold voltage Vt0. Based on the comparison result, determine the accuracy of the epitaxial wafer's thin film growth, and further determine whether the process precision of the manufacturing process, such as the thin film growth thickness and uniformity, doping, and metallization, meets the standards. The first wafer threshold voltage Vt0 serves as the reference for the second wafer threshold voltage Vt1.

[0049] Specifically, this method obtains a reference threshold voltage VS0 by weighted fusing the first dark-state threshold voltage V-dark1 and the first light-state threshold voltage V-light1 of the reference wafer, and obtains a deviation coefficient K by the ratio of the reference threshold voltage VS0 to the first wafer flotation threshold voltage Vt0. When judging the accuracy of epitaxial film growth, the same deviation coefficient K can be used to correct the measured threshold voltage V1 obtained by weighted fusing the second dark-state threshold voltage V-dark2 and the second light-state threshold voltage V-light2 of the epitaxial wafer, resulting in a relatively accurate second wafer flotation threshold voltage Vt1. The accuracy of epitaxial film growth is then judged by comparing the second wafer flotation threshold voltage Vt1 with the first wafer flotation threshold voltage Vt0.

[0050] It should be understood that since the deviation coefficient K reflects the ratio between the measured threshold voltage V1 of the epitaxial wafer and the threshold voltage Vt1 of the second wafer, the measured threshold voltage V1 of the epitaxial wafer can be corrected by the deviation coefficient K to obtain the threshold voltage Vt1 of the second wafer.

[0051] It should also be understood that both reference wafers and epitaxial wafers are wafers of thin film material grown on a substrate using epitaxial growth technology, exhibiting the same crystal orientation or lattice matching as the substrate. By selecting a batch of qualified epitaxial wafers (e.g., GaAs pHEMT epitaxial wafers) as a reference wafer and subjecting it to actual fabrication and electrical testing, the first fabrication threshold voltage Vt0 of the reference wafer can be obtained. The first fabrication threshold voltage Vt0 is the threshold voltage obtained from the actual testing of the reference wafer, representing a measured parameter from device design to actual manufacturing.

[0052] This application provides a method for judging the accuracy of thin film growth. The method fuses the dark-state threshold voltage and light-state threshold voltage of the epitaxial wafer and the reference wafer to obtain the second wafer flotation threshold voltage Vt1 of the epitaxial wafer. Then, the second wafer flotation threshold voltage Vt1 of the epitaxial wafer is compared with the first wafer flotation threshold voltage Vt0 of the reference wafer to obtain the threshold voltage deviation caused by epitaxial process fluctuations. This setup avoids the fluctuations in reference wafer testing and processes, improves the accuracy and consistency of threshold voltage judgment between batches, achieves batch-to-batch consistency control of GaAs pHEMT wafer fabrication, and improves the accuracy of process calibration during the mass production stage of epitaxial wafers.

[0053] Furthermore, by analyzing the epitaxial wafer testing stage, this method can quickly calculate the deviation of epitaxial growth, determine the optimal epitaxial growth conditions and uniformity, and greatly save time and material costs.

[0054] It should be noted that IV testing of the reference wafer and epitaxial wafer in both dark (V-dark) and light (V-light) states is affected by factors such as the testing environment and instrument accuracy. The measured values ​​in a single state are prone to random deviations. This solution reduces environmental interference by fusing the threshold voltages in the light and dark states, making the reference threshold voltage VS0 closer to the first fabrication threshold voltage Vt0 of the device, thus enhancing the reliability of the data.

[0055] Meanwhile, slight fluctuations exist in the fabrication processes (such as thin film growth rate and doping uniformity) of different batches of epitaxial wafers. This solution establishes a deviation coefficient using K = Vs0 / Vt0, which standardizes the process benchmark of the reference wafer for qualified batches. Subsequent batches of epitaxial wafers only need to convert the measured threshold voltage V1 with the second fabrication threshold voltage Vt1 using the deviation coefficient K. This allows for a comparison of the threshold voltage deviation between the reference wafer and the epitaxial wafer under a unified benchmark, avoiding misjudgments caused by batch process drift and ensuring better process stability for different batches of products.

[0056] Further, in steps S01 and S04, the current-voltage curve (I) in the electrochemical capacitance-voltage method (ECV) is used to... d -V g The first dark-state threshold voltage V-dark1 and the first light-state threshold voltage V-light1 of the reference wafer, and the second dark-state threshold voltage V-dark2 and the second light-state threshold voltage V-light2 of the epitaxial wafer are obtained respectively using curves. At this time, the reference threshold voltage VS0 of the reference wafer or the measured threshold voltage V1 of the epitaxial wafer is the weighted threshold voltage obtained by IV testing of the reference wafer or epitaxial wafer through ECV method, and the second fabrication threshold voltage Vt1 of the epitaxial wafer is the calculated value of the fabrication threshold voltage of the epitaxial wafer after correction by the deviation coefficient K.

[0057] Specifically, the electrochemical capacitance-voltage method (ECV) can be used to test and obtain the I values ​​of the reference wafer and the epitaxial wafer under both dark and light conditions (i.e., in a light state). d -V g Curves. Using the electrochemical capacitance-voltage method instead of traditional electrical testing to obtain the threshold voltage of the reference wafer and epitaxial wafer can improve detection accuracy, thereby enhancing the accuracy of the detection results.

[0058] In other embodiments, the first dark state threshold voltage V-dark1 and the first light state threshold voltage V-light1 of the reference wafer, as well as the second dark state threshold voltage V-dark2 and the second light state threshold voltage V-light2 of the epitaxial wafer, can also be obtained directly through IV testing. That is, by testing the curve of the source and drain current of the device with the gate voltage, and extracting the gate voltage corresponding to the current turn-on in the IV curve, V-dark1, V-light1, V-dark2 and V-light2 are obtained respectively.

[0059] Reference Figure 3 As shown, the dark state I of the reference image is... d -V g Curve (or optical state I) d -V g The curve is extrapolated to I=0, corresponding to V. g This is the first dark state threshold voltage V-dark1 (or the first light state threshold voltage V-light1). Similarly, the dark state I of the epitaxial wafer can be... d -V g Curve (or optical state I) d -V g The curve is extrapolated to I=0, corresponding to V. g This is the second dark state threshold voltage V-dark2 (or the second light state threshold voltage V-light2).

[0060] Preferably, step S05 further includes:

[0061] If the absolute value of the difference between the second wafer threshold voltage Vt1 and the first wafer threshold voltage Vt0 is less than or equal to the preset voltage value V0, i.e., |Vt1 - Vt0| ≤ V0, the film growth accuracy of the epitaxial wafer is deemed to meet the requirements. In other words, when the second wafer threshold voltage Vt1 is close to the reference value (the first wafer threshold voltage Vt0), it indicates that the film growth accuracy of the epitaxial wafer is relatively high.

[0062] If the absolute value of the difference between the second wafer threshold voltage Vt1 and the first wafer threshold voltage Vt0 is greater than the preset value V0, i.e., |Vt1-Vt0|>V0, it indicates that the process parameters of the reference wafer and the epitaxial wafer during epitaxial growth are significantly different, and the accuracy of the thin film growth of the epitaxial wafer is determined to be unsatisfactory.

[0063] It should be noted that the specific value of the voltage preset value V0 can be set as needed. For example, the voltage preset value V0 can be set to 0.05V. When the difference between the second wafer threshold voltage Vt1 and the first wafer threshold voltage Vt0 exceeds 0.05V, it is determined that the film growth accuracy of the epitaxial wafer does not meet the requirements. At this time, it is necessary to adjust the epitaxial growth process parameters so that the difference between the second wafer threshold voltage Vt1 of the epitaxial wafer and the first wafer threshold voltage Vt0 of the reference wafer is within 0.05V.

[0064] More preferably, in step S05, if it is determined that the accuracy of the epitaxial wafer's thin film growth does not meet the requirements, the process parameters (such as thin film thickness, doping concentration, and / or gate metal layer thickness) of the epitaxial wafer's growth layer are adjusted based on the second wafer threshold voltage Vt1 and the first wafer threshold voltage Vt0, so that the accuracy of the epitaxial wafer's growth film meets the requirements.

[0065] As a specific example, the thickness of the barrier layer and the doping in the epitaxial structure of the subsequent epitaxial wafer under test can be adjusted based on the difference between the second wafer threshold voltage Vt1 and the first wafer threshold voltage Vt0, thereby improving the accuracy of epitaxy and compensating for the fluctuations in threshold voltage caused by the reduction of the source furnace beam current during molecular beam epitaxy. Simultaneously, the thickness of the gate metal layer in the subsequent epitaxial wafer under test can also be adjusted based on the difference between the second wafer threshold voltage Vt1 and the first wafer threshold voltage Vt0 to regulate the threshold voltage of the GaAs pHEMT, reduce threshold voltage deviation, and improve the accuracy of the epitaxial film growth.

[0066] In a preferred embodiment, step S01, the step of obtaining the reference threshold voltage VSO of the reference chip based on the first dark state threshold voltage V-dark1 and the first light state threshold voltage V-light1, includes:

[0067] Set a first proportionality coefficient y1 between 0 and 1, where 0 < y1 < 1. Set the difference between 1 and the first proportionality coefficient y1 as the second proportionality coefficient y2, where y2 = (1 - y1). The reference threshold voltage VS0 is the sum of the product of the first dark-state threshold voltage V-dark1 and the first proportionality coefficient y1 and the product of the first light-state threshold voltage V-light1 and the second proportionality coefficient y2, i.e., VS0 = y1 * |V-dark1| + y2 * |V-light1|, or VS0 = y1 * |V-dark1| + (1 - y1) * |V-light1|.

[0068] It should be understood that the first proportional coefficient y1 is the dark state weight ratio adjusted in the reference threshold voltage VS0 according to process requirements, and the second proportional coefficient y2 is the light state weight ratio adjusted in the reference threshold voltage VS0 according to process requirements. The sum of the first proportional coefficient y1 and the second proportional coefficient y2 is 1.

[0069] It should be noted that the value of the first proportional coefficient y1 needs to be adjusted according to the application scenario of the device. For example, for high-frequency devices, the first proportional coefficient y1 can be increased to increase the dark state weight; for photosensitive devices, the second proportional coefficient y2 can be increased to increase the light state weight, thereby making the reference threshold voltage VS0 of the reference wafer closer to the first fabrication threshold voltage Vt0. The first proportional coefficient y1 usually needs to be adjusted and verified through multiple batches of fabrication to determine the optimal value of the first proportional coefficient y1 for the production line process.

[0070] Further, in step S04, the step of obtaining the measured threshold voltage V1 of the epitaxial wafer based on the second dark-state threshold voltage V-dark2 and the second light-state threshold voltage V-light2 includes:

[0071] The measured threshold voltage V1 is the sum of the product of the second dark-state threshold voltage V-dark2 and the first proportional coefficient y1 and the product of the second light-state threshold voltage V-light2 and the second proportional coefficient y2 (1-y1), that is, V1 = y1*|V-dark2| + y2* |V-light2|, or V1 = y1*|V-dark2| + (1-y1)*|V-light2|.

[0072] Generally, the reference threshold voltage VS0 of the reference wafer can be made close to or equal to the first wafer flotation threshold voltage Vt0 by repeatedly adjusting the first proportional coefficient y1 and the second proportional coefficient y2. If the first wafer flotation threshold voltage Vt0 still deviates from the reference threshold voltage VS0 after the first proportional coefficient y1 and the second proportional coefficient y2 are determined, the deviation coefficient K can be obtained, K = VS0 / Vt0. After obtaining the measured threshold voltage V1 of the epitaxial wafer, the second wafer flotation threshold voltage Vt1 of the epitaxial wafer is obtained through the same deviation coefficient K. That is, the calculated value of the wafer flotation threshold voltage is obtained by calculating the reference threshold voltage VS0 and the deviation coefficient K.

[0073] Reference Figure 4 and Figure 5 In some embodiments, the equivalent beam pressure of the source furnace (e.g., In, Ga, and Al source furnaces) before and after epitaxial growth is obtained using an ion gauge for the reference wafer and the epitaxial wafer, respectively. Preferably, it is determined whether the equivalent beam pressure exceeds an equivalent pressure threshold range, and the source furnace temperature is adjusted if the equivalent beam pressure exceeds the equivalent pressure threshold range.

[0074] It should be noted that molecular beam epitaxy (MBE) generates a beam by heating the source material at high temperatures. A beam is a stream of particles with a certain velocity and direction distribution. Beam equivalent pressure (BEP) is a physical quantity that reflects the magnitude of the beam generated by the source furnace, and is usually measured by the equivalent pressure generated by the beam in an ultra-high vacuum chamber.

[0075] Reference Figure 6 As shown, the effective beam pressure PE (i.e., the equivalent beam pressure) can be obtained by the difference between the test beam pressure PI and the test background pressure PB, i.e., PE = PI - PB. PE characterizes the beam magnitude of the source furnace. Before epitaxial growth, the beam current data of each source furnace at different temperatures can be determined by repeatedly adjusting the temperature of each source furnace and measuring the equivalent beam pressure. During epitaxial growth, material consumption and slight changes in the cavity environment within the source furnace may cause beam drift. After epitaxial growth: The equivalent beam pressure of each source furnace can be measured at the same temperature using the same method as before growth to verify the stability of the source furnace and correct deviations. Then, the beam current data before and after epitaxial growth are compared. If deviations exist, the source furnace parameters (e.g., source furnace temperature) can be adjusted in a timely manner to avoid affecting the film growth quality of the next batch of epitaxial wafers.

[0076] As a preferred embodiment, the diffraction peak of the epitaxial wafer can also be obtained by X-ray diffraction (XRD), and the diffraction peak is compared with the simulation test value to determine whether the composition ratio and film thickness of the epitaxial wafer meet the requirements.

[0077] More specifically, XRD testing of epitaxial growth of epitaxial wafers includes: using the simulation software built into XRD to perform simulation tests based on the growth structure, simulating the composition and thickness of the core epitaxial layers, such as the channel layer and the barrier layer, and comparing the simulation data with the XRD test data of the epitaxial wafer to confirm the simulation data.

[0078] Furthermore, XRD simulation software constructs a model based on dynamic diffraction theory or kinematic diffraction theory. By comparing the peak position, peak shape, and fringe period of the fitted experimental curve with the simulation curve, iterative optimization is performed to obtain the precise thickness of the film. Based on the film thickness, the structural parameters of the epitaxial layer (such as the composition and thickness of the channel layer and barrier layer) are corrected to obtain accurate parameters that match the simulation curve, thereby improving the yield of epitaxial wafers and the consistency of device performance.

[0079] Figure 7 This is a comparison chart of XRD simulation data and actual epitaxial growth test data. The horizontal axis represents the X-ray diffraction angle (2θ), and the vertical axis represents the intensity of the diffraction signal. Figure 7 The red curve represents the simulation data of the epitaxially grown thin film, and the blue curve represents the test data of the epitaxially grown thin film. If the blue data does not match the red data, the process parameters for growing the epitaxial layer (such as film thickness, doping concentration, and / or gate metal layer thickness) can be adjusted to ensure that the accuracy of the grown thin film meets the requirements.

[0080] By combining the electrochemical capacitance voltage method (ECV), source furnace beam equivalent pressure detection, and XRD simulation verification, the range of epitaxial film thickness values ​​can be determined, which helps to significantly reduce errors in the epitaxial growth stage. Furthermore, this method can compensate for beam attenuation in real time based on the reference values ​​provided by the calibration values ​​of beam current and film thickness, allowing for timely correction of epitaxial growth parameters and improving epitaxial production efficiency.

[0081] In a further preferred embodiment, the center wavelength of the epitaxial wafer can be obtained through photoluminescence spectroscopy, and the compositional uniformity of the epitaxial wafer can be obtained based on the center wavelength.

[0082] More specifically, during testing, the epitaxial layer of the epitaxial wafer needs to be scanned point by point to extract the center wavelength of the spectrum at each point, thus obtaining a center wavelength spectrum at each point. The photoluminescence wavelength spectrum is used to statistically represent the compositional uniformity distribution of the quantum wells (which can be regarded as channel layers) at various locations on the entire epitaxial wafer. Figure 8 The color differences in the epitaxial layers represent differences in component distribution. Figure 8 The color distribution of each test point can be used to measure the composition of the epitaxial wafer's thin film channel region and also to obtain the compositional uniformity of the epitaxial layer.

[0083] Preferably, the measurement results of photoluminescence spectroscopy and XRD test results can be compared to improve the accuracy of the measurement results and to comprehensively evaluate the film quality of the epitaxial wafer.

[0084] In a preferred embodiment, the sheet resistance value of each test point on the epitaxial wafer can be obtained by sheet resistance testing, thereby obtaining the sheet resistance uniformity of each test point.

[0085] Furthermore, sheet mapping can characterize the uniformity of thin film doping. By analyzing the surface distribution of resistance in micro-regions on the epitaxial layer, it can quantify and calibrate fluctuations in doping concentration. In actual measurements, a point-by-point scan is performed on the target region (such as the epitaxial channel region) of the epitaxial wafer, typically involving more than 50 test points. A constant current is applied to each test point, and the sheet resistance value for each point is calculated. After the scan is complete, a sheet resistance spectrum can be obtained (see reference). Figure 9 ). Figure 9 The curves in the figure represent the distribution of sheet resistance at each test point on the epitaxial wafer, thus providing statistical information on the sheet resistance distribution from the center to the edge of the epitaxial layer. The resistance at each test point reflects the doping concentration fluctuations, which allows for the determination of the doping concentration at various locations on the epitaxial wafer, thereby improving the accuracy of epitaxial wafer growth.

[0086] Preferably, the sheet resistance uniformity of each test point can be compared with the sheet resistance threshold range to determine whether the film doping uniformity of the epitaxial wafer meets the requirements.

[0087] Generally, if the sheet resistance uniformity (Wafer Uniformity Valve) at each test point on the epitaxial layer is less than or equal to the sheet resistance threshold range (e.g., 1%), i.e., sheet resistance uniformity (%) ≤ 1%, it indicates that the doping concentration fluctuation of the epitaxial layer has not exceeded the threshold. If the sheet resistance uniformity at each test point on the epitaxial layer is greater than the sheet resistance threshold range, i.e., sheet resistance uniformity (%) > 1%, it indicates that the doping concentration fluctuation of the epitaxial layer exceeds the threshold. In this case, parameters such as source furnace beam current stability and substrate temperature uniformity can be adjusted in conjunction with the epitaxial growth process to achieve accurate calibration of doping fluctuations.

[0088] This application employs ECV (Electrochemical Voltage Verification) to calibrate the threshold voltage of both the reference wafer and the epitaxial wafer. It also obtains the beam current magnitude for molecular beam epitaxy by measuring the equivalent beam pressure of the source furnace. Furthermore, it measures the film composition and thickness of the epitaxial wafer using XRD, and the compositional uniformity of the epitaxial film is measured using photoluminescence spectroscopy. Simultaneously, the doping uniformity of the epitaxial film is obtained through sheet resistance testing. This closed-loop process covers full parameter testing of the epitaxial wafer and establishes a standard procedure for calibrating the epitaxial accuracy of GaAs pHEMT, solving the problem of mass production consistency and possessing broad applicability.

[0089] In summary, this invention provides a method for determining the accuracy of thin film growth. This method fuses the dark-state threshold voltage and light-state threshold voltage of the epitaxial wafer and the reference wafer to obtain the second wafer flotation threshold voltage. Then, it compares the second wafer flotation threshold voltage with the first reference wafer flotation threshold voltage to identify threshold voltage deviations caused by epitaxial process fluctuations. This approach avoids the volatility of reference wafer testing and processes, improves the accuracy and consistency of threshold voltage determination between batches, achieves batch-to-batch consistency control of GaAs pHEMT wafer fabrication, and enhances the accuracy of process calibration during the mass production stage of epitaxial wafers.

[0090] Furthermore, by analyzing the epitaxial wafer testing stage, this method can quickly calculate the deviation of epitaxial growth, determine the optimal epitaxial growth conditions and uniformity, and greatly save time and material costs.

[0091] The above description is merely a description of preferred embodiments of the present invention and is not intended to limit the scope of the present invention in any way. Any changes or modifications made by those skilled in the art based on the above disclosure shall fall within the protection scope of the present invention.

Claims

1. A method for judging the accuracy of thin film growth, characterized in that, include: S01: Provide a reference wafer of a high electron mobility transistor, wherein the reference wafer is a standard sample with an epitaxially grown thin film, and obtain the first dark state threshold voltage of the reference wafer in the dark state and the first light state threshold voltage of the reference wafer in the light state, and obtain the reference threshold voltage of the reference wafer based on the first dark state threshold voltage and the first light state threshold voltage. S02: Measure and obtain the first wafer threshold voltage of the reference wafer; S03: Set the ratio of the reference threshold voltage to the first wafer threshold voltage as the deviation coefficient K; S04: Provide an epitaxial wafer of a high electron mobility transistor, wherein the epitaxial wafer is a test sample with an epitaxially grown thin film, and obtain the second dark-state threshold voltage of the epitaxial wafer in the dark state and the second light-state threshold voltage of the epitaxial wafer in the light state, respectively, obtain the measured threshold voltage of the epitaxial wafer based on the second dark-state threshold voltage and the second light-state threshold voltage, and obtain the second wafer fracturing threshold voltage based on the ratio of the measured threshold voltage to the deviation coefficient; S05: Compare the second wafer threshold voltage with the first wafer threshold voltage, and determine the accuracy of the thin film growth of the epitaxial wafer based on the comparison result.

2. The method for judging the accuracy of thin film growth as described in claim 1, characterized in that, In steps S01 and S04, the first dark-state threshold voltage and the first light-state threshold voltage of the reference wafer, as well as the second dark-state threshold voltage and the second light-state threshold voltage of the epitaxial wafer, are obtained by the current-voltage curves in the electrochemical capacitance-voltage method.

3. The method for judging the accuracy of thin film growth as described in claim 1, characterized in that, Step S05 also includes: If the difference between the second wafer threshold voltage and the first wafer threshold voltage is less than or equal to the preset voltage value, it is determined that the thin film growth accuracy of the epitaxial wafer meets the requirements. If the difference between the second wafer threshold voltage and the first wafer threshold voltage is greater than the preset voltage value, it is determined that the film growth accuracy of the epitaxial wafer does not meet the requirements.

4. The method for judging the accuracy of thin film growth as described in claim 3, characterized in that, In step S05, if it is determined that the accuracy of the thin film growth of the epitaxial wafer does not meet the requirements, the process parameters for growing the epitaxial layer of the epitaxial wafer are adjusted based on the second wafer threshold voltage and the first wafer threshold voltage.

5. The method for judging the accuracy of thin film growth as described in claim 1, characterized in that, In step S01, the step of obtaining the reference threshold voltage of the reference chip based on the first dark state threshold voltage and the first light state threshold voltage includes: A first scaling factor is set between 0 and 1, and the difference between 1 and the first scaling factor is set as a second scaling factor. The reference threshold voltage is the sum of the product of the first dark state threshold voltage and the first scaling factor and the product of the first light state threshold voltage and the second scaling factor.

6. The method for judging the accuracy of thin film growth as described in claim 5, characterized in that, In step S04, the step of obtaining the measured threshold voltage of the epitaxial wafer based on the second dark-state threshold voltage and the second light-state threshold voltage includes: The measured threshold voltage is the sum of the product of the second dark-state threshold voltage and the first proportionality coefficient and the product of the second light-state threshold voltage and the second proportionality coefficient.

7. The method for judging the accuracy of thin film growth as described in any one of claims 1 to 6, characterized in that, The equivalent beam pressure of the source furnace before and after epitaxial growth of the reference wafer and the epitaxial wafer was obtained by using an ion gauge.

8. The method for judging the accuracy of thin film growth as described in any one of claims 1 to 6, characterized in that, The diffraction peaks of the epitaxial wafer are obtained by X-ray diffraction, and the diffraction peaks are compared with the simulation test values ​​to determine whether the composition ratio and film thickness of the epitaxial wafer meet the requirements.

9. The method for judging the accuracy of thin film growth as described in any one of claims 1 to 6, characterized in that, The center wavelength of the epitaxial wafer is obtained by photoluminescence spectroscopy, and the compositional uniformity of the epitaxial wafer is obtained based on the center wavelength.

10. The method for judging the accuracy of thin film growth as described in any one of claims 1 to 6, characterized in that, Sheet resistance values ​​at each test point on the epitaxial wafer are obtained by sheet resistance testing to determine the sheet resistance uniformity of each test point; the sheet resistance uniformity of each test point is compared with the sheet resistance threshold range to determine whether the doping uniformity of the epitaxial wafer meets the requirements.

Images