Semiconductor process equipment and process formula optimization method thereof

Abstract

The invention discloses semiconductor process equipment and a process formula optimization method thereof, and the process formula optimization method comprises the steps: receiving an optimization instruction which comprises a to-be-optimized process formula and an optimization target and an optimization mode thereof, responding to the optimization instruction, and optimizing the to-be-optimized process formula according to the optimization target and the optimization mode of the to-be-optimized process formula. Obtaining process data matched with the to-be-optimized process formula from a process database, and calculating an optimization value of a to-be-optimized parameter matched with the optimization mode in the to-be-optimized process formula based on the obtained process data and a preset parameter calculation model matched with the optimization mode, and if the to-be-optimized process formula with the optimized value meets the optimization target, taking the optimized value as the value of the to-be-optimized parameter in the to-be-optimized process formula, so that the process formula can be automatically optimized by utilizing the semiconductor process equipment, and a standard process formula of the semiconductor process equipment is obtained. Therefore, the optimization efficiency and accuracy of the process formula can be improved, and the optimization cost of the process formula can be reduced.

Description

Technical Field

[0001] This invention relates to the field of semiconductor technology, and more specifically to a semiconductor process equipment and a method for optimizing the process formulation. Background Technology

[0002] With the development of advanced manufacturing processes, higher demands are being placed on the quality of thin films prepared by semiconductor process equipment such as LPCVD (Low Pressure Chemical Vapor Deposition). Because the process formulation is a key factor affecting the quality of thin films prepared by semiconductor process equipment, and different semiconductor process equipment using the same process formulation will produce thin films of varying quality, only by finding a standard process formulation that matches any given semiconductor process equipment can the production of high-quality thin films be ensured.

[0003] Currently, process engineers optimize and adjust the process formula based on test data from a large number of process experiments on any semiconductor process equipment to obtain a standard process formula that matches it. However, this optimization method mainly relies on the personal experience of process engineers, which carries the risk of optimization errors. Moreover, this optimization method requires a lot of time, manpower and equipment resources, which is not conducive to the rapid production of semiconductor process equipment. Summary of the Invention

[0004] This invention discloses a semiconductor process equipment and a method for optimizing the process formulation, so as to improve the optimization efficiency and accuracy of the process formulation and reduce the optimization cost of the process formulation.

[0005] In a first aspect, the present invention discloses a method for optimizing a process formulation of semiconductor process equipment, comprising: receiving an optimization instruction, the optimization instruction including a process formulation to be optimized, its optimization objective, and an optimization mode; responding to the optimization instruction, acquiring process data matching the process formulation to be optimized from a process database; the process database storing multiple sets of process data collected during the same thin film preparation process of the semiconductor process equipment; calculating, based on the acquired process data and a preset parameter calculation model matching the optimization mode, the optimized value of the parameter to be optimized in the process formulation to be optimized that matches the optimization mode; if the process formulation to be optimized having the optimized value satisfies the optimization objective, then the optimized value is used as the value of the parameter to be optimized in the process formulation to be optimized.

[0006] In some embodiments, obtaining process data matching the process formulation to be optimized from the process database includes: obtaining any set of process data from the process database; obtaining a data vector corresponding to the set of process data based on the set of process data, the optimization target, and the process formulation to be optimized; calculating the distance between the data vector and a preset reference vector; and determining the set of process data as process data matching the process formulation to be optimized if the distance is within a preset distance range.

[0007] In some embodiments, obtaining the data vector corresponding to the set of process data based on the set of process data, the optimization target, and the process formulation to be optimized includes: obtaining a temperature vector in the data vector corresponding to the set of process data based on the temperature values ​​of the intermediate temperature zones in the set of process data, the target temperature values ​​in the optimization target, and a preset temperature calculation model; obtaining each deposition rate vector in the data vector corresponding to the set of process data based on the film thickness values ​​and film formation times of each temperature zone in the set of process data, the target film thickness values ​​in the optimization target, and a preset deposition rate calculation model; obtaining a gas flow rate vector in the data vector corresponding to the set of process data based on the gas flow rate values ​​in the set of process data, the main process gas ratio in the process formulation to be optimized, and a preset gas flow rate calculation model; obtaining a gas pressure vector in the data vector corresponding to the set of process data based on the gas pressure values ​​in the set of process data, the gas pressure values ​​in the process formulation to be optimized, and a preset gas pressure calculation model; and determining a temperature change vector in the data vector corresponding to the set of process data based on the temperature change rate values ​​in the set of process data.

[0008] In some embodiments, the preset temperature calculation model is determined by the temperature values ​​of the intermediate temperature zone in the set of process data and the target temperature value in the optimization objectives; the preset deposition rate calculation model is determined by the film thickness values ​​of each temperature zone in the set of process data, the film formation time in the set of process data, and the target film thickness value in the optimization objectives; the preset gas flow rate calculation model is determined by the gas flow rate values ​​in the set of process data and the main process gas ratio in the process formulation to be optimized; and the preset gas pressure calculation model is determined by the gas pressure values ​​in the set of process data and the gas pressure values ​​in the process formulation to be optimized.

[0009] In some embodiments, the preset temperature calculation model includes: T = Z ij / T t Where T represents the temperature vector, Z ij T represents the temperature value of the intermediate temperature zone in this set of process data. t This represents the target temperature value in the optimization objectives; the preset deposition rate calculation model includes: DRz =(THK) z / Dt) / THK M , among which, DR z THK represents the various deposition rate vectors. z This represents the film thickness value for each temperature zone in this set of process data, Dt represents the film formation time in this set of process data, and THK represents the film thickness value for each temperature zone in this set of process data. M The target film thickness is represented in the optimization objective; the preset gas flow rate calculation model includes: MR = MFR / MFR M Where MR represents the gas flow vector, and MFR represents the gas flow value in this set of process data. M This represents the ratio of the main process gas in the process formulation to be optimized; the preset gas pressure calculation model includes: VP = Vac / Vac R Where VP represents the gas pressure vector, and Vac represents the gas pressure value in this set of process data. R This indicates the gas pressure value in the process formulation to be optimized.

[0010] In some embodiments, determining the set of process data as process data matching the process formulation to be optimized includes: performing an orthogonality check on the set of process data to obtain an orthogonality check result; if the orthogonality check result meets preset check requirements, then determining the set of process data as process data matching the process formulation to be optimized.

[0011] In some embodiments, calculating the optimized values ​​of the parameters to be optimized in the process formulation to be optimized that match the optimization mode based on the acquired process data and a preset parameter calculation model matching the optimization mode includes: if the optimization mode is a constant temperature setting optimization mode, then the optimized temperature value of each temperature zone in the process formulation to be optimized is calculated based on the temperature value and film thickness value of each temperature zone in the acquired process data and the first parameter calculation model; if the optimization mode is a variable temperature setting optimization mode, then the optimized temperature change rate value of each temperature zone in the process formulation to be optimized is calculated based on the temperature change rate value and intra-wafer uniformity value in the acquired process data and the second parameter calculation model.

[0012] In some embodiments, the first parameter calculation model is determined by the target temperature value in the optimization target, the target film thickness value of m temperature zones in the optimization target, the temperature value matrix of m temperature zones in n sets of process data, and the film thickness value matrix of m temperature zones in n sets of process data; the second parameter calculation model is determined by the optimized temperature change rate value of the middle (m+1) / 2 temperature zones in the process formulation to be optimized, the target intra-chip uniformity value of m temperature zones in the optimization target, the temperature change rate value matrix of m temperature zones in n sets of process data, and the intra-chip uniformity value matrix of m temperature zones in n sets of process data.

[0013] In some embodiments, the first parameter calculation model includes: Where Z1 to Zm represent the optimized temperature values ​​for m temperature zones in the process formulation to be optimized, and T t This represents the target temperature value in the optimization objective.

[0014] Thk1 to Thkm represent the target film thickness values ​​for the m temperature zones in the optimization objective. This is a matrix of temperature values ​​for m temperature zones from n sets of process data. This is a matrix of film thickness values ​​for m temperature zones from n sets of process data.

[0015] The second parameter calculation model includes:

[0016]

[0017] Where Ra1 to Ra((m+1) / 2) are the optimized temperature change rate values ​​for the middle (m+1) / 2 temperature zones in the process formulation to be optimized, WIWT1 to WIWTm are the target intra-piece uniformity values ​​for the m temperature zones in the optimization objective, and W′ is a generalized relation matrix constructed based on the generalized relation matrix W, where,

[0018] in, This is a matrix of temperature change rate values ​​for m temperature zones from n sets of process data. This is a matrix of intra-chip uniformity values ​​for m temperature zones in n sets of process data.

[0019] In some embodiments, the step of using the optimized value as the value of the parameter to be optimized in the process formulation to be optimized if the optimized value satisfies the optimization objective includes: if the product of the calculated optimized temperature change rate value and the film formation time in the process formulation to be optimized is within the target temperature range in the optimization objective, then the optimized temperature change rate value is used as the temperature change rate value in the process formulation to be optimized; if the product of the optimized temperature change rate value and the film formation time in the process formulation to be optimized is not within the target temperature range in the optimization objective, the film formation time in the process formulation to be optimized is divided into multiple sub-film formation times, and if the product of any sub-film formation time and the optimized temperature change rate value is within the target temperature range in the optimization objective, then the optimized temperature change rate value is used as the temperature change rate value in the process formulation to be optimized, and the number of sub-film formation times is used as the number of temperature changes in the process formulation to be optimized.

[0020] In a second aspect, the present invention discloses a semiconductor process apparatus, including a memory and a processor; the memory is used to store a computer program; the processor is used to execute a process formulation optimization method for the semiconductor process apparatus as described in any of the preceding claims, based on the computer program stored in the memory.

[0021] Thirdly, the present invention discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, performs a process formulation optimization method for any of the above semiconductor process equipment.

[0022] The semiconductor process equipment and its process recipe optimization method disclosed in this invention respond to an optimization instruction containing a process recipe to be optimized, its optimization target, and an optimization mode. It retrieves process data matching the process recipe to be optimized from a process database. Based on the retrieved process data and a preset parameter calculation model matching the optimization mode, it calculates the optimized values ​​of the parameters to be optimized in the process recipe that match the optimization mode. If the process recipe to be optimized with optimized values ​​satisfies the optimization target, the optimized values ​​are used as the values ​​of the parameters to be optimized in the process recipe. This allows for automatic optimization of process recipes using semiconductor process equipment, obtaining standard process recipes for the semiconductor process equipment, thereby improving the optimization efficiency and accuracy of process recipes and reducing the optimization cost. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention or the background art, the accompanying drawings used in the embodiments of the present invention or the background art will be described below.

[0024] Figure 1This is a flowchart of a method for optimizing the process formulation of semiconductor process equipment, as disclosed in an embodiment of the present invention.

[0025] Figure 2 This is a flowchart of a method for obtaining process data disclosed in an embodiment of the present invention.

[0026] Figure 3 This is a process effect diagram of a process formulation optimized through multiple cycles, as disclosed in an embodiment of the present invention.

[0027] Figure 4 This is a schematic diagram of the structure of a semiconductor process equipment disclosed in an embodiment of the present invention. Detailed Implementation

[0028] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0029] As an optional implementation of the disclosure of this invention, an embodiment of this invention discloses a method for optimizing the process recipe of semiconductor process equipment, such as... Figure 1 As shown, Figure 1 This is a flowchart of a process recipe optimization method for semiconductor process equipment disclosed in an embodiment of the present invention. The method includes:

[0030] S101: Receive optimization instructions, which include the process formulation to be optimized, its optimization objectives, and optimization modes.

[0031] When optimizing the process formulation of a certain film, the user first determines the process formulation to be optimized, as well as the optimization target and optimization mode of the process formulation. Then, the user sends the optimization instructions containing the process formulation to be optimized, its optimization target and optimization mode to the computer device so that the computer device can optimize the process formulation to be optimized.

[0032] In some embodiments, the optimization objectives include target temperature values, target film deposition times, target film thickness values, target intra-wafer uniformity values, and target temperature ranges that need to be achieved during the thin film preparation process. In some embodiments, the optimization modes include a constant temperature set (CTS) optimization mode and a dynamic temperature set (DTS) optimization mode, wherein the constant temperature set optimization mode is used to optimize inter-wafer uniformity (WWTW), and the dynamic temperature set optimization mode is used to optimize intra-wafer uniformity (WIW).

[0033] It should be noted that inter-wafer thickness uniformity refers to the uniformity of film thickness fabricated on different wafers, while intra-wafer thickness uniformity refers to the uniformity of film thickness fabricated on the same wafer. In isothermal optimization mode, the temperature during film fabrication remains constant; in variable temperature optimization mode, the temperature in each temperature zone during film fabrication changes (rises or falls) at a certain rate. Understandably, because there is a temperature difference between the edge and center of the wafer, the film thickness will also differ between the edge and center. Therefore, to improve film thickness uniformity, film growth needs to be performed during the cooling stage. By gradually cooling from the center to the edge of the wafer, the unevenness of the film on the wafer surface can be reduced.

[0034] S102: In response to the optimization command, retrieve process data from the process database that matches the process formulation to be optimized.

[0035] The process database stores multiple sets of process data collected during the fabrication of identical thin films using semiconductor processing equipment. Each set of process data includes temperature, gas flow rate, gas pressure, and film thickness values ​​during the thin film fabrication process. The semiconductor processing equipment can periodically collect this process data and store it in the process database. The process formulation to be optimized also includes temperature, gas flow rate, gas pressure, and film thickness values ​​during the thin film fabrication process.

[0036] Taking the preparation of silicon nitride thin films as an example, as shown in Table 1, Ru1 to Ru5 represent 5 sets of process data, such as the process data for preparing silicon nitride thin films on 5 wafers, Z1 to Z5 represent the temperature values ​​of 5 temperature zones of a wafer, THK1 to THK5 represent the film thickness values ​​of 5 temperature zones of a wafer, Dt represents the film deposition time of silicon nitride thin films on a wafer, MFR represents the gas flow rate, Vac represents the gas pressure, and Ra represents the temperature change rate.

[0037] Table 1

[0038]

[0039] In some embodiments, process data matching the process formulation to be optimized can be obtained from a process database based on temperature, film thickness, gas flow rate, and gas pressure values. For example, if the temperature, film thickness, gas flow rate, and gas pressure values ​​of a set of process data are equal to or similar to the temperature, film thickness, gas flow rate, and gas pressure values ​​of the process formulation to be optimized, then this set of process data is considered to be process data matching the process formulation to be optimized.

[0040] However, this method of acquisition cannot perform filtering with different focuses. For example, it cannot focus on temperature values ​​when filtering process data. Based on this, in some other embodiments, the distance between the data vector of the process data and the preset benchmark vector can be used to filter the process data, so as to improve the filtering efficiency while focusing on the optimization target.

[0041] Specifically, such as Figure 2 As shown, Figure 2 The present invention discloses a method for obtaining process data, comprising: S201: obtaining any set of process data from a database of semiconductor process equipment; S202: obtaining a data vector corresponding to the set of process data based on the set of process data, the optimization target, and the process formulation to be optimized; S203: calculating the distance between the data vector and a preset reference vector; S204: if the distance is within a preset distance range, determining the set of process data as process data matching the process formulation to be optimized.

[0042] In some embodiments, a temperature vector can be obtained from the data vector corresponding to the set of process data based on the temperature values ​​of the intermediate temperature zone in the set of process data, the target temperature value in the optimization objective, and a preset temperature calculation model. In some embodiments, the preset temperature calculation model is determined by the temperature values ​​of the intermediate temperature zone in the set of process data and the target temperature value in the optimization objective. In some embodiments, the preset temperature calculation model may include: T = Z ij / T t Where T represents the temperature vector in the data vector corresponding to this set of process data, and Z... ij This represents the temperature value of the intermediate temperature zone in this set of process data, where i indicates which set of process data it is, j indicates which intermediate temperature zone it is, and T t This represents the target temperature value in the optimization objective. Taking i=3 and j=1 as an example, as shown in Table 1, Z... ij This represents the temperature value of the third temperature zone Z3 of the wafer in the first set of process data Ru1.

[0043] In some embodiments, deposition rate vectors can be obtained from the data vector corresponding to the set of process data based on the film thickness and deposition time of each temperature zone in the set of process data, the target film thickness in the optimization objectives, and a preset deposition rate calculation model. In some embodiments, the preset deposition rate calculation model is determined by the film thickness of each temperature zone in the set of process data, the film deposition time in the set of process data, and the target film thickness in the optimization objectives. In some embodiments, the preset deposition rate calculation model includes: DR z =(THK) z / Dt) / THK M , among which, DR z This represents the deposition rate vectors in the data vector corresponding to this set of process data, THK. z This represents the film thickness value for each temperature zone in this set of process data, Dt represents the film deposition time in this set of process data, and THK represents the film thickness value for each temperature zone in this set of process data. M The target film thickness value in the optimization objective is represented by z, which takes values ​​of 1, 2, ..., m, where m is an integer greater than or equal to 2, and m represents the number of temperature zones on a wafer. Taking m equal to 5 as an example, as shown in Table 1, DR1 to DR5 represent the deposition rate vectors of the five temperature zones Z1 to Z5, respectively, and THK1 to THK5 represent the film thickness values ​​of the five temperature zones Z1 to Z5, respectively. Where DR1 = (THK1 / Dt) / THK M , ..., DR5 = (THK5 / Dt) / THK M .

[0044] In some embodiments, the gas flow vector in the data vector corresponding to the set of process data can be obtained based on the gas flow rate values ​​in the set of process data, the main process gas ratio in the process formulation to be optimized, and a preset gas flow rate calculation model. In some embodiments, the preset gas flow rate calculation model is determined by the gas flow rate values ​​in the set of process data and the main process gas ratio in the process formulation to be optimized. In some embodiments, the preset gas flow rate calculation model includes: MR = MFR / MFR M Where MR represents the gas flow vector in the data vector corresponding to this set of process data, and MFR represents the gas flow value in this set of process data. M This indicates the ratio of the main process gas in the process formulation to be optimized.

[0045] In some embodiments, a gas pressure vector can be obtained from the data vector corresponding to the set of process data based on the gas pressure values ​​in the set of process data, the gas pressure values ​​in the process formulation to be optimized, and a preset gas pressure calculation model. In some embodiments, the preset gas pressure calculation model is determined by the gas pressure values ​​in the set of process data and the gas pressure values ​​in the process formulation to be optimized. In some embodiments, the preset gas pressure calculation model includes: VP = Vac / Vac R Where VP represents the gas pressure vector in the data vector corresponding to this set of process data, and Vac represents the gas pressure value in this set of process data. R This indicates the gas pressure value in the process formulation to be optimized.

[0046] In some embodiments, the temperature change vector in the data vector corresponding to the set of process data can be determined based on the temperature change rate value in the set of process data. In some embodiments, if the temperature change rate value is equal to 0, that is, the temperature is constant, then the temperature change vector TR is equal to 1; if the temperature change rate value is less than 0, that is, the temperature is gradually decreasing, then the temperature change vector TR is equal to 2; if the temperature change rate value is greater than 0, that is, the temperature is gradually increasing, then the temperature change vector TR is equal to 3.

[0047] Based on this, the data vector R can be obtained. i ={T,DR1,DR2,DR3,DR4,DR5,MR,VP,TR}, where i is any value from 1, 2, ..., n, and n is the number of sets of process data that match the process formulation to be optimized, and n is an integer greater than or equal to 2. The five data vectors R1 to R5 corresponding to the five sets of process data Ru1 to Ru5 in Table 1 are shown in Table 2.

[0048] Table 2

[0049] Data vector T <![CDATA[DR1]]> <![CDATA[DR2]]> <![CDATA[DR3]]> <![CDATA[DR4]]> <![CDATA[DR5]]> MR VP TR <![CDATA[R1]]> 1.0 0.67 0.82 0.95 1.21 1.55 1.0 1.0 1.0 <![CDATA[R2]]> 1.0 0.82 0.94 1.02 1.12 1.16 1.0 1.0 1.0 <![CDATA[R3]]> 1.0 0.90 0.99 1.04 1.05 0.94 1.0 1.0 1.0 <![CDATA[R4]]> 0.95 1.26 1.11 1.01 0.93 0.84 0.75 0.5 1.0 <![CDATA[R5]]> 0.95 1.08 1.13 1.11 1.12 1.11 0.75 0.5 0.5

[0050] Then, the reference vector can be preset. Where K0, K1, K2, K3, K4, K5, K6, K7, K8 correspond to T, DR1, DR2, DR3, DR4, DR5, MR, VP, and TR respectively. K0, K1, K2, K3, K4, K5, K6, K7, and K8 are scaling factors for selecting data vectors. To simplify calculations, K0, K1, K2, K3, K4, K5, K6, K7, and K8 are usually set to 1, i.e.

[0051] Then, the data vector R can be calculated according to formula (1). i With reference vector distance in,

[0052]

[0053] z takes values ​​of 1, 2, ..., m in sequence. This is the sedimentary relation coefficient matrix calculated based on the Arrhenius formula. For example, if the focus is on calculating the temperature vector T, then α 11 =5, the rest are 0; if we focus on calculating both the temperature vector T and the deposition rate vector DR3, then α 11 =5, α 44 =5, and the rest are 0. Furthermore, if This set of process data matches the process formula to be optimized, and can be included in data set G1. Here, (0, D0] represents a preset distance range.

[0054] Taking the data vectors R1 to R5 in Table 2 as an example, the calculation is performed, and the depositional relationship coefficient matrix α is set. xy The elements on the main diagonal are taken as 5, 1, 1, 1, 1, 1, 5, 5, 5, respectively, while the remaining elements are taken as 0. The calculation yields... The values ​​are 0.5, 0.1, 0, 1.7, and 2.9, respectively, which correspond to the five data vectors R1 to R5 in Table 2 and the baseline vector. The distances are 0.5, 0.1, 0, 1.7, and 2.9 respectively. The process data Ru1 to Ru3 corresponding to the three smaller data vectors R1 to R3 are included in the data set G1, where G1 = {Ru1, Ru2, Ru3}. This data set G1 is the set of process data that matches the process formulation to be optimized.

[0055] In some embodiments, in order to further filter the process data in the data set G1 and avoid calculation deviations due to large deviations between the process data in the data set G1 and the process formula to be optimized, orthogonality checks are performed on multiple sets of process data in the data set G1 to obtain orthogonality check results. If the orthogonality check results meet the preset check requirements, the set of process data is determined as the process data that matches the process formula to be optimized.

[0056] Let G i ={T i1 ,T i2 ,T i3 ,T i4 ,…,T im}, where T i1 To T im Let m represent the temperature values ​​of the m temperature zones in the i-th set of process data, and then calculate the temperature difference for each zone. And so on, to obtain Then calculate GX i =X i ·XT i ―2 ·XT (i+1) ―2 If GX i If ≤GX0, then G i This refers to the data set that meets the preset verification requirements. Among them,

[0057]

[0058] Using the data in Table 1 as an example, we can calculate the GX corresponding to Ru1 to Ru3. i The values ​​are 0, 0, and 0.9997 respectively. Select GX. i The smaller Ru1 and Ru2 form a data set G2, where G2 = {Ru1, Ru2}. This data set G2 is the final set of process data that matches the process formulation to be optimized.

[0059] S103: Based on the acquired process data and the preset parameter calculation model that matches the optimization mode, calculate the optimized value of the parameter to be optimized that matches the optimization mode in the process formula to be optimized.

[0060] In some embodiments, the optimization mode is a constant temperature setting optimization mode or a variable temperature setting optimization mode. Based on this, if the optimization mode is a constant temperature setting optimization mode, the temperature optimization value of each temperature zone in the process formulation to be optimized is calculated based on the temperature value and film thickness value of each temperature zone in the acquired process data and the first parameter calculation model. If the optimization mode is a variable temperature setting optimization mode, the temperature change rate optimization value of each temperature zone in the process formulation to be optimized is calculated based on the temperature change rate value and intra-wafer uniformity value in the acquired process data and the second parameter calculation model.

[0061] In some embodiments, the first parameter calculation model is determined by the target temperature value in the optimization target, the target film thickness value of m temperature zones in the optimization target, the temperature value matrix of m temperature zones in n sets of process data, and the film thickness value matrix of m temperature zones in n sets of process data; the second parameter calculation model is determined by the optimized temperature change rate value of the middle (m+1) / 2 temperature zones in the process formulation to be optimized, the target intra-chip uniformity value of m temperature zones in the optimization target, the temperature change rate value matrix of m temperature zones in n sets of process data, and the intra-chip uniformity value matrix of m temperature zones in n sets of process data.

[0062] In some embodiments, the first parameter calculation model includes Where Z1 to Zm represent the optimized temperature values ​​for m temperature zones in the process formulation to be optimized, and T t This represents the target temperature value in the optimization objective.

[0063]

[0064] THK1 to THKm represent the target film thickness values ​​for m temperature zones in the optimization objective. This is a matrix of temperature values ​​for m temperature zones from n sets of process data. This is a matrix of film thickness values ​​for m temperature zones from n sets of process data.

[0065] Among them, it can be based on data relationships. get Then calculate Substitution

[0066] The above formula (2) is obtained. After normalizing the result of formula (2), it can be substituted into... Obtain Z1…Zm.

[0067] In some embodiments, the second parameter calculation model includes:

[0068]

[0069] Where Ra1 to Ra((m+1) / 2) are the optimized values ​​of the temperature change rate of the middle (m+1) / 2 temperature zones in the process formulation to be optimized, WIWT1 to WIWTm are the target intra-piece uniformity values ​​of the m temperature zones in the optimization target, and W′ is the generalized relation matrix constructed based on the generalized relation matrix W.

[0070] in,

[0071]

[0072] in, This is a matrix of temperature change rate values ​​for m temperature zones from n sets of process data. This is a matrix of intra-chip uniformity values ​​for m temperature zones in n sets of process data.

[0073] Among them, it can be based on data relationships. Obtain the generalized relation matrix

[0074] In some embodiments, the constructed generalized relation matrix Substituting the generalized relation matrix W′ into the above formula, Ra1 to Ra((m+1) / 2) can be calculated, and then the temperature change rate values ​​Ra1 to Ram for m temperature zones can be obtained.

[0075] Assuming m equals 5 and the target slice uniformity values ​​WIWT1 to WIWT5 are all equal to -0.005, using the data in Table 3 as an example, the following calculations are performed: Therefore, Ra1 to Ra3 can be calculated to be 1.8, 1.0, and 1.1 respectively, and Ra1 to Ra5 ​​can be calculated to be 1.8, 1.8, 1.0, 1.1, and 1.1 respectively. It is understandable that when the temperature difference between two adjacent temperature zones is small, their temperature change rates can be made equal. For example, the temperature change rate of the first and second temperature zones can both be set to 1.8, and the temperature change rate of the fourth and fifth temperature zones can both be set to 1.1.

[0076] Table 3

[0077]

[0078] In some embodiments, isothermal setting optimization is typically performed first. Based on the process results of the optimized process recipe executed by the semiconductor process equipment, it is determined whether the WTW (inter-wafer thickness uniformity) meets the requirements. Then, variable temperature setting optimization is performed, and based on the process results of the optimized process recipe executed by the semiconductor process equipment, it is determined whether the WIW (intra-wafer thickness uniformity) meets the requirements. In other embodiments, isothermal setting optimization and variable temperature setting optimization may be performed repeatedly, which will not be elaborated here. The process data of the optimized process recipe executed by the semiconductor process equipment is also stored in a process database for use in subsequent optimizations.

[0079] S104: If the process formulation to be optimized with an optimized value satisfies the optimization objective, then the optimized value shall be used as the value of the parameter to be optimized in the process formulation to be optimized.

[0080] The calculated optimized values ​​need to be verified before being updated to the process formulation to confirm whether they meet the expected requirements. Especially when preparing silicon nitride thin films, if a new cooling rate is adopted, it is necessary to avoid excessively large cooling ranges that could lead to drastic temperature changes on the wafer surface. Therefore, the cooling safety range is typically limited to (-Range, +Range), where Range represents the preset threshold of the cooling safety range. If the cooling safety range is exceeded, the cooling cycles are increased, and the temperature is gradually reduced through multiple cooling cycles.

[0081] In some embodiments, if the process formulation to be optimized with optimized values ​​satisfies the optimization objective, then using the optimized values ​​as the values ​​of the parameters to be optimized in the process formulation includes: if the product of the calculated optimized temperature change rate and the film formation time in the process formulation to be optimized is within the target temperature range in the optimization objective, then the optimized temperature change rate is used as the temperature change rate value in the process formulation to be optimized; if the product of the optimized temperature change rate and the film formation time in the process formulation to be optimized is not within the target temperature range in the optimization objective, the film formation time in the process formulation to be optimized is divided into multiple sub-film formation times, and if the product of any sub-film formation time and the optimized temperature change rate is within the target temperature range in the optimization objective, then the optimized temperature change rate is used as the temperature change rate value in the process formulation to be optimized, and the number of sub-film formation times is used as the number of temperature changes in the process formulation to be optimized. The target temperature range can be equal to 2Range.

[0082] Assuming the calculated optimal temperature change rate is Ramp, and the film formation time in the process formulation to be optimized is Depo Time, if Ramp*Depo Time ≤ 2Range, then the optimized temperature change rate value is used as the temperature change rate value in the process formulation to be optimized. If Ramp*Depo Time > 2Range, then let Depo Time′ = Depo Time / Count, where the value of Count can be gradually increased until Ramp*Depo Time′ ≤ 2Range. The final value of Count is equal to the number of sub-film formation times, and also equal to the number of temperature changes during the film formation process, such as the number of heating or cooling cycles.

[0083] In other words, with the temperature change rate (Ramp) remaining constant, by dividing a film deposition period into multiple sub-film deposition periods and a temperature change period into multiple temperature change sub-periods (e.g., dividing a 30-minute film deposition period into three 10-minute sub-film deposition periods and a 30-minute temperature change period into three 10-minute temperature change sub-periods), and ensuring a stabilization period (e.g., 2 minutes) between any two temperature change sub-periods, the duration of each individual temperature change sub-period (including the stabilization period) is effectively extended while maintaining a constant temperature change rate (Ramp). This helps prevent excessively large temperature variations during temperature change periods, which could lead to drastic temperature changes on the wafer surface and subsequent surface damage. The duration of the stabilization period can be set empirically and will not be elaborated upon here. It should be noted that even without adjusting the film deposition period, a new optimized temperature change rate value can be recalculated.

[0084] Subsequently, if the optimization mode is a constant-temperature setting optimization mode, the calculated temperature optimization values ​​for each temperature zone can be updated to the process recipe to be optimized; if the optimization mode is a variable-temperature setting optimization mode, the calculated temperature change rate optimization values ​​for each temperature zone can be updated to the process recipe to be optimized, thereby obtaining the standard process recipe for semiconductor process equipment. It should be noted that in this embodiment of the invention, only the optimization of temperature optimization values ​​under constant-temperature setting optimization mode and the optimization of temperature change rate optimization values ​​under variable-temperature setting optimization mode are used as examples for illustration. In other embodiments, other parameters in the process recipe can also be optimized, which will not be elaborated here.

[0085] The optimization method disclosed in this invention can automatically optimize process recipes using semiconductor process equipment, obtain standard process recipes for the semiconductor process equipment, improve the optimization efficiency and accuracy of process recipes, and reduce the optimization cost of process recipes. Furthermore, as shown in Table 4 and... Figure 3 As shown, after completing the target film thickness adjustment within 6 to 8 cycles, WIW≤1.5% and WTW≤1.5% can be achieved.

[0086] Table 4

[0087]

[0088] As an optional implementation of the disclosure of this invention, an embodiment of this invention discloses a semiconductor process apparatus, which includes a memory and a processor; the memory is used to store a computer program; the processor is used to execute the process recipe optimization method of the semiconductor process apparatus disclosed in any of the above embodiments according to the computer program stored in the memory. This semiconductor process apparatus includes semiconductor process equipment such as LPCVD.

[0089] In some embodiments, such as Figure 4 As shown, the semiconductor process equipment includes a Remote Group Control System (RGCS), which includes a memory and a processor; the memory is used to store a computer program; the processor is used to execute the process recipe optimization method for the semiconductor process equipment disclosed in any of the above embodiments according to the computer program stored in the memory.

[0090] The remote group control system can communicate with the industrial control systems of multiple semiconductor process equipment via a dedicated network. This industrial control system includes a host computer and slave computers. Specifically, the remote group control system can communicate with the host computers of multiple semiconductor process equipment via the production line's dedicated network. Therefore, the remote group control system can automatically collect process data from the semiconductor process equipment through the industrial control system. However, film thickness-related data, such as inter-wafer film thickness uniformity values ​​and intra-wafer film thickness uniformity values, still need to be imported from other film thickness measurement devices. Furthermore, after obtaining the optimized process formulation, the remote group control system will download it to the host computer of the corresponding semiconductor process equipment so that the semiconductor process equipment can prepare thin films based on the optimized process formulation. It can be understood that the remote group control system can optimize the process formulations of multiple semiconductor process equipment.

[0091] In this embodiment of the invention, remote automatic optimization of process formulations can be achieved, reducing the reliance on the personal experience of process personnel for process formulation optimization, reducing the impact of error risk, and improving the optimization efficiency of process formulations for semiconductor process equipment.

[0092] As an optional implementation of the disclosure of this invention, an embodiment of this invention discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the process formulation optimization method for semiconductor process equipment as disclosed in any of the above embodiments.

[0093] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0094] The above embodiments are merely illustrative of several implementation methods described in detail, but they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this specification, and these all fall within the protection scope of this specification. Therefore, the protection scope of this patent should be determined by the appended claims.

Claims

1. A method for optimizing the process formulation of semiconductor processing equipment, characterized in that, include: Receive optimization instructions, which include the process formulation to be optimized, its optimization objectives, and optimization modes; In response to the optimization instruction, process data matching the process formulation to be optimized is obtained from the process database; The process database stores multiple sets of process data collected during the preparation of the same thin film using the semiconductor process equipment. Based on the acquired process data and the preset parameter calculation model that matches the optimization mode, the optimized values ​​of the parameters to be optimized that match the optimization mode in the process formula to be optimized are calculated. If the process formulation to be optimized with the optimized value satisfies the optimization objective, then the optimized value is used as the value of the parameter to be optimized in the process formulation to be optimized.

2. The method according to claim 1, characterized in that, The process data obtained from the process database that matches the process formulation to be optimized includes: Obtain any set of process data from the process database; Based on the set of process data, the optimization objective, and the process formulation to be optimized, a data vector corresponding to the set of process data is obtained; Calculate the distance between the data vector and the preset reference vector; If the distance is within a preset range, then the set of process data is determined to be process data that matches the process formula to be optimized.

3. The method according to claim 2, characterized in that, The process of obtaining a data vector corresponding to the set of process data based on the set of process data, the optimization objective, and the process formulation to be optimized includes: Based on the temperature values ​​of the intermediate temperature zone in the set of process data, the target temperature value in the optimization objective, and the preset temperature calculation model, the temperature vector in the data vector corresponding to the set of process data is obtained. Based on the film thickness and film formation time of each temperature zone in the process data, the target film thickness in the optimization objective, and the preset deposition rate calculation model, the deposition rate vectors in the data vector corresponding to the process data are obtained. Based on the gas flow rate values ​​in the set of process data, the main process gas ratio in the process formula to be optimized, and the preset gas flow rate calculation model, the gas flow rate vector in the data vector corresponding to the set of process data is obtained. Based on the gas pressure values ​​in the set of process data, the gas pressure values ​​in the process formula to be optimized, and the preset gas pressure calculation model, the gas pressure vector in the data vector corresponding to the set of process data is obtained. Based on the temperature change rate value in the set of process data, determine the temperature change vector in the data vector corresponding to the set of process data.

4. The method according to claim 3, characterized in that, The preset temperature calculation model is determined by the temperature values ​​of the intermediate temperature zone in the set of process data and the target temperature value in the optimization objectives. The preset deposition rate calculation model is determined by the film thickness values ​​of each temperature zone in the set of process data, the film formation time in the set of process data, and the target film thickness value in the optimization target. The preset gas flow calculation model is determined by the gas flow value in the set of process data and the ratio of the main process gas in the process formula to be optimized. The preset gas pressure calculation model is determined by the gas pressure values ​​in the set of process data and the gas pressure values ​​in the process formula to be optimized.

5. The method according to claim 4, characterized in that, The preset temperature calculation model includes: T = Z ij / T t Where T represents the temperature vector, Z ij T represents the temperature value of the intermediate temperature zone in this set of process data. t This represents the target temperature value in the optimization objective; The preset deposition rate calculation model includes: DR z =(THK) z / Dt) / THK M , among which, DR z THK represents the various deposition rate vectors. z This represents the film thickness value for each temperature zone in this set of process data, Dt represents the film formation time in this set of process data, and THK represents the film thickness value for each temperature zone in this set of process data. M This represents the target film thickness in the optimization objective; The preset gas flow rate calculation model includes: MR = MFR / MFR M Where MR represents the gas flow vector, and MFR represents the gas flow value in this set of process data. M This indicates the ratio of the main process gas in the process formulation to be optimized; The preset gas pressure calculation model includes: VP = Vac / Vac R Where VP represents the gas pressure vector, and Vac represents the gas pressure value in this set of process data. R This indicates the gas pressure value in the process formulation to be optimized.

6. The method according to any one of claims 2 to 5, characterized in that, The process data used to determine that the set of process data matches the process formulation to be optimized includes: The orthogonality of this set of process data was checked, and the orthogonality check results were obtained. If the orthogonality check result meets the preset check requirements, then the set of process data is determined as the process data that matches the process formula to be optimized.

7. The method according to claim 1, characterized in that, The calculation of optimized values ​​for parameters matching the optimization mode in the process formulation to be optimized, based on the acquired process data and a preset parameter calculation model matching the optimization mode, includes: If the optimization mode is a constant temperature setting optimization mode, then based on the temperature value and film thickness value of each temperature zone in the acquired process data and the first parameter calculation model, the temperature optimization value of each temperature zone in the process formula to be optimized is calculated. If the optimization mode is a variable temperature setting optimization mode, then based on the temperature change rate value and intra-chip uniformity value in the acquired process data and the second parameter calculation model, the optimized temperature change rate value of each temperature zone in the process formulation to be optimized is calculated.

8. The method according to claim 7, characterized in that, The first parameter calculation model is determined by the target temperature value in the optimization target, the target film thickness value of m temperature zones in the optimization target, the temperature value matrix of m temperature zones in n sets of process data, and the film thickness value matrix of m temperature zones in n sets of process data. The second parameter calculation model is determined by the optimized temperature change rate values ​​of the middle (m+1) / 2 temperature zones in the process formulation to be optimized, the target intra-piece uniformity values ​​of the m temperature zones in the optimization target, the temperature change rate value matrix of the m temperature zones in the n sets of process data, and the intra-piece uniformity value matrix of the m temperature zones in the n sets of process data.

9. The method according to claim 8, characterized in that, The first parameter calculation model includes: Where Z1 to Zm represent the optimized temperature values ​​for m temperature zones in the process formulation to be optimized, and T t This represents the target temperature value in the optimization objective. Thk1 to Thkm represent the target film thickness values ​​for the m temperature zones in the optimization objective. This is a matrix of temperature values ​​for m temperature zones from n sets of process data. This is a matrix of film thickness values ​​for m temperature zones from n sets of process data. The second parameter calculation model includes: Where Ra1 to Ra((m+1) / 2) are the optimized temperature change rate values ​​for the middle (m+1) / 2 temperature zones in the process formulation to be optimized, WIWT1 to WIWTm are the target intra-piece uniformity values ​​for the m temperature zones in the optimization objective, and W′ is a generalized relation matrix constructed based on the generalized relation matrix W, where, in, This is a matrix of temperature change rate values ​​for m temperature zones from n sets of process data. This is a matrix of intra-chip uniformity values ​​for m temperature zones in n sets of process data.

10. The method according to claim 1, characterized in that, If the process formulation to be optimized with the optimized value satisfies the optimization objective, then using the optimized value as the value of the parameter to be optimized in the process formulation to be optimized includes: If the product of the calculated optimized temperature change rate and the film formation time in the process formulation to be optimized is within the target temperature range of the optimization target, then the optimized temperature change rate is taken as the temperature change rate value in the process formulation to be optimized. If the product of the optimized temperature change rate and the film formation time in the process formulation to be optimized is not within the target temperature range in the optimization objective, the film formation time in the process formulation to be optimized is divided into multiple sub-film formation times. If the product of any sub-film formation time and the optimized temperature change rate is within the target temperature range in the optimization objective, the optimized temperature change rate is taken as the temperature change rate value in the process formulation to be optimized, and the number of sub-film formation times is taken as the number of temperature changes in the process formulation to be optimized.

11. A semiconductor process apparatus, characterized in that, Including memory and processor; The memory is used to store computer programs; The processor is configured to execute the process formulation optimization method for semiconductor process equipment according to any one of claims 1-10, based on the computer program stored in the memory.

Images