Method and apparatus for determining electrical process parameters of a semiconductor device

Abstract

The application provides a semiconductor device electrical process parameter determination method and device, and belongs to the technical field of device manufacturing. The semiconductor device electrical process parameter determination method comprises the following steps: obtaining device manufacturing process parameters and current equipment state data; based on the device manufacturing process parameters, a preset process electrical property correlation model is used to determine an electrical property parameter, the preset process electrical property correlation model being used for determining associated device electrical property parameter information according to device manufacturing process parameter information; based on the current equipment state data, a preset uncertainty model is used to determine a process electrical property fluctuation range, the preset uncertainty model being used for determining the influence of equipment state on process fluctuation; and based on the electrical property parameter and the process electrical property fluctuation range, a device electrical property process parameter is obtained. The working efficiency is greatly improved, the determination process is simple and convenient, and the device can be reused.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of device manufacturing, in particular to a semiconductor device electrical process parameter determination method, a semiconductor device electrical process parameter determination device, a machine readable storage medium and an electronic device. BACKGROUND

[0002] The integrated circuit manufacturing process is precise and complex, and the device manufacturing process needs to go through hundreds of processes. The adjustment of parameters of each process will have more or less influence on the final device electrical property. Therefore, to obtain a device with excellent electrical property, the process parameters need to be continuously improved and iteratively optimized for multiple rounds.

[0003] Device electrical process parameters refer to a set of parameters used to describe the electrical performance and characteristics of a semiconductor device during manufacturing. These parameters reflect the electrical performance of the device under different working conditions, and are of great significance for evaluating device performance, optimizing process design and adjusting device performance.

[0004] Currently, when determining device electrical process parameters, theoretical parameters are adjusted to obtain the parameters. However, the adjustment of theoretical parameters often deviates from the expected results, which increases the number of experiments to continuously modify the results to obtain the final device electrical process parameters.

[0005] Therefore, the existing device electrical process parameter determination has the problems of complicated process and low efficiency. SUMMARY

[0006] The purpose of the embodiments of the present application is to provide a semiconductor device electrical process parameter determination method, a semiconductor device electrical process parameter determination device, a machine readable storage medium and an electronic device, which can provide device electrical process parameters that meet current process parameters, facilitate quick determination of device electrical process parameters, greatly improve work efficiency, and determine the process simply and conveniently.

[0007] To achieve the above purpose, the first aspect of the present application provides a semiconductor device electrical process parameter determination method, comprising:

[0008] Obtain device manufacturing process parameters and current equipment state data, the device manufacturing process parameters are various parameters used to control the characteristics and performance of semiconductor devices in the semiconductor device manufacturing process, and the current equipment state data is the state data of the equipment used in the current semiconductor device manufacturing process;

[0009] Based on the device manufacturing process parameters, a preset process electrical property correlation model is used to determine an electrical property parameter, the preset process electrical property correlation model being used to determine associated device electrical property parameter information according to device manufacturing process parameter information;

[0010] Based on the current device state data, a preset uncertainty model is used to determine a process electrical property fluctuation range, the preset uncertainty model being used to determine the influence of a device state on a process fluctuation according to the device state;

[0011] Based on the electrical property parameter and the process electrical property fluctuation range, a device electrical property process parameter is obtained, the device electrical property process parameter being used to describe electrical property performance and characteristics of a semiconductor device.

[0012] In the embodiments of the present application, the construction process of the preset uncertainty model includes:

[0013] Device historical process fluctuation data is obtained, the device historical process fluctuation data including multiple sets of fluctuation data, each set of fluctuation data including historical device state data and corresponding process fluctuation data;

[0014] Data mining is performed on the device historical process fluctuation data to obtain an association relationship, the association relationship being an association relationship between a device state and a process fluctuation;

[0015] Based on the association relationship, an uncertainty envelope model is used to describe the influence of a device state on a device manufacturing process fluctuation to obtain an uncertainty model.

[0016] In the embodiments of the present application, the uncertainty model is obtained by using the uncertainty envelope model to describe the influence of a device state on a device manufacturing process fluctuation, including:

[0017] Based on the association relationship, an uncertainty set U is constructed;

[0018] An uncertainty envelope model is used to describe the range of the uncertainty set U to obtain the uncertainty model, the uncertainty envelope model being: a linear programming problem is constructed as follows:

[0019] Wherein, x∈R n×1 , c∈R n×1 , X represents a bounded feasible region of a decision variable x, is a coefficient matrix of a constraint left end containing an uncertain parameter, is a constraint right end uncertainty parameter vector, R is a real number domain, m and n respectively represent different dimensions, is an element in , and is an element in the elements in the set, for the ith constraint, let and respectively:

[0020]

[0021] where a ij and b i are the nominal parts of the uncertain parameters, which are the device state data, and are the floating amplitudes, obtained from the association relationship, ξ ij and ξ i0 are the unit variables of the fluctuations, J i is the set of parameters affected by uncertainty in the ith constraint, and represent the upper and lower fluctuation ranges of the ith constraint, respectively, that is, the size of the influence of the device state on the process fluctuation.

[0022] In the embodiments of the present application, the process electrical property fluctuation range is determined based on the current device state data using a pre-set uncertainty model, including:

[0023] Based on the current device state data, a pre-set uncertainty model is used to determine the range description of the uncertainty set of the current device.

[0024] Based on the range description of the uncertainty set of the current device, the corresponding constraint upper and lower fluctuation ranges are determined.

[0025] Based on the corresponding constraint upper and lower fluctuation ranges, the process electrical property fluctuation range is obtained.

[0026] In the embodiments of the present application, the device historical process fluctuation data is mined to obtain the association relationship, including:

[0027] The XGBoost algorithm is used to mine the device historical process fluctuation data to obtain the association relationship.

[0028] In the embodiments of the present application, the construction process of the pre-set process electrical property related model includes:

[0029] A plurality of sample data are obtained, each sample data including sample process parameters and corresponding device basic parameters;

[0030] Based on the plurality of sample data, a deep learning algorithm is used for model training to obtain a process electrical property related model.

[0031] In the embodiments of the present application, the plurality of sample data are obtained, including:

[0032] obtaining device history data, the device history data comprising one or more of device history simulation data, device history DOE data, and device history electrical performance data;

[0033] performing feature extraction on the device history data to obtain a plurality of sample data.

[0034] In the embodiments of the present application, further comprising:

[0035] performing simulation based on the device manufacturing process parameters using a preset simulation system to obtain device electrical simulation parameters;

[0036] comparing the device electrical simulation parameters with the device electrical process parameters to obtain a comparison result.

[0037] The second aspect of the present application provides a semiconductor device electrical process parameter determination apparatus, comprising:

[0038] an obtaining module configured to obtain device manufacturing process parameters and current equipment state data, the device manufacturing process parameters being various parameters used to control characteristics and performance of semiconductor devices in a semiconductor device manufacturing process, and the current equipment state data being state data of equipment used in a current semiconductor device manufacturing process;

[0039] a parameter calculation module configured to determine electrical parameters based on the device manufacturing process parameters using a preset process electrical correlation model, the preset process electrical correlation model being used to determine associated device electrical parameter information according to device manufacturing process parameter information;

[0040] a fluctuation calculation module configured to determine a process electrical fluctuation range based on the current equipment state data using a preset uncertainty model, the preset uncertainty model being used to determine the influence of equipment state on process fluctuation according to the equipment state;

[0041] a determination module configured to obtain device electrical process parameters based on the electrical parameters and the process electrical fluctuation range, the device electrical process parameters being used to describe electrical performance and characteristics of semiconductor devices.

[0042] The third aspect of the present application provides an electronic device, comprising:

[0043] at least one processor;

[0044] a memory connected with the at least one processor;

[0045] wherein the memory stores instructions executable by the at least one processor, and the at least one processor implements the semiconductor device electrical process parameter determination method described above by executing the instructions stored in the memory.

[0046] The fourth aspect of the present application provides a machine readable storage medium, which stores instructions, when the instructions are executed by a processor, the processor is configured to perform the above semiconductor device electrical property parameter determination method.

[0047] Through the above technical solution, the device manufacturing process parameters and the current equipment state data are obtained; based on the device manufacturing process parameters, a preset process electrical property correlation model is used to determine the electrical property parameters, the preset process electrical property correlation model is used to determine the associated device electrical property parameter information according to the process parameter information; based on the current equipment state data, a preset uncertainty model is used to determine the process electrical property fluctuation range, the preset uncertainty model is used to determine the influence of the equipment state on the process fluctuation according to the equipment state; based on the electrical property parameters and the process electrical property fluctuation range, the device electrical property process parameters are obtained. The fluctuation of the process parameters can be obtained through the device manufacturing process parameters and the current equipment state data, the device electrical property process parameters that can meet the current process parameters are provided, the device electrical property process parameters can be quickly determined, compared with the existing scheme of increasing the number of experiments to adjust the parameters, the work efficiency is greatly improved, and the determination process is simple and convenient. At the same time, based on the current equipment state data, the process electrical property fluctuation range is determined, which considers the uncertainty factors caused by the equipment in the key manufacturing process, so that the obtained device electrical property process parameters are more accurate and can be reused. The device electrical property parameter information is obtained through the device manufacturing process parameters, so that the device electrical property can be estimated according to the process parameters, which is helpful for the research and design of the device.

[0048] Other features and advantages of the present application will be illustrated in the following detailed description. BRIEF DESCRIPTION OF DRAWINGS

[0049] The accompanying drawings are included to provide a further understanding of embodiments of the application, and constitute a part of the specification, and are used to explain the embodiments of the application together with the following detailed description, but do not constitute a limitation on the embodiments of the application. In the drawings:

[0050] Figure 1 The flowchart of a semiconductor device electrical property process parameter determination method according to an embodiment of the present application is schematically shown;

[0051] Figure 2 The generation step diagram of a semiconductor device electrical property process parameter according to an embodiment of the present application is schematically shown;

[0052] Figure 3 The structure diagram of a semiconductor device electrical property process parameter determination device according to an embodiment of the present application is schematically shown;

[0053] Figure 4 Fig. 1 schematically shows an internal structure diagram of a computer device according to an embodiment of the present application.

[0054] Legend of reference signs

[0055] 410 - acquisition module; 420 - parameter calculation module; 430 - fluctuation calculation module; 440 - determination module; A01 - processor; A02 - network interface; A03 - internal memory; A04 - display screen; A05 - input device; A06 - non-volatile storage medium; B01 - operating system; B02 - computer program. DETAILED DESCRIPTION

[0056] The specific embodiments of the present application will be described below in detail with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are merely intended to illustrate and explain the present application, and are not intended to limit the present application.

[0057] It should be noted that if the present application has any reference to directional indications (such as up, down, left, right, front, back, etc.), the directional indications are only used to explain the relative position relationship, movement condition, etc. between components in a certain posture (as shown in the drawings), and if the certain posture changes, the directional indications will also change accordingly.

[0058] In addition, if the present application has any description of "first", "second", etc., the description of "first", "second", etc. is only for description purposes, and cannot be understood as indicating or implying the relative importance of the indicated technical features or implicitly indicating the number of the indicated technical features. Therefore, the features limited by "first", "second" can explicitly or implicitly include at least one of the features. In addition, the technical solutions of each embodiment can be combined with each other, but it must be based on the realization of a person skilled in the art, and when the combination of technical solutions appears to be contradictory or unachievable, it should be considered that the combination of technical solutions does not exist, and is not within the protection scope claimed by the present application.

[0059] It should be noted that the equipment mentioned in the present embodiment refers to the equipment used in the process of manufacturing the device.

[0060] Please refer to Figure 1 , Figure 1 Fig. 1 schematically shows a flow diagram of a semiconductor device electrical process parameter determination method according to an embodiment of the present application. The present embodiment provides a semiconductor device electrical process parameter determination method, which comprises the following steps:

[0061] Step 210: obtaining device manufacturing process parameters and current equipment state data, the device manufacturing process parameters being various parameters used to control the characteristics and performance of semiconductor devices in the semiconductor device manufacturing process, and the current equipment state data being state data of equipment used in the current semiconductor device manufacturing process;

[0062] In the embodiment, the device manufacturing process parameters can be obtained from a device design of experiments (DOE) document or a device DOE table. Device DOE is a technique used in semiconductor device design and manufacturing to help optimize device performance and process parameters using experimental design methods. Through device DOE, a series of experiments can be designed to systematically study the impact of different factors on device performance, such as material selection, process conditions, size parameters, etc. Through statistical analysis and experimental data, it can be determined which factors play an important role in device performance and how to adjust these factors to achieve optimal performance.

[0063] The device manufacturing process parameters refer to various parameters used to control the characteristics and performance of semiconductor devices in the semiconductor device manufacturing process, including key process parameters in the device manufacturing process, such as ion implantation energy, dosage, angle, or heating process parameters. The current equipment state data refers to the state data of the equipment used in the current semiconductor device manufacturing process. The equipment state data refers to data recording various parameters and state information of the process equipment in the production process to monitor and control the equipment operating condition and ensure the normal operation of the production process. These data can include the values of various physical quantities such as the running time, temperature, pressure, speed, current, voltage of the equipment, as well as the failure information, alarm information, maintenance records and other related information of the equipment. By analyzing these data, equipment operation anomalies can be found in time, equipment failures can be predicted, production parameters can be optimized, and production efficiency and product quality can be improved. The current equipment state data can be obtained by real-time collection from each equipment or by direct input by personnel.

[0064] Step 220: based on the device manufacturing process parameters, determining the electrical property parameters using a pre-set process electrical property correlation model, the pre-set process electrical property correlation model being used to determine the associated device electrical property parameter information according to the device manufacturing process parameter information;

[0065] In the embodiment, the preset process electrical property correlation model is established in advance. For each type of device, a process electrical property correlation model corresponding to the device type is established. In a specific implementation, a corresponding process electrical property correlation model can be selected according to a device type to be manufactured. For example, a CMOS device corresponds to a process electrical property correlation model of the CMOS device, and a photoelectric device corresponds to a process electrical property correlation model of the photoelectric device. The process electrical property correlation model is a model that can calculate device performance through a process. The model can be trained by using a deep learning algorithm.

[0066] In some embodiments, in order to obtain an accurate process electrical property correlation model, a construction process of the preset process electrical property correlation model includes the following steps.

[0067] First, a plurality of sample data are obtained, each sample data including sample process parameters and corresponding device basic parameters.

[0068] In the embodiment, the sample data refer to sample data of the device, the sample process parameters refer to key process parameters in a device manufacturing process, and the device basic parameters refer to device performance parameters, such as a saturation threshold voltage, a linear threshold voltage, a saturation current, a leakage current, and the like, under the key process parameters. The number of sample data is not limited. The more the sample data, the more accurate the sample data, and the better the performance of the process electrical property correlation model obtained through training.

[0069] In order to obtain more accurate sample data, the plurality of sample data are obtained by:

[0070] First, device historical data are obtained, the device historical data including one or more of device historical simulation data, device historical DOE data, and device historical electrical performance data.

[0071] In the embodiment, the device historical DOE data can be obtained from a device historical DOE document, the device historical simulation data can be obtained from a simulation system, and the device historical electrical performance data can be obtained from a wafer acceptance test (WAT) test. The WAT is used to detect the quality and performance of a wafer in a manufacturing process, including an electrical test: testing electrical characteristics on a wafer, such as resistance, capacitance, and the like. An optical test: checking patterns, figures, and defects on a wafer. A characteristic test: testing the performance and function of a device on a wafer. A comprehensive test: comprehensively testing various performance indicators of a wafer to ensure that the quality of the wafer meets production requirements.

[0072] Second, feature extraction is performed on the device historical data to obtain a plurality of sample data.

[0073] In the embodiment, the feature extraction mentioned above refers to extracting process parameters and corresponding device basic parameters, and the feature extraction can be realized by using existing technologies, which will not be described herein.

[0074] By performing feature extraction from device history data including device history simulation data, device history DOE data or device history electrical performance data, more accurate sample data can be obtained, which helps to train a process electrical property related model with better performance.

[0075] Then, based on the plurality of sample data, a deep learning algorithm is used to train a model to obtain a process electrical property related model.

[0076] In the embodiment, the deep learning algorithm includes a deep neural network algorithm, a convolutional neural network algorithm, etc., and the learning and recognition of complex patterns and relationships are realized through multi-level nonlinear transformation and feature extraction. The model is trained by using a plurality of sample data, so that the model can determine the associated device electrical property parameter information according to the process parameter information.

[0077] In specific implementation, a fully connected network can be used to construct a model, and the output is a device basic parameter, such as a saturation threshold voltage Vtsat, a linear threshold voltage Vtlin, a saturation current Idsat, a leakage current Ioff, etc. The network input is a part of a key process parameter in a device manufacturing process, i.e., a device manufacturing process parameter, such as an ion implantation energy, a metering, an angle, or a heating process parameter. The network includes an input layer, a hidden layer and an output layer, and the nodes between each layer are fully connected. Through the training of a large number of sample data, a neural network model that can calculate the device performance by the process is finally trained. The process of training by using the fully connected network belongs to the prior art, which will not be described herein.

[0078] The deep learning algorithm can process large-scale, high-dimensional data and solve complex pattern recognition and decision-making problems. By using the deep learning algorithm to train a model, a more accurate process electrical property related model can be obtained.

[0079] Step 230: based on the current device state data, a preset uncertainty model is used to determine a process electrical property fluctuation range, and the preset uncertainty model is used to determine the influence of the device state on the process fluctuation according to the device state;

[0080] In the embodiment, the preset uncertainty model can be pre-established, and for each type of device, a corresponding uncertainty model is established. In specific implementation, the corresponding uncertainty model can be selected according to the type of device to be manufactured. The uncertainty model is used to determine the influence of the device state on the process fluctuation.

[0081] In some embodiments, in order to accurately describe the influence of the device state on the process fluctuation, the construction process of the preset uncertainty model comprises:

[0082] First, the device historical process fluctuation data is acquired, the device historical process fluctuation data comprising a plurality of sets of fluctuation data, each set of fluctuation data comprising historical device state data and corresponding process fluctuation data;

[0083] In the embodiment, the device historical process fluctuation data comprises fluctuation data of the device in the historical manufacturing process and the device state causing the process fluctuation. For example, the fluctuation data is that, from the time point of view, the device has been on the market for x years, has been operated at high intensity for y years, has not been maintained for z years, and has a historical number of a times of unannounced failure in the factory. These are the device states. The historical process fluctuation is affected by the size of the device state (x, y, z, a). The device historical process fluctuation data can be obtained from a manufacturing execution system (MES). The MES is a software system for monitoring and managing various links in the manufacturing process, which helps to monitor the production process in real time, collect and analyze production data, coordinate production plans, optimize resource utilization, and the like, so as to improve production efficiency, reduce production cost, and ensure product quality and compliance with regulations.

[0084] Then, data mining is performed on the device historical process fluctuation data to obtain an association relationship, the association relationship being an association relationship between the device state and the process fluctuation.

[0085] In the embodiment, the data mining is performed by analyzing a large amount of data to find potential rules, trends and patterns hidden therein, so as to find the association relationship between the process fluctuation data and the uncertainty data. The uncertainty data refers to the device state data. The data mining can be performed by using an optimized distributed gradient boosting library (XGBoost) algorithm, a random forest algorithm, or the like.

[0086] In some embodiments, the data mining on the device historical process fluctuation data to obtain the association relationship comprises: performing data mining on the device historical process fluctuation data by using the XGBoost algorithm to obtain the association relationship.

[0087] In the embodiment, in order to quickly obtain the accurate association relationship between the device state and the process fluctuation, the XGBoost algorithm can be used for data mining. The XGBoost algorithm is an ensemble learning algorithm based on decision trees, which has high efficiency and accuracy, so that the accurate management relationship can be quickly obtained.

[0088] In a specific implementation, some initial parameters are set first, including learning rate, number of trees, etc., and then iterative training is performed according to the device historical process fluctuation data and the objective function to gradually optimize the model parameters. The residuals are fitted by calculating the first-order derivative and the second-order derivative (Hessian matrix) of the loss function of each sample. Therefore, for the splitting of each tree node, the optimal split is made based on the gradient and Hessian matrix of the loss function. The greedy algorithm is used for tree growth, that is, the optimal split point is selected recursively from top to bottom to minimize the loss function. Through the greedy algorithm, XGBoost can efficiently build a tree model. In XGBoost, each leaf node predicts a numerical value, which represents the predicted value corresponding to the leaf node, and the final prediction result is obtained by summing the leaf nodes. XGBoost iteratively builds trees, trains new tree models according to the residuals each time, and adds the prediction results of each tree weightedly to obtain a learner. With this learner, the device historical process fluctuation data can be regressed to mine and fit the correlation of uncertain data, that is, the correlation between uncertain data and process fluctuations. It should be noted that the above process of data mining using XGBoost algorithm belongs to the prior art, and will not be described in detail here.

[0089] Finally, based on the correlation, an uncertainty envelope model is used to describe the influence of the device state on the process fluctuation to obtain an uncertainty model.

[0090] In this embodiment, after obtaining the correlation, in order to describe the influence of the device state on the process fluctuation, an uncertainty envelope model can be used for modeling. In the uncertainty envelope model, each data point of the data set is represented as an envelope rather than a certain point. This uncertainty of data can be caused by measurement error, ambiguity or other uncertain factors.

[0091] Specifically, through data mining, the correlation between the device state and the process fluctuation is obtained, and based on the correlation, an uncertainty set U is constructed.

[0092] Then, the uncertainty set U is described in range by using an uncertainty envelope model to obtain an uncertainty model, and the uncertainty envelope model is:

[0093] where x∈R n×1 , c∈R n×1 , X represents the bounded feasible region of the decision variable x, is a coefficient matrix containing uncertain parameters in the left end of the constraint, is the constraint vector, R is the real number field, and m and n represent different dimensions, respectively.

[0094] For any one constraint, such as the i-th constraint, is the element in is the element in is the element in is the element in and are respectively:

[0095]

[0096] where a ij and b i are the nominal parts of the uncertain parameters, and the uncertain parameters here refer to the equipment state data, and are the floating amplitudes, and the floating amplitudes are used for the convenience of unified description of the upper and lower fluctuation ranges, and the upper and lower fluctuation ranges can be obtained through the correlation relationship, ξ ij and ξ i0 are the unit variables of the fluctuations, which can be set in advance and are required when constructing the paradigm, and the upper and lower fluctuation ranges are expressed through the unit variables through an amplitude. J i is the parameter set affected by uncertainty in the i-th constraint, where, and represent the upper and lower fluctuation ranges of any one constraint, i.e., the influence of the equipment state on the process fluctuation.

[0097] Based on the above, the mathematical expression of the uncertainty set can be constructed according to the actual fluctuation of the equipment through different norms. Specifically, the uncertainty set can be constructed, and the range description of the uncertainty set can be constructed according to different norms. For example, the uncertainty set mathematical expression with the box type uncertainty set geometry can be used, i.e., by setting the adjustable parameter Ψ as the upper limit, so that all the unit variables ξ ij of the fluctuations do not exceed Ψ; the uncertainty set mathematical expression with the ellipsoid type uncertainty set geometry can also be used, i.e., by setting the adjustable parameter Ω as the upper limit, so that the sum of squares of all ξ ij does not exceed Ω 2 ; a polyhedron type can also be constructed. Specifically, the most suitable situation for the construction of the upper and lower limits can be selected according to the characteristics of the uncertainty set in the uncertainty set mathematical expression mode, and the range needs to be adjusted appropriately according to the demand of robustness to accommodate more low-frequency points.

[0098] The uncertainty set of the key equipment is constructed by the historical process fluctuation data of the device to determine the upper and lower limits of the equipment influence within the acceptable range, the multi-dimensional uncertainty set is described by the envelope method, and the uncertainty of each equipment is counted by data mining, so that the upper and lower limits of the influence of each equipment can be mapped to the influence on the electrical property of the device and the total influence is counted.

[0099] In some embodiments, the process electrical property fluctuation range is determined based on the current equipment state data and a preset uncertainty model, including the following steps:

[0100] Firstly, the range description of the uncertainty set of the current equipment is determined based on the current equipment state data and the preset uncertainty model.

[0101] In this embodiment, the preset uncertainty model describes the range of the correlation between the equipment state data and the process fluctuation, so that the range description of the uncertainty set of the current equipment can be obtained by substituting the current equipment state data, i.e. the range description of the current equipment state data.

[0102] Then, the corresponding constraint upper and lower limit fluctuation range is determined based on the range description of the uncertainty set of the current equipment.

[0103] Finally, the process electrical property fluctuation range is obtained based on the corresponding constraint upper and lower limit fluctuation range.

[0104] In this embodiment, the influence of the current equipment state on the process fluctuation is constructed according to different norms to describe the range of the uncertainty set, so that the corresponding norm is the process electrical property fluctuation range. By reading the current equipment state data, the uncertainty set of the current equipment can be obtained, and the corresponding norm is determined, i.e. the corresponding constraint upper and lower limit fluctuation range is obtained, so that the electrical property upper and lower fluctuation percentage is obtained, i.e. the process electrical property fluctuation range.

[0105] Step 240: obtaining the device electrical property process parameter based on the electrical property parameter and the process electrical property fluctuation range.

[0106] In this embodiment, the above-mentioned device electrical property process parameter is obtained by taking the process electrical property fluctuation range as the fluctuation range of the electrical property parameter. For example, the electrical property parameters include V1 and V2, and the above-mentioned process electrical property fluctuation range is the electrical property upper and lower fluctuation percentage, which can be expressed as ±10%, so that based on the electrical property parameter and the process electrical property fluctuation range, the device electrical property process parameter is V1±10% and V2±10%.

[0107] In the implementation process, the device manufacturing process parameters and the current equipment state data are acquired; based on the device manufacturing process parameters, a preset process electrical property correlation model is used to determine an electrical property parameter, the preset process electrical property correlation model being used to determine the associated device electrical property parameter information according to the process parameter information; based on the current equipment state data, a preset uncertainty model is used to determine a process electrical property fluctuation range, the preset uncertainty model being used to determine the influence of the equipment state on the process fluctuation according to the equipment state; and based on the electrical property parameter and the process electrical property fluctuation range, a device electrical property process parameter is obtained. The device manufacturing process parameters and the current equipment state data can be used to obtain the fluctuation of the process parameter, provide the device electrical property process parameter that can meet the current process parameter, facilitate the rapid determination of the device electrical property process parameter, greatly improve the work efficiency compared with the existing scheme of increasing the number of experiments to adjust the parameter, and the determination process is simple and convenient. Meanwhile, based on the current equipment state data, the process electrical property fluctuation range is determined, which considers the uncertainty factors caused by the equipment in the key manufacturing process, so that the obtained device electrical property process parameter is more accurate and can be reused. The device electrical property parameter information is obtained through the device manufacturing process parameter, so that the device electrical property can be estimated according to the process parameter, which is helpful for the research and development and design of the device.

[0108] In some embodiments, the method further comprises the following steps:

[0109] First, based on the device manufacturing process parameters, a preset simulation system is used for simulation to obtain a device electrical property simulation parameter;

[0110] In this embodiment, the simulation system can be used for simulation to obtain the device electrical property simulation parameter obtained based on the device manufacturing process parameters. The simulation system can also be used to evaluate other performances of the device. The simulation coefficient belongs to the prior art, and will not be described here.

[0111] Then, the device electrical property simulation parameter and the device electrical property process parameter are compared to obtain a comparison result.

[0112] In this embodiment, by comparing the device electrical property simulation parameter and the device electrical property process parameter, the comparison result can be the same or different. The same indicates that the device electrical property process parameter obtained by using the scheme is correct, and otherwise, it can be further evaluated.

[0113] By comparing the device electrical property simulation parameter and the device electrical property process parameter, the accuracy of the device electrical property process parameter can be evaluated, so as to ensure the reliability of the device electrical property process parameter.

[0114] For the convenience of the scheme, the following CMOS device electrical process parameter determination process is described in detail. Please refer to Figure 2 , Figure 2 The semiconductor device electrical process parameter generation step schematic diagram according to the embodiment of the application is schematically shown.

[0115] The relevant data including device DOE historical data, historical simulation data, electrical performance historical data and process engineering fluctuation historical data are extracted from the DOE document, the simulation system, the WAT and the MES system through the data acquisition module and stored in the database. The device DOE historical data, the historical simulation data and the electrical performance historical data are combined, and the process-electrical property correlation model is constructed through the experience calculation and the DNN deep learning method, corresponding to the process-electrical property correlation integration module. The process engineering fluctuation historical data is described by the envelope model and constructed by the data mining method to form an uncertainty set; this part corresponds to the uncertainty model establishment module. The above two models are integrated to obtain the device electrical property fluctuation calculation module, so as to obtain the device electrical process parameter generation device. The device electrical process reference parameter can be obtained by inputting the device DOE table and the equipment state into the device electrical process parameter generation device. Meanwhile, the device DOE table can be input into the simulation system for simulation, and the simulation result can be input into the device electrical process parameter generation device for comparison.

[0116] The embodiment provides a semiconductor device electrical process parameter determination device, please refer to Figure 3 , Figure 3 The structure schematic diagram of the semiconductor device electrical process parameter determination device according to the embodiment of the application is schematically shown. The semiconductor device electrical process parameter determination device comprises an acquisition module 410, a parameter calculation module 420, a fluctuation calculation module 430 and a determination module 440, wherein:

[0117] The acquisition module 410 is used for acquiring device manufacturing process parameters and current equipment state data. The device manufacturing process parameters are various parameters used for controlling the characteristics and performance of semiconductor devices in the semiconductor device manufacturing process. The current equipment state data is the state data of the equipment used in the current semiconductor device manufacturing process.

[0118] The parameter calculation module 420 is used for determining the electrical property parameters based on the device manufacturing process parameters and using the preset process-electrical property correlation model. The preset process-electrical property correlation model is used for determining the associated device electrical property parameter information according to the device manufacturing process parameter information.

[0119] The fluctuation calculation module 430 is configured to determine a process electrical fluctuation range based on the current device state data and a preset uncertainty model, the preset uncertainty model being used to determine an influence of the device state on the process fluctuation.

[0120] The determination module 440 is configured to obtain a device electrical process parameter based on the electrical parameter and the process electrical fluctuation range, the device electrical process parameter being used to describe electrical performance and characteristics of the semiconductor device.

[0121] The device electrical process parameter determination apparatus includes a processor and a memory, and the acquisition module 410, the parameter calculation module 420, the fluctuation calculation module 430 and the determination module 440 are all stored in the memory as program units, and the processor is configured to execute the program units stored in the memory to realize the corresponding functions.

[0122] The processor includes a core, and the core is configured to call the corresponding program units from the memory. The core can be one or more, and the core parameters are adjusted to quickly determine the device electrical process parameter.

[0123] The memory can include a non-permanent memory in a computer readable medium, a random access memory (RAM) and / or a non-volatile memory such as a read-only memory (ROM) or a flash memory (flash RAM), and the memory includes at least one memory chip.

[0124] The embodiment of the present application provides a machine readable storage medium, and a program is stored in the machine readable storage medium, and the program is executed by a processor to realize the semiconductor device electrical process parameter determination method.

[0125] The embodiment of the present application provides a processor, and the processor is used to run a program, and the program is executed to perform the semiconductor device electrical process parameter determination method.

[0126] In one embodiment, a computer device can be provided, which can be a terminal, and an internal structure diagram of the computer device can be as shown in FIG. 2. Figure 4As shown in the figure, the computer device includes a processor A01, a network interface A02, a display screen A04, an input device A05, and a memory (not shown) connected via a system bus. The processor A01 provides computing and control capabilities. The memory includes internal memory A03 and a non-volatile storage medium A06. The non-volatile storage medium A06 stores an operating system B01 and a computer program B02. The internal memory A03 provides an environment for the operation of the operating system B01 and the computer program B02 stored in the non-volatile storage medium A06. The network interface A02 is used for communication with external terminals via a network connection. When the computer program is executed by the processor A01, it implements a method for determining the electrical process parameters of a semiconductor device. The display screen A04 can be a liquid crystal display (LCD) or an e-ink display. The input device A05 can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0127] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0128] In one embodiment, the device electrical process parameter determination apparatus provided in this application can be implemented as a computer program, which can be implemented in the form of, for example... Figure 4 The computer device shown runs on this device. The computer device's memory can store the various program modules that make up the electrical process parameter determination device for this device, for example... Figure 3 The diagram shows an acquisition module 410, a parameter calculation module 420, a fluctuation calculation module 430, and a determination module 440. The computer program comprised of these modules causes the processor to execute the steps of the file system invocation methods described in the various embodiments of this application.

[0129] Figure 4 The computer device shown can be used as follows Figure 3 The device electrical process parameter determination device shown in the figure executes step 210 in the acquisition module 410, step 220 in the parameter calculation module 420, step 230 in the fluctuation calculation module 430, and step 240 in the determination module 440.

[0130] This application provides a device, which includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it performs the following steps:

[0131] obtaining device manufacturing process parameters and current equipment state data, the device manufacturing process parameters being various parameters used for controlling semiconductor device characteristics and performance in a semiconductor device manufacturing process, and the current equipment state data being state data of equipment used in a current semiconductor device manufacturing process;

[0132] determining electrical property parameters based on the device manufacturing process parameters and using a preset process electrical property correlation model, the preset process electrical property correlation model being used to determine associated device electrical property parameter information according to process parameter information;

[0133] determining a process electrical property fluctuation range based on the current equipment state data and using a preset uncertainty model, the preset uncertainty model being used to determine the influence of equipment state on process fluctuation according to equipment state;

[0134] obtaining device electrical property process parameters based on the electrical property parameters and the process electrical property fluctuation range, the device electrical property process parameters being used to describe electrical property performance and characteristics of a semiconductor device.

[0135] In one embodiment, the process of constructing the preset uncertainty model includes:

[0136] obtaining device historical process fluctuation data, the device historical process fluctuation data including multiple sets of fluctuation data, each set of fluctuation data including historical equipment state data and corresponding process fluctuation data;

[0137] performing data mining on the device historical process fluctuation data to obtain an association relationship, the association relationship being an association relationship between equipment state and process fluctuation;

[0138] obtaining an uncertainty model by using an uncertainty envelope model to describe the influence of equipment state on device manufacturing process fluctuation based on the association relationship.

[0139] In one embodiment, the process of obtaining an uncertainty model by using an uncertainty envelope model to describe the influence of equipment state on device manufacturing process fluctuation based on the association relationship includes:

[0140] constructing an uncertainty set U based on the association relationship;

[0141] obtaining an uncertainty model by using an uncertainty envelope model to describe the uncertainty set U, the uncertainty envelope model being: constructing a linear programming problem as follows:

[0142] wherein x∈R n×1 , c∈R n×1X represents a bounded feasible region of decision variable x, is a coefficient matrix with uncertain parameters on the left side of constraints, is an uncertain parameter vector on the right side of constraints, R is a real number field, and m and n represent different dimensions, is an element in is an element in is an element in for the ith constraint, let and are respectively:

[0143]

[0144] where a ij and b i are nominal parts of uncertain parameters, and the uncertain parameters are device state data, and are floating amplitudes, obtained from the correlation relationship, ξ ij and ξ i0 are unit variables of fluctuations, J i is a parameter set affected by uncertainty in the ith constraint, and respectively represent the upper and lower fluctuation ranges of the ith constraint, that is, the influence of device state on process fluctuation.

[0145] In one embodiment, the process electrical property fluctuation range is determined based on the current device state data using a preset uncertainty model, including:

[0146] Based on the current device state data, a range description of the current device's uncertainty set is determined using a preset uncertainty model;

[0147] Based on the range description of the current device's uncertainty set, a corresponding constraint upper and lower fluctuation range is determined;

[0148] Based on the corresponding constraint upper and lower fluctuation range, a process electrical property fluctuation range is obtained.

[0149] In one embodiment, the data mining of the device historical process fluctuation data to obtain the correlation relationship includes:

[0150] The XGBoost algorithm is used to perform data mining on the device historical process fluctuation data to obtain the correlation relationship.

[0151] In one embodiment, the construction process of the preset process electrical property correlation model includes:

[0152] Obtaining a plurality of sample data, each of the sample data comprising a sample process parameter and a corresponding device basic parameter;

[0153] Based on the plurality of sample data, a deep learning algorithm is used for model training to obtain a process electrical property related model.

[0154] In an embodiment, the obtaining a plurality of sample data comprises:

[0155] Obtaining device historical data, the device historical data comprising one or more of device historical simulation data, device historical DOE data, and device historical electrical property data;

[0156] Performing feature extraction on the device historical data to obtain a plurality of sample data.

[0157] In an embodiment, further comprising:

[0158] Based on the device manufacturing process parameter, a preset simulation system is used for simulation to obtain a device electrical property simulation parameter;

[0159] Comparing the device electrical property simulation parameter with the device electrical property process parameter to obtain a comparison result.

[0160] Those skilled in the art will understand that embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, etc.) containing computer usable program code.

[0161] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, as well as combinations of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 An apparatus for implementing each flow or a plurality of flows and / or blocks Figure 1 An apparatus for implementing each flow or a plurality of flows and / or blocks

[0162] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the Figure 1 function specified in the flow or flows and / or blocks Figure 1 of the block or blocks.

[0163] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions that are executed on the computer or other programmable apparatus provide steps for implementing the Figure 1 function specified in the flow or flows and / or blocks Figure 1 of the block or blocks.

[0164] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0165] The memory can include non-persistent memory and / or volatile memory, such as a random access memory (RAM) including a cache area for the temporary storage of data. The memory can also include non-volatile memory, such as read only memory (ROM), electrically programmable read only memory (EPROM), electrically erasable programmable read only memory (EEPROM), flash memory, or a combination of non-volatile memories in different forms. The memory is an example of computer readable storage media.

[0166] Computer readable media includes permanent and non-permanent, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory (ROM), electrically programmable read only memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital versatile disc (DVD), or other optical storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible to a computing device. According to the definition herein, computer readable media does not include transitory media, such as modulated data signals and carrier waves.

[0167] It should also be noted that the terms "comprising", "comprises" or other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can also include other elements not expressly listed or inherent to such process, method, article, or apparatus.

[0168] The above embodiments are only used to illustrate the present application, but not to limit it. Instead of the above, various modifications and changes can be made to the application by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the application shall fall into the scope of the claims of the application.

Claims

1. A method for determining electrical process parameters of a semiconductor device, characterized in that, include: Acquire device manufacturing process parameters and current equipment status data. The device manufacturing process parameters are various parameters used to control the characteristics and performance of semiconductor devices in the semiconductor device manufacturing process. The current equipment status data is the status data of the equipment used in the current semiconductor device manufacturing process. Based on the device manufacturing process parameters, electrical parameters are determined using a preset process electrical correlation model. The preset process electrical correlation model is used to determine the associated device electrical parameter information based on the device manufacturing process parameter information. Based on the current equipment status data, a preset uncertainty model is used to determine the range of process electrical fluctuations. The preset uncertainty model is used to determine the magnitude of the impact of the equipment status on the process fluctuations based on the equipment status. Based on the electrical parameters and the electrical fluctuation range of the process, the electrical process parameters of the device are obtained. These electrical process parameters are used to describe the electrical performance and characteristics of the semiconductor device. The process of constructing the pre-set uncertainty model includes: Historical process fluctuation data of the device is acquired. The historical process fluctuation data of the device includes multiple sets of fluctuation data. Each set of fluctuation data includes: historical equipment status data and corresponding process fluctuation data. The historical process fluctuation data of the device is obtained from the manufacturing execution system. Data mining is performed on the historical process fluctuation data of the device to obtain correlations, which are the correlations between equipment status and process fluctuations; Based on the aforementioned correlation, an uncertainty envelope model is used to describe the magnitude of the impact of equipment status on device manufacturing process fluctuations, thus obtaining the uncertainty model. Specifically, based on the aforementioned correlation, an uncertainty envelope model is used to describe the magnitude of the impact of equipment status on device manufacturing process fluctuations, resulting in an uncertainty model, including: Based on the aforementioned association, an uncertain set U is constructed; An uncertainty envelope model is used to describe the range of the uncertain set U, resulting in an uncertainty model. The uncertainty envelope model is as follows: The linear programming problem is constructed as follows: ,in, , Representing decision variables Bounded feasible domain, To constrain the coefficient matrix containing uncertain parameters on the left side, To constrain the uncertainty parameter vector on the right side, For the real number field, and They represent different dimensions. for The elements in for The elements in, for the first Let there be a constraint. and They are respectively: ， ， in, and These are the nominal portions of the uncertain parameters, which are equipment status data. and These are the floating ranges, obtained from the aforementioned correlation. and For fluctuating unit variables, For the first The set of parameters affected by uncertainty in a constraint. and They represent the first The upper and lower limits of the constraint fluctuation range, that is, the magnitude of the impact of equipment status on process fluctuation.

2. The method for determining the electrical process parameters of a semiconductor device according to claim 1, characterized in that, Based on the current equipment status data, a preset uncertainty model is used to determine the range of process electrical fluctuations, including: Based on the current device status data, a preset uncertainty model is used to determine the range description of the uncertainty set of the current device; Based on the range description of the uncertainty set of the current device, the corresponding upper and lower limit fluctuation ranges of the constraints are determined. Based on the corresponding upper and lower limit fluctuation ranges of the constraints, the fluctuation range of the process electrical properties is obtained.

3. The method for determining the electrical process parameters of a semiconductor device according to claim 1, characterized in that, The data mining of the historical process fluctuation data of the device to obtain correlations includes: The XGBoost algorithm was used to perform data mining on the historical process fluctuation data of the device to obtain correlations.

4. The method for determining the electrical process parameters of a semiconductor device according to claim 1, characterized in that, The construction process of the pre-set process electrical correlation model includes: Acquire multiple sample data, each of which includes sample process parameters and corresponding basic device parameters; Based on the aforementioned sample data, a deep learning algorithm is used to train the model, resulting in a process electrical correlation model.

5. The method for determining the electrical process parameters of a semiconductor device according to claim 4, characterized in that, The acquisition of multiple sample data includes: Acquire historical device data, which includes one or more of historical device simulation data, historical device DOE data, and historical device electrical performance data; Feature extraction was performed on the historical data of the device to obtain multiple sample data.

6. The method for determining the electrical process parameters of a semiconductor device according to claim 1, characterized in that, Also includes: Based on the device manufacturing process parameters, a preset simulation system is used to perform simulation to obtain the device's electrical simulation parameters; The electrical simulation parameters of the device are compared with the electrical process parameters of the device to obtain the comparison results.

7. A device for determining electrical process parameters of a semiconductor device, characterized in that, include: The acquisition module is used to acquire device manufacturing process parameters and current equipment status data. The device manufacturing process parameters are various parameters used to control the characteristics and performance of semiconductor devices in the semiconductor device manufacturing process. The current equipment status data is the status data of the equipment used in the current semiconductor device manufacturing process. The parameter calculation module is used to determine the electrical parameters based on the device manufacturing process parameters using a preset process electrical correlation model. The preset process electrical correlation model is used to determine the associated device electrical parameter information based on the device manufacturing process parameter information. The fluctuation calculation module is used to determine the range of process electrical fluctuations based on the current equipment status data and a preset uncertainty model. The preset uncertainty model is used to determine the magnitude of the impact of equipment status on process fluctuations based on the equipment status. The construction process of the preset uncertainty model includes: acquiring historical process fluctuation data of the device, which includes multiple sets of fluctuation data, each set including historical equipment status data and corresponding process fluctuation data; wherein the historical process fluctuation data is obtained from the manufacturing execution system; performing data mining on the historical process fluctuation data to obtain correlations, which are the correlations between equipment status and process fluctuations; and using an uncertainty envelope model to describe the magnitude of the impact of equipment status on the device manufacturing process fluctuations based on the correlations, thus obtaining an uncertainty model. The step of using an uncertainty envelope model to describe the magnitude of the impact of equipment status on the device manufacturing process fluctuations based on the correlations, thus obtaining an uncertainty model, includes: constructing an uncertainty set U based on the correlations; and using an uncertainty envelope model to describe the range of the uncertainty set U, thus obtaining an uncertainty model. The uncertainty envelope model is: [The text then abruptly shifts to a different topic:] Constructing a linear programming problem: ,in, , Representing decision variables Bounded feasible domain, To constrain the coefficient matrix containing uncertain parameters on the left side, To constrain the uncertainty parameter vector on the right side, For the real number field, and They represent different dimensions. for The elements in for The elements in, for the first Let there be a constraint. and They are respectively: , ,in, and These are the nominal portions of the uncertain parameters, which are equipment status data. and These are the floating ranges, obtained from the aforementioned correlation. and For fluctuating unit variables, For the first The set of parameters affected by uncertainty in a constraint. and They represent the first The upper and lower limits of the constraint fluctuation range, that is, the magnitude of the impact of equipment status on process fluctuation; The determination module is used to obtain device electrical process parameters based on the electrical parameters and the process electrical fluctuation range, wherein the device electrical process parameters are used to describe the electrical performance and characteristics of the semiconductor device.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; A memory connected to the at least one processor; The memory stores instructions executable by the at least one processor, which implements the semiconductor device electrical process parameter determination method according to any one of claims 1 to 6 by executing the instructions stored in the memory.

9. A machine-readable storage medium storing instructions thereon, characterized in that, When executed by a processor, this instruction causes the processor to be configured to perform the method for determining the electrical process parameters of a semiconductor device according to any one of claims 1 to 6.

Images