A method for calculating hole mobility by introducing a strain-dependent optical deformation potential

Abstract

This invention relates to the field of semiconductor technology and discloses a method for calculating hole mobility by introducing a strain-dependent optical deformation potential. The method includes obtaining the structural parameters and strain distribution parameters of a target semiconductor device; determining the strain-dependent optical deformation potential in the channel region of the target semiconductor device based on the strain distribution parameters, whereby the strain-dependent optical deformation potential characterizes the suppression effect of strain-induced crystal symmetry breaking on the electro-acoustic coupling strength; substituting the strain-dependent optical deformation potential into a carrier transport model and calculating the carrier mobility limited by the optical phonon scattering mechanism in conjunction with the structural parameters; simulating and predicting the electrical performance of the target semiconductor device based on the carrier mobility limited by the optical phonon scattering mechanism, and outputting macroscopic electrical performance indicators. This invention corrects the calculation bias of scattering rate under high strain conditions and improves the prediction accuracy of hole mobility in TCAD software.

Description

Technical Field

[0001] This invention relates to the field of semiconductor technology, specifically to a method for calculating hole mobility by introducing a strain-dependent optical deformation potential. Background Technology

[0002] Currently, semiconductor integrated circuits continue to evolve towards the nanoscale. Computer-aided design (CAD) software is widely used to shorten device development cycles and reduce manufacturing trial-and-error costs. Faced with the introduction of novel device structures and materials, CAD software has become a fundamental platform for evaluating semiconductor process technology routes. Strain engineering is a core technology of modern silicon-based transistors. This technology improves carrier mobility by applying mechanical stress to the channel region to alter the crystal structure. This provides an effective way to continue the development of Moore's Law.

[0003] Existing mainstream device simulation tools typically run a constant deformation potential approximation model when simulating strain effects. The system assumes that the acoustic and optical deformation potential constants of the semiconductor material remain constant and do not change with external mechanical stress. The related carrier simulation process mainly relies on strain-induced band structure changes to calculate the scattering rate. Specifically, this involves calculating the values ​​for stress-induced degeneracy of the light and heavy hole bands, or modifying the band curvature parameters to reduce the effective carrier mass. When the set strain-induced valence band splitting exceeds the optical phonon energy threshold, the system directly outputs the simulation results of intra-band scattering based on the preset constant deformation potential parameters, ignoring inter-band scattering.

[0004] Existing constant deformation potential calculation models do not incorporate the perturbation effect of lattice distortion on underlying microscopic physical processes. Under strong strain conditions, the coupling strength between holes and optical phonons within a semiconductor undergoes substantial changes. Fixed deformation potential parameters cannot characterize the suppression effect of strain-induced crystal symmetry breaking on electro-acoustic coupling. This results in underlying logic deviations in the optical phonon scattering rate output by the system under high stress. To compensate for these computational discrepancies, simulation workflows require extensive experimental data extraction for empirical parameter calibration at specific process nodes. Faced with advanced process technologies and novel structures lacking comprehensive test data, this artificial fitting approach based on non-physical parameters reduces the predictive power of simulation tools. The system struggles to accurately output hole mobility limited by optical phonon scattering, failing to provide accurate quantitative calculations of macroscopic electrical performance for process designs such as stress-substrate mismatch.

[0005] Therefore, this invention provides a method for calculating hole mobility in TCAD software by introducing strain-dependent optical deformation potential, in order to overcome the shortcomings of the prior art. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a method for calculating hole mobility in TCAD software by introducing a strain-dependent optical deformation potential. This method solves the problem that existing TCAD software, due to its use of a constant deformation potential approximation, cannot characterize the suppression effect of strain-induced crystal symmetry breaking on electroacoustic coupling, leading to inaccurate hole mobility calculations under high stress conditions and excessive reliance on empirical parameter calibration.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] In a first aspect, the present invention provides a method for calculating hole mobility in TCAD software by introducing strain-dependent optical deformation potential, using the following technical solution:

[0009] A method for calculating hole mobility in TCAD software by introducing strain-dependent optical deformation potential includes the following steps:

[0010] Obtain the structural parameters and strain distribution parameters of the target semiconductor device;

[0011] Based on the strain distribution parameters, the strain-dependent optical deformation potential in the channel region of the target semiconductor device is calculated and determined by calling the preset nonlinear function model. The strain-dependent optical deformation potential characterizes the suppression effect of strain-induced crystal symmetry breaking on the electro-acoustic coupling strength. The value of the strain-dependent optical deformation potential decreases with the increase of strain intensity.

[0012] The strain-dependent optical deformation potential is substituted into the carrier transport model, and the carrier mobility of the target semiconductor device channel region under the strain distribution parameters is calculated by combining the structural parameters and the optical phonon scattering mechanism. The calculation results of the scattering rate inside the carrier transport model are updated using the strain-dependent optical deformation potential.

[0013] The electrical performance of the target semiconductor device is simulated and predicted based on the carrier mobility limited by the optical phonon scattering mechanism, and the macroscopic electrical performance index of the target semiconductor device under strain distribution parameters is output.

[0014] By adopting the above technical solution, the optical deformation potential that characterizes the symmetry breaking of the crystal is calculated by calling a preset nonlinear function model, and the physical calculation benchmark of the scattering rate inside the carrier transport model is updated accordingly. Therefore, the effect of accurately reflecting the electroacoustic coupling suppression law under high stress and outputting high-precision semiconductor macroscopic electrical performance indicators is achieved.

[0015] Preferably, the process of obtaining the structural parameters and strain distribution parameters of the target semiconductor device includes: extracting the material type, geometric topology, doping concentration distribution, and electrode position information from the structural parameters, which are used to define the physical dimensions of the target semiconductor device and the geometry of the channel region of the target semiconductor device; and extracting the strain type, strain direction, and strain value from the strain distribution parameters, which are used to characterize the mechanical stress state inside the target semiconductor device and the channel region of the target semiconductor device.

[0016] By adopting the above technical solutions, the physical boundaries and material mechanical states of semiconductor devices can be accurately defined.

[0017] Preferably, the process of calculating and determining the strain-dependent optical deformation potential in the channel region of the target semiconductor device by calling a preset nonlinear function model includes: extracting the lattice distortion degree variable based on the strain distribution parameters in TCAD software; setting the scattering cross section between the hole state and the optical phonon mode in the central region of the Brillouin zone as a symmetry forbidden mapping variable that decays with decreasing crystal symmetry in the algorithm module based on the symmetry forbidden mapping variable; and performing the calculation of the strain-dependent optical deformation potential value based on the symmetry forbidden mapping variable.

[0018] By adopting the above technical solution, microscopic physical phenomena are transformed into mapping variables in software algorithms, thus establishing a reliable logical basis for the quantitative calculation of deformation potential within TCAD software.

[0019] Preferably, the process of calculating the strain-dependent optical deformation potential based on the symmetry forbidden mapping variable includes: in the momentum space discrete grid of the TCAD software, calculating the range parameter of the central region coupling forbidden zone extending outward according to the increase of strain intensity in the symmetry forbidden mapping variable; based on the range parameter, performing a reduction calculation on the number of optical phonon modes participating in scattering in the algorithm module; and based on the result of the reduction calculation, outputting the value of the strain-dependent optical deformation potential that decreases with the increase of strain intensity.

[0020] By adopting the above technical solution, a numerical calculation relationship between the forbidden zone range and phonon mode reduction is established in the discrete grid algorithm, thereby improving the accuracy of macroscopic deformation potential calculation and software execution capability.

[0021] Preferably, the process of outputting the strain-dependent optical deformation potential value that decreases with increasing strain intensity based on the result of the reduction calculation includes: calling a nonlinear function model, combining the result of the reduction calculation with the optical deformation potential under no strain, the first fitting parameter, the second fitting parameter, and the third fitting parameter for nonlinear mapping processing, so as to calculate the corresponding strain-dependent optical deformation potential value; wherein, the first fitting parameter, the second fitting parameter, and the third fitting parameter are values ​​that have been pre-acquired and stored in the material parameter library of the TCAD software, and the values ​​are determined according to the semiconductor material type and strain type of the target semiconductor device channel region.

[0022] By adopting the above technical solutions, the calculation efficiency of deformation potential values ​​under the influence of multiple parameters and the system call speed are improved.

[0023] Preferably, the parameter pre-acquisition process includes: applying strain of different magnitudes to the cell model of the target semiconductor material to establish atomic structure models under multiple strain states; calculating and acquiring the hole-optical phonon coupling matrix metadata in the Brillouin zone under multiple strain states using first-principles calculations; extracting the numerical values ​​of the equivalent optical phonon deformation potential corresponding to multiple strain states based on the calculated hole-optical phonon coupling matrix metadata; performing nonlinear fitting processing on the numerical values ​​of the equivalent optical phonon deformation potential to determine the first fitting parameter, the second fitting parameter, and the third fitting parameter in the nonlinear function model describing the strain-dependent optical deformation potential as a function of strain.

[0024] By adopting the above technical solutions, we can ensure that the empirical fitting parameters are supported by rigorous underlying quantum mechanical data.

[0025] Preferably, the process of calculating the carrier mobility of the target semiconductor device channel region under strain distribution parameters, which is limited by the optical phonon scattering mechanism, includes: substituting the strain-dependent optical deformation potential into the Fermi-Golden Rule calculation model inside the carrier transport model to analyze the optical phonon scattering physical process of carriers in different energy states; and calculating the scattering rate of holes and optical phonons by combining carrier energy, optical phonon energy, Bose-Einstein distribution function of phonons and band density of states.

[0026] By adopting the above technical solution, the dynamic interaction process between charge carriers and optical phonons can be fully analyzed.

[0027] Preferably, the process of analyzing the optical phonon scattering physical process of charge carriers in different energy states includes: calculating the scattering rate of the phonon absorption process that increases continuously with the increase of charge carrier energy; setting the energy threshold of the emission process according to the optical phonon energy, and calculating the scattering rate of the emission process triggered after the charge carrier energy crosses the energy threshold, thereby forming a numerical distribution of the total scattering process that exhibits a step-increase.

[0028] By adopting the above technical solution, the step characteristics of scattering rate in different transport energy ranges can be accurately characterized.

[0029] Preferably, the process of calculating the carrier mobility limited by the optical phonon scattering mechanism in the channel region of the target semiconductor device under strain distribution parameters further includes: extracting the reciprocal of the calculated scattering rate of holes and optical phonons as the relaxation time; substituting the relaxation time into the relaxation time approximation solution model based on the Boltzmann transport equation, and extracting the carrier mobility limited by the optical phonon scattering mechanism by performing weighted integration on different energy bands and different energy states.

[0030] By adopting the above technical solution, the quantitative conversion of microscopic scattering time parameters into macroscopic statistical carrier mobility can be achieved.

[0031] Preferably, the process of simulating and predicting the electrical performance of the target semiconductor device based on the carrier mobility limited by the optical phonon scattering mechanism includes: substituting the carrier mobility limited by the optical phonon scattering mechanism into a set of semiconductor device equations including the Poisson equation, the current continuity equation, and the drift-diffusion equation for iterative solution; and outputting the current-voltage characteristics and macroscopic electrical performance indicators of the target semiconductor device under strain distribution parameters, including the transfer characteristic curve, the output characteristic curve, the transconductance, the turn-on voltage, and the switching speed.

[0032] By adopting the above technical solution, device-level output results that fully reflect the characteristics of force-electric coupling are provided.

[0033] Preferably, after outputting the macroscopic electrical performance indicators of the target semiconductor device under strain distribution parameters, the hole mobility calculation method introduced in the TCAD software based on strain-dependent optical deformation potential also includes the process of setting process parameters for the target semiconductor device: determining the target strain parameters based on the macroscopic electrical performance indicators, mapping the target strain parameters to device structural parameters and material manufacturing parameters; establishing the mechanical stress field in the channel region of the target semiconductor device using the device structural parameters and material manufacturing parameters, and introducing uniaxial compressive strain or biaxial tensile strain in the channel region of the target semiconductor device to induce geometric deformation of the semiconductor crystal.

[0034] By adopting the above technical solutions, we can guide the forward-looking manufacturing and process parameter optimization of semiconductor devices through stress engineering.

[0035] Preferably, the process of establishing the mechanical stress field of the target semiconductor device channel region using device structure parameters and material manufacturing parameters includes: introducing stress into the target semiconductor device channel region by setting the lattice mismatch value of the stress substrate; and reducing the crystal symmetry of the semiconductor crystal to a state that triggers optical scattering prohibition by setting the intrinsic stress value of the stress capping layer and adjusting the composition ratio value of the epitaxial growth layer.

[0036] By adopting the above technical solutions, a feasible semiconductor physical process approach for achieving band structure control is established.

[0037] Secondly, the present invention provides a hole mobility calculation device that incorporates strain-dependent optical deformation potential into TCAD software, employing the following technical solution:

[0038] A hole mobility calculation device based on strain-dependent optical deformation potential is introduced into TCAD software, comprising:

[0039] The parameter acquisition module is used to acquire the structural parameters and strain distribution parameters of the target semiconductor device.

[0040] The deformation potential calculation module is used to calculate and determine the strain-dependent optical deformation potential in the channel region of the target semiconductor device based on the strain distribution parameters obtained by the parameter acquisition module and by calling a preset nonlinear function model.

[0041] The mobility calculation module is used to substitute the strain-dependent optical deformation potential determined by the deformation potential calculation module into the carrier transport model, and combine the structural parameters to calculate the carrier mobility of the target semiconductor device channel region under the strain distribution parameters, which is limited by the optical phonon scattering mechanism.

[0042] The performance prediction module is used to simulate and predict the electrical performance of the target semiconductor device based on the carrier mobility limited by the optical phonon scattering mechanism calculated by the mobility calculation module, and outputs the macroscopic electrical performance index of the target semiconductor device under the strain distribution parameters.

[0043] By adopting the above technical solution, and through the modular operation structure of multi-parameter transfer and physical mechanism calculation, which connects the extraction of front-end structural parameters with the iteration of electrical equations of back-end devices, the efficiency of the underlying core computing of the TCAD software system and its engineering application deployment capabilities are improved.

[0044] Thirdly, the present invention provides a computer device, which adopts the following technical solution:

[0045] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of introducing a hole mobility calculation method based on strain-dependent optical deformation potential in the TCAD software proposed in the first aspect.

[0046] By adopting the above technical solution, the high-concurrency calculation of complex physical models and the stable execution of the system are achieved by using the underlying hardware processing unit to execute the instructions for solving the highly nonlinear semiconductor physical transport equations and calling the pre-set crystal computing model in the memory.

[0047] Fourthly, the present invention provides a computer-readable storage medium, which adopts the following technical solution:

[0048] A computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the hole mobility calculation method based on strain-dependent optical deformation potential in the TCAD software proposed in the first aspect.

[0049] By adopting the above technical solution, the first-principles mapping table and carrier transport solution code are encapsulated in a non-volatile medium, thus freeing them from the limitations of the computing environment of specific physical devices. Therefore, the effect of persistent storage, version iteration, and large-scale distribution of high-precision TCAD core computing components can be achieved.

[0050] This invention provides a method for calculating hole mobility in TCAD software by incorporating strain-dependent optical deformation potential. It offers the following advantages:

[0051] 1. This invention accurately characterizes the suppressive effect of strain-induced crystal symmetry breaking on electroacoustic coupling strength by introducing a strain-dependent optical deformation potential into the carrier transport model. The method corrects the calculation deviation of scattering rate under high strain conditions caused by the fixed optical deformation potential in conventional models, and improves the prediction accuracy of hole mobility and macroscopic electrical performance indicators of devices in strained semiconductor structures using TCAD software.

[0052] 2. This invention utilizes a nonlinear function model obtained by fitting metadata from a first-principles coupling matrix to determine the deformation potential value, thus achieving a connection between microscopic quantum mechanical calculation results and macroscopic transport simulation processes. This approach, while maintaining the rigor of the underlying physical mechanisms, avoids directly performing large-scale quantum mechanical operations in device-level simulations, thereby improving the computational efficiency of iteratively solving the electrical equations of semiconductor devices.

[0053] 3. This invention establishes a physical link from microscopic scattering rate calculation and mobility extraction to macroscopic process parameter setting, providing a quantitative basis for stress engineering design. By correlating electrical performance indicators with device structural parameters and material manufacturing parameters, it can guide the process flow to induce a reduction in crystal symmetry by adjusting the stress substrate mismatch or the intrinsic stress of the capping layer, thereby achieving precise control of carrier transport characteristics. Attached Figure Description

[0054] Figure 1 This is a functional module architecture block diagram of the semiconductor device hole mobility simulation system of the present invention;

[0055] Figure 2 The flowchart of the hole mobility calculation method based on strain-dependent optical deformation potential introduced in TCAD software according to the present invention is shown below.

[0056] Figure 3 This is a schematic diagram showing the distribution of the hole-optical phonon coupling matrix elements in momentum space under different strain conditions according to the present invention.

[0057] Figure 4 This is a comparison curve of the strain-dependent optical phonon deformation potential of the present invention and the conventional constant deformation potential as a function of strain.

[0058] Figure 5 This is a comparison of the distribution of hole scattering rate as a function of energy under different models of the present invention (no strain, conventional model, and the model of the present invention);

[0059] Figure 6 This is a macroscopic verification diagram of the hole mobility calculation results under different strain types (biaxial strain and uniaxial strain) of the present invention.

[0060] Among them, 10 is the parameter acquisition module; 20 is the deformation potential calculation module; 30 is the mobility calculation module; and 40 is the performance prediction module. Detailed Implementation

[0061] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0062] See attached document Figure 1 The present invention provides a semiconductor device hole mobility simulation device, which is equipped with a system that may include a parameter acquisition module 10, a deformation potential calculation module 20, a mobility calculation module 30, and a performance prediction module 40.

[0063] The parameter acquisition module 10 is configured to acquire the structural parameters and strain distribution data of the target semiconductor device. The structural parameters encompass the material type used in the target semiconductor device, such as silicon, germanium, silicon-germanium alloys, or III-V compound semiconductor materials, and also include the device's geometric topology, doping concentration distribution, and electrode location information. The strain distribution data includes stress tensor components or strain components at various locations within the channel region, specifically involving uniaxial strain along the 001, 110, and 111 directions of the crystal, or biaxial strain data acting on the 001 plane. The parameter acquisition module 10 provides the basic data input for subsequent physical parameter calculations by reading pre-stored device design files or receiving stress distribution calculation results output by external semiconductor process simulation software.

[0064] The deformation potential calculation module 20 is connected to the parameter acquisition module 10 and is configured to calculate the optical phonon deformation potential values ​​at various locations in the channel region based on strain distribution data and a pre-set strain-dependent optical phonon deformation potential model. This module internally stores pre-fitted nonlinear model parameters for different semiconductor materials and strain types. During operation, the deformation potential calculation module 20 determines the dynamic value of the optical phonon deformation potential under the received strain amplitude. The core physics of this module lies in characterizing the suppression effect of strain-induced crystal symmetry breaking on the electro-acoustic coupling strength. When strain causes the crystal to change from its initial state... Cubic symmetry is reduced to At lower symmetry, the scattering between the hole state near the center of momentum space and the optical phonon mode becomes symmetry forbidden, resulting in the equivalent optical deformation potential value decreasing nonlinearly with increasing strain intensity.

[0065] The mobility calculation module 30 is connected to the deformation potential calculation module 20 and is configured to calculate the macroscopic carrier mobility using a carrier transport model that includes an optical phonon scattering mechanism. This module receives strain-dependent optical phonon deformation potential data output from the deformation potential calculation module 20 and substitutes it into a preset carrier transport physical model. The transport physical model can be a numerical solution model based on the Boltzmann transport equation or a Monte Carlo simulation model based on stochastic statistical processes. When calculating the scattering rate of each energy band, the mobility calculation module 30 accurately captures the effect of significantly weakened optical phonon scattering under strong strain by introducing a deformation potential value that dynamically adjusts with strain, thereby calculating the hole mobility after strain modulation. This module realizes a cross-scale mapping from microscopic electroacoustic coupling physical quantities to macroscopic transport parameters.

[0066] The performance prediction module 40 is connected to the mobility calculation module 30 and is used to simulate and predict the electrical performance of the target semiconductor device based on the carrier mobility output by the mobility calculation module 30. This module takes the calculated, high-precision hole mobility distribution as input parameters and substitutes it into the semiconductor device equations for iterative solution. The semiconductor device equations typically include the Poisson equation, the current continuity equation, and the drift-diffusion equation. The performance prediction module 40 ultimately outputs key electrical performance indicators of the device, such as the transfer characteristic curve, output characteristic curve, transconductance, turn-on voltage, and switching speed. The prediction results of the performance prediction module 40 can be directly used to guide the design and process optimization of gate-to-ring nanosheet field-effect transistors, nanowire transistors, or p-type metal-oxide-semiconductor field-effect transistors with strained channels.

[0067] The device provided in this embodiment corrects the optical phonon deformation potential, which was originally considered a constant in traditional simulation tools, into a physical quantity that changes dynamically with strain through logical coordination between modules. This eliminates the overestimation of scattering rate in traditional models under strong strain conditions and improves the prediction accuracy of TCAD simulation tools at advanced process nodes.

[0068] See attached document Figure 2 The present invention provides a method for calculating hole mobility by introducing strain-dependent optical deformation potential in TCAD software, which may include the following steps.

[0069] First, step S100 is executed to obtain the structural parameters and strain distribution parameters of the target semiconductor device. During semiconductor device simulation, the structural parameters define the physical dimensions of the target semiconductor device and the geometry of the channel region. The strain distribution parameters characterize the mechanical stress state experienced by the target semiconductor device and the channel region. The structural parameters and strain distribution parameters together constitute the input boundary conditions for subsequent calculations of carrier transport properties.

[0070] Subsequently, step S200 is performed to determine the strain-dependent optical deformation potential in the channel region of the target semiconductor device based on strain distribution parameters. The optical phonon deformation potential is configured to characterize the suppression effect of strain-induced crystal symmetry breaking on the coupling strength between charge carriers and phonons. The value of this optical phonon deformation potential decreases with increasing strain intensity. By determining this strain-dependent optical deformation potential, the electroacoustic scattering physical state of the semiconductor material under specific strain conditions can be quantified.

[0071] Next, step S300 is executed to substitute the strain-dependent optical deformation potential into the carrier transport model to calculate the carrier mobility in the channel region under the strain distribution. The optical phonon deformation potential determines the intensity of optical phonon scattering; substituting it into the carrier transport model updates the calculation results regarding the scattering rate within the transport model. Combined with the band structure parameters of the target semiconductor device, the hole mobility value limited by the optical phonon scattering mechanism is calculated.

[0072] Finally, step S400 simulates and predicts the electrical performance of the target semiconductor device based on the calculated carrier mobility. Carrier mobility is a physical parameter characterizing the conductivity of a semiconductor device. By inputting carrier mobility data, which includes the strain-dependent optical deformation potential mechanism, into the device-level simulation module of the computer-aided design software, the current-voltage characteristics and macroscopic electrical performance indicators of the target semiconductor device under given strain conditions can be output, thus providing data support for the structural adjustment and parameter setting of the target semiconductor device.

[0073] See attached document Figure 2 This invention provides a method for calculating hole mobility in TCAD software by introducing strain-dependent optical deformation potential, which may include:

[0074] In step S100, the parameter acquisition module 10 acquires the structural parameters and strain distribution parameters of the target semiconductor device. The structural parameters define the geometrical physical characteristics of the semiconductor device to be simulated, mainly including the device's topology, material composition, and doping concentration in each region. The target semiconductor device is selected from gate-to-ring nanosheet field-effect transistors, nanowire transistors, or p-type metal-oxide-semiconductor field-effect transistors with strained channels. In specific implementations, the channel region of the target semiconductor device is composed of specific semiconductor materials, including silicon, germanium, silicon-germanium alloys, or group III-V compound semiconductor materials. These structural parameters provide the necessary spatial boundary conditions and material basis for subsequent physical model establishment.

[0075] Strain distribution parameters are used to quantify the mechanical stress state within the channel region of a target semiconductor device and are a core input for calculating mobility. Specifically, strain distribution parameters include strain type, strain direction, and strain value. Strain types are classified into uniaxial strain and biaxial strain based on the manufacturing process and geometric characteristics that generate stress. The strain direction is defined based on the semiconductor crystallography coordinate system. In this embodiment, typical strain directions include uniaxial strain along the 001 direction of the crystal, uniaxial strain along the 110 direction of the crystal, uniaxial strain along the 111 direction of the crystal, and biaxial strain acting on the 001 crystal plane.

[0076] strain value The strain values ​​characterize the degree to which the lattice deviates from its equilibrium position; positive and negative values ​​represent the nature of the strain, with positive values ​​representing tensile strain and negative values ​​representing compressive strain. To ensure that the symmetry breaking effect has an observable impact on the scattering mechanism, the strain values ​​in the obtained strain distribution parameters are... The magnitude is typically greater than 0.1 percentage points. In optimized embodiments for advanced process devices, the strain value... The value ranges from 0.5 percentage points to 2 percentage points. These strain distribution data can be obtained from the process simulation module or preset according to the process parameters of the strain layer, stress substrate, or stress capping layer.

[0077] The parameter acquisition module 10 formats the acquired structural parameters and strain distribution parameters and passes them to the subsequent deformation potential calculation module 20. The strain distribution parameters can be uniformly distributed within the channel space, or they can be a non-uniform vector field or tensor field that varies with spatial coordinates. For non-uniform strain fields, the parameter acquisition module 10 maps them to the simulation mesh nodes of the device to ensure that the corresponding optical phonon deformation potential and carrier mobility can be determined for the local strain state at different locations within the channel in subsequent calculation steps. This parameter acquisition process provides underlying data support for cross-scale coupled simulation, enabling macroscopic device performance prediction to accurately reflect changes in the microscopic crystal structure.

[0078] The subsequent steps will determine the strain-dependent optical deformation potential in the channel region of the target semiconductor device based on the strain distribution parameters obtained in step S100, using the deformation potential calculation module 20. This process takes into account the suppression effect of strain-induced crystal symmetry breaking on the coupling strength between charge carriers and phonons, thereby correcting the physical deviation of using a fixed deformation potential constant in traditional TCAD tools.

[0079] See attached document Figure 3 The present invention provides a method for calculating hole mobility by introducing strain-dependent optical deformation potential in computer-aided design software, which may include step S200, determining the strain-dependent optical deformation potential based on strain distribution parameters.

[0080] Under strain-free conditions, the semiconductor material in the channel region of the target semiconductor device maintains cubic crystal symmetry, and the valence band apex and optical phonons satisfy a preset scattering selection rule. Electro-acoustic coupling is effective at this point, and the coupling matrix element between holes and optical phonons in the central region of the Brillouin zone is greater than 0.

[0081] When the acquired strain distribution parameters indicate the presence of biaxial tensile strain or uniaxial compressive strain in the channel region of the semiconductor material, the crystal structure of the semiconductor material is distorted, and its symmetry decreases from cubic to tetragonal. This reduction in crystal symmetry alters the aforementioned scattering selection rule, causing the scattering between the hole states in the central region of the Brillouin zone and the preset optical phonon mode to become symmetry-forbidden. In this state, the coupling matrix element value in the central region of the Brillouin zone decreases to 0, thereby forming a coupling-forbidden region in the central momentum space.

[0082] As the strain value in the strain distribution parameters increases, the coupling forbidden region expands from the center of the Brillouin zone outwards. The expansion of the forbidden region reduces the number of phonon modes involved in scattering, decreases the overall optical phonon scattering rate, and consequently reduces the value of the optical phonon deformation potential at the macroscopic level.

[0083] Specifically, Figure 3 Composed of a 4x3 two-dimensional heatmap matrix, it visually displays the absolute values ​​of the hole-optical phonon coupling matrix elements near the Brillouin zone center (Γ point) under different strain conditions. The distribution of momentum. The horizontal and vertical axes of the figure represent the momentum space. and (Unit: 2π / a), ranging from -0.06 to 0.06. The color bars at the bottom represent the coupling strength (unit: meV), transitioning from dark blue (0 meV, indicating extremely weak coupling or scattering forbidden) to dark red (100 meV, indicating extremely strong coupling). The three columns from left to right in the figure correspond to two transverse optical phonon modes (TO) and one longitudinal optical phonon mode (LO), respectively.

[0084] Regarding the evolution under no strain and small biaxial strain conditions Figure 3 The first row (ac) shows the strain-free state. At this point, the regions for the three optical phonon modes are predominantly red, indicating that cubic symmetry is maintained. In silicon, holes are strongly coupled to optical phonons. Figure 3 The second row (df) shows the result after applying a small (+0.10%) biaxial tensile strain. It is clear that as symmetry decreases, the center of momentum space ( =0, A distinct blue cross-shaped intersection region appears at (=0). This intuitively demonstrates that strain-induced symmetry breaking alters the scattering selection rule, causing the coupling matrix element in the central region of the Brillouin zone to plummet to zero, triggering the scattering forbidden effect.

[0085] Regarding the expansion of restricted areas under strong stress... Figure 3The third row (gi) shows the results of applying a large biaxial strain of +2.00%. At this point, for all TO and LO phonon modes, the blue forbidden zone in the central region expands sharply outward, covering the entire momentum space shown in the figure, appearing entirely dark blue. This indicates that under high biaxial strain, overall optical phonon scattering is almost completely shut off. Figure 3 The fourth row (jl) shows the results for +2.00% uniaxial tensile strain. Under this condition, both TO modes exhibit large areas of deep blue forbidden regions, indicating significantly suppressed scattering; while for the LO mode, although the central minimal region remains blue forbidden, a strong red coupling region is retained as it expands outward. In summary, Figure 3 The core microscopic mechanism of this invention is clearly revealed: strain not only changes the band structure, but also creates a coupling forbidden region (changing from red to blue) at the center of momentum space by breaking lattice symmetry. As the biaxial tensile strain increases, this forbidden region extends to the entire Brillouin zone, thereby weakening the macroscopic optical phonon deformation potential.

[0086] See attached document Figure 4 The present invention provides a method for calculating hole mobility by introducing strain-dependent optical deformation potential in computer-aided design software. This method may include extracting the nonlinear variation law of optical phonon deformation potential with strain decay based on the physical mechanism of scattering suppression caused by the above-mentioned symmetry breaking.

[0087] The equivalent optical phonon deformation potential is extracted from the metadata of the hole-optical phonon coupling matrix under different strain states. The optical phonon deformation potential reaches its maximum value when no strain is applied. After applying strain, the optical phonon deformation potential exhibits nonlinear decay due to the scattering forbidden effect triggered by symmetry breaking. The acoustic phonon deformation potential remains constant under different strain values.

[0088] See details Figure 4 , Figure 4 Composed of two side-by-side subplots (a) and (b), this diagram visually compares the differences between the deformation potential extracted from first-principles calculations and the assumptions of traditional models. The horizontal axis of both subplots represents the magnitude of strain, ranging from -2% (compressive strain) to 2% (tensile strain). The red hollow circles and solid red lines in the diagrams represent uniaxial strain, while the blue solid dots and solid blue lines represent biaxial strain. The multiple horizontal gray dashed lines in the background represent the constant deformation potential parameters that do not change with strain, as used in current mainstream TCAD simulation tools or traditional literature.

[0089] Figure 4 (a) demonstrates the strain-dependent optical phonon deformation potential. This refers to the dramatic nonlinear change in the optical phonon deformation potential with strain. In the unstrained (0%) state, It is at its peak (approximately 12.2 eV / Å). When strain is applied, whether uniaxial or biaxial, compressive or tensile, All exhibit a sharp decline trend similar to the Lorentz curve. In particular, at 2% biaxial tensile strain, the value drops to approximately 2.4 eV / Å, a decrease of about 5 times compared to the strain-free condition. This dynamic change trend, accurately fitted by the solid line, stands in stark contrast to the assumption of a constant in traditional models.

[0090] Figure 4 (b) demonstrates the acoustic phonon deformation potential. The variation with strain is quite different from the dramatic changes in optical deformation potential. Throughout the strain range of -2% to 2%, the value remains relatively stable at around 7-8 eV, regardless of whether it is uniaxial or biaxial strain. This is consistent with the constant value assumption in traditional physical models. Figure 4 The study intuitively demonstrated that strain has a negligible effect on the acoustic deformation potential, but it significantly suppresses electroacoustic coupling, leading to a substantial decrease in the optical deformation potential as strain increases. Existing constant deformation potential models are completely incapable of describing this microscopic physical mechanism.

[0091] During the execution of step S200, the strain-dependent optical deformation potential is determined by calculating it using a nonlinear function model. The system calculates the corresponding optical phonon deformation potential using a modified Lorentz model based on the strain type and strain value in the strain distribution parameters. This calculation model satisfies the following mathematical expression:

[0092] ;

[0093] in, Indicates the strain value. It represents the optical deformation potential under strain. It represents the optical deformation potential under no strain. Represents the first fitted parameter. Indicates the second fitting parameter. This represents the third fitting parameter. The first, second, and third fitting parameters mentioned above are all fixed numerical parameters determined based on the type of semiconductor material and the strain type.

[0094] When the semiconductor material is silicon and the strain type is biaxial tensile strain acting on a specific crystal plane, the optical deformation potential without strain is 12.2 eV / Å, the first fitting parameter is 0.998 eV / Å, the second fitting parameter is 4.039 eV / Å, and the third fitting parameter is 0.052.

[0095] When the semiconductor material is silicon and the strain type is biaxial compressive strain acting on a specific crystal plane, the optical deformation potential without strain is 12.2 eV / Å, the first fitting parameter is -2.022 eV / Å, the second fitting parameter is 5.824 eV / Å, and the third fitting parameter is 0.089.

[0096] When the semiconductor material is silicon and the strain type is uniaxial tensile strain along a specific crystal orientation, the optical deformation potential without strain is 12.2 eV / Å, the first fitting parameter is 1.616 eV / Å, the second fitting parameter is 5.824 eV / Å, and the third fitting parameter is 0.106.

[0097] When the semiconductor material is silicon and the strain type is uniaxial compressive strain along a specific crystal orientation, the optical deformation potential without strain is 12.2 eV / Å, the first fitting parameter is -0.636 eV / Å, the second fitting parameter is 4.039 eV / Å, and the third fitting parameter is 0.074.

[0098] See attached document Figure 5 The present invention provides a method for calculating hole mobility by introducing strain-dependent optical deformation potential in computer-aided design software, which may include step S300, substituting the strain-dependent optical deformation potential into the carrier transport model to calculate the carrier mobility of the target semiconductor device channel region under strain distribution.

[0099] Step S300 specifically involves analyzing the scattering physics of charge carriers at different energy states. The scattering process within a semiconductor device consists of a superposition of absorption and emission processes. Since charge carriers can absorb phonons at any energy state, the scattering rate of the absorption process exhibits a continuously increasing numerical distribution as the charge carrier energy increases. The emission process has a specific energy threshold characteristic. When the charge carrier energy is lower than the optical phonon energy of the semiconductor material itself, the charge carrier's own energy is insufficient to excite optical phonons, causing the scattering rate corresponding to the emission process to drop to zero. The emission process is only triggered when the charge carrier energy increases and crosses the optical phonon energy threshold. The numerical distribution of the total scattering process shows a step increase at the position crossing the optical phonon energy threshold.

[0100] In the absence of applied strain, holes in the channel region are subjected to strong optical phonon scattering. When biaxial tensile strain is applied to the channel region of the target semiconductor device, if the traditional constant deformation potential model is used for calculation, the final output scattering rate value will only decrease to a limited extent due to the fact that the traditional model sets the optical phonon deformation potential as a fixed constant that does not change with the strain distribution.

[0101] When step S300 is executed based on the technical solution of this invention, the system incorporates the strain-dependent optical deformation potential determined according to the actual strain distribution into the calculation model. Because this nonlinear model introduces a physical mechanism in its underlying logic that strain-induced crystal symmetry breaking leads to a significant attenuation of the optical phonon deformation potential, the final calculated optical scattering rate is reduced by more than an order of magnitude compared to the result calculated using the constant deformation potential model. This numerical result indicates that the electro-acoustic coupling strength is significantly suppressed under strong strain conditions.

[0102] like Figure 5 As shown, Figure 5 The diagram consists of three side-by-side subplots (a), (b), and (c), which visually compare the distribution of hole-optical phonon scattering rate as a function of hole energy under different models. The horizontal axis of each subplot represents carrier energy in electron volts (eV), ranging from 0.0 to 0.5 eV; the vertical axis represents optical scattering rate on a logarithmic scale (approximately from 10...). 10 Up to 10 14 The figure contains three different contrast curves: the thin solid line represents the no-strain state; the dashed line represents the 2% biaxial tensile strain state calculated using the traditional constant deformation potential model; and the thick solid line with solid dots represents the 2% biaxial tensile strain state calculated using the dynamic deformation potential model proposed in this invention.

[0103] The three sub-graphs respectively illustrate the complete scattering process and its two independent components. Figure 5 (b) illustrates the absorption process, and since charge carriers can absorb phonons at any energy, the curve rises continuously across the entire energy range. Figure 5 (c) illustrates the emission process, and a threshold characteristic can be clearly observed: when the energy is below about 0.06 eV (corresponding to the optical phonon energy of silicon, about 63.3 meV), the emission scattering rate drops sharply to zero because the hole energy is insufficient to excite an optical phonon; the emission process is only activated when the energy crosses this threshold. Figure 5 (a) shows the total scattering process, which is the superposition of absorption and emission processes, thus exhibiting a very obvious step jump at about 0.06 eV.

[0104] By comparing these three curves, the significant differences and conclusions of the model in this invention can be clearly demonstrated. In the unstrained state (thin solid line), the hole experiences extremely strong optical phonon scattering. When a 2% biaxial tensile strain is applied, if the conventional model is used (dotted line), the scattering rate only shows a moderate (less than one order of magnitude) decrease due to the change in band structure (density of states). However, when the model of this invention is introduced (thick solid line with dots), the calculated scattering rate shows a dramatic decrease—compared to the conventional model, its scattering rate is further reduced by more than one order of magnitude (e.g., in...). Figure 5 (a) Above the step point, from approximately 10 13 Reduced to as low as 10 12 (See below). This demonstrates that under strong strain conditions, the fixed deformation potential model used in traditional TCAD simulations significantly overestimates the optical phonon scattering rate, while the model of this invention can accurately capture the microscopic physical reality of the significantly weakened electro-acoustic coupling strength.

[0105] In performing the above calculations, the system extracts and calculates the scattering rates of holes and optical phonons based on a defined strain-dependent optical deformation potential using the Fermi gold rule. This scattering rate calculation model satisfies the following mathematical expression:

[0106] ;

[0107] in, Represents the scattering rate of holes and optical phonons. Represents the carrier energy. This represents the constant pi. This represents the reduced Planck constant. Represents the strain-dependent optical deformation potential. The density of a semiconductor material is indicated by its material density. Indicates the optical phonon angular frequency. This represents the optical phonon energy of semiconductor materials. The Bose-Einstein distribution function of phonons, This represents the density of band states of the final scattering state.

[0108] After obtaining the aforementioned scattering rate data, the system further extracts the reciprocal of the calculated scattering rates of holes and optical phonons as the relaxation time, and substitutes it into the relaxation time approximation model based on the Boltzmann transport equation. By performing weighted integration over different energy bands and different energy states, the system finally calculates and obtains the carrier mobility data limited by optical phonon scattering. This mobility integral calculation model satisfies the following mathematical expression:

[0109] ;

[0110] in, This represents the carrier mobility limited by optical phonon scattering. Represents the elementary charge. This indicates a summation operation on all band indices in a semiconductor material. This represents the relaxation time and is numerically equivalent to the reciprocal of the aforementioned hole and optical phonon scattering rates. This represents the squared value of the mean group velocity of the corresponding energy band on the isoenergetic surface. This represents the density of band states for the corresponding energy band. This represents the Fermi-Dirac distribution function of the corresponding energy band. The differential variable represents the integral over the carrier energy. Represents the Boltzmann constant. This indicates the lattice temperature of the environment in which the target semiconductor device is located.

[0111] See attached document Figure 6 The present invention provides a method for calculating hole mobility by introducing strain-dependent optical deformation potential in computer-aided design software, which may include: simulating and predicting the electrical performance of the target semiconductor device based on the calculated carrier mobility.

[0112] After obtaining the hole mobility limited by optical phonon scattering, a macroscopic device performance simulation verification step is performed. The hole mobility data limited by optical phonon scattering is input into the performance prediction module 40. Based on the carrier transport model and the calculated optical phonon deformation potential, the performance prediction module 40 calculates and outputs the macroscopic electrical performance parameters of the target semiconductor device.

[0113] Figure 6 The effect of applying biaxial strain along the zero-zero plane on hole mobility is demonstrated. Simulation results show that hole mobility is in a minimum range under strain-free conditions. When a positive tensile strain is applied to the target semiconductor material, hole mobility exhibits a nonlinear increasing trend with increasing strain amplitude. When a negative compressive strain is applied to the target semiconductor material, the increase in hole mobility shows a gradual characteristic. The simulation trends generated by the performance prediction module 40 are consistent with predictions from current state-of-the-art computing methods and data reported in the literature.

[0114] Further details Figure 6 The overall structure and data presentation. Figure 6 The vertical axis represents hole mobility, in cm. 2 / Vs; the horizontal axis represents biaxial strain, ranging from 0% (no strain) to 2% (tensile strain). In the figure, the red solid line and hollow dots represent the prediction results of the calculation model proposed in this invention, the gray solid line and five-pointed stars represent the prediction values ​​using the current advanced full first-principles calculation method, the black solid line and triangles represent the calculation results of the model without considering the influence of strain on the deformation potential, and the scatter dots of other shapes represent data from existing literature.

[0115] exist Figure 6 The effect of in-plane biaxial strain on hole mobility is shown in the figure. According to the prediction results of the present invention, the hole mobility reaches a minimum near no strain (0%); when a positive tensile strain is applied, the hole mobility exhibits a very obvious and sharp upward trend, reaching 4 × 10⁻⁶ at 2% tensile strain. 3 cm 2 / Vs or more. The predicted trends of this invention are in high agreement with current advanced algorithms and data from the literature by Fischetti et al.

[0116] The simulation results generated by the performance prediction module 40 are used to set the process parameters of the semiconductor device. For the target semiconductor device, the mechanical stress field in the channel region is configured using the simulation results. The target semiconductor devices include gate-around nanosheet field-effect transistors and P-type metal-oxide-semiconductor field-effect transistors with strained channels. In the device structure design, the physical parameters in the semiconductor manufacturing process are derived and set in reverse based on the stress type and stress amplitude indices determined by the performance prediction module 40.

[0117] The specific process setup steps involve mapping the target strain parameters to specific device structure and material manufacturing parameters. By configuring the lattice mismatch value of the stress substrate, setting the intrinsic stress value of the stress capping layer, and adjusting the composition ratio of the epitaxial growth layer, a specified uniaxial compressive strain or biaxial tensile strain is introduced into the channel region of the target semiconductor device. This strain configuration is used to induce geometric deformation in a specific direction in the semiconductor crystal, reducing the crystal symmetry level to a state that triggers optical scattering prohibition. This simulation-based process parameter setting method reduces the steps of empirically fitting and calibrating the underlying physical model, providing quantitative physical parameter basis for the device structure design of semiconductor transistors.

[0118] The present invention provides a method for determining the parameters of an optical phonon deformation potential model for semiconductor materials, which may include the following specific execution steps.

[0119] A series of preset strains of varying magnitudes are applied to the unit cell model of the target semiconductor material to establish atomic structure models under multiple strain states. The target semiconductor material includes silicon, germanium, or compound semiconductor materials. Crystal unit cell structures corresponding to different strain conditions are constructed by changing the lattice constant of the initial unstrained unit cell. The preset strains include uniaxial strain applied along a specific crystal direction and biaxial tensile and compressive strains acting on specific crystal planes. The execution device generates a series of three-dimensional atomic structure models with different degrees of deformation and different crystal symmetries by geometrically scaling the initial lattice vectors.

[0120] For each established atomic structure model under various strain states, first-principles calculations were used to obtain the coupling matrix elements between different electronic states and phonon modes within the Brillouin zone under that state. A density functional theory-based calculation module was employed to perform self-consistent field iterative calculations on the atomic structure models under various strain states to obtain the ground-state electronic structure data of the target material. Based on the converged electron charge density calculation, the lattice dynamics matrix was solved using density functional perturbation theory, thereby calculating the phonon dispersion relation corresponding to that specific strain state. Subsequently, the valence band hole energy state distribution information near the Brillouin zone center and the eigenvector data corresponding to various optical phonon modes were extracted. Based on the extracted hole energy state data and phonon eigenvector data, the scattering coupling matrix element values ​​between hole electronic states and various optical phonon modes were calculated.

[0121] The equivalent optical-phonon deformation potential (OPP) values ​​under strain conditions are extracted based on the calculated coupling matrix elements. According to the deformation potential scattering theory framework, the microscopic electro-acoustic coupling matrix element values ​​are transformed into macroscopic equivalent OPPP parameters. During data processing of the calculation results under strain conditions, characteristic data of the OPPPs near the Brillouin zone center point caused by strain-induced crystal symmetry breaking are extracted. The matrix element characteristic data under different energy states and momentum spaces are transformed into a discrete set of deformation potential values ​​varying with material deformation, thereby obtaining the absolute values ​​of the equivalent OPPPs for each independent strain state.

[0122] Nonlinear fitting was performed on multiple equivalent optical phonon deformation potential values ​​extracted under different strain states to determine the parameters of the functional model describing the change of optical phonon deformation potential with strain. A nonlinear least squares algorithm was used to perform curve fitting on a series of extracted discrete optical phonon deformation potential values. The objective function model used in the fitting operation directly characterizes the physical characteristic that the optical phonon deformation potential gradually decreases with increasing strain intensity. Iterative optimization calculations were used to accurately determine the various control parameters in the objective function model to ensure that the model curve closely approximates the discrete values ​​extracted by first-principles calculations. The determined underlying function model parameters were stored and output as the basic physical reference data for calculating the carrier mobility in the channel region of the corresponding semiconductor device in subsequent execution steps.

[0123] This invention provides a computer device, which may include a processor, a memory, a communication interface, and a bus. The processor, memory, and communication interface are connected via the bus and communicate with each other. The memory stores computer programs, and the processor calls the computer programs in the memory to execute the hole mobility calculation method for introducing strain-dependent optical deformation potential into TCAD software provided in various embodiments of this invention.

[0124] The present invention also provides a computer-readable storage medium having a computer program stored thereon. When executed by a processor, the computer program implements the steps of the above-described method. Specifically, during program execution, the structural parameters and strain distribution parameters of the target semiconductor device are first obtained. Then, based on the strain distribution parameters, the strain-dependent optical deformation potential in the channel region of the target semiconductor device is determined. Subsequently, the strain-dependent optical deformation potential is substituted into a carrier transport model to calculate the carrier mobility in the channel region under the strain distribution. Finally, based on the calculated carrier mobility, the electrical performance of the target semiconductor device is simulated and predicted.

[0125] This invention also provides a computer program product, including computer instructions. When executed by a computing device, the computer instructions cause the computing device to perform the aforementioned semiconductor device performance simulation method. The program product can exist in the form of software code on storage media such as magnetic tape, disk, optical disk, and semiconductor memory, or can be downloaded and installed onto a computing device via a network. During execution, the instructions call a preset nonlinear function relationship to calculate the corresponding optical phonon shape potential, and utilize pre-determined fitting parameters using first-principles methods to ensure simulation accuracy.

[0126] Within this framework of computer equipment and storage media implementation, by introducing a strain-dependent optical deformation potential model, this invention can quantify the weakening effect of strain-induced symmetry breaking on electroacoustic coupling strength, filling the gap in high-precision strain device simulation methods for advanced integrated circuit process nodes, and providing accurate physical model support for the design of next-generation high-speed, low-power electronic devices.

Claims

1. A method for calculating hole mobility by introducing a strain-dependent optical deformation potential, characterized in that, Includes the following steps: Obtain the structural parameters and strain distribution parameters of the target semiconductor device; Based on the strain distribution parameters, a preset nonlinear function model is called to calculate and determine the strain-dependent optical deformation potential in the channel region of the target semiconductor device. The strain-dependent optical deformation potential characterizes the suppression effect of strain-induced crystal symmetry breaking on the electroacoustic coupling strength. The value of the strain-dependent optical deformation potential decreases with the increase of strain intensity. The strain-dependent optical deformation potential is substituted into the carrier transport model, and the carrier mobility of the target semiconductor device channel region under the strain distribution parameters is calculated in combination with the structural parameters, which is limited by the optical phonon scattering mechanism. The strain-dependent optical deformation potential is used to update the calculation results of the scattering rate inside the carrier transport model. The electrical performance of the target semiconductor device is simulated and predicted based on the carrier mobility limited by the optical phonon scattering mechanism, and the macroscopic electrical performance index of the target semiconductor device under the strain distribution parameters is output. The process of calculating and determining the strain-dependent optical deformation potential in the channel region of the target semiconductor device by calling a preset nonlinear function model includes: Based on the strain distribution parameters, the degree of lattice distortion variable is extracted; Based on the lattice distortion degree variable, the scattering between the hole state and the optical phonon mode in the central region of the Brillouin zone is transformed into a symmetric forbidden state, forming a coupling forbidden region in the central region of momentum space and triggering the forbidden effect; Based on the increase in strain intensity and the expansion range of the forbidden effect, the number of optical phonon mode couplings involved in scattering is reduced. The nonlinear function model is invoked, and the results of the reduction calculation are combined with the optical deformation potential under no strain and the fitting parameters related to the semiconductor material type and strain type for nonlinear mapping processing to obtain the value of the strain-dependent optical deformation potential. The process of calculating the carrier mobility of the target semiconductor device channel region under the strain distribution parameters, limited by the optical phonon scattering mechanism, includes: Substituting the strain-dependent optical deformation potential into the Fermi-Gold Law calculation model within the carrier transport model, the optical phonon scattering physical process of carriers in different energy states is analyzed. By combining carrier energy, optical phonon energy, the Bose-Einstein distribution function of phonons, and the band density of states, the scattering rate of holes and optical phonons is calculated. The inverse of the scattering rate of holes and optical phonons obtained from the calculation is extracted as the relaxation time; Substituting the relaxation time into the approximate solution model based on the Boltzmann transport equation, the carrier mobility limited by the optical phonon scattering mechanism is extracted by weighted integration over different energy bands and different energy states.

2. The method for calculating hole mobility by introducing strain-dependent optical deformation potential according to claim 1, characterized in that, The process of obtaining the structural parameters and strain distribution parameters of the target semiconductor device includes: The material type, geometric topology, doping concentration distribution, and electrode location information are extracted from the structural parameters. These structural parameters are used to define the physical dimensions of the target semiconductor device and the geometry of the channel region of the target semiconductor device. The strain type, strain direction, and strain value are extracted from the strain distribution parameters. These strain distribution parameters are used to characterize the mechanical stress state experienced by the target semiconductor device and the channel region of the target semiconductor device.

3. The method for calculating hole mobility by introducing strain-dependent optical deformation potential according to claim 1, characterized in that, The process of calculating and determining the strain-dependent optical deformation potential in the channel region of the target semiconductor device by calling a preset nonlinear function model includes: In TCAD software, the degree of lattice distortion is extracted based on the strain distribution parameters. Based on the lattice distortion degree variable, the scattering cross section between the hole state and the optical phonon mode in the central region of the Brillouin zone is set as a symmetry forbidden mapping variable that decays as the crystal symmetry decreases. The calculation of the strain-dependent optical deformation potential is performed based on the symmetry forbidden mapping variable.

4. The method for calculating hole mobility by introducing strain-dependent optical deformation potential according to claim 3, characterized in that, The process of calculating the strain-dependent optical deformation potential value based on the symmetry forbidden mapping variable includes: In the momentum space discrete grid of the TCAD software, the range parameter of the central region coupling forbidden zone expanding outwards corresponding to the symmetry forbidden mapping variable is calculated based on the increase of the strain intensity. Based on the range parameter, the number of optical phonon modes involved in scattering is reduced in the algorithm module; Based on the results of the reduction calculation, the value of the strain-dependent optical deformation potential, which decreases as the strain intensity increases, is output.

5. The method for calculating hole mobility by introducing strain-dependent optical deformation potential according to claim 4, characterized in that, The process of outputting the numerical value of the strain-dependent optical deformation potential, which decreases with increasing strain intensity, based on the results of the reduction calculation includes: The nonlinear function model is invoked, and the results of the reduction calculation are combined with the optical deformation potential under no strain, the first fitting parameter, the second fitting parameter and the third fitting parameter for nonlinear mapping processing to calculate the corresponding strain-dependent optical deformation potential value. The first fitting parameter, the second fitting parameter, and the third fitting parameter are values ​​that are pre-acquired and stored in the material parameter library of the TCAD software. These values ​​are determined based on the type of semiconductor material and the strain type of the target semiconductor device channel region.

6. The method for calculating hole mobility by introducing strain-dependent optical deformation potential according to claim 5, characterized in that, The process of pre-acquiring the parameters includes: By applying strain of different magnitudes to the cell model of the target semiconductor material, atomic structure models under multiple strain states are established. Metadata of the hole-optic phonon coupling matrix in the Brillouin zone under the multiple strain states was obtained by first-principles calculation. The numerical values ​​of the equivalent optical phonon deformation potential corresponding to the multiple strain states are extracted based on the calculated hole-optical phonon coupling matrix metadata. The numerical value of the equivalent optical phonon deformation potential is subjected to nonlinear fitting processing to determine the first fitting parameter, the second fitting parameter, and the third fitting parameter in the nonlinear function model describing the strain-dependent optical deformation potential as a function of strain.

7. The method for calculating hole mobility by introducing strain-dependent optical deformation potential according to claim 1, characterized in that, The process of analyzing the optical phonon scattering physics of charge carriers at different energy states includes: Calculate the scattering rate of the phonon absorption process, which increases continuously with increasing carrier energy; The energy threshold of the emission process is set according to the optical phonon energy, and the scattering rate of the emission process triggered after the charge carrier energy crosses the energy threshold is calculated, thus forming a numerical distribution of the total scattering process that exhibits a step-increase.

8. The method for calculating hole mobility by introducing strain-dependent optical deformation potential according to claim 1, characterized in that, The process of simulating and predicting the electrical performance of the target semiconductor device based on the carrier mobility limited by the optical phonon scattering mechanism includes: The carrier mobility, which is limited by the optical phonon scattering mechanism, is substituted into a set of semiconductor device equations, including the Poisson equation, the current continuity equation, and the drift-diffusion equation, for iterative solution. The current-voltage characteristics of the target semiconductor device under the strain distribution parameters and the macroscopic electrical performance indicators are output. The macroscopic electrical performance indicators include the transfer characteristic curve, the output characteristic curve, the transconductance, the turn-on voltage, and the switching speed.

9. The method for calculating hole mobility by introducing strain-dependent optical deformation potential according to claim 1, characterized in that, After outputting the macroscopic electrical performance indicators of the target semiconductor device under the strain distribution parameters, the hole mobility calculation method based on strain-dependent optical deformation potential introduced in the TCAD software also includes the process of setting the process parameters for the target semiconductor device: The target strain parameter is determined based on the macroscopic electrical performance index, and the target strain parameter is mapped to device structure parameters and material manufacturing parameters. The mechanical stress field of the target semiconductor device channel region is established using the device structure parameters and material manufacturing parameters. Uniaxial compressive strain or biaxial tensile strain is introduced into the target semiconductor device channel region to induce geometric deformation of the semiconductor crystal.

10. The method for calculating hole mobility by incorporating strain-dependent optical deformation potential according to claim 9, characterized in that, The process of establishing the mechanical stress field of the channel region of the target semiconductor device using the device structure parameters and material manufacturing parameters includes: Stress is introduced in the channel region of the target semiconductor device by setting the lattice mismatch value of the stress substrate; By setting the intrinsic stress value of the stress capping layer and adjusting the composition ratio of the epitaxial growth layer, the crystal symmetry of the semiconductor crystal is reduced to a state that triggers the optical scattering ban.

11. A hole mobility calculation device incorporating strain-dependent optical deformation potential, characterized in that, A method for calculating hole mobility by incorporating a strain-dependent optical deformation potential, as described in any one of claims 1 to 10, comprises: The parameter acquisition module (10) is used to acquire the structural parameters and strain distribution parameters of the target semiconductor device; The deformation potential calculation module (20) is used to calculate and determine the strain-dependent optical deformation potential in the channel region of the target semiconductor device based on the strain distribution parameters obtained by the parameter acquisition module (10) and a preset nonlinear function model. The mobility calculation module (30) is used to substitute the strain-dependent optical deformation potential determined by the deformation potential calculation module (20) into the carrier transport model, and calculate the carrier mobility of the target semiconductor device channel region under the strain distribution parameters limited by the optical phonon scattering mechanism in combination with the structural parameters. The performance prediction module (40) is used to simulate and predict the electrical performance of the target semiconductor device based on the carrier mobility limited by the optical phonon scattering mechanism calculated by the mobility calculation module (30), and output the macroscopic electrical performance index of the target semiconductor device under the strain distribution parameter.

12. A computer device, characterized in that, The device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the hole mobility calculation method according to any one of claims 1 to 10, which introduces a strain-dependent optical deformation potential.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the hole mobility calculation method according to any one of claims 1 to 10, which introduces a strain-dependent optical deformation potential.

Images