Electronic device epitaxial structure optimization design method based on multi-physical field coupling simulation

Abstract

The application belongs to the technical field of electronic devices, and provides an electronic device epitaxial structure optimization design method based on multi-physical field coupling simulation, comprising: constructing a multi-physical field coupling model of a target electronic device, including a carrier transport model and an optical mode solving model, bidirectional coupling between the carrier transport model and the optical mode solving model, obtaining distribution data of carrier concentration in space and distribution data of optical field intensity in space; based on the carrier concentration spatial distribution data and the optical field intensity spatial distribution data, calculating the spatial offset between the peak position of the carrier and the peak position of the optical field, and determining whether the mismatch state of the current epitaxial structure is a trigger state according to the spatial offset; if it is determined to be a trigger state, a mismatch compensation strategy is executed. The problem of low gain efficiency and high threshold current caused by ignoring spatial matching is solved, and the electro-optical conversion efficiency and output power stability of the epitaxial structure are improved.

Description

Technical Field

[0001] This invention belongs to the field of electronic device technology, specifically a method for optimizing the design of epitaxial structures of electronic devices based on multi-physics field coupling simulation. Background Technology

[0002] The performance of electronic devices such as semiconductor lasers, light-emitting diodes, and high electron mobility transistors is essentially determined by their epitaxial structure. Epitaxial structures typically contain multiple functional thin films, such as quantum wells, waveguide layers, and confinement layers. The thickness, material composition, and doping concentration of each layer together determine the core indicators of the device, such as output wavelength, electro-optical conversion efficiency, reliability, and lifetime.

[0003] However, existing epitaxial structure optimization methods based on multiphysics coupling simulation still have shortcomings in practical applications. Specifically, existing methods typically use output power, electro-optic conversion efficiency, or threshold current as optimization targets, adjusting the epitaxial layer thickness, composition, and doping concentration through intelligent algorithms. However, they neglect the crucial physical constraint of spatial matching between the optical field distribution and the carrier distribution. In actual devices, the peak position of the optical field intensity is determined by the waveguide structure, while the peak position of carrier injection is affected by doping distribution, band structure, and transport mechanism. The two often have spatial offsets, especially in multi-quantum-well structures, where quantum wells closer to the P-side have higher [energy density]. The hole injection efficiency is low, and the carrier distribution between quantum wells is extremely uneven, resulting in the optical field peak and the carrier peak region not being able to effectively overlap. Existing optimization methods do not take this spatial matching as an explicit optimization target, nor do they establish an active compensation mechanism during the optimization process. As a result, although the optimized epitaxial structure has good output power in simulation, it has problems such as low gain efficiency and high threshold current in actual operation. More seriously, this spatial mismatch will aggravate local heat accumulation and carrier leakage, causing gain saturation, wavelength drift and reliability degradation, making it difficult to verify the optimization results through physical testing, which seriously weakens the practical value of virtual design methods.

[0004] To this end, the present invention provides a method for optimizing the design of epitaxial structures of electronic devices based on multiphysics field coupling simulation. Summary of the Invention

[0005] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0006] The technical solution adopted by this invention to solve its technical problem is: an optimization design method for the epitaxial structure of electronic devices based on multi-physics field coupling simulation, comprising:

[0007] A multiphysics coupling model of the target electronic device is constructed, including a carrier transport model and an optical mode solution model. The carrier transport model and the optical mode solution model are bidirectionally coupled to obtain spatial distribution data of carrier concentration and optical field intensity.

[0008] Based on the spatial distribution data of carrier concentration and the spatial distribution data of optical field intensity, the spatial offset between the peak position of carrier concentration and the peak position of optical field is calculated, and the mismatch state of the current epitaxial structure is determined according to the spatial offset.

[0009] If the state is determined to be triggered, the mismatch compensation strategy is executed, including calling a compensation scheme that matches the current mismatch state from the preset structural compensation scheme library;

[0010] The thickness or composition gradient of the quantum wells near the P and N sides in the variable space is automatically adjusted and optimized according to the compensation scheme to form an asymmetric quantum well structure.

[0011] The adjusted epitaxial structure parameters are iteratively executed until the adaptation state is non-triggered and meets the target wavelength, output power and reliability requirements, and the optimal epitaxial structure scheme is output.

[0012] The beneficial effects of this invention are as follows:

[0013] This invention obtains spatial distribution data of carrier concentration and light field intensity by constructing a bidirectional coupled iterative simulation of a carrier transport model and an optical mode solution model. It calculates the spatial offset between the carrier peak and the light field peak in real time and dynamically determines the mismatch state. When the mismatch exceeds the threshold, a compensation strategy is automatically executed. By differentially adjusting the thickness gradient or composition gradient of the quantum wells near the P-side and N-side, an asymmetric quantum well structure is formed, which forces the carrier injection peak and the light field peak to be spatially aligned. This solves the problems of low gain efficiency and high threshold current caused by ignoring spatial matching, and improves the electro-optical conversion efficiency and output power stability of the epitaxial structure.

[0014] This invention actively compensates for the spatial mismatch between the optical field and charge carriers, avoiding local heat accumulation and charge carrier leakage caused by the mismatch, reducing the risk of gain saturation and wavelength drift. At the same time, the insertion of a strain compensation layer offsets the additional lattice mismatch stress caused by thickness gradient or composition gradient adjustment, maintaining the overall strain balance of the epitaxial structure, effectively suppressing the generation and propagation of dislocation defects, improving the reliability and lifetime of the device, and enabling the optimized epitaxial structure to meet the requirements of high reliability application scenarios.

[0015] This invention uses spatial matching as an explicit optimization objective. Through a pre-set structural compensation scheme library, it achieves automatic matching and invocation of compensation schemes for different mismatch levels (mild, moderate, and severe). Combined with a closed-loop iterative optimization mechanism, the optimization process can quickly converge to the optimal epitaxial structure scheme that meets the requirements of wavelength, power, and reliability. This avoids the need for extensive physical wafer verification that traditional trial-and-error methods rely on repeatedly adjusting structural parameters based on experience. It shortens the R&D cycle, reduces R&D costs, and enhances the practical value of virtual design methods. Attached Figure Description

[0016] The invention will now be further described with reference to the accompanying drawings.

[0017] Figure 1 This is a flowchart of the steps in Embodiment 1 of the present invention;

[0018] Figure 2 This is a module architecture diagram of Embodiment 2 of the present invention. Detailed Implementation

[0019] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0020] Example 1

[0021] Please see Figure 1 As shown in the embodiment of the present invention, the method for optimizing the epitaxial structure of electronic devices based on multiphysics coupling simulation includes the following steps:

[0022] Step S1: Construct a multi-physics coupling model of the target electronic device, including a carrier transport model and an optical mode solution model. The carrier transport model and the optical mode solution model are bidirectionally coupled to obtain spatial distribution data of carrier concentration and spatial distribution data of light field intensity.

[0023] The specific process of step S1 includes: obtaining the epitaxial structure stacking order of the target electronic device, wherein the epitaxial structure includes at least a substrate layer, a buffer layer, an N-type confinement layer, an N-type waveguide layer, a quantum well active region, a P-type waveguide layer, a P-type confinement layer, and an ohmic contact layer;

[0024] Material property parameters are configured for each epitaxial layer. These parameters include the material type, thickness, material composition, doping concentration, and doping type of each epitaxial layer. The material property parameters also include the thermal conductivity, refractive index, absorption coefficient, carrier mobility, carrier lifetime, and band structure parameters of each material at different temperatures.

[0025] The geometric model and material property parameters are input into the multiphysics simulation solver as the basis for coupled solution;

[0026] The boundary conditions for setting the carrier transport model include: applying a first voltage bias to the outer boundary of the N-type confinement layer and applying a second voltage bias to the outer boundary of the P-type confinement layer. The difference between the first voltage bias and the second voltage bias constitutes the injection voltage.

[0027] Set the initial carrier concentration distribution for the carrier transport model;

[0028] The initial carrier concentration distribution is initialized based on the doping concentration of each epitaxial layer and the carrier statistical distribution under thermal equilibrium conditions.

[0029] Set boundary conditions for the optical mode solution model, including: setting absorption boundary conditions on the transverse boundary perpendicular to the epitaxial layer growth direction, and setting reflection boundary conditions or periodic boundary conditions on the longitudinal boundary parallel to the epitaxial layer growth direction.

[0030] Set the initial light field intensity distribution for solving the optical mode;

[0031] The initial light field intensity distribution is initialized based on the refractive index distribution of the epitaxial structure and the preset initial light field mode;

[0032] The bidirectional coupled iterative solution between the carrier transport model and the optical mode solution model is performed as follows:

[0033] First coupling step: Input the light field intensity distribution data calculated by the current optical mode solution model into the carrier transport model;

[0034] Specifically, based on the light field intensity distribution data, the stimulated radiative recombination rate and stimulated absorptivity at each spatial location are calculated; the stimulated radiative recombination rate and stimulated absorptivity are then used as additional carrier recombination terms and coupled into the carrier continuity equation of the carrier transport model.

[0035] In particular, the higher the light field intensity, the greater the contribution of stimulated radiative recombination rate or stimulated absorptivity, thus affecting the spatial distribution of carrier concentration.

[0036] Based on the coupled carrier transport model, the carrier continuity equation and Poisson equation are solved to obtain updated carrier concentration spatial distribution data, current density distribution data and electric field distribution data.

[0037] The second coupling step is to input the spatial distribution data of carrier concentration calculated by the current carrier transport model into the optical mode solution model.

[0038] Specifically, based on the spatial distribution data of carrier concentration, the material gain coefficient or material loss coefficient at each spatial location is calculated;

[0039] Within the active region of the quantum well, the higher the carrier concentration, the greater the material gain coefficient; outside the active region of the quantum well, the carrier concentration distribution affects the absorption loss of free carriers.

[0040] The material gain coefficient or material loss coefficient is superimposed on the complex refractive index distribution of the optical mode solution model to update the complex refractive index at each spatial location;

[0041] Based on the updated complex refractive index distribution, Maxwell's equations or wave equations are solved to obtain updated spatial distribution data of light field intensity and optical mode distribution data.

[0042] To determine whether the bidirectional coupled iteration has converged, the specific steps are as follows:

[0043] The relative difference between the spatial distribution data of light field intensity obtained in this iteration and the spatial distribution data of light field intensity obtained in the previous iteration is calculated and used as the first iteration error;

[0044] The relative difference between the spatial distribution data of carrier concentration obtained in this iteration and the spatial distribution data of carrier concentration obtained in the previous iteration is calculated as the second iteration error;

[0045] The first iteration error is compared with a preset first convergence threshold, and the second iteration error is compared with a preset second convergence threshold.

[0046] If the first iteration error is less than the first convergence threshold and the second iteration error is less than the second convergence threshold, then the bidirectional coupling iteration is determined to have converged, and the converged calculation result is output.

[0047] If the first iteration error is greater than or equal to the first convergence threshold, or the second iteration error is greater than or equal to the second convergence threshold, then the bidirectional coupling iteration is determined to have failed to converge, and the spatial distribution data of the light field intensity and the spatial distribution data of the carrier concentration obtained in this iteration are taken as the new current values.

[0048] The converged carrier concentration spatial distribution data is used as the final output of the carrier transport model. The carrier concentration spatial distribution data includes the electron concentration and hole concentration at each spatial location.

[0049] The converged spatial distribution data of the light field intensity is used as the final output of the optical mode solution model. The spatial distribution data of the light field intensity includes the light field intensity value and the peak position of the light field at each spatial location.

[0050] Step S2: Based on the spatial distribution data of carrier concentration and the spatial distribution data of light field intensity, calculate the spatial offset between the peak position of carrier concentration and the peak position of light field, and determine whether the current mismatch state of the epitaxial structure is a triggered state based on the spatial offset.

[0051] Among them, the trigger state is the state corresponding to when the spatial offset exceeds a preset threshold;

[0052] The specific process of step S2 includes: extracting carrier concentration spatial distribution data and light field intensity spatial distribution data; confirming that the acquired data covers the complete epitaxial structure region of the target electronic device;

[0053] Along the growth direction of the epitaxial structure, the carrier concentration value at each spatial coordinate point is extracted. The carrier concentration value is the larger of the electron concentration value and the hole concentration value, or the carrier concentration value is the square root of the product of the electron concentration value and the hole concentration value.

[0054] Traverse all spatial coordinate points, compare the carrier concentration values ​​at each coordinate point, determine the location of the spatial coordinate point with the largest carrier concentration value, and mark the location of this spatial coordinate point as the carrier peak position;

[0055] If multiple spatial coordinate points have the same maximum carrier concentration value, then the geometric center of these coordinate points shall be taken as the carrier peak position.

[0056] The carrier peak position includes at least a first directional coordinate value along the growth direction of the epitaxial structure, and a second directional coordinate value and a third directional coordinate value perpendicular to the growth direction;

[0057] Traverse all spatial coordinate points, compare the light field intensity values ​​at each coordinate point, determine the location of the spatial coordinate point with the largest light field intensity value, and mark the location of this spatial coordinate point as the peak position of the light field.

[0058] If multiple spatial coordinate points have the same maximum light field intensity value, then the geometric center of these coordinate points is taken as the peak position of the light field.

[0059] The peak position of the light field includes at least the first directional coordinate value along the growth direction of the epitaxial structure, and the second and third directional coordinate values ​​perpendicular to the growth direction;

[0060] Calculate the coordinate difference between the peak position of the charge carriers and the peak position of the optical field along the growth direction of the epitaxial structure, and use it as the first directional offset.

[0061] Calculate the coordinate difference between the peak position of the charge carriers and the peak position of the optical field in the first transverse direction perpendicular to the growth direction, and use it as the offset in the second direction;

[0062] Calculate the coordinate difference between the peak position of the charge carriers and the peak position of the optical field in the second transverse direction perpendicular to the growth direction, and use it as the third transverse offset;

[0063] The first direction offset, the second direction offset, and the third direction offset are vectorized to obtain the total spatial offset;

[0064] The preset spatial offset thresholds include a first directional threshold along the growth direction of the epitaxial structure, and a second and third directional thresholds perpendicular to the growth direction; these thresholds are set according to the type of the target electronic device, the operating wavelength, and the thickness of the quantum well active region; for electronic devices with a quantum well active region thickness of less than ten nanometers, the first directional threshold is set to be within the range of one-third to one-half of the thickness of the quantum well active region.

[0065] Compare the first direction offset with the first direction threshold;

[0066] If the offset in the first direction is greater than or equal to the threshold in the first direction, then the mismatch state of the current epitaxial structure is determined to be a triggered state.

[0067] If the first direction offset is less than the first direction threshold, then the second direction offset is compared with the second direction threshold, and the third direction offset is compared with the third direction threshold.

[0068] If the offset in the second direction is greater than or equal to the threshold in the second direction, or the offset in the third direction is greater than or equal to the threshold in the third direction, then the mismatch state of the current epitaxial structure is determined to be the trigger state.

[0069] If the offset in the first direction is less than the threshold in the first direction, and the offset in the second direction is less than the threshold in the second direction, and the offset in the third direction is less than the threshold in the third direction, then the mismatch state of the current extensional structure is determined to be an untriggered state.

[0070] Step S3: If the state is determined to be triggered, the mismatch compensation strategy is executed, including calling a compensation scheme that matches the current mismatch state from the preset structural compensation scheme library;

[0071] The specific process of step S3 includes: comparing the first direction offset with multiple preset offset threshold ranges to determine the current mismatch level;

[0072] Specifically, when the offset in the first direction is less than the product of the first direction threshold and the first proportional coefficient, it is determined to be a slight mismatch; when the offset in the first direction is greater than or equal to the product of the first direction threshold and the first proportional coefficient and less than the product of the first direction threshold and the second proportional coefficient, it is determined to be a moderate mismatch; when the offset in the first direction is greater than or equal to the product of the first direction threshold and the second proportional coefficient, it is determined to be a severe mismatch.

[0073] Wherein, the first proportional coefficient is less than the second proportional coefficient, and the first proportional coefficient and the second proportional coefficient are set according to the type and performance requirements of the target electronic device;

[0074] If the offset direction information indicates that the carrier peak position is located in the direction closer to the P side of the optical field peak position, then the compensation direction is determined to be either to enhance the carrier injection capability of the N-side quantum well or to enhance the optical field confinement capability of the P-side quantum well.

[0075] If the offset direction information indicates that the carrier peak position is located in the direction closer to the N side of the optical field peak position, then the compensation direction is determined to be either to enhance the carrier injection capability of the P-side quantum well or to enhance the optical field confinement capability of the N-side quantum well.

[0076] If the offset direction information indicates that the peak position of the charge carriers and the peak position of the optical field are also offset in the direction perpendicular to the growth direction, then the lateral compensation direction is further determined. The lateral compensation direction is used to indicate the direction in which the lateral refractive index distribution of the waveguide layer needs to be adjusted.

[0077] The pre-stored structural compensation scheme library contains multiple sets of compensation schemes, each set of compensation schemes corresponding to a specific mismatch level and compensation direction combination;

[0078] The construction process of the structural compensation scheme library includes: for different types of electronic devices and different epitaxial structure systems, establishing the mapping relationship between mismatch state and structural compensation parameters through pre-existing multiphysics simulation experiments or historical experimental data;

[0079] The compensation scheme includes at least one of the following parameters:

[0080] The first type of compensation parameter is the thickness gradient configuration of the quantum wells near the P side and the quantum wells near the N side, including the thickness value of each quantum well and the thickness variation law.

[0081] The second type of compensation parameter is the composition gradient configuration of the quantum wells near the P side and the quantum wells near the N side, including the material composition values ​​of each quantum well and the composition variation law.

[0082] The third type of compensation parameter: the thickness and composition of the strain compensation layer inserted between the quantum well and the waveguide layer;

[0083] The fourth type of compensation parameter: the refractive index distribution adjustment parameter of the waveguide layer, including the waveguide layer thickness adjustment value and the doping concentration adjustment value;

[0084] Based on the mismatch level and compensation direction, the execution plan is matched and invoked;

[0085] If it is a slight mismatch, the compensation scheme corresponding to the slight mismatch is called from the structural compensation scheme library, including fine-tuning the thickness of the quantum well near the P side or near the N side, with the adjustment range within the preset first adjustment range;

[0086] If the mismatch is moderate, a compensation scheme corresponding to moderate mismatch is called from the structural compensation scheme library, including differential adjustment of the thickness of the quantum well near the P side and near the N side, and gradient adjustment of the composition of the quantum well at the same time. The adjustment range is within the preset second adjustment range, and the second adjustment range is greater than the first adjustment range.

[0087] If the mismatch is severe, a compensation scheme corresponding to the severe mismatch is called from the structural compensation scheme library. This includes inserting a strain compensation layer between the quantum well and the waveguide layer, and simultaneously adjusting the thickness and composition of the quantum well near the P side and near the N side in a coordinated manner. The adjustment range is within a preset third adjustment range, and the third adjustment range is greater than the second adjustment range.

[0088] Step S4: Automatically adjust the thickness gradient or composition gradient of the quantum wells near the P and N sides in the optimization variable space according to the compensation scheme to form an asymmetric quantum well structure.

[0089] Among them, the asymmetric quantum well structure is used to force the carrier injection peak and the optical field peak to be spatially aligned;

[0090] The specific process of step S4 includes: obtaining the compensation scheme, which at least includes the compensation type identifier and structural compensation parameters;

[0091] The compensation type identifier is used to indicate the type of technical means used in this compensation, including thickness gradient adjustment type, composition gradient adjustment type, strain compensation layer insertion type or a combination thereof;

[0092] The structural compensation parameters include at least one of the following parameters: thickness adjustment value of each quantum well, composition adjustment value of each quantum well, thickness and composition information of the newly added layer, and refractive index distribution adjustment parameter of the waveguide layer;

[0093] When the compensation scheme includes thickness gradient adjustment, perform the following operations:

[0094] Obtain the number of quantum wells in the multi-quantum-well active region and the current thickness value of each quantum well in the current optimization variable space;

[0095] Based on the target thickness value of the quantum well near the P side in the compensation scheme, the thickness of one or more quantum wells near the P side is adjusted to the target thickness value.

[0096] Based on the target thickness value of the quantum wells near the N side in the compensation scheme, the thickness of one or more quantum wells near the N side is adjusted to the target thickness value.

[0097] Based on the thickness gradient variation law in the compensation scheme, the thickness value of the intermediate quantum well located between the P side and the N side is adjusted so that the thickness of each quantum well is distributed in a gradient increasing or gradient decreasing distribution from the N side to the P side.

[0098] The thickness gradient variation pattern includes at least one of linear gradient, exponential gradient, or step gradient.

[0099] When the compensation scheme includes component gradient adjustment, perform the following operations:

[0100] Obtain the number of quantum wells in the multi-quantum-well active region in the current optimization variable space and the current material composition value of each quantum well;

[0101] Based on the target material composition value of the quantum well near the P side in the compensation scheme, the material composition of one or more quantum wells near the P side is adjusted to the target material composition value;

[0102] Based on the target material composition value of the quantum well near the N side in the compensation scheme, the material composition of one or more quantum wells near the N side is adjusted to the target material composition value;

[0103] Based on the composition gradient change law in the compensation scheme, the material composition value of the intermediate quantum well located between the P side and the N side is adjusted so that the material composition of each quantum well is distributed in a gradient increasing or gradient decreasing from the N side to the P side.

[0104] The material composition includes aluminum, indium or gallium components in aluminum gallium indium arsenide materials, or aluminum or indium components in gallium nitride materials;

[0105] When the compensation scheme includes strain compensation layer insertion, perform the following operations:

[0106] According to the insertion position determined in the compensation scheme, a strain compensation layer is inserted between the active region of the quantum well and the waveguide layer or between two quantum wells.

[0107] The thickness of the strain compensation layer is set according to the target thickness value of the strain compensation layer in the compensation scheme;

[0108] The material composition of the strain compensation layer is set according to the target material composition value of the strain compensation layer in the compensation scheme;

[0109] The strain compensation layer is used to offset the additional lattice mismatch stress caused by thickness gradient adjustment or composition gradient adjustment, and to maintain the overall strain balance of the epitaxial structure.

[0110] When the compensation scheme includes waveguide layer refractive index distribution adjustment, perform the following operations:

[0111] Based on the fourth type of compensation parameter in the compensation scheme, obtain the thickness adjustment value and doping concentration adjustment value of the waveguide layer;

[0112] Adjusting the thickness of the N-type or P-type waveguide layer can change the spatial position of the peak optical field intensity perpendicular to the growth direction.

[0113] Adjusting the doping concentration distribution of the N-type or P-type waveguide layer changes the effective refractive index of the waveguide layer, thereby shifting the lateral position of the peak optical field intensity.

[0114] After completing the above adjustments, an updated optimization variable space is generated, which includes the adjusted thickness values ​​of each quantum well, material composition values, newly added strain compensation layer parameters, and adjusted waveguide layer parameters.

[0115] The updated optimized variable space is input into the multiphysics coupling simulation model (carrier transport model and optical mode solution model) in step S1 for verification simulation.

[0116] The verification simulation adopts a simplified mode that reduces the grid density or the number of iterations, which can preliminarily confirm in a shorter time whether the spatial offset between the adjusted carrier peak position and the optical field peak position has decreased.

[0117] If the simulation results show that the spatial offset does not decrease or even increases, the thickness gradient or composition gradient is fine-tuned according to the preset adjustment step size until the spatial offset shows a decreasing trend.

[0118] The adjusted and verified effective optimized variable space is taken as the output of step S4.

[0119] Step S5: Iterate through the adjusted epitaxial structure parameters until the adaptation state is non-triggered and meets the target wavelength, output power and reliability requirements, and output the optimal epitaxial structure scheme.

[0120] The specific process of step S5 includes: taking the optimization variable space as the current candidate epitaxial structure parameters, returning to step S1, re-executing the multiphysics coupled simulation (coupled solution), and obtaining the updated carrier concentration spatial distribution data and light field intensity spatial distribution data; repeating steps S2 to S4 to determine the mismatch state, call the compensation scheme and adjust the structure of the updated epitaxial structure to form a closed loop iteration;

[0121] After each iteration, record the current iteration number and the performance indicators of the current epitaxial structure. The performance indicators include at least the spatial offset, output power, target wavelength deviation, and reliability lifetime prediction.

[0122] Determine whether the iteration termination condition is met. The iteration termination condition includes:

[0123] First termination condition: The mismatch state is not triggered, that is, the spatial offset is lower than the preset threshold;

[0124] Second termination condition: Output power is greater than or equal to the preset power threshold;

[0125] Third termination condition: The target wavelength deviation is less than the preset wavelength tolerance;

[0126] Fourth termination condition: The predicted reliability lifetime value is greater than or equal to the preset lifetime threshold.

[0127] The iteration termination condition is determined to be satisfied when the first termination condition is simultaneously satisfied along with at least one of the second, third, and fourth termination conditions.

[0128] If the iteration termination condition is not met, execution continues until the preset maximum number of iterations is reached or the iteration termination condition is met.

[0129] When the iteration termination condition is met, the current epitaxial structure parameters are output as the optimal epitaxial structure scheme. The optimal epitaxial structure scheme includes at least the thickness, material composition, doping concentration of each epitaxial layer, and the thickness gradient and composition gradient configuration of the quantum well.

[0130] This embodiment uses a bidirectional coupled iterative simulation of a carrier transport model and an optical mode solution model to obtain spatial distribution data of carrier concentration and light field intensity. It calculates the spatial offset between the carrier peak and the light field peak in real time and dynamically determines the mismatch state. When the mismatch exceeds a threshold, it automatically calls a compensation scheme matching the mismatch level and compensation direction from the structural compensation scheme library. By differentially adjusting the thickness or composition gradient of the quantum wells near the P-side and N-side, an asymmetric quantum well structure is formed. Combined with the insertion of a strain compensation layer and the adjustment of the waveguide layer refractive index, the carrier injection peak and the light field peak are spatially forced to align. This effectively solves the problems of low gain efficiency, high threshold current, and degraded reliability caused by neglecting the spatial matching of the light field and carriers. It improves the electro-optic conversion efficiency and device reliability of the epitaxial structure, achieving a leap from "only optimizing power indicators" to "actively compensating for spatial mismatch," significantly reducing the number of physical prototypes and enhancing the practical value of the virtual design method.

[0131] Example 2

[0132] Based on the same inventive concept as the multiphysics coupling simulation-based electronic device epitaxial structure optimization design method in the foregoing embodiments, such as Figure 2 As shown, this application provides a system for optimizing the epitaxial structure design of electronic devices based on multiphysics coupling simulation, wherein the system specifically includes:

[0133] Multiphysics Coupled Simulation Module: Constructs a multiphysics coupled model of the target electronic device, including a carrier transport model and an optical mode solution model. The carrier transport model and the optical mode solution model are solved by bidirectional coupling iteration to obtain the spatial distribution data of carrier concentration and optical field intensity.

[0134] Mismatch sensing module: Based on the spatial distribution data of carrier concentration and the spatial distribution data of light field intensity, calculate the spatial offset between the peak position of carrier concentration and the peak position of light field, and determine whether the current mismatch state of the epitaxial structure is a triggered state based on the spatial offset.

[0135] Among them, the trigger state is the state corresponding to when the spatial offset exceeds a preset threshold;

[0136] Compensation decision module: If the state is determined to be triggered, the mismatch compensation strategy is executed, including calling a compensation scheme that matches the current mismatch state from the preset structural compensation scheme library;

[0137] Structural compensation execution module: Automatically adjusts or optimizes the thickness gradient or composition gradient of quantum wells near the P and N sides in the variable space according to the compensation scheme to form an asymmetric quantum well structure;

[0138] The asymmetric quantum well structure is used to spatially force the alignment of the carrier injection peak with the optical field peak.

[0139] Iterative optimization module: Iteratively executes the adjusted epitaxial structure parameters until the adaptation state is non-triggered and meets the target wavelength, output power and reliability requirements, and outputs the optimal epitaxial structure scheme.

[0140] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for optimizing the epitaxial structure design of electronic devices based on multiphysics coupling simulation, characterized in that: include: A multiphysics coupling model of the target electronic device is constructed, including a carrier transport model and an optical mode solution model. The carrier transport model and the optical mode solution model are bidirectionally coupled to obtain spatial distribution data of carrier concentration and optical field intensity. Based on the spatial distribution data of carrier concentration and the spatial distribution data of optical field intensity, the spatial offset between the peak position of carrier concentration and the peak position of optical field is calculated, and the mismatch state of the current epitaxial structure is determined according to the spatial offset. If the state is determined to be triggered, the mismatch compensation strategy is executed, including calling a compensation scheme that matches the current mismatch state from the preset structural compensation scheme library; The thickness or composition gradient of the quantum wells near the P and N sides in the variable space is automatically adjusted and optimized according to the compensation scheme to form an asymmetric quantum well structure. The adjusted epitaxial structure parameters are iteratively executed until the adaptation state is non-triggered and meets the target wavelength, output power and reliability requirements, and the optimal epitaxial structure scheme is output.

2. The method for optimizing the epitaxial structure design of electronic devices based on multiphysics coupling simulation according to claim 1, characterized in that: The process of obtaining spatial distribution data of carrier concentration is as follows: Obtain the epitaxial stacking order of the target electronic device and the material property parameters of each epitaxial layer; set the boundary conditions and initial carrier concentration distribution of the carrier transport model; input the light field intensity distribution data calculated by the optical mode solution model into the carrier transport model, calculate the stimulated radiative recombination rate and stimulated absorptivity based on the light field intensity distribution and couple them into the carrier continuity equation; solve the carrier continuity equation and Poisson equation to obtain the spatial distribution data of carrier concentration; repeat the iteration until convergence and output the spatial distribution data of carrier concentration.

3. The method for optimizing the epitaxial structure design of electronic devices based on multiphysics coupling simulation according to claim 2, characterized in that: The process of obtaining spatial distribution data of light field intensity is as follows: Set the boundary conditions and initial light field intensity distribution of the optical mode solution model; input the carrier concentration spatial distribution data calculated by the carrier transport model into the optical mode solution model, calculate the material gain coefficient or material loss coefficient based on the carrier concentration distribution and superimpose it into the complex refractive index distribution; Solve Maxwell's equations to obtain spatial distribution data of light field intensity; repeat the iteration until convergence, and output the spatial distribution data of light field intensity.

4. The method for optimizing the epitaxial structure design of electronic devices based on multiphysics coupling simulation according to claim 1, characterized in that: The process of obtaining spatial offset is as follows: The coordinate difference between the peak position of the charge carriers and the peak position of the optical field along the growth direction of the epitaxial structure is calculated as the first directional offset; the coordinate difference in the first transverse direction perpendicular to the growth direction is calculated as the second directional offset; and the coordinate difference in the second transverse direction perpendicular to the growth direction is calculated as the third directional offset. The first direction offset, the second direction offset, and the third direction offset are vectorized to obtain the total spatial offset.

5. The method for optimizing the epitaxial structure design of electronic devices based on multiphysics coupling simulation according to claim 4, characterized in that: The process of obtaining the peak positions of charge carriers and the peak positions of the optical field is as follows: The carrier concentration values ​​at each spatial coordinate point are extracted along the growth direction of the epitaxial structure. The carrier concentration value is the maximum of the electron concentration value and the hole concentration value. Traverse all spatial coordinate points and determine the position of the spatial coordinate point with the largest carrier concentration value as the carrier peak position. If there are multiple spatial coordinate points with the same maximum carrier concentration value, take the geometric center position of these coordinate points as the carrier peak position. Traverse all spatial coordinate points and determine the position of the spatial coordinate point with the largest light field intensity value as the peak position of the light field. If there are multiple spatial coordinate points with the same maximum light field intensity value, then take the geometric center position of these coordinate points as the peak position of the light field.

6. The method for optimizing the epitaxial structure design of electronic devices based on multiphysics coupling simulation according to claim 1, characterized in that: The process of determining whether the current mismatch state of the epitaxial structure is a triggered state based on the spatial offset is as follows: If the offset in the first direction is greater than or equal to the threshold in the first direction, it is determined to be a triggered state; If the first direction offset is less than the first direction threshold, then the second direction offset is compared with the second direction threshold, and the third direction offset is compared with the third direction threshold. If the offset in the second direction is greater than or equal to the threshold in the second direction, or the offset in the third direction is greater than or equal to the threshold in the third direction, then it is determined to be a triggered state.

7. The method for optimizing the epitaxial structure design of electronic devices based on multiphysics coupling simulation according to claim 1, characterized in that: The process of calling a compensation scheme that matches the current mismatch state from the preset structural compensation scheme library is as follows: The first directional offset is compared with multiple preset offset threshold ranges to determine the current mismatch level, including mild, moderate and severe mismatch. The compensation direction is determined based on the offset direction information. Based on the mismatch level and compensation direction, the corresponding compensation scheme is called from the structural compensation scheme library. For mild mismatch, a scheme that fine-tunes the thickness of the quantum well near the P side or near the N side is called. For moderate mismatch, a scheme that differentiates the thickness of the quantum well near the P side and near the N side and simultaneously adjusts the composition is called. For severe mismatch, a scheme that inserts a strain compensation layer and simultaneously adjusts the thickness and composition of the quantum well is called.

8. The method for optimizing the epitaxial structure design of electronic devices based on multiphysics coupling simulation according to claim 1, characterized in that: The process of forming an asymmetric quantum well structure is as follows: Obtain the compensation type identifier and structural compensation parameters in the compensation scheme; when thickness gradient adjustment is included, adjust the thickness of the quantum wells near the P side and near the N side according to the target thickness value, and make the thickness of the intermediate quantum wells gradient-distributed; when composition gradient adjustment is included, adjust the composition of the quantum wells near the P side and near the N side according to the target material composition value, and make the composition of the intermediate quantum wells gradient-distributed; when strain compensation layer insertion is included, insert a strain compensation layer between the quantum wells and the waveguide layer or between the quantum wells; when waveguide layer refractive index distribution adjustment is included, adjust the thickness or doping concentration distribution of the waveguide layer.

9. The method for optimizing the epitaxial structure design of electronic devices based on multiphysics coupling simulation according to claim 8, characterized in that: After automatically adjusting the thickness or composition gradient of the quantum wells near the P and N sides in the optimization variable space according to the compensation scheme, the process also includes a step to verify the adjustment effect: The adjusted optimized variable space was input into the multiphysics coupling simulation model for verification simulation, and it was initially confirmed whether the spatial offset between the adjusted carrier peak position and the optical field peak position was reduced. If the simulation results show that the spatial offset does not decrease or increases, then the thickness gradient or composition gradient is fine-tuned according to the preset adjustment step size until the spatial offset shows a decreasing trend; the optimized variable space that has been adjusted and verified to be effective is used as the output result.

10. The method for optimizing the epitaxial structure design of electronic devices based on multiphysics coupling simulation according to claim 1, characterized in that: The process of outputting the optimal epitaxial structure scheme is as follows: The adjusted optimization variable space is used as the current candidate extensional structure parameters. The coupling solution is then re-executed, and the mismatch state determination, compensation scheme invocation and structural adjustment are repeatedly executed to form a closed loop iteration. After each iteration, record the current iteration number and the performance metrics of the current epitaxial structure; Determine whether the iteration termination condition is met. The iteration termination condition includes at least one of the following: the mismatch state is not triggered, the output power is greater than or equal to the preset power threshold, the target wavelength deviation value is less than the preset wavelength tolerance, and the reliability lifetime prediction value is greater than or equal to the preset lifetime threshold, and the mismatch state is not triggered at the same time. When the iteration termination condition is met, the current epitaxial structure parameters are output as the optimal epitaxial structure scheme.

Images