A simulation method and system for the evolution of irradiation defects in a semiconductor device

Abstract

The present invention provides a simulation method and system for the evolution of irradiation defects in a semiconductor device, belonging to the technical field of simulation. The method includes: using the molecular dynamics method to simulate the defect evolution of the semiconductor device and obtaining the defect healing rate of the semiconductor device; obtaining the initial defect concentration of the semiconductor device and obtaining the steady-state defect concentration of the semiconductor device according to the initial defect concentration and the defect healing rate; using the steady-state defect concentration of the semiconductor device as an input parameter to perform performance simulation of the semiconductor device and obtaining the relationship between the performance of the semiconductor device and the defect healing rate. After obtaining the defect healing rate by using the molecular dynamics method, the present invention uses the healing rate as an input parameter to perform semiconductor device performance simulation, and intuitively and accurately characterizes the defect healing rate of semiconductor device irradiation through the performance of the semiconductor device.

Description

Technical Field

[0001] The present invention relates to the field of simulation technology, and in particular, to a method and system for simulating the evolution of irradiation defects in semiconductor devices. Background Art

[0002] When semiconductor electronic devices are in the space environment for a long time, ray particles such as neutrons, protons, heavy ions, etc. interact with atoms in semiconductor materials, including the collision process between ray particles and material atoms, the defect formation process, and the microstructure evolution process, resulting in radiation damage, causing changes in the physical properties, mechanical potential energy, and tissue composition and structure of the material. The defects caused by radiation damage are mainly generated through ionization damage and displacement damage. For most materials, it is mainly displacement damage. When an incident particle collides with the nucleus in a crystal material, part of the energy is converted into the recoil kinetic energy of the target atom. When this kinetic energy exceeds the binding energy of the lattice position, the atom can leave the lattice position, which is called displacement, and the resulting damage is called displacement damage. The atom directly collided by the incident particle to cause displacement is called a primary knock-on atom (PKA). If the primary knock-on atom has enough kinetic energy to continue to collide with other lattice atoms, this process is called a cascade collision. The atom leaving the lattice point becomes an interstitial atom, and the original lattice point forms a vacancy due to the lack of an atom. At a certain temperature, the interstitial atom and the vacancy can migrate and then recombine with each other, causing a certain degree of defect healing.

[0003] Since the time scale of the cascade process triggered by PKA occurs in the femtosecond to picosecond level, it is difficult to observe experimentally, and this time scale is within the range that can be processed by molecular dynamics (MD) simulation. Therefore, the molecular dynamics method can be used to simulate the evolution of defects caused by PKA. However, currently, only the defect situation of the structure at the molecular dynamics steady state is considered, while ignoring the influence of the defect healing rate on the device performance, making the simulation relatively abstract and the accuracy relatively low. Summary of the Invention

[0004] The present invention solves the problem that the irradiation defect healing rate in semiconductor devices cannot be intuitively and accurately characterized.

[0005] To solve the above problems, the present invention provides a method for simulating the evolution of irradiation defects in semiconductor devices, including:

[0006] Using the molecular dynamics method to simulate the defect evolution of a semiconductor device and obtaining the defect healing rate of the semiconductor device;

[0007] Obtaining the initial defect concentration of the semiconductor device, and obtaining the steady-state defect concentration of the semiconductor device according to the initial defect concentration and the defect healing rate;

[0008] Taking the steady-state defect concentration of the semiconductor device as an input parameter, perform performance simulation of the semiconductor device to obtain the relationship between the performance of the semiconductor device and the defect healing rate.

[0009] Preferably, the defect evolution simulation of the semiconductor device by using the molecular dynamics method and obtaining the defect healing rate of the semiconductor device include:

[0010] Obtain the peak state structure when the number of generated defects is the largest and the steady-state structure when the number of generated defects is stable during the defect evolution simulation process, and respectively obtain the defect information in the peak state structure and the steady-state structure;

[0011] According to the defect information in the peak state structure and the defect information in the steady-state structure, obtain the defect healing rate.

[0012] Preferably, the obtaining the defect healing rate according to the defect information in the peak state structure and the defect information in the steady-state structure includes:

[0013] Obtain the defect healing rate c according to the following formula:

[0014]

[0015] where A is the number of defects in the peak state structure and B is the number of defects in the steady-state structure.

[0016] Preferably, the obtaining the initial defect concentration of the semiconductor device includes:

[0017] Use SRIM software to calculate the initial defect concentration of the semiconductor device, where the initial defect concentration refers to the defect concentration in the semiconductor device without considering defect healing.

[0018] Preferably, the obtaining the steady-state defect concentration of the semiconductor device according to the initial defect concentration and the defect healing rate includes:

[0019] Obtain the steady-state defect concentration according to the following formula:

[0020] n S =n 0 (1 - c);

[0021] where n 0 is the initial defect concentration, n S is the steady-state defect concentration, and c is the defect healing rate.

[0022] Preferably, the defect evolution simulation of the semiconductor device using the molecular dynamics method includes a primary evolution stage, a secondary evolution stage, and an ultimate evolution stage. The time step of the primary evolution stage is 0.01 fs, the time step of the secondary evolution stage is 0.1 fs, the time step of the ultimate evolution stage is 1 fs, and the total evolution time of the primary evolution stage, the secondary evolution stage, and the ultimate evolution stage is 20 ps.

[0023] Preferably, the defect evolution simulation of the semiconductor device using the molecular dynamics method further includes: setting the coordinates, kinetic energy, and displacement of each atom in the output evolution region during the defect evolution simulation.

[0024] Preferably, before the defect evolution simulation of the semiconductor device using the molecular dynamics method, it further includes: performing a meshing process on the semiconductor device and establishing a system model of the same size as the mesh.

[0025] Preferably, the simulation ensemble of the molecular dynamics method includes the NVE ensemble, and a two-temperature model is used to control the temperature of the simulation system of the molecular dynamics method.

[0026] The advantage of the simulation method for irradiated defect evolution in the semiconductor device of the present invention compared with the prior art is that:

[0027] After obtaining the defect healing rate using the molecular dynamics method in the present invention, the healing rate is used as an input parameter for semiconductor device performance simulation. Since the performance calculation results of the device can be easily compared with experiments, therefore, the method of the present invention can better connect the molecular dynamics simulation and the semiconductor device performance simulation, and intuitively and accurately characterize the defect healing rate of the irradiated semiconductor device through the performance of the semiconductor device.

[0028] The present invention also provides a simulation system for irradiated defect evolution in a semiconductor device, including:

[0029] Molecular dynamics evolution module: used to perform defect evolution simulation on the semiconductor device using the molecular dynamics method;

[0030] Acquisition module: used to acquire the defect healing rate, initial defect concentration, and steady-state defect concentration of the semiconductor device, and also used to acquire the relationship between the performance parameters of the semiconductor device and the defect healing rate;

[0031] Performance simulation module: used to use the steady-state defect concentration of the semiconductor device as an input parameter to perform performance simulation of the semiconductor device.

[0032] The advantages of the simulation system for the evolution of irradiation defects in the semiconductor device of the present invention compared with the prior art are the same as those of the simulation method for the evolution of irradiation defects in the semiconductor device, which will not be elaborated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a flowchart of the simulation method for the evolution of irradiation defects in the semiconductor device in the embodiment of the present invention;

[0034] Figure 2 It is a diagram showing the calculation results of the defect evolution of Si material and GaN material in the embodiment of the present invention;

[0035] Figure 3 It is a schematic structural diagram of the AlGaN / GaN-HEMTs device in the embodiment of the present invention;

[0036] Figure 4 It is a diagram showing the threshold voltage shift of the GaN material before and after radiation in the embodiment of the present invention;

[0037] Figure 5 It is a diagram showing the calculation results of the electrical performance of the AlGaN / GaN-HEMTs device after being irradiated in the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0038] In order to make the above objects, features and advantages of the present invention more obvious and understandable, the following detailed description of the specific embodiments of the present invention will be given with reference to the accompanying drawings.

[0039] Please refer to Figure 1 As shown, a simulation method for the evolution of irradiation defects in a semiconductor device according to an embodiment of the present invention includes:

[0040] Step S1, using the molecular dynamics method to simulate the defect evolution of the semiconductor device, and obtaining the defect healing rate of the semiconductor device;

[0041] Step S2, obtaining the initial defect concentration of the semiconductor device, and obtaining the steady-state defect concentration of the semiconductor device according to the initial defect concentration and the defect healing rate;

[0042] Step S3, taking the steady-state defect concentration of the semiconductor device as an input parameter, performing performance simulation on the semiconductor device, and obtaining the relationship between the performance parameters of the semiconductor device and the defect healing rate.

[0043] The devices in this embodiment include various semiconductor devices, such as semiconductor silicon (Si) materials and GaN-based materials. Among them, GaN, namely gallium nitride, belongs to the third-generation semiconductor materials. The process of incident particle irradiation causing displacement damage to semiconductor devices is mainly due to the incident particles hitting the lattice atoms in the semiconductor device materials to generate primary knock-on atoms (PKAs). Then, the primary knock-on atoms trigger cascade collisions, high-temperature annealing, and finally tend to be stable. During the cascade collision process, interstitial atoms and vacancies are continuously generated. In the subsequent process, under certain temperature conditions, the interstitial atoms can recombine with the vacancies, or diffuse to dislocations, grain boundaries, or surfaces and annihilate, or aggregate into vacancy clusters or form dislocation loops.

[0044] Since the time scale of the cascade process triggered by primary knock-on atoms occurs in the femtosecond to picosecond level, in the prior art, the molecular dynamics (MD) method is used for simulation. During the implementation process, an input file readable by LAMMPS is written according to the specific parameters of the secondary particles obtained by Geant4. The energy, space, and angular distributions of PKAs, secondary charged particles, and recoil nuclear particles are used as input parameters to simulate the microscopic damage process of PKA cascade collisions in the material by LAMMPS. Before introducing PKA, the energy of the system is minimized and relaxed by methods such as CG (conjugate gradient). Based on the simulation of the cascade process, relevant defect identification methods, such as the Wigner-Seits method, can identify the defects (vacancies, interstitial atoms, etc.), so as to analyze the types and properties of the defects.

[0045] Although the molecular dynamics simulation can quantitatively obtain the change of the number of defects with time during the defect evolution process of semiconductor devices, however, it is often relatively abstract to characterize the healing rate only using the number of defects in the steady-state structure and the maximum number of structural defects during the evolution process, and it is not closely combined with the device performance. Therefore, in this embodiment, the defect evolution of the semiconductor device is first simulated by the molecular dynamics method to calculate the defect healing rate of the semiconductor device. Then, according to the initial defect concentration and the defect healing rate of the device, the steady-state defect concentration of the device is calculated, that is, the defect concentration at the steady-state structure completed by the molecular dynamics method evolution. Finally, using semiconductor process simulation and device simulation tools, the electrical performance of the device is calculated according to the steady-state defect concentration. Thus, it is very convenient to compare the calculation results of the electrical performance with the experiment, so that the defect healing rate of the semiconductor device can be intuitively characterized.

[0046] After obtaining the defect healing rate using the molecular dynamics method in this embodiment, the healing rate is used as an input parameter for semiconductor device performance simulation. Since the calculation results of the device performance can be easily compared with the experiment, the method of the present invention can better connect the molecular dynamics simulation and the semiconductor device performance simulation, and intuitively and accurately characterize the defect healing rate of semiconductor device irradiation through the performance of the semiconductor device.

[0047] In some embodiments, before the defect evolution simulation of the semiconductor device by using the molecular dynamics method in step S1, it further includes: performing a meshing process on the semiconductor device and establishing a system model with the same size as the mesh.

[0048] Among them, when performing the meshing process on the device, to make the mesh density in the semiconductor device moderate and for subsequent calculation convenience, the mesh (box) size is set to 70nm * 70nm * 70nm. After the mesh division is completed, the meshes without PKA are cleared and the meshes containing PKA are retained. Then, the information of the PKA contained in each mesh after the incident particle is incident is statistically analyzed, and this information includes the number, energy, position, incident direction, etc. of the PKA.

[0049] Among them, establishing a system model with the same size as each mesh includes: establishing a pure silicon system model in LAMMPS, and expanding the Si unit cell by an appropriate multiple in the x, y, and z directions in LAMMPS to establish a Si structure with the same size as the box for subsequent simulation processes. For example, the Si unit cell is expanded 129 times in the x, y, and z directions to obtain a pure Si model of 70nm * 70nm * 70nm.

[0050] In a preferred example, according to the actual conditions, doping particles into the pure Si model, such as doping trace amounts of C, O, etc. or doping atoms such as B, P, etc. into the pure Si model, to ensure the conformity of the simulation model with the actual situation.

[0051] In some of these embodiments, in step S1 when performing the molecular dynamics simulation, parameters such as PKA information, ensemble, simulation step size, and evolution time need to be set, select an appropriate force field according to the different systems corresponding to the semiconductor device, and use LAMMPS to perform defect evolution simulation calculations. The incident velocity magnitude of the PKA is given by the energy, and random numbers are used to determine the angles between the PKA velocity and the x, y, and z directions; the NVE (constant number of system particles, volume, and energy) ensemble is selected for the simulation, and the two-temperature (TTM) model is used to control the temperature of the simulation system; the MD defect evolution calculation is divided into a primary evolution stage, a secondary evolution stage, and an ultimate evolution stage. A variable time step is selected to reflect the defect evolution process, and the primary evolution stage is set to 0.01fs, the secondary evolution stage is set to 0.1fs, and the ultimate evolution stage is set to 1fs. The variable step size can capture the peak and stable values of the number of defects in the structure while saving calculation time.

[0052] Preferably, after inputting the PKA information in the box as a parameter, first perform system energy minimization and relaxation, and then perform MD defect evolution calculation. The total evolution time of the three stages of MD defect evolution reaches the stable state of the structure under the MD time scale in about 20 ps. During the MD calculation process, parameters such as the coordinates, kinetic energy, and displacement of each atom in the evolution region need to be set to facilitate observing the evolution process and statistical results.

[0053] In some embodiments, the method for simulating defect evolution of a semiconductor device by using the molecular dynamics method and obtaining the defect healing rate of the semiconductor device includes:

[0054] Obtain the peak state structure when the number of defects generated is the largest and the steady state structure when the number of defects generated is stable during the defect evolution simulation process, and respectively obtain the defect information in the peak state structure and the steady state structure;

[0055] According to the defect information in the peak state structure and the defect information in the steady state structure, obtain the defect healing rate.

[0056] Specifically, first obtain the defect healing rate c according to the following formula, and the obtained healing rate is a percentage less than 1;

[0057]

[0058] where A is the number of defects in the peak state structure and B is the number of defects in the steady state structure.

[0059] Then use the SRIM software to calculate the initial defect concentration of the semiconductor device, where the initial defect concentration refers to the defect concentration in the semiconductor device without considering defect healing.

[0060] Finally, obtain the steady state defect concentration according to the following formula:

[0061] n S =n 0 (1 - c);

[0062] where n 0 is the initial defect concentration, n S is the steady state defect concentration, and c is the defect healing rate.

[0063] In some embodiments, in step S3, the steady-state defect concentration is input into TCAD for device electrical performance calculation, and the relationship between the performance parameters of the semiconductor device and the defect healing rate is obtained, and the defect healing situation of the semiconductor device after irradiation is characterized by electrical performance degradation. Among them, TCAD is the abbreviation of Technology Computer Aided Design, which refers to semiconductor process simulation and device simulation tools. It is a numerical simulation tool based on semiconductor physics. It can simulate different process conditions to replace or partially replace expensive and time-consuming process experiments; it can also optimize different device structures to obtain ideal characteristics; it can also simulate circuit performance and electrical defects, etc.

[0064] Defect evolution simulations are respectively carried out on semiconductor Si materials and semiconductor GaN-based materials. After these two semiconductor device materials are irradiated, the change trend of the number of defects with time evolution is as Figure 2 shown. It can be seen that within the initial time of irradiation, the number of defects in the device surges rapidly and reaches a peak value, that is, the number of defects A in the peak state structure, simply referred to as the defect peak; as the subsequent evolution process continues, the defects gradually heal, and at this time the number of defects in the device begins to decrease. During the healing process, the number of defects will decrease to a relatively stable value. At this time, the remaining defects in the device are the number of defects B in the steady-state structure, simply referred to as the steady-state defects. After subtracting the number of defect peaks from the number of steady-state defects and then dividing by the number of defect peaks, the defect healing rate is obtained.

[0065] Fluorine irradiation tests are carried out on AlGaN / GaN-HEMTs devices under different fluence conditions. Among them, the structural schematic diagram of the AlGaN / GaN-HEMTs device is as Figure 3 shown. The test data are statistically analyzed and made into a relationship diagram of the gate voltage and the drain current as Figure 4 shown. Figure 4 In Figure 4 , the abscissa "gate voltage" represents the gate voltage, and the ordinate "drain current" represents the drain current. 11 ion·cm -2 In 12 ion·cm -2 In 12 ion·cm -2 In 12 ion·cm -2 In 12 ion·cm -2 respectively represent the voltage-current conditions of the device when it is not irradiated (unirradiated) and when the particle fluence is 7×10 Figure 4The arrow in [Figure] indicates that as the incident particle fluence increases, for the same leakage current, the gate voltage gradually increases.

[0066] The Figure 4 relationship diagram between the gate voltage and leakage current at different particle fluences obtained from the experimental data in [Figure] is transformed into a relationship diagram between the particle fluence and the threshold voltage shift. The results are as Figure 5 shown in [Figure], where the curve represented by the legend 30MeV fluorine irradiation is shown. Figure 5 In [Figure], the abscissa ion fluence represents the ion fluence, with the unit of (cm -2 ), and the ordinate shift of threshold voltage represents the shift of the threshold voltage, with the unit of volts (V).

[0067] Using the simulation method for the evolution of irradiation defects in the semiconductor device of the present invention, first use SRIM to calculate the defect concentration at different fluences, and then subtract the healed defect number (i.e., the initial defect concentration multiplied by the healing rate) from the defect concentration to obtain the steady-state defect concentration. Inputting the steady-state defect concentration into TCAD can obtain the corresponding threshold voltage shift, thereby obtaining the relationship between different particle fluences and the threshold voltage shift, as shown by the three curves in Figure 5 [Figure]. The data markers are respectively hollow squares, hollow circles, and hollow triangles, and the legends are respectively "defects only in all bulk", "defects only in GaN", and "defects only in AlGaN". These three curves almost completely overlap, so it seems to be one curve shown. These three curves respectively represent the relationship between the ion fluence and the threshold voltage shift obtained by simulating the evolution of defects at different positions. Among them, Figure 5 in [Figure], "defects only in all bulk" means that the simulated defect distribution is in the AlGaN and GaN regions of the device as shown in Figure 3 [Figure], "defects only in GaN" means that the simulated defect distribution is in the GaN region of the device as shown in Figure 3 [Figure], and "defects only in AlGaN" means that the simulated defect distribution is in the AlGaN region of the device as shown in Figure 3 [Figure].

[0068] It can be seen from Figure 5 [Figure] that the results calculated by using the simulation method of this embodiment (data markers are hollow squares, hollow circles, and hollow triangles) are in good agreement with the data obtained through experiments (data marker is a solid square).

[0069] This embodiment realizes an intuitive and accurate characterization of the irradiation defect healing rate in semiconductor devices. During the entire simulation process, it closely adheres to the actual situation. The irradiation source information and the reduction degree of the device system are very close to the experimental situation. The calculation results are also in good agreement with the experimental data. Moreover, the simulation method has clear logic, simple steps and is easy to operate.

[0070] A simulation system for the evolution of irradiation defects in a semiconductor device according to another embodiment of the present invention includes:

[0071] A molecular dynamics evolution module: used to simulate the defect evolution of a semiconductor device by using the molecular dynamics method;

[0072] An acquisition module: used to acquire the defect healing rate, initial defect concentration and steady-state defect concentration of the semiconductor device, and also used to acquire the relationship between the performance parameters of the semiconductor device and the defect healing rate;

[0073] A performance simulation module: used to perform performance simulation of the semiconductor device with the steady-state defect concentration of the semiconductor device as the input parameter.

[0074] Although the present disclosure is disclosed as above, the protection scope of the present disclosure is not limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present disclosure, and these changes and modifications will all fall within the protection scope of this embodiment.

Claims

1. A simulation method for the evolution of irradiation defects in a semiconductor device, characterized in that, it includes: Step S1, using the molecular dynamics method to simulate the defect evolution of the semiconductor device, and obtaining the defect healing rate of the semiconductor device; Obtaining the peak state structure when the number of generated defects is the largest and the steady state structure when the number of generated defects is stable during the defect evolution simulation process, and respectively obtaining the defect information in the peak state structure and the steady state structure; According to the defect information in the peak state structure and the defect information in the steady state structure, obtaining the defect healing rate; Step S2, obtaining the initial defect concentration of the semiconductor device, and obtaining the steady state defect concentration of the semiconductor device according to the initial defect concentration and the defect healing rate; wherein, the initial defect concentration refers to the defect concentration in the semiconductor device without considering defect healing; Step S3, using the steady state defect concentration of the semiconductor device as an input parameter, performing performance simulation of the semiconductor device, and obtaining the relationship between the performance parameters of the semiconductor device and the defect healing rate.

2. The simulation method for the evolution of irradiation defects in a semiconductor device according to claim 1, characterized in that, The obtaining the defect healing rate according to the defect information in the peak state structure and the defect information in the steady state structure includes: Obtaining the defect healing rate c according to the following formula: ； where A is the number of defects in the peak state structure and B is the number of defects in the steady state structure.

3. The simulation method for the evolution of irradiation defects in a semiconductor device according to claim 2, characterized in that, The obtaining the initial defect concentration of the semiconductor device includes: Using SRIM software to calculate the initial defect concentration of the semiconductor device.

4. The simulation method for the evolution of irradiation defects in a semiconductor device according to claim 3, characterized in that, The obtaining the steady state defect concentration of the semiconductor device according to the initial defect concentration and the defect healing rate includes: Obtaining the steady state defect concentration according to the following formula: ； where n 0 is the initial defect concentration, n S is the steady-state defect concentration, and c is the defect healing rate.

5. The simulation method for the evolution of irradiation defects in a semiconductor device according to any one of claims 1-4, characterized in that, The using the molecular dynamics method to simulate the defect evolution of the semiconductor device includes a primary evolution stage, a secondary evolution stage and an ultimate evolution stage. The time step of the primary evolution stage is 0.01 fs, the time step of the secondary evolution stage is 0.1 fs, the time step of the ultimate evolution stage is 1 fs, and the total evolution time of the primary evolution stage, the secondary evolution stage and the ultimate evolution stage is 20 ps.

6. The simulation method for the evolution of irradiation defects in a semiconductor device according to claim 5, characterized in that, The using the molecular dynamics method to simulate the defect evolution of the semiconductor device further includes: setting the coordinates, kinetic energy and displacement of each atom in the output evolution region during the defect evolution simulation process.

7. The simulation method for the evolution of irradiation defects in a semiconductor device according to claim 5, characterized in that, Before the defect evolution simulation of the semiconductor device using the molecular dynamics method, it further includes: performing meshing on the semiconductor device and establishing a system model with the same size as the mesh.

8. The simulation method for irradiated defect evolution in the semiconductor device according to claim 5, wherein, the simulation ensemble of the molecular dynamics method includes the NVE ensemble, and a two-temperature model is used to control the temperature of the simulation system of the molecular dynamics method.

9. A simulation system for irradiated defect evolution in a semiconductor device, wherein, it includes: Molecular dynamics evolution module: used to simulate the defect evolution of the semiconductor device using the molecular dynamics method; Acquisition module: used to acquire the defect healing rate, initial defect concentration, and steady-state defect concentration of the semiconductor device, and is also used to acquire the relationship between the performance parameters of the semiconductor device and the defect healing rate; Specifically used to acquire the peak state structure when the number of generated defects is the largest and the steady-state structure when the number of generated defects is stable during the defect evolution simulation process, and respectively acquire the defect information in the peak state structure and the steady-state structure; According to the defect information in the peak state structure and the defect information in the steady-state structure, obtain the defect healing rate; wherein, the initial defect concentration refers to the defect concentration in the semiconductor device without considering defect healing; Performance simulation module: used to perform performance simulation of the semiconductor device with the steady-state defect concentration of the semiconductor device as the input parameter.

Images