Defect evolution simulation method and system for semiconductor devices with different doping systems

Abstract

The present invention provides a method and system for simulating the defect evolution of devices with different doping systems, which belongs to the field of simulation technology. The method includes: establishing a pure silicon system model; introducing doping elements into the pure silicon system model to obtain a doping system model; based on the doping system model, using the information of primary ejected particles generated after the incident particles irradiate the device as input conditions, using the molecular dynamics method to simulate the defect evolution process of the device and output the evolution structure; statistically analyzing the defect information in the evolution structure, and obtaining the relationship between the defect information and the doping information and the information of the primary ejected particles in the doping system model. The present invention realizes the simulation of the defect evolution of semiconductor devices with different doping systems through the molecular dynamics method. The simulation process fits the actual situation as closely as possible. The simulation results can be compared with the experimental data. The calculation method has clear logic, simple steps and is easy to operate.

Description

Technical Field

[0001] The present invention relates to the field of simulation technology, and in particular to a method and system for simulating defect evolution of devices with different doping systems. Background Art

[0002] During space irradiation, damage to semiconductor electronic devices caused by high-energy, high-mass charged particles is primarily caused by displacement damage. To fully understand the generation and formation mechanisms of displacement defects, it is necessary to study the entire process of defect evolution, from the cascade of primary ejected particles to high-temperature annealing and ultimately stabilization. Specifically, the displacement effect produced by radiation in a material is typically the direct generation and cascade of primary ejected particles. The high energy of primary ejected particles in the early stages of transport results in a large collision mean free path. As the cascade progresses, the terminal primary ejected particles can collide with a large number of lattice atoms within a short distance until their energy is insufficient to displace lattice atoms. Furthermore, diffusion and annealing caused by thermal effects also influence the defect structure. Therefore, the entire cascade of primary ejected particles progresses from initial intense collisions to subsequent high-temperature annealing and ultimately to equilibrium.

[0003] At present, the simulation methods used in simulating and calculating radiation defects in semiconductor devices are relatively crude. For example, the model used in the simulation calculation is usually a single element system similar to single-crystal silicon, which is quite different from the actual situation of semiconductor devices and cannot accurately simulate the evolution process of radiation defects in semiconductor devices. Summary of the Invention

[0004] The problem solved by the present invention is that the model currently used for simulating the radiation defects of devices is quite different from the actual situation and cannot accurately simulate the evolution process of the radiation defects.

[0005] To solve the above problems, the present invention provides a method for simulating defect evolution in devices with different doping systems, comprising:

[0006] Step S1, establishing a pure silicon system model;

[0007] Step S2, introducing doping elements into the pure silicon system model to obtain a doping system model;

[0008] Step S3, based on the doping system model, using information of primary ejected particles generated after the incident particles irradiate the device as input conditions, using molecular dynamics methods to simulate the defect evolution process of the device, and outputting the evolved structure;

[0009] Step S4: Count the defect information in the evolved structure and obtain the relationship between the defect information, the doping information in the doping system model, and the information of the primary ejected particles.

[0010] Preferably, the doping element includes at least one of C, O, B and P elements, wherein C and O elements are doped into the pure silicon system model as interstitial atoms, and B and P elements are doped into the pure silicon system model as substitutional atoms.

[0011] Preferably, the doping concentrations of the C and O elements are both 10 12 / cm 3 —10 14 / cm 3 , and / or, the doping concentrations of the B and P elements are both 10 18 / cm 3 —10 20 / cm 3 .

[0012] Preferably, the information of the primary ejected particles includes the energy of the primary ejected particles, and the defect information includes the number of the defects.

[0013] Preferably, the defect evolution process of the device simulated by the molecular dynamics method includes: a primary evolution stage, a secondary evolution stage and an ultimate evolution stage, wherein the time step of the primary evolution stage is 0.01fs, the time step of the secondary evolution stage is 0.1fs, the time step of the ultimate evolution stage is 1fs, and the evolution time of the primary evolution stage, the secondary evolution stage and the ultimate evolution stage is 20ps in total.

[0014] Preferably, the simulating the defect evolution process of the device by using a molecular dynamics method includes: performing defect evolution simulation calculations using LAMMPS.

[0015] Preferably, the simulation ensemble of the molecular dynamics method includes an NVE ensemble.

[0016] Preferably, the simulation force field of the molecular dynamics method includes one of SW potential, MORSE potential, Tersoff potential and LJ potential.

[0017] Preferably, before simulating the defect evolution process of the device by using a molecular dynamics method, the method further includes: using a dual-temperature relaxation model to perform temperature control on the doping system model to minimize the energy of the doping system model.

[0018] The advantages of the present invention over the prior art are:

[0019] During space irradiation, incident particles irradiate the device material and generate primary ejected particles. The primary ejected particles interact with the lattice atoms in the material to produce a displacement cascade, thereby causing irradiation damage to the material. The doping elements present in the device will also have an important impact on defect evolution. However, the existing technology often uses a single element system model similar to single crystal silicon when simulating defect evolution, ignoring the influence of doping elements in the device on defect evolution. Therefore, in order to investigate the influence of different doping systems on the defect evolution of the device, the present invention provides a simulation method that can reflect the influence of doping element effects. Specifically, the simulation system is doped according to the actual situation, so that the simulation conditions are closer to the experimental conditions, and the calculation results are more consistent with the experimental data. The molecular dynamics method is used to simulate the defect evolution of devices with different doping systems. The relationship between defect information, doping information and primary ejected particle information is used to reflect the influence of doping elements in the device on defect evolution.

[0020] When using the molecular dynamics method to simulate device defect evolution, the present invention takes into account the influence of doping elements in the device, realizes the simulation of the defect evolution of semiconductor devices with different doping systems, and the simulation process is extremely consistent with the actual situation. The simulation results can be compared with experimental data, and the calculation method has clear logic, simple steps and is easy to operate.

[0021] The present invention also provides a defect evolution simulation system for devices with different doping systems, comprising:

[0022] Modeling module, used to build pure silicon system model;

[0023] a doping module, configured to introduce doping elements into the pure silicon system model to obtain a doping system model;

[0024] Molecular dynamics module: used to perform molecular dynamics simulation on the doping system model and output the evolved structure;

[0025] An acquisition module is used to collect statistics on defect information of the evolved structure and obtain the relationship between the defect information and the doping information of the doping system model and the information of primary ejected particles after the incident particles irradiate the device.

[0026] The advantages of the defect evolution simulation system for devices with different doping systems of the present invention over the prior art are the same as those of the defect evolution simulation method for devices with different doping systems, which will not be described in detail here. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 Flowchart of a defect evolution simulation method for devices with different doping systems according to an embodiment of the present invention;

[0028] Figure 2Schematic diagram of the system after element doping into the pure silicon system model in an embodiment of the present invention;

[0029] Figure 3 Graph showing defect evolution calculation results when only B doping or P doping is considered in Examples 1 and 3 of the present invention;

[0030] Figure 4 Graph showing defect evolution calculation results when C, O, B, and P doping are considered in Examples 3-5 of the present invention;

[0031] Figure 5 Graph showing defect evolution calculation results during PN junction doping in Examples 6 and 7 of the present invention. DETAILED DESCRIPTION

[0032] During space irradiation, incident particles irradiate device materials and generate primary knockout particles (PKAs). PKAs interact with lattice atoms in the material to produce a displacement cascade, thereby causing irradiation damage to the material. PKA cascade damage is at a microscopic scale. Therefore, in order to accurately describe the microscopic dynamic evolution of the PKA cascade, including the cascade transport of PKA, the balance of system energy, and the relaxation process, a molecular dynamics method including defect identification can be used. However, in the prior art, when simulating defect evolution, a single-element system model similar to single-crystal silicon is often used, ignoring the effect of doping elements in the device on defect evolution. The doping elements present in the device actually have a significant impact on defect evolution, and due to different doping systems, the force fields used to reflect the interactions between multiple elements during molecular dynamics simulations are also different. However, there is currently a lack of methods that can accurately introduce doping elements in modeling, nor is there a force field that can describe the interactions between elements when multiple elements coexist. Therefore, in order to investigate the influence of different doping systems on the defect evolution of devices, the present invention provides a simulation method that can reflect the influence of doping elements on the defect evolution of devices, especially semiconductor devices, and attempts to find an accurate force field that can describe the interaction between each element when multiple doping elements exist, so as to reflect the influence of doping elements in the device on the defect evolution.

[0033] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0034] See also Figure 1 As shown, a defect evolution simulation method for devices with different doping systems according to an embodiment of the present invention includes:

[0035] Step S1, establishing a pure silicon system model;

[0036] Step S2, introducing doping elements into the pure silicon system model to obtain a doping system model;

[0037] Step S3, based on the doping system model, using the information of primary ejected particles generated after the incident particles irradiate the device as input conditions, using the molecular dynamics method (MD) to simulate the defect evolution process of the device and output the evolved structure;

[0038] Step S4: Count the defect information in the evolved structure and obtain the relationship between the defect information and the doping information and the information of the primary ejected particles in the doping system model.

[0039] This embodiment uses molecular dynamics to simulate device defect evolution, taking into account the influence of doping elements in the device. The simulated system is doped according to real conditions, making the simulation conditions closer to the experimental conditions and the calculated results more consistent with the experimental data. Molecular dynamics is used to simulate the evolution of device defects in different doping systems. The relationship between defect information, doping information, and primary ejected particle information is used to demonstrate the influence of doping elements in the device on defect evolution. The calculation method is logically clear, the steps are simple, and it is easy to operate.

[0040] In some embodiments, before modeling in step S1, the method further includes: obtaining information on PKA generated after the incident particle irradiates the device, and meshing the entire device based on the PKA information. For example, the grid (box) size is set to 70nm*70nm*70nm, and then the information on PKA contained in each box after the incident particle is incident is counted, including the amount, energy, position and incident direction of the PKA.

[0041] In some embodiments, establishing a pure silicon system model in step S1 includes: establishing a pure silicon system model in LAMMPS, and expanding the Si unit cell in the x, y, and z directions by appropriate multiples to establish a Si structure of the same size as the box for subsequent simulation. For example, to obtain a 70nm*70nm*70nm pure Si model, the Si unit cell is expanded 129 times in the x, y, and z directions. LAMMPS is a large-scale atomic and molecular parallel simulator primarily used for molecular dynamics-related calculations and simulations.

[0042] In some of the embodiments, the device includes various types of semiconductor devices, and the doping elements include at least one of C, O, B and P elements. In order to make the doping closer to the actual situation, in step S2, the doping elements are introduced into the pure silicon system according to the actual doping method and doping rate (also referred to as doping concentration) of each element. In terms of doping method, C and O elements are doped into the pure silicon system in the form of interstitial atoms, and B and P elements are doped into the pure silicon system in the form of substitutional atoms. Both doping methods are easy to implement in LAMMPS. The doping rate is calculated in the following way. Specifically, let the doping rate of a certain element be D, and the unit is / cm 3, first convert the doping rate unit to / nm 3 , per nm in Si system 3 contains 50 Si atoms, and then divide the converted value by the number per nm 3 The number of Si atoms contained in can be used to obtain the doping rate n expressed as a percentage. The calculation formula for the doping rate is shown below:

[0043]

[0044] In this embodiment, the doping concentration of C and O elements is 10 12 / cm 3 —10 14 / cm 3 , the doping concentration of B and P elements is 10 18 / cm 3 —10 20 / cm 3 , the structure after doping is as follows Figure 2 shown.

[0045] In some embodiments, in step S3, defect evolution simulations are performed using LAMMPS. The NVE (number, volume, and energy constant) ensemble is selected as the simulation ensemble. Parameters such as the PKA information, ensemble, simulation step size, and evolution time are set. PKA velocities are assigned by varying energy values, and the direction of the velocity is controlled by setting the angle between the PKA velocity and the xyz axis using random numbers. During the molecular dynamics calculations, parameters such as the coordinates, kinetic energy, and displacement of each atom in the output evolution region are set to facilitate statistical analysis in step S4.

[0046] In a preferred embodiment, after inputting the PKA information, the doping system is first energy minimized and relaxed to ensure that the system is stable, for example, using a two-temperature relaxation model (TTM model) to control the system temperature, and then performing MD defect evolution calculations.

[0047] The MD defect evolution calculation includes the primary evolution stage, the secondary evolution stage and the ultimate evolution stage. The time step of the primary evolution stage is 0.01fs, the time step of the secondary evolution stage is 0.1fs, and the time step of the ultimate evolution stage is 1fs. The total evolution time of the three stages is about 20ps to reach the stable state of the structure under the MD time scale.

[0048] Because different elements are doped into the pure silicon system during modeling in this example, it is necessary to select an appropriate force field based on the types of elements ultimately contained in the different systems. In this example, a hybrid force field format is selected in LAMMPS, using different force fields coupled to characterize the interactions between different elements. For example, the SW potential can be used to describe Si-Si and Si-B interactions, the MORSE potential can be used to describe Si-P and PP interactions, the Tersoff potential can be used to describe Si-C, Si-O, CC, and OO interactions, and the LJ potential can be used to describe the remaining interactions.

[0049] In some implementation methods, in step S4, in the defect information of the statistically evolved structure, the defect information includes the type and quantity of defects, or the type and concentration of defects.

[0050] The present invention is described below with reference to specific embodiments.

[0051] Example 1

[0052] A defect evolution simulation method for devices with different doping systems, comprising:

[0053] Step S1: Establish a pure Si system model in LAMMPS, and expand the Si unit cell by 129 times in the x, y, and z directions to form a Si structure with a box size of 70nm*70nm*70nm.

[0054] Step S2: Introduce B element into the pure Si system, wherein B element is doped into Si in the form of substitutional atoms, with a doping concentration of 10 18 / cm 3 —10 20 / cm 3 .

[0055] Step S3: Set parameters such as PKA information, ensemble, simulation step size, and evolution time. Select the NVE ensemble as the simulation ensemble. Use the TTM model to control the system temperature. Use the SW potential to describe the Si-Si and Si-B interactions. Use LAMMPS to perform defect evolution simulation calculations.

[0056] Example 2

[0057] The difference from Example 1 is that:

[0058] In step S2, P element is introduced into the pure Si system, wherein P element is doped into Si in the form of substitutional atoms, with a doping concentration of 10 18 / cm 3 —10 20 / cm 3 .

[0059] In step S3, the MORSE potential is used to describe the interaction between Si and P and the interaction between PPs.

[0060] Example 3

[0061] The difference from Example 1 is that:

[0062] In step S2, C and O elements are introduced into the pure Si system, wherein C and O elements are doped into the pure silicon system in the form of interstitial atoms, and the doping concentrations of C and O elements are both 10 12 / cm 3 —10 14 / cm 3 .

[0063] In step S3, the Tersoff potential is used to describe the Si-C interaction, the Si-O interaction, the CC interaction, and the OO interaction.

[0064] Example 4

[0065] The difference from Example 1 is that:

[0066] In step S2, B, C and O elements are introduced into the pure Si system, wherein C and O elements are doped into the pure Si system in the form of interstitial atoms, and B element is doped into the pure Si system in the form of substitutional atoms. The doping concentrations of C and O elements are both 10 12 / cm 3 —10 14 / cm 3 , the doping concentration of B element is 10 18 / cm 3 —10 20 / cm 3 .

[0067] In step S3, the Tersoff potential is used to describe the Si-C interaction, the Si-O interaction, the CC interaction, and the OO interaction, and the SW potential is used to describe the Si-Si interaction and the Si-B interaction.

[0068] Example 5

[0069] The difference from Example 1 is that:

[0070] In step S2, P, C and O elements are introduced into the pure Si system, wherein C and O elements are doped into the pure Si system in the form of interstitial atoms, and P element is doped into the pure Si system in the form of substitutional atoms. The doping concentrations of C and O elements are both 10 12 / cm 3 —10 14 / cm 3 , the doping concentration of P element is 10 18 / cm 3 —10 20 / cm 3 .

[0071] In step S3, the Tersoff potential is used to describe the Si-C interaction, the Si-O interaction, the CC interaction, and the OO interaction, and the MORSE potential is used to describe the Si-P interaction and the PP interaction.

[0072] Example 6

[0073] The difference from Example 1 is that:

[0074] In step S2, in the pure silicon system, the upper half is doped with B element, and the lower half is doped with P element, wherein B and P elements are doped into the pure silicon system in the form of substitutional atoms, and the doping concentrations of B and P elements are both 10 18 / cm 3 —10 20 / cm 3 .

[0075] In step S3, the SW potential is used to describe the Si-Si interaction and the Si-B interaction, and the MORSE potential is used to describe the Si-P interaction and the PP interaction.

[0076] Example 7

[0077] The difference from Example 6 is that:

[0078] In step S2, in the pure silicon system, the upper half is doped with P element and the lower half is doped with B element.

[0079] Example 8

[0080] On the basis of Examples 1-7, the number of defects contained in the system after the molecular dynamics simulation calculation of Examples 1-7 was completed was counted, and the relationship between the defect data and the doping elements and PKA information was obtained. The results are as follows: Figure 3-5 As shown. Among them, Figure 3 The relationship between the number of defects in the device and the PKA energy is shown when the device is doped with B and P elements respectively. Figure 4 The relationship between the number of defects in the device and the PKA energy is shown when the device is doped with C and O elements, B, C and O elements, and P, C and O elements. Figure 5 The relationship between the number of defects and the PKA energy when B and P are doped at different locations in the device is shown. This allows us to understand the impact of the doping system on the defect evolution of semiconductor devices.

[0081] It should be noted that when calculating the relationship between defect information, doping information and PKA information, the defect information can also use other parameters, such as defect type and concentration, and the PKA information can also use other parameters, such as PKA quantity, position or incident direction.

[0082] The present invention uses molecular dynamics to simulate the defect evolution of semiconductor devices with different doping systems. The simulation system considers doping according to actual conditions, making the simulation conditions closer to the experimental conditions and the calculated results more consistent with the experimental data. This method has clear calculation logic, simple steps, and is easy to operate. The entire simulation process closely matches the actual situation, and the simulation results can be compared with experimental data. This method has obvious advantages and broad prospects in the design and application research of different semiconductor devices.

[0083] Another embodiment of the present invention provides a defect evolution simulation system for devices with different doping systems, comprising:

[0084] Modeling module, used to build pure silicon system model;

[0085] Doping module, used to introduce doping elements into the pure silicon system model to obtain a doping system model;

[0086] Molecular dynamics module: used to perform molecular dynamics simulation on the doping system model and output the evolved structure;

[0087] The acquisition module is used to statistically analyze the defect information of the evolving structure and obtain the relationship between the defect information and the doping information of the doping system model and the information of the primary ejected particles after the incident particles irradiate the device.

[0088] Although the present disclosure is disclosed as above, the protection scope of the present disclosure is not limited thereto. Those skilled in the art may make various changes and modifications without departing from the spirit and scope of the present disclosure, and these changes and modifications will fall within the protection scope of the present invention.

Claims

1. A defect evolution simulation method for devices with different doping systems, characterized in that: include: Step S1, establishing a pure silicon system model; Step S2, introducing doping elements into the pure silicon system model to obtain a doping system model; Step S3, based on the doping system model, using information of primary ejected particles generated after the incident particles irradiate the device as input conditions, using molecular dynamics methods to simulate the defect evolution process of the device, and outputting the evolved structure; Step S4: Count the defect information in the evolved structure and obtain the relationship between the defect information, the doping information in the doping system model, and the information of the primary ejected particles.

2. The defect evolution simulation method for devices with different doping systems according to claim 1, characterized in that: The doping elements include at least one of C, O, B and P elements, wherein C and O elements are doped into the pure silicon system model in the form of interstitial atoms, and B and P elements are doped into the pure silicon system model in the form of substitutional atoms.

3. The defect evolution simulation method for devices with different doping systems according to claim 2, characterized in that: The doping concentrations of C and O elements are both 10 12 / cm 3 —10 14 / cm 3 , and / or, the doping concentrations of the B and P elements are both 10 18 / cm 3 —10 20 / cm 3 .

4. The defect evolution simulation method for devices with different doping systems according to claim 1, characterized in that: The information of the primary ejected particles includes the energy of the primary ejected particles, and the defect information includes the number of the defects.

5. The defect evolution simulation method for devices with different doping systems according to claim 1, characterized in that: The defect evolution process of the device simulated using the molecular dynamics method includes: a primary evolution stage, a secondary evolution stage and a final evolution stage, wherein the time step of the primary evolution stage is 0.01fs, the time step of the secondary evolution stage is 0.1fs, the time step of the final evolution stage is 1fs, and the evolution time of the primary evolution stage, the secondary evolution stage and the final evolution stage is 20ps in total.

6. The defect evolution simulation method for devices with different doping systems according to claim 1, characterized in that: The simulating the defect evolution process of the device by using the molecular dynamics method includes: using LAMMPS to perform defect evolution simulation calculation.

7. The defect evolution simulation method for devices with different doping systems according to claim 5, characterized in that: The simulation ensemble of the molecular dynamics method includes the NVE ensemble.

8. The defect evolution simulation method for devices with different doping systems according to claim 1, characterized in that: The simulation force field of the molecular dynamics method includes one of SW potential, MORSE potential, Tersoff potential and LJ potential.

9. The defect evolution simulation method for devices with different doping systems according to claim 1, characterized in that: Before simulating the defect evolution process of the device by using the molecular dynamics method, the method further includes: using a dual-temperature relaxation model to perform temperature control on the doping system model to minimize the energy of the doping system model.

10. A defect evolution simulation system for devices with different doping systems, characterized in that: include: Modeling module, used to build pure silicon system model; a doping module, configured to introduce doping elements into the pure silicon system model to obtain a doping system model; Molecular dynamics module: used to perform molecular dynamics simulation on the doping system model and output the evolved structure; An acquisition module is used to collect statistics on defect information of the evolved structure and obtain the relationship between the defect information and the doping information of the doping system model and the information of primary ejected particles after the incident particles irradiate the device.

Images