Simulation method of predicting a breakdown voltage of a semiconductor device

Abstract

The present invention relates to a method for simulating a semiconductor device. The simulation method comprises: (a) a step of simulating the semiconductor device without applying an impact ionization model; (b) a step of setting target values ​​and initial values ​​for impact ionization parameters of the impact ionization model; (c) a step of performing device simulation by applying the impact ionization model while gradually varying the impact ionization parameters in a direction that increases impact ionization at various simulation voltages; and (d) a step of detecting a simulation voltage at which the result of the device simulation converges at the target value of the impact ionization parameters, and setting the detected simulation voltage as the predicted breakdown voltage. By utilizing the change in breakdown characteristics according to the magnitude of impact ionization, the present invention enables not only efficient prediction of the breakdown voltage of the semiconductor device but also rapid and robust simulation performance.

Description

Technology Field

[0001] The present invention relates to a method for simulating a semiconductor device, and more specifically, to a method for simulating a semiconductor device configured to predict the breakdown voltage of a power semiconductor device more robustly and efficiently. Background Technology

[0002] Semiconductor device simulation is a technology that uses computer programs to predict the performance of semiconductor devices, and it is essential for reducing time and costs during the development process. A semiconductor device simulator, which performs these simulations, is a numerical analysis program that calculates and outputs the electrical characteristics of a semiconductor device by numerically solving governing equations representing the movement of electrons and holes within the device. Typical examples of the output electrical characteristics include IV characteristic relationships. Since the governing equations considered in semiconductor device simulation are nonlinear, a complete solution cannot be obtained in a single step. Consequently, semiconductor device simulators first assume an approximate initial solution and iteratively improve the solution by reducing the error caused by this initial solution to obtain the complete solution.

[0003] Since the simulation process by a semiconductor device simulator involves solving nonlinear equations, it is necessary to know a good approximation that is close to the correct answer. However, because it is difficult to know a good approximation for the voltage conditions the user wants to know in advance, it is inevitable to find a solution iteratively by starting from an initial voltage and gradually increasing the voltage. Consequently, most of the execution time of semiconductor device simulation is consumed in calculating these intermediate answers; although techniques such as parallel computing are incorporated to address this, it ultimately requires massive computing resources.

[0004] Meanwhile, power semiconductor devices are components used to process power and are essential for humanity's efficient energy use. Furthermore, the breakdown voltage of a semiconductor device is the voltage at which the reverse current suddenly increases when a reverse electromotive force is applied to a PN junction. This breakdown voltage represents the maximum voltage at which a power semiconductor device can withstand current flow, making it one of the most critical performance indicators. Therefore, it is necessary to predict the breakdown voltage during the actual device design phase.

[0005] Conventional methods for predicting breakdown voltage observe the increase in current while gradually increasing the applied voltage at a specific terminal and determine the voltage at which a rapid increase in current occurs as the breakdown voltage. However, when the mechanism causing the breakdown phenomenon is considered in computer simulations, the simulation may become unstable or stop midway even if the breakdown phenomenon does not occur. Generally, simulation is a process of finding the correct answer by iteratively refining the solution using the Newton-Raphson method. In this simulation process, if the correct answer is not found within a predetermined number of iterations, convergence fails, and in this case, the simulation stops midway.

[0006] Figure 1 is a graph showing the results of device simulation performed according to conventional technology. Referring to Figure 1, a breakdown voltage is observed at approximately 1400 Volts. However, it can be seen that computer simulations for voltage regions lower than the breakdown voltage take a very long time. If the simulation of the semiconductor device is stopped before reaching the breakdown voltage, information regarding the breakdown voltage cannot be obtained, even though a significant amount of time has already been spent on the simulation. Prior art literature

[0007] Korean Registered Patent Publication No. 10-0716912 Korean Registered Patent Publication No. 10-2572752 The problem to be solved

[0008] To solve the aforementioned problems, the present invention aims to provide a simulation method for a semiconductor device configured to predict the breakdown voltage of the semiconductor device more robustly and efficiently by utilizing the change in breakdown characteristics according to the magnitude of impact ionization. means of solving the problem

[0009] A method for simulating a semiconductor device according to the first feature of the present invention for achieving the aforementioned technical problem relates to a method for simulating a semiconductor device using a semiconductor simulation program configured to input information on a semiconductor device to be simulated and physical model information for predicting the breakdown voltage of a semiconductor device, comprising: (a) a step of setting a maximum value of a predictable range of the breakdown voltage of the semiconductor device; (b) a step of simulating the device without applying a shock ionization model to the semiconductor device up to the maximum value of the predictable range of the breakdown voltage of the semiconductor device; (c) a step of setting a target value and an initial value for a shock ionization parameter that determines the magnitude of shock ionization in the shock ionization model; (d) a step of setting an initial value of the simulation voltage to the maximum value of the predictable range of the breakdown voltage and setting the shock ionization parameter to the set initial value; (e) a step of performing device simulation by applying a shock ionization model while gradually varying the shock ionization parameter in a direction that increases shock ionization at various simulation voltages. (f) a step of detecting a simulation voltage at which the result of the device simulation converges at the target value of the impact ionization parameter, and setting the detected simulation voltage as the predicted breakdown voltage; wherein the step (e) is performed repeatedly while gradually decreasing the simulation voltage from the initial value of the predicted breakdown voltage until the result of the device simulation converges at the target value of the impact ionization parameter, thereby enabling efficient prediction of the breakdown voltage of the semiconductor device.

[0010] In the simulation method of a semiconductor device according to the first feature described above, the maximum value of the predictable range of the breakdown voltage of the semiconductor device is preferably a voltage arbitrarily set by a user or a simulation program, and is the maximum value of the predictable range of the breakdown voltage for the semiconductor device.

[0011] In the simulation method of a semiconductor device according to the first feature described above, it is preferable that the semiconductor device is a power semiconductor device.

[0012] A method for simulating a semiconductor device according to the second feature of the present invention relates to a method for simulating a semiconductor device using a semiconductor simulation program configured to input information about the semiconductor device to be simulated and physical model information for predicting the breakdown voltage of the semiconductor device, comprising: (a) a step of setting a maximum value of the predictable range of the breakdown voltage of the semiconductor device; (b) a step of simulating the device without applying a shock ionization model to the semiconductor device up to the maximum value of the predictable range of the breakdown voltage of the semiconductor device; (c) a step of setting a target value and an initial value for a shock ionization parameter that determines the magnitude of shock ionization in the shock ionization model; (d) a step of setting an initial value of the simulation voltage to the maximum value of the predictable range of the breakdown voltage and setting the shock ionization parameter to the initial value set above; (e) a step of performing device simulation by applying a shock ionization model while gradually varying the shock ionization parameter in a direction that increases shock ionization at the simulation voltage; (f) a step of reducing the simulation voltage and repeating step (e) if the result of the simulation fails to converge in step (e); (g) a step of setting the simulation voltage as the predicted breakdown voltage of the semiconductor device when the result of the simulation converges in step (e) above and the shock ionization parameter is the set target value; thereby enabling efficient prediction of the breakdown voltage of the semiconductor device. Effects of the invention

[0013] The semiconductor device simulation method according to the present invention, having the aforementioned configuration, performs a simulation without shock ionization up to the upper limit of the breakdown voltage as a first step. Accordingly, not only can the simulation be prevented from being interrupted during the intermediate process due to shock ionization, but the voltage region of the intermediate process, which takes a long time, can also be passed quickly. As a result, the semiconductor device simulation method according to the present invention can rapidly perform a simulation up to the upper limit of the breakdown voltage.

[0014] In addition, the semiconductor device simulation method according to the present invention attempts to simulate again after lowering the simulation voltage if convergence failure occurs in the second step, and this step is repeated until the impact ionization parameter reaches a target value. Therefore, even if convergence failure of the simulation occurs during the intermediate process, it does not have a negative effect on the calculation of the breakdown voltage. Rather, the semiconductor device simulation method according to the present invention possesses algorithmic robustness by enabling a closer approach to the breakdown voltage when convergence failure of the simulation occurs during the intermediate process. Brief explanation of the drawing

[0015] Figure 1 is a graph showing the results of performing device simulation according to conventional technology. FIG. 2 is a flowchart illustrating a method for simulating a semiconductor device according to a preferred embodiment of the present invention. FIG. 3 is a graph exemplarily illustrating breakdown voltages according to the magnitude of impact ionization parameters in a simulation method of a semiconductor device according to the present invention. FIG. 4 is a graph showing the trajectory according to a simulation performed while varying the impact ionization parameter and the simulation voltage in a simulation method for a semiconductor device according to the present invention. Specific details for implementing the invention

[0016] The simulation method for a semiconductor device according to the present invention is characterized by utilizing the change in breakdown characteristics according to the magnitude of impact ionization to perform the simulation. In particular, the simulation method according to the present invention performs device simulation while stepwise increasing or decreasing the value of the impact ionization parameter at each simulation voltage, and sets the corresponding simulation voltage as the predicted breakdown voltage when the simulation result converges to a preset target value of the impact ionization parameter.

[0017] Hereinafter, a simulation method for a semiconductor device capable of efficiently predicting breakdown voltage according to a preferred embodiment of the present invention will be described in detail with reference to the attached drawings.

[0018] FIG. 2 is a flowchart illustrating a method for simulating a semiconductor device according to a preferred embodiment of the present invention.

[0019] Referring to FIG. 2, a simulation method for a semiconductor device capable of efficiently predicting breakdown voltage according to the present invention inputs information on a semiconductor device to be simulated and physical model information for predicting the breakdown voltage of a semiconductor device using a semiconductor simulation program, and starts a simulation of the semiconductor device.

[0020] First, the maximum value of the predictable range of the breakdown voltage of the semiconductor device is set (step 100). The maximum value of the predictable range of the breakdown voltage of the semiconductor device can be arbitrarily set by a user or a simulation program.

[0021] Next, device simulation is performed on the semiconductor device without applying the shock ionization model up to the maximum value of the predictable range of the breakdown voltage set above (step 110). In this case, by simulating the semiconductor device without applying the shock ionization model, device simulation can be performed without numerical instability caused by the shock ionization model, and as a result, the possibility of failure of convergence in the simulation can be minimized, and the simulation time can be shortened.

[0022] FIG. 3 is a graph exemplarily illustrating the breakdown voltage according to the magnitude of the impact ionization parameter in a simulation method for a semiconductor device according to the present invention. Referring to FIG. 3, when the Bologna model is used as the impact ionization model, it can be seen that as the magnitude of hd0, which is the impact ionization parameter determining the magnitude of impact ionization, increases, the impact ionization decreases and the breakdown voltage increases. In this specification, the parameter determining the magnitude of impact ionization is defined as the impact ionization parameter. Furthermore, the impact ionization parameter may be set differently depending on the type of impact ionization model.

[0023] Next, target and initial values ​​are set for impact ionization parameters that determine the magnitude of impact ionization in the impact ionization model (step 120). Here, the initial value of the impact ionization parameter is a value that prevents impact ionization from occurring, and the target value of the impact ionization parameter is a value corresponding to the magnitude of impact ionization that is ultimately to be applied to the device simulation. As shown in FIG. 3, the initial value of the impact ionization parameter hd0 can be set to, for example, about 4e6 V / cm, and the target value can be set to, for example, about 1.2e6 V / cm, indicated as the Default parameter. However, the initial and target values ​​of the impact ionization parameter may be set differently depending on the characteristics of the semiconductor device.

[0024] Next, to simulate the breakdown voltage, the initial value of the simulation voltage is set to the maximum value of the predictable range of the breakdown voltage, and the shock ionization parameter is set to the initial value set above (step 130).

[0025] FIG. 4 is a graph showing the trajectory of a simulation performed while varying the impact ionization parameter and the simulation voltage in a simulation method for a semiconductor device according to the present invention. Referring to FIG. 4, it can be seen that the simulation voltage is set to an initial value of 20V and the simulation is started.

[0026] Next, as illustrated in FIG. 4, at the simulation voltage, the device simulation is repeatedly performed by applying the impact ionization model while gradually varying the impact ionization parameter in the direction of increasing impact ionization (steps 140 and 142). The aforementioned steps involve repeatedly performing the device simulation until the result of the device simulation fails to converge or the impact ionization parameter of the impact ionization model reaches the set target value.

[0027] If the simulation result fails to converge in the aforementioned step, the simulation voltage is reduced, and the aforementioned step is repeated (step 150).

[0028] If, in the aforementioned step, the result of the simulation converges and the impact ionization parameter is the set target value, the simulation voltage is set as the predicted breakdown voltage of the semiconductor device (step 160).

[0029] With the above-described configuration, the simulation method for a semiconductor device according to the present invention enables efficient prediction of the breakdown voltage of the semiconductor device.

[0030] Meanwhile, the simulation method of a semiconductor device according to the present invention can be applied particularly to power semiconductor devices.

[0031] Although the present invention has been described above with reference to preferred embodiments, this is merely illustrative and does not limit the invention. Those skilled in the art will understand that various modifications and applications not exemplified above are possible within the scope of the essential characteristics of the invention. Furthermore, differences related to such modifications and applications should be interpreted as being included within the scope of the invention as defined in the appended claims.

Claims

Claim 1 A method for simulating a semiconductor device, wherein a semiconductor simulation program configured to input information on a semiconductor device to be simulated and physical model information for predicting the breakdown voltage of a semiconductor device is executed by one or more processors of a computer, comprising: (a) a step in which the processor sets a maximum value of a predictable range of the breakdown voltage of the semiconductor device; (b) a step in which the processor simulates the semiconductor device without applying a shock ionization model up to the maximum value of the predictable range of the breakdown voltage of the semiconductor device; (c) a step in which the processor sets a target value and an initial value for a shock ionization parameter that determines the magnitude of shock ionization in the shock ionization model; (d) a step in which the processor sets an initial value of the simulation voltage to the maximum value of the predictable range of the breakdown voltage and sets the shock ionization parameter to the set initial value; (e) a step in which the processor performs device simulation by applying a shock ionization model while gradually varying the shock ionization parameter in a direction that increases shock ionization at various simulation voltages. (f) a step in which the processor detects a simulation voltage at which the result of the device simulation converges at the target value of the impact ionization parameter and sets the detected simulation voltage as the predicted breakdown voltage; wherein the step (e) is repeatedly performed while gradually decreasing the simulation voltage from the initial value of the predicted breakdown voltage until the result of the device simulation converges at the target value of the impact ionization parameter, thereby enabling efficient prediction of the breakdown voltage of the semiconductor device. Claim 2 A method for simulating a semiconductor device capable of efficiently predicting a breakdown voltage, wherein, in claim 1, the maximum value of the predictable range of the breakdown voltage of the semiconductor device is a voltage that is set by the processor receiving input from a user or arbitrarily set by a simulation program, and is the maximum value of the predictable range of the breakdown voltage for the semiconductor device. Claim 3 A method for simulating a semiconductor device capable of efficiently predicting breakdown voltage, characterized in that, in claim 1, the semiconductor device is a power semiconductor device. Claim 4 A method for simulating a semiconductor device, wherein a semiconductor simulation program configured to input information on a semiconductor device to be simulated and physical model information for predicting the breakdown voltage of a semiconductor device is executed by one or more processors of a computer, comprising: (a) a step in which the processor sets a maximum value of a predictable range of the breakdown voltage of the semiconductor device; (b) a step in which the processor simulates the semiconductor device without applying an impact ionization model up to the maximum value of the predictable range of the breakdown voltage of the semiconductor device; (c) a step in which the processor sets a target value and an initial value for an impact ionization parameter that determines the magnitude of impact ionization in the impact ionization model; (d) a step in which the processor sets an initial value of the simulation voltage to the maximum value of the predictable range of the breakdown voltage and sets the impact ionization parameter to the set initial value; (e) a step in which the processor performs the device simulation by applying an impact ionization model while gradually varying the impact ionization parameter in a direction that increases impact ionization at the simulation voltage. (f) a step in which, if the result of the simulation in step (e) fails to converge, the processor reduces the simulation voltage and repeats step (e); (g) a step in which, if the result of the simulation in step (e) converges and the shock ionization parameter is the set target value, the processor sets the simulation voltage to the predicted breakdown voltage of the semiconductor device; wherein the method of simulating a semiconductor device is configured to efficiently predict the breakdown voltage of the semiconductor device. Claim 5 A method for simulating a semiconductor device according to claim 4, wherein the semiconductor device is a power semiconductor device.

Images