A method for calibrating and optimizing CMOS doping parameters

Abstract

The application discloses a CMOS doping parameter calibration and optimization method, comprising the following steps: obtaining the working conditions, characteristic data, initial doping parameters and structure parameters of multiple CMOS devices with different sizes; selecting original first end characteristic values and original second end characteristic values from the characteristic data; determining at least two groups of parameters based on the working conditions, structure parameters and initial doping parameters, through simulation of the first characteristic of each CMOS device and adjustment of the initial doping parameters; the error value between the first simulation end characteristic value of each CMOS device and the first original end characteristic value of the CMOS device is less than a first preset threshold under the at least two groups of parameters; simulating the second characteristic of each CMOS device by using the at least two groups of parameters, and obtaining the second simulation end characteristic value of each CMOS device according to the simulation result; and selecting a group of target parameters from the at least two groups of parameters according to the second original end characteristic value and the second simulation end characteristic value of the multiple CMOS devices.

Description

Technical Field

[0001] This invention belongs to the field of semiconductor technology, specifically relating to a method for CMOS doping parameter process-design collaborative calibration and optimization. Background Technology

[0002] With the continuous development of semiconductor technology, the most widely used process in integrated circuits is currently complementary metal-oxide-semiconductor (CMOS) technology. This process has many advantages, such as low power consumption, high speed, high integration density, and strong anti-interference capability. CMOS technology incorporates both NMOS and PMOS transistors, involving multiple doping processes. Different doping concentrations lead to different device and circuit characteristics. For example, increasing the doping concentration in the drift region reduces the on-resistance of the device. According to Ohm's law, I... dlin Increasing the doping concentration leads to a higher critical breakdown electric field, making the device more prone to breakdown and reducing the breakdown voltage. Therefore, obtaining accurate doping parameters is crucial for developing novel semiconductor devices and designing complex circuit systems.

[0003] In actual manufacturing processes, Gaussian doping is frequently used for device doping. This is a finite surface-source diffusion method where the total amount of impurities within the silicon wafer remains constant throughout the diffusion process. Diffusion occurs only from a limited number of impurity atoms already deposited on the silicon wafer surface before diffusion. As the diffusion time increases, the surface impurity concentration decreases, and diffusion continues to propagate inwards. For CMOS devices, a longer diffusion time results in a greater diffusion depth in the drift region, a smaller peak doping concentration, and a decrease in the surface doping concentration. dlin Decreasing the drift region increases the breakdown voltage. Furthermore, the length of the drift region also affects device characteristics. For transfer characteristics, a longer drift region results in a longer path for carriers from the source to the drain, leading to a higher on-resistance and increased Ik. dlin It will decrease; for breakdown characteristics, the longer the drift region, the greater the voltage it can withstand, and the greater the breakdown voltage will be.

[0004] Currently, manufacturing processes and device structures are becoming increasingly complex, and the specific doping process is difficult to control precisely. Experimental testing often only yields terminal characteristics such as current or voltage, making it difficult to determine accurate information such as peak doping concentration and diffusion parameters. The inability to determine the specific parameters of the device during actual doping can affect the accuracy of device characteristic analysis, potentially leading to significant errors. Summary of the Invention

[0005] To address the aforementioned problems in related technologies, this invention provides a method for co-calibrating and optimizing CMOS doping parameters based on process design. The technical problem to be solved by this invention is achieved through the following technical solution:

[0006] This invention provides a method for co-calibrating and optimizing CMOS doping parameters based on process and design, comprising:

[0007] Obtain the operating conditions, first characteristic data, second characteristic data, initial doping parameters, and structural parameters of multiple CMOS devices of different sizes;

[0008] From the first characteristic data and the second characteristic data of each CMOS device, select the terminal characteristic value under preset conditions to obtain the original first terminal characteristic value and the original second terminal characteristic value;

[0009] Based on the operating conditions, the structural parameters, and the initial doping parameters, at least two sets of simulated doping parameters are determined by simulating the first characteristic of each CMOS device and adjusting the initial doping parameters; under the at least two sets of simulated doping parameters, the error between the first simulated terminal characteristic value and the first original terminal characteristic value of each CMOS device is less than a first preset threshold.

[0010] Based on the at least two sets of simulated doping parameters, the second characteristic of each CMOS device is simulated, and the second end characteristic value of each CMOS device under the preset condition is selected from the simulation results to obtain the second simulated end characteristic value.

[0011] Based on the second original terminal characteristic value and the second simulated terminal characteristic value, a target parameter is selected from the at least two sets of simulated doping parameters; under the target parameter, the error between the second simulated terminal characteristic value of each CMOS device and the second original terminal characteristic value of the CMOS device is minimized, and the error is less than or equal to the second preset threshold.

[0012] In some embodiments, the first characteristic is either a breakdown characteristic or a transfer characteristic, and the second characteristic is the other of a breakdown characteristic or a transfer characteristic; correspondingly, the original first terminal characteristic value is either the original breakdown voltage or the original drain current of the linear region, and the original second terminal characteristic value is the other of the original breakdown voltage or the original drain current of the linear region.

[0013] In some embodiments, the structural parameters and doping parameters of the plurality of CMOS devices are the same, while the drift region lengths are different.

[0014] In some embodiments, each set of simulated doping parameters includes: the doping concentration of the drift region each time and the doping concentration of the channel region each time; the doping concentration of at least one drift region or channel region is different between different sets of simulated doping parameters; based on the operating conditions, the structural parameters, and the initial doping parameters, at least two sets of simulated doping parameters are determined by simulating the first characteristic of each CMOS device and adjusting the initial doping parameters, including:

[0015] The operating conditions, structural parameters, and initial doping parameters are input into the simulation platform to simulate and obtain the multiple CMOS devices of different sizes, as well as the first initial simulation characteristic data of each CMOS device.

[0016] Select the first terminal characteristic value of each CMOS device under the preset condition from the first initial simulation characteristic data to obtain the first initial simulation terminal characteristic value;

[0017] Based on the difference between the original first end characteristic value and the first initial simulation end characteristic value, the initial doping parameters are adjusted multiple times, and after each adjustment, the first characteristic of each size of the simulated CMOS device is simulated using a set of simulation doping parameters obtained from the adjustment, so as to obtain the first simulation characteristic data of each size of the CMOS device under the set of simulation doping parameters.

[0018] From the multiple sets of simulated doping parameters corresponding one-to-one to the multiple first simulated characteristic data, the first terminal characteristic value of each CMOS device under the preset condition in each first simulated characteristic data is taken as a set of first simulated terminal characteristic values, thus obtaining multiple sets of first simulated terminal characteristic values.

[0019] Based on the variation relationship between the multiple sets of first simulation terminal characteristic values, and the correspondence between the multiple sets of simulation doping parameters and the multiple sets of first simulation terminal characteristic values, the influence relationship between the doping parameters and the first simulation terminal characteristic values ​​is determined.

[0020] The initial doping parameters are adjusted according to the influence relationship, and the at least two sets of simulated doping parameters are determined based on the adjusted doping parameters.

[0021] In some embodiments, the initial doping parameters are adjusted multiple times based on the difference between the original first end characteristic value and the first initial simulation end characteristic value. After each adjustment, a set of adjusted simulation doping parameters is used to simulate the first characteristic of the simulated CMOS device of each size, thereby obtaining the first simulation characteristic data of the CMOS device of each size under the set of simulation doping parameters. This includes:

[0022] Based on the difference between the original first-end characteristic value and the first initial simulation end characteristic value, the initial doping parameters are adjusted for the i-th time to obtain the i-th set of simulation doping parameters; i is an integer from 1 to m-1; m is an integer greater than or equal to 2;

[0023] The first characteristic of each size of CMOS device is simulated using the i-th set of simulation doping parameters to obtain the i-th first simulation characteristic data of each size of CMOS device;

[0024] Select the first terminal characteristic value of each CMOS device under the preset condition from the i-th first simulation characteristic data to obtain the i-th first simulation terminal characteristic value;

[0025] Based on the difference between the original first end characteristic value, the first initial simulation end characteristic value, and the i-th first simulation end characteristic value, the initial doping parameters are adjusted for the (i+1)th time to obtain the (i+1)-th set of simulation doping parameters.

[0026] The first characteristic of each size of CMOS device is simulated using the (i+1)th set of simulation doping parameters to obtain the (i+1)th first simulation characteristic data of each size of CMOS device.

[0027] Based on the difference between the original first end characteristic value, the first initial simulation end characteristic value, the i-th first simulation end characteristic value, and the i+1th first simulation end characteristic value in the (i+1)th first simulation characteristic data, the initial doping parameters are adjusted for the (i+2)th time to obtain the (i+2)th set of simulation doping parameters. This process is repeated until the m-th first simulation characteristic data is obtained.

[0028] In some embodiments, adjusting the initial doping parameters according to the influence relationship and determining the at least two sets of simulated doping parameters based on the adjusted doping parameters includes:

[0029] When the initial doping parameters are adjusted multiple times according to the influence relationship, and multiple sets of doping parameters are obtained, the first characteristic of each size of CMOS device is simulated using each set of doping parameters, and the first simulation characteristic data of each size of CMOS device under the set of doping parameters is obtained.

[0030] The first terminal characteristic value of each CMOS device in the first simulation characteristic data under the preset condition is taken as the first simulation terminal characteristic value;

[0031] Calculate the error between the first original terminal characteristic value of each CMOS device and the first simulated terminal characteristic value of the CMOS device under the set of doping parameters, and obtain the error result corresponding to the CMOS device under the set of doping parameters.

[0032] When the error results of multiple CMOS devices under this set of doping parameters are all less than the first preset threshold, the set of doping parameters is determined to be a set of simulated doping parameters, thereby obtaining the at least two sets of simulated doping parameters.

[0033] In some embodiments, selecting a set of target simulation doping parameters from the at least two sets of simulation doping parameters based on the second original end characteristic value and the second simulation end characteristic value includes:

[0034] Calculate the error between the second original terminal characteristic value of each CMOS device and the second simulated terminal characteristic value of the CMOS device under each set of simulated doping parameters;

[0035] When the error values ​​corresponding to multiple CMOS devices under a set of simulated doping parameters are all the minimum error values ​​among the error values ​​corresponding to multiple CMOS devices under at least two sets of simulated doping parameters, and the minimum error values ​​are all less than or equal to a second preset threshold, the set of simulated doping parameters is taken as the target parameter.

[0036] In some embodiments, the step of simulating the second characteristic of each CMOS device based on the at least two sets of simulation doping parameters, and selecting the second end characteristic value of each CMOS device under the preset conditions under each set of simulation doping parameters from the simulation results to obtain the second simulation end characteristic value includes:

[0037] Using the selected simulation doping parameters for each group, the second characteristic of the simulated CMOS device of each size is simulated to obtain the second simulation characteristic data of the CMOS device of each size under the set of simulation doping parameters.

[0038] From the second simulation characteristic data of each size of CMOS device under the set of simulation doping parameters, the second terminal characteristic value of the CMOS device under the preset conditions is selected accordingly, thereby obtaining the second simulation terminal characteristic value of each of the multiple CMOS devices under the set of simulation doping parameters.

[0039] In some embodiments, the first characteristic is a transfer characteristic, and the second characteristic is a breakdown characteristic; correspondingly, the original first terminal characteristic value is the original drain current in the linear region, the original second terminal characteristic value is the original breakdown voltage, the first simulated terminal characteristic value is the simulated drain current in the linear region, and the second simulated terminal characteristic value is the simulated breakdown voltage.

[0040] The present invention has the following beneficial technical effects:

[0041] The optimization method provided by this invention can not only determine the specific doping concentration information of a device through simulation when the device's operating conditions, first characteristic data, second characteristic data, initial doping parameters, and structural parameters are known, but also, by simulating multiple devices of different sizes, it can reduce the difficulty in obtaining accurate doping concentration information of a device because a single device is often affected by more parameters, resulting in non-unique values ​​for parameters such as drift region diffusion depth, impurity distribution, and drift region length. Thus, accurate doping concentration information of the device can be obtained, achieving accurate calibration of the device's doping concentration information.

[0042] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0043] Figure 1 A flowchart of a CMOS doping parameter process-design co-calibration and optimization method provided in an embodiment of the present invention;

[0044] Figure 2 Exemplary I provided for embodiments of the present invention dlin With the length L of the drift region diff A diagram illustrating the inverse proportional relationship between them;

[0045] Figure 3A An exemplary embodiment of the present invention illustrates the effect of changing the peak concentration of a certain Gaussian doping in the drift region on the device's Ig. dlin A diagram illustrating the impact;

[0046] Figure 3B An exemplary embodiment of the present invention illustrates the effect of changing the peak concentration of a certain Gaussian doping in the channel region on the I of the device. dlin A diagram illustrating the impact;

[0047] Figure 4 The original I of the exemplary LDMOS devices with different drift region lengths provided in the embodiments of the present invention dlin With Simulation I dlin A comparison diagram between them;

[0048] Figure 5 For devices with different drift region lengths, as exemplified in this embodiment of the invention, after adjusting the peak doping concentration, three groups of peak doping concentrations that meet the requirements are selected for simulation, resulting in Simulation I. dlin With the original I dlin A contrasting line chart;

[0049] Figure 6 The three exemplary conditions provided for embodiments of the present invention all satisfy I. dlin and L diff A schematic diagram of a broken line representing an inverse proportional relationship;

[0050] Figure 7 This is an exemplary diagram showing the comparison between the simulated breakdown voltage and the original breakdown voltage of LDMOS devices with different drift region lengths under each set of simulated doping parameters, after simulating LDMOS devices with different drift region lengths using three selected sets of simulated doping parameters.

[0051] Figure 8 The exemplary LDMOS devices with different drift region lengths provided in the embodiments of the present invention are shown in a line graph of the breakdown voltage of the device as a function of the peak doping concentration when the peak concentration of a certain Gaussian doping in the drift region is changed during the simulation process.

[0052] Figure 9A This is an exemplary schematic diagram showing the comparison between the simulated breakdown voltage and the original breakdown voltage as the drift region length of the device changes, provided for an embodiment of the present invention.

[0053] Figure 9B As an exemplary embodiment of the present invention, simulation I shows the effect of varying drift region length of the device. dlin With the original I dlin A diagram showing the changes. Detailed Implementation

[0054] The present invention will be further described in detail below with reference to specific embodiments, but the implementation of the present invention is not limited thereto.

[0055] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0056] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. In addition, those skilled in the art can combine and integrate the different embodiments or examples described in this specification.

[0057] Although the invention has been described herein in conjunction with various embodiments, those skilled in the art will understand and implement other variations of the disclosed embodiments by reviewing the accompanying drawings, disclosure, and appended claims in carrying out the claimed invention. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude a plurality. A single processor or other unit can implement several functions listed in the claims. While different dependent claims may recite certain measures, this does not mean that these measures cannot be combined to produce good results.

[0058] Figure 1 This is a flowchart of a CMOS doping parameter process-design co-calibration and optimization method provided in an embodiment of the present invention, as shown below. Figure 1 As shown, the method includes the following steps:

[0059] S101. Obtain the operating conditions, first characteristic data, second characteristic data, initial doping parameters, and structural parameters of multiple CMOS devices of different sizes.

[0060] The CMOS doping parameter process-design co-calibration and optimization method provided by this invention can be executed by a computer device with simulation software installed. Furthermore, the computer device can receive some instructions or information input by the user through input components (touch screen, mouse, keyboard, etc.) and execute the CMOS doping parameter process-design co-calibration and optimization method provided by this invention according to these instructions or information.

[0061] Here, the operating conditions, primary characteristic data, secondary characteristic data, initial doping parameters, and structural parameters of multiple CMOS devices of different sizes can be obtained by acquiring their process documents. The process document for a CMOS device can be the specification document accompanying the device at the factory. This specification document may include the operating conditions, transfer characteristic curves, breakdown characteristic curves, doping parameters (i.e., initial doping parameters, which are often not precise) for each region of the CMOS device at a specific doping number, as well as structural parameters such as the channel length, source length, and drain length.

[0062] Here, the first characteristic data can be either the transfer characteristic curve or the breakdown characteristic curve, and correspondingly, the second characteristic data can be the other one of the transfer characteristic curve and the breakdown characteristic curve.

[0063] Here, multiple CMOS devices of different sizes can be multiple CMOS devices with different drift region lengths but identical other process parameters. Specifically, they can be multiple LDMOS devices with different drift region lengths but identical other process parameters. For example, they can be: 18V HS NLDMOS, 20V HS NLDMOS, 24V HS NLDMOS, and 30V HS NLDMOS, with corresponding drift region lengths of 1.5μm, 1.65μm, 2μm, and 2.4μm, respectively.

[0064] S102. Select terminal characteristic values ​​under preset conditions from the first characteristic data and the second characteristic data of each CMOS device to obtain the original first terminal characteristic value and the original second terminal characteristic value.

[0065] Here, when the first characteristic data is the transfer characteristic curve, the second characteristic data is the breakdown characteristic curve, and the original first terminal characteristic value is the original drain current in the linear region, hereinafter referred to as the original I. dlin The original characteristic value of the second terminal is the original breakdown voltage, hereinafter referred to as the original BV.

[0066] Here, the raw I of each CMOS device dlin It is I selected from the transfer characteristic curve of the CMOS device under a preset condition. dlin The original BV of each CMOS device is a BV under a preset condition selected from the breakdown characteristic curve of that CMOS device.

[0067] Here, the preset conditions can be determined according to the different types of devices. For example, when multiple LDMOS devices of different sizes are: 18V HS NLDMOS, 20V HS NLDMOS, 24V HS NLDMOS and 30V HS NLDMOS, the preset conditions may include: 1) V d =0.1V, and V g =5V; 2)I d =1μA, therefore, the selected original first-end characteristic value can be when V d =0.1V, and V g I at 5V dlin Furthermore, the selected original second-end characteristic value can be when I d Voltage at 1μA.

[0068] S103. Based on the working conditions, structural parameters and initial doping parameters, at least two sets of simulated doping parameters are determined by simulating the first characteristic of each CMOS device and adjusting the initial doping parameters; under at least two sets of simulated doping parameters, the error between the first simulated terminal characteristic value and the first original terminal characteristic value of each CMOS device is less than a first preset threshold.

[0069] Here, each set of doping parameters can be a set of doping parameters obtained by adjusting the initial doping parameters multiple times. Each set of doping parameters can include: the doping concentration in the drift region each time and the doping concentration in the channel region each time; the doping concentration in the drift region or channel region is different at least once between different sets of doping parameters.

[0070] Here, since these multiple CMOS devices differ only in the length of their drift regions while having the same other process parameters, they can share each set of doping parameters in each simulation.

[0071] Here, the simulation platform used for simulation can be the Sentaurus TCAD platform.

[0072] Specifically, the operating conditions, structural parameters, and initial doping parameters can be input into the simulation platform first to simulate multiple CMOS devices of different sizes, and obtain the first initial simulation characteristic data for each CMOS device. Then, the first end characteristic value of each CMOS device under preset conditions is selected from the first initial simulation characteristic data to obtain the first initial simulation end characteristic value. Next, based on the difference between the original first end characteristic value and the first initial simulation end characteristic value, the initial doping parameters are adjusted multiple times. After each adjustment, the first characteristic of each size of the simulated CMOS device is simulated using the adjusted set of simulation doping parameters to obtain this set of simulation doping parameters. First simulation characteristic data of CMOS devices of each size are obtained; among the multiple sets of first simulation characteristic data corresponding one-to-one with multiple sets of simulation doping parameters, the first terminal characteristic value of each CMOS device under preset conditions in each set of first simulation characteristic data is taken as a set of first simulation terminal characteristic values, thus obtaining multiple sets of first simulation terminal characteristic values; based on the variation relationship between the multiple sets of first simulation terminal characteristic values ​​and the correspondence between the multiple sets of simulation doping parameters and the multiple sets of first simulation terminal characteristic values, the influence relationship between the doping parameters and the first terminal characteristic values ​​is determined; the initial doping parameters are adjusted according to the influence relationship, and the at least two sets of simulation doping parameters are determined based on the adjusted doping parameters.

[0073] Specifically, when adjusting the initial doping parameters multiple times, firstly, based on the difference between the original first-end characteristic value and the first initial simulated end characteristic value, the initial doping parameters can be adjusted for the i-th time to obtain the i-th set of simulated doping parameters; i is an integer from 1 to m-1; m is an integer greater than or equal to 2; then, the first characteristic of each size of CMOS device is simulated using the i-th set of simulated doping parameters to obtain the i-th first simulated characteristic data of each size of CMOS device; from the i-th first simulated characteristic data, the first-end characteristic value of each CMOS device under preset conditions is selected to obtain the i-th first simulated end characteristic value; based on the original first-end characteristic value and the first initial simulated end characteristic value, Based on the difference between the i-th first simulation end characteristic value and the initial doping parameters, the initial doping parameters are adjusted for the (i+1)th time to obtain the (i+1)-th set of simulation doping parameters. The first characteristic of each size of CMOS device is simulated using the (i+1)-th set of simulation doping parameters to obtain the (i+1)-th first simulation characteristic data of each size of CMOS device. Based on the difference between the original first end characteristic value, the first initial simulation end characteristic value, the i-th first simulation end characteristic value, and the (i+1)-th first simulation end characteristic value in the (i+1)-th first simulation characteristic data, the initial doping parameters are adjusted for the (i+2)-th time to obtain the (i+2)-th set of simulation doping parameters. This process is repeated until the m-th first simulation characteristic data is obtained.

[0074] Here, for each CMOS device, when the drain voltage is low, the channel has not yet reached saturation, and the on-resistance R... dlin Satisfying the formula: (1); where R ch R is the resistance in the channel region. diff R is the resistance in the drift region. ohmic Let be the ohmic contact resistance. The channel resistance satisfies: (2); where N ch The doping concentration in the channel region is μ. ch L represents the migration rate of the channel area. G Let S be the length of the channel region, and S be the cross-sectional area of ​​the channel. The drift region resistance satisfies the formula: (3); where N diff μ represents the doping concentration in the drift region. diff L represents the migration rate of the drift region. diff Let I be the length of the drift region. If ohmic contact resistance is ignored, and it is assumed that the channel cross-section is uniform everywhere, then I... dlin Satisfying the formula: (4); where V dlin It remains constant. From formula (4), we can derive I. dlin With the length L of the drift region diff The relationship is an inverse proportional function; please refer to the diagram. Figure 2As the doping concentration in the channel region increases, the channel mobility μ... ch The Io decreases, but the mobility of minority carriers increases, resulting in a lower Io for the same drift region length. dlin As the doping concentration in the drift region decreases, the curve in Figure 3 shifts to the left; as the doping concentration in the drift region increases, the drift region mobility μ... diff Reduce I for the same drift region length dlin Increase Figure 2 The curve shifts towards the origin. For example, when the devices of each size described in this invention undergo 4 and 3 Gaussian doping processes in the drift region and channel region, respectively, Figure 3A To vary the peak concentration of Gaussian doping during a specific Gaussian doping process in the drift region (incrementally changing to 1e...) 15 cm -3 1e 16 cm -3 1e 17 cm -3 1e 18 cm -3 ) to I dlin The impact, Figure 3B To determine the change (in sequence, becoming 1e) during a certain Gaussian doping process in the channel region. 17 cm -3 1e 18 cm -3 1e 19 cm -3 The peak concentration of this Gaussian doping on I dlin The impact, according to Figure 3A and Figure 3B It can be seen that as the peak doping concentration increases, I dlin Increase. The peak doping concentration of the remaining sub-Gaussian doping has an effect on I. dlin The effects can also be derived using a similar method. Therefore, when adjusting the initial doping parameters for the i-th time, the i-th set of simulated doping parameters can be obtained by changing the doping concentrations in the channel region and drift region of the device within the initial doping parameters. Furthermore, the i-th set of simulated doping parameters can enable simulations of devices with multiple different sizes. dlin With their respective original I dlin near.

[0075] Here, by adjusting the initial doping parameters m times and performing m corresponding simulations on the resulting m sets of simulated doping parameters, we can determine how the first-end characteristic values ​​of each device change with the doping parameters based on the first-end characteristic values ​​obtained from these m simulations and the m sets of simulated doping parameters used for simulation. In other words, we obtain the influence relationship between the doping parameters and the first-end characteristic values. Specifically, when the first-end characteristic value is I... dlinThe first original end characteristic value is the original I. dlin The first simulation terminal characteristic value is simulation I. dlin Then, based on the simulation I of each device obtained from these m simulations... dlin And the m sets of simulation doping parameters used for simulation, determining the relationship between the doping parameters and I. dlin When considering the relationship between doping concentration and I, a line graph can be plotted to find the relationship between doping concentration and I. dlin The relationship is that, and this relationship is established by adjusting the doping concentration each time, I dlin The corresponding changes are summarized. For example, when performing a Gaussian doping in the drift region, the peak concentration of that Gaussian doping is changed (in turn to 1e). 15 cm -3 1e 16 cm -3 1e 17 cm -3 1e 18 cm -3 Each change in doping concentration yields a graph with the drift region length as the horizontal axis and I as the vertical axis. dlin The broken line along the vertical axis shows that as the doping concentration changes at this point, I... dlin How exactly does the doping concentration change as the doping concentration increases? Specifically, for example... Figure 4 The line graph S1 shown represents the final target I that needs to be adjusted to. dlin —A line graph showing the drift region length, where S2 is the I value simulated from the initial doping parameters. dlin —A line graph showing the drift region length. Comparing these two lines, we can see that as the drift region length changes, the I obtained from the simulation of the original doping parameters... dlin The slope of the broken line (S2) is greater than that of the target broken line (S1), so subsequent adjustments will focus on how to reduce the slope of S2. This can be achieved by observing the above... Figure 3A It was found that increasing the peak concentration of the Gaussian doping in this instance increases the slope of the curve, so the adjustment for this Gaussian doping should be to increase the doping concentration. This analysis considers the entire curve, but sometimes changing the doping concentration of a particular doping level affects different segments of the curve differently. For example, increasing the doping concentration of a particular doping level might increase the slope of the curve segment between 1.5 μm and 1.65 μm, but decrease the slope of the curve segment between 2 μm and 2.4 μm. Therefore, when exploring the relationship between doping concentration and Idlin, the specific circumstances of each doping level should be fully considered, and adjustments should be made based on a comprehensive assessment. Furthermore, adjusting the doping concentration not only affects the slope of the curve but also its height (i.e., the numerical value at each point). Therefore, there may be many sets of conditions satisfying Idlin... dlinThe doping concentration under the constraint conditions will yield the aforementioned "at least two sets of simulated doping parameters." Therefore, further BV simulation is needed through S104~S105 to determine the final doping concentration. If the BV simulation finds that the target parameter does not exist, it is because the I... dlin The determined doping concentration is not comprehensive enough (i.e., "at least two sets of simulated doping parameters" is not comprehensive enough), and should be returned to I. dlin The simulation is performed by finding at least two sets of doping concentrations that meet the first preset threshold using the above method, so as to continue the BV simulation until the target parameters are obtained.

[0076] Here, when the first characteristic is the transfer characteristic, the operating conditions, structural parameters, and initial doping parameters can be input into the simulation platform to simulate multiple CMOS devices of different sizes, as well as the initial simulated transfer characteristic curve of each CMOS device. Then, the Ig value of each CMOS device under preset conditions can be selected from the initial simulated transfer characteristic curve. dlin The initial simulation I was obtained. dlin Then, based on the original I dlin Compared with the initial simulation I dlin The differences between them were investigated, and the initial doping parameters were adjusted multiple times. After each adjustment, the transfer characteristics of each CMOS device of different sizes were simulated using the adjusted set of simulated doping parameters. This yielded simulated transfer characteristic curves for each CMOS device of different sizes under these simulated doping parameters. From the multiple simulated transfer characteristic curves corresponding to the multiple sets of simulated doping parameters, the Ig of each CMOS device under preset conditions was analyzed. dlin As a set of simulation I dlin Multiple sets of simulation I were obtained. dlin According to multiple simulations I dlin The relationship between the changes, and the relationship between multiple sets of simulated doping parameters and multiple sets of simulated I dlin The correspondence between them is determined to establish the relationship between the doping parameters and I. dlin The influence relationship between them is determined; the initial doping parameters are adjusted according to the influence relationship, and at least two sets of simulated doping parameters are determined based on the adjusted doping parameters.

[0077] In some embodiments, the aforementioned at least two sets of simulated doping parameters may be multiple sets of doping parameters obtained by adjusting the initial doping parameters multiple times according to the aforementioned influence relationship, and at least two sets of doping parameters are selected based on the simulation results of each set of doping parameters by simulating the first characteristic of the device multiple times. Specifically, during selection, the first characteristic of each size of CMOS device can be simulated using each set of doping parameters to obtain the first simulated characteristic data of each size of CMOS device under the set of doping parameters. The first end characteristic value of each CMOS device under preset conditions in the first simulated characteristic data is taken as the first simulated end characteristic value. Then, the error value between the first original end characteristic value of each CMOS device and the first simulated end characteristic value of the CMOS device under the set of doping parameters is calculated to obtain the error result corresponding to the CMOS device under the set of simulated doping parameters. Then, when the error results corresponding to multiple CMOS devices under the set of doping parameters are all less than a first preset threshold, the set of doping parameters is determined to be a selected set of doping parameters, thereby selecting at least two sets of doping parameters from multiple sets of doping parameters, thus obtaining at least two sets of simulated doping parameters.

[0078] Here, the error value can be the difference between the first simulated end characteristic value and the first original end characteristic value, or it can be the percentage between the difference between the first simulated end characteristic value and the first original end characteristic value, etc. The embodiments of the present invention do not limit this.

[0079] Here, the first preset threshold can be set according to actual needs. For example, when the error value is the percentage between the difference between the first simulation end characteristic value and the first original end characteristic value and the first original end characteristic value, the first preset threshold can be 7%.

[0080] For example, when the first characteristic is a transfer characteristic, the first original end characteristic value is the original I. dlin The first simulation terminal characteristic value is simulation I. dlin At that time, the raw It of each CMOS device can be calculated. dlin The simulation I of the CMOS device under each set of simulated doping parameters dlin The error value between them is used to obtain the I value of the CMOS device under the set of simulated doping parameters. dlin The error results; then, when the I of multiple CMOS devices under this set of simulation doping parameters is... dlin When all the error results are less than the first preset threshold, the set of doping parameters can be determined as a selected set of doping parameters, thus allowing at least two sets of doping parameters to be selected from multiple sets of doping parameters. For example, when the above... Figure 4 Curve S2 in the figure represents the simulation results of LDMOS devices with different drift region lengths. dlinWhen constructing a broken line, the original I corresponding to the same drift region length of an LDMOS device can be calculated. dlin With Simulation I dlin The error between them is calculated to obtain the error result; and when S2 and S1 reach a certain degree of overlap, a set of doping parameters corresponding to S2 can be used as a set of selected doping parameters.

[0081] Through the above selection steps, at least two sets of peak doping concentrations in the channel region and the drift region can be obtained. Applying these peak doping concentrations to the simulation yields the following relationship: I. Simulation I dlin and L diff The inverse proportional relationship; second, comparing the original I of CMOS devices of different sizes with those of CMOS devices of different sizes. dlin The relationship between the broken lines and the simulation of CMOS devices of different sizes. dlin When comparing the relationship lines on the same coordinate system, the two relationship lines should approximately coincide. For example, when using three selected sets of peak doping concentrations that satisfy the above relationship for simulation, the simulation results of CMOS devices of different sizes will be obtained. dlin and the original I dlin The comparison chart between them can be as follows Figure 5 As shown. Figure 5 In the middle, the broken line "obtained I" dlin "This represents the original I of LDMOS devices with different drift region lengths." dlin The broken line "satisfies I" dlin and L diff The inverse proportional relationship of doping concentration 1” and the broken line “satisfies I” dlin and L diff The inverse proportional relationship between doping concentration 2” and the broken line “satisfies I” dlin and L diff The inverse proportionality relationship of doping concentration 3" refers to the simulation results of LDMOS devices with different drift region lengths obtained by performing simulations using three sets of peak doping concentrations that satisfy the above relationship. dlin .

[0082] S104. Based on at least two sets of simulated doping parameters, simulate the second characteristic of each CMOS device, select the second end characteristic value of each CMOS device under preset conditions from the simulation results, and obtain the second simulated end characteristic value.

[0083] Here, the operating conditions of the device during simulation are consistent with the operating conditions recorded in the process documentation.

[0084] Specifically, the second characteristic of each size of CMOS device can be simulated using the selected simulation doping parameters for each group, thereby obtaining the second simulation characteristic data of each size of CMOS device under the set of simulation doping parameters. Then, the second end characteristic value of the CMOS device under the preset conditions can be selected from the second simulation characteristic data of each size of CMOS device under the set of simulation doping parameters, thereby obtaining the second simulation end characteristic values ​​of multiple CMOS devices under the set of simulation doping parameters.

[0085] For example, when the second characteristic is the breakdown characteristic and the second simulated characteristic data is the simulated breakdown characteristic curve, the second simulated terminal characteristic value is the simulated breakdown voltage. Thus, the breakdown characteristics of each size of CMOS device can be simulated first using the selected simulated doping parameters of each group, and the simulated breakdown characteristic curve of each size of CMOS device under the set of simulated doping parameters can be obtained. Then, the breakdown voltage of the CMOS device under the preset conditions can be selected from the simulated breakdown characteristic curve of each size of CMOS device under the set of simulated doping parameters, so as to obtain the simulated breakdown voltage of each of the multiple CMOS devices under the set of simulated doping parameters.

[0086] Here, due to the presence of ohmic contact resistance and electrode series resistance in the actual device, through I dlin and L diff The inverse proportional relationship yields the channel region doping concentration N. ch and drift region doping concentration N diff Not unique (i.e., multiple sets of simulated doping concentrations are selected), for example, when there are three sets of simulated doping concentrations, and the I values ​​corresponding to these three sets of simulated doping concentrations are... dlin and L diff Inverse proportional relationship such as Figure 6 As shown, according to Figure 6 It can be seen that all three broken lines satisfy I. dlin and L diff The inverse proportional relationship means that only a few sets of conditions satisfying I can be obtained at this point. dlin The doping concentration of the constraint conditions can be determined, but it is impossible to determine which specific group it belongs to. Therefore, it needs to be uniquely determined by the breakdown voltage BV.

[0087] S105. Based on the second original terminal characteristic value and the second simulated terminal characteristic value, select a set of target parameters from at least two sets of simulated doping parameters; under the target parameters, the error between the second simulated terminal characteristic value and the second original terminal characteristic value of each CMOS device is minimized, and the error is less than or equal to the second preset threshold.

[0088] Specifically, the error value between the second original terminal characteristic value of each CMOS device and the second simulated terminal characteristic value of the CMOS device under each set of simulated doping parameters can be calculated first. Then, when the error values ​​corresponding to multiple CMOS devices under a set of simulated doping parameters are all the minimum error values ​​among the error values ​​corresponding to multiple CMOS devices under at least two sets of simulated doping parameters, and the minimum error values ​​are all less than or equal to a second preset threshold, the set of simulated doping parameters is taken as the target parameter. The first preset threshold can be equal to the second preset threshold, for example, it can be 5%.

[0089] Here, when selecting the target parameter from at least two sets of simulated doping parameters, if the error values ​​corresponding to multiple CMOS devices under one set of simulated doping parameters are all the minimum error values ​​among the error values ​​corresponding to multiple CMOS devices under these at least two sets of simulated doping parameters, but one or more minimum error values ​​are greater than the second preset threshold, it indicates that these at least two sets of simulated doping parameters are not comprehensive enough. In this case, the above-mentioned method for determining these at least two sets of simulated doping parameters can be used to re-determine at least two new sets of simulated doping parameters, and the target parameter can be re-selected from these at least two new sets of simulated doping parameters until the target parameter is selected.

[0090] For example, when the second characteristic is a breakdown characteristic, the second original terminal characteristic value is the original breakdown voltage value, and the second simulated terminal characteristic value is the simulated breakdown voltage, the error value between the original breakdown voltage of each CMOS device and the simulated breakdown voltage of that CMOS device under each set of simulated doping parameters can be calculated first. Then, when the error values ​​corresponding to multiple CMOS devices under a set of simulated doping parameters are all the minimum error values ​​among the error values ​​corresponding to multiple CMOS devices under at least two sets of simulated doping parameters, and all are less than 5%, this set of simulated doping parameters is used as the target parameter. For example, such as... Figure 7 As shown, when at least two sets of simulated doping parameters selected by S104 are: "satisfying I dlin and L diff The inverse proportional relationship of doping concentration 1" and "satisfying I" dlin and L diff The inverse proportional relationship between doping concentration 2 and "satisfies I" dlin and L diff When simulating the three sets of doping parameters with inverse proportional relationships (doping concentration 3), we can plot the relationship between LDMOS devices with different drift region lengths and their respective original breakdown voltages, obtaining the "obtained BV" plot. We can also plot the relationship between LDMOS devices with different drift region lengths and their respective simulated breakdown voltages under these three sets of simulated doping parameters, obtaining the "satisfying I" plot. dlin and L diff The inverse proportional relationship of doping concentration 1" and "satisfying I"dlin and L diff The inverse proportional relationship between doping concentration 2 and "satisfies I" dlin and L diff The three broken lines represent the inverse proportional relationship of doping concentration 3". According to... Figure 7 It can be seen that the error between the original breakdown voltage and each simulated breakdown voltage for LDMOS devices with the same drift region length can be calculated, and, as... Figure 7 As shown, because "satisfies I" dlin and L diff The line segment showing the inverse proportional relationship between doping concentration 1 and the line segment showing the obtained BV has the highest overlap (i.e., "satisfying I"). dlin and L diff The error values ​​corresponding to different drift region lengths of LDMOS devices under the simulated doping parameters of "inverse proportional relationship I" are all the minimum error values ​​among the error values ​​corresponding to different drift region lengths of LDMOS devices under these three sets of simulated doping parameters, and the minimum error value is less than or equal to 5%. Therefore, "satisfying I" can be considered as... dlin and L diff The set of simulated doping parameters with an inverse proportional relationship of doping concentration 1" is used as the final set of simulated doping parameters. Furthermore, this set of simulated doping parameters is used to simulate the device's doping concentration 1. dlin With the original I dlin This is a compromise solution where the goodness of fit between the simulated breakdown voltage and the original breakdown voltage are relatively good.

[0091] Here, the device mainly experiences avalanche breakdown during breakdown, with the same mechanism as PN junction avalanche breakdown. The expression for avalanche breakdown of a single-sided abrupt junction is: (5); where N D To determine the doping concentration in the lower doping region, applying this formula to this embodiment yields the following formula: (6). Therefore, according to formula (6), when the drift region length is fixed, decreasing the drift region doping concentration can increase the device breakdown voltage; when the drift region doping concentration is fixed, increasing the drift region length can increase the device breakdown voltage. When the device undergoes a certain Gaussian doping in the drift region, the changes (in sequence 1e) are as follows: 15 cm -3 1e 16 cm -3 5e 15 cm -3 The effect of the peak concentration of this Gaussian doping on the breakdown voltage (BV) of devices with different drift region lengths can be found in [reference needed]. Figure 8The effect of the peak doping concentration of other Gaussian dopants on the breakdown voltage BV of devices with different drift region lengths can also be derived using a similar method. Based on the above formulas and... Figure 8 The relationship between the peak doping concentration in the drift region and the device's breakdown voltage (BV) shows that, among the multiple sets of peak doping concentrations in the channel region and drift region determined in S104, finding the set that satisfies the breakdown voltage of devices with different drift region lengths, and finally determining the unique peak doping concentrations in the channel region and drift region (target simulation doping parameters), is feasible. Plotting the simulated breakdown voltages of devices with different drift region lengths under the finally determined set of peak doping concentrations in the channel region and drift region, along with the original breakdown voltages, on the same coordinate system yields a line graph as shown below. Figure 9A As shown, the line segment "BV (Test Data)" represents the relationship between the original breakdown voltage of devices with different drift region lengths and the drift region length of the device itself; the line segment "BV (Final Simulation Result)" represents the relationship between the simulated breakdown voltage of devices with different drift region lengths under a final set of channel region peak doping concentrations and drift region peak doping concentrations and the drift region length of the device itself; and the simulated breakdown voltage of devices with different drift region lengths under a final set of channel region peak doping concentrations and drift region peak doping concentrations is shown in Figure I. dlin With the original I dlin Plotting the relationship within the same coordinate system yields a line graph as follows: Figure 9B As shown, the broken line “Idlin (test data)” represents the original I of the device with different drift region lengths. dlin A line graph showing the relationship between the device's drift region length and the device's own drift region length; the line graph "Idlin (final simulation result)" represents the final set of simulation results for devices with different drift region lengths under different peak doping concentrations in the channel and drift regions. dlin A broken line showing the relationship between the length of the device's drift region and the device's own drift region. According to... Figure 9A and 9B As shown, under the finally determined set of peak doping concentrations in the channel region and peak doping concentrations in the drift region, simulation results for devices with different drift region lengths are presented. dlin With the original I dlin Furthermore, the simulated breakdown voltage and the original breakdown voltage are both well-fitted, indicating that the peak doping concentration in the channel region and the peak doping concentration in the drift region are accurate.

[0092] As described above, the doping concentration of CMOS devices can be divided into drift region doping concentration and channel region doping concentration. The breakdown voltage is only related to the drift region doping concentration. dlin It is related to both the doping concentration in the drift region and the doping concentration in the channel region. For a known breakdown voltage, I... dlinFor devices with a fixed drift region width, the drift region doping concentration can be determined using the doping concentration-breakdown voltage relationship curve. Based on the determined drift region doping concentration, the doping concentration can be further reduced using the doping concentration-I... dlin The relationship curve is used to determine the doping concentration in the channel region. This allows for the optimization design of the doping process and the acquisition of doping parameters that meet the requirements.

[0093] The following simulation data further illustrates the effectiveness of the optimization method provided by this invention. Table 1 shows the simulation I values ​​determined using the optimization method of this invention for devices with different drift region lengths. dlin (Simulation data) and the original I of devices with different drift region lengths. dlin Table 2 shows the comparison and error between the simulated breakdown voltage (simulation data) determined by the optimization method of this invention for devices with different drift region lengths and the original breakdown voltage (test data) for devices with different drift region lengths.

[0094]

[0095] Table 1

[0096]

[0097] Table 2

[0098] This invention utilizes the TCAD simulation tool and only uses the I of multiple dimensional devices. dlin The precise doping concentration of a CMOS device can be determined by the breakdown voltage (BV), allowing TCAD simulation results to reproduce the test results of key CMOS device parameters. The calibrated simulation parameters are then compared with the test parameters. dlin Both the breakdown voltage and the accuracy of the CMOS device reached a high level with an error of less than 5%, thus enabling the calibration of the doping concentration of the CMOS device.

[0099] The optimization method provided by this invention not only allows for the determination of specific doping concentration information of a device through simulation, given its operating conditions, first characteristic data, second characteristic data, initial doping parameters, and structural parameters, but also reduces the difficulty in obtaining accurate doping concentration information by simulating multiple devices of different sizes. This is because a single device is often affected by more parameters, resulting in non-unique values ​​for parameters such as drift region diffusion depth, impurity distribution, and drift region length. Therefore, accurate doping concentration information can be obtained, achieving accurate calibration of the device's doping concentration information. Ultimately, this improves the accuracy of device characteristic analysis and facilitates better design and optimization of device and circuit parameters. This invention enables doping parameter calibration from process to design, providing a reference for process designers.

[0100] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A method for calibrating and optimizing CMOS doping parameters, characterized in that, include: Obtain the operating conditions, first characteristic data, second characteristic data, initial doping parameters, and structural parameters of multiple CMOS devices of different sizes; From the first characteristic data and the second characteristic data of each CMOS device, select terminal characteristic values ​​under preset conditions to obtain the original first terminal characteristic value and the original second terminal characteristic value; the first characteristic is either the breakdown characteristic or the transfer characteristic, and the second characteristic is the other one of the breakdown characteristic and the transfer characteristic; correspondingly, the original first terminal characteristic value is either the original breakdown voltage or the original drain current of the linear region, and the original second terminal characteristic value is the other one of the original breakdown voltage and the original drain current of the linear region; Based on the operating conditions, the structural parameters, and the initial doping parameters, at least two sets of simulated doping parameters are determined by simulating the first characteristic of each CMOS device and adjusting the initial doping parameters; under the at least two sets of simulated doping parameters, the error value between the first simulated terminal characteristic value of each CMOS device and the original first terminal characteristic value of the CMOS device is less than a first preset threshold. Each set of simulated doping parameters includes: the doping concentration in the drift region and the doping concentration in the channel region for each time; the doping concentration in the drift region or channel region is different at least once between different sets of simulated doping parameters; Based on the at least two sets of simulated doping parameters, the second characteristic of each CMOS device is simulated, and the second end characteristic value of each CMOS device under the preset condition is selected from the simulation results to obtain the second simulated end characteristic value. Based on the original second-end characteristic value and the second simulated end characteristic value, a set of target parameters is selected from the at least two sets of simulated doping parameters; under the target parameters, the error between the second simulated end characteristic value of each CMOS device and the original second-end characteristic value of the CMOS device is minimized, and the error is less than or equal to the second preset threshold. Based on the operating conditions, the structural parameters, and the initial doping parameters, at least two sets of simulated doping parameters are determined through simulation of the first characteristic of each CMOS device and adjustment of the initial doping parameters, including: The operating conditions, structural parameters, and initial doping parameters are input into the simulation platform to simulate and obtain the multiple CMOS devices of different sizes, as well as the first initial simulation characteristic data of each CMOS device. Select the first terminal characteristic value of each CMOS device under the preset condition from the first initial simulation characteristic data to obtain the first initial simulation terminal characteristic value; Based on the difference between the original first end characteristic value and the first initial simulation end characteristic value, the initial doping parameters are adjusted multiple times, and after each adjustment, the first characteristic of each size of the simulated CMOS device is simulated using a set of simulation doping parameters obtained from the adjustment, so as to obtain the first simulation characteristic data of each size of the CMOS device under the set of simulation doping parameters. From the multiple sets of simulated doping parameters corresponding one-to-one to the multiple first simulated characteristic data, the first terminal characteristic value of each CMOS device under the preset condition in each first simulated characteristic data is taken as a set of first simulated terminal characteristic values, thus obtaining multiple sets of first simulated terminal characteristic values. Based on the variation relationship between the multiple sets of first simulation terminal characteristic values, and the correspondence between the multiple sets of simulation doping parameters and the multiple sets of first simulation terminal characteristic values, the influence relationship between the doping parameters and the first simulation terminal characteristic values ​​is determined. The initial doping parameters are adjusted according to the influence relationship, and the at least two sets of simulated doping parameters are determined based on the adjusted doping parameters.

2. The method for calibrating and optimizing CMOS doping parameters according to claim 1, characterized in that, The structural and doping parameters of the multiple CMOS devices are the same, but the drift region lengths are different.

3. The method for calibrating and optimizing CMOS doping parameters according to claim 1, characterized in that, Based on the difference between the original first-end characteristic value and the first initial simulation end characteristic value, the initial doping parameters are adjusted multiple times. After each adjustment, the first characteristic of the simulated CMOS device of each size is simulated using the adjusted set of simulation doping parameters to obtain the first simulation characteristic data of the CMOS device of each size under the set of simulation doping parameters, including: Based on the difference between the original first-end characteristic value and the first initial simulation end characteristic value, the initial doping parameters are adjusted for the i-th time to obtain the i-th set of simulation doping parameters; i is an integer from 1 to m-1; m is an integer greater than or equal to 2; The first characteristic of each size of CMOS device is simulated using the i-th set of simulation doping parameters to obtain the i-th first simulation characteristic data of each size of CMOS device; Select the first terminal characteristic value of each CMOS device under the preset condition from the i-th first simulation characteristic data to obtain the i-th first simulation terminal characteristic value; Based on the difference between the original first end characteristic value, the first initial simulation end characteristic value, and the i-th first simulation end characteristic value, the initial doping parameters are adjusted for the (i+1)th time to obtain the (i+1)-th set of simulation doping parameters. The first characteristic of each size of CMOS device is simulated using the (i+1)th set of simulation doping parameters to obtain the (i+1)th first simulation characteristic data of each size of CMOS device. Based on the difference between the original first end characteristic value, the first initial simulation end characteristic value, the i-th first simulation end characteristic value, and the i+1th first simulation end characteristic value in the (i+1)th first simulation characteristic data, the initial doping parameters are adjusted for the (i+2)th time to obtain the (i+2)th set of simulation doping parameters. This process is repeated until the m-th first simulation characteristic data is obtained.

4. The method for calibrating and optimizing CMOS doping parameters according to claim 1, characterized in that, The step of adjusting the initial doping parameters according to the influence relationship, and determining the at least two sets of simulated doping parameters based on the adjusted doping parameters, includes: When the initial doping parameters are adjusted multiple times according to the influence relationship, and multiple sets of doping parameters are obtained, the first characteristic of each size of the simulated CMOS device is simulated using each set of doping parameters, and the first simulation characteristic data of each size of the CMOS device under the set of doping parameters is obtained. The first terminal characteristic value of each CMOS device in the first simulation characteristic data under the preset condition is taken as the first simulation terminal characteristic value; Calculate the error between the original first-terminal characteristic value of each CMOS device and the simulated first-terminal characteristic value of the CMOS device under the set of doping parameters, and obtain the error result corresponding to the CMOS device under the set of doping parameters; When the error results of multiple CMOS devices under this set of doping parameters are all less than the first preset threshold, the set of doping parameters is determined to be a set of simulated doping parameters, thereby obtaining the at least two sets of simulated doping parameters.

5. The method for calibrating and optimizing CMOS doping parameters according to claim 1, characterized in that, The step of selecting a target parameter from the at least two sets of simulated doping parameters based on the original second-end characteristic value and the second simulated end characteristic value includes: Calculate the error between the original second-terminal characteristic value of each CMOS device and the simulated second-terminal characteristic value of the CMOS device under each set of simulated doping parameters; When the error values ​​corresponding to multiple CMOS devices under a set of simulated doping parameters are all the minimum error values ​​among the error values ​​corresponding to multiple CMOS devices under at least two sets of simulated doping parameters, and the minimum error values ​​are all less than or equal to the second preset threshold, the set of simulated doping parameters is taken as the target parameter.

6. The method for calibrating and optimizing CMOS doping parameters according to claim 1, characterized in that, The step of simulating the second characteristic of each CMOS device based on the at least two sets of simulation doping parameters, selecting the second end characteristic value of each CMOS device under the preset conditions under each set of simulation doping parameters from the simulation results, and obtaining the second simulation end characteristic value includes: Using the selected simulation doping parameters for each group, the second characteristic of the simulated CMOS device of each size is simulated to obtain the second simulation characteristic data of the CMOS device of each size under the set of simulation doping parameters. From the second simulation characteristic data of each size of CMOS device under the set of simulation doping parameters, the second terminal characteristic value of the CMOS device under the preset conditions is selected accordingly, thereby obtaining the second simulation terminal characteristic value of each of the multiple CMOS devices under the set of simulation doping parameters.

7. The method for calibrating and optimizing CMOS doping parameters according to claim 1, characterized in that, The first characteristic is the transfer characteristic, and the second characteristic is the breakdown characteristic; correspondingly, the original first terminal characteristic value is the original drain current in the linear region, the original second terminal characteristic value is the original breakdown voltage, the first simulated terminal characteristic value is the simulated drain current in the linear region, and the second simulated terminal characteristic value is the simulated breakdown voltage.

Images