TCAD-based nbiti effect recovery phase simulation method

Abstract

The application discloses a TCAD-based NBTI effect recovery stage simulation method, and the implementation steps are as follows: generating a target device structure and defining a model in TCAD simulation software, performing degradation simulation on the target device, performing recovery simulation on the degraded target device, respectively calculating the average value of the charged state ratio of the ABDWT model at the channel region and the gate oxide layer interface of the degraded target device and the target device at different recovery time points, calculating the ratio of the charged state ratio at each recovery time point, and finally multiplying the threshold voltage degradation amount of the degraded target device to obtain the threshold voltage degradation amount at each recovery time point. The application has the advantages of being completely developed based on the TCAD simulation tool, being simple in operation and calculation, being capable of reflecting the real behavior of micro particles, and being small in data error.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of electronic devices, and further relates to a negative bias temperature instability (NBTI) effect recovery stage simulation method based on technology computer aided design (TCAD). BACKGROUND

[0002] In semiconductor device engineering, as the device size becomes smaller and smaller, the reliability problem of the device becomes more and more serious, and the most representative problem among the reliability problems is the NBTI effect. At present, the simulation of the NBTI effect recovery stage mainly adopts two methods of TCAD simulation and mathematical formula representation, wherein the mathematical formula representation has disadvantages of complex formula and lack of physical basis. Although the TCAD simulation can overcome these disadvantages, at present, the TCAD simulation of the NBTI effect recovery stage has problems of inaccurate simulation. Therefore, the equivalent simulation of the NBTI effect recovery stage based on the TCAD simulation software can simply and accurately describe the NBTI effect recovery stage.

[0003] Parihar in his published paper "BTI Analysis Tool—Modeling of NBTI DC, AC Stress and Recovery Time Kinetics, Nitrogen Impact, and EOL Estimation" (DOI number: 10.1109 / TED.2017.2780083) discloses a simulation method of NBTI effect recovery stage based on mathematical model. The implementation steps of the method include: user simulates NBTI effect degradation stage in TCAD simulation software; exports trap concentration at channel region and gate oxide interface; sets formula parameters; recalculates trap concentration at interface using transient trap occupancy model (TTOM); finally obtains the change of NBTI effect recovery stage. The mathematical model TTOM disclosed by the method uses three components: the traps not recovered in the previous cycle are recovered by hydrogen passivation, the traps not recovered in the previous cycle are recovered by trapping electrons, and the newly generated traps in the current cycle are recovered by trapping electrons to describe the NBTI effect recovery stage. However, the method still has the following shortcomings: the TTOM model is too complex, and there are too many parameters in the model, and any change of any parameter will affect the final result. And the formula lacks physical theory support and cannot reflect the actual physical behavior of microscopic particles. And the TTOM model is not integrated in the TCAD simulation software, so the data needs to be exported for further processing, which is low in efficiency and easy to produce errors.

[0004] Choudhury in his published paper "Modeling of DC-AC NBTI Stress-Recovery Time Kinetics in P-Channel Planar Bulk and FDSOI MOSFETs and FinFETs" (DOI number: 10.1109 / JEDS.2020.3023803) discloses a simulation method for simulating the recovery stage of NBTI effect by using an activated barrier double well thermionic (ABDWT) model instead of the TTOM model. The implementation steps of the method include: the user simulates the degradation stage of NBTI effect in the TCAD simulation software; the trap concentration at the channel and gate oxide interface is derived; the formula parameters are set; the occupation state of the method at the interface is calculated using the ABDWT model; the occupation state and the trap concentration at the interface are associated; and finally the change of the recovery stage of NBTI effect is obtained. The method uses the ABDWT model with a physical theoretical basis to simulate the recovery stage of NBTI effect. However, the method still has the deficiency that the model is not integrated in the TCAD simulation software, and the data still needs to be exported for further processing, which is easy to produce errors. And the model does not participate in the simulation process of the TCAD simulation software, and cannot accurately reflect the real behavior of the micro-particles. SUMMARY

[0005] The purpose of the present application is to overcome the deficiencies of the prior art, and to provide a NBTI effect recovery stage simulation method based on TCAD simulation software, which aims to solve the problems of complex mathematical model calculation process, easy to produce errors, lack of physical theory support, not integrated in the TCAD simulation software and cannot reflect the real behavior of the micro-particles, and the problem that the TCAD simulation in the prior art cannot accurately describe the actual test results of the NBTI effect recovery stage.

[0006] The idea for achieving the object of the present application is that the present application establishes a model of a target device in TCAD simulation software, applies an ABDWT model built in the TCAD simulation software to the target device to simulate a charged state ratio of the ABDWT model at a channel and gate oxide interface of the target device, and equivalently solves the problem that the previous simulation method cannot be integrated in the TCAD and cannot reflect the behavior of real micro particles. The present application biases the target device in a stress state to obtain a degradation amount of the device under the action of the NBTI effect and an average value of the charged state ratio of the ABDWT model at the channel and gate oxide interface after the stress ends, then, biases the target device in a recovery state to obtain average values of the charged state ratio of the ABDWT model at the interface under different recovery times, makes a ratio of the average values of the charged state ratio under different recovery times and the average value of the charged state ratio after the stress ends, and multiplies the ratio of the average values of the charged state ratio under different recovery times with the degradation amount of the target device under the action of the NBTI effect degradation stage to obtain a recovery curve of the recovery stage of the target device, thereby solving the problems that the previous simulation method has a complex calculation process and needs to export data for processing again and the problem that the previous TCAD simulation cannot reflect the actual change in the recovery stage of the NBTI effect.

[0007] The steps of the present application include the following:

[0008] Step 1, defining a model according to the generated target device structure:

[0009] Using TCAD simulation software, generating a target device structure according to actual process parameters of a target device to be simulated; defining a multi-state configuration hydrogen transport degradation model at the interface of a channel region and a gate oxide layer of the target device, and defining related parameters in a.par file; defining an ABDWT model at the interface of the channel region and the gate oxide layer of the target device, and defining related parameters in the.par file;

[0010] Step 2, performing degradation simulation on the target device:

[0011] Using TCAD simulation software, simulating a normal working state of the target device to obtain a transfer characteristic curve of the target device; performing degradation transient simulation on the target device in the normal working state to obtain a degraded target device; simulating a normal working state of the degraded target device to obtain a transfer characteristic curve of the degraded target device; and calculating a degradation amount ΔV0 of a threshold voltage of the degraded target device by using a formula ΔV0=V1-V0, wherein V1 and V0 respectively represent threshold voltages extracted from the transfer characteristic curve of the degraded target device and the transfer characteristic curve of the target device in the normal working state by using a threshold voltage extraction tool inside the TCAD simulation software;

[0012] Step 3, the degenerated target device is simulated to recover:

[0013] TCAD simulation software is used to simulate the NBTI effect recovery stage of the degenerated target device at different times, and the target device at different recovery times is obtained;

[0014] Step 4, calculate the average value of the charged state ratio:

[0015] Step 4.1 uses TCAD simulation software to open the.tdr file of the degenerated target device and the target device at different recovery times respectively;

[0016] Step 4.2 respectively views the charged state ratio of the ABDWT model of the degenerated target device and the target device at different recovery times;

[0017] Step 4.3 uses the tangent tool of TCAD simulation software to make tangent at the channel region and the gate oxide layer interface of the degenerated target device and the target device at different recovery times, respectively, to obtain the charged state ratio of the ABDWT model of the degenerated target device and the target device at different recovery times, and calculate the average value respectively.

[0018] Step 5, according to the following formula, the ratio of the charged state ratio at each recovery time is calculated:

[0019]

[0020] Wherein, A t The ratio of the charged state ratio at recovery time t is s t And s0 respectively represent the average value of the ABDWT model charged state ratio at the channel region and the gate oxide layer interface of the target device at recovery time t and the degenerated target device;

[0021] Step 6, according to the following formula, the threshold voltage degradation amount at each recovery time is calculated:

[0022] ΔV2(t) = ΔV0 x A t

[0023] Wherein, ΔV2(t) represents the threshold voltage degradation amount at recovery time t.

[0024] Compared with the prior art, the present application has the following advantages:

[0025] Firstly, the simulation method is based on a commercial TCAD simulation software, the ABDWT model is calculated to participate in the simulation process, and the real micro-particle behavior can be reflected, so that the problem that the existing technology cannot be integrated into the TCAD simulation software and cannot reflect the real micro-particle behavior is overcome. The application has the advantages of being able to reflect the real behavior of micro-particles and being integrated into the TCAD simulation software.

[0026] Secondly, the application can be built into a commercial TCAD simulation software, the trap occupancy probability at the interface between the channel and the gate oxide layer of the target device is obtained by the simulation result of the TCAD simulation software, and the problem that the interface trap concentration needs to be derived from the TCAD simulation software and then the occupancy probability is calculated in the prior art is overcome. The application has the advantages of simple operation and small data error.

[0027] Thirdly, the application obtains the simulation curve of the NBTI effect recovery stage by calculating the average value of the charged state ratio distribution of the ABDWT model at the interface between the channel and the gate oxide layer of the device at the end of stress and in the recovery process, calculating the ratio of the charged state ratio in the recovery process to the charged state ratio at the end of stress, and multiplying the threshold voltage degradation amount after the device is stressed. The application overcomes the problem that the mathematical formula calculation process is complex and error-prone in the prior art. The application has the advantage of simple calculation. BRIEF DESCRIPTION OF DRAWINGS

[0028] Figure 1 is a block diagram of the application;

[0029] Figure 2 is a structure diagram of a 28nm p-type MOSFET device used in the specific embodiment of the application;

[0030] Figure 3 is a trend diagram of the ratio of the ABDWT model occupancy probability at the interface between the channel region and the gate oxide layer of the 28nm p-type MOSFET device in the specific embodiment of the application to the recovery time;

[0031] Figure 4 is a comparison diagram of the recovery curve obtained by the traditional TCAD method and the recovery curve obtained by using the ABDWT method in the specific embodiment of the application and experimental data. DETAILED DESCRIPTION

[0032] The application will be further described below in combination with the drawings and specific embodiments.

[0033] Reference Figure 1 and examples, the implementation steps of the application will be further described.

[0034] Step 1, an embodiment of the present application is to use Sentaurus TCAD simulation software, according to the actual process parameters of 28nm p-type MOSFET, as shown in Table 1, using scheme language in Sentaurus SDE tool to establish simulation program, running the simulation program to get the corresponding 28nm p-type MOSFET device structure, as shown in Figure 2

[0035] Table 1 28nm p-type MOSFET process parameter table

[0036] Source drain region doping 3 x 10 20 cm -3 ]]> Substrate region doping 6 x 10 17 cm -3 ]]> Halo region doping 6 x 10 18 cm -3 ]]> Gate length 30 nm SiO2layer thickness 0.6 nm [CAT] Hf02 layer thickness 2 nm

[0037] Figure 2 The upper left of the 28nm p-type MOSFET device is the source, the upper right of the device is the drain, the middle is the gate of the device, and the dashed line is the interface of the channel and the gate oxide layer of the device;

[0038] Step 2, define the position and parameters of the multi-state configuration hydrogen transport degradation model:

[0039] Step 2.1, using Sentaurus TCAD simulation software, define the multi-state configuration hydrogen transport degradation model at the interface of the channel region and the gate oxide layer of the 28nm p-type MOSFET device. The multi-state configuration hydrogen transport degradation model is a self-contained model of Sentaurus TCAD simulation software;

[0040] Step 2.2, according to the following formula, calculate the forward reaction rate and reverse reaction rate of the multi-state configuration hydrogen transport degradation model:

[0041]

[0042]

[0043] Wherein, k 01 represents the forward reaction rate of the multi-state configuration hydrogen transport degradation model, σ represents the capture cross section of holes in the channel, v th represents the average value of all hole thermal velocities in the channel, n represents the hole concentration in the channel of 28nm p-type MOSFET which is updated in real time by Sentaurus TCAD simulation software during simulation, exp[.] represents the exponential operation with natural base e, γ represents the pre-factor of electric field enhancement in the insulating layer determined by the process parameters of 28nm p-type MOSFET, F represents the electric field intensity in the insulating layer, V / cm represents the electric field intensity per unit in the insulating layer, ρ represents the electric field enhancement index in the insulating layer determined by the process parameters of 28nm p-type MOSFET, W f ​represents the activation energy of the forward reaction, q represents the electronic charge, a represents a factor that accounts for the reduction of the potential barrier due to the electric field in the insulating layer, which is determined by the process parameters of the 28 nm p-type MOSFET device, k represents the Boltzmann constant, which has a value of 1.380649 x 10 -23 J / K, T represents the temperature of the 28 nm p-type MOSFET device, which has a value of 398 K 10 represents the reverse reaction rate of the multi-state configuration hydrogen transport degradation model, k0 represents a deactivation rate constant, which has a value of 5 x 10 -6 / s, exp(.) represents an exponential operation with a base of the natural exponential e, W r represents the activation energy of the reverse reaction, [H] represents the hydrogen concentration at the interface between the channel region and the gate oxide layer of the 28 nm p-type MOSFET device, which is updated in real time during simulation by the Sentaurus TCAD simulation software, 1 / cm 3 represents the hydrogen concentration per unit.

[0044] In the present embodiment, but not limited to, a = 3 x 10 -10 , v th = 3 x 10 -6 , g = 4 x 10 -7 , W f = 0.22, a = 1.2 x 10 -9 , k0 = 5 x 10 -6 , W r = 0.12, T = 398;

[0045] Step 3, define the position and parameters of the ABDWT model:

[0046] Step 3.1, using the Sentaurus TCAD simulation software, define the ABDWT model at the interface between the channel region and the gate oxide layer of the 28 nm p-type MOSFET device. The ABDWT model is a built-in model of the Sentaurus TCAD simulation software.

[0047] Step 3.2, calculate the forward reaction rate and the reverse reaction rate of the multi-state configuration ABDWT model according to the following formula:

[0048]

[0049]

[0050] wherein, k 12 represents the rate of forward conversion of the ABDWT model, v represents the attempt frequency of forward or reverse reaction, exp(.) represents an exponential operation with a base of the natural exponential e, E Brepresents the barrier height between the charged and uncharged states of the ABDWT model, E1represents the energy of the uncharged state of the ABDWT model, γ represents a fitting factor determined by the process parameters of the 28 nm p-type MOSFET device, E ins represents the electric field strength in the gate oxide layer, k represents the Boltzmann constant, which has a value of 1.380649 x 10 -23 J / K, T represents the temperature of the 28 nm p-type MOSFET device, which has a value of 398 K, k 21 represents the rate of reverse conversion of the ABDWT model, exp[.] represents the exponential operation with the natural base e, E2represents the energy of the charged state of the ABDWT model, m represents a fitting factor determined by the process parameters of the 28 nm p-type MOSFET device.

[0051] where E B is calculated according to the following formula, following a Gaussian distribution related to temperature: B

[0052]

[0053]

[0054] where <E B > represents the average value of E B , <E B0 > represents the initial average value of E B , exp(.) represents the exponential operation with the natural base e, E A represents the activation energy related to temperature, k represents the Boltzmann constant, which has a value of 1.380649 x 10 - 23 J / K, T represents the temperature of the 28 nm p-type MOSFET device, which has a value of 398 K, σ EB represents the variance of E B , σ EB0 represents the initial variance of E B .

[0055] E2follows a Gaussian distribution, with an average value and a variance of: E 2_mean , E 2_spread .

[0056] In this embodiment, but not limited to, v = 1 x 10 13 , γ = 4 x 10 -9 , m = 1.5, <E B0 > = 0.9, σ EB0 = 0.24, E A = 0, E 2_mean = 0.27, E​2_spread = 0, T = 398;

[0057] Step 4, extracting threshold voltage in normal working state:

[0058] Step 4.1, biasing the drain and gate of the 28nm p-type MOSFET device at -1.0V in turn;

[0059] Step 4.2, opening the.plt output file of the 28nm p-type MOSFET device using the inspect tool of Sentaurus TCAD simulation software, generating the transfer characteristic curve of the device with the gate voltage of the device as the horizontal axis and the drain current of the device as the vertical axis;

[0060] Step 4.3, extracting the threshold voltage V0 in normal working state from the transfer characteristic curve of the 28nm p-type MOSFET device using the threshold voltage extraction tool in the inspect tool of Sentaurus TCAD simulation software;

[0061] Step 5, reducing the drain voltage of the 28nm p-type MOSFET device to 0, biasing the gate voltage at -1.5V, and then performing a degradation transient simulation for 1000s to obtain the degraded 28nm p-type MOSFET device;

[0062] Step 6, using the Svisual tool of Sentaurus TCAD simulation software to view the InterfaceABDWTState2 characteristic in the.tdr file of the degraded 28nm p-type MOSFET device, using the tangent tool of Svisual to draw a tangent at the channel region and gate oxide interface of the device, obtaining the distribution s0(x) of the charge state ratio of the ABDWT model at the channel region and gate oxide interface of the device, where x represents the position at the interface;

[0063] Step 7, extracting threshold voltage after degradation:

[0064] Step 7.1, biasing the drain and gate of the degraded 28nm p-type MOSFET device at -1.0V in turn;

[0065] Step 7.2, opening the.plt output file of the degraded 28nm p-type MOSFET device using the inspect tool of Sentaurus TCAD simulation software, generating the transfer characteristic curve of the device with the gate voltage of the device as the horizontal axis and the drain current of the device as the vertical axis;

[0066] Step 7.3, the threshold voltage extraction tool in the inspect tool of Sentaurus TCAD simulation software is used to extract the threshold voltage V1 of the degraded 28 nm p-type MOSFET device from the transfer characteristic curve of the degraded 28 nm p-type MOSFET device;

[0067] Step 8, the degradation amount AV0 of the threshold voltage of the degraded 28 nm p-type MOSFET device is calculated by the formula AV0 = V1 - V0;

[0068] Step 9, the drain bias voltage of the 28 nm p-type MOSFET device is reduced to 0 V, and the gate bias voltage is increased to 0.6 V, respectively, and the recovery transient simulation is performed for 10 -6 s, 10 -5 s, 10 -4 s, 10 -3 s, 0.01 s, 0.1 s, 1 s, 10 s, 100 s, 1000 s, to obtain the 28 nm p-type MOSFET device at different recovery times;

[0069] Step 10, the Svisual tool of Sentaurus TCAD simulation software is used to view the InterfaceABDWTState2 characteristics in the.tdr file of the 28 nm p-type MOSFET device at different recovery times, and the tangent tool of Svisual is used to draw a tangent line at the channel and gate oxide interface of the device at different recovery times, to obtain the charged state ratio distribution s t (x) of the ABDWT model at the interface at different recovery times, wherein t represents the recovery time, and x represents the position at the interface;

[0070] Step 11, the threshold voltage at different recovery times is extracted:

[0071] Step 11.1, the drain and gate of the 28 nm p-type MOSFET device at different recovery times are sequentially biased at -1.0 V;

[0072] Step 11.2, the.plt output file of the 28 nm p-type MOSFET device at different recovery times is opened by using the inspect tool of Sentaurus TCAD simulation software, and the transfer characteristic curve of the device at different recovery times is generated by taking the gate voltage of the device at different recovery times as the horizontal axis and the drain current of the device at different recovery times as the vertical axis;

[0073] Step 11.3, the threshold voltage V2(t) at different recovery times is extracted from the transfer characteristic curve of the 28 nm p-type MOSFET device at different recovery times by using the threshold voltage extraction tool in the inspect tool of Sentaurus TCAD simulation software, wherein t represents the recovery time.

[0074] Step 12, the threshold voltage degradation amount ΔV1(t) of the 28nm p-type MOSFET device at different recovery times is calculated by using the formula ΔV1(t) = V2(t) - V0, and the trend of ΔV1(t) with the recovery time is obtained as shown in the solid line in FIG. 2, with the recovery time as the horizontal axis and ΔV1(t) as the vertical axis. Figure 4

[0075] Step 13, s0(x) and s t (x) are averaged with respect to x, respectively, to obtain s0 and s t , respectively, wherein s0 represents the average value of the ABDWT model charged state ratio of the 28nm p-type MOSFET device at the channel region and the gate oxide layer interface after degradation, wherein s t represents the average value of the ABDWT model charged state ratio of the 28nm p-type MOSFET device at the channel region and the gate oxide layer interface at the recovery time t.

[0076] Step 14, the ratio of the charged state ratio at each recovery time is calculated according to the following formula:

[0077]

[0078] , wherein A t represents the ratio of the charged state ratio at the recovery time t, with the recovery time as the horizontal axis and A t as the vertical axis, and the trend of A t with the recovery time is obtained as shown in the dashed line in FIG. 2. Figure 3

[0079] Step 15, the threshold voltage degradation amount at each recovery time is calculated according to the following formula:

[0080] ΔV2(t) = ΔV0 x A t

[0081] , wherein ΔV2(t) represents the threshold voltage degradation amount at the recovery time t, with the recovery time as the horizontal axis and ΔV2(t) as the vertical axis, and the trend of ΔV2(t) with the recovery time is obtained as shown in the dashed-dotted line in FIG. 2. Figure 4

[0082] The present application will be further described in combination with the simulation experiment of the present application.

[0083] 1. Simulation experiment conditions.

[0084] The software platform of the simulation experiment of the present application is: Red Hat 7.9 operating system and Sentaurus TCAD2021.

[0085] 2. Simulation content and result analysis.​​​

[0086] The simulation experiment of the application has two.

[0087] The simulation experiment 1 is to simulate the structure of a 28nm p-type MOSFET device by using the SDE tool of the Sentaurus TCAD simulation software.

[0088] The Sentaurus TCAD simulation software used in the simulation experiment 1 of the application is developed by Synopsys Company, and the version number is 2021.06.

[0089] The simulation experiment 1 of the application simulates the structure of a 28nm p-type MOSFET device by using the process parameters shown in Table 1, and the obtained device structure is shown in Figure 2 .

[0090] The simulation experiment 2 is to simulate the NBTI effect degradation stage and recovery stage of a 28nm p-type MOSFET device by using the application and one prior art (NBTI effect recovery stage MSC simulation method), obtain the ratio of the charged ratio of the ABDWT model at different recovery times and the change of the threshold voltage degradation amount in the recovery stage, draw the relationship between the obtained ratio of the charged ratio of the ABDWT model and the recovery time into the curve shown in Figure 3 , and draw the relationship between the obtained threshold voltage degradation amount in the recovery stage and the recovery time into the curve shown in Figure 4 .

[0091] In the simulation experiment 2, one prior art refers to:

[0092] The prior art NBTI effect recovery stage MSC simulation method refers to the simulation method of the NBTI effect recovery stage based on the multi-state configuration hydrogen transport degradation model proposed by Tiwari et al. in the paper “A 3-D TCAD Framework for NBTI—Part I: Implementation Details and FinFET Channel Material Impact, IEEE TRANSACTIONS ON ELECTRON DEVICES, vol. 66, no. 5, pp. 2086-2092, May. 2019 ”, which is referred to as NBTI effect recovery stage MSC simulation method.

[0093] The effects of the application will be further described below in combination with the simulation diagrams of Figure 2 , Figure 3 , Figure 4 .

[0094] Figure 2 The top left of the image shows the source of a 28nm p-type MOSFET device, the top right shows the drain of the device, the middle shows the gate of the device, and the dashed line shows the interface between the channel region and the gate oxide layer of the device.

[0095] Figure 3 The horizontal axis represents the recovery time during the recovery process of the NBTI effect, in seconds, and the vertical axis represents the ratio of the charge ratios in the ABDWT model. Figure 3 In this context, 0.6V represents the gate voltage during the recovery process, and 398K represents the temperature of a 28nm p-type MOSFET device.

[0096] from Figure 3 As can be seen, the charge ratio of the ABDWT model obtained using the method of this invention decreases continuously with the increase of recovery time. In the initial stage of the recovery phase (10... -6 From 1 s to 1 s, the ratio of the charge ratio in the ABDWT model decreases rapidly with increasing recovery time. In the later stage of the recovery phase (from 1 s to 1000 s), the ratio of the charge ratio in the ABDWT model decreases slowly with increasing recovery time.

[0097] Figure 4 The horizontal axis represents the recovery time during the NBTI effect recovery process, in seconds, and the vertical axis represents the threshold voltage degradation, in mV. Circles represent the relationship between threshold voltage degradation and recovery time during the NBTI effect recovery stage as measured in actual experiments; solid lines represent the relationship curve between threshold voltage degradation and recovery time during the NBTI effect recovery stage obtained using the existing MSC simulation method; and dashed lines represent the relationship curve between threshold voltage degradation and recovery time during the NBTI effect recovery stage obtained using the method proposed in this invention. Figure 4 In this context, 0.6V represents the gate voltage during the recovery process, and 398K represents the temperature of a 28nm p-type MOSFET device.

[0098] from Figure 4 As can be seen, the threshold voltage degradation during the NBTI effect recovery stage obtained by the method of the present invention decreases continuously with the increase of recovery time. Under the same recovery time conditions, the threshold voltage degradation during the NBTI effect recovery stage obtained by the method of the present invention decreases faster and by a larger amount, which is closer to the trend of actual experimental measurement results and better reflects the true state of the device during the NBTI effect recovery stage.

Claims

1. A simulation method for the recovery phase of the NBTI effect based on TCAD, characterized in that, The ratio of the charged states in the ABDWT model at the channel region and gate oxide interface of the target device under different recovery times is calculated using TCAD simulation software to equivalently calculate the NBTI effect recovery stage. The specific steps of this method are as follows: Step 1: Define a model based on the generated target device structure: Using TCAD simulation software, the target device structure is generated based on the actual process parameters of the target device to be simulated; a multi-state configuration hydrogen transport degradation model is defined at the interface between the channel region and the gate oxide layer of the target device, and the relevant parameters are defined in the .par file; an ABDWT model is defined at the interface between the channel region and the gate oxide layer of the target device, and the relevant parameters are defined in the .par file. Step 2: Perform degradation simulation on the target device: Using TCAD simulation software, the normal operating state of the target device is simulated to obtain the transfer characteristic curve of the target device; a degradation transient simulation is performed on the target device under normal operating state to obtain the degraded target device; the normal operating state of the degraded target device is simulated to obtain the transfer characteristic curve of the degraded target device; the degradation amount ΔV0 of the threshold voltage of the degraded target device is calculated using the formula ΔV0=V1–V0, where V1 and V0 represent the threshold voltage extracted from the transfer characteristic curve of the degraded target device and the transfer characteristic curve of the target device under normal operating state using the threshold voltage extraction tool inside the TCAD simulation software, respectively. Step 3: Perform recovery simulation on the degraded target device: Using TCAD simulation software, transient simulations of the recovery phase of the NBTI effect of the degraded target device were performed at different times to obtain the target device at different recovery times. Step 4, calculate the average state-of-charge ratio: Step 4.1 Use TCAD simulation software to open the .tdr files of the degraded target device and the target device at different recovery times; Step 4.2 Examine the charge state ratios of the ABDWT models of the degraded target device and the target device at different recovery times; Step 4.3 Use the tangent tool of TCAD simulation software to make tangents at the channel region and gate oxide interface of the target device after degradation and at different recovery times, respectively, to obtain the ABDWT model charge state ratio after degradation and at different recovery times, and calculate the average value respectively. Step 5: Calculate the ratio of the charged state ratios at each recovery time using the following formula: Among them, A t The ratio of the proportions of the charged states at recovery time t, s t s0 and s0 represent the average values ​​of the ABDWT model charged state ratios at the channel region and gate oxide interface of the target device at recovery time t and the degraded target device, respectively. Step 6: Calculate the threshold voltage degradation at each recovery time using the following formula: ΔV2(t)=ΔV0×A t Where ΔV2(t) represents the threshold voltage degradation at recovery time t.

2. The simulation method for the recovery phase of the NBTI effect based on TCAD as described in claim 1, characterized in that, The transfer characteristic curve of the target device mentioned in step 2 refers to the transfer characteristic curve of the target device with the gate voltage of the target device as the horizontal axis and the drain current of the target device as the vertical axis.

3. The simulation method for the recovery phase of the NBTI effect based on TCAD as described in claim 1, characterized in that, The degradation transient simulation mentioned in step 2 refers to simulating transient times customized according to user requirements.

4. The simulation method for the recovery phase of the NBTI effect based on TCAD as described in claim 1, characterized in that, The recovery transient simulation mentioned in step 3 refers to performing simulation based on a transient time that is customized according to user requirements.

Images