A Method for Establishing IV Models and Extracting Parameters for Enhanced GaN HEMT Devices

Abstract

The application discloses a kind of enhanced GaN HEMT device I-V model establishment and parameter extraction method, belong to power device field.First, the transfer characteristic and output characteristic curve of enhanced GaN HEMT device under different irradiation doses are measured by experiment;Subsequently, based on ASM-HEMT model, the I-V model of enhanced GaN HEMT device considering total dose effect is established, and the model parameters are sorted and classified;Then, the value range of the to-be-extracted parameters is preliminarily determined by deep level transient spectrum test and manual fitting experimental curve;Finally, the optimal value of the to-be-extracted parameters is found by global fitting, and the fitting result is derived.The I-V model of enhanced GaN HEMT device established by the application can accurately predict the change of electrical characteristics of enhanced GaN HEMT device under the action of total dose effect, and the parameters derived based on the parameter extraction method proposed by the application can be used for the preparation of GaN-based integrated circuit simulation file, so as to predict the effect of total dose effect on the performance of GaN-based integrated circuit, and provide technical support for subsequent anti-total dose irradiation design.

Description

Technical Field

[0001] This invention relates to the field of power device technology, and in particular to a method for establishing an IV model and extracting parameters for an enhanced GaN HEMT device. Background Technology

[0002] A typical GaN HEMT device structure contains a heterojunction formed by an AlGaN barrier layer and a GaN channel layer. A triangular potential well is formed at the heterojunction interface due to band discontinuity. Under the influence of piezoelectric polarization and spontaneous polarization, polarization charges are induced at the heterojunction interface. These polarization charges are confined within the potential well and, through lateral movement, form a conductive channel connecting the drain and source, i.e., a two-dimensional electron gas. Even with zero gate-source bias, a two-dimensional electron gas conductive channel can still form between the drain and source, thus the device is in a normally-on state, i.e., a depletion-mode device. Researchers both domestically and internationally typically use the method of growing a p-GaN cap layer on top of the AlGaN barrier layer to deplete the two-dimensional electrons and achieve enhancement-mode devices. Compared to depletion-mode devices, enhancement-mode devices exhibit lower static power consumption and superior switching performance.

[0003] Compared to traditional silicon-based devices, enhancement-mode GaN HEMTs offer faster switching speeds, higher breakdown voltages, superior thermal conductivity, and lower on-resistance, making them promising for applications in deep space exploration, satellite communications, and other aerospace fields. Semiconductor devices used in aerospace applications are exposed to high-energy particle radiation in space for extended periods, inevitably suffering from ionizing radiation damage. The total dose effect (TDE) is a cumulative dose-induced ionizing radiation effect that describes the performance degradation caused by radiation-induced trap charges. Studying the impact of the TDE on the performance of enhancement-mode GaN HEMT devices, and further analyzing its effect on the overall performance of GaN-based integrated circuits, is of profound significance for promoting the development of radiation-hardened design for GaN-based integrated circuits and their application in the aerospace field. Existing test results show that, under high-dose gamma irradiation, enhancement-mode GaN HEMT devices exhibit performance degradation phenomena such as negative threshold voltage drift and increased leakage current due to the TDE. However, few studies have reported on the changes in the performance of GaN-based integrated circuits caused by the TDE.

[0004] EDA technology serves as the link between semiconductor devices and circuit design, and semiconductor device models are one of the core components of EDA technology. Specifically, a semiconductor device model is a series of mathematical equations that characterize the electrical properties of a device by modeling various physical effects within the device. Using a reliable device model allows for rapid and accurate prediction of circuit behavior, effectively shortening the R&D cycle and reducing R&D costs. The reliability of a device model depends not only on the model's quantification of complex internal physical effects but also on the extraction process of relevant parameters. The more accurate the parameters extracted by fitting experimental test curves, the smaller the difference between the circuit characteristics simulated using the device model and the tested characteristics of the finished product. Currently, the ASM-HEMT model is a certified, industry-standard compact model for GaN power devices based on surface potentials, but it does not model the total dose effect, and no other enhancement-mode GaN HEMT device IV models considering the total dose effect have been reported. Therefore, how to establish a reliable enhancement-mode GaN HEMT device IV model considering the total dose effect and propose a method for extracting relevant parameters from this model has become a pressing technical problem in this field. Summary of the Invention

[0005] The purpose of this invention is to provide a method for establishing an IV model and extracting parameters for an enhanced GaN HEMT device, so as to solve the problems in the background art.

[0006] To address the aforementioned technical problems, this invention provides a method for establishing an IV model and extracting parameters for an enhanced GaN HEMT device, comprising:

[0007] Step 1: Experimentally measured the transfer characteristic curves and output characteristic curves of the enhanced GaN HEMT under different irradiation doses;

[0008] Step 2: Based on the ASM-HEMT model, establish an enhanced GaNHEMT device IV model that considers the total dose effect, organize the model parameters, and classify them into: structural parameters and parameters to be extracted;

[0009] Step 3: Develop corresponding parameter extraction software;

[0010] Step 4: Filter the experimental data obtained in Step 1 and import it into the parameter extraction software;

[0011] Step 5: Input the structural parameters of the enhanced GaN HEMT into the parameter extraction software, and set the fitting accuracy and the number of fitting iterations;

[0012] Step 6: Initially determine the value range of the parameters to be extracted;

[0013] Step 7: Globally fit and compare the model's predicted curve with the experimental test curve in real time to find the optimal value of the parameter to be extracted;

[0014] Step 8. Fitting complete. Observe and export the fitting results, specifically: fitting data and fitting parameters.

[0015] In one embodiment, step 1 uses an average energy of 1.25 MeV. 60 Irradiation tests were conducted using a Co gamma-ray source; the irradiation dose rate was set to 50 rad(Si) / s, and irradiation dose points were selected at 0 M rad(Si), 0.3 M rad(Si), 0.6 M rad(Si), and 1 M rad(Si). Two enhanced GaN HEMT devices from the same batch were used at the same irradiation dose point to ensure that at least two sets of data were compared and analyzed under each radiation condition.

[0016] In one embodiment, in step 1, an on-state bias is applied to the enhancement-mode GaN HEMT device using a low-voltage DC power supply, i.e., V0. GS =3V, V DS =0.5V, where V GS V is the gate-source voltage. DS This is the drain-source voltage.

[0017] In one embodiment, in step 2, based on the ASM-HEMT model, the enhancement-mode GaN HEMT device IV model considering the total dose effect is established as follows:

[0018]

[0019] Among them, I D Where w is the leakage current, l is the gate width, and C is the gate length. g V is the effective conductance between the gate and the two-dimensional electron gas. th ψ is the thermal voltage, λ is the channel length modulation coefficient; DS =ψ D -ψ S , ψ m =(ψ D +ψ S ) / 2, ψ D With ψ S These are the surface potentials of the drain and source, respectively; μ eff,sat To account for mobility degradation and velocity saturation effects, the effective electron mobility is compared with the low-field mobility μ0 and the first-order mobility attenuation coefficient μ. a Second-order mobility attenuation coefficient μ b And related to the electron velocity saturation coefficient thesat; ΔV TID The threshold voltage drift caused by the total dose effect is expressed as follows:

[0020]

[0021] Where the negative sign indicates the negative shift in the threshold voltage of the enhancement-mode GaN HEMT device caused by the total dose effect, q is the electron charge, and N e c represents the concentration of free electrons near the p-GaN cap layer / AlGaN barrier layer interface within the p-GaN cap layer. p c n e p e n These are the hole capture coefficient, electron capture coefficient, hole emission rate, and electron emission rate, respectively, with hole emission rate e0. p and electron emission rate e n Both are related to the hole trap energy level E introduced by total dose irradiation within the p-GaN cap layer. t Related to; P t N represents the hole trap density within the p-GaN cap layer near the p-GaN cap layer / AlGaN barrier layer interface after total dose irradiation. h P represents the concentration of radiation-induced and escape recombination holes in the p-GaN cap layer. t With N h They can be represented as:

[0022] P t =P0+a·DOSE b (3)

[0023] N h =DOSE·g0·f y ·t cap (4)

[0024] Where P0 is the hole trap density in the p-GaN cap layer before total dose irradiation, DOSE is the total irradiation dose, and t cap denoted as p-GaN cap layer thickness; a and b are the total dose irradiation-induced hole trap density parameters, where b is related to the formation process of irradiation-induced hole traps; g0 is the number of electron-hole pairs generated in the p-GaN cap layer per unit dose irradiation; f y The hole yield for escape recombination is related to the electric field strength within the p-GaN cap layer.

[0025] In one embodiment, the structural parameters in step 2 are specifically: gate length l, gate width w, gate index nf, and p-GaN cap thickness t. cap and the thickness t of the AlGaN barrier layer bar ;

[0026] The parameter to be extracted is specifically: escaped composite hole production f. yHole trap density P0 in the p-GaN cap layer before total dose irradiation; hole trap density parameters a and b induced by total dose irradiation; hole trap energy level E introduced by total dose irradiation in the p-GaN cap layer. t Resistance temperature detector (RTD) R th Threshold voltage V off Threshold voltage temperature dependence coefficient kt1, subthreshold slope coefficient nfactor, drain-source voltage scaling coefficient vdscale related to drain-induced barrier reduction effect, drain voltage-induced subthreshold slope change coefficient cdscd, drain-induced barrier reduction effect coefficient eta0, effective drain voltage exponential coefficient delta, electron mobility temperature dependence coefficient ute, low-field mobility μ0, first-order mobility decay coefficient μ a Second-order mobility attenuation coefficient μ b The channel length modulation coefficient lambda, the electron saturation velocity vsat, the electron velocity saturation coefficient thesat, and the temperature dependence coefficient at of the electron saturation velocity.

[0027] In one implementation, step 3 involves developing parameter extraction software TID-HEMT using MatlabApp Designer.

[0028] In one embodiment, in step 4, two sets of test data at the same irradiation dose point are compared horizontally. If a large difference is found between the two, the test data at different irradiation dose points are compared vertically. The trend of the curve changing with the increase of irradiation dose is observed, individual abnormal abrupt curves are screened out, and appropriate curve data is selected and imported into the parameter extraction software TID-HEMT.

[0029] In one implementation, in step 5, the fitting accuracy is set to 10. -20 The fitting iteration count was 100.

[0030] In one implementation, the specific method for initially determining the range of values ​​for total dose irradiation-related parameters in step 6 is as follows:

[0031] (1) The positions and densities of hole traps in p-GaN under different irradiation doses were measured using deep-level transient spectroscopy. The hole trap density P0 in the p-GaN cap layer before total dose irradiation, the hole trap density parameters a and b induced by total dose irradiation, and the hole trap energy level E introduced by total dose irradiation in the p-GaN cap layer were preliminarily determined by fitting the equation (3). t The range of values ​​for;

[0032] (2) Through the gate-source voltage V GS With p-GaN cap thickness t cap The electric field intensity E inside the p-GaN cap layer is estimated by the ratio of the two values, using the empirical formula f. y=[(|E|+E1) / (|E|+E2)] m E1 = 10 -6 MV / cm, E2=0.2MV / cm, m=0.9, preliminarily determine the hole production f of escape recombination. y The range of values ​​for .

[0033] In one implementation, the specific method for initially determining the value range of other parameters in step 6 is as follows:

[0034] (1) Import the experimental data of the enhanced GaNHEMT device under different irradiation doses into the image fitting interface of the parameter extraction software TID-HEMT, and plot I D -V GS Curve selection: Manual fitting; Adjust threshold voltage V. off Low-field mobility μ0, first-order mobility attenuation coefficient μ a and the second-order mobility attenuation coefficient μ b By observing and comparing I D -V GS The model's predicted values ​​and experimental values ​​in the image are used to preliminarily determine V. off μ0, μ a and μ b The range of values ​​for;

[0035] (2) In the image fitting interface of the parameter extraction software TID-HEMT, import the experimental data of the enhanced GaNHEMT device under different irradiation doses, and plot Log(I D )-V GS Curve selection: Manual fitting; Adjust threshold voltage V. off The following parameters were considered: subthreshold slope coefficient nfactor, drain-source voltage scaling factor vdscale related to the drain-induced barrier reduction effect, drain voltage-induced subthreshold slope change coefficient cdscd, and drain-induced barrier reduction effect coefficient eta0. These parameters were compared by observing Log(I... D )-V GS The model's predicted values ​​and experimental values ​​in the image are used to re-determine V. off The range of values ​​for nfactor, vdscale, cdscd, and eta0 is initially determined.

[0036] (3) Import the experimental data of the enhanced GaNHEMT device under different irradiation doses into the image fitting interface of the parameter extraction software TID-HEMT, and plot I D -V DS For the curve, select manual fitting and adjust the thermal resistance R. thThe threshold voltage temperature dependence coefficient kt1, the electron mobility temperature dependence coefficient ute, the channel length modulation coefficient lambda, the electron saturation velocity vsat, the electron velocity saturation coefficient thesat, and the electron saturation velocity temperature dependence coefficient at are observed and compared. D -V DS The model's predicted values ​​and experimental values ​​in the image are used to preliminarily determine R. th The range of values ​​for kt1, ute, lambda, vsat, thesat, and at.

[0037] In one implementation, in step 7, the upper and lower bounds of the parameter values ​​to be extracted, determined in step 6, are entered in the parameter setting interface of the parameter extraction software TID-HEMT. Then, the fitting is started, and the optimal values ​​of all parameters to be extracted are found by comparing the model prediction curve and the experimental test curve in real time through global fitting.

[0038] In one implementation, in step 8, when the error between the model prediction value and the experimental test value is less than the fitting accuracy or the maximum number of fitting iterations is reached, a "Fitting End" prompt box pops up. The model prediction curve and the experimental test curve displayed on the TID-HEMT image fitting interface of the parameter extraction software are observed and compared. If the difference between the two is large, the parameter setting interface is returned to modify the fitting accuracy and the number of fitting iterations, and the fitting is performed again until the model prediction curve and the experimental test curve meet the requirements, and the fitting is completed.

[0039] In one implementation, in step 8, the user selects to export the fitted leakage current data I in the TID-HEMT parameter extraction software data processing interface. D _fit and the parameter values ​​obtained from the fitting.

[0040] The present invention provides a method for establishing an IV model and extracting parameters for an enhanced GaN HEMT device, which has the following beneficial technical effects:

[0041] (1) The IV model of the enhanced GaN HEMT device considering the total dose effect established in this invention has a clear physical meaning for each parameter, which can accurately predict the changes in the electrical characteristics of the enhanced GaN HEMT device under the total dose effect.

[0042] (2) The model parameters derived from this invention can be used to write relevant circuit simulation files, predict the effect of total dose on the performance of GaN-based integrated circuits, provide technical support for subsequent anti-total dose irradiation design, and effectively shorten the R&D cycle and reduce R&D costs.

[0043] (3) This invention can be implemented by developing parameter extraction software, which is easy to operate, provides intuitive results, and has high reusability. Attached Figure Description

[0044] Figure 1 A flowchart of a method for establishing an IV model and extracting parameters for an enhanced GaN HEMT device considering the total dose effect, provided by the present invention;

[0045] Figure 2 The parameter setting interface for the parameter extraction software TID-HEMT;

[0046] Figure 3 The interface for manually fitting the parameters to be extracted in the parameter extraction software TID-HEMT;

[0047] Figure 4 The interface for exporting fitted data and fitted parameters for the parameter extraction software TID-HEMT;

[0048] Figure 5 This is a comparison of the model-predicted values ​​and experimental values ​​of the transfer characteristic curves of enhanced GaN HEMT devices under different irradiation doses after fitting.

[0049] Figure 6 This is a comparison chart of the model predictions and experimental test values ​​of the output characteristic curves of the enhanced GaN HEMT device under different irradiation doses after fitting. Detailed Implementation

[0050] The following detailed description, in conjunction with the accompanying drawings and specific embodiments, provides a more detailed explanation of the method for establishing an enhanced GaN HEMT device IV model and extracting parameters proposed in this invention. The advantages and features of this invention will become clearer from the following description. It should be noted that the accompanying drawings are all in a very simplified form and use non-precise scales, and are only used to facilitate and clarify the illustration of the embodiments of this invention.

[0051] This invention provides a method for establishing an IV model and extracting parameters for enhancement-mode GaN HEMT devices considering the total dose effect, the flowchart of which is shown below. Figure 1 As shown, it includes the following steps:

[0052] Step 1. Experimentally measure the transfer characteristic curves and output characteristic curves of the enhancement-mode GaN HEMT device under different irradiation doses.

[0053] Specifically, an average energy of 1.25 MeV is used. 60 Irradiation tests were conducted using a Co gamma-ray source, with the irradiation dose rate set at 50 rad(Si) / s. Irradiation dose points were selected at 0 M rad(Si), 0.3 M rad(Si), 0.6 M rad(Si), and 1 M rad(Si).

[0054] Irradiation experiments were conducted using domestically produced commercial p-GaN gate-enhanced AlGaN / GaN HEMT devices (650V, 11A, 120mΩ) of the same batch, model NP20G65D6. Two devices were used at the same irradiation dose point to ensure that at least two sets of data were available for comparative analysis under each radiation condition.

[0055] The enhancement-mode GaN HEMT device was soldered onto a TO-247 adapter board, and an on-state bias was applied to the device using a low-voltage DC power supply, i.e., a 3V gate-source voltage and a 0.5V drain-source voltage were applied.

[0056] Step 2. Based on the ASM-HEMT model, establish an enhanced GaNHEMT device IV model that considers the total dose effect, organize the model parameters, and classify them into: structural parameters and parameters to be extracted.

[0057] The mechanism by which the total dose effect affects the electrical properties of enhancement-mode GaN HEMT devices is as follows: Accumulated dose irradiation induces electron-hole pairs in the p-GaN cap layer, while simultaneously introducing hole traps. Induced electrons, due to their higher mobility, are rapidly swept out of the p-GaN cap layer under the influence of the electric field. Induced holes, after initial recombination, move towards the p-GaN cap layer / AlGaN barrier layer interface under the influence of the electric field, where they are captured by hole traps near the p-GaN cap layer / AlGaN barrier layer interface, forming positive trap charges. These trap charges weaken the depletion effect of the p-GaN cap layer on the two-dimensional electron gas, leading to a negative shift in the device threshold voltage and an increase in leakage current. Based on the ASM-HEMT model, an IV model of the enhancement-mode GaN HEMT device considering the total dose effect is established as follows:

[0058]

[0059] Among them, I D Where is the leakage current, w is the gate width, l is the gate length, and V is the leakage current. GS V is the gate-source voltage. DS V is the drain-source voltage. off For the threshold voltage, C g V is the effective conductance between the gate and the two-dimensional electron gas. th ψ is the thermal voltage, λ is the channel length modulation coefficient; DS =ψ D -ψ S , ψ m =(ψ D +ψ S ) / 2, ψ D With ψ S These are the surface potentials of the drain and source, respectively; μ eff,sat To account for mobility degradation and velocity saturation effects, the effective electron mobility is compared with the low-field mobility μ0 and the first-order mobility attenuation coefficient μ.a Second-order mobility attenuation coefficient μ b And related to the electron velocity saturation coefficient thesat; ΔV TID The threshold voltage drift caused by the total dose effect is expressed as follows:

[0060]

[0061] Where the negative sign indicates the negative shift in the threshold voltage of the enhancement-mode GaN HEMT device caused by the total dose effect, q is the electron charge, and N e c represents the concentration of free electrons near the p-GaN cap layer / AlGaN barrier layer interface within the p-GaN cap layer. p With c n Let c be the hole trapping coefficient and the electron trapping coefficient. p N h With c n N e These represent the hole capture rate and the electron capture rate, respectively; e p With e n These are the hole emission rate and electron emission rate, respectively, both of which are related to the hole trap energy level E introduced within the p-GaN cap layer by the total dose irradiation. t Related to; P t N represents the hole trap density within the p-GaN cap layer near the p-GaN cap layer / AlGaN barrier layer interface after total dose irradiation. h P represents the concentration of radiation-induced and escape recombination holes in the p-GaN cap layer. t With N h They can be represented as:

[0062] P t =P0+a·DOSE b (3)

[0063] N h =DOSE·g0·f y ·t cap (4)

[0064] Where P0 is the hole trap density in the p-GaN cap layer before total dose irradiation, DOSE is the total irradiation dose, and t cap denoted as p-GaN cap layer thickness; a and b are the total dose irradiation-induced hole trap density parameters, where b is related to the formation process of irradiation-induced hole traps; g0 is the number of electron-hole pairs generated in the p-GaN cap layer per unit dose irradiation; f y The hole production rate for escaping recombination is related to the electric field strength within the p-GaN cap layer.

[0065] Therefore, the structural parameters are: gate length l, gate width w, gate index nf, and p-GaN cap thickness t.cap and the thickness t of the AlGaN barrier layer bar .

[0066] The parameter to be extracted is: escaped composite hole production f. y Hole trap density P0 in the p-GaN cap layer before total dose irradiation; hole trap density parameters a and b induced by total dose irradiation; hole trap energy level E introduced by total dose irradiation in the p-GaN cap layer. t Resistance temperature detector (RTD) R th Threshold voltage V off Threshold voltage temperature dependence coefficient kt1, subthreshold slope coefficient nfactor, drain-source voltage scaling coefficient vdscale related to drain-induced barrier reduction effect, drain voltage-induced subthreshold slope change coefficient cdscd, drain-induced barrier reduction effect coefficient eta0, effective drain voltage exponential coefficient delta, electron mobility temperature dependence coefficient μ te Low-field mobility μ0, first-order mobility attenuation coefficient μ a Second-order mobility attenuation coefficient μ b The channel length modulation coefficient lambda, the electron saturation velocity vsat, the electron velocity saturation coefficient thesat, and the temperature dependence coefficient at of the electron saturation velocity.

[0067] Step 3. Develop corresponding parameter extraction software.

[0068] The parameter extraction software is TID-HEMT, developed using MatlabApp Designer.

[0069] Step 4. Filter the experimental data obtained in Step 1 and import it into the parameter extraction software.

[0070] By comparing two sets of test data at the same irradiation dose point horizontally, if a significant difference is found, then by comparing test data at different irradiation dose points vertically, observe the trend of the curve as the irradiation dose increases, filter out individual abnormal abrupt curves, and select appropriate curve data for importing into the parameter extraction software TID-HEMT, including the gate-source voltage V. GS Drain-source voltage V DS Total radiation dose (DOSE) and leakage power source (I) D data.

[0071] Step 5. Input the structural parameters of the enhanced GaN HEMT into the parameter extraction software, and set the fitting accuracy and the number of fitting iterations.

[0072] like Figure 2 As shown, the input parameters for the enhanced GaN HEMT structure are: gate length l = 1 μm, gate width w = 50 μm, gate index nf = 2, and p-GaN cap thickness t.cap =60nm, and the AlGaN barrier layer thickness t bar =15nm.

[0073] Specifically, such as Figure 2 As shown, the fitting accuracy is set to 10. -20 The fitting iteration count was 100.

[0074] Step 6. Preliminarily determine the value range of the parameters to be extracted.

[0075] The process of initially determining the value range of the total dose irradiation-related parameters is as follows:

[0076] (1) By using deep-level transient spectroscopy to test the hole trap energy levels in the p-GaN cap layer before and after total dose irradiation, one energy level was found to be located at E. v A hole trap at +3.28 eV has a pre-irradiation density of approximately 1 × 10⁻⁶. 10 cm -2 That is, the value of P0 is in the range of 1×10. 10 cm -2 The hole trap is related to the + / 0 charge state transition of nitrogen vacancies, and its formation process is a one-time process, so the value of parameter b is around 1; and the energy level density of the hole trap increases monotonically with the increase of total irradiation dose. By fitting with equation (3), the value range of parameter a of the hole trap density induced by total dose irradiation is initially determined to be 3×10. 5 cm -2 / rad;

[0077] (2) Through the gate-source voltage V GS With p-GaN cap thickness t cap The electric field strength within the p-GaN cap layer is estimated using the ratio E = 3V / 60nm = 0.5MV / cm, and then calculated using the empirical formula f. y =[(|E|+E1) / (|E|+E2)] m E1 = 10 -6 MV / cm, E2=0.2MV / cm, m=0.9, the hole production f of escape recombination is preliminarily determined. y The value range is around 0.7.

[0078] The process of initially determining the value range of other parameters is as follows:

[0079] (1) As Figure 3 As shown, experimental data of enhanced GaN HEMT devices under different irradiation doses were imported into the TID-HEMT image fitting interface of the parameter extraction software, and I was plotted. D -V GS Curve selection: Manual fitting; Adjust threshold voltage V. offLow-field mobility μ0, first-order mobility attenuation coefficient μ a and the second-order mobility attenuation coefficient μ b By observing and comparing I D -V GS The model's predicted values ​​and experimental values ​​in the image are used to preliminarily determine V. off μ0, μ a and μ b The range of values ​​for;

[0080] (2) Figure 3 As shown, experimental data of enhanced GaN HEMT devices under different irradiation doses were imported into the image fitting interface of the parameter extraction software TID-HEMT, and Log(I) was plotted. D )-V GS Curve selection: Manual fitting; Adjust threshold voltage V. off The following parameters were considered: subthreshold slope coefficient nfactor, drain-source voltage scaling factor vdscale related to the drain-induced barrier reduction effect, drain voltage-induced subthreshold slope change coefficient cdscd, and drain-induced barrier reduction effect coefficient eta0. These parameters were compared by observing Log(I... D )-V GS The model's predicted values ​​and experimental values ​​in the image are used to re-determine V. off The value range is around 1.28V, and the value ranges of nfactor, vdscale, cdscd, and eta0 are initially determined.

[0081] (3) Figure 3 As shown, experimental data of enhanced GaN HEMT devices under different irradiation doses were imported into the TID-HEMT image fitting interface of the parameter extraction software, and I was plotted. D -V DS For the curve, select manual fitting and adjust the thermal resistance R. th The threshold voltage temperature dependence coefficient kt1, the electron mobility temperature dependence coefficient ute, the channel length modulation coefficient lambda, the electron saturation velocity vsat, the electron velocity saturation coefficient thesat, and the electron saturation velocity temperature dependence coefficient at are observed and compared. D -V DS The model's predicted values ​​and experimental values ​​in the image are used to preliminarily determine R. th The range of values ​​for kt1, ute, lambda, vsat, thesat, and at.

[0082] Step 7. Globally fit and compare the model prediction curve with the experimental test curve in real time to find the optimal value of the parameter to be extracted.

[0083] like Figure 2As shown, in the parameter setting interface of the parameter extraction software TID-HEMT, input the upper and lower bounds of the parameters to be extracted determined in step 6, select start fitting, and find the optimal values ​​of all parameters to be extracted by comparing the model prediction curve and the experimental test curve in real time through global fitting.

[0084] Step 8. Fitting complete. Observe and export the fitting results, specifically: fitting data and fitting parameters.

[0085] When the error between the model prediction and the experimental test value is less than the fitting accuracy or the maximum number of fitting iterations is reached, a "Fitting End" prompt box will pop up. Observe and compare the model prediction curve and the experimental test curve displayed on the TID-HEMT image fitting interface of the parameter extraction software. If the difference between the two is large, return to the parameter setting interface to modify the fitting accuracy and the number of fitting iterations, and fit again until the model prediction curve and the experimental test curve meet the requirements, and the fitting is completed.

[0086] like Figure 4 As shown, in the data processing interface of the parameter extraction software TID-HEMT, select to export the fitted leakage current data I. D _fit and the parameter values ​​obtained from the fitting.

[0087] Based on the exported fitted data, it can be plotted Figure 5 and Figure 6 : Figure 5 This is a comparison of the model-predicted values ​​and experimental values ​​of the transfer characteristic curves of enhancement-mode GaN HEMT devices under different irradiation doses after fitting. The horizontal axis represents the gate-source voltage V. GS (V), with the vertical axis representing the leakage current I. D (mA), solid lines represent experimental test values, and hollow circles represent model prediction values; Figure 6 This is a comparison of the model-predicted values ​​and experimental values ​​of the output characteristic curves of enhancement-mode GaN HEMT devices under different irradiation doses after fitting. The horizontal axis represents the drain-source voltage V. DS (V), with the vertical axis representing the leakage current I. D (mA), solid lines represent experimental test values, and hollow circles represent model predicted values. Observation Figure 5 and Figure 6 It can be observed that the proposed method for establishing and extracting parameters for enhanced GaN HEMT devices considering the total dose effect has high parameter fitting accuracy and can accurately predict the changes in the electrical characteristics of enhanced GaN HEMT devices under the influence of the total dose effect. The model parameters derived from the method proposed in this invention can be used to write relevant circuit simulation files, predict the effect of the total dose effect on the performance of GaN-based integrated circuits, provide technical support for subsequent total dose irradiation-resistant design, and effectively shorten the R&D cycle and reduce R&D costs.

[0088] The above description is merely a description of preferred embodiments of the present invention and is not intended to limit the scope of the present invention in any way. Any changes or modifications made by those skilled in the art based on the above disclosure shall fall within the protection scope of the claims.

Claims

1. A method for establishing an IV model and extracting parameters for an enhanced GaN HEMT device, characterized in that, include: Step 1: Experimentally measured the transfer characteristic curves and output characteristic curves of the enhanced GaN HEMT under different irradiation doses; Step 2: Based on the ASM-HEMT model, establish an enhanced GaN HEMT device IV model that considers the total dose effect, organize the model parameters, and classify them into: structural parameters and parameters to be extracted; Step 3: Develop corresponding parameter extraction software; Step 4: Filter the experimental data obtained in Step 1 and import it into the parameter extraction software; Step 5: Input the structural parameters of the enhanced GaN HEMT into the parameter extraction software, and set the fitting accuracy and the number of fitting iterations; Step 6: Initially determine the value range of the parameters to be extracted; Step 7: Globally fit and compare the model's predicted curve with the experimental test curve in real time to find the optimal value of the parameter to be extracted; Step 8: Fitting complete. Observe and export the fitting results, specifically: fitting data and fitting parameters; In step 2, based on the ASM-HEMT model, the enhancement-mode GaN HEMT device IV model considering the total dose effect is established as follows: （1） Among them, I D Where w is the leakage current, l is the gate width, and C is the gate length. g V is the effective conductance between the gate and the two-dimensional electron gas. th ψ is the thermal voltage, λ is the channel length modulation coefficient; DS = ψ D – ψ S , ψ m = (ψ D + ψ S ) / 2, ψ D With ψ S These are the surface potentials of the drain and source, respectively; μ eff,sat To account for mobility degradation and velocity saturation effects, the effective electron mobility is compared with the low-field mobility μ0 and the first-order mobility attenuation coefficient μ. a Second-order mobility attenuation coefficient μ b And related to the electron velocity saturation coefficient thesat; ΔV TID The threshold voltage drift caused by the total dose effect is expressed as follows: （2） Where the negative sign indicates the negative shift in the threshold voltage of the enhancement-mode GaN HEMT device caused by the total dose effect, q is the electron charge, and N e c represents the concentration of free electrons near the p-GaN cap layer / AlGaN barrier layer interface within the p-GaN cap layer. p c n e p e n These are the hole capture coefficient, electron capture coefficient, hole emission rate, and electron emission rate, respectively, with hole emission rate e0. p and electron emission rate e n Both are related to the hole trap energy level E introduced by total dose irradiation within the p-GaN cap layer. t Related to; P t N represents the hole trap density within the p-GaN cap layer near the p-GaN cap layer / AlGaN barrier layer interface after total dose irradiation. h P represents the concentration of radiation-induced and escape recombination holes in the p-GaN cap layer. t With N h They can be represented as: （3） （4） Where P0 is the hole trap density in the p-GaN cap layer before total dose irradiation, DOSE is the total irradiation dose, and t cap denoted as p-GaN cap layer thickness; a and b are the total dose irradiation-induced hole trap density parameters, where b is related to the formation process of irradiation-induced hole traps; g0 is the number of electron-hole pairs generated in the p-GaN cap layer per unit dose irradiation; f y The hole yield for escape recombination is related to the electric field strength inside the p-GaN cap layer; In step 6, the specific method for initially determining the value range of total dose irradiation-related parameters is as follows: (1) The energy level positions and densities of hole traps in p-GaN under different irradiation doses were measured using deep-level transient spectroscopy. The hole trap density P0 in the p-GaN cap layer before total dose irradiation, the hole trap density parameters a and b induced by total dose irradiation, and the hole trap energy level E introduced in the p-GaN cap layer by total dose irradiation were preliminarily determined. t The range of values ​​for; (2) Through the gate-source voltage V GS With p-GaN cap thickness t cap The electric field intensity E inside the p-GaN cap layer is estimated by the ratio of the two values, using the empirical formula f. y = [(|E| + E1) / (|E| + E2)] m E1 = 10 -6 MV / cm, E2 = 0.2 MV / cm, m = 0.9, preliminarily determine the hole production f of escape recombination. y The range of values ​​for .

2. The method for establishing and extracting parameters of an enhanced GaN HEMT device IV model as described in claim 1, characterized in that, In step 1, an average energy of 1.25 MeV is used. 60 Irradiation tests were conducted using a Co gamma-ray source; the irradiation dose rate was set to 50 rad(Si) / s, and irradiation dose points were selected at 0 M rad(Si), 0.3 M rad(Si), 0.6 M rad(Si), and 1 M rad(Si). Two enhancement-type GaN HEMT devices from the same batch were used at the same irradiation dose point to ensure that at least two sets of data were compared and analyzed under each radiation condition.

3. The method for establishing and extracting parameters of an enhanced GaN HEMT device IV model as described in claim 2, characterized in that, In step 1, a low-voltage DC power supply is used to apply an on-state bias to the enhancement-mode GaN HEMT device, i.e., V GS = 3V, V DS = 0.5 V, where V GS V is the gate-source voltage. DS This is the drain-source voltage.

4. The method for establishing and extracting parameters of an enhanced GaN HEMT device IV model as described in claim 3, characterized in that, In step 2, the structural parameters are specifically: gate length l, gate width w, gate index nf, and p-GaN cap thickness t. cap and the thickness t of the AlGaN barrier layer bar ; The parameter to be extracted is specifically: escaped composite hole production f. y Hole trap density P0 in the p-GaN cap layer before total dose irradiation; hole trap density parameters a and b induced by total dose irradiation; hole trap energy level E introduced by total dose irradiation in the p-GaN cap layer. t Resistance temperature detector (RTD) R th Threshold voltage V off Threshold voltage temperature dependence coefficient kt1, subthreshold slope coefficient nfactor, drain-source voltage scaling coefficient vdscale related to drain-induced barrier reduction effect, drain voltage-induced subthreshold slope change coefficient cdscd, drain-induced barrier reduction effect coefficient eta0, effective drain voltage exponential coefficient delta, electron mobility temperature dependence coefficient ute, low-field mobility μ0, first-order mobility decay coefficient μ a Second-order mobility attenuation coefficient μ b The channel length modulation coefficient lambda, the electron saturation velocity vsat, the electron velocity saturation coefficient thesat, and the temperature dependence coefficient at of the electron saturation velocity.

5. The method for establishing and extracting parameters of an enhanced GaN HEMT device IV model as described in claim 4, characterized in that, In step 3, the parameter extraction software TID-HEMT is developed using Matlab App Designer.

6. The method for establishing and extracting parameters of an enhanced GaN HEMT device IV model as described in claim 5, characterized in that, In step 4, two sets of test data at the same irradiation dose point are compared horizontally. If a large difference is found between the two, the test data at different irradiation dose points are compared vertically. The trend of the curve changing with the increase of irradiation dose is observed, individual abnormal abrupt curves are screened out, and appropriate curve data is selected and imported into the parameter extraction software TID-HEMT.

7. The method for establishing and extracting parameters of an enhanced GaN HEMT device IV model as described in claim 6, characterized in that, In step 5, the fitting accuracy is set to 10. -20 The fitting iteration count was 100.

8. The method for establishing and extracting parameters of an enhanced GaN HEMT device IV model as described in claim 7, characterized in that, In step 6, the specific method for initially determining the value range of other parameters is as follows: (1) In the TID-HEMT image fitting interface of the parameter extraction software, import the experimental data of the enhanced GaN HEMT device under different irradiation doses, and plot I D -V GS Curve selection: Manual fitting; Adjust threshold voltage V. off Low-field mobility μ0, first-order mobility attenuation coefficient μ a and the second-order mobility attenuation coefficient μ b By observing and comparing I D -V GS The model's predicted values ​​and experimental values ​​in the image are used to preliminarily determine V. off μ0, μ a and μ b The range of values ​​for; (2) In the TID-HEMT image fitting interface of the parameter extraction software, import the experimental data of the enhanced GaN HEMT device under different irradiation doses, and plot Log(I D )-V GS Curve selection: Manual fitting; Adjust threshold voltage V. off The following parameters were considered: subthreshold slope coefficient nfactor, drain-source voltage scaling factor vdscale related to the drain-induced barrier reduction effect, drain voltage-induced subthreshold slope change coefficient cdscd, and drain-induced barrier reduction effect coefficient eta0. These parameters were compared by observing Log(I... D )-V GS The model's predicted values ​​and experimental values ​​in the image are used to re-determine V. off The range of values ​​for nfactor, vdscale, cdscd, and eta0 is initially determined. (3) Import the experimental data of the enhanced GaN HEMT device under different irradiation doses into the TID-HEMT image fitting interface of the parameter extraction software, and plot the I... D -V DS For the curve, select manual fitting and adjust the thermal resistance R. th The threshold voltage temperature dependence coefficient kt1, the electron mobility temperature dependence coefficient ute, the channel length modulation coefficient lambda, the electron saturation velocity vsat, the electron velocity saturation coefficient thesat, and the electron saturation velocity temperature dependence coefficient at are observed and compared. D -V DS The model's predicted values ​​and experimental values ​​in the image are used to preliminarily determine R. th The range of values ​​for kt1, ute, lambda, vsat, thesat, and at.

9. The method for establishing an enhanced GaN HEMT device IV model and extracting parameters as described in claim 8, characterized in that, In step 7, in the parameter setting interface of the parameter extraction software TID-HEMT, the upper and lower bounds of the values ​​of the parameters to be extracted determined in step 6 are entered, and the fitting is started. The optimal values ​​of all parameters to be extracted are found by comparing the model prediction curve and the experimental test curve in real time through global fitting.

10. The method for establishing and extracting parameters of an enhanced GaN HEMT device IV model as described in claim 9, characterized in that, In step 8, when the error between the model prediction value and the experimental test value is less than the fitting accuracy or the maximum number of fitting iterations is reached, a "Fitting End" prompt box pops up. Observe and compare the model prediction curve and the experimental test curve displayed on the TID-HEMT image fitting interface of the parameter extraction software. If the difference between the two is large, return to the parameter setting interface to modify the fitting accuracy and the number of fitting iterations, and fit again until the model prediction curve and the experimental test curve meet the requirements, and the fitting is completed.

11. The method for establishing and extracting parameters of an enhanced GaN HEMT device IV model as described in claim 10, characterized in that, In step 8, in the TID-HEMT parameter extraction software data processing interface, select to export the fitted leakage current data I. D _fit and the parameter values ​​obtained from the fitting.

Images