How to Use SILVACO Software to Simulate Diode Characteristics

Abstract: In order to shorten the reverse recovery time, traditional fast recovery diodes usually use electron irradiation to reduce the minority carrier lifetime in the base region. However, electron irradiation reduces the device's reverse recovery time and also makes the on-state voltage Increased. In this paper, the dual-proton irradiation localized lifetime control method is used to simulate the characteristics of the diode using SILVACO software. The different locations of the local low life region in the diode are discussed. The reverse recovery current of the Fast Recovery Diode is reversed. The softness factor and the effect of the on-state voltage drop provide a theoretical basis for the actual production of fast recovery diodes.

1 Introduction

In high-voltage and large-current loops, the switching devices mainly use self-shutdown power semiconductor devices such as GTOs, MOSFETs, IGBTs, or thyristors that use external current-triggered turn-off. These switching devices require a power in parallel with them. The fast Recovery Diode is mainly used to form a loop with the load through the excess current in the circuit load, reduce the storage and disappearance time of the charge on the capacitor in the switching device, and weaken the high voltage induced by the parasitic inductance caused by the reverse of the load current. Delayed the aging of switching devices and increased device lifetime [1].

Therefore, this requires that the fast recovery diode softness factor S is large, the reverse recovery time trr is short, and the reverse recovery current Irr is small. For a long time, overall life-cycle control technologies have been adopted, such as gold-expanding, platinum-enlarging, and electron-irradiation technologies. However, while these techniques reduce the device's reverse recovery time, they also increase the turn-on voltage drop and reverse leakage current. Big. With the deepening of research, the local area life control technology has received extensive attention as a new life-cycle control technology. Local area life control technology is the frontier of today's international life-cycle control technology [2].

The local area life control technology is selectively introduced into the recombination center within the FRD to form a local low life area. With ion implantation of hydrogen ions or helium ions, high concentrations of defects occur at different depths (near the end of the range) at different energies, resulting in a localized low-life region [3, 4]. The local area life control technology can effectively solve the conflict between the reverse recovery time and turn-on voltage drop, reverse recovery time and leakage current of the fast recovery diode. In this paper, the dual-proton irradiation simulation of PIN diodes is performed using SILVACO software. The influence of the location of the low-life area under different energy conditions on the static and dynamic characteristics of PIN diodes is discussed.

2 PIN diode reverse recovery characteristics principle analysis

At the time t0, the reverse conduction voltage is applied to the forward conduction PIN diode. The current and voltage waveforms are shown in Fig. 1, and the plasma distribution curve in the base period is shown in Fig. 2. The reverse recovery process is divided into the following stages.

Phase 1 (t0-t1): The current flowing through the diode changes with an almost constant rate of change di/dt. The magnitude of di/dt depends on the applied reverse voltage VDC and the parasitic inductance Li in the circuit. Because di/dt is very high, the time interval between t0 and t1 is very short, so when the current passes 0 (t1), there is a large amount of unbalanced carriers in the diode.

Phase 2 (t1~t2): After the current passes 0, the excess carriers in the n− region still keep the diode in conduction, and the voltage on the diode is still very small. Thus current I(t) continues to increase at the same di/dt speed. The reverse current is maintained by extracting the n-zone non-equilibrium carriers. During the extraction of the non-equilibrium carriers, holes are discharged through the anode and electrons are discharged through the cathode, making the n-zones at both edges. The plasma concentration decays rapidly.

Stage 3 (t2 to t3): At t2, the plasma concentration at the pn-junction falls to 0, so a depletion layer can be formed at the pn-junction. Simply put, this stage is the process of depletion layer expansion. At t3, the diode voltage reaches VDC and di/dt drops to zero. The reverse recovery current reaches its maximum value and is referred to as the reverse recovery peak current Irr.

Phase 4 (t3 to t4): After t4, the non-equilibrium carrier concentration gradient at the boundary of the space charge zone will decrease. Therefore, the reverse current will decrease after t3, and the related current change rate is called recovery di/dt. Negative di/dt causes the inductor Li to build a negative voltage, which causes the diode to overvoltage.

Stage 5 (t5-t6): The depletion of the plasma causes the reverse current to drop to zero and the recovery di/dt decreases, so that the voltage on the diode falls back to VDC. At t6, the reverse recovery process ends.

The intersection of the connection between the reverse peak current Irr and 0.25 Irr on the time axis is referred to as time t5, and the time from the time t0 to time t5 is referred to as the reverse recovery time trr. T1 ~ t3 is the time when the non-equilibrium carriers stored in the base region are swept out by the reverse voltage, which is called the non-equilibrium carrier storage time ta. The time from t1 to t2 is called composite time tb. The method of characterizing the reverse recovery softness of a diode is usually expressed as the ratio of tb to ta, ie, S = tb/ta [5].

Diode main dynamic parameters: Irr, trr and S. From the previous analysis, we can see that to reduce Irr and trr, increase S at the same time, and minimize the forward voltage drop. The base region should have an ideal plasma distribution, ie the plasma concentration on the left side should be as low as possible and the plasma concentration on the right side should be as high as shown in Figure 2. The axial carrier lifetime distribution obtained by irradiation with two different energy protons as shown in FIG. 3 [6] allows the base region to have an ideal plasma distribution as shown in FIG. 2 .

3 Determination of structural parameters

The structure of FRD is shown in Fig. 4. The doping concentration of N-layer is 2×1014 cm-3, the width of N-layer is 90 μm, the depth of N+ region is 60 μm, and the doping concentration is 5×1019 cm-3. The junction depth in the P+ region is 50 μm and the doping concentration is 1×10 19 cm −3 .

The literature shows that bimodal proton irradiation is more beneficial to the improvement of the reverse recovery characteristics of FRD [7]. In this paper, two protons are used to form a bimodal trap region. One peak trap region is located in the N-region, one peak trap region is located in the P+ region, and the influence of the proton irradiation parameters on the FRD characteristics is simulated and analyzed.

4 Simulation Tools and Model Selection

The simulation tool used in this study is the ATLAS device simulator in Silvaco software. The physical models selected according to the research needs are the Shockley-Reed-Hall compound model (SRH); the narrowed model with restricted band when heavily doped (BGN); the Auger compound model with large infusion (AUGER); Mobility model for the effect of parallel electric field on carrier mobility (FLDMOB); Impact ionization model (IMPACT SELB). Among them, the Shockley-Reed-Hall compound model is:

(1)

Among them, ETRAP is the difference between the composite center energy level and the intrinsic Fermi energy level; TL is the lattice temperature at the Kelvin temperature; TAUN0, TAUPO are the lifetimes of electrons and holes, respectively.

5 Proton irradiation simulation analysis

The depth of proton irradiation depends on the energy of irradiation, and the radiation dose is controlled within the range of 1×1011 to 5×1014 cm-2 [8]. This paper simulates the characteristics of different diodes by irradiating the irradiated energy. The effect of radiation dose selected here is 1×1012 cm-2.

5.1 Effect of Radiation Depth on Soft Factor of Fast Recovery Diode

The softness factor is an important parameter to measure the diode reverse recovery characteristics. In order to achieve soft recovery of the diode, it is required to be at the end of reverse recovery (t3 to t6), and there are a large number of non-equilibrium carriers in the drift region near the cathode. The reverse recovery characteristic has a long tail, ie, the di/dt is small and the curve changes slowly.

The irradiation enters from the anode area. From the previous analysis, the first irradiation depth is set in the drift zone, ie, the irradiation depth is in the range of 50 to 140 μm, and the second irradiation depth is in the anode area, ie, irradiated. The depth is 0-50 μm.

Figure 5 shows the softness factor and the first case when the second depth is 10, 20, 30, 40, and 50 μm in the case where the test conditions are IF = 5A, VR = 30V, and di/dt = 200 A/μs. The depth of the sub-irradiation relationship. As can be seen from Figure 5, when the first irradiation depth is 120 μm, the second irradiation depth is 20 μm, and the softness factor is 2.06.

The P+ anode region is introduced into the recombination center by proton irradiation, which can reduce the anode injection efficiency. Proton irradiation in the N-drift region reduces the lifetime of the left side of the N-zone. Both of the proton irradiations reduce the plasma concentration on the left side of the drift region, and the carrier distribution tends to be idealized as shown in Fig. 2, which shortens the time interval (tb) from t1 to t3. Since the proton irradiation only reduces the lifetime of the left side of the N-zone, the lifetime of the right side remains high, so the time interval (ta) between t3 and t4 will not be shortened due to proton irradiation, so the softness The factor is larger.

5.2 Effect of Irradiation Depth on Reverse Recovery Peak Current of Fast Recovery Diodes

Figure 6 shows the reverse recovery peak current when the second depth is 10, 20, 30, 40, and 50 μm when the test conditions are IF = 10 A, VR = 100 V, and di/dt = 200 A/μs. The relationship between size and depth of first irradiation.

It can be seen from Fig. 6 that when the first irradiation depth is 60 μm and the second irradiation depth is 30 μm, the reverse recovery peak current is the smallest, which is 16.6 A. When the first irradiation depth is 120 μm and the second irradiation depth is 20 μm, the reverse recovery peak current is 32 A. The closer the two irradiation positions are to the PN junction, the smaller the stored charge amount in the drift region near the anode region side, and the number of carriers extracted on the anode side in the reverse recovery process is decreased, so that the reverse recovery peak current is reduced.

5.3 Effect of energy of proton irradiation on the pressure drop across the FRD

At a forward conduction current density of 100 A/cm2, the forward recovery voltage drop of fast recovery diodes corresponding to different proton irradiation depths was extracted. Figure 7 shows the second depths of 10, 20, 30, 40, and 50, respectively. In μm, the relationship between the forward voltage and the first irradiation depth.

When the first irradiation depth is 120 μm and the second irradiation depth is 20 μm, the diode's turn-on voltage is 1.18 V. It is far less than the forward pressure drop across the entire area that reduces life.

6 Conclusion

In this paper, the influence of dual proton irradiation on the FRD characteristics is studied. The following conclusions are drawn: Under the condition that the proton irradiation dose is not changed, the P+ anode region and the N-drift region are realized by adjusting the irradiation depth of proton irradiation. Introduce local low-life areas. Through analysis, the energy of diproton irradiation is 3.3 MeV and 0.8 MeV, respectively, and when the corresponding proton radiation irradiation depth is 120 μm and 20 μm respectively, the FRD has the best characteristics.