Analysis of Defects and Electrical Characteristics of Variable-Temperature Proton-Irradiated 4H-SiC JBS Diodes

Abstract

The defects and electrical characteristics of 4H-SiC JBS diodes irradiated by 2 MeV protons under irradiation temperatures of 100–400 K were studied. Forward and reverse current–voltage (I–V), capacitance–voltage (C–V), and deep-level transient spectroscopy (DLTS) measurements were performed to study the changes in the characteristics of the device before and after variable-temperature proton irradiation. As the irradiation temperature increased from 100 to 400 K, the on-resistance decreased from 251 to 204 mΩ, and the carrier concentration gradually increased. The reverse current–voltage experiment results showed that the leakage current increased after proton irradiation at each irradiation temperature compared to before irradiation. The DLTS spectra analyses showed that proton irradiation mainly introduced a carbon vacancy related to the Z1/2 center (E_0.68 and E_0.72), which may have been the main reason for the changes in the forward and reverse electrical characteristics. The intensity of the DLTS spectrum decreased with the increasing irradiation temperature, indicating that the concentration of defects gradually decreased, due to the increase in the radius of the recombination of a vacancy with a related interstitial atom.

1. Introduction

Silicon carbide is an attractive semiconductor material for application in power devices and integrated circuits due to its wide band gap, high electron drift velocity, high breakdown voltage, and high thermal conductivity [1,2,3]. Silicon carbide with irradiation hardness has recently attracted widespread attention because of its important applications in the field of next-generation spacecraft and advanced nuclear reactors [4,5,6,7,8]. The performance of these 4H-SiC devices when implemented in the aerospace field will be significantly affected by the complex irradiation environment of space, with different types of irradiation sources and changes in ambient temperature. For example, semiconductor devices operating under high radiation and extreme temperatures (77 K) need to be investigated for the US ERI 2.0, and semiconductor devices located in the dark region of the moon might face irradiation and extreme low-temperature environments in lunar exploration projects. Research on the variable-temperature irradiation environments of space, including extreme low-temperature environments, is very meaningful [9,10,11]. Many studies have been carried out on the irradiation damage effects of electrons [12,13,14], protons [15,16], neutrons [17,18], and alpha particles on silicon carbide devices [19]. However, most of these studies have focused on the effects of irradiation at room temperature.

Silicon carbide devices have a wide range of requirements under irradiation and variable-temperature environments. Some researchers have conducted work in the field of variable-temperature irradiation. For example, research carried out on 15 MeV proton-irradiated Schottky diodes at elevated temperatures showed that the irradiation resistance of the diodes decreased from 9.5 × 10⁷ to 300 Ω as the irradiation temperature increased from 23 to 500 °C [20]. The study of 1700 V Schottky diodes irradiated by 0.9 MeV electrons at a high temperature also demonstrated that the resistance of the diodes decreased from 1 × 10⁶ Ω to 1 Ω as the irradiation temperature increased from 23 to 500 °C [21]. Shaomin Wang et al. reported that 4H-SiC Schottky barrier diodes were irradiated by 6 MeV Au ions at 77 K and indicated that series resistance and effective impurity concentration were recovered at 5 × 10¹⁵ ions/cm² [9]. The increase in the effective impurity concentration was due to the increase in the compensation role of the Au²⁺ ions when the concentration of the secondary electron increased with the increase in the irradiation fluence. To the best of our knowledge, there are few studies on low-temperature irradiation for 4H-SiC junction barrier Schottky (JBS) diodes. The defects and reverse current–voltage characteristics of silicon carbide devices irradiated at low temperatures are not clearly understood. When semiconductor devices are implemented in an aerospace environment that is rich in irradiation or has an extremely low temperature, the electronic characteristics might change. To promote the application of 4H-SiC power devices in aerospace, it is necessary to study the defects and electrical characteristics of 4H-SiC JBS diodes irradiated at different temperatures.

The main advantages of JBS devices are Schottky-like on-state and off-state characteristics similar to PiN devices, including a large forward current, a high breakdown voltage, and a low reverse breakdown current. In the present study, the effect of the temperature with 2 MeV protons on the characteristics of 4H-SiC JBS diodes was examined at irradiation temperatures ranging from 100 to 400 K and a fluence of 1 × 10¹³ cm⁻². Current–voltage (I–V), capacitance–voltage (C–V), and deep-level transient spectroscopy (DLTS) measurements were carried out on Schottky barrier diodes to study the change in the performance of the devices before and after proton irradiation.

2. Experimental Details

The 4H-SiC diode dies based on multiple field-limiting rings (M-FLR) were prepared as the test samples. The structure of the JBS diode is shown in Figure 1a. An N-type drift epilayer with a doping concentration of 8 × 10¹⁵ cm⁻³ was employed, and the thickness of the N-type drift epilayer was 11 μm. In the active region, the width (W) and depth (X_j) of the P⁺ rings were 2 μm and 0.5 μm, and the spacing of the adjacent P⁺ rings (S) was 2 μm. The anode of all the diodes consisted of a Schottky contact at the W/N−SiC interface and a thick AlSi metallic electrode. All the 1200 V/10 A 4H-SiC JBS samples were manufactured in the same batch of processes. The fabricated 4H-SiC JBS dies are shown in Figure 1b. The irradiation experiments were performed on a 2 MeV proton accelerator at Beijing Normal University. The temperature was regulated using a temperature regulation target stage that was customized by American Instec. It could achieve accurate temperature control in a range of 100 to 873 K with liquid nitrogen [22]. For reliable temperature management, computer software controlled the heating module and liquid nitrogen loop behind the sample stage (the external liquid nitrogen pump adjusted the liquid nitrogen flow rate). The accuracy of maintaining the temperature of the sample during irradiation was ± 1K. The samples were irradiated with 2 MeV protons with a fluence of 1 × 10¹³ cm⁻². The samples were irradiated at 100 K, 200 K, 300 K, and 400 K, respectively. There were three samples for each irradiation temperature. In addition, two experiments were conducted, and the experimental results that were consistent with the trend of the average were selected in this paper.

The I–V measurements were carried out using a curve tracer CT-3200 at room temperature. The on-resistance values were obtained from forward I–V measurements. The initial concentration of electrons in the base layer of unirradiated diodes determined from C–V measurements was 7.18 × 10¹⁵ cm⁻³. The I–V characteristics and C–V characteristics of all samples were measured at room temperature. Irradiation defects in the diodes were measured utilizing a Phys Tech FT1030 (high energy resolution analysis deep-level transient spectroscopy) system before and after irradiations. The DLTS spectrometer was measured with a temperature scanning from 400 to 50 K. The DLTS scans were performed with a reverse bias voltage of −20 V, a pulse voltage (V_p) of −0.1 V, and a fill pulse of 19.9 V for 4H-SiC diodes. The test period (T_w) was 19.2 ms, and a fill pulse width (t_p) of 0.1 ms was chosen to ensure complete trap filling.

3. Results and Discussion

3.1. I–V and C–V Characteristics at Room Temperature

Figure 2 shows the forward current–voltage characteristics of the diodes measured at room temperature after irradiation with protons at a fluence of 1 × 10¹³ cm⁻². This figure shows that the applied voltage mainly dropped in the base of the diode, which indicates an increase in on-resistance [23] after proton irradiation. Figure 3 shows the relationship between on-resistance and irradiation temperature. For the irradiation temperatures of 100, 200, 300, and 400 K, the on-resistance values were 251, 239, 221, and 204 mΩ, respectively. The on-resistance gradually decreased as the irradiation temperature increased from 100 to 400 K. In this experiment, the on-resistance gradually decreased as the irradiation temperature increased from 300 K to 400 K, which is consistent with the trend of decreasing resistance with increasing irradiation temperature reported by Kozlovski et al. for high-temperature proton irradiation [24]. As the irradiation temperature increased from 100 to 300 K, the results illustrated in Figure 3 show that the on-resistance gradually decreased. Compared with the initial resistance of 98 mΩ before irradiation, the on-resistance after irradiation at 100 K increased by about three times to 251 mΩ. The reason may be that some defects induced by proton irradiation can trap or capture free carriers, resulting in a decrease in carrier concentration.

Figure 4b shows the plot of (1/C²) as a function of reverse voltage measured at room temperature. The free carrier concentration N_D and the barrier height Φ_c–v can be calculated as follows [9]:

$$\frac{1}{{\mathrm{C}}^2}=\frac{2}{{\mathrm{q}\mathsf{\epsilon }\mathrm{N}}_{\mathrm{D}}{\mathrm{A}}^2}\left({\mathrm{V}}_{\mathrm{bi}}-\mathrm{V}-\frac{\mathrm{kT}}{\mathrm{q}}\right)$$

(1)

$${\mathrm{N}}_{\mathrm{D}}=-\frac{2}{{\mathrm{q}\mathsf{\epsilon }\mathrm{A}}^2}\frac{1}{\frac{\mathrm{d}\left({\mathrm{C}}^{-2}\right)}{\mathrm{dV}}}$$

(2)

$${\mathsf{\Phi }}_{\mathrm{c}\text{-}\mathrm{v}}={\mathrm{V}}_{\mathrm{bi}}+\frac{\mathrm{kT}}{\mathrm{q}}+\frac{\mathrm{kT}}{\mathrm{q}}\ln \frac{{\mathrm{N}}_{\mathrm{C}}}{{\mathrm{N}}_{\mathrm{D}}}$$

(3)

where ε is the dielectric constant of the 4H-SiC semiconductor material, C is capacitance, V_bi is the built-in voltage potential, k is Boltzmann’s constant, N_C is the effective charge density of states in the conduction band, T is temperature, kT/q is the thermal voltage, and A is the area of the device.

The capacitance of the sample decreased with a decrease in irradiation temperature, as shown in Figure 4a. At a reverse bias of −3 V, the capacitances obtained before and after irradiation at an irradiation temperature of 100 K were 942 and 684 pF, respectively. The decrease in the capacitance of the irradiated diodes may have been the result of the increase in the width of the semiconductor depletion layer due to the introduction of defects with acceptor states in the band gap, which reduced the free carrier concentration of the devices after irradiation. Figure 5 shows the curve of N_D and Φ_c-v with irradiation temperature. The Schottky barrier height Φ_c-v can be obtained from Equation (3). The value of the barrier height obtained from the C–V measurements before irradiation was 1.75 eV. The barrier height values were 1.76, 1.78, 1.82, and 1.77 eV at irradiation temperatures of 100, 200, 300, and 400 K, respectively. It can be seen that the barrier height changed slightly after irradiation. The free carrier concentration N_D was determined from the slope of the plots, using the standard Equation (2). The free carrier concentrations of the irradiated samples were 3.78 × 10¹⁵, 4.47 × 10¹⁵, 4.88 × 10¹⁵, and 4.97 × 10¹⁵ cm⁻³ at irradiation temperatures of 100, 200, 300, and 400 K, respectively. The defects with acceptor states in the band gap had the effect of compensating the free carriers, resulting in a decrease in the carrier concentration.

The change of the on-resistance could be caused by the change in the carrier concentration. According to the theoretical knowledge of semiconductor physics, the conductivity of semiconductor materials is a positive correlation function of carrier concentration and mobility, and resistance and conductivity are inversely proportional. On the basis of previous research [9,24,25], it was assumed that the effect of irradiation on mobility in this work was negligible. Therefore, the change of free carrier concentration could explain the change of on-resistance.

The reverse leakage current curves obtained before and after proton irradiation at different irradiation temperatures are presented in Figure 6. Compared to the leakage current value before irradiation, the leakage current increased after irradiation, as shown in Figure 6, and the defects induced by irradiation are suggested to be the main reason for this [26]. According to previous studies, defects in semiconductors can cause an increase in the surface electric field and tunneling electric field, which leads to the enhancement of the tunneling effect and barrier-lowering effect and finally causes an increase in the tunneling current [23,26,27,28,29,30,31]. However, the increase in the leakage current is small after proton irradiation, especially in the high-voltage region, which may be due to the small change in the barrier height derived from the experimental results and the injection annealing effect of the proton irradiation of SiC [32,33].

3.2. DLTS Analysis

Figure 7a shows the DLTS spectra of the 4H-SiC JBS diodes at irradiation temperatures of 100, 200, 300, and 400 K. The activation energy of each defect and the capture cross-section was determined from the Arrhenius plot in Figure 7b [34]. The electronic properties are tabulated in Table 1.

The level labeled E_0.39 has been attributed to a silicon vacancy (V_Si) [35], and the peak intensity decreased with increasing irradiation temperature. E_0.42 was the defect introduced by proton irradiation, which has been attributed to a silicon vacancy (V_Si) [36,37], and the intensity of the peak decreased gradually as the irradiation temperature increased from 100 to 400 K.

Previous work has shown that E_0.68 and E_0.72 are well-known Z1/2 centers associated with carbon vacancy [38,39,40], and the signal intensity of the peaks decreased with increasing irradiation temperature. Z1/2 is a well-known deep level that is assigned to a transition between the double-negative and neutral charge states of a carbon vacancy [40]. In particular, E_0.72 disappeared at the irradiation temperature of 300 K, and E_0.68 had a broad peak. The broad and asymmetrical peak may have been due to the presence of several states or defects with closely spaced emission rates, based on previous results reported in the literature [31,41]. It can be seen that the negative peak appeared in the low-temperature region, which was due to minority carrier trapping [42].

It can be seen from Figure 7a that the peaks of E_0.39, E_0.42, and E_0.68 in the DLTS spectra gradually decreased as the irradiation temperature increased from 100 to 400 K, which indicates that the number of defects gradually decreased. Since the irradiation temperature was variable in this experiment, the increase in the defect concentration was closely related to the irradiation temperature. As the irradiation temperature increased, the radius of the recombination of a vacancy with a genetically related interstitial atom became larger [43]. Therefore, the fraction of vacancies that escaped recombination and created deep acceptor levels may have significantly decreased with the increasing irradiation temperature. Compared with other irradiation temperatures, the intensity of the DLTS spectra was lower at the irradiation temperature of 400 K. These deep-level defects could act as the capture centers of carriers, leading to a decrease in the carrier concentration and thus an increase in the on-resistance, and they are also the main reason for the increase in the reverse leakage current.

Table 1. Defect parameters obtained from the DLTS measurements.

Irradiation Temperature	Defect Label	ET (eV)	σn (cm−3)	Identity	Reference
Before irradiation	E0.39	EC − 0.39	9.91 × 10−14	VSi	[35]
	E0.68	EC − 0.68	5.76 × 10−15	Z1/2	[38,44]
100 K	E0.39	EC − 0.39	2.89 × 10−13	VSi	[35]
	E0.42	EC − 0.42	3.30 × 10−15	VSi	[36,37]
	E0.68	EC − 0.68	1.04 × 10−14	Z1/2	[38,44]
	E0.72	EC − 0.42	1.67 × 10−15	Z1/2	[39,40]
200 K	E0.39	EC − 0.39	1.53 × 10−13	VSi	[35]
	E0.42	EC − 0.42	4.72 × 10−15	VSi	[36,37]
	E0.68	EC − 0.68	9.89 × 10−15	Z1/2	[38,44]
	E0.72	EC − 0.72	1.80 × 10−15	Z1/2	[39,40]
300 K	E0.39	EC −0.39	1.02 × 10−13	VSi	[35]
	E0.42	EC − 0.42	1.67 × 10−15	VSi	[36,37]
	E0.68	EC − 0.68	3.15 × 10−15	Z1/2	[38,44]
400 K	E0.39	EC − 0.39	1.11 × 10−13	VSi	[35]
	E0.42	EC − 0.42	2.51 × 10−15	VSi	[36,37]
	E0.68	EC − 0.68	4.90 × 10−15	Z1/2	[38,44]
	E0.72	EC − 0.72	1.92 × 10−15	Z1/2	[39,40]

Table 1. Defect parameters obtained from the DLTS measurements.

Irradiation Temperature	Defect Label	ET (eV)	σn (cm−3)	Identity	Reference
Before irradiation	E0.39	EC − 0.39	9.91 × 10−14	VSi	[35]
	E0.68	EC − 0.68	5.76 × 10−15	Z1/2	[38,44]
100 K	E0.39	EC − 0.39	2.89 × 10−13	VSi	[35]
	E0.42	EC − 0.42	3.30 × 10−15	VSi	[36,37]
	E0.68	EC − 0.68	1.04 × 10−14	Z1/2	[38,44]
	E0.72	EC − 0.42	1.67 × 10−15	Z1/2	[39,40]
200 K	E0.39	EC − 0.39	1.53 × 10−13	VSi	[35]
	E0.42	EC − 0.42	4.72 × 10−15	VSi	[36,37]
	E0.68	EC − 0.68	9.89 × 10−15	Z1/2	[38,44]
	E0.72	EC − 0.72	1.80 × 10−15	Z1/2	[39,40]
300 K	E0.39	EC −0.39	1.02 × 10−13	VSi	[35]
	E0.42	EC − 0.42	1.67 × 10−15	VSi	[36,37]
	E0.68	EC − 0.68	3.15 × 10−15	Z1/2	[38,44]
400 K	E0.39	EC − 0.39	1.11 × 10−13	VSi	[35]
	E0.42	EC − 0.42	2.51 × 10−15	VSi	[36,37]
	E0.68	EC − 0.68	4.90 × 10−15	Z1/2	[38,44]
	E0.72	EC − 0.72	1.92 × 10−15	Z1/2	[39,40]

Figure 8 illustrates the band diagram and capture of carriers in the 4H-SiC diodes after proton irradiation. The defects produced by proton irradiation were silicon vacancies (E_0.39 and E_0.42) and carbon vacancies (E_0.68 and E_0.72), which were located in the upper half of the band gap. The acceptor defect produced by irradiation could capture carriers from the conduction band, which led to a reduction in the carriers in the conduction band.

3.3. Discussion

The reason for the large number of vacancies and interstitials (Frenkel pairs) was the atomic displacement caused by the collision of protons incident on the 4H-SiC diodes with silicon and carbon atoms in their semiconductor materials. The separation distances between the interstitials and vacancies were generally short, since long replacement collision sequences rarely occur. Figure 9 is a schematic diagram of the defect location. Interstitials or vacancies may migrate in the crystal lattice during variable-temperature proton irradiation. When interstitial defects migrate from one site to another site adjacent to a vacancy, they may recombine with the vacancies. The time (t) required for an interstitial to migrate to adjacent points can be expressed as [45,46]:

$$\mathrm{t}=\frac{1}{\mathsf{\nu }}{e}^{{\mathrm{E}}_1/{\mathrm{kT}}_1}$$

(4)

where E₁ is the minimum potential barrier required for vacancy migration, ν is the vibration frequency of the atoms adjacent to the vacancies, and T₁ is the temperature during irradiation. It can be seen that as the temperature increases, the vibration frequency of the atoms increases. The time required for vacancy migration becomes short, and the probability of vacancy recombination with interstitial atoms becomes high.

The number of primary defects (vacancies and interstitial atoms) produced by proton irradiation at a fluence of 1 × 10¹³ cm⁻² was almost the same in all samples. The possibility of recombination was greater at a high irradiation temperature, which resulted in more interstitials and vacancies recombining [47]. Therefore, the defect concentration was lower at the irradiation temperature of 400 K, and the defect concentration was higher at the irradiation temperature of 100 K. The defect concentration in the semiconductor device decreased as the irradiation temperature increased from 100 to 400 K, which was consistent with the results of the DLTS measurements.

4. Conclusions

In this paper, the defects and electrical characteristics of 4H-SiC JBS diodes were studied after proton irradiation under irradiation temperatures of 100–400 K. The experimental results showed that as the irradiation temperature increased from 100 to 400 K, the on-resistance decreased from 251 to 204 mΩ, and the free carrier concentration increased from 3.78 × 10¹⁵ to 4.97 × 10¹⁵ cm⁻³. The reverse leakage current increased after proton irradiation at each irradiation temperature compared to before irradiation. The changes in the forward and reverse properties were caused by silicon vacancy and carbon vacancy defects generated after proton irradiation. The DLTS spectra showed that the Z1/2 defects (E_0.68 and E_0.72) after variable temperature proton irradiation were dominant, and the intensity of the peaks increased gradually with the decreasing irradiation temperature. As the irradiation temperature increased from 100 K to 400 K, the concentration of carbon vacancies also decreased gradually due to the increased recombination radius of the carbon vacancies with related interstitial atoms. The results of this work could facilitate the application of SiC power devices in an aerospace environment that is rich in irradiation or in an extreme low-temperature environment.