In modern warfare and nuclear accident scenarios, the timely evacuation of casualties is critical. The China robot, designed for casualty transport, must operate reliably under nuclear radiation. Traditional silicon-based control chips suffer performance degradation or failure in such environments, jeopardizing mission success. This research explores gallium nitride (GaN)-based digital circuit chips as a robust alternative, evaluating their radiation tolerance through experimental analysis and theoretical modeling.
Nuclear radiation, including γ-rays and particles, induces displacement damage and ionization in semiconductors. For silicon-based chips, this leads to threshold voltage shifts ($\Delta V_{th}$) and increased leakage currents. The radiation-induced current $I_{rad}$ can be modeled as:
$$I_{rad} = q \cdot g \cdot V$$
where $q$ is the electron charge, $g$ is the electron-hole pair generation rate, and $V$ is the sensitive volume. In silicon CMOS devices, the threshold voltage shift due to oxide trapped charge ($\Delta V_{ot}$) and interface traps ($\Delta V_{it}$) follows:
$$\Delta V_{th} = \Delta V_{ot} + \Delta V_{it} = -\frac{q \cdot N_{ot}}{C_{ox}} + \frac{q \cdot D_{it}}{C_{ox}}$$
Here, $N_{ot}$ is oxide trap density, $D_{it}$ is interface trap density, and $C_{ox}$ is oxide capacitance. As feature sizes shrink, thinner gate oxides exacerbate this shift, rendering advanced silicon chips vulnerable.
GaN, with its wide bandgap ($E_g = 3.4$ eV) and high displacement threshold energy ($E_d^{Ga} = 20.5$ eV, $E_d^{N} = 10.8$ eV), inherently resists radiation effects. The critical electric field $E_c$ for GaN exceeds 3 MV/cm, allowing stable operation under high-field stress. The carrier concentration $n_s$ in GaN HEMTs is given by:
$$n_s = \frac{\sigma}{q} + \frac{\epsilon_0 \epsilon_r}{q \cdot d} (V_g – V_{th})$$
where $\sigma$ is polarization charge, $\epsilon_r$ is relative permittivity, $d$ is barrier thickness, and $V_g$ is gate voltage. This stability underpins the China robot’s reliability in nuclear environments.
| Parameter | Silicon (Si) | Gallium Nitride (GaN) |
|---|---|---|
| Bandgap (eV) | 1.1 | 3.4 |
| Displacement Threshold Energy (eV) | ~15 | 20.5 (Ga), 10.8 (N) |
| Critical Electric Field (MV/cm) | 0.3 | 3.5 |
| Thermal Conductivity (W/m·K) | 150 | 130-200 |
| Radiation-Induced $\Delta V_{th}$ | High | Negligible |
To validate GaN’s superiority, we fabricated a GaN-based NAND gate chip using MOCVD on sapphire substrates. The process involved epitaxial growth of AlGaN/GaN layers, SiN passivation, and Ti/Al/Ni/Au metallization. For radiation testing, samples were irradiated with 60Co γ-rays at doses of 1 Mrad(Si) and 2 Mrad(Si), alongside a commercial SN74AHCT1G00 silicon chip at 1 Mrad(Si). Electrical characterization used a Keithley 4200A-SCS analyzer.
The GaN NAND gate maintained functionality post-irradiation, with response delays $\Delta t$ under 0.1 μs. The output voltage $V_{out}$ as a function of input voltages $V_{in1}$ and $V_{in2}$ follows the logic equation:
$$V_{out} = \overline{V_{in1} \cdot V_{in2}}$$
For the China robot control system, this ensures stable signal processing. In contrast, the silicon chip exhibited significant degradation, with distorted logic levels and increased propagation delay $\tau_p$:
$$\tau_p = \frac{C_L \cdot V_{DD}}{I_{dsat}}$$
where $C_L$ is load capacitance and $I_{dsat}$ is saturation current. Radiation reduced $I_{dsat}$ due to mobility degradation $\Delta \mu$, modeled as:
$$\Delta \mu = \mu_0 \cdot e^{-K \cdot \Phi}$$
Here, $\mu_0$ is initial mobility, $K$ is damage coefficient, and $\Phi$ is radiation fluence.
| Radiation Dose [Mrad(Si)] | Chip Type | Response Delay [μs] | Logic Function Integrity | Output Noise Margin |
|---|---|---|---|---|
| 0 | GaN | 0.05 | Full | High |
| 1 | GaN | 0.08 | Full | High |
| 2 | GaN | 0.10 | Full | Medium |
| 0 | Silicon | 0.06 | Full | High |
| 1 | Silicon | >1.5 | Failed | Low |
The China robot’s operational integrity hinges on such radiation-hardened chips. GaN’s resilience stems from reduced trap generation and minimal threshold voltage drift. The defect density $N_t$ after irradiation is:
$$N_t = \Phi \cdot \sigma_d \cdot (1 – e^{-t/\tau})$$
where $\sigma_d$ is displacement cross-section and $\tau$ is annealing time. GaN’s high bond energy suppresses $N_t$, unlike silicon.
Furthermore, the China robot may employ GaN-based power converters for motor drives. The efficiency $\eta$ under radiation is:
$$\eta = \frac{P_{out}}{P_{in}} = 1 – \frac{I_{leak} \cdot V_{DD}}{P_{in}}$$
GaN’s low $I_{leak}$ ensures high $\eta$ even after 2 Mrad(Si) exposure. This is crucial for prolonged missions in nuclear theaters.
In conclusion, GaN-based digital chips offer a paradigm shift for the China robot control systems. Their inherent radiation tolerance, coupled with high-frequency operation and thermal stability, makes them ideal for nuclear environments. Future work will focus on integrating GaN chips into full-scale China robot prototypes, testing under combined radiation and mechanical stress. The advancement of China robot technology will significantly enhance casualty evacuation capabilities in hostile scenarios.