Schottky Diodes SDs on Undoped GaN Templates

SDs were fabricated on both undoped and doped GaN templates with varying SiN deposition time to study the effect of SiN coverage. The Si doped GaN layers were deposited to achieve reasonable capacitance for the DLTS measurement. 

Planar SDs were fabricated using standard photolithography on 0, 3,4 and 5 min SiN samples (undoped). Before metallization, all samples were cleaned in acetone, methanol, and deionized (DI) water in an ultrasonic bath, followed by boiling aqua regia cleaning for 20 min and 5 min DI water rinse. Ti/Al/Ti/Au (30/100/30/100 nm) ohmic contacts were deposited by e-beam and thermal evaporation, followed by a 60 s rapid thermal annealing (RTA) at 900 °C in nitrogen ambient. Finally, 200 pm diameter Ni/Au (30/120 nm) SDs were deposited by e-beam evaporation. The distance between SDs and ohmic contacts was 50 pm. 

Using Equations (6.1) and (6.2), we calculated the barrier height and ideality factor which are listed in Table 6.2. For the sample without the 

SiN nanonetwork, the barrier height is 0.76 eV. When the SiN deposition time is increased, the barrier height increases from 0.84 (3 min SiN ) to 1.13 eV (5 min SiN (). At the same time, the ideality factor reduces from 1.3 (no SiN ) to 1.06 (5 min SiN ) which indicates that the SDs are nearly ideal in samples grown with the SiN nanonetwork. Incidentally, this improved value is consistent with the work function of Ni (5.2 eV) and the electron affinity of GaN (4.1 eV). In the literature, a value of 1.099 eV (Ni) barrier height was achieved only after the GaN surface was treated with (NH S [12], which is known to passivate the surface defects albeit temporarily. Our results indicate that the Ni Schottky barrier height is very sensitive to the crystalline quality and the excess current 

Temperature dependent I-V measurements in the temperature range 300-500 K, which allowed determination of the activation energy, were also performed, and the results are listed in Table 6.2. The barrier heights calculated from the Richardson plot using Equation (6.2) are consistent with the room temperature values for each sample. However, the calculated effective Richardson constant is much smaller than the theoretically expected value which is commonly reported in the literature for GaN [2]. As we can see in Table 6.2, the effective Richardson constant is also related to the crystal quality, and obviously further work is required to shed light on the discrepancy between the measured and theoretical values endemic to GaN. 

---