Figure 4. Aging of black phosphorus FETs.

(a-c) I_d-V_g plot of BP devices with various capping schemes. The drain voltage V_d = −100 mV for all measurements. (a) FETs with thin capping (5 nm ALD-deposited Al₂O₃) experienced significant degradation, such as reduced ON/OFF ratio and the increase of gate hysteresis. (b) Thick-cap devices (25 nm ALD Al₂O₃) showed less aging effect. (c) Devices with double-layer capping (25 nm ALD Al₂O₃ followed by spin-casting of DuPont Teflon-AF fluoropolymer) showed the least aging effect within the same duration. (d–e) Statistical analysis of BP FET devices with different capping methods. The mean values of thin-, thick-, and double-cap devices are presented as red diamonds, blue squares, and green circles, respectively. Vertical bars indicate the range of one standard deviation. (d) The change of ON current (I_on) with respect to the values at Day 1. (e) Hysteresis in V_g between forward and reverse sweeps. For the two performance metrics shown here, the thin-cap FETs exhibited the largest fluctuations and the best stability is obtained in devices with double-layer capping. (f) I_on/I_off ratio of three selected devices. Thin capped device showed sharp degradation, resulting in I_on/I_off ratio of ~10× after a month. In contrast, I_on/I_off ratio of thick capped device was preserved to ~70% compared to Day 1. Double capped device showed the best preservation of I_on/I_off ratio, with negligible change after 79 days. Initial mobility values of the three selected devices are ~200 cm²/V-s as shown in Supplementary Note 9.