Effects of NH/sub 3/ plasma passivation on N-channel polycrystalline silicon thin-film transistors

Hydrogenation Effects on the Hot-Carrier Endurance of Metal Induced Laterally Crystallized n-Type Polycrystalline Silicon Thin Film Transistors

“…Hydrogen passivation is effective to reduce defect states at GBs and at the poly-Si/ SiO 2 interface in poly-Si TFTs, leading to significant improvement of device characteristics. [3][4][5][6][7][8]11,[14][15][16][17][18][19] Hot-carrier (HC) induced device degradation is a key reliability issue for n-type poly-Si TFTs. [5][6][7][8][9][10]12,15,16,18,[20][21][22][23][24] However, it seems that most previous work agreed that hydrogenated TFTs are more prone to HC degradation due to breaking of weak Si-H bonds introduced by hydrogenation.…”

“…[3][4][5][6][7][8]11,[14][15][16][17][18][19] Hot-carrier (HC) induced device degradation is a key reliability issue for n-type poly-Si TFTs. [5][6][7][8][9][10]12,15,16,18,[20][21][22][23][24] However, it seems that most previous work agreed that hydrogenated TFTs are more prone to HC degradation due to breaking of weak Si-H bonds introduced by hydrogenation. 5,6,8,16) These studies were found in solid phase crystallized (SPC) 16) or ELC 5,6,8) poly-Si TFTs, mainly using NH 3 plasma hydrogenation.…”

Comparison of the effects of downstream H2- and O2-based plasmas on the removal of photoresist, silicon, and silicon nitride

“…For example, although we found that the N atoms by themselves do not etch Si (at least, at low temperatures), N atoms have been shown to interact with silicon in various ways, such as forming Si-N complexes, 58,59 passivating Si defects, 60 and diffusing rapidly in bulk Si. It is possible that the H x N y species 5,57 present in the H 2 /N 2 plasma could react with the poly-Si surface and occupy the binding sites that would be available for the H atoms to adsorb on the Si surface.…”

Nitrided HfTiON/Ga2O3(Gd2O3) as stacked gate dielectric for GaAs MOS applications

“…The positive shift of V fb indicates the presence of negative oxide charges in the dielectric film. The smaller positive shift of V fb for the GGO-NH 3 sample (1.09 V) than for the GGO-N 2 sample (1.28 V) should be attributed to the fact that more nitrogen incorporation into the HfTiON/GGO stack dielectric can effectively occupy its oxygen vacancies, resulting in a reduction in the density of defect traps in the film and near the GGO/GaAs interface, 19,20) which is also the main reason why the Q ox of the GGO-NH 3 sample (1.0 © 10 13 cm ¹2 ) is smaller than that of the GGO-N 2 sample (1.3 © 10 13 cm ¹2 ). Moreover, the samples with GGO IPL exhibit a smaller D it than those without GGO IPL, with the smallest value for the GGO-NH 3 sample (1.1 © 10 12 cm ¹2 eV ¹1 ) owing to the passivation role of H and N atoms decomposed from NH 3 during the PDA on the dangling bonds near/at the interface.…”

“…For the GGO-NH 3 sample, the smallest hysteresis voltage (50 mV) is obtained, implying the lowest density of defects in the HfTiON/GGO stack dielectric and near/at the GGO/GaAs interface owing to the reductions in oxygen defects and As outdiffusion associated with the GGO IPL and NH 3 annealing. [19][20][21] Furthermore, a stretchout occurs in the C-V curves of the control samples, especially for the control-N 2 sample, indicating a high interface-trap density at the conduction-band edge of GaAs caused by a considerable number of As-O, As-As, and Ga-O bonds at the HfTiON/GaAs interface. [22][23][24] Fortunately, the stretchout effect is weakened for the two samples with GGO IPL, with the minimum for the NH 3 -annealed sample, which should be ascribed to the formation of N-related strong bonds at/near the GGO/GaAs interface and the blocking role of the GGO IPL against oxygen diffusion from the HfTiON gate dielectric to the surface of the GaAs substrate.…”