Amorphous Silicon Process
The amorphous silicon process deposits a-Si — silicon atoms with no long-range crystalline order at all, grown from silane at temperatures low enough to coat glass, not just wafers. That single property, a deposition temperature compatible with glass substrates, is what separates it from the polysilicon process covered elsewhere: LPCVD polysilicon needs 580–650°C and a silicon wafer; PECVD amorphous silicon runs as low as 200–300°C and can be deposited directly on display-grade glass, which is why a-Si became the backbone material of the flat-panel display industry rather than a logic-chip material.
As-deposited amorphous silicon is good enough to switch a pixel, but not good enough to drive serious logic, and that trade-off is exactly what made LCD backplanes practical. Electron mobility in amorphous silicon is roughly 1 cm²/V·s — orders of magnitude below crystalline silicon's hundreds of cm²/V·s — which rules a-Si out for anything resembling a logic transistor. But a simple a-Si thin-film transistor (TFT) only has to switch one pixel's liquid crystal on and off, a job that low mobility handles just fine, and a-Si's low deposition temperature meant that job could be done directly on large sheets of ordinary glass at a cost crystalline silicon on glass could never match. That combination — low performance requirement, low-cost glass-compatible process — is the entire reason a-Si TFT became the dominant LCD backplane technology for decades.
Laser annealing exists specifically because glass can't survive the temperature a thermal recrystallization would require. Converting amorphous silicon into usable polysilicon normally wants furnace temperatures well above what display glass tolerates, so excimer laser annealing (ELA) instead delivers the energy in nanosecond pulses, melting only a thin surface layer of the a-Si film and letting it recrystallize into a fine-grained polysilicon before the heat has time to conduct into and damage the glass underneath. The resulting low-temperature polysilicon (LTPS) reaches mobilities tens to over a hundred times higher than as-deposited a-Si, high enough to integrate actual pixel-driving logic directly on the display substrate — the technology behind the high-resolution, high-refresh-rate OLED and LCD panels in modern smartphones, something a-Si TFT backplanes alone could never support.
The same base chemistry also deposits the absorber layer in thin-film amorphous silicon solar cells, for a reason unrelated to transistors entirely. Amorphous silicon absorbs visible light far more strongly per unit thickness than crystalline silicon does, because its disordered structure relaxes the strict momentum-conservation rules that make crystalline silicon an inefficient (indirect-bandgap) light absorber. A solar cell only needs a few hundred nanometers of a-Si to absorb most usable sunlight, versus tens to hundreds of microns of crystalline silicon for the same job — trading some conversion efficiency and long-term stability (a-Si solar cells degrade through the Staebler-Wronski effect under light exposure) for a dramatically thinner, cheaper, flexible-substrate-compatible film.
Read the amorphous silicon process as proof that giving up crystal order on purpose opens doors a crystalline process keeps closed: a-Si's low deposition temperature put transistors on glass for the first time, its light absorption put a usable solar cell in a few hundred nanometers instead of a few hundred microns, and when a laser pulse recrystallizes it back into tiny polysilicon grains without ever heating the glass underneath, the same disordered starting film becomes the high-mobility backbone of every modern smartphone display.