The concept of phase-change memory using chalcogenide glasses dates back to the 1960s, as pioneered by Ovshinsky. [1] Starting from late 1980s, Ge-Sb-Te alloys along the GeTe-Sb 2 Te 3 pseudo-binary line (Group I), such as Ge 2 Sb 2 Te 5 , Ge 1 Sb 2 Te 4 and Ge 8 Sb 2 Te 11 , [2][3][4][5] doped Sb 2ĂŸx -Te alloys (Group II), such as Ag-, In-doped Sb 2ĂŸx Te (AIST), and alloyed Sb (Group III), [6][7][8][9][10] such as Ge 15 Sb 85 , were identified as suitable PCMs. These alloys were initially used for the rewritable optical data storage industry. [11] More recently, the Ge 1 Sb 2 Te 4 (GST) alloy has been used as the core material for electronic non-volatile memories, [12][13][14][15][16][17][18][19] e.g., 3D Xpoint. Moreover, PCMs can be used for neuro-inspired computing and in-memory computing, [20][21][22][23][24][25][26][27] and could also enable various nonvolatile photonic applications, including memory and computing devices, switches, and displays. [28][29][30][31][32][33][34][35][36] PCMs utilize the significant contrast in electrical resistivity or optical reflectivity between their crystalline and amorphous phase to encode data. [11] The switching between the two logic states is achieved by rapid and reversible phase transitions, namely, SET (crystallization) and RESET (melt-quenched amorphization). In addition to binary storage, multiple resistivity or reflectivity states can be obtained within a PCM cell via partial amorphization (iterative RESET) and crystallization (accumulative SET), enabling multilevel storage [37,38] and neuro-inspired computing applications. [39,40]