The Science and Practice of Metal-Organic Vapor Phase Epitaxy (MOVPE)

Surface chemistry models for GaAs epitaxial growth and hydride cracking using reacting flow simulations

“…These experiments were conducted with a Dynamic-Hydride Vapor phase epitaxy (D-HVPE) reactor 12 which addresses manufacturing challenges posed by HVPE. Highly efficient thin-film solar cells typically consist of several layers, each one made of a different III-V compound 13 . When the growth rate of the solar cell is low enough and the reactor equilibrates quickly, the substrate can be kept idle in the same chamber while different external gas are injected over time.…”

Low-Cost Approaches to III–V Semiconductor Growth for Photovoltaic Applications

“…This has historically provided low growth rates but afforded excellent control over junction formation, doping profiles, and buffer grading between III–V’s. , As a result, MOVPE semiconductor growth costs are high not only due to the expense and low utilization of its toxic and pyrophoric precursors but also due to the slow growth leading to large contributions of safety infrastructure, reactor depreciation, maintenance, and labor costs per layer . The same level of materials control has been demonstrated in molecular beam epitaxy (MBE), but production costs for this technique are even higher due to lower growth rates and limits on scalability imposed by high-vacuum growth environments. , …”

“…Because HVPE is operated near equilibrium, addition of HCl at the reaction zone via a second gas flow can slow or reverse deposition, while growth rates can reach >100 μm h –1 by increasing precursor supersaturation or mass transport to the substrate. , While there are examples of III–V-based solar cells in the HVPE , and related chloride vapor-phase epitaxy literature, III–V PV device growth by HVPE was initially hampered by difficulty producing abrupt junctions, controlling metal chloride chemistry, particularly that of Al-containing species, and preventing surface degradation . As a result, MOVPE became preferred over HVPE, particularly for depositing low-dimensional structures such as superlattices . HVPE remained prominent for growth of LEDs because of its high growth rates and, more recently, its compatibility with III–nitrides, and has been used for anisotropic growth of III–nitride micro- and nanostructures for complex transistor and lasing devices …”

“…4 The same level of materials control has been demonstrated in molecular beam epitaxy (MBE), 35 but production costs for this technique are even higher due to lower growth rates and limits on scalability imposed by high-vacuum growth environments. 33,36 Recent economic analysis shows that III−V solar cells produced by MOVPE cannot become cost-competitive for nonconcentrator applications without a radical change in approach; 2,4 techniques that are inherently less expensive while providing the same material quality for high-efficiency devices are required. 37 Ideally, such systems would be flexible enough to produce multijunction devices, but demonstrated record efficiencies for single-junction thin-film GaAs are sufficiently high (η = 28.8% at 1 Sun) 38 that even single-junction PVs produced by an alternative route could be costeffective.…”

“…59 As a result, MOVPE became preferred over HVPE, particularly for depositing low-dimensional structures such as superlattices. 36 HVPE remained prominent for growth of LEDs because of its high growth rates and, more recently, its compatibility with III−nitrides, and has been used for anisotropic growth of III−nitride micro-and nanostructures for complex transistor and lasing devices. 52 HVPE is being reinvestigated for III−V growth as the cost barriers for MOVPE-grown III−V solar cells have become more apparent.…”