Large-area, low-cost substrates are envisioned for next-generation HgCdTe infrared focal-plane arrays (IRFPA). Si, GaAs, Ge, and InSb have been previously examined as potential candidates. Fabrication of IRFPAs based on these substrates is limited by fundamental materials properties that potentially lead to lower detector performance and operability. Lattice and thermal mismatch between the substrate and epilayer are just two of several material factors that must be considered. We have reviewed these factors in the context of more recent data, and determined it worthwhile to revisit the use of GaAs substrates for epitaxial growth of HgCdTe. Our study starts with an evaluation of the surface quality (epireadiness) of commercially available (211) B-oriented GaAs substrates. Molecular beam epitaxial growth of CdTe buffer layers and subsequent HgCdTe absorber layers are performed in separate vacuum-interconnected chambers. The importance of optimization of the CdTe buffer layer growth for high-quality HgCdTe is detailed through surface morphology and x-ray studies. x-Ray diffraction rocking-curve full-width at half-maximum values as low as 52 arcsec have been obtained. Long-wave infrared Hg1−xCdxTe (x = 0.23) has been grown on these buffer layers, producing cross-hatch-dominated surface morphologies, with dislocation densities as low as ∼3 × 106 cm−2. We have also obtained (for optimized layers), 80-K Hall-effect n-type carrier concentration and electron mobility of approximately ~1.5 × 1015 cm−3 and 1 × 105 cm2 V−1 s−1, respectively. Finally, we briefly compare GaAs and Si in light of our preliminary investigation.