Mismatch, in terms of two materials with cubic crystalline structure, means that their lattice constant are different. Some semiconductor alloys can be grown to have the same lattice constant as another, in that case you have a lattice "matched" system, otherwise you are "mismatched".
Due to the strain accumulation, a critical thickness exists in the case where different atoms are involved in the growth of strained layer or mismatched layers. below that thickness It's possible to grow crystalline material. however beyond the critical thickness, the strain relaxation occurs in most of cases by defects generation (excluding the case of Stranski Krastanov growth mode) always.
In fact the growth of this kind of materials, is highly dependent on the growth parameters like temperature, deposition rate....
If we consider your technical (practical) point of view, growing films of about some dozens nanometers of thickness of crystalline quality is impossible. What about simulating this for studding optical properties (absorbency) for photovoltaic applications? Is possible to enhance efficiency (optical) by using thick films instead ?
there are now several attempt to compensate the strain towards successful growth of lattice matched materials having the same band gap tunebility as the highly strained materials (mismatched materials).
For examples InGaAs is highly strained on GaAs . Now you have InGaAsN that can be lattice matched to GaAs.
Generally when you need thick material you have to avoid lattice mismatch.