Anonymous @ on
the lattice constant for tight binding should be really used in the generation of the heterostructure
Right now the lattice constant is set to the GaAs lattice constant of 0.56. That means that the barriers should be automatically set to equal multiples of that lattice constant – one cannot have a 5nm barrier for example in this case. So the tight binding calculation should really report the actual band edge diagram that is being used in the calculation. I.e. barrier and well heights need to be rounded to the closest multiple of the lattice constant and the user should receive a report of this rounding.
As of now with a lattice constant of 0.56nm the 2 methods actually compute two different geometries and therefore the transmission constants are quite different. However the user has no idea that the geometries are indeed different since only the continuum model band edges are being displayed.
That means that there probably needs to be a representation of 2 different band edge diagrams, indicative of the two different methods.
Since this is primarily an educational tool, one should probably set the lattice constant to 0.5nm anyhow such that the default layer thicknesses that are typically multiples of the even nm lengths.