Electrical simulation of Metal-oxide-ITO-semiconductor structure


I need to check the band diagram and accumulated charge layer of the EOM structure in this paper:

The structure simply looks like this:

To give an example of the results I aim to get from the simulations, here is a similar approach applied to a different ITO structure:

Attached are my trials to obtain such results. I modeled the ITO (Indium-tin-oxide) as a semiconductor of large energy gap (around 4.5 eV) and then added a high doping region of around 1e19 cm-3.
However, I faced a fundamental problem; I can’t plot the band structure inside the oxide. I don’t know why. In addition, the charge monitor displays the charge distribution inside the silicon only. I can see nothing inside the ITO layer. So, any help is highly appreciated.


moscap (1).ldev (5.7 MB)

Electrostatic Potential Convergence Problem

The CHARGE solver only reports band diagram in semiconductors. As a result it will not show any band diagram in SiO2.

The monitor does report charge inside ITO. However because of the mismatch in the work function of the metal, ITO, and silicon, the ITO layer gets depleted. Since the layer is so thin, it gets almost completely depleted and the charge inside the layer basically becomes zero. If you look at the charge in log scale you will see that the monitor does report the charge in ITO.

One thing that I noticed is that the work function that you have defined for ITO is quite small (3 eV). I would expect this value to be much larger (more like 5 eV). Please note that the work function that you define in the material database should be for the intrinsic material. The CHARGE solver will automatically recalculate the work function based on the doping you have used when running the simulation.

When I set the work function of ITO to 5.5 eV and ran the simulation I got a much smaller band bending in the ITO layer and the charge monitor shows a larger electron density in the layer.


Thanks for your help.
However, I was expecting that under the biasing condition of +2.5 V, I should get an accumulation layer not a depletion region. What simulation/material parameter could be responsible for this?


Hi @aya_zaki, When you put two semiconductors in contact with each other, the Fermi levels in the two materials get aligned. Depending on the difference in work function between the two materials, electrons/holes flow from one material into the other to make the Fermi levels aligned. This is nothing to do with applied bias.

That said, once I changed the work function of ITO to 5.5 eV (more in line with existing literature) the Fermi levels in the two semiconductors become more comparable to each other and the ITO do not get depleted that much (I must have made some mistake above). If you apply a bias sweep you will the surface of the ITO going into accumulation.


I got a request for the modified file so here it is: .moscap MOD.ldev (5.7 MB)