Graphene model in DEVICE


#1

In DEVICE, Graphene hadn’t defined as a material, so how we can define it as a 2-D or 3-D material like FDTD solution? Since even the two-dimensional rectangular is not defined in the Structures button!

Best Regards,
Hananeh.


Problem with simulation time for a graphene interdigitated-MSM
Inactive buttons in DEVICE
#2

DEVICE does not support the 2D rectangle. So if you want to model graphene in DEVICE, it will have to be through a 3D volumetric model. The default material database of DEVICE does not have a model for graphene. However, in this KB example (electro-optic modulator using graphene), a 3D equivalent model of graphene has been used in electrical simulation. However, this model has limitations. For some more details on this check out this other KX post: https://kx.lumerical.com/t/simulation-of-a-graphene-based-transistors/2035/2?u=aalam.


#3

Hi,
Thanks a lot. I have some questions about use materials in DEVICE.

What is the limitations of graphene in DEVICE? How can we eliminate these limitations?

Can we simulate a structure with this model of graphene in DEVICE and is the correct answer will be obtained?

And my last question: in FDTD some more materials have defined. for example 3 type of gold(palik-CRC-johnson & christy) but in DEVICE only 1 type (CRC) has defined. in according to material explorer in FDTD, the characteristics (real & imaginary part) of 2 type of gold (CRC-johnson & christy) compare together are some different. in other hand,in DEVICE only one electrical parameter (Work function) has defined to change! So,how can we change characteristics of gold (CRC) to johnson & christy like FDTD?

Best Wishes,
Hananeh.


#4

Hi. As mentioned in the referred post, the model that we had used for graphene in the modulator example is a fitted model. The DOS is a 3D DOS that has a finite bandgap but the parameters have been tuned so that the material has (approximately) the same charge density as given by the 2D DOS of graphene. This gives us reasonable results when only the charge-chemical potential relationship is important as in the modulator (no transport). However, this model is not going to be accurate enough to model transport in graphene.

In order to have a proper model for graphene, DEVICE needs to be able to support 2D DOS models that it currently does not. So unfortunately there is no way to overcome this limitation at the moment.

Regarding the material models, FDTD has different models for gold because there are different index values reported for the material from different sources. However, in a (semiclassical) electrical simulator like DEVICE, gold and all other metals are simply used to define an electrical contact and the only parameter that is needed to model this contact (between metal and semiconductor) is the work function of the metal. The work function of gold is well documented and we only need to use a single value. Please note that the FDTD models and the DEVICE models are totally unrelated and irrespective of which model of gold you are using in FDTD, you should be able to use the default model for gold in DEVICE for your electrical simulation.


#5

Thank you very much for explanation and helping.