MIM tunnel junction near and far field patten



I am trying to simulate the Metal-Insulator-metal junction from the paper - " Highly efficient on-chip direct electronic–plasmonic transducers" (attached). More specifically I am trying to simulate the structure in fig. S10-S13 given in its supplementary paper(attached). I am facing difficulties with the results not matching the paper. I am attaching my .fsp file and would like you to have a look at it.

I’d also like to replicate the graph, Fig 2e from the main paper. Any insights regarding the approach for different voltages applied would be helpful.

P.S- I’ve used all the sources and monitors in the same file. Enable and disable them according to the simulation


Du_Highly efficient on-chip direct electronic–plasmonic transducers_NaturePhotonics2017.pdf (1.7 MB)
Supplimentary section_Highly efficient on chip.pdf (3.1 MB)
MIM_1.fsp (320.7 KB)


Hi @nke,

It’s always tricky to try and duplicate results from a paper, as often, the actual simulation settings are not fully provided. As a starting point, I would review the plasmon simulation methodology and the common simulation consideration pages in our Knowledge Base.

I would have a few comments regarding your simulation file:

  • Mesh: plasmons are very sensitive to the mesh size nearby the metal interfaces. This will be tricky since the \(AlO_x\) layer is only 2nm thick. In your simulation, the mesh in x direction is too big to resolve the thickness. You can add different mesh override regions to cover each 3 elements and resolve their thickness.
  • Distance to PML: make sure there is at least half a wavelength between the metal interfaces and the FDTD boundaries as, if the PML are too close, they may perturb the plasmons.
  • Mesh refinement: in your file, you are using the Yu-Mittra method 2. We usually advise to use the Conformal Mesh variant 1 when metals are present, with careful convergence testing as discussed here.

Additionally, you can also check that you are using the same material properties as in the paper.

I hope this will help!


Thank you for your reply.