Using Klaassen model for impurity scattering




In semiconductor material model, it is mentioned in the KnowledgeBase that “… the Klaassen model accounts for temperature dependence separately, therefore a constant value should be used for the lattice scattering mobility…” I am using silicon material at 300K. I need to confirm that, does it mean that I must disable the temperature dependent model for lattice scattering?




You are correct. The Klaassen model already includes the effect of temperature on mobility. The input it needs therefore is the electron and hole mobility of silicon at 300 K. So the mobility entry for both electron and hole should be a constant and not a function of temperature when using this model.


Thank you @aalam for your previous confirmatory reply.

However, I am experiencing another relevant problem. I am using previous version of DEVICE (4.6.590) to simulate silicon pn junction, using the Klaassen model for impurity scattering (for both electrons and holes). I am using Klaassen model because it is mentioned in the KB that this model is specifically tuned for silicon at 300K. However, when I run my simulation, I get errors saying “Bad reference material” and “solver failed”. For exactly the same structure and simulation setup the simulation would not fail if I’d use the default silicon material from the material database (instead of my copy of silicon material where I am using various models - let’s call it mySilicon). The simulation would also run without this error if I only change from Klaassen to Masetti in mySilicon material. Please advise me where am I going wrong. My suspicion, however, is that the Klaassen model is using very different parameters for electrons as compared to holes (I found out by clicking “…” button in front of the model name). The parameters for holes have all the values of r (total 6) and s (total 7) set to zero and all the other parameters are set to 1 (except for Nref1, NrefA and NrefD which are set to 1e+30). Should not we use the same parameter values of Klaassen model for electrons and holes? If yes, which one should be used (the one specified for electrons or the one specified for holes)?



Hi. Sorry for the delayed response. I think the problem is that some of the values of the parameters for both electrons and holes are missing. I can share a DEVICE project file with the correct parameters however it will be in the new version of DEVICE and will not open with your older version. If you are okay with upgrading to the newer version then let me know and I will share the file with you.


Thank you @aalam for your response. Yes please share the file as I also have one license for the newer device version. I believe the same parameters would also work on the previous version of DEVICE (the one I am using now).



Hi @ajs11_eee138, I have updated the Klaassen model parameters for the Silicon material in this project file Si_Klaassen_model.ldev (117.7 KB). Let me know if this solves your problem.


Hi @aalam I have used new project you have shared but still I get Bad reference material as error


Hi @ee13s067, Please note that in the project file there is only one material, silicon. The way you can bring this material into your simulation is to import the material from the material database of your own project file (take a look at the screenshot below).

When you import the material it will try to overwrite your existing silicon model since the names are identical. You will have to allow the overwrite (replace) and then you should be able to use the new silicon material in your simulation.

If you still have trouble then another thing to check would be whether you are using the latest version of DEVICE. If you are using an older version then please upgrade to the latest version and that should solve the problem. You can get the latest version from here:


Thank you for the reply. I used the import material n replaced original
material. But it still showed bad reference material. I am yet to try with
new version of DEVICE.


Hi @ee13s067, Let me know if upgrading to the latest version does not solve the problem. If you would like me to take a look at the file then please feel free to share it with me.