Hello, I am trying to simulate the plasmonic properties of a gold nanorod (20*40 nm), e.g. scattering cross-section and absorption cross-section. I used the model of rounded cylinder in the structure. And I set the override mesh size as 1 nm and the polarization perpendicular to the length axis. But the scattering spectrum showed strong roughness, and the absorption spectrum showed three peaks which is greatly different from the single absorption peak in references. I attached the simulated spectra. So, I am wandering if there is special parameters in simulating gold nanorod, since I have got good results on gold nanosphere. I notice the model of rounded cylinder in structure is made of two hemispheres and a cylinder, should I specially consider the boundary between these different elements?
I think using the rounded cylinder to simulation a nanorod is reasonable. The way to simulate a nanorod should be very similar to one would do it with a nanosphere. Obviously, you will need to adjust the size of the simulation region, source span, etc…
It would be hard to me to make any specific comments without looking at your simulation file. If you can share your file and reference here, I am happy to take a look at it.
I modified some parameters, and attached the results as follows. The spectra are better, but still show roughness and additional peaks.
model.fsp (303.2 KB)
It is important to make sure that auto shutoff is triggered. It tells you if the simulation is done or not. I have increased the simulation time to 300 fs and the auto shutoff is triggered at around 30%. It means that you will need at least 100 fs for the simulation to finish. Please refer to this post for more information
On the other hand, I used finer mesh and more PML layer in this simulation and it is able to obtain smoother curve. If you want more information about testing convergence, please refer to this page.
model (1).fsp (301.9 KB)
How can I check at what percentage auto shutoff is triggered after running the simulation?
You can find this information in the log file <simulation_name>.log that is generated when you run your simulation. This file can be opened in a simple text editor and among other things, it records the autoshutoff progress as shown here:
If I choose a auto-cutoff of 10^-7, then doesn’t it mean that even if the auto-cutoff only reaches 10^-6 when it’s 100% complete (which means the auto cutoff is not triggered), it’s still better than the calculation where the auto cutoff is triggered but the cutoff is 10^-5?
The Auto Shutoff indicates the amount of energy remaining in the simulation volume. This feature allows the software to terminate the simulation once a desired shutoff level is reached. The default value of Auto Shutoff min is 1e-5. The default value should work for most typical simulations. It is recommended to stick with the default value. That said, this value can be adjusted by the users when necessary. However, as you can see that this option is in the Advanced tab, you should not need to change this value unless there is a reason.