Question in simulation of charge transport plasmon mode

[Jaekak Yoo]

Hi, 
I am trying to simulate the charge transport plasmon mode using my dimer system.

(I am very new at this simulation, so I am under struggle to simulating my system)

Using ADDA, can we describe measured charge transport plasmon mode, which obtained by dark field scattering measurement?

I tried to simulate Au nanocube dimer encapsulated by Ag with 4 nm. In the case of Au, the roundness was set with ~ 4 nm, and Ag was 8 nm.

Can you give me an advise this situaiton?
Here is the parameters what I used to this simulations. (Only size was changed depending on the structures)
ADDA_PARAMS = {
    'size': 0.094,
    'eps': 5,
    'maxiter': 10000000,
    'pol': 'ldr',
    'refractive_index_sets': [
        ['n_100', 'k_100']
    ],
    'store_dip_pol': True,
    'store_int_field': True

Thank you so much!

Jaekak Yoo

[Maxim Yurkin]

Dear Jaekak,

I am not sure that I fully understand your question. In principle, ADDA is capable of simulating these systems, but some
technical issues are always possible. I guess, you're expecting to see some second peak, but don't observe it. If yes,
could you describe what exactly you expect? Maybe, share some figure from some previous publication.

Then, the details that you provided are not sufficient to reproduce your simulations (hence, we cannot fully debug
them). I understand, that full spectrum contains a lot of data (and input files), so for debugging I recommend to focus
on a single wavelength. For instance, you may choose the one where you expect to see the second peak and share the whole
simulation of ADDA for this wavelength (all input and output files).

The parameters that you specified seems to be fine, so I suggest, first, to check the shape file and the values of
refractive index.

Maxim.

[Jaekak Yoo]

Dear Maxim,

Thank you for your kind response for my question.
Before diving into the dimer simulation, I tried to simulate the extinction, absorption and scattering spectra of Au nanospheres of 20 / 80 nm using ADDA.
Because I want to compare my experimental results with theoretical simulations, so that is why I am trying to simulate according to the wavelength variance besides of single wavelength value.

Here is the input option what I used for my simulation.

        $MPI_EXEC $MPI_PROCESSES $ADDA_BIN/mpi/adda_mpi \
            -shape sphere \
            -eq_rad (0.04 for 80 nm, 0.01 for 20 nm) \
            -pol ldr \
            -lambda $(echo "scale=3; $LAMBDA/1000" | bc) \
            -m $REFRAC_VALUES \
            -maxiter 10000 \
            -dir $LAMBDA_PATH \
            -eps 0.00001 \
            -store_dip_pol \
            -store_int_field

I wanted to find proper refractive index sets, so I tried to simulate same structures (Au nanosphere 20 nm / 80 nm) by altering the refractive indices.
However, I found that the max wavelength of 80 nm Au NS does not changes a lot, which compared to the previous DDA based reference (https://pubs.acs.org/doi/10.1021/jp057170o, Title: Calculated Absorption and Scattering Properties of Gold Nanoparticles of Different Size, Shape, and Composition:  Applications in Biological Imaging and Biomedicine).
According to the reference, the 80 nm of nanosphere has over 550 nm of resonance wavelength of scattering but my result shows below 550 nm even located around 520-530 nm.
In the case of refractive index, I obtained from this website (https://refractiveindex.info/), and tested with Yakubovsky, Werner, Rosenblatt, and Rakic datasets.

Could you give me any advice to get similar result with the reference -- If I change the -pol with other options, can I get similar result or not?
Also, if I need to change any options, and refractive indices, please let me know.

I am attaching my simulation results.
Thank you so much for your kind advises!
Best regards,

Jaekak Yoo

[Jaekak Yoo]

I forgot to attach the image from the reference.

Calculated spectra of the efficiency of absorption Qabs (red dashed), scattering Qsca (black dotted), and extinction Qext(green solid) for gold nanospheres (a) D = 20 nm, (b) D = 40 nm, (c) D = 80 nm, and polystyrene nanospheres (d) D = 300 nm.

[Maxim Yurkin]

Dear Jaekak,

The reference that you provided uses the refractive indices of gold from Johnson & Christy with size correction. The
latter is probably not important for sizes larger than 10-20 nm, but the first step for you is to use Johnson & Christy
instead of the alternative sources that you considered.

Second, and probably more important, the references considers gold nanospheres in water (which is known to shift the
resonance positions to larger wavelengths). Your simulation seems to be for vacuum host medium, so you need to do
https://github.com/adda-team/adda/wiki/FAQ#how-to-simulate-light-scattering-by-a-particle-embedded-in-a-homogeneous-non-absorbing-medium
.

These are my current observations. If that doesn't help, please provide at least some of ADDA log files, since it helps
a lot to understand which parameters were actually used in the simulations.

Maxim.

[Jaekak Yoo]

Dear Maxim,

Thank you for you kind response.

Upon your advice, I used Johnson & Christy refractive indices instead of the others.

Also, I changed the -pol option from ldr to fcd to simulate the nobel metal nanoparticles efficiently based on the previous reference (Journal of Nanophotonics, Vol. 4, 041585 (2010) -- I just realized that was your paper). 

Regardless of the presence of medium I think that my simulation results shows too blue-shifted compared to both of the other method based simulation results and experimental results. 
In my case, the max wavelength of scattering was around 520 nm for 80 nm Au nanosphere, which significantly differ from the conventional references located above 550-560 nm.

To this end, I tried to simulate by changing the 1 nm, 5 nm and 10 nm of voxel size for the 20 nm, 40 nm, 60 nm, and 80 nm of Au nanospheres.

However, I got the weird results from those simulations.

Can you give me some advises?
I attached one of my log file and config file (contains the parameter settings), and examples of simulated structures.

I really look forward to your response!
Best,

Jaekak

2025년 9월 16일 화요일 오전 1시 33분 38초 UTC-5에 yur...@gmail.com님이 작성:

[Maxim Yurkin]

Jaekak, if you aim to reproduce the results for gold water, then you missed the main part in my previous message. For
instance, the simulation for the vacuum wavelength of 400 nm, should have:
-lambda .29873 -m 1.09664 1.45855
(assuming 1.3390 for water refractive index) instead of:
-lambda .400 -m 1.4683647798742139 1.9529811320754717

However, some published results are for gold in vacuum. For instance, that is the case for our 2010 paper. And the
position of your peaks seems to agree with Fig.1 (but the latter is not that convenient due to the used log scale).

Finally, for such tests of simulation workflow, I recommend to first use built-in ADDA shape generation (e.g., -shape
sphere -grid 80 -size 0.08) to avoid the potential issues with external shape generation (which needs to be tested
separately).

Maxim.