Imaging an isolated structure


#1

I’m trying to simulate microscope images. This is similar to the example given here https://kb.lumerical.com/en/micro_and_lith_pcm.html, except that I have an isolated structure and I’m interested only in bright field microscope images.

Since it is an isolated defect on substrate, I use TFSF source and PML BC. Illumination angle will be set to 8 different angles distributed around the circular annulus. I’m not sure if this can be done with TFSF source. Can anyone please let me know if the setup is right for simulating microscope images of isolated structures.

Script and simulation files are attached.phase_contrast_batch_run.lsf (1.5 KB)
phase_contrast_analysis.lsf (6.2 KB)
phase_contrast1.fsp (238.5 KB)


#2

I checked the simulation file setup, and one issue that I noticed is that the TFSF source does not fully enclose the “grating” rectangle structure which is the object that is scattering the light that gets imaged. The following page mentions that for the TFSF source to work properly, the boundaries of the source cannot extend through the scattering object:
https://kb.lumerical.com/en/index.html?ref_sim_obj_tfsf_sources_usage.html

To correct this, you could extend the y span of both the source and simulation region so that they are large enough to enclose the full length of the “grating” object.

The x and y span of the monitor may also be increased to make sure that the size of the monitor is large enough to collect all of the scattered light, even light travelling at steep angles.

It is still possible to use the phase_contrast_batch_run.lsf script file to vary the “angle phi” setting of the TFSF source to simulate the source incident at different azimuthal angles from the annulus.

The phase_contrast_analysis.lsf script would need some modification since it is using the grating projection script commands which are meant for periodic simulations (gratingpolar, gratingu1, gratingu2), and these would need to be replaced by near to far field projection script commands which are meant for standalone structures (farfieldpolar3d, farfieldux, farfielduy). The details about these near to far field projection script commands are here:
https://kb.lumerical.com/en/index.html?ref_scripts_near_to_far_field_projections.html

And there is also an example showing simulation of microscopy imaging using these script commands here:
https://kb.lumerical.com/en/index.html?micro_and_lith_imaging.html

Rather than modifying the phase_contrast_analysis.lsf script file, it may be easier to modify the scripted analysis from the microscopy imaging example linked above and add a for loop to loop through each simulation file with different source angle and sum the resulting E intensity images from all of the files together to get the final result.

Hopefully this helps!


#3

I have been able to simulate successfully according to the settings you have suggested in the script. I have couple of other questions.

  1. Power monitor outside the TFSF source collects only the light from the isolated scattering object right. I want to capture the uniform amplitude and phase added to the image by the substrate as well. I need this for subsequent image processing. To achieve this, would it be ok If I move the power monitor inside the TFSF box.

  2. As far as I understand, defocus parameter will project the image at a distance Zo from the near field. Is it the same as adjusting the focusing knob of the microscope to displace the image plane?


#4

You are correct that the outside of the TFSF source region, only the light scattered from the isolated scattering object will be present, and any light directly reflected by the substrate will not be measured by the monitor outside of the source region.

You are also right that the defocus parameter in the example script for the microscopy image application shifts the position of the image plane. In a standard compound microscope the focus knob adjusts the height of the stage relative to the microscope body so it should have an analogous effect.