Increasing carbon's van der Waals radius to delete water molecules

I used charmm-gui for building micelle-protein complex.
I increased carbon radius for removing water inside the micelle. but nothing happend! Should I do something with charmm-gui to build a correct structure without water inside it?
the following picture showes water density in the box, water density should be zero at the center of the micelle!!

You are best directing this question to those who provide the charmm-gui, this is not a GROMACS issue.

I know it isn’t gromacs issue. But I thought maybe some body had the same problem who can help me now.
Anyway, Thank you.

Hi - depending on what you’re trying to study, if you just run the simulation for a while the water will eject eventually. You could restrain the micelle atoms in the meantime.

Thanks for your response. I ran the simulation for more 100 ps but water density increased at the center of the structure even though my structure seems more compact than before!!
my first structure:

after simulation:

You wouldn’t expect to see ejection on that timescale, more like 10s to 100s of nanoseconds. Also what is your coordinate/command for gmx density?

Thank you so much…
So you think nothing is wrong with the initial structure and it’s just about timescale.
This is my command:
gmx density -s nvt.tpr -f nvt.xtc -o dens_nvt.xvg -d Z

If you’re slicing along the z dimension, but your micelle doesn’t cover the entire box in that dimension (which it wouldn’t because it has finite size), gmx density is also picking up the waters in those z-slabs outside of the micelle. I think you’re just misinterpreting gmx density

Thank you kevin. All bests for you:)