I had the same problem. It was solved in a very simple way: we dropped the concentrated solution of the nanoparticles on the carbon-coated copper mesh (used for TEM samples as a support) and dried it in a desiccator. As we had metal nanoparticles they were visible on the copper surface at even high magnification without further carbon or metal coating.
The CEM method scans solid samples. Filter or centrifuge nanoparticles, rinse, dry. For the study of dispersions, use SAXS, DLS, SERS, UV- visible spectroscopy ...
In our case, a drop of NPs solution was dropped on a normal glass slide (small pieces made with slide cutter) and dried in an hot air oven at 40 0C. Before inserting of this slide, the SEM operator coated with gold (conductivity purpose) in case of biological samples.
Liquid samples it is not possible due to the necessary characteristics of the technique (vacuum, high voltage, etc).
What you can do is that it was mentioned above depositing a drop on a substrate (with different characteristics in comparison with the material that you want to analyse) crystal silicon, copper mesh, for instance.