Given they only need to specify two properties of the material (density, and ability to be suspended in water), they should be spoilt for choice; and would be pretty stupid if they didn't pick something inert. This is not like semiconductor manufacturing or something where they need to find a material with obscure properties.
My guess is the main reason for not specifying the material is that they will pick whatever is cheapest locally to each development.
The particles would have to be extremely small to not fall out of suspension, so I don't think they'd be very abrasive.
The thickness of the viscous part of the boundary layer in water is maybe 0.1 microns (if the water is moving at 10 m/s). Particles smaller than this would be slowed before striking the surface. If the particles are making the liquid more viscous the viscous part of the boundary layer will be proportionally thicker.
