It seems to me that the particles that will interfere with playing may be so small that frequently their mass may be immeasurable. A vey tiny piece of dust may become wedged in a groove at just the right angle to deflect a cartridge temporarily. Those dust particles may have accumulated over multiple playing and then over the years become deeply embedded. Their removal may improve replay but if they weigh a millionth of a milligram I wouldn’t trust any pre and post clean weight of either the lp or the cleaning solution.
What makes more sense to me would be a digital graph of the waveform produced as the stylus traverses the record. It should be possible to find deflections caused by debris. It should also be possible to correlate noises heard by listeners, such as pops and clicks, to a deflection on the graph.
It then should be possible to do the following test. Take 200 LPs. Play prior to cleaning and measure the events. Perhaps the average will be 3 of these events per side. Clean half the LPs with a machine such as Degritter and the other 100 ultra sonically. Hypothetically, the former will reduce the events to 2 per side, and ultra sound to 1 per side. Or not.
This to me would be more meaningful than comparing pre and post weights. Even if these weights could be measured accurately, there is always the possibility that the Delta in the weight is due to musically irrelevant factors. For example, what if one of the technologies is more prone to remove some vinyl from the edge of the recording before any music is playing? I am more interested in what happens in the grooves