There was no doping of the mwcnts. The cnts were made to a spec, mainly average length and average diameter, mwcnts being the targeted morphology. We took pains to clean any tramp materials from the cnts. All cnts were boiled in nitric and hydrochloric acids, and washed in hydrofluoric (I think some were boiled as well, but it was difficult). They were then washed in DI water. Since they were primarily intended for water 'filters' that was the end of the pre-processing.
Problems arose when the filters were put on test during the day (100% testing), and left pressurized but not subject to flow overnight. The next day, when testing was restarted, we noted the flow rate was markedly less initially, but climbed back to normal flow rates after a while. I was tasked with finding out what the problem was.
With the clear filter housings a mass of gas bubbles was apparent, visible on the surface of the cnt filter. I began by isolating elements present in the filter. The simplest test combined cnts and the same water used during testing. Gas bubbles formed a blanket over the cnts overnight. The vessel was a flask with a tapered ground glass seal, with a mating glass valve as a seal. Nothing fancy. My intention was to do a 'pop' test at the valves exit. When I did so, I got the (actually unexpected) pop, but it also unseated the valve.
