Cleaning of nanofiber filter

A preloaded nanofiber filter can be effectively cleaned, respectively, by backpulse (BP), backblow (BB), and backpulse followed by backblow (BP–BB) to remove submicron aerosols and nanoaerosols trapped in the nanofiber filter by depth and cake filtration. Three different stages of cleaning behavior as reflected by the pressure drop across the filter can be distinctly identified. The first stage is when the cake adhered to the filter is being removed for which the pressure drop across the loaded filter reduces significantly and precipitously over a short period. The second stage is when aerosols inside the filter are being purged. This is accompanied by only modest decrease in pressure drop. The third stage is when residual aerosols are removed gradually with very little decrease in pressure drop. Loosened aerosols from the fibers in the filter by BP/BB/BP–BB, if not immediately removed from the filter by convection flow, can be recaptured by the fibers in the filter once more during the second and/or third stages. The pressure drop ultimately reaches an equilibrium residual level that is slightly higher than that of the clean filter indicating some residual aerosols being trapped in the dead pores of the filter during the prior loading-cleaning processes. Three key parameters play an important role in using reversed air jet to clean a preloaded filter. Higher pressure in the pressurized air source provides more favorable conditions for cleaning by BP, BB, and BP–BB. The advantages have been demonstrated for pressurized sources at 3, 4, and 6.5 bar, respectively. The impingement of the jet on the backside of the loaded filter needs to be distributed uniformly about the filter area to avoid high-velocity air jet in concentrating at a certain spot that can break locally the nanofiber mat. For BP, it has been found that longer pulse duration of 0.5 s is better than shorter pulse duration as it provides a stronger inertial force on cleaning the preloaded filter. A simple empirical model has been developed using two parameters a and b, where parameter a is related to the detachment-to-attachment force and the recapture of the pre-loosened aerosols, while parameter b is related to the adhesion-to-detachment force of the cake onto the filter. The results from testing on the five groups of parameters compare reasonably well with the model. This is especially for the first and third stages of cleaning using reversed air jets. Three of these five groups of parameters concern the delivery of an effective driving force to remove the trapped aerosols in a preloaded filter: the most appropriate cleaning mode BP/BB/BP–BB, pulse duration, and applied overpressure. The remaining two groups of parameters address the “retention of aerosols” for a given filter configuration, namely, increase in the specific surface area of the fibers exposed to the aerosols (from reduced fiber diameter) and increase in the filter thickness or fiber packing density to recapture detached and loose aerosols. These parameters can be tuned in various combinations to achieve the final residual solids as represented by the residual pressure drop in the filter.