Johns Hopkins University Test

 

DETERMINATION OF CLEAR-VUE CYCLONE CUTOFF SIZE USING A FLUORESCENCE METHOD

 

Introduction

A high volume cyclone was tested at the Aerosol Laboratory, to verify the lower cutoff size (50% aerodynamic diameter) specified by the manufacturer. The cyclone (Clear-vue mini CV06) was designed as an air-cleaning device for small hobbyist wood shops with a theoretical cutoff size of 2.5 μm when run with a 6 HP shop-vac at 117 cfm.

.

Objective

Establish the cutoff size of the cyclone at the flow rate specified by the manufacturer.

.

Materials and Methods

The method used to determine the cutoff size is based on the fluorescence of a monodisperse aerosol. Based on fluorescence, it is possible to determine the percentage of particles captured in the cyclone body and the cyclone cup, as well as the percentage of particles that go through the cyclone. Using different particle sizes, we constructed a plot relating particle diameter with percent fluorescence retained both in the cyclone body and cyclone cup. From this plot we can estimate the particle diameter for which there is a 50% particle retention efficiency (known as the cyclone lower cutoff point).

.

An ink-jet aerosol generator (IJAG) (Bottiger et al, 1998) developed at Edgewood Chemical and Biological Center (ECBC, Aberdeen MD) was used for this test. Monodisperse particles are generated by desiccating liquid droplets of sodium hydroxide mixed with sodium fluorescein. By manipulating the proportion of ingredients in solution, it is possible to generate different particle diameters. The sodium hydroxide-fluorescein solution is released into a vertical cylinder using an ink-jet printer nozzle. The number of particles released by the nozzle is calculated by a laser source using light scattering, and particle counts are controlled with a program written in Visual Basic. After the particles are counted, they travel through a charge neutralizer, and a heated vertical cylinder that evaporates the liquid, leaving solid particles of a single size that are released from the cylinder at 1 L/min. Particle diameter is confirmed using an Aerodynamic Particle Sizer (APS 3320, TSI inc. MN). Particles were generated at a rate of 300 particles/second during 100 seconds, for a total of 30,000 particles per run, and each run was repeated 3 times. Cyclone testing was conducted in a clean room environment. Background particle concentration in the room was confirmed by APS.

.

The first step in the cyclone testing was to determine a reference level of fluorescence for the number of particles generated in each particle size run. This was done by capturing the particles on a filter placed at the bottom of the generator’s outer nozzle. The filter was then placed in a 50 ml centrifuge tube, where 20 ml of a 0.1% solution of ammonium hydroxide in de ionized water (recovery solution) was added to recover the fluoresce in from the filter, and capped for future fluorescence reading. The average fluorescence of three replicate filters served as the reference level (i.e., 100% fluorescence level) for each particle size.

.

Immediately after the reference filters were collected, the cyclone’s inlet was placed approximately 5 cm away from the generator’s outer nozzle. Since the flow through the cyclone was high and the particle sizes small, it was expected that all the particles released from the generator were suctioned by the cyclone’s inlet. After each generation run, the pumps were turned off and the cyclone’s cup was washed using 20 ml of the recovery solution. The washout of the cup was collected in a 50 ml centrifuge tube for future fluorescence reading. This process was repeated three times for each particle size.

.

After the three runs described above, the inside of the cyclone’s body was washed using 40 ml of recovery solution, and the process repeated 3 times. Each 40 ml washout was collected in a 50 ml centrifuge tube for future fluorescence reading. A preliminary test was conducted to determine how many cyclone body washouts were required to capture all the particles. This was done by washing the cyclone with different volumes of the recovery solution, and determining the fluorescence level of each washout.

.

Blank samples of the room and clean cyclone were collected by running the cyclone for up to 30 minutes. The cup and body washes were collected in centrifuge tubes with 20 ml recovery solution.

.

Fluorescence was determined by pipetting a 3ml aliquot from each sample in a spectrophotometer cuvette ( Model 450, Sequoia-Turner, Mountain View,Calif ). Blanks of the recovery solution were run every 3 samples to zero the instrument. 

Results and Discussion

The preliminary test performed to determine how many washouts of the cyclone’s body were required showed that 40 ml was an adequate volume, and that three washouts were sufficient to recover >95% of the particles retained in the cyclone body.

.

Retention efficiency for the Clear-vue cyclone was found to be 60%  (Table 1) for the smallest particle size that the IJAG can generate, 1.7 μm. Thus, to determine the 50 % cutoff size it will be necessary to conduct further experiments with smaller particle sizes (1.0 μm and 0.5 μm).

.

.

A probable cut-off size of between 1.5 and 2 μm for the Clear-vue cyclone can be estimated by extrapolating two likely trend curves (estimated from the available data), as seen in Figure 1..

Based on these results, we currently know that the cutoff size of the cyclone is <1.7 μm, when run at the specified flow rate, and that we need to perform additional tests with smaller particle sizes in order to determine the actual cutoff size, and with one larger size (9 um) to confirm 100% retention.

.

Conclusions

The Clear-Vue cyclone has a 60% retention efficiency of 1.7 um particle sizes, and a 100% efficiency at 4.7 um at the test flow rate of  117 cfm.

The fluorescence method is a reliable way to determine cyclone cutoff sizes. It allows testing of the retention levels in both the cyclone cup and cyclone body. Since the particles used are water soluble, the recovery of particles is high. Another major advantage is that the method uses solid particles, so particle bounce and interception are similar to what can be expected with solid environmental samples.

Applying simple mass balance equations it was possible to confirm that the cyclone was capturing all the particles released by the particle-generating instrument. Mass balance also confirmed that washing both the cyclone cup and cyclone body was effective to capture the particles.