Particle-fluid separation: Solving the CPC problem

by Robert P. Donovan

This month's column illustrates yet another technical problem with a solution based on particle-fluid separation brought about by mechanisms other than fiber or membrane filtration. In this application, the gas component of an aerosol is changed from a reactive gas, such as oxygen or hydrogen, to a less reactive gas without significantly altering the aerosol particle component of the aerosol.

By putting this application to work you can solve the problem of using a condensation particle counter (CPC) to measure particle concentrations in oxygen or hydrogen—both present safety concerns if used with butyl alcohol, the working fluid used in the most common, commercially available CPC.

Figure 1. McDermott’s Matrix Gas Exchanger (adapted from Ref 1).
Click here to enlarge image

One solution to this problem would be to replace the working fluid in a CPC with an inert liquid. This approach is available but penalizes performance with respect to the butyl alcohol analyzer and introduces compatibility problems of its own. The solution that follows applies to particles entrained in a number of potentially hazardous gases but retains the performance advantages of CPCs using butyl alcohol as their working fluid.

This is achieved by changing the gas matrix of the aerosol beam from its initial, reactive composition to that of a benign gas, such as nitrogen, which presents no problems for a butyl alcohol CPC. This matrix switch is accomplished without significantly altering the concentration of most aerosol particles in the initial aerosol stream.

Figure 1 illustrates the design described by McDermott [ref 1]. This design depends on the differing diffusion coefficients of gases and particles to convert an aerosol particle stream entrained in a reactive gas (shown entering the tube at the left-hand side of the figure) to an aerosol particle stream of close to the same particle concentration but entrained in nitrogen gas (shown exiting the tube on the right-hand side). What has transpired along the tube length is the replacement of many of the initial aerosol gas molecules with nitrogen molecules from a sheath flow.

Two gas streams enter the left end of the flow tube. The first is a stream of aerosol particles in a reactive gas matrix that enters along the tube centerline. The second stream is a filtered nitrogen stream that enters as sheath gas flow surrounding the aerosol stream. The flow rates of the aerosol stream and the sheath stream are adjusted so that the entering velocities of the two streams are the same. This condition preserves laminar flow along the tube length that minimizes turbulent mixing of the two streams.

As the two streams flow down the tube, however, the oxygen or hydrogen molecules in the aerosol matrix diffuse out of the aerosol stream, mixing into the nitrogen sheath stream.

Conversely, nitrogen molecules in the sheath stream diffuse into the centerline aerosol stream.

Aerosol particles in the aerosol stream do not appreciably diffuse but stay in the center-line stream because of their smaller diffusion coefficients (a 2-nm particle has a diffusion coefficient ~ 1/10 that of an oxygen molecule in nitrogen and ~1/100 that of a hydrogen molecule in nitrogen, according to McDermott [ref 1]). At some downstream location, the molecular exchange process renders the aerosol stream suitable for evaluation by a butyl alcohol CPC. Figure 1 indicates this condition by labeling the centerline aerosol stream exiting the flow tube as being in a gas matrix composed mostly of nitrogen molecules. Similarly, the vent gas is now a mixture of nitrogen with most of the oxygen or hydrogen molecules of the initial aerosol stream.

Hydrogen's larger diffusion coefficient means that the required tube length for safe CPC operation is shorter for an aerosol stream in a hydrogen matrix than for an aerosol stream in an oxygen matrix. McDermott notes, however, that hydrogen's low density also means that the diluter tube should be oriented vertically to prevent buoyancy effects from moving the aerosol stream off center.

While particle losses from the exiting centerline aerosol stream increase with decreasing particle size because particle diffusion coefficients increase with decreasing particle size, in most matrix gases these particle losses will be minor for flow tube lengths compatible with safe CPC operation.

Robert P. Donovan is a process engineer assigned to the Sandia National Laboratories as a contract employee by L & M Technologies Inc., Albuquerque, NM. His Sandia project work is developing technology for recycling spent rinse waters from semiconductor wet benches.


  1. McDermott, W. T., “A Gas Diluter for Measuring Nanometer-Sized Particles in Oxygen or Hydrogen,” JIEST 41(4), July/August 1998, pp 17-23 (IEST, 940 E. Northwest Highway, Mount Prospect, IL 60056;


Easily post a comment below using your Linkedin, Twitter, Google or Facebook account. Comments won't automatically be posted to your social media accounts unless you select to share.