A plot published by Don Grant several years ago graphically illustrated an important filtration phenomenon. It is a plot of particle penetration through a filter vs. particle diameter, similar to the plot discussed in last month's column (“Fibrous filter performance: Particle control success; AMC failure,” CleanRooms, August 1999, page 10).
From the parameters of the filter construction, the predicted particle penetration, assuming that the primary particle capture mechanisms are diffusion and interception, is the curve sketched in the lower left of the figure. The data points the upper right corner represent the actual performance. These data disagree with the predicted diffusion/interception performance but match up well with particle penetration determined by the filter pore size. They show that this filter under these test conditions behaves as a sieve, despite last month's pronouncement that “filters are not sieves.”
Predicted and actual liquid-borne particle penetration through a filter.1
The fluid in which the test particles were entrained was water not air, as was true in last month's plot. This change of fluid media can make a huge difference.
The mechanisms of diffusion and interception still apply in the liquid environment, although particle diffusion coefficients are significantly lower in liquids than in gases. Particle capture by diffusion should thus be reduced but, as Grant points out, the longer fluid transit times through the filter that are typical of liquid filtration compensate for the lower diffusion coefficient so that a distinct diffusion-dominated branch of the particle penetration curve can be observed under the right conditions (not the conditions of the plot). Interception depends only on particle diameter and is equally applicable in gases and liquids. The Van der Waals forces that hold a particle to a surface are also typically lower in liquids, so reentrainment should be more likely in a liquid. These effects, however, are minor compared to the impact of an interaction not typically important in air/gas filtration.
Electrical interaction arises from the charge separation that in general occurs at the interface between any two phases. In particular, an ionic charge separation takes place when a solid surface, such as a particle or a filter fiber, is immersed in a liquid. Differential ion adsorption on the solid surface, differential ionization of specific surface groups and perhaps other interactions create a charge separation and a potential difference across the liquid-fiber interface and also across the particle-liquid interface. It is when both the filter fibers and the particles become similarly charged that the electric fields so created repel each other and thwart the traditional, purely physical mechanisms of particle capture. This electrostatic repulsive force prevents all particle contact with the filter surfaces except when the fluid flow carries a particle into a pore that is too small for the particle to fit through.
The same forces exist in a wafer/particle/liquid system. Here the electrical repulsion forces act to prevent, or at least minimize, particle deposition on a wafer. Cleaning prescriptions, notably from Ohmi2, include steps designed to introduce such repulsive forces into a wafer-cleaning recipe, thereby minimizing particle deposition during wet cleans, (see Ref. 3).
by Robert P. Donovan
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.
- Grant, D. C., B. Y. H. Liu, W. G. Fisher and R. A. Bowling, “Particle Capture Mechanisms in Gases and Liquids: An Analysis of Operative Mechanisms,” pp 464 – 473 in 1988 Proceedings of the Institute of Environmental Sciences.
- Ohmi, T., “Total Room Temperature Wet Cleaning for Si Substrate Surface,” J. Electrochem. Soc., v. 143, No. 9, September 1996, pp. 2957 – 2964.
- Riley, D. J. and R. G. Carbonell, “Mechanisms of Particle Deposition from Ultrapure Chemicals onto Semiconductor Wafers: Deposition from Bulk Liquid during Wafer Submersion,” J. Colloid Interface Sci. 158, 1993, p. 259, “Deposition from a Thin Film of Drying Rinse Water,” J. Colloid Interface Sci. 158, 1993, p. 274