by Robert P. Donovan
 |
My column in the November 1999 issue (p. 8), entitled “Fickle particles,” alluded to what I found to be surprising particle behavior. “Fickle” behavior, I called it then, in that some particles are clearly easy to dislodge from a tubing surface by a simple finger flick at the same time that their adhesion forces to this surface are predicted to be quite large.
This observation implies that measurements used to certify cleanrooms, for example, are at the mercy, or at least the technique, of the particle counter operator. A revisit of this puzzling behavior seems in order.
Part of the explanation for the discrepancy between predictions and observations, I think, is that the particles re-entrained by the finger flick are part of an accumulated multilayer particle buildup on the surface and do not originate from a particle monolayer or less in which single particles are in direct contact with the tubing surface. The 2nd, 3rd and subsequent layers of particles accumulating on a surface form a three-dimensional network, each particle adhering to its nearest neighbors at a discrete number of particle-to-particle contact points rather than to a planar surface. Often the particles themselves are hard and not given to plastic flow and deformation under the van der Waals force between them—unlike the walls of a sampling tube, which are usually made of materials that deform more easily, thereby increasing the contact area and the adhering force.1 Even without any deformation, the van der Waals force between two spherical particles of the same diameter is predicted to be about one-half that between a sphere of the same diameter and a plane (a sphere of infinite diameter).2, 3
Thus the particles in the top layers of the accumulated particle load on a surface have smaller forces binding them to their nearest neighbors than the particles in direct contact with the surface. However, these van der Waals interparticle forces can bind a large number of particles into an agglomerate so that when accelerated under the finger flick, an agglomeration of adhering particles gets accelerated as a single, large particle mass, which breaks free of its underlying neighbors at the points of lowest van der Waals force.
The removal force in the case of finger-flick acceleration is proportional to mass which varies as diameter cubed; the adhering van der Waals forces holding the particles of the agglomerate to each other vary only linearly with diameter. Thus the ratio of the removal force to the adhesion force increases as the mass of the agglomerate increases.
Shouldn't the particle counter classify these particle agglomerates as much larger particles than the single particles adhering to the surface? Not necessarily, because the agglomerates easily breakup in transit through collisions with the tube walls and with each other.
Dan Milholland (Pentagon Technologies Inc.) tells me not to worry about the practical consequences of the finger-flick phenomenon. It can be avoided altogether by suitable precleaning of the tubing used to collect samples. A liquid flush of such tubing proves sufficient to remove all particles vulnerable to the finger flick actions. Indeed the test for adequate flushing is to demonstrate that a finger flick produces no increase in the particle counts recorded by the particle counter attached to the end of the flicked tubing. Thus the finger flick observation becomes more than just a puzzle in understanding particle behavior. It's also a useful technique for verifying that one potential source of error in sampling has been eliminated.
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.
References
- Bowling, R. A., “An Analysis of Particle Adhesion on Semiconductor Surfaces,” J. Electrochem. Soc., September, 1985, pp 2208-2214.
- Hamaker, H. C., “The London-Van Der Waals Attraction between Spherical Particles,” Physica IV (10), 23 November, 1937, pp 1058-1072.
- Rumpf, H., “Particle Adhesion,” Chap. 7, pp 97-129 in Agglomeration 77, American Institute of Mining, Metallurgical and Petroleum Engineers, 1977.