June 18, 1941
Dr. James B. Conant
National Defense Research Committee
1530 P Street NW
Dear Dr. Conant:
I was very glad to receive your letter of June 13, since I have plenty of time to think about things and like to
have a problem to work on. I wish to suggest a method for determining particle size in fogs. So far as I know, the method
is new, and after making some calculations I think that it should be satisfactory. We ought to be able to build an apparatus
which will register with considerable accuracy particle size and particle-size distribution in a fog such as described in
The proposed method involves giving an electric charge to the particles or to a fraction of them and then measuring
the rate at which they are attracted to a condenser plate. The rate at which particles with a given charge are pulled to a
condenser plate varies inversely with the radius of the particle, as shown by Stokes' law. Two methods of distributing the
charge on the particles might be used. The first would involve placing the same charge on all particles or on a representative
fraction of them. The second would involve charging the particles to extents determined by their sizes. I think that the second
method might be the better one. If electrons or ions were passed through the fog in such a way that each particle received
a charge proportionate to its effective cross-sectional area, the large particles would then move faster in the field than
the smaller particles. Some study of the method of charging the particles most effectively should be made.
After the particles had been charged, a stream of the charged fog would be liberated in the middle of a stream of
air moving non-turbulently between the plates of a condenser, with plate separation perhaps 2 centimeters and field of a few
thousand volts per centimeter. The charged particles would then move toward one plate from their original level at rates proportional
to their charges and inversely proportional to their radii. They would strike collectors, placed in sequence along one of
the condenser plates. By amplifying the currents flowing from the different collectors the size and size distribution for
the particles in the fog would be found.
I have made various order-of-magnitude calculations which look favorable to the success of the method. With a few
thousand volts between the condenser plates the Stokes' law speed of the particles is of the order of 1 centimeter per second,
which is reasonable. The current reaching each collector plate is also measurable, being such that it could easily be measured
with a one vacuum tube amplifier, F.P. 54. For field work I would suggest a simple apparatus with three or five collectors
and with variable condenser voltage. Probably one amplifier could be used for all of the collectors, being shifted along the
series to make the readings. For careful laboratory work an instrument with more collectors might be useful.
If your study of the field has shown that an instrument of this sort has not been built, I think that it would be
worth while to follow up this idea. I could, I think, arrange to have the work done here.
My address from June 23 to July 18 will be Chemistry Department, University of Chicago, Chicago Illinois. I am leaving
here Friday night, June 20.
With best regards, I am