polluted water. Three modular designs are being 
tested at plant sizes ranging from 1,000 to 5,000 
gpd. It is believed that this process will have 
commercial application in sizes suitable for major 
municipal and industrial use in the next two to 
three years. 
a. Reverse Osmosis The unique membranes 
utilized in the reverse Osmosis process resist the 
passage of most dissolved contaminants. As a 
result, the process promises to be useful in a 
variety of applications. The rejection of ordinary 
salinity, the major sea water contaminant, and of 
hardness, scale and alkalinity factors predominant 
in many brackish waters make its use obvious for 
such purposes. Furthermore, the membranes hold 
back organic matter, including detergents, a major 
constitucnt of waste water. The membrane also 
rejects bacteria and virus so that the product is 
sterile. Mine drainage water also may be purified 
by this process as well as water contaminated by 
chemical, bacteriological, and radioactive 
agents.** 
One major problem in this technique is the 
development of longer-life membranes. 
Virtually all potential advantages of the reverse 
osmosis process derive from its room temperature 
operation. No intermediate formation of steam or 
ice is required which, for the distillation and 
freezing processes respectively, leads to the ex- 
penditure of large quantities of energy. Reverse 
osmosis, on the other hand, requires only pressuri- 
zation energy, and the energy cost is relatively 
low. Furthermore, there is practically no scaling 
and little corrosion resulting in low maintenance 
costs; these advantages combine to promise an 
extremely low total operation cost.** 
OSW has defined 10 types of brackish water 
typically found in the United States. It plans to 
develop data concerning the best process for the 
particular type of water. 
b. Electrodialysis Figure 65 is a photograph of 
the OSW Test Bed at Webster, South Dakota. It is 
rated at 250,000 gpd and uses the electrodialysis 
principle. Electrodialysis is currently in use at 125 
44 Senate Hearings, May 1965, testimony by Dr. B. 
Keilin, Aerojet-General Corporation, p. 81. 
4° Thid., p. 81. 
VI-206 
installations around the world, according to Senate 
testimony in 1965.*° It was further stated that: 
~ 
Figure 65. Electrodialysis test bed at Webster, 
South Dakota. In operation since 1962, plant 
converts brackish well water to 250,000 gal- 
lons of fresh water daily. (Office of Saline 
Water photo) 
Electrodialysis has the great virtue of being a 
simple process with high operating reliability. It is 
good for small towns in that the conscientious 
personnel available in small towns can do the job 
easily as shown in Buckeye and in Coalinga, Calif. 
In Buckeye, the operator spends a good bit of his 
day with other duties, such as repairing cars and 
trucks. Constant attention to the plant is not 
needed and no attention is provided during the 
two night shifts. Since it does not involve compii- 
cated, high-pressure equipment, but electricity, the 
local utility can be called on for help when 
needed. 
Furthermore, the only sure, long-range source of 
water for most communities will be water that is 
reused. Water once used by a community has its 
mineral content increased by 300 to 400 parts per 
million. Normal water treatment plants remove 
suspended and organic matter, but not the miner- 
als. Mild salinity of this order of magnitude is ideal 
for economic processing by electrodialysis. 
Although electrodialysis works well in Arizona 
and South Dakota, it requires a substantial chem- 
ical engineering effort to learn how to pretreat 
these brackish waters so that the process operates 
efficiently. Thus, if iron manganese is present in 
46 Senate Hearings, May 1965, testimony by R. L. 
Haden, Jr., President of Ionics, Inc., pp. 130-131. 
