296 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1958 
that around 97 percent of the salts of sea water can be removed in one 
pass through a membrane, such as cellulose acetate, but at slow rate. 
Investigation aimed at increasing the durability and flow rate of 
membranes is continuing. 
FREEZING 
Salt-water separation by freezing has been the subject of a number 
of experimental researches [86, 37, 38, 39, 40]. The use of freezing 
has certain inherent advantages such as a lesser tendency toward 
scaling and corrosion because of the low temperatures involved and the 
lower value of the heat of fusion as compared to the heat of vapori- 
zation. There is also the advantage of low-temperature differentials. 
However, the occlusion of brine with ice crystals has been a major 
difficulty in development of a feasible process. 
A promising approach to the utilization of freezing as a means 
of saline-water conversion and elimination of the brine from the 
crystals is a combination of freezing with evaporation being investi- 
gated by the Carrier Corp. of Syracuse, N.Y. Experimental appa- 
ratus and washing techniques have been developed so that now it is 
possible to produce practically salt-free ice from sea water in a 
continuous manner. 
Figure 5 illustrates a version of the process. Chilled saline water 
is admitted to a chamber under high vacuum. In this low pressure 
about one-seventh of the water flashes to vapor, further chilling the 
remainder which freezes to an ice-brine slurry. The slurry flows 
through a separation column for counter-current washing. The vapor 
formed in the freezing operation is compressed and condenses on the 
ice. The melted ice becomes the fresh-water product. Part of the 
product is used for washing the ice. 
The experimental program on this process utilizing a small shop- 
size pilot plant has not disclosed any technical problems that could 
not be solved or that would render the process impracticable. Opera- 
tion of the equipment is continuing and preliminary designs of a 
larger pilot plant of about 15,000 gallons daily capacity are being 
made. The next step is construction of that pilot plant incorporating 
all components of the complete process. 
In another approach to the use of freezing for demineralizing saline 
water, an immiscible refrigerant such as isobutane is vaporized in 
direct contact with the saline solution. The development of this prin- 
ciple is under way at Cornell University [41], Ithaca, N.Y. Because 
most equipment necessary for this process could be of compara- 
tively simple design, it may be particularly adaptable to large-scale 
installations. 
Some rather basic research has been conducted in an attempt to 
adapt the zone refining technique to convert saline water. A frozen 
