SPERRY-RAND IN HOT WATER

Another method of tapping geothermal energy is to pump water down a hole deep enough to reach hot rock, and to harness the resulting steam. The Los Alamos Labs, for example, are investigating a variant of this method where only an initial cavity would have to be created; the cold water would further fracture the hot rock, and the whole process is hoped to be self-perpetuating: While the heat is being extracted from one cavity, another becomes available as the first is cooled. The energies to be gained from these methods are gigantic: The heat contained in a block of hot granite 4 by 2 by 5 miles equals the annual US energy consumption.

The most easily reached sources of geothermal energy are the numerous deposits of hot brine, usually between 1,000 and 8,000 feet below the surface.

One way to solve the fouling and corrosion problem is to use a heat exchanger in which the brine transfers its heat to clean water (for generation of electricity) without physically coming into contact with it. But this leads to other problems: The primary side of the heat exchanger (the coils carrying the brine) scale up rapidly, impairing the heat transfer and causing energy losses. Still, the method is a reasonable compromise, for periodic replacement and cleaning of the primary side of the heat exchanger is not an insurmountable problem, and the efficiency percentage is still in the upper sixties, almost twice as good as that of the average fossil-burning power plant.

This sounds too good to be true, and indeed, in most cases it isn't. For there is an additional large energy loss in the case of natural brine deposits. Geothermal steam, when given an outlet, comes up all by itself; but if it's just hot brine, it has to be pumped to the surface, using up part of the energy. For hot brine deposits, therefore, efficiency is the critical issue.

Research engineers of the Sperry Rand Corporation have now come up with an idea that may substantially increase the efficiency of tapping hot brine deposits. They still have to pump the brine up, of course, but the trick is to power the pump by the heat of the brine directly, rather than by electric power which suffers additional conversion losses and turns out to be an unnecessary intermediary.

It works like this: A boiler and superheater, steam turbine and pump are placed at the end of a pipe which is lowered in the well casing until submerged in the hot brine. Clean water pumped from the surface is converted to steam by the heat of the brine (which has a higher boiling point than clean water) and drives the turbine which, in turn, drives an impeller to pump the hot brine to the surface.

The Sperry Rand system thus has three water circuits: The brine up from the well, through a heat exchanger on the surface and down into a reinjection sink; clean water down the well, through the boiler and turbine, up the same well and into a condenser and a feed pump on the surface; and a conventional water-steam circuit for power generation, entirely on the surface, through the secondary side of the heat exchanger.

If the brine is not hot enough to turn water into steam, presumably (we could obtain no confirmation) a working liquid with boiling point lower than water can be used.