Practical Machinist Forum of American Machinist

Mercury/Water Binary Cycle

Toshiyori posted 08-28-2006: In the 1920's & 30's GE designed & marketed a Mercury cycle for power generation. I have descriptions of the cycles, I am looking for "Heat Balance Diagrams" for the cycles.
Anyone out there have such a thing? Or can direct me to it?
Thermo1 08-28-2006: I can give you a few particulars, and point you to one obscure reference, and one very obscure reference.

There were 5 mercury steam plants: one experimental plant at Dutch Point Station in Hartford, Connecticut, a full size plant at the South Meadow plant in Hartford, full size plants at Shiller Station near Portsmouth NH, and at the Kearny (NJ) Station of Public Service Electric and Gas Co. I believe there was also a plant at the GE works in Schenectady, NY.

The following is from Connecticut, the Industrial Incubator, 1982, published by the History and Heritage Committee of the Hartford section of the Americian Society of Mechanical Engineers. You might contact them to see how to get a copy of the relevant pages of the book.

This is Note 855, page 98.

There is more data on the earlier, experimental plant, built in 1923, than there is in the later plant, so I will give data on that first. In the 1923 plant, 20 tons of liquid mercury was heated to mercury vapor at 35 psig, and 812F. It condensed at 1.5 psia, and 485F. It generated steam in excess of 200 psig to drive steam turbine. To improve efficiency, the steam was reheated in the mercury boiler to about 700F.

The South Meadow plant (built in 1928) had a 10,000 kW mercury turbine with 5 stages. There is no data on the temperature of the mercury vapor on this plant. The mercury was condensed in two condenser/boilers that generated 129,000 pounds of steam per hour, at 280 psi and 700F (persumably the steam was reheated in the boiler).

The plant ran from 1928 to 1947. In 1947, the mercury unit was rebuilt, and the generator rewound for 15,000 kW.

The plant had a heat rate of 10,200 btu/kwhr, which is probably 10% to 20% better than any steam plant built in 1927. There is a mention of a new boiler in 1947 where 70% of the available furnace heat was absorbed in the convection region of the furnace side walls and back wall. This does not state what the actual boiler efficiency was.

As near as I can figure, the mercury steam plant cycle was heating mercury in a boiler, expanding it in a turbine, and condensing it to make steam. The steam was superheated in the boiler, and expanded through another turbine. It is likely that there were two or three stages of feedwater heating for the steam turbine.

From this data, you might be able to rough out a heat balance. I have no idea what the efficiency of the mercury turbine would be. Steam turbines of the day were probably 65% to 70% efficient. The boiler was probably 80% efficient.

The other reference is cited in the first:
Douglass & Hackett, Modern Mercury Power Plant, ASME Power Division Meeting, New London, Connecticut, May 1949.

You might see if General Electric has any historical material available to the general public.

08-30-2006 I find that the 6th edition (1958)of Marks Mechanical Engineer's Handbook has a table of thermodynamic properties of mercury vapor. Later editions do not have the table, so you will have to look through libraries to find the old edition. Incidently, the table was put together by L.A. Sheldon, of the General Electic Company. Hope this helps.
I hope this is of some help. Good luck on tracing the references.

Toshiyori 08-31-2006: My Marks is the 7th edition. I contacted GE, no response, I am now requesting books from the Library of Congress one such as: "The possibilities of mercury as a working substance for binary fluid turbines". This book just may have some heat balance info. There are others at the Library of Congress.
surplusjohn 09-01-2006:This is interesting, seems like a good artical for "history of Invention and Technology". Persumably this was more efficient and was promissing enought to build a few plants but then was abandonded, Why? Toxic issues are certainly a concern, tons of mercury vapor in my back yard? yeah right! Does not seem like that was an issue back in that era. In this area we are dealing with the clean up of tons of mercury that leaked from Allied Chemicals mfg of chlorine from the 40s to the 80s. There is a 20-30 acre site that is being encapsulated in clay and Onadaga Lake's setiment is full of HG so bad that fish are not to be eaten. I often wonder about the HG Io played with when I was a kid. My Grandfather had a wooden jar of it [it was probably a glass jar inside a wooden jar, but I just remember the wood] that was too heavy to pick up easily, we played with that more than once. I also wonder where that disappeared to! 09-06-2006 I passed this discussion onto History of Invention and Technology as a potential artical. http://www.americanheritage.com
ttok 09-01-2006: I seem to recall that the use of mercury in a power plant cycle, while it was more thermodynamically efficient, was impractical from a metal erosion standpoint. Maybe the mercury metal eroded the metal container (boiler and heat exchangers) too fast to be economical. Don't know for sure, though. What did they use to store it in? We used to have it in 760-pound steel containers called "flasks" - the usual method of transporting it. But it was at room temperature in these. What alloy I do not know. Much erosion of steel at elevated temperatures??
Toshiyori 09-06-2006:To date, with help from persons such as yourself & others, I am roughly 50% toward a "Heat Balance Diagram" for a "Mercury Binary Cycle". 09-07-2006:The mercury cycle does, in fact, still exist, at least in design considerations. An article in: "Energy Citations Database" dated 6/3/01 describes a mercury boiler & turbine/alternator designed & developed for producing electricity for extended periods in satellites. The system was designated: SNAP-1. I don't know if SNAP-1 ever got off the ground or not! When I search for "SNAP-1 on the enternet, I get x rated returns.
Thermo1 09-06-2006 07: A couple of quick comments about the mercury water steam cycle. It was developed to overcome the limitations of power plant cycles as they existed in the 1920s. Mercury had two big advantages, the pressures were very low at high temperatures, so the stresses in the piping were low; and the amount of heat required to heat liquid mercury up to boiling temperature was a smaller fracton of the total heat input than it was for water.

By the late 1940s, and certainly by the mid 1950's, materials and cycle designs had improved to where it was possible to build a conventional steam plant with better efficiency than the mercury steam cycle. By 1959, supercritical power plants (where the steam pressure is abouve the critical pressure of 3200 psi) had been built, with efficiencies 20% better than the mercury water cycle. At this point, there was nothing to be gained by investing in improvements to the mercury cycle, when advanced steam cycles did so well. 09-07-2006: I have vague memories of power units designated SNAP a very long time ago. I think I remember that they used the decay heat of radioactive materials to supply heat to a power cycle. A mercury plant would be great in space, as it could operate with a high condensing temperaure, as necessary for radiative cooling, and the working fluid would not freeze up.
From: http://www.practicalmachinist.com/cgi-bin/ubbcgi/ultimatebb.cgi?ubb=get_topic;f=15;t=000802;p=0
Found: 11/21/06