Historical Background to Liquid Mirror Telescopes.  By Peter Abrahams. 

   A liquid mirror telescope (LMT) uses a rotating basin of mercury to form a 
parabolic reflecting surface that is directed towards the zenith.  In the 
1990s, 2.7 and 3 meter LMTs were built, and a 6 meter LMT is under construction 
as of 2002.

   Isaac Newton described the use of rotating liquid to form a parabolic 
surface.

   Ernesto Capocci (1798-1864) at the Naples Observatory, wrote a letter to the 
Royal Academy of Sciences in Brussels, Belgium, read at a meeting 02 November 
1850, that contained the first published description of a mercury mirror 
telescope, but Capocci did not pursue the idea.

   Friedrich Wilhelm Christian Krecke (1812-1882), an astronomer in Utrecht and 
Rotterdam, built a liquid mirror with a mechanical drive.  Krecke had read 
Capocci's communication to Belgium and replied with an account of problems with 
ripples and of his development of the idea that molten metal could be 
solidified into a paraboloid surface.  T.W. Webb wrote in 'Nature' about 
Capocci and Krecke in 1872.

   David Brewster noted in his diary in June of 1857 that Leon Foucault had 
told him about an American, Mr. Buchan, who proposed a liquid mirror telescope 
and had deduced that a true parabola would only be obtained when located at the 
North or South Pole (due to Coriolis force, a minor factor in the design of 
large modern LMTs).

   George Perkins, in Utica, N.Y., wrote a paper on rotating mercury telescope 
mirrors, 'A Fluid Parabolic Mirror', published in Mathematical Monthly 1 (1859) 
79-82.  David Todd of Amherst referred to Perkin's work circa 1873-1874, 
proposing that the idea be tested before a decision was reached on the estate 
of James Lick.

   An unsuccessful, steam driven LMT built in 1868 by Robert Carrington (1826-
1875) in Frensham, Surrey, England, was the subject of a note by Robert Wood; 
but Carrington's papers do not describe this apparatus.

   Henry Skey (1836-1914) at Dunedin Observatory in Otago, New Zealand, built 
the first known working LMT, publishing papers in 1872 and 1874 with 
descriptions of his approximately 35 cm aperture telescope and calculations of 
focal length as related to angular velocity.  He exhibited the telescope to the 
New Zealand Institute on 19 November, 1872.  Skey developed two methods of 
driving the rotating basin, an electric motor with a regulator using a conical 
pendulum, and also a water-driven turbine.  Both provided good imaging quality 
and allowed him to change the focal length by changing rate of rotation.  A 
diagram of the telescope shows a plane mirror with a drive to allow observing 
away from the zenith.  Skey wrote in the 1874 paper that he had built an LMT 
telescope in England during the late 1850s, before emigrating to New Zealand in 
1860.

   Alonzo Jackman, librarian and professor of civil engineering, military 
tactics, mathematics, and natural philosophy at Norwich University, Vermont, 
wrote an unpublished and undated manuscript in the 1870s entitled 'A Novel 
Telescope'.  This paper was intended for the Norwich U. 'Reville' campus 
newspaper.  Jackman describes a shallow vessel or cistern of mercury, with a 
diameter of twenty feet, spun at a uniform velocity by 'suitable machinery' to 
produce a parabolodial surface.  The central portion of the mercury vessel was 
replaced by an elevating 'contrivance' with an observer's chair that could be 
carried to the focus of the optical surface.  This focal point could be 
adjusted by altering the speed of revolution of the vessel.
   Jackman's telescope design included large flat mirrors, to allow tracking of 
celestial objects away from the zenith.  The first mirror was placed 
horizontally, adjacent to the vessel of mercury, and was tiltable to be aimed 
at any star.  The second mirror was placed directly over the mercury vessel, at 
an angle to receive light from the first mirror.  Starlight would reflect off 
the first mirror, to the second, to the mercury, and thence to focus, 
presumably through a hole bored in the middle of the second mirror.  An 
observer's assistant would use a small telescope mounted on the first mirror to 
aim the system for the primary observer.  Jackman's manuscript is incomplete at 
this point, possibly because he realized that the difficulties of implementing 
this design might be insurmountable; the second mirror would be a flat twenty 
feet in diameter, the first mirror would be much larger.  It seems that either 
Jackman was misguided or possibly a modern account is mistaken in interpreting 
Jackman's incomplete manuscript.  However, the design could be implemented, if 
only a portion of the full aperture of the mercury vessel was used.

   Physicist Robert Wood (1868-1955), at Johns Hopkins University, conducted 
the first extensive tests of liquid mirror optics for a telescope, publishing 
three papers in 1909 on his attempts to build a functioning astronomical 
research instrument.
   A 7 inch mercury mirror was placed in a empty barrel to eliminate air 
currents.  To isolate the mirror from motor vibrations, a ring of magnets was 
attached to the bottom the basin, and a ring of magnets was attached on top of 
a wheel driven by the motor, set just underneath the basin so the lower ring of 
magnets could engage those attached to the basin.  Other ripples were found to 
be from the bearings used on the revolving shaft of the mirror; out-of-level 
conditions which caused larger waves in the mercury; and the most intractable, 
variations in angular velocity of the rotating basin.
   A 20 inch LMT, drawn by Wood, adapted by W. Warner, and built by Warner & 
Swasey, was installed in a 4.3 meter deep pit on Wood's Long Island country 
property, to reduce vibrations from city traffic & factories.   Built in the 
lab, it had a focal ratio of f1.7 to f3, and as used in the pit, it was f9.  A 
drive belt made of a quantity of thin rubber cords was found to be simpler & as 
effective as the magnets.  This prototype was successful in eliminating ripples 
except those caused by the AC motor.  Using this telescope, Woods viewed the 
Milky Way and the Andromeda nebula without an eyepiece, and observed stars 
using an eyepiece.  He discovered an unexpected effect, a periodic fluctuation 
in the focal length of the mirror of plus or minus 2 cm, caused by changes in 
rotational velocity.  He could resolve double stars with separations as small 
as 2.3 arcsec, the first astronomical observations made with an LMT.  To 
further examine sources of vibration, a 5 inch Clark refractor was mounted at 
the rim of the pit, pointed at the mercury, and reflected stars were examined.  
With the motor running, no vibrations were detected; but detectable effects 
were caused by human footsteps at 50 yards, a horse & carriage at one eighth of 
a mile, and surf after a storm at one quarter of a mile distance.
   A vivid description of viewing with this telescope is provided by Wood: 
'especially interesting was the slowness with which the velocity of the dish 
was communicated to the fluid....first at the rim....crawling in gradually from 
the edge toward the center....The appearance of the room as viewed in the 
mirror while it is getting up to speed is very striking.  The rafters of the 
roof recede to an enormous height and the whole room appears to expand in a 
most remarkable manner.'
   Very precise adjustments were required of the leveling of the rotating 
basin.  If not exactly level, the out of focus image of a star showed a type of 
coma.  Photographs of star trails were used for further diagnostics of the 
fluctuations in the drive system; instead of the trail being a thin line, it 
was a series of 1 millimeter blobs with a fine dot between each - this from 
changes in focal length caused by changes in rotation velocity.
   Wood cast a resin on top of the mercury, producing a convex paraboloid, and 
considered replicating the mercury and coating the hardened resin with metal.  
Small mirrors were made with a focal ratio as low as 0.25.  Further research 
involved layers of castor oil or glycerine on top of the surface to dampen 
vibrations.  These were transparent enough to avoid image degradation and 
suppressed ripples after 1 to three wavelengths of travel.  Wood considered 
using a very thin layer of mercury on a shaped copper vessel.  Successful 
elimination of all surface waves was never achieved.  This, combined with the 
limits of observation at zenith only, caused Wood to move on to other fields of 
inquiry after three published texts.

   In the early 1920s, 'B.A. McA.' proposed a 50 foot aperture, f5.6 to f12 
LMT, to be used at the bottom of a mine shaft in Chile during the 1924 
opposition of Mars.  This design was apparently completely unworkable, and was 
denounced in a 1922 article in 'School Science and Mathematics'.  

   In the 1960s, liquid epoxy was cast into a paraboloid by rotating a basin of 
the material as it set, to be used in radio and infra-red (Cal Tech 62 inch) 
astronomy.  Molten glass was cast into a paraboloid in the 1980s and 1990s, 
most ambitiously in the 8 meter glass mirrors from the University of Arizona.

   Vasil'ev, at Khar'kov State University in Russia, circa 1985, built a 14 cm 
LMT using castor oil as a reflecting surface.  To dampen vibration, the mirror 
basin was placed in a rotating IDL (intermediate damping liquid), water that is 
itself spinning in a larger vessel.  The inner & outer vessel attain 
equilibrium and spin at the same rate, and the inner vessel centers itself and 
levels itself.  The very even support provided by floating in liquid helps the 
designer when attempting large instruments.

   In 1982, Ermanno Borra of Laval University, Quebec, published the first of 
many papers that led to a major expansion of LMT projects.  In the mid-1980s 
into the 1990s, telescopes built at Laval and at the University of British 
Columbia by Borra, Paul Hickson, and Luc Girard, included a 1 meter f /1.6, 1m 
f /4.7, 1.2m f /0.89, 1.2m f /4.58, 1.5m f /2, 1.65m f /0.89, and 2.7m f /1.9.  
Paul Hickson led an effort to design & build 3 meter LMTs, for UBC, NASA, and 
UCLA.  A 6 meter Large Zenith Telescope is under construction at UBC.
   New technology included using oscillator stabilized power to synchronous 
motors driving air bearings, and Null Offner corrector lenses.  By 1989, the 
Airy disc of the 1.5m f /2 mirror was observed.  Diffraction rings allowed 
clear observation of particular problems, leading to further improvements until 
the 1.5m f /2 liquid mirror was judged essentially diffraction-limited, 
confirmed by interferometry.  Borra writes that interferometric tests have 
shown LMT surface quality as good as 1/30 wavelength of light.

   LMTs were found to be useful for optical testing, because of the high 
quality parabolas formed by rotation.  LMTs have been used for LIDAR studies of 
the atmosphere.  NASA's Orbital Debris Observatory near Cloudcroft has a 3 
meter f1.75 'Liquid Mercury Telescope' built by NASA & Lockheed, used to track 
space debris.
   These modern instruments can simulate tracking by using CCD chips and 
computers in 'time delay and integrate' mode.  However, zenith instruments have 
a 'Time Delay Integration' (TDI) distortion.  Stars rotate around Polaris, and 
in a photograph form arcs instead of straight lines, but CCDs read out in 
straight rows and columns. 
   LMTs will not replace glass optics, but they do have some advantages.  Costs 
per unit aperture are far lower.  Many complex systems are eliminated, 
including elaborate mirror support systems, tracking systems, and tube 
stiffening designs.  Seeing conditions are optimal at zenith.  LMTs can have 
excellent quality optical surfaces, very low scatter, very low or very high f 
ratios, and a precisely controlled variable focus.

   Modern research includes minimizing the quantity of mercury, using layers of 
mercury as thin as one half mm.  Since mercury is very heavy and very toxic, 
other materials have been explored.  Gallium melts at 30°C, but can be 
supercooled to remain liquid at lower temperatures.  Gallium-Indium melts at 
16°C, is 2.2 times less dense than mercury, and has a reflectivity superior to 
mercury.   Finally, a liquid mirror in space under investigation.  Thrust can 
act as gravity does with a terrestrial LMT.  In space an LMT is not limited to 
any one orientation and weightlessness would eliminate elaborate bearings.  

=========
References:

Gibson, Brad.  Liquid Mirror Telescopes: History.  Journal of the Royal 
Astronomical Society of Canada 85:4 (August 1991) 158-171.

Olsson-Steel, Duncan.  A Note on the History of the Liquid Mirror Telescope.   
Journal of the Royal Astronomical Society of Canada 80:3 (1986) 128-133. 

Parker, Gary.  Alonzo Jackman's Design for 'A Novel Telescope'.  Vermont 
History 50:1 (1982) 13-22.

Wood, R.W.  The Mercury Paraboloid as a Reflecting Telescope.  Astrophysical 
Journal 29 (1909) 164-176.

Wood, R.W.  The Mercury Telescope.  Scientific American 100 (March 27, 1909) 
240-241.

=======================
home page:    http://home.europa.com/~telscope/binotele.htm