Solidifying space mirrors in space?

Not open for further replies.


Tell me if this is feasible. Liquid mirrors may cost only 1/100th the
cost of a comparable solid mirror:

Friday, June 22, 2007
Molten Mirrors.
Liquid mirrors could enable more-powerful space telescopes.
By Katherine Bourzac

A disadvantage is that they must be pointed up because they need
gravity pointing downward along their vertical axis to operate. Still
their simplicity and low cost is what led their being proposed to be
put on the Moon. The above article is actually about putting such a
mirror on the Moon because the liquid mirrors need gravity to
But couldn't we just form the parabolic shape of the mirror in space
by rotating a molten substrate while at the same time creating the
gravity by accelerating the mirror by a propulsion method? We would
then let the mirror cool so that we would wind up with a solid mirror
that no longer needed to be rotated or accelerated to hold its shape.
The advantage of this is that after the acceleration is cut off the
mirror would be in zero gravity and therefore would not have to have
the thickness required to hold its shape as for mirrors on Earth. Then
we might be able to get mirrors of much greater size then for current
Earth bound mirrors. We could also then point it in any direction
because it would be a solid mirror.
I was thinking about this first for glass mirrors since rotating
molten blanks is how large mirrors on Earth are currently formed:

Making a Giant Mirror to Scour the Skies.
by Ted Robbins
All Things Considered, July 27, 2005. ... Id=4773461

As described in this article, the building holding the mirrors is two
stories tall and the glass weighs 20 tons. However, it may be this can
be shrunk in the zero gravity environment of space. The glass has to
be heavier for an Earth mirror because it has to hold its shape after
the rotation and after it is allowed to solidify. This wouldn't be the
case for a space mirror so its mass would be much less. Therefore the
structure holding it probably also could be much smaller.
However, as indicated in this article you need three months for the
glass to solidify so you would need to provide the acceleration for
this length of time. However, it probably is the case you could make
the acceleration much smaller than 1 g for this to work. On the Moon
for instance it's only 1/6 g. Still though you would need a great deal
of power for the heating elements at the temperature required to keep
the glass melted.
Instead could we just use mercury for the substrate? The temperature
could be even less than 0 C for the mercury to become liquid. Then
when we cut off the heat the mercury would rapidly solidify in the
cold of space, presumable maintaining it's parabolic shape in zero g.
So you wouldn't have to provide the acceleration for a great length of
time, perhaps only hours or days.
A couple of problems. If the mercury were exposed directly to space
at near zero pressure it might boil or evaporate off despite the cold
temperature. So you might have to provide some background air pressure
for it. You could have a very thin transparent cover to maintain the
air pressure. Likely the pressure required would not have to be very
high so we could make the cover very thin. Also, if you pointed the
mirror too close to the Sun the mercury would rise in temperature
again to melt. You would avoid this but avoiding looking in the Suns
direction during observations. This is not that severe a limitation.
Hubble has to do the same thing because of its sensitive optics.
A potentially severe problem though is whether or not the mirror
would need polishing after it solidified. The glass mirrors for
example require a year of polishing after they solidify. It's not
clear if the mercury mirrors would require polishing after they
solidify. They obviously don't require it as liquid mirrors on Earth.
It may be possible to do the polishing using some type of automated
nanometer-scale deposition method. For instance, this method allows
deposition at 100 nm accuracy:

Versatile Nanodeposition of Dielectrics and Metals by Non-Contact
Direct-Write Technology.
H.D. Wanzenboeck, H. Langfischer, S. Harasek, B. Basnar, H. Hutter,
and E. Bertagnolli
Vienna University of Technology, Austria ... ion=detail

Using the recently developed "superlenses" it might be possible to do
better than this since they allow microscopy at subwavelength

Bob Clark


This report suggests using an ion engine to create the required
acceleration to generate the parabolic shape on rotation:

Space Based Liquid Ring Mirror.
Jon A Magnuson, Ph.D/Boeing
Duncan C Watson, Ph.D/Boeing
Robert J States, Ph.D/Boeing ... gnuson.pdf

The authors suggest having the acceleration operate continuously and
having the mirror remain liquid. The Deep Space 1 mission suggests
this should be possible for an ion engine.
They argue that only a 1/1000th g acceleration would suffice for a 30
meter liquid mirror. However, to accelerate a liquid mirror this size
and its attendant mass would require an ion engine of perhaps two
orders of magnitude higher thrust than the Deep Space 1 engine.
On the other hand if the engine only had to operate for a few hours
or conceivably only for minutes considering that the liquid layer
might only need to be 1/2 mm thick, then we could use commonly
available chemical propulsion methods.
For the roughness of the surface on solidifying, it seems reasonable
that this is partly due to the material's vapor condensing on the
surface. This for example is what happens when ice freezes.
Perhaps then we could keep the vapor pressure very low to prevent
this from happening. This could be done by making the entire
surrounding pressure low or by removing only the materials vapor above
the surface.
However, very low surrounding pressure or vapor pressure would make
the material as liquid rapidly evaporate or boil off. This might be
alleviated though by including solutes that lower the melting point.
These have the property that they also lower the vapor pressure
required for the material to remain liquid. This would have the
additional advantage of decreasing the amount of power for heating
that had to be used to keep the material liquid.
Another contributing factor in the roughness of the surface on
solidifying is undoubtedly unevenness in temperature at different
positions on the surface. The thinness of the layer would help to
limit this. We might also limit this by using very sensitive thermal
infrared imaging at high resolution on the surface during the cooling
process and using lasers to heat up portions at the nanometer scale to
have a uniform temperature across the surface.
After a web search I also found an example of ice with a mirror like
smoothness that occurs naturally under specialized conditions called

Also: Firn Spiegel, firn mirror, glacier fire
"An ice crust formed on the snow surface on sunny, cold days. The
sun's heat penetrates the surface snow layers and causes melt around
the grains beneath the surface but meltwater at the surface is
refrozen forming a thin layer of ice. This requires just the right
heat balance.
Firnspiegel is highly reflective and mirror-like, giving it the
colloquial names of firn mirror and glacier fire.. " ... piegel.php

This might give us clues on how to accomplish this for our mirrors on

Bob Clark


You've got to look at the practical engineering first.

What I envision is series of segments (hexagon, like the Keck?), slowly assembled on the ground, and spot-on perfect when they're lofted / transported. Then cover the hexes with a highly reflective material, perhaps a nano-material. From their, use good, real-time data processing to keep the mirror in shape (flexing due to temperature differentials, etc.).

Keck, without the intervening atmosphere, in short.
Not open for further replies.