Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: [Phys-L] absorbers, selective and otherwise; was: potatoes



On 01/12/2015 10:58 PM, Bernard Cleyet wrote:
I suspect the “roughness” is too fine for the 6 + micron to "notice it”.

That's true. Such suspicions are easily confirmed. We can
do the physics.

Executive summary: It is hard to make anything thin and black.
It's also hard to make anything thin and white
https://www.av8n.com/physics/white.htm
but let's leave that aside for now.

Suppose instead of black we have shiny. The wave comes in from
the left, hits a piece of perfect conductor, and reflects. There
is no field inside the conductor:
______________
| |
~ ~ ~ ~ ~ ~| mirror |
|______________|


If you try to make a perfect sudden absorber, what you get
is indistinguishable from a mirror. There is no field inside
the absorber:
______________
| |
~ ~ ~ ~ ~ ~| absorber |
|______________|

The point is, a perfect sudden absorber wouldn't absorb at
all; it would reflect just as much as the perfect conductor.
This is the only possible solution to the wave equation.

You can understand this in terms of Huygens-style radiators
on the surface of the object (mirror or absorber). They
radiate a something that cancels the field inside the object.
However they also radiate in the opposite direction, which
gives us the reflected wave. There is nothing you can do
with a /thin/ layer of radiators that will give a different
result.

Now if you have a layer λ/4 thick, you can fiddle with the
phases so that it radiates rightward but not leftward. This
works only for normal incidence, and only for one wavelength,
but it's a start. If you want it to work for a range of angles
and a range of wavelengths, you need even more thickness.

This is basically the same physics as for designing a high-
gain directional antenna for radio, TV, GPS, cell phones,
et cetera. If you make it small compared to the wavelength,
it cannot possibly be directional. The wave equation is
quite tyrannical about this.

Contrast this with (say) light bouncing off the moon, where
the roughness is thousands or millions of times larger than
the wavelength. In that case, sure, the texture has a big
effect on the emissivity.

===================

There are such things as "selective surface" materials,
notably stuff that is black at optical wavelengths and
shiny at thermal wavelengths. This stuff is somewhat
expensive to make, but it useful e.g. to put at the
focus of a solar collector for driving a heat engine.
https://en.wikipedia.org/wiki/Selective_surface

It would be nice if off-the-shelf aluminum foil would
act as a selective surface, but it doesn't ... not to
any appreciable degree.