1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Reverse solar trap

  1. Jun 11, 2012 #1
    I've been thinking about this on and off for awhile now. It first started with the idea of a solar trap.

    Solar trap being something like having a solar collector focusing light on a container of water, having an infra red barrier around the container, and a black body at the centre. Incoming solar energy passes by the IR reflector, absorbed by the black body and re-emitted as radiation which is reflected by the IR reflector.

    I was pleasantly surprised some form of this technique is used in the high efficiency solar water heaters.

    So what stops one doing this in reverse? having a visible reflector, IR transmitting layer.

    Could this allow solar cooling during the day? It would of course depend on the net flux of radiation, perhaps looking at natural sunlight spectrum there maybe a band strongly absorbed by the atmosphere that is within the range put out by a black body.
  2. jcsd
  3. Jun 11, 2012 #2


    User Avatar
    Gold Member

  4. Jun 11, 2012 #3
    found this


    Germanium seems a good candidate, for a better view

    http://rmico.com/technical-notes/bk7-quartz-ge-si [Broken]

    in particular


    based on this calculator

    peak radiance of a 300 K object is 10 μm


    there's some absorbance around that wavelength
    Last edited by a moderator: May 6, 2017
  5. Jun 11, 2012 #4
    So, after a bt of googling. This turns out to exist.

    A cold mirror


    Although, they don't quite cover the spectrum that would be optimal, the limit currently for commerical mirrors is up to 2500nm,This is usually layered dialectric coatings.

    really want that to be transmissive between 2500nm and 15000nm.

    Germanium does a good job, although the ROI would be terrible, silicon, which does about 1600-7000nm would be ok, though im guessing cost wise it wouldn't be much better.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook