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!

B IR spectra (in thermal scope vs IR spectroscopy)

  1. Jul 27, 2017 #1
    In thermal imager vs IR spectroscopy, they both look at the IR radiation of an object. What is the difference between the two.. I know IR spectroscopy look at the dipole moments of molecules. But in thermal imager.. do they also image the dipole moments of molecules? Also why you see images in the IR range of the thermal imager while you don't see any images in the IR spectroscopy? Could IR frequencies be involved such that if you reverse the two, you can see images in IR spectroscopy and see dipole moments in thermal scope? Thank you.
  2. jcsd
  3. Jul 27, 2017 #2

    Andy Resnick

    User Avatar
    Science Advisor
    Education Advisor

    Spectroscopy and imaging are, in one sense, complimentary optical methods. Crudely, imaging provides high spatial resolution and low spectral resolution while spectroscopy provides poor spatial resolution and high spectral resolution (imaging spectrometers and hyperspectral imaging systems notwithstanding...).
  4. Jul 27, 2017 #3
    Thanks for this idea that gives me clue how to compare them.
    Here's an actual data.
    A typical thermal imager has range of 7.5um to 13um wavelength (spectral range).
    A typical IR spectrometer has range of 350 cm-1 to 8000 cm-1 or between 28um and 1.25 um or 1.25um to 28um
    so a typical IR spectrometer has greater spectral range.

    for typical IR spectrometer, the spectral resolution is 0.5 cm-1.
    for typical thermal imager, spectral resolution is ___ .. I can't find the spec.

    I own a Flir i7 thermal imager. I don't use it often because when you use it on shiny surface, the reading is no longer accurate.. how come we don't have emissivity adjustment in IR spectrometer??
  5. Jul 27, 2017 #4


    User Avatar
    Science Advisor
    Homework Helper

    As has been pointed out by Andy that the data acquired by imaging is (mostly) of spatial type. Thermal imager maps the heat distribution from an object through the emission of infrared photons, it retrieves spatial distribution of heat on a surface that's why the knowledge about the spectral composition is neglected and hence its spectral resolution. If your thermal imager device does not output wavelength-related data, there is no need to specify the spectral resolution.
    As far as I know for IR spectrometers used for measuring laser bandwidth, you tune the source's brightness rather than the measuring device, e.g. by placing a tunable iris before the spectrometer.
  6. Jul 27, 2017 #5

    Andy Resnick

    User Avatar
    Science Advisor
    Education Advisor

    Again, thermal imagers do not have spectral resolution- they are (typically) broadband absorbers. IR spectrometers (for example, an FTIR) have dispersive elements (gratings, etc.) that allow pixels in the detector to sample a very narrow spectral 'slice'. IR spectrometers are more typically used to measure molecular vibrations: energy differences between vibrational states are in the IR. "emissivity adjustments" for an IR spectrometer doesn't make a lot of sense.
  7. Jul 27, 2017 #6
    what did you mean by "narrow spectral 'slice'"? But if you will see IR graph like the following:

    you will see that the IR spectrometer has spectral range of 500to 4000 cm-1 or about 2.5um to 20um which are even more than the spectral range of thermal imager which are usually about 7.5um to 13um. Maybe you meant something else by "narrow spectral 'slice"".. what is it? Thanks.
    Last edited by a moderator: Jul 27, 2017
  8. Jul 28, 2017 #7

    Andy Resnick

    User Avatar
    Science Advisor
    Education Advisor

    I mean each individual pixel of the detector records light within a very narrow spectral band. The detector (a 1-D array of pixels), 'slices' the broadband field into an array of narrowband fields.
  9. Jul 28, 2017 #8
    Thanks for that. I'm puzzled about something. IR spectroscopy and Raman spectroscopy are based on two distinct phenomena.. but how come the Raman scattering spectrum and infrared absorption spectrum for a given species often resemble one another quite closely in terms of observed frequencies? For example:

    Raman scattering spectrum are supposed to be shifted in frequency due to the Stroke and Antistroke mechanism, and this having to do with polarization of molecules while IR spectroscopy is from the dipole moment.. these two are distinct so how much the frequencies (as detailed above) resemble so closely??
  10. Jul 31, 2017 #9


    User Avatar
    Science Advisor

    Although the mechanisms of interaction are different (and by the way, it's Stokes, not Stroke), the frequencies are determined by the molecular vibrational modes. Energy can only be absorbed by exciting one of these modes. So the spectra often look similar, especially for molecules of low symmetry like SBR. Differences arise because (i) the intensity of a particular mode in IR and Raman is often different, because the probability of excitation of that mode is a function of the mechanism of interaction; (ii) in molecules of higher symmetry, the symmetry selection rules are different for IR and Raman, so you can have modes that are active in one but not the other (or even in neither). In molecules with a centre of inversion, there are no modes that are active in both, as Raman only excites modes that are symmetric with respect to inversion, and IR only antisymmetric modes, so the IR and Raman spectra look very different.
  11. Aug 6, 2017 #10
    Hyperspectral imaging systems mean it can detect the presence of individual spectra of any elements or compounds.. was this how the Mars Rover identified whether there was methane on mars? but why do they need to use laser on the sample?

    Can we use this to scan for any gunpowder residue on hands of anyone in public places (like airport) by using these hyperspectral imaging systems?
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted