2

I am currently studying the textbook Infrared and Raman Spectroscopy, 2nd edition, by Peter Larkin. In a section entitled Infrared Absorption Process, the author says the following:

The typical IR spectrometer broadband source emits all IR frequencies of interest simultaneously where the near-IR region is 14,000 - 4000 cm$^{-1}$, the mid-IR region is 4000 - 400 cm$^{-1}$, and the far-IR region is 400 - 10 cm$^{-1}$.

I know that lasers are approximately, more or less, monochromatic. Obviously, wavelength is not the same thing as frequency, but the idea that a light source could simultaneously emit a broad range of frequencies made me uncomfortable, given the aforementioned fact about monochromaticity.

In doing research, I found the following explanation (see 10.6.2 Sensing Mechanisms):

Optical sensors are typically interfaced with an optical module, as shown in Figure 10.38. The module supplies the excitation light, which may be from a monochromatic source such as a diode laser or from a broadband source (e.g., quartz-halogen) that is filtered to provide a narrow bandwidth of excitation.

It seems to me that this explanation confuses the concepts of "broadband" and "monochromaticity", or it at least writes in such a way that makes it seem so:

The module supplies the excitation light, which may be from a monochromatic source such as a diode laser or from a broadband source (e.g., quartz-halogen) ...

I would greatly appreciate it if people would please take the time to explain how a light source can be "broadband" (that is, emit a broad range of frequencies), and clarify what the above explanation is saying

The Pointer
  • 849
  • 1
    Are you familiar with blackbody radiation? Thermal emission is broadband (multiple frequencies at once). An incandescent lightbulb is like this. It seems to me that the references you provided make a proper distinction between monochromatic emission from laser diodes, and broadband emission from thermal sources. – kleingordon May 13 '20 at 16:27
  • Note also that the quoted sentence that mentions the quartz-halogen source indicates that its broadband emission must be filtered in order to make it behave more like a monochromatic source - without the filter, it doesn't have that property – kleingordon May 13 '20 at 16:29
  • 1
    @kleingordon Ahh, yes, I forgot about blackbody radiation and thermal emission. With regards to the reference, in light of your response, I think I was misreading it. Thank you for the clarification. Feel free to post an answer, so that I can accept it and close this question. – The Pointer May 13 '20 at 16:34

2 Answers2

2

Thermal emission (often idealized as blackbody radiation) is broadband, i.e. emitting multiple frequencies at once. An incandescent lightbulb is like this. The references provided in the question make a proper distinction between monochromatic emission from lasers, and broadband emission from thermal sources.

The quoted sentence that mentions the quartz-halogen source indicates that its broadband emission must be filtered in order to make it behave more like a monochromatic source - without the filter, it doesn't have that property.

kleingordon
  • 6,634
2

There are two possible routes to making a broadband light source.

  1. Incoherent sources

    The first way is to make a source that emits light at a bunch of different frequencies without any coherence between those different frequency components. The easiest way to do this is via thermal sources (like the quartz-halogen lamps mentioned as examples in your quote) but there are also other mechanisms such as e.g. fluorescence.

  2. Coherent sources

    Or, in other words, pulsed lasers. To be blunt, when you say

    I know that lasers are approximately, more or less, monochromatic

    you 'know' wrong. Lasers are indeed often described as producing light of only one frequency, but this is a misconception. Lasers come from the amplification of light via stimulated emission, and this amplification can happen for any frequency where the gain medium has available transitions. For solid-state systems, these transitions often occur over broad frequency bands, so those gain media can be used as 'tunable' lasers (where you use the cavity to select the frequency that gets amplified), but you can also set them up so that all of the frequencies are amplified at the same time, fixing their relative phase so that they are coherent with each other, which results in a series of (often very short) laser pulses.

Because of the added layers of technology, pulsed lasers are often more expensive than thermal sources (which can be extremely simple), so if you don't need the coherence (such as e.g. in a spectrometer) then you just use thermal sources when you need broadband light.

Emilio Pisanty
  • 132,859
  • 33
  • 351
  • 666
  • If I'm not mistaken, monochromaticity refers to electromagnetic radiation of a single wavelength, not frequency. And my understanding is that lasers are often described as producing light of only one wavelength -- not frequency. Am I incorrect? You seem to be confusing wavelength, of which monochromaticity concerns, and frequency, which is a different property. As per my link, monochromaticity is a physically impossible phenomenon, which lasers approximate, which is why I phrased it as I did. Am I misunderstanding something here? – The Pointer May 13 '20 at 17:32
  • Nice, I hadn't accounted for broadband emission via pulsed lasers in my answer, and this is indeed relevant to the question – kleingordon May 13 '20 at 17:32
  • 2
    @ThePointer as with any wave phenomenon there is a one-to-one relationship between frequency and wavelength, given by dispersion relation. For electromagnetic light traveling in vacuum or near-vacuum the relation is simply frequency * wavelength = speed of light (to good approximation). So it's common to identify the two quantities with each other and mention "wavelength" or "frequency" interchangeably when distinguishing emission at different wavelengths or frequencies – kleingordon May 13 '20 at 17:44
  • @kleingordon Oh, ok, I see. This is very interesting. So, does this mean that we can have a laser device that simultaneously emits multiple different wavelengths across a broad spectrum? For instance, could we have a laser device that emits a single beam that contains wavelengths of 500nm and 1000nm, or 1500nm and 2000nm, simultaneously? Or am I misunderstanding Emilio's answer? – The Pointer May 13 '20 at 17:47
  • 1
    Yes, as Emilio said, pulsed lasers can emit coherent light spanning a range of wavelengths simultaneously. Different devices can be set up to do this at different wavelength ranges. In practice one might be limited by the wavelengths that an available pulsed laser system can produce. – kleingordon May 13 '20 at 17:55
  • @kleingordon Very interesting. Thank you for the clarification. – The Pointer May 13 '20 at 17:57
  • @ThePointer Yes, at its correct. As always, the real world imposes practical restrictions on just how wide your bandwidth can be (and, in particular, the octave-spanning spectrum in your first example would be a significant practical challenge, though there are groups that achieve similar spectra in the IR quite routinely) but there's nothing preventing that in principle. – Emilio Pisanty May 13 '20 at 23:41