RamanOxxiusLaser
Theoretical knowledge

How to choose a laser for Raman Spectroscopy

by

In specific spectroscopy techniques like the Raman or LIBS (Laser-Induced Breakdown Spectroscopy) ones, the choice of the color and of the optical specifications of the excitation laser can be very important.

The Raman spectrum of a material or a gas sample represents the energy shift of its vibration modes in relation to the laser excitation and is usually expressed in wavenumbers (cm-1 unit).

In Raman spectroscopy, the effective radiated power varies as the excitation frequency to the fourth (due to the electric dipole nature of light), therefore proportionally to λ0-4 of the wavelength λ0 of the Raman laser. Thus it may be interesting to use shorter wavelengths to increase the signal to be analyzed, typically with 514 nm or 532 nm green Raman lasers.

Visible Raman lasers are especially used for inorganic material experiments and for Surface Enhanced Raman Scattering (SERS).

However, the excitement of some samples may be accompanied by a phenomenon of fluorescence that can interfere, even exceed the Raman signal. This problem can generally be solved by considering a lower laser energy, and therefore a higher wavelength, typically with 785 nm or 830 nm near-infrared Raman lasers. Recent developments can extend this consideration to the infrared range with 980 nm or 1064 nm infrared Raman lasers.

Red and NIR Raman lasers are especially beneficial for the study of bio-molecules where good fluorescence suppression is needed.

The possibility of multiplexing measurements at different wavelengths – 532 nm, 785 nm and 1064 nm – could also be a considerable advantage for the analysis of the sample properties.

Image1

Figure 1. Typical Raman spectra of Diamond (left) and Cyclohexane (right) samples with 785 nm laser excitation (spectral density versus Raman shift): diamond spectrum presents a single-mode peak on a mean fluorescence signal; cyclohexane spectrum is a multi-mode signal.

In addition to the wavelength and the power of the laser, several optical specifications are important to consider while choosing your Raman laser:

  • the stability in power is an important criterion in the choice of the exciting laser, especially for repeatable and long duration measurements: good Raman lasers have a power stability better than ± 1 % peak-to-peak in a 8 hour period.
  • the laser linewidth and the stability in wavelength (“laser drift”) of the Raman laser are also important for the quality of the spectrum, especially for high spectral resolution reconstruction: good Raman lasers have less than 0.5 % rms of optical noise.
  • the lifetime of the Raman laser is another critical parameter to check: 10,000 hours of expected operation is a minimum to have.
  • for specific applications, the polarization extinction ratio of the Raman laser must be controlled: a ratio greater than 100:1 is generally expected.

Image3

Figure 2. Typical high-resolution spectrum of a low-end multimode high power Raman laser: the linewidth provided by the vendor corresponds to many modes and may be quite unstable.

Tags

/ / / / / / /

Read other articles

5 Comments

  • amayas says:
    August 11, 2016 at 0 h 00 min

    Very nice paper !
    But, no think is said about the linewidth of the laser … any indication?
    0.1 nm is good? 0.5 nm?

    Reply
    • Fabrice Thomas says:
      August 11, 2016 at 15 h 16 min

      Dear Amayas,

      Thank you very much for your comment.
      You are right, the linewidth can be an important specification to consider for a Raman laser, and must be linked with the spectral resolution you need on the Raman spectrum.

      Good single-mode Raman laser diodes can typically provide very narrow linewidths, less than 100 MHz FWHM (i.e. 0.0002 nm at 785 nm) with a wavelength stability better than 0.010 nm, but are relatively low in power (50 to 100 mW); combined with a high-resolution spectrometer, they can be very valuable for high-coherence Raman spectroscopy to differentiate very small Raman shifts.
      Multi-mode Raman laser diodes provide higher power (up to 600 mW), but their linewidth can be as wide as 1 nm; so this results in all Raman peaks being 1 nm wide, regardless of the resolving power of the spectrometer.

      A good tradeoff can be a single-mode Raman laser with 0.1 nm linewidth and ~ 300 mW intermediate power.

      Hoping that it will be useful for you,
      Best regards,

      Fabrice

      Reply
  • Fabrice Thomas says:
    August 12, 2016 at 9 h 22 min

    Dear Amayas,

    Yes, 300-600 dollars is currently the price of chinese low-end multi-mode lasers ; they may be generally sufficient to spectrometric studies with 0.5 nm resolution (20 cm-1 Raman shifts at 532 nm).
    The difference may come from the higher power and the stability of the BeamQ. Better quality will require a price > 2 k$ .
    If you use fibered or bulk-optics setups, be careful that all these lasers are generally free-beam and not necessarily collimated (see beam divergence and quality), so you will need collimating optics (with basically 40-60% coupling efficiency).

    If you want more information, please feel free to send me a direct email to : fabrice.thomas@resolutionspectra.com

    Best regards,
    Fabrice

    Reply
    • amayas says:
      August 12, 2016 at 11 h 22 min

      Dear Fabrice !

      THanks a thousand for these precious information.

      I’ll be back to you by email

      All the best for the MyLaserSpectrum community

      Amayas

      Reply

Leave a Reply

Your email address will not be published. Required fields are marked *