Raman dependence on exciter wavelength

[Tim Holmes-2]

A physicist colleague explained to me that I can “excite” ANY compound/sample with a 250 nm laser or a 1 micrometer laser, and the

reflected Raman shifts/spectra will be the same in either case, except that the sensitivity/reflection/number-of-photons-returned

will be 64X for the 250 nm laser because there is a 1/(wavelength**4) relationship in this reflection/scatter sensitivity.

 

Could someone tell me what is correct and not correct about the above statement?

 

Thanks a lot

Tim Holmes

Lickenbrock Technologies

 

[Emmanuel Gustin]

Hello Tim,

 

When I was a master's student and did my first Raman experiments, I also suggested using a blue or even an UV laser line to have more efficient Raman scattering, relying on this steep wavelength dependence. I have to admit that my more experienced colleagues considered this a silly idea. There are a number of complications to it.

 

Raman scattering generates a fixed shift in frequency, not in wavelength. Raman spectroscopists commonly think in wavenumbers or 1/cm: Calculated as 10^7 / wavelength if wavelength is expressed in nanometer. The stretch vibration of an O-H bond can be around 3600 1/cm, and that is about the highest frequency you will find. Excitation at 250 nm (40,000 1/cm) gives you a Stokes Raman signal from it at 275 nm (36,400 1/cm). But excitation at 1000 nm (10,000 1/cm) gives a Raman signal at 1563 nm (6400 1/cm). And because spectrometer gratings disperse light in function of wavelength, the resolution of a Raman spectrum is better using red excitation.

 

The other serious objection against using an UV laser is that normal Raman scattering is about a million times less efficient than fluorescence. So one wants to avoid any background fluorescence that might drown the signal, and that usually is a good reason not to use an UV laser, but a red or near-infrared laser. You can get away with having some fluorescence from the sample if you study the anti-Stokes Raman signal, light that is Raman-shifted to higher frequencies, but there are limits even to that because spectrometers don't have perfect stray light suppression.

 

Nevertheless people do use UV lasers for Raman spectroscopy, but for a different reason, and that is resonant enhacement. Forget about the paltry 64x boost the wavelength dependence of the scattering gives you: If the laser wavelength coincides with an electronic excitation of the molecule, the Raman scattering can be enhanced by up to a factor 10^6. If the sample is not fluorescent, it can be very advantageous to have a laser line in an absorption band. That makes UV lasers attractive for the study of proteins and DNA.

 

Best Regards,

 

Emmanuel

 

 

-----Original Message-----
From: Confocal Microscopy List [mailto:[hidden email]]On Behalf Of Tim Holmes
Sent: Thursday, February 19, 2009 18:09
To: [hidden email]
Subject: Raman dependence on exciter wavelength

A physicist colleague explained to me that I can “excite” ANY compound/sample with a 250 nm laser or a 1 micrometer laser, and the

reflected Raman shifts/spectra will be the same in either case, except that the sensitivity/reflection/number-of-photons-returned

will be 64X for the 250 nm laser because there is a 1/(wavelength**4) relationship in this reflection/scatter sensitivity.

 

Could someone tell me what is correct and not correct about the above statement?

 

Thanks a lot

Tim Holmes

Lickenbrock Technologies

 

[Andreas Bruckbauer]

I disagree with Emmanuel, UV Raman is an established technique, fluorescence
seems to be a problem above 300 nm, Semrock has some information on their
website. However you need special lasers, optics, spectrometer and detectors
which are more expensive and you might damage your sample easily. For
biological samples near IR is the wavelength of choice.
In your sentence you would say "scattered" instead of "reflected" Raman
signal, it scatters can go in all directions.

Andreas

[Emmanuel Gustin]

Hi Andreas,

I didn't want to suggest that UV Raman is not used, but to point
out that about the only practical reason to do so is resonant
enhancement. The boost resonance gives to the Raman signal is so
large, that it can override the otherwise serious complications.
For biological samples NIR causes less damage but UV can greatly
improve sensitivity.

Admittedly, there is one other reason people use UV excitation
for Raman, and that is to study material surfaces, profiting
from the very shallow penetration of the UV.

Best Regards,

Emmanuel


-----Original Message-----
From: Confocal Microscopy List
[mailto:[hidden email]]On Behalf Of Andreas Bruckbauer
Sent: Thursday, February 19, 2009 22:58
To: [hidden email]
Subject: Re: Raman dependence on exciter wavelength


I disagree with Emmanuel, UV Raman is an established technique, fluorescence
seems to be a problem above 300 nm, Semrock has some information on their
website. However you need special lasers, optics, spectrometer and detectors
which are more expensive and you might damage your sample easily. For
biological samples near IR is the wavelength of choice.
In your sentence you would say "scattered" instead of "reflected" Raman
signal, it scatters can go in all directions.

Andreas