Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Maximum Fluorescence Energy vs Excitation

  1. Mar 16, 2013 #1
    I have taken a time dependent fluorescence series of spectra for an ethylene glycol solution using a 532nm laser as the excitation source. I have attached the spectra to this post. Nothing looks unusual-there is a sharp notch near the 532nm line--this is due to a filter I have used to protect the spectrometer from the laser.

    However--note that the spectrum has points to the left of the laser line-e.g. at wavelengths lower (energies higher) than the laser. How does this happen? I don't believe this could be attributed to Raman /anti-stokes processes due to the large separation--usually these shifts are smaller yes? Even though we are dealing with a complex solution here, we cannot expect electrons to become promoted to levels higher than the excitation photons to this extent.


    Attached Files:

    • anti.jpg
      File size:
      25.4 KB
  2. jcsd
  3. Mar 17, 2013 #2


    User Avatar
    Science Advisor

    That is difficult to say without knowing what your laser looks like. What kind of laser is it? 532 nm cw? pulsed? If pulsed, fs, ps or ns pulses? It might be worthwhile to check the laser spectrum under identical conditions. To me it seems that at the low intensities considered here, that might just be the width of your laser extending beyond the filter.
  4. Mar 17, 2013 #3
    I am using a Qswitched Quantel CFR200 532nm NdYAG set at about 20mJ. Pulses are around 7-9ns. The laser is "tight". The green "tower" in the middle of the filter notch is the laser line as seen by our spectro. I am using a spectro with a resulution of bettter than 0.03nm. I see the same effect with a cheap OceanOptics spectrometer.

    I have taken some spectra with the power choked down quite a bit (probably ~1mJ). I don't expect that we are seeing thermal effects in the sample contributing here but that is all conjecture of course.

  5. Mar 23, 2013 #4
    You have to eliminate the possibility that the antiStokes line is an artifact caused by spontaneous fluorescence of the lasing material. The laser that you are using to excite the sample may not be monochromatic.

    I recommend that you use a "sanity check". Replace your sample with either an empty cuvette or a scatter plate. A frosty piece of glass (i.e., scatter plate) should have no fluorescence lines. If you still still see the same anitStokes fluorescence line, then you have an artifact.

    You may not have a frosty piece of glass. However, you could use almost any white thing as a sanity check. You could even use a white piece of paper. Warning, white paper does have a broad fluorescence spectrum. However, white paper probably doesn't have the same antiStokes fluorescence line.

    One should always use at least one blank for comparison. I recommend a few blanks with different scattering properties. Analytical chemists are trained to always use a blank.

    One thing you could do is place a narrow band filter at 532 nm right in front of the laser. You can use a multilayer filter, a diffraction grating or even a prism in front of the laser. If you use a diffraction grating or a prism, you also need an aperture to eliminate the stray reflections at oblique angles. If the antiStokes line disappears, then it was probably an artifact.

    Spontaneous emission of the lasing material is a chronic problem with gas lasers. Gas lasers often have strong emission bands with wavelengths shorter than the main band. I made that mistake often when I was doing my thesis. The problem is not as severe with solid state lasers. However, even solid state lasers have some spontaneous emission at wavelengths shorter than the lasing wavelength.

    I would even suggest that you always use a dispersive device in front of your laser to eliminate or reduce emission at wavelengths other than the desired laser line. Lasers are more monochromatic than other other sources of light, but they still have extraneous emission at different wavelengths. Fluorescence and Raman signals are very weak compared to the laser line. Therefore, even a small emission from the laser at the wrong wavelength can be mistaken for a fluorescence of Raman signal.

    Never believe those people who tell you that the laser is monochromatic. Not even the manufacturer. Not even your thesis adviser. Trust no one!

    Since your antiStokes shift is larger than the typical antiStokes Raman line, it should be easy to find a dispersive device that can eliminate the spontaneous emission from the laser medium.

    It may be useful for you to use a scatter plate to examine the spectrum of your laser. A l laser can have more than one lasing line in addition to spontaneous emission. If your "antiStokes" line matches one of the lines of your laser, then maybe you are looking at a laser artifact.

    If your anti-Stokes signal is actually from the sample (i.e., "real"), then the real fun begins. If you have done a number of tests to eliminate the possibility of an error, then you should do a number of other tests to determine what that line really is.

    You still won't know if your antiStokes band is from scattering or fluorescence. Use different laser wavelengths to see how that line changes. Lines due to scattering have a constant shift, lines due to fluorescence have a constant wavelength.Use neutral density filters in front of the laser to determine how the intensity of the band changes with laser intensity. Check polarization of the antiStokes band.

    Once you have reduced the possibility of artifacts, you should examine the weird possibilities.
    Last edited: Mar 23, 2013
  6. Mar 23, 2013 #5
    Thanks for your well thought out reply! I always shoot some "inert" samples to check out the system. The stray spectrum below the laser line is not visible in that case. I believe the laser line to be better than 0.3nm. I took spectra of it a long time ago. This is a solid state high end laser. The spectral lines below 532nm in the attachment cover around 50nm.

    The possibility of harmonic generation is quite real. The laser in use is 1032nm NdYag fronted with a harmonic crystal for 532 but I would expect at least some 266nm in there.

    I will be using a 405nm CW diode laser in the next few weeks. It will be interesting to compare.

    I appreciate your response here-some good ideas
  7. Mar 24, 2013 #6
    Your "antiStokes band" could be a Stokes band from another component of your laser.

    If there is a 266 nm component to your laser, then it can excite photoluminescence emission in your sample. UV at 266 nm can be absorbed by many materials. Therefore, your "antiStokes" band maybe be Stokes fluorescence or even Raman from your sample or the cuvette.

    There are other components that can be generated by nonlinear optical processes. There can be a 355 nm component generated which can also cause fluorescence.

    If you are using a diffraction grating in your laser optics, you have to be careful of the high order diffraction. High order diffraction can make a high frequency component of the laser look like an antiStokes line. Second order diffraction can make that 266 nm component appear like a 532 nm component. There are also diffraction "ghosts" that can make that 266 nm look like a narrow spectral band.

    An ordinary glass window blocks all wavelengths shorter than 290 nm while allowing through all wavelengths longer than 400 nm. Ordinary glass is inexpensive.

    When using a diffraction grating to disperse light, you should have a filter in front of your collection optics to block off laser line components. You should at least have a filter in front of the sample that blocks that 266 nm component. If scattered light at 266 nm enters a spectrometer, the spectrometer can read it as a 532 nm band. A plate made of ordinary glass would be sufficient to block the 266 nm component.

    If you have to work cheap, then you may be restricted to inexpensive filters. You can buy a long pass filter that blocks all light with wavelength shorter than 400 nm. Such a filter would have a high transmittance at 532 nm.

    Filtering of the laser beam can be a problem. Since spontaneous fluorescence is isotropic, a narrow aperture in front of the laser can reduce the amount of spontaneous fluorescence from your laser beam. There are broad band filters that transmit all wavelengths from 500 nm to 900 nm. An inexpensive glass prism can separate your 532 nm line from the 355 nm and 1066 nm components in your laser line. However, you have to give the beam plenty of distance if you want greater resolution from that prism.

    I don't know the pulse width of your laser. However, a laser with very large peak power can generate self phase modulation (SPM) within the laser. Femtosecond lasers can cause SPM continua in almost any material.

    I always place a band filter in front of the laser and a long pass filter in front of the sample. The band filter may be a problem since narrow band filters have low transmittance. However, a wide band filter in front of the laser can eliminate many false signals without reducing the signal significantly. There are many companies that sell long pass filters.

    What I am suggesting is standard part of optical studies. I have mistaken optical artifacts for spectra several times. Maybe you won't need the extra optics once you have proven the signal is real. However, the optics that you buy will be useful for the next unknown material that you work with.

    Your antiStokes line may be an interesting phenomenon. I think that you should follow up on it. However, I recommend that you should first prove that it is a "real" antiStokes line from the sample.
  8. Jun 4, 2013 #7
    Ethylene Gycol "anomalous" fluorescence

    You say that you have an ethylene glycol "solution" - What else is in it besides ethylene glycol?

    The suggestion of placing a narrow-band filter in front of the laser is an excellent one (if you can find a filter that will withstand those energies), as is the suggestion to record "blank" spectra, e.g with frosted glass.

    I believe your laser is flash lamp pumped. I am wondering if you are seeing either part of the flash lamp spectrum, or possibly fluorescence excited by the shorter wavelength components of the flash lamp.

    The reference below suggests that ethylene glycol has no visible absorbance, so what you are seeing as emission must come from some other source (e.g. uranine or other dye added as a colorant):

    Guang Pu Xue Yu Guang Pu Fen Xi. 2007 Jul;27(7):1381-4.
    [Study on the absorption and fluorescence spectra of ethylene glycol and glycerol].
    [Article in Chinese]
    Xu H, Zhu T, Yu RP.

    School of Communication and Control Engineering, Southern Yangtze University, Wuxi 214122,

    The absorption and fluorescence spectra of ethylene glycol and glycerol solution induced by UV light were studied respectively in the present paper. The most intense absorption wavelength for both of them was located at 198 nm. Moreover, fluorescence was detected when induced by suitable UV light, and the corresponding fluorescence spectra were listed. But there is no obvious relationship found between the fluorescence intensity and the excited wavelength, and a further research should be done. From the first derivative fluorescence spectra of ethylene glycol, it was concluded that under the UV light of 210 nm, the variation speed for relative intensity proved to be the fastest. In contrast, when excited by 225 nm, the speed proved to be the slowest. In addition, based on the quantum calculation and the transition from HOMO to LUMO of electronics in one-dimensional quantum well, the authors attempted to give out the value of absorption wavelength. In consideration of the bond-length variety brought out by the chain processing, the error between the experimental and calculation values should be apprehensible, and the latter can serve as some reference value in theory.

    Lots of possibilities here...

    Best of luck!
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook