I am tasked with building a low cost back scattering probe that can be used for fluorescence or Raman alike. My expertise lies in front end electronics design and I haven't had a lot of exposure to optics and optical instrument design. My setup right now looks like this as shown in "Setup.jpg". I am using a free space 784nm laser with ~2nm FWHM and about 2mm beam diameter. I am feeding this to a cube beamsplitter (https://www.thorlabs.com/thorproduct.cfm?partnumber=CCM1-BS014). The beam exiting is fed through a long pass filter @800nm cut on (https://www.thorlabs.com/thorproduct.cfm?partnumber=FELH0800). The output beam is then fed to a spectrometer using a focusing optic. With this setup, the spectrum collected looks as shown in "784nm_Spectrum.png". I don't know what to make of this spectrum. There is the peak @784nm due to laser but I would have expected there to be a "flat" spectrum after 800nm. Instead I see this weird "decaying" pattern. The spectrometer has been calibrated and should not cause any error. Please have a look and let me know if you have run into something similar or what you think it could be. I am a bit clueless at this point.
Dear Anirban,
did you shut off all ambient light sources in your lab?
For example incandescent lamps and sun light have got large amount of light above 800nm *), but you can't see this light and might think that there is no such light.
Especially ambient light coming from the left will enter your beam splitter cube and will be partially directed to your pick up lense and the fibre.
So please shut off all ambient light or at least shield your optical set up hermetically in order to prevent any light pick up of your optical system other than the laser light.
*) Please have a look at the first Fig. in:
http://www.roperld.com/science/electromagneticspectraoflightbulbs.htm
Since you don't mention any other light sources, I assume that the 50-50 BS is throwing away half of your laser light and not accepting any new light (i.e. florescence or back-scattered sources. So the only light going into your fiber is the transmitted 784 nm beam. It is possible, if your laser is intense enough that you might be seeing florescence from the BK7 glass that is in both the BS cube and the focusing lens that you are using. You don't specify your fiber type or material, but it can also be a source of converted light. Self-phase modulation (especially in the fiber) is a possibility too. Is your laser short-pulse (i.e. sub-ps pulse duration)? What is your average laser power? If pulsed, what is the energy and duration of each pulse? Try turning your laser power down and see if the signal goes away? If it does, does it do so proportional to the energy?
Hi Anirban,
With the 800nm cut-on long-pass filter in place you shouldn't really see any laser light at 784nm at all. Looking at the spectrum in your plot, the amplitude at 784nm looks like in might be saturating the detector in the spectrometer (it's flattened at the peak). So, I suspect that your laser power is so high that it's breaking through the filter (which will have a maximum power intensity specification). The spectral output beyond 800nm could be fluorescence or might even be due to everything getting hot.
So in consequence of Dustin's and Ray's answers; please shut off your laser and see what happens with/in your spectrum above 800nm.
Does the hump above 800nm also disappear when the laser is down?
Gerhard Martens Dustin Offermann thanks for your suggestions, I can confirm that I see this spectrum only when the laser is on. With the ambient dark, I see the noise floor of the spectrometer, the instant the laser turns on, I see the spectrum. So this must be an effect due to the laser alone. This is a cw laser running at 80mW. I will modify the laser power and note any change that it brings.
Ray Simpkin I think the filter gives an OD5 at 785nm. You could be right about too much light entering the spectrometer. I will try toning down the power but not sure if that will help alleviate any fluorescence trouble from the optics.
After some thoughts, I think the tail you are measuring is just the remaining of your laser peak. Even though it is a laser, it still has a linewidth which can be several nm wide (cheap lasers anyway, 2 nm in your case). What you are measuring is the convolution of the laser output spectrum and your filter.
If you remove your filter and replace it with a neutral density (otherwise you might burn your spectrometer) you should measure a nice lorentzian shape peak.
at 10 times your linewidth you still have around 0.1 % of the peak power.
The filter you are using is OD4 so it means that the peak is attenuated by 10000.
So basically the fact that you still see your peak means that your measuring signal attenuated by 10000 and then around 800 nm this attenuation jumps to almost 0. Since the power you receive from the tail is of the order of your attenuated peak you measure this strange looking spectrum.
Philippe Velha Interesting. If this is the case, if I cascaded two such filters I should not be seeing this weirdness then. I will run this experiment and update this thread.
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Physical Phenomena
- Luminescence
- Fluorescence