The Pulsar Engineering Community Board

Considering the noise level in the spectrum shown, I would say that the approximation of ~5°C is not that far from reality.

I really have to take a moment to read that paper to get a better understanding of the experiment!

I have looked at another raman scattering experiment for the students. The used a sample of sulfur powder to get the temperature measurement. They then fitted a gaussian to the peaks and read their intensity off those. I have tried to get a raman spectrum of sulfur (I am aware that there is an OpenRAMAN powder analyzing cuvette, however I did not see anything that would justify it over simply using the normal cuvette holder) and I got this spectrum. As you can see there is a tall peak in the stokes region and one smaller but visible in the antistokes region. If I use the local minima of the behind the stokes peak as a baiseline and I got -1,88 °C for this sample. Does anybody have an idea on how to better take fluorecence into account? Or how to avoid it alltogether?

The generic background correction method is not ideal for quantitative work.

The easiest is to export the raw spectrum in csv and, using Matlab or excel, remove a slanted line at the base of the peak. It works very well for isolated sharp peaks but it's more tricky for broader peaks.

Got interested in this discussion. I worked with Anti-Stokes luminescence (not Raman, but plasmon-enhanced luminescence).

In my setup I used a 90:10 beamsplitter since I had excess laser power, I preferred to optimize collection. I also used a short-pass RazorEdge from Semrock (be prepared to open up your wallet) and one or two notch filters.

For the Stokes/Anti-Stokes ratiometric measurements, I used two notch, so I could see both sides of the spectrum.

In the scattering measurements you showed, I wouldn't expect to see the Raman signal, be aware that it is orders of magnitude dimmer than the laser power, so it would be swamped. Side-bands can be generated by the laser itself (you may check the spectrum of the laser directly), by stray light reaching the camera, or it can be some interference pattern on the camera cover glass itself (hard to identify once you see 1-D data).

It would be fantastic if the OpenRaman is able to measure line broadening and anti-Stokes! It adds a new dimension to the experiments that can be done.