Respected Professors
I am working on low temperature plasma. I am Interested on diagnostics of plasma parameters through optical means. We have an optical emission spectrometer from Ocean Optics (USB 2000+). How do i calculate the instrumental width of my USB 2000+ instrument? I calculated the electron density using stark broadening method only and i got the value in the order of 10^16/cm3. However, I was suggested by the reviewers from the journal that " a spectrometer from Ocean Optics Inc. (USB 2000+) probably provides resolution insufficient for directly applying eq (3). I would suggest discussing instrumental width of the USB 2000+ instrument and how this was considered in your line-profile analysis. Note, that Ne estimated using equation of stark broadening method to line profile shown in figure is too high for the DBD discharge in argon ". I would be helpful if you could help me in this regards. I have attached the images and the data that I have got from the USB 2000+ spectrometer.
Sincerely
Rajesh Prakash Guragain
Kathmandu University, Nepal
Email: [email protected]
Dear Rajesh,
unfortunately, the reviewer is right (just a disclaimer - I am NOT the reviewer) - I had the exact same problem with an Ocean Optics 2000+HR and I wasted a lot of time with trying to fix this. The resolution is indeed too low for Stark broadening measurements. It is additionally unfortunate that the Ocean Optics company was very secretive about their instrument function (or instrumental width as your referee calls it), which I wanted to use in order to get at least some information about n_e with Stark broadening. However, after some conversation with the company they at least told me that usually they use a Gaussian fit with 2 pixels FWHM for this particular spectrometer but, again, this is far to low for a suitable resolution. In addition, the company told me that this is also just an approximation since the apparatus function of your spectrometer might as well be assymetric to some degree. This again worsens the reliability of your measurements....
So, to conclude the Ocean Optics 2000 is a very nice toy for things like process monitoring and it comes with a very reasonable price tag but for real optical plasma diagnostics, especially for high pressure plasma and measurments that are connected to line broadening of some kind it is not that suited.
I am just writing this here on RG instead of sending you a PM because after I read that you experience the same problems that I had, I thought it perhaps might be helpful for other people as well and save them the time I wasted on this problem.
Any time you are trying to measure a spectral line width, you must consider the resolution of the instrument. The apparent peak will be the convolution of the actual line shape with the instrument‘s line shape. You may safely ignore it IF you find that the instrument’s spectral resolution is much much narrower than the line width you are measuring. Otherwise you must deconvolve the true line shape from instrumental line shape. If the instrumental line shape is much wider than what you hope to measure, then deconvolving the true shape from the instrument‘s line is tricky and easy to get wrong. This is particularly true for an Ocean Optics type spectrometer where the spectrum is captured by discrete pixels. It can be very difficult or often impossible to confidently back out the true line shape.
Calculating the expected instrument line width is straight forward and described any number of places. For example this post from Newport: https://www.newport.com/t/resolution-of-monochromators-and-spectrographs
or you can use the calculator provided by Ocean optics (who have apparently changed their name to ocean insight)
https://www.oceaninsight.com/support/knowledge-hub/spectrometer-range--resolution-calculator/
However, if the instrument line width turns out to be significant compared to what you want to measure, then you are going to want to deconvolve a true line shape. In that case you will want more than the instrument resolution. You will want a line shape. This too can be calculated with a little more full description of the spectrometer optics. However, real spectrometers are imperfect enough (slightly out of focus, axis of cylindrical optics, grating, slits not perfectly aligned, etc etc) that it’s better to measure the line shape directly. This requires a source with a much narrower width than the spectrometer resolution. Stable CW gas lasers have nice narrow lines. Some atomic emission lines in calibration lamps can be quite narrow. (see https://www.newport.com/f/pencil-style-calibration-lamps ). It’s best if the calibration line is somewhere near your measurement wavelength. With a narrow source, the shape of the measurement is taken to be the instrument line shape. That is then used to deconvolve the line shape from your measurement of the source you are investigating.
Dear Rajesh,
Stark broadening is a powerful tool to get the electron number density. However, there are several other mechanisms that can broaden atomic lines. Some are physical (Van der Waals, resonant, Stark and Doppler) and the instrument itself has a limited resolution (instrumental broadening), as your reviewer noted.
I recommend you to:
Article Fully ionized nanosecond discharges in air: the thermal spark
If you find that another broadening mechanism is dominating, you can at least provide a higher bound of your electron number density.
Good luck with the review
Nicolas
- Badges
- Science topic
- Similar topics
- Instrumentation Engineering
- Instruments