Cyclohexane Raman Spectra: Analyzing the 802 cm-1 and 2853 cm-1 Peaks

[new6ton]

Who is goodly familiar with the 802 cm-1 and 2853 raman peaks of Cyclohexane or the raman spectra of Cyclohexane in general?

There are conflicting results about it. And even after hours of google searches. I can't figure out what peak should be higher.

In above, the 802 cm-1 peak is higher. But elsewhere (for example from https://www.princetoninstruments.com/userfiles/files/assetLibrary/Technical%20posters/Poster-Aberration-Free-Raman-Spectroscopy-with-the-IsoPlane.pdf ) the 2853 cm-1 peak is higher.

Based on the molecular structure and dynamics of the ring breathing at 802 cm-1 and symmetric stretch at 2853 cm-1, which should be higher in peak with respect to each other?

 

[TeethWhitener]

Raman intensity will depend on the excitation laser wavelength and the way the sample is prepared, among other things.

 

[new6ton]

Raman intensity will depend on the excitation laser wavelength and the way the sample is prepared, among other things.

I know. But given the same laser source (532nm) and same sample. is the 802 cm-1 or 2853 cm-1 peak higher *with respect* to each other?

 

[TeethWhitener]

Without specifying the exact way this hypothetical sample is prepared, I imagine this question doesn’t have one answer. For one thing, you could probably choose a photonically active surface that resonantly enhances either wavelength.

 

[new6ton]

Without specifying the exact way this hypothetical sample is prepared, I imagine this question doesn’t have one answer. For one thing, you could probably choose a photonically active surface that resonantly enhances either wavelength.

So there is no molecular basis why the ring breathing at 802 cm-1 is higher in peak than the symmetric stretching of CH2 at 2853 cm-1 or vice versa? But is it not electron polarizability is quantitative and intensity means difference in the bonding or molecular behavior?

Cyclohexane is supposed to be the reference material for calibration in most raman spectrometers.

 

[TeethWhitener]

So there is no molecular basis why the ring breathing at 802 cm-1 is higher in peak than the symmetric stretching of CH2 at 2853 cm-1 or vice versa?

Of course there is. Transitions can easily be symmetry forbidden, for example. But the molecule by itself is not the only factor.

Cyclohexane is supposed to be the reference material for calibration in most raman spectrometers

Yes, you basically calibrate it so that the spectra from different wavelengths look the same, even though they aren’t. Here’s an example:
https://www.kosi.com/na_en/products/raman-spectroscopy/raman-accessories/calibration-accessory.php

 

[new6ton]

Of course there is. Transitions can easily be symmetry forbidden, for example. But the molecule by itself is not the only factor.

Yes, you basically calibrate it so that the spectra from different wavelengths look the same, even though they aren’t. Here’s an example:
https://www.kosi.com/na_en/products/raman-spectroscopy/raman-accessories/calibration-accessory.php

In the site you provided, this is the calibrated Cyclohexane spectrum:

So it means the 802 cm-1 peak is lower than the 2853 cm-1 peak?

And do you mean in other spectrometers, it's the opposite? Or based on the electron polarizability, the 802 cm-1 peak being lower than the 2853 cm-1 peak is the correct one?

 

[TeethWhitener]

The point of calibration is to try to mitigate differences across instruments and instrument settings. Ideally, if you know the instrument response exactly, the absolute Raman cross section of a peak (the part most closely corresponding to $\langle f| \alpha | i \rangle$) is the integrated area of the peak. But this still is dependent on wavelength, sample geometry, etc., etc. Here’s what looks like a decent reference on Raman calibration at first glance:
https://www.thevespiary.org/library...tometric.Standards.for.Raman.Spectroscopy.pdf
You can see that the community has pretty much come to a consensus about the value of a handful of cross sections (the article mentions the benzene 992 resonance in particular) and calibrate all instruments according to those.

The reason this question is so tough is because the scattering cross sections and the instrument responses both change with wavelength. So let’s say for example you have an excitation wavelength of 532 nm. The resonance at 800 cm^-1 is at a wavelength of around 560 nm, while the resonance at 2800 cm^-1 is at around 625 nm. You need a standard at each of those wavelengths to know the instrument response, but you need to know the instrument response to know how the standard should behave at those wavelengths. It’s a bootstrapping problem.

 

[new6ton]

The point of calibration is to try to mitigate differences across instruments and instrument settings. Ideally, if you know the instrument response exactly, the absolute Raman cross section of a peak (the part most closely corresponding to $\langle f| \alpha | i \rangle$) is the integrated area of the peak. But this still is dependent on wavelength, sample geometry, etc., etc. Here’s what looks like a decent reference on Raman calibration at first glance:
https://www.thevespiary.org/library...tometric.Standards.for.Raman.Spectroscopy.pdf
You can see that the community has pretty much come to a consensus about the value of a handful of cross sections (the article mentions the benzene 992 resonance in particular) and calibrate all instruments according to those.

The reason this question is so tough is because the scattering cross sections and the instrument responses both change with wavelength. So let’s say for example you have an excitation wavelength of 532 nm. The resonance at 800 cm^-1 is at a wavelength of around 560 nm, while the resonance at 2800 cm^-1 is at around 625 nm. You need a standard at each of those wavelengths to know the instrument response, but you need to know the instrument response to know how the standard should behave at those wavelengths. It’s a bootstrapping problem.

My wine purity 532nm molecular scanner has the following Cyclohexane raman spectra. And I can't decide which is the right one. Both has 200ms exposure at one frame.

In the above the 802 cm-1 peak is at 40000 counts, while the 2853 cm-1 peak is lower. If I move the focus 3mm forward, this is the spectrum:

Here, the 802 cm-1 peak becomes lower (at 30000 counts) than the 2853 cm-1 peak.

Which do you think is the right one? The manufacturer themselves got confused when I asked them.

 

[TeethWhitener]

If I move the focus 3mm forward, this is the spectrum:

Which do you think is the right one?

You've essentially used two different instruments, so they'll have different responses. Neither are wrong. There are standard calibration procedures that you can undergo that will make the two spectra look more similar, but as I (and the links) have basically laid out already: these procedures are essentially choosing a winner and tweaking all the other spectra to look like it.

 

[new6ton]

You've essentially used two different instruments, so they'll have different responses. Neither are wrong. There are standard calibration procedures that you can undergo that will make the two spectra look more similar, but as I (and the links) have basically laid out already: these procedures are essentially choosing a winner and tweaking all the other spectra to look like it.

Can the tweaking be done by software or does it involve the hardware mostly? In this guide "How to calibrate your spectrometer"

http://georaman2014.wustl.edu/previ...rnationalschool/003 Calibration_MC-Caumon.pdf
It mentioned: "The scattered Raman Intensity IRaman:
Depends on optics, grating, filters, wavelength, detetor, polarization, objectives."

Why didn't it mention the software? So the wavelength tweakings are all done on hardware mostly?

 

[TeethWhitener]

It's almost always software. The instrument response depends on the hardware, but the calibration is done to eliminate the contribution from the instrument response by correcting the spectrum with (usually) a polynomial function of some kind.

 

[new6ton]

It's almost always software. The instrument response depends on the hardware, but the calibration is done to eliminate the contribution from the instrument response by correcting the spectrum with (usually) a polynomial function of some kind.

Did you mean software as in their PC application software or firmware inside the hardware?

When spectra is readout from CCD? Is processing done inside unit or in the application software? They describe the unit as having an onboard programmable microcontroller that provides flexibility in controlling the spectrometer and accessories.

 

[TeethWhitener]

Did you mean software as in their PC application software or firmware inside the hardware?

You’ll have to check your specific spectrometer for the answer to that.

Is processing done inside unit or in the application software?

Yes, all of the above. Probably. Again, this is a question to ask your spectrometer vendor.

 

[new6ton]

You’ll have to check your specific spectrometer for the answer to that.

Yes, all of the above. Probably. Again, this is a question to ask your spectrometer vendor.

In this guide there is this page about ASTM E 1840-96. Which is the standard for calibration. http://georaman2014.wustl.edu/previ...rnationalschool/003 Calibration_MC-Caumon.pdf

I can't find a text of ASTM E 1840-96. I think it's been sold for $67. I don't know if the spectrum above comes from ASTM E 1840-96 itself. But the 802 cm-1 peak has so called "Standard Deviation" of 0.96. How do you interpret this? Does it really mean it can change by over 96%?

If you (or others reading) go to your lab today or tomorrow. Please take a look at your cyclohexine and please share the raman spectrum you are getting. For the 802 cm-1 peak described as ring breathing. What is the exact atomic identity or molecular structure or bonding involved in it? Why can it vary so much?

 

[TeethWhitener]

But the 802 cm-1 peak has so called "Standard Deviation" of 0.96. How do you interpret this? Does it really mean it can change by over 96%?

No. It means that to within one standard deviation, you will find the peak in question within 0.96 cm^-1 of 801.3 cm^-1. Note also that this is a peak position measurement and not a peak intensity measurement.

 

[new6ton]

No. It means that to within one standard deviation, you will find the peak in question within 0.96 cm^-1 of 801.3 cm^-1. Note also that this is a peak position measurement and not a peak intensity measurement.

This is a company selling a sealed cyclohexane reference for testing. https://www.starna.com/cyclohexane#typical-excitation-and-emission-spectra

"This reference material may be used to qualify the wavelength scale of Raman Spectrometers between 380 cm-1 and 3000 cm-1. It is cited for this purpose in ASTM E1840 - 96(2014).
Reagent grade cyclohexane, permanently sealed by heat fusion into a 10 mm quartz cuvette "

It shows this spectrum when they used a 512nm raman analyzer.

If different raman spectrometers scan it. What do you think are the results? They vary too or should the intensity look identical with the 802 cm-1 peak higher. If different raman spectrometers won't show the same result. Then what's the use of it for reference?

 

[new6ton]

This is a company selling a sealed cyclohexane reference for testing. https://www.starna.com/cyclohexane#typical-excitation-and-emission-spectra

"This reference material may be used to qualify the wavelength scale of Raman Spectrometers between 380 cm-1 and 3000 cm-1. It is cited for this purpose in ASTM E1840 - 96(2014).
Reagent grade cyclohexane, permanently sealed by heat fusion into a 10 mm quartz cuvette "

It shows this spectrum when they used a 512nm raman analyzer.

View attachment 251296
If different raman spectrometers scan it. What do you think are the results? They vary too or should the intensity look identical with the 802 cm-1 peak higher. If different raman spectrometers won't show the same result. Then what's the use of it for reference?

Teethwhitener. I have read your reference 3 times. I think most raman spectrometers now were calibrated with luminescence source. Do you think some weren't? Since most are calibrated already. Then they should all show the same Cyclohexane peaks above. Still no?

from your reference:

 

[mjc123]

If different raman spectrometers scan it. What do you think are the results? They vary too or should the intensity look identical with the 802 cm-1 peak higher. If different raman spectrometers won't show the same result. Then what's the use of it for reference?

The quoted blurb says it's for calibrating the wavelength, not the intensity. It can still do that even if the intensities are for some reason not the same.

 

[new6ton]

The quoted blurb says it's for calibrating the wavelength, not the intensity. It can still do that even if the intensities are for some reason not the same.

In the paper TeethWhitener provided. It mentioned about calibration via luminescent sources. The paper seems to suggest these are accurate enough to get the intensity. So if all manufacturers do this. Why can't we have a standard intensity at least for 532nm raman systems?

This is the quote regard the figure in my last message:

"
Correction for instrument response requires either an
accurate knowledge of the wavelength dependence of all
components in the spectrometer, or a standard source with
known output. The former approach is difficult and generally
impractical, so several approaches based on standard
light sources or luminescent standards have been
reported.6,20–23 Consider Figure 6, which shows the perturbation
of a standard source output (omegaL) by the response
function (R) to yield an observed spectrum (SL). If omegaL
is known, then acquisition of SL suffices to determine R.
Once R is known, any uncorrected sample spectrum (Ss)
may be divided by R to yield a corrected sample spectrum
(omegaL). Provided the standard source output is known and
available in digital form, the response correction procedure
may be automated through the spectrometer software."

I assume that all 532nm raman systems use standard source with known output in the calibration, so all should have similar and accurate intensity? Or does it mean only some do that? Or even with standard source with known output, it is still not accurate? Why? The paper didn't give say it is still not accurate even with this.

If all do it. Why are the intensity still not accurate? I know 785nm raman systems have more variable so let's ignore 785nm systems for now.

 

[new6ton]

The point of calibration is to try to mitigate differences across instruments and instrument settings. Ideally, if you know the instrument response exactly, the absolute Raman cross section of a peak (the part most closely corresponding to $\langle f| \alpha | i \rangle$) is the integrated area of the peak. But this still is dependent on wavelength, sample geometry, etc., etc. Here’s what looks like a decent reference on Raman calibration at first glance:
https://www.thevespiary.org/library...tometric.Standards.for.Raman.Spectroscopy.pdf
You can see that the community has pretty much come to a consensus about the value of a handful of cross sections (the article mentions the benzene 992 resonance in particular) and calibrate all instruments according to those.

The reason this question is so tough is because the scattering cross sections and the instrument responses both change with wavelength. So let’s say for example you have an excitation wavelength of 532 nm. The resonance at 800 cm^-1 is at a wavelength of around 560 nm, while the resonance at 2800 cm^-1 is at around 625 nm. You need a standard at each of those wavelengths to know the instrument response, but you need to know the instrument response to know how the standard should behave at those wavelengths. It’s a bootstrapping problem.

I'm aware of this bootstrapping problem. But this can be wholly addressed by luminescent source calibration, right? Or still not complete and why? I read in your reference:

"
The article describes three steps toward achievement of
a corrected Raman spectrum which accurately represents
the scattering intensity of a given sample as a function of
Raman shift:
1. Reproducibility of observed scattering intensity.
2. Correction for variation of instrument response across
a Raman spectrum.
3. Determination of absolute scattering intensity and absolute
Raman cross-sections.
As discussed below, the first two objectives may be
achieved by straightforward calibration, while the third is
much more involved. Fortunately, the great majority of
Raman applications do not require assessment of absolute
intensity, with the accompanying experimental difficulties.
Reproducible Raman intensities may be achieved with sensible
design of sampling optics and reasonable experimental
care. Although response function correction is not yet routine,
it is readily applied and quite useful. The result of
these two calibration steps is a Raman spectrum which
accurately reflects relative Raman scattering intensities and
is useful for library searching, quantitative analysis, and
comparison of spectra between laboratories. Current applications
of Raman spectroscopy can be broadened significantly
without calibrating absolute intensity, and absolute
measurements of cross-sections are generally left to the
specialist."

 

[TeethWhitener]

I think I already answered this in post 8:

Ideally, if you know the instrument response exactly, the absolute Raman cross section of a peak (the part most closely corresponding to $\langle f| \alpha | i \rangle$) is the integrated area of the peak. But this still is dependent on wavelength, sample geometry, etc., etc.

I’m not sure I can add anything to that or the link I provided.

 

[new6ton]

I think I already answered this in post 8:

I’m not sure I can add anything to that or the link I provided.

In that message. Your first paragraph was optimistic, but the second paragraph was pessimistic. So I thought the pessimistic ruled over the optimism. Your optimisic 1st paragraph line was:

"The point of calibration is to try to mitigate differences across instruments and instrument settings."

It was followed by the second pessimistic paragraph line which was:

"The reason this question is so tough is because the scattering cross sections and the instrument responses both change with wavelength."

For a beginner. He may think it is hopeless. Had you reversed them, then the meaning changes. Which is:

"Without calibration, the scattering cross sections and the instrument responses both change with wavelength.

But with Calibration. It can address it and mitigate differences across instruments and instrument settings."

Correct? :) Thanks anyway.

 

[TeethWhitener]

I’m not sure I see the difference, but ok sure.

 

[new6ton]

I’m not sure I see the difference, but ok sure.

In page 9 of your reference is this:

For 785nm raman, the right peak is higher. The system is calibrated using luminescent sources. This means it is the accurate peaks of Cyclohexane. They don't have data yet for 532nm systems. But if the 532nm system is superbly calibrated in the luminescent source. It should also produce the near identical peaks with the 2853 cm-1 higher than 802 cm-1, right?

 

[TeethWhitener]

What makes you think the Raman cross section is NOT wavelength dependent?

 

[new6ton]

What makes you think the Raman cross section is NOT wavelength dependent?

Even if the Raman cross section is wavelength dependent. If the 532nm and 785nm systems are both calibrated using accurate luminescent sources. They can adjust both the instrument response and show similar results. This is what is supposed to be the purpose of calibration. This was how I understood the paper after reading it 5 times. They didn't say or emphasize this. But are you saying that even after calibration, it still won't take into account this raman cross section being wavelength dependent, and even calibration can't solve it?

 

[new6ton]

What makes you think the Raman cross section is NOT wavelength dependent?

You agreed with this summary:

"Without calibration, the scattering cross sections and the instrument responses both change with wavelength.

But with Calibration. It can address it and mitigate differences across instruments and instrument settings."

But you agreed it only worked for either 532nm or 785nm, etc. instrument when calibrated on the *same* analyzer?

And when you compare 532nm to 785nm raman systems, it no longer works because of the scattering cross section dependent on wavelength?

So does the following statement apply to the former or latter? I guess the latter also so calibration doesn't really solve it (across different raman system like 532nm vs 785nm?) It's very subtle but important distinction that one can't surmiss from your message #8 or even the paper itself.

I understand that:

 

[TeethWhitener]

even calibration can't solve it?

Why should calibration solve something that isn’t a problem? Raman cross sections are wavelength dependent. Why do you want to discard that information?

 

[TeethWhitener]

I’ve reached the end of what I can tell you about Raman calibration standards. Maybe someone else here can better address your concerns.