Vibrational spectroscopy (NIR,MID, Ra... Log Out | Topics | Search
Moderators | Register | Edit Profile

NIR Discussion Forum » Bruce Campbell's List » I need help » Vibrational spectroscopy (NIR,MID, Raman) versus electronic (UV-Vis) spec. « Previous Next »

Author Message
Top of pagePrevious messageNext messageBottom of page Link to this message

Javier (Lazarillo)
New member
Username: Lazarillo

Post Number: 4
Registered: 3-2006
Posted on Friday, March 17, 2006 - 5:56 pm:   

Hi guys,

This post is getting hot now! Pity most of you were on hols at the beginning of the week! My question basically originated in a research work I am doing at University and which I had to hand in last Wednesday. But all your discussions are very helpful!

I didn't really look at UV spec. because I read somewhere the absorption bands were not very good. In my conclusions I have suggested that NIR would do the job appropriately. Second came Raman. I decided this way because NIR is a much more mature technology and therefore cheaper. But basically both would do the job. There is loads of overlapping in the NIR spectra but the calibration is good if the data is regressed by using multivariate data analysis (partial least squares). The Raman spectra is cleaner and tend to give slightly better regression results but the equipment is more expensive.

I do have a couple of research articles comparing both techniques. Both spectra are shown there. You have been very helpful; if someone its really interested in getting more details I will be happy to forward them to you. Please, just send me a private email: [email protected].

Thanks a lot for your help.
Top of pagePrevious messageNext messageBottom of page Link to this message

(Unregistered Guest)
Unregistered guest
Posted on Friday, March 17, 2006 - 3:40 pm:   

An advantage of UV-Vis is that they've been around a long time and some models are VERY inexpensive - less than $1-2K, I think. More important in an academic environment than in industry.

\o/
/_\
Top of pagePrevious messageNext messageBottom of page Link to this message

Gabi Levin (Unregistered Guest)
Unregistered guest
Posted on Friday, March 17, 2006 - 3:33 pm:   

Hi Lois,

Thanks a lot, it does help to better understand the issue, not that I want to become a UV guy. Do you have something you can post as an attachment?

And do you expect to have same distinction when they will be mixed with each other?

I also undersatnd that scanning UV spectrometers will resolve the peaks, if so, the originator can take his pick between UV and NIR - his call of course - but I would still go with NIR, but I am of course very "impartial" on that subject. From the practical aspect, NIR does offer the ability to add compounds that do not respond so well to UV, if such a potential situation can arise.


Gabi Levin
Top of pagePrevious messageNext messageBottom of page Link to this message

(Unregistered Guest)
Unregistered guest
Posted on Friday, March 17, 2006 - 3:33 pm:   

Besides, Lois, all you have to do is click on the link at the bottom of any of these messages and you can read the WHOOOOOOLE thing!!!

\o/
/_\
Top of pagePrevious messageNext messageBottom of page Link to this message

Howard Mark (Hlmark)
New member
Username: Hlmark

Post Number: 13
Registered: 9-2001
Posted on Friday, March 17, 2006 - 3:31 pm:   

See, there's nothing like someone who knows what they're talking about!!

\o/
/_\
Top of pagePrevious messageNext messageBottom of page Link to this message

Lois Weyer (Lois_weyer)
New member
Username: Lois_weyer

Post Number: 12
Registered: 7-2002
Posted on Friday, March 17, 2006 - 3:25 pm:   

Hi guys,

I missed the beginning of this thread, but I have to add that the UV spectra of liquid xylenes are quite distinctive due to superimposed vibrational fine structure. The peaks are nice and sharp, especially when derivatives are used. And the absorptivities are around 2 l/g-cm. I used to measure them dissolved in various solvents.
Top of pagePrevious messageNext messageBottom of page Link to this message

(Unregistered Guest)
Unregistered guest
Posted on Friday, March 17, 2006 - 3:10 pm:   

BTW - the eighth hit on that same search page is an article (in .PDF format) on NIR spectra of xylenes.

\o/
/_\
Top of pagePrevious messageNext messageBottom of page Link to this message

Howard Mark (Hlmark)
New member
Username: Hlmark

Post Number: 12
Registered: 9-2001
Posted on Friday, March 17, 2006 - 3:06 pm:   

Gabi - I'm not such an expert on UV either, I just remembered that conjugated and aromatic electrons are good for UV absorbance. I always thought of the electrons in these conjugated and resonant systems as acting like little antennas to pick up the local E-M "signal", that's how I remembered about it. I missed the part about their being vapor-phase spectra, I guess.

When the atomic environment is constant from one molecule to the next, which is what you can get in gases (at low enough pressure so that each molecule is isolated) and crystalline solids (where the next atom over is always the same and in the same place) then you can get sharp bands. But in liquids, where the environment changes continuously from molecule to molecule, and for each molecule over time as other atoms and molecules come and go, then there is continuum of environemnts and the sharp structures get spread out by the continuum of interactions with the other molecules, giving rise to broadening of the bands.

As an analogy, we can consider the difference between sucrose NIR spectra in the solid versus when the sucrose is dissolved in water, then the same effect due to the liquid is very obvious.

I expected something of that sort when I said somewhat early on in this thread that UV should be good for total xylene content, as opposed to NIR or Raman which would be better for the individual xylenes.

\o/
/_\
Top of pagePrevious messageNext messageBottom of page Link to this message

Gabi Levin (Unregistered Guest)
Unregistered guest
Posted on Friday, March 17, 2006 - 2:40 pm:   

Hi Howard,

I guess I am not much of a UV man at all, but the reference you indicated is for vapor phase - I am not sure what woud the liquid phase be like - because the situation shall be different even for the electronic transitions, not just the vibrational. By the way - the difference in wavelength between the two 0,0 transitions of the two forms reported is only 2.5nm, how well the conventional UV spectrometers will resolve them? I don't have experience with those, definitely regular filter based analyzers will not do the job. The spectrometer they used in the study says 3meter - must be that long to provide sufficient resolution. Scanning UV spectrometers - I am not familiar with them, could you comment on that?

I assume the path length will be long enough for practical implementation.


Thanks,

Gabi Levin
Top of pagePrevious messageNext messageBottom of page Link to this message

Howard Mark (Hlmark)
New member
Username: Hlmark

Post Number: 11
Registered: 9-2001
Posted on Friday, March 17, 2006 - 12:53 pm:   

'Fraid not, Gabi - the resonance in the ring makes for very nice UV absorbances. I did a google search for "xylene spectrum" and quickly (second or third hit) came up with the following discussion of the UV spectra of xylenes (indicating more differences than I expected, either) with references:

http://scitation.aip.org/getabs/servlet/GetabsServlet?prog=normal&id=JCPSA6000020000008001248000001&idtype=cvips&gifs=yes

Howard

\o/
/_\
Top of pagePrevious messageNext messageBottom of page Link to this message

Gabi Levin (Unregistered Guest)
Unregistered guest
Posted on Friday, March 17, 2006 - 11:45 am:   

Just a few words - those chemicals are all totally transparent in the visible - they have no spectrum of ineterst there.
To get into elecronic transitions of molecular orbitals will require UV, this will probably require deep UV, not practical and probably the differecne in energy will be very small because of the almost identical isomer structure of the electronic orbitals. The NIR to my taste and experience with some ortho, meta and para phenols is a better chance because there is sufficient shift of the C-H wavelength strech wavelengths.
I just checked the Atlas of Infra Red Spectra - I found O, and M xylene, no para, but it does not matter - it will differ from the other two anyway - and there are sufficient differences in peak position - over 5nm and also in shape of peaks. The ethyl benzene is way out in wavelengths and shapes of the peaks from the xylenes - more than 30nm apart.

So, best choice will be NIR because you can have a long path length, over 2 mm and get good signal, low noise, better than Mid-Ir where the pathlength will be short, and better than Raman because of very strong absorption peaks.

If you contact me directly I will send you information

Gabi Levin
Brimrose Corp.
[email protected]
Top of pagePrevious messageNext messageBottom of page Link to this message

hlmark (Unregistered Guest)
Unregistered guest
Posted on Thursday, March 16, 2006 - 9:39 pm:   

Javier - don't believe everything David says about how wonderful it was in Orlando. If I wanted to be hot and sticky, I could have stayed in New York and waited for Summer!

Regarding your question, however: I'm pretty sure that both Mid-IR and Raman spectroscopy are going to be more selective than UV. I don't know what the relative absorption coefficients are for your sample materials in the UV compared to, say, the Mid-IR, but even if the Mid-IR absorption is less intense, the greater selectivity may make up for it. It's not a simple comparison, by any means.

Near-IR will be somewhat less selective as well as being somewhat less sensitive than Mid-IR. To some extent, that's compensated for by the greater S/N achievable with Near-IR, plus the fact that the material available for use (for, e.g., cells) are easier to handle and deal with.

Raman is relatively insensitive, compared to any of the other methods.

The choice will also depend on what types of samples you intend to measure. If the xylenes will be present in trace quantities, then mid-IR or UV is your best bet. The choice between those can be determined by whether you want to measure each type of xylene individually (in which case you should use mid-IR) or the total xylene content (in which case you should use UV).

If the xylenes are present in large concentrations, (i.e., at least several percent) then you could probably use any of the techniques.

Howard

\o/
/_\
Top of pagePrevious messageNext messageBottom of page Link to this message

David W. Hopkins (Dhopkins)
New member
Username: Dhopkins

Post Number: 70
Registered: 10-2002
Posted on Tuesday, March 14, 2006 - 2:22 pm:   

Javier,

I think that would be an eminantly doable application in NIR, or IR. I'm not familiar with the spectra in the UV, but I'd guess they absorb in the UV too, below 280 nm. I'd recommend that you contact vendors and find out about cost and features of the instrument, and pick the features you need.

Yes, it is pretty warm and humid in Orlando, but it was in the 60's when I left Battle Creek, still a pleasant change....

Best wishes,
Dave
Top of pagePrevious messageNext messageBottom of page Link to this message

Javier (Lazarillo)
New member
Username: Lazarillo

Post Number: 3
Registered: 3-2006
Posted on Tuesday, March 14, 2006 - 12:09 pm:   

Hi

Thanks for your answers.

I am dealing with a mixture of xylenes (meta, ortho and para) and ethyl benzene.

Apart from the sensitivity question I raised before I now have another doubt. I have read a few papers dealing with the measurement of those compounds; all of them use a vibrational technique (i.e., either NIR, MIR or Raman spectroscopy)�

Is that because that mixture is organic and transparent in the UV/Vis region? Or is it because UV/Vis would not be able to distinguish between the isomers due to broad and overlapping peaks?

Thanks�

PS. Lucky those guys in Orlando. It is snowing here, in the North of England!
Top of pagePrevious messageNext messageBottom of page Link to this message

David W. Hopkins (Dhopkins)
New member
Username: Dhopkins

Post Number: 69
Registered: 10-2002
Posted on Tuesday, March 14, 2006 - 9:38 am:   

Javier,

We need more information to respond. What application are you interested in? What is the nature of the samples, and what parameters do you want to measure?

Best wishes,
Dave
Top of pagePrevious messageNext messageBottom of page Link to this message

MPDC (Unregistered Guest)
Unregistered guest
Posted on Tuesday, March 14, 2006 - 7:40 am:   

That will entirely be dependent on what you are trying to measure. Good luck trying to monitor RON in petrol with a UV-VIS. If you however want to measure how much water you just added to your coca-cola syrup, I would imagine VIS would be perfect... P.S. don't expect a lot of answers here, this week is PITTCON week, and everybody is soaking up sun in Orlando...
Top of pagePrevious messageNext messageBottom of page Link to this message

Javier (Lazarillo)
New member
Username: Lazarillo

Post Number: 1
Registered: 3-2006
Posted on Monday, March 13, 2006 - 8:56 am:   

Hello there

When it comes to in-situ process analysis, someone just mentioned to me that vibrational spectroscopy (NIR, MIR or Raman) is likely to be more sensitive than UV/Vis spectroscopy...


My problem is that I dont quite understand that. I would greatly appreatiate if someone could either discuss this or point me
to a good reference (either book or in the web).

Thanks a lot

Add Your Message Here
Posting is currently disabled in this topic. Contact your discussion moderator for more information.