Skip to main content

Analytical Chemistry

A Rebirth for IR Spectroscopy?

As new technologies develop, they can end up bringing back some old ones. That might be the case for infrared spectroscopy. Most organic chemists use it infrequently – in my case, years go by between taking an IR spectrum. There are infrared sensors that can go right into a reaction mixture (or flow stream), and these can be very useful, but they’re also quite expensive – not something that everyone has sitting around ready to go.

But nano materials can do a lot of odd things with electromagnetic waves, and it turns out that you can make surfaces that respond very robustly to changes in infrared wavelengths through a combination of effects (piezoelectric, nano-scale plasmonic). This opens up the way to doing extremely small-scale analysis in real time, directly from solutions or on crude material without any sample prep. This new paper in Angewandte Chemie demonstrates this down into the picogram range with indomethacin as a test compound (here’s another paper from the same group. To put it mildly, you are not going to get that kind of sensitivity with any classic IR technique, and this would seem to open the door to putting such sensors in all sorts of apparatus. Flow chemistry, LC purifications, tandem analysis with mass spec and other techniques – if these things really are that tiny, you could drop them into all sorts of applications, and there are really are situations where a vibrational spectroscopy fingerprint could tell you things that other techniques can’t.

So bring on the high-tech nanomaterials. I just wonder how readily they foul under real world conditions, and what range of solvents and pHs they’ll stand up to, but the field is still very young.

34 comments on “A Rebirth for IR Spectroscopy?”

  1. John Wayne says:

    I love IR spectroscopy! Insert joke about age here. It is a nice technique that compliments NMR data very well, and probably should be more widely utilized.

    1. Nicholas Yee says:

      It is simple, cheap, robust, and easily learnt and interpreted. Really good for teaching undergraduate students (like me 2 years ago) the ropes. Nearly all the educational labwork I do has UV spec as well as NMR to conjure up various problems and demonstrate different lab techniques to students.

      1. UudonRock says:

        I do enjoy IR with NMR and GCMS. Used to like AA way back when. Wonder if that’s due for a come-back…

      2. not so old timer says:

        I remember my undergrad Spectroscopy course final was NMR, IR, and chemical formula. Interpret and draw the molecule. That was the last time I looked at an IR spectra. LCMS and NMR are good for me now, although I could use GCMS currently, darn capital request restraints.

    2. Anonymous says:

      That may be true of big research departments at well-funded universities or national labs, but in industrial R&D labs and the resesrch labs of small colleges who don’t have a few million lying around to burn on a cryogenic NMR system, IR has never fallen out of favor.

  2. Ksr15 says:

    Maybe this new advance in IR spectrometery will allow them to get a better look at that one molecule derived from n-ammino azidotetrazole, C2N14!

  3. anon the II says:

    I say phooey on this IR stuff. I’ve run enough IRs in my life and I hope never to run another. I’ve made Nujol mulls, KBr pellets and even run samples in the sacred IR salt cells that live in the locked cabinet less some moron put water in them. You never got information back close to the rate you put work in. If it hadn’t been for those damn JOC requirements, it’s have died even faster. I’d rather have a box of TLC plates than an IR machine. Even FT-IR, which made those old prep methods obsolete and reduced the compound requirements to zilch, couldn’t save it. And LC/MS put the final nail in the coffin.

    IR is dead! Let it rest in peace.

    1. John Wayne says:

      The ATR detector removed all the annoying prep from the technique, but it does have some usually minor drawbacks.

      1. Paus says:

        What do you find the drawbacks to me? We’re looking at getting one for our lab

        1. DRP says:

          As an inorganic chemist, only great if you can get one in your glovebox!

          1. anon says:

            There are sealed IR cells that you can put your sample in a glove box and take it out, run it out in the air.

        2. John Wayne says:

          There are (at least) two things that can confound the use of an ATR plate:
          1. The ATR material absorbs more strongly than a KBr or salt plate, so the signal to noise isn’t as good as the old techniques. This is usually minor, but people accustomed to looking at old technique spectra need to get used to the new appearance. Different ATR materials have different absorption areas and strengths, so you can pick something complimentary to your use.
          2. If your sample does something odd, like recrystallize, when you put it on the plate you can get some odd results. If you have an oil or noncomplex solid it is faster to get an IR than run a TLC.

    2. Chemperor says:

      Interesting take. I work with metal carbonyl complexes (CO ligands for you organic types) and IR is probably the most useful tool in my toolbox. NMR is all but worthless in many cases, and $16M gets you a perfectly serviceable workhorse instrument. Long live IR!

      1. Nick K says:

        Quite so. If you’re doing functional group manipulations on carbonyl compounds, terminal acetylenes, nitriles and the like, IR is really clean and diagnostic, unlike NMR. In any case sample preparation with a solution cell is trivial.

      2. Isidore says:

        $16M?! Does it come with its own power plant?

        1. Chemperor says:

          Hmm. I forget not everyone spent time in industry. $#M means thousands for most financial folks, $MM means millions.

          1. Isidore says:

            Interesting, I spent time in both industry and academia, and I have used and seen others use K for thousands but since I’ve never bough anything that cost more than $600,000 I don’t know what people would use for millions.

  4. CJR says:

    Another neat enhancement technique: Khine lab used shrinkydinks to concentrate material onto plasmonic arrays (built by metal buckling on the shrinkydinks) before IR spectroscopy :

  5. Patrick Sweetman says:

    Could they be incorporated into something like a FitBit to continuously monitor plasma drug concentrations including endogenous drugs such as hormones?

    1. anon says:

      I was thinking about similar things and assuming that they already got some patents. But, “The authors declare no conflict of interest.”

      1. Anonymous says:

        Hah! The campus news mill where I used to be raved about a new development (a small clinical trial in humans) by a research group and how they were starting a company to exploit and market it.
        The journal article contained this disclaimer:
        Conflict Of Interest Statement: Guarantor of the article : [Professor X]
        Potential competing interests: None.
        Submitted / Received: [Date]
        THE PATENTS WERE FILED 2 years before the submission date!

        I suggested to the guy teaching the “ethics for student researchers” course that he include that as an example. (I don’t know if he did or did not.)

  6. regdoug says:

    Interesting to me that it is not in common use in chemistry. I’m a mechanical engineer and FTIR is usually the first tool we grab to analyze unknown polymers.

  7. anon electrochemist says:

    Derek and the rest of the organic crowd are mentally underestimating the sensitivity of classical FTIR here, because they don’t think of it as a quantitative tool.

    I’m a surface guy who does infrared microscopy. A typical sample can be 5 picograms of polymer (a 100nm thick film 10x10um) giving a peaks of maybe 0.2 Abs units, which is considered a lot of material. That’s using a crappy old 90’s instrument, and the field hasn’t progressed much since the 80s, even with synchrotron sources. People have been studying adsorbed monolayers using IR since the earliest days, which is low-femtogram range. It’s freakin’ sensitive.

  8. Professor Electron says:

    As well as the sensitivity, the use of liquid state samples is a big advantage because it can be automated and reduces problems of inhomogeneity. Haven’t been able to read the paper yet (paywall) so don’t know about what solvents they used. LC-IR would be a great complement to LC-MS for low-level analysis if pg levels are really detectable in solvent mixtures (MeOH/ACN/H20). Interpretation isn’t as easy as NMR, but DFT calculations are an option for distinguishing isomers – and the current method for pg amounts is MS/MS which isn’t easy to interpret either.

  9. David Stone says:

    Quite a few researchers at my institution are using FTIR-ATR on a routine basis. With a diamond ATR element you can go to pretty low wave-numbers. Not the pg sensitivity being described in the article, but way better than the old nujol mull and KBr pellet techniques from when I was an undergraduate. Plus, you can run aqueous solutions too…

  10. Jason says:

    ATR makes IR so convenient. The latest ones are portable and can even include a microscope camera so you can see which part of an inhomogeneous sample you are analyzing. Robust too: last year I had an undergraduate student looking at the hardening kinetics of coatings. She was supposed to be getting them to set on a slice of germanium (which is transparent to IR) sat on top of the ATR. She later sheepishly admitted her best kinetic runs involved direct formation of the coating on the ATR window, once she learnt she could use solvent and “gentle rubbing” to clean it afterwards. Said ATR is working fine.

    1. tangent says:

      Jason, how thick is the germanium slice for that, like a fraction-of-a-wavelength film? I would have figured even a thin spacer would put the sample out of reach of the evanescent wave.

      (The whole exploitation of evanescent waves seems like a cheat code, to be honest.)

      1. anon says:

        I suspect that the refractive index of germanium is high enough to support total internal reflection at the germanium-sample interface, but that it’s difficult to squeeze it down tightly enough to prevent reflective losses at the ATR crystal-germanium interface.

  11. Anonymous says:

    Harry H. Wasserman (RIP) told stories of his early days getting his research started at Yale (1949-early 50s; pre-NMR, of course). There was no IR at Yale so he and students would drive up to Cambridge where Woodward let him use his IR in long, marathon sessions. It wasn’t an automatic instrument. You had to manually adjust the wavelength and manually record the absorbance at each setting and then plot your data to see a nice curve (that we all now take for granted).

    Most useful IR book on my shelf: Nakanishi and Solomon, Infrared Absorption Spectroscopy 2nd Edition. Great tables and explanations.

    FTIR-ATR – so simple! If you have access to one, why not get the data when it’s so easy to get?

    1. Curious Wavefunction says:

      There’s a photo on the internet somewhere featuring Woodward studying an IR spectrum with a magnifying glass. That’s how he would compare the synthesized compound with the natural one, and would insist that his students do so too.

  12. Anon says:

    Back in undergrad our entire stock of NaCl IR plates went missing. Turns out the lab tech left them to soak overnight in a water bath.

    What *is* that Nujol stuff anyway?

  13. bcpmoon says:

    IR is the most common technique for identity – one fingerprint and that´s it. Fast and easy. And I had once a polymorph problem where I was lucky to have a diagnostic peak to discern both polymorphs. It was just 2221 vs. 2223 wavelengths but worked like a charm as a Monitoring. Much better than doing XRPD in production.
    IR is just a great tool – not a wonder potion but under the right circumstances very versatile.

  14. TWS says:

    Have to echo many of the comments here – I think IR is seriously underestimated by some. While I agree that mulls and NaCl plates were a complete nightmare (still remember with hatred from my undergrad), ATR makes everything so much easier. The initial instrument cost and upkeep are on a different order of magnitude to NMR. While I’m sure NMR beats IR hands down for organic synthetic chemists, from the comments here (and my own experience) I’m not so sure this is true for some other areas of chemical / materials / polymer science.

    To add another area to the list of fields which people have mentioned, we use FTIR for gas-phase in situ monitoring of various biological processes and more fundamental phys. chem work on hydrogen bonded dimers in the gaseous state. You can do the former with GC-MS, but a high resolution, line-resolved IR spectrum is a thing of beauty to look at…..

  15. Albert says:

    We have FT-IR system in our department (industry), but I personally rarely use it. On the other hand ReactIR and Raman probes are invaluable for our physical organic department. Those guys who investigate mechanisms, kinetics etc. Believe it or not, but there are such people in industry as well.

Comments are closed.