Josh P. Roberts has an M.A. in the history and philosophy of science, and he also went through the Ph.D. program in molecular, cellular, developmental biology, and genetics at the University of Minnesota, with dissertation research in ocular immunology.
We’re all familiar with UV/visible (UV/Vis) spectrometry, which excites certain bonding electrons of molecules like proteins and nucleic acids. The spectrometer measures how much light of a given wavelength (or ratio of wavelengths) is absorbed by a particular type of material—such as protein or nucleic acid—from which researchers can determine how much of that material the sample contains.
Molecules also absorb in the IR range. But in this case lower energy and longer wavelengths induce vibrational changes in various functional groups. Carbonyls, ethers, esters and carbon double and triple bonds all have their own characteristic vibrational frequencies, for example, and the absorbance of a CH3 “stretch” will differ from that of a CH3 “bend.” Querying the IR spectrum can indicate the presence of protein in a hazmat/biosecurity setting, probe the interactions of lipids in cell membranes and help the pharmaceutical industry discern the secondary structure of drug candidates and metabolites.
Principle of FT-IR
IR spectra typically are measured using a Fourier transform-IR (FT-IR) spectrometer. Such instruments share fundamental principles with UV/Vis spectrometry: Light interacts with the sample, and the energy that reaches a detector is recorded. This is compared to what is detected when there is no sample present, and the difference between these is plotted against wavelength to give an absorbance spectrum. (Absorbance is calculated directly from transmittance.)
But rather than collecting the spectrum by scanning a narrow band of wavelengths at a time, FT-IR uses a Michelson interferometer to gather the entire spectrum at once. The interferometer splits the incoming light, sending it to both a fixed and an oscillating mirror. The path length of light from the oscillating mirror will vary, and as it recombines with that of the fixed mirror, it creates a changing pattern of interference. The resulting interferogram—generated in a matter of seconds and looking like an EKG spike—is uninterpretable to the human eye. Thus a complex formula called a Fourier transform is used to convert the raw interferogram into the familiar plot of wavelength vs. absorbance. “The word ‘Fourier transform’ is just mathematical magic in a box,” quips Michael Bradley, marketing manager for FT-IR products at Thermo Scientific.
Because FT-IR collects the entire spectrum simultaneously in a relatively short time, many sequential runs are typically collected in an FT-IR instrument. “You can do a number of scans—8, 16, 32—in about the same time [it would take a non-FT instrument to do so] and average them together,” points out Mark Talbott, the molecular spectroscopy product manager at Shimadzu.
“The noise averages out, and the signal average increases, so you get better signal-to-noise ratio.”
Fleshing out the structure
Complex molecules, with their multiple chemical pendants, typically produce not a single FT-IR peak but rather “a fingerprint, or a somewhat unique set of absorptions,” Bradley explains. The majority of users will search for a match to that spectrum in a digital reference library. Some spectral databases are maintained by the equipment vendors, and proprietary, public and commercial libraries exist, as well. (For a listing of resources, see www.lib.utexas.edu/chem/info/spectra.html.)
An advanced user can also look at the spectrum ab initio (that is, without a reference), and, by knowing which absorption peaks correspond to which functional groups, build a picture of the molecule responsible for the spectrum.
Yet that may not be enough. There may be five or six carbonyls from a molecule that make up a single peak, for example. “I can then begin to look [at] what happens to those individual carbonyl moieties as the molecule is changed. Either addition of water, removal of water, addition of heat, removal of heat [or] vacuum pressure [will give] me the ability to actually figure out what’s going on in the molecule. What’s the shape? What affects what?” Talbott says.
Researchers also can supplement FT-IR with orthogonal techniques like nuclear magnetic resonance to garner information about neighboring protons, for example, and mass spectrometry to identify molecular fragmentation patterns.
The “IR” in FT-IR typically refers to midrange IR, around 350 to 4,600 cm-1; water, in which many biological samples are prepared, absorbs strongly in this range. To remove this signal, many FT-IR instruments can be equipped to query in the near-IR range (approximately 4,600 to 12,000 cm-1) by changing the source, beam splitter and detector. “If you move up into the near-IR, you begin to look at overtones—for example, NH overtones, CH overtones and OH overtones,” Talbott says. “So by looking at the overtones, you’ve kind of moved water out of the way.” Some instruments can also be equipped to query in the far-IR range (approximately 180 to 350 cm-1), which is more useful for metals and inorganics.
Accessorize
Until the last 10 to 20 years, solid samples were typically prepared by grinding them with potassium bromide (KBr) salt (which is transparent to IR radiation), melting the mixture under pressure to essentially turn it into a thin glass disc and mounting that disc to the instrument. Liquids could be assayed by placing them between salt-plate “windows” (also often comprised of KBr) which are then mounted into a cell and placed in the instrument.
More recently, attenuated total reflectance (ATR) has supplanted KBr transmittance as the leading technique for most FT-IR analyses, says Bradley. In ATR, the light is sent into a crystal prism where it is reflected internally at an angle. Wherever there is a reflection in the prism, it produces an evanescent wave. That wave only extends outside the prism a very small distance, but it’s enough to penetrate into and be absorbed by any sample that is pressed intimately against it, allowing an absorbance spectrum to be generated just as with transmission.
“If you’re doing routine FT-IR on routine organic compounds—liquids, solids—then you would probably want to get a really basic FT-IR with a diamond ATR accessory, because that allows you to run samples very quickly with minimal sample prep,” notes James Windak, supervisor of instrument services in the chemistry core lab at the University of Michigan. As an added bonus, such a setup attenuates the “massive -OH water band,” adds Windak, whose facility houses two FT-IRs from PerkinElmer.
Other applications, such as looking at samples as small as a few microns square, may indicate a microscope accessory—which can be used in either reflectance or transmission mode— Windak says. This, in turn, may require a special liquid-nitrogen-cooled detector, “because of the sensitivity you need if you’re looking at extremely tiny apertures.”
And there are many other accessories available for FT-IR, as well, including gas cells that introduce extreme heat or pressure. It’s best to consider what kinds of samples and applications you anticipate running and to make sure your instrument will be able to handle the appropriate accessories. “Is the sample compartment large enough?” Talbott asks, pointing out that this is especially pertinent when it comes to the new portable/handheld instruments. “Do I have the software control that I need?”
Of course, the usual litany of considerations also applies when deciding on an FT-IR, such as budget, footprint, upgradability, vendor reputation, regulatory compliance and even ergonomics. Many manufacturers offer several levels of instrument. Typically the higher-bracket instruments are more flexible and offer higher performance in terms of resolution and signal-to-noise ratio, but may require greater skill or training to operate.
Image: Schematic of a Michelson interferometer, from Wikipedia.