Principles
Within the electromagnetic spectrum ( Figure 1), the energy range below the UV/ vis is the infrared region, encompassing the wavelength range of about 700 nm to 25 μm, and thus reaching from the red end of the visible to the microwave region. The absorption of infrared light by a molecule results in transition to higher levels of vibration ( Figure 2).
Fig1. The electromagnetic spectrum and its uses for spectroscopic methods.
Fig2. Energy diagram for a diatomic molecule exhibiting rotation and vibration, as well as an electronic structure. The distance between two masses m 1 and m 2 (nuclear displacement) is described as a Lennard–Jones potential curve with different equilibrium distances ( R eq ) for each electronic state. Energetically lower states always have lower equilibrium distances. The vibrational levels (horizontal lines) are superimposed on the electronic levels. Rotational levels are superimposed on the vibrational levels and not shown for reasons of clarity.
For the purpose of this discussion, the bonds between atoms can be considered as flexible springs, illustrating the constant vibrational motion within a molecule (Figure 3). Bond vibrations can thus be either stretching or bending (deformation) actions. Theory predicts that a molecule with N atoms will have a total of [3 N −6] fundamental vibrations ([3 N −5], if the molecule is linear): [2 N −5] bending, and [ N −1] stretching modes ( Figure4).
Fig3. Possible stretching vibrations in acetaldehyde.
Fig4. Normal vibrational modes for CO 2 . For symmetric molecules that possess a centre of symmetry, bands that appear in the IR do not appear in the Raman spectrum. A linear molecule with N atoms possesses (3 N – 5) different normal vibrational modes; thus, CO 2 possesses the 3 × 3 – 5 = 4 modes shown.
Infrared and Raman spectroscopy (see below) give similar information about a molecule, but the criteria for the phenomena to occur are different for each type. For asymmetric molecules, incident infrared light will give rise to an absorption band in the infrared spectrum, as well as a peak in the Raman spectrum. However, as shown in Figure 4, symmetric molecules, such as CO2, that possess a centre of symmetry, show a selective behaviour: bands that appear in the infrared spectrum do not appear in the Raman spectrum, and vice versa:
• An infrared spectrum arises from the fact that a molecule absorbs incident light of a certain wavelength which will then be ‘missing’ from the transmitted light. The recorded spectrum will show an absorption band.
• A Raman spectrum arises from the analysis of scattered light. The largest part of an incident light beam passes through the sample (transmission). A small part is scattered isotropically, i.e. uniformly in all directions (Rayleigh scatter), and possesses the same wavelength as the incident beam. The Raman spectrum arises from the fact that a very small proportion of light scattered by the sample will have a different frequency from the incident light. As different vibrational states are excited, energy portions will be missing, thus giving rise to peaks at lower frequencies than the incident light (Stokes lines). Notably, higher frequencies are also observed (anti Stokes lines); these arise from excited molecules returning to ground state. The emit ted energy is dumped onto the incident light, which results in scattered light of higher energy than the incident light.
The criterion for a band to appear in the infrared spectrum is that the transition to the excited state is accompanied by a change in dipole momentum, i.e. a change in charge displacement. Conversely, the criterion for a peak to appear in the Raman spectrum is a change in polarisability (the distortion of the electron distribution in the molecule) of the molecule during the transition.
Infrared Spectroscopy
The fundamental frequencies observed are characteristic of the functional groups concerned, hence the term fingerprint. Figure 5 shows the major bands of an FT IR spectrum of the drug phenacetin. As the number of functional groups increases in more complex molecules, the absorption bands become more difficult to assign. However, groups of certain bands regularly appear near the same wavelength and may be assigned to specific functional groups. Such group frequencies are thus extremely helpful in structural diagnosis. A more detailed analysis of the structure of a molecule is possible, because the wavenumber ν ∼ associated with a particular functional group varies slightly, owing to the influence of the molecular environment. For example, it is possible to distinguish between C–H vibrations in methylene (–CH 2 –) and methyl groups (–CH 3).
Fig5. FT–IR spectrum of phenacetin, historically the first synthetic fever reducer to go on the market. Bands at the appropriate wavenumbers (in cm −1 ) are shown, indicating the bonds with which they are associated, and the type (s, stretching; b, bending).
Raman Spectroscopy
The assignment of peaks in Raman spectra usually requires consideration of peak position, intensity and form, as well as depolarisation . This allows identification of the type of symmetry of individual vibrations , but not the determination of structural elements of a molecule. The depolarisation is calculated as the ratio of two intensities with perpendicular and parallel polarisations with respect to the incident beam. The use of lasers as the light source for Raman spectroscopy easily facilitates the use of linearly polarised light. Practically, the Raman spectrum is measured twice; in the second measurement, the polarisation plane of the incident beam is rotated by 90°.
Instrumentation
The most common source for infrared light is white-glowing zircon oxide or the so-called globar made of silicium carbide with a glowing temperature of 1500 K. The beam of infrared light passes a monochromator and splits into two separate beams: one runs through the sample, the other through a reference made of the substance the sample is prepared in. After passing through a splitter alternating between both beams, they are reflected into the detector. The reference is used to compensate for fluctuations in the source, as well as to cancel possible effects of the solvent. Non covalent materials must be used for sample containment and in the optics, as these materials are transparent to infrared. All materials need to be free of water, because of the strong absorption of the O–H vibration.
Analysis using a Michelson interferometer enables Fourier transform infrared spectroscopy (FT-IR). The entire light emitted from the source is passed through the sample at once, and then split into two beams that are reflected back onto the point of split (interferometer plate). Using a movable mirror, path length differences are generated between both beams, yielding an interferogram that is recorded by the detector. The interferogram is related to a conventional infrared spectrum by a mathematical operation called Fourier transformation. Historically, liquid or suspension samples were delivered to the IR instrument held as layers between NaCl planes or solids were pressed into KBr discs or prepared in thick suspensions (mulls) such as nujol. Most contemporary instruments combine the FT-IR technique with a sample probing based on attenuated total reflection (ATR). ATR employs the features of an evanescent fi eld, the same technology used in surface plasmon resonance. The infrared light enters the internal reflection element (a glass prism similar to the one described for SPR) and probes the sample on the surface of the prism. The fact that the sample is probed multiple times has the advantage of yielding stronger absorbances. ATR-FT-IR has become a very popular method since small amounts of solid and liquid samples can be measured conveniently without lengthy preparation.
For Raman spectroscopy, aqueous solutions are frequently used, since water possesses a rather featureless weak Raman spectrum. The Raman effect can principally be observed with bright, monochromatic light of any wavelength; however, light in the visible region of the spectrum is normally used due to few unwanted absorption effects. The ideal light source for Raman spectrometers is therefore a laser. Because the Raman effect is observed in light scattered off the sample, typical spectrometers use a 90° configuration.
Applications
The use of infrared and Raman spectroscopy has typically been in chemical and bio chemical research of small-molecule compounds such as drugs, metabolic intermediates and substrates. Examples are the identification of synthesised compounds, or identification of sample constituents (e.g. in food) when coupled to a separating method such as gas chromatography (GC-IR).
However, FT–IR is also used for analysis of peptides and proteins. The peptide bond gives rise to nine characteristic bands, named amide A, B, I, II, III, …, VII. The amide I (1600–1700 cm −1) and amide II (1500–1600 cm −1) bands are the major contributors to the protein infrared spectrum. Both bands are directly related to the backbone conformation and have thus been used for assessment of the secondary structure of peptides and proteins. The interpretation of spectra of molecules with a large number of atoms usually involves deconvolution of individual bands and second-derivative spectra.
Time-resolved FT-IR enables the observation of protein reactions at the submilli second timescale. One of the first applications of this technique was the investigation of the light-driven proton pump bacteriorhodopsin. For instance, the catalytic steps in the proton-pumping mechanism have been validated with time-resolved FT–IR, and involve transfer of a proton from the Schiff base (R1R2 C=N–R3) to a catalytic aspartate residue, followed by reprotonation of a second catalytic aspartate residue.
