Infrared and Raman spectroscopy are very useful tools for measuring zone-center phonon frequencies. Both these techniques are based on the interaction of electromagnetic waves with the lattice waves or phonons. But these techniques are also suitable to probe the internal motions of the molecules as well. There are selection rules for Raman scattering and infrared absorption and they are complimentary and therefore both should be employed to study the dynamics of molecules and solids. These techniques are however not unable to measure dispersions of phonons. Neutron spectroscopy should be used for complete investigation of the dynamics of molecules and solids.

There are excellent text books and review articles for both Raman and infrared spectroscopy. For Raman sectroscopy you can consult series "Light Scattering in Solids" edited by M. Cardona and G. Guentherrodt, Springer Verlag. There are several text book on Infrared spectroscopy. You an for example consult the book by C.N.R. Rao. There are nice review articles on infrared spectroscopy in Handbuch der Physik.

Both interact with phonons. In IR, the infrared photons are absorbed, and the energy of the absorbed IR photon is related to the phonon in the material. Raman is more complex: here, an optical photon is absorbed, exciting a carrier to an excited state. Most excited carriers decay to the ground state directly, but some switch to a different "virtual" energy level by absorption or emission of a phonon before decaying to the ground state. Since they decay to the ground state from the different level, the photon they emit is of shorter or longer wavelength than the incident photon. This energy difference is measured in Raman and related to the phonon that must have been absorbed to produce the shift.

Dear Manuel, you explain: "Raman is more complex: here, an optical photon is absorbed, exciting a carrier to an excited state. Most excited carriers decay to..." I'm very sorry, but that's rather (please, no offense) "vulgar" interpretation of Raman effect. Perhaps it is more appropriate for description of fluorescence. Contrary to IR absorption, in Raman the energy of exciting photon (it may be from UV, VIS or NIR range) usually does not correspond to any electronic transition energy of the system under study, and the photon of energy hω is not absorbed in the strict spectroscopic sense. The role of the incident photons in Raman is to perturb the molecule or the lattice and make feasible spectroscopic transitions other than direct absorption. Once again, Raman is a light-scattering phenomenon. All light-scattering processes are characterized by the fact that, unlike direct absorption processes (e.g., selective resonant absorption of IR photons in IR absorption spectroscopy), the energy of an incident photon is not required to be equal to the energy corresponding to the difference between two discrete energy levels of the material system. It is experimentally observed however, that as the energy of exciting (laser) photons gets closer to energy of a transition from the ground electronic state to an excited electronic state of the material system, the intensity of the scattering is enhanced. Such enhanced scattering is called resonance scattering. But, strictly speaking, the characteristic properties of resonance Raman scattering markedly differ from those of normal scattering.

Turning back to Sridhar question, I'd resume my answer as follows: there are similarities between IR and Raman, but the differencies between the two may be even more interesting, informative and important. As an example, please compare the uploaded IR and NIR-FT Raman spectra of the same compound - single crystalline polydiacetylene, poly(PTS). IR absorption spectrum represents vibrations of the polymer side substituents only, while the Raman one is actually a pure phonon spectrum of the conjugated backbone.

Dear Alexander Shchegolikhin,

Could you explain me "why"?

When the frequency of the radiation matches the frequency of the vibrating molecule the IR radiation is absorbed.

consider if the absorbed frequency (resonance frequency or wave number) is 3000 cm-1 which corresponds to the energy E1.

Such molecules are not absorbing other frequency or wave number ( higher than the resonance frequency or wave number) say 4000 cm-1 corresponds to the energy E2.

Here E2 is greater than the E1. I dont know why E 2 is not absorbed by such molecules even though E2 is higher than E1.

Dear Sridhar, since i'm not sure i fully understood your last question, i'd better just remind a few classical points: (i) matter does not interact with energy in a continuous form; (ii) accoring to Einstein, Planck, Bohr et. al., electromagnetic radiation can be regarded as a stream of particles (or quanta) for which the energy, E, is given by the Bohr equation: E = hν, where h is the Planck constant and ν is equivalent to the classical frequency; (iii) processes of change, including those of vibrations in IR spectroscopy, can be represented in terms of quantized discrete energy levels E0, E1, E2, etc.; (iv) each atom or molecule in a system must exist in one or other of these levels; (v) in a large assembly of molecules, there will be a distribution of all atoms or molecules among these different energy levels; (vi) the latter are a function of an integer (the quantum number); (vii) whenever a molecule interacts with radiation, a quantum of energy (or photon) is either emitted or absorbed; (viii) in every case, the energy of the quantum of radiation must exactly fit the energy gap E1 − E0 or E2 − E1, etc. And it doesn't matter whether E2>E1 or E1>E2.

What is really matters is that, in order to show IR absorptions, a molecule (a chemical bond between atoms) must possess a specific feature, i.e. an electric dipole moment of the molecule (the bond) must change during the vibration. This is the so called selection rule for IR spectroscopy. At the same time, an example of an ‘infrared-inactive’ molecule is a homonuclear diatomic molecule – just because its dipole moment remains zero no matter how long or short the bond is. However, such molecule should be Raman-active. In principle, an understanding of molecular symmetry, group theory and all this stuff is important when initially assigning infrared bands. Detailed description of the IR and Raman theory may be found in many fascinating text-books including those named by Tapan Chatterji above. Fortunately, it is not necessary to work from first principles each time a new infrared spectrum is obtained. Good Luck!

Thank you very much sir

One more thing also very important. IR is absorption, it's a sum of imformation gethered along the path. On the other hand, Raman is scattering, it takes information from a "point" in the space. Sometime, this difference is crucial.

What is complementary and why Raman is complementary to IR?

Will both measurement give complete information?