Methane $\ce{CH4}$ is symmetrical and non polar, and so is oxygen $\ce{O2}$. But $\ce{O2}$ doesn't absorb IR. Why is that? People say that if a molecule can change its polarity then it will absorb IR, but I don't understand how a molecule changes its polarity while absorbing IR.

Please try to keep your answer simple — I don't have a strong background in infrared spectroscopy.


In basic terms, for a molecule to absorb radiation there has to be an oscillating dipole being produced. This can occur by nuclear motion (vibrations, rotations) or electronic motion to produce electronically excited states. You ask about IR radiation, which does not have sufficient energy to produce electronic excitation, so this is ignored in the text below.

The restriction on producing a moving dipole exists because radiation consists of oscillating electric and magnetic fields and to allow energy to go from the radiation into the molecule, there has to be an interaction between radiation and the molecule and the origin of this is the oscillating molecule's dipole. (Of course the radiation also has to have the correct energy to bridge the gap between any two energy levels, but this is just another way of saying that the molecular dipole and the radiation oscillate at the same frequency, i.e. are in resonance.)

Thus in any homonuclear diatomic there is no IR absorption because these molecules have no intrinsic dipole or one caused by vibrations. Also as they have no permanent dipole there is obviously no dipole generated when molecules rotate. Heteronuclear diatomics, HCl for example, all have a permanent dipole and one that is changed by vibrational motion, thus there is a vibrational and rotational spectrum in all heteronuclear diatomics.

All homo-nuclear diatomics, $\ce{N2}$, $\ce{O2}$ for example, are transparent to IR radiation and this is of prime importance in global warming, since if these molecules obtain energy, say by collision with water or carbon dioxide that has been excited by IR radiation, they cannot radiate the energy away and so must heat the atmosphere by collisions with other molecules.

In a molecule such as $\ce{CO2}$ although it has no overall dipole it does have moving (oscillating) dipoles due to the way it vibrates. One mode is to stretch one bond at the same time as the other contacts (an asymmetric stretch), and so an oscillating dipole is formed and this interacts with radiation and $\ce{CO2}$ has an IR vibrational-rotational spectrum. The bending vibration is also IR active.

A similar effect occurs in methane, the way the molecule vibrates develops oscillating dipoles and these interact with the radiation and so absorb energy. Molecular group theory is usually used to sort out which of the $3N-6$ ways molecules can vibrate can have IR spectra.

There are complicated higher order effect even in homonuclear diatomics which may generate absorption but these effect are absolutely tiny and can be ignored for most every purpose. The relative magnitudes of electric dipole, quadrupole and magnetic dipole intensities are $I_\mathrm{dip}:I_\mathrm{quad}:I_\mathrm{mag} \approx 1:10^{-5}:10^{-5}.$

We can measure the vibrational frequency of homonuclear as well as heteronuclear diatomics by using Raman spectroscopy, which is a scattering effect (no photons are absorbed), and depends on the shape of the 'electron cloud' in the molecule. This is more properly called its polarisability.

  • $\begingroup$ Could you please clarify some physics - why to resonate the frequencies of the dipole and radiation must be the same? They can't resonate if the frequencies differ a bit? $\endgroup$ – Stanislav Bashkyrtsev May 1 '17 at 8:42
  • $\begingroup$ This is normally explained as the Bohr condition; energy of photon matches transition energy gap. As $E=h\nu$ you can think of this also as frequency match. If the energy is not exact the chance of absorption rapidly diminishes in accordance with the line shape of the transition. A typically homogenous linewidth (full width half max) is $ 10^{-3}\pu{ cm^{-1}}$ but can be wider due to Doppler effect, typically $\approx 0.1 \pu { cm^{-1}}$ still very narrow if the transition frequency is $1500 \pu{ cm^{-1}}$ $\endgroup$ – porphyrin May 1 '17 at 10:19

Infrared radiation is absorbed by a molecule when the frequency of the radiation matches that of one of the vibrational modes of the molecule. It is also necessary that the molecule have a vibrational mode for which there is a change in dipole moment. If the energy of a photon does not meet both these criteria, then it will be transmitted rather than absorbed.

According to this Wikipedia article:

A molecule can vibrate in many ways, and each way is called a vibrational mode. For molecules with $N$ number of atoms, linear molecules have $3N - 5$ degrees of vibrational modes, whereas nonlinear molecules have $3N - 6$ degrees of vibrational modes (also called vibrational degrees of freedom). As an example $\ce{H2O},$ a non-linear molecule, will have $3 × 3 - 6 = 3$ degrees of vibrational freedom, or modes.
Simple diatomic molecules have only one bond and only one vibrational band. If the molecule is symmetrical, e.g. $\ce{N2},$ the band is not observed in the IR spectrum.

So $\ce{N2}$ (or $\ce{O2}$) only has a single vibrational mode $(3 × 2 - 5 = 1),$ no permanent dipole, and can only have a weak induced dipole, so it's absorption in the IR is very small. $\ce{CO}$ also has only a single vibrational mode, but there is a permanent dipole moment and therefore the molecule has the ability to absorb in the IR spectrum.

I've tried to balance "keeping it simple" with giving a meaningful explanation. Don't hesitate to ask for clarifications in the comments.


Probably all molecules with two or more atoms find a way to absorb at least some IR radiation. A molecule with only nonpolar bonds, such as a nitrogen molecule, may be nonpolar if the electrons are perfectly centered in their orbitals, but if the electrons move there could be a weak temporary dipole.

Still that is much weaker absorption than what you would get with a permanent dipole like water or the individual bonds in carbon dioxide. Lucky us, otherwise an atmosphere that is 78% nitrogen would launch a huge greenhouse effect.


Very simple answer: IR radiation (in very first approximation) interacts with the individual bond. While methane is unpolar, the bonds in it are polar.

Not terribly polar, but still.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.