The purpose of this paper is to compare and contrast the IR and Raman spectra of various functional groups, based on the Lambert Organic Structural Spectroscopy text. The ranges seen in Raman and IR spectra are different and this contributes to their complementary nature. IR reaches a much higher range of frequencies, but in the lower ranges, Raman can be clearer and more understandable (this is the "fingerprint" range in the IR). Raman spectroscopy is possible beyond the 2500 cm-1 we can measure, but I will generally not discuss bands outside of this range. The different theories underlying the techniques also lead to their complementary nature. IR spectroscopy is based on changing dipole moment, while Raman is based on changing polarizability.

Perhaps the most obvious region in the IR spectrum is what Lambert refers to as the CH stretching region. I tend to think of this as the alkane region. There are several sharp peaks just below 3000 cm-1 when an alkyl group is present. Aromaticity is indicated in the IR by small, sharp peaks between 3100 and 3000 cm-1, but these are often lost in the noise of the alkyl activity. Alkene activity is also obscured in this region. When seen very close up, methyl, methylene, and occasionally methine stretches can be distinguished. Because of the differences in range, one cannot see in the Raman the hydrocarbon region around 3000 cm-1 that is seen in the IR. Raman frequencies for alkanes include 14756 to 1450 cm-1, 1350 to 1300 cm-1, and 340 to 230 cm-1 (below the range the IR can reach).

Alkene activity (C=C) is found in the Raman at 1650 cm-1. This is a strong peak; the IR equivalent is weak and sometimes not present. The substitution around the double bond can affect frequency and give important clues as to the kind of substitution present. Also, this band can give important clues to the nature of cyclic compounds. There is also a band between 1450 and 1200 cm-1 (Ch in-plane deformation).

There are several more subtle points to be made concerning alkyl groups. Methyl groups absorb in the IR in a strong band between 1380 and 1360 cm-1; if this band is a doublet it can indicate branching. Isopropyl and t-butyl groups can give characteristic absorptions. Isopropyl groups absorb in the IR as a doublet at 1385 and 1370 cm-1. The t-butyl group appears at 1370 cm-1 and a smaller peak at 1395 cm-1. Also, the methylene group can appear as a strong IR band between 1480 and 1440 cm-1 (this is the bending/scissoring motion) and also at 1350 and 1150 cm-1 (these are wagging and twisting bands). There is also a rocking band near 725 cm-1 found in the IR. Unsaturated compounds have specific bands, which are detailed in Lambert. Vinyl and vinylidene compounds, cyclic alkenes, etc. have their own frequencies.

Alkynes give very distinctive absorptions in both the IR and Raman. The terminal carbon carbon triple bond appears near 2100 cm-1 but can be shifted higher by substitution. This peak is stronger than in the IR and is always present; symmetrical alkynes may not give an IR absorption in this region. A disubstituted alkyne can have a Fermi doublet in the Raman.

Aromaticity is usually indicated by a strong band at 1000 cm-1; there may also be a band between 750 and 610 cm-1. There is also a band around 1440 cm-1 in both spectra. Aromatic compounds can be unclear in the IR. There are various bands: "3100-3000 cm-1 (CH stretching), 2000-1700 cm-1 (overtones and combinations), 1650-1430 cm-1 (C=C stretching), 1275 -1000 cm-1 (in-plane CH deformation), and 900 to 690 cm-1 )out of plane CH deformation)." (Lambert, J. B., et. al., Organic Structural Spectroscopy, 1998.) Raman spectra are often clearer. Most obvious is the ring breathing mode found at 1000 cm-1. This is a strong band found in most aromatic compounds (it is absent in o- and p- disubstituted compounds, such as p-xylene, and also some trisubstitued compounds). There is also a peak around 1600 cm-1 (much clearer than in the IR); monosubstituted compounds have a medium peak between 1030 and 1010 cm-1 and another, weaker one between 625 and 605 cm-1.

There are several expected bands for oxygen compounds. An alcohol gives a very broad IR band around 3600 to 3200 cm-1. This band is even larger for a carboxylic acid, encompassing the hydrocarbon stretch around 3000 cm-1. Bands may be seen for ethers and alcohols between 1300 and 1000 cm-1, however, this is not a reliable stretch. Alcohols can appear at 1450 to 1350 cm-1, 1150 to 1050 cm-1, and 970 to 800 cm-1 in the Raman.

The carbonyl of an aldehyde, ketone, acid, ester, etc. appears around 1710 cm-1 in both spectra. It is much stronger in the IR than in the Raman, owing to the large dipole moment of the group. The IR peak is also much sharper than the Raman. A carbonyl can absorb in the range of 1850 to 1650 cm-1 at the extreme. Some of the compounds that absorb in the far ends of this region include acid chlorides (around 1800 cm-1), aromatic carbonyl compounds (aromatic acids and ketones: 1700 to 1680 cm-1 1) and conjugated ketones (1700-1650 cm-1). Compounds with multiple carbonyls can lead to some interesting absorptions. Aldehydes also give a CH doublet around 2820 and 2710 cm-1. This is referred to as a Fermi doublet. However, the main feature of an aldehyde in the Raman spectrum is the carbonyl around 1710 cm-1. Acids may give a Raman stretch at 1680 to 1640 cm-1 due to the dimer. Esters can give a 1100 to 1025 cm-1 C-O-C stretch.

Nitrogen compounds also appear around 3500-3300 cm-1, but usually have a different shape than the -OH stretch. Amides appear as a doublet in the IR near 1640 cm-1. In the Raman the bands are more separate, appearing at 1650 and 1400 cm-1. Nitriles appear near 2200 cm-1 in both spectra. This line is much stronger in the Raman than in the IR. The IR band is general near 2250 cm-1, while the Raman is closer to 2230 cm-1. Broad bands can indicate amino acids and amine hydrohalides (from 3000 to 2200 cm-1); an amine or amide may give rise to a stretch around 1000 cm-1. Raman spectra of heterocyclic nitrogen compounds are usually stronger than IR.

Halongen substitution can affect frequencies. Carbon-fluorine stretching is strong in IR (1350 to 1100 cm-1), and can hide some other features. Because this stretching is weak in Raman, it can reveal these features.

Raman and IR spectroscopy can serve as valuable complements, especially as Raman technology becomes affordable.