12.7 Interpreting Infrared Spectra

Interpreting Infrared Spectra

Infrared (IR) spectroscopy identifies functional groups by measuring which wavelengths of IR light a molecule absorbs. Each functional group vibrates at characteristic frequencies, producing a unique spectral fingerprint. By reading an IR spectrum, you can figure out what functional groups are present, distinguish between similar compounds, and confirm the structure of a synthetic product.

Principles of IR Spectroscopy

When a molecule absorbs IR radiation, the energy causes its bonds to vibrate (stretching, bending, etc.). Stronger bonds and lighter atoms vibrate at higher frequencies, which is why an O-H stretch appears at a much higher wavenumber than a C-O stretch.

• Wavenumber (cm$^{-1}$) is the x-axis of an IR spectrum. Higher wavenumbers correspond to higher-energy vibrations.
• Transmittance (%) is the y-axis. Absorption peaks point downward on most IR spectra, so a "strong" peak means a deep dip in transmittance.
• Fourier transform IR (FTIR) is the standard modern technique. It collects all frequencies simultaneously, making measurements faster and more sensitive than older dispersive instruments.
• The Beer-Lambert law relates absorbance to concentration and path length, but for routine functional group identification you're mostly focused on peak position and shape, not precise intensity.

A useful rule of thumb: the region above ~1500 cm$^{-1}$ is where most diagnostic functional group peaks appear. The region below ~1500 cm$^{-1}$ is called the fingerprint region because it contains many overlapping peaks unique to each molecule, making it harder to assign individual vibrations but useful for matching against reference spectra.

Functional Groups in IR Spectra

The table below summarizes the major absorptions you need to know. After the table, each group is discussed in more detail.

The O-H stretch is one of the easiest peaks to spot because of its breadth. Alcohols and phenols absorb around 3200–3600 cm$^{-1}$ as a broad band caused by hydrogen bonding (e.g., ethanol, phenol). Carboxylic acids show an even broader O-H absorption from roughly 2500–3300 cm$^{-1}$ that often swallows the C-H peaks underneath. If you see a huge, rounded absorption spanning that entire region, think carboxylic acid.

N-H stretching vibrations

Amines and amides absorb between 3300–3500 cm$^{-1}$. A primary amine (like methylamine) or primary amide (like acetamide) typically shows two peaks in this region, while a secondary amine or amide shows one. This "two vs. one" pattern is a quick way to distinguish them.

C-H stretching vibrations

The 3000 cm$^{-1}$ line is a critical divider:

• Below 3000 (2850–3000 cm$^{-1}$): sp$^3$ C-H stretches, found in alkanes like hexane and cyclohexane.
• Above 3000 (3010–3100 cm$^{-1}$): sp$^2$ C-H stretches, found in alkenes (ethene, 1-butene) and aromatics (benzene, toluene).
• Around 3300 cm$^{-1}$: sp C-H stretch of terminal alkynes (ethyne, 1-butyne). This peak is sharp and narrow, unlike the broad O-H peak in the same region.

C=O stretching vibrations

The carbonyl stretch is usually the strongest, most obvious peak in an IR spectrum. Its exact position tells you which type of carbonyl you have:

• Esters: 1735–1750 cm$^{-1}$ (ethyl acetate, methyl benzoate)
• Aldehydes: 1720–1740 cm$^{-1}$ (acetaldehyde, propanal)
• Ketones: 1705–1725 cm$^{-1}$ (acetone, cyclohexanone)
• Carboxylic acids: 1700–1730 cm$^{-1}$ (formic acid, propionic acid)
• Amides: 1640–1690 cm$^{-1}$ (formamide, acetamide)

Notice the trend: amides absorb at the lowest frequency because nitrogen donates electron density into the carbonyl, giving it more single-bond character and lowering the stretching frequency.

C=C and C≡C stretching vibrations

Alkene C=C stretches appear at 1620–1680 cm$^{-1}$ (1-pentene, cyclopentene). Aromatic ring C=C stretches show up as two or three bands between 1450–1600 cm$^{-1}$ (naphthalene, toluene). Alkyne C≡C stretches fall at 2100–2260 cm$^{-1}$ (1-hexyne, diphenylacetylene). Keep in mind that a symmetric internal alkyne may show no C≡C peak at all because the vibration doesn't change the dipole moment.

C-O stretching vibrations

Alcohols, ethers, and esters all show C-O stretches between 1050–1300 cm$^{-1}$. This peak alone isn't very diagnostic, but combined with other evidence (like an O-H or C=O peak), it helps confirm the functional group.

Comparison of Similar Compounds

One of the trickiest parts of IR interpretation is telling apart compounds that look similar. Here are the key distinctions:

Alcohols vs. Phenols

Both show a broad O-H stretch, but phenols also display aromatic C=C stretching bands around 1450–1600 cm$^{-1}$ and aromatic C-H stretches just above 3000 cm$^{-1}$. If you see a broad O-H plus aromatic ring peaks, you're looking at a phenol rather than a simple alcohol.

Aldehydes vs. Ketones

Both have a strong C=O stretch, but aldehydes have a telltale pair of C-H stretching peaks around 2700–2850 cm$^{-1}$ (sometimes called "Fermi resonance doublet"). Ketones lack these peaks entirely. If you see a carbonyl and no peaks in the 2700–2850 cm$^{-1}$ region, it's likely a ketone.

Aldehyde C=O also tends to absorb at a slightly higher frequency (~1720–1740 cm$^{-1}$) than ketone C=O (~1705–1725 cm$^{-1}$).

Primary vs. Secondary vs. Tertiary Alcohols

All three classes show a broad O-H stretch, but they differ in the C-O stretching region:

1. Primary alcohols have a strong C-O stretch around 1050 cm$^{-1}$ (1-butanol, 1-hexanol).
2. Secondary alcohols show a C-O stretch around 1100 cm$^{-1}$ (2-propanol, 2-pentanol).
3. Tertiary alcohols display a C-O stretch around 1150 cm$^{-1}$ (2-methyl-2-propanol, 2-methyl-2-butanol).

The shift to higher wavenumber from primary to tertiary is subtle, so in practice this distinction is less reliable than other IR comparisons.

Structural Deduction from IR Data

When you sit down with an IR spectrum, work through it systematically:

1. Scan the O-H / N-H region (3200–3600 cm$^{-1}$). A broad absorption here means an alcohol, phenol, or carboxylic acid. A medium peak (or pair of peaks) suggests an amine or amide. No peak means none of these groups are present.

2. Check the C-H region around 3000 cm$^{-1}$. Peaks below 3000 indicate sp$^3$ C-H bonds. Peaks above 3000 suggest sp$^2$ (alkene or aromatic) or sp (alkyne at ~3300) C-H bonds.

3. Look for a carbonyl peak (1640–1760 cm$^{-1}$). If present, narrow down the type using the exact position and the presence or absence of other peaks (O-H for carboxylic acid, aldehyde C-H doublet for aldehyde, N-H for amide).

4. Check the triple bond region (2100–2260 cm$^{-1}$). A peak here points to an alkyne (or possibly a nitrile, C≡N, around 2200–2260 cm$^{-1}$).

5. Examine the fingerprint region (below 1500 cm$^{-1}$). Look for C-O stretches and use this region to compare against known reference spectra if you need to confirm identity.

Beyond this checklist, a few additional patterns are worth knowing:

• Conjugation lowers stretching frequencies. A conjugated carbonyl (like in benzoic acid) absorbs at a lower wavenumber than an isolated one (like in acetic acid) because conjugation delocalizes electron density and weakens the C=O bond slightly. The same applies to conjugated C=C bonds.
• Hydrogen bonding broadens and shifts peaks. The broad O-H stretch in ethanol is a direct result of intermolecular hydrogen bonding. In the gas phase (no H-bonding), the same O-H stretch would be sharp and at a higher frequency.
• Symmetry can make vibrations IR-inactive. A vibration is only IR-active if it changes the molecule's dipole moment. For example, the symmetric stretch of $CO_2$ doesn't appear in the IR spectrum because the dipole moment doesn't change, while the asymmetric stretch does appear.