Sign in to save

Bookmark this page so you can find it later.

Sign in to save

Bookmark this page so you can find it later.

This cheat sheet covers the three most common 2D NMR experiments used in organic structure determination: COSY, HSQC, and HMBC. Students need it because each spectrum answers a different connectivity question, and confusing them can lead to incorrect structures. The reference emphasizes what each experiment correlates, how to read the axes, and how to combine the data efficiently.

Key Facts

  • COSY shows through-bond proton-proton correlations, most often between hydrogens separated by 3JHH^{3}J_{HH} coupling.
  • HSQC shows direct one-bond carbon-proton correlations, usually 1JCH^{1}J_{CH}, so each cross peak links a proton to the carbon it is attached to.
  • HMBC shows longer-range carbon-proton correlations, usually 2JCH^{2}J_{CH} and 3JCH^{3}J_{CH}, so it helps connect fragments through quaternary carbons, carbonyls, and heteroatom-substituted centers.
  • In a typical 1H^1\mathrm{H}-1H^1\mathrm{H} COSY spectrum, both axes are proton chemical shifts, so a cross peak at (δH=2.10,δH=4.25)(\delta_H = 2.10, \delta_H = 4.25) indicates coupling between those two proton signals.
  • In HSQC and HMBC spectra, the proton dimension is usually F2F_2 and the carbon dimension is usually F1F_1, so a point is read as (δH,δC)(\delta_H, \delta_C).
  • A chemical shift difference is reported as Δδ=δAδB\Delta\delta = \delta_A - \delta_B, and values are measured in ppm\mathrm{ppm} rather than Hz\mathrm{Hz}.
  • The frequency separation between two signals is Δν=Δδ×ν0\Delta\nu = \Delta\delta \times \nu_0, where ν0\nu_0 is the spectrometer frequency in MHz\mathrm{MHz} and Δν\Delta\nu is in Hz\mathrm{Hz}.
  • A strong structure assignment usually uses COSY for proton spin systems, HSQC for attached CH\mathrm{C-H} pairs, and HMBC for connecting those fragments into the carbon skeleton.

Vocabulary

COSY
Correlation spectroscopy is a 2D NMR experiment that shows which protons are scalar-coupled to each other through bonds.
HSQC
Heteronuclear single quantum coherence is a 2D NMR experiment that correlates each proton with its directly attached carbon, usually through 1JCH^{1}J_{CH}.
HMBC
Heteronuclear multiple bond correlation is a 2D NMR experiment that correlates protons with carbons two or three bonds away, usually through 2JCH^{2}J_{CH} or 3JCH^{3}J_{CH}.
Cross peak
A cross peak is an off-diagonal signal in a 2D NMR spectrum that indicates a correlation between two nuclei.
Chemical shift
Chemical shift, written δ\delta, is the position of an NMR signal relative to a reference standard and is measured in ppm\mathrm{ppm}.
Scalar coupling
Scalar coupling, written JJ, is through-bond magnetic interaction between nuclei and is measured in Hz\mathrm{Hz}.

Common Mistakes to Avoid

  • Using HMBC as proof of direct attachment is wrong because HMBC usually shows 2JCH^{2}J_{CH} and 3JCH^{3}J_{CH} correlations, not one-bond 1JCH^{1}J_{CH} correlations.
  • Ignoring the axis labels is wrong because COSY uses two proton dimensions, while HSQC and HMBC usually pair δH\delta_H with δC\delta_C.
  • Treating every missing correlation as structural absence is wrong because weak coupling, overlap, relaxation, or experiment settings can hide expected cross peaks.
  • Assigning quaternary carbons from HSQC alone is wrong because quaternary carbons have no directly attached hydrogens and usually require HMBC or 13C^{13}\mathrm{C} data.
  • Confusing diagonal peaks with COSY cross peaks is wrong because diagonal peaks are self-correlations, while off-diagonal peaks reveal proton-proton connectivity.

Practice Questions

  1. 1 A COSY spectrum shows a cross peak between δH=1.25 ppm\delta_H = 1.25\ \mathrm{ppm} and δH=3.65 ppm\delta_H = 3.65\ \mathrm{ppm}. What does this suggest about the two proton signals?
  2. 2 An HSQC cross peak occurs at (δH=4.10 ppm,δC=62.0 ppm)(\delta_H = 4.10\ \mathrm{ppm}, \delta_C = 62.0\ \mathrm{ppm}). What carbon-proton relationship does this indicate?
  3. 3 On a 400 MHz400\ \mathrm{MHz} spectrometer, two proton signals differ by 0.035 ppm0.035\ \mathrm{ppm}. Calculate their separation in Hz\mathrm{Hz} using Δν=Δδ×ν0\Delta\nu = \Delta\delta \times \nu_0.
  4. 4 A molecule has two proton spin systems from COSY, but no COSY cross peak between them. Explain how HMBC could still connect the fragments into one structure.

Understanding 2D NMR (COSY, HSQC, HMBC) Reference

A two dimensional NMR spectrum is best treated as a map of evidence, not as a finished molecular structure. Each spot, called a cross peak, must agree with the one dimensional proton and carbon spectra. Start by making a signal list.

Record each proton shift, its integration, and its splitting pattern. Then use an HSQC spectrum to give every proton signal a carbon partner. This immediately separates methyl, methylene, and methine sites when combined with integration and carbon data.

Carbon signals with no HSQC peak are especially important. They are often quaternary carbons, carbonyl carbons, or carbons whose attached hydrogen is replaced by another atom.

COSY is useful for building local proton networks. A chain of neighboring proton signals can reveal part of an alkyl group, a ring, or a substituted aromatic system. The diagonal contains each proton correlated with itself, so the useful information lies away from that diagonal.

COSY peaks normally appear in matching pairs across the diagonal. Weak or missing peaks do not always mean that two hydrogens are unrelated. Overlapping signals, very small couplings, exchangeable hydrogens, and poor signal strength can hide a real correlation.

Students should compare COSY connections with the splitting patterns in the proton spectrum. A proposed neighboring relationship should make sense in both experiments.

HMBC supplies longer connections, but it requires careful judgment. A proton can show an HMBC peak to more than one carbon nearby in the structure. This is helpful for locating a carbonyl carbon or joining two fragments, yet it can create several possible structures.

Count bonds along a proposed path from the proton to the carbon. Two-bond and three-bond paths are the most dependable.

Four-bond peaks sometimes occur, especially in conjugated or aromatic systems, but they should be treated as supporting evidence rather than proof. A missing HMBC correlation is not proof that a pathway is absent because coupling size and experimental settings affect visibility.

Chemical shift gives chemical context before any correlations are assigned. Protons near oxygen, nitrogen, halogens, double bonds, or aromatic rings often appear farther downfield than simple alkyl protons. Carbon shifts provide similar clues.

A carbonyl carbon, an aromatic carbon, and a saturated carbon occupy very different regions of a carbon spectrum. These ranges are guides, not fixed rules. Solvent, concentration, hydrogen bonding, charge, and molecular shape can move signals.

The same shift difference in parts per million represents a larger separation in hertz on a higher field instrument. This matters when comparing coupling patterns or deciding whether two close signals can be resolved.

A reliable assignment is built by testing consistency. Begin with clear proton groups and their HSQC partners. Extend each group through COSY.

Use HMBC to connect those groups through carbons that lack attached hydrogens. Finally, check the molecular formula, degree of unsaturation, integrations, and expected functional groups. Do not force every small peak into the first structure that seems plausible.

Solvent peaks, impurities, folded signals, overlap, and noise can mislead even experienced users. The strongest conclusion is one supported independently by several observations.