Questions: NMR Spectroscopy for Structure Elucidation

Question 1

A researcher runs a COSY experiment on an unknown organic compound and observes a cross-peak between a signal at 3.5 ppm and a signal at 1.2 ppm. What does this cross-peak indicate?

• A. The protons at 3.5 ppm and 1.2 ppm are bonded to the same carbon
• B. The protons at 3.5 ppm and 1.2 ppm are within 5 Å of each other in three-dimensional space
• C. The protons at 3.5 ppm and 1.2 ppm are on adjacent carbons and show J-coupling through bonds
• D. The carbons bearing these protons are directly bonded to each other with no intervening atoms

Answer: (C)

COSY (Correlation SpectroscopY) reveals H–H correlations through J-coupling, which propagates most efficiently over two or three bonds (geminal or vicinal protons). A cross-peak means the two proton signals are scalar-coupled through bonds — typically on adjacent carbons. HSQC, not COSY, establishes one-bond C–H connections. NOESY, not COSY, reveals through-space proximity.

Question 2

A proton signal in ¹H NMR appears as a singlet. This means the proton has no neighboring hydrogen atoms on adjacent carbons.

• T. True
• F. False

Answer: False

A singlet means no J-coupling splitting is observed, not that no neighbors exist. This can happen when neighboring protons are symmetrically equivalent (e.g., the six protons of benzene all couple to each other but, being equivalent, produce a singlet). It can also occur when coupling constants are too small to resolve, or when neighboring hydrogens are on heteroatoms (OH, NH) that exchange rapidly. Always consider molecular symmetry before concluding a singlet means isolation.

Question 3

Why is ¹³C NMR generally non-quantitative in standard experiments, and what would a researcher need to do to obtain quantitative ¹³C data?

Model answer: Standard ¹³C NMR uses proton decoupling, which creates NOE enhancements that differ in magnitude for each carbon, making peak intensities non-uniform. Additionally, different carbons have very different relaxation times (T1), so if pulses are applied too rapidly, quaternary carbons with long T1 values are underrepresented. To get quantitative data, the researcher must use inverse-gated decoupling (to eliminate NOE) and long relaxation delays between pulses (5×T1 of the slowest-relaxing carbon), or use specific qNMR pulse sequences.

This is a major practical limitation. ¹H NMR integration is reliable because the NOE and relaxation effects are relatively uniform across protons. For ¹³C, the variability is large enough that you cannot compare peak areas to count equivalent carbons the way you can with ¹H. Many chemists mistakenly assume peak count = carbon count, which is only valid if no two carbons are equivalent — and even then, peak heights vary.