TOCSY: Whole Spin-System Mapping

Introduction

COSY links a proton only to its direct coupling partners, so identifying a long chain means walking step by step, and one overlapped signal can halt the walk. TOCSY takes a different approach. During a mixing period called isotropic mixing, magnetisation spreads through the entire coupled network, so a single well-resolved proton can reveal every other proton in its spin system along one line of the spectrum. This makes TOCSY especially powerful for sugars, peptides and other molecules built from repeating units.

Core explanation

A TOCSY experiment follows the usual 2D pattern: preparation, evolution for t₁, mixing, detection. The difference lies in the mixing step. Instead of a single pulse, the spectrometer applies a spin-lock , a continuous train of composite radiofrequency pulses such as MLEV-17 or DIPSI-2, lasting typically 20 to 120 ms. During the spin-lock, the large chemical-shift differences between protons are effectively suppressed, because the radiofrequency field keeps refocusing them. What remains is the scalar coupling. Under these conditions all the coupled protons behave as a strongly coupled system, and magnetisation oscillates back and forth between them, a process described by the isotropic mixing Hamiltonian.

Crucially, transfer does not stop at nearest neighbours. Magnetisation that moves from proton A to proton B during the first part of the spin-lock can continue from B to C, then from C to D. The result is a set of relayed correlations : cross-peaks between A and every proton connected to it by an unbroken chain of resolved couplings, even if A and D share no direct coupling. Reading along the row or column of one proton therefore displays the whole spin system at once.

The mixing time controls how far transfer spreads. With a short spin-lock of about 10–20 ms, transfer reaches mainly directly coupled protons, so the spectrum resembles a COSY. With 60–100 ms, magnetisation propagates through five or six bonds' worth of couplings and fills out the whole network. Too long a mixing time loses signal to relaxation in the rotating frame, and it can allow unwanted cross-relaxation (ROE) effects to add or subtract intensity. Because transfer efficiency depends on the size of each coupling, a small coupling anywhere in a chain acts as a bottleneck: protons beyond it may appear only weakly, even with long mixing.

TOCSY spin systems stop exactly where COSY chains stop: at atoms without protons and at heteroatoms that interrupt coupling. This is a strength. In a peptide, each amino acid residue is an isolated spin system, because the carbonyl carbon and the nitrogen of the peptide bond break the proton coupling chain. The amide NH proton of each residue, usually well resolved between 7 and 9 ppm, shows a column of relayed cross-peaks to its own Hα, Hβ and side-chain protons, giving a characteristic fingerprint for each amino-acid type.

Phase-sensitive TOCSY cross-peaks are largely in-phase, unlike the antiphase peaks of DQF-COSY, so they do not cancel when couplings are small relative to linewidth. The trade-off is that TOCSY alone cannot say which proton is directly bonded next to which within a spin system. Combined with COSY, it tells you both membership and order.

A one-dimensional selective version, 1D TOCSY, excites a single resolved proton and returns only its spin system, a quick way to pull one component out of a crowded spectrum.

Step-by-step reasoning

1. Choose a resolved proton, such as an amide NH or an anomeric H-1. 2. Read along its row or column in the TOCSY spectrum. 3. List every cross-peak: these protons belong to the same spin system. 4. Compare with COSY to place them in bonding order. 5. Repeat for other resolved protons until every signal is assigned to a system.

Visual explanation

Imagine a TOCSY plot of a tripeptide. In the amide region at the left, three columns rise vertically. Each column is a ladder of spots at the Hα, Hβ and side-chain shifts of one residue. A valine column shows spots near 4.1, 2.1 and 0.9 ppm; a glycine column shows only Hα spots near 3.9 ppm.

Real-world analogy

COSY is like asking each person whom they are standing next to. TOCSY is like shouting a message into one end of a group linked by hands: the message travels along the chain to everyone who is connected, but stops where the chain of hands breaks. A shorter shout reaches only close neighbours.

Real-world example

Protein chemists use TOCSY together with NOESY to assign solution NMR spectra of small proteins and peptides. Carbohydrate chemists use it to separate the overlapping ring signals of different sugar units in an oligosaccharide, starting from each distinct anomeric proton.

Why?

Why does TOCSY show correlations that COSY misses? The spin-lock suppresses chemical-shift differences, so coupled spins exchange magnetisation continuously during mixing. Magnetisation that has arrived at one proton can pass on again, relaying through several couplings before detection.

Common misconception

"A TOCSY cross-peak means the two protons are directly coupled or adjacent." A TOCSY cross-peak shows only that both protons belong to the same connected network. It may be relayed through several intermediate spins.

Worked example

Question: In a glucose-containing disaccharide, the H-1 of one unit at 4.5 ppm shows TOCSY cross-peaks at 3.3, 3.5, 3.4 and 3.45 ppm with 80 ms mixing, but only a cross-peak at 3.3 ppm with 15 ms mixing. Interpret this.

Reasoning: At short mixing, transfer reaches only the directly coupled H-2, so 3.3 ppm is H-2. Longer mixing relays magnetisation to H-3, H-4 and H-5, all in the same ring spin system.

Answer: 3.3 ppm is H-2; the other peaks are H-3 to H-5 of the same ring, whose order must be fixed with COSY.

Quick check

1. What happens to the number of TOCSY cross-peaks for a proton as the mixing time is increased from 15 ms to 80 ms? Answer: It increases, because magnetisation relays further along the coupled network.

Exam focus

Contrast TOCSY with COSY: TOCSY shows whole spin systems through relayed transfer; COSY shows direct couplings. Explain the role of the spin-lock and mixing time. Remember that spin systems break at protonless atoms and heteroatoms.

Advanced insight

Under isotropic mixing, transfer between two spins follows approximately sin²(πJτ), so each coupling sets its own timescale and complex networks show oscillating, non-monotonic build-up. Spin-lock sequences are designed to minimise ROE transfer, which has opposite sign to TOCSY transfer and can cancel cross-peaks; clean TOCSY variants add delays that balance the two effects.

Summary

TOCSY uses a spin-lock mixing period to make coupled protons exchange magnetisation continuously. Transfer relays along unbroken coupling chains, so one proton reveals its whole spin system. Mixing time sets how far transfer spreads, small couplings act as bottlenecks, and spin systems stop at protonless atoms. TOCSY gives membership; COSY gives order.

Practice questions

1. Why is each amino-acid residue a separate TOCSY spin system in a peptide? Answer: The carbonyl carbon and peptide nitrogen break the chain of proton–proton couplings between residues. 2. What does the spin-lock do to chemical-shift differences during mixing? Answer: It effectively suppresses them, leaving scalar coupling to drive transfer so the protons behave as strongly coupled. 3. A proton at the far end of a chain is missing from a TOCSY row even with 100 ms mixing. Suggest a reason. Answer: A small coupling somewhere along the chain limits transfer, and relaxation removes signal before it can pass through that bottleneck. 4. Why are TOCSY cross-peaks less affected by cancellation than DQF-COSY cross-peaks? Answer: TOCSY cross-peaks are mainly in-phase, whereas DQF-COSY peaks are antiphase and cancel when coupling is smaller than the linewidth.