The COSY Multiple-Quantum Filtered (COSY-MQF) experiment is an alternative version of the COSY experiment, in which a multiple-quantum filter is inserted to allow the detection of signals from all coupled spin systems but suppresses signals arising of lower coherence levels. Thus, a COSY with a double-quantum filter (2D COSY-DQF experiment) experiment efficiently suppress single-quantum coherences from singlet uncoupled signals as, for instance, those of methyl groups or solvents.

Easy implementation on any AVANCE spectrometer.

The pulse sequence of the COSY-MQF experiment is derived from the conventional COSY pulse sequence by incorporating a MQF after the variable evolution period ( 82JACS6800 , 83BRC479 , and 86JACS6482 ). Thus, the COSY-MQF pulse sequence is a three 90º pulse sequence, with the first two separated by the variable evolution period and the last two separated by a very short delay (3 microseconds) (see COSY-MQF block) . Specific phase cycling is used to select the particular coherence level of interest.

The advantages of the COSY-DQF experiment over the conventional 2D COSY experiment is that diagonal peaks show absorptive antiphase lineshapes and they are attenuated from uncoupled resonances. On the other hand, sensitivity is lower and long phase cycle are required. Because COSY-DQF removes problems associated with dispersive tails from the diagonal resonances, the experiment is suitable to extract coupling constant information. The COSY-DQF experiment can be performed in magnitude or phase-sensitive modes by selecting the appropriate phase programs and transform algorithm. However, phase-sensitive data is usually recommended.

Higher-orders multiple-quantum coherences can be selected by appropiate phase cycling. Gradient-enhanced versions have been also proposed (see ge-2D COSY-MQF experiment).

An E.COSY (Exclusive COSY) experiment can be thought of as the result of combining double-, triple and sometimes quadruple-filtered COSY spectra. Combining these spectra causes cross-peak components not arising from coherence-transfer processes between connected transitions to cancel out ( 85JACS6394 , 86JCP6837 , 87JMR474-75 and 93JMRA106-104 ). This experiment has been largely used to extract accurate of very small coupling constant values in overcrowded regions. The same procedure has been extensively expanded to multidimensional heteronuclear experiments to measure homonuclear ans heteronuclear coupling constants in labeled biomolecules.

The COSY-DQF experiment is usually recorded in routine/automation modes. Only 90º 1H pulses and a short fixed delay must be specified. In spectrometers equipped with gradient technology, gradient-based COSY versions are highly recommended.
 
Details on practical implementation of several variants of the 2D COSY-DQF experiment on AVANCE spectrometers can be found in the corresponding tutorials:

Tutorials: 2D homonuclear experiments

Tutorials: 2D gradient-based homonuclear experiments

Tutorials: 2D gradient-based homonuclear experiments in H₂O

A COSY-MQF spectrum correlates chemical shifts of the same nucleus in both dimensions. Both cross-peaks and diagonal multiplets have the same anti-phase absorption character. However, a COSY-DQF experiment generates only half as much signal intensity and requires four times as many transients to achieve the same signal-to-noise as a normal 2D COSY experiment. In general, the basic selection rules for a pQF are: (i) the active spins gives a diagonal multiplet when it is coupled to at least p-1 nonequivalent spins; (ii) a cross-peak between two active spin appears when both are coupled to the same set of at least p-2 passive spins.

The theoretical background of the COSY experiment has been extensively described and discussed in many books and review articles. Extensive descriptions can be found in 88BOOK1 (pages 58-94) and 94BOOK1 (pages 295-315).

Related experiments:

2D homonuclear experiments

2D gradient-based homonuclear experiments