NOAH supersequences

Much of the time taken to acquire multi-dimensional NMR spectra is consumed by the relaxation delay. NOAH (NMR by Ordered Acquisition using ¹H-detection) is a method developed to interleave several experiments and use a single relaxation delay for the whole group, thereby reducing the time taken to acquire a series of experiments.

Most NMR experiments consist of (i) a relaxation delay, where the sample is allowed to relax back to magnetic equilibrium; (ii) a train of pulses and delays, to manipulate the magnetisation; and (iii) the acquisition time, in which the data is recorded. Typically, the relaxation delay is 1000 to 2000 ms, the pulse sequence around 100 ms, and the acquisition time between 50 and 300 ms. Clearly, most of the time spent recording NMR data is actually used waiting for the magnetisation to return to equilibrium. The ASAP technique addresses this problem by mixing manipulated and unmanipulated magnetisation to allow the use of short relaxation delays. The NOAH supersequences take a different approach, using a single relaxation delay before the pulse sequence and acquisition period of several different experiments.

The top line of the figure below shows HMBC, HSQC and COSY experiments, along with the approximate time taken for the relaxation delay, pulse sequence, and acquisition period in each. The bottom line of the figure shows the NOAH3-BSC experiment which combines all three spectra, but uses only one relaxation delay at the start of the supersequence. Summing the times beneath the sequences gives the time taken to acquire a a single scan of a single row. For the individual experiments the sum is 3240 ms, while for the NOAH supersequence the total is 1240 ms. In this example, NOAH allows the same data to be acquired in roughly a third of the time taken for the standard implementation.

The order used to combine experiments into a NOAH supersequence is important. Modules that use the least magnetisation, such as the HMBC, are placed near the start, with the homonuclear experiments near the end. It is also important to leave unused magnetisation along the +z axis wherever possible. In the example above, the HMBC module uses only the magnetisation of ¹H nuclei bound to ¹³C, or 1.1% of the total. During the HMBC acquisition period some relaxation takes place, allowing the more sensitive HSQC module to detect signals. Finally, the COSY module detects the untouched 98.9% of ¹H nuclei, as well as those that have relaxed sufficiently during the HSQC acquisition period. Where a COSY and NOESY module are included it is possible to incorporate the COSY module within the NOESY mixing period for even more time savings. For processing, the data is split into separate raw data files then processed normally.

Numerous NOAH supersequences have been developed. The table below lists those available on the Skaggs NMRs. Most of these have been tested and found to work. The optimisations are similar to those used by the standard parameter sets. The HSQC experiments use adiabatic ¹³C pulses, including a matched sweep inversion pulse, for more uniform signal recovery, and most have the option to use multiplicity editing. The HMBC experiments employ a two-fold J-filter. The 2BOB and H2OBC experiments produce constant time HMQC and H2BC spectra. Many of the experiments are compatible with NUS acquisition.

Name	Experiments
NOAH2_BO	HMBC and 2BOB
NOAH2_BO1	HMBC and H2OBC
NOAH2_BO2	HMBC and 2BOB
NOAH2_BS	HMBC and HSQC
NOAH2_SC	HSQC and COSY
NOAH2_SCqf	HSQC and DQF-COSY
NOAH3_BSC	HMBC, HSQC and COSY
NOAH3_BSCclip	HMBC, HSQC and CLIP-COSY
NOAH3_BSCqf	HMBC, HSQC and DQF-COSY
NOAH3_BSN	HMBC, HSQC and NOESY
NOAH3_BSR	HMBC, HSQC and ROESY
NOAH3_BSRad	HMBC, HSQC and adiabatic ROESY
NOAH3_BST	HMBC, HSQC and TOCSY
NOAH3_SCN	HSQC, COSY and NOESY
NOAH4_BSCN	HMBC, HSQC, COSY and NOESY

One of the big advantages of the NOAH supersequences is that all the different spectra needed for structure elucidation can be collected at once in an interleaved fashion. This ensures that any sample or instrument instabilities are averaged over all the experiments, making them more uniform and better suited for automated analysis.

The NOAH experiments that I have tried all worked remarkably well, however, I used high concentration samples. I also found that sweep widths and number of scans have to be the same for all modules. For example, the experiment outlined in the first figure is acquired with the same number of scans for the COSY as for the HMBC, which may not be ideal. Furthermore, the ¹³C sweep widths of the HSQC and HMBC spectra must be the same. There may be ways to get around this, but I think the NOAH experiments are best used for routine data collection of high concentration samples.

References

NOAH – NMR Supersequences for Small Molecule Analysis and Structure Elucidation.
Ēriks Kupče and Tim D. W. Claridge.
Angew. Chem. Int. Ed., 2017 56 11779-11783

Molecular Structure from a Single NMR Supersequence.
Ēriks Kupče and Tim D. W. Claridge.
Chem. Comm., 2018 54 7139-7142

New NOAH Modules for Structure Elucidation at Natural Isotopic Abundance.
Ēriks Kupče and Tim D. W. Claridge
J. Magn. Reson., 2019 307 106568

Triplet NOAH supersequences optimised for small molecule structure characterisation.
Tim D. W. Claridge, Maksim Mayzel, and Ēriks Kupče.
Magn. Reson. Chem., 2019 57 946-952