SABRE was initially demonstrated with a single substrate of interest. 81 Shortly afterwards, the potential of SABRE-based hyperpolarisation for analytical applications was illustrated by Lloyds et al. 111 A key feature of SABRE is the reversible nature of the hyperpolarisation process, which can occur multiple times for a given substrate. As a result, an array of classic 2D NMR experiments can be used with little modifications to record spectra with enhanced sensitivity. Experimental complications, however, result from the fact that the polarisation transfer occurs spontaneously only at low magnetic field (in the milliTesla range of 1H and in the microTesla range for 15N). The sample thus has to be moved between a polarisation region and the NMR spectrometer. The most used methods consist of dropping the sample in the magnet after shaking, or using a flow system to shuttle the sample between a polarisation region (where parahydrogen gas is bubbled) and the detection region. 112 P. N. Reardon, C. L. Marean-Reardon, M. A. Bukovec, B. E. Coggins and N. G. Isern, Anal. Chem., 2016, 88, 2825 CrossRef CAS PubMed .

L. Guduff, D. Kurzbach, C. van Heijenoort, D. Abergel and J.-N. Dumez, Chem. – Eur. J., 2017, 23, 16722 CrossRef CAS PubMed . Fig. 8 SPEN 3D DOSY of a mixture of 3 alcohols (methanol, ethanol, propanol) and an amino-acid ( L-valine), at a concentration of ∼100 mM in D 2O. (a–e) COSY-type spectra obtained as slabs of the 3D ( D, δ direct, δ indirect) dataset resulting from DOSY processing. The selected range in D is shown in each panel. The slice of the ( z, δ direct, δ indirect) dataset with the lowest diffusion gradient area is shown in (f). The experiment was carried out with a 600 MHz spectrometer equipped with a triple-axis gradient high-resolution probe, and lasted 12 min. Adapted from ref. 161 with permission from the Royal Society of Chemistry. The complexity of the NMR spectra for mixtures has several origins. First, similar compounds have similar spectra and as a result the spectrum of a mixture may be complex even with a moderate number of detectable compounds. Second, the number of compounds can be very large. This is for example the case in biofluids or extracts of biological samples. In the spectra of mixtures, signal overlap prevents the reliable identification and accurate quantification of signals. The assignment of each signal in the spectrum to a specific compound, even resolved one, is also not trivial. Note that samples such as natural organic matter, which comprise tens of thousands of compounds that are mostly unknown, and call for the use dedicated methods, are outside the scope of the review. 4 R. Freeman, Spin Choreography: Basic Steps in High Resolution NMR, Oxford University Press, 1998 Search PubMed .A. Dawn, M. Mirzamani, C. D. Jones, D. S. Yufit, S. Qian, J. W. Steed and H. Kumari, Soft Matter, 2018, 14, 9489–9497 RSC. B. Schade, A. K. Singh, V. Wycisk, J. L. Cuellar-Camacho, H. von Berlepsch, R. Haag and C. Böttcher, Chem. – Eur. J., 2020, 26, 6919–6934 CrossRef CAS PubMed.

Jean-Nicolas Dumez is a CNRS researcher (Centre National de la Recherche Scientifique) currently working at Nantes Université. He received his PhD in 2011 from Ecole Normale Supérieure de Lyon, and carried out post-doctoral research at the Weizmann Institute of Science and Southampton University, before moving in 2014 to Université Paris-Saclay as a CNRS researcher, then to Nantes in 2018. His research interests include the development of methods for the analysis of mixtures, NMR pulse sequences, and spin dynamics.

2D 1H-13C HSQC (multiplicity edited)

A possible approach to improve the separation of overlapping spectra is to use multivariate processing methods. They consist of algorithmically solving the problem: S. Balayssac, S. Trefi, V. Gilard, M. Malet-Martino, R. Martino and M. A. Delsuc, J. Pharm. Biomed. Anal., 2009, 50, 602 CrossRef CAS PubMed . Nuclear magnetic resonance (NMR) spectroscopy is a powerful approach for the analysis of mixtures. 1,2 The most common NMR experiment has several essential features that are highly favourable for mixture analysis. It relies on a “pulse-acquire” pulse sequence, carried out at thermal spin equilibrium, to yield 1D spectra, most commonly for 1H spins. This experiment is simple to implement, it can be fast, and it already yields a wealth of structural information, which is useful to identify components. It is also non-invasive and non-destructive, meaning that the sample can be retrieved after analysis for storage or other analyses, and also, in the case of monitoring applications, that experiments can be repeated multiple times on the same sample as a function of time. 1D 1H NMR also provides straightforward access to quantitative information on concentrations. L. Sellies, R. Aspers, M. C. Feiters, F. Rutjes and M. Tessari, Angew. Chem., Int. Ed., 2021, 60, 26954 CrossRef CAS PubMed . A. Ustinov, H. Weissman, E. Shirman, I. Pinkas, X. Zuo and B. Rybtchinski, J. Am. Chem. Soc., 2011, 133, 16201–16211 CrossRef CAS PubMed.