Institut de Chimie Moléculaire et des Matériaux d'Orsay

Laboratoire de RMN en milieu orienté - LRMN



Projet gNMR

Our project gNMR has been supported by a grant from the Agence Nationale de la Recherche (ANR 2011 JCJC SIMI 8) :"Development of gradient spatially encoded NMR : toward a novel generation of correlation experiments.". A description of this project is still available here.

Resolution vs acquisition rate : " squaring the cirle " ...

 Since the advent of Fourier transform NMR, and the development of multidimensional experiments, a vast number of pulse sequences have been specifically designed, that allow to gather all the interactions between active nuclei together on the same correlation spectrum. Unfortunately, in most of the systems that are of interest to the scientific community nowadays, the size or the complexity of the molecular architecture which is probed often leads to overcrowded spectra whose resolution is too low to give access to their analytical content.

 COSY and J-resolved spectra recorded on a strychnine sample at 600 MHz. Gathering accurate structural information, for instance about proton site H12, requires that a great number of datasets are recorded, and most of all a long analysis process (assignment, coupling measurements ...).


 In this context, considerable methodological developments have been devoted to the design of multi-dimensional experiments which aim either at reducing the number of correlations, or at simplifying their lineshape.
 On the one hand, sequences which employ hard pulses can be designed so that they give rise to different evolutions for different kinds of interactions in the spin network (chemical shift, dipolar or scalar coupling …): COSY and J-resolved experiments illustrate for instance in what extent chemical shift and scalar coupling informations can be separated. Unfortunately, the use of hard pulses creates as many coherences as there are spin interactions, and yields often overcrowded spectra.
 On the other hand, one way to reduce the number of correlations which contribute to the structure of NMR spectra consists in using semi-selective pulses (low power r.f. fields). Selective refocusing experiments have for instance opened the way to a site-specific measurement of each spin-spin interaction, out of unresolved spectra.(1-2) In this latter case however, reducing the number of interactions which contribute to the multiplet structure of a given signal, without reducing the number of signals is almost impossible. Consequently, the longer experimental time which is needed to collect site-specifically all the measurements from a given molecular assembly remains problematic.
 Finally, neither the use of very high field spectrometers, nor the development of pulse sequences that combine broadband and selective irradiations have allowed to fully address this problem so far.

Spatial Frequency Encoding (SFE) spectroscopy

  We develop original concepts which aim at enhancing resolution in spectra of fully coupled systems, to a point where their analysis becomes a faster, easier, and a more accurate process: we have recently proposed a novel appraoch which consists in carrying out a parallel acquisition of different experiments using a single-receiver-coil system.(3-4) For this purpose, we create a spatial frequency encoding of the sample, in a way which is similar to magnetic resonance imaging. This gradient encoded spectroscopy(5-11) has recently aroused a great interest in very different fields of Nuclear Magnetic Resonance, which has has led to a wide range of achievements as diverse as the recording of single scan multidimensional experiments(7), or pure J-resolved spectroscopy ...


 The methodology behind these applications can be divided in two groups. In the first of them, the combination of frequency swept pulses with a field gradient results in the acquisition of different evolution periods τ for molecules which are in different parts of the sample. It constitutes now the basis for Ultrafast (or single scan) NMR spectroscopy(8) or more specific purposes, such as implementing Zero Quantum filters(9), or QQ-HSQC experiments.(8)
 The second group is referred to as spatial frequency encoding. It is based on the use of semi-selective pulses, still in the presence of a field gradient, which allow to handle different spin coherences in different parts of the sample.  In the same way as the introduction of an additional evolution delay in multidimensional experiments allows to separate spin evolutions into different time dimensions, we have shown that a spatial frequency encoding can be considered as a novel spectral dimension in itself, insofar as it allows to perform different spin evolutions on spectrally resolved spins, at the same time. This methodology has notably been used for instance to acquire broadband homodecoupled 1H 2D spectra.(10)

Gradient encoded selective refocusing spectroscopy (G-SERF)

 More recently, we have presented the Gradient encoded homonuclear SElective ReFocusing experiment (G-SERF), which originates from this latter approach: a spatial frequency encoding is created along the sample, which allows to select, in separate cross sections, each interaction which is involved in the coupling network around a given proton site. In other words, spatially edited selective echoes are generated along the sample.


 On the resulting 2D spectrum, each coupling can be straightforwardly assigned and measured, on a fully resolved multiplet, at the resonance frequency of the coupling partner. The G-SERF pulse sequence allows to probe all the coupling network around a given proton site within a single, fully resolved 2D spectrum: although being intrinsically less sensitive, acquiring this single experiment requires less experimental time than recording the series of SERF spectra that would give the same structural information. In addition, it provides chemists with the assignement of the spin network that is being probed.(11)
 The great simplication of these pure J-resolved spectra which are acquired using this approach, paves the way for the investigation of complex spin networks.

Combining spin evolutions into fully tailored correlation spectra

We have also developed pulse sequences that take advantage of the combination of pure shift and J-edited techniques. On the one hand, we have shown that it is possible to combine pure shift and J-edited spin evolutions, by coding them along different gradient axes.This approach allows to trigger, in different parts of the sample, different spin evolutions in a selective and fully controlled manner. [12]


This PCR-COSY experiment is general, and yields 2D spectra in which the whole proton network appears as a series of fully resolved and straightforwardly assignable doublets, triplets or quartets, which gives access, on a single spectrum, to a fully edited –and assignable- measurement of the whole proton coupling network.


On the other hand, we have reported a new correlation experiment cited as "push-G-SERF" that meets the challenge of combining the analytical potential of multi-dimensional correlation spectra with the resolution enhancements provided by pure shift and J-edited spectroscopies. In the resulting phased 2D spectrum, the chemical shift information is selected along the direct dimension through a pure shift real time acquisition scheme, while scalar couplings involving a selected proton nucleus are edited in the indirect domain using a J-edited method.


This experiment allows for fully resolving both dimensions of the spectrum that yields a straightforward assignment and measurement of the coupling network around a given proton in the molecule. The robustness of this pulse sequence has been demonstrated on model compounds with increasing structural and spectral complexity, notably on an oligomeric saccharide showing a highly crowded 1H spectrum.[13]

Optimizing resolution and sensitivity in spatial frequency encoding NMR spectroscopy : from theory to practice

In order to understand the key features of NMR spectra acquired under spatial frequency encoding, we have developed a theoretical formalism to describe the spatial properties of the NMR signals that are locally created throughout the sample. The underlying idea is that the free induction decay that is detected by the receiver coil is the sum of the local spin evolutions that are triggered in different regions of the sample by gradient encoded selective irradiations.


We have studied the slice selection process that is occuring during gradient encoded excitation or refocusing pulses that are cornerstones of modern high resolution NMR pulse sequence. Notably, we have described the spatial properties of the NMR signal produced at key steps of state-of-the-art SFE experiments. The influence of experimental parameters such as the strength of the encoding gradient, or the selectivity of the selective pulse on spatial as well as spectral resolution has also been adressed.

The amplitude profiles calculated at the end of a gradient encoded excitation step, for an E-Burp2 selective pulse of duration from 20 ms to 120 ms.

We have shown that sensitivity and resolution of pure shift and J-edited experiments based on a spatial frequency encoding can be optimized to a point where high-resolution techniques based on a spatial frequency encoding approach show optimal performance compared to other methods.[14,15]


(1) Facke, T., Berger, S., J. Magn. Res. 113: 114-116 (1995)
(2) Beguin, L., Giraud, N., Ouvrard, JM., Courtieu, J. & Merlet, D. J. Magn. Res. 89 (1): 41-47 (2009)
(3) Kupce, E., Freeman, R., John, B.K., J. Am. Chem. Soc. 128: 9606-9607 (2006)
(4) Freeman, R., Kupce, E., J. BioMol. NMR 27: 101-113 (2003)
(5) Pell A.J., Keeler, J., J. Magn. Res. 189: 293-299 (2007)
(6) Morris, G., Aguilar, J., Evans, R., Haiber, S., Nilsson, M., J. Am. Chem. Soc. 132: 12770-12772 (2010)
(5) Frydman, L., Lupulescu, A., Scherf, T., J. Am. Chem. Soc. 125: 9204-9217 (2003)
(7) Pelupessy, P., Duma, D., Bodenhausen, G., J. Magn. Res. 194: 169-174 (2008)
(8) Thrippleton, M.J., Keeler, J., Ang. Chem. Int. Ed. 42: 3938-3941 (2003)
(9) Peterson, D.J., Loening, N.M., Magn. Res. Chem. 45, 937-941 (2007)
(10) Giraud, N., Joos, M., Courtieu, J. & Merlet, D. , Magn. Res. Chem. 47 (4): 300-306 (2009)
(11) Giraud, N., Beguin, L., Courtieu, J. & Merlet, D., Ang. Chem. Int. Ed. 49 (20): 3481-3484 (2010)
(12) Giraud, N., Pitoux, D., Ouvrard, J.M. & Merlet, D., Chem. Eur. J., 19: 12221-12224 (2013)
(13) Pitoux, D., Plainchont, B., Merlet, D., Hu, Z., Bonnaffé, D. Farjon, J., & Giraud, N.*, Chemistry –A European Journal, 21: 9044-9047 (2015)
(14) Plainchont, B., Pitoux, D., Hamdoun, G., Ouvrard, J.M., Merlet, D., Farjon, J., & Giraud, N.*, Physical Chemistry Chemical Physics, 18: 22827 - 22839 (2016)
(15) Plainchont, B., Farjon, J., & Giraud, N.*, Encyclopedia of Magnetic Resonance (eMagRes), Vol 5: 1377–1382 (2016)

Email: <- Page Précédente