Elsevier

Carbohydrate Polymers

Volume 298, 15 December 2022, 120104
Carbohydrate Polymers

Assessment of cellulose interactions with water by ssNMR: 1H->13C transfer kinetics revisited

https://doi.org/10.1016/j.carbpol.2022.120104Get rights and content

Abstract

To evaluate cellulose interactions with water, 1H->13C polarization transfer kinetics during Variable Contact Time CP-MAS NMR spectroscopy were studied and modelled using cellulose of different origins. The increase in the temporal resolution of the plot relating signal intensity to contact-time made it possible to compare different physical models for use in fitting the kinetic curve. These models involve combinations of variables, such as proton spin diffusions, that require a better understanding of their physicochemical and structural bases. To that end, hydrogen interactions were modulated by adding water, first by varying cellulose water content, second by exchanging hydroxyl protons with D2O, and last by varying the spinning rate.

The results demonstrate that this approach makes it possible to probe interactions of polysaccharides with structural water, as well as to follow the evolution of the proton-proton interactions during hydration through spin diffusion times.

Introduction

Cellulose is a linear chain of β-(1->4) glucose units. This simple structure becomes more complex when several glucan chains associate and then come together to form a fibre network, as encountered in plant cell walls (Khodayari et al., 2021). Cellulose is used in a wide range of applications, such as packaging, paper, composite materials, adhesives, absorbents, electronics, medical, pharmaceutical and cosmetic products (Sharma et al., 2019). To further improve our understanding of the many functions of plant cell walls and to expand the uses of cellulose, a detailed knowledge of the relationships between its structures at different levels (elementary fibres, fibre aggregates…) and functional properties is required.

Cellulose fibres are made up of crystalline and amorphous regions (Mariano et al., 2014). Their length varies from 0.1 to several tens of microns, while their height and width ranges from 3 to 50 nm (Beck-Candanedo et al., 2005). The fibres are built on the aggregation of elementary fibres made of 18 glucan chains extruded from “rosettes” made of 6 transmembrane cellulose synthase complexes (Allen et al., 2021; Cosgrove, 2022). The cellulose amphiphilicity is thought to play a key role in the mechanism governing interactions with water (Chami Khazraji & Robert, 2013). Chen et al. (2022) have produced thermodynamic evidence of the presence of water confined within and/or strongly bonded to the interfaces between cellulose elementary fibrils in cellulose fibril aggregates. Their results are in line with a previous 2H NMR study (Lindh & Salmén, 2017) that showed slow and anisotropic reorientation of part of the water in cellulose. Depending on the origin of cellulose, the crystalline domains can be of different sizes and shapes (Moon et al., 2011). Indeed, the crystal cross section can be a square, a cylinder, rectangular or a diamond. Moreover, native cellulose is a composite of two distinct crystal modifications, namely Iα and Iβ, whose fractions vary depending on the origin of the cellulose. The Iα and Iβ structures are associated, respectively, with a triclinic chain and two monoclinic chains (Sugiyama et al., 1991). Recent ssNMR studies on cell walls have shown a more complicated organization, such as different packing environment for internal cellulose and surface chains (Ghassemi et al., 2021; Shekar, 2022).

Two major structural models for cellulose have been proposed in the literature. The first is the “core-shell” model, often used in the interpretation of NMR characterisations of cellulose (Larsson & Westlund, 2005). It takes the view that cellulose chains form a crystalline core surrounded by less structured chains approaching the outside of the cellulose aggregate (Villares et al., 2017). The second model proposes the presence of mesomorphous-crystalline nanofibrils (Ioelovich, 2016) with a paracrystalline monomolecular surface layer. This monolayer is superimposed on an alternation of crystalline and amorphous regions. These models illustrate the structural diversity of cellulose indicating the potential complexity of interactions occurring at the nanoscale.

Solid-state Nuclear Magnetic Resonance spectroscopy (ssNMR) has become a major tool in the study of macromolecular complexes (Opella, 2015; Wang & Hong, 2016). For cellulose, it is used as a reference technique to determine the degree of crystallinity (Park et al., 2010). It is also the only technique to provide access to the relative proportions of alpha and beta alloforms (Newman, 1999).

The study of cellulose by ssNMR is mainly performed using 13C nucleus resonance, whose signal is intensified by the magnetization transfer of protons. An approach used for a long time and still relevant today consists in carrying out a partial or total isotopic labelling of the sample to be analysed. This significantly increases the 13C signal. Nevertheless, it is a costly approach, and is difficult to apply outside of model plants (Zhao et al., 2021). A more recent approach, the Magic-Angle Spinning Dynamic Nuclear Polarization (MAS-DNP), makes it possible to solve this problem, but involving rather considerable operating costs (Ghassemi et al., 2021; Smith et al., 2019).

However, there is a less costly approach that allows for the analysis of a larger number of samples. This experiment, known as cross-polarization (CP), is carried out by varying the contact time (CT) or spin lock (SL) duration from a few microseconds to tens of milliseconds (see below). In addition to increasing the carbon signal, the kinetics of polarization transfer as a function of CT or SL times has long been studied in solid-state NMR to characterize the organization of macromolecules (Stejskal et al., 1981; Willis & Herring, 1987).

Three main NMR pulse sequences are available to access all or part of the polarization transfer kinetics for various substrates (Abelmann et al., 2004; Jurkiewicz & Maciel, 1994; Smernik et al., 2002): VCT-CP (Variable Contact Time), VSL-CP (Variable Spin Lock) and Direct VSL, making it possible to transfer magnetization from protons to carbons. The last two allow access only to the decreasing part of the kinetics, which provides the characteristic proton rotating-frame relaxation, TH. The first stage of the kinetics (a few microseconds) can be measured using the VCT-CP sequence using different contact times. This sequence gives access to the initial polarization transfer, which provides clues on the macromolecular environment (Dagys et al., 2018; Jarvis et al., 1996).

Most dynamical studies of plant macromolecular assemblies with natural abundant 13C are restricted to the measurement of TH, which provides information on the molecular order within a sphere of a few tens of nanometers in diameter around the studied carbons (Newman, 1992; Yuris et al., 2019). Other accessible parameters, such as the mean dipolar coupling between proton and carbon (TCH) and the proton spin-diffusion time (THH), are sensitive to strong homonuclear coupling and can complement TH allowing interactions to be located more accurately (Lahaye et al., 2020). Given that the term “spin diffusion” was coined by Bloembergen (Bloembergen, 1949) to characterize the process of polarization exchange between spins exhibiting strong homonuclear couplings, spin diffusion is likely to be affected by the mobility of water molecules. All these parameters are, however, under-used due to the time-consuming nature of this VCT-CP sequence in terms of both acquisition time and spectra processing. For this reason, most kinetics experiments are performed with a small number of contact times (CT) (Hjertberg et al., 1986; Lahaye et al., 2020), though use of multiple CT points was reported to produce a marked improvement in the quality of the kinetic parameters (Dagys et al., 2018; Dagys et al., 2020). To these acquisition times can be added the considerable time taken to process the data. Indeed, low spectral resolution combined with the natural abundance of 13C results into high numbers of overlapping lines, most of which require spectral decomposition to be carried out in what is usually a manual process.

Polarization transfer is a complex phenomenon involving several mechanisms. This fact has given rise to the use of different physical models (Kolodziejski & Klinowski, 2002) to describe it based on different theories of polarization transfer, resulting in the definition of many different variables.

The present study adopts the hypothesis that polysaccharides-water interactions in polysaccharide assemblies can be characterized by studying the 1H->13C polarization transfer. To test this hypothesis, acquisition and processing conditions of VCT-CP MAS 13C NMR were optimized for the acquisition of 460 CT points from which the polarization transfer kinetics was studied. Such high CT numbers made it possible to test different model equations to determine which of the latter achieved the best fit with the experimental data. To further identify the origins of the different variables in these equations, notably the different THH, several celluloses of different size and crystallinity were investigated and a number of conditions were modulated, namely, water content, deuterium exchange between the cellulose sample's hydroxyl groups, and spinning rate. Cellulose as the major component of complex polysaccharides assemblies in nature has been chosen as a simplified version of these assemblies to evaluate the analytical methods.

Section snippets

Theory

Magnetization transfer from protons to carbons is a complex phenomenon involving several mechanisms and is characterized by different durations. In summary, the experiment (Fig. 1a) consists in applying a radiofrequency pulse to flip the orientation of the protons to 90° in relation to field B0, and then placing the H-C spin system in the so-called Hartmann-Hahn condition (Hartmann & Hahn, 1962). This is achieved by applying a radiofrequency pulse simultaneously to the protons and carbons in

Materials

Native cotton cellulose nanofibrils (CNF), cotton nanocrystals (native CNCI and mercerized CNCII) and tunicate nanocrystals (TNC) were supplied from the laboratory collection (Favier et al., 1995; Haouache et al., 2022). Avicel® PH-101 (powdered; 99.7 % purity; mean particle size of 50 μm) was supplied by Sigma-Aldrich.

Basic characteristics of the different substrates are as follows:

For CNCI: 3–6 nm of cross section, 50–250 nm of length, a crystallinity index of 54 % and a water content of

Modelling relaxation and spin diffusion dynamics in cellulose nanocrystals

To improve the VCT-CPMAS spectra signal-to-noise ratio and thereby better explore the polarization transfer kinetics from protons to carbons in cotton cellulose nanocrystals (CNCI), the sum of the intensities of signals C1 to C5, all of which are CH, was used. Having improved the signal-to-noise ratio of the kinetics through this sum, a further step was required to better understand the phenomena of polarization transfer, since CH and CH2 must be differentiated as they yield different kinetics (

Conclusion

The significant gain in temporal resolution achieved through the use of optimized acquisition and processing parameters for the VCT-CPMAS experiment allowed the testing of complex physical models of polarization transfer kinetics to enrich our understanding of cellulose interactions with water. The use of two equations with four spin diffusion times allowed a better fit for the results obtained for different celluloses. There are in fact three different spin-diffusion times at three different

CRediT authorship contribution statement

X. Falourd: Conceptualization, Methodology, Formal analysis, Investigation, Writing – original draft. M. Lahaye: Conceptualization, Methodology, Writing – review & editing. C. Rondeau-Mouro: Conceptualization, Methodology, Writing – review & editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

Thanks go to Isabelle Capron for kindly providing the CNCI, CNCII and NFCI and to Bruno Pontoire for kindly providing tunicate cellulose. The authors thank Lucie Birault and Camille Jonchère for their Karl-Fisher measurements.

References (61)

  • M. Lahaye et al.

    Cellulose, pectin and water in cell walls determine apple flesh viscoelastic mechanical properties

    Carbohydrate Polymers

    (2020)
  • P.T. Larsson et al.

    Line shapes in CP/MAS 13C NMR spectra of cellulose I

    Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy

    (2005)
  • G. Metz et al.

    Ramped-amplitude cross polarization in magic-angle-spinning NMR

    Journal of Magnetic Resonance, Series A

    (1994)
  • T.Í.S. Oliveira et al.

    Bionanocomposite films based on polysaccharides from banana peels

    International Journal of Biological Macromolecules

    (2017)
  • S.J. Opella

    Solid-state NMR and membrane proteins

    Journal of Magnetic Resonance (San Diego, Calif.: 1997)

    (2015)
  • M. Paris et al.

    NMR local range investigations in amorphous starchy substrates : II-dynamical heterogeneity probed by 1H/13C magnetization transfer and 2D WISE solid state NMR

    International Journal of Biological Macromolecules

    (2001)
  • R.J. Pugmire et al.

    Structural evolution of matched tar-char pairs in rapid pyrolysis experiments

    Fuel

    (1991)
  • A. Sharma et al.

    Commercial application of cellulose nano-composites – A review

    Biotechnology Reports

    (2019)
  • R.J. Smernik et al.

    Determination of T1ρH relaxation rates in charred and uncharred wood and consequences for NMR quantitation

    Solid State Nuclear Magnetic Resonance

    (2002)
  • I. Wawer et al.

    Solid state NMR study of dietary fiber powders from aronia, bilberry, black currant and apple

    Solid State Nuclear Magnetic Resonance

    (2006)
  • H. Yang et al.

    Quantitative characterization of coal structure by high-resolution CP/MAS 13C solid-state NMR spectroscopy

    Proceedings of the Combustion Institute

    (2021)
  • A. Yuris et al.

    The interactions between wheat starch and Mesona chinensis polysaccharide : A study using solid-state NMR

    Food Chemistry

    (2019)
  • L.B. Alemany et al.

    Cross polarization and magic angle sample spinning NMR spectra of model organic compounds. 3. Effect of the carbon-13-proton dipolar interaction on cross polarization and carbon-proton dephasing

    Journal of the American Chemical Society

    (1983)
  • M. Bardet et al.

    Dynamics property recovery of archaeological-wood fibers treated with polyethylene glycol demonstrated by high-resolution solid-state NMR

    Cellulose

    (2012)
  • S. Beck-Candanedo et al.

    Effect of reaction conditions on the properties and behavior of wood cellulose nanocrystal suspensions

    Biomacromolecules

    (2005)
  • A. Chami Khazraji et al.

    Interaction effects between cellulose and water in nanocrystalline and amorphous regions: A novel approach using molecular modeling

    Journal of Nanomaterials

    (2013)
  • P. Chen et al.

    Water as an intrinsic structural element in cellulose fibril aggregates

    Journal of Physical Chemistry Letters

    (2022)
  • R.L. Cook et al.

    Structural characterization of a fulvic acid and a humic acid using solid-state ramp-CP-MAS 13C nuclear magnetic resonance

    Environmental Science & Technology

    (1998)
  • D.J. Cosgrove

    Building an extensible cell wall

    Plant Physiology

    (2022)
  • L. Dagys et al.

    Processing of CP MAS kinetics : Towards NMR crystallography for complex solids

    The Journal of Chemical Physics

    (2016)
  • Cited by (3)

    View full text