NMR insights into the role of urea in cellulose dissolution: A stoichiometric study of cellobiose in NaOH/urea aqueous solution

Abstract

The mechanism underlying the enhanced dissolution of cellulose in NaOH/urea aqueous solutions has long been a subject of debate. In this study, high-resolution solution-state ¹H and ¹³C DEPTq NMR spectroscopy was employed to probe the intermolecular interactions within this ternary system, using cellobiose as a molecular model of cellulose. Chemical shift titration experiments revealed that urea does not directly interact with cellobiose via hydrogen bonding in neutral aqueous environments, but it can form a complex with cellobiose and NaOH characterized by a specific stoichiometric ratio of 1:2:1 (cellobiose/NaOH/urea) in the alkaline regime. The urea weakens the NaOH-cellobiose interaction and effectively shields the reducing-end hemiacetal groups from the ionic environment. It is suggested that the urea-modified solution shell consisting of Na⁺, OH⁻, urea, and water is responsible for the dissolution of cellulose.

Graphical abstract

1. Introduction

Cellulose, recognized as the most abundant renewable biopolymer on Earth, holds immense potential for the development of sustainable materials and bio-based chemicals [[1], [2], [3]]. However, its inherent recalcitrance, stemming from a robust network of inter- and intramolecular hydrogen bonds and van der Waals forces, renders it insoluble in water and most common organic solvents. This “dissolution bottleneck” severely restricts its efficient utilization. Among the emerging solvent systems, the pre-cooled NaOH/urea aqueous system has garnered significant attention as a promising, environmentally benign solvent that is less toxic than traditional CS₂ processes [4].

It is well-established that the addition of urea significantly enhances both the solubility of cellulose in NaOH aqueous solution and the thermal stability of the resulting solution, but the precise molecular mechanism driving this synergy remains a subject of intense debate [5]. Two primary schools of thought dominate the current discourse. The first hypothesis suggests the formation of a “channel inclusion complex”, where urea molecules form a host lattice that encages the cellulose macromolecules. The second hypothesis posits that urea acts primarily as a chaotropic agent or a water-structure breaker, stabilizing the cellulose-NaOH interactions by binding to water molecules rather than interacting directly with the polymer chains [[6], [7], [8]].

To distinguish between these mechanisms, researchers have employed various techniques, including solid-state ¹³C CP/MAS NMR and molecular dynamics (MD) simulations [[9], [10], [11]]. However, direct spectroscopic evidence characterizing the dynamic interactions and specific stoichiometry of the complex in the solution state remains scarce. The broad linewidths associated with high-molecular-weight cellulose in solution-state NMR often obscure the fine structural details required to quantify these interactions [12,13].

In this work, we address this challenge by utilizing cellobiose as a molecular probe and model compound for cellulose. Cellobiose retains the essential glycosidic linkage and the capability for intramolecular hydrogen bonding (Fig. 1), yet its low molecular weight allows for high-resolution solution-state NMR analysis. By systematically monitoring the ¹H and ¹³C DEPTq (Distortionless Enhancement by Polarization Transfer, including the detection of quaternary nuclei) chemical shift perturbations, we aim to map the interaction sites on the glucose ring and determine the precise stoichiometric ratio of the species involved [14,15].

Fig. 1.

Structure of cellobiose and the possible intramolecular hydrogen bonding connections indicated by dashed lines.

2. Experimental section

2.1. Materials and sample preparation

Cellobiose, NaOH, and Urea were purchased from Sinopharm Chemical Reagent Co., Ltd. (China). Deuterium oxide (D₂O) was obtained from Cambridge Isotope Laboratories, Inc. (USA). NMR samples were prepared in D₂O with a constant cellobiose concentration of 12 mmol/L. Three series of samples were prepared to independently investigate the effects of NaOH (Series I, NaOH/cellobiose molar concentration ratios from 0:1 to 20:1) and urea (Series II & III, varying urea content with and without fixed NaOH). After each addition of titrant (NaOH or urea), the samples were allowed to equilibrate for 15 min at 298 K before acquiring NMR spectra.

2.2. NMR spectroscopy

All NMR experiments were performed on a Bruker Avance III 600 MHz spectrometer equipped with a 5 mm TCI CryoProbe at 298 K.

¹H NMR spectra were recorded using a pre-saturation pulse sequence (zgpr). The relaxation delay was 1.5 s, and the acquisition time was 3 s. The ¹H spectral widths were 9090.9 Hz. The number of scans (ns) and dummy scans (ds) for ¹H were 16 and 2. The data point for ¹H was 64 k. The procpars were as follows: size of real spectrum (SI) = 64 k, line broadening (LB) = 0.30 Hz.

¹³C DEPTq NMR spectra were acquired using the deptgpq sequence to enhance sensitivity and multiplicity editing. The relaxation delays were 2 s, and the acquisition time was 1 s. The ¹³C DEPTq spectral widths were 30120 Hz. The number of scans (ns) and dummy scans (ds) for ¹³C DEPTq were 1024 and 4. The data point for ¹³C DEPTq was 64 k. The procpars were as follows: size of real spectrum (SI) = 64 k, line broadening (LB) = 1.0 Hz.

2D HSQC (Heteronuclear Single Quantum Coherence) experiments were conducted for signal assignment. HSQC data were acquired with 2048 × 256 data points (t2 × t1). The other experiment conditions are as follows: the pulse program was hsqcetgpprsisp2.2, center of the ¹H spectrum (o1p) = 4.70 ppm, center of the ¹³C spectrum (o2p) = 100.00 ppm, spectral width in the ¹H dimension (2sw) = 15.00 ppm, spectral width in the ¹³C dimension (1sw) = 216.00 ppm. The HSQC procpars were as follows: size of real spectrum in F2 (SI2) and F1 (S11) were both set to 2 k, window functions in F2 (WDW2) and F1 (WDW1) were both set to QSINE, sine bell shift in F2 (SSB2) and F1 (SSB1) were both set to 2. All NMR data were analyzed using TopSpin 4.1.4.

3. Results and discussion

3.1. Signal assignment and interaction between NaOH and cellobiose

To establish a baseline for interaction studies, the ¹H and ¹³C DEPTq NMR signals of neat cellobiose (12 mmol/L) in D₂O were first unambiguously assigned. This was achieved using 2D ¹H–¹³C HSQC spectroscopy (Fig. S1) combined with database references. The complete assignment of chemical shifts for both the α and β anomers is summarized in Table S1.

With the assignments confirmed, we investigated the initiation of dissolution by titrating NaOH into the cellobiose solution (Series I). The evolution of the ¹H and ¹³C DEPTq NMR spectra with increasing NaOH concentration is displayed in the stack plots in Fig. S2 and Fig. S3, respectively. Significant and systematic perturbations were observed: as the molar ratio of NaOH/cellobiose increased from 0:1 to 20:1, most alkyl proton resonances shifted upfield (shielding), whereas the corresponding carbon resonances shifted downfield (deshielding) (Fig. 2 and Fig. 3) [16].

Fig. 2.

Changes of the proton chemical shift difference in cellobiose (12 mmol/L) with increasing NaOH in D₂O (Δδ = δ−δ_neat).

Fig. 3.

Changes of the carbon chemical shift difference in cellobiose (12 mmol/L) with increasing NaOH in D₂O (Δδ = δ−δ_neat).

A detailed analysis of these shifts reveals site-specific interactions. The atoms adjacent to the reducing end hemiacetal group—specifically C1′/H1′ and C2′/H2′—displayed the most pronounced chemical shift changes compared to the non-reducing end. This behavior can be attributed to the higher acidity of the hemiacetal hydroxyl group (pK_a approx 12–13) compared to aliphatic hydroxyls. In alkaline conditions, the deprotonation or strong hydrogen-bonding interaction at the anomeric center (C1′) significantly perturbs the electron density of the adjacent C2′ position. In addition to the most pronounced shifts observed at the reducing-end positions (C1′/H1′ and C2′/H2′), significant chemical shift changes are also evident for other resonances, such as H5′/C5′. Notably, the H5′ proton and the corresponding carbon signal exhibits an upfield shift—changes that, although less pronounced than those at C1′/H1′ and C2′/H2′, suggest a broader influence of NaOH on the electronic environment of the glucose ring. Even distant carbons, including those not directly adjacent to the reducing end, show subtle but consistent shifts. This indicates that the perturbation induced by NaOH propagates through the glycosidic linkage and affects the overall electron density distribution across the ring. Furthermore, the relative stability of the C3′/H3′, C4′/H4′ and C6′/H6′ resonances suggests that the intramolecular hydrogen bonds (O3′–H…O5 and O2–H⋯O6′) largely sustain the conformation of the glucose rings even in the presence of NaOH.

3.2. The synergistic role of urea and complex stoichiometry

To isolate the specific role of urea, we first performed a control experiment by titrating urea into a solution of neat cellobiose in neutral D₂O (Series II). As shown in the spectra in Fig. S4 (¹H) and Fig. S5 (¹³C DEPTq), no observable variations in both ¹H and ¹³C chemical shifts occurred even at high urea concentrations. This crucial result confirms that urea does not interact with cellobiose via direct hydrogen bonding in neutral water, corroborating MD simulation findings that urea requires an activated cellulose surface or an alkaline environment to bind effectively [17].

In sharp contrast, when urea was added to the alkaline cellobiose solution (12 mmol/L cellobiose + 24 mmol/L NaOH, Series III), it induced a dramatic reversal in the chemical shift trends. The original spectra for this titration are presented in Fig. S6 and Fig. S7. The proton signals, which had initially shifted downfield due to NaOH, moved back (e.g., upfield) in a step-like manner upon urea addition, while carbon signals displayed the opposite behavior (Fig. 4, Fig. 5). The above result indicates that the NaOH-cellobiose interactions are weakened or partially “shielded” by the addition of urea to a certain concentration above. Such a “shielding” effect suggests that urea modifies the solvation shell of the NaOH-cellobiose adduct, likely by displacing water molecules or forming a shielding layer that reduces the direct exposure of the sugar ring to the bulk ionic environment.

Fig. 4.

Changes of the proton chemical shift difference in cellobiose (12 mmol/L) with increasing urea in NaOH/D₂O solution (The molar ratio of NaOH/cellobiose is 2:1; Δδ = δ−δ_neat).

Fig. 5.

Changes of the carbon chemical shift difference in cellobiose (12 mmol/L) with increasing urea in NaOH/D₂O solution (The molar ratio of NaOH/cellobiose is 2:1; Δδ = δ−δ_neat).

Most significantly, the titration curves in Fig. 4 and Fig. 5 reveal step-like variations at a urea/cellobiose molar ratio between 1.2:1 and 1.6:1. Beyond this ratio, the chemical shifts plateau, indicating the saturation of the interaction sites. Given the experimental conditions (fixed NaOH/cellobiose ratio of 2:1) and the observed saturation point, the data strongly supports the formation of a stable ternary complex with a defined stoichiometry. Based on these results, we propose a stoichiometric ratio for the cellobiose: NaOH: urea complex of 1:2:1 [18].

3.3. The synergistic role of urea and complex stoichiometry

These spectroscopic findings allow us to construct a refined molecular picture for the dissolution of cellobiose in aqueous NaOH/urea solution. The process begins with NaOH interacting with specific hydroxyl groups to form a hydrated intermediate, as evidenced by the dramatic shifts in Fig. S2 and Fig. S3. Urea then associates with this intermediate, not through random co-solvation, but by assembling a structured inclusion complex. The absence of interaction in Fig. S4 and Fig. S5 highlights that NaOH “activates” the cellobiose for urea binding. The stoichiometry of 1 urea per cellobiose unit suggests that urea molecules likely occupy specific pockets created by the NaOH-induced conformational changes, effectively “capping” the hydrophobic faces of the glucose rings via van der Waals forces. This inclusion complex acts as a molecular chaperone, preventing the re-association of cellulose chains and thereby enhancing the stability of the solution.

4. Conclusion

This work provides definitive spectroscopic evidence elucidating the specific role of urea in cellulose dissolution. By utilizing cellobiose as a molecular probe to circumvent line-broadening issues, we identified a stable ternary complex with a rigorous stoichiometry of 1:2:1. The NMR data clarify a synergistic mechanism wherein urea does not bind to the carbohydrate backbone via hydrogen bond in neutral water but selectively stabilizes the alkali-cellulose adduct once formed, specifically by shielding the reducing end. These insights not only support the “inclusion complex” model but also provide quantitative structural parameters essential for the rational design and optimization of urea-based “green” solvent systems for cellulose processing. Although the use of cellobiose as a model provides high-resolution NMR insights, it does not fully capture the polymeric nature of cellulose, including chain-chain packing, crystallinity, and long-range intermolecular interactions. Furthermore, our experiments were conducted under dilute conditions, whereas industrial dissolution typically involves higher concentrations where kinetic and thermodynamic factors may differ significantly. Nevertheless, we believe that the molecular-level insights and quantitative stoichiometric parameters obtained here offer a valuable foundation for understanding cellulose dissolution mechanisms.

CRediT authorship contribution statement

Bowen Han: Writing – original draft, Methodology, Investigation, Data curation. Dandan Guo: Validation, Formal analysis. Xing Zhang: Writing – review & editing, Supervision. Dongsheng Zhang: Validation, Formal analysis. Shaohua Huang: Writing – review & editing, Supervision, Project administration, Funding acquisition.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Shaohua Huang is an editorial board member of Magnetic Resonance Letters but was not involved in the editorial review or the decision to publish this article.

Biographies

Bowen Han received the B.S. degree in chemistry from Zhengzhou University in 2016, and M.S. degrees in organic chemistry from Xiamen University in 2019. He is currently working in the Institute of Drug Discovery Technology, Ningbo University. His research interests include DOSY NMR application and chiral recognition.

Dandan Guo received the B.S. degree in chemistry from Zhejiang University in 2014 and the Ph.D. degree in chemistry from Zhejiang University in 2019. She is currently working at the Institute of Drug Discovery Technology, Ningbo University. Her major research interests include chromatographic analysis, stationary phase materials, and chiral separations.

Xing Zhang received the Ph.D. degree from Huazhong Agricultural University in 2022. She is currently working at the Institute of Drug Discovery Technology, Ningbo University. Her research focuses on natural products chemistry and the deep processing of agricultural products. Her work primarily includes the analysis of key components of bioactive proteins, research on their functional nutrition, and comprehensive utilization.

Dongsheng Zhang, born in Hebei in 1974, received his Ph D. degree in Forestry Engineering from Chinese Academy of Forestry in 2005. He is currently the Deputy Director General of the International Bamboo and Rattan Organization(INBAR). His major research interests include bamboo and rattan, soil and water Conservation, forest cultivation, sustainable agriculture, forestry engineering, and law.

Shaohua Huang was born in Jiangsu in 1976. He received the Ph.D. degree in analytical chemistry from Wuhan Institute of Physics and Mathematics, Chinese Academy of Sciences in 2008. He is currently a full professor with the Institute of Drug Discovery Technology, Ningbo University. His major research interests include chirality, chiral separation, chiral amplification, NMR methodology and HPLC methodology.

Appendix A. Supplementary data

The following is the Supplementary data to this article: