Effective dispersion of aqueous clay suspension using carboxylated nanofibrillated cellulose as dispersant

Abstract

Using a natural dispersant to prepare a well-dispersed aqueous clay suspension with remarkable fluidity and stability is of significance because of the gradual depletion of fossil-based raw materials. In this study, nanofibrillated cellulose (NFC) prepared by 2,2,6,6-tetramethylpiperidine-1-oxyl oxidation method was used as a green dispersing agent to effectively disperse clay particles in water. The carboxylated NFC dispersant increased the maximum solid content of an aqueous clay suspension while reducing its static viscosity. As compared to a sodium polyacrylate-dispersed clay suspension, the NFC-dispersed clay suspension not only exhibited similar maximum solids content, particle-size distribution, and rheology, it also had better water retention value and stability. In addition, we found that the increased carboxyl content of NFC had a significant positive impact on the dispersing effect of clay suspensions. This green and effective carboxylated NFC shows its potential to disperse aqueous clay particles with excellent performance for advanced uses.

---

Introduction

Clay is the most widely used mineral in human civilization. It is used in applications from common paper and ceramic industries to state-of-the-art end uses such as separation, nanocomposites, emulsifiers, and biomedicine due to its lamellar particle shape, non-toxicity, good color, low surface area, chemical inertness, and fine particle size.^1–9 Good dispersion and stability of aqueous clay suspensions are extremely important to yield a final product with the best performance.^10,11 However, the heterogeneity of the charged edges and faces of clay invokes electrostatic face-to-face, edge-to-edge, and edge-to-face attraction, leading to the formation of card-house agglomerates at high clay concentrations. To achieve well-dispersed and stable clay suspensions with excellent fluidity at a high solids contents for cutting-edge applications, a variety of dispersants, such as sodium hexametaphosphate,¹² sodium polyacrylate,¹³ and sodium polyphosphate,¹⁴ are applied to modify the particle surfaces to enhance the electrostatic repulsion between clays.^15,16 Moreover, chemical modification of clay has been applied extensively to improve its dispersity in water.^10,17 However, chemical functionalization increases the production cost of clay. Additionally, the introduction of dispersants may deteriorate the interaction between clay and desirable substances.

Cellulose nanofibers (CNs) assembled by cellulose chains have unsurpassed quintessential physical and chemical properties^18–21 that make them a fantastic material ingredient for numerous cutting-edge applications ranging from electronic and energy storage devices to filtration, hybrid nanocomposites, and functional materials.^22–30 It has been extensively proven that CNs exhibit amphiphilic characteristics resulting from the presence of hydrophobic C–H moieties at the equatorial positions and hydrophilic hydroxyl groups at the axial positions of glucopyranose rings.^31,32 Hence, this characteristic enables CNs to uniformly disperse carbon nanotubes^33,34 and two-dimensional (2D) materials^35,36 (graphene, molybdenum disulfide, boron nitride) in water to prepare high-performance nanocomposites for advanced applications. In addition to the amphiphilicity of CNs, they have the ability to form a three-dimensional (3D) network configuration in the continuous phase that provides steric hindrance of aggregation through entrapment of particles or emulsion droplets,^37,38 which could be further enhanced by introducing charged carboxyl groups onto the surfaces of CNs to improve electrostatic repulsive forces.^35,39

The aim of this study was to achieve a well-dispersed aqueous clay suspension with desirable stability and fluidity by using carboxylated nanofibrillated cellulose (NFC) as a green and effective dispersing agent. One wt% NFC dispersions were initially obtained by a combination of a 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO) oxidization system and homogenization. The shear-thickening behavior of these NFC suspensions, measured by ACA automated ultra-high shear viscometry in the high shear rate area between 3 × 10⁵ and 19 × 10⁵ s⁻¹, was first identified. TEMPO-oxidized NFC was then used as a dispersing agent to disperse clay in water in a disperser and the particle-size distribution, maximum solids content, rheology, and stability of the NFC-dispersed clay suspension was investigated.

Experimental

Materials

Commercial bleached eucalyptus pulp (Arauco Eucalyptus, Chile) was used as the raw material for the preparation of TEMPO-mediated NFC. Clay primarily made of kaolinite was obtained from Maoming City in Guangdong province of China. Laboratory grade sodium bromide, TEMPO, and 10–13% sodium hypochlorite solution were purchased from Sigma Aldrich. Sodium polyacrylate with 40 wt% solid content was kindly provided by SNF (Disperse 3000, SNF China).

Preparation of TEMPO-oxidized NFC dispersant

TEMPO-treated wood pulp was prepared according to our previous publication.²⁵ The obtained 1% TEMPO-oxidized wood pulp was disintegrated by a homogenizer (Nano DeBee, BEE International, USA) with different size nozzles several times. The pulp was then passed once through a homogenizer with a D10 nozzle. The resulting aqueous NFC was then passed through a D4 nozzle for 1, 3, 5, and 7 cycles. The detailed disintegration procedure for the NFC is given in Table 1.

Table 1 Procedure for NFC preparation

Sample	Passes (D10 at 5000 psi)	Passes (D4 at 20 000 psi)
a	1	1
b	1	3
c	1	5
d	1	7

---

Clay dispersion

Clay particles and 0.2 wt% NFC (relative to the clay particles) were mixed together into distilled water (100 mL) in a beaker. The mixed dispersion was stirred at 2000 rpm in a disperser (Shanghai, China) for 30 min. It was then stirred at 700 rpm for 5 min in order to eliminate air bubbles. The clay dispersed by sodium polyacrylate was produced using the same procedure.

Characterization of clay dispersion

The water retention value (WRV) of the clay dispersion was measured by the dynamic water retention measurement (DWR dynamic water retention analyzer, ACA systems Oy, Finland).⁴⁰ First, the clay dispersion was poured into the sample holder-coating barrel. Following this, a piece of filter membrane (diameter: 50 mm, pore size: 0.22 μm, Sigma-Aldrich) was put on the testbed and a piece of pre-weighed filter paper with a diameter of 50 mm and an average pore size of ∼150 μm (Whatman cellulose filter paper, Sigma-Aldrich) was put on the filter membrane. The inside pressure of the coating barrel was set to no more than 0.5 MPa, and then the testing machine started running. After testing, the weight of the filter paper was measured. WRV is expressed as grams of water per area of filter paper, and it was calculated according to the following equation:
where m_t is the final weight of the filter paper, m₀ is the initial weight of the filter paper, and A (=1735 mm²) is the active area of the filter paper. The calculation was repeated several times until there was little difference between the results. The average value of the most relevant results was used as the final water retention value.

The viscosity of the clay dispersion at ultra-high shear rates was measured on an ACAV A2 capillary rheometer (ACA systems, Oy, Finland). The capillary dimensions were a diameter of 0.5 mm and a length of 50 mm. First, 1 L of the clay dispersion was placed in the cylindrical container of the ultra-high shear viscometer. Following this, the parameters and the kinetic correction factor were chosen carefully. Afterwards, the range of pressures was set between 0 and 60 MPa and the step number was set as 18. Particle size distribution of clay dispersion was analyzed by the Mastersizer 3000 laser diffraction particle size analyzer.

Results and discussion

Characterization of NFC dispersion

TEMPO oxidation was considered a promising approach to oxidizing the C6 primary hydroxyls at the surface of solid cellulose into carboxyl groups without impairing their original crystallinities.⁴¹ The charged wood fiber facilitates the subsequent mechanical disintegration due to the introduction of electrostatic repulsion between adjacent cellulose nanofibrils. A TEMPO/NaBr/NaClO oxidation system was utilized to pretreat the softwood pulp prior to mechanical treatments. The TEMPO-treated wood pulp was then passed once through a homogenizer with a D10 nozzle, and the resulting aqueous NFC dispersion was disintegrated a further 1, 3, 5, and 7 times using a homogenizer with a D4 nozzle. The fibrous morphology of the resulting NFC presented in Fig. 1a–d illustrates that both the fiber length and fiber width tended to decrease with increasing number of cycles through the homogenizer. The average fiber width of NFC with 2, 4, 6, and 8 treatment cycles was approximately 24, 19, 16, and 16 nm, respectively. As the number of passes increased from 2 to 6, the fiber width decreased from 24 to 16 nm, but there was no significant decrease in the fiber width as the number of passes increased from 6 to 8.

Fig. 1 AFM images depicting the morphologies of various TEMPO-oxidized NFC obtained by passing the fibers through a homogenizer for (a) 2, (b) 4, (c) 6, and (d) 8 passes. Scale bar is 200 nm.

In addition to the fibrous morphology, the water retention value (WRV) and crystallinities of the NFC (carboxyl content: 1.157 mmol g⁻¹) with varying numbers of passes are displayed in Fig. 2a. WRV is a useful parameter to investigate the interaction between the hydroxyl groups of cellulose materials and water that plays an important role during the physical and chemical treatment and isolation of cellulose.^42,43 Herein the WRV was used to evaluate the ability of an NFC suspension to retain water. The higher the WRV of the NFC dispersion, the better its water retention capacity. Overall, the crystallinity of NFC increased somewhat with increasing number of cycles. Nevertheless, the WRV decreased sharply until the 6th cycle, indicating the damage of amorphous region of NFC due to the strong mechanical integration.

Fig. 2 (a) WRV and crystallinity of NFC suspensions as a function of treatment times. (b) Shearing dilatant behavior of 1 wt% TEMPO-oxidized NFC with varying numbers of passes at the shear rates ranging from 3 × 10⁵ to 19 × 10⁵ s⁻¹.

A number of works have reported extensively on the rheology of NFC suspensions,^44–46 but their rheology at ultra-high shear rates up to 1 800 000 s⁻¹ has not been investigated. An automated ultra-high shear viscometer was utilized to measure the rheological behavior of various NFC dispersions. Fig. 2b illustrates the rheological behavior of four types of 1 wt% NFC dispersions with varying numbers of passes at shear rates between 3 × 10⁵ and 19 × 10⁵ s⁻¹. All samples exhibit shearing rheopectic behavior with increasing shear rate, but their apparent viscosity was lower than 8 Pa s. As noted from the graph, the apparent viscosity of the NFC suspension is proportional to the number of cycles through the homogenizer as they increased from 2 to 6, and the viscosity curves for NFC suspensions with 6 and 8 passes nearly overlap due to their similar fiber morphologies.

Amphiphilic characteristics of cellulose

Fig. 3a shows the chemical structure of a cellulose chain that includes hydrophilic hydroxyls and hydrophobic C–H moieties, which results in the amphiphilic character of cellulose.³⁴ The location of hydrogen atoms of the C–H bonds on the axial positions of the glucopyranose ring makes the axial orientation of the ring hydrophobic (see Fig. 3b). On the other hand, three hydroxyls on the glucose are located on the equatorial positions of the ring, leading to hydrophilicity in the equatorial direction of the ring (Fig. 3c). Intermolecular hydrogen bonds and van der Waals bonds between oxygens and hydroxyls of neighboring cellulose molecules lead to parallel stacking of cellulose chains that form cellulose nanofibers (CNs) during biosynthesis.²⁰ Mazeau demonstrated that there were three different families of surfaces in the external morphology of cellulose nanofibers.³¹ Two of the families with different surface roughnesses displayed hydrophilicity due to the exposure of the hydroxyls in the external morphology of the CNs. The remaining family exhibited hydrophobic characteristics because of the exposure of the C–H moieties. Hence, CNs are an amphiphilic material that has the potential to act as a dispersant in clay suspensions.

Fig. 3 (a) Chemical structure of cellulose molecule. (b) End view and (c) front view of 3D glucopyranose ring plane. Note that C, H, O, and –OH are shown in black, aqua, indigo, and orange, respectively. The red dash circle denotes the hydrophobic C–H moiety, whereas the green dash circle represents the hydrophilic –OH moiety.

Dispersion of clay using NFC as a dispersant

A clay suspension without agglomeration is extremely desirable because of its characteristics in cutting-edge applications.^10,11 NFC is one type of CNs directly extracted from wood fibers by pretreatments and mechanical disintegration, thereby exhibiting amphiphilic characteristics.^31,35 Here, we used amphiphilic NFC as a natural dispersant to prepare a well-dispersed and stable clay suspension (sample 3), and its dispersing effect was evaluated in terms of particle-size distribution and corresponding maximum solids content. The original particle morphology of the clay was measured by AFM (Fig. 4a and b), which showed particle sizes of <10 μm. Two control samples were also prepared. One was the clay suspension without dispersant (sample 1), the other was the clay suspension with sodium polyacrylate as the dispersing agent, which is widely used in the dispersion of clay particles in water (sample 2). The maximum concentrations and viscosities of the various clay suspensions are listed in Table 2. Note that the maximum solids content is denoted herein as the critical concentration of clay suspension in water when the size of all clay particles is <10 μm.

Fig. 4 (a) and (b) AFM images of well-dispersed clay particles. (c) Particle-size distribution of clay suspension without dispersant as a function of solids content.

Table 2 Maximum solids content and corresponding viscosity of various clay suspensions

Sample	Dispersant	Maximum solids content (%)	Viscosity (cP)
1	No	28.3	1433
2	Sodium polyacrylate	34.5	9
3	NFC3	34.0	10

---

The maximum solids content of sample 1 was approximately 28.3% and its viscosity was approximately 1433 cP. As shown in Fig. 4c, there are two bumps in the particle-size distribution curve for the clay suspension with a consistency of 28.3%, and the diameter of the clay particles was <10 μm. As the solids content increased to 29.2%, an extra bump occurs in the range of 30 to 150 μm, indicating the occurrence of undispersed aggregates. Sample 2 is the clay suspension with sodium polyacrylate as the dispersant. This sample had a maximum solids content of 34.5% and a viscosity of 9 cP. Fig. S1a† shows the particle-size distribution curves of sodium polyacrylate-dispersed clay suspensions with different solids content. When the solids content was ≤34.5%, overlapping particle-size distribution curves are observed due to the well-dispersed clay particles. As the solids content reaches 35%, large aggregates appeared in the clay suspension. Sample 3 is the clay suspension with NFC dispersant (carboxyl group content: 1.157 mmol g⁻¹, cycles: 2), which shows a similar maximum solids content (34%) and viscosity (10 cP) as sample 2. Because the solids content is over 34.0%, individual clay particles tend to aggregate after with the removal of strongly mechanical shear forces (Fig. S1b†). Note that the volumetric peak of 34% NFC3 dispersed clay suspension at ∼1 μm is much higher than that of other NFC3 dispersed clay suspensions at ∼1 μm. This may contribute to the synergistic effect of charged carboxyl groups and steric hindrance of NFC.

The particle-size distribution curves of the 28.3% clay suspension without dispersant, the 34.5% sodium polyacrylate-dispersed clay suspension, and 34.0% NFC-dispersed clay suspension are demonstrated in Fig. 5a. The samples showed average particle sizes of 2.03, 1.89, and 1.78 μm, respectively. In comparison to sample 1 without dispersant, the clay suspensions with dispersants exhibited higher maximum solids contents, lower viscosities, and smaller particle diameters. The samples with dispersants exhibited overlapping particle-size distribution curves, revealing the excellent dispersing ability of the NFC dispersant. Also, the NFC-dispersed clay exhibited a higher WRV (74 g m⁻²) than the sodium polyacrylate-dispersed suspension (58 g m⁻²) due to the extreme hygroscopicity of NFC.

Fig. 5 (a) Particle-size distribution of 28.3% clay suspension without dispersant, 34.0% clay suspension with NFC dispersant, and 34.5% clay suspension with sodium polyacrylate dispersant. (b) Rheological behavior of clay suspension using NFC and sodium polyacrylate as dispersants. (c) Digital photographs of various clay suspensions after sitting for 2 months. From left to right: 28.3% clay suspension without dispersant, 34.5% sodium polyacrylate-dispersed clay suspension, and NFC-dispersed clay suspension (carboxyl group content: 1.157 mmol g⁻¹, cycles: 2).

The rheological properties of clay suspensions with different dispersants at high shear rates, which is important for coating applications, were also studied (Fig. 5b). As the shear rate increased from 3 × 10⁵ to 19 × 10⁵ s⁻¹, the viscosities of both clay suspensions increased from 4.5 to 18 cP. The apparent viscosity of the clay suspension with the NFC dispersant was slightly lower than that of the clay suspension with sodium polyacrylate as the dispersant. Additionally, we investigated the stability of clay suspensions without and with dispersants. The three samples showed sedimentary phenomena after standing still for 3 months, but the NFC-dispersed clay suspension had less supernatant than the other clay suspensions (Fig. 5c), revealing that the NFC dispersant not only assists in dispersing clay particles, it also prevents the individual clay particles from agglomerating due to electrostatic repulsion and steric hindrance of the charged NFC.

Influence of carboxyl content of NFC on its dispersing ability

The carboxyl content of NFC could be adjusted by altering the dosage of sodium hypochlorite during TEMPO oxidation, which contributes significantly to the static repulsion between neighboring clay particles. In this study, two NFC dispersions with different carboxyl contents were used as dispersing agents in clay suspension. Their respective characteristics are presented in Table S2.† Fig. 6a shows the particle-size distribution curve for the NFC3-dispersed clay suspension (sample 3) and the NFC4-dispersed clay suspension (sample 4). In comparison to sample 3, sample 4 exhibits a lower viscosity, a smaller average particle size, and a higher water retention value (Table 3). Their rheological properties are shown in Fig. 6b. Sample 4 has a lower viscosity than sample 3 in the shear rate range between 3.5 × 10⁵ and 11 × 10⁵. With increasing shearing rate, their viscosity curves tend to overlap. In addition, the stabilities of the two samples were evaluated by extracting supernatant from both samples after standing still for 1 month. As we can see from Fig. 6c, the supernatant on the left is more turbid than the supernatant on the right, which is attributed to the higher carboxyl content of NFC, which increases the electrostatic repulsion between clay particles. Hence, the increased carboxyl content of NFC has a positive effect on the particle-size distribution and stability of a clay suspension while showing a negligible impact on rheological properties.

Fig. 6 (a) Particle-size distribution of clay suspension using NFC with different carboxyl contents as dispersants and (b) their corresponding rheological behavior at high shear rate. (c) Supernatant extracted from the NFC3-dispersed clay suspension (left) and NFC4-dispersed clay suspension (right) after standing still for 1 month. Note: NFC3 represents a carboxyl content of 1.16 mmol g⁻¹ and NFC4 denotes a carboxyl content of 0.88 mmol g⁻¹.

Table 3 Viscosity, WRV, and average particle size of NFC-dispersed clay suspensions

Samples	Viscosity (cP)	WRV (g m^−2)	Average particle size (μm)	Dispersant
3	10	73.95	1.78	NFC3
4	65	38.62	1.94	NFC4

---

Conclusion

In summary, carboxylated NFC prepared by the TEMPO oxidation method was used as a dispersant to prepare a well-dispersed aqueous clay suspension with desirable stability and fluidity. In comparison to the sample without a dispersant, the clay dispersion with NFC dispersant demonstrated a higher solids content and a much lower static viscosity. Moreover, a well-dispersed clay suspension with carboxylated NFC dispersant demonstrated a similar maximum solids concentration, particle-size distribution, and rheological performance. However, it showed a higher stability and water retention value as compared to sodium polyacrylate. The increased carboxyl content of NFC has a positive impact on the average particle size and stability of clay suspensions. Carboxylated NFC thus has the potential to act as a green and effective dispersant to prepare well-dispersed clay suspensions with desirable fluidity and stability for high-tech applications.

Acknowledgements

Siyi Ming acknowledges the support from National Training Program of Innovation and Entrepreneurship for Undergraduates (201310561001) and 2016 Guangdong "Climbing" Program for Undergraduate. Zhiqiang Fang would like to acknowledge the financial support from China Postdoctoral Science Foundation (2015M570716), the Fundamental Research Funds for the Central Universities (2015ZM156).

References

1. A. C. Balazs, T. Emrick and T. P. Russell, Science, 2006, 314, 1107–1110.
2. Y. Dou, S. Xu, X. Liu, J. Han, H. Yan, M. Wei, D. G. Evans and X. Duan, Adv. Funct. Mater., 2014, 24, 514–521.
3. M. A. Priolo, D. Gamboa, K. M. Holder and J. C. Grunlan, Nano Lett., 2010, 10, 4970–4974.
4. A. Liu, A. Walther, O. Ikkala, L. Belova and L. A. Berglund, Biomacromolecules, 2011, 12, 633–641.
5. C. W. Chiu, T. K. Huang, Y. C. Wang, B. G. Alamani and J. J. Lin, Prog. Polym. Sci., 2014, 39, 443–485.
6. P. Podsiadlo, A. K. Kaushik, E. M. Arruda, A. M. Waas, B. S. Shim, J. Xu, H. Nandivada, B. G. Pumplin, J. Lahann and A. Ramamoorthy, Science, 2007, 318, 80–83.
7. M. W. Möller, T. Lunkenbein, H. Kalo, M. Schieder, D. A. Kunz and J. Breu, Adv. Mater., 2010, 22, 5245–5249.
8. J. Dong, A. J. Worthen, L. M. Foster, Y. Chen, K. A. Cornell, S. L. Bryant, T. M. Truskett, C. W. Bielawski and K. P. Johnston, ACS Appl. Mater. Interfaces, 2014, 6, 11502–11513.
9. O. Owoseni, Y. Zhang, Y. Su, J. He, G. L. McPherson, A. Bose and V. T. John, Langmuir, 2015, 31, 13700–13707.
10. K. K. R. Datta, A. Achari and M. Eswaramoorthy, J. Mater. Chem. A, 2013, 1, 6707–6718.
11. Y. A. Dzenis, Structural nanocomposites, 2008, vol. 319, pp. 419–420.
12. F. Morsy, S. El-Sherbiny, M. Hassan and H. Mohammed, Powder Technol., 2014, 264, 430–438.
13. M. Loginov, O. Larue, N. Lebovka and E. Vorobiev, Colloids Surf., A, 2008, 325, 64–71.
14. A. Papo, L. Piani and R. Ricceri, Colloids Surf., A, 2002, 201, 219–230.
15. D. Penner and G. Lagaly, Appl. Clay Sci., 2001, 19, 131–142.
16. F. Demir, Appl. Clay Sci., 2015, 105, 41–47.
17. A. J. Patil, M. Li, E. Dujardin and S. Mann, Nano Lett., 2007, 7, 2660–2665.
18. S. J. Eichhorn, A. Dufresne, M. Aranguren, N. E. Marcovich, J. R. Capadona, S. J. Rowan, C. Weder, W. Thielemans, M. Roman, S. Renneckar, W. Gindl, S. Veigel, J. Keckes, H. Yano, K. Abe, M. Nogi, A. N. Nakagaito, A. Mangalam, J. Simonsen, A. S. Benight, A. Bismarck, L. A. Berglund and T. Peijs, J. Mater. Sci., 2010, 45, 1–33.
19. D. Klemm, F. Kramer, S. Moritz, T. Lindström, M. Ankerfors, D. Gray and A. Dorris, Angew. Chem., Int. Ed., 2011, 50, 5438–5466.
20. R. J. Moon, A. Martini, J. Nairn, J. Simonsen and J. Youngblood, Chem. Soc. Rev., 2011, 40, 3941–3994.
21. Y. Zhang, T. Nypelö, C. Salas, J. Arboleda, I. C. Hoeger and O. J. Rojas, J. Renewable Mater., 2013, 1, 195–211.
22. Y. Zhang, R. G. Carbonell and O. J. Rojas, Biomacromolecules, 2013, 14, 4161–4168.
23. K. Junka, J. Guo, I. Filpponen, J. Laine and O. J. Rojas, Biomacromolecules, 2014, 15, 876–881.
24. T. Inui, H. Koga, M. Nogi, N. Komoda and K. Suganuma, Adv. Mater., 2015, 27, 1112–1116.
25. Z. Fang, H. Zhu, C. Preston, X. Han, Y. Li, S. Lee, X. Chai, G. Chen and L. Hu, J. Mater. Chem. C, 2013, 1, 6191–6197.
26. L. Hu, G. Zheng, J. Yao, N. Liu, B. Weil, M. Eskilsson, E. Karabulut, Z. C. Ruan, S. H. Fan, J. T. Bloking, M. D. McGehee, L. Wagberg and Y. Cui, Energy Environ. Sci., 2013, 6, 513–518.
27. H. Zhu, Z. Xiao, D. Liu, Y. Li, N. Weadock, Z. Fang, J. Huang and L. Hu, Energy Environ. Sci., 2013, 6, 2105–2111.
28. Y. Li, H. Zhu, H. Gu, H. Dai, Z. Fang, N. Weadock, Z. Guo and L. Hu, J. Mater. Chem. A, 2013, 1, 15278–15283.
29. Y. Jung, T. Chang, H. Zhang, C. Yao, Q. Zheng, V. W. Yang, H. Mi, M. Kim, S. J. Cho and D. W. Park, Nat. Commun., 2015, 6, 7170.
30. H. Ma, C. Burger, B. S. Hsiao and B. Chu, J. Mater. Chem., 2011, 21, 7507–7510.
31. K. Mazeau, Carbohydr. Polym., 2011, 84, 524–532.
32. L.-S. Johansson, T. Tammelin, J. M. Campbell, H. Setälä and M. Österberg, Soft Matter, 2011, 7, 10917–10924.
33. M. M. Hamedi, A. Hajian, A. B. Fall, K. Håkansson, M. Salajkova, F. Lundell, L. Wågberg and L. A. Berglund, ACS Nano, 2014, 8, 2467–2476.
34. C. Olivier, C. l. Moreau, P. Bertoncini, H. Bizot, O. Chauvet and B. Cathala, Langmuir, 2012, 28, 12463–12471.
35. Y. Li, H. Zhu, F. Shen, J. Wan, S. Lacey, Z. Fang, H. Dai and L. Hu, Nano Energy, 2015, 13, 346–354.
36. J. M. Malho, P. Laaksonen, A. Walther, O. Ikkala and M. B. Linder, Biomacromolecules, 2012, 13, 1093–1099.
37. I. Kalashnikova, H. Bizot, B. Cathala and I. Capron, Langmuir, 2011, 27, 7471–7479.
38. M. Andresen and P. Stenius, J. Dispersion Sci. Technol., 2007, 28, 837–844.
39. I. Besbes, S. Alila and S. Boufi, Carbohydr. Polym., 2011, 84, 975–983.
40. X. Zhu, C. Yan and J. Chen, Appl. Clay Sci., 2012, 55, 114–119.
41. Y. Okita, T. Saito and A. Isogai, Biomacromolecules, 2010, 11, 1696–1700.
42. D. Klemm, B. Philip, T. Heinze, U. Heinze and W. Wagenknecht, Comprehensive Cellulose Chemistry-Fundamentals and analytical methods, Wiley-VCH, Weinheim, 1998.
43. Q. Cheng, J. Wang, J. McNeel and P. Jacobson, BioResources, 2010, 5, 1945–1954.
44. M. Mohtaschemi, A. Sorvari, A. Puisto, M. Nuopponen, J. Seppälä and M. J. Alava, Cellulose, 2014, 21, 3913–3925.
45. M. Mohtaschemi, K. Dimic-Misic, A. Puisto, M. Korhonen, T. Maloney, J. Paltakari and M. J. Alava, Cellulose, 2014, 21, 1305–1312.
46. A. Naderi, T. Lindström and J. Sundström, Cellulose, 2014, 21, 1561–1571.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra03935a

---