Letter www.acsami.org

Individually Dispersed Wood-Based Cellulose Nanocrystals Huibin Chang,†,‡ Jeffrey Luo,†,‡ Amir A. Bakhtiary Davijani,† An-Ting Chien,† Po-Hsiang Wang,† H. Clive Liu,†,‡ and Satish Kumar*,†,‡ †

School of Materials Science and Engineering and ‡Renewable Bioproducts Institute, Georgia Institute of Technology, Atlanta, Georgia 30332, United States S Supporting Information *

ABSTRACT: Good dispersion of cellulose nanocrystals (CNCs) in the polymer matrix is one of the key factors to obtaining good properties in the resulting nanocomposites. However, the preparation of individually dispersed CNCs in solvents or in polymer matrices has been a challenge. In this study, individually dispersed wood-based CNCs have been successfully prepared in solvents, including dimethylformamide (DMF), H2O, and a mixture of H2O/DMF, by sonication of moisture-containing CNCs. The CNCs dispersions were characterized by dynamic light scattering (DLS). It is found that CNCs containing above about 3.8 wt % moisture is critical for achieving individually dispersed CNC in solvents. Hydrodynamic radius (Rh) of CNCs is smaller in H2O/DMF cosolvent mixture than that in pure DMF or in pure H2O under same sonication treatment conditions. Experimental results have been corroborated using molecular simulation study. KEYWORDS: cellulose nanocrystals, hydrodynamic radius, co-solvent, dispersibility


calculated as follows, which can be used to assess the dispersibility of CNCs in solvents. The hydrodynamic radius, Rh, is defined as the size of a diffusion-equivalent sphere according to the Stokes−Einstein relation

ellulose nanocrystals (CNCs) have attracted significant interest as reinforcement materials in the nanocomposite field because of their nanoscale dimensions, high aspect ratio, high mechanical properties, biorenewability, and sustainability.1 Individual CNCs exhibit tensile strength up to 7.5 GPa and tensile modulus in the range of 110−220 GPa.2−4 Despite these high mechanical properties of CNCs, it is difficult to obtain significant property enhancement in nanocomposites. This is especially true in a hydrophobic polymer matrix system, because of the poor dispersibility of CNCs in organic solvents.5 In polymer composites, good dispersion of fillers in the polymer matrix is one of the key factors for obtaining good mechanical properties.6 Even under sonication, it is hard to obtain individual CNCs in organic solvents without surface modification.5,7 Different stabilizing methods such as surfactants,8 silylating agents,9 grafting of polyethylene glycol (PEG),10 grafting of maleated polypropylene,11 and acylation12 have been explored to improve CNC dispersibility in organic solvents. However, the surface modification may have negative impact on mechanical properties of the composite.13 It is reported that a small amount of water is critical to disperse CNC in organic solvents such as DMF and DMSO.5,14 However, there is no systematic study to investigate the effect of water on the dispersibility of CNCs in organic solvents. In this work, the effect of water on the dispersibility of CNCs in DMF is systematically investigated. CNCs dispersions are systematically characterized by dynamic light scattering (DLS). In this study, hydrodynamic radius (Rh) of individual CNC is © 2016 American Chemical Society

Rh =

KT 6πηDt


Where K is Boltzmann’s constant, T is the absolute temperature, η is the viscosity, and Dt is the translational diffusion coefficient. For rod-shaped macromolecules with a given length (L) and diameter (d) in dilute solution, the translational diffusion coefficient (Dt) can be approximated by the Kirkwood and Riseman results15

Dt =

KT ⎛⎜ L ⎞⎟ ln 3πηL ⎝ d ⎠


Combining eqs 1 and (2), results in Rh =


( Ld )



Received: January 5, 2016 Accepted: February 22, 2016 Published: February 22, 2016 5768

DOI: 10.1021/acsami.6b00094 ACS Appl. Mater. Interfaces 2016, 8, 5768−5771


ACS Applied Materials & Interfaces

moisture, Figure S2) were dispersed at concentration of 75 mg/ 100 mL in pure DMF, and in 50−50 volume percent H2O/ DMF mixture, and these dispersions were still cloudy even after bath sonication for 1 week, as CNCs were not dispersible (Figure S3). For 3.8 wt % moisture-containing CNCs, hydrodynamic radius of about 22 nm was obtained in a broad solvent concentration range of 5 to 75% water in DMF (Figure 2). After 6 h sonication, hydrodynamic radius (Rh) of

Wood-based CNCs used in this study are reported to have a diameter of 5 nm and length is in the range of 150−200 nm (Figures S1 and S2, Table S3). Using eq 3, the hydrodynamic radius of the CNCs is calculated to be between 22 and 27 nm. If the CNCs are dispersed as individual CNCs in a solvent, then the experimental measurements should result in hydrodynamic radius values in this range. In one set of experiments, CNCs with 2.6, 3.2, and 3.8 wt % moisture were dispersed in pure DMF by bath sonication for different times (moisture is determined by TGA as shown in Figure S2). The results show that hydrodynamic radius (Rh) of CNCs decreased with increasing sonication time (Figure 1).

Figure 2. Hydrodynamic radius of CNCs in H2O/DMF mixture. In addition to the water value shown in the solvent mixture on the x-axis, CNCs also contain 3.8 wt % moisture. Sonication times are indicated in the figure. CNC concentration was 75 mg/100 mL.

CNCs (3.8 wt % moisture-containing) in pure DMF and pure H2O are 50 and 60 nm, respectively, which are significantly higher than the calculated Rh of individual CNCs. However, for 5 and 25 volume percent H2O in DMF, predominantly individually dispersed CNCs were achieved within 6 h of sonication, where R h is below 27 nm. CNCs were predominantly individually dispersed in all co-solvents (when water concentration was in the range of 5 to 75 vol %) within 12 h of sonication. For pure DMF or pure water, even after 18 h of sonication, it does not reach to individual CNC status (Figure 2). We also investigated the dispersibility at relatively high CNC concentration of 3g/100 mL (CNCs contained 3.8 wt % water) (Figure 3). For 25−75 volume percent H2O/DMF mixture, mostly individually dispersed CNCs could be achieved in 24 h of bath sonication. However, at this CNC concentration, in pure water, sonication had little effect on the Rh of CNC, when sonication time increased from 24 to 66 h. Rh of CNC in pure water is about 3 times larger than that for 25 and 50 volume percent H2O co-solvents after 66 h sonication. In a subsequent experiment, different CNC concentrations were dispersed in 25−75 and 50−50 volume H2O/DMF co-solvent as well as in pure water (Figure. 3b). After 24 h of bath sonication, individually dispersed CNCs were achieved in H2O/DMF cosolvents at all CNC concentrations (75 mg to 3 g/100 mL). On the other hand, in pure water, Rh of CNC is slightly larger than that in co-solvents at low CNC concentration, and it was significantly higher at high CNC concentrations. This suggests that CNCs were not individually dispersed in water, and that they can be individually dispersed very effectively in H2O/DMF co-solvent at high concentration and for relatively low sonication duration. TEM images obtained from extremely dilute dispersions are included in Figure S4, and the average

Figure 1. Hydrodynamic radius of CNC with different moisture contents dispersed in DMF for different sonication time: (a) 2.6 wt % moisture; (b) 3.2 wt % moisture; (c) 3.8 wt % moisture. CNC concentration was 75 mg/100 mL.

For low moisture-containing (≤3.2 wt %) CNCs, even after 7 days of sonication, Rh is still higher (38 and 31 nm for 2.6 and 3.2 wt % moisture containing CNCs, respectively) than the calculated value (22 to 27 nm). At moisture content of 3.8 wt %, hydrodynamic radius of 23 nm was achieved after 24 h sonication. This is consistent with the previous study, where it was reported that a moisture content of 4 wt % in CNC was necessary to fully disperse them in water.16 The binary mixture of solvents is gaining more interest because of combined properties from each solvent. Because the above data show that moisture significantly affects the dispersibility of CNCs in DMF, CNC dispersibility was explored in broad range of H2O/DMF mixtures. Also, to further confirm the importance of moisture for CNC dispersion, CNC moisture was removed by heating in oven at 105 °C for 48 h under vacuum. Dried CNCs (0.6 wt % 5769

DOI: 10.1021/acsami.6b00094 ACS Appl. Mater. Interfaces 2016, 8, 5768−5771


ACS Applied Materials & Interfaces

Figure 4. CNC repeat unit (black) dispersion in solvents: (a) pure DMF (green); (b) pure H2O (red); (c) 25−75 H2O/DMF co-solvent.

Figure 3. Hydrodynamic radius of 3.8 wt % moisture-containing CNCs in solvents (a) under different sonication time (concentration 3 g/100 mL); (b) different concentrations after 24 h sonication.

CNC dimensions based on the TEM images are given in Table S3. Molecular simulation study was done to understand the CNC dispersion (Videos S1, S2, and S3). For this purpose cellulose repeat units were dispersed in 100% DMF, 100% water, and in 25−75 volume H2O/DMF mixture. In pure DMF, CNC repeat units aggregate in the center, while in pure water CNC repeat units are mostly on the outside (Figure 4a, b). On the other hand, in H2O/DMF mixture, CNC repeat units were found to be individually dispersed (Figure 4c). These are consistent with our experimental results. Hydrogen bond (H-bond) is formed through hydroxyl groups in CNCs, resulting in CNC aggregates. Water molecules interact with cellulose to form H-bond, and thus results in reduced number of intermolecular hydrogen bonds in cellulose.17 In moisture-containing CNCs, cellulose-water Hbond forms on the CNC surface, and thus weakens the interaction between two CNCs, thus moisture containing CNCs result in better dispersion than CNCs without moisture. Water−water hydrogen bonds have lower binding energy than cellulose-water hydrogen bonds.16,18 Also, because of the strong interaction between water and amide, it is easy to form H-bond between the DMF and water. On the CNC surface, H-bond forms between DMF and H2O. DMF surrounded individual CNC can effectively prevent the H-bond formation between CNCs. The proposed mechanism is shown in Figure 5. WaterDMF molecules have strong interaction. This is well-known and is verified by FTIR spectroscopy (Figure S5). H2O/DMF molecules diffuse in between cellulose nanocrystals in the H2O/ DMF solvent mixture. Therefore, under the same sonication conditions, it is easier to separate CNCs to the individual CNC

Figure 5. Proposed mechanism to disperse CNCs in H2O/DMF cosolvent (black bars, CNC; green circles, DMF; red circles, H2O).

in H2O/DMF co-solvent when compared to 100% H2O or 100% DMF. In conclusion, individually dispersed CNCs in solvents including pure DMF, pure water, and mixture of H2O/DMF have been successfully prepared. The critical moisture (≥3.8 wt %) in CNC is required to disperse individual CNCs. Compared with pure solvents such as DMF and H2O, H2O/DMF cosolvent can more effectively disperse individual CNCs. This cosolvent system will help in achieving better dispersions of CNCs in both hydrophilic and hydrophobic polymer matrices.


S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.6b00094. Experimental methods, simulation methods, and additional figures including TEM images relating to the dispersed cellulose nanocrystals (PDF) Video S1, CNC repeat unit in pure DMF (AVI) Video S2, CNC repeat unit in water (AVI) 5770

DOI: 10.1021/acsami.6b00094 ACS Appl. Mater. Interfaces 2016, 8, 5768−5771


ACS Applied Materials & Interfaces

(16) Beck, S.; Bouchard, J.; Berry, R. Dispersibility in Water of Dried Nanocrystalline Cellulose. Biomacromolecules 2012, 13 (5), 1486− 1494. (17) Khazraji, A. C.; Robert, S. Interaction Effects Between Cellulose and Water in Nanocrystalline and Amorphous Regions: A Novel Approach Using Molecular Modeling. J. Nanomater. 2013, 2013, 44. (18) Fengel, D. Wegener, G. Wood: Chemistry, Ultrastructure, Reactions; Walter de Gruyter: New York, 1989; pp 77−81.

Video S3, CNC repeat unit in 25−75 H2O−DMF cosolvent (AVI)


Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

ACKNOWLEDGMENTS Financial support for this work from Renewable Bioproducts Institute at the Georgia Institute of Technology and a grant from the Air Force Office of Scientific Research (FA 9550-14-10194) are gratefully acknowledged.


(1) Moon, R. J.; Martini, A.; Nairn, J.; Simonsen, J.; Youngblood, J. Cellulose Nanomaterials Review: Structure, Properties and Nanocomposites. Chem. Soc. Rev. 2011, 40 (7), 3941−3994. (2) Azizi Samir, M. A. S.; Alloin, F.; Dufresne, A. Review of Recent Research into Cellulosic Whiskers, Their Properties and Their Application in Nanocomposite Field. Biomacromolecules 2005, 6 (2), 612−626. (3) Dri, F. L.; Hector, L. G., Jr; Moon, R. J.; Zavattieri, P. D. Anisotropy of the Elastic Properties of Crystalline Cellulose Iβ from First Principles Density Functional Theory with Van der Waals Interactions. Cellulose 2013, 20 (6), 2703−2718. (4) Diddens, I.; Murphy, B.; Krisch, M.; Müller, M. Anisotropic Elastic Properties of Cellulose Measured Using Inelastic X-ray Scattering. Macromolecules 2008, 41 (24), 9755−9759. (5) Viet, D.; Beck-Candanedo, S.; Gray, D. G. Dispersion of Cellulose Nanocrystals in Polar Organic Solvents. Cellulose 2007, 14 (2), 109−113. (6) Lu, J.; Askeland, P.; Drzal, L. T. Surface Modification of Microfibrillated Cellulose for Epoxy Composite Applications. Polymer 2008, 49 (5), 1285−1296. (7) van den Berg, O.; Capadona, J. R.; Weder, C. Preparation of Homogeneous Dispersions of Tunicate Cellulose Whiskers in Organic Solvents. Biomacromolecules 2007, 8 (4), 1353−1357. (8) Heux, L.; Chauve, G.; Bonini, C. Nonflocculating and ChiralNematic Self-Ordering of Cellulose Microcrystals Suspensions in Nonpolar Solvents. Langmuir 2000, 16 (21), 8210−8212. (9) Goussé, C.; Chanzy, H.; Excoffier, G.; Soubeyrand, L.; Fleury, E. Stable Suspensions of Partially Silylated Cellulose Whiskers Dispersed in Organic Solvents. Polymer 2002, 43 (9), 2645−2651. (10) Araki, J.; Wada, M.; Kuga, S. Steric Stabilization of A Cellulose Microcrystal Suspension by Poly (Ethylene Glycol) Grafting. Langmuir 2001, 17 (1), 21−27. (11) Ljungberg, N.; Bonini, C.; Bortolussi, F.; Boisson, C.; Heux, L.; Cavaillé, J.-Y. New Nanocomposite Materials Reinforced with Cellulose Whiskers in Atactic Polypropylene: Effect of Surface and Dispersion Characteristics. Biomacromolecules 2005, 6 (5), 2732− 2739. (12) Yuan, H.; Nishiyama, Y.; Wada, M.; Kuga, S. Surface Acylation of Cellulose Whiskers by Drying Aqueous Emulsion. Biomacromolecules 2006, 7 (3), 696−700. (13) Grunert, M.; Winter, W. T. Nanocomposites of Cellulose Acetate Butyrate Reinforced with Cellulose Nanocrystals. J. Polym. Environ. 2002, 10 (1−2), 27−30. (14) Azizi Samir, M. A. S.; Alloin, F.; Sanchez, J.-Y.; El Kissi, N.; Dufresne, A. Preparation of Cellulose Whiskers Reinforced Nanocomposites From An Organic Medium Suspension. Macromolecules 2004, 37 (4), 1386−1393. (15) Riseman, J.; Kirkwood, J. G. The Intrinsic Viscosity, Translational and Rotatory Diffusion Constants of Rodlike Macromolecules in Solution. J. Chem. Phys. 1950, 18 (4), 512−516. 5771

DOI: 10.1021/acsami.6b00094 ACS Appl. Mater. Interfaces 2016, 8, 5768−5771

Individually Dispersed Wood-Based Cellulose Nanocrystals.

Good dispersion of cellulose nanocrystals (CNCs) in the polymer matrix is one of the key factors to obtaining good properties in the resulting nanocom...
1MB Sizes 1 Downloads 4 Views