Hydrophobic cellulose nanocrystals modified with quaternary ammonium salts

Abstract

An environmentally friendly procedure in aqueous solution for the surface modification of cellulose nanocrystals (CNCs) using quaternary ammonium salts via adsorption is developed as inspired by organomodified layered silicates. CNCs with a high carboxylate content of 1.5 mmol g⁻¹ were prepared by a new route, direct hydrochloric acid hydrolysis of 2,2,6,6-tetramethylpiperidine-1-oxyl radical (TEMPO)-oxidized nanofibrillated cellulose from a softwood pulp, and characterized by atomic force microscopy (AFM) and X-ray diffraction (XRD). Four quaternary ammonium cation surfactants bearing long alkyl, phenyl, glycidyl, and diallyl groups were successfully used to modify CNCs carrying carboxylic acid groups as characterized by Fourier transform infrared spectroscopy (FTIR). The modified CNCs can be redispersed and individualized in an organic solvent such as toluene as observed by scanning transmission electron microscopy (STEM). One may envision removing excess surfactant to obtain CNC with a monolayer of surfactant. The toluene suspension of the modified CNCs showed strong birefringence under crossed polars but no further chiral-nematic ordering was observed. The model surface prepared by the CNCs modified with quaternary ammonium salts bearing C18 alkyl chains showed a significant increase in water contact angle (71°) compared to that of unmodified CNCs (12°). This new series of modified CNCs can be dried from solvent and have the potential to form well-dispersed nanocomposites with non-polar polymers.

---

Introduction

A critical challenge in the fabrication of nanostructured composite materials is to make nanoscale building blocks with a particular surface structure, charge, and functionality. Surface properties will control the interactions between the nanoscale building blocks as well as with their matrix, which ultimately dictates the structure and macroscale properties of the bulk composite material. Cellulose nanocrystals (CNCs), which have also been named as cellulose nanowhiskers or nanocrystalline cellulose, are rod-like or whisker shaped particles with lateral dimensions three orders of magnitude smaller than plant cellulosic fibers. CNCs are ideal nanoscale building blocks due to their lightweight, high aspect ratio, large specific surface area, reactive surface of –OH groups, high mechanical strength and stiffness, and ability to form superstructures.^1–4 In order to improve the performance of CNC-based composite materials, there has been a strong desire to develop an environmentally friendly procedure for tailoring the surface functionality of CNCs.

CNCs are generally prepared by sulfuric acid hydrolysis of native cellulose fibers, resulting in the formation of sulfate esters on the surface,^5,6 and further reaction with epoxypropyltrimethylammonium chloride will render the surface cationic.⁷ Alternatively, CNCs with surface carboxylic acid groups have also been prepared by the combination of hydrochloric acid or hydrobromic acid hydrolysis and oxidative carboxylation (2,2,6,6-tetramethylpiperidine 1-oxyl radical (TEMPO)-mediated oxidation).^8–11 Subsequent grafting of amine-terminated poly(ethylene glycol) to carboxylated cellulose microcrystals improved the stability of the microcrystal suspensions through steric stabilization.⁸ These CNCs, with either surface charges or grafted pedant polymers, form stable colloidal suspensions in water and can be well-dispersed in water soluble polymer matrices^1,12–15 and aqueous dispersed polymers, i.e., latexes.¹⁶

In order to broaden the application scope, various covalent and non-covalent modification approaches have been developed for CNCs to make functional nanocrystalline cellulose particles or to improve their dispersion in hydrophobic polymer matrices.¹⁷ Functional groups such as initiators^18–21 for living radical polymerizations, alkyne and azide groups,⁹ epoxy and amine groups²² have been previously decorated onto the surface of CNCs, which facilitates the applications of graft polymerization, click chemistry, and fluorescence labeling on CNCs. The covalent attachments are generally achieved via the reaction of isocyanates,^23,24 epoxides,⁷ acid halides,^18,20,21 acid anhydrides,²⁵ and chlorosilanes^26–28 with the hydroxyl groups of CNCs, or via carbodiimide-mediated formation of an amide linkage between the carboxylic acid groups on CNCs and moieties carrying an amine group,^8,9,29 or direct ring opening polymerization using the hydroxyl groups on CNCs as the initiators.^30,31 These covalent approaches generally require organic solvents as reaction media, thus, a tedious solvent exchange step for CNCs or freeze drying of CNCs from very dilute aqueous suspension has to be employed. In the case of acid anhydride, fresh emulsions have to be prepared immediately before use every time due to the instability in both the dispersion state and the chemical nature.

Regarding non-covalent modification, cationic polyelectrolytes such as poly(amideamine) epichlorohydrin (PAE),³² polydimethyldiallylammonium chloride (PDMAC),³³ and poly(allylamine hydrochloride) (PAH) have been used to modify CNCs and layer-by-layer self-assembled multilayer films have been made.³⁴ The anionic surfactant, an acid phosphate ester of alkyl phenol ethoxylate, has been used to coat the surface of CNCs and effectively aid their dispersion in organic solvents, poly(propylene), and polylactic acid.^35–38 Non-ionic surfactant sorbitan monostearate has also been utilized to improve the dispersion of CNCs in polystyrene.^39,40 The cationic surfactant, cetyltrimethylammonium bromide (CTAB), has been utilized not only as a stabilizer of metallic nanoparticles but also as a vehicle for the positioning of these particles on the CNCs surface.⁴¹ In our previous work, we have developed a method for the efficient incorporation of chemical functionality onto cellulose surfaces via chemo-enzymatic modified xyloglucan, which has a naturally high affinity for cellulose.^42,43 Further, a triblock copolymer based on xyloglucan oligosaccharides has been synthesized and used for the surface modification of CNCs, resulting in chiral nematic ordered CNCs in toluene.⁴⁴

The inorganic cations of layered silicates are generally exchanged by ammonium or phosphonium cations bearing at least one long alkyl chain, and possibly other substituted groups.⁴⁵ The purpose is to facilitate the dispersion of individual silicate platelets in the polymer matrix. By modifying the layered silicate, silicate platelets become more compatible with the matrix. This influences the distribution of platelets in the final nanostructure, and consequently the properties of the nanocomposites. Previously, the adsorption of cationic surfactants (quaternary ammonium salts) with different alkyl chain lengths onto microcrystalline cellulose and TEMPO oxidized cellulose fibers as well as the coadsorption of different organic solutes has been extensively studied.^46–49 Only a slight reduction in the wettability was observed for the films prepared by adsorption of CTAB on TEMPO-oxidized cellulose nanofibrillated fibrils.⁵⁰ However, the modification of CNCs using quaternary ammonium salts and their redispersibility in organic solvent has not been previously investigated.

As inspired by organomodified layered silicates, the objective of the present study is to develop an environmentally friendly procedure in aqueous suspension for the surface modification of CNCs using quaternary ammonium salts via an ionic exchange process. In order to obtain higher surface charge density on CNCs, a new route for the preparation of CNCs bearing carboxylic acid groups is developed. Quaternary ammonium cations bearing long alkyl, phenyl, glycidyl, and diallyl groups are used to modify CNCs to render their surface hydrophobic. This will make the CNCs more compatible with polymers with hydrophobic characteristics, and facilitate CNC dispersion in such matrices. The structure of unmodified and modified CNCs and their dispersibility in organic solvent are characterized by X-ray diffraction (XRD), atomic force microscopy (AFM), Fourier transform infrared spectroscopy (FTIR), scanning transmission electron microscopy (STEM), and optical microscopy. The surface hydrophobicity of the modified CNCs is studied by contact angle measurements.

Experimental

Materials

A commercial never-dried softwood sulphite pulp provided by Nordic Paper was used for native cellulose fibers. The cellulose content was 86% and the rest was mostly hemicelluloses. TEMPO (Sigma-Aldrich), sodium bromide (Sigma-Aldrich), sodium hypochlorite (VWR), sodium hydroxide (Sigma-Aldrich), hydrochloric acid (37%, Sigma-Aldrich), toluene (Sigma-Aldrich), tetrahydrofuran (THF, Sigma-Aldrich), chloroform (Fisher Scientific), acetone (Fisher Scientific), stearyltrimethylammonium chloride (97%, Chemtronica), glycidyl trimethylammonium chloride (technical, Sigma-Aldrich), diallyldimethylammonium chloride (97%, Sigma-Aldrich), phenyltrimethylammonium chloride (97%, Sigma-Aldrich), and poly-L-lysine (0.1 w/v% aqueous solution, Ted Pella Inc.) were used without further purification.

TEMPO-mediated oxidation of wood pulp

TEMPO-mediated oxidation of wood pulp was done according to the method described by Saito et al.⁵¹ In brief, the wood pulp fibers were suspended in water containing TEMPO and sodium bromide. The TEMPO-mediated oxidation was initiated by adding the desired amount of sodium hypochlorite solution and was continued at room temperature while stirring. The pH was kept at 10 by adding sodium hydroxide until no further sodium hydroxide consumption was observed. The TEMPO-oxidized cellulose fibers were thoroughly washed with water by filtration and centrifugation. The TEMPO-oxidized cellulose fibers were then dispersed in water at a concentration of 1 wt% and disintegrated by a homogenization process, i.e. only one pass through 200 and 100 μm chambers at a pressure of 1600 bar at room temperature (21 °C) using a Microfluidizer M-110 EH (Microfluidics Ind., Newton, MA). The resulting suspension of TEMPO-oxidized nanofibrillated cellulose (TO-NFC) was stored at 4 °C before use.

Preparation of cellulose nanocrystals

Cellulose nanocrystals were prepared by hydrochloric acid hydrolysis of TO-NFC. Typically, 100 g of wet TO-NFC (corresponding to 1 g dry weight of cellulose) was dispersed in 316 mL deionized water using shear forces (13 500 rpm, Ultra Turrax, IKA, Germany) for 2 min and stirred with magnetic stirrer overnight. The suspension was placed into a three-neck round bottom flask. Hydrochloric acid was added dropwise until a final concentration of 2.5 M was obtained. The mixture was then heated up to 105 °C and refluxed for 3, 5 and 7 h, and the resulting samples were coded as CNC-3h, CNC-5h, and CNC-7h, respectively. The reaction was stopped by dilution with deionized water. The resulting suspension was centrifuged at 4166 g for 20 min. The precipitates were collected and diluted with approximately 100 mL of water, and subsequently dialyzed against water using Spectrum Spectra/Por regenerated cellulose dialysis membrane with a molecular weight cut off of 12 000–14 000 for 5–7 days. After dialysis, the suspensions were sonicated for 10 min (Sonics Vibracell, USA) to obtain individual nanocrystals. The dry content of the suspensions and the yield of the reactions were determined by gravimetric analysis. The yield of the CNCs was 85, 70, and 69% after 3, 5, and 7 hours hydrolysis reaction, respectively.

Surface modification of cellulose nanocrystals with quaternary ammonium salts

As illustrated in Fig. 1, four quaternary ammonium salts, stearyltrimethylammonium chloride (1), phenyltrimethyl-ammonium chloride (2), glycidyl trimethylammonium chloride (3), and diallyldimethylammonium chloride (4) were utilized to modify CNCs bearing carboxylic acid groups on the surface. The pH of the CNC suspension (0.1 wt%) was adjusted to 10 using NaOH aqueous solution in order to have the carboxyl groups on the surface of the nanocrystals fully dissociated.⁵² The CNC suspension was then added dropwise into the desired amount of aqueous quaternary ammonium salt solution (0.1 wt%) with stirring at 60 °C. The suspension was kept at 60 °C for 3 h, then stirred at room temperature overnight. The suspension was then dialyzed against deionized water to remove any NaCl formed during the adsorption and any unbound quaternary ammonium salts. The final suspension was freeze-dried and redispersed in toluene (1 wt%) using an ultrasonic treatment (Sonics Vibracell, USA) for 1 min. These suspensions were then centrifuged for 10 min at 20 000 g to remove any possible excess of quaternary ammonium salts that were aggregated on the surface of CNCs. The resulting pellet was easily redispersed in the desired amount of organic solvent with an ultrasonic treatment for a few tenths of a second.

Fig. 1 Reaction of cellulose nanocrystals with quaternary ammonium salts.

Characterization

The carboxylate content of the CNCs was determined by conductometric titration. Typically, 50 mg of CNCs was suspended in 50 mL of water containing 1 mM NaCl. The titration was performed using 0.0114 M solution NaOH (solution was standardized against oxalic acid) and the conductivity was monitored with a Mettler Toledo conductivity meter (USA).

The shape and size distribution of the CNCs was determined by AFM. The samples were prepared according to a previous method⁶ with minor modification. The original cellulose suspension was diluted to approximately 0.01 wt%. A 20 μL drop of 0.1% w/v solution of poly-L-lysine was placed on a piece of freshly cleaved mica for 3 min, and then rinsed with water and dried. Subsequently, a 20 μL drop of CNC suspension was allowed to stand on mica for 3 min, then rinsed with water and dried. The mica substrate was attached to a specimen holder and analyzed with tapping-mode AFM (Multimode IIIa, Veeco Instruments, Santa Barbara, CA). RTESP silica cantilevers (Veeco, tip radius 8 nm and spring constant 40 N m⁻¹) were oscillated at a resonance frequency that ranged between 200 and 400 Hz. Topographic (height) and phase images were recorded under controlled air conditions (23 °C and 50% relative humidity). No additional image processing except flattening was made.

XRD measurements were made on disks prepared by pressing freeze-dried samples, using a Philips X'Pert Pro diffractometer (model PW 3040/60). Diffractograms were recorded in the reflection mode in a 2θ angular range 5–30° by steps of 0.05°. The Cu Kα radiation (λ = 1.5418 Å) generated at 45 kV and 40 mA was monochromatized using a 20 μm Ni filter. Diffractograms were recorded from rotating specimen using a position sensitive detector.

FTIR spectra were recorded using Perkin-Elmer Spectrum 2000 FTIR equipped with a MKII Golden Gate, single reflection attenuated total reflectance (ATR) system (Specac Ltd., London, UK). The ATR crystal was a MKII heated diamond 45° ATR top plate. Samples were prepared by solvent casting of CNC suspension.

The dispersion quality of modified CNCs in toluene was confirmed by visualization of the individualized nanocrystals using field emission scanning electron microscopy (FE-SEM). A drop of the suspension was placed on a mica wafer and the solvent was slowly evaporated. The mica substrate was attached to a SEM stub and was sputtered with a thin layer of gold/palladium alloy. FE-SEM (Hitachi S-4800, Japan) operated at 1 kV acceleration voltage was used to capture secondary electron images. A suspension of unmodified CNC in water and a suspension of C18 modified CNC in toluene were deposited on the copper grid. A Hitachi S-4800 scanning electron microscope equipped with transmitted electron detector was used to capture a transmitted electron image to further characterize the morphology of the particles.

The CNC aqueous suspension and the toluene suspension of modified CNCs were sealed in a flat-sided capillary (a rectangular cross-section glass capillary tube with an optical path length of 0.4 mm and width of 4 mm, VitroCom Inc., NJ) and observed with optical microscopy between crossed polars (Leica Ortholuc equipped with a Leica DC 300 camera, Germany).

Contact angle measurements were carried out using an easy drop commercial instrument (CAM 200, KSV Instrument, Finland). A drop of Milli-Q water (3 μL) was added onto each surface using a Hamilton syringe at 25 °C and the contact angle was calculated using the sessile drop fitting method.

Results and discussion

Preparation of cellulose nanocrystals

One of the main disadvantages presented by the preparation of an aqueous suspension of cellulose nanocrystals by sulfuric acid hydrolysis is that the resulting sulfate moieties at the nanocrystal surfaces are rather labile, being in particular readily removed under mild alkaline conditions.¹⁰ Cellulose nanocrystals prepared by hydrochloric or hydrobromic acid hydrolysis have limited dispersibility in water and tend to flocculate. In order to improve their dispersion in water, subsequent TEMPO-mediated oxidation has been employed to convert C6 primary hydroxyls to carboxylic acid groups, which are more chemically stable than the sulfate esters. By using this route, i.e. TEMPO-mediated oxidation on HCl or HBr hydrolyzed cellulose, a maximum degree of oxidation (DO)⁵³ value of ca. 0.20, corresponding to a carboxyl group content of 1.2 mmol g⁻¹, can be achieved.^8,9 However, the maximum DO values of TEMPO-oxidized nonhydrolyzed cotton linters and sugar beet pulp were higher than those of TEMPO-oxidized HCl-hydrolyzed samples.¹¹ Inspired by this observation, an alternate new route, HCl hydrolysis of TEMPO-oxidized cellulose nanofibrils (TO-NFC) from softwood pulp was applied to prepare cellulose nanocrystals, since a higher surface charge density is essential for further functionalization on the surface of CNCs. TO-NFC with a carboxylate content of 1.46 mmol g⁻¹ corresponding to a DO value of 0.25 was used to evaluate this new route.

The morphology and size distribution of the CNCs were characterized by AFM. The height mode AFM images for the samples of TO-NFC and CNCs prepared at different hydrolysis times are shown in Fig. 2. To eliminate the effect of tip radius on width measurements, the heights of the nanofibrils and nanocrystals, which are not subject to peak broadening artifacts, were measured and the results are summarized in Table 1. The width of TO-NFC nanofibrils is 2.4 ± 0.7 nm as measured from the height of AFM image, the lengths of TO-NFC are possibly several micrometers and nanofibril ends are not apparent (Fig. 2A). As shown in Table 1, after HCl acid hydrolysis, the polydispersity of nanocrystal length distribution is reduced with increasing hydrolysis time, while the mean value decreased from 331 to 288 nm corresponding to a hydrolysis time of 3 and 7 h, respectively. The widths of the nanocrystals did not change significantly as compared to that of the starting TO-NFC. There is a slight increase in width of the sample CNC-7h, but this is most likely due to the larger amount of agglomerates present in the suspension. As determined by using an electric conductivity titration method,⁵⁴ the CNC samples with different hydrolysis times have almost the same total carboxylate content (1.50 mmol g⁻¹) as the starting TO-NFC (1.46 mmol g⁻¹), indicating that the carboxyl groups on the surfaces of the cellulose microfibrils are not affected by the HCl acid hydrolysis.

Fig. 2 Height mode AFM images of (A) TO-NFC, (B) CNC-3h, (C) CNC-5h, and (D) CNC-7h (Scan size 5 × 5 μm).

Table 1 Properties of the CNCs prepared by HCl-hydrolysis of TO-NFC from softwood pulp

	Crystal size (nm)^a			CrI^c	Nanocrystal dimensions^b
	C 1	C 2	C A		Width (nm)	Length (nm)
a The C1 and C2 are crystal sizes of the planes corresponding to d-spacings of 0.60–0.61 and 0.53–0.54 nm, respectively. The CA is the average value of C1 and C2. b Data obtained from height mode AFM images. c Crystallinity index. d Data not available.
Softwood pulp	3.3	2.8	3.1	61%	—^d	—^d
TO-NFC	2.8	2.4	2.6	66%	2.4 ± 0.7	—^d
CNC-3h	2.6	2.4	2.5	79%	2.3 ± 0.1	331 ± 140
CNC-5h	2.5	2.5	2.5	80%	2.7 ± 1.1	278 ± 115
CNC-7h	2.6	2.4	2.5	78%	3.6 ± 1.1	288 ± 94

---

The crystal structure of the CNC samples was investigated by XRD. The CNC samples show a typical diffraction pattern of cellulose I (Fig. 3), just as the original pulp and TO-NFC. Two peaks centered at about 14.8 and 16.8 in the XRD patterns, which correspond to d-spacing of 0.60–0.61 and 0.53–0.54 nm, respectively, were separated by curve-fitting using pseudo-Voigt function.⁵⁵ The crystal sizes of the corresponding planes were calculated from full widths at half heights of the diffraction peaks by Scherrer's equation.⁵⁶ The crystallinity index (CrI) was determined from the ratio of the separated crystalline peak area to the total reflection area including background. The results are summarized in Table 1. The average crystal size of TO-NFC (2.6 nm) is slightly smaller than that of the original softwood pulp (3.1 nm) due to the treatment with high-pressure homogenizer. The crystal sizes of the CNCs samples did not change as compared to TO-NFC, which is comparable to the results of AFM. The crystallinity of the CNCs samples is in the range of 78–80%, which is higher than the value for TO-NFC and the original softwood pulp due to the break down of the non-crystalline region by HCl acid hydrolysis. Altogether these results suggest that CNCs bearing a high content of carboxyl groups on the surfaces have been successfully prepared for the first time via direct HCl acid hydrolysis of TO-NFC.

Fig. 3 X-Ray diffraction patterns of the original pulp fibers, TO-NFC, and CNCs obtained at different hydrolysis times.

Modification of cellulose nanocrystals with quaternary ammonium salts

The CNC-3h sample having a carboxylate content of 1.5 mmol g⁻¹ was used for the reaction with quaternary ammonium salts. Stearyltrimethylammonium chloride (1) (STAC) was first used to screen the ionic exchange conditions, where three different molar ratios, 1 : 1, 1 : 2, and 1 : 4, between the carboxyl groups of CNC and STAC were studied, respectively. It was found that a lower ratio of 1 : 1 was not sufficient to obtain a good dispersion of STAC-modified CNCs in the organic solvents, while a higher amount with a ratio of 1 : 4 did not improve the quality of the dispersion. Therefore, a molar ratio of 1 : 2 between the carboxyl groups of CNCs and the quaternary ammonium salts was also used for phenyltrimethylammonium chloride (2), glycidyl trimethylammonium chloride (3), and diallyldimethylammonium chloride (4).

FTIR spectra of unmodified CNCs and CNCs modified with four different quaternary ammonium salts are shown in Fig. 4. The band at around 1050 cm⁻¹, which is the most intensive in the spectrum of unmodified CNCs, has been identified in all modified CNC samples as corresponding to the pyranose ring ether band of cellulose. For unmodified CNCs, the band at around 1730 cm⁻¹ corresponds to the C=O stretching frequency of carbonyl groups in their acidic form, which is generated by HCl-hydrolysis of TO-NFC.¹⁰ This band is shifted to approximately 1607 cm⁻¹ corresponding to carboxylate groups for all modified CNCs. For STAC (1) modified CNCs, strong bands at 2900 and 2850 cm⁻¹ corresponding to asymmetrical and symmetrical CH₂ stretches from the long alkyl chain of stearyltrimethylammonium chloride, are observed.²⁴ The successful ionic exchange of quaternary ammonium salt with Na⁺, the counter ion of carboxylate group at pH 10 was proved by the presence of a new peak at 1480 cm⁻¹, corresponding to the trimethyl groups of the quaternary ammonium, which is identified in all modified CNC samples.

Fig. 4 FTIR spectra of unmodified CNCs and CNCs modified with quaternary ammonium salts bearing long alkyl (1), phenyl (2), glycidyl (3), and diallyl (4) groups.

The STAC-modified CNCs were not dispersible in acetone and THF. However, a stable translucent suspension in chloroform and a transparent suspension in toluene were obtained. The dispersion quality of the STAC-modified CNCs was confirmed by visualization of the individualized nanocrystals using STEM (Fig. 5). The unmodified CNCs are well individualized in water, with typical dimensions of 300–500 nm in length and 3–4 nm in width (Fig. 5A). A good dispersion was observed for the toluene suspension of the CNCs modified with the quaternary ammonium salt STAC (Fig. 5B). The modified nanocrystals have similar lengths compared with unmodified ones, but their widths are wider (around 15 nm) due to the aggregates of a few nanocrystals in the form of bundles. A similar level of dispersion of CNCs in toluene with bundles of nanocrystal aggregates was also observed in the surfactant-assisted dispersion method.³⁵

Fig. 5 STEM images of (A) unmodified CNCs in water suspension and (B) STAC-modified CNCs in toluene suspension. Scale bar is 1 μm.

The aqueous suspension of unmodified CNCs formed a strong gel at around 5% solid content and showed birefringence under crossed polars (Fig. 6A and B), but no chiral nematic phase separation was observed due to the very high viscosity, similar to previously reported TEMPO-mediated oxidation of cellulose whiskers resulting from HCl hydrolysis of tunicin¹⁰ and cationic CNCs.⁷ The toluene suspension of STAC-modified CNCs with a solid content of 20% also showed strong birefringence under crossed polars (Fig. 6C). A close examination (Fig. 6D) indicates that the birefringence is not uniform, but consists of domains of various sizes and colors. These observations indicate that the STAC-modified CNCs are organized locally into oriented domains, but the absence of either tactoids or fingerprints indicates that the more perfect chiral-nematic order could not be reached. This lack of further organization is perhaps due to the fact that the size of the C18 long alkyl chain of STAC is much smaller than that of pedant poly(ethylene glycol) 1000 grafted on CNCs⁸ and the triblock copolymers adsorbed on CNCs,⁴⁴ thus, STAC-modified CNCs can not be perfectly sterically stabilized to form chiral-nematic ordering.

Fig. 6 Optical microscopy images viewed between crossed polars: (A) and (B), unmodified CNCs in water (5% solid content); (C) and (D), STAC-modified CNCs in toluene (20% solid content). The overview images (A and C) were taken with a Nikon dissecting microscope and the width of the images is 4 mm. The zoom-in images (B and D) were taken with a Leica optical microscope and the scale bar is 50 μm.

Contact angle measurement

The STAC-modified CNCs formed a powder as dried from their toluene suspension. This is a very useful form of CNC since it has the potential to form nanocomposites either through mixing with non-polar polymer melts or prepolymer/monomer liquids. Herein, model surfaces were prepared by using a spin coating technique to demonstrate the hydrophobicity of the CNCs modified with the quaternary ammonium salts. A plasma treated Si-wafer was dipped into poly-L-lysine solution (0.1% w/v) for 3 min. The surface was then washed with water and dried with nitrogen gas. Smooth model films (approximately 30 nm thick) were prepared by spin-coating (Chemat Technology, KW-4A) the aqueous suspension of unmodified CNCs onto the pretreated Si-wafer substrates at 700 rpm for 15 s followed by 2500 rpm for 60 s. Afterwards the surfaces were cleaned using nitrogen gas and the substrate was dipped into STAC aqueous solution (0.1% w/v) at pH 10 for 3 min, washed with water and dried using nitrogen, and finally washed with toluene and dried in the hood. As shown in Fig. 7, the model surface containing unmodified cellulose nanocrystals gave a water contact angle of 12°. The model surface prepared from C18-modified CNCs washed with water showed a higher contact angle with water of 48°. The model surface of C18-modified CNCs washed with toluene showed, on the other hand, a much higher contact angle with water of 71°, indicating that CNCs modified with quaternary ammonium salts bearing C18 alkyl chains are rather hydrophobic. This also shows the importance of the washing step, during which the admicelles reorganize themselves and any excess surfactant is removed.

Fig. 7 AFM image of model surface from unmodified CNCs (A), droplet of water on the film of unmodified CNCs (B), AFM image of model surface from the STAC-modified CNCs rinsed with water (C) and rinsed with toluene (E), droplet of water on the STAC-modified CNCs rinsed with water (D) and rinsed with toluene (F) (Scan size of the AFM images, 5 × 5 μm).

Conclusions

CNC is an environmentally friendly bio-based material that could serve as a valuable renewable resource on which to base a new biopolymer composites industry. The surface chemical functionality of CNCs dictates their degree of dispersion in liquids, but also in nanocomposite fabrication processes which is important. Well-dispersed nanoparticles in the polymer matrix is a prerequisite for obtaining favorable nanocomposite properties.

CNCs bearing high carboxylate content (1.5 mmol g⁻¹) have been successfully prepared by hydrochloric acid hydrolysis of nanofibrillated cellulose obtained from TEMPO-mediated oxidation of softwood pulp. The resulting nanocrystals have similar surface charge density and lateral dimensions as compared to the starting TO-NFC. The density of carboxylic groups introduced on the cellulose nanofibres surface can be tailored in the range approximately 0.5–2.0 mmol g⁻¹ by TEMPO-mediated hypohalite oxidation as reported previously,⁵¹ indicating that CNCs with high charge densities between 1.5 and 2.0 mmol g⁻¹ can be obtained. This value is higher than the maximum value (1.2 mmol g⁻¹) reported previously by TEMPO-mediated oxidation on HCl or HBr hydrolyzed cellulose.^8,9 The advantages of having carboxylic groups are: (1) the primary hydroxyl groups on the surface of the CNCs are selectively converted to carboxylic acids; (2) carboxylic acid groups are chemically more stable than the sulfate esters generated by sulfuric acid hydrolysis.

Furthermore, surface modification of these CNCs with high surface charge densities of carboxylic acid groups have been carried out using quaternary ammonium salt surfactants via an ionic exchange process in aqueous solution as inspired by organomodified layered silicates. Four quaternary ammonium cations bearing long alkyl, phenyl, glycidyl, and diallyl groups were successfully used to modify CNCs to alter their surface hydrophobicity. The modified CNCs can be redispersed and individualized in organic solvents such as toluene and chloroform. One may envision removing excess surfactant to obtain CNCs with a monolayer of surfactant.

This surface modification method for CNCs, combining adsorption of quaternary ammonium salts and the new route for the production of CNCs, is simple and flexible with high yield. A broad range of functional groups can be introduced on the surface of the cellulose nanocrystals in this fashion. The density of the functional groups can also be tuned by controlling the density of the carboxylic acid groups on the surface of CNCs. This new series of modified CNCs can be dried from solvent in non-agglomerated form and have the potential to form well-dispersed nanocomposites with non-polar polymers, through mixing with thermoplastic melts or liquid prepolymer/monomers.

Acknowledgements

The authors thank the Wallenberg Wood Science Center (WWSC, http://wwsc.se) for support. Q.Z. acknowledges the Swedish Research Council Formas for additional salary funding via CarboMat—The KTH Advanced Carbohydrate Materials Consortium (A Formas Strong Research Environment). We thank Spectral Solutions AB for help with the STEM analysis.

References

1. M. A. S. A. Samir, F. Alloin and A. Dufresne, Biomacromolecules, 2005, 6, 612–626.
2. Y. Habibi, L. A. Lucia and O. J. Rojas, Chem. Rev., 2010, 110, 3479–3500.
3. R. J. Moon, A. Martini, J. Nairn, J. Simonsen and J. Youngblood, Chem. Soc. Rev., 2011, 40, 3941–3994.
4. D. Klemm, F. Kramer, S. Moritz, T. Lindstrom, M. Ankerfors, D. Gray and A. Dorris, Angew. Chem., Int. Ed., 2011, 50, 5438–5466.
5. J. F. Revol, H. Bradford, J. Giasson, R. H. Marchessault and D. G. Gray, Int. J. Biol. Macromol., 1992, 14, 170–172.
6. S. Beck-Candanedo, M. Roman and D. G. Gray, Biomacromolecules, 2005, 6, 1048–1054.
7. M. Hasani, E. D. Cranston, G. Westman and D. G. Gray, Soft Matter, 2008, 4, 2238–2244.
8. J. Araki, M. Wada and S. Kuga, Langmuir, 2001, 17, 21–27.
9. I. Filpponen and D. S. Argyropoulos, Biomacromolecules, 2010, 11, 1060–1066.
10. Y. Habibi, H. Chanzy and M. R. Vignon, Cellulose, 2006, 13, 679–687.
11. S. Montanari, M. Rountani, L. Heux and M. R. Vignon, Macromolecules, 2005, 38, 1665–1671.
12. D. Dubief, E. Samain and A. Dufresne, Macromolecules, 1999, 32, 5765–5771.
13. A. Dufresne, M. B. Kellerhals and B. Witholt, Macromolecules, 1999, 32, 7396–7401.
14. M. N. Angles and A. Dufresne, Macromolecules, 2000, 33, 8344–8353.
15. A. Dufresne, Molecules, 2010, 15, 4111–4128.
16. V. Favier, H. Chanzy and J. Y. Cavaille, Macromolecules, 1995, 28, 6365–6367.
17. B. L. Peng, N. Dhar, H. L. Liu and K. C. Tam, Can. J. Chem. Eng., 2011, 89, 1191–1206.
18. J. O. Zoppe, Y. Habibi, O. J. Rojas, R. A. Venditti, L. S. Johansson, K. Efimenko, M. Osterberg and J. Laine, Biomacromolecules, 2010, 11, 2683–2691.
19. Q. X. Xu, J. Yi, X. F. Zhang and H. L. Zhang, Eur. Polym. J., 2008, 44, 2830–2837.
20. J. Majoinen, A. Walther, J. R. McKee, E. Kontturi, V. Aseyev, J. M. Malho, J. Ruokolainen and O. Ikkala, Biomacromolecules, 2011, 12, 2997–3006.
21. G. Morandi, L. Heath and W. Thielemans, Langmuir, 2009, 25, 8280–8286.
22. S. P. Dong and M. Roman, J. Am. Chem. Soc., 2007, 129, 13810–13811.
23. L. Rueda, B. F. d'Arlas, Q. Zhou, L. A. Berglund, M. A. Corcuera, I. Mondragon and A. Eceiza, Compos. Sci. Technol., 2011, 71, 1953–1960.
24. G. Siqueira, J. Bras and A. Dufresne, Langmuir, 2010, 26, 402–411.
25. H. H. Yuan, Y. Nishiyama, M. Wada and S. Kuga, Biomacromolecules, 2006, 7, 696–700.
26. C. Gousse, H. Chanzy, G. Excoffier, L. Soubeyrand and E. Fleury, Polymer, 2002, 43, 2645–2651.
27. C. Gousse, H. Chanzy, M. L. Cerrada and E. Fleury, Polymer, 2004, 45, 1569–1575.
28. A. H. Pei, Q. Zhou and L. A. Berglund, Compos. Sci. Technol., 2010, 70, 815–821.
29. E. Feese, H. Sadeghifar, H. S. Gracz, D. S. Argyropoulos and R. A. Ghiladi, Biomacromolecules, 2011, 12, 3528–3539.
30. J. O. Zoppe, M. S. Peresin, Y. Habibi, R. A. Venditti and O. J. Rojas, ACS Appl. Mater. Interfaces, 2009, 1, 1996–2004.
31. Y. Habibi, A. L. Goffin, N. Schiltz, E. Duquesne, P. Dubois and A. Dufresne, J. Mater. Chem., 2008, 18, 5002–5010.
32. S. Ahola, M. Osterberg and J. Laine, Cellulose, 2008, 15, 303–314.
33. F. Aloulou, S. Boufi and D. Beneventi, J. Colloid Interface Sci., 2004, 280, 350–358.
34. E. D. Cranston and D. G. Gray, Biomacromolecules, 2006, 7, 2522–2530.
35. L. Heux, G. Chauve and C. Bonini, Langmuir, 2000, 16, 8210–8212.
36. C. Bonini, L. Heux, J. Y. Cavaille, P. Lindner, C. Dewhurst and P. Terech, Langmuir, 2002, 18, 3311–3314.
37. N. Ljungberg, C. Bonini, F. Bortolussi, C. Boisson, L. Heux and J. Y. Cavaille, Biomacromolecules, 2005, 6, 2732–2739.
38. I. Kvien, B. S. Tanem and K. Oksman, Biomacromolecules, 2005, 6, 3160–3165.
39. J. Kim, G. Montero, Y. Habibi, J. P. Hinestroza, J. Genzer, D. S. Argyropoulos and O. J. Rojas, Polym. Eng. Sci., 2009, 49, 2054–2061.
40. O. J. Rojas, G. A. Montero and Y. Habibi, J. Appl. Polym. Sci., 2009, 113, 927–935.
41. S. Padalkar, J. R. Capadona, S. J. Rowan, C. Weder, Y. H. Won, L. A. Stanciu and R. J. Moon, Langmuir, 2010, 26, 8497–8502.
42. H. Brumer, Q. Zhou, M. J. Baumann, K. Carlsson and T. T. Teeri, J. Am. Chem. Soc., 2004, 126, 5715–5721.
43. Q. Zhou, L. Greffe, M. J. Baumann, E. Malmstrom, T. T. Teeri and H. Brumer, Macromolecules, 2005, 38, 3547–3549.
44. Q. Zhou, H. Brumer and T. T. Teeri, Macromolecules, 2009, 42, 5430–5432.
45. S. S. Ray and M. Okamoto, Prog. Polym. Sci., 2003, 28, 1539–1641.
46. F. Aloulou, S. Boufi and M. Chakchouk, Colloid Polym. Sci., 2004, 282, 699–707.
47. F. Aloulou, S. Boufi, N. Belgacem and A. Gandini, Colloid Polym. Sci., 2004, 283, 344–350.
48. S. Alila, S. Boufi, M. N. Belgacem and D. Beneventi, Langmuir, 2005, 21, 8106–8113.
49. J. Penfold, I. Tucker, J. Petkov and R. K. Thomas, Langmuir, 2007, 23, 8357–8364.
50. K. Xhanari, K. Syverud, G. Chinga-Carrasco, K. Paso and P. Stenius, Cellulose, 2011, 18, 257–270.
51. T. Saito, S. Kimura, Y. Nishiyama and A. Isogai, Biomacromolecules, 2007, 8, 2485–2491.
52. L. Wagberg, G. Decher, M. Norgren, T. Lindstrom, M. Ankerfors and K. Axnas, Langmuir, 2008, 24, 784–795.
53. D. D. Perez, S. Montanari and M. R. Vignon, Biomacromolecules, 2003, 4, 1417–1425.
54. T. Saito and A. Isogai, Biomacromolecules, 2004, 5, 1983–1989.
55. M. Wada, T. Okano and J. Sugiyama, Cellulose, 1997, 4, 221–232.
56. Y. Okita, T. Saito and A. Isogai, Biomacromolecules, 2010, 11, 1696–1700.

---