Fabrication of a uniaxial cellulose nanocrystal thin film for coassembly of single-walled carbon nanotubes

Abstract

A solution-based, low-cost dip-coating method was explored to assemble cellulose nanocrystals (CNCs) into a uniaxial thin film, as confirmed by multiple microscopic tools on multiple length scales. Furthermore, this approach is readily modified for the coassembly of CNCs and a tough 1D nanoitem – single-walled carbon nanotubes into a uniaxial array.

---

Cellulose is the most abundant biopolymer source on the planet. One recent emerging trend is to isolate its fibrils and whiskers that are nanometers in diameter, namely cellulose nanocrystals (CNCs), and to utilize these 1D nanoitems for fabrication of cellulose-based advanced materials for their low density, variable aspect ratio, high strength and stiffness, and rich assembly behaviors.^1–5 Interestingly, the properties of typical CNC-based materials can be highly related to the assembly behavior of CNCs. Usually, CNCs in concentrated dispersions⁶ and iridescent thin films^7,8 assemble into left-handed cholesteric domains, which reflect the intrinsic chirality of crystalline cellulose and the helical interfacial distribution of hydroxylic and sulfonic groups. The assembly behavior can largely enrich the synthetic approaches to achieving functional porous materials with tunable cholesteric architectures, where cholesteric CNC xerogels serve as the framework or sacrificing template.^9–11 For example, numerous pioneering researches show that the inclusion of the cholesteric CNC assembly into a sol–gel process leads to chiral porous ceramics.^10,12,13 Another emerging direction regarding the CNC assembly focuses on its coassembly with anisotropic nanocrystals. For instance, the coassembly of Au nanorods with CNCs nicely combines the advantage of the polarization-sensitive property of CNCs and the plasmatic one of metal nanocrystals.^14,15

Interestingly, CNC assemblies have multiple response possibilities to the external fields, which enriches their applicational opportunities. From the applicational point of view, assembling CNCs into the uniaxial thin film is significantly intuitive. Indeed, the shearing,^16–18 electric field,^18,19 and Langmuir membranes^20,21 have been used separately or together to guide the formation of uniaxial CNC assemblies. Though the local ordering in the Langmuir membrane-supported densely-packed CNC thin films can be achieved after the trough compression, it is extremely difficult to achieve the uniform uniaxial CNC thin film across the macroscopic distance.^20,21 Electric field provides another means of aligning CNCs because their dielectric anisotropy is positive. For instance, a uniaxial CNC film containing sparse particles suspended in cyclohexane was obtained in the presence of an electric field, where the long axis of CNCs points to the direction of the electric field.²² In another case study, Habibi et al. used the AC field to obtain a uniaxial CNC thin film.¹⁹ Nevertheless, the electric field-guided uniaxial alignment of CNCs into a densely-packed thin film across the macroscopic distance remains challenging. A third but effective field is the shearing force, featuring its simplicity and large-scale operation. For instance, solvent evaporation within a rolling cylindrical glass vial generated a uniaxial CNC thin film in situ.¹⁶ The shearing force can also serve in the presence of other fields to precisely control the uniaxial alignment of CNCs,¹⁸ as verified by the atomic force microscopy (AFM) tool. Overall, the controversial success has been achieved to fabricate uniaxial, densely-packed CNC thin films macroscopically with the abovementioned external fields separately or together.

Herein, a controllable dip-coating method was introduced to achieve the uniaxial CNC thin film with the uniform domain on the macroscopic scale. Two parameters, namely the lifting rate and the CNC concentration in the dispersion have been studied systematically to optimize processing parameters to achieve this goal. Structural transformation from the nematic to uniaxial assemblies was confirmed by two independent microscopic techniques, namely the polarized optical microscopy (POM) and AFM on the macroscopic and mesoscopic length scales, respectively. We highlight that the use of multiple tools is highly desirable for the structural characterization of uniaxial CNC thin films on multiple length scales. Furthermore, the dip-coating process can be readily modified for the coassembly of CNCs and single-walled carbon nanotubes (SWNTs), the latter of which are a typical tough 1D nanoitem and hence, notorious for being aligned to its uniaxial assembly in a densely-packed thin film. The uniaxial coalignment of CNCs and SWNTs was confirmed by microscopic tools and the polarized resonant Raman spectroscopy as well. Exploring controllable assembly approaches to fabricating uniaxial CNC thin films expands the applicational opportunities in assembling functional 1D nanoitems.

The CNC dispersion, prepared by the hydrolysis of the microcrystalline cellulose pulps in the presence of the concentrated sulphuric acid solution,^12,23 was used throughout all dip-coating experiments. The histogram of the length and thickness of CNCs based on AFM images determines that the values are 161 ± 65 and 8.0 ± 1.0 nm (N = 5, where N represents the total number of AFM images used to derive error bars), respectively (see ESI†).

CNC thin films show the distinct assembly behaviors in the presence of different external fields. In a dip-coating process with the lifting rate at 53 mm min⁻¹, a macroscopically-sized uniaxial CNC thin film, composed of densely-packed CNCs, was successfully achieved when a 0.046 g mL⁻¹ CNC dispersion was used. The weak and uniform contrast under the polarizers suggests the formation of the uniaxial nematic phase uniform across the macroscopic distance (Fig. 1a). The weak signal can be attributed to the low thickness value. For instance, an AFM imaging analysis verifies that the thickness of the abovementioned sample is 15.0 ± 1.6 nm (N = 4). The AFM image collected in an ambient area unambiguously probes that CNCs in the thin film align with their long axis preferentially along the lifting direction (Fig. 1d). As comparison, a solution casting procedure performed in the ambient condition led to a nematic CNC thin film by casting a 0.010 g mL⁻¹ CNC dispersion on a 1 cm² glass slide. The Schlieren texture detected under polarizers confirms the nematic nature of the film (Fig. 1b), where different contrasts under the crossed polarizers hint multiple orientational preference in different nematic domains. A zoomed-in AFM image indicates the curved alignment of CNCs (Fig. 1e), which is faithfully ascribed to the typical curved pattern of a 1D nanoitems-based nematic domain.²⁴ We note that the increase of the shearing force in a dip-coating process gradually switches the nematic domain to the uniaxial one. For instance, the use of a relatively slow lifting rate at 27 mm min⁻¹ in a dip-coating process caused the appearance of the mixed domains characteristic of the uniform birefringent contrast and the Schlieren textures (Fig. 1c). A corresponding AFM image confirms that in the curved CNC assembly, CNCs averagely point to the lifting direction (Fig. 1f).

Fig. 1 POM (a–c) and AFM (d–f) images of uniaxial (a and d), nematic (b and e), and partially aligned (c and f) CNC thin films. White arrows in multiple images point to the lifting direction in a dip-coating process. See experimental details in ESI.†

The uniaxial character is unambiguously strengthened from the rotational POM characterization. It clearly evidences that the entire contrast switched on and off simultaneously when the sample is rotated under the crossed polarizers (Fig. 2a–d). Nevertheless, sporadic gray spots are still visible, which indicates that few structural defects were inevitably formed assumedly due to the presence of turbulence during the dip-coating process. We demonstrate that the uniform birefringent contrast can be used as an important criterion to evidence the uniaxial character in the CNC film macroscopically, which can be well paired by AFM images microscopically.

Fig. 2 A series of POM images taken at the same area show a typical uniaxial CNC thin film under crossed polarizers. The white arrow in each image points to the lifting direction in a dip-coating process. The black crossed arrows show the polarization direction. The lifting rate was 53 mm min⁻¹, and the CNC concentration in the dispersion phase was 0.046 g mL⁻¹.

The histogram analysis based on AFM images provides a quantitative method to determine the in-plane nematic order parameter (S) in CNC thin films with distinct assembly behaviors. For instance, an exemplary histogram analysis on an exemplary uniaxial CNC thin film confirms that the S value is 0.89 ± 0.08 (N = 5) along the lifting direction (Fig. 3a–b; see the calculation equations in ESI†). This result is reasonable because 1D nanoitems are prone to align as such to cancel out the external shearing force. As comparison, the localized S value of a partially-aligned CNC thin film is decreased to 0.61 ± 0.19 (N = 5) along the lifting direction (Fig. 3c–d). We note that for the above partially-aligned sample, the histogram analysis based on multiple AFM images taken in random areas is required to cancel the orientational inhomogeneity in the CNC thin film, and hence, to provide the relatively precise S value.

Fig. 3 (a and b) Exemplary AFM ((a), the amplitude mode) image and the histogram (b) of a uniaxial CNC thin film. (c and d) Exemplary AFM ((c), the amplitude mode) image and histogram (d) of a partially-aligned CNC thin film. The statistics in the histogram was based on the number percentage of CNCs in leading angles with a reference of the dip-coating direction. At least three AFM images were analyzed for statistics. White arrows in both AFM images point to the lifting direction in a dip-coating process.

Next, experimental conditions in a dip-coating process are discussed. The formation of the uniaxial CNC thin film is largely dependent on a net fluid/CNC flow towards the fluid meniscus driven by evaporation, based on the idea of the contact-line deposition of colloids.^25–27 Two main factors, namely the lifting rate and the concentration of CNCs in the dispersion, were studied to optimize the structural outcome of the CNC thin film in a dip-coating process. The lifting rate is the first typical factor in determining the assembly behavior of CNCs. Phenomenally, the size percentage of uniaxial domains increased when the lifting rate was increased up to 53 mm min⁻¹ (Fig. 1a and d, and ESI†). Particularly, the histogram based on AFM images discloses the nice interplay between the lifting rates and the S values of the CNC thin film along the lifting direction (see ESI†). Nevertheless, increasing the lifting rate further caused the reappearance of nematic domains, as revealed in POM images (see ESI†). We infer that the fast lifting rate left the aqueous CNC dispersion limited time to evaporate at the meniscus area. Hence, nematic domains could be assumably formed subsequently during the evaporation of the remaining dispersion layer.

The CNC concentration in the dispersion phase is another crucial factor to obtain the uniaxial CNC thin film. The POM scanning verifies that the size percentage of uniaxial domains rises with the increase of the CNC concentration value in the dispersion (ESI†). The high CNC concentration in the evaporating thin layer close to the meniscus area, plus the presence of the shearing force, is deemed to guide the uniaxial alignment of CNCs. Nevertheless, the gelation appeared in the CNC dispersion when the CNC concentration was as high as about 0.050 g mL⁻¹, in line with the threshold gelation concentration of the CNC dispersion obtained from the microcrystalline cellulose and filter paper sources.^23,28 Unlike the assembly of spherical colloids in a dip-coating process, where a wide combination of the lifting rate and the colloidal concentration leads to colloidal crystals,²⁹ the assembly of CNCs requires more strict experimental conditions, namely the high CNC concentration and the fast lifting rate. In the current case study, the uniaxial CNC thin film can be achieved when the CNC concentration and the lifting rate are 0.046 ± 0.003 g mL⁻¹ and 53 ± 3 mm min⁻¹.

Particularly, the uniaxial CNC thin film achieved in a dip-coating process was formed not from the very beginning. Instead, the nematic domains were formed initially, and the gradual structural transformation to the uniaxial character occurred when the meniscus line moved forward in the next hundreds of micrometers (see ESI†). Afterwards, the uniaxial CNC thin film was generated continuously. This gradual structural change evidences that the presence of the strong shearing force, generated by using high lifting rate, is prerequisite for obtaining the uniaxial CNC thin film in the current study.

The abovementioned procedure can be readily modified to coalign CNCs and other 1D nanoitems – SWNTs for example. A POM image nicely shows the uniaxial coalignment of both ingredients macroscopically (Fig. 4a). The SEM image evidences the uniform coverage of SWNTs on the substrate along the shearing direction after the removal of the CNC ingredient by using a heating procedure (Fig. 4b). We hypothesize that the uniaxial CNC alignment at the meniscus during evaporation assisted the local confinement of SWNTs with their tubular axis along the lifting direction. In a previous study, the shearing force in obtaining the uniaxial thin film was strengthened by the presence of the patterned hydrophilic stripes micrometers in width.³⁰ We expect that the presence of CNCs – an inert ingredient – should have the limited interference on the functional outcomes of the SWNT-based hybrid thin film. The current study provides a simple, solution-based approach to align SWNTs into an in-plane uniaxial, densely-packed thin film across the macroscopic distance.

Fig. 4 (a) POM image of the uniaxial CNC–SWNT hybrid thin film; (b) SEM image of the uniaxial SWNT thin film after removal of the CNC ingredient; (c) three characteristic SWNT bands, namely the radial breathing mode, G-band, and G′-band, in the polarized resonant Raman spectra obtained in the CNC–SWNT hybrid thin film. In two curves, the polarized light is perpendicular (solid) and parallel (dotted) to the long axis of SWNTs. White arrows in images (a and b) point to the lifting direction in a dip-coating process. The black crossed arrows show the polarization direction.

The polarized resonant Raman spectroscopy was used to quantify the in-plane uniaxial alignment of SWNTs within the hybrid thin film.^31,32 The orientational-dependent measurements reveal that the intensities of SWNT scattering signals of the radial breathing mode, G, and G′-bands regimes are angular-dependent (Fig. 4c). The maximized and minimized intensities are obtained when SWNTs align parallel and vertical to the polarization of the incident laser, respectively. The analyses of Raman data allow for calculating the quantitative value of the local S value in the SWNT thin film. According to a previous study, an approximation was made by using the Raman mode intensities I in the SWNT thin film directly for calculation.³³ The result shows that the S value is about 0.60 ± 0.05 (N = 10), as the values at 1 and 0 correspond to the perfect crystalline order and the total disorder, respectively. The slightly curved SWNTs in the hybrid thin film undermines the S value in the measurement.

To conclude, our study puts forward the versatility of a simple dip-coating process for obtaining the uniaxial CNC thin film across the macroscopic distance. Such in-plane oriented CNC thin film can function as the colloidal confining agent for the uniaxial alignment of SWNTs, which are among important functional 1D nanoitems in various fields. In short, the method explored herein is an enabling technique that may lay the function for complex hybrids and even devices, where the uniaxial alignment is a key factor.

Acknowledgements

The work was supported by the National Natural Science Foundation of China (21303144), Science Foundation of the Fujian Province, China (2014J0101), the “111” Project (B16029), National Nature Science Foundation of China (No. U1405226), Fujian Provincial Department of Science & Technology (2014H6022), and the 1000 Talents Program from Xiamen University. Special thanks to Mr Mingfeng Liu for his assistance in performing supplementary experiments.

Notes and references

1. 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., 2009, 45, 1–33.
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. P. Lagerwall, C. Schütz, M. Salajkova, J. H. Noh, J. Hyun Park, G. Scalia and L. Bergström, NPG Asia Mater., 2014, 6, e80.
6. J.-F. Revol, H. Bradford, J. Giasson, R. H. Marchessault and D. G. Gray, Int. J. Biol. Macromol., 1992, 14, 170–172.
7. J.-F. Revol, L. Godbout and D. G. Gray, J. Pulp Pap. Sci., 1998, 24, 146–149.
8. J. H. Pan, W. Y. Hamad and S. K. Straus, Macromolecules, 2010, 43, 3851–3858.
9. E. Dujardin, M. Blaseby and S. Mann, J. Mater. Chem., 2003, 13, 696–699.
10. J. A. Kelly, M. Giese, K. E. Shopsowitz, W. Y. Hamad and M. J. MacLachlan, Acc. Chem. Res., 2014, 47, 1088–1096.
11. M. Giese, L. K. Blusch, M. K. Khan and M. J. MacLachlan, Angew. Chem., Int. Ed., 2015, 54, 2888–2910.
12. K. E. Shopsowitz, H. Qi, W. Y. Hamad and M. J. Maclachlan, Nature, 2010, 468, 422–425.
13. K. E. Shopsowitz, A. Stahl, W. Y. Hamad and M. J. MacLachlan, Angew. Chem., Int. Ed., 2012, 51, 6886–6890.
14. Q. Liu, M. G. Campbell, J. S. Evans and I. I. Smalyukh, Adv. Mater., 2014, 26, 7178–7184.
15. A. Querejeta-Fernandez, G. Chauve, M. Methot, J. Bouchard and E. Kumacheva, J. Am. Chem. Soc., 2014, 136, 4788–4793.
16. N. Yoshiharu, K. Shigenori, W. Masahisa and O. Takeshi, Macromolecules, 1997, 30, 6395–6397.
17. I. Hoeger, O. J. Rojas, K. Efimenko, O. D. Velev and S. S. Kelley, Soft Matter, 2011, 7, 1957–1967.
18. L. Csoka, I. C. Hoeger, P. Peralta, I. Peszlen and O. J. Rojas, J. Colloid Interface Sci., 2011, 363, 206–212.
19. Y. Habibi, T. Heim and R. Douillard, J. Polym. Sci., Part B: Polym. Phys., 2008, 46, 1430–1436.
20. Y. Habibi, L. Foulon, V. Aguié-Béghin, M. Molinari and R. Douillard, J. Colloid Interface Sci., 2007, 316, 388–397.
21. Y. Habibi, I. Hoeger, S. S. Kelley and O. J. Rojas, Langmuir, 2010, 26, 990–1001.
22. D. Bordel, J.-L. Putaux and L. Heux, Langmuir, 2006, 22, 4899–4901.
23. X. M. Dong, J.-F. Revol and D. G. Gray, Cellulose, 1998, 5, 19–32.
24. J. C. Gabriel and P. Davidson, Adv. Mater., 2000, 12, 9–20.
25. R. D. Deegan, O. Bakajin, T. F. Dupont, G. Huber, S. R. Nagel and T. A. Witten, Nature, 1997, 389, 827–829.
26. J. X. Huang, A. R. Tao, S. Connor, R. R. He and P. D. Yang, Nano Lett., 2006, 6, 524–529.
27. M. Ghosh, F. Q. Fan and K. J. Stebe, Langmuir, 2007, 23, 2180–2183.
28. S. Beck-Candanedo, M. Roman and D. G. Gray, Biomacromolecules, 2005, 6, 1048–1054.
29. Y. N. Xia, B. Gates, Y. D. Yin and Y. Lu, Adv. Mater., 2000, 12, 693–713.
30. S. Park, G. Pitner, G. Giri, J. H. Koo, J. Park, K. Kim, H. L. Wang, R. Sinclair, H. S. P. Wong and Z. N. Bao, Adv. Mater., 2015, 27, 2656–2662.
31. R. Saito, T. Takeya, T. Kimura, G. Dresselhaus and M. S. Dresselhaus, Phys. Rev. B: Condens. Matter, 1998, 57, 4145–4153.
32. G. S. Duesberg, I. Loa, M. Burghard, K. Syassen and S. Roth, Phys. Rev. Lett., 2000, 85, 5436–5439.
33. J. P. Lagerwall, G. Scalia, M. Haluska, U. Dettlaff-Weglikowska, S. Roth and F. Giesselmann, Adv. Mater., 2007, 19, 359–364.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental details and supplementary images. See DOI: 10.1039/c6ra06574k

---