pubs.acs.org/Langmuir © 2010 American Chemical Society
Elaboration of Spin-Coated Cellulose-Xyloglucan Multilayered Thin Films Carole Cerclier,† Fabrice Cousin,‡ Herve Bizot,† Celine Moreau,† and Bernard Cathala*,† †
UR1268 Biopolym eres Interactions Assemblages, INRA, F-44316 Nantes, France, and Laboratoire L eon Brillouin, CEA/CNRS Saclay, 91191 Gif-sur-Yvette cedex, France
‡
Received February 5, 2010. Revised Manuscript Received September 8, 2010 In the context of developing a biomimetic model of the primary cell wall, our aim was to produce multilayered thin films composed of cellulose nanocrystals (CN) and xyloglucan (XG). We investigated the effect of XG concentrations ranging from 0.5 g/L to 10 g/L. The choice of concentration was based on rheological investigation of the XG solutions which indicated that the two lower concentrations (0.5 and 1 g/L) correspond to a semidilute regime where the polymer chains are not entangled, whereas they are entangled at the highest concentrations (5 and 10 g/L). Several processes of film preparation were tested (dipping or spin-coating, with or without a rinsing step). The film growth profiles obtained for different XG concentrations by mechanical profilometry showed that spin-coating without rinsing was the most efficient process. Results showed that at high XG concentrations (XG = 5 g/L and XG = 10 g/L) plateau values were reached after the formation of 3 or 4 bilayers, whereas growth of the multilayer structure was linear at the lower XG concentrations (XG = 0.5 g/L and XG = 1 g/L). The thickness of one CN/XG bilayer corresponded to a single layer of CN covered by a thin XG layer, despite the absence of a rinsing step between successive coatings. The importance of the XG concentration was confirmed by determining by neutron reflectivity the film architecture obtained from four XG solutions after eight successive paired coatings. The results are discussed in relation to the role of XG in the plant cell wall.
Introduction The increasing interest of using plant cell wall polymers to produce new biobased materials or biofuels has prompted scientists to develop mimetic assemblies in order to gain both a better insight into plant cell wall organization and to develop new strategies for improving biopolymer use. The rules governing the interactions between polymers in plant assemblies are far from being fully understood at present. Biologists are using mutagenesis to alter the activity of genes involved in biosynthesis and to identify the role of individual components,1-3 whereas physical chemists aim to elucidate the cell formation process by designing in vitro biomimetic models made with purified molecules. Such studies have already provided valuable information to clarify the interactions occurring between cell wall polymers,4-8 to propose new hypotheses concerning plant cell wall structure,9 and to understand the reactivity of cell wall components.10,11 *Author e-mail address:
[email protected]. (1) Cannon, M.; Terneus, K.; Hall, Q.; Tan, L.; Wang, Y.; Wegenhart, L.; Chen, L.; Lamport, D. T. A.; Chen, Y. R.; Kieliszewski, M. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 2226–2231. (2) Bouton, S; Leboeuf, E; Mouille, G; Leydecker, M. T.; Talbotec, J; Granier, F; Lahaye, M; Hofte, H; HN, T. Plant Cell 2002, 14(10), 2577–2590. (3) Mouille, G.; Ralet, M.; Cavelier, C.; Eland, C.; Effroy, D.; Hematy, K.; McCartney, L.; Truong, H.; Gaudon, V.; Thibault, J.; Marchant, A.; Hofte, H. Plant J. 2007, 50(4), 605–614. (4) Whitney, S. E. C.; Gothard, M. G. E.; Mitchell, J. T.; Gidley, M. J. Plant Physiol. 1999, 121, 657–663. (5) Chanliaud, E.; Gidley, M. J. Plant J. 1999, 20(1), 25–35. (6) Touzel, J. P.; Chabbert, B.; Monties, B.; Debeire, P.; Cathala, B. J. Agric. Food Chem. 2003, 51, 981–986. (7) Lairez, D.; Cathala, B.; Monties, B.; Bedos-Belval, F.; Duran, D.; Gorrichon, L. Biomacromolecules 2005, 6, 763–774. (8) Barakat, A.; Gaillard, C. x.; dric; Lairez, D.; Saulnier, L.; Chabbert, B.; Cathala, B. Biomacromolecules 2008, 9(2), 487–493. (9) Zykwinska, A. W.; Ralet, M. C. J.; Garnier, C. D.; Thibault, J. F. J. Plant Physiol. 2005, 139(1), 397–407. (10) Barakat, A.; Chabbert, B.; Cathala, B. Phytochemistry 2007, 68(15), 2118–2125. (11) Barakat, A.; Winter, H.; Rondeau-Mouro, C.; Saake, B.; Chabbert, B.; Cathala, B. Planta 2007, 226(1), 267–281.
17248 DOI: 10.1021/la102614b
Cellulose is the main structural element of woody plant cell walls and is therefore one of the most abundant biopolymers available in nature. The long unramified cellulose polymer chains are designed for self-association and consist of flat ribbons of β-1,4-linked D-glucose rings, which adopt an extended 2-1 helical conformation with successive pyranose rings exposing alternatively their most hydrophobic face every second residue. The resulting packs of parallel chains, assembled upon biosynthesis, rely mainly on van der Waals attractions that provide cohesion perpendicularly to the ribbon surfaces and on hydrogen bonds between equatorial hydroxyls within the sheets of ribbons. Detailed quasi-atomic unit cell coordinates have recently been refined for the two allomorphs encountered in natural celluloses: monoclinic type I β with staggered chains12 and triclinic type I R with tiered chains.13 The resulting assemblies are hierarchical with crystalline microfibrils constituting the central element of a complex network responsible for the plant’s mechanical properties once arranged along tissue-specific orientations. In vivo, microfibrils are lubricated or associated with hemicelluloses, transiently or permanently, so as to form interconnected networks responsible for the morphogenesis and the dynamic load-bearing properties of the cell walls.14 The most widely studied polysaccharides interacting with cellulose belong to the xyloglucan (XG) family.15,16 Xyloglucan may be considered a hairy β 1-4 glycan, i.e. a statistical multiblock copolymer based on decorated cellotetraose (or lower oligo) segments. Typically, in such blocks three out of four successive glucose units carry a flexibly R-(1,6) linked D-xylosyl substituent, either as a single sugar or as β-D-Gal-(1,2)-R-D-Xyl/or R-l-Fuc-(1,2)-β-D-Gal-(1,2)-R-D-Xyl (12) Nishiyama, Y.; Langan, P.; Chanzy, H. J. Am. Chem. Soc. 2002, 124(31), 9074–9082. (13) Nishiyama, Y.; Sugiyama, J.; Chanzy, H.; Langan, P. J. Am. Chem. Soc. 2003, 125(47), 14300–14306. (14) Carpita, N.; Gibeaut, D. Plant J. 1993, 3(1), 1–30. (15) Hayashi, T. Annu. Rev. Plant Physiol. Plant Mol. Biol. 1989, 40, 139–168. (16) Vincken, J. P.; Keizer, A.; Beldman, G.; Voragen, G. Plant Physiol. 1995, 108, 1579–1585.
Published on Web 09/30/2010
Langmuir 2010, 26(22), 17248–17255
Cerclier et al.
side chains. However, considerable variability has been identified among plants. The resulting ribbon is topologically dissymmetric, one edge bearing two times more side chains than the other, this being modulated by the variable concentration of substituents. The implications of such a structure, regarding coiling and wrapping abilities, have not been clarified. These polysaccharides are considered to play a crucial role in cell wall elongation. Their adsorption on cellulose is mediated by hydrogen bonding and van der Waals forces.17 Besides their biological role, XGs are used as non-retrograding thickening agents in foods.18 They also have a dispersing effect on cellulose that improves paper manufacture.19 The ability of intermediate molar mass XGs to interact quasi-irreversibly with cellulose has also been used to develop molecular anchors which give new chemical functions to cellulose crystals.20 Here, we took advantage of this interaction ability to produce nanostructured films composed of XG and cellulose nanocrystals (CN). Our aims were as follows: (i) to gain a better understanding of cellulose/XG interactions, and thus a better knowledge of the design principles implemented in primary plant cell walls, and (ii) to propose new strategies for constructing materials based on bioresources. Studies of cellulose model surfaces have been of constant interest for decades, since surface phenomena are critical for controlling a wide range of material properties.21 Recently, much research has been carried out on films or coatings with tunable internal multilayered organization since such films offer new potential for adaptive applications. For instance, multilayered controlled deposition of cellulose coatings has led to the development of optically active surfaces displaying antireflective,22 semireflective23,24 or birefringent25 properties, depending on the deposition process and/or cellulosic elements morphology. The most popular method used to produce multilayered films is the layer-by-layer method developed by Decher et al. in the early 1990s.26,27 In the case of polyelectrolytes, this approach consists of alternately dipping a charged surface into a solution of a polyanion and a polycation in order to successively reverse the charge of the surface. However, multilayered thin films can also be successfully prepared through nonelectrostatic interactions.28 The various approaches have included specific interactions such as DNA hybridization,29,30 chemical reaction,31-33 or hydrogen bonds.34,35 Recently, in order to construct a surface in which the structural architecture was closer to that of the cell wall, (17) Hanus, J.; Mazeau, K. Biopolymers 2006, 82(1), 59–73. (18) Yamatoya, K.; Shirakawa, M. Curr. Trends Polym. Sci. 2003, 8, 27–72. (19) Yan, H. W.; Lindstrom, T.; Christiernin, M. Nord. Pulp Paper Res. J. 2006, 21(1), 36–43. (20) Zhou, Q.; Rutland, M. W.; Teeri, T. T.; Brumer, H. Cellulose 2007, 14(6), 625–641. (21) Kontturi, E.; Tammelin, T.; Osterberg, M. Chem. Soc. Rev. 2006, 35(12), 1287–1304. (22) Podsiadlo, P.; Sui, L.; Elkasabi, Y.; Burgardt, P.; Lee, J.; Miryala, A.; Kusumaatmaja, W.; Carman, M. R.; Shtein, M.; Kieffer, J.; Lahann, J.; Kotov, N. A. Langmuir 2007, 23(15), 7901–7906. (23) Cranston, E. D.; Gray, D. G. Biomacromolecules 2006, 9(7), 2522–2530. (24) Wagberg, L.; Decher, G.; Norgren, M.; Lindstrom, T.; Ankerfors, M.; Axnas, K. Langmuir 2008, 24(3), 784–795. (25) Cranston, E. D.; Gray, D. G. Colloids Surf., A 2008, 325(1-2), 44–51. (26) Decher, G.; Hong, J. D.; Schmitt, J. 5th International Conf on LangmuirBlodgett Films, Paris, France, Aug 26-30, 1991; pp 831-835. (27) Decher, G. Science 1997, 277(5330), 1232–1237. (28) Quinn, J. F.; Johnston, A. P. R.; Such, G. K.; Zelikin, A. N.; Caruso, F. Chem. Soc. Rev. 2007, 36(5), 707–718. (29) Johnston, A. P. R.; Read, E. S.; Caruso, F. Nano Lett. 2005, 5(5), 953–956. (30) Johnston, A. P. R.; Mitomo, H.; Read, E. S.; Caruso, F. Langmuir 2006, 22(7), 3251–3258. (31) Kohli, P.; Blanchard, G. J. Langmuir 2000, 16(10), 4655–4661. (32) Kohli, P.; Blanchard, G. J. Langmuir 2000, 16(22), 8518–8524. (33) Liu, Y. L.; Bruening, M. L.; Bergbreiter, D. E.; Crooks, R. M. Angew. Chem., Int. Edit. Engl 1997, 36(19), 2114–2116. (34) Sukhishvili, S. A.; Granick, S. J. Am. Chem. Soc. 2000, 122(39), 9550–9551. (35) Sukhishvili, S. A.; Granick, S. Macromolecules 2002, 35(1), 301–310.
Langmuir 2010, 26(22), 17248–17255
Article
Jean et al.36 explored new multilayered films made of XG and cellulose that could be prepared using a layer-by-layer construction approach. In this case, the formation of multilayers is driven by hydrogen bonding and van der Waals interactions. The recent publication of these results prompted us to test and compare different ways of constructing multilayered films of cellulose and XG by spin-coating. Among the methods tested, spin-coating without a rinsing step was shown to be the most efficient procedure, as compared to spin-coating plus rinsing and dipping procedures with or without rinsing. The other parameter investigated was the concentration of the XG solutions. Semidilute polymer solutions were selected according to their rheological properties, i.e. unentangled (0.5 and 1 g/L) or entangled (5 and 10 g/L). Growth, thickness and inner structure of the films varied with the XG concentration regime and were characterized by mechanical profilometry and neutron reflectivity. Film growth was linear for the lowest XG concentrations, whereas a plateau value was reached, with concentrations corresponding to the entangled regime, after deposition of a few bilayers. The interaction mechanisms occurring between CN and XG as the multilayer built up are discussed.
Materials and Methods Materials. Poly(L-lysine) (PLL40000-60000 g/mol) was used for the anchoring layer. It was provided by Sigma-Aldrich and used without further purification. Deionized water (18.2 MΩ, Millipore Milli-Q purification system) was used, without pH adjustment, to prepare aqueous PLL solutions at a concentration of 0.5 g/L. XGs from Tamarindus indica were provided by Dainippon Pharmaceutical Corporation, Osaka, Japan, and purified according to Gidley’s method.37 The weight-average and number-average molar masses of XGs, determined by size exclusion chromatography coupled with laser light scattering, were Mw = 670 000 g/mol and Mn = 394 000 g/mol, respectively (columns, PL aquagel OH Guard - PL aquagel OH 60 - PL aquagel OH 40; solvent, NaNO3 0.1 M þ 0.02% NaN3; flow rate, 0.7 mg/mL). The gyration radius was 51 nm. All solutions were filtered prior to use. Preparation of Sulfated CN. The protocol was derived from Revol et al.38 Cotton linters from Whatman 20CHR filter paper were redispersed in water by homogenizing for 2 min in a Waring Blender at maximum speed. Hydrolysis was carried out with 50% (w/w) sulfuric acid at 70 C for 40 min with constant stirring. Immediately after acid hydrolysis, the reaction was quenched by diluting the suspension 10-fold with deionized water. The suspension was centrifuged at 10 000 rpm for 10 min to concentrate the cellulose and remove excess aqueous acid. The resulting precipitate was repeatedly rinsed and recentrifuged until a colloidal suspension was obtained. Subsequent dialysis against water until the pH matched that of deionized water within 1 unit required about 1 week. Mixed-bed research-grade resin (Sigma TMD-8) was added to the cellulose nanocrystal (Whiskers) suspension for 48 h to remove sulfate ester groups residual counterions. The resulting aqueous suspension contained approximately 1% cellulose by weight. Sodium azide (0.01%) was added to the opalescent suspension. The length of nanocrystals was determined by image analysis of the TEM images using Image J software. The lengths (36) Jean, B.; Heux, L.; Dubreuil, F.; Chambat, G.; Cousin, F. Langmuir 2009, 25(7), 3920–3923. (37) Gidley, M. J.; Lillford, P. J.; Rowlands, D. W.; Lang, P.; Dentini, M.; Crescenzi, V.; Edwards, M.; Fanutti, C.; Reid, J. S. G. Carbohydr. Res. 1991, 214(2), 299–314. (38) Revol, J. F.; Bradford, H.; Giasson, J.; Marchessault, R. H.; Gray, D. G. Int. J. Biol. Macromol. 1992, 14(3), 170–172.
DOI: 10.1021/la102614b
17249
Article
of about 400 nanocrystals were measured. The number average length (Ln), length-weighted average length (Ll), and length polydispersity index (PL) were, respectively, Ln = 215 nm, Ll = 243 nm, PL = 1.13. The CN charges were determined by conductometric titration with 0.1 mM NaOH and were found to be equal to 0.04 e/nm2. Viscosity measurements were performed over a range of shear rates from 1 10-4 up to 100 s-1 for concentrations of xyloglucan solutions ranging from 0.01% to 1% (w/v) at 20 C. The xyloglucan solutions were filtered prior to use. Newtonian viscosities were obtained by extrapolation. The instruments used to cover the concentration range were a Contraves Low Shear 40 viscosimeter (Couette geometry) and a RFS II rheometer (coneplate geometry). The value of the intrinsic viscosity was determined by extrapolation at infinite dilution from measurements on dilute solutions using an Ostwald-type capillary viscosimeter (AVS 310, Schott Ger€ate). Quartz Crystal Microbalance (QCM). Measurements were obtained with a Maxtek Research QCM Apparatus. The sensing element was a 1 in. disk-shaped, AT-cut piezoelectric quartz crystal sandwiched between two gold electrodes, provided by Maxtek. The crystal was excited at its fundamental resonant frequency (F0 ≈ 5 MHz) which changed as mass was deposited onto, or depleted from, the gold sensing surface. The quartz crystals were first cleaned in a piranha solution H2SO4:H2O2 (7:3) for 30 min, then rinsed with deionized water and dried under N2. PLL was spin-coated on the gold quartz surface followed by CN deposition (using the procedure described in the next section). The crystal was then clamped in the holder (CHC-100, Maxtek) and immersed in deionized water for appropriate frequency adjustments before XG solution (0.5, 1, 5, and 10 g/L). The adsorption times on the bottom layer of CN were monitored for 1-12 h depending on the XG solutions studied. Multilayered Film Preparation. Both spin-coated and dipped multilayer films were assembled on single-sided polished quarters of 2 in. silicon Æ100æ wafers obtained from ACM (Villiers Saint Frederic, France). The wafers were first cleaned in a piranha solution for 30 min, then repeatedly rinsed with purified water. The samples were prepared immediately after substrate cleaning. Spin-Coating Procedure. Spin-coated films were produced with a homemade spin coater. The CN suspension (2 mL) was poured on a wafer previously coated by PLL, which was then accelerated after 5 min of adsorption (acceleration speed: 360 rpm/s) and spun at 3600 rpm for 40 s. The XG solution (2 mL) was poured directly on the cellulose layer and spun (3600 rpm, acceleration 360 rpm/s), following the same procedure. This procedure was performed with or without intermediate rinsing steps with deionized water. A bilayer was defined as a single CN deposition step followed by an XG deposition step. Dipping Procedure. The wafer was dipped in the PLL bath (0.5 g/L) for 10 min and rinsed with deionized water to obtain the cationic anchoring layer. The wafers were then dipped for 5 min in the CN suspension (1 g/L) and in the XG solutions (0.5, 1, 5, or 10 g/L) to be deposited, with or without intermediate rinsing steps with deionized water. This sequential dipping procedure was repeated 8 times. After the last rinsing step, all the samples were dried under a stream of N2. Mechanical Profilometry. Thicknesses were obtained by measuring the groove depth on a surface scratched with a razor blade. The step measurements of samples were obtained with a stylus-based mechanical profilometer (Dektak 8, Veeco). The scratched surfaces were scanned along multiple straight lines using a 2.5 μm-radius hemispherical tip carbide stylus and a contact 17250 DOI: 10.1021/la102614b
Cerclier et al. Table 1. Thicknesses of Films Measured after 8 Bilayers Deposition and Thicknesses per Bilayer as a Function of XG Concentration and the Process Useda preparation method
concentration (g/L)
thickness (nm)
dipping with rinsing
0.5
