Supporting Information

Transcription

1 Supporting Information Metal Ion Mediated Cellulose Nanofibrils Transient Network in Covalently Cross-linked Hydrogels: Mechanistic Insight into Morphology and Dynamics Jun Yang *, Feng Xu, Chun-Rui Han Beijing Key Laboratory of Lignocellulosic Chemistry, Beijing Forestry University, Beijing, , China * Correspondent author: yangjun11@bjfu.edu.cn. Tel: Contents Table S1 Mechanical parameters of hydrogels Table S2 Nanoindentation results Figure S1 Morphology of CNF by TEM Figure S2 POM images of hydrogels Figure S3 Dynamic rheology properties of hydrogels Figure S4 Scheme of resting time dependent rheology Figure S5 Tensile properties and TEM of hydrogels after EDTA treatment Figure S6 Cyclic loading-unloading curves at various strains Figure S7 Scheme of force-displacement in nanoindentation measurement Figure S8 Influence of holding time on creep Figure S9 Digital images of notched hydrogels during deformation Figure S10 SEM images of hydrogels at different valences Cryo-TEM for morphological evolution in Figure 9 s-1

2 Table S1 Mechanical properties of IPN hydrogels (25, strain rate s -1 ) Sample Fracture strength Fracture elongation Young s modulus Fracture energy (kpa) (%) (kpa) (kj/m 3 ) PAAm 37 ± ± ± ± 0.12 PAAm-CNF ± ± ± ± 4.3 PAAm-CNF-0.5-Ca ± ± ± ± 6.1 PAAm-CNF-0.5-Zn ± ± ± ± 5.7 PAAm-CNF-0.5-Al ± ± ± ± 6.3 PAAm-CNF-0.5-Cs ± ± ± ± 6.4 Table S2 Mechanical results of nanoindentation tests Sample Elastic modulus a) Hardness Plasticity index (GPa) (GPa) PAAm 0.21 ± ± ± PAAm-CNF ± ± ± PAAm-CNF-0.5-Ca ± ± ± PAAm-CNF-0.5-Zn ± ± ± PAAm-CNF-0.5-Al ± ± ± PAAm-CNF-0.5-Cs ± ± ± a) The modulus data by nanoindentation and rheology have the same tendency but different absolute values. This discrepancy may stem from the different frequency and loading direction applied for the measurements, where the loading direction on the samples are along with compression molding direction and perpendicular to the compression molding direction for indentation and tensile test, respectively. s-2

3 Extraction of Cellulose Nanofibrils The cellulose pulp suspension (5 wt%) was mechanically stirred for 24 h. Subsequently, the suspension was homogenized using an Ultra-Turrax (IKA) at 12,000 rpm for 1 h. This initially refining step was performed to increase accessibility of cell wall for the following mechanical shearing treatment. The pulp slurry (3 wt%) was pumping into a high-pressure fluidizer with two different Z-shaped chamber pairs. First, the slurry was passed into a chamber pair with a diameter of 400 and 200 µm for 16 passes at 110 MPa, then passed through a chamber pair with a diameter of 200 and 75 µm for 12 passes at 170 MPa. The mixture was sonicated for 10 min (300 W) and converted to gel-like suspension with a solid content of 1.2 wt%. TEMPO-Mediated Oxidation of Cellulose Nanofibrils Conversion of CNF to COOH-CNF, by oxidation of primary alcohol groups to carboxylic acid groups, was accomplished by the procedure reported in literature. 1 About 5 g of CNF was suspend in deionized water (400 ml) containing TEMPO (2,2,6,6-Tetramethylpiperidine-1-oxyl, C9H18NO) g, 0.4mM) and NaBr (0.4 g, 4 mm) and homogenized by ultrasonic treatment at an ice-water bath for 10 min. Then certain amount of NaClO aqueous solution (10 wt%, 5 mm per gram of cellulose) was slowly added to the suspension to start oxidization at room temperature with mechanical stirring (300 rpm). The ph of the mixture was maintained at 10 by adding 0.5 M NaOH throughout the reaction. The reaction was quenched by adding ethanol (10 ml), and 0.5 M HCl was slowly dropped into the mixture to adjust ph to 7. Finally, the oxidized CNF was throughout dialyzed against water. The TEMPO-oxidized CNF has a carboxylic group content of 1.47 mmol/g, which was determined by conductivity titration. The average dimensions of CNFs were measured to be 45 ± 5 nm wide 600 ± 40 nm long with TEM observation (Figure S1). Figure S1. TEM image and dimension distribution histograms of CNFs. s-3

4 Polarized optical microscopy (POM) was performed using an Olympus System Microscope, BX41, where the composite gels were initially confined between two glass slides and then immersed into metal chloride solution for 2 h before observation. Figure S2. POM images of hydrogels ( a, a for PAAm-CNF-0.5, b, b for PAAm-CNF-0.5-Zn 2+, and c, c for PAAm-CNF-0.5-Al 3+ ). The CNFs in composite gels and ionic gels exhibit the featured pre-cholesteric structure and marble-like texture, respectively. s-4

5 Figure S3. Dynamic frequency sweep of hydrogels (red for PAAm-CNF-0.5, blue for PAAm-CNF-0.5-Zn 2+, pink for PAAm-CNF-0.5-Ca 2+, green for PAAm-CNF-0.5-Al 3+, and wine for PAAm-CNF-0.5-Cs 3+ ). To further examine the time-dependent rheological recovery, resting time evolution of frequency sweep is performed to check if the mechanical testing is in the linear region. The protocol of this test is described in Figure S4 (a) and the resting time dependent storage moduli (G ) and loss moduli (G ) are present in Figure S4(b). The results show that the cation mediated CNF bundle networks build up some internal stress and relax towards equilibrium due to the transient ionic cross-linking associations. Figure S4. (a) Scheme of shear-rest time measurement and (b) resting time dependent rheological properties of PAAm-CNF-0.5-Zn 2+ as a function of angular frequency (1% strain) (G solid, G open, black for initial moduli, red for moduli after resting 10 2 s, green for moduli after resting 10 3 s, and blue for moduli after resting 10 5 s). s-5

6 Figure S5. Tensile stress-strain curves of PAAm-CNF-0.5-Al 3+ gels after EDTA treatment (0.1 M, 4h) and its TEM image. Figure S6. Tensile loading-unloading test of (a) PAAm-CNF-0.5 and (b) PAAm-CNF-0.5-Al 3+ under different strains. Modulus, hardness, and plasticity by nanoindentation (1) Modulus The elastic modulus was calculate by fitting the unloading curve to a powder law relation (Oliver-Pharr equation 2 ), and the slope of the initial part of the unloading curve gives the stiffness (S) dp S= dh where p is the applied load and h is the displacement. According to Sneddon et al., 3 the indentation of an elastic half space by any punch can be described as a solid of revolution of a smooth function, a geometry becomes independent with contact stiffness, contact area. Thus, the elastic modulus can be derived by following equation S = A β π 2 Er s-6

7 where β is a constant depending on tip geometry (β =1.034 for Berkovich indenter), A is the projected area of indentation. For a Berkovich indenter, the relation between A and the depth h can is given by The elastic modulus E r is defined by A= 24.5 h υ 1-υ ( ) specimen + ( i = ) E E E r 2 i indenter The contribution of elastic modulus comes from the specimen, with elastic modulus E and Poisson s ratio, v, and the indenter, with elastic modulus and Poisson s ratio of E i (1140 GPa) and v i (0.07), respectively. (2) Hardness Hardness (H) is defined as the indentation load divided by the projected area of the indentation, and can be calculated from the load-displacement curves: H = where A is the projected contact area. For a given indenter (Berkovich tip in this work), the projected contact area is a function of contact depth. P max A (3) Plasticity index According to Briscoe et al., 4 the plasticity index (χ), a parameter to characterize the relative plastic/elastic behavior of the material when it undergoes external stress, was determined as follows: A1 χ = A + A 1 2 where A 1 is the area encompassed between the loading and unloading curves (corresponds to the plastic work dome during the indentation), and A 2 is the area encompassed by the unloading curve (corresponds to viscoelastic recovery). Thus, it follows χ = 0 (A 1 =0) for an ideal elastic deformation, 0<χ<1 for a viscoelastic deformation, and χ =1 (A 2 =0) for an ideal plastic deformation. s-7

8 Figure S7. Schematic illustration of a typical force-displacement of a nanoindentation test. P max = maximum load force, h max = maximum penetration depth, h f = permanent penetration depth Effect of holding time Polymers tend to show creep at the load due to their unique viscoelasticity 5 (e.g. as a result of the diffusion and motion of atoms or movement of dislocations in the indentation stress field), the holding segment at the peak load is necessary to allow for dissipation of creep displacement. Since the indentation creep rate decreases with increasing holding time under the peak load, the effect of indentation creep can be minimized by providing enough holding time. Thus, find an optimum hold time is important for creep behavior. In this study, the holding time intervals of s are considered and the typical force-displacement-time curves for 25 and 40 s are illustrated in Figure S8 (PAAm-CNF-0.5-Al 3+ ). One can note that the displacement of nanoindenter tip occurs at constant maximum load for few seconds and after that the displacement is almost constant for subsequent time at the peak load. The point at which the creep displacement diminishes implies the minimum time for the hold segment, the peak load holding time of 25 s is selected as optimum time to avoid the creep influence of the unloading curve that is used to calculate the elastic modulus. Figure S8. Loading and corresponding displacement for a holding time of (a) 25 s and (b) 40 s. s-8

9 Figure S9. Photos of uniaxial tensile test for the PAAm-CNF-0.5-Al 3+ with a notch. With increasing in strain, the single notch gradually transform to many smaller size cracks (represent by red dash curves). This phenomenon indicates that the disruption of the ionic coordination effectively dissipates energy around the crack and transfers the stress to neighboring large process zone, leading to the decreased stress concentration and thus increases crack propagation resistance. Figure S10. SEM images of fracture morphology of (a) PAAm-CNF-0.5-Ca 2+ PAAm-CNF-0.5-Cs 3+. and (b) s-9

10 Cryo-TEM measurement in Figure 9 Cryo-TEM is a powerful technique for obtaining in situ images of gels micro-morphology evolution. The hydrogels were stretched uniaxially using Zwick Z005 under different strains and fixed with binder clips to maintain their elongated state. The stretched samples were then cryo-fractured by immersing into liquid nitrogen. Ultrathin sections ( 100 nm) were cut using a Leica cold knife at 80 C. The sections were collected on carbon grids and transferred to a cryoholder (D626, Gatan, Inc.) without any further modification, and then the vitrified sample was observed and imaged using a JEOL-2010 microscope at 140 C. References (1) Saito, T.; Kimura, S.; Nishiyama, Y.; Isogai, A. Cellulose Nanofibers Prepared by TEMPO-Mediated Oxidation of Native Cellulose. Biomacromolecules 2007, 8, (2) Oliver, W. C.; Pharr G. M. An Improved Technique for Determining Hardness and Elastic Modulus Using Load and Displacement Sensing Indentation Experiments. J. Mater. Res.1992, 7, (3) Sneddon, Ian N. The Relation Between Load and Penetration in the Axisymmetric Boussinesq Problem for A Punch of Arbitrary Profile. Int. J. Eng. Sci. 1965, 3, (4) Briscoe, B. J.; Fiori, L.; Pelillo, E. J. Nano-indentation of Polymeric Surfaces. J. Phys. D: Appl. Phys. 1998, 31, (5) Oyen, M. L. Sensitivity of Polymer Nanoindentation Creep Measurements to Experimental Variables. Acta Mater. 2007, 55, s-10