Chemically Cross-Linked Cellulose Nanocrystal Aerogels with Shape

Subscriber access provided by UNIV OF NORTH CAROLINA

Article

Chemically Cross-Linked Cellulose Nanocrystal Aerogels with Shape Recovery and Superabsorbent Properties Xuan Yang, and Emily Dawn Cranston Chem. Mater., Just Accepted Manuscript • DOI: 10.1021/cm502873c • Publication Date (Web): 24 Sep 2014 Downloaded from http://pubs.acs.org on September 25, 2014

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Chemistry of Materials is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemistry of Materials

Chemically Cross-Linked Cellulose Nanocrystal Aerogels with Shape Recovery and Superabsorbent Properties Xuan Yang, Emily D. Cranston* Department of Chemical Engineering, McMaster University, Hamilton, Canada L8S 4L7 *To whom correspondence should be addressed. E-mail: [email protected].

ABSTRACT Cellulose nanocrystals (CNCs) are entering the marketplace as new high strength nano-additives from renewable resources. These high aspect ratio particles have potential applications as rheological modifiers, reinforcing agents in composites, coatings, and porous materials. In this work, chemically cross-linked CNC aerogels were prepared based on hydrazone cross-linking of hydrazide and aldehyde-functionalized CNCs. The resulting aerogels were ultra-lightweight (5.6 mg/cm) and highly porous (99.6%) with a bimodal pore distribution (mesopores 1 µm). Chemically cross-linked CNC aerogels showed enhanced mechanical properties and shape recovery ability, particularly in water, compared to previous reports of physically cross-linked CNC aerogels. Specifically, the aerogel shape recovered more than 85% after 80% compression, even after 20 compress and release cycles. These CNC aerogels can

ACS Paragon Plus Environment

1


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 34

absorb significant amounts of both water (160 ± 10 g per g aerogel) and dodecane (72 ± 5 g per g aerogel) with cyclic absorption capacity. We demonstrate that CNC aerogels can be used as superabsorbents and for oil/water separations and they may also find application as insulating or shock-absorbing materials. The cross-linking technology developed here presents new ways to design CNC networked structures and suggests an alternate route to incorporate CNCs into matrix materials, such as epoxies and foams. KEYWORDS: cellulose nanocrystals; chemical cross-links; aerogels; shape recovery; superabsorbent

INTRODUCTION Aerogels are highly porous materials derived from gels, in which the liquid component of the gel has been replaced with a gas. Currently, aerogels based on silica are the most widely used due to their high porosity (up to 99%), low density (4 to 500 mg/cm3) and large surface area (100 to 1000 m2/g).1 Applications of aerogels include catalysts,2,3 temperature and sound insulation,4,5 and electronics,6 however, aerogels based on silica often suffer from fragility which limits their use when mechanical robustness is needed. Aerogels from synthetic or natural polymers have been proposed to overcome the challenge of fragility, since polymers are generally less brittle, more flexible and easy to process/handle.7 Compared to synthetic polymers derived from declining petrochemical resources, renewable biopolymers like cellulose, are an economic and environmentally-friendly alternative. Specifically, nanocelluloses show particular promise for use in aerogel materials due to their high strength, low density, liquid crystalline properties, biodegradability and general biocompatibility.8–11 The nanocellulose family is a new class of sustainable nanomaterials which


2

Page 3 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


includes nanofibrillated cellulose (NFC), bacterial cellulose (BC) and cellulose nanocrystals (CNCs).12 Examples of nanocellulose aerogels in the literature generally focus on NFC because the nanofibers are many microns long and possess regions of crystalline and amorphous cellulose making them very flexible and easily entangled.13–23 NFC aerogels based on physical crosslinking have been proposed as oil absorbents, filters, and antibacterial substrates. Similarly, aerogels based on BC have been prepared and have potential applications as supercapacitors, catalyst supports and sensors.24–26 This work focuses on CNCs which are highly crystalline rigid rod-like particles that are 5-10 nm in cross-section and 100-200 nm long.27,28 CNCs have recently gained attention in the media (and in many new patents and scientific articles) due to their industrial production and commercial availability in Canada, USA and Europe.29,30 Compared to NFC and BC, CNCs have enhanced mechanical properties due to their high crystallinity and are the only member of the nanocellulose family approved as safe by various standards.31,32 Thus aerogels based on CNCs are promising to be used when strength/stiffness is required and in biomaterials. The challenge of creating aerogels from CNCs stems from the inability to form effective entanglements between individual rigid particles. To use CNCs in gel nanomaterials (both wet and dry) there are essentially three possible routes: (1) use CNCs as a nano-filler within a polymer matrix to give a reinforced gel,33–35 (2) allow CNCs to self-assemble into liquid crystalline structures which can act as substrates or templates for gels,36–38 or (3) devise a way for CNCs alone to form a networked structure through entanglements of CNC agglomerates or sheets,39–41 or by cross-linking individual CNC particles with physical cross-links (hydrogen bonding and ionic interactions),42 or by chemically cross-linking, as demonstrated here.


3


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 34

To the best of our knowledge, only a few examples of aerogels comprised of CNCs only (i.e., with no other matrix components) exist in the literature and they take a different approach from our chemical cross-linking method which relies on surface modified CNCs, freezing and critical point drying. The three examples are briefly summarized here: Heath et al.39 produced CNC aerogels from CNC hydrogels by critical point drying. Their method employed a large starting concentration of CNCs (>8 wt. %) and high power sonication to create the initial hydrogel structure. Their aerogels displayed impressive properties such as low density (78 mg/cm3) and high specific surface area (605 m2/g). Next, Dash et al.40 prepared a CNC aerogel by directional freeze casting of CNC suspensions. These aerogels were based on sheets of CNCs which assembled into a networked structure where the pore size was dependent on the size of the ice crystals formed during the freezing procedure (pore diameter >5 µm). Finally, Fumagalli et al.41 reported a CNC aerogel prepared by freeze-drying a CNC suspension in a water and tert-butanol mixture. Afterwards, their CNC aerogels were functionalized through a gas-phase palmitoylation reaction and the CNCs were then re-dispersed into organic solvents. The aerogel was thus used as a high surface area solid to enable hydrophobization of CNCs. In these three cases, mechanical properties and the ability to use these aerogels in liquid environments was not discussed. We have reasonable concern that CNC aerogels based on hydrogen bonding and physical cross-links may suffer in mechanical performance and may permanently collapse or redisperse in water. Furthermore, these aerogels have macropores which suggests that some of the advantages of using nano-sized CNCs as a starting material are lost. In this paper, we present the first chemically cross-linked CNC aerogels which possess both meso and macropores. The CNCs are modified through straightforward water-based reactions to have hydrazide and aldehyde surface functionality allowing for hydrazone cross-link bonds to


4

Page 5 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


form when orthogonally modified CNCs come into contact. Hydrazone cross-linking chemistry stands out because it has a fast gelation time, low toxicity, leads to materials with good mechanical properties, and does not require any type of external stimulus or initiator for bond formation. 43–45 The overall goal of this work was to introduce a new cross-linking method for CNCs and create networked nanocellulose materials with tailorable mechanical performance and novel functionality. We demonstrate that these ultra-lightweight, highly porous and flexible aerogels show superabsorbent capacities and shape recovery abilities. The development of cross-linked CNC structures, and the ability to extend this technology to cross-link CNCs within other materials, may open new application potential for CNCs in protective/sports equipment, construction and insulation, energy/gas production and storage, marine applications, personal care, sensors, and reinforced epoxies, to name just a few.

EXPERIMENTAL SECTION Materials. Whatman cotton ashless filter aid was purchased from GE Healthcare Canada, and sulfuric acid (95 ~ 98%) was purchased from VWR Canada. Adipic acid dihydrazide (ADH, 98%), N-hydroxysuccinimide (NHS, 97%), N′-ethyl-N-(3-dimethylaminopropyl)-carbodiimide (EDC, commercial grade), ethylene glycol (99.8%), silver(I) oxide (≥99.99% trace metals basis), sodium periodate (NaIO4, >99.8%), nile blue dye (≥75%) and dodecane (≥99%)

were all

purchased from Sigma Aldrich (Oakville, ON, Canada). Sodium hydroxide was purchased from EMD Millipore Germany, and hydrogen chloride was purchased from LabChem Inc. Anhydrous ethyl alcohol (EtOH) was purchased from Commercial Alcohols Inc. All water used was purified Type I water with a resistivity of 18.2 MΩ cm (Barnstead NANOpure DIamond system, Thermo


5


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 34

Scientific, Asheville, NC). Dimethyl sulfoxide (DMSO, reagent grade) was purchased from Caledon Laboratory Chemicals (Georgetown, ON, Canada). Preparation of Cellulose Nanocrystals. CNCs were obtained from the acid hydrolysis of cotton ashless filter aid using 64 wt.% sulfuric acid at 45 °C for 45 min, as described previously.47 The CNC precipitate was then rinsed, centrifuged and dialyzed until constant neutral pH in the effluent was achieved. CNC suspensions were sonicated (Sonifier 450, Branson Ultrasonics, Danbury, CT) and placed over acidic cation exchange resin (Sigma-Aldrich, Milwaukee, WI). The final suspension was ca. 1 wt. % cellulose in water with a pH of 2.8 and was used in its acid form. The sulfuric acid hydrolysis reaction introduces sulfate half-esters (OSO3- groups) onto the surface of the CNCs, OSO3- density was measured to be 0.22 ± 0.01 mmol/g by conductometric titration with 0.05 M NaOH (Supporting Information, Figure S1). The average particle dimensions for CNCs were measured to be 133 × 5 nm from TEM images. Our lab-made CNCs are virtually indistinguishable from industrially produced CNCs from wood pulp and cotton using sulfuric acid hydrolysis, however other sources of cellulose, such as bacteria or tunicin produce higher aspect ratio particles and adjusting the preparation method/conditions can also lead to varied nanoparticle properties. CNC suspensions were concentrated by evaporation under ambient conditions and the weight-percent concentration was measured by gravimetric analysis. Sulfuric acid-hydrolyzed CNCs were used to prepare the aldehyde-modified CNCs as described below. Cellulose nanocrystals with surface carboxyl groups (COOH-CNCs) and no sulfate half-esters were supplied by BioVision Inc. (New Minas, NS, Canada). BioVision’s COOH-CNCs were prepared by oxidation of cotton with ammonium persulfate.46 The average crystal dimensions of COOH-CNCs is 153 nm × 4 nm from TEM images, and the COOH density was measured to be


6

Page 7 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


0.31 ± 0.04 mmol/g by conductometric titration with 0.05 M NaOH (Supporting Information, Figure S2). The commercially obtained COOH-CNCs were used to prepare the hydrazidemodified CNCs as described below. Aldehyde-Modified CNC Synthesis (CHO-CNC). NaIO4 (0.3g) was dissolved in 1 wt. % CNC aqueous suspension (50 mL). The mixture was allowed to stir for 2 h at room temperature before ethylene glycol (0.15 mL) was added quickly to the suspension. The suspension was dialyzed as described above, and stored in suspension format at 4 °C. Hydrazide-Modified CNC Synthesis (NHNH2-CNC). ADH (0.6 g) was dissolved in 1 wt.% COOH-CNC (oxidized CNCs from BioVision Inc.) aqueous suspension (200 mL). NHS (0.014 g) was suspended in a 1:1 DMSO:H2O solution (1 mL). EDC (0.06 g) was dissolved in a 1:1 DMSO:H2O solution (1 mL). NHS solutions and EDC solutions were added dropwise to the flask sequentially. The pH of the NHNH2-CNC suspension was then adjusted to pH 6.8 using NaOH and HCl solutions until the pH of the solution no longer rapidly changed. The suspension was dialyzed as described above, and stored in suspension format at 4 °C. Degree of Functionalization. The amount of hydrazide functionality on NHNH2-CNCs was quantified by conductometric titration. The COOH groups on CNCs and NHNH2-CNCs were titrated using 0.05 M NaOH to indirectly determine the amount of -NHNH2 groups that had reacted with the CNCs. The number of aldehyde groups on CHO-CNCs was determined by conductometric titration after selectively oxidizing the aldehyde groups to carboxylic acid groups using silver (I) oxide, as reported by Campbell et al.48,49 Here CHO-CNCs (0.05 g) and NaOH (0.0248 g, 0.62 mmol) were dissolved in 10 mL of purified water, and silver (I) oxide (0.193 g) was added to the solution which was allowed to stir overnight. 5 mL of the oxidized reaction


7


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 34

mixture was diluted with purified water (80 mL) and 1 M HCl (0.85 mL). Finally, the solution was titrated using 0.05 M NaOH. Preparation of Aerogels. 1: Vortex Method. CHO-CNCs (5 mL) and NHNH2-CNCs (5 mL), in suspension at the same concentration (0.5, 1.0, 1.5 or 2.0 wt. %), were mixed in a glass vial (20 mL) using a Vortex mixer for 2 min (Level 8, Analog Vortex Mixer, VWR, Canada). The hydrogels (mixtures) were transferred into a cylindrical glass vial (8 mm in diameter) and allowed to set for another 10 min before transfer into the freezer (-4 °C). After 12 hours, hydrogels were directly transferred into 100% ethanol at room temperature. Hydrogels were solvent exchanged to ethanol during the melting process without changing the size and shape. Finally, after changing the ethanol daily for 5 days, aerogels were obtained by drying in a critical point dryer. 2: Blending Method. A double-barrel syringe was used to make the chemically cross-linked aerogels which was similar to our previous work.33 Barrel A and barrel B contained CHO-CNCs and CHO-CNCs, respectively, in suspension at identical concentrations (0.5, 1.0, 1.5 or 2.0 wt. %). The suspensions in barrel A and B were well mixed in the “blending area” of the double-barrel syringe and then were extruded out through the needle (tip 20G) into a cylindrical glass vial. The freezing, solvent exchange and critical point drying procedures are the same as described above for the “Vortex Method”. 3: Sonication Method. Aerogels prepared using the sonication method follow the “Vortex Method” procedure except that mixing the CHOCNC and NHNH2-CNC suspensions was performed under sonication for 5 or 15 min (35 kHz, 144W, Ultrasonic bath, VWR). Particle Sizing of Sol Suspensions. 0.5 wt. % of CHO-CNC and NHNH2-CNC suspensions were mixed using the different mixing methods described. The particle size and distribution of


8

Page 9 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


each diluted suspension (0.1 wt. %) was measured using a Malvern Zetasizer Nano particle analyser. The temperature was maintained at 20 °C during all measurements. Turbidity of Sol Suspensions. 0.5 wt. % of CHO-CNC and NHNH2-CNC suspensions were mixed under Vortex for 2 min. The turbidity of each suspension (CHO-CNCs, NHNH2-CNCs, supernatant of the mixture and lower turbid portion of the mixture) was measured at a wavelength of 450 nm using the Beckman Coulter DU® 800 UV/Visible spectrophotometer. The temperature was maintained at 25 °C during all measurements. Atomic force microscopy (AFM). 0.1 wt. % of CHO-CNC, NHNH2-CNC suspensions or mixtures of these two suspensions (1:1, v/v) was spin coated onto Si wafers at 4000 rpm for 30 seconds. A Nanoscope IIIa MultiMode Scanning Probe Microscope with an E scanner (Bruker AXS, Santa Barbara, CA) was used in tapping mode to image the samples, using probes from Nano World (FMR type 2.8 N/m spring constant and 75 kHz resonance frequency). Density. Aerogel densities (ρ ) were determined by measuring the weight and volume of each individual aerogel. The weight of an aerogel was measured by an analytical balance (readability 0.0001 g, Analystical Balance, Mettler Toledo) and the dimensions of an aerogel were measured by a digital caliper at four different positions. Five samples were used for density determination for aerogels with the same starting CNC concentration. Porosity. Aerogel porosities (P) were calculated using the density of the aerogel (ρ ) and density of cellulose nanocrystals (ρ ) using equation (1), obtained from the simple rule of mixtures and assuming the gas density was negligible. P (%) = 100 × (1 − ρ / ρ )

(1)

Brunauer–Emmett–Teller (BET) Analysis. The BET surface area was determined by N2 adsorption and desorption using a Micrometric ADAP 2020 automated system (Micromeritics, USA). A sample was first degassed at 115 oC for 4 h prior to the analysis followed by BET


9


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 34

analysis at -196 oC. The BET surface area and the Barrett-Joyner-Halenda (BJH) pore size distribution were calculated from the obtained isotherms. Morphology and Nano-structure. A JEOL 7000F SEM (JEOL Ltd., Japan) was used to image aerogels. Aerogels were immersed in liquid nitrogen and fractured to image the inner sections. A 5 nm thick platinum coating was applied before SEM imaging. Mechanical Properties and Shape Recovery Ability. Aerogels with different CNC contents were tested using a compress and release routine on a Mach-1Mechanical Testing System (Biomomentum, Canada). Each aerogel was made into a cylinder shape with a diameter of 8 mm and a height of 10 mm. The tests were carried out either in open-air or purified water, the maximum compression strain (ε) was set at 0.2, 0.4, 0.6, 0.8 or 0.85, the speed of compressing and releasing was set at 1 mm/s. The shape recovery ability was assessed by calculating the shape recovery percentage using equation (2), in which the original height (h ) and final height (h ) were measured: Shape Recovery Percentage (%) =

h &h × 100%

(2)

Liquid Absorption Capacity. To determine the liquid absorption capacity of the CNC aerogels, different solvents were used including water, ethanol, DMSO and dodecane. A 0.5 wt. % CNC aerogel was submerged in a given solvent for 1 min until the absorbance reached the maximum loading. The weight of the aerogels before (Wbefore) and after (Wafter) solvent uptake were measured, and the absorption capacity was calculated according to equation (3): Absorption capacity (g/g) =

W,- &W .

(3)

Reusability. To check the reusability of aerogels as liquid absorbents, two different methods were used: (1) Compression: The absorbed solvent was removed by compressing the aerogels


10

Page 11 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


between parallel glass slides whereby >70% strain was applied. The weight of the aerogels after compression during each cycle was measured to ensure that more than 70% of the absorbed solvent was removed. (2) Evaporation: The solvent-saturated aerogel was removed from the solvent bath and dried in the fume hood until it reached the original dry weight. The weight of aerogels before solvent absorption, after absorption, and after removing the solvent was measured. A minimum of four samples and 20 cycles were tested for each experiment. Error Calculations. All error bars and error intervals shown are confidence intervals (Δx) calculated from the estimated small data set standard deviation (Sx) of repeat measurements (N), that is, Δx = Sx × t-value(N)1/2, where the t-value is the Student’s t-distribution at a confidence level of 95% for N−1 degrees of freedom for individual measurements (unless otherwise noted).

RESULTS AND DISCUSSION Characterization of Aerogel Components Hydrazone cross-linking occurs between hydrazide (or hydrazine) and carbonyl groups (i.e., ketones or aldehydes); these chemical functionalities were effectively added to CNCs through water-based chemical reactions. The degree of functionalization of hydrazide-modified CNCs (NHNH2-CNCs)

and

aldehyde-modified

CNCs

(CHO-CNCs)

was

characterized

by

conductometric titration to determine the total number of potential cross-link sites (Table 1). For NHNH2-CNCs, 44 ± 7 % of the carboxyl groups on carboxylated CNCs (COOH-CNCs) were converted to hydrazide groups, corresponding to 0.14 ± 0.02 mmol of NHNH2 groups per gram of NHNH2-CNCs. Aldehyde groups were generated by oxidation of sulfuric acid-hydrolyzed CNCs with sodium periodate resulting in CHO-CNCs with 0.9 ± 0.1 mmol of CHO groups per gram of CHO-CNCs. As CNCs are considered highly crystalline and generally intractable


11


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 34

particles, the functionalization is assumed to occur on the outer surface and a surface area per functional group can be calculated as indicated in brackets in Table 1. FTIR spectroscopy also supports the chemical functionalization of both types of modified CNCs, as shown in the Supporting Information (Figures S3 and S4).

Table 1. Dimensions and degree of functionalization of chemically-modified CNCs.

length × width* [nm]

aldehyde groups [mmol/g]

hydrazide groups [mmol/g] 0.14 ± 0.02

NHNH2-CNCs

155 × 4.1

—

(one group occupying ~7.4 nm2)

0.9 ± 0.1 CHO-CNCs

129 × 4.8

(one group occupying ~1.2 nm2)

—

*Average length and width measured from TEM images (Supporting Information, Figure S5) for >100 individual CNCs, typical size distribution is 100-260 nm × 2-7 nm for NHNH2-CNCs, and 60-220 nm × 2-9 nm for CHO-CNCs. Characterization of unmodified CNCs and COOHCNCs is presented in the Experimental Section and Supporting Information, Figures S1 and S2.

Aerogel Assembly The chemically cross-linked “all-CNC” aerogels were successfully prepared following a simple sol-gel process consisting of mixing, freezing, solvent exchange and critical point drying. No additional initiators, chemicals, or polymers were required. First, NHNH2-CNC and CHOCNC suspensions were mixed to form the initial sol whereby hydrazone cross-links form immediately upon mixing, creating small networked aggregates (Figure 1a). The sol suspension was then frozen to promote ice crystal growth (Figure 1b). The growing ice crystals push the cross-linked aggregates together to form a gel-like structure (Figure 1c),50,51 the ice crystals are


12

Page 13 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


removed through solvent exchange with ethanol, and the final aerogel is obtained by critical point drying (Figure 1d).

Figure 1. Schematic representation of chemically cross-linked CNC aerogels prepared by a solgel process: (a) formation of initial sol from NHNH2-CNCs and CHO-CNCs, (b) sol suspension, (c) gel formation with the growth of ice crystals and (d) aerogel formation after solvent exchanging and critical point drying. Different methods to mix the modified CNC suspensions were tested including Vortex mixing, sonication and double-barrel syringe blending. For samples prepared by sonication or doublebarrel syringe mixing, the gel-like structures shrink substantially (>50%) during freezing and/or disintegrate into pieces during the solvent exchange step. We believe that these mixing methods do not provide enough mixing for NHNH2-CNCs and CHO-CNCs to form efficient aggregates,


13


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 34

since the hydrazone cross-linking reaction needs the orthogonally-modified CNCs to come into close contact (sub-nanometer separations). We hypothesize that for CNCs at dilute concentrations (i.e., 0.5 wt. %), diffusion and light mixing is insufficient to bring enough particles into contact to form the large cross-linked aggregates required for robust aerogel formation. Dynamic light scattering (DLS) was used to track apparent particle and aggregate sizes of NHNH2-CNCs, CHO-CNCs and mixtures from the three mixing methods. (The term apparent is used in recognition that this is a Z-average hydrodynamic diameter obtained from diffusion coefficients, assuming spherical particles, in a salt-free suspension and does not give absolute dimensions.) The results show that the aggregate size from sonication ranges from 150 to 3500 nm with a distribution peak at 185 nm with 81% intensity. The aggregate size from double-barrel syringe blending ranges from 160 to 2800 with a distribution peak at 213 nm with 83% intensity. Unmixed NHNH2-CNC and CHO-CNC particle sizes are 141 nm and 135 nm, respectively; therefore most sol aggregates from these two mixing methods are small and only contain a few CNCs. On the contrary, the distribution peak of Vortex mixed samples is 1470 nm with 99.9% intensity, indicating that the aggregates are large and more uniform in size. Thus, the Vortex mixing method provides better mixing and allows for more efficient cross-linking, better gelation and a more extensive network structure in the final aerogel. We believe that hydrazone bonds are the main cross-links in our aerogels and are responsible for the network assembly; however, hydrogen bonding may play a secondary role. This is supported by DLS measurements which show no change in particle or aggregate size when “bare” CNCs, (i.e. unmodified sulfuric acid-hydrolyzed CNCs or COOH-CNCs) are mixed with either NHNH2-CNCs or CHO-CNCs. Furthermore, gels made with “bare” CNCs alone or “bare”


14

Page 15 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


CNCs with one type of modified CNCs break down during solvent exchange and thus cannot be used to prepare aerogels using our method. Interestingly, the sol suspension obtained by Vortex mixing orthogonally-modified CNCs is colloidally stable and does not show further aggregation or sedimentation over one month, as demonstrated by negligible changes in turbidity (Supporting Information, Figure S6). Atomic force microscopy (AFM) was used to image the structure of the modified CNCs (Figures 2a) and the sol suspensions (Figure 2b). Both NHNH2-CNCs and CHO-CNCs appear as shown in Figure 2a, whereby CNCs are drawn together and align due to capillary forces as water evaporates during AFM sample preparation. Conversely, the Vortexed sol does not allow CNCs to reorient/align because their structure is set by the chemical cross-links between CNC particles. Similar results were visible by transmission electron microscopy (TEM) as shown in the Supporting Information, Figure S5.

Density and Porosity The effect of CNC concentration on the density and porosity of CNC aerogels was investigated. Figure 3a shows that the average density of our aerogels ranges from 5.6 to 21.7 mg/cm3 for aerogels prepared from 0.5 to 2.0 wt. % CNC suspensions. These densities are considerably lower than previously reported values for other CNC aerogels (78-155 mg/cm3 from Heath et al.39 and 22-31 mg/cm3 from Dash et al.40). Figure 3b shows that the average aerogel porosity decreases linearly with the starting CNC concentration from 99.6% to 98.6%. Lines in Figure 3 are predictions calculated using the theoretical density52 and porosity and assuming no shrinkage during the solvent exchange and drying processes. As such, we estimate the shrinkage of our aerogels to be 11% which agrees with the reported range for nanocellulose


15


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 34

aerogels (6.5 to 30%).26,39 It is noteworthy that our starting CNC concentrations are much lower than for the work which reported shrinkage of 6.5% prepared from 8 wt. % NFC.39 We believe that both the large modulus8,11 of highly crystalline CNCs and the strong cross-links from covalent bonds can significantly prevent shrinkage. Remarkably, the final ultra-lightweight aerogel can stand on top of a dandelion without even bending the flower’s seed heads (Figure 3c).

Figure 2. AFM height images of dried CNC suspensions on silicon wafers: (a) CHO-CNCs and (b) a Vortexed mixture of orthogonally-modified CNCs.


16

Page 17 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3. (a) Aerogel density and (b) porosity as a function of starting CNC concentration. The lines are theoretical predictions for no shrinkage during solvent exchange and drying. (c) A low light photograph of a piece of CNC aerogel (prepared from a 0.5 wt. % CNC suspension) standing on top of a dandelion flower.


17


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 34

Specific Area and Pore Structure The specific surface area of the CNC aerogels was determined to be 250 ± 80 m²/g by Brunauer-Emmett-Teller (BET) analysis. This high surface area is similar to that reported by Heath et al. (78-605 m²/g).39 The surface area can be theoretically calculated to be 512 m²/g for individual CNCs assuming a length of 130 nm, width of 5 nm and density of 1.59 g/cm3.52 The overlap of CNC particles due to cross-linking, and the sheet-like structures that arise from freezing, will reduce the total specific surface area of the aerogels, making the measured value of 250 m²/g reasonable. We anticipated that our aerogels would have a bimodal pore distribution with micro or mesopores between CNC cross-links and macropores templated from ice crystal formation, as depicted in Figure 1. CNC aerogels from 0.5 wt. % CNC suspensions were found to have a mesoporous volume of 0.25 cm3/g, while the total calculated pore volume was 178 cm3/g. Thus mesopores make up only ~0.14% of the total pores. This is somewhat expected as CNCs occupy only 0.4% of the total aerogel volume, and are responsible for the cross-link-generated mesopores. Interestingly, the pore size distribution (Figure 4) from Barrett-Joyner-Halenda (BJH) analysis of BET data shows a single peak at 8 nm which indicates that our mesopores are uniform and small. Both chemical cross-linking between orthogonally-modified CNCs and the size uniformity of CNCs themselves are likely responsible for the uniform pore structure and minimal collapse of the aerogels during drying. In contrast, Heath et al. prepared physically cross-linked CNC aerogels which displayed two peaks at 4.3 and 15.5 nm,39 however, the origin of the two pore sizes is unclear.


18

Page 19 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 4. Mesopore size distribution in aerogels prepared from 0.5 wt. % CNC suspensions from BJH analysis based on pore area.

Aerogel Morphology Scanning electron microscopy (SEM) images of aerogels prepared from different starting concentrations of CNCs show the internal nano and micro-structure of the aerogels (Figure 5). All of the aerogels exhibit a similar internal pore structure with smaller pores (< 50 nm) and bigger pores (>1 µm) as expected from BET analysis and density/porosity measurements. Specifically, macropores are seen at low magnification (Figure 5, top), while the mesopores are seen by zooming in on the “walls” which form the macropores (Figure 5, bottom). Additionally, the mesopore structure is consistent with the TEM images of the sol suspension (Supporting Information, Figure S5). As the starting CNC concentration increases, aerogels tend to have more macropores with smaller sizes. However, the mesopore structure stays the same. We believe the mesopores are coming from the cross-links between CNC particles which are more or less independent of the starting CNC concentration. On the other hand, increasing the


19


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 34

concentration of CNCs means CNCs occupy more space in the mixture and ice crystals will be smaller which results in smaller macropores when the ice crystals are removed.

Figure 5. SEM images of aerogels with different starting CNC concentrations: (a, a’) 0.5 wt. %, (b, b’) 1.0 wt. %, (c, c’) 1.5 wt. % and (d, d’) 2.0 wt. %. Macropores are visible at low magnifications (top, a-d) and mesopore uniformity is observed as higher magnification (bottom, a’-d’).

Shape Recovery and Mechanical Properties The CNC aerogels are strong and remain intact when compressed in both air and liquid environments, which is superior to typically brittle silica aerogels which easily break under pressure.7 Figure 6a shows the compressive stress-strain behaviour of aerogels in air, where two regions are apparent: a slowly increasing stress response below 80% strain, and an exponentially increasing stress response above 80% strain. At 80% strain, the compressive stress of CNC aerogels prepared from 0.5, 1.0, 1.5 and 2.0 wt. % CNCs is 8.9, 11.6, 15.7 and 20.5 kPa,


20

Page 21 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


respectively. This is similar to NFC aerogels of Jiang et al.14 which exhibited a compressive stress of 8.5 kPa at 80% strain. Furthermore, at 95% strain, the compressive stress of each aerogel is 17.5, 26.8, 53.0 and 105.3 kPa. From these results, we believe that aerogels from higher starting concentrations of CNCs will have a larger proportion of chemically cross-linked parts (i.e., more “solid-like” parts compared to hollow parts) and thus they will exhibit higher stress at the same compressive strain compared to aerogels prepared from low starting concentrations. Figure 6b shows that water saturated CNC aerogels display excellent wet strength as well. The compressive stress is low below 80% strain, but monotonically increases from 80% to 95% strain. At 80% strain, the compressive stress of aerogels prepared from 0.5, 1.0, 1.5 and 2.0 wt. % CNCs is 0.8, 1.7, 2.2 and 3.7 kPa, respectively. These values are within range of other reports, such as 1.2 kPa for NFC aerogels14 and 12.1 kPa for carbon aerogels53 compressed 80% in water. At 95% strain, the compressive stress of each aerogel is 15.2, 26.6, 30.7 and 44.4 kPa which is more than 10 times larger than the stress at 80% strain.


21


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 34

Figure 6. Compressive stress-strain curves of aerogels prepared from 0.5, 1.0, 1.5 and 2.0 wt. % CNC suspensions, from 0 to 95% strain in (a) air and (b) water. The compressive stress of aerogels in water (i.e., aerogels reconstituted into cross-linked hydrogels) is much lower than in air. Water acts as a plasticizer and softens the hydrogel by forming hydrogen bonds between water molecules and CNCs, and disrupts some CNC-CNC hydrogen bonding.54 Combining the morphology results and the mechanical measurements of these aerogels, the main deformation appears to be primarily due to the bending, or even collapsing, of the macropores, implying that the resistance to compression stems from hydrogen


22

Page 23 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


bonds and physical cross-links. Compared with synthetic polymer aerogels which show a horizontal plateau region at low strain, our aerogels maintain their load-bearing capacity and show monotonically increasing compressive stress behaviour.55,56 At high strain, the aerogel deformation is mainly due to the bending or damage of the mesopores whereby the covalent bonds (hydrazone cross-links) are compressed or broken. In this case, due to the densification of the porous structure, the walls of the macropores start to touch each other and become the loading-bearing portion. The chemical cross-links “lock” the CNCs together more strongly and prevent deformation and fiber slippage which is the main failure mechanism in paper and fiberbased materials. Cyclic compression tests were undertaken to determine the shape recovery ability of CNC aerogels after 20 cycles in air or in water. Figure 7a shows that all CNC aerogels in air can return to ~90% of their original height after being compressed 20%, but the shape recovery percentage decreases with increasing compressive strain. Above 95% strain in air, the aerogels cannot fully recover their original shape. On the contrary, Figure 7b shows that all CNC aerogels in water recover at least 85% of their original size, and furthermore, the shape recovery percentage is above 50%, even at 95% strain. Specifically, the aerogels from 0.5 wt. % CNC suspensions can recover 85% of their original size after 95% compression which has not been shown previously for other nanocellulose aerogels. In fact, the literature shows that aerogels are generally not tested above 80% strain and show inferior shape recovery abilities even at these relatively low compressions.14,53


23


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 34

Figure 7. Shape recovery percentage of aerogels prepared from 0.5, 1.0, 1.5 and 2.0 wt. % CNC suspensions under different compressive strains after 20 cyclic compressions: (a) in air and (b) in water. The linear relationship between shape recovery % and strain in air implies that the ratio of broken to bent hydrogen bonds (plastic to elastic deformation) is constant and that the breakage may occur primarily in the compressive strain direction. We believe that water acts as a plasticizer and improves the recovery by helping to reform the hydrogen bonded network when


24

Page 25 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


the compressive strain is released. Aside from compressive testing cycles performed all in air or all in water, we observed that an aerogel compressed up to 80% in air recovers in water to the same % as if it was compressed in water. Conversely, if the sample is compressed in water and dried in its compressed state, it will not recover its shape in air or water when the strain is removed. A video demonstrating the shape recovery ability of CNC aerogels in water is included in the Supporting Information, Video 1. The cyclic compression tests (Figure 7) are consistent with the compressive stress-strain curves (Figure 6): the aerogels recover to their original height if compressive stress does not exceed the flat region of the stress-strain curve, i.e., when the chemical cross-links are not broken. It is also noteworthy that the recovery speed of aerogels in water is ~1 mm s-1 and the aerogels can return to their original height from 95% compression in under 11 seconds. This recovery is similar to NFC aerogels of Zhang et al.57 We believe that the unique pore structure in our cross-linked CNC aerogels aids in fast shape recovery; the mesopores absorb the water quickly due to capillary action and the macropores store and transfer the water.

Superabsorption Capacity Cellulose has an abundance of hydroxyl groups in its chemical structure which leads to an overall affinity for water and other polar liquids. CNC aerogels made from 0.5 wt. % CNC suspensions were used to determine the absorption capacity of aerogels in various solvents. This sample was chosen since it has the highest porosity and the best shape recovery ability. Figure 8 shows that 160 ± 10 g of water, 130 ± 10 g of ethanol or 134 ± 8 g of DMSO can be absorbed per gram of aerogel (on the first absorption cycle). Based on the calculated pore volume of 178 cm3 g-1, about 87%, 89% and 68% (v/v) of the total pores are filled with these three solvents,


25


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 34

respectively. Interestingly, 72 ± 5 g of dodecane (a hydrophobic oil) can also be absorbed, filling about 54% of the total pore volume. We believe that the capillary forces imparted by the mesopores of the aerogels help with the absorption of dodecane yet the larger pores are not fully filled because CNCs and dodecane are incompatible based on their surface energy. Interestingly, the aerogels formed stable “solvo-gels” upon soaking in different solvents, without noticeable volume change or structural disintegration. Furthermore, it takes less than 10 seconds for the aerogels to absorb the solvents tested to their maximum loading. This ultra-fast absorbing quality is another unique feature of our aerogels. Cyclic absorption capacity was tested to check the reusability of aerogels made from 0.5 wt. % CNC suspensions. Two desorbing methods were used: compressing the solvent out (Figure 8a) and letting the solvent evaporate (Figure 8b). By using the former method, the absorption capacity of aerogels slightly decreases which is consistent with the shape recovery tests. More precisely, the absorption capacity of aerogels decreases from 155 to 121 for water, 125 to 100 for ethanol, 134 to 111 for DMSO and 72 to 58 for dodecane (absorption capacity values are reported as g of solvent per g of aerogel). By the evaporation method, the absorption capacity of aerogels is very similar in ethanol, DMSO and dodecane, however, this is not the case for water. Water is not easily evaporated from the aerogels under ambient conditions and the absorption capacity decreases significantly from 155 to 18, on the 20th cycle. Strong hydrogen bonding of water to cellulosic materials is well known and furthermore, the hydration of the aerogels to hydrogels likely results in a rearrangement of the larger pores and some internal collapse of the structure. Ethanol, DMSO and dodecane appear to have less ability to disrupt the original aerogel structure and hydrogen bonding overall.


26

Page 27 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 8. Cyclic absorption capacity of aerogels prepared from 0.5 wt. % CNC suspensions for different solvents using different desorbing methods: (a) compressing the solvent out and (b) letting the solvent evaporate. Interestingly, the absorption preferences of CNC aerogels may give them the ability to separate oil and water. Figure 9 shows a series of images of an aerogel picking up water from a dodecane/water mixture, where water is denser and sits at the bottom of the vial. As an aerogel is immersed into the vial, it first absorbs dodecane, however, as it comes into contact with water,


27


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 34

the dodecane is displaced (in under 1 second) and the water-filled aerogel can be removed from the vial.

Figure 9. Chronological images of an aerogel picking up water from a dodecane/water mixture. Water has been dyed with blue color for easier visualization.

CONCLUSIONS We have presented a new type of chemically cross-linked “all-CNC” aerogel based on hydrazone cross-links between NHNH2-CNCs and CHO-CNCs. The final aerogel is ultralight and highly porous, surpassing other CNC aerogels reported to date. This CNC aerogel displays bimodal pore distribution as confirmed by BET analysis and SEM. We believe the mesopores come from the CNC cross-links and the macropores result from ice templating during the freezing process. These CNC aerogels shows great mechanical properties and shape recovery abilities, especially in water. Furthermore, water, ethanol, DMSO and dodecane are all quickly, and reversibly absorbed by the aerogels. Based on these results, chemically cross-linked CNC aerogels have potential use as superabsorbents and in oil/water separation technologies, among other traditional and nontraditional aerogel applications. Chemically cross-linked CNCs have shown enhanced mechanical properties over physically cross-linked nanocellulose materials and although our work focuses on hydrazone cross-link chemistry because it is a fast and easy process, other


28

Page 29 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


cross-linking chemistries may be used as well, and may bring about other benefits. This work is one step towards tailored nanomaterials that take advantage of modified CNCs that can impart new or enhanced functionality to commercial products. Supporting Information Available: Fourier transform infrared spectroscopy data and transmission electron microscopy for CNCs and modified CNCs, UV/Vis spectroscopy of CNC suspensions, video of cyclic compression testing of aerogels. This material is available free of charge via the Internet at http://pubs.acs.org.

ACKNOWLEDGEMENTS The authors thank Professors T. Hoare, R. Pelton, S. Zhu, M. Thompson, and J. Moran-Mirabal for access to equipment and expertise, and Marcia Reid, Michael Reid, G. de Silveira, E. Bakaic, S. Campbell and E. Gustafsson for sample analysis or training. Useful discussions with Dr. T. Abitbol, H. Marway, L. Graham, A. Palermo, Z. Hu, D. LeClair and S. Kedzior are gratefully acknowledged. Special thanks to BioVision Technology Inc. for the generous donation of COOH-CNCs. This work was supported in part by the Natural Sciences and Engineering Research Council of Canada and the Faculty of Engineering at McMaster University.


29


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 34

REFERENCES (1)

Hüsing, N.; Schubert, U. Angew. Chemie Int. Ed. 1998, 37, 22.

(2)

Pajonk, G. M. Appl. Catal. 1991, 72, 217.

(3)

Hirashima, H.; Kojima, C.; Imai, H. J. Sol-Gel Sci. Technol. 1997, 8, 843.

(4)

Reim, M.; Körner, W.; Manara, J.; Korder, S.; Arduini-Schuster, M.; Ebert, H.-P.; Fricke, J. Sol. Energy 2005, 79, 131.

(5)

Oh, K.; Kim, D.; Kim, S. Fibers Polym. 2009, 10, 731.

(6)

Lee, Y. J.; Kim, G.-P.; Bang, Y.; Yi, J.; Seo, J. G.; Song, I. K. Mater. Res. Bull. 2014, 50, 240.

(7)

Pääkkö, M.; Vapaavuori, J.; Silvennoinen, R.; Kosonen, H.; Ankerfors, M.; Lindström, T.; Berglund, L. A.; Ikkala, O. Soft Matter 2008, 4, 2492.

(8)

Moon, R. J.; Martini, A.; Nairn, J.; Simonsen, J.; Youngblood, J. Chem. Soc. Rev. 2011, 40, 3941.

(9)

Eichhorn, S. J. Soft Matter 2011, 7, 303.

(10)

Azizi Samir, M. A. S.; Alloin, F.; Dufresne, A. Biomacromolecules 2005, 6, 612.

(11)

Habibi, Y.; Lucia, L. A.; Rojas, O. J. Chem. Rev. 2010, 110, 3479.

(12)

Klemm, D.; Kramer, F.; Moritz, S.; Lindström, T.; Ankerfors, M.; Gray, D.; Dorris, A. Angew. Chemie Int. Ed. 2011, 50, 5438.

(13)

Jin, H.; Nishiyama, Y.; Wada, M.; Kuga, S. Colloids Surfaces A Physicochem. Eng. Asp. 2004, 240, 63.

(14)

Jiang, F.; Hsieh, Y.-L. J. Mater. Chem. A 2014, 2, 6337.

(15)

Zhang, Z.; Sèbe, G.; Rentsch, D.; Zimmermann, T.; Tingaut, P. Chem. Mater. 2014, 26, 2659.

(16)

Chen, W.; Yu, H.; Li, Q.; Liu, Y.; Li, J. Soft Matter 2011, 7, 10360.

(17)

Cervin, N.; Aulin, C.; Larsson, P.; Wågberg, L. Cellulose 2012, 19, 401.

(18)

Aulin, C.; Netrval, J.; Wågberg, L.; Lindström, T. Soft Matter 2010, 6, 3298.


30

Page 31 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(19)

Jin, H.; Kettunen, M.; Laiho, A.; Pynnönen, H.; Paltakari, J.; Marmur, A.; Ikkala, O.; Ras, R. H. A. Langmuir 2011, 27, 1930.

(20) Kettunen, M.; Silvennoinen, R. J.; Houbenov, N.; Nykänen, A.; Ruokolainen, J.; Sainio, J.; Pore, V.; Kemell, M.; Ankerfors, M.; Lindström, T.; Ritala, M.; Ras, R. H. A.; Ikkala, O. Adv. Funct. Mater. 2011, 21, 510. (21) Korhonen, J. T.; Kettunen, M.; Ras, R. H. A.; Ikkala, O. ACS Appl. Mater. Interfaces 2011, 3, 1813. (22)

Sehaqui, H.; Zhou, Q.; Berglund, L. A. Compos. Sci. Technol. 2011, 71, 1593.

(23)

Díez, I.; Eronen, P.; Österberg, M.; Linder, M. B.; Ikkala, O.; Ras, R. H. A. Macromol. Biosci. 2011, 11, 1185.

(24)

Haimer, E.; Wendland, M.; Schlufter, K.; Frankenfeld, K.; Miethe, P.; Potthast, A.; Rosenau, T.; Liebner, F. Macromol. Symp. 2010, 294, 64.

(25)

Wu, Z.-Y.; Li, C.; Liang, H.-W.; Chen, J.-F.; Yu, S.-H. Angew. Chemie Int. Ed. 2013, 52, 2925.

(26)

Liebner, F.; Haimer, E.; Wendland, M.; Neouze, M.-A.; Schlufter, K.; Miethe, P.; Heinze, T.; Potthast, A.; Rosenau, T. Macromol. Biosci. 2010, 10, 349.

(27)

Rånby, B. G. Acta Chem. Scand. 1949, 3, 649.

(28)

Revol, J.-F.; Bradford, H.; Giasson, J.; Marchessault, R. H.; Gray, D. G. Int. J. Biol. Macromol. 1992, 14, 170.

(29)

Jacoby, M. Chem. Eng. News 2014, 92, 9.

(30)

Future Markets Inc. Market Report: The Global Market for Nanocellulose to 2020; 2nd ed.;, 2012.

(31)

O′Connor, B. NCC: Environmental Health and Safety Update. 2012 TAPPI Nanotechnology Conference for Renewable Materials, Montréal, QC, June 4–7, Abstract 1837; 2012.

(32)

Government of Canada; Minister of the Environment. Domestic Substance List 2013. http://gazette.gc.ca/rp-pr/p2/2012/2012-11-21/html/sor-dors229-eng.html#archived%20, accessed: Sept, 2014 http://gazette.gc.ca/rp-pr/p2/2012/2012-11-21/html/sor-dors229eng.html#archived (accessed Sept, 2014).

(33)

Yang, X.; Bakaic, E.; Hoare, T.; Cranston, E. D. Biomacromolecules 2013, 14, 4447.


31


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(34)

Page 32 of 34

Xu, X.; Liu, F.; Jiang, L.; Zhu, J. Y.; Haagenson, D.; Wiesenborn, D. P. ACS Appl. Mater. Interfaces 2013, 5, 2999.

(35) Coulibaly, S.; Roulin, A.; Balog, S.; Biyani, M. V; Foster, E. J.; Rowan, S. J.; Fiore, G. L.; Weder, C. Macromolecules 2013, 47, 152. (36)

Shopsowitz, K. E.; Qi, H.; Hamad, W. Y.; Maclachlan, M. J. Nature 2010, 468, 422.

(37)

Kelly, J. A.; Giese, M.; Shopsowitz, K. E.; Hamad, W. Y.; MacLachlan, M. J. Acc. Chem. Res. 2014, 47, 1088.

(38)

Kelly, J. A.; Shukaliak, A. M.; Cheung, C. C. Y.; Shopsowitz, K. E.; Hamad, W. Y.; MacLachlan, M. J. Angew. Chemie Int. Ed. 2013, 52, 8912.

(39)

Heath, L.; Thielemans, W. Green Chem. 2010, 12, 1448.

(40)

Dash, R.; Li, Y.; Ragauskas, A. J. Carbohydr. Polym. 2012, 88, 789.

(41)

Fumagalli, M.; Sanchez, F.; Boisseau, S. M.; Heux, L. Soft Matter 2013, 9, 11309.

(42) Way, A. E.; Hsu, L.; Shanmuganathan, K.; Weder, C.; Rowan, S. J. ACS Macro Lett. 2012, 1, 1001. (43)

Sivakumaran, D.; Maitland, D.; Hoare, T. Biomacromolecules 2011, 12, 4112.

(44)

Smeets, N. M. B.; Bakaic, E.; Patenaude, M.; Hoare, T. Chem. Commun. 2014, 50, 3306.

(45)

Patenaude, M.; Hoare, T. ACS Macro Lett. 2012, 1, 409.

(46)

Leung, A. C. W.; Hrapovic, S.; Lam, E.; Liu, Y.; Male, K. B.; Mahmoud, K. A.; Luong, J. H. T. Small 2011, 7, 302.

(47)

Beck-Candanedo, S.; Roman, M.; Gray, D. G. Biomacromolecules 2005, 6, 1048.

(48)

Campbell, S. B.; Patenaude, M.; Hoare, T. Biomacromolecules 2013, 14, 644.

(49)

Thomason, S. C.; Kubler, D. G. J. Chem. Educ. 1968, 45, 546.

(50)

Abitbol, T.; Johnstone, T.; Quinn, T. M.; Gray, D. G. Soft Matter 2011, 7, 2373.

(51)

Lee, J.; Deng, Y. Soft Matter 2011, 7, 6034.

(52)

Lyons, W. J. J. Chem. Phys. 1941, 9, 377.

(53)

Liang, H.-W.; Guan, Q.-F.; Chen, L.-F.; Zhu, Z.; Zhang, W.-J.; Yu, S.-H. Angew. Chemie Int. Ed. 2012, 51, 5101.


32

Page 33 of 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(54)

Cranston, E. D.; Eita, M.; Johansson, E.; Netrval, J.; Salajková, M.; Arwin, H.; Wågberg, L. Biomacromolecules 2011, 12, 961.

(55)

Sehaqui, H.; Salajkova, M.; Zhou, Q.; Berglund, L. A. Soft Matter 2010, 6, 1824.

(56)

Gibson, L. J.; Ashby, M. F. Cellular Solids: Structure and Properties; 2nd ed.; Cambridge University Press, Cambridge, 1999.

(57)

Zhang, W.; Zhang, Y.; Lu, C.; Deng, Y. J. Mater. Chem. 2012, 22, 11642.


33


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 34 of 34

Table of contents graphic:


34

Chemically Cross-Linked Cellulose Nanocrystal Aerogels with Shape

Recommend Documents