Letters

Films made of cellulose nanofibrils: surface modification by adsorption of a cationic surfactant and characterization by computer-assisted electron microscopy

Description
Films made of nanofibrils were modified by adsorption of a cationic surfactant directly on the film surfaces. The nanofibrils were prepared by 2,2,6,6-tetramethylpiperidinyl-1-oxyl (TEMPO)-mediated oxidation and mechanical fibrillation, and were
Categories
Published
of 10
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  RESEARCH PAPER Films made of cellulose nanofibrils: surface modificationby adsorption of a cationic surfactant and characterizationby computer-assisted electron microscopy K. Syverud  • K. Xhanari  • G. Chinga-Carrasco  • Y. Yu  • P. Stenius Received: 4 November 2009/Accepted: 30 August 2010/Published online: 15 September 2010   Springer Science+Business Media B.V. 2010 Abstract  Films made of nanofibrils were modifiedby adsorption of a cationic surfactant directly on thefilm surfaces. The nanofibrils were prepared by 2,2,6,6-tetramethylpiperidinyl-1-oxyl (TEMPO)-mediatedoxidation and mechanical fibrillation, and were rela-tively homogeneous in size. The average nanofibrildiameter and surface porosity was quantified based oncomputer-assisted field-emission scanning electronmicroscopy (FE-SEM). The cationic surfactant usedin the adsorption was  n -hexadecyl trimethylammoni-um bromide (cetyltrimethylammonium bromide,CTAB). The adsorption of CTAB was confirmed byFourier transform infrared (FTIR) spectroscopy andhigh-resolution transmission electron microscopy(HRTEM) analyses. It was shown that the adsorbedlayer of CTAB increased the hydrophobicity, withoutaffecting the tensile index significantly. This capabil-ity, combined with the antiseptic properties of CTAB,may be a major advantage for several applications. Keywords  Microfibrillated cellulose   Surface analysis    High-resolutionelectron microscopy    Image analysis   Surface modification    Nanofibres    Wood products Introduction During the past 10 years there has been a significantand continuous interest in research on microfibrillatedcellulose. This relatively new material that srcinatesfrom the disintegration of cellulose fibres into sub-micron and nanofibrils was introduced in the begin-ning of the 1980s (Turbak et al. 1983; Herrick et al.1983). The material has been given several additionaldenominations such as nanofibrillated cellulose,nanocellulose, nanofibres and nanofibrils. For clari-fication purposes, in this study we use the followingscale definitions: (i) sub-micron fibrils being approx-imately between 100 and 1,000 nm, (ii) nanofibrilshaving diameters of roughly below 100 nm and (iii)microfibrils having diameters of approximately3.5 nm (Ohad and Danon 1964; Emons 1988; Kennedy et al. 2007). The term fibril includes allthe three scales. For a detailed description of thedifferent terms commonly applied to the fibrillatedmaterial, see Mo¨rseburg and Chinga-Carrasco (2009).Several applications have been envisaged, basedon the enormous strength of crystalline cellulose andthe capability of the fibrils to adsorb water. Benefitswith respect to improved paper properties (Eriksen K. Syverud ( & )    G. Chinga-CarrascoPaper and Fibre Research Institute (PFI AS),Høgskoleringen 6b, 7491 Trondheim, Norwaye-mail: kristin.syverud@pfi.noK. Xhanari    P. SteniusUgelstad Laboratory, Norwegian University of Scienceand Technology (NTNU), Trondheim, NorwayY. YuDepartment of Materials Science and Engineering,Norwegian University of Science and Technology(NTNU), 7491 Trondheim, Norway  1 3 J Nanopart Res (2011) 13:773–782DOI 10.1007/s11051-010-0077-1  et al. 2008; Syverud and Stenius 2009; Mo ¨rseburgand Chinga-Carrasco 2009) and composite materials(Iwatake et al. 2008; Seydibeyoglu and Oksman2008) have been reported. However, efficient utili-zation of nanofibrils has been significantly limited bythe costs of production, which involves large amountsof energy and/or cumbersome procedures such asmechanical (Eriksen et al. 2008), chemical (Saitoet al. 2006; Abe et al. 2007) and enzymatic treatments (Pa¨a¨kko¨ et al. 2007).Cellulose nanofibrils are hydrophilic, with greatability to form interfibrillar hydrogen bonds and,hence, strong thin films. The fibrils may also bemodified so that their surface becomes hydrophobic.Such capability suggests new applications within,i.e. packaging, composites or emulsions. Hydropho-bic nanofibrils can be achieved by reactions with thehydroxyl groups in the glucose units of cellulose, e.g.by silylation (Andresen et al. 2006), or by grafting of polymer brushes and layers (Stenstad et al. 2008).These modifications are time-consuming and fre-quently involve solvent exchange to organic solvents.This makes them little realistic for industrial appli-cations. Fukuzumi et al. (2009) demonstrated that it ispossible to modify films made from nanofibrilsproduced with 2,2,6,6-tetramethylpiperidinyl-1-oxyl(TEMPO)-mediated oxidation by soaking them in analkyl ketene dimer (AKD) suspension. The surface of TEMPO-oxidized nanofibrils carries substantial amountsof acid groups. It is thus possible to modify the fibrilsurfaces by ionic bonds with hydrophobic counterions.  N  -hexadecyl trimethylammonium bromide, Cetyl-trimethylammonium bromide (C 16 H 33 N(CH 3 ) 3 Br,CTAB) is a hydrophobic cationic surfactant thatmay increase the hydrophobicity of the nanofibrilsurfaces. CTAB has also been used previously formodification of cellulose fibres (Alila et al. 2005). Inaddition, CTAB is an effective antimicrobial agent(Salton 1951; Denyer and Hugo 1977). A material composed of nanofibrils requires ded-icated procedures for assessing their chemical andstructural properties. Numerous methods have beenproposed for structural characterization of nanofibrils,including several microscopy techniques. Continuousadvances have been made for quantification of thedimensions of fibril structures, based on atomic forcemicroscopy (AFM) (Fukuzumi et al. 2009), trans-mission electron microscopy (TEM) (Pa¨a¨kko¨ et al.2007; Wa˚gberg et al. 2008) and field-emissionscanning electron microscopy (FE-SEM) (Abe et al.2007; Chinga-Carrasco and Syverud 2009). Individ- ually, none of these methods gives a completedescription of these tiny structures. However, it isvaluable to clarify their advantages and limitationsand thus exploit their complementary capabilities. Inaddition, in most of the cases direct assessment bymicroscopy techniques is based only on subjectiveevaluations. The full potential of computerized imageanalysis for quantifying nanostructures has not beenexploited completely (Chinga-Carrasco and Syverud2009). Finally, modern approaches apply microscopyand computer-assisted image analysis for processing,visualization and quantification of structures, whichis most valuable. Hence, for simplification purposesthe term computer-assisted electron microscopy isapplied, which focusses on the computer assistancein the complete process from the image acquisition tothe quantification.The purpose of this study was to demonstratemethods for preparation, modification and character-ization of films made of cellulose nanofibrils. Themodification was performed on the film surface, andnot on the fibrils in suspension before film formation.The hydrogen bonding capability of the fibrils in thebulk of the film is retained, and thus also the filmstrength. Effects of surface modification on the filmsproperties were verified. The surface-modified fibrilswere characterized with computer-assisted electronmicroscopy techniques. The approach presented inthis study exemplifies the advantages of computer-assisted microscopy for moving towards automaticobjective quantification of nanostructures and thusreducing subjective evaluation, which is applied inmost studies reported in the literature. Experimental section Chemi-mechanical production of nanofibrilsNever-dried fully bleached kraft pulp fibres madefrom softwood were pretreated by TEMPO-mediatedoxidation according to the procedure described bySaito et al. (2006), where the TEMPO radicalcatalyses oxidation of primary alcohol groups usingNaClO. The method gives carboxylic acid groups inthe C6 position of the glucose unit. All chemicalsused in the oxidation were laboratory grade. 774 J Nanopart Res (2011) 13:773–782  1 3  3.8 mmol NaClO per gram of cellulose was used inthe oxidation, and pH was kept at 10.5 by titration of NaOH. The reaction time was 50 min. The pretreat-ment was followed by mechanical fibrillation usingan Ultra TURRAX at 24,000 rpm (160 Watt) for6 min giving a fibril suspension of about 0.8%. Nocentrifugation or filtration was performed on theresulting batch of fibrils.Charge determinationThe carboxylate content of the nanofibrils wasdetermined by conductometric and potentiometrictitrations. To a suspension of 0.3 g dried cellulosefibrils in water, a certain volume of 0.01 M NaCl wasadded as described by Saito and Isogai (2004). Themixture was then stirred to obtain well-dispersedslurry. The pH of the suspension was adjusted to2.5–3.0 by addition of 0.1 M HCl and then titratedwith a 0.04 M NaOH solution up to pH  =  11. Thecarboxylate content (amount of weak acid groups)was determined from the break points in theconductivity curves whilst Gran plots were used forthe potentiometric titrations (Rossotti and Rossotti1965; Gran 1952). Manufacturing of model filmsModel films made of fibrils were prepared from a0.1% suspension poured into a cylindrical mould witha filter paper and a supporting wire in the bottom,similar to the procedure described by Syverud andStenius (2009). The water was removed by evapora-tion and free suction without vacuum. The films weredried by evaporation at room temperature (approxi-mately 24 h) and 50   C (2 h). The basis weight of thefilms was 20 g/m 2 .Surface modificationBefore the films were removed from the moulds, aknown amount of dissolved CTAB was applied to thetop side of the films. Two concentrations of CTABsolution were used in the adsorption, 0.93 and 4 timesthe critical micelle concentration (CMC, 0.9 mmol/ dm 3 , Furst et al. 1996), i.e. 0.81 and 3.6 mmol/L. Thesame total amount of CTAB, 3.55 g/m 2 , was thusused in both cases. This corresponds to approxi-mately one CTAB molecule for every 10th of thetotal cellulose monomers, assuming that there is anexcess of CTAB at the film surface. The water wasallowed to evaporate and the films were dried at50   C. After this procedure, the films were removedfrom the moulds and substrate. The films were rinsedby dipping 10 times in distilled water for 1–2 s anddried again at 50   C before testing. Table 1 gives adescription of five different cellulose nanofibril filmsprepared in this study.Analysis of model films propertiesContact angle measurements were performed using aDAT 1100-FIBRO system using deionized waterwith a droplet size of 4  l L. The tensile index wasmeasured with a Zwick material tester (T1-FRxxMOD.A1K).FTIR spectroscopyThe infrared spectra were obtained with a BIORADExcalibur Spectrometer (Series FTS 3000). The filmswere analysed using transmission mode.Electron microscopyThe films were embedded in epoxy resin forscanning electron microscopy (SEM) cross-sectionalanalysis. The device was a Hitachi 3000, low-vacuummicroscope. The film thicknesses were quantifiedaccording to Chinga et al. (2007). The surfaces of thefilms were determined by the solid–air interfaces.10 images were acquired in backscatter electron Table 1  Model films prepared from cellulose nanofibrilspretreated with TEMPOSample DescriptionT1 Nanofibrils onlyT2 Nanofibrils  ?  CTAB (4  9  CMC)T3 Nanofibrils  ?  CTAB (4  9  CMC),rinsed with distilled waterT4 Nanofibrils  ?  CTAB (0.93  9  CMC)T5 Nanofibrils  ?  CTAB (0.93  9  CMC),rinsed with distilled waterTwo concentrations of the CTAB solution were applied forcomparison purposes, i.e. 0.9 and 4 9  the critical micelleconcentration (CMC)J Nanopart Res (2011) 13:773–782 775  1 3  imaging mode (BEI) at 1,000 9  magnification. Theimages had sizes of 2,560  9  1,920 pixels and aresolution of 50 nm.Surface images of the films were obtained with aZeiss Ultra field-emission (FE)-SEM with the in-lenscapability. The model films were prepared and imagedas described by Chinga-Carrasco and Syverud (2009).Images from areas apparently without a metallicconductive layer were acquired. The accelerationvoltage was 2 kV and the working distance was1 mm. The applied magnification was 50,000 9  andthe resolution was 2.22 nm.TEM studies were performed using a JEM 2010TEM with an accelerating voltage of 200 kV. Unmod-ified nanofibrils and CTAB-modified nanofibrils weredispersed in distilled water. A drop of the dispersionwas applied on a carbon-coated Cu grid for TEMinvestigations. The high-resolution transmission elec-tron microscopy (HRTEM) images were acquireddigitally in a 2 k   9  2 k Gatan CCD camera. Theapplied magnification was 100,000 9  and the effectivepoint-to-point resolution was 0.23 nm.Image processing and analysisThe public domain program ImageJ (Rasband 1997)was applied for processing and analysis. Estimation of fibril dimensions and surface porosity is most valuablein different applications for quality control. Aseries of FE-SEM images from the model film without CTABmolecules were processed with a Fast Fourier Trans-form (FFT) bandpass filter (Walter 2007), for facilitat-ing the segmentation of the nanofibrils and porestructure (Fig. 1). The average nanofibril diameterwas quantified according to Chinga-Carrasco andSyverud (2009). A simple procedure for estimation of surface porosity is introduced in this study. A back-ground-symmetry algorithm (Young et al. 1998) wasapplied to the surface images for segmenting thesurface pore structure. The algorithm is based onfindingadistinctive background,which isgivenby themaximum value of the grey level histogram. A  p %value on the non-object side is then determined andapplied for segmenting the surface pore structure.TheTEMimagesoftheCTAB-modifiednanofibrilswere thresholded. Noise particles were removed auto-matically. A Voronoi routine (Schmid 2008) wasapplied for estimating the CTAB intermolecular dis-tances.Theroutineyieldsaskeletonofthebackgroundcontaining the distances to the nearest particle edges.The Feret diameter was applied for estimating thelength of the CTAB molecules. The Feret diametercorresponds to the longest axis of a given object. Results and discussion Properties of surface-modified model filmsThe contact angles of the films before (T1) and afterCTAB adsorption (T2 and T4) are shown in Fig. 2. Inaddition, the contact angle was also measured onfilms after washing with distilled water (T3 and T5).A significant increase in contact angle was measuredafter adsorption of CTAB (Fig. 2). The surfaceadsorption was performed using two concentra-tions of CTAB solution: 4    CMC (T2 and T3) and0.93    CMC (T4 and T5), but with the same totalamounts. The contact angles of the model films T4and T5 were similar to the contact angles of T2 andT3, thus indicating that the difference in concentra-tion yielded CTAB layers of similar structures and Fig. 1  Segmentation of surface pore structures of model films.  a  An FE-SEM surface image.  c  Image of the same area showed in a  after processing the image with an FFT band pass filter.  b  and  d  are the corresponding segmentation results776 J Nanopart Res (2011) 13:773–782  1 3  did not affect the contact angles significantly. Itseems, however, that the contact angle was reducedslightly after rinsing with water. This may be due toan excess of CTAB (compared to available anionicsites on the fibril surfaces) that is removed afterrinsing. Preliminary studies by XPS of fibrils equili-brated with CTAB solutions indicate that the adsorp-tion sites of the fibril surfaces may indeed have beensaturated with CTAB for both concentrations.Although a significant increase in contact anglewas obtained by the CTAB adsorption, the contactangles may be even higher with an increased numberof acid groups on the fibril surfaces. In this study weobtained 0.5 mmol/g acid groups. This may beincreased by using higher amounts of NaClO in theoxidation of fibres, thus increasing the number of negative sites on the surface susceptible for adsorp-tion of CTAB molecules.The tensile strengths of the films T1–T5 are shownin Fig. 3. The treatment with CTAB followed bywashing tended to reduce the tensile index slightly,although the decrease was not statistically significant.However, the reduction in strength depended on theprocedure applied for adsorbing CTAB. The reduc-tion was more pronounced at the lower CTABconcentration, i.e. using larger amounts of waterduring surface modification. The films were in thiscase exposed to water for a longer period of time. Thewater and CTAB molecules penetrated into the filmnetwork, facilitated by its porous structure (Fig. 5).The presumptive presence of CTAB molecules ininterfibril spaces limited the formation of hydrogenbonds between the nanofibrils during the subsequentdrying. Hence, such a reduction of nanofibrilsinterbonds reduced the strength of the model films.Theseresultsindicatethatitispossibletoexploitthehigh degree of surface oxidation of the cellulosenanofibril films. Specific functionality may thus beadded to the fibrillated material by a relatively simplesurface modification technique. However, the properdesign of surface modification procedures for a givenapplication requires an extensive understanding of the interaction between cellulose nanofibrils andthe applied surfactant, i.e. CTAB in this case. Thefollowing sections describe thus the direct assessmentof interactions between cellulose nanofibrils andCTAB molecules.Estimating cellulose–CTAB interactionsFourier transform infrared spectroscopy (FTIR) wasapplied to verify the adsorption of CTAB on themodel film surfaces (Fig. 4). According to Sui et al.(2006), CTAB has symmetric and asymmetric CH 2 vibrations of the alkyl chain at 2,850 and 2,917 cm - 1 .The FTIR spectra confirmed the absorbance at * 2,850 cm - 1 for modified films and at both wave-numbers when CTAB was adsorbed on single fibrils.Note that the peak at approximately 2,850 cm - 1 isevidenced in both modified (sample T2) and modi-fied  ?  rinsed (sample T3) films. This confirms theabsorption of CTAB molecules on the film surfaces.However, the detection of CTAB in sample T3 is lessthan in sample T2. This agrees with the measured 0102030405060708090T1 T2 T3 T4 T5    C  o  n   t  a  c   t  a  n  g   l  e   (   d  e  g  r  e  e   ) Fig. 2  Contact angle of model films made with TEMPO-basedcellulose fibrils with and without surface modification withCTAB 020406080100120140T1 T2 T3 T4 T5    T  e  n  s   i   l  e   i  n   d  e  x   (   N  m   /  g   ) Fig. 3  Tensile index of model films made with TEMPO-basedcellulose fibrils with and without surface modification withCTABJ Nanopart Res (2011) 13:773–782 777  1 3
Search
Similar documents
View more...
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks