Short Stories

AKD-Modification of bacterial cellulose aerogels in supercritical CO2

Description
AKD-Modification of bacterial cellulose aerogels in supercritical CO2
Categories
Published
of 13
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
  ORIGINAL PAPER AKD-Modification of bacterial cellulose aerogelsin supercritical CO 2 Axel Russler  • Marcel Wieland  • Markus Bacher  • Ute Henniges  • Peter Miethe  • Falk Liebner  • Antje Potthast  • Thomas Rosenau Received: 8 March 2012/Accepted: 15 May 2012/Published online: 29 May 2012   Springer Science+Business Media B.V. 2012 Abstract  Different approaches towards hydropho-bic modification of bacterial cellulose aerogels withthe alkyl ketene dimer(AKD) reagent are presented. If AKD modification was performed in supercriticalCO 2 , an unexpectedly high degree of loading wasobserved. About 15 % of the AKD was boundcovalently to the cellulose matrix, while the other partconsisted of re-extractable AKD-carbonate oligomers,which are novel chemical structures described for thefirst time. These oligomers contain up to six AKD andCO 2  moieties linked by enolcarbonate structures. Thehumidity uptake from environments with differentrelative humidity by samples equipped with up to30 % AKD is strongly reduced, as expected due to thehydrophobization effect. Samples above 30 % AKD,and especially at very high loading between 100 and250 %, showed the peculiar effect of increasedhumidity uptake which even exceeded the value of unmodified bacterial cellulose aerogels. Keywords  Cellulose aerogel    Bacterial cellulose   Surface modification    Hydrophobization   Supercritical carbon dioxide    Alkyl ketene dimer(AKD)    AKD-CO 2 -oligomers    Humidity uptake Introduction Aerogels from bacterial cellulose (BCAG) are fasci-nating ultra-lightweight structures, exhibiting a hier-archical, three-dimensional porous morphology madeof pure cellulose, combined with extraordinarily highsurface area and toughness (Liebner et al. 2010). However, for some application fields, the character-istics of pure bacterial cellulose do not translate intothe desired material properties: The natural hydrophi-licity may be a drawback for some applications wherean interaction with hydrophobic structures is neces-sary or hydrophilic substances have to be preventedfrom interaction with the BCAG structure. Modifica-tion of bacterial cellulose generally can follow therules of ordinary cellulose modification when themorphology of the product is of no further interest.However, modification of BCAG under simultaneouspreservation of the inherent structural featuresrequires special approaches. In principle, both deriv-atization (covalent reagent bonding) and coating(reagents are just physically attached) are suitablestrategies for BCAG modification. Because of thefragile nature of BCAGs, techniques based on A. Russler    M. Wieland    M. Bacher   U. Henniges    F. Liebner    A. Potthast    T. Rosenau ( & )Christian Doppler Laboratory for Advanced CelluloseChemistry and Analytics, University of Natural Resourcesand Life Sciences Vienna, Muthgasse 18, 1190 Vienna,Austriae-mail: thomas.rosenau@boku.ac.atP. MietheFZMB GmbH Forschungszentrum fu¨r Medizintechnik und Biotechnologie, Bad Langensalza, Germany  1 3 Cellulose (2012) 19:1337–1349DOI 10.1007/s10570-012-9728-y  supercritical carbon dioxide (scCO 2 ) have to beapplied for appropriate drying to largely avoidchanges of the natural BC structure (Liebner et al.2010; Haimer et al. 2008; Liebner et al. 2009, 2007, 2008; Gavillon and Budtova 2008). Due to its intrinsically low viscosity and special dissolutionproperties, scCO 2  can be used as multi-functionalmedium for processing of (bacterial) cellulose aero-gels: drying, chemical reactions, and solvent forimpregnation or fixation of various substances withinthe porous material without altering the naturalmorphology (Liebner et al. 2010; Russler et al.2011). Combining supercritical drying with derivati-zation or surface coating in one step would represent ahighly efficient and advantageous approach.Alkyl ketene dimer (AKD,  1 ) is a well known andwidely used sizing agent in paper production that isused to hydrophobize the cellulose surface to increasewater repellency and thus to enable paper to be printedand written on in a way that applied inks do not leachinto the paper material and the print remains sharp anddetailed (Roberts 1996; Ek et al. 2009). It is added in the head box of the paper machine in rather smallamount of 0.1–0.5 %. It is prepared e. g. from naturalfatty acids and stearic acid. Conventional AKD is awaxy material with a melting point of about 50   C,neverthelessalsoformulationsexistwhichareliquidatroom temperature. Having alkyl chains of normally16–18 C-atoms, AKD wax is not soluble in water and,whenaddedtotheheadboxinpaperproduction,isusedin the form of an aqueous dispersion in cationicallymodified starch. Application is possible within a pHrange of about 7–10, so AKD became quite importantfor neutral and alkali paper formation systems, whichcompare favorably to acidic systems with regard tolong-term stability of the produced paper.AKD is proposed to react with the hydroxyl groupsof cellulose to form a  b -keto ester moiety ( 2 ). Thereaction with ethanol to the corresponding ethyl ester 2a  is of limited predictive power as a model for thereaction with cellulose: the esterification is in factsimilar, but the accessibility of the cellulose surface isof course largely different from that of the low-molecular weight alcohol. AKD will slowly react withwater to a  b -keto acid ( 3 ) that readily decarboxylatesto produce a ketone ( 4 ) (see Fig. 1). Why the reactionwith the OH-groups of cellulose is favored over thereaction with excess water under normal applicationconditions is not clear until now (Odermatt et al. 2003;Seo et al. 2008a, b; Lindfors et al. 2007). Nevertheless it is known that the sizing effect is not an instant one,but needs some time to develop (Shen and Parker2001; Shen et al. 2005; Zhang et al. 2007). It is assumed that in this time the AKD is spreading overthe cellulosic surface.AKD is a well-known, well-established and cheapagent for hydrophobization of cellulose in papermak-ing—we were interested in its applicability in mod-ifying BCAG by means of a novel application systemwith scCO 2 . Novel cellulosic aerogels with morehydrophobic surfaces will be the result. These mate-rials will be better compatible with synthetic-organicmatrices than non-modified cellulose aerogels, whichwould open up new fields of applications. Differentprocesses and process step sequences are possible forthe modification of BCAG with AKD. They differ incomplexity, limitations in reagent (AKD) loading,effectiveness and homogeneity of the modification.Naturally, it should be the aim to reduce the technicalcomplexity and work input while maintaining theBCAGs 0 above-discussed structural features. Thedifferent processes are given in Fig. 2 schematically.All of these methods contain at least one processstep involving scCO 2 , which we have recently shownto be a promising medium for production of aerogelsalso from bacterial celluloses (Liebner et al. 2010). As AKD is soluble in scCO 2 , it is possible to perform alsothe reagent-loading itself in scCO 2  as the solvent, notonly the drying step. Some modifications of cellulosewith AKD applying scCO 2  are known in literature, butthe techniques applied and the cellulose types usedwere different from our approach (Hutton and Parker2009; Quan et al. 2009). OORROCellOROOHR'ORO OROEtRRROROCell-OHH 2 O+CO 2 123 42a Et-OH Fig. 1  Reaction of AKD with cellulose (Cell-OH) or ethanol,and the hydrolytic reaction with water, R and R 0 being C 14 -alkylor C 16 -alkyl1338 Cellulose (2012) 19:1337–1349  1 3  Experimental GeneralAll chemicals were available from commercial sup-pliers. Ethanol (absolute, Merck) and  n -hexane(Merck) of reagent grade were used as received.Distilled water was used throughout. TLC wasperformed using Merck silica gel 60 F 254  pre-coatedplates.FlashchromatographywasperformedonBakersilica gel (40  l m particle size). All products werepurified to homogeneity (TLC/GC analysis). Bacterialcellulose was obtained from FZMB GmbH, BadLangensalza, Germany (see Liebner et al. 2010;Russler et al. 2011 for preparation and purification).AKD starting materialAlkyl ketene dimer (AKD, technical grade, Herkules)was further purified by flash chromatography using  n -hexane as the solvent. GCMS analysis confirmed theremainder to be pure AKD of different chain lengths,without any impurities of AKD ketone ( 4 ) or esters.NMR and GCMS 1 H NMR spectra were recorded at 400 MHz for  1 Hand at 100 MHz for  13 C NMR in CDCl 3 . Chemicalshifts, relative to TMS as internal standard, are givenin  d  values, coupling constants in Hz.  13 C peaks wereassigned by means of APT, H,H-COSY, HMQC andHMBC spectra. As AKD is a mixture of severalcompounds (due to the different chain lengths of theused fatty acids), resonances are sometimes superim-posed, and multiplicities are only given when clearlydiscernible. GCMS analysis was carried out on anAgilent 6890 N/5975B in the ESI (70 eV) ionizationmode.Compound characterization by NMRAKD ( 1 , starting material).  1 H NMR:  d  0.87 (t, 6H,CH 3 ), 1.22–1.40 (m, br, CH 2 –), 1.39 (m, 2H, CH 2 –CH 2 –CH–COO), 1.77 (q, 2H, CH 2 –CH 2 –CH–COO),2.12 (q, 2H, CH 2 –CH=C), 3.94 (t, 1H, CH 2 –CH–COO), 4.68 (CH 2 –CH=C).  13 C NMR:  d  14.1 (CH 3 ),22.7 (CH 3 –CH 2 ), 24.6 (CH 2 –CH 2 –CH–COO), 26.3(CH 2 –CH 2 –CH–COO), 27.5 (CH 2 –CH=C), 29.1–29.7(CH 2 ), 31.9 (CH 3 –CH 2 –CH 2 ), 53.7 (=C–CH–COO),101.7 (CH 2 –CH  =  C), 145.6 (CH=C), 169.7 (COO).AKD ethyl ester ( 2a ).  1 H NMR:  d  0.87 (t, 6H, CH 3 in alkyl), 1.20–1.36 (m, br, CH 2 –), 1.28 (3H, super-imposed by m, O–CH 2 –CH 3 ), 1.51–1.64 (m, 2H,CH 2 –CH 2 –CO), 1.78–1.84 (q, 2H, CH 2 –CH 2 –CH–COO), 2.38 (t, 2H, CH 2 –CH 2 –CO), 3.40 (t, 1H, CO–CH–COO), 4.18 (q, 2H, O–CH 2 –CH 3 ).  13 C NMR:  d 14.1 (CH 3  in alkyl), 14.2 (O–CH 2 –CH 3 ), 22.7 (CH 3 –CH 2 –CH 2 ), 23.8 (CH 2 –CH 2 –CO), 28.2 (CH 2 –CH–COO), 29.1–29.9 (CH 2 ), 32.0 (CH 3 –CH 2 –CH 2 ), 43.7(CH 2 –CH 2 –CO), 59.3 (CO–CH–COO), 169.8 (COO),205.7 (CO).AKD-ketone ( 4 ).  1 H NMR:  d  0.88 (t, 6H, CH 3  inalkyl), 1.22-1.34 (m, br, CH 2 –), 1.55 (pent, 2H, CO–CH 2 –CH 2 ), 2.38 (t, 2H, CO–CH 2 –CH 2 ).  13 C NMR:  d 14.2 (CH 3  in alkyl), 22.8 (CH 3 –CH 2 –CH 2 ), 24.0 BCG BCAG scCO 2 Drying  mBCAG AKDExtraction scCO 2 DryingscCO 2 Loading BCG  LoadingBath  mBCAG AKDSolvent Exchange scCO 2 Drying BCG BCAG scCO 2 Drying  mBCAG AKDscCO 2 DryingLoadingBath Solvent Exchange 132 Fig. 2  Overview of thedifferent process approachesfor AKD modification of BCAGs (see theexperimental section for adetailed description), BCG:bacterial cellulose gel,BCAG, bacterial celluloseaerogel, mBCAG: AKD-modified bacterial celluloseaerogel)Cellulose (2012) 19:1337–1349 1339  1 3  (CH 2 –CH 2 –CO), 29.4–29.8 (CH 2 ), 32.0 (CH 3 –CH 2 –CH 2 ), 42.9 (CH 2 –CH 2 –CO), 211.9 (CO).AKD-CO 2  hexamer ( 5 ). ‘‘Terminal unit’’ denotesthe AKD unit with the ketone moiety, ‘‘proximal unit’’the one with the free acid moiety.  1 H NMR:  d  0.87(36H, CH 3 ), 1.20–1.42 (CH 2 –), 1.39 (m, 10H, CH 2 –CH 2 –CH–COO), 1.51–1.59 (2H, CH 2 –CH 2 –CO interminal unit), 1.78 (q, 10H, CH 2 –CH 2 –CH–COO),1.78–1.82 (q, 2H, CH 2 –CH–COO in terminal unit),2.12(q,10H,CH 2 –CH=C),2.38(t,2H,CH 2 –CH 2 –COin terminal unit), 3.40 (t, 1H, CO–CH–COO interminal unit), 3.95 (t, 5H, CH 2 –CH–COO), 4.72 (t,5H, CH 2 –CH=C).  13 C NMR:  d  14.1 (CH 3 ), 22.7(CH 3 –CH 2 ), 23.8 (CH 2 –CH 2 –CO in terminal unit),24.4–24.6 (CH 2 –CH 2 –CH–COO), 26.3–26.4 (CH 2 –CH 2 –CH–COO), 27.5 (CH 2 –CH=C), 28.2 (CH 2 –CH–COO in terminal unit), 28.9–30.4 (CH 2 ), 31.9 (CH 3 –CH 2 –CH 2 ), 43.7 (CH 2 –CH 2 –CO in terminal unit),53.7-53.8 (=C–CH–COO), 59.3 (CO–CH–COO interminal unit), 101.7–101.9 (CH 2 –CH=C), 145.5–145.7 (CH=C), 169.3–169.6 (COO), 169.9 (COOHin proximal unit), 205.7 (CO).scCO 2  systemsFor the drying and AKD-modification of BCAG, ahigh-pressure system was used. CO 2  from a gas bottlewas pressurized in a high pressure pump and chargedinto a 500 cm 3 reactor. Temperature was set to 40   Cand was controlled by a thermostat. The reactor wasequipped with a stirrer and a grid to store the aerogelsseparated from the stirrer. The stirrer was not neces-sary for the drying of the aerogels alone, but isrequired for AKD-modification to insure uniformdistribution of the reagent. The system was equippedwith a two-stage separatorunit and with the possibilityto recycle the gas.Before supercritical drying and AKD modification,bacterial cellulose aquogels have to undergo solventexchange. A solvent exchange against absolute etha-nol was performed three times for at least 24 h. Thesubsequent supercritical drying of the BCGs wasperformed in the pressure vessel at 100 bar, 40   C anda CO 2  flow of 3.5 kg/h with no stirring. Treatmenttimes were 1 h per centimeter of aerogel diameter. Forproperties and morphology of the BC aerogels see:Liebner et al. 2010; Russler et al. 2011. In the case of AKD-modification, the same proce-dure as for the supercritical drying was used, with thefollowing modifications: pressure and time werevaried (see text), no CO 2  flow was applied, andstirring was applied at 350 min - 1 .The simple setup of the pressure system guaranteedproper cleanability and good repeatability. Neverthe-less, since precipitated AKD can block valves andtubes, it is advisable to flush the apparatus withsuitable organic solvents (chloroform, petrol ether)from time to time, and to release the pressure by avalve directly, without separators (Odermatt et al.2003; Seo et al. 2008; Hutton and Parker 2009; Quan et al. 2009).SEMScanning electron microscopy (SEM) was performedon a Phillips XL 30 ESEM (Environmental ScanningElectron Microscope, ESEM) at an accelerationvoltage of 10 kV with different magnifications.FTIRFourier transform infrared spectroscopy (FTIR) wasperformed on a Bruker Vertex 70 HTS-XT in theAttenuated Total Reflectance (ATR) mode with 32scans per measurement between 4000 and 400 cm - 1 .Humidity uptakeThe humidity uptake of the samples was measuredgravimetrically after storage in desiccators of con-trolledhumidityforatleast 72 h.Thedesiccatorswerefilled with phosphorous pentoxide for 0 % relativehumidity, with saturated solutions of calcium chloridefor 30 %, of ammonium nitrate for 65 % and of potassium sulfate for 98 % relative humidity, andwere stored at room temperature.Modification approaches according to Fig. 2In the case of direct supercritical loading, the AKDwax was charged directly into the pressure vesseltogether with BCAG. The BCAG samples were putintotheinternalwirecasketstopreventthemfromfreefloating in the supercritical fluid. The system was thenpressurized at different pressure levels for 60 min.The stirring was started when supercritical conditionswere reached, and was stopped during pressurerelease beforedroppingbelowsupercriticalconditions. 1340 Cellulose (2012) 19:1337–1349  1 3  TheAKD-modifiedBCAGwasimmediatelysubjecttoanalyticalcharacterization,extracted aftersomecuringtime (1 or 2 weeks at room temperature) or furtherstored at ambient conditions. The extraction of AKDfromloadedsamples(totestforphysicaladsorptionvs.covalent binding) was performed batchwise with neat n -hexane(20-foldsamplevolume)for24 hinashaker,and thisextraction was repeated twice. As indicated bythe dotted line in Fig. 2, a combination of BCG dryingwith AKD loading is possible if the scCO 2  apparatus isequipped with separate pressure chambers connectedwith valves,so that after drying scCO 2 -dissolved AKDcan be introduced.According to the second approach of Fig. 2, theBCG was pretreated by solvent exchange to ethanoland then subject to a loading bath, containing differentamounts of AKD dissolved in  n -hexane. The modifiedBCG was directly dried supercritically in the highpressure system. Alternatively, another solventexchange step to ethanol, a non-solvent for AKD,was done beforehand. The conditioning in the loadingbath should lastlongenough, dependingon thesampledimensions,toensurehomogeneousdistributionoftheloading medium throughout the sample, at least 1 hper centimeter of aerogel.The third way in Fig. 2 for the loading with AKDwas starting from already dried BCAG. Modificationof this aerogel with AKD was performed by AKDsolutions in  n -hexane of set concentrations, so that adefined amount of AKD was homogeneously distrib-uted throughout the BC matrix. To obtain aerogelsfrom the solvogels, it was necessary to apply anotherscCO 2 -drying step, either with or without preliminarysolvent exchange to ethanol. Results and discussion BCAG modification with AKD (‘‘AKD loading’’)The supercritical loading of AKD onto BCAG (path 1in Fig. 2) was found to be a very flexible approach.The degrees of loading, i.e. the ratio between mass of AKD to mass of srcinal, non-modified BCAG, wasvariedinawiderangefromabout30 %toover250 %.When the pressure in the vessel is reduced belowsupercritical conditions, the dissolved AKD would beexpected to precipitate homogeneously within thewhole pressurized volume. If, for example, the densityof BCAG is 10 mg/cm 3 and the density of the agentto be precipitated into the system is 2 mg/cm 3 , amaximum degree of loading of about 20 % isexpected. Such a result coming close to the ‘‘theoret-ical’’ value is obtained, for instance, for inert long-chain aliphatics and triglycerides as the substances tobe deposited within the aerogel matrix. However, inthe case of AKD, we found in all cases a higherloading of the BCAG than calculated from the mass of AKD putinto the pressure system, with values rangingup to 250 %.Threemechanismscanbesummonedtoaccountforthis peculiar behavior. The first is the reaction of AKDwith the cellulose. The contribution of this mechanismto the enrichment effect, however, must be rathersmallsince85  ±  3 %ofthedepositedAKDcanbere-extracted with  n -hexane from the cellulosic bodies(independent of the degree of loading), so that amaximumof15  ±  3 %ofthe initiallydepositedAKDcan be covalently bound to cellulose, thus being non-extractable. The second mechanism of the enrichmentas described might be a preferred precipitation withinthe nano-scale pore system of the BCAG due tocapillary effects. However, it is not clear why sucheffects would be operative for AKD, but not forparaffin waxes or triglycerides (which do not show theenrichment effect). Hence, this mechanism is ratherunlikely. The third mechanism is an oligomerizationof AKD. By polymer–polymer (polymer-oligomer)interactions, these bigger molecules would be prefer-ably adsorbed on the cellulose surface, so that as theneteffecttheaerogelisenrichedwithAKDmoietiesatthe same time depleting the supercritical solvent of AKD. The initial trigger for assuming that such areaction could have occurred was the observation thatre-extracted AKD was no longer monomeric, butoligomeric. This prompted us to look into the mech-anism more closely.Mass spectrometric analysis of the  n -hexaneextracts of the deposited AKD, showed the presenceof larger AKD-derived molecules ( 5 ), which howeverwere not just AKD-oligomers, but appeared to containadditional CO 2 -moieties, readily identifiable by amass difference of 44 and its multiples. Apparently,the supercritical medium did not behave as inertsolvent for AKD, but participated in a reaction,promoting an oligomerization. Such oligomerizationis not observed in aqueous suspension or in organicsolutions in common solvents, so that this effect must Cellulose (2012) 19:1337–1349 1341  1 3
Search
Similar documents
View more...
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