jhon alejandro Ávila ramírez

8/10/2019 Jhon Alejandro vila Ramrez

1/8

Carbohydrate Polymers 114 (2014) 416423

Contents lists available at ScienceDirect

Carbohydrate Polymers

j ournal homepage: www.elsevier .com/ locate /carbpol

Surface esterification ofcellulose nanofibers by a simpleorganocatalytic methodology

Jhon Alejandro vila Ramrez a,b, CamilaJuan Surianoa, Patricia Cerrutti a,c,Mara Laura Foresti a,b,

a Grupode Biotecnologa y BiosntesisInstituto de Tecnologa en Polmeros y Nanotecnologa (ITPN)Facultad de Ingeniera,Universidad de BuenosAires,Las Heras 2214 (CP1127AAR), BuenosAires, Argentinab Consejo Nacional de Investigaciones Cientficas y Tcnicas (CONICET), Argentinac Departamento de Ingeniera Qumica, Facultad de Ingeniera, Universidad de BuenosAires, Argentina

a r t i c l e i n f o

Article history:Received 12 June 2014Received in revised form 11 August 2014Accepted 17 August 2014Available online 23 August 2014

Keywords:Bacterial celluloseSurface esterificationOrganocatalysisCharacterization

a b s t r a c t

Bacterial cellulose nanofibers were esterified with two short carboxylic acids by means of a simpleand novel organic acid-catalyzed route. The methodology proposed relayed on the use ofa non-toxicbiobased -hydroxycarboxylic acid as catalyst, and proceeded under moderate reaction conditions insolventless medium. By varying the esterification interval, acetylated and propionized bacterial cellu-lose nanofibers with degree of substitution (DS) in the 0.020.45 range could be obtained. Esterifiedbacterial cellulose samples were characterized by means ofSolid-State CP/MAS 13 C Nuclear MagneticResonance spectroscopy (CP/MAS 13C NMR), Fourier Transform Infrared spectroscopy (FTIR), X-rayDiffraction (XRD), Thermogravimetric Analysis (TGA) andchosen hydrophobicity test assays. TGA resultsshowed that the esterified nanofibers had increased thermal stability, whereas XRD data evidenced thatthe organocatalytic esterification protocol did not alter their crystallinity. The analysis of the ensuingmodified nanofibers by NMR, FTIR, XRD and TGA demonstrated that esterification occurred essentiallyat the surface ofbacterial cellulose microfibrils, something highly desirable for changing their surfacehydrophilicity while not affecting their ultrastructure.

2014 Elsevier Ltd. All rights reserved.

1. Introduction

Cellulose is the main component of wood and most naturalfibers, and it is also produced by tunicates and certain bacteria.In the last decade, cellulose microfibrils and cellulose nanocrys-tals obtained from the mentioned sources have received increasingattentionbasedon their uniqueproperties such as high water hold-ing capacity, high crystallinity, low thermal expansion coefficientin the axial direction, and high tensile strength and Young mod-ulus. Among nanocellulose applications, its use as reinforcementof nanocomposites is a continuously growing field (Hubbe, Rojas,Lucia,& Sain, 2008; Siro & Plackett,2010; Siqueira,Bras,& Dufresne,2010). The relative low cost, low density, availability, renewabil-ity and environmentally benign character of cellulose, as well asthe proved high mechanical properties of nanocellulosics, haveencouragedtheiruse as matrix reinforcement. However, the highlypolar surface of native cellulose nanoparticles associatedwith their

Corresponding author. Tel.: +54 11 45143010; fax: +54 11 45143010.E-mail address:[email protected] (M.L. Foresti).

OH-rich structure leads to low interfacial compatibility withnon-polar matrices which results in inadequate mechanical per-formance of the corresponding composites; high moisture uptakewhich leads to loss of fiber strength and composite deformation;and inter-fiber aggregation by hydrogen bonding which results inpoor fiber dispersion in the composite (Ifuku et al., 2007; Tomet al., 2011;Yuan, Nishiyama, Wada, & Kuga, 2006;Berlioz,Molina-Boisseau, Nishiyama, & Heux, 2009).

To overcome the mentioned difficulties, chemical modificationof cellulose nanofibers can be attempted. In particular, esterifica-tion reactions in which an acylant reacts with the hydroxyl groupsof cellulose nanofibers to produce less hydrophilic ester groupsis a common technique used to turn the cellulose surface into amore hydrophobic one. Esterification of cellulose nanofibers can beaccomplished by use of homogeneous or heterogeneous processes.In the homogeneous mechanism, mainly applied in cellulose acet-ylation processes, the esterified cellulose microfibrils are solublein the medium, and thus the chains located at the cellulose surfacediffuse into the surrounding solvent as soon as they are substan-tially derivatized (Sassi & Chanzy, 1995). On the other hand, in theheterogeneous or fibrous process, esterification is carried outin a

http://dx.doi.org/10.1016/j.carbpol.2014.08.0200144-8617/ 2014 Elsevier Ltd. All rights reserved.
http://localhost/var/www/apps/conversion/tmp/scratch_10/dx.doi.org/10.1016/j.carbpol.2014.08.020http://www.sciencedirect.com/science/journal/01448617http://www.elsevier.com/locate/carbpolmailto:[email protected]://localhost/var/www/apps/conversion/tmp/scratch_10/dx.doi.org/10.1016/j.carbpol.2014.08.020http://localhost/var/www/apps/conversion/tmp/scratch_10/dx.doi.org/10.1016/j.carbpol.2014.08.020mailto:[email protected]://crossmark.crossref.org/dialog/?doi=10.1016/j.carbpol.2014.08.020&domain=pdfhttp://www.elsevier.com/locate/carbpolhttp://www.sciencedirect.com/science/journal/01448617http://localhost/var/www/apps/conversion/tmp/scratch_10/dx.doi.org/10.1016/j.carbpol.2014.08.020


2/8

J.A. vila Ramrez et al. / Carbohydrate Polymers 114 (2014) 416423 417

mediumthatisanonsolventforbothnativeandesterifiedcellulose;and cellulose nanofibers morphology is preserved (Berlioz et al.,2009). In the last decade, different heterogeneous methodologiesforthe esterification of cellulose nanoparticles havebeen proposed,including acetylation of compressed bacterial cellulose pelliclescatalyzed by perchloric acid in toluene medium (Ifuku et al., 2007),surface acylation of cellulosewhiskers withalkyenylsuccinic anhy-dride aqueous emulsion followed by heat annealing (Yuan et al.,2006), solvent-free gas-phase esterification of cellulose whiskersand bacterial cellulose microfibrils with palmitoyl chloride (Berliozet al., 2009), simultaneous isolation and Fischer esterification ofhydroxyl groups of cellulose nanowhiskers in aqueous mediumusing acetic and butyric acids (Braun & Dorgan, 2009), hetero-geneous esterification of cellulose whiskers with long-chain fattyacids in p-tosyl chloride/pyridine medium (Uschanov, Johansson,Maunu, & Laine, 2011), solvent-free acetylation of bacterial cellu-lose using acetic anhydride in the presence of iodine as a catalyst(Hu, Chen, Xu, & Wang, 2011), and heterogeneous esterification ofbacterialcelluloseusing ionicliquids as solvent media andcatalysts(Tom et al., 2011), among others.

In the current manuscript, bacterial nanocellulose (BNC) washeterogeneously esterified in the absence of solvents using anoperationally simple eco-friendly methodology. Esterification withacetic and propionic acids as acylants was accomplished by useof an organic-acid-catalyzed heterogeneous route developed byHafrn and Crdova (2005) for the organocatalytic ring-openingpolymerization of -caprolactone from cotton and paper cel-lulose. The mentioned authors also applied this route for theesterification of cotton fibers with hexadecanoic and pentynoicacids catalyzed by tartaric acid. The organocatalytic esterificationmethodology described has recently been applied to the synthesisofstarchacetatesandbutyrateswithvaryingacylationlevels( Tupa,Maldonado, Vazquez, & Foresti, 2013). However, at the authorsbest knowledge the esterification of cellulose nanofibers by use ofthe described organocatalytic route has not been reported before.Acetylated and propionized cellulose nanofibers obtained hereinwere characterized by means of CP/MAS13C NMR, FTIR, DRX, TGA

andchosen hydrophobicity test assays, in order to demonstrate thesuitability of the organic acid-catalyzed route for BNC esterifica-tion, and study the effect of derivatization on particular propertiesof BNC.

2. Materials andmethods

2.1. Microorganism

The bacterial strain ofGluconacetobacter xylinus (syn.Acetobac-teraceti subsp.xylinus, Acetobacter xylinum) NRRLB-42 usedin thiswork was gently provided by Dr. Luis Ielpi (Fundacin InstitutoLeloir, Buenos Aires, Argentina).

2.2. Materials

Acetic acid (99.5%) and propionic acid (99.5%) were pur-chased from Dorwil. Hydrochloric acid (36.538%) and sodiumhydroxide were reagent grade chemicals purchased from Ane-dra and BioPack, respectively. Potassium hydrogen phthalateand sodium carbonate were bought from Laboratorios Cicarelliand Mallinckrodt, respectively. For BNC production, anhydrousdextrose (Biopack), meat peptone and yeast extract (Britania, Lab-oratorios Britania S.A.), disodium phosphate (Anhedra), citric acid(Merck), glycerol (Sintorgan) and corn steep liquor (Ingredion)

were used.

2.3. Bacterial cellulose production

Inocula were cultured for 48h in Erlenmeyers flasks contain-ing Hestrin and Schramm (HS) medium (%, w/v): glucose, 2.0;peptone, 0.5; yeast extract, 0.5; disodium phosphate, 0.27; citricacid, 0.115. The pH was adjusted to 6.0 with dil. HCl or NaOH.Agitation (200rpm) was provided by an orbital shaker. For BNCproduction, static incubations at 28C were performed in Erlen-meyers flasks containing 4.0% w/v glycerol and corn steep liquorrespectively, and keeping a flask/autoclaved medium v/v ratio of5:1. After 14 days bacterial nanocellulose pellicles were harvested,rinsed with water to remove the culture medium, and homoge-nized in a blender for 5min. The microfibrils thus obtained weretreated for 14h in a 5%w/v KOH solution in order to eliminate thebacterialcellsfromthe cellulosematrix,and then rinsedwithwatertill neutralization.

2.4. Organocatalytic esterification of BNC

Homogenized bacterial nanocellulose pellicles (0.5g, dryweight) were solvent exchangedfrom waterto theacylant by soak-ing and centrifugation (three times, 20mL). The acylant (0.53mol,i.e. 30mLof acetic acid or 40mLof propionic acid), tartaric acid

(0.47 g) and soaked BNC were mixed in an oven-dried 100 mLglassflask equipped with a magnetic stir bar, and a reflux condenser toprevent the loss of the acylant. The mixture was heated to 120 Cunder continuous agitation in a thermostatized oil bath. When themixture achieved the target temperature, all tartaric acid dissolvedand this was considered as the beginning of reaction. Esterifica-tion was run for different reaction times (1, 3, 4, 5, 6 and 8 h). Afterthe chosen reaction time, the mixture was allowed to cool down toroom temperature,and thesolidproduct wasseparated by vacuumfiltration in a Buchner funnel. Several washings of the recoveredsolid with distilled water were performed in order to guaranteethe removal of the catalyst and the unreacted acid.

The acyl content of the esterified nanofibers was determinedby heterogeneous saponification and back titration with HCl, as an

adaptation of theapproachused fordeterminingthe acyl content incellulose acetate (ASTM D871-96, Standard Test Methods of Test-ing Cellulose Acetate). With this purpose, esterified BNC sampleswere made into pellicles by drying, grinded under liquid nitrogenflux, and carefully dried again in order to have a precise measureof the mass of sample used in saponification. With the describedsample preparation protocol the method proved simple and accu-rate to quantitatively assay the bulk level of substitution achievedin BNC without specific instrumentation. Dried grinded samples(0.10g, 105 C, 2 h) were thustransferred to100mLErlenmeyerstowhich 10mLof ethyl alcohol (75%) was added. Flasks were heatedloosely stoppered for 30min at50 C. Afterwards, the suspensionswere brought to slightly basic pH by addition of a few drops of0.1 N NaOH using phenolphthalein as indicator. Aliquots of 10mL

of 0.1N NaOH were then added to each flask, and heated again at50 C for 15min. Erlenmeyers were finally allowed to stand tightlystoppered atroomtemperature for 72h. Atthe end of this time, theexcess NaOH present in the flasks was back titrated with 0.1N HCl,using phenolphthalein as the end-point indicator. A blank deter-mination (grinded native BNC) was carried through the completeprocedure. NaOH and HCl solutions were standardized using pre-viously dried standard potassium hydrogen phthalate and sodiumcarbonate, respectively.

The acyl content was then calculated by:

Acyl (%) = [(VB VS)NHCl Macyl 103 100]/W (1)

where VB (mL) is the volume of HCl required for titration of theblank; VS(mL) is the volume of HCl required to titrate the sample;

NHCl is the normality of the HCl solution; Macyl is the molecular


3/8

418 J.A. vila Ramrez et al. / Carbohydrate Polymers 114 (2014) 416423

weight of the acyl group (43 for acetyl and 57 for propionyl), andW(g) is the mass of sample used.

Using the Acyl % value, the substitution degree of esterifiedBNCsamples was calculated by:

DS = (162Acyl%)/[Macyl 100 ((Macyl 1)Acyl%)] (2)

where 162 is the molecular weight of the anhydroglucose units.

2.5. Characterization of esterified BNC

Solid-state 13C NMR measurements: High-resolution 13C solid-state spectra of grinded samples were recorded using the ramp{1H} {13C} CP/MAS pulse sequence (cross-polarization andmagic angle spinning) with proton decoupling during acquisition.All experiments were performed at room temperature in a BrukerAvance II-300 spectrometer equipped with a 4-mm MAS probe.The operating frequency for protons and carbons was 300.13 and75.46MHz, respectively. Glycine was used as an external referencefor the 13C spectra and to set the Hartmann-Hahn matching con-dition in the cross-polarization experiments. The recycling timevariedfrom5to6saccordingtothesample.ThecontacttimeduringCP was 2ms for all of them. The SPINAL64 sequence (small phase

incremental alternation with 64 steps) was used for heteronucleardecoupling during acquisition with a proton field H1H satisfying1H/2 =HH1H= 62kHz. The spinning rate for all the sampleswas 10kHz. The crystallinity index (CI) of chosen samples wasdetermined by separating the C4 region of the spectrum into crys-talline and amorphous peaks, and calculated by dividing the areaof the crystalline peak (8793ppm) by the total area assigned tothe C4 peak (8093ppm) as described elsewhere (Newman, 2004;Park, Baker, Himmel, Parrilla, & Johnson, 2010).

Fourier transform infrared spectroscopy (FTIR): Fourier transforminfrared spectra of native and esterified grinded samples wereacquired on an IR Affinity-1 Shimadzu Fourier Transform InfraredSpectrophotometer in transmission mode. Carefully dried (10 mg,105 C, 1h) grinded samples of native and acetylated BNC were

mixed with previously dried (130

C, overnight) KBr in the ratio1:25. The prepared pellets were further dried overnight at 105Cand spectra were finally collected with 40 scans in the range of4000700 cm1 witharesolutionof4cm1. Samples were normal-ized against the band found at 1165cm1 assigned to the (C O C)link of cellulose (Ilharco, Gracia, daSilva, & Ferreira,1997; Lee et al.,2011).X-ray diffraction (XRD): The structure of esterified BNC grinded

samples was analyzed with a Rigaku D/Max-C Wide Angle auto-mated X-ray diffractometer with vertical goniometer. The X-raydiffraction pattern was recorded in a 2angle range of 1040 ata step size of 0.02. The wavelength of the Cu/Ka radiation sourceused was 0.154 nm, generated at accelerating voltage of 40kV anda filament emission of 30mA. In order to calculate the crystallinityindex of native and esterified BNC samples Segals method wasapplied (Segal, Creely, Martin, & Conrad, 1959):

CI = (I002 Iam)/I002 100 (3)

where I002corresponds to the maximum intensity of the 002lattice diffraction, and Iamcorresponds to the intensity at 2= 18.

Thermogravimetric analysis (TGA): Thermogravimetric analysisof dried grinded samples (2.53 mg, 105 C, 1h) was conducted ina TGA-50 Shimadzu instrument. Temperature programs were runfrom 25C to 550 C at a heating rate of 10 C/min, under nitro-gen atmosphere (30 mL/min) in order to prevent thermoxidativedegradation.Analysis of hydrophobicity: Native and esterified grinded BNC

samples were qualitatively tested for hydrophobicity by water

flotation in water filled vials, as well as by distribution in test tubes

containing two immiscible polar/non polar liquid phases (i.e. dis-tilled water/petroleum ether). Neat and esterified pellicles werealso subjected to water drop tests, for which 10L water dropletswere manually placed on the pellicles surface using a syringe, andthe time required for complete absorption was determined.

3. Results and discussion

3.1. Organocatalytic esterification of BNC

Fig. 1 illustrates the evolution of the degree of substitution (DS)determined for acetylated and propionized BNC samples as a func-tion of time. As it is shown,using both acids the substitutiondegreeincreased with reaction time, reaching a DS value of 0.45 for theacetylation reaction after 8 h of reaction. Using propionic acid asacylant, a lower substitution degree was achieved, reaching a DSof 0.23 in 8h of reaction. The lower acylation level observed withpropionic acid is explained in terms of a lower reactivity of thisacylant due to a bulkier fatty acylium ion with higher steric hin-drance. Similar tendencies have been reported previously by otherauthors who studied the heterogeneous esterification of cellulosenanofibers with acylants of different chain length (Uschanov et al.,

2011; Tom et al., 2011; Lee et al., 2011). The contribution of theuncatalyzed reaction was in all cases lower than 10% of the mea-sured DS.

Literature dealing with cellulose nanofibers esterification byother methodologies has demonstrated that for a number of appli-cations low acyl contents have succeeded in altering cellulosenanofibers properties in the desired way. Nogi et al. (2006), pre-pared a nanocomposite consisting of an acrylic resin reinforcedwith acetylated BNC nanofibers with a DS of 0.17. The authorsobserved that the DS achieved effectively reduced the hygroscop-icity of BNC nanocomposites and improved thermal degradationresistance regarding transparency. In the surface acetylation ofbacterial cellulose nanofibers using perchloric acid as catalyst, byvarying the volume of acylant used Ifuku et al. (2007) obtained

acetylated nanofibers with DS in the 01.76 interval, which werealso used as acrylic resins reinforcement. The authors demon-strated that the moisture content of the nanocomposites wasminimizedat intermediate DS, i.e. 0.450.56, which they explainedin terms of two opposite effects: change in nanofibers surfacehydrophobicity and water penetration into the internal spaces ofthe esterified nanofibers upon bulky ester groups introduction.

0 1 2 3 4 5 6 7 8

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

0.50 Acetic acid

Propionic acid

DS

time (h)

Fig.1. Organocatalytic esterification of bacterialnanocelluloseusing acetic acidand

propionic acid as acylants, 120

C.


4/8


Scheme 1. Proposed mechanism forthe esterification of bacterial cellulose nanofibers with acetic and propionic acids catalyzed by l-tartaric acid.

Although it has not been completely elucidated, the catalyticmechanism of esterifications catalyzed by-hydroxy acids such astartaric acid, has been proposed to proceed analogously to that oflipase/esterase-catalyzed reactions (Domnguez de Mara, 2010).According to the mechanism proposed, the acetylation and pro-pionization reactions described herein (Scheme 1) initiate with anucleophilic attack of the acylant by the hydroxyl group of tar-taric acid. The intermediate formed is then further nucleophilicallyattacked by the hydroxyl groups of bacterial cellulose nanofibers,resulting in transesterification (Domnguez de Mara, 2010; Hafrn

& Crdova, 2005).

3.2. CP/MAS13C NMR spectroscopy

Esterification of BNC by the proposed organocatalytic method-ologywasconfirmedbyuseofCP/MAS13CNMRspectroscopy. Fig.2shows the CP/MAS 13CNMRspectra obtained for native, acetylated(DS = 0.22), and propionized (DS = 0.18) BNC. Native BNC spectrumshowed carbon peaks typical of cellulose I, i.e. C1: 105 ppm, C4:89ppm, C4: 84ppm, cluster C2C3C5: 7080ppm, C6: 65.3ppm,

and C6: 62.7ppm (Earl & VanderHart, 1980). As the esterificationreaction proceeded, several new resonances appeared on the spec-tra (Fig. 2b and c). In the spectrum of the acetylated BNC a newpeak was observed at 174 ppm, assignable to the resonance of thecarbonyl ester peak. For BNC esterification with propionic acid thissignal was observed at 176.2 ppm. Both samples also showed theappearanceofnewsignalsbetween9and30ppm,whicharosefromtheterminalmethylcarbonsofacetic(21.3ppm)andpropionicacid(9.6ppm) (Yamamoto, Horrii, & Hirai, 2005). The signal observedat 27ppm in the propionized sample corresponds to CH2. Signals

attributable to tartaric acid acting as grafting molecule were notobserved.Crystalline and disorderedcomponents of cellulose are detected

in solid-state13 C NMRspectra as downfield and upfield lines forthe C4 or C6 carbons, respectively (Yamamoto et al., 2005; Jandura,Kokta,&Riedl,2000). Asshownin Fig.2, nativeBNCexhibitedbroadnon-crystalline resonances centered at 84 and 62.7ppm (upfieldlines, i.e. C4 and C6, respectively). Downfield signals centeredat 89 and 65.3ppm (C4 and C6, respectively) have been assignedto ordered cellulose structures (Atalla & VanderHart, 1999;

Fig. 2. CP/MAS13

C NMRspectra of neat and esterified BNC samples. (a) Neat BNC, (b) acetylated BNCDS= 0.22 and (c) propionized BNCDS= 0.18.


5/8


Table 1

Crystallinity index of neat and esterified BNC calculated by Newmans (NMR) and634 Segals (XRD) method.

Sample CINMR(%) CIXRD (%)

Native BNC 71.8 91.8AcetylatedBNC, DS=0.14 90.1AcetylatedBNC, DS=0.22 72.3 AcetylatedBNC, DS=0.45 91.9Propionized BNC,DS= 0.09 92.3

Propionized BNC,DS= 0.18 69.0 Propionized BNC,DS= 0.23 90.7

Park et al., 2010). The crystallinity index of native, acetylated(DS = 0.22) and propionized BNC (DS = 0.18) was calculated usingthepeakseparationmethodofNewman(2004), bydividingtheareaof the crystalline peak by the total area of assigned to the C4 peaks.The CI values obtained suggested that no significant change in thecrystallinityofBNCtookplaceduringesterification( Table1, CINMR).

3.3. FTIR spectroscopy

BNC esterification was further qualitatively confirmed by FTIRspectroscopy of acylated samples. Fig. 3 shows the FTIR spectra

of acetylated and propionized BNC with chosen DS. The spec-trum of native BNC was also included for comparison. NativeBNC spectrum showed bands typical of cellulose I, showingthe presence of hydroxyl groups (3360 cm1); C H stretch-ing (2895cm1); H O H bending vibration of absorbed watermolecules (1647cm1); CH2 symmetrical bending (1427cm1);cellulose C O C bridges (1165 cm1); C O bond stretching(1118cm1); ether C O C functionalities (1061cm1); and theband at 897cm1 which is typical of-linked glucose polymers(Castro et al., 2011; Moosavi-Nasab & Yousefi, 2011; Rani, Navin, &Appaiah, 2011; Ashori, Babaee, Jonoobi, & Hamzed, 2014).

Spectra of acetylated and propionized BNC provided evidenceof esterification by appearance of a new band at 1730cm1 whichis assigned to the stretching of the ester carbonyl groups (C O)

introduced (Lee et al., 2011; Tom et al., 2011; Ashori et al., 2014;Yuan et al., 2006; Braun & Dorgan, 2009). As shown in Fig. 3, in thespectra of samples with lower DS the intensity of the C O esterwas found to increase with the DS measured by saponification.On the other hand, although esterification implied substitution ofa fraction of hydroxyl groups by less polar ester groups, no clearreduction of the intensity of the band associated with the vibration

of OH groups located at 3300cm1 was observed. This behaviorhas previously been attributed to BNC derivatization occurringessentially on the accessible hydroxyl groups of the surface of thenanofibers or in the amorphous fraction of the cellulose (Lee et al.,2011). The absence of absorption at 1700cm1 attributed to car-boxylic groupsindicated that no unreactedacylant remained in theesterified BNC.

3.4. X-ray diffraction (XRD)

Cellulose has several crystalline polymorphisms (I, II, III, IV).Cellulose I is the crystalline polymorphism that is naturally pro-duced by a variety of organisms, i.e. trees, plants, tunicates,algae and bacteria; and it is the crystal structure with thehighest axial elastic modulus (Moon, Martini, Nairn, Simonsen,& Youngblood, 2011). Diffraction patterns obtained for nativeBNC, acetylated BNC (DS= 0.14 and DS= 0.45) and propionizedBNC (DS= 0.09 and DS= 0.23) are shown in Fig. 4. Native BNCshowed four diffraction peaks at 2=14.4 (101), 16.7 (10-1),22.5 (002), and 34.0 (040) which confirmed that only cellu-lose I was present in all samples (Vazquez, Foresti, Cerrutti, &Galvagno, 2013; Johnson Ford, Mendon, Thames, & Rawlins, 2010;

Terinte, Ibbett, & Schuster, 2011; Gea et al., 2011). For the DSattained in this contribution, esterified cellulose nanofibers main-tained the diffraction pattern of native BNC, with no evidenceof peak shifting or appearance of new peaks. This is further evi-dence that the organocatalytic reaction involved essentially theOH groups on the surface or in the amorphous regions of BNC,not affecting in a great extent its ultrastructure (Tom et al.,2011; Lee et al., 2011). The degree of crystallinity of native andmodified BNC was determined using the Segal equation (Segalet al., 1959). Results included in Table 1 (CIXRD), further confirmedthat no significant change in BNC crystallinity took place duringorganic-acid-catalyzed esterification of BNC. As expected, absoluteCI values differed from those obtained by NMRpeak separationmethods, since CI values are known to vary significantly depend-ing on the measurement method used to determine them (Parket al., 2010; Terinte et al., 2011).

3.5. Thermogravimetric analysis (TGA)

Esterified BNC samples were characterized by thermogravimet-ric analysis to study the impact of organocatalytic esterification on

i) ii)

3900 3600 3300 3000 2700 2400 2100 1800 1500 1200 900

(f)

(e)

(c)

(b)

(a)

Absorbance

Wavenumber [cm-1]

(d)

3900 3600 3300 3000 2700 2400 2100 1800 1500 1200 900

Absorbance

Wavenumber [cm-1]

(e)

(c)

(b)

(a)

(d)

Fig. 3. FTIR spectra of neat andesterified BNC samples. Acetylated BNC (left,i): (a)neat BNC;(b) DS= 0.05; (c)DS = 0.12; (d)DS =0.14; (e)DS =0.39; (f)DS = 0.45. Propionized

BNC (right, ii): (a) neat BNC; (b) DS= 0.06; (c) DS= 0.09; (d) DS= 0.12; (e) DS= 0.23.


6/8


10 15 20 25 30 35 40

(040)

(002)

(101)

2

(e)

(b)

(a)

(d)

(c)

(101)

-

Fig. 4. X-ray diffraction pattern of neat andesterified BNC.(a) Neat BNC;(b) propi-onizedBNCDS=0.09;(b)propionizedBNCDS=0.23;(c)acetylatedBNCDS=0.14;(d) acetylated BNCDS= 0.45. For clarity purposes, baselines of XRD patterns havebeen moved up in they axes.

the thermal properties of BNC. Fig. 5 shows the TG and DTG curvesobtained for (i) acetylated BNC and (ii) propionized BNC sampleswith increasing DS. DTG data have been normalized with respectto the initial sample mass, and results have been moved up in they axes for clarity purposes. The TG and DTG curves of native BNChave also been included for comparison.

Neat BNC pyrolysis showed a single weight loss stage, with anextrapolated onset temperature (Tonset) of 313 C. Tonset was cal-culated as the intersection of the extrapolated pre-decompositionordinate value and a tangential line drawn to the point of steepestslope of the weightloss curve in thedecomposition region. Accord-ing to DTG data, the maximum weight loss rate of neat BNC (Tmax)took place at 329 C. Thermogravimetric data of neat BNC showeda minimal weight loss assignable to BNCdehydration (2.3%), due tosample preconditioning at 105C during 1 h. The measured weightloss (%) during samples preconditioning decreased as the DS ofthe samples increased, which was a consequence of the effect ofesterification on BNC hydrophilicity.

Asitisshownin Fig.5-i, acetylatedbacterial cellulosenanofibersdescomposed in a single step process. Irrespectively of acyla-tion level achieved, acetylated nanofibers started to decompose at

50 100 150 200 250 300 350 400 450 500 550

0

10

20

30

40

50

60

70

80

90

100 (a)

(b)

(c)

(d)

(e)

(f)

(g)

m/m

0[%]

Temperature [C]

50 100 150 200 250 300 350 400 450 500 550

0

10

20

30

40

50

60

70

80

90

100

m/m

0[%]

Temperature [C]

(a)

(b)

(c)

(d)

(e)

(f)

(g)

50

100 150

200

250 300

350

400 450

500

550

DTG

Temperature [C]

(e)

(b)

(a)

(d)

(c)

(g)

(f)

50

100 150

200

250 300

350

400 450

500

550

DTG

Temperature [C]

(e)

(b)

(a)

(d)

(c)

(g)

(f)

i)

ii)

Fig. 5. TG and DTG curves of neat and esterified BNC samples. (i) Acetylated BNC: (a) neat BNC; (b) DS=0.05; (c) DS=0.12; (d) DS=0.14; (e) DS=0.22; (f) DS=0.39; (g)

DS =0.45. (ii) Propionized BNC: (a) neat BNC; (b) DS= 0.02; (c) DS= 0.06; (d) DS= 0.09; (e) DS= 0.12; (f) DS= 0.18;(g) DS= 0.23.


7/8


0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50

300

310

320

330

340

350

360

370

Neat BNC - Tonset

Acetylated BNC - Tonset

Propionized BNC - Tonset

Neat BNC - Tmax

Acetylated BNC - Tmax

Propionized BNC - Tmax

Tonse

t,T

max

(C)

DS

Fig.6. Tonset(C) andTmax(C)valuescalculatedfor neat andesterified BNCsampleswith increasing DS.

substantially higher temperatures than neat BNC, with Tonsetval-

ues that were found to increase with the DS of the sample. Tmaxvalues of acetylated samples also showed a positive correlationwith DS.Similartendencieswere found forpropionizedBNC (Fig.5-i). The increase ofTonset and Tmax with the DS of acetylated andpropionized samples is illustrated in Fig. 6. Results included inFig. 6 also suggested that the length of the ester group introducedshowedno significant effecton theTonsetandTmaxvaluesregisteredfor acetylated and propionized samples with similar DS. Higherthermal stability of modified celluloses for increasing substitu-tion degrees has previously been reported (Jandura, Riedl, & Kokta,2000; Morgado & Frollini, 2011; Uschanov et al., 2011). Increasedthermal stability of esterified samples has been attributed tothe replacement of hydroxyl groups by more stable ester groups(Ashori et al., 2014; Lin, Huang, Chang, Feng, & Yu, 2011).

Among the methodologies used in the literature to hetero-geneously esterify BNC, a number of authors have reported areduction in the thermal stability of cellulose nanoparticles uponmodification, characterized by onset and maximum temperaturedecomposition peaks found at lower temperatures than those ofneatBNC(Tom et al., 2011; Uschanov et al., 2011; Lee et al., 2011).The mentioned behavior was attributed to the reduction of cellulose crystallinity induced by esterification, and, lessfrequently, to the high liability of the ester linkages introducedin the cellulose structure, or to the reduced number of effectivehydrogen bonds remaining among BNC nanofibers upon esterifi-cation (Tom et al., 2011; Uschanov et al., 2011; Lee et al., 2011).

In this context, the increase in thermal stability of BNC shown inFig. 5 is in agreement with previous characterization data whichevidenced that the organocatalytic esterification of cellulosenanofibers took place in their outmost layers, while keeping theultrastructure unaffected. Specially when used as reinforcementfor polymers whose melting processing require elevated temper-atures, it is important that alterations of the surface chemistryof cellulose nanofibers do not decrease the onset temperature ofthermal degradation (Braun & Dorgan, 2009).

In the case of propionized BNC samples a second high temper-ature decomposition peak was observed in the 430550C range(Fig. 5-i). TheDTG peak wasthe resultof a secondthermal degrada-tion step registered in TG curves of propionized samples, which ledto weight residues fractions much lower than those observed forneat BNC in the same temperature interval. The behavior observedsuggested that organocatalytic propionization induced changesthat increased the volatility of the solid residue of BNC samplesat high temperatures. Further studies on the nature and reasonsfor the mentioned effects of organocatalytic propionization of BNCare currently going on.

3.6. Surface hydrophobicity

Grinded acetylated and propionized BNC samples were quali-tatively tested for hydrophobicity by water flotation, distributionin polar/non-polar biphasic systems, and water drop tests. Allassays confirmed that acetate or propionate groups covalentlyattached to the surface of BNC altered its polarity. Sample distribu-tion in test tubes containing equal volumes of distilled water andpetroleum ether (0.67g/mL) is illustrated in Fig. 7 for chosen sam-ples. When added to the biphasic liquid system, native BNC fell tothe lower aqueous phase, absorbed water and sank immediatelyto the bottom of the test tube. Contrarily, acetylated (DS =0.45)and propionized (DS =0.23) BNC grinded samples fell across thenon polar phase and remained floating in the non-polar/polarinterphase without crossing to the aqueous phase (Fig. 7c and f,

respectively). The described sample distribution did not changeupon vertically shaking of test tubes. For both acylants, esterifiedsamples with lower DS showed intermediate behaviors (Fig.7b ande, respectively). In water flotation tests, samples were placed onthe surface of water-filled vials. In the case of native BNC the sam-ple absorbed water and sank immediately to the vial bottom. Incontrast, esterified samples corresponding to DS= 0.45 (acetylatedBNC) and DS= 0.23 (propionized BNC), did not absorb water andremained floating in the water surface (data not shown). In waterdrop tests 10L water droplets were manually placed on the sur-faces of native, acetylated (DS = 0.14) and propionized (DS = 0.12)BNC pellicles. Fig. 7gi shows the appearance of the droplets after

Fig. 7. Surface hydrophobicity assays. (a, d, g) Neat BNC; (b) acetylated BNCDS= 0.14; (c) acetylated BNCDS= 0.45; (e) propionized BNCDS= 0.09; (f) propionized

BNCDS = 0.23; (h)acetylated BNCDS= 0.14; (i) propionized BNCDS= 0.12.


8/8
http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0185http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0185http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0185http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0185http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0185http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0185http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0185http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0185http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0185http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0185http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0185http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0185http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0185http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0185http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0185http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0185http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0180http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0180http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0180http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0180http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0180http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0180http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0180http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0180http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0180http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0180http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0180http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0180http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0180http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0180http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0180http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0180http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0180http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0180http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0180http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0180http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0180http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0180http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0180http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0180http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0175http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0175http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0175http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0175http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0175http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0175http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0175http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0175http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0175http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0175http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0175http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0175http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0175http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0175http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0175http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0175http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0175http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0175http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0175http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0175http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0175http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0175http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0175http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0175http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0175http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0170http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0170http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0170http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0170http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0170http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0170http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0170http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0170http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0170http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0170http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0170http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0170http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0170http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0170http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0170http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0170http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0165http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0165http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0165http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0165http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0165http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0165http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0165http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0165http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0165http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0165http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0165http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0165http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0165http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0165http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0165http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0165http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0165http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0165http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0160http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0160http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0160http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0160http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0160http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0160http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0160http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0160http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0160http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0160http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0160http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0160http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0160http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0160http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0160http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0160http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0160http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0160http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0160http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0160http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0160http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0160http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0160http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0155http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0155http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0155http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0155http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0155http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0155http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0155http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0155http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0155http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0155http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0155http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0155http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0155http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0155http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0155http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0155http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0155http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0155http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0155http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0155http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0155http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0155http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0155http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0155http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0155http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0155http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0150http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0150http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0150http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0150http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0150http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0150http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0150http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0150http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0150http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0150http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0150http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0150http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0150http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0150http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0150http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0145http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0145http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0145http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0145http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0145http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0145http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0145http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0145http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0145http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0145http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0145http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0145http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0145http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0145http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0145http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0145http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0140http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0140http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0140http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0140http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0140http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0140http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0140http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0140http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0140http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0140http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0140http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0140http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0140http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0140http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0140http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0140http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0140http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0140http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0140http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0140http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0140http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0140http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0140http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0140http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0140http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0135http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0135http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0135http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0135http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0135http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0135http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0135http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0135http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0135http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0135http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0135http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0135http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0135http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0130http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0125http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0125http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0125http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0125http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0125http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0125http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0125http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0125http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0125http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0125http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0125http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0125http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0125http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0125http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0125http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0125http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0125http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0125http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0125http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0125http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0125http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0120http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0120http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0120http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0120http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0120http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0120http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0120http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0120http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0120http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0120http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0120http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0120http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0120http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0120http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0120http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0120http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0120http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0120http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0115http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0115http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0115http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0115http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0115http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0115http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0115http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0115http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0115http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0115http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0115http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0115http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0115http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0115http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0115http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0115http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0110http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0110http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0110http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0110http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0110http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0110http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0110http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0110http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0110http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0110http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0110http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0110http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0110http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0110http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0110http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0110http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0110http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0110http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0110http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0110http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0105http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0105http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0105http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0105http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0105http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0105http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0105http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0105http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0105http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0105http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0105http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0105http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0105http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0105http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0105http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0105http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0105http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0105http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0105http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0105http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0105http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0100http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0100http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0100http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0100http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0100http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0100http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0100http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0100http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0100http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0100http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0100http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0100http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0100http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0100http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0100http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0100http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0100http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0095http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0095http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0095http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0095http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0095http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0095http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0095http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0095http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0095http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0095http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0095http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0095http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0095http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0095http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0095http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0095http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0095http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0095http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0095http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0095http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0095http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0090http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0090http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0090http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0090http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0090http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0090http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0090http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0090http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0090http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0090http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0090http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0090http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0090http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0090http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0090http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0090http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0085http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0085http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0085http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0085http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0085http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0085http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0085http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0085http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0085http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0085http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0085http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0085http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0085http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0085http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0085http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0085http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0085http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0085http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0085http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0085http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0085http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0085http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0085http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0085http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0080http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0080http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0080http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0080http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0080http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0080http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0080http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0080http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0080http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0080http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0080http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0080http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0080http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0080http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0080http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0080http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0080http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0080http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0080http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0080http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0080http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0080http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0080http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0080http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0080http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0075http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0070http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0070http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0070http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0070http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0070http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0070http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0070http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0070http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0070http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0070http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0070http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0070http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0070http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0070http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0070http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0070http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0070http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0070http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0070http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0070http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0070http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0070http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0070http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0070http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0065http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0065http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0065http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0065http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0065http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0065http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0065http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0065http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0065http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0065http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0065http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0065http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0065http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0065http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0065http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0065http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0065http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0065http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0065http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0065http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0065http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0065http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0065http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0065http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0065http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0060http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0060http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0060http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0060http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0060http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0060http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0060http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0060http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0060http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0060http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0060http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0055http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0055http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0055http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0055http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0055http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0055http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0055http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0055http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0055http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0055http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0055http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0055http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0055http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0055http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0055http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0055http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0050http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0050http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0050http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0050http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0050http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0050http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0050http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0050http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0050http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0050http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0050http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0050http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0050http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0050http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0045http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0045http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0045http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0045http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0045http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0045http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0045http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0045http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0045http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0045http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0045http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0045http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0045http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0045http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0045http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0045http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0045http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0045http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0045http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0045http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0045http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0045http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0045http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0045http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0040http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0040http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0040http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0040http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0040http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0040http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0040http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0040http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0040http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0040http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0040http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0040http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0040http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0040http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0040http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0040http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0040http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0040http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0040http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0040http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0040http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0040http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0040http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0040http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0035http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0035http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0035http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0035http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0035http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0035http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0035http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0035http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0035http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0035http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0035http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0035http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0035http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0035http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0035http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0035http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0035http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0035http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0035http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0030http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0030http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0030http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0030http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0030http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0030http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0030http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0030http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0030http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0030http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0030http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0030http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0030http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0030http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0030http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0030http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0030http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0030http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0030http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0030http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0030http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0030http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0025http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0025http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0025http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0025http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0025http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0025http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0025http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0025http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0025http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0025http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0025http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0025http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0025http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0025http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0025http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0025http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0025http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0020http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0020http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0020http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0020http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0020http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0020http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0020http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0020http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0020http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0020http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0020http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0020http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0020http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0020http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0020http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0010http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0005http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0005http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0005http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0005http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0005http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0005http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0005http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0005http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0005http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0005http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0005http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0005http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0005http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0005http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0005http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0005http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0005http://refhub.elsevier.com/S0144-8617(14)00789-9/sbref0005

jhon alejandro Ávila ramírez

Documents