NC State
Žepič, V., Poljanšek, I., Oven, P., Škapin, A. S., and Hančič, A. (2015). "Effect of drying pretreatment on the acetylation of nanofibrillated cellulose," BioRes. 10(4), 8148-8167.


The aim of this study was to evaluate the effect of different morphologies of solvent-exchanged (NFCSE), spray-dried (NFCSD), and freeze-dried (NFCFD) nano-fibrillated cellulose on the susceptibility to surface modification with the acetic anhydride/pyridine system. The degree of substitution (DS), morphology, degree of crystallinity (Icr), hydrophobicity, and thermal stability of acetylated products were examined. Acetylated NFCSD and NFCFD had higher DS than acetylated NFCSE, suggesting that drying pre-treatment increased the susceptibility of NFC for acetylation. The morphology of acetylated NFCFD and NFCSD with higher DS was different from unmodified samples, while that of NFCSE was not affected by acetylation. Microspheres of acetylated NFCSD started to dissolve when the highest DS was reached. As opposed to unmodified NFCFD, the nanofibrillar units of acetylated NFCFD became individualised at lower DS. Acetylated samples had lower Icr than the unmodified samples. A significant increase in the contact angle was observed at higher DS of acetylated NFC samples. Acetylation markedly elevated the thermal stability of the acetylated NFC samples.

Full Article

Effect of Drying Pretreatment on the Acetylation of Nanofibrillated Cellulose

Vesna Žepič,a Ida Poljanšek,b,* Primož Oven,b,* Andrijana Sever Škapin,c and Aleš Hančič a

The aim of this study was to evaluate the effect of different morphologies of solvent-exchanged (NFCSE), spray-dried (NFCSD), and freeze-dried (NFCFD) nano-fibrillated cellulose on the susceptibility to surface modification with the acetic anhydride/pyridine system. The degree of substitution (DS), morphology, degree of crystallinity (Icr), hydrophobicity, and thermal stability of acetylated products were examined. Acetylated NFCSD and NFCFD had higher DS than acetylated NFCSE, suggesting that drying pre-treatment increased the susceptibility of NFC for acetylation. The morphology of acetylated NFCFD and NFCSD with higher DS was different from unmodified samples, while that of NFCSE was not affected by acetylation. Microspheres of acetylated NFCSD started to dissolve when the highest DS was reached. As opposed to unmodified NFCFD, the nanofibrillar units of acetylated NFCFD became individualised at lower DS. Acetylated samples had lower Icr than the unmodified samples. A significant increase in the contact angle was observed at higher DS of acetylated NFC samples. Acetylation markedly elevated the thermal stability of the acetylated NFC samples.

Keywords: Acetylation; Freeze dried; Hydrophobicity; Nanofibrillated cellulose; Properties; Spray dried

Contact information: a: TECOS, Slovenian Tool and Die Development Centre, Kidričeva 25, SI-3000 Celje, Slovenia; b: University of Ljubljana, Biotechnical Faculty, Department of Wood Science and Technology, Jamnikarjeva 101, SI-1000 Ljubljana, Slovenia; c: Slovenian National Building and Civil Engineering Institute, Dimičeva 12, SI-1000 Ljubljana, Slovenia;

* Corresponding author:;


In recent years, increased attention has been directed to the development of sustainable, green, and environmentally friendly materials, a field in which cellulose plays an important role (Miao and Hamad 2013; Rebouillat and Pla 2013). Cellulose is the most abundant natural polymer on earth, having a unique molecular structure and properties (Dufresne 2010). Cellulose is a linear chain composed of β (1→4) linked D-glucopyranosyl units. Cellulose chains are hydrogen bonded into microfibrils and microfibrillar aggregates, where highly ordered regions alternate with disordered regions (Zugenmaier 2008).

A significant breakthrough in cellulose applicability has been made by the development of efficient procedures for the disintegration of cellulose fibres (Zimmermann et al. 2004; Chakraborty et al. 2005; Iwamoto et al. 2005; Saito et al. 2006; Henriksson et al. 2007; Pääkkö et al. 2007; Chen et al. 2011) into the product referred to as microfibrillated cellulose (MFC) (Herrick et al. 1983; Turbak et al.1983). MFC may be composed of nanofibrils, fibrillar fines, fibre fragments, and fibres (Chinga-Carrasco 2011), whereas nano-structures represent a main component of MFC (Abdul Khalil et al. 2014). In this work, these cellulose portions will be referred to as nanofibrillated cellulose (NFC).

NFC can compete with other synthetic reinforcing agents in terms of strength to weight ratio (Azizi Samir et al. 2005; Šturcova et al. 2005; Eichhorn et al. 2010), whereas its hydrophilic character represents a major obstacle for its use in combination with hydrophobic polymers. With a chemical modification of the NFC surface, its hydrophilicity and a tendency toward hornification can be drastically reduced (Eyholzer et al. 2010; Tingaut et al. 2010). Acetylation is a commonly used chemical modification procedure for wood (Hill et al. 1998) and cellulose fibres (Fengel and Wegner 1984), and was successfully applied to MFC as well (Tingaut et al. 2010).

It has been demonstrated that grafting of acetyl moieties on the surface of various cellulose nanomaterial creates a hydrophobic surface (Jonoobi et al. 2010; Tingaut et al. 2010; Jonoobi et al. 2012) and hence reduces water wettability (Jonoobi et al. 2010; Lin et al. 2011). Grafting also reduces the degree of crystallinity (Hu et al. 2011) and the average size of nanofibers (Rodionova et al. 2011), and it preserves the nanofibrillar structure (Hu et al. 2011; Missoum et al. 2012). Acetylation of NFC has been used for various purposes. Rodionova et al. (2011) used acetylation to increase the barrier properties of NFC films. Acetylation was also used for hydrophobization of the NFC surface to improve the dispersion of modified NFC in nonpolar PLA solution for production of nanocomposite cast films (Bulota et al. 2012). In addition to NFC with plant origins, acetylation was successfully employed for the esterification of bacterial cellulose (Kim et al. 2002; Ifuku et al. 2007; Lee et al. 2011; Berlioz et al. 2009) and tunicin cellulose whiskers (Berlioz et al. 2009).

In a previous study, the authors investigated a range of properties of nanofibrillated cellulose obtained after different drying techniques (Žepič et al. 2014). It was demonstrated that aggregation phenomena are significantly reduced in freeze-dried specimens compared to the powder obtained by spray-drying. That study also showed that re-dispersed freeze-dried NFC had practically the same rheological properties as never-dried NFC. It was also reported that freeze-dried NFC powder retained the initial morphological structure of NFC (Žepič et al. 2014).

To the best of our knowledge there has been no study on the influence of different drying methods on the accessibility for acetylation reaction. Although Lee and Bismarck (2012) recently reported the susceptibility towards organic acid esterification of never-dried and freeze-dried bacterial cellulose, the susceptibility towards acetylation of freeze- and spray-dried in comparison with the susceptibility towards acetylation of solvent-exchanged NFC has not yet been explored.

In this study, the susceptibility of solvent-exchanged (SE), freeze-dried (FD), and spray-dried (SD) NFC for a surface modification was examined by employing the acetic anhydride/pyridine system under heterogeneous reaction conditions. In addition, the influence of the catalyst concentration and the reaction time was studied. The degree of substitution, morphology, degree of crystallinity, hydrophobicity, and thermal stability of acetylated products were examined by Fourier transform infrared spectroscopy (FT-IR), field emission scanning electron microscopy (FE-SEM), X-ray powder diffraction analysis (XRPD), measurements of contact angle, and thermogravimetric analysis, respectively.



Nanofibrillated cellulose (NFC) was supplied by the Centre for Biocomposite and Biomaterial Processing, University of Toronto, Canada, as a water suspension with a solid content of 1.6 wt%. The homogenised NFC suspension, obtained through mechanical disintegration of softwood pulp, consisted of cellulose nanofibrils with diameters in the range of 20 to 60 nm. The pure cellulose content in this sample was 91%, the lignin content was less than 0.3%, and the remaining components were primarily hemicelluloses. Acetic anhydride (Ac2O, ≥99%), N,N-dimethylformamide (DMF, anhydrous, 99.8%), pyridine (anhydrous, 99.8%), toluene (99.5%), chloroform (anhydrous, ≥99%, with 0.5-1.0% ethanol as stabiliser), ethanol (96%), and acetone (99.5%) were all purchased from Sigma-Aldrich (Steinheim, Germany). The reagents were used as received without further purification.

Pre-treatment of NFC

An aqueous NFC suspension was solvent-exchanged using five sequential centrifugations and re-dispersion operations at 7830 rpm and 10 °C for 20 min, first into acetone and then into DMF at a final concentration of 1 wt%. In separate experiments, the NFC suspensions were dried using freeze- and spray-drying processes. Prior to freeze drying, the NFC suspension was frozen using liquid nitrogen and then lyophilised (LyoQuest freeze dryer, Telstar) for 72 h. The pressure within the freeze drying system was set to 0.040 mbar, the temperature of the plates to 22 °C, and the temperature of the condenser to -50 °C. The spray drying of the NFC suspension (0.5 wt%) was performed using a Büchi B-290 lab-scale mini-spray-drying unit (Büchi Corporation, Switzerland) with a water evaporating capacity of 1 L h-1 under the following conditions: inlet temperature 160 ± 10 °C, outlet temperature 50 ± 10 °C, spray flow 750 L h-1, aspirator rate 100%, and peristaltic pump 50% (corresponds to approx. 8 mL min-1). The powdery sample was collected from a separation cyclone with a yield of 68% based on the dry matter. The dried forms of NFC were dispersed into DMF at a concentration of 1 wt% using homogenisation (Ultra Turrax T 25 basic, IKA-Werke, Staufen, Germany) and high intensity ultrasonication (Ultrasonic Vibra cell VC500 (Sonics and Materials, USA), 19 mm needle probe tip, 60% output amplitude, for 10 and 5 min, respectively. The DMF/NFC suspensions obtained are hereafter referred to as solvent exchanged (NFCSE), freeze dried (NFCFD), and spray dried (NFCSD) nanofibrillar cellulose.

Heterogeneous Acetylation of NFC

A 100 mL suspension of DMF/NFC (1% based on the dry weight) was mixed with 35 mL (37.8 g; 0.37 mol) of acetic anhydride (Ac2O). A predetermined amount of pyridine, which served as a catalyst, was added into the beaker and mixed thoroughly with the reaction mixture under slow homogenisation for 10 min. The pre-treated NFC sample was added into a round-bottomed flask equipped with a condenser and a magnetic stirrer. The reaction was performed under a nitrogen flow and kept at the required temperature of 105 ± 5 °C. At the specific reaction intervals of 30, 60, 300, 600, 900, and 1200 min, samples were withdrawn from the reaction mixture. A series of acetylated products, namely ANFCSE, ANFCSD, and ANFCFD, with different degrees of substitution (DS) were obtained. The effect of the pyridine concentration on the extent of acetylation was studied by adding 1% (P1; 0.02 mol), 2% (P2; 0.04 mol), and 3% (P3; 0.06 mol) (v/v) of pyridine. Acetylated samples treated with different pyridine concentration were marked as ANFCn P1, ANFCn P2, and ANFCn P3, where n indicates the solvent exchanged (SE), spray-dried (SD), and freeze-dried (FD) NFC. At the end of every experiment, the mixture of NFC and chemical reagents was cooled to room temperature and then the acetylation by-products and remaining reagents were removed through five repeated centrifugation and re-dispersion steps with an 800 mL toluene/ethanol/acetone suspension (4/1/1 by v/v/v). The modified samples were dried in a laboratory oven with circulating air at 105 °C for 24 h and stored in desiccators prior to analysis.

Characterization of the Unmodified and Acetylated NFC

Determination of the acetyl content using FT-IR spectroscopy

The infrared spectra of the unmodified and acetylated NFCSE and NFCFD sheets and powder in the case of NFCSD were recorded using a Spectrum One FTIR spectrometer (Perkin Elmer, USA) in attenuated-total-reflection (ATR) mode on a ZnSe crystal. The spectra were collected at a resolution of 4 cm-1, over the range from 650 to 4000 cm-1. A total of 64 scans were used to collect each spectrum. According to Tingaut et al. (2010), the peak heights at 1060 cm-1 (H1060) and 1740 cm-1 (H1740) were calculated using baselines constructed between 1500 and 860 cm-1 and between 1790 and 1690 cm-1, respectively. The peak areas at 1740 cm-1 (A1740) and 1370 cm-1 (A1370) were measured using baselines constructed between 1790 and 1690 cm-1 and between 1394 and 1347 cm-1, respectively. The acetyl content of the modified NFCs was calculated from the ratios A1740/ H1060, H1740/ H1060, and A1370/ H1060 (Tingaut et al. 2010), and the average values are reported in this study. Bands of FT-IR spectra were assigned according to Tingaut et al. (2010), Adebajo and Frost (2004 a, b), and Sun et al. (2002). The acetyl content (Ac %) was determined according to Tingaut et al. (2010) using FT-IR calibration curves and the degree of substitution (DS) was then calculated using Eq. 1 (Fordyce et al. 1946):


It needs to be emphasized that Eq. 1 considers all hydroxyl groups in NFC and not just those on the surface.

Field emission scanning electron microscopy (FE-SEM)

The morphology of the samples was observed in secondary electron mode using a field-emission scanning electron microscope (Zeiss ULTRA plus, Germany) with an acceleration voltage (EHT) of 1 and 2 kV and a working distance of 4.7 mm. The 0.5 wt% suspensions of unmodified and acetylated NFCs (NFCSE, NFCSD, NFCFD, and ANFCSE, ANFCSD, ANFCFD) in CHCl3 were sonicated at room temperature for 3 min using an Ultrasonic Vibra cell VC500 (Sonics and Materials, USA). A droplet of the suspension was deposited on a glass plate and dried for 1 h at room temperature. Samples were coated with a highly conductive film of gold using a BAL-TEC/SCD_500.

Thermogravimetric analysis (TGA)

The thermal stability of unmodified and acetylated NFC samples was determined using a thermogravimetric analyser (Mettler Toledo TGA/SDTA 851e). The samples were heated in a corundum crucible with a diameter of 8 mm from 25 to 500 °C under a dynamic argon atmosphere (100 mL min-1) and a heating rate of 20 K min-1. The baseline was subtracted in all measurements.

X-ray powder diffraction analysis (XRPD)

X-ray powder diffraction (XRPD) patterns were collected using a Siemens D5000 X-ray powder diffractometer (XRPD) equipped with Cu Kα1,2 radiation (λ = 1.5418 Å, 40 kV, and 30 mA), working in reflection mode, from 2θ = 5° to 2θ = 50° with a step size of 0.040°. The unmodified and acetylated NFCSE and NFCFD sheets and the NFCSD powders were uniaxially pressed to form pellets with thicknesses of 1 mm, and their surfaces were analysed. Simulated diffraction patterns of cellulose Iβ obtained by using Mercury 3.3 software from the Cambridge Crystallographic Data Centre (French 2014) were compared to the experimental patterns.

The internal reference peak height method, developed by Segal and co-workers (1959), was used for comparing the relative differences between the unmodified and acetylated samples. The estimated crystallinity index (Icr) was determined using Eq. 2, where I200 (2θ = 22.7°) represents both crystalline and amorphous material, while Iam (the minimum between the (200) and (110) peaks, 2θ = 18°) represents amorphous material.


The crystallite size and the number of cellulose chains for both experimental and simulated diffraction patterns were determined by the Scherrer equation (Eq. 3) and by dividing the crystallite size by 3.9 Å representing the thickness of the cellulose chain (Nishiyama et al. 2012; French and Santiago Cintrón 2013), respectively.


In Eq. 3, K is a constant (1.0), λ is the X-ray wavelength, β is the full width at half maximum in radians, and θ is the half of the plotted 2θ value at the position of the studied peak.

Hydrophobicity measurements

Static contact angle measurements were performed at room temperature using a Krüss, DSA100 contact angle system. Prior to contact angle measurements, pellets were prepared with a smooth surface and a diameter of 11 mm by pressing the unmodified and acetylated NFC on a hydraulic pressing machine with a force of 10 tons. A droplet of distilled water (vol. 1.5 µL) was then deposited onto the surface of the pellets and the contact angle was measured after 1 s. Five droplets were deposited on different positions of the sample to reduce the possible influence of the heterogeneity of the surface. The average values of the contact angle measurements are reported.

Determining the residual water content of NFC/DMF suspensions

The water content in the initial DMF/NFCs suspensions was analysed according to the principle of dry-coulometric Karl-Fischer back titration using a Mettler Toledo C30 Coulometer device with a Stromboli sample changer oven (Mettler- Toledo GmbH, Greifensee, Switzerland). The temperature was set to 105 °C. The sample was weighed at a precision of 0.1 g and swept with a molecular-sieve dried air stream for approximately 8 min.


FT-IR Spectra of Unmodified and Acetylated NFC

FT-IR spectra for all acetylated NFC samples (Fig. 1) showed three main ester bands at 1740 (carbonyl C=O stretching of ester), 1369 [C-H in -O(C=O)-CH3], and 1234 cm-1 (C-O stretching of acetyl group). The intensity of these bands increased with increasing reaction time. On the other hand, the intensities of the -OH stretching band at 3337 cm-1 and the -OH in-plane bending bands at 1337 and 1310 cm-1 decreased, which was much more obvious in the cases of ANFCFD and ANFCSD. The absence of typical twin bands of acetic anhydride, which appear in the region from 1840 to1760 cm-1, indicated that there was no reactant present in acetylated NFCs. The absence of a peak at 1700 cm-1for a carboxylic group in the spectra of the acetylated samples also indicated that the acetylated products were free of the acetic acid by-product (Sun et al. 2002). When comparing the FT-IR spectra of the acetylated NFCs, it was found that both the reaction time and sample pre-treatment had an influence on the DS (Fig. 2). ANFCSD and ANFCFD exhibited pronounced acetylation peaks, particularly at longer reaction times, compared to ANFCSE samples. The observed differences may have been caused by trace water in the reaction mixtures, originating from the possibly incomplete solvent exchange procedure for the NFCSE samples. Another reason for the more pronounced intensities of the -OH groups in the case of ANFCSE can be ascribed to the erosion mechanism of microfibrils (Sassi and Chanzy 1995) caused by acetylation where new -OH groups are exposed on the surface. Sassi and Chanzy (1995) proposed the erosion mechanism where a part of a cellulose chain located at the surface is sufficiently acetylated. It then becomes soluble in the acetylating medium and is lifted from the surface of the crystalline lattice.

Fig. 1. FT-IR spectra of the unmodified and acetylated (a) NFCSE, (b) NFCSD, and (c) NFCFDobtained after different reaction times: 0 (unmodified), 60, 600, and 1200 min. Pyridine concentration of 1 vol.% and 3 vol.% at 1200 min is labelled by P1 and P3, respectively.

Degree of Substitution for Acetylated NFC

The catalyst concentrations, the values of DS, and the residual water contents for each acetylated product are listed in Table 1. The esterification reaction in the presence of low catalyst concentrations (1 vol.%) gave DS values in the range of 0.48, 0.85, and 1.00 for ANFCSE, ANFCSD, and ANFCFD, respectively (Table 1). When the same reaction was carried out in the presence of higher catalytic amounts, 2 vol.% and 3 vol.%, higher values of DS were obtained for ANFCSE and ANFCSD (Table 1). According to Sassi and Chanzy (1995), the acetylation reaction starts within the amorphous region and in a second step the acetylation occurs in the crystallites as an erosion mechanism. It was assumed that other available hydroxyl groups, which might be located deeper in the nanofibrillar structures, were involved in the reaction as well, especially at the highest pyridine concentration for NFCSD. Higher pyridine concentration had no impact on the DS for ANFCFD.

Table 1. Degree of Substitution of the Acetylated NFCSE, NFCSD, and NFCFD Samples at Different Catalyst Concentration and Reaction Time 1200 min. *

* Residual water content is given for NFC/DMF samples.

Figure 2 shows the DS as a function of the reaction time for NFCSE, NFCSD, and NFCFD. For all three materials, the DS increased with reaction time (Fig. 2).

Fig 2_biores

Fig. 2. The degree of substitution as a function of reaction time for the acetylated NFCSE, NFCSD, and NFCFD at 1 vol.% of pyridine

The DS values increased rapidly during the first hour of acetylation. Thereafter, the DS increased gradually until 900 min, where an additional rise in the DS was observed for all acetylated NFC samples, regardless of the applied pre-treatment. This was pronounced for ANFCSD and ANFCFD, where DS was higher than for the solvent exchanged ANFCs, suggesting that drying increased the susceptibility of NFC for acetylation. It is assumed that the drying pre-treatment significantly increased the swelling ability of the cellulose and the diffusion rate of the acetic anhydride and pyridine, which was evaluated from the measurements of the size of the average diameter of the nanofibrils (Fig. 3).