Introduction

Acetylation is a common chemical modification in which acetyl groups (CH3CO−) react with the surface hydroxyl groups (OH) of cellulose, making its surface less hydrophilic. The acetylation process depends on the fiber accessibility and the susceptibility of OH groups in the crystalline and less crystalline domains of cellulose (Kalia et al. 2014). The greater the accessibility, the easier it is for the reactants to diffuse into the interior of the fibers. The generic methods for acetylation are those in heterogeneous (in fiber dispersions) and homogeneous (in solution) phase. The heterogeneous acetylation process is performed in the presence of a non-solvent, such as toluene, benzene or carbon tetrachloride. The reaction product (cellulose acetate) is insoluble and thereby, this process preserves the morphological structure of the fiber. In contrast, cellulose acetate is dissolved during the homogeneous acetylation and, therefore, it demands solvents capable of deconstructing the crystalline network and interacting with the anhydroglucose units of cellulose. This is usually done by reducing or eliminating inter and intra-molecular hydrogen interactions. In the homogenous phase, cellulose begins to react with acetic anhydride, with the initial reaction occurring mainly in the amorphous regions of the structure. Sulfuric acid is used as a catalyst and it combines with the cellulose, forming sulfate linkages; however, most of these are removed during acetylation via exchange with acetyl groups. It is important that the final cellulose acetate contains only a very small amount of sulfate groups because they affect the properties adversely, especially the color. When acetylation is virtually complete, the product of reaction is viscous and clear. The excess of acetic anhydride is then neutralized by adding aqueous acetic acid, which helps to desulfate the residual sulfate linkages (LaNieve and Richard 2007; Luo et al. 2013).

The extent to which the available hydroxyl groups in the repeating unit of cellulose are substituted, the degree of substitution (DS), does not quite reach the maximum of three units per anhydroglucose unit (as in cellulose triacetate). Cellulose triacetate (DS > 2.8) displays a limited solubility in acetone and is reported for use in a relatively narrower number of commercial applications (Cao et al. 2007). Diacetates with a DS from 2.2 to 2.7 (also named secondary acetates) are the most commonly reported cellulose esters. They are soluble in acetone and other organic solvents (Steinmeier 2004; Fischer et al. 2008; Wan Daud and Djuned 2015), and can be used in applications such as coatings, films, textiles, synthetic polymeric membranes, among others.

Following a heterogeneous route, it is possible to obtain more crystalline and less biodegradable cellulose acetates (CA) than those produced through homogeneous routes (Barud et al. 2008). On the other hand, the advantages of acetylation in homogeneous phase include the excellent control of the degree of substitution (DS) and the possibility of a uniform distribution of the functional groups along the polymer chain (Ass et al. 2004).

Importantly, CA is usually produced from high quality cellulose fibers, namely, dissolving grades derived from cotton or wood (α-cellulose content of > 95%) (Saka and Matsumura 2004; Roselli et al. 2014; Wan Daud and Djuned 2015). In the case of cotton sources, issues related to the large land area required for farming and water required for irrigation, result in high economic and environmental burdens. Further, the so-called “cotton gap” motivates a need for more extensive utilization of dissolving-grade fibers derived from wood. According to FAO (2012), dissolving-grade fibers constitute a small share of the global pulp production, but prospective consumer markets indicate that this share will increase in the coming decades. Based on this scenario, new technologies are being suggested as alternative to traditional dissolving pulp production processes. In previous studies (Quintana et al. 2013, 2015a), the laccase–mediator system was used to bleach sulfite pulp and the conversion to dissolving-grade was achieved by cellulase treatment. According to the results, the obtained chemo-enzymatic dissolving-grade fibers exhibited suitable characteristics for use in the synthesis of cellulose derivatives.

In the present work, fibers obtained via chemo-enzymatic treatments of biobleached fibers (termed herein as LE and LCE) were investigated as far as their suitability to synthesize acetylated cellulose. A bleached commercial dissolving-grade fiber, used as a reference and termed “Com”, was used for comparison. This study focuses on the surface acetylation reactions, typical of heterogeneous acetylation, while homogeneous acetylation was also carried out for comparison purposes. In terms of surface acetylation, given doses of acetic anhydride (Ac2O) were tested and the degree of acetylation was evaluated by FTIR spectroscopy. Paper handsheets were produced from fibers that were acetylated on the surface (heterogeneous reaction) and characterized in terms of contact angle, mechanical strength and surface morphology. Samples obtained by homogeneous acetylation were used to prepare transparent films via solvent casting and characterized in terms of the tensile strength and contact angle. This work, therefore, aims at determining if chemo-enzymatic treatment is a suitable alternative for the synthesis of materials with low hydrophilicity via heterogeneous and homogeneous acetylation.

Materials and methods

Precursor fibers

As starting fiber material, unbleached sulfite cellulose fibers were used and obtained as a mixture of 60% Norway spruce (Picea abies) and 40% Scots pine (Pinus sylvestris) (Domsjö Fabriker mill, Sweden). Fiber characteristics included a kappa number of 4.2 ± 0.2, ISO brightness of 61.25 ± 0.6% and viscosity of 511 ± 11 mL/g. The carbohydrate content, as determined by high-performance liquid chromatography (HPLC), was 88.5 ± 0.3% glucan, 6.0 ± 1.3% mannan, 2.4 ± 0.4% xylan and 0.3 ± 0.2% rhamnan. As a reference fiber source, a totally chlorine-free (TCF) bleached sulfite dissolving-grade pulp was employed. This pulp was obtained from the unbleached fibers (as indicated above). It has an ISO brightness of 91.70 ± 0.15% and viscosity of 474 ± 1 mL/g. The carbohydrate composition, also determined by HPLC, included 95.1 ± 0.3% glucan, 2.8 ± 0.2% mannan, 0.8 ± 0.0% xylan, 0.2 ± 0.2% rhamnan, 0.2 ± 0.2% arabinan, 0.3 ± 0.1% glucuronic acid and 0.2 ± 0.1% acetic acid. The bleached fibers were obtained by sulfite digestion followed by chemical bleaching at the Domsjö Fabriker mill (Sweden). These fibers, which are used commercially, are thereafter referred to as Com.

Enzyme treatment

A laccase (Trametes villosa, TvL) was supplied by Novozymes® (Denmark) with an activity of 746 U/mL and used for biobleaching the fibers. The laccase activity was measured as the extent of oxidation of 5 mM 2,20-azinobis(3-ethylbenzothiazoline-6-sulphonic acid) (ABTS) to its cation radical (ε436 = 29,300 M−1 cm−1) in 0.1 M sodium acetate buffer (pH 5) at 24 °C. One activity unit (U) was defined as the amount of enzyme converting 1 µmol of ABTS per min. Violuric acid (VA), the mediator used for this enzymatic treatment, was purchased from Sigma-Aldrich and used as received. A hydrolytic treatment was also applied involving an endoglucanase produced from Cerrena unicolor (supplied by Fungal Bioproducts®, Spain). The activity measured as U/g dry enzyme powder of the cellulase preparation was 1700 CMCase U/g and 680 U/g for the cellulase and xylanase activity, respectively. The activity was determined in our laboratory using the Somogyi–Nelson method.

For biobleaching, unbleached sulfite fibers were first conditioned at pH 4 adjusted with H2SO4, stirred at 2% solids content for 30 min and washed with de-ionized water in a glass filter funnel. This step was needed to remove contaminants and metals, and also to bring the fiber dispersion to the pH required for the enzymatic treatment. The biobleaching process included a sequence denoted as LVA(PO)(PO), where LVA denotes an enzymatic (laccase) treatment and PO the hydrogen peroxide stage assisted with oxygen. The enzymatic stage was carried out with the laccase–violuric acid system in an oxygen pressurized reactor (0.6 MPa) at stirring rate of 30 rpm, using 50 mM sodium tartrate buffer (pH 4) to adjust 5% (w/w) fiber content, at 50 °C for 4 h. The enzyme dose was 20 U/g odp (oven dry weight of fibers) of laccase and 1.5% odp of violuric acid (Quintana et al. 2013). The enzymatic treatment was followed by a chemical bleaching stage involving hydrogen peroxide assisted with oxygen. PO was carried out at 5% (w/w) solids in an oxygen pressurized (0.6 MPa) reactor at a stirring rate of 30 rpm under the following conditions: 3% odp H2O2, 1.5% odp NaOH, 0.3% odp DTPA and 0.2% odp MgSO4, at 90 °C for 1 h. Treated fibers were washed extensively with deionized water, and then followed with another hydrogen peroxide stage assisted with oxygen. The treatment was performed under same conditions described above but 2.5% odp H2O2 and 3 h of reaction were used. The chemical stage was finished by washing the bleached fibers with deionized water.

The resulting biobleached fibers (LVA(PO)(PO)), denoted here as L, for simplicity, were used in two different additional treatments to produce the chemo-enzymatic samples used later for acetylation reactions. One was subjected to enzymatic hydrolysis with an endoglucanase (resulting in fibers that are denoted thereafter as LE). The other included the application of cold caustic extraction before endoglucanase treatment (resulting in fibers that are denoted thereafter as LCE). The purpose of introducing an endoglucanase treatment was to improve fiber reactivity. By its side, cold caustic extraction was a purification stage where hemicelluloses were removed and, as a result, fiber quality was improved. Both enzymatic treatments were performed in polyethylene bags that were placed in a laboratory water bath, at 10% solids (w/w) in 0.05 M sodium acetate buffer at pH 5.5 at 55 °C for 1 h and with 12 U/g odp enzyme. The samples were periodically kneaded and the reaction was stopped by washing the fibers with de-ionized water in a porous glass filter funnel of porosity grade 2. The cold caustic extraction was also conducted in a polyethylene bag. The treatment was performed at 10% (w/w) solids adjusted with 9% (w/v) NaOH at 25 °C for 1 h. Treated fibers were washed with de-ionized water until the filtrate pH was neutral (Quintana et al. 2015a).

LE, LCE, and Com fiber analysis

The commercial dissolving grade and chemo-enzymatic fiber samples (Com, LE and LCE) were characterized in terms of kappa number, brightness and viscosity according to ISO 302:2004, ISO 2470:2009, ISO 5351:2004, respectively. The cellulose reactivity of the fiber samples was determined according to slightly modified version of Fock’s method (Fock 1959; Köpcke et al. 2010). This is a micro-scale method simulating the industrial viscose process for manufacturing regenerated cellulose. Prior to analysis, the samples were dried at 50 °C and conditioned in a climate room at 23 °C and 50% RH overnight. Carbohydrate composition of treated fibers was determined using high performance liquid chromatography (HPLC). Samples were studied by duplicate using a modified version of TAPPI 249 cm-09 test method. Prior to HPLC analysis, samples were filtered using a 0.45 µm pore size Whatman membrane. Chromatographic analysis was performed using a 1200 Agilent HPLC instrument furnished with a Biorad Aminex HPX-87H ion-exchange column. Concentrations were calculated by interpolation in calibration curves ran from standards of glucose, xylose, rhamnose and arabinose. In order to resolve xylose, mannose and galactose peaks, the hydrolyzed effluents were neutralized with barium carbonate (BaCO3), then were filtered through a membrane of 0.45 µm pore size and then were analyzed with a Biorad Aminex HPX-87P column. The chromatographic determination was performed with the following conditions: mobile phase, 6 mmol/L (acid samples) or ultrapure water (neutralized samples); flow rate, 0.7 mL/min; column temperature, 60 °C (acid sample) or 80 °C (neutralized sample).

13C-CP/MAS NMR spectra were recorded in a Bruker AMX-300 instrument operating at 7.05 T and at 75.5 MHz for 13C. Samples were immersed in deionized water for at least 2 h. All measurements were performed at 290 ± 1 K. The magic angle spinning (MAS) rate was 4 kHz. The cross-polarization contact time was 1 ms and the recycle delay time 2.5 s. Acquisition time was 98.3 ms and sweep-width was 31.2 kHz. The number of scans was 5100.

Heterogeneous (surface) and homogenous (bulk) acetylation

In the heterogeneous phase acetylation, 2.0 g oven dried pulp (odp) of each type (Com, LE, LCE) was disintegrated and then filtered using a filter paper (Whatman 1) for water removal. The samples were then placed in a glass beaker containing a mixture of 20 mL of acetic acid (99.7% w/w) and 35 mL toluene. The dispersion was stirred for 5 min and 0.2 mL sulfuric acid (95% w/w) was added. Then, a desired amount of acetic anhydride (Ac2O) was added and the mixture was stirred for 1 h at room temperature. The specific conditions for acetylation reactions and the nomenclature used were as follows: 0.53 g (lowest), 2.67 g (low), 5.35 g (medium) and 10.7 g (high) Ac2O per gram of dried fiber sample (Com, LE, LCE). The “lowest” conditions were not applied to the LCE pulp. The reaction was quenched by adding 6 mL of distilled water and ethanol, 3:7 v/v. The mixture was allowed to stand for 20 min and then washed 3 times with methanol and finally with water until neutral pH (Fig. 1).

Fig. 1
figure 1

Outline of experimental procedures and samples studied. Biobleached (L or LVA(PO)(PO)) sulfite fibers were subjected to cellulase treatment (1a, LE) or cellulase treatment after cold caustic extraction, CCE (1b, LCE). The fibers after LE, LCE treatment were subjected to heterogeneous (surface) or homogeneous acetylation reactions. Fiber handsheets or films were prepared and characterized. Bleached commercial dissolving fibers (Com) were used as a reference, and same heterogeneous and homogeneous acetylation reactions were performed on such reference fibers

Full size image

Homogeneous acetylation was performed as reference. For this purpose, 2.5 g odp of respective fiber type (Com, LE, LCE) was disintegrated and then filtered using a filter paper for water removal. Then, 50 mL of acetic acid was added to the sample, stirred 5 min and then filtered. This step was done by duplicate. After filtration, 45 mL of acetic acid and 0.25 mL sulfuric acid was dropped into the sample and stirred for 1 min. Then, 5.35 g Ac2O/g dried fiber (~ 12.5 mL Ac2O) was added and continuously stirred for 30 min at room temperature. The reaction was quenched with the addition of 6.25 mL of distilled water and acetic acid at a ratio of 3:7 v/v, respectively. Finally, cellulose acetate (CA) was obtained by pouring the viscous reaction mixture into distilled water obtaining a continuous droplet and with constant stirring. With precipitation, cellulose acetate was regenerated. The obtained product was washed with distilled water until neutrality and subsequently dried using a freeze-drying (Fig. 1).

The acetylated samples were analyzed by Fourier transform infrared spectroscopy (FTIR) by using a Nicolet Avatar 360 spectrophotometer (Nicolet Instrument Corporation). The samples were prepared by mixing 1 mg of the sample in a matrix of 300 mg of KBr followed by pressing. The spectrum was recorded in the range of 400–4000 cm−1 and 32 scans were run at 4 cm−1 resolution.

Determination of acetyl group content of acetylated cellulose

The nominal degree of substitution was determined according to ASTM D871-96 (2010). Firstly, the respective acetylated sample was ground and 100 mg (oven dried) were weighed accurately and placed into 20 mL of 75% v/v of ethanol in an Erlenmeyer flask. The bottle, loosely stoppered, was heated to 50–60 °C for 30 min for better swelling of the material. Then, 20 mL of 0.5 N NaOH solution was added to the sample and the mixture was heated to 50–60 °C for 15 min. A blank was also conducted but in absence of fiber sample. The flasks were stoppered tightly and allowed to stand at room temperature for 72 h. The excess alkali was then titrated with 0.5 N HCl using phenolphthalein as indicator. An excess of about 1 mL of 0.5 N HCl was added and allowed the NaOH to diffuse from the regenerated cellulose overnight. The small excess of HCl was titrated with 0.5 N NaOH to a phenolphthalein end point. The percentage of acetyl groups was calculated as follows:

$$ Acetyl\,groups,\,{\text{\% }} = \left[ {\left( {D - C} \right)Na + \left( {A - B} \right)Nb} \right]\cdot\left( {F/W} \right) $$

where A and B are the volumes (mL) of the NaOH solution (normality = Nb) required for titration of the sample and the blank, respectively. C and D are the volumes (mL) of the HCl solution (normality = Na) used for the titration of the sample and the blank, respectively.

F is a constant (4.305) for acetyl and W the mass (g) of the sample used.

Handsheets from surface acetylated fibers and cellulose acetate films

Fibers obtained by surface acetylation were used for preparing handsheets. For sheet manufacture 1 g of each sample at 1% solids was disintegrated and poured into an over-pressurized device (< 1 bar pressure difference) allowing few minutes drainage to obtain a web or handsheet of the acetylated fibers. The device was equipped with open mesh fabric screen (Sefar Nitex 03-10/2, mesh opening of 10 µm with open area of 2%) to remove the excess water and retain the fibrils. The webs were pressed between two blotting papers using a metal roller (10 kg) and then dried at 80 °C, for 1 h in a tumble drier. The obtained sheets were then stored in a conditioned room (23 °C and 50% relative humidity) until further use.

Surface acetylated handsheets were used to determine different properties. The morphological characteristics of fibers (viz., length, width and curl), and fine content were determined in accordance with TAPPI T 271 on a Metso kajaaniFS300 fiber analyzer. High-resolution imaging of surfaces (handsheets were taken on a JEOL JSM-6400 scanning electron microscope (SEM). Samples were placed on the SEM sample holding stub with the aid of conductive double side sticky carbon film and coated with Au/Pd alloy prior to analysis. The wetting characteristics of the acetylated handsheets was determined by the initial water contact angle (WCA) using a Dataphysics OCA15EC contact angle goniophotometer (Dataphysics, USA). A 4 µL water drop was dropped to the sample surface, and an image capture ratio of 25 frames/s was used to calculate the initial contact angle. A minimum of ten readings were taken on every sample to reduce possible influence of the heterogeneity of the surface. Also, changes in contact angle were monitored until complete absorption of each water drop. Wet and dry tensile strength of the surface acetylated sheets were measured on a MTS 400/M Vertical Tensile Tester equipped with a 50 N load cell, in accordance with ISO 1924-3:2005.

Cellulose acetate (CA) obtained from homogeneous acetylation reaction was used for preparing transparent films by means of a casting technique. Dried cellulose acetate was dissolved in given amounts of acetone in order to obtain a concentration of 8 wt%. The solutions for film casting were firstly centrifuged at 6000 rpm for 10 min. The supernatant was carefully transferred and centrifuged again at 2000 rpm for 5 min. The films were cast by pouring the transparent solution on a glass plates, well distributed and followed by drying in a vacuum desiccator for at least 2 h. The film samples were finally kept in a desiccator. Tensile strength tests for CA films resulted from homogeneous acetylation reactions were performed on a MTS 400/M vertical Tensile Tester, with a cross-head speed of 40 mm/min. Specimen strips presented 10 mm width and 40 cm length. Note that comparison of the handsheet (paper) and film samples is not possible since they are quite different systems. The water contact angle, water drop test (Tappi standard T835 om-08) and dry zero-span strength (ISO 15361:2000) were also determined.

Results and discussion

Precursor fiber characterization

The effect of acetylation on the quality of the systems obtained from LE, LCE or Com was evaluated. The characteristics of respective treated fibers are indicated in Table 1. All applied sequences resulted in similar lignin content, as assessed by the kappa number, ISO brightness and viscosity. However, the commercial dissolving fibers (Com) exhibited the highest ISO brightness. Similar values of Fock solubility were found for all the fibers. Some authors obtained a higher Fock solubility if endoglucanases were applied after a cold caustic extraction stage (concentration > 8 wt%). This was due to the transformation of cellulose I into cellulose II and the fact that endoglucanases have a greater affinity for the latter allomorph (Engström et al. 2006; Köpcke et al. 2008; Gehmayr and Sixta 2011; Quintana et al. 2015b). However, although LCE solubility tended to be higher compared to LE, no significant differences were produced. It is also known that lower viscosity can influence cellulose solubility (i.e. reactivity); however, in general, all fibers presented comparable viscosity.

Table 1 Main characteristics (mean ± SD) of LE, LCE fibers as well as Com reference
Full size table

The carbohydrate composition was determined by HPLC, with special attention to the hemicelluloses content (Fig. 2). The endoglucanase treatment applied to the biobleached fibers (L) to obtain LE reduced the amount of hemicelluloses by 35%, especially the mannan and galactan fractions. The introduction of a cold caustic extraction followed by hydrolytic treatment (LCE) further decreased the amount of hemicelluloses by 46.2%. To be precise, compared to LE, LCE treatment contribution amounted to 11.2%, resulting in a smaller xylan fraction and similar mannan and galactan content. The lowest hemicellulose content was measured in Com.

Fig. 2
figure 2

Hemicellulose composition of biobleached fibers before (L) and after chemo-enzymatic treatment (LE and LCE). The composition of commercial dissolving fibers (Com) is also indicated. The total content of hemicelluloses is indicated on top of each column. Mn mannan, Xn xylan, Glt galactan, Rn rhamnan, Others it includes acetic acid, glucuronic acid and galacturonic acid

Full size image

Solid state 13C-NMR spectra of biobleached fibers (L), biobleached fibers followed by endoglucanase treatment (LE), and biobleached fibers submitted to cold caustic extraction (9% (w/v) NaOH) followed by endoglucanase treatment (LCE) are included as Supporting Information (Fig. S1). LCE treatment presented slightly different polymorphic form from unbleached and LE sample. Caustic extraction converted cellulose I to cellulose II: the C-6 signal at 64 ppm increased, obtaining two peaks with nearly identical heights at 66 and 64 ppm. However, a shoulder at 108 ppm of C-1 signal, which is characteristic of cellulose II, was not observed (Janzon et al. 2008a). Note that the small proportion of cellulose II in LCE is associated with the similar Fock solubility values between samples (Krässig 1993; Janzon et al. 2008b).

The enzymatically-treated pulps displayed properties comparable to those of commercial dissolving pulp. The environmental advantages of enzymatic technologies have been reported previously via Life Cycle Assessments (LCA), which indicated a reduced contribution to global warming. In addition, a reduced contribution to acidification, eutrophication, photochemical ozone formation and energy were noted (Jegannathan and Nielsen 2013). Skals et al. (2008) also reported that the introduction of xylanase in biobleaching contributed to reduce global warming. Zhi Fu et al. (2005) showed that the introduction of a laccase–mediator stage in biobleaching reduced the contribution to ozone depletion and acidification, as well as reducing solid waste generation and energy consumption. Related work highlighted the benefits of producing the enzyme and mediator at the point of use.

Heterogeneous acetylation

Different degrees of surface acetylation were achieved by varying the concentration of acetic anhydride (Ac2O) in the nonpolar solvent used. The Ac2O loading correspond to relative activities denoted thereafter as “lowest”, “low”, “medium” and “high”. The effect of acetylation on LE, LCE and Com samples was assessed via FTIR spectroscopy (Fig. 3). Changes in non-acetylated and acetylated samples were identified. Specifically, the structural changes of acetylated fibers were confirmed by the appearance of three new bands characteristic of the acetyl group vibration at about 1735–1740, 1368–1375 and 1259–1277 cm−1. The peaks located at 1735–1740 cm−1 were attributed to the C=O stretching of carbonyl in the ester bonds. The peaks located at 1368–1375 cm−1 were assigned to C–H symmetrical deformation in methyl group. The vibration peaks between 1259 and 1277 cm−1 corresponded to C–O stretching of the acetyl group. The absence of peaks in the 1840-1760 cm−1 region demonstrated that there was no residual, unreacted acetic anhydride in the acetylated fibers (Rodionova et al. 2011; Cunha et al. 2014; Muhammad Djuned et al. 2014; Mashkour et al. 2015).

Fig. 3
figure 3

FTIR spectra for LE, LCE and Com (commercial dissolving fibers) samples at different acetylation levels (lowest, low, medium and high)

Full size image

In the case of Com fibers, it is noted that by increasing the amount of Ac2O used for acetylation resulted in a higher intensity of the C=O band at 1735 cm−1; at the same time, a decrease in the C–O band at 1235 cm−1 was clear. Although the C–H band at 1375 cm−1 is characteristic of acetylated fibers, no variation in absorption was evident for the different acetylated conditions. A similar observation applies to LE and LCE fibers but differences in the intensity peak at 1735 cm−1 with respect to the acetylation conditions were less pronounced. For both, LE and LCE, a high intensity peak at 1735 cm−1 was observed when a high (10.7 g) Ac2O level was introduced.

The degree of acetylation (i.e. acetyl group content) was determined by titration with NaOH and HCl (Fig. 4). LE sample did not show differences in terms of acetyl content after reaction with lowest and medium Ac2O levels, but a significant gain in acetyl group content was observed for the high dose level (10.7 g). In fact, from all studied fibers, the highest acetyl group content (~ 24%) was determined for the LE sample. In general, compared to LE, acetylation of LCE fibers yielded a smaller amount of acetyl groups. The application of “low” Ac2O levels was not effective in incorporating enough acetyl groups. Only by using two or four-fold the Ac2O dosage level introduced a suitable amount of acetyl groups. Specifically, 13 and 15% of acetyl group content were measured for medium and high Ac2O dosages. In the case of Com sample, a gradual improvement in acetyl group content was observed from the lowest (0.53 g Ac2O) to the medium (5.35 g Ac2O) Ac2O addition. Unexpectedly, a high Ac2O addition produced a relatively small acetylation degree. Com subjected to medium conditions resulted in 15% acetyl group substitution, while only 7.9% was measured at high Ac2O levels (a 47.3% reduction). These results were not in agreement with FTIR data that indicated that the sample treated under “high” conditions displayed the highest peak intensity in the C=O band. Actually, the same acetyl content was found using low and high amounts of Ac2O. Several reasons can explain these observations. For example, the distribution of the functional groups along the polymer chain may not be uniform after heterogeneous acetylation, which introduces artifacts in the determination of acetyl group content.

Fig. 4
figure 4

Content of acetyl group (%) as a function of acetic anhydride used in the respective acetylation reactions for LE (diagonal bar), LCE (dotted bar) and Com (filled bar) samples

Full size image

Changes in fiber morphology upon chemo-enzymatic treatment and acetylation

Endoglucanase treatment (LE) of the biobleached fibers (L) caused a significant reduction (~ 65%) in fiber length and increase of the fines content (Table S1 of Supporting Information). The fiber length decreased from 1.8 mm (L) to 0.68 and 0.62 mm for LE and LCE samples, respectively. A further length reduction was observed for LE, LCE and Com samples upon acetylation (from lowest to high Ac2O reaction levels). Specifically, Com sample consisted at first of longer fibers than those in LE and LCE, but at medium and high acetylation conditions (5.35 and 10.7 g Ac2O) the fiber length reduction and fines generation was more severe for the Com sample. A high acetylation degree was achieved in Com by using medium level conditions (5.35 g Ac2O). Moreover, fiber length was reduced by about 77% (from 1.33 to 0.30) and fines increased to 65%. However, the greatest reduction in fiber length (87%) and amount of fines generated (> than 90% of fines) took place when high Ac2O levels were used (10.7 g), indicating the strong degradation of fibers under these conditions. Importantly, the acetyl groups incorporated on the cellulose surface are associated with an increase of mass (coarseness results) and fiber width, and with a reduction in curl. In fact, the strongest effect in these properties was also produced under the “high” conditions of acetylation (coarseness and fiber width increased by 384 and 22% respectively, and fiber curl decreased by 66%). Therefore, the effects on fiber morphology correlate with FTIR results, which indicated an increased acetylation at the “high” conditions. The low values measured for acetyl content in Fig. 4 may be explained by the high fines content measured in the sample.

LE and LCE also suffered a reduction in fiber length with increasing acetylation degree but, to a lesser extent if compared to the Com sample. In particular, at the highest acetylation level (23.6% of acetyl group) of LE, a mass gain (coarseness) of about 163%, a fiber reduction of about 68% and an increase of fines up to 82% were observed compared to the initial value. Meanwhile, similar values for LCE at the highest acetylation level (15.2% of acetyl groups) were measured (178, 68 and 84.5%, respectively) (Table S1 of Supporting Information).

Surface changes in handsheets of acetylated fibers

The change in the surface morphology of the acetylated fibers was evaluated by scanning electron microscopy (SEM). A clear fiber degradation due to acetylation reactions was confirmed by fiber morphology and also by SEM analyses. As can be seen for all samples treated at the medium Ac2O level, fiber length was reduced; in addition, the greater amount of fines produced yielded a more entangled structure, with smaller pore size. The greatest changes were observed for fibers subjected to more severe acetylation conditions. In this case, in fact, whole fibers were not observed at the given SEM magnification and the pattern of the mesh used for web preparation was observed (Fig. 5). In addition, the increase in bulk density observed from non-acetylated to the high acetylation conditions indicated a denser and more compact structure (data not shown).

Fig. 5
figure 5

SEM images of handsheets produced from LE, LCE and Com fibers that were subjected to heterogeneous acetylation at different Ac2O levels, as indicated

Full size image

Mechanical properties of the fibers webs

The effect of fiber morphology and acetylation degree was assessed as far as the mechanical properties of the corresponding handsheets (Fig. 6). A high acetylation level is expected to limit hydrogen bonding capacity since acetyl groups substitute –OH’s otherwise available for bonding in the cellulose network (Ernest-Saunders et al. 2014). Moreover, the strong deterioration of fibers during acetylation (much shorter and with higher fines content), may yield weaker bonding. Additionally, sheet formation (spatial distribution of mass) was limited and large variations in the measured physical properties were noted between samples. Generally, a negative effect in tensile strength was observed upon acetylation (Fig. 6a). Fibers subjected to high Ac2O reaction levels suffered a strength loss of about 70 and 90% for the LCE and Com fibers, respectively. In addition, the observed strength loss for LE and LCE samples correlated with a reduced bulk density. Acetylated fibers from LE and LCE samples produced slightly higher wet strength compared to that on non-acetylated fibers (Fig. 6b).

Fig. 6
figure 6

Dry (a) and wet (b) tensile strength of webs produced with non-acetylated and high acetylated fibers as a function of acetyl group content

Full size image

Wetting properties of acetylated fibers

The effect of acetylation treatment on the hydrophilicity of the fibers was examined by means of initial water contact angle (WCA) of the respective handsheets (Figure S2a of Supporting Information). In general, the water absorption of paper depends on the porous structure of the sheet and the nature of the interactions that occur between fibers and the fluid (Mashkour et al. 2015). Acetyl groups were expected to reduce the hydrophilicity of fibers and lower the interfiber bonding. The different WCA observed between non-acetylated and acetylated samples confirm the effect of chemical surface modification. Samples with the highest degree of acetylation presented twice the WCA value compared to non-acetylated ones. To be precise, a WCA of 64°, 58° and 55° were obtained for acetylated Com, LE and LCE samples, confirming that the acetylation reactions reduced the hydrophilicity of fibers.

Changes in WCA are mainly due to absorption in the sheet structure and to evaporation—the latter, however, is only relevant for relative long absorption times (Cusola et al. 2013). Water drops were absorbed rapidly for non-acetylated samples, giving an equilibrium WCA close to 0° (Figure S2b of Supporting Information). Acetylated LE and LCE samples also showed fast drop absorption, 2.4 and 56 s respectively (Figure S2c of Supporting Information). In contrast, Com acetylated sample indicated no change in water drop during 2 min and a WCA of 45° was recorded after 20 min. Finally, after about 36 min the water drop was fully absorbed.

Homogeneous acetylation

Homogeneous acetylation was conducted in order to evaluate the dissolution behavior of fibers treated chemo-enzymatically (LE and LCE). The results were compared to Com reference fibers. In the absence of toluene in the acetylation medium (homogeneous acetylation), a higher percentage of substituted acetyl groups are determined relative to the results from heterogeneous acetylation. FTIR spectroscopy confirmed that acetylation reactions were substantial, as indicated by the fingerprint peak at 1730 cm−1 (Fig. 7).

Fig. 7
figure 7

FTIR spectra for acetylated cellulose from LE, LCE and Com films

Full size image

Quantification of the degree of substitution by titration showed similar acetylation degrees for all studied fibers. Values between 33 and 36% of acetyl substituted groups were found (Table 2), indicating a high level of acetylation comparable to commercial available cellulose acetate (from Sigma-Aldrich ~ 39%).

Table 2 Acetyl group % determined by the titration method and dry tensile strength index of films produced by solvent casting of LE, LCE and Com samples after homogenous acetylation reaction
Full size table

Fibers obtained after acetylation were freeze-dried, then dissolved in acetone and the resulting viscous solution was used to prepare films via solvent casting. Films made from the chemo-enzymatic samples presented notably greater strength values than those of heterogeneous acetylation reaction at the highest acetylation level (Table 2). However, despite the fact that a similar acetyl content was measured for all samples, Com films presented a tensile strength three times higher than those measured for the samples acetylated after chemo-enzymatic treatment, this can be explained by the differences in fiber morphology of precursor fibers prior to acetylation. In terms of dry zero-span tensile strength, Com and chemoenzymatic acetylated fibers showed values in the same range. As observed with heterogeneous acetylation reactions, the presence of acetyl groups reduced the hydrophilic character, giving a contact angle between 67° and 76°. Although high hydrophobicity was not achieved (contact angle < 90°), water drops remained long time on the surface until complete absorption as WDT assay showed. Overall, cellulose acetate fibers with new functional groups and high strength-related properties were achieved.

Conclusions

Various surface acetylation conditions were studied from a dissolving fiber grade (Com) and from a set of newly introduced fibers obtained by chemo-enzymatic treatment of sulfite fibers (LE, LCE). The respective precursor fibers presented different hemicellulose content, crystallinity and fiber morphology. As a result, upon given heterogeneous reaction conditions, different acetylation degrees were achieved. FTIR and acetyl group content titrations confirmed the fact that much higher acetyl group content was developed for the more severe acetylation conditions. Morphological studies revealed that acetyl groups were introduced via heterogeneous reactions on the surface of the fibers, as indicated by the gain in coarseness that was observed. Generally, the fiber length decreased with the acetylation degree and a larger amount of fines were produced. Notably, the greatest fiber degradation was observed for Com sample under high acetylation conditions giving a 86% fiber length reduction and a gain of about 115% of fines. Handsheets obtained with acetylated fibers exhibited lower dry tensile strength and lower hydrophilicity (determined by contact angle) compared to the non-acetylated grades. Compared to the heterogeneous acetylation, homogeneous reactions led to higher acetyl group degree of substitution. These samples exhibited good solubility in acetone and produced transparent films (via solvent casting) with enhanced dry strength, less hydrophilic character and long time absorption resistance. In conclusion, the synthesis of cellulose esters from the unbleached fibers after the chemo-enzymatic treatment in heterogeneous or homogenous phase (surface or bulk acetylation, respectively) was demonstrated.