Composite films prepared from agricultural by-products

Carbohydrate Polymers 156 (2017) 77–85

Contents lists available at ScienceDirect

Carbohydrate Polymers journal homepage: www.elsevier.com/locate/carbpol

Composite films prepared from agricultural by-products a,∗ ˇ Ivan Simkovic , Ivan Kelnar b , Raniero Mendichi c , Tomáˇs Bertok a , Jaroslav Filip a a b c

Institute of Chemistry, Slovak Academy of Sciences, Dúbravská cesta 9, 845 38 Bratislava, Slovak Republic, Slovak Republic Institute of Macromolecular Chemistry, Academy of Sciences of the Czech Republic, Heyrovsky Square 2, 162 06 Prague, Czech Republic Istituto per lo Studio delle Macromolecole (CNR), Via E. Bassini 15, 20133 Milan, Italy

a r t i c l e

i n f o

Article history: Received 23 May 2016 Received in revised form 14 July 2016 Accepted 5 September 2016 Available online 6 September 2016 Keywords: Sugar beet residue Bagasse Holocellulose Composite film properties

a b s t r a c t In our study we used holocelluloses from sugar beet and bagasse for film preparation. Films from sugar beet holocellulose have better mechanical properties than from bagasse holocellulose. By subsequent carboxymethylation of bagasse holocellulose, films with better properties were produced. Specimens prepared from combined sugar beet and bagasse carboxymethylated holocellulose had the best mechanical properties. The results could be explained by the ratios of cellulose, arabinan, polygalacturonan and xylan content in individual films, based on the elemental analysis data. The use of microwaves to prepare holocellulose film speed up the process, but negatively affected the mechanical properties. Lignin content of the sugar beet holocellulose and bagasse samples was low and did not affect the mechanical properties. Both types of agricultural by-products could be used for preparation of composite film with high strength and stiffness suitable for broad range of applications. © 2016 Elsevier Ltd. All rights reserved.

1. Introduction Many authors consider sugar beet residue (SBR) and bagasse (B) are important agricultural byproducts that could be used for composite and film preparation, with possible applications in civil engineering, packaging, and the food industry (Ghaderi, Mousavi, Yousefi, & Labbafi, 2014; Dufresne, Cavailé, & Vignon, 1997; Heux et al., 1999; Agoda-Tandjawa, Durand, Gaillard, & Doublier, 2012). Also chitosan composite films were studied for film applications (Tripathi, Mehrotra & Dutta, 2009a; 2009b; Dutta, Tripathi, Mehrotra, & Dutta, 2009). Chitosan films are known for their antimicrobial and antifungal activity however, their weak mechanical properties, gas and water permeability limit its uses (van den Broek, Knoop, Kappen, & Boeriu, 2014; Elsabee & Abdou, 2013). For that reason composites with improved mechanical properties must be prepared. Also the combination of chitosan with synthetic polymers is not environmentally safe. It is known that cellulose-containing composites have better mechanical properties than chitosan composites (Khalil et al., 2016). Besides cellulose, non-cellulosic polysaccharides in SBR and B might also support ˇ composite formation (Simkovic, 2012, 2013). The non-cellulosic polysaccharides of SBR mostly form a pectin/arabinogalactan network that is hard to characterize due to polysaccharide branching

∗ Corresponding author. ˇ E-mail address: [email protected] (I. Simkovic). http://dx.doi.org/10.1016/j.carbpol.2016.09.014 0144-8617/© 2016 Elsevier Ltd. All rights reserved.

ˇ ˇ et al., 2009; Simkovic, Uhliariková, Yadav, & Mendichi, (Simkovic 2010). Attempts to combine sugar beet pulp with plastics like polylactic acid (PLA) seem to be economically non-optimal for making composites since the preparation of PLA requires energy to both transform starch to lactic acid and to polymerize it to PLA (Li et al., 2012). New methods are needed for preparation of composites solely from SBR and similar agricultural resources are of high interest without the use of synthetic polymers. Bagasse hemicelluloses are also intensively studied (Bian et al., 2012Bian, Peng, Peng, Xu, Sun, & Kennedy, 2012). They can be isolated from the source by the holocellulose method. This method is well established for delignification studies on wood (Tetard, Passian, Farahi, Davison, Jung, Ragauskas, Lereu, & Thundat, 2011) as well as for bagasse cellulose studies (Yue et al., 2015). Also lignin composites are studied (Doherty, Mousavioun, & Fellows, 2011). The use of chitosan and lignin for composite preparation methods could be useful also for other polysaccharide composites (Rai & Mehrotra, 2016). In the present work we have used the holocellulose procedure and subsequent introduction of carboxymethyl-groups for film composite preparation from SBR and B. The goal was to learn about their properties with the help of mechanical testing, elemental analysis, SEC – MALS, TG/DTG/DTA and AFM techniques. Also microwave procedure was applied to improve the procedure for possible application in the packaging industry.

78

I. Sˇ imkovic et al. / Carbohydrate Polymers 156 (2017) 77–85

2. Experimental 2.1. Materials and methods 2.1.1. Holocellulose preparation SBR (Beta vulgaris, L.; 10 g of mealed powder; C, 39.98; H, 5.82; N, 1.28; 6.85% Klason lignin; 8.00% of proteins) from local sources was mixed with water (320 ml) in 1 l Erlenmeyer flask. After adding 1 ml of concentrated acetic acid and 3 g of NaClO2 (recalculated on pure sodium chlorite from 80%; Sigma-Aldrich, 144155-100G) the mixture was tempered at 85 ◦ C/500 RPM for seven hours. Subsequently the mixture was cooled and stirred at room temperature (RT) overnight and separated into soluble part and insoluble residue by centrifuge (10 000 RPM/30 min). The soluble part was dialyzed (12–14 kDa MWCO; SERVA 44126, 49 mm diameter), preconcentrated with a vacuum evaporator and poured on Petri dishes to dry at RT to constant weight (SBSH, 1.8816 g, 19%; C, 40.38; H, 5.54; N, 1.31; 8.19% of proteins; 0.1% Klason lignin). The insoluble residue was also poured on a Petri dish and dried (6.8256 g; 68%, SBH; C, 38.65; H, 5.42; N, 1.42; 8.88% of proteins; 0.5% Klason lignin). The experiment was repeated several times after which the samples were lyophilized and used for chemical modification or for film preparations by mixing one gram of material with 30 ml of water. After one hour of stirring at 1250 RPM/RT the suspension was poured onto plastic Petri dishes and dried to constant weight in a refrigerator at 5 ◦ C. When the soluble part was not separated and the whole mixture was dialyzed according to an analogous treatment, SBH’ was prepared (6.8521 g; C, 31.53; H, 4.76; 69%; N, 1.15; 7.19% of proteins; 0.6% of Klason lignin). The microwave-assisted experiments on SBR were run on a CEM Discover® S-class instrument (Matthews, NC 28106-0200, USA) with a home-made glass adapter equipped with air cooler and glass rod. To ten grams of mealed sample 10 g of NaClO2 was added and mixed with 101 ml water solution containing 1 ml of acetic acid. The mixture was treated at 50 W constant power, high stirring for one hour with occasional water (cca 30 ml) washing down of the foam and mechanical releasing of gases by glass rod. After the treatment the pH of the mixture was increased from 6.0 to 7.2, and the sample was dialyzed and lyophilized (6.5 g; C, 39.95; H, 5.64; N, 1.37; 8.6% protein; 3.1% lignin). Subsequently the sample was separated into a soluble part and insoluble residue on fritted glass (4–16 ␮m) and treated as above. SBH (C, 40.56; H, 5.61; N, 0.13; 0.8% protein; 0.7% lignin) represented 52% of the starting material and 41% of soluble holocellulose (C, 40.61; H, 5.83; N, 2.34; 14.6% protein; 0.1% of Klason lignin). When the components were not separated, then SBH’(C, 37.81; H, 5.36; N, 0.85; 5.3% protein; 0.4% lignin; 76% yield) was obtained. Analogous holocellulose was also prepared from bagasse (Saccharum officinarum, L.; 5 g; C, 45.21; H, 6.05; 27.8% Klason lignin; Davies Hamakua Company, P.O.Box 250, Paaulino, Hawaii 96776, USA), but the treatment and addition of CH3 COOH/NaClO2 was repeated five times. The resulting mixture was dialyzed and poured on Petri dish to form a film (0.4408; BH’; C, 38.78; H, 5.53) and part of it also separated to soluble part (0.6960 g; C, 40.06; H, 5.35) and insoluble residue (2.0548 g; 41%, BH; C, 39.86; H, 5.66; 1.10% Klason lignin) and poured on Petri dishes. The optimal microwave treatment for bagasse was ten grams on ten grams of NaClO2 at 40 W power/high magnetic stirring and 110 ml of water supply with additional 150 ml added during the run. 2.1.2. Carboxymethylation of holocellulose SBH (2 g) was mixed with water (40 ml) containing NaOH (4 g) and subsequently 8 g of ClCH2 COONa was added and reacted at 60 ◦ C/500 RPM/24 h and dialzyed until the pH dropped to 6.75. Then the mixture was lyophilized or cast on plastic or glass Petri

dishes and dried at RT to constant weight (2.1072 g, CMSBH’; C, 34.88; H, 4.92; 99% total yield). Alternatively when SBH (1.6 g) was mixed with water (32 ml) containing 3.2 g NaOH and subsequently 6.4 g of ClCH2 COONa and carboxymethylated as above. After dialysis the soluble part was filtered off and both the soluble part (CMSBS; 1.1101 g; 69%) and insoluble residue (CMSBH; 0.7846 g, 49%; C, 35.06; H, 5.27) were cast separately onto Petri dishes for film preparation. Under alternative treatment when the soluble part was not filtered off CMSBH’ (2.1075 g; C, 34.88; H, 4.92; 0.1% of Klason lignin) was prepared. BH (2 g) was carboxymethylated using water (40 ml) containing NaOH (4 g) and subsequently 8 g of ClCH2 COONa was added and reacted at 60 ◦ C/500 RPM/24 h and dialyzed until the pH reached 5.54. Subsequently part of the sample was cast onto plastic (CMBH’; 0.8286 g; C, 36.10; H, 5.49; 0.1% of Klason lignin) or glass (1.855 g) Petri dishes to prepare composite films, while the rest of the sample was lyophilized (0.4162 g; C, 38.42; H, 5.42; 99% total yield). Alternatively, CMBH was prepared after dialysis when the suspension was separated on fritted glass and only the insoluble residue was cast on a Petri dish (0.6387 g, 64%; C, 37.84%; H, 5.79). 2.1.3. Extraction of holocellulose SBH (5 g) was extracted with 10% KOH in 100 ml of water at RT/500 RPM/11 h, filtered through fritted glass (4–16 ␮m), preconcentrated with a vacuum evaporator and cast on Petri dish and treated as above (4.52 g of ESBH film; C, 33.89; H, 5.07). Analogously, by extraction of BH, the EBH specimen was also prepared (4.1927 g; C, 40.21; H, 6.02). 2.2. Analytical methods ˇ All the analytical methods were described previously (Simkovic, ˇ Kelnar, Tracz, Kelnar, Uhliariková, & Mendichi, 2014; Simkovic, Uhliariková, Mendichi, Mandalika, & Elder, 2014). Mechanical testing was performed on cut dog-bone specimens (type B, ISO 527-2) with working part length of 10 mm. Tensile tests were carried out at 22 ◦ C and 40% relative humidity (RH) using an Instron 5800 apparatus at a crosshead speed 1 mm/min. At least eight specimens were tested for each sample. The stress-at-break (␴b ), strain-at-break (␧b ) and Young’s modulus (E) were evaluated. The film thickness was measured by the micrometer with 0.001 mm accuracy. AFM images were performed using a BioScope Catalyst (Bruker, Santa Barbara, USA) with the PeakForce quantitative nanomechanical mapping technique in air and evaluated by NanoScope Analysis 1.40 software (Bruker). The conditions/parameters for these images were: silicon nitrate cantilever tip curvature radius, R = 12 nm (Max), 2 nm (Nom); 0.3 sample Poisson’s ratio; 15◦ tip front angle, 25◦ back angle and 17.5◦ side angle; 0.4 N/m nominal cantilever spring constant [0.2 N/m (Min.), 0.8 N/m (Max.)]; cantilever resonant frequency 70 kHz (Nom.) and 0.250401 Hz scan rate. The lignin content of samples was determined by the modified Klason method (Bunzel, Schüßler, & Tchetseubu Saha, 2011). Briefly, to one gram of sample 15 ml of 72% sulfuric acid cooled to 5 ◦ C was added and the mixture stirred at RT for two hours. Subsequently the mixture was diluted with 560 ml of water and refluxed for four hours at 121 ◦ C and the insoluble part filtered on fritted glass (4–16 ␮m). The amount of collected sample on fritted glass was vacuum-dried in oven at 105 ◦ C for two hour, cooled under vacuum to RT, balanced and determined as lignin content in% from the amount used for determination. 3. Results and discussion 3.1. Film preparations and mechanical properties Mechanical properties of all prepared holocellulose-based films are summarized in Table 1. It is obvious that very strong (ten-

I. Sˇ imkovic et al. / Carbohydrate Polymers 156 (2017) 77–85 Table 1 Mechanical properties of prepared films. Sample

E (MPa)

SBHa SBH’b MSBHc MSBH’d CMSBHe CMSBH’f BHg BH’h MBHi MBH’j CMBHk CMBH’l EBHm ESBHn SBH/BHo MSBH’/MBH’p MSBH/MBHq CMSBH’/CMBH’r CMSBH/CMBHs

5030 ± 470 3900 ± 210 4500 ± 450 3250 ± 400 4800 ± 400 5960 ± 670 3510 ± 500 5750 ± 250 3700 ± 600 3550 ± 400 7400 ± 600 6250 ± 700 900 ± 250 4500 ± 500 1270 ± 220 4450 ± 350 4000 ± 750 8000 ± 800 7500 ± 450

Thickness (␮m) 80 160 180 180 150 150 300 90 180 175 100 110 460 160 280 160 190 90 50

␴b (MPa)

␧b (%)

55.0 ± 6.8 42.5 ± 12.0 58.0 ± 8.0 31.7 ± 10.0 80.0 ± 20.0 35.0 ± 10.0 50.2 ± 6.0 42.3 ± 4.6 36 ± 0.4 43.6 ± 9.3 120 ± 10.0 86 ± 27.0 6.05 ± 2.3 64.6 ± 11.0 7.5 ± 2.5 56 ± 11 32.2 ± 12 44 ± 36 83 ± 9.4

1.5 ± 0.23 1.5 ± 0.60 1.7 ± 0.20 1.3 ± 0.47 2.6 ± 0.10 0.85 ± 0.40 2.00 ± 0.50 0.90 ± 0.10 1.3 ± 0.3 1.5 ± 0.3 2.40 ± 1.1 1.44 ± 0.3 1.65 ± 0.67 2.80 ± 0.9 0.8 ± 0.2 1.5 ± 0.5 1.1 ± 0.3 0.60 ± 0.4 2.15 ± 0.5

a

Sugar beet holocellulose. SBH containing also water-soluble fraction. Microwave-assisted preparation of SBH. d Microwave-assisted preparation of SBH’. e CM-holocellulose from sugar beet after separation of soluble part. f CM holocellulose from sugar beet containing both soluble part and insoluble residue. g Bagasse holocellulose fraction insoluble in water. h Bagasse holocellulose containing also soluble part. i Microwave-assisted preparation of BH. j Microwave-assisted preparation of BH’. k CM-holocellulose from bagasse after elution of soluble part. l CM holocellulose from bagasse containing both soluble part and insoluble residue. m Extracted bagasse holocellulose. n Extracted sugar beet holocellulose. o SBH/BH. p MSBH’/MBH = 1/1 mixture. q MSBH/MBH = 1/1. r CMSBH’/CMBH’ = 1/1. s CMSBH/CMBH = 1/1. b c

140

k

Tensile stress (MPa)

120

l

100

r

80 60

f s

40

e a

b

g

h

20 0 -20 0

1

2

Tensile strain (%)

3

Fig. 1. Stress-strain curves of sugar beet and bagasse films. The abbreviations are the same as in Table 1.

sile stress >100 MPa) and rigid (E modulus >7000 MPa) materials can be obtained. The selected representative stress-strain curves in Fig. 1 indicate that all films have very low ductility with absence of yielding (i.e. of plastic deformation). The original holocellulose procedure does not use the soluble part of the acetic acid/sodium chlorate mixture (Wise, Murphy,

79

& D’Addieco, 1947). In case of SBR after filtration of the watersoluble fraction it was dialyzed and lyophilized (SBSH, 23% yield). According to the elemental analysis (C, 40.65; H, 5.75, N, 2.29; 0.1% lignin) this fraction consists of 14.31% proteins (Selvedran & O’Neill, 1987). Unfortunately, this sample did not form a film, splitting into small particles unsuitable for mechanical testing. As we know from previous studies where the chemical structure was proved by NMR spectroscopy. This fraction consists of branched arabinan molecules that cannot interact well with each other due to insufficient hydrogen bond formation between the branched non-linear ˇ ˇ et al., 2009; Simkovic et al., 2010). Additional facchains (Simkovic tor is the molar mass of SBSH. According to SEC-MALS analysis the molecular mass was 5.9 kg/mol (Mw /Mn = 5.71 at 71.4% mass recovery with 0.1 M sodium carbonate buffer at pH 10.0). This result is supported by the observation that decreasing the branching of the water-soluble arabinoxylan the tensile strength is decreasing (Höije, Sternemalm, Heikkinen, Tankanen, & Gatenholm, 2008). Elemental analysis of SBH film prepared from water-insoluble fraction (C, 39.20; H, 5.55; N, 1.28) suggests that 8% of this fraction is protein. The lignin content was only 0.5%. The modulus of the composite films prepared from SBH was 5030 MPa, with tensile strength at 55 MPa and 1.5% elongation at break (Table 1). When the film was prepared from a mixture of soluble and insoluble part obtained by the alternative holocellulose procedure (SBH’), the modulus decreased to 3900 MPa with a slight decrease of tensile strength (to 42.5 MPa) and identical elongation at break. The decrease of modulus and tensile strength could be explained by the presence of the soluble component in SBH’, which could be adsorbed on the surface of the insoluble cellulose component. In this way the cellulose blocks could not interact with each-other through hydrogen bonds, which might result in lower modulus and tensile strength. Similar plasticization effect was observed when xylitol or sorbitol were added to xylan solutions, which resulted in films with less ¨ favorable mechanical properties (Grondahl, Eriksson, & Gatenholm, 2004). The similar effect is also possible in our case, when the cellulose fiber component of the composite film is insoluble. Similar adsorption of arabinoxylan on cellulose surfaces was studied before ¨ (Kohnke, Östlund, & Brelid, 2011). Additional factor is probably the branching chain of the arabinan, which could not effectively form hydrogen bonds with the cellulose. When the holocellulose was prepared by microwave assistance, the MSBH and MSBH’ fractions were prepared (Table 1). Under optimal conditions the experiment could be run within one hour at 50W power, but the mechanical properties of MSBH and MSBH’ films were worse than the previous film made with regular heating. The lignin content of MSBH’ (C, 37.63; H, 5.43; N, 0.95; 5.9% protein content) was low (1.3%) and did not affect the mechanical properties. The soluble holocellulose fraction prepared by microwave assistance (MSBSH) was also not suitable for film preparation, probably also for the same reasons as SBSH, due to branched structure ˇ and small molar mass (Simkovic et al., 2010). The lignin content of MSBSH was equal to zero. Generally the Klason lignin values for all studied samples was so low that the effect of this component could not be evaluated. We think that due to this fact the presence of lignin did not affect the properties of the samples. According to the known data of mechanical properties of lignin films, they depend upon content of some polymeric matrix and molar mass of lignin at 85% lignin concentration (Li & Mlynár Sarkanen, 1997). That is why under conditions when the lignin content was lower than 5% it could not play a significant role. Carboxymethylation of SBH improved its mechanical properties. When only the insoluble part of the product was tested (CMSBH), then the measured modulus (4800 MPa) was smaller, while the tensile strength (45 MPa) and elongation at the break were bigger in comparison to the sample consisting of both soluble and insoluble parts (CMSBH’; Table 1). This confirms the concept of

80

I. Sˇ imkovic et al. / Carbohydrate Polymers 156 (2017) 77–85

interaction of soluble polysaccharides with insoluble cellulose during the drying process of the film composites. The CMSBH’ specimen had the highest modulus value of all sugar beet samples at the same 150 ␮m film thickness. Based on its zero nitrogen content this fraction does not contain protein. Average formula of CMSBH’ was [(C6 H10 O5 )1.0 (C5 H8 O4 )0.3 (C5 H7 O4 )0.5 (C6 H7 O6 Na)0.3 (C2 H2 ONa)2.0 (H2 O)2 ] = 524.5 (calculated: C, 34.78; H, 4.96; found: C, 34.88; H, 4.92) and for CMSBH it was [(C6 H10 O5 )0.5 (C5 H8 O4 )0.2 (C5 H7 O4 )0.1 (C2 H2 O2 Na)1.0 (H2 O)1 ] = 219.5 (calculated: C, 35.54; H,5.15; found: C, 35.06; H, 5.27). It means the different results have been attributed also to other factors. Under these conditions when the elemental composition did not indicate the presence of nitrogen and also lignin was not present in the sample, the content of individual polysaccharide components could be calculated. According to our knowledge there is no other method which could provide better or only the average chemical formulas. On polysaccharides all results are just average due to the polydispersity of the macromolecule. The CMSBH’ sample contains 23% of cellulose and 36% of arabinan/galacturonate pectin with 31% of carboxymethyl groups, while CMSBH contains 37% of cellulose and 18% of arabinan without polygalacturonan component, but with 37% of carboxymethyl groups. It indicates that the CMSBH’ sample with its smaller amount of cellulose but bigger amount of pectin and with similar amount of carboxymethyl groups has better mechanical properties than CMSBH. It is probably because the high content of carboxymethyl groups causes better solubility and mixing of the polysaccharides. All our mechanical property values are better than those for films prepared from sugar beet cellulose microfibrils (Dufresne et al., 1997). When the analogous films were prepared from bagasse, slightly different results were obtained. The water-soluble holocellulose fraction from bagasse (BHS) represented 12% of the material after dialysis (see Experimental section). Unfortunately this fraction could also not be used for film preparation, because during the drying process it split into small particles that could not be tested. The molar mass determined by SEC-MALS (Mn = 3.7 kg/mol; Mw /Mn = 2.6; 83% mass recovery from 0.1 M sodium carbonate buffer at pH 10.0) was lower than the molar mass determined for the analogous sugar beet fraction. The bagasse holocellulose fraction (BH) that was insoluble in water and represents 50% of the material consists according to elemental analysis of 20% cellulose and 53% of xylan with average summary formula: [(C6 H10 O5 )0.3 (C5 H8 O4 )0.3 (C5 H7 O4 )0.1 (C5 H6 O4 )0.2 C7 H10 O6 Na)0.5 (H2 O)1.0 ] = 240.8; (calculated: C, 41.36; H, 6.15; found: C, 41.99; H, 5.92). Sugarcane bagasse does not contain polygalacturonic acid (Li et al., 2014). A film could be prepared from this fraction with slightly smaller modulus (3500 MPa), tensile strength (50.2 MPa) and slightly better elongation (2.0%) than from SBH (Table 1). Elemental analysis indicates that the BH film consists of 53% of branched xylan with high uronic acid content and 20% cellulose content. It does not have mechanical properties as good as the SBH film formed from 23% of PGA, 38% of cellulose and 30% of arabinan. Thus, worse mechanical properties result from a decreased amount of cellulose and the presence of branched xylan which is not able to interact with cellulose due to branching. When the BH’ film containing also the water-soluble fraction was prepared, slightly improved mechanical properties were produced in comparison to BH (Table 1). According to the elemental analysis, the film summary formula is [(C6 H10 O5 )0.3 (C5 H8 O4 )0.2 (C5 H7 O4 )0.2 (C5 H6 O4 )0.3 (C7 H10 O6 Na)0.6 (H2 O)2.0 ] = 329.8; (calculated: C, 38.93; H, 5.94; found: C, 38.78; H, 5.53). This composite film contains 15% of cellulose and 74% of arabino-(4-O-methyl-dglucurono)-d-xylan. It means a lower cellulose content but a higher amount xylan with bigger content of uronic acids improves the mechanical properties in comparison to BH. We think that the improvement is caused by the fact that xylan is interacting with

cellulose by sorption on the fiber surface than arabinan in the case of SBH’, which results in better mechanical properties. When BH was prepared by microwave assistance under optimal conditions at 40 W power and a two hour treatment under stirring, the mechanical properties were again less good (Table 1). While the modulus values were bigger for MSBH than for MBH, the tensile strength and elongation at break were bigger for MBH. The values for MSBH’ and MBH were very close. The lignin content was 3%. After elution of the water-soluble part, carboxymethylated BH (CMBH) film had the summary formula: [(C6 H10 O5 )0.3 (C6 H9 O5 )0.1 (C2 H2 O)0.4 Na0.4 (C5 H8 O4 )0.2 (C5 H7 O4 )0.21 (C7 H10 O6 Na)0.01 (CH3 CO)0.1 (H2 O)1.0 ] = 175.44; (calculated: C, 37.76; H, 6.31; found: C, 37.84; H, 5.79). Alternatively CMBH’ film with formula: [(C6 H10 O5 )0.5 (C6 H9 O5 )0.04 (C2 H2 O)0.5 Na0.5 (C5 H8 O4 )0.3 (C5 H7 O4 )0.46 (CH3 CO)0.1 (H2 O)2.0 ] = 268.1; (calculated: C, 36.57; H, 6.06; found: C, 36.10; H, 5.49) was produced when the watersoluble part was not separated. According to these results the CMBH’ film contains 33% of cellulose and 38% of xylan, while CMBH contained 37% cellulose and 34% of xylan. The CMBH film with a thickness 100 ␮m gave E = 7400 MPa, ␴b = 120 MPa and ␧b = 2.40%. It had slightly better mechanical properties than CMBH’ film with 110 ␮m thickness: E = 6250 MPa, ␴b = 86 MPa and ␧b = 1.44%. In this case the mechanical properties were better for CMBH than for CMBH’ due to the larger amount of cellulose in CMBH. Additionally polygalacturonic/arabinan pectin was replaced with arabino-(4O-methyl-d-glucurono-d-xylan). The modulus obtained for the CMBH composite (7400 MPa) is the highest in comparison to previous samples. When BH was extracted with 10% KOH and the film prepared from the insoluble residue, the modulus and tensile strength decreased while the elongation value slightly increased (Table 1). This drop of the mechanical properties was more dramatical than in SBH. According to the elemental analysis EBH consisted of 71% cellulose and 21% xylan [(C6 H10 O5 )1.0 (C5 H8 O4 )0.2 (C7 H10 O6 Na1 )0.1 (H2 O)1.0 ] = 227.7; calculated: C, 40.58; H, 6.41; found: C, 40.21; H, 6.02]. The mechanical properties also dropped in comparison to CMBH sample. It is probably because the xylan portion could not interact with amorphous cellulose. When SBH was extracted with KOH the mechanical properties also slightly decreased in comparison to SBH, but were much better in comparison to both EBH and slightly better than on SBH’ (Table 1). From the elemental composition of ESBH it follows that this film consists of 38% of arabinan, 26% of polygalacturonan and 17% of cellulose [(C6 H10 O5 )0.2 (C5 H6 O4 )0.55 (C6 H7 O6 Na)0.25 (H2 O)2 ] = 173.75; calculated: C, 34.53; H, 5.83; found: C, 33.89; H, 5.07]. While ESBH due to alkaline treatment consisted from smaller amount of cellulose but bigger amount of pectin polygalacturonan/arabinan mixture than EBH where more cellulose was present and surrounded by (4-O-methyl-d-glucurono)-d-xylan without the presence of polygalacturonic acid/arabinan pectin. This resulted in the much better mechanical properties for ESBH than for EBH. The explanation might be the fact that the ESBH cellulose matrix at the excess of pectin is better reinforced than in EBH where more cellulose is present than is the amount of xylan. Mixing SBH and BH in 1:1 ratio resulted in decrease of mechanical properties of the SBH/BH specimen in comparison to SBH, SBH’ and BH (Table 1). It is because both components were not soluble in water and they were holding together only by physical interaction of the fibers after drying from the suspension. It indicates that the fibers of individual samples could not associate together better than with their own cellulose fibers. Mechanical properties of the MSBH’/MBH’ specimen were improved in comparison to SBH/BH film. This might be caused by the presence of soluble components of both species that could be homogeneously distributed in the film and act as reinforcement constituent. When only insoluble holocellulose fractions were mixed, then the measured mechanical

I. Sˇ imkovic et al. / Carbohydrate Polymers 156 (2017) 77–85

properties were worse. The mixing of CMSBH’/CMBH’ = 1/1 resulted in the specimen with the best modulus (8000 MPa) of all samples, while ␴b and ␧b were bigger for the CMSBH/CMBH film. This is probably due to introduction of CM-group, which increased the solubility of all polysaccharides and allowed their mixing of the reinforcing components around the cellulose matrix.

3.2. TG/DTG/DTA analysis According to TG/DTG/DTA results in air environment (Table 2) the onset temperature (OT) of sugar beet pup (SBP) is at 176 ◦ C with first DTG peak at 243 ◦ C which relates to dehydration of arabinan/pectin fraction. The second DTG maximum is at 327 ◦ C due to cellulose dehydration. The overall thermo-oxidation is accompanied with an exotherm with DTA peak at 343 ◦ C. The exothermic degradation of the formed gases is expressed with DTG maximum at 406 ◦ C and an exotherm at 429 ◦ C. The final carbonization of the residue results in the last DTG maximum at 516 ◦ C accompanied with DTA peak at 586 ◦ C, with 25% residue at 500 ◦ C and 4% at 1000 ◦ C. SBS holocellulose fraction has its OT at 143 ◦ C with sharp DTA exotherm at 182 ◦ C (Table 2). This exotherm is accompanied by a DTG maximum at 168 ◦ C with a shoulder at 170 ◦ C due to pectin and arabinan dehydration. Subsequently two small exotherms at 307 and 398 ◦ C follow, accompanied with two small DTG peaks at the same temperatures due to cellulose degradation. The thermooxidation is accompanied with a broad exotherm with a DTA maximum at 485 ◦ C and the DTG maximum at 476 ◦ C. The residue at 500 ◦ C is 32%, while at 1000 ◦ C the residue decreases only to 22%. The dramatically higher residue at 1000 ◦ C is due to the formation of carbonates from ions incorporated into pectin during the preparation process. The onset temperature of SBH is at 185 ◦ C. The thermo-oxidation process of this film consists of four steps starting with first exotherm with peak at 200 ◦ C related to first DTG peak at 246 ◦ C probably related to arabinan degradation. The second exotherm with peak at 336 ◦ C is related to degradation of cellulose component with DTG peak at 325 ◦ C with maximal rate of degradation at 1.77 mg/min. The remaining two steps of degradation are characterized with two exotherms at 518 and 596 ◦ C related to two DTG peaks at 504 and 596 ◦ C. These responses are related to the gasification of the residue, which is 6% at 1000 ◦ C. The SBH’ has the lowest onset temperature from all the studied samples at 137 ◦ C. The first DTG is already at 166 ◦ C followed by DTA at 180 ◦ C with 87% residue at 200 ◦ C. At this stage the dehydration reactions on pectin arabinan/polygalacturonan part took place. The biggest DTG peak is at 232 ◦ C due to pectin degradation. Then the cellulose component degrades with DTG maximum at 317 ◦ C and DTA at 327 ◦ C. Subsequently, combustion of the residue took place resulting in a DTG peak at 493 ◦ C accompanied with DTA peak at 514 ◦ C. The residue at 1000 ◦ C is 13%, which indicates that all the PGA carboxyls were in sodium cycle. By carboxymethylation the OT (182 ◦ C) of the CMSBH film slightly decreased in comparison to SBH (185 ◦ C). The first peak on DTA curve is an endotherm at 203 ◦ C probably due to decarboxylation of introduced carboxyl group because it was not observed on SBH. This process is related the first DTG peak at 200 ◦ C with residue at 200 ◦ C at 82%. The next step of degradation is the DTG peak at 247 ◦ C related to DTA peak at 255 ◦ C, which is due to splitting of arabinan glycosidic bonds. The next step is at DTG peak (289 ◦ C) due to splitting of cellulose glycosidic bonds related to the DTA exotherm at 301 ◦ C. Then combustion and subsequent gasification take place expressed by exotherms at 385, 419, 689, 751, 767 and 778 ◦ C, related to DTG maxima at 385, 440, 683, 750, 765 and 778 ◦ C. The residue at 1000 ◦ C is the biggest from all studied samples due

81

to the contribution of the carboxymethyl group introduced by the chemical modification in sodium form. The carboxymethylated holocellulose film formed from a mixture of the soluble part and insoluble residue (CMSBH’) has onset temperature at 163 ◦ C. Its first DTG is already at 189 ◦ C, which is probably due to decarboxylation of introduced carboxymethyl groups. We conclude this hypothesis from the fact that on SBH’ the first DTG took place at lower temperature and the residue at 200 ◦ C was slightly bigger than for SBH’. The decarboxylation is related to complete pectin degradation because the shape is different than for SBH’ where there was an additional DTG peak at 232 ◦ C. Also DTG (247 ◦ C) and DTA (255 ◦ C) peaks observed on CMSBH were not observed on CMSBH’. It might be due to changes taking place on arabinan during the carboxymethylation process. The next DTG peak is at 298 ◦ C, related to the DTA peak at 311 ◦ C, and is due to cellulose degradation. It takes palace at higher temperature than on CMSBH’, but at lower temperature than on SBH and SBH’. Subsequently five DTA peaks at 364, 428, 585, 608 and 646 ◦ C related to four DTG peaks at 367, 428, 586 and 616 ◦ C are taking place. The residues at 400, 500 and 1000 ◦ C are slightly smaller than on CMSBH’, which might be due to the fact that the formed fraction smaller than 12 kDa was dialyzed. The onset temperature of the film formed after KOH extraction (ESBH) is lower (177 ◦ C) than on SBH film. This indicates that some of the non-cellulose polysaccharides were extracted. The rest of the pectin degrades at the maximal rate at 235 ◦ C accompanied by DTA maximum at the same temperature. The maximal rate of cellulose degradation for ESBH is 1.74 mg/min at 291 ◦ C accompanied with DTA peak at 310 ◦ C. The subsequent DTA peaks at 367 and 709 ◦ C and DTG peak at 662 ◦ C, are related to subsequent degradation reactions taking place at increasing temperature. The residues are bigger than for SBH due to higher cellulose content, with 15% residue at 1000 ◦ C. The bagasse starting material (B; Table 2) has OT at 158 ◦ C with the first DTG peak at 288 ◦ C. This is due to xylan dehydration. The second DTG maximum at 316 ◦ C is due to cellulose dehydration and parallel thermo-oxidation which results in overall DTA peak at 379 ◦ C. Subsequent carbonization of the residue results in last DTG maximum at 478 ◦ C and an accompanied DTA maximum at 495 ◦ C. The residue at 1000 ◦ C is similar as observed for SBF. The bagasse holocellulose (BH) film has OT (155 ◦ C) at lower temperature than SBH, but higher than for SBH’. Its xylan dehydration is accompanied with two DTA exotherms at 166 and 225 ◦ C followed by DTG peak at 300 ◦ C due to cellulose degradation accompanied with DTA peak at 318 ◦ C. The residue at 300 ◦ C is 61%, which is five degrees smaller than observed on SBH. The further degradation takes place in two DTA (403 and 491 ◦ C) and two DTG (383 and 473 ◦ C) peaks with residues at 500 ◦ C (7%) and 1000 ◦ C (4%) representing just inorganic salts. The EBH film has the identical OT as BH film with residue at 200 ◦ C at 93%. It is due to slightly higher cellulose content than for BH. The DTG peak related to cellulose degradation is at 307 ◦ C. Also the residue at 300 ◦ C is higher than observed on BH. The subsequent combustion of the formed residue is accompanied with exotherms at 373 and 491 ◦ C with DTG peaks at 374 and 491 ◦ C. The residue at 1000 ◦ C is 2%. The CMBH film was having the lowest OT from all studied sample (106 ◦ C). Similarly like for CMSBH there is an endotherm but at 223 ◦ C, which is twenty degrees higher in comparison to CMSBH. Then there is a DTG peak at 309 ◦ C due to cellulose degradation followed by exotherm at 325 ◦ C. Further the thermo-oxidation is characterized by two exotherms at 547 and 600 ◦ C accompanied with DTG peaks at 544 and 599 ◦ C. The residue at 1000 ◦ C is 5%. The CMBH’ film has higher OT than CMBH at 136 ◦ C with 88% residue at 200 ◦ C. It indicates the thermal degradation of xylan component with DTA and DTG peaks at 253 ◦ C. The subsequent cellulose

I. Sˇ imkovic et al. / Carbohydrate Polymers 156 (2017) 77–85

82

Table 2 TG/DTG/DTA data of studied sugar beet and bagasse composite films. Sample

OTa ◦

[ C] SBPb

SBS

SBH

SBH’

CMSBH

CMSBH’

ESBH

B

DTA peaks

EBH

CMBH

CMBH’

◦

 [K/mg]

[ C] [mg/min]

200

300

400

500

1000

1.30 1.28 0.35 0.48

90

62

37

25

4

343 0.1122 429 0.1200 586 0.2151

243 327 406 516

5.23 5.13 0.09 0.19 0.44

49

47

44

32

22

307 0.0478 398 0.0875 485 0.1931

168 170 307 398 476

185

200 0.0181 336 0.0555 518 0.1732 596 0.1738

246 325 504 596

0.94 1.77 0.66 0.35

91

66

39

26

6

137

180 0.0318

0.56 1.10 0.88 0.52

87

61

47

34

13

327 0.0275 514 0.1660

166 232 317 493

203 −0.0026 255 0.0413 301 0.0622 385 0.0585 419 0.0662 689 0.1548 751 0.0926 767 0.0673 778 0.0768

200 247 289 385 440 683 750 765 778

0.29 0.93 1.18 0.27 0.17 0.79 0.47 0.27 0.23

82

56

46

41

18

189 298 367 428 586 616

0.23 1.97 0.23 0.22 0.50 0.50

89

60

43

37

14

235 291

1.06 1.74

92

54

41

37

15

93

68

30

12

3

92

61

25

7

4

93

64

21

3

2

90

66

36

28

5

88

59

38

33

10

176

143

182

182 0.1824

163 311 0.0489 364 0.0312 428 0.0260 585 0.1330 608 0.1380 646 0.1451 177

235 0.0137 310 0.0486 367 0.0520 709 0.1298

158 379 0.2010 495 0.2011

BH

Residue [%] at [◦ C]

DTG peaks

155

136

0.51

288 316

1.55 2.52

478

0.59

166 0.0076 225 0.0302 318 0.1485 403 0.2168 491 0.1757

300 383 473

2.56 0.76 0.51

373 0.1785 491 0.1616

307 374 491

3.06 0.68 0.40

223 −0.0135 325 0.0352 547 0.1614 600 0.1614

309 544 599

2.58 0.54 0.47

253 303 429 575 615 675 716 754

1.02 1.98 0.19 0.21 0.34 0.40 0.39 0.27

155

106

662

253 0.0248 317 0.0568 436 0.0400 585 0.0935 639 0.1135 645 0.1138 657 0.1139 669 0.1111 676 0.1082 726 0.0847 755 0.0673

I. Sˇ imkovic et al. / Carbohydrate Polymers 156 (2017) 77–85

83

Table 2 (Continued) Sample

OTa ◦

SBH/BH

CMSBH/CMBH

DTA peaks

a b

◦

[ C]

 [K/mg]

[ C] [mg/min]

152

218 −0.0076 320 0.1632 382 0.1684 472 0.1799

289 369

2.34 0.61

451

0.63

309

2.49

507 593

0.69 0.43

208 254

0.29 0.92

616

0.44

194 326 0.0600 508 0.1987 593 0.2154

CMSBH’/CMBH’

Residue [%] at [◦ C]

DTG peaks

181

201 −0.0086 254 0.0250 327 0.0548 616 0.1829

200

300

400

500

93

55

28

10

1000 6

91

68

40

29

9

90

64

39

33

7

Onset temperature. Sugar beet pulp; the remaining sample abbreviations are according to Table 1.

Fig. 2. AFM image and section analysis of SBH (a), BH (b) SBH/BH (c), CMSBH/CMBH (d) and CMSBH’/CMBH’ (e) films.

component is having DTG maximum at 303 ◦ C accompanied with DTA peak at 317 ◦ C. Subsequently the most complicated combustion process from all studied samples is characterized with nine exotherms and six DTG peak with 10% residue at 1000 ◦ C. This is probably due to lignin condensation and subsequent gasification and sodium inorganics formation in the residue. The SBH/BH film from combined holocelluloses had smaller OT than both components (152 ◦ C). There was an endotherm observed at 218 ◦ C which might be due to some intermolecular dehydration of non-cellulosic components, which was not observed on either component. The thermolysis of these components takes place with a maximal rate at 289 ◦ C, while the celluloses from both sources are accompanied with DTA (320 ◦ C) and DTG (369 ◦ C) peaks. At higher temperatures the process is characterized with two DTA peaks at 382 and 472 ◦ C and one DTG peak at 451 ◦ C. The residue at 1000 ◦ C is 6%. The film prepared from carboxymethylated holocelluloses, CMSBH/CMBH at 1/1 ratio was the most thermally stable composite with OT at 194 ◦ C. It is probably because part of the cellulose was also carboxymethylated, which contributed to the matrix formation interacting with non-cellulose polysaccharides. The first DTG

maximum is at 309 ◦ C, which is due to cellulose degradation, the predominant component of the mixture. The first DTA exotherm is at 326 ◦ C due to thermo-oxidation of the residue formed. Subsequently the formed gaseous products are combusting which results in two DTA and DTG peaks at 508 and 593, respectively 507 and 593 ◦ C. The residues at 200, 300, 400, 500 and 1000 ◦ C were bigger than those for SBH/BH films due to higher cellulose content. It is related due to sodium ions linked to carboxymethyl groups and carboxyls of (4-O-methyl-d-glucurono)-d-xylan. The composite prepared from CMSBH’ and CMBH’ at 1/1 ratio has the second highest OT of all bagasse composites (181 ◦ C) and only CMSBH has slightly higher OT at 182 ◦ C. Similarly like CMSBH it has a small endotherm at 201 ◦ C that is probably related to decarboxylation and subsequent dehydration of SB pectin/arabinan (208 ◦ C) and B xylan at 254 ◦ C, with overlapping DTA and DTG at 254 ◦ C. The cellulose component then degrades with a peak at 327 ◦ C with subsequent carbonization with maximum at 616 ◦ C. The residue at 1000 ◦ C is 7%, due to present salts. The differences between the courses of thermal degradation of CBSBH/CMBH and CMSBH’/CMBH’ are due presence of arabinan/pectin fraction in the later sample which lowers the thermal stability of the film.

84

I. Sˇ imkovic et al. / Carbohydrate Polymers 156 (2017) 77–85

3.3. AFM analysis According to AFM analysis of the height 10 × 10 ␮m image of top surfaces of the SBH holocellulose (Rmax = 220 nm) and section analysis (Fig. 2a) and analogous image of BH (Rmax = 239 nm; Fig. 2b), both composites are having similar overall shape of the surface with holes up to 100 nm deep. But the SBH section analysis curve is slightly smoother than the BH curve shape. The small dots on bagasse film surface height image might be interpreted as nanocrystals of cellulose. The specimen made from the SBH/BH = 1/1 mixture with dramatically better mechanical properties than both components of the film (Table 1), has a much smoother surface with Rmax = 12.2 nm (Fig. 2c). The film prepared by mixing CMSBH/CMBH at 1/1 ratio with the highest modulus has Rmax = 829 nm and according to the section analysis the holes are as deep as 200 nm (Fig. 2d). The image of the analogous sample prepared from the sample containing mixture of soluble and insoluble carboxymethylated sugar beet and bagasse holocelluloses (CMSBH’/CMBH’) with similar mechanical properties has Rmax = 762 nm and similar shape of section analysis profile (Fig. 2e). According to these results it could be concluded that the mechanical properties of the films are not related to the surface shape of the specimens. It is evident that by the carboxymethylation of the holocelluloses part of the cellulose component was solubilized and also the crystalline structure was disordered which resulted in separation of bigger crystalline cellulose blocks. Also the different non-cellulosic polysaccharide structures from sugar beet and bagasse were solubilized and removed from the supramolecular structure present in the original plant cell walls. All these contributed to the surface shape of the films. The resulting synergistic effect of improved mechanical properties produced by mixing of two different plant species could be explained by the role of the bagasse xylan component that is able form strong interactions with structurally similar linear cellulose chains from both types of plant species cell wall fragments. 4. Conclusions The modulus values of films made from sugar beet pulp and bagasse residues were between 8000 and 900 MPa. The CMSBH’/CMBH’ and CMSBH/CMBH samples with best modulus values properties, also exhibited a synergism in comparison to the values measured on individual components. They also gave the highest onset temperature values at 182 and 194 ◦ C. On the other side, the tensile strength values were bigger on CMBH and CMBH’ (120 and 86 MPa), than on CMSBH and CMSBH’ films (45 and 35 MPa). The strain at the break values of the obtained films were between 0.6 and 2.80%. According to AFM the surface of both SBH and BH films contained holes as deep as 100 nm and the BH surface was covered with nanocrystal-like structures. It was also concluded that the film surface textures are not related to their mechanical properties, which mostly depend upon interaction of cellulose linear structures with structurally related xylan chains. Acknowledgements State programs # 2003SP200280203 and The 2003SP200280301; Slovak Granting Agency VEGA (Project No 2/7030/7, 2/0087/11, 2/0007/13 and 2/0100/14) for the support. We acknowledge Dr. Maria Mastihubová for providing the microvawe instument (“Centre of Excellence on Green Chemistry Methods and Processes (CEGreenI)”, Contract No. 26240120001, supported by the Research & Development Operational Programme ˇ funded by the ERDF); and Eva Spyrková-Hadzimová for running the TG/DTG/DTA experiments. This contribution is also the result of

the project implementation: Centre of excellence for white-green biotechnology, ITMS 26220120054, supported by the Research & Development Operational Program funded by the ERDF and also the result of the project implementation: Applied research in the field of industrial biocatalysis, ITMS code: 26240220079 supported by the Research & Development Operational Programme funded by the ERDF.

References Agoda-Tandjawa, G., Durand, S., Gaillard, C., Garnier, C., & Doublier, J. L. (2012). Properties of cellulose/pectin composites: Implication for structural and mechanical properties of cell wall. Carbohydrate Polymers, 90, 1081–1092. Bian, J., Peng, F., Peng, X.-P., Xu, F., Sun, R.-C., & Kennedy, J. F. (2012). Isolation of hemicelluloses from sugarcane bagasse at different temperatures: Structure and properties. Carbohydrate Polymers, 88, 638–645. Bunzel, M., Schüßler, A., & Tchetseubu Saha, G. (2011). Chemical characteriyation of Klason lignin preparations from plant-based foods. Journal of Agricultural and Food Chemistry, 59, 12506–12513. Doherty, W. O. S., Mousavioun, P., & Fellows, C. M. (2011). Value-adding to cellulosic ethanol: Lignin polymers. Industrial Crops and Products, 33, 259–276. Dufresne, A., Cavailé, J.-Y., & Vignon, M. R. (1997). Mechanical behavior of sheets prepared from sugar beet cellulose microfibrils. Journal of Applied Polymer Science, 64, 1185–1194. Dutta, P. K., Tripathi, S., Mehrotra, G. K., & Dutta, J. (2009). Perspectives for chitosan based antimicrobial films in food applications. Food Chemistry, 114, 1173–1182. Elsabee, M. Z., & Abdou, E. S. (2013). Chitosan based edible films and coatings: A review. Materials Science and Engineering C, 33, 1819–1841. Ghaderi, M., Mousavi, M., Yousefi, H., & Labbafi, M. (2014). All-cellulose nanocomposite film made from bagasse cellulose nanofibers for food packing applications. Carbohydrate Polymers, 104, 59–65. ¨ Grondahl, M., Eriksson, L., & Gatenholm, P. (2004). Material properties of plasticized hardwood xylans for potential application as oxygen barrier films. Biomacromolecules, 5, 1528–1535. Höije, A., Sternemalm, E., Heikkinen, S., Tankanen, M., & Gatenholm, P. (2008). Material properties of films from enzymatically tailored arabinoxylans. Biomacromolecules, 9, 2042–2047. ¨ Kohnke, T., Östlund, & Brelid, H. (2011). Adsorption of arabinoxylan on cellulosic surfaces: Influence of degree of substitution and substitution pattern on adsorption characteristics. Biomacromolecules, 12, 2633–2641. Khalil, A. H. P. S., Saurabh, C. K., Adnan, A. S., Nural Fazita, M. R., Syakir, M. I., Davoudpour, Y., et al. (2016). A review on chitosan-cellulose blends and nanocellulose reinforced chitosan biocomposites: Properties and their applications. Carbohydrate Polymers, 150, 216–226. Li, Y., & Mlynár Sarkanen, S. (1997). The first 85% kraft lignin-based thermoplastics. Journal of Polymer Science B: Polymer Physics, 35, 1899–1910. Li, W., Coffin, D. R., Jin, T. Z., Latona, N., Liu, C. K., Liu, B., Zhang, J., & Liu, L. S. (2012). Biodegradable composites from polyester and sugar beet pulp with antimicrobial coating for food packing. Journal of Applied Polymer Science, 126, E361–E372. Li, J., Zhou, P., Liu, H., Xiong, C., Lin, J., Xiao, W., et al. (2014). Synergism of cellulase, xylanase: And pectinase on hydrolyzing sugarcane bagasse resulting from different pretreatment technologies. Bioresource Technology, 155, 258–265. ˇ Simkovic, I., Nunez, A., Strahan, G. D., Yadav, M. P., Mendichi, R., & Hicks, K. (2009). Fractionation of sugar beet pulp by introduction ion-exchanging groups. Carbohydrate Polymers, 78, 806–812. ˇ Simkovic, I., Uhliariková, I., Yadav, M. P., & Mendichi, R. (2010). Branched arabinan obtained from sugar beet pulp by quaternization under acidic conditions. Carbohydrate Polymers, 82, 815–821. ˇ Simkovic, I. (2012). Flame retarded composite panels from sugar beet residues. Journal of Thermal Analysis and Calorimetry, 109, 1445–1455. ˇ Simkovic, I. (2013). Unexplored possibilities of all-polysaccharide composites. Carbohydrate Polymers, 95, 697–715. ˇ Simkovic, I., Kelnar, I., Uhliariková, I., Mendichi, R., Mandalika, A., & Elder, T. (2014). Carboxymethylated-: hydroxypropylsulfonated- and quaternized xylan derivative films. Carbohydrate Polymers, 110, 464–471. ˇ Simkovic, I., Tracz, A., Kelnar, I., Uhliariková, I., & Mendichi, R. (2014). Quaternized and sulfated xylan film derivatives. Carbohydrate Polymers, 99, 356–364. Rai, S., & Mehrotra, G. K. (2016). Lignin-chitosan polymer composites: A state-of-the-art short review. Asian Chitin Journal, 11, 23–30. Selvedran, R. R., & O’Neill, M. A. (1987). Isolation and analysis of cell walls from plant material. Methods of Biochemistry Analysis, 32, 25–153. Tetard, L., Passian, A., Farahi, R. H., Davison, B. H., Jung, S., Ragauskas, A. J., et al. (2011). Nanometrology of delignified Populus using mode synthesizing atomic force microscopy. Nanotechnology, 22, 465702, 9 pp. Tripathi, S., Mehrotra, G. P., & Dutta, P. K. (2009a). Preparation and physicochemical evaluation of chitosan/poly (vinyl alcohol)/pectin ternary film for food-packing applications. Carbohydrate Polymers, 79, 711–716. Tripathi, S., Mehrotra, G. P., & Dutta, P. K. (2009b). Physicochemical and bioactivity of cross-linked chitosan-PVA film for food packing applications. International Journal of Biological Macromolecules, 45, 372–376.

I. Sˇ imkovic et al. / Carbohydrate Polymers 156 (2017) 77–85 van den Broek, L. A. M., Knoop, R. J. I., Kappen, F. H. J., & Boeriu, C. G. (2014). Chitosan films and blends for packing material. Carbohydrate Polymers, 116, 237–242. Wise, L. E., Murphy, M., & D’Addieco, A. A. (1947). Chlorine holocellulose, its fractionation and bearing on summative wood analysis and on studies on hemicelluloses. Paper Trade Journal, 122(2), 35–43.

85

Yue, Y., Han, J., Han, G., Zhang, Q., French, A. D., & Wu, Q. (2015). Characterization of cellulose I/II hybrid fibers isolated from enerycane bagasse during the delignification process: Morphology, crystallinity and percentage estimation. Carbohydrate Polymers, 133, 438–447.