α-Cellulose-based films: effect of sodium lignosulfonate (SLS) incorporation on physicochemical and antibacterial performance

A homogeneous α-cellulose film was prepared by a regeneration method from ZnCl2/CaCl2/cellulose mixed system and was further combined with sodium lignosulfonate (SLS) via crosslinking through intermolecular hydrogen bonds and “bridge linkages”. The physicochemical and antibacterial performance of prepared films were investigated and results showed that the modified film exhibited stronger tensile strength, higher thermal stability, lower hydrophilic effect, better UV shielding as compared with the original one, and especially, better antibacterial ability derived from the presence of phenolic hydroxyl and sulfonate groups in SLS. This study proposed a simple and sustainable method for fabricating a multifunctional and environmentally friendly composite film using two main lignocellulose resources as raw materials.


Introduction
Cellulose, as one of the three main ingredients in lignocellulosic biomass, is a famous biopolymer and has great potential for preparing lm-based materials (Khalili et  Recently, the regenerated cellulose lm derived from regenerated method is a promising alternative to polymeric one (Puspasari et al., 2018). Many works have focused on the investigation of regenerated method due to its excellent bio-compatibility, and nontoxicity (Lee & Jeong, 2015;Pang et al., 2015). For instance, Weng et al. prepared the regenerated cellulose lm through phase inversion, where Nmethylmorpholine-N-oxide (NMMO) was used as a solvent for dissolving cellulose, and the obtained lm exhibited high rejection rate (99.7 %) for Congo red (Weng et al., 2017). Nevertheless, the pure regenerated cellulose lm is hydrophilic, poor thermal stability, bacteria prevention, and mechanical properties (Mao et al., 2006;Ruan et al., 2004). Chemical modi cation or blended with other polymers is effective pathway for addressing this issue through introducing different functional groups and modifying the inherent structure.
Sodium lignosulfonate (SLS) is a by-product produced via the production of wood pulp by using sul te pulping processes. Approximately 50 million tons of SLS can be obtained every year, where 95 % of this resource is consumed by direct combustion for low-e ciency electricity supply (Kim & Um, 2020;Xu et al., 2020), therefore it is meaningful to achieve high e ciency of SLS. Fortunately, abundant presence of active groups, such as hydroxyl (O-H) and sulfonate (R-SO 3 − Na + ), and various linkages, including C-O and C-C, make SLS reactive for oxidation, reduction, and polymerization reactions. Recently, an inorganic salt system (zinc chloride/calcium chloride) was researched for preparing cellulose lms . Therefore, we prepared regenerated lms (including pure and mixed lms with different content of SLS) through inorganic co-system (ZnCl 2 /CaCl 2 ), and modi ed these obtained lms through crosslinking with assistance of succinic anhydride and triethanolamine. The result showed that the modi ed mixed lm (cellulose/SLS-M) with assistance of succinic anhydride and triethanolamine acting as crosslinking agents exhibited excellent properties (such as thermal stability, hydrophobic effect, mechanical performance, and UV transmission) as compared with those of original one, and the improved performance of bacteriostasis was mainly derived from the addition of SLS.

Materials
In this study, α-cellulose with particle size of 250 µm used was purchased from Aladdin (Shanghai, China

Preparation of mixed cellulose lms
These cellulose lms (CFs, including pure and mixed cellulose lms which contained different amounts of SLS) were prepared via a regeneration method reported in a previously reported literature (Hou et al., 2019). Typically, 0.45 g α-cellulose was added into 1.05 g deionized water to obtain a cellulose suspension liquid. Afterwards, continuous stir was performed in a beaker (10 mL) for 30 min to achieve uniform dispersion of cellulose, which was recorded as solution 1. Then, 9.87 g ZnCl 2 and 0.30 g CaCl 2 were mixed into 3.63 g deionized water. After 30 min stir, different amounts of SLS (10, 20, 30, and 40 wt%, based on the input of α-cellulose) were added into this homodisperse metal precursor solution, and were stirred at 75 o C for 30 min together, which was named as solution 2. After that, solution 2 was dropwise added into solution 1 with intense stir and was further stirred at room temperature for 12 h to make these two solutions mix well. Finally, the obtained solution was placed under vacuum atmosphere to remove air bubbles and was casted on a glass plate to prepare the lm with thickness of 450 µm, and then it was immersed in 450 mL absolute ethanol for 1 h to separate the lm from glass plate, which was immediately xed with two splints and dried at room temperature overnight. The prepared lm was tailored with size of 2 cm × 2 cm for the next characterization and was denoted as SLS x -CF, where x means the mixed addition of SLS (wt%). For instance, SLS 0 -CF represents the pure CF without extra addition of SLS, while SLS 10 -CF means the mixed CF with the addition of 10 wt% SLS. . Typically, it is di cult to realize mutual crosslinking between two macromolecules through chemical pathways due to their complex structures and unclear reaction mechanism. Therefore, the introduction of "bridge monomers" (succinic anhydride and triethanolamine) is bene cial for improving the interconnection between these two macromolecules. Generally, the polymerization reaction occurs between carboxylic acids and alcohols at a certain temperature to form polymeric esters. In theory, it is feasible to combine cellulose and lignin through this mechanism mentioned above. In this section, succinic anhydride and triethanolamine were utilized as "bridge monomers" (i.e., crosslinking agents) to improve the interconnection between α-cellulose and SLS to form the crosslinking structure, and the speci c modi cation steps are as follow.

Modi cation process
First, succinic anhydride (0.18 g), triethanolamine (0.45 g), and different volumes of absolute ethanol (100, 200, and 300 mL) were blended uniformly to prepare organic precursor solutions with different concentrations. Afterwards, the pure CF (i.e., SLS 0 -CF) was immersed into different organic precursor solutions for different time (1, 2, and 3 days). Finally, the obtained lm after immersion was cured at different temperatures (50, 60, and 70 o C) for 6 h to make the modi cation complete. Finally, the lm after modi cation was washed by ethanol for three times to remove unreacted molecules, and dried at curing temperatures (50, 60, and 70 o C) for 12 h, which was named as CF-M. The performance of obtained lms was preliminarily evaluated via their mechanical properties to optimize the modi cation condition, and results were presented in the next discussion. Other blending lms, which contained different amounts of SLS (10, 20, 30, and 40 wt%), were all modi ed by the optimized condition (300 mL ethanol, 70 o C curing temperature, and 1 day immersing time), which were assigned as SLS x -CF-M. During the curing at mild conditions (below 100 o C), succinic anhydride was hydrolyzed as succinic acid, which can be polymerized with polyols (such as triethanolamine, and hydroxyl groups in structures of cellulose and lignin) to achieve the interconnection of several molecules, and the illustration of modi cation process is presented in Scheme 1. From the structure displayed in Scheme 1, two internal bonding (interaction hydrogen bonding and crosslinking derived from "bridge monomers") can be obtained to fabricate the crosslinking structure of modi ed blending lms.

Film characterization
The lm thickness was measured by an ID-C112XBS micrometer (Mitutoyo Corp., Tokyo, Japan) and presented as average value of ve points. FTIR (Fourier transform infrared spectrometer) was performed on a VERTEX 70 spectrometer. The range of measured wavenumbers was between 500-4000 cm − 1 , and 32 scans per spectrum were collected with a resolution of 4 wavenumbers. The surface morphology was recorded by S-4800 cold eld emission SEM (scanning electron microscope). The thermal stability of prepared lms was carried out on a TA Instruments TGA (thermo gravimetric analyzer) Q500 (TA Instruments, USA) with a xed heating rate of 10 o C/min from room temperature to 600 o C with nitrogen ow of 40 mL/min. Tensile test was conducted on the lm with a strain rate of 30 mm/min at 25 o C via an auto tensile tester (SANS CMT4000). The lm specimens contained width and length of 10 and 35 mm, respectively. The water contact angle was determined by Kruss DSA100. Light transmission was carried out through the lm (2 cm × 2 cm) on a Lambda 950 UV-vis spectrometer in the wavelength range of 200-600 cm − 1 .
The water absorption was measured via Eq. (1): where W t was the mass of lm measured at different immersion time at room temperature after excess surface water was removed.
Water vapor permeance (WVP) was tested by the standard ASTM method E96. Typical bottles with diameter of 2.7 cm and volume of 20 mL were selected to determine the WVP of lms. First, the lm was cut into a circle with a diameter of 2.8 cm (slight larger than bottle used). After that, a determined amount of anhydrous CaSO 4 (RH = 0 %) was added into the bottle until two-thirds volume (approximately 3 g), which was covered by different lms. Each bottle was placed into a desiccator (a 100 mL beaker) containing saturated Na 2 CO 3 solution at the bottom. Excess amount of Na 2 CO 3 was added to ensure the All measuring data (e.g., water absorption, WVP, and the diameter of inhibition zone) was recorded in triplicates for each group.

FTIR
FTIR spectroscopy is a common characterization for recording the structural change of testing materials, which is presented in Fig. 1. As shown in Fig. 1a, the pure CF exhibits a strong and broad absorbance at around 3300-3350 cm − 1 , which is assigned to O-H stretching derived from the structure of cellulose. The

SEM
The surface morphology of pure CF, CF-M, and SLS x -CF-M was observed by SEM and shown in Fig. 2.
Clearly, the surface of pure CF is homogeneous, while the obvious variation on surface morphology appears after modi cation to make the lm heterogeneous, which is due to the co-existence of different crystalline and amorphous regions derived from different components. With the addition of SLS, the lm becomes more heterogeneous and phase separation occurs as the content of SLS increases from 10 to excess addition of SLS is not bene cial for the effect derived from "bridge monomers" to make the interconnection between cellulose and lignin close, which may be the cause of variation in mechanical property. Figure 3 shows thermosgrams of pure CF and other modi ed blending lms. As presented in Fig. 3, the thermal degradation process is mainly divided into three stages, including moisture evaporation (stage 1, . Furthermore, not only the improvement of thermal stability during main decomposition process can be achieved, but also the content of condensed phase obtained after pyrolysis also increases, which is caused by structural collapse of cellulose backbone and lignin aromatic units to form the thermostable char with higher crosslinking density. TGA results illustrate that the thermal stability of pure CF is improved obviously, and predictable reasons are as follows: 1) abundant polar bonds (hydroxyls) were introduced with the addition of lignin, which could interact with cellulose via interaction hydrogen bonds to form crosslinking network; 2) the introduction of "bridge monomers" was bene cial to form the interconnection between lignin and cellulose via "bridge linking" bonds (the polymerization of carboxylic acid and alcoholic hydroxyl), thus strengthening the stability of mixed system; 3) three-dimensional structure of lignin was more stable as compared with cellulose, which was harder to be decomposed at high temperatures, as more stable condensed phase (char) with higher thermal stability was generated due to the repolymerization caused by the instability of active functional groups and the regeneration of carbon-carbon bonds. For obtaining the optimized condition of modi cation in Sect. 2.3, orthogonal experiments for testing tensile strength were performed with three variables (i.e., ethanol volume/mL, immersing time/day, and curing temperature/ o C), and each tensile strength presented in Table 1 was the average of three records.

Mechanical properties
From Table 1, the tensile strength of 64.6 MPa was obtained in 300 mL ethanol at 70 o C for 1 day. More ethanol and less immersion time are bene cial for the lm regeneration and maintaining the integrity of obtained lms, respectively. Excessing immersing time may slightly destroy the interaction bonding, thus declining the degree of crosslinking . Similarly, higher temperatures will accelerate the crosslinking to strengthen the internal interconnection.
In addition, all lms exhibit high tensile strength and relevant strain-to-failure values (see Fig. 4). The pure CF shows a tensile strength of 55.1 MPa and a strain-to-failure value of 18.5 %. After modi cation, these two values all increase signi cantly to 66.7 MPa and 20.8 %, respectively, which demonstrates that strong crosslinking bonds are formed after "bridge monomers" addition to extremely improve the mechanical properties. With the incorporation of 10 wt% SLS, the value of tensile strength rises obviously to 67. 8 MPa, while that of strain-to-failure declines to 10.6 %. Further increasing SLS addition, both tensile strength and strain-to-failure value all dropped rapidly, which indicated that the strong interaction hydrogen bonding is created between SLS and cellulose lm, as proven in FTIR, but phase separation is subsequently formed (see SEM images in Fig. 2), breaking the compatibility of blending system, thus leading to the degradation of mechanical properties. As for SLS 10  Nevertheless, the worse hydrophilic capacity of mixed lms is obtained after SLS addition, and the main reason is that hydroxyl groups in both SLS and cellulose can be interacted with each other to form crosslinking structure. In addition, the formation of "bridge linking" is positive for the decrease of hydroxyl groups, thus lowering hydrophilic ability (Llorens et al., 2015). The excess addition of SLS is negative for the blending lm to obtain a compatible system as separation phase (SLS region) is formed on the surface of lm, therefore increasing their hydrophilic capacity.

Light transmittance
The UV transmission of pure CF and modi ed mixed lms was recorded in the range of 200-600 nm, and their transmission curves are shown in Fig. 7. The transmission of pure CF reaches 69.6 % at 280 nm, while that of SLS 10 -CF-M is 1.2 %, which almost appears opaque. Further increasing the SLS addition from 10 to 40 wt% signi cantly decreases the transmission of blending lms, almost close to 0 %, showing excellent UV blocking performance. These observations suggest that the extra addition of SLS makes the lm surface bumpy (see SEM images), thereby inducing optical scattering and refraction. Therefore, with assistance of "bridge linking" and interaction hydrogen bonding, the presence of SLS plays an important role in blocking UV transmission, which make the blending lms potential in UVshielding application.

Antibacterial property
The antibacterial ability of different samples was evaluated via measuring the diameter (mm) of inhibition zones, a classic method, which adopted lm samples on the central location of agar medium coated with bacteria, and observed the inhibition of growth and propagation of bacteria (E. coli) to form a blank circle region around the samples, which was generally named as inhibition zone. According to the result in Fig. 8, the original CF presented as a control sample almost exhibits no inhibition zone against E. coli, while an apparent inhibition zone is observed after 10 wt% SLS addition in Fig. 8b, indicating the important role of SLS on antibacterial performance. Increasing the addition of SLS from 10 to 40 wt%, the obtained inhibition zone becomes larger and larger (from 8.0 mm for SLS 10

Conclusions
In this study, SLS x -CF-M composite lms were fabricated, and FTIR result demonstrated the completion of blending lm preparation. The crosslinking structure formed by interaction hydrogen bonds and "bridge linkages", and separation phase induced by the increasing addition of SLS were two main reasons for the change of mechanical and hydrophobic properties, while the UV transmission and antibacterial performance were all in uenced by the addition of SLS. Treating SLS as a renewable alternative resource of other synthetic polyols or bacteriostatic agent for lm modi cation meets the requirement of sustainable development Declarations