Introduction

Since 1924, cellulose films are well known under the trade name cellophane (Inskeep and Horn 1952). In the 1980s polyolefin films, such as polypropylene (PP) and polyethylene terephthalate (PET), displaced cellophane and started dominating the film market (Morris 2017). Despite its current minor position in the global consumption, cellophane is still commonly employed in applications such as packaging, adhesive tapes and battery separators, among others (Fink et al. 2001; Futamura Group 2022). Cellophane is considered as an attractive sustainable film since it is created from cotton or wood cellulose. However, its production is based on the viscose process, which poses high-production costs and environmental risks due to the use of toxic chemicals, including carbon disulfide, caustic soda and sulfuric acid (Fink et al. 2001; Ma et al. 2019).

Synthetic films are leading the market and, in addition, almost any utensils of our daily lives are wholly or partially made of plastic (Cronin et al. 2022), due to its high versatility, high durability, low-cost production, light weight and suitability for food packaging (Andrady and Neal 2009). Consequently, the global consumption is immense and is expected to grow from 460 million tons in 2019 to 1231 million tons in 2060 (OECD 2022). However, plastic usage also implies serious drawbacks. First, the contribution to the depletion of fossil resources, and second, the environmental pollution caused by insufficient global recycling rates, since most of the plastic is incinerated, ends up in landfills, or in the worst case, in water resources, and only 9% is recycled (OECD 2022). The misuse and mismanagement of plastic waste is one of the main reasons for the increasing contamination of the aquatic environment. The degradation of plastics into microplastics (size < 5 mm) presents an immediate danger for the biodiversity of animals and ecosystems, and further for human health (Laskar and Kumar 2019; Thushari and Senevirathna 2020). Therefore, more sustainable alternatives are needed to tackle these problems.

In response to these environmental challenges, the European Union issued the Single-Use-Plastic (SUP) Directive, which came into effect in July 2021. Several SUP products are banned gradually, such as cutlery, plates, straws, and lightweight plastic bags, among others. This measure envisions a twofold impact: reducing the contamination of fresh water resources through plastic debris and encouraging the development of new eco-friendly alternatives originating from renewable resources, e.g. cellulose (European Commission 2021). Cellulose is the most abundant biopolymer on earth present in a wide spectrum of plants and microorganisms (Koch 2006).

Due to its extensive inter- and intramolecular hydrogen bond network, cellulose is not susceptible to thermoplastic processing. Shaping processes require prior dissolution of cellulose either via (intermediate) chemical derivatization or by the use of direct solvent systems. Cellophane is produced via the former approach, which is intermediate xanthate formation. Direct dissolution of cellulose has been accomplished with various solvent systems, for instance, N-methylmorpholine N-oxide (NMMO) monohydrate (Fink et al. 2001), sodium hydroxide/urea solutions (NaOH/CO(NH2)2) (Zhang et al. 2001; Zhou and Wang 2021) and lithium chloride/N,N-dimethylacetamide (LiCl/DMAc) (Zhou and Wang 2021). However, the utilization of these solvents is accompanied by disadvantages like high toxicity, limited dissolution capacity and high process costs (Gericke et al. 2009; Heinze and Liebert 2001). Among the mentioned solvents, only NMMO is commercially used in the preparation of regenerated cellulose fibers and films. The process, known as Lyocell, is based on a dry-jet wet spinning technology where a cellulose solution in NMMO – H2O is extruded via an air gap into a coagulation bath, where the cellulose regenerates (Fink et al. 2001; Hummel et al. 2015). Lenzing AG has demonstrated that by using a slit nozzle it is possible to produce flat films with tensile strengths up to 300 MPa in machine direction (Gspaltl and Schloss-Nikl 2000). By changing the spinneret to a circular blow nozzle, tubular blown films can be generated by stretching in both transversal and longitudinal directions (Fink et al. 2001; Gspaltl et al. 1999). However, the utilization of NMMO bears the risk of side and runaway reactions at high temperatures due to its inherent thermal instability and requires stabilizers (Gericke et al. 2009; Holding 2016).

In recent years, the high cellulose dissolution capacity of ionic liquids (IL), demonstrated by Swatloski et al. (2002), opened a door to develop new and innovative sustainable film production processes. ILs are salts with melting points below 100 oC, low vapor pressures and high thermal stabilities. The potential of the ILs to manufacture cellulosic fibers and films has been demonstrated numerous times (Hermanutz et al. 2008). Mainly imidazolium-based ILs have been investigated to produce films, such as 1-ethyl-3-methylimidazolium acetate [emim][OAc] (Liu et al. 2013; Pang et al. 2015; Wawro et al. 2014), 1-allyl-3-methylimidazolium chloride [amim][Cl] (Pang et al. 2013), 1-butyl-3-methylimidazolium chloride [bmim][Cl] (Zheng et al. 2019) and 1-ethyl-3-methylimidazolium chloride [emim][Cl] (Pang et al. 2014). Thereby, films with tensile strengths of 55 MPa up to 152 MPa were produced by means of casting. However, imidazolium based ionic liquids are not inert under typical process conditions. They can undergo side reactions such a retro-alkylation (Zweckmair et al. 2015) and they can react with cellulose causing degradation of the polymer chains and side reactions (Clough et al. 2015; Ebner et al. 2008; Wendler et al. 2012). These challenges are not solved yet. Therefore, non-imidazolium based ionic liquids have been investigated thoroughly during the past decade.

Superbase-based ILs, such as 1,5-diazabicyclo[4.3.0]non-5-enium acetate, [DBNH][OAc], 7-methyl-1,5,7-triazabicyclo[4.4.0]dec-5-enium acetate, [mTBDH][OAc], and 1,8-diazabicyclo[5,4,0]undec-7-enium acetate, [DBUH][OAc] (Elsayed et al. 2020a, b) have shown to be excellent solvents for Lyocell-type dry-jet wet spinning of high-performance fibers (Michud et al. 2015b; Moriam et al. 2021). This fiber spinning process is known as the Ioncell® technology (Sixta et al. 2015). Various cellulose-based materials, e.g. hardwood pulp, pulp blends (Ma et al. 2015; Trogen et al. 2021), cardboards, paper, newsprints and various textile wastes (Haslinger et al. 2019a, b; Ma et al. 2016, 2018) were successfully converted in fibers and filaments for textile and technical applications.

This study represents the consequential next step of this development and addresses the continuous extrusion of films from pulp cellulose-[DBNH][OAc] solutions. Through comprehensive characterization of the film properties the process parameters (pulp concentration and extrusion temperature) were correlated with the structural features. The obtained films exhibit mechanical properties that outperform those of commercial cellophane.

Experimental

Materials

Pre-hydrolysis kraft (PHK) dissolving pulp from birch (Enocell, Stora Enso, Finland) was received in the form of sheets and was finely ground using a Fritsch mill. [DBNH][OAc] (Fig. 1), was synthesized by the slow addition of an equimolar amount of acetic acid (100%, C2H4O2, CAS: 64-19-7, M = 60.06 g/mol, Merck, Germany) to 1,5-diazabicyclo[4.3.0]non-5-ene (99%, DBN, C7H12N2, CAS: 3001-72-7, M = 124.18 g/mol, Fluorochem, UK) at 70 °C, under constant stirring to prevent the ionic liquid to solidify (Sixta et al. 2015). All numerical graphs in the Result and Discussion section were created using Origin software (Edwards 2002).

Fig. 1
figure 1

Chemical structure of [DBNH][OAc]

Full size image

Dope preparation

Prior to the preparation of the cellulose solutions (dopes), the IL was melted in a water bath at 80 °C. Then, respective amounts of cellulose (8, 10 and 13 wt% dry pulp consistency) were dissolved in [DBNH][OAc] in a vertical kneader at 80 °C (90 min), under constant stirring and vacuum pressure (< 100 mbar). Finally, the solution was filtrated by the use of a hydraulic press (using a GDK Ymax2 metal filter, with 5 μm nominal mesh size) at 80 °C and 180 MPa in order to remove the remaining undissolved solids and impurities (Ma et al. 2015).

Rheology of cellulose solutions

The viscoelastic properties of the dopes were examined using an Anton Paar Physica MCR 302 rheometer with a 25 mm parallel-plate geometry and 1 mm gap. The storage modulus (G’), loss modulus (G″) and complex viscosity (ƞ*) were determined via oscillatory shear measurements, with a 1% shear strain and a frequency range of 0.01 to 100 rad/s from 50 to 100 °C (Michud et al. 2015a). Anton Paar Rheocompass software was used to calculate the zero-shear viscosity (ƞ0*), according to the Cross model (Cross 1965), and the crossover points of the dynamic moduli (G′=G″) for each temperature (Mezger 2014).

Film production

Cellulose films were produced by means of a modified dry-jet wet spinning unit (Fourné Polymertechnik, Germany) reported previously by Sixta et al. (2015). The dope was positioned in a vertical cylinder equipped with a triple layer filter plate (280/170/100 mesh), a breaking plate and a slit nozzle (length: 60 mm, width: 0.25 mm). The dope was liquefied at 80 °C before adjusting the targeted spinning temperature (50–100 °C) in order to study the effects of the temperature/viscosity on the spinning process. The films were extruded through the slit nozzle at a rate of 0.133 m/min and immersed into a regeneration bath filled with cold water (10–14 °C) by passing a 1 cm air gap. The generated films were guided by a freely moving roll to a motor-driven Teflon collecting roll at a speed of 0.356 m/min ensuring a draw ratio (DR = take-up velocity / extrusion velocity) of 2.7. Subsequently, the films were washed thoroughly to remove the remaining ionic liquid by attaching them to a smooth glass surface, which was then immersed in a water bath at 60 °C for 30 min. Finally, the films were dried in an oven at 40 °C for 25 min.

Mechanical properties

The thickness of the films was measured by the use of a Mitutoyo IP 65 micrometer, following the standard ISO 4593:1993. The mechanical properties of the regenerated cellulose films were determined according to the standard ISO 527-3:1995 with an MTS 400 tensile and compression testing device using a 200 N cell and an elongation rate of 12 mm/min. Therefore, the films were cut to specimens with a fixed width of 15 mm and different lengths, depending on whether the measurement was in machine (MD) or in transversal (TD) direction. For machine direction, the length of the samples was 100 mm and the gauge length was set to 60 mm. For transversal direction, the length of the films depended on the width of the original film (minimum 24 mm) with a set gauge length of 20 mm. Stress and strain at breaking point, Young Modulus and toughness (Hearn 1997) were calculated using Matlab (Hunt et al. 2014). Prior to the measurements, samples were conditioned at 23 °C and 50% HR overnight.

Wide angle X-ray scattering (WAXS)

WAXS experiments were performed in transmission mode using a Xenocs Xeuss 3.0 CuKα X-ray instrument operated at 50 kV and 0.6 mA (λ = 1.5406 Å). The film samples were placed vertically on the sample holder for X-ray irradiation at reduced pressure (P = 0.16 mbar). The scattering intensity was recorded on a 2D Dectris Eiger2 R 1 M detector with a sample-to-detector distance of 56 mm and the background was corrected by a blank run. The 2D scattering patterns were processed using pyFAI software (Ashiotis et al. 2015). For oriented cellulose film samples, the 2D scattering patterns (Fig. S6) were processed to obtain structural parameters of cellulose crystals as previously described for cellulose fibers by Trogen et al. (2021); Elsayed et al. (2020b), with a modification for analyzing less oriented cellulose films: i.e. crystallite size and crystallinity index were estimated from the intensity profiles averaged for azimuthal angle of 360 degree for each scattering vector (SI, Sect. 4).

Transmittance

The transparency of the films was determined by measuring the optical transmittance in the visible light spectrum, from 400 to 800 nm (Liu et al. 2013), using a Shimadzu UV-2600 spectrometer.

Scanning electron microscopy (SEM)

The surface and cross-section of the films were examined via SEM in a Zeiss Sigma VP microscope. For the cross-section, the cellulose films were subjected to cryo-fracture method in liquid nitrogen to obtain an undistorted analysis surface. The samples were fixed to the holders using carbon tape and then sputter-coated with Au/Pd (80/20) (90 s, 20 mA). Finally, the film images were taken using an operating voltage of 3 kV (Ma et al. 2015).

Elemental analysis

The efficiency of the washing step is linked to the retained ionic liquid inside the regenerated films, which was monitored through the residual amount of nitrogen in the samples. The amount of nitrogen was measured by elemental analysis using a Thermo Fisher Flash Smart CHNSO Elemental Analyzer, in accordance with the standard EN 15104:2011. Helium was used as the carrier gas and sulfanilamide as standard. Further, the amounts of other elements (Mg, Ca, Mn, Fe, Cu, Si, Al) were determined via ICP-OES (Inductively coupled plasma optical emission spectroscopy) analysis using an Agilent 5900 SVDV system according to ISO 14869-3:2017. Lastly, the ash content of the films was quantified following ISO 1762:2001.

Chemical composition

The chemical composition of the produced films and pulp was investigated according to the two-step acid hydrolysis method described in the NREL/TP-510-42618 standard. Monosaccharides in the hydrolyzed solution were quantified by High-Performance Anion Exchange Chromatography (HPAEC-PAD) using a Thermo Fisher Dionex ICS-3000 system equipped with a Carbopac PA20 column. The amount of cellulose and hemicelluloses in the samples were then calculated from the amount of monosaccharides using the Janson formulas (Janson 1970). The Klason lignin, acid insoluble lignin (AIL), retained in the crucible after the filtration of the hydrolyzed suspension was washed with water and dried at 105ºC overnight, and then measured gravimetrically. Lastly, the acid-soluble lignin was quantified by measuring the absorbance of the hydrolyzed solution at a wavelength of 205 nm (spectrophotometer Shimadzu UV-2550).

Molecular mass distribution

The molar mass distribution (MMD) of the raw material and the cellulose films was determined by Gel Permeation Chromatography (GPC), by the use of a Thermo Fisher Dionex Ultimate 3000 system equipped with four PLgel MIXED-A columns, refractive index (Shodez RI-101) and multi-angle light scattering (Viscotek SEC-MALS 20) detectors, with 9 g L−1 LiCl/DMAc as eluent. The samples were subjected to an activation step, prior to the analysis, by a solvent exchange sequence of water – acetone – DMAc to facilitate the dissolution. Then, they were dissolved in 90 g L−1 LiCl/DMAc at room temperature under constant gentle stirring. Lastly, the solutions were subsequently diluted to 9 g L−1 LiCl/DMAc (sample concentration ~ 1 mg/mL), filtered with 0.2 μm syringe filters and injected for GPC analysis (Pitkänen and Sixta 2020; Potthast et al. 2015).

Contact angle

The hydrophobicity of the regenerated cellulose films was studied by means of a Biolin Scientific Theta Flex optical tensiometer. Sessile drop experiments were conducted according to standard ASTM D5725-99, where 4.5 µL water drops were placed on the surface of the films and the first contact angle, which represents the angle between the film and the free water drop after deposition, at ambient temperature and humidity, was determined by the Young-Laplace method.

Thermal stability

Thermal stability of the pulp and cellulose films was determined by thermogravimetric analysis (TGA) using a Netzsch STA 449 F3 Jupiter® thermal analyzer. The mass changes were evaluated from 40 to 600 °C at a heating rate of 10 °C /min, under a protective helium flow of 70 ml/min (Schlapp-Hackl et al. 2022).

Results and discussion

Film production

The cellulose films were produced by dry-jet wet spinning technique using a similar set-up as reported by Sixta et al. (2015) for producing man-made cellulose fibers. However, instead of a multi-hole spinneret a single slit nozzle spinneret (length: 60 mm, width: 0.25 mm) was used for film extrusion. In addition, the coagulation bath was tailored towards film spinning. A scheme of the setup is shown in Fig. 2. The extruded cellulosic films are regenerated in (a) an aqueous coagulation bath and collected at (b) a motor-driven roll, after passing (c) a freely movable roll. In addition, the speed of the motor driven roll is adjustable to produce films at different DRs.

Fig. 2
figure 2

Schematic illustration of the spinning line for the production of films, where a represents the coagulation bath, b the motor-driven roll, and c the freely movable roll

Full size image

The optimum spinning conditions (temperature, speeds, DRs) and bath set-up (bath arrangement, air gap distance) were evaluated using a 13 wt% cellulose - IL solutions, in analogy to earlier fiber spinning studies (Elsayed et al. 2020b; Hummel et al. 2015; Michud et al. 2015b). Several extrusion velocities were tested, with 0.133 m/min being the optimum speed for the used setup. Due to the viscoelastic properties of the 13 wt% dope, higher velocities resulted in melt fracture, a phenomenon by which high shear rates during the extrusion of viscoelastic polymers introduce distortions and irregularities on the extrudate surface, sometimes also referred to as shark skin (Bertola et al. 2003; Schramm 2004).

The distinct asset of dry-jet wet spinning is the possibility to exert elongational stress on the solution when passing the air gap (Sixta et al. 2015). Uniaxial stress in longitudinal direction affects the film dimension in transversal direction notably. After the immersion of the films into the water, cellulose coagulates instantly and further stretching is severely impaired (Hauru 2017). Consequently, the role of the air gap length was assessed in more detail. Air gaps from 0.5 to 7 cm were tested, causing differences in the film width. An increase of the stretching field led to a reduction of the film width. A too short air gap (0.5 cm) created some instabilities in the process due to the closeness of the bath and the slit, even though the shape of the films was not compromised. Overall, a minimum reduction of the film width in transversal direction and a stable spinning process was achieved by a 1-cm air gap, which is in compliance with the air gap used in fiber spinning (Sixta et al. 2015).

The stretching in the air gap is defined by the draw ratio (DR = take-up velocity / extrusion velocity) and controlled by the speed of the collector roll (Fig. 2: b) in relation to the piston velocity, which determined both the width and thickness of the final films (Fink et al. 2001; Guizani et al. 2020). By the use of low DRs wide and thick films are generated, whereas higher DRs led to narrower and thinner films. Different DRs were tested by keeping the extrusion speed constant and adjusting the speed of the collecting roll in order to have as wide films as possible within a certain range of thickness. The best draw ratio for this set-up was 2.7, which refers to a collector roll (production) speed of 0.356 m/min. By means of this DR, films with a thickness of 12–21 μm, thinner than cellophane (21–31 μm) (Futamura Group 2022), and widths of 2.4–2.9 cm have been produced. The production speed is limited by the used setup. However, Lenzing AG proved that the production of thin flat films can be achieved at even higher velocities at similar DRs (Gspaltl and Schloss-Nikl 2000; Schloss-Nikl et al. 1998).

The regeneration of cellulose in the aqueous spin bath is affected by the water temperature (Hauru et al. 2014). First, to possibly reduce the production costs, water at room temperature (18–22 °C) was tested. At this temperature, the regeneration process was impaired, probably due to the relaxation of the polymers during solidification (Hauru et al. 2014), which resulted in inhomogeneities in terms of thickness and width. This effect was inhibited by a reduction of the temperature below 18 °C. Temperatures in the range of 5–15 °C ensured good results and led to homogeneous films, which is in accordance with the fiber spinning studies (Sixta et al. 2015). Therefore, the water temperature was adjusted between 10 and 15 °C to find a compromise between film forming stability and energy demand.

Cellulose films show notable shrinkage during drying due to water evaporation and the collapse of the pores that it leaves behind (Ibrahim et al. 2022). Tailored washing and drying protocols are needed to overcome or minimize shrinking of the obtained films. To maintain the appearance and shape of the films, they were placed onto smooth glass surfaces and then submerged in water for washing. Different temperatures, from room temperature to 100 °C, were tested and the effect on the shape and the washing efficiency was evaluated. The simple width and residual nitrogen content were measured. Low temperatures did not remove the ionic liquid efficiently from the films while high temperatures caused shrinkage. It was found that submerging the films in water at 60 °C for 30 min was sufficient to remove the IL without affecting the width of the films. Regarding the drying, several temperatures were examined, from room temperature to 45 °C, with the films still attached to the glass surface. Temperatures up to 40 °C did not evoke shrinkage. Thereby, the drying time ranged from at least 12 h at room temperature to 25 min in the oven at 40 °C. At higher temperatures, a strong shrinkage, and therefore, a reduction in width was observed.

Once the ideal spinning and bath settings for the production of films from a 13 wt% cellulose solution were found, the impact of dope concentration and dope viscosity on the properties was studied. Thus, films were generated by the use of 13, 10 and 8 wt% dopes, and for each dope, the optimum conditions for film production have been found by varying the spinning temperature from 50 to 100 °C. The cellulose concentration and the spinning temperature had the biggest impact on the zero-shear (Table 1, Fig. S1) and complex viscosities of the dopes (Fig. 3). The viscoelastic properties of the solutions were adjusted by choosing an appropriate spinning temperature. In the case of fiber spinning, the optimum rheological conditions for 13 wt% dopes are zero-shear viscosities between 25,000 and 35,000 Pa·s, and a cross-over point (COP) at a modulus of ~ 4000 Pa and angular frequency of 1 s−1 (Sixta et al. 2015). For the film production, those conditions were not adequate to obtain high-quality films. High zero-shear viscosities triggered melt fracture, and a wavy pattern in transversal direction was observed (Bertola et al. 2003; Schramm 2004). The temperatures allowing stable spinning and the production of high-quality films from 13 wt% solutions were 80 to 100 °C, which corresponds to zero-shear viscosities of 19,000–8000 Pa·s, and COP of 3900–4200 Pa at angular frequencies of 1.3–3.4 s−1 (Table S1, Fig. S2). For 10 and 8 wt% solutions, the viscosities are reduced due to the reduction of the cellulose concentration. Consequently, for 10 wt% dopes, films have been successfully produced at temperatures from 60 to 80 °C at zero-share viscosities of 9400–3000 Pa·s and COP of 2500–2700 Pa at angular frequencies of 1.6–5.8 s−1 (Table S1, Fig. S3). In case of 8 wt% dopes, spinning took place at 50–70 °C, with zero-shear viscosities of 7600–2000 Pa·s, and a COP at ~ 1700 Pa at angular frequencies of 1.3–5.5 s−1 (Table S1, Fig. S4). However, at higher temperatures, above 70 °C for 8 wt% and above 80 °C for 10 wt% dopes, the process was affected by the reduced viscosities of the dopes. Too low viscosities caused an extrusion flow that was difficult to control and resulted in an inefficient stretching at the corresponding DR.

Table 1 Zero-shear viscosities (ƞ0*) of the solutions of 8, 10 and 13 wt% cellulose at different temperatures (50–100 °C)
Full size table
Fig. 3
figure 3

Double-logarithmic plot of the complex viscosities (ƞ*) of the spinning dopes (8, 10 and 13 wt% cellulose) versus frequencies (0.01 to 100 rad/s) at the indicated spinning temperatures

Full size image

Film properties

Mechanical properties

The main target for the produced cellulose films concerning their mechanical properties was to overcome the properties of commercial uncoated cellophane. The reference cellophane film showed tenacities of 125 MPa with 22% elongation in MD and 70 MPa with 70% elongation in TD, at a thickness of 21–31 μm (Futamura Group 2022). Table 2 summarizes the thickness and stress-strain behavior of films generated from dopes with different cellulose concentrations (8, 10 and 13 wt%) by varying the spinning temperatures between 50 and 100 °C. Figure 4 displays the stress–strain curves of selected samples. All the produced films are thin (12–21 μm) and present higher stress values compared to commercial cellophane in machine direction. The highest stress values, up to 210 MPa, were reached with a low polymer concentration (8 wt% dope). Similar values have been achieved by Lenzing AG with a spinneret of equal geometry by the use of a 12 wt% dope (Gspaltl and Schloss-Nikl 2000). Overall, these values are in contrast to the observations made in fiber spinning studies earlier in which an increasing polymer concentration led to higher strength values (Sixta et al. 2015). In terms of elongation, an average of 19% strain, close to cellophane, was obtained when spinning a 13 wt% dope at ~ 7800 Pa s. The elongation of the remaining films was significantly lower. This might be attributed to the higher orientation of the cellulose chains in these films (Table 3) and partly due to their reduced thickness. In terms of Young’s modulus, the films achieved values between 9 and 13 GPa, compared to the ca. 3 GPa of cellophane (Leppänen et al. 2019).

The extrusion temperature also affected the mechanical properties. Consistently for all concentrations, films produced at the lowest temperature, and therefore, the highest solution viscosity, present higher stress and lower strain values. Likewise, higher extrusion temperatures increased the strain values of the resulting films, at the expense of the tensile strength though.

Upon extrusion, the films can be solely stretched in longitudinal direction, not in transverse direction. Thus, the mechanical properties are not isotropic. However, similarly to the vales in MD, the highest values in TD were reached with 8 wt% dopes (95 MPa) and the lowest with 13 wt% (50 MPa). In contrast to the trend for MD, both the strength and elongation values increased with the extrusion temperature. Films from 8 wt% dopes spun at 50 °C exhibited values of 69 ± 4 MPa stress and 16 ± 2% strain and at 70 °C, 95 ± 8 MPa stress and 20 ± 5% strain, respectively. In case of 13 wt% dopes, extrusion at 80 °C produced films with a stress-strain behavior of 50 ± 2 MPa and 13 ± 7%, which increased to 88 ± 4 MPa and 34 ± 8% when heated to 100 °C. Although notably lower than in MD, these values are surprisingly high compared to cellophane, taking into account that the films were not biaxially stretched.

Table 2 Mechanical properties of cellulose films produced from 8, 10, 13 wt% dopes at the corresponding spinning temperatures
Full size table
Fig. 4
figure 4

Stress-strain curves of films produced from 8 (black), 10 (red) and 13 (blue) wt% dopes, at selected dope viscosities

Full size image
Table 3 Orientation, crystallinity and crystallite size of films prepared from 8, 10, 13 wt% dopes and the initial PHK pulp determined via XRD
Full size table

Figure 5 shows the diffraction pattern of the initial PHK pulp (cellulose I) and the obtained films (cellulose II). The pattern of the pulp corresponded to cellulose Iβ, where the main peaks (1–10), (110), (200) and (004) can be found at 2θ values of approximately 15o, 17o, 23o and 35o, respectively. The change from cellulose I to cellulose II after the regeneration of the cellulose was clearly observed in the diffraction pattern of the films. Peaks of 2θ values at around 13o, 20o, 22o and 35o were attributed to the planes (1–10), (110), (020) and (004) (French 2014). The overlapping (110)/(020) found in the regenerated films differs from the ideal pattern for randomly oriented cellulose II (French 2014) in the significant decrease of the 020 peak, observed in a sharp (110)/(020) reflection. Films frequently display this preferred orientation, together with the preferred orientation of the crystallographic (110) lattice plane (Gindl et al. 2006). Herman’s orientation, crystallinity and crystallite dimensions of the films produced at different cellulose concentrations and temperatures are summarized in Table 3. For all concentrations studied, the films produced at lower temperatures, and therefore higher viscosities, showed a higher orientation which decreased with higher extrusion temperatures. This is in line with the trend observed for the mechanical properties. Higher orientation leads to an increase in strength with a slight decrease in elongation. Consequently, due to the high Hermans orientation of 0.48, films produced with 8% cellulose consistency reached the highest stress (210 MPa), but the lowest strain values (8%) (Tables 2 and 3). Regarding the crystallinity of the films (34–38%), neither a change of concentration nor spinning temperatures showed a very significant effect. Moreover, these values are in consonance with the crystallinity of regenerated cellulose fibers produced by means of the same process (Elsayed et al. 2020b). In comparison, cellophane exhibits a crystallinity of 45% (Fink et al. 2001). In terms of crystallite dimensions, a preference in the planes (110) and (020) was noticed, whereas the crystal size in the plane (1–10) was very small. This differs significantly from the crystal orientation and crystallite size of cellophane. Cellulose crystallites in cellophane are mostly oriented in planes (1–10) and (110), with crystallite sizes of 53 and 52 Å, respectively, whereas the crystallite size in (020) is 36 Å (Fink et al. 2001). This difference in crystallinity and crystallite size between our cellulose films and cellophane could be caused by the way they are produced. Cellophane films are wet extruded, meaning that the solutions are extruded inside the spin bath and a complex interplay of regeneration and coagulation triggers instant solidification with limited stretching possibilities.

Fig. 5
figure 5

Diffraction pattern of the PHK pulp (orange) and the produced films from dopes with 8 wt% (black), 10 wt% (red) and 13 wt% (blue) cellulose consistency

Full size image
Table 4 Transmittance, chemical characterization, thermal stability and contact angle of regenerated cellulose films produced from 8, 10, 13 wt% dopes and the initial PHK pulp (* = Limit of quantification)
Full size table

Physical appearance of films

The obtained films are completely see-through, colorless and homogeneous, as can be seen in Fig. 6, with transmittance values close to 91% at 800 nm (Table 4). Compared to cellophane, which present values of 85–90%, the optical properties of these films are slightly improved (Qi et al. 2009; Zhang et al. 2001).

Fig. 6
figure 6

Illustration of the appearance of a produced film covering a wood log

Full size image

The surface and cross-sections of the produced films are shown in Fig. 7. The films are highly homogeneous, illustrating a smooth and uniform surface (bottom row). In addition, the cross-sections of the films (upper row) are even, compact and dense, with no visible holes or void spaces. In contrast, Fink et al. (1999) determined that the cross-section of cellophane presents a thin layer of spindle-shaped voids on each edge of the film and some in the central part oriented in machine direction, due to its aforementioned manufacturing process.

Fig. 7
figure 7

SEM images of the cross-sections (upper row) and surfaces (bottom row) of cellulose films produced from (left) 8, (center) 10, and (right) 13 wt% cellulose solutions

Full size image

In summary, despite being thinner, more homogeneous (Fig. 7), stronger (Table 2), and highly oriented (Table 3) the produced films are similar in touch and handling as cellulose films commercially available on the market (cellophane).

Chemical characterization

Elemental analysis was conducted to determine the amount of nitrogen introduced by IL residues. Inorganic elements (Al, Ca, Cu, Fe, K, Mg, Mn, Na, Si) were quantified by ICP-OES to determine the amount of impurities present in the films (Table 4). The nitrogen content of the films was 0.08–0.09%, in comparison to the initial PHK pulp 0.01%. Considering that each IL molecule contains 2 nitrogen units (Fig. 1) the remaining IL concentration can be estimated at ~ 0.04%. Therefore, most molecules have been successfully removed during the introduced washing step. The amount of aluminum, calcium, magnesium, sodium and silicon increases along with the cellulose consistency of the dopes. Copper, iron, potassium and manganese are relatively unaffected. The composition of the pulp changes from batch to batch, and the salts present in the water used during the production and washing process may cause a variation of those elements. Besides, some inorganics might originate from the equipment. Nevertheless, these results are in accordance with the values found for bleached pulp. The total amount of inorganic compounds was determined via ash content analysis (Table 4). The PHK dissolving-grade pulp contained 0.16% of inorganics, which is slightly higher than the values found in literature (0.07–0.08% by Sixta (2006). However, the ash amount found in the produced films, was in an expected range of 0.05–0.09%.

The carbohydrate composition of the initial PHK pulp and the produced films is gathered in Table 4. The cellulose, hemicellulose and lignin content of the produced films remained nearly the same as in the PHK pulp, which is in compliance with the results for fibers produced from the same substrate (Trogen et al. 2021).

Figure 8 presents the molar mass distribution of the obtained films and the initial PHK pulp. Different average molar masses calculated therefrom are summarized in Table 4. In Fig. 8, the curves corresponding to the films are slightly shifted to lower molecular weights compared to the raw material. The films experienced a decrease of the cellulose fraction with a DP > 2000 and an increase of the fraction with a DP < 100 compared to the starting material. Nevertheless, the cellulose degradation is minimal and was observed in earlier studies (Elsayed et al. 2020a; Ma et al. 2015).

Fig. 8
figure 8

Molar mass distribution of the films produced from 8 (black), 10 (red) and 13 wt% (blue) dopes compared to the initial PHK raw material (orange)

Full size image

Other properties

The surface properties of the films were assessed through contact angle measurements (Table 4). Measurements were conducted for each film produced from different dopes (8, 10 and 13 wt% cellulose consistency) at different spinning temperatures. The contact angle varied from 52o, corresponding to films produced from 8 wt% solutions, to almost 60o for films produced from 10 to 13 wt% dopes. However, the spinning temperature showed no significant influence (Table S2). Pictures of the water drops illustrating the first contact angle are presented in Fig. 9. According to Liukkonen (1997), cellophane reaches a value of 34o. The difference between the contact angle of cellophane and the created films might be explained by their morphology. As shown by the SEM images, the cross-section of the IL-derived films is compact and dense, whereas cellophane is less homogeneous (e.g. holes) (Fink et al. 1999), allowing water to penetrate more easily across films.

Fig. 9
figure 9

Illustration of the contact angle measurements of the man-made Ioncell® films produced from (left) 8, (center) 10 and (right) 13 wt% cellulose dopes

Full size image

The thermogravimetric analyses of all produced films, obtained from 8, 10 and 13 wt% dopes and the initial PHK pulp are plotted in Fig. 10. All films exhibit a high thermal stability and an almost similar degradation profile. The onset of the main mass loss starts at 310–330 °C and levels off at 375–390 °C (Table 4). Under inert conditions, around 30% of mass is left at 600 °C. The PHK pulp shows a similar trend, however, after the main reaction only 13% of mass is left. This difference might be justified by the increased crystallinity and orientation as well as by the different types of polymorphs (cellulose I and II). These values are almost identical to those of cellophane and in the range of lyocell, modal and viscose fibers (Carrillo et al. 2004; Li et al. 2018).

Fig. 10
figure 10

Thermogravimetric analyses curve (solid) and first derivative (dash) of the PHK pulp (orange) and the films produced from the 8 (black), 10 (red) and 13 wt% (blue) cellulose solutions

Full size image

Conclusions

Highly transparent cellulose films with excellent mechanical properties were prepared through direct dissolution of cellulose in a superbase-based ionic liquid and subsequent dry-jet wet extrusion. Unlike in stationary film casting, continuous film extrusion requires defined rheological properties of the solutions to maintain stable film formation. The extrusion velocity had to be adjusted to the viscoelasticity of the cellulose-IL system, which is governed by the cellulose concentration and processing temperature, to avoid flow instabilities and shark skin formation. Furthermore, the temperature of the aqueous coagulation bath had to be slightly below room temperature for continuous, homogeneous film formation. All films showed tensile strength values in MD higher than commercial cellophane (125 MPa), and values up to 210 MPa have been reached at thicknesses as low as 12–21 μm. Through tailoring the processing parameters, it was possible to vary the polymer orientation and thus, selectively increase either the tensile strength or elongation of the films. Due to their wide property spectrum these films hold great potential as sustainable alternative to petrochemical-based films for packaging, membranes in pressure-driven processes, and separators in batteries.