1 Introduction

In recent years, interest in the development of environmentally friendly bio-based films for packaging has increased (Wang et al. 2016; Wilpiszewska et al. 2020). Due to its biodegradability and renewability, cellulose has gained attention as a potential replacement for fossil-derived plastics made from fossil raw materials (such as polyethylene, PE, or polypropylene, PP) especially in packaging applications, (Klemm et al. 2005; Wang et al. 2013; Almeida et al. 2023).

The dissolution of cellulose is usually performed with dissolving pulps, which are generally made from wood (85–88%) and cotton linter (10%). Compared to paper pulp, dissolving pulps have a lower proportion of hemicellulose (3–6%), a higher proportion of cellulose (92–96%), and minimal amount of lignin and impurities (Chen et al. 2016); Sixta 2006). In the literature, these pulps are usually used for the production of regenerated cellulose fiber and films.

Cellulose has been dissolved using various solvents, such as lithium chloride/N, N-dimethylacetamide,, ionic liquids, tetra butyl ammonium fluoride/dimethyl sulfoxide, and alkali/urea solutions. Ionic liquids and alkali/urea solutions have attracted the most attention recently. Ionic liquids are chemically and thermally stable, but their corrosiveness, high cost, and difficulty in recycling limit their industrial use (Tu et al. 2021); Ribeiro et al. 2021). Among the alkali/urea systems, the NaOH/Urea/H2O solvent (7:12:81 wt% at -12.0 ºC)(Cai and Zhang 2005) is cost-effective in comparison with ionic liquids and less toxic than LiCl/DMAc, or NMMO/H2O (Olsson and Westm 2013; Tu et al. 2021; Delgado et al. 2023).

In the literature, anti-solvents (also called non-solvents), used to regenerate dissolved cellulose, are usually referred to as coagulants. Since the regeneration process is related to the diffusion between the solvent and the coagulant, there are many factors that can influence it, such as, the nature of the coagulant, the temperature, the diffusion coefficient in water, etc. On the other hand, the orientation of the cellulose fibers in solution is very important to create ordered structures by intra/inter molecular interactions that improve the final mechanical properties (Yang et al. 2011; Wang et al. 2016; Huang et al. 2022). The most commonly used anti-solvents in the regeneration of dissolved cellulose are acidic water (with CH3COOH, H2SO4, H2SO4/Na2SO4), distilled water, alcohols, or acetone (Zhang et al. 2001; Li et al. 2012; Wang et al. 2016; Wei et al. 2024). However, for the production of cellulose films by regeneration with anti-solvents it is necessary that the cellulose solution has sufficient viscosity to spread on glass plates. The temperature of the regeneration bath can also influence the final properties of the regenerated films (Liu and Zhang 2009; Liu et al. 2011). Thermal gelation of cellulose dissolved in NaOH/Urea/H2O has been reported before, but the method was not used to produce cellulose film, but to evaluate the storage capacity of cellulose solutions (Cai and Zhang 2006). Other authors studied the water evaporation during preparation of cellulose films at 25 ºC, without evaluating the effects of temperature (Wang et al. 2019).

The molecular structure of cellulose undergoes significant changes upon dissolution and regeneration, leading to the formation of biodegradable films (Jiang et al. 2014; Tu et al. 2021). However, the films produced during the dissolution/regeneration process often exhibit brittleness due to the high number of hydrogen bonds and physical stacking in cellulose chains (Medronho and Lindman 2015). To overcome this problem, plasticizers can be used to penetrate the interstices between the cellulose chains, preventing the inter-chain hydrogen bonding.

Glycerol is known for its low cost, non-toxicity, and environmental friendliness, and has been widely used as plasticizer in the research of cellulose film (Xiao et al. 2003; Cielecka et al. 2019; Cazón et al. 2020b; Van Nguyen and Lee 2021). Xiao et al. (2003) used glycerol and α-monoglycerides for the plasticization of cellulose films, but they did not study the effect of glycerol concentration or the effects of hot-pressing operation . Cazón et al. (2018) investigated the use of glycerol as a plasticizer but used polyvinyl alcohol (PVA) and acetic acid in the regeneration step . Rumi et al. (2021) used glycerol to plasticize films derived from cotton fibers dissolved in DMAc/LiCl solvent with hot-pressing step . Van Nguyen and Lee (2021) used glycerol in microfibrillated cellulose wet films. However, none of those authors have considered the use of thermally regenerated cellulose films followed by plasticization with glycerol and subsequent hot-pressing.

In this study, cellulose films were prepared by dissolving cellulose pulp in NaOH/Urea/H2O solvent (at -12 ºC) followed by thermal regeneration at 50 ºC for 2 h – an approach attempted in this work for the first time. The resulting materials were washed with water to remove salts, air dried at room temperature, plasticized by immersion in glycerol solutions of different concentrations (10–70 wt%) and hot-pressed to obtain the flexible films. The films so obtained were characterized by ATR-FTIR, DSC, TGA, XRD, contact angle measurements, transparency analysis, light, and gas permeabilities (oxygen and water vapor) aiming to assess the films’ potential for packaging applications.

2 Materials and methods

2.1 Materials

Industrial cellulose pulp (bleached eucalyptus kraft pulp, BEKP, DP≃1100) was kindly supplied by the Biotek cellulose company (Altri, SGPS, SA) and used without further purification. Deionized water was obtained by reverse osmosis. Sodium hydroxide (NaOH, 99%) and ethanol (C2H5OH, 96.0%) were purchased from JGMS. Urea ((NH2)2CO, 99%) and sulfuric acid (H2SO4, 95–97%) were bought from ChemLab. Potassium hydroxide (KOH, 94.0%) was purchased from EMPURA and sodium bromide (NaBr, 99.5%) was acquired from TCI. Glycerol (Gly, 99+%) was purchased from Alfa Aesar. All chemicals were used as received.

2.2 Methods

2.2.1 Kraft pulp pretreatment

Cellulose kraft pulp was firstly hydrolyzed with sulfuric acid (Rebelo et al. 2024). Briefly, the pulp was mechanically defibrillated (in dry) using in coffee grinder to obtain a cotton-like material. Then, 150 g of ground pulp were hydrolyzed with sulfuric acid (10.5 ml) during 72 h, at room temperature, with mechanical stirring. The solution was neutralized with KOH/ethanol solution (0.5 wt%) washed with ethanol and vacuum filtered. The degree of polymerization (DP) of resultant cellulose powder was 267, obtained by intrinsic viscosity method [η] at 25 ± 0.5 °C, using a described equation (Cai and Zhang 2006). Final DP was similar to other hydrolyzation methods reported in the literature (Setu et al. 2014; Paula et al. 2019) and similar to commercial MCCs (Rojas et al. 2011; Alves et al. 2016). The cellulose powder obtained was maintained in an oven (50 °C) until use.

2.2.2 Dissolution of cellulose

The hydrolyzed kraft cellulose powder was dissolved in the NaOH/Urea/H2O (7:12:81 wt%) solvent as previously reported (Qi et al. 2009). The desired amount of cellulose (12.0 g) was slowly added to 200 ml of solvent cooled to -12 ºC. The mixture was stirred for 3 h to achieve a yellowish transparent cellulose solution of 6% (w/v).

2.2.3 Preparation of cellulose/glycerol films

Cellulose-based films were prepared as illustrated in Fig. 1. Briefly, the cellulose solution (6%, w/v) was spread on glass plates (18 × 30 cm) using a micrometric film applicator (1000 μm, Zehntner ZUA 2000 series). Then, the glass plates were placed in an oven at 50 ºC for 2 h to obtain cellulose gels. The resulting gels were washed with distilled water until their surface reached a pH = 7 (3–4 washes). To prevent shrinkage during drying, the resulting gels were fixed over a plastic film with adhesive tape. The films were then air-dried at room temperature for 48 h to obtain cellulose control films (Cel0). To incorporate glycerol (as plasticizer), cellulose films were immersed in aqueous glycerol solutions of different concentrations (10, 20, 40 and 70 vol%) for 30 min (at room temperature). Then, the soaked materials were dried by hot-pressing (at 105 °C for 30 min and 0.1 MPa) to obtain the plasticized cellulose films (Cel10, Cel20 Cel40 and Cel70, respectively).

Fig. 1
figure 1

Experimental methodology used in the preparation of cellulose/glycerol films

Full size image

The samples were labelled as Celϰ where ϰ corresponds to the glycerol concentration in the soaking bath, i.e., a film labelled “Cel20” is a film obtained upon soaking a cellulose film in a glycerol 20% (v/v) solution. A control sample (without glycerol) was also prepared and denoted as “Cel0”.

2.2.4 Glycerol quantification

The glycerol content (GC) of final films was evaluated by two different methods: (1) by gravimetry, and (2) by TGA analysis (Ribeiro et al. 2021). Gravimetrically, samples of all films were dried for 24 h in a vacuum oven at 40 ºC and weighed repeatedly until a constant weight was reached. This procedure was repeated before soaking in the glycerol solutions and after the hot pressing. The glycerol content (GC, wt%) was calculated by the mass ratio as follows,

$$\:\text{G}\text{C}\:(\text{\%}.\text{w}\text{t})=\:\frac{{\text{w}}_{\text{f}}-{\text{w}}_{\text{i}}}{{\text{w}}_{\text{i}}}\:\times\:100$$

where Wi and Wf are the initial dry weight (before soaking) and final dry weight (after hot-pressing) of cellulose films, respectively.

By TGA, the GC was calculated by the drop of mass loss (wt%) corresponding to glycerol evaporation step (125–290 ºC) in the thermogravimetric curve.

2.2.5 Films thickness measurement

The film thicknesses were measured using a thickness Syntek electronic micrometer (accuracy 0.001 mm) at 3 different locations, and the average value was used for characterization/calculations purposes.

2.3 Characterizations

2.3.1 Fourier transform infrared spectroscopy (FTIR)

Chemical interactions of cellulose/glycerol films were evaluated by FTIR in ATR mode in 4000 –600 cm− 1 wavenumber range, with 4 cm− 1 resolution and 64 accumulations using Carey 630 spectrometer (Agilent Technologies) equipped with a Golden Gate Single Reflection Diamond ATR, at room temperature.

2.3.2 Morphology

The film surface and cross-section morphologies were examined by scanning electron microscopy (SEM). The film samples were secured with adhesive carbon tape, coated with gold, and observed in a field emission scanning electron microscope (FESEM), ZEISS MERLIN (Compact/VPCompact, Gemini II, Germany), using an acceleration voltage of 2 kV. To perform the cross-section observations, the samples were firstly frozen in liquid nitrogen and fractured to prevent the risk of plastic deformations.

2.3.3 Crystallinity

The crystallinity variations were analyzed by X-ray diffraction (XRD) patterns acquired via an automated multipurpose X-ray diffractometer (Rigaku SmartLab®) with a copper anode, CuKα radiation of Kα (1.5406 Å), at 40 kV and 50 mA, in 2θ range from 5º to 50º at a scanning rate of 2.0º min− 1.

The spectra were smoothed using a 5th order polynomial function of the Savitzky-Golay method. The crystallinity index (CrI, %) was estimated using deconvolution method of Gaussian model, by the ratio between the crystalline peaks area and the pattern total area (Park et al. 2010; Yao et al. 2020; Rebelo et al. 2023) after subtracting the background signal of the substrate measured without any sample, as follows.

$$\:\text{C}\text{r}\text{I}\:\left(\text{\%}\right)=\:\frac{\text{c}\text{r}\text{y}\text{s}\text{t}\text{a}\text{l}\text{l}\text{i}\text{n}\text{e}\:\text{p}\text{e}\text{a}\text{k}\text{s}\:\text{a}\text{r}\text{e}\text{a}}{\text{t}\text{o}\text{t}\text{a}\text{l}\:\text{a}\text{r}\text{e}\text{a}}\:\times\:100$$

2.3.4 Thermal studies

The thermal stability of cellulose-based films was analyzed by thermogravimetric analysis (TGA) using NETZSCH STA 44F5 (Netzsch, Germany). The samples were tested up to 500 ºC using a heating rate of 10 ºC⋅min− 1 with continuous nitrogen purge flow (250 ml∙min− 1).

The thermal behavior was also evaluated by differential scanning calorimetry (DSC) using DSC 204 F1 Phoenix model (Netzsch, Germany). The samples (5–10 mg) were conditioned in aluminum pans closed with aluminum lids, heated from 25 ºC to 100 ºC followed by cooling to -90 ºC, and then heated again up to 100 ºC using a heating/cooling rate of 10 ºC·min− 1. A nitrogen environment with a purge flow of 40 ml-min− 1 was used. The presented results are from 2nd heat flux curve.

2.3.5 Tensile testing

The tensile tests were performed as described in a previous work (Ribeiro et al. 2021). Succinctly, cellulose film rectangular-shaped specimens (50 mm × 10 mm) were tested at a rate of 5 mm·min− 1 until failure in an Instron 5944 mechanical tester (Instron, USA) equipped with a 500 N load cell. Two types of samples were tested: (1) samples dried in a vacuum oven at 40 ºC for 24 h before testing (dry samples) and (2) samples maintained in a desiccator with sodium bromide at 50 ºC for 7 days to a provide relative humidity (RH) of ≃50%. The presented results are means and standard deviations of five valid tests.

2.3.6 Contact angle

The water contact angles were determined using an Optical Tensiometer Theta Flex (Biolin Scientific, Manchester, UK) using the sessile drop technique. A water droplet of 3.0 µL volume was applied over the sample surfaces, and the images were captured with an image resolution of 1984 × 1264 pixels, 3009 fps maximum measuring speed and ± 0.1º accuracy, in a controlled temperature room (23 ºC). To limit the influence of highly hydrophilic characteristics of cellulose and glycerol, two different times were used to acquire images of droplets (1 s and 5 s after dropping), and respective profiles were analyzed using the Young-Laplace equation. All measurements were performed in triplicate.

2.3.7 Light barrier properties and transparency

The light barrier properties of films were analyzed in a V-530 spectrophotometer (Jasco Inc, Japan) using transmittance mode in UV-Vis range (800 –250 nm) at 1 nm intervals. The transparency of films was evaluated following the equation (Han and Floros 1997; Guo et al. 2014; Cazón et al. 2018).

$$\:\text{T}\text{r}\text{a}\text{n}\text{s}\text{p}\text{a}\text{r}\text{e}\text{n}\text{c}\text{y}=\:\frac{\text{log}{\text{T}}_{\text{x}}}{\text{l}}$$

where \(\:\text{log}{\text{T}}_{\text{x}}\) is the percent transmittance at x nm, and 1 is the film thickness (mm).

2.3.8 Oxygen and water vapor permeabilities

The oxygen transmission rates (OTR) were measured in a Labthink C301 equipment following the ASTM 3985 standard, while the water vapor transition rates (WVTR) were measured in a Labthink C201 equipment following the ASTM F1249 standard. The oxygen permeability (OP) was calculated from the respective values of OTR, as follows:

$$\:\text{O}\text{P}=\frac{\text{O}\text{T}\text{R}\cdot\text{l}}{\varDelta\:\text{P}}$$

where l being the film thickness (mm), and ∆P partial pressure of oxygen (0.667 atm) at the measurement conditions (T = 23 ºC and RH = 0%).

The water vapor permeability (WVP) was calculated from the respective values of the WVTR, as follows:

$$\:\text{W}\text{V}\text{P}=\frac{\text{W}\text{V}\text{T}\text{R}\cdot\:\text{l}}{\text{P}\cdot\:(\varDelta\:\text{R}\text{H})}$$

where l is the film thickness (mm), P is the saturation pressure (0.0655 atm) and ∆RH is the relative humidity difference between both sides of the membrane at the measurement conditions (T = 38 ºC and RH = 90%) (Almeida et al. 2021).

3 Results and discussion

3.1 Preparation of cellulose-based films

The conversion of cellulose from a fibrous structure into a continuous/homogeneous film requires the deconstruction of the organized cellulose fibers at the molecular level (cellulose I) during the dissolution process, followed by recovering of the solid structure by regeneration of the cellulose chains into a gel state which after drying process originates a different and more stable crystalline structure (cellulose II). In this process, several parameters can affect the properties of the final cellulose films, such as the cellulose source, the cellulose DP, the solvent used for dissolution, the concentration of the cellulose solution, the regeneration method (to the gel state), the drying process, and the presence of additives (such as plasticizers). Several works have addressed these aspects, and the final materials could differ in several properties (Xiao et al. 2003; Cheng et al. 2019; Wang et al. 2019; Huang et al. 2022). In this work, a cellulose solution of NaOH/Urea/H2O (7:12:81 wt%) with a mass concentration of 6% (w/v) was used to produce films. This cellulose concentration was used because below this concentration the viscosity of the solution did not allow the spreading without spillage. Cellulose regeneration is usually achieved by film immersion in anti-solvents. Such approaches were attempted in this work (for instance, immersion in distilled water, ethanol, and acetone) but they all resulted in weak and brittle gels, which did not allow the washing without damaging gel structure. For this reason, thermal regeneration using an oven at 50 °C for 2 h was attempted. This regeneration (by water evaporation) allowed the transformation of the dissolved cellulose solution into a manageable gel (≃ 0.2 mm thick) that had sufficient mechanical integrity for the washing steps. The duration of this procedure was also investigated, and 2 h was considered optimal, as the gels were too weak below this period and adhered to the glass after this time. After washing, the gels were fixed and dried slowly at room temperature. Fixation at this stage was very important because drying gels tended to shrink considerably and became brittle. The hot-pressing step was crucial to obtain good cellulose films. Several tests were carried out and 0.1 MPa and 105 ºC for 30 min proved to be the best conditions. The addition of glycerol by soaking followed by the hot-pressing results in compact, transparent, and easily foldable films (Fig. 2 and video in online resources).

Fig. 2
figure 2

Photos of cellulose/glycerol films rolled and foldable without losing its integrity

Full size image

3.2 Glycerol content in films

During soaking, glycerol molecules occupy the empty spaces between the cellulose chains, and during hot-pressing the water molecules leave the structure while the glycerol remains in the cellulose matrix, resulting in a plasticizing effect (Van Nguyen and Lee 2021). To understand how different amounts of plasticizer could affect cellulose film properties, the glycerol content in films was evaluated by two different methods: (1) gravimetry of dried films before and after immersion in glycerol/water mixtures and (2) by TGA via weight loss in the range 125–290 °C (Ribeiro et al. 2021; Rumi et al. 2021). Table 1 lists the glycerol contents of the prepared films evaluated by the two methods.

Table 1 Glycerol content in cellulose films calculated by gravimetry and TGA analysis (see Thermal properties)
Full size table

The results obtained with both techniques were very consistent as shown in Table 1. Since the soaking time (30 min) was the same for all glycerol concentrations tested, it can be concluded that the glycerol content of the films increased with the glycerol concentration in the soaking solution. Interestingly, the glycerol concentration of the soaking bath was close to the final glycerol content of the films, showing that it was a fast process, except in the case of Cel70. When the glycerol concentration of the soaking bath was increased to 70% (w/w), the glycerol content of the films did not follow the same pattern and leveled off at around 50% (w/w), indicating that these conditions were close to the maximum absorption capacity of the films or, in other words, close to the complete filling of the empty spaces between the cellulose chains. In general, the use of glycerol or plasticizers is described in the literature, but in most cases their content is not quantified (Xiao et al. 2003) so, unfortunately, it was not possible to compare our results with those of other authors.

3.3 Fourier transform infrared spectroscopy

FTIR analysis is a useful analytical technique for studying the chemical composition and interactions between glycerol and cellulose in films. The FTIR spectra of the cellulose films with and without glycerol, are shown in Fig. 3.

Fig. 3
figure 3

FTIR spectra of Cel0, Cel10, Cel20, Cel40 and Cel70, from top to bottom, respectively

Full size image

All the spectra were very similar, and the main events observed were the differences in relative intensities and some small shifts. This could mean that the presence of glycerol caused limited changes of the chemical environment, particularly the degree of -OH interactions. The broad band between 3700 and 3000 cm− 1 was attributed to the stretching vibrations of the -OH group, characteristic of cellulose-based films (Ribeiro et al. 2021; Rebelo et al. 2023). The apparent increase in absorbance and sharpness of this band with increasing glycerol content may be due to the contribution of the -OH groups of glycerol. The wavenumber corresponding to the maximum of this band also experienced a small shift, suggesting the existence of hydrogen bonding between glycerol and cellulose chains (Xiao et al. 2003). The band at 2877 cm− 1 was attributed to the stretching vibrations of -CH, which became more intense and clearly developed into a double peak in films with higher glycerol content (Xiao et al. 2003). The -OH bending of absorbed water was associated with the band at 1643 cm− 1 (Fan et al. 2012), and its relative intensity increased with increasing glycerol content (Cazón et al. 2018). The characteristic peaks of cellulose at 1027 cm− 1 corresponded to C-O-C anhydroglucose ring stretching vibrations, and the 896 cm− 1 band corresponded to the glycosidic -CH deformations with ring vibration and O-H bending contributions, which are characteristic of regenerated cellulose films (Liu et al. 2011). The appearance of the vibration of the C–C skeleton of glycerol bonds at 856 cm− 1 (Vahur et al. 2016; Basiak et al. 2018) indicated the presence of a higher amount of glycerol in the film composition as the peak intensity increased with higher glycerol concentrations.

3.4 Scanning electronic microscopy

SEM images were used to study cellulose films morphology and Fig. 4 shows the surface and cross-section views of the films.

Fig. 4
figure 4

SEM images of top view (left) and cross-section (right) perspectives of cellulose/glycerol films

Full size image

The presence of glycerol did not affect the morphology of the plasticized films, comparably to Cel0. The surface morphology was homogenous in all prepared films. Dense and homogenous internal structures without cracks, holes, or pores could be observed. The small agglomerates observed on top views could be related to residual fiber fragments (Rumi et al. 2021). The dense and compact internal structures were confirmed in cross-section views, which showed no evidence of the original fibrous structures. These results suggest that the films were formed from a new supramolecular rearrangement resulting from the complete dissolution and thermal regeneration of the cellulose fibers (Rebelo et al. 2023). Overall, glycerol had no effect on the morphology of the films, therefore, they all looked alike in SEM images.

3.5 Thermal properties

The thermal stability and thermal behavior of cellulose-based films were studied by TGA and DSC, respectively. Figure 5 shows the weight loss (A) and the corresponding derivative (DTG) (B), and the derived calculated values are presented in Table 2.

Table 2 Thermal properties and crystallinity index (CrI) of cellulose-based films
Full size table

All cellulose films exhibited a small dehydration step below 125 ºC (corresponding to less than 5 wt%), and a degradation step above 300 ºC attributed to cellulose degradation. The films with glycerol revealed one additional mass loss step associated with glycerol degradation (125–290 ºC) (Dorris and Gray 2012).

Fig. 5
figure 5

Weight loss curve of cellulose-based films (A) and the respective derivative of weight loss with respect to time (DTG) (B)

Full size image

Figure 5 shows that the presence of glycerol did not change the profile of cellulose degradation, and only a slight shift in the maximum degradation temperature was observed. As the amount of glycerol in the films increased, the corresponding mass loss levels increased accordingly. This was evidenced by the increase in the intensity of the peaks between 125 and 290 ºC (Fig. 5B) and by the decrease in the intensity of the peaks corresponding to the cellulose degradation. In addition, the DTG peaks of glycerol shifted to higher temperatures closer to the boiling point of the plasticizer. Based on the mass losses, the glycerol content of the films was calculated, as mentioned above. The residual mass of Cel0 was about 23%, which is likely due to the presence of inorganic salts as a result of incomplete washing or the presence of small amounts of lignin in the industrial kraft pulp used. It should be noted that lignin decomposes in a broader range of temperature (280–800 ºC) with more intense decomposition between 350 and 650 ºC (Brebu and Vasile 2010; Zhao et al. 2014). Overall, the residual weights decreased as the glycerol contents were higher.

The thermal behavior was also considered to investigate how glycerol interacts with the cellulose in the films (Fig. 6). Cel0 showed no thermal transitions, as expected (Rebelo et al. 2023). Cel10 and Cel20 also showed no thermal transition, in contrast to Cel40 and Cel70. The small amounts of glycerol (in the Cel10 and Cel20 films) were not sufficient to affect the cellulose structure or their own transitions. With increasing glycerol content, a glass transition appeared around − 70 ºC which has been attributed to the Tg of glycerol (Averous et al. 2000; Arık Kibar and Us 2013), and at a slightly lower temperature in Cel70 due to the higher glycerol content.

Fig. 6
figure 6

DSC thermograms of cellulose films Cel0, Cel10, Cel20, Cel40 and Cel70, from top to the bottom, respectively (second heat flux)

Full size image

3.6 Crystallinity (XRD)

XRD analyses were performed to investigate the influence of glycerol on the crystallinity of the films, and the resulting patterns are shown in Fig. 7. In contrast to the original cellulose pulp, which has shown to be cellulose I (Rebelo et al. 2023), the regenerated cellulose films exhibited the crystalline form of cellulose II with characteristic diffraction peaks around 2θ = 12, 20 and 22°, attributed to the crystallographic planes (1–10), (110), and (020), respectively (Chiriac et al. 2014; French 2014).

Fig. 7
figure 7

The XRD patterns of cellulose films Cel0, Cel10, Cel20, Cel40 and Cel70, from top to the bottom, respectively

Full size image

The calculated crystallinity indexes are listed in Table 2. The crystallinity of all cellulose films was very similar, which, was not expected a priori since the addition of plasticizers can make the cellulose molecules more ordered as previously reported by Xiao et al. (2003). Crystallinity indexes varied within a small range (51–55%), probably since the reorganization of cellulose molecules was achieved by thermal regeneration, rather than by soaking with glycerol solutions and hot drying. Nevertheless, the resulting cellulose/glycerol films exhibited similar crystallinities to the films regenerated in water baths at similar temperatures (Cheng et al. 2019).

3.7 Mechanical properties

Tensile tests were performed in samples stabilized at 0% relative humidity (RH) (dried samples), as well as in samples stabilized at RH = 50%, and the results are shown in Fig. 8.

Fig. 8
figure 8

Tensile tests: dried samples (black bars) and samples at RH = 50% (red bars)

Full size image

Cellulose has a high tendency to absorb water (which is related to the high hydroxyl content), making the structures fragile by destroying the inter-bonding between the cellulose chains. It was interesting to observe the effects of water vapor on the mechanical properties of cellulose films. The effect of water vapor was more pronounced in the samples with glycerol (Cel10, Cel20, Cel40 and Cel70) than in the samples without glycerol (Cel0), as shown in Fig. 8. Interestingly, this sample showed no significant mechanical changes in the presence of moisture (RH = 50%), which could also be related to its lower permeability to water vapor (see Sect. 3.10).

The addition of glycerol drastically reduced the tensile strength and stiffness of the films. Cel70 exhibited eight-times less tension than the film without glycerol (Cel0). On the other hand, a higher glycerol content led to films with improved strain at break (SaB). Accordingly, the stiffness of the plasticized films decreased as the glycerol content of the films was increased. This general behavior was expected due to the presence of the plasticizer, as observed by other authors (Hidayati et al. 2021). The observed results could be due to the formation of intermolecular hydrogen bonds between glycerol and cellulose, which weakened the hydrogen inter-bonds between the cellulose chains, and this process could have been favored by hot-pressing. This weakening of the chain-chain interaction allowed the cellulose chains to move more freely, resulting in very malleable films especially those with higher glycerol content (Xiao et al. 2003; Van Nguyen and Lee 2021).

The plasticizing effect of moisture in cellulose films has already been reported (Wei et al. 2019; Cazón et al. 2020b). Thus, when the films were conditioned at RH = 50%, some water molecules could have been absorbed by the films, in addition to the plasticization by glycerol, and act as plasticizers. It was expected that the presence of glycerol would increase water absorption (Kumar et al. 2014; Cazón et al. 2020a). However, the synergistic effect of plasticization by water was only pronounced for films with glycerol. In the case of Cel0, the results obtained in dried (RH = 0%) and RH = 50% were very similar. Besides the absolute value of tensile strength (TS) was slightly higher at RH = 50% than at RH = 0%, such value was within the standard deviation (SD) of the dried samples. The dried films had a higher SD compared to the conditioned ones, which could be related to water absorption from the air during sample preparation and mechanical testing. This effect was not pronounced in samples with RH = 50% as they were conditioned at a relative humidity close to that of the ambient air. On the other hand, Cel70 had the highest glycerol content and the effect of plasticizing with glycerol was more pronounced than the effect of water. This means that the glycerol effect overlapped the water effect and probably it is near its maximum (as seen in Sect. 3.2). Therefore, it seems that glycerol was near its maximum capacity to fill the empty space between cellulose chains. The water vapor permeability of the films increased with the glycerol content (see Sect. 3.10). Consequently, the combination of these two plasticizers (water and glycerol) had strong effects on the tensile properties of the films by drastically decreasing the tensile strength and Young’ modulus and increasing the elongation.

3.8 Transparency and light barrier properties

The light transmittance of the cellulose films prepared in this work is presented in Fig. 9.

Fig. 9
figure 9

UV-Vis spectra of cellulose/glycerol films, from 250 to 800 nm

Full size image

The transmittances values observed in this work are similar to those previously reported by other authors (Cheng et al. 2019). It has already been observed that the incorporation of glycerol in cellulose films leads to a decrease in films transmittance (Cazón et al. 2018). The presence of glycerol decreased the transmittance of the films by 5–10%, compared to the films without glycerol (Cel0). The transmittance at 250 nm was less than 10% for the films plasticized with glycerol. It was interesting to note that the transmittance profile of Cel70 was considerably different than the profiles of the other films. Gonçalves and co-workers investigated the effects of incorporating glycerol into cellulose acetate films and found that there were only significant differences when the glycerol concentration in the films was around 50%. Considering the glycerol quantification shown in Table 1, the Cel70 film had a similar glycerol content, so this transmittance result is consistent with the previously reported result (Gonçalves et al. 2019).

The literature reports that transmittance is affected by the moisture content, therefore, values should decrease significantly in dried or full swollen films (Cazón et al. 2020b). The low transmittances observed may indicate a UV light semi-protective effect caused by the glycerol added to the cellulose films. Additional UV light protection can be achieved by adding other compounds to the film formulation, as those traditionally used in the plastic industry (Cherif Lahimer et al. 2017), or natural-based anti-UV additives (Zhang et al. 2022b; Sun et al. 2023).

Transparency is an important property in packaging applications. The transparency of films was also determined, as it depends on the film thickness and a higher transmittance does not necessarily corresponds to higher transparency (Elsabee and Abdou 2013). The wavelength used for the calculation was 550 nm, which corresponds to the average wavelength of the visible light spectrum and one of the most intense irradiations of the sunlight spectrum. The results are shown in Table 3.

Table 3 Transmittance and corresponding transparency at 550 nm for cellulose-based films
Full size table

The results presented in Table 3 are within the range of the polymers most used in packaging films, such as polyethylene (PE) and polypropylene (PP). Considering that the transparency of low-density PE film ranges is 15–20 and PP films is 35–40 (Han and Floros 1997; Lee et al. 2008), the cellulose-based films prepared in this work were within the range of transparencies of synthetic polymers most used in packaging, which is a promising property.

3.9 Contact angle

Water contact angle (WCA) is a common measurement used to study the hydrophilicity/hydrophobicity of packaging films. The water contact angles of the films prepared in this work are shown in Fig. 10.

Fig. 10
figure 10

Water contact angle of cellulose/glycerol films, after 1s (black) and 5s (red) of droplets drop

Full size image

The contact angles measured after 1 s and 5 s were remarkably similar, indicating some slowing down in water absorption process. The smooth surfaces and dense structures observed by SEM could explain this fact. For the cellulose sample (Cel0) the observed WCAs were 48.6 ± 1.2° and 49.1 ± 2.0° at 1 and 5 s, respectively, suggesting a hydrophilic surface. These values were unexpected and higher than those previously reported by other authors, namely, 22° (De Bon et al. 2022), 24° (Rebelo et al. 2023), 32° (Cheng et al. 2019), and 39° (Van Nguyen and Lee 2021). The contact angle depends on the surface properties, structure density and organization (Van Nguyen and Lee 2021). The high contact angles of the control samples were attributed to the hot-pressing used in this work, which resulted in a denser and smother surface, compared to the other mentioned works where the films were only air dried.

The WCAs at 1s and 5s were quite similar for all formulations, demonstrating the low water absorption of cellulose films. At lower glycerol concentrations (Cel10 and Cel20), the effect was negligible, and no substantial differences were observed. On the other hand, the effect was remarkable at higher glycerol concentrations (Cel40 and Cel70), for which the WCAs increased with the glycerol content. The WCA of Cel70 was almost two-fold higher (88.0°) in comparison with the blanks (Cel0) and films with low glycerol concentrations (Cel10 and Cel20). On the other hand, the WCA of the Cel70 approached the typical hydrophobic behavior (> 90º). The WCA depends on the availability of hydroxyl groups on the surface of the films, and both cellulose and glycerol can form hydrogen bonds with water and with each other as well (Van Nguyen and Lee 2021). The results observed in higher glycerol content samples suggested that the films prepared in this work had higher amount of glycerol molecules at surface, which created a “coating” layer on the film surface, preventing the formation of hydrogen bonds between cellulose and water.

3.10 Permeability to oxygen and water vapor

A particularly important characteristic for packaging applications is the water vapor and oxygen permeability of the films. Therefore, it is of outmost importance to use adequate equations for the calculation of permeabilities as some confusion of concepts has been observed, even in textbooks (Cooksey et al. 1999). The results of oxygen and water vapor permeabilities of Cel0, Cel20 and Cel40 films are presented in Table 4. Cel20 and Cel40 were selected because they were shown to have the most promising properties for packaging applications, while Cel0 was used as control.

Table 4 Oxygen permeability (OP) and water vapor permeability (WVP) of Cel0, Cel20 and Cel40
Full size table

In the literature, the reported permeabilities of cellulose films vary widely, and depend on the way the films were prepared. For instance, OP values of 0.0003–0.148 cm3⋅mm⋅m− 2⋅d− 1⋅atm− 1 and WVP values of 110–739 g⋅mm⋅m− 2⋅d− 1⋅atm− 1 have been reported (Bras et al. 2007; Yang et al. 2011; Ghasemlou et al. 2013; Paunonen 2013; McKeen 2017; Su et al. 2022). Unfortunately, it was not possible to measure the OTR of the films without glycerol (Cel0) because they have blown-up repeatedly during the measurements, probably due to an excessive drying of the material caused by the passing of gases during the measurement. The OP decreased by 13-fold when the glycerol concentration was doubled (Table 1). These results may be attributed to the low solubility of oxygen in glycerol (Kutsche et al. 1984), which resulted in a much lower permeation of this gas through the plasticized films. Hence, the molecules of glycerol simultaneously interacted with cellulose (plasticization effect) and formed a barrier to oxygen molecules.

On the other hand, water vapor permeability was higher in the films that have higher glycerol content, which may be attributed to the hygroscopic properties of both cellulose and glycerol. As expected, the sample with high glycerol content was the most permeable to water vapor.

Polyethylene (PE) and polypropylene (PP) are the most commonly plastics used in packaging applications. PE films have OPs of 50–200 cm3⋅mm⋅m− 2⋅d− 1⋅atm− 1 and WVPs of 0.5–2 g⋅mm⋅m− 2⋅d− 1⋅atm− 1 (Domininghaus 1993; Abbott 2017; Okchem 2020), while PP films have OPs of 50–100 cm3⋅mm⋅m− 2⋅d− 1⋅atm− 1 and WVPs of 0.2–0.4 g⋅mm⋅m− 2⋅d− 1⋅atm− 1 (Domininghaus 1993; Abbott 2017; Okchem 2020; Zhang et al. 2022a). Comparing these values of PE and PP with those in Table 4, it can be said that the films prepared in this work could be more effective in protecting packaged goods from oxygen but would perform worst in the case of water vapor. However, depending on the packaging application, this characteristic is not necessarily a disadvantage. For example, when packaging fresh food it is important to avoid dehydration (Siracusa 2012) whereas bakery products benefit from a high water vapor permeability to avoid the development and growth of fungi. Also, the low oxygen permeability is a plus in such application as it protects the bakery products from lipid oxidation and rancidity (Kim and Seo 2018; Qian et al. 2021; Benitez et al. 2024). Therefore, the films prepared in this research could be a promising alternative for use in some packaging applications.

4 Conclusion

In this work, a thermal regeneration method for cellulose solutions was attempted for the first time. This method has the advantage of facilitating cellulose regeneration, because if the cellulose solution does not possess the suitable viscosity, the formation of cellulose gel in anti-solvent systems can be extremely difficult, and it can prove extremely challenging to obtain large films. The cellulose films prepared in this work were plasticized by immersion in glycerol solutions of various concentrations and dried by hot-pressing. Glycerol acted as a plasticizer, improving the flexibility of the films, and its content was quantified by gravimetry and TGA analysis. The films had different glycerol contents which led to different mechanical properties. The elongation at break increased and the tensile strength decreased with increasing plasticizer content. In general, the films exhibited dense and compact structures, good transparencies and their surfaces became more hydrophobic with the increase in the glycerol content. The presence of glycerol decreases the oxygen permeability but increased water vapor permeability. Regarding all the presented results, the films prepared in this work showed similar performances compared to those reported in the literature, proving that thermal regeneration can be used effectively when an anti-solvent system is problematic or undesired.