A simple and low-cost strategy to develop antibacterial composite film of nanocellulose and stilbite zeolite particles

Abstract

Herein, we report a detailed preparation method for a low-cost antibacterial film made from sustainable materials- cellulose nanofibers (CNF) and stilbite zeolite using a simple solvent casting technique. The preparation process involved a series of steps: isolation of cellulose from Teff straw, purification and ultra-sonication of cellulose to CNF, purification of stilbite, and the fabrication of CNF/PVA/stilbite composite films. The results indicated that the carefully optimized inclusion and processing of CNF and stilbite particles led to improvements in the mechanical strength, thermal resistance, morphological properties and antibacterial activity of the nanocomposite films when compared to bare PVA film. Besides, the CNF networks maintained homogenous blend of the components and uniformity of the subsequent composite films. Interestingly, the integration of stilbite (STI) into the composite films led to high antibacterial activity against both gram-positive and gram-negative bacterial strains, with clear inhibition zones of up to 12.33 mm for E. coli and 13.30 mm for S. aureus. The activities are higher or comparable to commercial Amoxicillin antibiotics that demonstrated zones of inhibition of 13.0 and 9.8 mm against E. coli and S. aureus, respectively. The unusual antibacterial property of STI particles could be ascribed to the presence of free metal ions in STI that can form metal hydroxides in the interfacial structures between STI particles and CNF/STI polymer networks. The rough morphology of the composite film due to the inorganic clay fillers can contribute to reduce biofilm formation, which is essential in water treatment and other disinfection applications. These low cost cellulose-scaffolded clay composite film hold promise as a well-performing sustainable material alternative for various applications like food packaging, surface disinfection coatings, and water treatment membranes.

Introduction

Microbes pose growing concerns in the health care¹, water treatment², food³, and textile industries⁴. One major challenge is the control of the proliferation of many micro-organisms as they develop resistance to most commercial antibiotics, potentially posing increased threat to both human health and the ecological environment⁵. Establishing safe, low-cost and an efficient means to control pathogenic microbes, has therefore become a primary challenge among the scientific community⁶. On the other hand, the emergence of nanotechnology and the fabrication of various nano-assisted material architectures has offered promising alternatives in developing efficient antimicrobial systems⁷. Thin films are suitable platforms to design drug delivery systems, allowing the integration and immobilization of antimicrobial agents with substrates for controlled active agent release⁸. Such systems are capable of enhancing drug efficacy, target drug delivery and the onset of drug action, while minimizing microbial resistance and side effect to the host⁹. Various classes of antimicrobial thin film materials, such as metallic, metal oxide and clay-based nanocomposites, have been investigated either to kill or inhibit the growth of bacteria⁸. Nanocomposite films offer the integration of different microbicidal agents for improved or synergetic efficacy. However, the scale up and wide range application of antimicrobial materials is still limited, requiring the search for versatile and affordable nanocomposite films.

As the most abundant, biocompatible, biodegradable and renewable source, cellulose is a versatile alternative film forming polymer material with desired mechanical, thermal and optical properties. It is widely used in many diverse applications such as paper making, food packaging, optically transparent films, coatings, membranes for water treatment, textiles and medical supplies^10–12. Despite these attractive characteristics, cellulose has some major drawbacks like high water absorption and lack of antibacterial action^12,13, limiting its range of applications. Although the issues that limits its utility can be solved to some extent by chemical modification of the cellulose fibers, recent research is focusing on the conversion of cellulose into nanocellulose and incorporating bactericidal agents for high throughput antimicrobial applications¹⁴. Amongst the commonly reported antimicrobial agents that has been incorporated into cellulose and nanocellulose matrix include metallic nanoparticles (NPs) such as Ag, Au and Cu, metal oxide (CuO, TiO₂, etc.) NPs and metal sulfide (MgS, MnS, etc.) NPs^8,11. However, the use of such NPs raises safety concerns, especially in applications such as water disinfection and food packaging mainly because their ultimate ecological fate is not fully understood¹⁵.

Natural and abundant clay materials have been found to be safe and are therefore promoted as viable alternatives for large scale applications¹⁶. They also possess important characteristics like mechanical strength, high surface area, absorption ability, cation exchange capacity, non-toxicity and environmental friendliness. Stilbite (STI), a natural zeolite mineral abundant in the rift valley regions of Ethiopia, is reported to have good adsorptive, cation exchange and charge transfer properties^17,18, which could potentially impart antibacterial properties into cellulosic films. STI is a hydrated aluminosilicate mineral, with Si: Al ratio ~ 3, that also contains Na and Ca ions as extra-framework cations¹⁸. Gómez-Hortigüela et al. demonstrated that the Ethiopian stilbite possessed water defluoridation abilities¹⁷. In addition, thick composite films of Ca-STI and Mg-STI are reported to have ethanol sensing properties¹⁹. However, to the best of our knowledge, there is no report on the investigation of stilbite’s antimicrobial properties, nor its incorporation into cellulose nanofiber-based composite films.

The above motivated the present work on the incorporation of stilbite into cellulose nanofiber (CNF)/polyvinyl alcohol (PVA) scaffold and its antibacterial study. It is envisaged that the outcomes will serve as a benchmark for developing tailored antimicrobial thin films for diverse applications such as in food packaging, wound dressing and water disinfection. Herein we report the fabrication of a low cost cellulose-based antibacterial composite film incorporating naturally abundant STI using a simple solvent-casting method to broaden the applications of cellulose.

Experimental

Chemicals and materials

Teff strew (Eragrostis teff), the source biomass for the preparation of cellulose, was collected from Kilinto area farms, Addis Ababa. Polyvinyl alcohol (PVA), Sodium hypochlorite (NaOCl, 5.0%) and Ethanol (99.4%) were purchased from Fine Chemicals Inc., Germany. NaOH from Alpha Chemika, India; HCl (35–37%), Toluene (99.5%) and CH₃COOH (99.8%) from Pentokey Organy, India. All the reagents used in this work were of analytical grade.

Biomass preparation

Teff straw was first washed with distilled water, followed by sun drying for 5 days and oven drying for 6 h at 60℃. The straw sample was then grinded and sieved to obtain 350 μm mesh size particles.

Proximate analysis of Teff straw

Oven drying method was used to determine moisture content by heating the Teff straw at a temperature of 103℃ for 2 h. Then it was calculated using Eq. 1²⁰.

1

where and refers to initial (before drying) and final mass (after drying), respectively.

To determine ash content of the Teff straw, 3 g of dried straw was placed in a pre-weighed crucible which was incinerated in a furnace at 600℃ for 2 h. When complete ash was achieved, the crucible was transferred into a desiccator for cooling. The ash content was calculated using Eq. 2²⁰.

2

where Wa refers to the weight after complete ashing + crucible, Wb refers to the weight before ashing + crucible, and Wc refers to the weight of the empty crucible.

Compositional analysis of Teff straw

Teff straw (10 g) was taken, mixed with 200 ml of distilled water, and the mixture was kept in a water bath at 80℃ for 3 h, followed by cooling and vacuum filtration. The solid sample was then oven dried at 105℃ until a constant weight. To determine the ethanol extractives, 5% ethanol was prepared and the water extractive free sample (5 g) was added and mixed. Then, the same treatment like that of the water extractive sample was used. The amounts of total extractives were calculated using Eq. 3²¹.

3

where Mi is initial mass and Mf is final mass of sample.

Hemicellulose content

To the dried extractive free sample (1 g), 0.5 M NaOH solution (10 ml) was added, and the mixture was kept in a water bath at 80℃ for 3 h. Thereafter, the solid residue was washed with distilled water and filtered until neutral pH was achieved. After drying the sample at 105℃, the content was calculated using Eq. 4²².

4

where Wi is the initial weight and Wf is the final weight.

Lignin content

Hydrochloric acid (6 M, 20 ml) was added drop-wise with constant stirring to the extractive sample (2 g). This mixture was then allowed to stand overnight at room temperature. It was subsequently transferred to a round flask, diluted to 3% HCl, and boiled for 4 h. The lignin was filtered on a pre-weighed filter paper and washed with hot distilled water until neutral pH. The lignin was dried at 105℃ for 6 h and its yield estimated using Eq. 5²³.

5

where Wd is the weight of the dried sample and Wi is the initial weight.

Cellulose content

The cellulose content was determined by considering the other constituents of lignocellulose, as a difference of the sum of extractives, hemicellulose, and lignin from hundred (Eq. 6).

6

Fabrication of CNF from Teff straw

Isolation and purification of cellulose

Purified cellulose was obtained by removing the non-cellulosic components in Teff straw powder²⁴. Dewaxing of straw sample (5 g) was carried out in Soxhlet extractor using a 2:1 (v/v) mixture of toluene/ethanol for 7 h. Afterwards, bleaching was performed to remove lignin from the samples using an acidified NaOCl, with pH adjusted to 3.0–4.0 using CH₃COOH, at 75 for 1 h. The process was repeated six times, resulting in holocellulose, which was then treated with KOH (2 wt%) at 90 for 2 h to remove hemicelluloses, residual pectin and starch. The samples were washed after each treatment with distilled water. Purified cellulose powder was obtained by lyophilizing in a vacuum freeze drier²⁵. The yield of cellulose was determined as the ratio of weight of cellulose obtained to the weight of Teff straw used.

Preparation and characterization of CNF

Cellulose (5 g) was hydrolyzed with a 6 M HCl solution under continuous mechanical stirring (500 rpm) at 35 ℃ for 30 min. The ratio of cellulose to HCl was kept at 1:25 g/ml. To stop the hydrolysis process, distilled water (2500 ml) was added to the reaction mixture. Then the suspension was washed repeatedly with distilled water until the pH of the wash liquid became neutral. After that it was sonicated at 15 kHz for 45 min to obtain uniformly dispersed cellulose nanofiber suspension, avoiding agglomeration of the nanocellulose particles. Finally, the yield of CNF was calculated by dividing the weight of CNF obtained with the weight of cellulose used²⁶. The dimension and structural analysis of the prepared CNF were determined using DLS (Malvern, MAL 1149420) and FTIR (Thermo scientific iS50 ABX, USA), respectively.

Preparation of CNF/PVA/STI nanocomposite films

The CNF/PVA/STI nanocomposite film was prepared by a simple casting method²⁷. In brief, the CNF, STI and PVA were blended by varying the loading of STI zeolite while keeping the percentage of both PVA and CNF constant at 2% (w/v) in solution. Aqueous dispersions of 3%, 2%, and 1% STI zeolite were prepared, and mixed with previously prepared 2% (w/v) each of CNF/PVA dispersion, mixed homogenously for 2 h in distilled water and stirred until a uniform suspension obtained. Finally, the mixture of CNF, PVA and STI zeolite was drop casted on a Petri dish. The resulted film was then cured at room temperature for 24 h, followed by oven drying at 50 °C. Similarly, CNF, PVA, and CNF/PVA films were prepared under the same conditions, and used for comparative studies.

Characterization of thin films

Surface morphology of the samples was observed under Scanning Electron Microscope (SEM, COXEM CX-200Plus) and 3D-optical microscope (Malvern Morphology^® G3S), while their thickness was measured using electronic digital Pointed-Jaw-caliper. Light absorbance and transmittance of the films was evalusted by UV/Vis spectrophotometer (V-770, JASCO International Co. Ltd., Japan). Functional group analysis was conducted by Fourier Transform Infrared Spectroscopy (FTIR, Nicolet^Tm IS50). The spectra were taken in 4000–400 cm^− 1 wave numbers. Thermal properties were studied by Thermogravimetric Analyzer (TGA, PerkinElmer Diamond TG/DTA) and Differential Scanning Calorimeter (DSC; TA Instruments, USA). The tensile strength and elongation at-break of the films were also studied by Universal Testing Machine (UTM, 766 tensile testers).

Antibacterial test

The antimicrobial activity of the prepared membranes was evaluated by an inhibition zone assay method against two different pathogenic microbes as test strains. Escherichia coli (E. coli) was used as a gram-negative bacteria and Staphylococcus aureus (S. aureus) as gram-positive bacteria. The agar diffusion assay was prepared by autoclaving 25 mL of 38 g/L solution of Mueller Hinton Agar at 121 °C for 15 min, followed by cooling to 45 °C. Using a sterile pipette, 1000 µL of the bacterial broth culture of the test organism was inoculated with 10 mL aliquot of nutrient broth, spread evenly with a sterile spreader onto sterile petri dishes to get a bacterial lawn, and incubated at 37 °C for 24 h. The diameters of the inhibition zones were measured (in mm) on the inside of the petri dish. The inhibition zone diameters were measured using millimeter in comparison to inhibition diameters of amoxicillin and reported as mean ± standard deviations²⁸. Pure CNF film was used as a control.

Result and discussion

Isolation of cellulose from Teff straw

The cellulose employed for nanocomposite film development was extracted from Teff straw using alkali and sodium hypochlorite treatments as shown in Fig. 1. The extraction process removed hemicellulose, lignin and other impurities with yield of cellulose and nanocellulose being 75.25% and 37.7%, respectively. The yield of both cellulose and nanocellulose were comparable with previous studies; for example, Bacha et al. reported a yield of 78.7% for cellulose and 42.8% for nanocellulose fibers²⁶.

Fig. 1.

Schematic diagram of the isolation and purification of cellulose from bamboo.

To determine whether a particular lignocellulosic material is suitable source of cellulose, it is crucial to analyze its composition. The nanocellulose yield of the finished product is significantly influenced by the straw’s cellulose, hemicellulose, and lignin contents. Teff straw has comparable/higher cellulose contents than common biomasses used for similar applications (Table 1).

Table 1.

Composition analysis of Teff straw, and its comparison with common biomasses.

Biomass source	Average value (%)						Reference
	Extractives	Cellulose	Hemicellulose	Lignin	Moisture content	Ash content
Teff straw	7.32	49.00	23.00	28.00	7.96	6.03	This study
Teff straw	8.5	36.5	29.5	17.5	6.4	5.1	29
Teff straw	–	31.56	29.80	22.67	8.76	7.23	30
Rice straw	3.22	35	24.3	17.73	–	–	31
Sugarcane bagasse	3.37	40.84	30.79	25	–	–	32
Coffee husk	–	33.25	23.9	25.07	8.45	6.20	30
Corn cob	–	41.15	32.02	20.11	8.64	2.79	30
Sorghum stalk	–	36.42	30.1	19.85	6.40	3.94	30

Preparation and characterization of cellulose nanofiber

Cellulose nanofiber was prepared by ultrasonication of the purified cellulose, with a yield of 41.16%. The 2D and 3D-optical surface profile (with magnification of 20x) of cellulose extract and subsequently prepared CNFs are shown in Fig. 2. Figure 2a1 and a2 show alkaline hydrogen peroxide (APH) treated fiber bundles, whereas Fig. 2b1 and b2 illustrate acid hydrolyzed fiber, which has less size than AHP treated fiber. More in-depth observation under SEM (Fig. 7) further confirmed the successful preparation and appearance of the CNFs.

Fig. 2.

2D and 3D optical images of cellulose (a₁ and a₂) and CNF (b₁ and b₂).

Fig. 7.

3D Optical microscope images (a-d) of Pristine PVA (a), C2P (b), C4P (c) and C6P (d), and SEM images of CNF (e), C2P (f) and CP-STI2 (g).

The average particle size, size distribution and polydispersity index (PDI) of the CNF suspension were also investigated using DLS (Fig. 3). The result indicated that over 99% (by volume) of the fibrous materials was under 10 nm diameter. The PDI value, a measure of size uniformity of a material, of the suspended fibers was 1, which is higher than 0.1 and indicated that there is less uniformity in the obtained nanocellulose fibers.

Fig. 3.

Particle size distribution (A) and Z-Average particle size (B) obtained from DLS measurement.

Furthermore, the structural and compositional investigation confirmed the formation of cellulose and CNF. For instance, the FTIR spectra (Fig. 4) verified the disappearance of key functional groups during the isolation and purification of cellulose and nanocellulose. The peak at 1736 cm^− 1 due to C = O stretching vibration of COOH groups in hemicellulose and lignin; the peaks at 1505 cm^− 1 (associated to C = C stretching vibration of aromatic ring) and 1234 (from C-O stretching vibration of aryl group) of lignin revealed the removal of lignin and hemicellulose during Teff straw processing. Similar findings were reported by Wulandari et al. for the removal of lignin and hemicellulose from sugarcane bagasse³³. On the other hand, the existence of key functional groups in the FTIR peaks of nanocellulose, cellulose and Teff straw demonstrated that the chemical and ultrasonic treatments did not damage the cellulosic units. The band around 3340 cm^− 1 shows the stretching of –OH groups, while that of 1636 cm^− 1 indicates C = O stretching of aldehyde group of cellulose in all the three samples. The increase in intensity of the peaks around 1013 cm^− 1 (due to –C–O–C– pyranose ring skeletal vibration of cellulose) and 3340 cm^− 1 is because of the increased cellulose composition.

Fig. 4.

FTIR spectra of Teff straw, cellulose and cellulose nanofiber.

CNF/PVA/STI nanocomposite films

Optimization of thin film composites

At first, the ratio of CNF/PVA combination in the scaffolding matrix was optimized by varying the weight% of CNF as 2%, 4% and 6%, where a 2%:2% (w/v) ratio resulted in a film with desired mechanical features like uniformity, transparency and flexibility. Then, tri-composite films of varying stilbite loading (1%, 2% and 4%) were prepared as summarized in Table 2.

Table 2.

Nanocomposite films of different compositions.

Film Composition (w/v)			Film Designation
CNF	PVA	Stilbite
–	2%	–	PVA
2%	2%	–	C2P
4%	2%	–	C4P
6%	2%	–	C6P
2%	2%	1%	CP-STI1
2%	2%	2%	CP-STI2
2%	2%	3%	CP-STI3
2%	–	2%	C-STI2

The resulting nanocomposite films prepared by casting the aforementioned compositions are presented in Fig. 5. The bare PVA film (Fig. 5a) was highly transparent, smooth and flexible, while the CNF/PVA films were partially transparent with decreasing transparency, uniformity and flexibility as the CNF loading increased (Fig. 5b and d). For example, C2P (Fig. 5b), which contained 2% each of CNF and PVA was fairly transparent with homogeneously dispersed CNFs in the PVA matrix with no evidence of agglomeration. The UV/Vis transmittance data (Fig. 6) further confirmed our observation, where PVA, C2P, CP-STI, and SP-STI2 films exhibited a decreasing transmittance of 27.08%, 16.88%, 16.093%, and 11.76% respectively. This indicates that the loading of CNF and stilbite hindered light transmission due to increased scattering and absorption. Similar observation was reported by Abdullah et al., in that halloysite nanotubes (HNT) induced a decrease in light transmittance of PVA/starch/glycerol/HNT nanocomposite films³⁴.

Fig. 5.

Digital image of nanocomposite thin films prepared from CNF, PVA and Stilbite (STI). (a) Pristine PVA, (b) C2P, (c) C4P, (d) C6P, and (e) CP-STI1, (f) CP-STI2, and (g) C-STI2.

Fig. 6.

UV/Vis transmission spectra of PVA, C2P, CP-STI, and SP-STI2 films.

On the other hand, the presence of CNF improved the mechanical behaviour of the film, making it more stable and easy to manipulate. In nanocomposite films that contained higher CNF content, i.e., C4P and C6P (Fig. 5c and d respectively), there was a decline in the dispersion of fibers, leading to defects like bumps (indicative of agglomeration) and development of cracks in the films. Furthermore, the nanocomposite film C6P containing 6% (w/v) CNF was rough and not transparent (Fig. 5d). Therefore, careful optimization of CNF content and processing condition was found to be crucial factor to maintain film uniformity and desired performance.

The incorporation of stilbite zeolite to CNF/PVA films introduced several distinct properties and advantages. For instance, the films became harder and stiffer with rough surfaces. The roughness and stiffness of the nanocomposite films increased with STI loading (Fig. 5e and f), in parallel with a decline in transmittance (Fig. 6). In order to see the effect of PVA in film formation, a CNF/STI film was prepared in the absence of PVA. The obtained rigged and non-transparent film (Fig. 5g) revealed that PVA played a pivotal role as a plasticizer for introducing flexibility to the CNF/PVA/STI nanocomposite films.

Morphological features

In addition to the digital images in Fig. 5, the optical microscope and SEM observations of the different films (Fig. 7) illustrated the variation in surface morphologies due to the nature and amount of nanofiber and zeolite loadings. Accordingly, the observations confirmed that the pure PVA film was smooth and uniform (Fig. 7a) while pure CNF forms fibrous network structures with less agglomeration (Fig. 7e), confirming the successful dispersion of CNF suspension in water. The CNF/PVA scaffolding films of 2% and 4% CNF loading formed uniform composite structures with uniform distribution of the CNF networks in the continuous PVA matrix (Fig. 7b, c and f). However, in the case of 6% CNF loading (Fig. 7d), bumps and irregular surface morphologies were observed, demonstrating non-uniform coating.

Furthermore, as shown in the SEM micrograph of CNF/PVA/STI (Fig. 7g), up on the introduction of stilbite, polygonal particles and porous structures are seen distributed throughout CNF/PVA scaffolds. The CNF networks held the zeolite particles immobilized and pores are observed between the imbedded particle surfaces and scaffolding nets, which are believed to be the origin of the porous effects. It is also seen that STI particles were imbedded evenly, and the inclusion of STI particles do not affect the network structure of the CNF scaffolds. Even though the porosity was neither optimised nor controlled, the porous features resulting from the embedding of stilbite particles would enable the nanocomposite film to be used in membrane filtration and breathable packaging applications.

Functional group analysis

The FTIR spectra of CNF, PVA, STI, CNF/PVA, and CNF/PVA/STI are shown in Fig. 8. A shift or disappearance of characteristic FTIR peaks of constituent materials or the appearances of new peaks indicate new interactions in composite material formation. Two dominant changes can be considered to illustrate this: the shift in OH stretching vibration peak present in the composite and constituents, and the change in the intensity of C = O symmetric stretching vibration. According to the result, the OH absorption peak of CNF shifted from 3320 cm^− 1 to 3317 cm^− 1 in CNF/PVA, and then to 3307 cm^− 1 upon addition of STI. This progressive shift to lower energies could indicate the increase in hydrogen bonding between CNF and PVA due to the inclusion of STI, which is in agreement with similar studies³⁵. The shorter wavelength OH absorption peak of PVA at 3290 cm^− 1 should be due to crystallinity of PVA in the film – reflecting stronger hydrogen bonding interactions. The crystallinity of PVA decreased in the presence of CNF and STI, as evidenced by the spectral shift to longer wavelengths. This could be because the nanomaterial fillers restrict the intermolecular H-bonding among PVA polymer chains.

Fig. 8.

FTIR spectra of CNF, PVA, STI, CNF/PVA and CNF/PVA/STI.

In the second case, the gradual decrease in the intensity of C = O symmetric stretching vibration due to the unhydrolyzed ester functional group present on the PVA backbone³⁶ from pure PVA film (appeared at 1730 cm^− 1) to CNF/PVA (appeared at 1727 cm^− 1) and then to CNF/PVA/STI (appeared at 1733 cm^− 1) could be associated to the alkaline-mediated hydrolysis of vinyl acetate group of PVA. The ionic interactions could change the acetate group to hydroxyl group³⁵, which weakens the absorption peak of C = O, leading to disappearance of the peak in the CNF/PVA/STI spectrum. Therefore, from the two observations it can be concluded that enhanced H-bonding between PVA and CNF exist in the CNF/PVA/STI nanocomposite film. The reduced crystallinity of PVA could indicate an even distribution of STI and CNF fillers in the composite film, which at the same time enhanced the hydrolysis of PVA, collectively contributing to the formation of a uniform stable film.

Thermal properties

Thermogravimetric analysis curves exhibiting the effects of temperature on the degradation of the CNF/PVA/STI nanocomposite films and source materials are presented in Fig. 9. All composite films were more thermally stable than pure PVA film, indicating improved stability up on composite formation with CNF and STI. According to the thermograms, weight loss of samples can be categorized into several stages. The first transition observed over the temperature range between 40 and 120 °C is ascribed to the evaporation of moisture. For PVA, the second transition that accounted for about 60% of weight loss occurred in the temperature range between 223 and 389 °C. This transition is predominately the characteristic degradation of PVA via dehydration (or elimination of water from the PVA molecules), which causes the formation of polyene intermediate³⁷. Besides, an average of about 10% mass lost during the evaporation of residual water. Degradation of PVA occurred in a first transition stage beginning near 46℃ with a maximum degradation at approximately 389℃. The data is in agreement with previous reports³⁸.

Fig. 9.

Thermo-gravimetric (a) and differential scanning calorimetry (b) curves of PVA, CNF/PVA, CP-STI1 and CP-STI2 composites.

The thermal degradation of cellulose nanofiber includes processes of dehydration and depolymerization, followed by the degradation of glycosyl units to produce a residual char^39,40. As depicted in Fig. 9, the degradation of CNF/PVA occurred in three stages. The first degradation occurred beginning near 120℃ up to 268℃, while the second stage occurs in the temperature range of 268–400℃. The final transition occurred in the temperature range of 400–550℃. These multiple stages may represent structural degradation of CNF/PVA film with the formation of CO₂, H₂O, CO and solid char as characterized by Thi’s group⁴¹.

Films containing stilbite exhibited improved thermal stability when compared to CNF/PVA films. CNF/PVA/STI films decomposed in three general stages (Fig. 9): moisture evaporation between 40 and 120 °C, structural degradation in the range 120 to approximately 360℃, and further chain scission, cyclization and molecular decomposition in the range 360–580℃, which resulted in the formation of residual solid or char³⁷.

Mechanical properties

The mechanical properties (tensile strength and elongation at break, ε) of composite films reinforced with varying compositions of CNF and STI were investigated using UTM, and results are presented in Table 3; Fig. 10.

Table 3.

Thickness and mechanical properties (TS = tensile strength, E = elongation) of CNF/PVA with different STI contents.

Types of membrane	Thickness (µm)	TS (MPa)	E (%)
CNF/PVA	290 ± 0a	15.45 ± 0.84a	4.73 ± 0.42a
CP-STI1	360 ± 90b	13.33 ± 0.95a	2.03 ± 0.74a
CP-STI2	420 ± 0a	13.56 ± 2.25b	1.86 ± 0.14a

^aindicates that values are significantly different at P < 0.05.

^bindicates no significant difference at P > 0.05.

Fig. 10.

Histogram showing the mechanical properties of pristine PVA, CNF/PVA, and CP-STI (1 and 2 wt%) composite films.

As can be seen from the data, when the content of stilbite increased from 1 to 2 wt%, tensile strength was slightly improved from 13.33 MPa to 13.56 MPa whereas elongation at break decreased from 2.03 to 1.8%. This phenomenon may be caused by coordination bonds of STI with CNF and PVA^23,42, which can improve inter-molecular forces. The increased interaction was also evidenced by the FTIR analysis (Fig. 8), where a shift to lower wavelength of OH absorption peak (from 3317 cm^− 1 in CNF/PVA to 3307 cm^− 1 in CNF/PVA/STI) was observed.

Antimicrobial properties

The antimicrobial activity of the developed composite films was evaluated against gram-negative (E. coli) and gram-positive (S. aureus) bacterial strains using disc diffusion assay, and the results are shown in Fig. 11. After the incubation period, the antibacterial activities of the films (pure PVA, CNF/PVA and CNF/PVA/STI composite films with varying stelibite loadings) were carried out by measuring zone of inhibition around the disk. Accordingly, PVA and CNF/PVA showed no activity towards both bacterial strains. Interestingly, the incorporation of stilbite to CNF/PVA brought about clear zones of inhibition, with measured values as high as 12.33 mm against E. coli and 13.30 mm against S. aureus. The films demonstrated strong antibacterial activities, which was higher for S. aureus than E. coli⁴³. When compared to standard antibiotics, the composite film showed good activities comparable to or higher than Amoxicillin. Amoxicillin exhibited an inhibition zone of 13.0 and 9.8 mm against E. coli and S. aureus, respectively.

Fig. 11.

Antibacterial activity of CNF (-ve control), CP-STI1 (Sample 1), CP-STI2 (Sample 2) and Ampicillin (+ ve control) against E. coli (left) and S. aureus (right) using disk diffusion method.

Even though some zeolite particles, such as montmorillonites, possess inherent and acquired antimicrobial properties in combination with other active agents like metal nanoparticles¹³ and hexadecyl pyridine bromide⁴⁴, the observed high antibacterial property of stilbite zeolite in this work is unique. The unique activity could be ascribed to the presence of free metal ions (Ca and Na) on STI surfaces, which would potentially lead to the formation of metal hydroxides in the interfacial structures of CP-STI composite films. Metal hydroxides such as NaOH are known to have high antibacterial activity induced by the alkaline conditions⁴⁵. In addition, we speculate that the availability of the free metal ions might have been enhanced by the presence of CNF and PVA in the composite film.

Conclusion

This work presents a low cost and robust alternative for the preparation of nanocomposite films from naturally abundant cellulose and stilbite zeolite using a simple solvent casting method. It involves effective and efficient extraction of nanocellulose fiber from Teff straw with a nanofiber yield of 37.7% and subsequent preparation of CNF-scaffolded stilbite composite films. Cellulose was extracted by alkaline hydrogen peroxide, followed by acid hydrolysis and untrasonication to obtain the nanocellulose. Besides, the resulting CNF/PVA/STI uniform composite films exhibited good mechanical properties, demonstrating an effective integration of STI particles into PVA crosslinked CNF network structures. The developed films also showed an interestingly high antibacterial property against both gram-positive and gram-negative bacterial strains, with clear inhibition zones of up to 12.33 mm against E. coli and 13.30 mm against S. aureus. These unique antibacterial properties caused by the presence of STI particles could be ascribed to the formation of metal hydroxides in the interfacial structures between STI particles and CNF/STI polymer networks. Overall, the observed characteristic features indicate the potential of the cellulose nanofiber-scaffolded stilbite films for practical applications, such as active packaging, surface disinfection and water treatment.

Author contributions

M.Y. and D.B.: Wrote original draft, review and editing, investigation; K.M. and G.K.: Review & editing; R.M.: Conceptualization & review; A.M.: Conceptualization, investigation, review and editing.

Funding

The work was supported by Addis Ababa Science and Technology University, under Grant Number EA-552/4 − 1/20.

Data availability

Data will be available from the corresponding author upon reasonable request.

Declarations

Competing interests

The authors declare no competing interests.