Introduction

Food items, whether fresh or processed, must be securely packaged to protect them from external threats such as pollutants, spoilage-inducing microorganisms, mechanical loads, and physical damage [1]. Studies have indicated that the usage of plastic packaging still accounts for roughly 37% of the world's plastic demand [2] despite global awareness of the environmental impact of plastic waste disposal. Over recent years biopolymers have generated major interest as a new generation of environmentally friendly packaging materials. Starch-based sheets are unique among all biopolymers due to their biodegradability, affordability, and abundance and have the potential to serve as a main packaging material [3]. In line with their biodegradability characteristics, however, starch-based packaging materials are prone to microbial contamination [4].

Active packaging, designed to extend the shelf life of food, works by either absorbing (scavenging) or releasing substances that prevent spoilage. These substances, which include antioxidants and antimicrobials, help maintain the safety and quality of foods. Typically, these active agents are embedded in polymer films. From there, they can move to the surface of the food, protecting against spoilage [5,6,7]. It is expected that by 2024, the market value of smart packaging, which includes active packaging, will grow to $26.7 billion [8].

The antimicrobial qualities of silver nanoparticles (AgNPs) have made them an intriguing element of biodegradable biopolymers [9], enabling the creation of active food packaging materials that extend food products shelf lives [10,11,12]. AgNPs have been incorporated into biopolymers, for example dextran [13], cellulose [14], chitosan [15], pectin [16], polycaprolactone, polylactic acid [17], in which the resulting composite films have shown antimicrobial effects on a spectrum of bacterial strains such as B. cereus, S. aureus, S. pneumoniae, L. monocytogenes, E. coli, S. Typhimurium, P. aeruginosa, and K. pneumonia [10]. According to a published review, AgNPs-biopolymer composites have been fabricated through two distinct methods: ex situ, in which polymeric matrix functions mainly as a dispersion and stabilisation medium for independently pre-synthesised AgNPs, and in situ, in which polymeric matrix serves as a reaction medium for the AgNP formation process and functions as a stabilising agent for the AgNPs [10]. Both techniques described in the literature result in AgNPs that are incorporated inside the polymeric matrix bulk. As the initial interaction between packaging materials and their surroundings primarily occurs at the surface, substrates with surface functionalisation may prove more efficient in averting the initial wave of microbial contamination. Additionally, surface functionalisation of the polymeric substrate is likely to demand a lesser quantity of AgNPs compared to bulk functionalisation.

Plasma technology is a substrate-independent technique that allows for tailoring surface properties without affecting the bulk properties of the materials [18,19,20,21,22,23,24]. Reactive functional groups can be added to biomaterial surfaces through thin-layer deposition using plasma polymerisation, which can then bind bioactive molecules and nanoparticles efficiently. This simple plasma treatment stands out in that it does not require solvents or initiators, requires very little amount of precursors, and produces no liquid organic waste, all of which makes it environmentally friendly [18]. The process is also environmentally sustainable since all energy required can be obtained from renewable sources such as solar panels or windmills. Plasma polymers have been extensively used for tailoring the surfaces of biomaterials, including for immobilisation of ligands and nanoparticles [18,19,20,21,22,23,24]. For example, an approximately 50-nm plasma film based on 2-methyl-2-oxazoline has been deposited on a glass coverslip for the attachment of AuNPs [25]. The authors reported that polyoxazoline films, which were combined with a specific type of ligand bearing —COOH groups like AuNP and antibody, remained reactive even after being stored for 12 months. This storage was done under conditions without air (vacuum) and in the dark to preserve their qualities. Other authors reported a plasma film thickness of approximately 24 nm for most bioapplications and < 20 nm for the attachment of AuNPs [26]. Other biomolecules and nanoparticles have been covalently immobilised using polyoxazoline (POx) for sensing and other uses [27].

To generate environmentally friendly and sustainable active packaging development, here a plasma-deposited polyoxazoline (POx) thin film based on 2-methyl-2-oxazoline is used to covalently immobilise AgNPs of starch sheets, as illustrated in Scheme 1. Plasma polymerisation utilising POx has garnered significant interest owing to its distinctive features, including low susceptibility to biofouling, biocompatibility, and stability [28,29,30,31]. The essence of the functionality of plasma-deposited polyoxazoline lies within the plasma phase. An additional noteworthy characteristic of POx is its capacity to modify surfaces through the attachment of biologically active molecules, such as nanoparticles, antibodies, and aptamers [32,33,34]. This will improve the substrate’s biological or physicochemical characteristics. To the best of the authors’ knowledge, there have been no reports so far on the immobilisation of AgNPs on plasma-treated cassava starch sheet surfaces.

Scheme 1
scheme 1

Schematic representation of creating plasma-nanoengineered active bioplastics through the attachment of mercaptosuccinic acid (MSA)-functionalized silver nanoparticles onto cassava starch-based bioplastics coated with polyoxazoline plasma film.

Full size image

This study introduces a novel approach in the field of active bioplastics by employing plasma nanoengineering. Unlike traditional methods that typically involve bulk incorporation of NPs, this technique is based on a surface functionalisation process using a 2-methyl-2-oxazoline-based plasma-deposited thin film for the immobilisation of AgNPs on starch sheets. This approach is particularly advantageous for active packaging applications, as it focuses on the initial interaction between packaging materials and their surroundings, which primarily occurs at the surface. This study will contribute significantly to the field of sustainable packaging technologies since it introduces a novel plasma-based approach for the development of active bioplastics, potentially reducing reliance on traditional plastics and minimising food spoilage.

The novel plasma-nanoengineered active bioplastics approach, characterised by immobilising silver nanoparticles on starch sheets’ surfaces to enhance antimicrobial properties without generating liquid organic waste, represents a significant advancement in sustainable packaging technologies. Its versatility, process efficiency, minimal precursor requirement, and green chemistry alignment offer clear advantages over existing solutions such as in situ synthesis and immobilisation [10, 35], electrostatic layer-by-layer [36], thiol chemistry [37], sol–gel [38], and electrospinning [39, 40] processes. However, it faces challenges in the diversity of plasma sources, i.e. the working gas, the electrode arrangement, and the power source that generates the plasma are unique for each set-up, cost, and integration challenges, alongside regulatory and safety considerations due to nanoparticle use. Despite the drawbacks, its potential to provide tailored antimicrobial responses and overcome the inherent weaknesses of biodegradable materials makes it a promising contender in the transition to more sustainable packaging options, provided that its antimicrobial spectrum is further optimised and expanded. The significance of plasma technology in enhancing food packaging has been extensively discussed in recent literature [41]. Recently, our research has highlighted the application of plasma technology for embedding natural dyes onto bioplastic surfaces [42]. Moving forward, our next phase aims to synergize natural dyes with AgNPs on bioplastics, aspiring to create an innovative form of active and intelligent packaging. This packaging will not only actively eliminate bacteria but also alter its colour in response to pH variations, paving the way for dynamic and interactive packaging solutions.

Materials and methods

Cassava starch, 2-methyl 2-oxazoline, sodium borohydride (NaBH4), silver nitrate, and 2-mercaptosuccinic acid were bought from Sigma-Aldrich (MO, USA). Tryptone Soy Broth (TSB) was procured from Oxoid. In the context of antimicrobial experiments, we utilised Eshercia coli (E. coli, ATCC 11303), Staphylococcus aureus (S. aureus, ATCC 25923), and Candida albicans (C. albicans, ATCC 10231). Ultra-pure Milli-Q water was used for all experiments. Silicon wafers (100) were bought from ProSciTech (Queensland, Australia). Fibroblast NIH3T3 (ATCC) cell was used for cell study. For the cell’s medium culture, Dulbecco’s modified Eagle medium was supplemented with Bovine Calf Serum. Both were purchased from Merck (Darmstadt, Germany). The mixture of Penicillin, Streptomycin, and Amphotericin B was purchased from Gene Dire X (Taiwan) and Cell Counting Kit 8 from Abbkine (Georgia, USA).

Synthesis of AgNPs

Silver nanoparticles (AgNPs) were generated through a conventional process involving the use of silver nitrate and sodium borohydride as reducing agent [18]. Briefly, 2 mM of silver nitrate (12 mL) was prepared to which 5 ml of 2 mM mercaptosuccinic acid (MSA) was added under ice-cold conditions to minimise excess heat generated during the process. 0.5 M of NaBH4 (0.5 mL) was added dropwise to the solution till it turned to reddish brown colour. The reaction proceeded in the dark, and the nanoparticle solution was subsequently stored at 4 °C.

Plasma polymerisation

Cassava starch bioplastic films were cut into defined shapes and placed inside a custom-built plasma reactor chamber functioning at 13.56 MHz. The starch sheets were subjected to air plasma cleaning at 50W for 5 min and a pressure of 1 × 10−1 mbar. Then the films were coated with polyoxazoline at 50 W for 2 min at a monomer precursor pressure of 2.0 × 10−1 mbar. The POx-coated samples were stored under vacuum-sealed conditions.

Surface immobilisation

The POx plasma-coated cassava starch sheets (5 mm diameter) were incubated with nanoparticle solution (1 mL) for 5 and 25 h, resulting in Ag-5 h and Ag-25 h samples, respectively. After the immobilisation period, the samples were washed thrice with Milli-Q water, dried using nitrogen gas, and stored in vacuum-sealed conditions.

Scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDS)

The surface morphology was imaged on an SEM Hitachi SU7000 Ultra-High-Resolution FE-SEM (Hitachi, Tokyo, Japan), equipped with an EDS detector at an accelerated voltage of 10 kV, working distance of 5 mm, and using an upper detector. The SEM micrographs were captured within the magnification range of 500 to 10,000. To compare the qualitative elemental analysis between the untreated starch, POx-treated starch, Ag-5 h, and Ag-25 h samples, spectra were acquired using AZTEC EDS software version 6.1 (Oxford Instruments, Oxfordshire, UK).

To determine the density of the immobilised AgNPs on the POx plasma-coated cassava starch, we analysed the EDS-SEM images using ImageJ 1.54 h (NIH, MD, USA). The 'Find Maxima' tool was utilised to enumerate AgNPs within 100 µm2 across three random areas of the image. A Student's t test was conducted to assess the significance between Ag-5 h and Ag-25 h.

X-ray photoelectron spectroscopy (XPS)

The surface chemistry of the plasma-coated cassava starch sheets was evaluated using a Kratos AXIS Ultra DLD spectrometer (Manchester, UK) operated at 15 keV and 15 mA. A monochromatic Al Kα source was used for the experiment. The survey spectra were obtained in the range of 0–1100 eV after setting pass energy of 160 eV and resolution of 0.5 eV. On the other hand, the high-resolution spectra were collected at a pass energy of 20 eV. Data analysis was conducted using Casa XPS software.

Water contact angle

The water contact angle of surface-functionalised cassava starch sheets was carried out using an RD-SDMO2 goniometer dropping 2 µL of Milli-Q water on the surface. The contact angle values from drop images were calculated using the DropSnake plugin toolbar installed in ImageJ software. The average values of contact angles were tabulated.

Atomic force microscopy (AFM)

Specimens were attached to a 1.5-cm-diameter holder using adhesive tape. Utilising a Park System NX10 Atomic Force Microscope in a non-contact setting with an AC160TS cantilever, the surface roughness data on a 5 × 5 µm area were acquired. The acquisition was facilitated through the SmartScan™ software specific to Park AFM, while analysis was carried out with Park System Corp's XEI software, version 5.1.6 Build 1, dated 20,200,313.

UV–Vis spectroscopy

For the UV–Vis spectroscopy analysis, a Shimadzu UV-1800 spectrophotometer was employed. This instrument facilitated the examination of samples across a broad wavelength range that encompasses both ultraviolet (UV) light, from 200 to 350 nm, and visible light, from 350 to 800 nm, allowing for a comprehensive analysis of the sample's light absorption characteristics. Measurements at 200–800 nm were taken using a bioplastic sheet sample dimensions of 1 × 5 cm. The scanning was performed in slow mode (0.1) to maximise the precision of the wavelength detection and absorbance measurements.

ICP-MS

The AgNP release was performed using 5-mm-diameter samples, immersed in 2 ml ultrapure water inside the wells of a 24-well plate, for 5 h and 25 h. After removal of the samples from the wells, released Ag in water was dissolved in 2% nitric acid to completely dissolve all released Ag. The Ag ion concentration was then measured using Agilent 8800 Triple Quad inductively coupled mass spectroscopy (ICP-MS).

Mechanical testing

The tensile strength and elongation at break were assessed following the ASTM D882 standard. These measurements were taken using a robust Zwick/Roell Z020 Universal Testing Machine. During the tests, the speed was consistently maintained at 10 mm/min, and the standard gauge length used for measuring the extension was 50 mm.

Antimicrobial testing

E. coli, S. aureus, and C. albicans were cultured in tryptic soy broth (TSB; Oxoid, ThermoFisher, Waltham, MA, USA) overnight at 37 °C. Before starting the bacterial experiments, the colony-forming units (CFU) / mL were determined for an optical density of 1. Subsequently, the bacteria and fungi were cultured to the logarithmic phase, adjusted to 1 × 106 CFU/mL, and 1 mL was introduced into a 24-well plate containing untreated starch sheet (CTL), Ag-5 h, and Ag-25 h in triplicate and for 20 h at 37 °C. Following that, the specimens were stained with the LIVE/DEAD® BacLight™ Bacterial Viability Kit (ThermoFisher Scientific, Waltham, MA, USA) according to the manufacturer's instructions. Subsequently, analysis was performed using an Olympus FV3000 laser confocal microscope (CLSM; Olympus, Tokyo, Japan). The samples were inverted onto a glass coverslip, and three randomly chosen regions were captured. SYTO 9 and propidium iodide fluorescence excitation/emission were observed at 480/500 nm and 490/635 nm, respectively. Viability assessments were conducted using ImageJ software v1.53a (NIH, Bethesda, MD, USA).

In vitro toxicity testing

The effect of the bioplastic samples on living cells was evaluated by studying the viable cell percentage after 24 h of interaction. The embryonic mouse fibroblast cell line NIH3T3 was cultured in a Dulbecco’s modified Eagle medium (DMEM) supplemented with 10% bovine calf serum and 1% mixture of 10.000 units/ml Penicillin, 10.000 µg/ml Streptomycin, and 25 µg/ml Amphotericin B in a 37 °C incubator with 5% CO2. After 90% confluence, cells were detached using Trypsin–EDTA and transferred into 24-well transwell plate. Each plate was dwelled by 6 × 104 cells/ml. The bioplastic sheets were punched to get 6-mm-diameter samples and placed on the upper part of the transwell inserts, allowing the released product from the samples to run through its 40-µm pores for 24 h at 37 °C. Viable cell determination is based on the dehydrogenase activity detection by reacting a highly soluble 2-(2-methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2, 4-disulfophenyl)-2H-tetrazolium, monosodium salt (Cell Counting Kit-8) for 2 h with the cell culture. This resulted in a water-soluble formazan dye, whose absorbance (Ab) was detected using a microplate reader (Multiskan Sky, Thermoscientific) at 460 nm. Finally, the cell viability percentage was calculated according to the formula below:

$${\text{Viable}}\,{\text{cell}}\left( \% \right) =\frac{{({\text{Ab}}\,{\text{of}}\,{\text{cell}}\,{\text{culture}}\,{\text{in}}\,{\text{a}}\,{\text{medium}}\,{\text{that}}\,{\text{has}}\,{\text{been}}\,{\text{exposed}}\,{\text{to}}\,{\text{the}}\,{\text{sample}}\,{-}\,{\text{Ab}}\,{\text{of}}\,{\text{medium}}\,{\text{only}})}}{{({\text{Ab}}\,{\text{of}}\,{\text{cell}}\,{\text{culture}}\,{\text{in}}\,{\text{a}}\,{\text{medium}}\,{-}\,{\text{Ab}}\,{\text{of}}\,{\text{medium}}\,{\text{only}})}}\times 100$$

Data from 6 (six) replications of each bioplastic were evaluated and compared. To determine cytotoxicity, a threshold of 80% cell viability was applied.

Statistical analysis

Graphical data were presented as the mean with its corresponding standard deviation. Statistical analyses were conducted using GraphPad Prism version 9.0.0.0 for Windows, developed by GraphPad Software (La Jolla, CA, USA, www.graphpad.com). XPS data were graphed using Casa XPS software. The water contact angle, tensile strength, elongation at break, and cell viability were subjected to analysis through one-way ANOVA, adjusted for multiple comparisons using the Dunnett method. Antibacterial data were analysed using a two-way ANOVA corrected for multiple comparisons using the Tukey method. A p-value of ≤ 0.05 was considered statistically significant, and each experiment was conducted in triplicate.

Results

The study embarks on the fabrication of plasma-nanoengineered active bioplastics utilising a POx-based plasma film and AgNPs. The POx-coated plasma film serves as the adhesive (crosslinking) layer between the bioplastic substrate surface and the AgNP-COOH antimicrobial compound [29], as illustrated in Scheme 1.

Morphology and elemental mapping: SEM–EDS

SEM and EDS were performed to assess the surface morphology and AgNP mapping, respectively, of the prepared samples. As comparisons, bare and POx-coated starch before immobilisation of AgNPs was included. Figure 1 shows that while the morphology of bare starch and POx-coated starch look similar (Fig. 1a, b), it changes after the immobilisation of silver nanoparticles for 5 h (Ag-5h) and 25 h (Ag-25h) (Fig. 1c, d). The deposition of thin plasma polymer film maintained the overall morphological characteristics of the cassava starch sheet. This was expected since plasma techniques are known for being capable of uniformly applying ultrathin, pinhole-free polymer-like coatings from almost a range of monomers onto various substrates [24, 43]. The immobilisation of AgNPs on the plasma-deposited POx film led to altered morphology.

Figure 1
figure 1

SEM images, colour-mapped images of Ag, and EDS analysis on (a, e, i) untreated, (b, f, j) POx-coated, (c, g, k) Ag-5h, and (d, h, l) Ag-25h starch sheets; (m) AgNP density on Ag-5h and Ag-25h sheets (ns = not significant, p = 0.10).

Full size image

The Ag elemental mapping images show no Ag on bare and POx-coated starch (black background in Fig. 1e, f, 0 weight % in Fig. 1i, j), while they show a prominent appearance of Ag on both Ag-5h and Ag-25h surfaces (brown spots in Fig. 1g, h 0.5 and 1 weight % for Ag-5h and Ag-25h, respectively, in Fig. 1k, l). Following immobilisation, the scaffolds exhibited a light brownish hue (as shown in scheme 1) attributed to the localised surface plasmon resonance of the AgNP [18]. Figure 1m shows the AgNP surface density values, derived from the elemental mapping images of Ag-5h and Ag-25h through ImageJ analysis, which are 26.6 ± 3.0 and 22.5 ± 1.3 AgNP/100µm2, respectively.

It is interesting to note that although the Ag-5h and Ag-25h starch sample’s surface density values are statistically similar (Fig. 1m), their mapping images appear different, with the former showing nicely dispersed AgNPs and the latter displaying some aggregations and surface inhomogeneity (Fig. 1g, h). We conclude that extended immersion times result in more surface-attached AgNPs; nevertheless, they can also cause aggregation, which leaves the AgNP's surface density uneven. In this scenario, we are examining the behaviour of nanoparticles (NPs) over a specific time frame in relation to their density and the time of immobilisation. Initially, within the first 5 h, these nanoparticles tend to bind to the POx-functionalised surface, forming a uniform layer. This phase represents the initial stage of NP-POx-functionalised surface interaction. As time progresses, specifically at the 25-h mark, there is a notable shift in the dynamics of these interactions. The interaction between nanoparticles themselves (NP-NP) starts to become more prominent, overshadowing the interaction between POx and nanoparticles (POx-NP). It should be noted that a crucial factor in this process is the small size of the nanoparticles, which inherently possess high surface energy. This high surface energy drives the nanoparticles to aggregate, a natural response aimed at reducing their surface energy. Ostwald ripening [44, 45] could also explain the observed agglomeration. Such aggregation is a critical aspect of the scenario, influencing the overall behaviour and formation of nanoparticles over the observed period.

The aggregation of silver nanoparticles (AgNPs) can influence the performance of the modified bioplastics, potentially reducing their antimicrobial efficiency, altering mechanical and optical characteristics, and compromising their barrier functions. Efforts to achieve a uniform nanoparticle coating include strategies like surface-modifying AgNPs (such as with MSA) to prevent clumping, incorporating capping agents during synthesis, and controlling the immobilisation time.

Chemical characterisations: XPS

XPS was employed to validate the surface chemical composition and reinforce the EDS results. Two peaks that corresponded to C1s at 285 eV and O1s at 532 eV were visible in the survey spectra (Fig. 2A) of the untreated starch sheets, which is in line with the polymer's chemical structure. Furthermore, the survey spectrum showed a prominent peak corresponding to N1s at roughly 400 eV when the starch was coated with POx plasma polymer, indicating that the surface modification of the starch sheet surface had been successful. Importantly, a strong peak at 368 eV that corresponds to Ag 3d was observed after immobilisation of AgNPs. Figure 2B demonstrates the atomic percentages of the four abundant elements (O, C, N, and Ag) that were also detected by EDS. Ag atomic percentage increased with increasing immobilisation time from 5 to 25h. This observed phenomenon agrees with the elemental mapping and calculation from EDS as described above. Figure 2C (1) shows that C1s contain mostly C–C and C–H in all studied samples, which fits the chemical structure of starch. After Ox plasma treatment [Fig. 2C (2)], POx-coated starch shows an increase in O-containing species (C–O–C, C–O–H, and C = O) and a decrease in C–C and C–H species, when compared to untreated starch. This is due to the abundant oxazoline groups on top of the starch surface, as hypothesised in Scheme 1. Five different components, C–C, C–C (285 eV), C–N/C–O (286.5 eV), C = O/C = N/NCO (288.2 eV), and O–C = OH, O–C = OR (289.7 eV with a β-shift peak at 285.9 eV) can be fitted in the high-resolution C1s spectrum to explain POx, which is in line with the literature [46].

Figure 2
figure 2

(A) XPS survey spectra, (B) atomic concentration, and (C) high-resolution C1s spectra of (1) untreated, (2) POx-coated, (3) Ag-5h, and (4) Ag-25h starch sheets.

Full size image

Furthermore, XPS detects 10 nm from the surface. The size of our particle is 12 ± 3 nm, aligning with our prior findings [47]. In the event of agglomeration, the size may increase to approximately 50 nm. In essence, it identifies Ag nanoparticles and a small amount of underlying POx. This explains the decrease in N1, which corresponds to C = O and O–C = O in Fig. 2C (3) and (4), originating from carboxyl-functionalised Ag nanoparticles. These values increase proportionally with the concentration of Ag nanoparticles.

Hydrophilicity and roughness: water contact angle and AFM

Generally, there is a direct correlation between the water contact angle and moisture barrier properties. Materials with a high water contact angle (hydrophobic surfaces) tend to have better moisture barrier properties. This is because hydrophobic surfaces repel water, reducing the likelihood of moisture penetration. Such materials are preferred in applications requiring high resistance to moisture to ensure product longevity and integrity. By manipulating the water contact angle, manufacturers can tailor the moisture resistance of packaging materials to meet specific requirements [48]. The water contact angle (WCA) data demonstrate how water interacts and wet the starch sheet surface upon the first/immediate contact. In Fig. 3A, it is evident that the application of POx coating resulted in a substantial decrease in the water contact angle (WCA) value of the starch surface, dropping from 94.7 ± 1.3° to 37.3 ± 2.2° (P < 0.0001). The same phenomenon was observed in our previous study [42]. Starch consists of lengthy glucose molecule chains that create a dense and intricate arrangement. The surface is thus covered with water-repelling aromatic groups. After POx deposition, the film covering the surface is rich with oxygen and nitrogen-containing groups, resulting in a lower observed WCA [42]. Subsequent immobilisation of 2-mercaptosuccinic acid-decorated AgNPs on the POx coating increased the WCA to some extent (43.7 ± 1.9° for Ag-5h and 47.7 ± 0.8° for Ag-25h), however, still significantly lower than the WCA of untreated starch. Apart from altered surface roughness, the higher polarity of oxazoline and 2-mercaptosuccinic acid (MSA) groups compared to starch should be responsible for these surface energy changes.

Figure 3
figure 3

(A) Water contact angle, (B) AFM images, and (C) UV–Vis spectra of untreated, POx-coated, Ag-5h, and Ag-25h starch sheets.

Full size image

Building on the above discussion of the effects of POx coating and decoration with MSA-decorated AgNPs on the hydrophilicity of the starch surface, it is relevant to investigate the relationship between these modifications and AFM-measured surface roughness. The roughness data in Fig. 3B indicate that untreated starch sheets possess a roughness value of 0.46, which significantly decreases to 0.25 upon the application of the POx coating. This was anticipated given that plasma techniques are recognised for their ability to consistently create ultrathin, pinhole-free coatings resembling polymers from nearly a wide range of monomers onto a variety of surfaces [24, 43]. This reduction in roughness correlates with the substantial decrease in WCA, suggesting a smoother surface allows for better water spreading, consistent with Wenzel’s law of wetting [42]. The POx coating, rich in oxygen and nitrogen-containing groups, not only alters the chemical composition but also the physical topography of the starch surface, making it more conducive to water spreading. Interestingly, the subsequent immobilisation of MSA-decorated AgNPs on the POx-coated starch surface resulted in an increase in roughness values to 0.39 for Ag-5h and 0.36 for Ag-25h. Despite this increase in roughness compared to the POx-coated surface, the roughness values are still lower than that of untreated starch. This partial reversal in smoothness corresponds with the observed increase in WCA, though the angles remain significantly lower than those of untreated starch. It suggests that while the decoration with AgNPs does increase surface roughness to a degree, the chemical nature of the coating—namely, the higher polarity of oxazoline and 2-mercaptosuccinic acid (MSA) groups—maintains a lower WCA compared to the untreated starch.

The relationship between hydrophilicity, as evidenced by WCA measurements, and surface functionalisation as well as surface roughness underscores the complex interplay between chemical composition and physical topography in determining the wettability of starch sheets. The decrease in roughness with POx coating enhances hydrophilicity by providing a smoother surface for water spreading. However, the introduction of AgNPs, despite increasing the surface roughness, does not revert the hydrophilicity to levels of untreated starch due to the overriding effect of the chemical characteristics of the coatings. This suggests that while roughness is a factor in determining hydrophilicity, the chemical composition and the presence of polar groups have a more significant impact on the water–starch surface interaction.

Importantly, while a high water contact angle is beneficial for moisture resistance, the specific requirements of the packaging application will dictate the ideal surface characteristics. For example, like in our case, controlled wettability was desired for AgNP immobilisation purposes, balancing the need for a moisture barrier with other functional requirements such as antimicrobial functionality.

Light absorption: UV–Vis spectroscopy

UV–Vis light absorption relates to a material's ability to absorb light in the ultraviolet (100–400 nm) and visible (400–700 nm) spectrum, crucial for packaging materials designed to protect sensitive products. The packaging materials help prevent photodegradation by blocking or significantly reducing light transmission. Thus, higher absorbance indicates higher light barrier ability of the sheet [49]. The synthesised MSA-capped AgNP solution has a characteristic peak around 400–450 nm due to plasmon resonance adsorption [47]. The specific wavelengths at which AgNPs exhibit strong absorption depend on their size, shape, the surrounding medium, and any modifications or coatings. When AgNPs are immobilised on a substrate, their light absorption characteristics might slightly alter due to changes in the surrounding medium and possible interactions with the substrate material [50,51,52]. This peak is present in Ag-5h and Ag-25h samples, while it is absent in POX and untreated samples, which confirms the successful immobilisation of AgNPs on plasma-functionalised surfaces (Fig. 3C). The minor peaks at 340 and 362 nm in all the samples were attributed by cassava starch films. Importantly, Fig. 3C shows that after nanoparticle immobilisation, both Ag-5h and Ag-25h samples exhibit higher absorbance, which means less light transmission or higher light barrier.

Tensile strength and elongation at break

Tensile strength and elongation at break tests were conducted to determine whether the surface treatments had an impact on the bulk properties. A material's ability to tolerate stretching forces before breaking is determined by its tensile strength (TS). The length difference between the sample's initial and altered lengths following breakage is called elongation at break (EB) or fracture strain. It illustrates the sample's capacity to withstand shape changes without breaking [53]. Figure 4A, B shows that compared to untreated starch, all surface treatments (POx, Ag-5h, Ag-25h) increased both TS (P < 0.05) and EB (P < 0.001). This indicates that the treatments enhanced the material’s (both surface and bulk) properties, most probably due to the crosslinking events upon plasma treatments to the thin starch sheets.

Figure 4
figure 4

Comparison of (A) tensile strength and (B) elongation at break of untreated starch, POx-treated, Ag-5h-treated, and Ag-25h-treated starch sheets. Data plotted as mean ± SD and n = 3, ns = not statistically significant, * P < 0.05, ** P < 0.01, *** P < 0.001, and **** P < 0.0001 = statistically significant.

Full size image

However, an intriguing phenomenon can be seen in the EB data (Fig. 4B), where the EB was greatly enhanced by POx coating (P < 0.0001). Subsequently, the application of Ag coating led to a reduction in the EB, reaching a nearly identical level as the untreated starch. Greater EB signifies higher ductility in the material. Materials with high ductility tend to stretch or bend rather than break, showing a capacity for deformation under tensile stress, while those with low ductility are more brittle, tending to crack or shatter before significantly bending or stretching [54]. In comparison with the POx-coated starch sheet, Fig. 4B shows that the addition of inorganic nanoparticles, or AgNPs, on top of the POx plasma film induces lower ductility to the starch sheet. Scheme 1 suggests that although AgNPs can interact with POx film through covalent bonds, they interact between themselves through weak hydrogen bonds. The combination presents a thin brittle AgNP coating on a ductile POx-coated starch sheet. This may contribute to the more brittle characteristic of the AgNP-coated starch in general. Tensile strength and elongation at break are among the crucial aspects of a packing material's mechanical characteristics [55]. Figure 4A, B shows that our treatments did not weaken the material's mechanical strength; in fact, there was a slight improvement.

Release of Ag+ ions, antimicrobial activity, and in vitro toxicity

The amount of Ag+ ions that were released into the solution upon immersion of Ag-5h and Ag-25h starch sheets for 1 day (24 h) and 7 days was measured using ICP-MS. The results are plotted in Fig. 5A. As expected, Ag-25h samples released more Ag+ ions compared to Ag-5h, due to the greater number of silver nanoparticles immobilised on the surface.

Figure 5
figure 5

(A) Concentration of Ag+ ion in the ultrapure water used to immerse the untreated, POx, Ag-5h, and Ag-25h samples (it is worth noting that the untreated and POx samples exhibited 0 values), (B) cell viability of C. albicans, E. coli, and S. aureus of cells that were not exposed to any starch sheets (CTL) and those exposed to Ag-5h and Ag-25h starch sheets for 24 h (overnight), (C) the corresponding exemplary FDA/PI images, and (D) illustration of antibacterial mechanism of AgNPs. The data are plotted as mean ± SD, n = 3, ns = not statistically significant, * p < 0.05, and ** p < 0.01 = statistically significant.

Full size image

The last part of the study involved examining the antimicrobial efficacy of the starch sheets coated with AgNP against three microbial strains: C. albicans (fungi), E. coli (Gram-negative bacteria), and S. aureus (Gram-positive bacteria). Cells that were not exposed to any starch sheets were employed as the control. While E. coli [56] and Candida species in general [57] are food-related bacteria and food-borne pathogens, respectively, S. aureus is a model for food-related Gram-positive bacteria. Figure 5B, C shows that AgNP-coated starch did not exhibit antifungal properties (no significant reduction of viability against C. albicans). However, a prominent reduction of viability was observable when Ag-5h and Ag-25h were compared to the untreated starch for both Gram-negative (P < 0.05) and Gram-positive bacteria (P < 0.001). When exposed to an aqueous medium, AgNPs undergo oxidation, resulting in the release of Ag+ ions. These ions bind to bacterial DNA, hindering replication and cell metabolism, ultimately leading to bacterial lysis [58, 59]. This mechanism is illustrated in Fig. 5D. One of the serious concerns in food applications involving Ag+ ions is that the released ions should induce bacterial death without causing toxicity to the consumer. Notably, our film releases only trace amounts, and the concentration of released Ag+ in parts per billion (according to ICP-MS results) was sufficient to induce bacterial lysis. In packaging applications, these trace amounts help limit the migration of silver ions to food as well as the daily intake of silver ions by humans. Group restriction of 50 μg silver/kg food has been proposed by the Scientific Panel on food additives, flavourings, processing aids, and materials in contact with food (AFC), while an acceptable daily intake (ADI) of 0.9 μg silver ions/kg body weight per day has been set by the European Chemicals Agency (ECHA) [60].

The findings underscore the distinctive ability of plasma coating to function as an adhesive layer that covalently binds the AgNPs with the starch substrate surface, controlling the release of metal ions and preventing excessive leaching. With increased surface loading of AgNPs (i.e. 25h), we noted a continued reduction in bacterial viability for both E. coli and S. aureus, pointing to the tunable character of our coatings and the opportunity to tailor them as per the specific requirements of the application. In the future, further surface optimisation will be carried out to improve the surface’s antifungal properties (for example against C. albicans).

Antimicrobial activity often implies cytotoxicity, as agents effective at eliminating bacteria can also harm living cells. Studies indicate that AgNPs can cause cell damage, DNA alterations, and inflammation, posing substantial health risks [10, 61]. Nonetheless, there are strategies to reduce these harmful effects. Encapsulating silver nanoparticles in biocompatible materials can reduce direct exposure to cells, minimising toxicity while preserving antimicrobial efficacy [62]. Surface modification of nanoparticles with biocompatible ligands can improve their stability and reduce their reactivity with biological molecules [63]. Employing controlled release mechanisms ensures that silver ions are released at therapeutic concentrations, avoiding excessive accumulation that could harm cells [64]. Additionally, optimising the size, shape, and dose of silver nanoparticles can further reduce cytotoxic effects, as these factors significantly influence their interaction with cellular structures [63].

Despite the above concerns, our study showed that the percentage values of viable cells after 24-h interaction with all bioplastic samples, were above 80% (Fig. 6), implying that no bioplastic variant was cytotoxic. Interestingly, it is seen that POx, Ag-5h, and Ag-25h samples increased cell viability by about 20 to 40% (P = 0.000). This result is supported by previous study by Subbiah et al. [65] who investigated AgNP sized 19.5 ± 7.7 nm. The authors reported that AgNPs provided an inhibition effect against S.aureus, E.coli, E. faecalis, S. epidermidis, B. subtilis, S. typhimurium, and S.enterica. However, a concentration of the AgNPs up to 10 µg/mL did not influence NIH3T3 cell viability. Future research stages are expected to include detailed safety evaluations, focussing on further analysis of antimicrobial activities and toxicity studies, building upon the preliminary confirmation of antimicrobial effectiveness and the absence of toxic effects.

Figure 6
figure 6

The percentage of viable cells after 24 h of NIH3T3 cell culture with untreated, POx, Ag-5h, and Ag-25h bioplastic sheet samples at 37°C.

Full size image

Moreover, our research effectively addresses the environmental risks associated with nanoparticle dispersion into water streams and soil upon waste disposal, by anchoring silver nanoparticles (AgNPs) onto cassava starch sheets through a plasma-deposited film. This process drastically minimises their potential release into ecosystems. Moreover, the controlled release mechanism designed within the coatings further limits environmental exposure, ensuring that the antimicrobial properties are utilised effectively without compromising ecological safety.

While initial investments in plasma technology may be higher compared to conventional methods, the long-term benefits justify the costs. The specific benefits of plasma polymerization include a) rapid deposition, independent of substrate types; b) formation of pinhole-free, highly crosslinked structures ensuring strong substrate-to-coating adhesion; c) absence of solvents, reducing potential cytotoxicity; d) effective deposition within 3D porous structures through plasma diffusion; e) deposition without the need for pre-treatment; f) chemically stable and physically robust coatings; g) precise control over thickness, reaching down to tens of nanometres for biodegradable applications; and h) non-destructive nature to prevent alterations in the substrate biomaterial's structure.

Conclusions

The innovative approach of using an oxazoline-based plasma film for the attachment of AgNPs to starch-based sheets has demonstrated a promising opportunity for generations of active packaging materials. The successful immobilisation of AgNPs was evidenced by SEM–EDS, XPS, as well as UV–Vis analyses, which also suggested an even distribution of nanoparticles at a 5-h immersion time of the POx-coated starch in a solution containing AgNPs. Longer immersion time resulted in the aggregation of AgNPs on the surface, although the surface density was not significantly altered. The water contact angle measurements revealed a significant reduction in the contact angle after the application of the plasma film and subsequent MSA-decorated AgNPs, suggesting an increase in surface hydrophilicity. AFM data showed changes in surface roughness, with a decrease in roughness values following the POx coating application, leading to a smoother surface that facilitates better water spreadability. However, the addition of AgNPs resulted in a slight increase in roughness, which, despite being higher than the POx-only coating, still represented a surface smoother than the untreated starch sheets. The analysis highlights the critical role of both surface chemistry and topography in designing starch-based materials with tailored wettability and moisture barrier properties. By manipulating these parameters, it is possible to engineer starch sheets with specific hydrophilic or hydrophobic characteristics suitable for packaging applications. UV–Vis spectroscopy implied enhanced light barrier properties of the treated sheets, which could provide additional protection against photodegradation of packaged goods. Mechanical testing results demonstrated an improvement in both tensile strength and elongation at break for the treated starch sheets compared to the untreated control. This improvement signifies that the surface modification process does not compromise the mechanical integrity of the packaging material; instead, it potentially enhances it, making the treated sheets more suitable for practical packaging applications. While the antimicrobial efficaciousness against bacteria exhibited potential without harming living cells, the lack of antifungal activity indicates a possible direction for further investigation and refinement. In conclusion, our research advances sustainable packaging options to reduce food deterioration and plastic waste.

Implementing the suggested technology into practice on a large industrial scale encounters significant challenges, notably in cost-effectiveness, plasma source diversity, scalability, and integration with existing packaging production processes. The technology's reliance on advanced procedures such as plasma deposition and nanoparticle synthesis necessitates specialised equipment and potentially pricey materials, escalating production costs. Scaling these processes to meet industrial demands poses additional difficulties, with the potential for variability in product quality and challenges in achieving uniform nanoparticle distribution across extensive batches. Moreover, integrating this novel approach into current manufacturing set-ups requires overcoming logistical and technical hurdles, potentially disrupting established workflows, and necessitating substantial modifications to accommodate the new technology. Overcoming these hurdles through innovation and research is crucial for widespread adoption, aligning with the growing preference for eco-friendly options.

Supplementary information

Not applicable.