Impact of Whey and Sucrose Concentrations on Bacterial Cellulose Characteristics for Functional Food Applications

Abstract

Bacterial cellulose (BC) shows high potential for food applications, yet its scalable production using whey-based substrates remains challenging. The aim of the study was to evaluate the influence of whey (40–100% v/v) and sucrose (0–50 g·L⁻¹) concentrations on BC synthesis in a medium formulated with black tea. Static cultures (28 ± 2 °C, 15 days) were carried out using an inoculum of 25% (v/v) of an adapted microbial consortium and compared to a whey-free control. The structural, physicochemical, and functional properties of BC were characterized by dry mass yield, hygroscopicity, FTIR, XRD, SEM, transparency, and mechanical tests. Although it did not alter the chemical structure of BC, whey exerted a strong impact on its synthesis: the formulation with the highest whey and sucrose contents showed the highest yield (9.56 ± 1.76 g·L⁻¹), with fibrils ranging in diameter from 50 to 100 nm. Although crystallinity decreased (57.30%), this result did not impair mechanical performance; on the contrary, such a treatment resulted in the highest tensile strength (13.43 ± 2.30 Mpa). Thus, modulating whey and sucrose concentration proves to be an effective strategy for adjusting the structural and functional properties of BC, highlighting the potential of the selected byproducts as low-cost substrates for technological applications in the food sector.

1. Introduction

The use of biopolymers in advanced food technologies is increasingly recognized as a promising alternative to synthetic polymers of petrochemical origin (SPPOs), which are widely used in different stages of the agri-food chain, with their environmental impacts and potential risks to human health being widely reported [1,2]. SPPOs exhibit high chemical stability and low degradation rates, which favor their persistence in the environment and the formation of microscopic fragments, such as micro- and nanoparticles, that have been found in food, water, and air, harming human health and increasing environmental impact [3]. Despite the technological advancements associated with these polymers, global waste management remains insufficient, resulting in significant environmental accumulation, incineration, or limited reuse [1]. In addition, the presence of functional additives incorporated into synthetic polymer matrices, such as bisphenol A and phthalates, is a concern, since these compounds can be released under certain processing or storage conditions, favoring their migration into food systems [4]. This scenario highlights the need for alternative materials with equivalent functional performance, capable of gradually replacing conventional polymers without compromising technological efficiency or product quality.

Polymers of biological origin, such as bacterial cellulose (BC), exhibit structural and functional properties favorable to applications in food technology, including the formation of films, edible coatings, and colloidal systems, such as particle-stabilized emulsions. BC is an exopolysaccharide produced by acetic acid bacteria, consisting of linear β-1,4-D-glucose chains, organized in highly crystalline micro and nanofibrils with a high degree of purity, whose three-dimensional network confers high mechanical strength and water retention capacity [5]. This fiber organization provides stability under temperature variations, a necessary characteristic for its application in industrial processes that can be improved with process modifications [6]. In addition, BC exhibits biocompatibility, non-toxicity, and biodegradability, essential characteristics for materials intended for direct contact with food. The surface of BC is rich in hydroxyl groups, allowing chemical modifications and incorporation of functional compounds, facilitating the optimization of properties such as antimicrobial activity, gas barrier properties, hydrophilicity, and emulsion stability, which expands its potential use in different food technology platforms [7].

Although the potential of BC as coatings, films and emulsions is widely recognized, there are still technical and scientific challenges that need to be overcome for its production to achieve industrial viability [5,8,9]. Studies indicate that the cost, availability, and quality of raw materials directly affect the scalability of fermentation processes and the cost–benefit ratio of large-scale production [10]. In this context, Kombucha fermentation, which uses black and green tea infusions and a microbial consortium predominantly composed of acetic acid bacteria (AAB) and yeasts, has been widely explored as an alternative strategy to enhance BC production [11]. Tea infusions act as a basal medium, providing phenolic compounds, minerals, and accessory sources of nitrogen that promote microbial growth and the biosynthesis of the cellulosic matrix [12]. Furthermore, the metabolic interaction between yeast and AAB, associated with the conversion of ethanol to acetic acid, promotes an acidic environment favorable to BC synthesis and the inhibition of undesirable microorganisms, while tea polyphenols can modulate the structural organization and physicochemical properties of the cellulose produced [13].

The use of alternative nutrient sources, especially agro-industrial byproducts, has stood out for offering bioavailability, favorable nutritional composition, and low cost. Dairy byproducts can serve as culture media rich in lactose, proteins, organic acids, and B vitamins, thereby promoting microbial growth and metabolic activity [14,15]. Cheese whey, formed after protein coagulation and curd separation, has a variable composition depending on the origin of the milk, which is also influenced by factors such as animal breed, feed, and seasonality, as well as the type of technology used in the manufacture of different cheeses [16,17]. Depending on the coagulation mechanism, this byproduct can be classified as sweet whey, originating from the action of rennet (rennin enzyme), or acid whey, resulting from chemical or microbial acidification, presenting only small compositional variations [17]. Both have similar amounts of protein, fat, and lactose; however, sweet whey has a higher pH value (between 6 and 7) and lower levels of minerals and lactic acid, while acid whey has a lower pH (between 4.5 and 5.8) and higher contents of minerals and lactic acid [17,18,19].

Based on this composition, cheese whey has been used as a base for feeding acetic acid bacteria to obtain BC [18,20]. However, most studies employ whey at high proportions (often 100%), without prior adaptation steps for the microbial consortium, and predominantly resort to medium supplementation to overcome nutritional and metabolic limitations. These studies generally report lower yields of bacterial cellulose, such as [21], which has been associated with the limited capacity of acetic acid bacteria to efficiently metabolize lactose as the main carbon source and the lower quantity of the Lactobacillus genus in the initial consortium.

The substitution of black tea with other constituents, such as coffee husk tea, has already been reported in the literature [22]. However, gaps remain regarding the complete replacement of tea infusions with alternative nutritional sources, since the available studies focus mainly on supplementing the classic kombucha medium, instead of systematically evaluating alternative formulations [23]. In this context, the hypothesis is that the prior adaptation of the microbial consortium, associated with the controlled variation in the proportions of cheese whey (40–100%) and sucrose concentration (0–50 g/L), may favor the metabolic utilization of the substrate and modulate the biosynthesis of the cellulosic matrix, reflecting on the yield and properties of the bacterial cellulose produced. The selection of these concentration ranges allows for controlled investigation of the nutritional and metabolic limits of the microbial consortium in a system inspired by the classic kombucha medium, as well as the role of sucrose as a complementary carbon source. Thus, the present work aimed to evaluate the influence of whey and sucrose concentrations on the production and properties of bacterial cellulose.

2. Materials and Methods

2.1. Microorganisms and Culture Media

The Symbiotic Culture of Bacteria and Yeast (SCOBY), previously cultivated at the Advanced Institute of Technology and Innovation (IATI, Recife, Brazil), was used as inoculum to produce bacterial cellulose (BC). Prior to the experimental assays, the microbial consortium was subjected to a stepwise adaptation protocol to a sweet cheese whey-based medium, as schematically illustrated in Figure 1. This adaptation strategy was designed to promote the gradual replacement of black tea by cheese whey through serial inoculations with increasing whey proportions. The fermentation assays were conducted using a kombucha-based process adapted from Silva Júnior et al. [7]. The microbial consortium (25% v/v) was added to 5 L Schott flasks containing 3 L of culture medium (containing whey and black tea) supplemented with 50 g/L of sucrose. The cultivation was conducted under static conditions at 28 ± 2 °C, with adaptation over 30 days [5,10].

The cheese whey used in the adaptation and fermentation assays was supplied by Campo da Serra (Pombos, Brazil) and was obtained from Minas padrão cheese production by enzymatic coagulation with rennet, presenting an initial pH of 6.38. The adapted microbial consortium maintained at IATI comprised dominant bacterial genera, including Komagataeibacter, Gluconobacter, Lactobacillus, and Pediococcus, as well as yeasts such as Saccharomyces and Zygosaccharomyces [24,25].

Culture Conditions and BC Production

After adaptation, this consortium was used as inoculum in culture media formulated to evaluate the effect of whey and sucrose concentrations on bacterial cellulose (BC) production. The culture medium was prepared from black tea by infusing 10 g·L⁻¹ of leaves in water, followed by adding 1.15 g·L⁻¹ of citric acid. Whey and sucrose were chosen as independent variables and examined using a 2² factorial experimental design with a central point, aiming to assess their individual and combined effects on BC production. Whey concentration was tested at two factorial levels (40% and 100% v/v), while sucrose concentration was tested at 0 and 50 g·L⁻¹ (

Table 1. Factors and levels evaluated in the experimental matrix.

Factor	Level
	−1	0	+1
Whey concentration (% v/v)	40	70	100
Sucrose concentration (g·L−1)	0	25	50

). A central point (EC) (70% v/v whey and 25 g·L⁻¹ sucrose) was included and performed in triplicate.

Culture media were prepared according to

Table 2. Experimental matrix for evaluating whey and sucrose concentrations.

Sample	Whey Concentration (% v/v)	Sucrose Concentration (g·L−1)
E1	40	0
E2	100	0
E3	40	50
E4	100	50
EC	70	25

. For each experimental condition, 300 mL of culture medium was transferred to 500 mL Schott flasks and inoculated with 25% (v/v) of the previously adapted microbial suspension. A medium containing only tea infusion was used as the control (EP). Cultures were conducted under static conditions at 28 ± 2 °C for 15 days.

After fermentation, the BC formed was subjected to a purification process. Initially, the membranes were washed with running water and subsequently immersed in a 1 M NaOH solution at 90 °C for 2 h [26]. Then, a second wash with water was performed until the pH of the membranes reached values close to 7. The wet membranes were weighed on an analytical balance (ATX224R, Shimadzu, Kyoto, Japan), placed to dry in an incubator oven (TE-371/240L, Tecnal, Piracicaba, Brazil) at 30 °C until a constant mass was reached. The dry mass yield (Y), used as the response of the design, was determined as the dry mass of BC produced per liter of fermented broth (g·L⁻¹) according to the following equation [7]:

$$Y \left(\mathrm{g}·{\mathrm{L}}^{-1}\right)={\displaystyle \frac{\mathrm{BC} \mathrm{membrane} \mathrm{dry} \mathrm{mass} \left(\mathrm{g}\right)}{\mathrm{Volume} \mathrm{of} \mathrm{culture} \mathrm{medium} \left(\mathrm{L}\right)}}$$

(1)

2.2. Characterization of BC Membranes

2.2.1. Water Retention Capacity (WRC)

The water retention capacity (WRC) of BC membranes was determined from the masses of the BC before and after drying, according to the following equation:

$$WCR \left(\%\right)={\displaystyle \frac{\mathrm{Wet} \mathrm{mass} \mathrm{of} \mathrm{membrane} \left(\mathrm{g}\right)-\mathrm{Dry} \mathrm{mass} \mathrm{of} \mathrm{membrane} \left(\mathrm{g}\right)}{\mathrm{Wet} \mathrm{mass} \mathrm{of} \mathrm{membrane} \left(\mathrm{g}\right)}} \times 100$$

(2)

2.2.2. Determination of Water Contact Angle and Swelling Rate

The contact angle with water was determined by the sessile drop method [27]. Briefly, dry BC membranes were cut into 10 × 5 mm sized pieces, a drop of water was deposited on their surface, and the angle was measured after 1 s of spreading using a goniometer coupled to a mirrorless digital camera (XT10, Fujifilm, Tokyo, Japan). The dry membranes (10 × 5 mm) were weighed, immersed in distilled water at 25 °C for 24 h, and weighed again, and the swelling rate (SR) was determined from the mass variation according to Equation (3):

$$SR \left(\%\right)={\displaystyle \frac{\mathrm{Swollen} \mathrm{membrane} \mathrm{mass}\left(\mathrm{g}\right)-\mathrm{Initial} \mathrm{membrane} \mathrm{mass}\left(\mathrm{g}\right)}{\mathrm{Initial} \mathrm{membrane} \mathrm{mass}\left(\mathrm{g}\right)}} \times 100$$

(3)

2.2.3. Fourier Transform Infrared (FTIR) Spectroscopy

The FTIR spectra of samples were recorded in transmittance mode, at room temperature (22 °C), from the accumulation of 16 scans at 4 cm⁻¹ in the wavenumber range of 4000 to 650 cm⁻¹, using a Spectrum 400 spectrometer (PerkinElmer, Waltham, MA, USA) equipped with an attenuated total reflectance (ATR) diamond sensor.

2.2.4. Thermogravimetric Analysis (TGA)

The thermal stability of BC membranes was evaluated using a thermal analyzer (DTG-60H, Shimadzu, Kyoto, Japan). Approximately 6 mg of the dry sample was placed in an aluminum crucible and heated from 35 to 600 °C at a 10 °C/min rate under a nitrogen flow of 20 mL/min.

2.2.5. X-Ray Diffractometry (XRD)

The crystallinity index (CrI) and BC crystallite size were studied by XRD using a D2 Phaser analytical diffractometer (Bruker, Billerica, MA, USA), with Cu Kα radiation generated at 30 kV and 10 mA. Data were recorded in reflection mode at a scan range (2θ) of 5° to 80° and an angular step of 0.05°. CrI was determined using Equation (4):

$$CrI\left(\%\right)= {\displaystyle \frac{\mathrm{Ac}}{\mathrm{Ac} + \mathrm{Aam}}} \times 100$$

(4)

where A_c is the area of the crystalline peaks, and A_am is one of the amorphous halos, which were calculated using 8.5 software (OriginLab Corporation, Northampton, MA, USA) [27].

2.2.6. Scanning Electron Microscopy (SEM)

Dried BC membranes were mounted on metal supports with conductive carbon tape and coated with gold for 30 s (SC-701 Quick Coater, Tokyo, Japan). Morphological analysis was performed using a scanning electron microscope (MIRA3 LM, Tescan, Warrendale, PA, USA).

2.2.7. Transparency

The transparency of membranes was determined by UV-Vis spectrophotometry. Rectangular samples were positioned directly in the optical beam, and absorbance was recorded at 600 nm using a UV-Vis spectrophotometer (UV-M51, BEL Photonics, São Paulo, Brazil) [28]. Relative transparency was calculated as the inverse of the average absorbance, with higher values indicating greater transparency of the material.

2.2.8. Mechanical Tests

The mechanical properties of BC were evaluated by tensile testing, determining tensile strength (Mpa), Young’s modulus (Mpa), and maximum deformation (%). Samples were cut into strips of 7 × 3 cm, were 0.04 ± 0.01 mm in thickness, and were analyzed at 25 °C, at 0.5 m/min using a 1 kN load cell in a universal testing machine (Instron 5969, Norwood, MA, USA). In addition, samples were subjected to a bending strength test, being bent and unfolded or 100 cycles along the same axis, and their flexibility was classified as poor (<20 cycles), fair (20–49 cycles), good (50–99 cycles), or excellent (≥100 cycles) [26,29].

2.3. Statistical Analyses

Statistical analyses were performed using the mean values obtained from experimental triplicates. The results were subjected to analysis of variance (ANOVA) using RStudio software (R 3.6.0+) and Tukey’s multiple comparisons test, executed using PAST 4.03 software (Oslo, Norway), adopting a significance level of 5% (α ≤ 0.05).

3. Results and Discussion

3.1. Influence of Whey and Sucrose Concentrations on Bacterial Cellulose Yield

It was visually observed that changes in the nutritional composition of the medium directly influenced the biosynthesis and morphological aspects of BC (Figure 2). Membranes produced exclusively in whey showed a lighter color and homogeneous surface, probably due to a greater incorporation of residual components from the dairy substrate, such as soluble proteins and lipids. On the other hand, those obtained in the medium containing black tea exhibited a darker tone, consistent with the natural compounds resulting from the chemical modifications of polyphenols as a consequence of microbial growth [30].

Wet mass, dry mass, and dry mass yield varied consistently among treatments (

Table 3. Masses of bacterial cellulose before and after drying and dry mass yield (Y). E1–E4: different concentrations of cheese whey and sucrose; EC: concentrations of cheese whey and sucrose at the central point. E1: 40% (v/v) whey and 0 g·L⁻¹ sucrose; E2: 100% (v/v) whey and 0 g·L⁻¹ sucrose; E3: 40% (v/v) whey and 50 g·L⁻¹ sucrose; E4: 100% (v/v) whey and 50 g·L⁻¹ sucrose; and EC (corresponding to samples E5, E6 and E7): 70% (v/v) whey and 25 g·L⁻¹ sucrose. EP: control medium containing only the tea medium.

Sample	Wet Mass (g)	Dry Mass (g)	Y (g·L−1)
E1	10.45 ± 0.67	0.35 ± 0.01	1.15 ± 0.05
E2	8.56 ± 1.35	0.56 ± 0.06	1.85 ± 0.21
E3	16.28 ± 2.15	0.91 ± 0.07	3.04 ± 0.23
E4	28.10 ± 2.95	2.87 ± 0.07	9.56 ± 1.76
EC	5.90 ± 0.27	0.31 ± 0.03	1.18 ± 0.26
EP	10.00 ± 4.19	0.21 ± 0.11	0.70 ± 0.37

), reflecting the direct influence of the nutritional composition of the medium on BC biosynthesis. Taking the intermediate level (EC = average of the intermediate points), corresponding to 70% (v/v) whey and 25 g·L⁻¹ sucrose, as a reference, it was observed that the simultaneous increase in whey concentrations to 100% (v/v) and sucrose to 50 g·L⁻¹ (condition E4) resulted in the highest yield, reaching 9.56 ± 1.76 g·L⁻¹ (p < 0.001). Although direct quantitative comparisons with the literature data must be interpreted with caution due to differences in microbial strains, substrate composition, fermentation duration, and cultivation strategies, the BC yield obtained in this study is within or above the range reported for dairy-based systems under static conditions. Liu et al. [31], for example, reported 2.75 g·L⁻¹ after four days of static fermentation using SCOBY, while Rollini et al. [32] reported 6.77 ± 0.14 g L⁻¹ in a medium supplemented with β-galactosidase (0.5 U mL⁻¹) over seven days using Komagataeibacter xylinus DSM 2325. Kolesovs et al. [33] also reported 6.97 ± 0.17 g L⁻¹ in an enzymatically hydrolyzed milk medium supplemented with corn steep liquor. The higher BC yield observed may be associated with prior microbial adaptation to dairy substrates, which can enhance metabolic efficiency and cellulose biosynthesis in mixed microbial systems [15].

No significant differences in dry mass yield (Y) were observed among E1, E2, E3, and EC (p > 0.05). EC showed an intermediate yield (1.18 ± 0.26 g·L⁻¹), similar to that observed for E1 (40% whey, 0 g·L⁻¹ sucrose; 1.15 ± 0.05 g·L⁻¹) and E2 (100% whey, 0 g·L⁻¹ sucrose; 1.85 ± 0.21 g·L⁻¹), indicating that the absence of sucrose significantly limits the ability of microbial cells to convert the substrate into cellulosic biomass. Treatment E3 (40% whey and 50 g·L⁻¹ sucrose), in turn, guaranteed the second highest Y value (3.04 ± 0.23 g·L⁻¹) among the experiments, showing a significant difference only from E4 and EP according to Tukey’s test (95% confidence). This behavior demonstrates that an increase in sucrose level exerts a positive effect on cellulose synthesis, even when whey content is reduced, reinforcing the importance of the availability of assimilable carbon in the process [34]. The standardized effects Pareto chart (Figure 3) confirms that sucrose concentration was the most influential factor in BC yield (SQ = 69.072; F-value = 27.13; p = 7.09 × 10⁻⁵), followed by whey concentration (SQ of 38.92, F-value of 15.29 and p = 0.0011) and their interaction (SS = 25.317; F-value = 9.95; p = 0.0058). All effects exceeded the statistical significance threshold (p < 0.05), corroborating the ANOVA results and indicating that BC production is governed by the combined availability of fermentable carbon and milk-derived nutrients. In contrast to these results, the control medium (EP), containing only tea, led to the lowest yield (0.70 ± 0.37 g·L⁻¹), belonging to the same statistical group as E1, E2, and EC (p > 0.05). This low performance can be attributed to the fact that tea does not provide sufficient quantities of essential nutrients for the biosynthetic phase, which limits the production of the exopolysaccharide.

These findings corroborate the technical viability of using whey as the main or sole component of the fermentation medium for biocellulose production. The statistically significant dependence of the response on the whey concentration and its interaction with sucrose reinforces the potential of this dairy byproduct as a sustainable substrate for biotechnological applications.

3.2. Influence of Whey and Sucrose Concentrations on the Physicochemical and Structural Characteristics of BC

3.2.1. Water Retention Capacity, Swelling Rate, and Droplet Contact Angle

The values of water retention capacity (WRC) indicated that all formulations exhibited high water holding properties, an intrinsic characteristic of the highly hydratable nanofibrillar network of BC [26,35]. However, variations in WRC could be observed using different medium formulations (

Table 4. Results of water retention capacity (WRC), swelling rate (SR), and droplet angle. E1–E4: different concentrations of whey and sucrose; EC: concentrations of whey and sucrose at the central point. E1: 40% (v/v) whey and 0 g·L⁻¹ sucrose; E2: 100% (v/v) whey and 0 g·L⁻¹ sucrose; E3: 40% (v/v) whey and 50 g·L⁻¹ sucrose; E4: 100% (v/v) whey and 50 g·L⁻¹ sucrose; and EC (corresponding to samples E5, E6, and E7): 70% (v/v) whey and 25 g·L⁻¹ sucrose. EP: control medium containing only the tea medium.

Sample	WRC (%)	SR (%)	Droplet Angle (°)
E1	96.67 ± 0.35	64.10 ± 0.03	22.27 ± 1.57
E2	93.43 ± 1.07	56.15 ± 0.10	39.99 ± 0.64
E3	94.29 ± 1.20	78.64 ± 0.01	16.35 ± 1.62
E4	89.68 ± 1.34	45.23 ± 0.06	22.48 ± 1.58
EC	94.91 ± 0.70	24.42 ± 0.23	18.83 ± 0.56
EP	97.95 ± 0.24	47.74 ± 0.00	21.93 ± 0.87

). Tukey’s analysis showed that condition E4 differed significantly from E1, EC, and EP (p < 0.05), while no significant differences were found between E4 and E2 or E3 (p > 0.05). The EP condition allowed obtaining a cellulosic membrane with the highest WRC value (97.95 ± 0.24%), while condition E4 exhibited the lowest retention (89.68 ± 1.34%). The remaining treatments displayed intermediate WRC values. According to Kolesovs et al. [33], cellulosic fibers produced in whey-based medium tend to have larger diameters than those produced in Hestrin–Schramm (HS) medium, which may promote greater network compaction and reduce water retention. This behavior is consistent with the lower WRC observed for condition E4, which was formulated with 100% whey and a high sucrose concentration. In contrast, the medium containing only tea (EP) produced a matrix with smaller-diameter nanofibers, higher porosity, and a higher density of hydrophilic sites, resulting in the highest retention rates observed.

The swelling rate (SR) represents the ability of dry membranes to reabsorb water after dehydration, being a parameter directly influenced by the structural reorganization of the material during the drying process [36]. As discussed by Muthu & Rathinamoorthy [37], the partial collapse of the three-dimensional nanofibril network throughout drying significantly reduces porosity, limiting rehydration. Although all BCs in this study underwent the same drying method, significant variation was observed among the evaluated conditions, indicating intrinsic differences in the degree of compaction and interfibrillar interactions. That is, heat affected the structure of these fibers differently. Condition E3 allowed for the highest SR value (78.64 ± 0.01%; p < 0.05), demonstrating that its nanofibrillar structure remained preserved after the drying process. On the other hand, EC proved to be more susceptible to heat, with an SR value of 24.42 ± 0.23% (p < 0.05). The values of E1, E2, E4, and EP (p > 0.05) showed similar behavior to that reported by Amorim et al. [26] for BC produced in HS medium. In food systems, such as emulsions or packaging, a greater rehydration capacity can be advantageous by favoring moisture retention, releasing active ingredients in a controlled manner, and contributing to extending the shelf life of products [38].

The droplet contact angle provides an indirect measure of surface wettability. All samples showed values below 40°, indicating hydrophilic surfaces, as expected for cellulosic materials, which usually exhibit angles in the range of 17 to 47° [39,40]. E3 exhibited the smallest angle (16.35 ± 1.62°) (p < 0.05), indicating a highly wettable surface, associated with the highest observed swelling rate, confirming that it is the most hydrophilic BC among those produced in the present work. E2 showed the largest angle (39.99 ± 0.64°) (p < 0.05), revealing the lowest surface affinity for water. Such a result may be related to the reduced number of accessible hydroxyl groups on the BC surface due to the adhesion of other components in the whey, limiting interaction with water. Samples E1, E4, EC, and EP showed intermediate angles, but still within the range of microbial cellulosic materials.

3.2.2. Fourier Transform Infrared (FTIR) Spectroscopy

Figure 4 shows structural changes resulting from variations in whey and sucrose concentrations.

The Fourier transform infrared spectroscopy spectra, obtained for all samples using the Attenuated Total Reflectance (ATR) accessory, showed the presence of bands characteristic of BC [41,42]. The spectral similarities indicate that the fermentation process maintained the integrity of the cellulose polysaccharide network, regardless of the formulation used, consistent with the result found by Revin et al. [20], who did not observe any significant effect of whey on BC chemical structure [42,43]. The main bands observed, as well as their respective vibrational assignments, are summarized in

Table 5. Main bands observed in the FTIR-ATR spectra of bacterial cellulose (BC) membranes and their assignments.

Wavenumber (cm−1)	Assignment	Meaning in the BC Context
3300	O–H/N–H stretching	Cellulose hydroxyls, bound water, and the possible contribution of whey proteins
2894	C–H stretching	Aliphatic chains (whey lipids and organic compounds)
1640	Amide I (C=O)/H–O–H	Adsorbed water
1427	C–H/COO− deformation	Aliphatic structures and possible carboxylates
1160	C–O–C stretching	Polysaccharide structure of BC

.

The bands at ~3300 cm⁻¹ (O–H/N–H) and 1160 cm⁻¹ (C–O–C) remain well-defined in all samples, reflecting the predominance of the polysaccharide matrix of BC. The observed punctual variations in intensity, especially in formulations with a higher proportion of whey, are mainly related to differences in physical interaction and water retention, and not to chemical alterations of the cellulosic chain [44,45]. In E2, a relative intensification of the bands occurred around 2920 and 2850 cm⁻¹, which can be assigned to the asymmetric and symmetric stretching of the CH₂ of residual organic compounds originating from the culture medium [46]. The bands located at ~2894 cm⁻¹ (aliphatic C–H) and ~1427 cm⁻¹ (C–H/COO⁻ deformation) show more evident variations in this sample, suggesting the presence of surface-associated aliphatic organic components [44,45]. Furthermore, the absence of sucrose may have promoted the adhesion of these additional components to the matrix, considering that run E4 was performed at higher concentrations of whey and sucrose and did not show this peak. Thus, the absence of easily assimilated sugars may cause other components present in the medium to aggregate to the membrane. In EP, these bands are less intense, highlighting the direct influence of whey on the surface composition of the formed material [41].

The band around 1640 cm⁻¹, attributed to water deformation, also shows differences in intensity among the samples. Its progressive increase in formulations with lower whey contents indicates a probable greater interaction of water with the BC network. This behavior suggests a favorable interaction between the polysaccharide matrix and the biomolecules present in the fermentation medium, even after the alkaline purification process. In samples with higher whey concentration, this band is less pronounced, reinforcing the lower presence of hydrophilic groups on the surface [20].

3.2.3. Thermogravimetric Analysis (TGA)

Thermogravimetric analysis (TGA) curves (Figure 5;

Table 6. Degradation temperature of bacterial cellulose samples prepared in different systems. E1–E4: different concentrations of cheese whey and sucrose; EC: concentrations of cheese whey and sucrose at the central point. E1: 40% (v/v) whey and 0 g·L⁻¹ sucrose; E2: 100% (v/v) whey and 0 g·L⁻¹ sucrose; E3: 40% (v/v) whey and 50 g·L⁻¹ sucrose; E4: 100% (v/v) whey and 50 g·L⁻¹ sucrose; and EC (corresponding to samples E5, E6 and E7): 70% (v/v) whey and 25 g·L⁻¹ sucrose.

Sample	Stage 1			Stage 2			Stage 3			Mass Loss at 600 °C (%)
	Tmax	Tonset	Tendset	Tmax	Tonset	Tendset	Tmax	Tonset	Tendset
E1	140.3	30.2	158.7	350.7	160.5	380	517.2	430.2	599.3	95.7
E2	154.3	32.3	159	357.7	161.2	380	520.2	430	599.7	90.6
E3	140.3	30.2	159.5	356.2	161.2	378.2	515.2	430	599.7	83.6
E4	135.3	30.2	156.5	361.7	160.2	380	515.2	430.5	599.3	81.2
EC	140.3	30.2	159.7	343.2	161.2	379.2	514.7	430	599.7	83.5
EP	135.3	32.7	158.8	356.2	160.7	380	526.7	430.2	598.8	82.5

) reveal a characteristic thermal profile of cellulosic materials, with three distinct stages of mass loss. The first corresponds to the elimination of adsorbed and structural water at low temperatures; the second reflects the main degradation of the cellulose chain; and the third involves the decomposition of the carbonized residue and more stable fractions at higher temperatures, a behavior widely described for BC and its nanocrystalline derivatives [47]. In the first stage, common to all formulations, it is observed that the onset T and endset T values are in the ranges 30–33 °C and 156–160 °C, respectively, with T max varying from 135.3 to 154.3 °C. This range confirms that the main contribution in this interval is associated with the loss of free water and water bound by hydrogen bonds to the cellulose fibril network. Sample E2 exhibits the highest Tmax value (154.3 °C), suggesting slightly stronger water retention, possibly due to differences in fibrillar network packing and pore distribution provided by specific cultivation conditions. From an application standpoint, the fact that all samples maintain structural stability well above 100 °C indicates that the membrane can withstand usual drying and mild sterilization processes employed in packaging systems without degradation, consistent with reviews that highlight the high thermal resistance of BC compared to other biopolymers used in food technologies [48,49].

The second stage, approximately between 160 and 380 °C, corresponds to the main degradation of the cellulose matrix, involving depolymerization, dehydration of anhydro-glucose units, rupture of β-1,4 linkages, and formation of volatile species concomitant with the generation of a more stable carbonized residue [47]. In this range, the T max values vary between 343.2 °C (EC) and 361.7 °C (E4), while T onset remains very similar for all samples (~160 °C), showing that the beginning of degradation occurs at similar temperatures, but that the maximum mass loss rate is modulated by the microstructure developed under each cultivation condition. E4 membrane has the highest T max value in stage 2 (361.7 °C), indicating greater thermal stability of the cellulose matrix. E2, E3, and EP show intermediate values (~356–358 °C), consistent with BC produced in conventional media and reported in the literature, with main degradation peaks between 330 and 360 °C. Finally, the EC sample shows a slightly lower value of Tmax (343.2 °C), suggesting that its structure decomposes slightly earlier. These values are fully within the range reported for BC membranes and nanocomposites with high crystallinity, developed for packaging and reinforcement applications in polymers [47,48,49].

In the third stage, delimited between approximately 430 and 600 °C, the decomposition of the carbonized residue and more thermally stable fractions occurs, as well as the contribution of inorganic components remaining from the culture medium. The Tmax values are between approximately 514.7 and 526.7 °C, with T onset around 430 °C for all samples, which reveals a fairly homogeneous behavior in this temperature range and confirms the presence of a relatively stable carbonized residue. The most striking differences appear in the residual mass at 600 °C: while E1 shows the greatest mass loss (95.7%), the residue progressively grew to 9.4% in E2, 16.0–16.5% in E3, EC, and EP, and even 18.8% in E4. This increase in residual mass from E1 to E4 suggests that formulations with a higher proportion of whey and sucrose lead to BC with a higher content of non-volatile material, either in the form of more structured carbonized residue or through a greater contribution of mineral components originating from the whey. Similar trends are reported in studies evaluating BC produced in alternative nutrient-rich media, in which differences in morphology, fibrillar packing density, and crystalline phase content impact both thermal stability and residue content at high temperatures [50,51]. In systems aimed at applications in packaging materials and composites, the presence of a stable carbonized residue at high temperatures has been associated with better overall thermal resistance and, in some cases, more favorable behavior in relation to flame propagation, which reinforces the relevance of the results observed for formulation E4, which is more thermally robust [48,49,52].

3.2.4. X-Ray Diffraction (XRD)

X-ray diffraction (XRD) analysis of BC produced in the different formulations (Figure 6) revealed characteristic profiles of type I cellulose, with three clearly identifiable main peaks near 2θ ≈ 14–16°, 22–23° and 34–35°, which can be attributed, respectively, to the crystallographic planes (1–10/110), (200) and (004), as also reported in other studies on BC [53,54].

The 2θ values and respective crystallographic assignments are also in agreement with results reported for BC produced by SCOBY in green tea-based media (2θ = 14.2°, 16.8°, and 22.3°) [39]. The crystalline structure of cellulose I is composed of highly ordered microfibrils, conferring high mechanical rigidity and lower water accessibility to the crystalline regions.

The compositional variation in the medium consistently impacted the degree of crystalline ordering, which was corroborated by the calculated values of the crystallinity index (CrI) (

Table 7. Crystallinity index values of the samples.

Sample	CrI (%)
E1	75.96
E2	63.02
E3	76.00
E4	57.30
EC	68.85
EP	62.27

). Sample E1 maintained the most intense peak (200), associated with the highest CrI value (75.96%), suggesting that moderate concentrations of whey constituents favor the biosynthesis of more well-oriented microfibrils. On the other hand, increasing the whey concentration to 100 g·L⁻¹ (E2 and E4, with CrI of 63.02% and 57.30%, respectively) correlated with reductions in CrI. This behavior may be related to metabolic effects induced by the greater availability of nutrients derived from dairy products, such as changes in carbon flow and in the production of organic acids (metabolism of lactose, hexoses, and monocarboxylic acids), which may influence the kinetics of cellulose polymerization, fibril packing, increasing the apparent amorphous fraction [38,55].

The addition of sucrose (E3, E4, and EC) also influenced the crystallinity degree. Although E3 showed a high CrI (76.00%), lower values were observed for E4 (57.30%) and EC (68.85%), indicating that higher sucrose availability may accelerate cellulose biosynthesis, leading to less ordered fibril assembly due to kinetic limitations during polymer deposition. This behavior is aligned with the literature, which describes that high sucrose supplementation can intensify metabolic activity and generate byproducts that affect fibril organization, rather than directly modifying cellulose chemistry [56].

Kolesovs & Semjonovs [50] and Revin et al. [20] reported that BC produced in whey generally exhibits lower crystallinity (~50%) than that prepared in HS medium (~79%), a trend attributed to metabolic modulation arising from complex nutrient matrices. The present results follow this trend, reinforcing that carbon source complexity and metabolic balance play a decisive role in determining crystalline ordering during BC biosynthesis [57]. Comparable trends were described by Salari et al. [58], who observed a reduction in crystallinity (from 79.07% to 61.86%) associated with the presence of galactose in enzymatically treated whey. However, it has also been reported that the extent of this effect is strain-dependent, with some microorganisms exhibiting changes in yield without proportional alterations in crystallinity [56].

In these systems, adjustments to the medium composition, such as the addition of sucrose, can mitigate the structural effects associated with lower crystallinity, contributing to preserving or even improving mechanical properties. According to Revin et al. [20], BC synthesized in cheese whey tends to have fibrils with slightly larger diameters, which can attenuate or even mask the direct influence of crystallinity on water retention and other hygroscopic behaviors. Additionally, the presence of organic acids in the medium has a direct impact on the structural organization of the cellulosic network. As reported by Chen et al. [38], treatment with organic acids preferentially affects the amorphous surface regions, reducing the crystallinity of BC. Thus, the formation of acids during fermentation may have intensified these structural changes, simultaneously modulating crystallinity and other properties of the material.

3.2.5. Transparency and Scanning Electron Microscopy

The dry membranes (Figure 7) exhibited color patterns consistent with their wet counterparts. Higher whey proportions yielded clearer, more homogeneous bacterial cellulose membranes, whereas formulations containing black tea produced darker, more heterogeneous surfaces, consistent with literature reports on the influence of phenolic compounds from plant extracts on membrane coloration and uniformity [50,59].

E4 showed the highest transparency (25.18 ± 0.74%; p < 0.05), followed by EC (17.24 ± 0.20%) and E2 (16.22 ± 0.11%), indicating less light scattering and therefore a more uniform matrix. Lin et al. [60] reported the presence of greater transparency in cellulose membranes with more compacted fibers and smooth surfaces, similar to that found in the present work. In contrast, E1 (8.89 ± 0.56%) and E3 (3.56 ± 0.05%) exhibited lower values, suggesting greater internal heterogeneity, possible retention of residual components, and greater optical scattering associated with irregular fibrillar networks. These results reinforce that the interaction between cheese whey and sucrose, as well as between whey and tea, directly affects the microstructural organization of BC and, consequently, its optical properties.

This microstructure can be visualized in Figure 8, from images generated by scanning electron microscopy (SEM) at a magnification of 5000×. E1 and E3 exhibited relatively flat surfaces with distinct microcracks, which are often linked to drying-induced stresses. These samples exhibited fiber diameters predominantly in the nanometric range, varying from 60 to 132 nm for E1 and 60–90 nm for E3, which is consistent with the formation of thinner and less compacted microfibrillar networks. Similar results have been reported in whey protein isolate (WPI) films with low plasticizer content, which tend to exhibit surface cracks [61,62,63].

Exhibited a more detailed morphology, with larger fibrillar aggregates and fiber diameters ranging from about 250 to 800 nm, indicating the formation of thicker and more diverse structures compared to other conditions. These features may reflect the aggregation of residual organic components from whey, especially in the absence of sucrose, similar to observations in protein films with limited network stabilization [51,61]. Even so, it is possible to identify interwoven fibers between the aggregates, which confirms the formation of the cellulosic network.

The EC sample (tea only) showed less compacted fibers and open regions, which highlights the fragility of the network with the smallest fiber diameters among all conditions, ranging from 25 to 50 nm, which is indicative of a loosely organized microfibrillar structure. Looser and more heterogeneous structures are typical of matrices with low availability of structuring macromolecules, as observed in less stabilized protein composites [62,64]. The E4 BC showed the most compact surface, characterized by densely deposited fibers with diameters in the range of 50–100 nm, which are comparable to those observed for E1, E3 and BC microfibrils produced in HS medium (60–90 nm) [20]. This corroborates the report by Catarino et al. [55], in which higher whey contents may contribute to a buffering effect of the fermentation medium, helping to maintain conditions favorable to BC production and ensuring more uniform fiber formation and structural stability, rather than altering the intrinsic dimensions of the microfibrils.

The comparison between E2 and E4 reinforces the fundamental role of sucrose in modulating fiber formation and surface organization. While E2 presented surface particles and aggregates associated with thicker fibers and retention of organic molecules in a poorly stabilized matrix, E4 exhibited a cohesive, compact, and continuous microstructure, consistent with sucrose’s ability to favor intermolecular interactions and reduce surface disorganization [59,62]. This contrast is even more evident in Figure 9, which presents higher magnification (500×) images of the membrane surfaces.

3.2.6. Mechanical Tests

BC samples exhibited high flexibility, maintaining structural integrity even after being manually bent for more than 100 cycles at the same point. This characteristic is crucial for applications in primary packaging and bioactive systems, favoring adaptation to the food surface and efficiency in the controlled release of preservative compounds [26]. As described by Wu et al. [65], such high flexibility stems from the nanofibrillar structure of BC, which allows repetitive deformation without loss of cohesion or mechanical worsening. Tensile strength showed wide variation depending on the tested conditions, reflecting intrinsic differences in fibrillar density, degree of crystallinity, and organization of the nanocellulosic network (Figure 10). These results demonstrate that the composition of the culture medium directly influences the mechanical architecture of the membranes formed.

Samples E1, E2, and E3 showed similar tensile strength values (6.56 ± 1.15 Mpa, 6.46 ± 0.01 Mpa, and 6.35 ± 2.30 Mpa, respectively) but had distinct mechanical behaviors due to structural differences. E2 and E3 showed similar behavior, exhibiting lower specific deformation and a high modulus of elasticity, characterizing them as more brittle materials, possibly influenced by microstructural heterogeneities. E1 combined high strength with a lower modulus and deformation, indicating a less rigid network.

The E4 membrane stood out as the most mechanically robust material, consistent with SEM micrographs that showed a more densely packed network. Its tensile strength (13.43 ± 2.30 Mpa) is close to the values reported for BC grown in HS medium, while its modulus of elasticity (59.00 ± 2.50 MPa) far exceeds those of the other formulations [26]. The specific deformation (17.00 ± 1.67%) was also among the highest, indicating that the film combines stiffness, strength, and ductility, a characteristic profile of strongly interconnected microstructures. These results suggest that the lower crystallinity did not negatively impact mechanical performance and that the greater width of nanofibers may have contributed to pore filling, reinforcing the matrix, and increasing strength. Together, these factors give the E4 BC a behavior close to that of high-performance materials within the scope of biotechnological packaging.

The EC BC exhibited moderately low strength (4.35 ± 0.09 Mpa) and high stiffness (34.79 ± 0.73 MPa), resulting in a rigid but structurally brittle membrane with low deformation capacity (7.06 ± 0.15%). This behavior indicates a less integrated matrix, in which an increase in modulus does not translate into effective strength. The EP BC, on the other hand, showed the worst mechanical performance, with a strength of 3.18 ± 0.20 Mpa, a modulus of 26.89 ± 3.60 MPa, and intermediate deformation (11.07 ± 0.59%), reflecting a poorly consolidated microstructure, consistent with the SEM image, with a lower density of fibrillar interactions, which favors failures under stress.

This variability confirms the high sensitivity of BC to synthesis conditions and highlights the possibility of modulating its mechanical properties according to the intended application, allowing for transitions between more resistant or more flexible structures. As highlighted by Gao et al. [66], producing cellulose films that simultaneously combine high transparency and high mechanical strength remains a challenge. In the present study, the membrane prepared using the highest concentration of whey and sucrose stood out for allowing precisely this balance, combining greater transparency with greater strength, indicating a significant advance towards high-performance materials for packaging development.

4. Conclusions

The results confirm that cheese whey can be effectively used as a fermentation medium for the production of bacterial cellulose, with the highest dry mass yield obtained in the system containing 100% (v/v) whey and 50 g·L⁻¹ sucrose. Variations in the medium composition influenced yield and microstructural organization, preserving the chemical integrity of the cellulose matrix. The structural differences between the samples were mainly associated with fiber packing and surface organization. Furthermore, the reduction in crystallinity observed at higher whey concentrations reflects the influence of complex nutritional matrices on fibril assembly during biosynthesis, which, in turn, affects the water-related properties of the membranes. In this context, the findings provide relevant technical support for the development of bacterial cellulose-based materials aimed at applications in food technologies, especially in the formulation of films, coatings, and structuring systems obtained from agro-industrial substrates.