Food grade wheat gluten (WG), moisture content = 9.08 ± 0.10% (determined by drying WG in a convection oven at 120°C until constant weight was achieved), technical grade glycerol and thyme essential oil (TEO), extracted from the plant Thymus vulgaris, were purchased in local shops (Mar del Plata/Buenos Aires, Argentina) [18]. β-cyclodextrin (β-CD) (Cavamax W7 food grade) was provided by Wacker Biochem, Eddyville, IA, USA.

Encapsulation efficiency (EE) indicates the amount of active component entrapped in the encapsulation systems (i.e., load capacity) and is therefore directly related to the controlled release characteristics of the core material [28]. The EE value of TEO:β-CD microcapsules (Eq. (1)), was 72.22 ± 2.82%, which indicated the effective entrapment of thyme essential oil by the β-CD carrier. This value was comparable with those obtained in other studies using the same method of inclusion, such as 73.5% by loading lemongrass [28], 71.66% by embedding TEO [41] and 77.74% by loading clove into β-CD [38]. Besides, Tao and co-workers [19] measured an encapsulation efficiency of 74.57% for TEO in β-CD microcapsules obtained by freeze-drying.

The average diameter of TEO:β-CD microcapsules was found to be 7.3 ± 0.2 µm. These values are in the same range as those obtained by Tao and co-workers [19] and Chen and co-workers [28] by loading TEO and lemongrass into β-CD.

Scheme 1 summarizes the methodology applied to obtain active and control WG based films. From the image, it can be noticed that films are yellowish/light brown, optically opaque, and do not present notable visual differences that can be attributed to the incorporation of free or microencapsulated TEO.

Fig. 1 presents the thermogravimetric curves obtained for pure β-cyclodextrins, neat and microencapsulated TEO. The great volatility of the thyme essential oil can be recognized by the sudden loss of mass that occurs as the test temperature increases, resulting in the entire evaporation/degradation of the TEO at 150°C. The neat β-CDs show several stages of mass loss or decomposition, which are evidenced by a change in the slope of the TGA vs. temperature curve or, more clearly, as downward peaks in the DTG curve. The early one occurs in the temperature range of 50°C to 190°C and shows a relatively low mass loss of 10.4%, mainly attributed to the evaporation of water from the inclusion complexes, as indicated in related papers [42]. The major loss of mass was traced in the degradation stages following the loss of adsorbed water: 12.6% mass loss in the second stage (DTG peak between 190°C and 280°C), 46.5% mass loss in the third (280°C to 400°C, from DTG peak), and fourth stage (400°C to 600°C), in which the rest of the mass is lost, leaving almost no carbonaceous residue. The mass loss in these stages was attributed to the breakdown of the eight-membered cyclodextrin structure, as reported by Ahmed et al. [42]. On the other hand, the microencapsulated oil presents a mass loss pattern very similar to that of the β-CDs, clearly indicating that the encapsulated TEO degrades at a higher temperature and slower rate than neat TEO, which confirms the molecular encapsulation of the essential oil inside the β-CD cavity (i.e., complexation between essential oil and CDs), after the encapsulation process. Furthermore, the mass loss in the first (up to 190°C) and second (up to 280°C) stages of the β-CDs-TEO-15% sample are 12.3% and 14.2%, respectively, that is, higher than the mass lost by the neat β-CDs in the same steps. However, for the following heating steps this behavior reversed, showing in the third stage a 41.6% mass loss, lower than the 46.5% exhibited by the neat β-CDs, and leaving a low but noticeable amount of carbonaceous residue (1.5% at 620°C) in the last stage, indicating that the encapsulated β-CD retained higher masses compared to the neat β-CD powder [42]. This corroborates the usefulness of microencapsulation in preserving the volatile active principles of TEO against agents such as heat.

The morphology of the films incorporating free and encapsulated TEO was evaluated by scanning electron microscopy (SEM). As shown in Fig. 2, the fractured surfaces of films containing no or free TEO are smooth, neat and compact, and do not exhibit evidence of phase separation, which could indicate that free TEO is highly compatible and miscible with wheat gluten plasticized matrices. However, sample preparation for SEM observation, which involves a vacuum stage for covering the sample with a conductive metal, should also be considered before drawing conclusions. In fact, the free TEO incorporated into the wheat gluten films would probably evaporate during this stage and this would be the reason why the fracture surface of WG-Control (Fig. 2a) and those containing free essential oil (Fig. 2c,d) are so similar. On the other hand, the encapsulation into the β-CD should protect the TEO from evaporation during sample preparation for SEM observation. Although this statement cannot be verified from the SEM images, it is clear that the morphology of the fractured surfaces of these samples is different than the previous ones, which can be attributed to the presence of β-CD. In fact, films carrying β-CD (control) exhibit an irregular and rough fracture surface, in which the presence of relatively large aggregates embedded in the polymeric matrix stands out. These inclusions are attributed to a certain degree of β-CD aggregation that took place during the film formation process.

On the other hand, films containing β-CD-TEO (Fig. 2e,f) also exhibit more heterogeneous and uneven fracture surfaces than those corresponding to pure wheat gluten or films incorporated with free TEO (Fig. 2a–c). Even when no large agglomerates are noticed in the WG-β-CD-TEO-10% film, its surface exhibits a heterogeneous and rough morphology, evidenced by the presence of ridges and valleys that could have originated during the brittle fracture of the film, at points where β-CD inclusions diverted crack propagation toward an easier fracture path. Furthermore, the SEM image of the WG-β-CD-TEO-15% film also evidences that for higher loadings, a certain degree of β-CD aggregation occurred and therefore Fig. 2f combines the features of WG-β-CD-TEO-10% and WG-β-CD-control films, albeit with slightly smaller aggregates than the latter. This decrease in aggregate size can be attributed to the methodology used to prepare the TEO-β-CD complex, which involves dissolution, inclusion and co-precipitation steps. Similar trends were found in related papers [43–45].

Table 1 presents the equilibrium moisture content, the total soluble mass values obtained from two different methodologies (i.e., WET and DRY) and the water vapor permeability of the different wheat gluten films. The equilibrium moisture content decreases from about 46% for the glycerol plasticized wheat gluten (WG-Control) film to around 40% for the active films (i.e., samples containing thyme oil in any form), which can be ascribed to the slightly more hydrophobic nature of the TEO compared to plasticized WG, although this property does not show a clear trend with respect to thyme essential oil concentration. On the other hand, it is also clear that the incorporation of β-CD into the WG matrix also affects the equilibrium moisture content, since the WG-β-CD-control sample exhibits the lowest value of the entire series, while it does not improve the performance of the TEO (i.e., β-CDs do not act by enhancing the reduction of adsorbed moisture when the oil is microencapsulated). Regarding total soluble matter, when taking into account the standard deviation of the WET TSM results, it becomes evident that the content of free thyme oil almost does not affect the total soluble mass of the corresponding films when TSM is determined from samples not previously dried. Otherwise, adding β-CDs to the protein matrix increases the solubility of the resulting films largely when the TEO is not microencapsulated into them. The microencapsulation of thyme oil leads to a reduction in this parameter, but the active films still remain more soluble than the WG-Control sample. Evidently, the amount of active ingredient, the interaction developed among film components, and the consequential film microstructure, affect the solubility of the resulting films, as indicated in related works [43]. The values of the total soluble mass determined from dried films (DRY TSM) are smaller than those corresponding to the wet samples in all cases, supporting previous observations indicating that increased cross-linking of gluten proteins takes place during drying at 105°C [18,36]. However, this effect is more noticeable for the films without oil (control films) or containing free oil (reductions ranging from 0.54 to 0.66) than in those containing β-CDs (reductions ranging from 0.75 to 0.87), indicating that the cyclic oligosaccharides do not participate in the cross-linking reactions. In both cases, the influence of the essential oil, if any, is not clear. Similar results can be found in related papers [43].

Regarding water vapor permeability, the incorporation of free oil results in a slight (WG-TEO-10%) or a significant increase (WG-TEO-15%) in WVP values compared to that of the WG-Control sample, probably because its inclusion into the wheat gluten network leads to the reduction of the interactions between protein chains. On the other hand, the films incorporating microencapsulated TEO display significantly (WG-β-CD-TEO-10%) or slightly (WG-β-CD-TEO-15%) reduced WVP values with respect to that of the β-CD-control sample, that could be ascribed to the increase in network heterogeneity, leading to a more tortuous course for the passage of water vapor. However, it is clear that the increase in the content of microencapsulated TEO above 10% does not contribute to a further decrease in this property. This effect can be explained considering the changes introduced in the structure of the films by the β-CD-TEO inclusion complex when its amount is excessive, which, according to related works [43], leads to polymer networks with decreased intermolecular interactions between protein chains, which results in increased segmental motions as a result of the increased free volume of the system. In summary, moderate loading of microencapsulated TEO leads to a decrease in the WVP because the effect of increasing vapor path due to increased tortuosity predominates over the effect caused by increasing free volume of the system, but the opposite occurs for higher or excessive contents.

The optical properties of films are important characteristics that could determine the acceptability and applicability of foods in packaging systems. Light-permeable materials have absolute transparency as their characteristic optical property. The transparency of a package could affect the appearance and marketability of the food and therefore could determine the suitability of the film for various demands [46]. Therefore, the opacity values of the different wheat gluten-based films are shown in Table 2. Both control films exhibited lower opacity values than those containing free or microencapsulated TEO, which is attributed to the existence of an oil-dispersed phase in the films containing free essential oil that favors light scattering and therefore reduces transparency. On the other hand, samples including microencapsulated oil show a higher increase in opacity values compared to the analogous control samples, almost certainly due to both the greater opacity of the β-CD phase when it contains TEO, and the higher heterogeneity of the polymeric network, as was also noticed in other works [26].

The mechanical properties of the active wheat gluten films obtained from tensile tests were also studied. From the results presented in Table 2, it can be seen that the films containing free or microencapsulated TEO are less stiff than the corresponding control ones, since a decrease in tensile modulus (E) and strength (σ_m) as well as an increase in the elongation at break (ε_b) are observed as the TEO content increases. Evidently, the oily nature of the active component added to its less polar nature reduces the interactions developed between protein chains, leading to softer and more flexible films. The variations in the mechanical properties suggest that the polymeric network has a lower cohesion, almost certainly because the lipid phase partially hinders the formation of protein-protein intermolecular bonds, while favoring the formation of hydrophobic interactions during the drying stage of the film [47]. Regarding samples containing β-CD, it is evident that they have a less compact and less cohesive structure than films made only from plasticized wheat gluten, since these oligosaccharides also limit protein interactions, which become more evident when TEO is microencapsulated into them. Similar performance was noted in previous [18] and related works, i.e., wheat gluten films incorporating potassium sorbate obtained by compression molding [47]. The mechanical behavior of the films is influenced by the type and concentration of EO, the nature of the polymeric matrix and the specific interactions developed among components [48], which dictate the actual adhesion forces between the polymeric and oily phases. Usually, if immiscible and/or non-polymeric components such as essential oils or β-CDs (as in the present case), are included in excessive amounts into the polymeric matrix, the obtained films will comprise a large amount of dispersed phase, generating a structure that is too heterogeneous and contains a large number of discontinuities.

The dynamical mechanical behavior presented in Fig. 3 indicates that all materials exhibit a predominantly elastic character, since G′ higher than G″ over the entire temperature range studied. Moreover, these results also support the finding that TEO and β-CD behaves as co-plasticizers for the wheat gluten films, since their addition caused a reduction in both viscoelastic moduli that is more pronounced in the glassy region. This effect can again be attributed to the decrease of the protein-protein interactions due to the addition of both the TEO essential oil and the cyclic oligosaccharides [49], as discussed in previous sections. The glass transition temperature corresponding to the glycerol-rich phase (peaks in the loss modulus curves) decreased from 11°C for the control sample to about −6°C (films containing β-CD) and −12°C (films containing free TEO). Moreover, the curves of storage modulus clearly show that active wheat gluten films are less rigid than the control sample, although the observed trend does not exactly match the tensile behavior.

The concentration of thymol incorporated into film formulations prior to processing and the remained concentration after thermoplastic processing were calculated (data not shown). Results revealed a thymol retention after processing of approximately 71% when TEO was incorporated in its free form [18]. Encapsulation further reduces thymol losses, resulting in WG active films with ≥90% retention of thymol content after processing (93.1% and 90.2% for WG-β-CD-TEO-10% and WG-β-CD-TEO-15%, respectively). Thymol losses are mainly due to evaporation or degradation of the bioactive during high temperature and high pressure film processing. Similar results were obtained by Ramos et al. [50] by including thymol to polylactic acid (PLA) films prepared by compression molding and by Hwang et al. [51] by including resveratrol and α-tocopherol into poly(lactic acid)/starch blend films.

The antimicrobial activity of control and active WG films incorporated with free and microencapsulated TEO against selected pathogens is presented in Table 3. Control films were not effective against any of the tested bacteria, while both microencapsulation and TEO concentration affect significantly the antimicrobial performance of the active films (p < 0.05). As higher the TEO concentration (in any form), the larger is the film antimicrobial activity. The antimicrobial activity of thyme essential oil has been well demonstrated and is directly related to the synergistic effect of its main components thymol and carvacrol [19,52], as stated above. The action mechanism of TEO could be associated to the lipophilic characteristics of these major compounds, which attach themselves to the phospholipid bilayer of the cell membrane, increasing its permeability, therefore causing damage to the cellular enzymatic system by altering the cytoplasmic structure [53].

Wheat gluten films with encapsulated TEO presented higher inhibition zones compared to those containing free TEO oil against all pathogenic bacterial strains tested, and were also significantly less effective against Gram (−) bacteria than against Gram (+) bacteria, as also found in related works [54]. These results can be attributed to the higher thymol retention after film processing due to encapsulation, which sustains their antimicrobial activity. The use of encapsulated thymol for preserving the antimicrobial activity of films against E.coli and S.aureus was also reported by Yuan and co-workers by developing starch based composite films incorporated with thyme essential oil microcapsules and/or microemulsion [55].

Fig. 4 shows the antioxidant capacity and total phenolic content of WG films containing free and encapsulated TEO. The DPPH (2,2-diphenyl-1-picrylhydrazyl) radical scavenging rate of control films was close to zero, suggesting these films had no antioxidant capacity. Also in this case, both the encapsulation and concentration of TEO significantly affected the total polyphenol content and the antioxidant capacity of the active films. The antioxidant performance of WG films significantly increased (p < 0.05) as the concentration of TEO (incorporated in both free and microencapsulated forms) increases. On the other hand, the DPPH radical scavenging activity of films containing encapsulated TEO was significantly higher than that of samples prepared with free TEO, denoting that encapsulation preserves TEO from light, heat and pressure during film processing, improving the stability of the essential oil and therefore boosting the action of the antioxidant polyphenols contained in it. The use of microencapsulation for preserving the antioxidant activity of films was reported by other authors for starch films developed by casting and incorporated with black tea extract encapsulated in a pectin-sodium caseinate mixture [56], polyvinyl alcohol (PVA)-starch films obtained by casting and incorporated with β-cyclodextrin inclusion complex embedding lemongrass oil [28] and PVA films obtained by casting and incorporated with tea polyphenols-loaded starch microcapsules [57].

The excellent antioxidant activity of EOs, mainly attributed to its phenolic compounds content, has been reviewed by Adel et al. [23] and Bai et al. [30]. As reported in a previous publication [18], the total phenolic content of the pure thyme essential oil used in this work is high, 74.54 mg GAE/g and therefore it is not surprising that the TPC in the WG films increases significantly (p < 0.05) as the thyme essential oil concentration increases, in agreement with the results obtained for the free radical scavenging activity.

Fig. 5 shows the cumulative release of thymol as a function of time obtained from compression-molded active WG films containing free and microencapsulated TEO and immersed in 50% ethanol. The release rate showed no significant differences between films with different TEO concentrations for both types of films, i.e., containing free or encapsulated thyme essential oil. Moreover, in the release curves of the active component, two stages can be clearly distinguished: burst or fast release (0–25 h) and sustained slow release (25–360 h). The performance of the encapsulated TEO during the burst period could be ascribed to the presence of unencapsulated thymol in the surface of the β-CD microcapsules as well as to the part of the active component that is located very close to the inner surface of the capsules and therefore could be released in the early stages of the assay [39]. Furthermore, release kinetics showed that thymol release from WG films in the food simulant system is more sustained when TEO is incorporated as an inclusion complex compared to adding TEO directly into the films. The cumulative release rates at 24 h from the active WG films were approximately 73% and 65% for films containing free and microencapsulated TEO, respectively. Alinaqi and co-workers [58] reported results in agreement with those found in this work: a release of approximately 60% of clove essential oil encapsulated in zein nanoparticles and incorporated in a starch bionanocomposite film after 24 h using 50% ethanol as food simulant. Film formulations containing free TEO showed a more accelerated release, reaching equilibrium release rates of 95% (for both TEO concentrations) within 72 h, indicating that almost all the bioactive was released from the films after 3 days. On the other hand, encapsulated TEO was released from the active films in a more controlled way, compared to free TEO, supporting that microencapsulation of the active agent can delay the liberation rate of TEO. In this way, the TEO can be maintained into the film for longer periods of time, thus enabling the packaging to stay active for much longer. Similar results were also reported for whey protein edible films containing carvacrol and eugenol β-cyclodextrin inclusion complexes [59] and for gelatin films incorporated with β-cyclodextrin/curcumin complexes [60].

Release of compounds with low molecular weight from a polymeric matrix can be described in terms of the following stages: (1) diffusion of the solvent from the external solution into the polymer network; (2) swelling/relaxation of the polymeric network; (3) diffusion of the compound from the swollen macromolecular matrix to the external solvent solution [61,62]. In this case, once the WG matrix is swollen, the encapsulated thymol must first be released from the β-CD inclusion complexes, then migrate to the polymeric matrix and finally, pass through the film into the ethanol simulant until both phases reach thermodynamic equilibrium.

The experimental release results were fitted to short-term (Eq. (5)) and long-term (Eq. (6)) diffusion models. Both models fitted the experimental data quite well (R² > 0.9), which therefore allowed the calculation of the diffusion coefficients (D) and release rate constants (k) presented in Table 4. Wheat gluten films containing encapsulated TEO presented lower D and k values (42% and 35%, respectively) compared to those exhibited by films prepared with free TEO, regardless the amount of TEO incorporated into the films, confirming that β-CD had a significant slow-release effect on the thyme essential oil.