The bovine collagen fiber was supplied by Novapron Food Ingredients (São Paulo, Brazil). Polyvinyl alcohol was purchased from Kuraray (São Paulo, Brazil). Essential oils were donated by doTERRA (São Paulo, Brazil). Other reagents, all of analytical quality, included brain heart infusion (BHI) agar (Kasvi, Spain), sodium chloride (Synth, Brazil), acetic acid (Vetec, Brazil), Tween 80 (InduLab, Brazil), glycerol (Dinâmica, Brazil), potassium nitrate (Dinâmica, Brazil), chloroform (Dinâmica, Brazil), peptone water (Kasvi, Spain), 1,1,3,3-Tetraethoxypropane (Sigma Aldrich, USA), thiorbituric acid (Merck, Germany), trichloroacetic acid (Dinâmica, Brazil), sulfanilamide (Vetec, Brazil), butylated hydroxytoluene (Vetec, Brazil), 2,2-diphenyl-1-(2,4,6-trinitrophenyl)hydrazin-1-yl (DPPH) (Sigma Aldrich, USA), and eosin methylene blue agar (Kasvi, Spain).

The compounds in the EOs were characterized according to the method described by Thiel et al. [20] . The chemical characterization of the essential oils was performed by gas chromatography coupled to mass spectrometry (GC/MS Shimadzu QP2010 Plus; Shimadzu Corporation, Tokyo, JP). It used a ZB-5 MS capillary column (30 m × 0.25 m; 0.25 mm, flow rate 1 mL/min), and ramp conditions were 40˚C for 1 min and 230˚C at a rate of 4˚C/min for 3 min and the injector at 250˚C. Identification was made by comparing mass spectra from the National Institute of Standards and Technology (NIST) and experimental LRI with those available in the literature.

The films were produced using the method proposed by Thiel et al. [20] . Film-forming solution: collagen fiber (5.0% w/v), polyvinyl alcohol (5.0% w/v), Tween 80 (1.0%), and glycerol (1.30%). The pH is corrected to 3.0. The solution was homogenized and heated at 90˚C for 20 min. After the heating time, 2%, 4%, and 6% concentrations of the OEO or REO were added. Each film-forming solution was inverted in glass plates (29.7 × 21.0 cm) and placed in an oven with air circulation for roughly 24 h. The films were then stored until the respective analyses in desiccators (RH 60% and 25˚C).

Each film was cut into smaller pieces and weighed at 0.3 g, and 1 mL of chloroform was added. The vials were then sealed and placed in a water bath at 30˚C for 1 hour. After this period, the extract was removed and transferred to vials. After removing the extract, an additional 1 mL of chloroform was added, and the procedure was repeated. The chemical characterization was conducted in accordance with Section 2.2.

The samples were cut, weighed to 0.5 g, and immersed in 50 mL of 95% ethyl alcohol for 4 days. Aliquots of 2 mL were then mixed with 2 mL of DPPH solution in 95% ethyl alcohol (75 µM). After stirring, the solutions were left in the dark for 30 minutes. The absorbance was subsequently measured at 570 nm using a UV–vis spectrophotometer (BEL-UV-M51, BEL Equipment, Italy) [21] . The antioxidant activity was quantified as the percentage inhibition of the DPPH radical (Equation (1)).

I (%) = AC − Asample/AC × 100(1)

where AC is the absorbance of the DPPH solution, and Asample is the absorbance of the sample solution with DPPH.

Tensile strength (TS) (MPa), elongation at break (EB) (%), tenacity (T) (MJ/M³), and Young’s modulus (YM) (MPa) were determined according to the ASTM D882-18 method [22] . The TA.XT plus texture analyzer (Stable Micro Systems, UK) was utilized. The films were cut into dimensions of 20 × 80 mm, with a claw distance of 30 mm and a speed of 8.33 mm/s. The thickness of the films was measured at the time of analysis using a digital micrometer (Digimess, Brazil).

The sliced mortadella was purchased in sealed 1 kg packages from a local shop in Santa Maria (RS/Brazil), and the same brand was consistently used throughout the experiment. However, different batches were utilized for each repetition.

The films, measuring 310 × 200 mm, were folded in half and sealed at both ends; the slices were then placed inside, and the end was heat-sealed at 60˚C for 5 seconds using a Cetro PFS200 Manual Sealer. The width of the sealing area measured 0.6 mm, resulting in the final package dimensions of 155 × 100 mm ( Figure 1 ). The slices, measuring 7.1 mm and 1.8 mm in thickness, were placed in the package, totaling 14 slices per package. The packages were stored at 7˚C for 9 days, and samples were collected on days 0, 3, 6, and 9 to analyze moisture content, water activity, pH, color, and thiobarbituric acid reactive substances (TBARS). Before conducting the analyses, the samples were ground for 10 seconds using a Philips Walita RI2622 mini processor.

The moisture content [23] and water activity of the products were assessed using an Aqualab Series 4 TEV (Decagon Devices, USA) at a temperature of 20˚C. A digital potentiometer (Tecnopon Mpa 210) facilitated the electrometric analysis of pH. The analysis of color employed the CIE L^*, a^*, and b^* color space, with six readings taken using a CM700D colorimetric spectrophotometer (Konica Minolta Sensing, Japan). This analysis was conducted utilizing a D65 illuminant and a 10˚ observation angle, with the specular component excluded and the measurement area set to 8 mm. Lipid oxidation was evaluated through the TBARS assay, following the methodology described by Bruna et al. [24] with modifications. For that, 5 g samples were homogenized in 1 mL of 0.15% BHT, 1 mL of 0.5% Sulfanilamide, and 20 mL of 5% Trichloroacetic Acid. Homogenize 3 times for 20 seconds by vortex. The homogenate was centrifuged at 4˚C for 10 minutes at 3000 rpm, and filter the solution. An aliquot of 2 mL of the filtrate is placed in a tube with 2 mL of TBA. Homogenize in vortex for 20 seconds and heated at 40˚C/80 minutes in an ultra-thermostatic bath (Solab, Model SL 152) after cooling (20˚C) by immersion in water. Finally, the absorbance was measured at 532 nm (BEL-UV-M51, BEL Equipment, Italy). The results were expressed in milligrams of malondialdehyde per kilogram (mg MDA/kg). A standard curve using 1,1,3,3-tetraethoxypropane was established for each analysis.

1) Preparation of the bacteria

The E. coli strain (ATCC 8739) was stored at freezing conditions (−18˚C) in a mixture of tryptone soy broth and glycerol, serving as stock cultures until required for the experiment. Approximately 24 hours prior to analysis, the strain was reactivated by inoculation into a tube containing BHI agar, followed by incubation in a microbiological oven at 37˚C. Subsequently, a cell suspension (10⁸ CFU/mL) was introduced, achieving a final concentration of approximately 10⁶ CFU/mL.

2) Sample preparation in the active packaging

The slices of mortadella were exposed for 20 minutes to UV light from a 30 W Germicidal Lamp (diameter 26.0 mm, length 893.0 mm) inside a laminar flow chamber (TROX TFL, Brazil), with the procedure repeated for both sides of the slices [25] . Following this, an aliquot of 100 µL from the adjusted solution containing 10⁶ CFU/mL of the E. coli strain was inoculated onto each side of the meat product slices. After 20 minutes, the slices were placed in their respective packaging, which varied by EO (Essential Oil) concentration (0%, 2%, 4%, and 6% OEO (Oregano Essential Oil) or REO (Rosemary Essential Oil), were heat-sealed using a Cetro PFS200 Manual Sealer, and then immediately stored under refrigeration at 7˚C.

3) E. coli count

The samples were diluted in peptone water (0.1%), homogenized using a stomacher (BagMixer 400 CC, Interscience, France), and subjected to decimal dilutions ranging from 10⁻¹ to 10⁻¹⁰. Subsequently, aliquots were inoculated onto eosin methylene blue agar plates, and the resulting bacterial counts were expressed as log (CFU/g) [25] . This procedure was performed at 0, 3, 6, and 9 days of storage at 7˚C).

The qualitative method proposed by Jaramillo et al. [26] was employed. Vegetable compost was poured into a tray measuring 60 cm by 60 cm. The films, each measuring 2 cm by 3 cm, were buried in the soil to a depth of 2 cm. They were maintained at room temperature, and approximately 250 mL of water was sprayed daily to preserve humidity. The samples were carefully removed and photographed on days 0, 3, 6, and 9.

The experimental design was completely randomized with a factorial arrangement of 2 × 4 × 4 (2 types of EOs × 4 levels of EOs × 4 storage times), totaling 32 treatments with three replicates each, according to Equation (2):

$Y_{ijkl}=\mu +{\alpha }_i+{\beta }_j+{\gamma }_k+{\left(\alpha \beta \right)}_{ij}+{\left(\alpha \gamma \right)}_{ik}+{\left(\beta \gamma \right)}_{jk}+{\left(\alpha \beta \gamma \right)}_{ijk}+{\epsilon }_{ijkl}$ (2)

where $Y_{ijkl}$ is the value observed in the i-th type, j-th level, k-th time, and l-th repetition, μ is the general average of the response variable; α_i is the fixed effect of the i-th type, ${\beta }_j$ is the fixed effect of the j-th level, ${\gamma }_k$ is the fixed effect of the k-th time; ${\left(\alpha \beta \right)}_{ij}$ = is the fixed effect of interaction between type and level; ${\left(\alpha \gamma \right)}_{ik}$ = is the fixed effect of interaction between type and time, ${\left(\beta \gamma \right)}_{jk}$ is the fixed effect of interaction between level and time ${\left(\alpha \beta \gamma \right)}_{ijk}$ = is the fixed effect of interaction between type, level, and time, and ${\epsilon }_{ijkl}$ = is the random effect associated with observation $Y_{ijkl}$, assuming ${\epsilon }_{ijkl}{}_{~}^{iid}N\left(0,{\sigma }^2\right)$.

For the mechanical analyses, antioxidant activity, and volatile compounds, the design was completely randomized with a factorial arrangement of 2 × 4 (2 types of EO × 4 levels of EO), resulting in a total of 8 treatments with five replicates for mechanical analyses and three replicates for antioxidant activity and volatile compounds for each treatment.

The data were subjected to a univariate analysis of variance (ANOVA) using the general linear model (GLM); their means were adjusted using the ordinary least squares method with the LSMEANS statement and compared using the Student Newman Keuls test. Additionally, the levels’ and times’ linear and quadratic trends were evaluated using orthogonal contrast analysis based on the coefficients for interpolating the orthogonal polynomials, estimated using the Interactive Matrix Language (IML) procedure.

Polynomial regression analysis was then conducted to investigate changes in dependent variables as a function of the quantitative independent variables (level and time). The parameters of the regression models were estimated using the GLM (simple) and response surface regression—RSREG (multiple) procedures. The values of the coefficient of determination (R²) were expressed in relation to the source treatments (regression + lack of fit).

Subsequently, a multivariate analysis of variance (MANOVA) was conducted, wherein the matrices of sums of squares and products were obtained: T, H, and E, where T refers to total, H to treatments, and E to residual. The Wilks (λ), Pillai (V), Hotelling-Lawley (U), and Roy (F₀) tests were used to test the hypothesis that the treatment mean vectors were null.

In addition, a cluster analysis of the treatments was conducted using the DISTANCE, CLUSTER, and TREE procedures, employing the average Euclidean distance as a measure of dissimilarity and Ward’s method as a clustering method. A cluster analysis of the dependent variables was also conducted using the VARCLUS and TREE procedures, using the correlation matrix as input.

Principal Component Analysis (PCA) was performed using the PRINQUAL, PRINCOMP, and FACTOR procedures [27] . Statistical analyses were performed on the SAS analytical platform Viya for Learners v. 4.0 (SAS Institute, USA) [28] at a 5% significance level.

In the analysis of oregano EO (Essential Oil), 17 compounds were identified ( Table 1 ). The predominant compound found was carvacrol, followed by p-cymene, a finding that corroborates the results reported by Lee et al. [29] . Notably, carvacrol, thymol, γ-terpinene, and linalool exhibit antioxidant activities. Furthermore, carvacrol is recognized in the literature for its comprehensive antimicrobial activity [30] .

For rosemary EO, 26 compounds were identified. This composition aligns with the findings of Giarratana et al. [31] . The literature presents varying accounts of the biological activities of rosemary EO, reporting intermediate antioxidant and antimicrobial activity [32] . The biological effects are primarily attributed to the compounds 1,8-cineole and camphor. However, a study by Ribeiro-Santos et al. [33] did not report any activity against E. coli and Staphylococcus aureus, with the primary compounds identified in the EO being $\alpha$-pinene and 1,8-cineole.

There was a significant effect (p < 0.05) of the level of EO (Essential Oil) on TS (Tensile strength), EB (Elongation at break), and YM (Young’s modulus). Additionally, the type of EO significantly affected YM (p < 0.05). In contrast, no significant difference was observed for T (Tenacity) (p > 0.05). The increase in EOs led to a reduction (p < 0.05) in TS and YM and an increase (p < 0.05) in EB ( Figure 2 ). Moreover, rosemary EO exhibited a significantly higher YM (p < 0.05) compared to oregano.

Zhang et al. [34] reported a reduction in TS and an increase in EB with the addition of thyme EO to polyvinyl alcohol and curdlan films. Incorporating EO promotes a reduction in polymer-polymer interactions, thereby facilitating the movement of the polymer chains and consequently increasing the film’s flexibility [34] . Conversely, Shen et al. [35] observed decreased EB values upon adding oregano EO to starch-based films, highlighting that the film matrix and its interaction with the EO significantly influence EB enhancement. Typical food packaging materials generally possess a TS of 5 - 15 MPa and an EB of approximately 200%. The films examined in this study exhibited higher values, thereby satisfying the requirements for their application in packaging [34] .

The reduction in YM renders the films more pliable, as suggested by Sánchez-Soto et al. [36] . A significant difference (p < 0.05) was noted in the average flexibility between the films incorporating oregano and rosemary EOs. Increasing the concentration of EOs in the films, composed of collagen fiber and polyvinyl alcohol, enhanced their flexibility; notably, films with oregano EO were more flexible than those with rosemary EO.

The interplay between the type and concentration markedly affected the DPPH free radical scavenging activity (p < 0.05). Films containing oregano exhibited the highest antioxidant activity, reaching a peak at approximately 4.2% EO ( Figure 3 ). The addition of 2% - 6% oregano resulted in significantly greater (p < 0.05) antioxidant activity. Shen et al. [35] reported similar inhibition rates in films enhanced with oregano EO.

The multivariate analysis of the films ( Figure 4 ) revealed four distinct groups: I. control films without the inclusion of EOs and with 2% rosemary EO, characterized by the highest TS, YM, and the lowest antioxidant activity, EB, as well as practically no volatile organic compounds; II. films with 2%, 4%, and 6% oregano EO, characterized by the highest antioxidant activity (determined by DPPH), EB, and the lowest YM, as well as an association with the compounds carvacrol, thymol, thymol acetate, pseudolimonene, γ-terpinene, and ρ-cymene; III. film with 4% rosemary EO, characterized by an abundance of 1,8-cineole, camphor, and a scarcity of carvacrol and thymol, as well as intermediate antioxidant activity; IV. film with 6% rosemary EO, differing essentially from group III by the magnitude of the volatile organic compounds present.

The analysis of the film matrix is critical to ensure the incorporation of oils essentials, given that their production involves the use of temperatures above room temperature and that they are sensitive to heat, thus increasing their volatility with elevated temperatures. The composition of the films revealed that all compounds present in oregano EO were retained in their respective films to the extent that they formed a single, exclusive group ( Figure 4 ) due to their similar chemical composition and mechanical properties. However, the films containing rosemary EO were found to contain additional compounds, which may be attributed to the film matrix, processing temperatures, and the thermal stability of rosemary EO.

The type of EO (E) significantly affected (p < 0.05) the TBARS values of sliced mortadella. In contrast, the concentration of EO (L) significantly influenced (p < 0.05) the humidity, water activity (Aw), pH, and E. coli counts. Additionally, the storage duration (T) had a significant impact (p < 0.05) on the moisture content, Aw, pH, TBARS, and color attributes (L^*, a^*, b^*) of the sliced mortadella. Notably, there was a significant interaction (p < 0.05) between the type and concentration of (E × L) on TBARS values and between the concentration of and storage time (L × T) on Aw, TBARS, and E. coli counts.

The highest moisture content observed was 52.1%, corresponding to the addition of 3.1% EO, and the pH showed a significant decrease (p < 0.05) with an increase in the EO concentration ( Figure 5 ). Over the storage period, there was a significant reduction (p < 0.05) in humidity, pH, brightness (L^*), and paleness (b^*), along with an increase in redness (a^*) ( Figure 6 ).

The color of meat products is an important attribute and closely relates to the acceptability of the product [37] . According to Viuda-Martos et al. [38] , the L^* and a^* parameters are associated with the water content present in the meat product, its surface availability, and a tendency for humidity to decrease during storage, resulting in a decrease in L^* and an increase in a^*. The behavior of b^* may depend on pH, degree of oxidation, and water activity [38] . The principal components analysis revealed a correlation between humidity, water activity, L^*, and b^*.

Changes in the pH of meat products can occur due to the metabolic activity of microorganisms and chemical reactions, such as protein breakdown, which increases the pH [39] . Meat products packed in active packaging composed of collagen fiber and polyvinyl alcohol prevented the pH from rising, which is attributable to the biological activity of the EOs and inhibited bacterial growth and the progression of chemical reactions. Gedikoglu [40] and Quesada et al. [19] observed lower pH values in meat products with coatings and chitosan films infused with EOs and extracts. During storage, a decrease in pH was observed until approximately the sixth day, after which there was a stabilization followed by a gradual increase in pH around the eighth day. Qiu et al. [3] reported a pH increase in chicken burgers coated with gelatin-chitosan and infused with clove EO and rosemary extract from the fourth day of storage, with the increase becoming significant by the eighth day.

The decline in Aw at 9 days of storage was significantly lower (p < 0.05) for films with 2% - 4% levels of EO. Packaging protection is pivotal in mitigating the exchange between the food storage environment and relative humidity, which directly influences Aw. Environments with low relative humidity are known to reduce the Aw of food. The refrigerator’s internal environment is notably dry, leading to a tendency for refrigerated products to lose water to the environment [41] [42] . A significant effect (p < 0.05) was observed between the level of EO and time on the Aw of the product in the packaging. At 9 days of storage, the packaging without the addition of EO (control) and with 6% EO obtained a lower Aw value, likely due to the greater loss of water in these products ( Figure 7 ).

The response surface model for Aw as a function of the level and time variables is shown in Equation (3).

$\hat{y}=0.9615+0.0042L+0.0003T-0.0008L^2-0.0009T^2+0.0001LT\left(r^2=0.95\right)$(3)

The incorporation of EOs into the collagen fiber and polyvinyl alcohol matrix contributes to the reduction of packaging permeability through the presence of various highly hydrophobic compounds, which aid in decreasing water passage. Nonetheless, excessive incorporation of EOs into the matrix (6%) may adversely affect permeability, attributed to the OE’s agglomeration and heightened structure sensitivity [43] .

Lipid oxidation plays a crucial role in determining the quality of meat products, being responsible for the loss of nutritional quality. The secondary products formed during storage alter sensory characteristics [44] . In this study, the observed values were lower than those reported in the literature, which is known to impact sensory quality (2 - 2.5 MDA/kg) [45] . The inclusion of 6% rosemary EO in the films was associated with a greater (p < 0.05) susceptibility to lipid oxidation (TBARS) ( Figure 8 ). In contrast, the addition of 2% - 6% oregano EO was found to reduce (p < 0.05) susceptibility to lipid oxidation.

The results of this study, along with the in situ application, corroborate the findings regarding the inactivation of the DPPH radical in vitro. As anticipated, excluding EO from the films and prolonged storage time resulted in a higher susceptibility to lipid oxidation, as indicated by TBARS measurements. In 0 - 3 days, adding EO at concentrations above 2% acted as a pro-oxidant. However, its inclusion proved effective in delaying lipid oxidation from day 3 onward ( Figure 7 ). The response surface model for TBARS as a function of the level and storage time is presented in Equation (4).

$\hat{y}=0.2225-0.0009L+0.0048T+0.0013L^2+0.0007T^2-0.0016LT\left(r^2=0.32\right)$(4)

Similar outcomes were observed by Gedikoglu [40] , with the application of pectin coatings enriched with EOs and extracts from T. vulgaris and T. spicata on slices of mortadella (0.235 - 0.392 μmol MDA/g) over 21 days. Moreover, Zheng et al. [46] reported values of 0.37 - 0.26 mg MDA/kg during a 6-day storage period of chicken breasts in chitosan and oregano EO films.

Oregano EO is renowned in the literature for its antimicrobial and antioxidant activities, attributed to phenolic compounds such as carvacrol, γ-terpinene, and thymol [29] . In the PCA ( Figure 4 ), these compounds and ρ-cymene demonstrated a significant association with the DPPH free radical inactivation analysis.

The overall composition of the EO is crucial. As noted by Amorati et al. [47] , synergistic or antagonistic behaviors can be anticipated depending on the specific makeup of the EO and experimental conditions. Non-phenolic terpenoids, particularly unsaturated ones, may self-oxidize similarly to lipids, a notable example being $\alpha$-pinene. When these compounds are integrated with lipids from the food matrix, both are subject to auto-oxidation. There are exceptions, such as $\gamma$-terpinene, which can alter the propagation and termination phases of the chain reaction, thereby accelerating the termination process, shortening the chain, and reducing the overall oxidation rate [47] [48] . Research conducted by Guo et al. [48] has confirmed that in the presence of mono- or polyphenolic compounds, $\gamma$-terpinene exhibits a synergistic effect, thus enhancing antioxidant activity. In context with monophenolic compounds like carvacrol and thymol, the regeneration mechanism of the intermediate phenoxyl radical by γ-terpinene is suggested, which extends the inhibition period [48] .

The increased oxidation observed at the 6% rosemary EO concentration may be ascribed to certain compounds in the EO that possibly promote co-oxidation within the product.

Discrepancies exist in the literature concerning the antimicrobial activity in food matrices with high-fat content. Some studies suggest that such activity against spoilage and pathogenic bacteria is diminished in higher-fat environments [49] . However, this study found the antimicrobial effect against E. coli to remain unaffected by the fat content in meat products, which may contain up to 30% - 35% fat in the formulation [50] . Elevating the concentration of EOs in the films lowered the microbial contamination (E. coli) throughout the storage period ( Figure 7 ). With prolonged storage, the control packaging exhibited greater bacterial growth. The response surface model for the E. coli count as a function of EO concentration and time is performed using Equation (5).

$\hat{y}=6.1651-0.0856L-0.0366T+0.0089L^2+0.0104T^2-0.0207LT\left(r^2=0.83\right)$(5)

Ready-to-eat meat products, available in the refrigerated sections of supermarkets, can be sliced at the branding company or directly at bakeries or supermarkets before sale. Here, the cleanliness of the slicer, handler, and premises plays a critical role in preventing contamination [51] . Strains of E. coli O157:H7, known for their biofilm-forming capabilities, exhibit increased resistance to sanitizers when forming a biofilm network [52] . Sheen and Hwang [51] demonstrated that experimental contamination of slicing equipment blades with 8 log CFU of E. coli resulted in significant contamination transfers during ham slicing, ranging between 7 and 6 log CFU. Active packaging can serve as a protective barrier for these products by inhibiting pathogenic bacterial growth.

The PCA summarizes all analyses, accounting for 85.7% of the total variance, with 68.3% captured by the PC1 and 17.4% by PC2. Moisture content, Aw, L^*, b^*, and pH were positively correlated, with average EO content (2% - 4%) and shorter storage times. Conversely, TBARS, E. coli counts, and a^* were negatively correlated with each other but positively related to longer storage times (6 days).

A qualitative test of the films’ biodegradability was carried out, and images were taken in 0, 3, 6, and 9 days ( Figure 9 ).

Assessing the visual aspect, on the third day, there were already changes in the appearance of the films, which were more wrinkled and fragile due to swelling. Generally, the films containing oregano or rosemary EO were more sensitive to the biodegradation process, which was more pronounced with higher EOs. Similar results were found by Wang et al. [43] , starch films with higher additions of Zanthoxylum armatum EO showed a higher rate of biodegradation. The films with no or lower levels of EO presented a smoother surface, limiting water attack due to the increase in how it enters the polymer matrix. The greater incorporation of EOs leads to uneven dispersion due to weaker areas that promote water absorption, swelling, and microbial decomposition [43] .