3.1. Influence of Ultrasound Pre-Treatment on the Drying Kinetics of Chicken Breast Meat
One of the main purposes of meat pre-treatment using the US is the improvement in mass transfer during processing, which is expected to occur due to structural changes following the treatment. To evaluate the effect of the US on the hot air drying at 60 °C and freeze-drying at 40 °C processes, the kinetics of the moisture content (MR) were determined and presented in
Figure 1a,b, respectively. The green line marked on both graphs indicates the critical moisture content required for dried meat-based snacks, such as beef jerky, which was established at 0.2 [
52]. On the other hand, using pork and beef as a matrix [
53] suggested that the optimum MR for freeze-dried meat was 0.04. Based on that, drying times until the MR in chicken breast meat reached 0.2 and 0.04 were determined for both methods in
Table 3. Statistical analysis of the drying times showed that the US only slightly impacted the process, considering that the time required to reach MR equaled 0.2. The application of contact US treatment resulted in a prolongation of the drying time by 39% compared to the control sample (175 min). An increased frequency also caused the HA drying time to rise by 30%. However, it was still in the range similar to the material treated with 21 kHz and the control sample. Comparable dependencies occurred in the following stage of drying that was carried on until the MR was reduced to 0.04 (
Figure 1a,
Table 3). Compared with the literature regarding plant tissue, the results obtained in this research can be explained by the peculiarities of the material, which is muscle tissue. The fibrous structure of meat absorbs ultrasonic waves, which makes it difficult to exert an effect on the tissue. Moreover, meat lacks air-filled pores, as in plant tissues, resulting in less intense ultrasonic effects [
54]. The longer HA drying time might result from structural changes in fibrous structure (see Figure 4), particularly protein, caused by the US. This method of pre-treatment destroys the cellular structure of the muscle. However, it was also observed that the water-holding capacity of US-treated poultry meat increases due to myosin gelling induction [
37,
55]. Moreover, that phenomenon might have favored crust formation on the surface of the meat, which trapped water inside the material and worsened mass transfer during processing. Additionally, [
56] found that the temperature of 70 °C gives better drying rates and quality of the products than 60 or 80 °C. Hence, for further improvement in the HA drying of chicken breast meat, the US pre-treatment may be tested in combination with higher drying temperatures.
In the case of the freeze-drying process, the drying time varied from 181.3 to 267.5 min. The US pre-treatment did not affect the processing time significantly (
p > 0.05), but the treated material tended to attain the desired moisture content (0.04) faster, especially after subjection to contact treatment. The reduction in the drying time after pre-treatment was 5–32% (
Figure 1b,
Table 3).
Faster water removal after the US application was related to the damage made by the electromagnetic waves in the material’s cellular structure. Breaking the internal barriers naturally existing in the tissue made water removal easier. Thus, the drying process was shorter [
57]. According to established drying times, freeze-drying provided more effective and faster water removal in the examined material than HA drying. Similar results were obtained before for turkey breast meat [
58]. The US treatment was recognized as working appositively in both dehydration methods tested in this study. Freeze-drying combined with the unconventional US pre-treatment gave beneficial and promising results regarding drying kinetics and time, contrary to HA-drying. The most possible explanation is the difference between hot-air and freeze-drying mechanisms. Mass transfer in HA drying is based on the moisture content difference within the sample volume. Water contained in the material migrates to its surface, from which it is removed by the drying agent (hot air). However, exposure to the hot drying agent in HA is associated with structure collapsing and crust formation on the surface of the materials, especially those built out and prone to thermal degradation compounds such as protein. In freeze-drying, moisture removal is driven by a pressure difference between the sample and its surroundings. Moreover, during dehydration, the material is preserved in a frozen state, reducing the risk of unfavorable changes exacerbating the process [
57,
59].
3.2. Influence of Ultrasound Pre-Treatment on the Water Content and Water Activity of Dried Chicken Breast Meat
Water content and water activity are jointly reliable indicators of the effectiveness of food processing and preservation, as well as its microbiological safety. Raw meat consists of around 75% water. Therefore, the degree and rate of its removal during drying can be criteria for the shelf life of dried meat products [
60,
61].
The water content and water activity (
aw) in chicken breast samples ranged from 2.81 to 8.20% and 0.099 to 0.436, respectively (
Table 4). This is consistent with the requirements for low-moisture foods, which generally contain no more than 25% water [
15]. In turn, the water activity for low-moisture food products with extended shelf-life (even without refrigeration) should not exceed 0.6 [
62]. Moreover, as seen in
Table 4, both the water content and water activity in the FD meat were significantly lower than in the HA-dried samples. During HA drying, meat dehydration is constrained over time by the shrinkage of the muscle myofibril network and connective tissue and, thus, by surface hardening [
63,
64]. However, during the freeze-drying process, water is gently and gradually removed through micro and macro capillaries of the tissue. The integrity of the muscle fibers is relatively retained, although they become denser and shorter. Importantly, maintaining the porosity of dehydrated muscle tissue increases the effectiveness of its rehydration, which is a relevant quality trait of dried food [
23].
The HA-dried meat pre-treated with 300 W ultrasonic power showed about 20% lower water content than the untreated HA-dried sample. Applied ultrasounds could disrupt the cell walls and improve the transfer of water from the intercellular to the extracellular space of the tissue, hence facilitating the further drying process, even with less water activity [
37]. In addition, a significant positive linear relationship was found between the water content and water activity of dried chicken meat, which is described by the equation
Wc% = 11.140 ∙
aw + 2.5725 and confirmed by the Pearson correlation coefficient
r = 0.93,
p = 0001 (
Figure S1). However, this relationship is not directly proportional and should be considered individually for each product type [
20,
22,
63]. For example, in this study, HA-dried and FD chicken breast meat pre-treated with 300 W US were statistically similar in water content but significantly different in water activity (
Table 4). Importantly, these effects arise from two main factors: the method and parameters of ultrasonic processing, as well as the drying technique. During freeze-drying and hot-air drying, different mechanisms occur and lead to greater or lesser changes in the structure and physical properties of the dried product [
34,
41,
58]. Ultrasound significantly influenced the mass changes in raw chicken breasts even before the drying process. As shown in
Table 1, the US contact method resulted in a mass loss, while the US immersion method (regardless of frequency and power) resulted in a mass gain compared to the untreated samples. According to Huang et al. [
16] and Ricce et al. [
21], the water absorption by tissue during ultrasonic treatment may even hinder its subsequent drying, especially at relatively low temperatures, e.g., ≤ 40 °C. In the present study, this can be observed, for example, in freeze-dried meat pre-treated with ultrasound at a frequency of 21 kHz using the immersion method (
Table 1,
Table 4). However, it can be stated that the drying method predominantly influenced the water content and water activity of the dried chicken breast meat.
3.3. Influence of Ultrasound Pre-Treatment on the Rehydration Ratio and Hygroscopic Properties of Dried Chicken Breast Meat
The process of evaporation of water from the material, commonly known as drying, is associated with the simultaneous occurrence of various biochemical effects, different types of chemical reactions, and modification of the physical properties of the dried material. Physical modifications of the material after drying, e.g., the occurrence and scale of drying shrinkage, decreased or increased porosity, range of water absorption and adsorption capacities (rehydration and hygroscopicity, respectively), and the amount of damage at the microstructural level depend on the characteristics of the matrix, the method of treating the material before drying, and the selected drying method as well as process conditions [
65,
66,
67,
68].
Figure 2a shows how the values of the RR parameter increased depending on the time of immersing dried chicken breast meat in distilled water. As can be seen, at each of the analyzed times, all dried HA samples exhibited a similar ability to absorb water. It increased over time until reaching its maximum after approximately 90 min. Higher variation in RR values occurred in FD samples (
Figure 2a). Additionally, regardless of time, soluble solid loss (SSL) in all analyzed dried HA samples remained similar (0.009 ± 0.001—
Figure 2b). After 5 and 15 min of analysis, the FD_cUS_25_250 sample showed significantly lower RR and SSL than the other samples, which, as mentioned above, could be due to the more damaged structure of this sample caused by its direct contact with the sonotrode [
69,
70]. After 30 min of analysis, the untreated FD sample was characterized by a slightly lower RR and SSL than the US-pretreated samples. Taking into account only the analysis time of 30 min, it can be seen that the FD samples had more than twice the values of the RR parameter than the HA samples (
Figure 2a,b). As explained above, the reduction in the water absorption capacity of the HA samples could result from the occurrence of shrinkage, and in turn, the significant porosity of the FD samples intensified the water absorption [
71,
72].
Table 5 contains the rehydration ratio (RR) values obtained by untreated and US-pretreated HA-dried and FD chicken breast meat after 30 min of immersing the materials in distilled water. As can be seen, all samples dried with hot air showed similar rehydration properties, as evidenced by the fact that these samples belong to one homogeneous group. The application of preliminary ultrasonic treatment, regardless of the parameters used or the method of supplying ultrasonic waves to the treated material, did not significantly change the RR of dried HA samples (
p > 0.05). In the case of FD, all US-pretreated samples did not differ significantly in RR from samples untreated and dried in the same way. However, it was observed that the sample to which ultrasound was delivered in a contact manner (FD_cUS_25_250) exhibited RR lower by 23.8–28.6% than samples treated with the US using water (
Table 5). Direct exposure of the material to the sonotrode (contact method) could have led to greater damage to its surface (e.g., to contraction of muscle fibers), which could have contributed to the reduced ability of this sample to reabsorb water [
69,
70]. Moreover, FD samples had more than twice as high values of the RR parameter as HA samples. This result is the effect of probable differences in the physical properties of the analyzed materials—higher shrinkage of the HA samples, which prevented water absorption, and higher porosity of the FD samples, which enhanced this process [
71,
72].
The second important parameter determining the quality of the dried material is hygroscopicity (H).
Figure 3 shows the increase in the value of the H parameter of all dried chicken breast meat obtained depending on the time these samples remain above distilled water. At each of the analyzed times, e.g., 0, 1, 3, 6, 9, and 24 h, the HA samples exhibited similar hygroscopicity. Nevertheless, slightly higher H for untreated samples (HA) than US-pretreated samples can be observed. More variation in results occurs in the case of FD drying, especially after 6 h or more of analysis. Samples FD and FD_cUS_25_250, e.g., untreated and treated with contact-supplied US, respectively, achieved noticeably higher values of the H parameter than the other samples. The results may indicate that the disruption of the tissue structure by ultrasound and its transmission through water reduced the treated samples’ ability to adsorb water vapor [
73].
As shown in
Table 5, a one-way analysis of variance showed that neither the application of ultrasound as a preliminary treatment before drying HA and FD (interpreted separately), the modification of the parameters of this process, nor even the change in the method of supplying ultrasonic waves to the treated chicken breast meat, caused statistically significant differences in hygroscopic properties determined after 1 h of testing of the obtained dried materials (
p > 0.05). In the case of HA drying, this trend also continued after 24 h of analysis. Nevertheless, when analyzing FD drying, a significant difference can be observed in the values of the H parameter (after 24 h of analysis) between the US-treated samples (immersion method) and the US-treated sample (contact method), whose H (1.253 ± 0.011) was comparable to the H achieved by the untreated sample (1.241 ± 0.010). The ability of a given material to adsorb water depends not only on its surface properties but also on its chemical composition. Delivering ultrasonic waves to the material using water could cause significant changes in its chemical characteristics and thus modify its sorption behavior [
47]. Similar to the rehydration properties, the FD samples showed higher hygroscopicity than the HA samples, which can also be observed in
Figure 3. The obtained results can be explained by the probable differences in the scale of drying shrinkage and porosity of materials dried by both methods [
71,
72].
3.4. Influence of Ultrasound Pre-Treatment on the Structure of Chicken Breast Meat
Drying is a complex process in which the material undergoes chemical and physical modifications. Changes in physical characteristics include, among others, shrinkage and porosity [
65].
Figure 4 shows images of obtained untreated and US-pretreated HA-dried and FD chicken breast meat taken using the scanning electron microscope. In general, the structure of the samples dried by hot air (regardless of the use of ultrasonic pre-treatment or the differentiation of this process parameters) was relatively dense and compact, with visible shrinkage, which can also be observed on the macroscopic photographs (
Figure 5). However, the sample structure to which ultrasound was supplied in a contact manner (HA_cUS_25_250) was slightly more porous than others, characterized by samples that dried similarly. In the case of FD, the effect of ultrasound on chicken breast meat was more visible. One of the effects of ultrasound is the creation of microchannels in the treated tissue [
74]. As can be seen in
Figure 4, all FD samples with US application before drying had free spaces in their structure. The absence of these channels in the untreated sample (FD) structure was also noticeable. The phenomena that accompany the action of ultrasound, e.g., sponge effect, cavitation, and other accompanying effects, are responsible for the modifications of the treated material. However, their scale depends on the method of supplying ultrasonic waves, selected ultrasonic processing parameters, and the characteristics of the matrix itself [
16]. By comparing the images obtained for HA-dried and FD samples, an apparent shrinkage of the former and a much more porous structure can be observed for the latter. This is due to different drying mechanisms. FD involves removing water from the material through the process of ice sublimation. The ice crystals formed in this process compress the material, causing the formation of a porous and spongy structure, a process which, in turn, mitigates the occurrence of drying shrinkage. In turn, during HA drying, water is evaporated from the surface of the dried material, which leads to a pressure difference, and, as a result, the structure collapses [
75].
3.5. Influence of Ultrasound Pre-Treatment on the Color Parameter of Chicken Breast Meat
The quality and shelf life of meat is determined by many factors, including animal-specific (e.g., breed, genetics, age, feeding, and pre-slaughter handling), product-specific (e.g., acidity, moisture, texture), process-specific (e.g., ripening, process technology, and heating techniques), and environmental (e.g., temperature, time, packaging, and storage) factors [
76]. For consumers, color is still the main indicator of the quality and freshness of meat and the final purchasing decision. Therefore, meat science research frequently examines this quality parameter using objective instrumental analysis methods [
77].
Figure 5 shows the visual differences in the color of the HA-dried and FD chicken breast meat samples. Assessing only the appearance of the samples, there was no meaningful effect of US pre-treatment on the color of dried meat, while the drying method caused significant changes. However, the objective evaluation was based on the results of instrumental analysis, expressed in the CIE L*a*b* system, presented in
Table 6.
There were no significant differences (
p > 0.05) in the values of the
L* (lightness),
a* (redness), and
b* (yellowness) color parameters alone between all undried meat samples. The lightness of all the samples, in a range of 53.1–59.1 (
Table 6), was typical for fresh chicken breast muscles [
7,
9]. In turn, the total color difference (Δ
ERAW), ranging from 3.1 to 7.2, showed a quite significant color change in the US pre-treated chicken breasts compared to the untreated one. When ∆
E exceeds the level of 5, the observer can have an impression of two different colors [
78,
79]. For example, [
36] categorized the total color change in US-treated raw chicken meat as detectable by the human eye at the level of Δ
E > 5.5. Intriguingly, the slightest color change (Δ
ERAW~3.0) was caused by immersive US treatment at a power of 300 W (
Table 6). This is consistent with the findings of [
80], who studied the color changes in carp muscles during immersive US-assisted thawing.
As shown in
Table 5, the dried samples exhibited an increase in both
a* and
b* color coordinates, compared to raw meat, irrespective of US treatment and drying method. In turn, the lightness (
L*) of the meat samples notably increased during FD and slightly decreased as a result of HA drying (
p ≤ 0.05). The variation in color coordinates values has been reflected in the total color difference Δ
ERAW, which varied from 17.0 to 35.4, indicating the perception of two different colors of chicken breast meat. It was concluded that these color changes mainly depended on the drying process and only slightly on the US pre-treatment. It was suggested that the heat generated during ultrasonic processing is insufficient to denature proteins and pigments in meat and, consequently, does not affect its color [
81,
82]. However, attention was also drawn to the fact that despite a specific total color change in the raw meat caused by ultrasounds, these differences are unnoticeable after thermal treatment [
83,
84].
When comparing the dried meat, FD samples differed significantly from HA-dried samples in terms of lightness (
p ≤ 0.05), showing an approximately 2-fold higher value of the
L* color parameter. In turn, the HA-dried meat was characterized by a more significant redness (
a*) than FD meat (
Table 5). As reported by [
22], these are the characteristic features differentiating the color of air-dried and freeze-dried meat and meat products. The lighter surface of FD chicken breast meat can be explained by structural changes in myofibrillar and connective tissue proteins caused by their denaturation, leading to optical masking of heme proteins and higher light scattering intensity of the surface. Due to low drying temperature, this phenomenon cannot be explained by the denaturation of heme proteins [
5,
85] but can be connected with transformations of myoglobin [
86]. On the other hand, the increase in redness and darkening of chicken breasts during hot-air drying (
Table 5) indicated browning reactions [
87] due to high drying temperatures and contact with oxygen-rich air [
8], accelerating oxidative reactions, metmyoglobin formation [
88], and Maillard reactions [
89]. These changes have been visualized numerically using the Browning Index (
BI), calculated from the
L*,
a*, and
b* color coordinates. As a result of the statistical analysis, HA-dried chicken breasts showed approximately twice the BI value than FD samples. At the same time, ultrasonic pre-treatment did not have any significant effect on the degree of browning of chicken breast samples during drying (
p > 0.05) (
Table 6). However, it was noticed that FD meat pre-treated with contact US treatment at 25 kHz and 250 W was similar to HA-dried meat in
BI values. In addition, using this method caused the greatest total change in the color of FD meat (Δ
EFD). Presumably, it could be caused by the method of applying ultrasounds that triggered an increase in the sample temperature, which resulted in the adhesion of the meat to the sonotrode sieve. Similar observations were made by [
70] when examining the impact of contact ultrasound treatment on apple tissue.
3.6. Multivariate Statistical Analysis
Principal Component Analysis led to the extraction of the two first principal components (PCs), factors 1 and 2, which had eigenvalues higher than 1.0 and accounted for 78.85% and 9.48% of the variance, respectively. This means that 88.33% of the total variance for the quality of dried meat in the 13 variables considered can be condensed into two new variables, e.g., PCs.
Figure 6a shows that all of the variables had similar proportions in PC1 except for the b* color parameter. HA-dried and FD samples were localized on the positive and negative sides of the PC1 axis, respectively. HA-dried meat pre-treated with contact US contributed most positively along PC1, whereas FD meat pre-treated with immersive US contributed most negatively. The PCA loading plot also suggested the parameters that most contributed to the quality of particular dried chicken breast meat samples. HA-dried meat was positively influenced by drying time, color parameters a* and BI, water activity, and water content. In turn, FD samples were negatively influenced by hygroscopic (
H1,
H24) and rehydration properties (RR, SSL), lightness (
L*), and total color difference (Δ
ERAW). The position and angles of the quality parameters (variables) vectors, projected on the plane of the two first principal components, indicate the direction and strengths of the relationships between the variables (
Figure 6a). However, the matrix of correlation coefficients and their significance level were examined to illustrate the correlations between variables better, as presented in
Table S1. All statistically significant correlations showed at least moderate correlation strength (
r ≥ 0.4). At the same time, no significant correlation of the
b* color coordinate (yellowness) with other quality parameters of dried meat was observed (
p > 0.05). In turn, the longer the meat is drying, the higher the final water content and water activity of the dried meat (positive correlation), the lower the rehydration capacity, and the worse the hygroscopicity (negative correlation). As previously mentioned, these are characteristics typical of HA-dried meat—
Figure 6a. Moreover, the browning index was strongly negatively correlated with the total color difference compared to raw meat (Δ
ERAW), which proves the dominant influence of surface browning on the total change in meat color.
The results obtained in PCA were reflected in the results of HCA, in which the dried meat variants were divided into six clusters (binding distance: y = 2.45), as shown in the tree diagram in
Figure 6b. The clusters were characterized in terms of mean values of the analyzed qualitative parameters (13 variables). HA_US_21_180 (as cluster 1) exhibited the lowest
L* and
b* color coordinates as well as total color difference compared to raw meat (Δ
ERAW). FD (as cluster 2) showed the lowest browning index, while FD_cUS_25_250 (as cluster 3) had the shortest drying time, the lowest final water content, and the greatest hygroscopicity within 1 h. However, HA and HA_US_21_300 samples (as cluster 4) exhibited the longest drying time until MR = 0.04 and the highest browning index. The other FD meat (FD_US_21_300, FD_US_21_180, and FD_US_40_180) as a cluster 5 were the lightest (
L*) and the most different from raw meat in terms of total color change and simultaneously had the highest rehydration rate and soluble solids loss. Cluster 6 involved HA_US_40_180 and HA_cUS_25_250 samples with the longest drying time until MR = 0.2, the highest final water content, and the lowest rehydration rates and hygroscopicity within 1 h. Based on the coefficients of variation (
Vc), the degree of differentiation of dried meat samples in the cluster concerning a given quality parameter was determined (
Table S2). For example, cluster 5 (FD meat pre-treated with immersive US method) and cluster 6 (HA-dried meat pre-treated by US contact method at 25 kHz or US immersion at 40 kHz) were heterogeneous in terms of water content (
Vc > 20%) and a* color parameter (
Vc > 30%). Nevertheless, in most cases,
Vc did not exceed 20%, which indicates a high degree of homogeneity of the clusters identified in this multivariate statistical analysis.