4.1. Stress Sweep (Amplitude Sweep)
As shown in
Figure 1, the formulations exhibited a crossover of the storage (G′) and loss (G″) moduli around 0.5 Pa, indicating predominantly elastic behavior following the aging process. This response was more pronounced in the samples formulated with vegetable fat (T1) and UHT cream (T3), suggesting a more structured viscoelastic network capable of sustaining deformation under stress. An exception was observed for formulation T2 (butter), which showed a higher crossover point near 1.0 Pa, indicating reduced elastic dominance at lower stress levels and potential structural fragility. These findings are consistent with Brazmi et al. [
16], who demonstrated that variations in lipid composition can significantly affect the rheological profile of frozen systems, with direct implications for stability and textural performance.
Formulations T3 (UHT cream) and T4 (fresh cream) displayed lower values for both G′ and G″, which may be attributed to their compositional characteristics, particularly the higher proportion of unsaturated lipids and reduced solid fat content (SFC). These factors contribute to less organized fat crystal networks, resulting in decreased resistance to deformation and reduced mechanical integrity. In the case of UHT cream, thermal processing may promote partial denaturation of whey proteins and alter fat–protein interfacial interactions, potentially impairing the development of a cohesive microstructure. The greater fluidity of these formulations at near-zero temperatures likely undermines emulsion stability upon partial melting, contributing to more flow-prone behavior under stress conditions.
In the case of UHT cream, the ultra-high-temperature sterilization process can cause changes in protein structures and fat globules, leading to a loss of stability at higher temperatures. This phenomenon, when observed in the rheological properties of ice cream, can result in a more fluid product that does not resist mechanical stress, such as during processing and freezing [
1]. This behavior was discussed by [
17], who pointed out that stabilizers present in ice cream emulsions play a crucial role in modulating rheological characteristics, controlling the viscosity and texture of the final product.
Ice water plays a fundamental role in the rheological behavior and microstructure of ice creams. The presence of solid water, i.e., ice, dictates the rigidity of the sample, which is expressed in the elastic modulus (G′). Ice creams with higher amounts of ice water tend to exhibit greater hardness and rigidity, which are crucial for ensuring proper handling during consumption and for maintaining stability over time. The formation of ice during freezing and its interaction with the lipid and protein phases of the ice cream matrix also affect the product’s microstructure, as noted by [
18], who discussed the importance of this solid fraction in the product’s behavior.
The effect of ice water can also be linked to the rheological behavior of ice cream emulsions. The formation of ice and its interaction with lipid and protein phases influences not only viscosity but also texture properties. The hardness of ice cream is a critical parameter, particularly in the context of sales and consumption, where texture can be a key determinant of product acceptance. Studies by [
19] suggest that modulating the amount of ice water in the system is crucial for controlling the sensory characteristics of ice cream, such as creaminess and optimal melt performance.
The crystallization of triacylglycerols (TAGs), the primary components of fats in ice cream, is a crucial phenomenon for determining the product’s structural and textural properties. The crystallization behavior of TAGs is greatly influenced by factors such as fat composition (saturated vs. unsaturated) and melting temperature. As noted by [
20], trans fatty acids crystallize faster than their cis isomer counterparts, which can lead to a better structuring and formation of a more stable lipid network in products like ice cream.
The presence of unsaturated fats can modify the crystallization rate of TAGs, as discussed by [
21]. The formation of smaller fat globules, which occurs with a higher content of unsaturated fatty acids, tends to result in more stable and smoother emulsions, which are desirable for ice cream products. However, controlling the crystallization rate is essential, as too-rapid crystallization can lead to the formation of large ice crystals, resulting in a grainy texture.
After the maturation of ice cream emulsions at low temperatures, the physical and rheological properties change due to the hydration of stabilizers. As discussed by [
22], the hydration of stabilizers like guar gum, xanthan gum, or gelatin increases the viscosity of emulsions, promoting improved stability and texture during storage. The presence of emulsifiers and stabilizers, such as glycerol esters and milk proteins, can significantly influence flow and resistance properties, leading to a dynamic balance between the elastic and viscous moduli [
18,
22].
The different rheological behaviors observed between formulations with butter, fresh cream, and UHT cream reflect not only differences in physical properties of these ingredients but also in crystalline structures and melting/solidification processes [
18]. Saturated fats, like butter, offer greater control over texture and rheological stability, while unsaturated and processed fats, such as those in UHT cream, have a more fluid behavior, with less stability under certain conditions. Differences in crystallization properties and the stability of emulsifiers result in significant variations in final texture and behavior during the handling and storage of ice cream [
17,
22].
4.2. Frequency Sweep
The variation in material properties with respect to strain and stress rates was investigated using an oscillatory frequency sweep test. This test demonstrates how the elastic and viscous behavior of a material oscillates, and it is possible to analyze the complex viscosity as a function of frequency, describing the total resistance to dynamic shear. In this test, the complex viscosity either increases or decreases while maintaining constant tension [
23]. Based on the parameters tested, all formulations indicated that frequency could be proportionally increased with the increase in the elastic and viscous moduli, as well as the decrease in the complex viscosity curve. This suggests the subordination of the moduli to frequency, thus characterizing the viscoelastic fluid behavior, where the molecules tend to orient themselves in the direction of the applied force. Under shear stress, the material typically behaves as a pseudoplastic fluid, i.e., a non-Newtonian fluid where the viscosity decreases with increasing shear rate [
23]. This behavior was observed in most samples, as reported by [
20,
22,
24] for ice cream systems.
Among the formulations tested (
Figure 2), those produced with fresh cream and UHT cream presented the lowest frequency resistance, being below 1 Hz. Additionally, these formulations showed the lowest loss modulus (G″) and storage modulus (G′), indicating a less rigid structure compared to other formulations. This suggests that the fat composition and the manufacturing process (such as the UHT treatment of UHT cream) significantly influence the rheological properties of the system. The lower G′ and G″ values of these formulations indicate a weaker gel structure, with less resistance to deformation under applied stress.
These findings are consistent with the observations made by [
16], who emphasized the structural sensitivity of ice cream mixtures based on dairy fat during the maturation phase at 4 °C. At this stage, emulsions undergo structural changes due to the reorganization of fat globules and the hydration of stabilizers, which affects the elastic and viscous properties of the ice cream base. The UHT cream, for instance, can undergo significant changes in its fat globule membrane during processing, leading to a destabilized fat phase, which exhibits more fluid-like behavior under oscillatory shear tests.
It is important to note that the maturation process plays a crucial role in altering the rheological properties of ice cream formulations, as pointed out by [
19,
22]. During maturation, emulsifiers and stabilizers hydrate, which increases viscosity and alters the gel network. As a result, formulations with low-trans-fat content exhibit higher G′ values and a more elastic behavior, while formulations with fresh cream and UHT cream, due to their weaker lipid structures and lower G′ values, may not exhibit the same level of rigidity after maturation, resulting in higher tan δ values and a behavior more dominated by viscosity. This behavior is consistent with [
16], who noted that emulsions made with dairy fat exhibit a prevalence of the elastic modulus after maturation at low temperatures.
The shear-thinning behavior observed in the formulations, especially those containing fresh cream and UHT cream, suggests that these systems are well suited for processing under high shear conditions, which are typical of ice cream production. The results from the oscillatory shear test reflect the ability of these emulsions to flow more easily under high shear stress, contributing to a smoother, creamier texture when whipped. However, the lower G′ values observed in these formulations might indicate a softer texture in the final product, which could affect the mouthfeel and structural integrity of the ice cream.
In contrast, samples with higher fat content or with more stable fat systems demonstrated elevated G′, G″, and complex viscosity values. This behavior indicates enhanced fat crystal packing and greater contribution of partial coalescence to the viscoelastic network, conferring superior firmness and stability. The balance between elasticity (G′) and viscosity (G″) is critical for maintaining the microstructural integrity of ice cream throughout processing, storage, and consumption, ultimately ensuring consistent texture and consumer acceptability.
There was a significant reduction in complex viscosity at high frequencies, which caused the disaggregation of fat particles in the system. It was clear that the viscosity of the mixtures depends on the type of fat and hydrocolloids added to the formula and the size of these particles in the matrix. It was possible to verify that the ice creams produced with low-
trans fat (T1) and butter (T2) were the most rigid products and had the most ice crystals among the formulations tested. Therefore, we highlight the thesis that low-
trans vegetable fat suffers less membrane destabilization because it has a large amount of palmitic acid (16:0), ranging from 35 to 47%. This high level of saturated fatty acids is disadvantageous for the destabilization of fat globules, as this destabilization will be greater the more unsaturated and longer the chains of vegetable fat used in the formulation. Furthermore, mono- and diacylglycerol emulsifiers can interact better through the hydrophobic region of the molecules [
24].
However, formulations T2 and T3, produced with milk fat, demonstrated similar and lower values (<1) of tan δ, thus indicating the existence of a remaining microstructure. This can be attributed in part to the presence of proteins in association with stabilizers. Furthermore, the fact that these formulations achieved similar G′ values above 0.5 Hz frequency suggested that interactions between milk proteins and fat occurred to ensure a structured product. The tan δ values were below 1 at high frequencies, G′ was greater than G″, proposing that the molecules have less mobility, and the product has solid and not liquid properties. The ice cream prepared with milk fat (T2, T3, and T4) presented a lower slope of the curves and differentiated profiles, indicating greater stability.
4.3. Overrun Determination
The calculated overrun values for the evaluated ice cream formulations are depicted in
Figure 3. Overrun, defined as the volume increase due to air incorporation during freezing and whipping, is a paramount quality parameter influencing the texture, mouthfeel, and overall consumer acceptance of frozen desserts [
25,
26]. The extent of air incorporation is inherently dependent on the interplay between the fat globule network, milk proteins, emulsifiers, stabilizers, and ice crystal morphology, all of which modulate the viscoelastic properties and stability of the continuous phase [
5,
27].
Among the tested formulations, T3 (UHT cream) and T1 (low-trans vegetable fat) exhibited the highest overrun values, indicating superior air entrapment and retention capacities. The elevated overrun in T3 is attributable not only to the relatively stable crystalline structure of the UHT cream fat fraction but also to the intrinsic presence of hydrocolloidal stabilizers incorporated during industrial processing [
28,
29]. These stabilizers synergistically interact with emulsifiers and anticrystallization agents added in the ice cream mix, enhancing the interfacial stability of fat globules and improving the viscosity of the unfrozen serum phase, which collectively promote the formation and maintenance of a cohesive partial coalescence fat network [
18,
19].
While total fat content remains a significant factor in determining overrun, controlled destabilization of the fat membrane during mechanical agitation is critical for partial coalescence, which stabilizes air cells within the matrix and confers elasticity to the continuous phase [
20,
30]. In contrast, formulation T2 (butter) presented the lowest overrun, a phenomenon consistent with its comparatively weaker fat network, as corroborated by lower values of the storage (G′) and loss moduli (G″) in rheological frequency sweep tests (
Figure 2). These rheological findings indicate diminished structural strength and elastic behavior, which compromise the product’s ability to trap and stabilize air bubbles, thus limiting overrun [
12,
21].
Moreover, the rheological profiles reveal a positive correlation between higher elastic modulus and increased overrun, emphasizing the role of fat–protein interactions and stabilizer systems in dictating air incorporation efficiency [
1,
22]. Stabilizers present in both the UHT cream and the ice cream base reduce excessive ice crystal growth by inhibiting water mobility and recrystallization, leading to a smoother texture and further aiding the stabilization of the aerated structure [
14,
23].
Consequently, the superior overrun observed in T3 reflects the technological advantages conferred by the integration of lipid stabilizers and complex emulsifier systems, which collectively optimize the viscoelastic properties of the ice cream matrix. This formulation strategy enhances the retention of incorporated air, improves creaminess and sensory perception, and contributes to extended shelf-life through increased storage stability [
17,
18]. Thus, the use of UHT cream enriched with stabilizers, in combination with a tailored rheological formulation, represents an effective approach for the production of premium ice cream products with enhanced structural and sensory qualities.
4.4. Melting Behavior
The melting behavior of ice cream constitutes a pivotal quality attribute, intimately linked to both its sensory perception and physicochemical stability, as it directly reflects the microstructural integrity of the frozen matrix and the interactions among ice crystals, fat networks, and serum phases [
16,
18].
Figure 4 presents the cumulative melting profiles of the four formulations over a 45-min interval under rigorously controlled ambient conditions, with the initial sample masses standardized to 100 g to enable valid comparative analyses of absolute melt loss and temporal melting kinetics [
1].
Formulation T1, incorporating low-trans vegetable fat, exhibited the highest melting rate, losing approximately 63 g of mass (63%) over the test duration. This pronounced melt susceptibility can be ascribed to the intrinsic physicochemical characteristics of industrially modified vegetable fats, which generally display a fractionated triglyceride composition and possess less stable polymorphic crystalline structures compared to dairy fats [
31]. Consequently, these fats exhibit diminished capacity for partial coalescence and the formation of an interconnected fat globule network during the freezing process, thereby compromising the mechanical rigidity and melt resistance of the ice cream matrix [
18,
20]. The limited formation of interglobular bridges results in a weaker three-dimensional fat network that inadequately immobilizes unfrozen water, accelerating melting kinetics [
31,
32].
Conversely, formulation T3, based on UHT cream, demonstrated the most robust melting resistance, with only about 13 g (13%) of mass loss, indicating a highly cohesive and stable frozen structure. This enhanced performance is likely mediated by the denaturation of whey proteins induced during the ultra-high-temperature treatment, which facilitates the formation of protein–fat aggregates [
15,
31]. These aggregates contribute to the reinforcement of the fat–protein matrix, effectively entrapping free water within the microstructure and mitigating phase separation during melting [
14,
17,
31]. Additionally, the UHT process may alter emulsification dynamics, enhancing fat globule interface stability and favoring a more resilient network formation upon freezing [
19,
25].
The intermediate melting profiles observed for T2 (butter; ~50% melt) and T4 (fresh cream; ~46% melt) reflect their compositional and processing attributes. Butter, rich in short- and medium-chain saturated triglycerides, can form rigid crystalline networks; however, its structural contribution is highly dependent on emulsification efficiency and crystallization kinetics, which are factors that govern the extent of fat partial coalescence and network continuity [
18]. Fresh cream, while retaining native emulsifying milk proteins such as caseins, undergoes minimal thermal treatment, which may limit protein denaturation and subsequent aggregate formation necessary for robust fat–protein network stabilization [
20,
22]. Moreover, variations in total solids and serum viscosity between fresh and UHT cream further influence matrix rheology and melting resistance [
23,
33].
Another relevant factor contributing to the superior melt resistance of T3 is its lower overrun, resulting in a denser microstructure with reduced air cell volume. The diminished air incorporation increases the thermal mass and reduces the specific surface area exposed to ambient temperature, thereby retarding heat transfer and melting rates [
14,
22]. This attribute is especially desirable in premium or regionally targeted ice cream products intended for warm climates, where prolonged shape retention and slow melting are critical for consumer satisfaction and product differentiation [
27,
34].
In summary, the melting profiles elucidate the fundamental role of fat source composition, thermal processing, and microstructural organization in determining the melt stability of ice cream. The findings affirm that the structural functionality of protein–fat networks, modulated by ingredient characteristics and processing parameters, is paramount for controlling melting behavior, thereby influencing sensory attributes and shelf-life performance [
14,
35].
4.5. Color Determination
Color is a critical sensory attribute in the acceptance of ice cream, as it directly shapes consumer expectations regarding flavor, freshness, and overall product quality [
14,
36]. Objective measurements were obtained through reflectance colorimetry using the
CIE Lab* color space, which quantifies the three-dimensional coordinates
L* (lightness),
a* (red–green axis), and
b* (yellow–blue axis). This approach provides reproducible and comparable parameters for visual characterization and is widely employed in dairy-based colloidal systems due to its ability to capture subtle compositional and processing-induced variations [
5,
37,
38,
39].
The
L* values were consistently high across all formulations, reflecting the typical optical behavior of dairy colloids in which light scattering is mainly governed by fat globule dispersion and casein micelle distribution within the aqueous matrix [
5,
37]. However, treatment T3 (UHT cream) exhibited a significantly reduced
L* value (77.51) compared with T1 (81.01), T2 (80.25), and T4 (79.36). This darker appearance may be associated with the formation of early Maillard reaction products and non-enzymatic browning intermediates generated during severe heat treatment [
2,
27]. Compounds such as hydroxymethylfurfural (HMF) and furosine are known to absorb visible light, thereby lowering diffuse reflectance [
40,
41,
42]. Since no such markers were quantified in this study, the explanation remains hypothetical and requires confirmation by targeted chemical analyses.
The
a* coordinate was slightly negative in all samples, indicating a faint greenish hue typical of dairy products and which is often attributed to trace chlorophyll derivatives or subtle scattering effects [
39,
43]. T3 showed the most negative
a* value (−2.09), which could be linked to microstructural modifications from UHT processing that alter light absorption and scattering [
41]. Nonetheless, the small absolute magnitude of these values suggests that such differences are unlikely to be perceptible without complementary visual or spectrophotometric evaluation.
All samples exhibited positive
b* values consistent with the characteristic yellowish coloration of fat-rich dairy matrices [
41,
43]. The butter-based formulation (T2) showed the highest
b* (28.88), attributable to the presence of β-carotene and other lipophilic pigments concentrated in milk fat [
41]. The greater yellowness in T2 may also be accentuated by reduced emulsification efficiency, leading to larger fat globule clusters that intensify light absorption [
14,
19]. Conversely, T3 displayed a reduced
b* (25.74), plausibly due to carotenoid degradation during sterilization, in line with documented thermal sensitivity of these pigments [
40,
41].
It should be emphasized that this study did not include pigment-specific analyses such as high-performance liquid chromatography (HPLC) or UV–Vis spectrophotometry. As a result, the interpretation of
L* reduction and
b* shifts as outcomes of Maillard intermediates or carotenoid degradation should be considered with caution. Future investigations combining instrumental colorimetry with molecular-level characterization and sensory validation are warranted [
22,
36].
In summary, the colorimetric data demonstrate that fat source exerts a decisive influence on the visual attributes of ice cream. Variations in
L* and
b* values reflect intrinsic differences in lipid composition, pigment content, and thermal processing history of the fat ingredients. These findings underscore the strategic role of fat selection in shaping the final visual appeal of ice cream and, consequently, its acceptance in the marketplace [
14,
41].
4.6. Texture Determination
The texture of ice cream is a pivotal quality attribute that profoundly influences consumer perception and overall product acceptability. This attribute emerges from the complex interplay among ice crystals, fat globules, air cells, and the frozen serum matrix, which collectively determines the microstructure and mechanical integrity of the product [
4,
14]. Instrumental texture analysis provides objective quantification of critical mechanical parameters such as firmness, cohesiveness, and resistance to deformation, which closely correlate with sensory attributes including creaminess, hardness, and melt profile [
3,
22]. Understanding and controlling these parameters is essential for formulating ice creams that meet both technological requirements and consumer expectations.
In this study, the extrusion test was selected to evaluate ice cream firmness or hardness due to its superior sensitivity in mimicking the consumer’s scooping action compared to the conventional cone penetration test [
4,
19]. Furthermore, extrusion testing engages a larger sample volume, enhancing representativeness in heterogeneous semi-solid systems such as ice cream, wherein the tri-dimensional structure results from the interaction between air cells, fat crystal networks, and ice crystals [
22,
27]. This methodological advantage allows for a more comprehensive assessment of structural nuances that critically influence rheological and sensory outcomes.
Statistical analysis revealed significant differences (
p < 0.05) in firmness across formulations (
Figure 5). The butter-based formulation (T2) exhibited the highest maximum extrusion force, indicative of a more rigid and mechanically robust matrix, followed sequentially by fresh cream (T4), UHT cream (T3), and vegetable fat (T1), which presented the lowest firmness values. These differences can be attributed to the distinct lipid composition and crystallization behaviors inherent to each fat source [
14,
25]. Butter, characterized by a high solid fat content at freezing temperatures (~−18 °C) and polymorphic fat crystallization, forms a stable crystalline network that restricts water mobility and enhances mechanical strength [
4], thus reinforcing the ice cream’s structural integrity.
Additionally, the butter formulation demonstrated the lowest overrun, which correlates with increased density and fewer air-induced structural discontinuities, thereby elevating resistance to deformation [
22,
27]. Conversely, the vegetable fat-based formulation, although plastically stable, possesses a reduced solid fat fraction at freezer temperatures and contains emulsifiers that impede partial coalescence of fat globules, limiting the formation of aggregated fat structures essential for mechanical reinforcement [
14,
25]. Consequently, this formulation yields a softer texture with higher deformability under compressive forces.
Intermediate firmness values observed for the UHT cream (T3) and fresh cream (T4) formulations likely reflect differences induced by thermal processing. The reduced firmness of the UHT cream may be associated with protein denaturation caused by high-temperature treatment, which can compromise fat emulsification and water-holding capacity, resulting in lower matrix stability and mechanical strength [
19,
29]. However, no direct measurements of protein denaturation (such as solubility assays or protein fraction analyses) were performed in this study; thus, this explanation should be considered a plausible mechanism supported by literature rather than a confirmed finding. Fresh cream, by retaining native milk proteins, favors greater stabilization of the continuous phase and facilitates the formation of a more cohesive fat network, resulting in slightly higher firmness despite its lower fat content compared with butter [
32].
The results presented here exclusively reflect the measurement of hardness through the extrusion test. Other relevant texture parameters—such as cohesiveness, elasticity, and adhesiveness—were not determined. This means that the data describes the resistance to compression with precision but do not capture the multi-dimensional nature of ice cream texture. Future investigations should incorporate a broader instrumental texture profile, combined with sensory analyses, to provide a more comprehensive understanding of the relationship between structure, mechanical properties, and consumer-perceived quality [
44,
45].
4.7. Nutritional Composition
The comprehensive nutritional profiling of the ice cream formulations (T1, T2, T3, and T4), detailed in
Table 3, underscores significant variability in macronutrient distribution and lipid composition. These parameters are paramount in both product development and nutritional risk assessment [
19,
26,
45]. The precise quantification of energy content, protein, carbohydrates, and total and saturated fats, alongside sodium levels, enables a multi-dimensional evaluation of the formulations’ nutritional quality and their alignment with dietary recommendations and public health policies [
46,
47].
The energy values were distinctly heterogeneous, with formulation T2 demonstrating the lowest caloric density (117.5 kcal/100 g), contrasted by T3′s highest caloric load (126.4 kcal/100 g). This variance reflects the differential contributions of lipid and protein fractions, which are the predominant macronutrients influencing energy density in frozen dairy matrices [
7,
18]. The carbohydrate content across samples remained generally consistent, save for T3, which exhibited a relatively elevated carbohydrate concentration (23.3 g/100 g). This anomaly is plausibly linked to the inclusion of polysaccharide-based stabilizers or other carbohydrate-rich additives, which are commonly utilized in UHT cream formulations to modulate rheology and shelf-life [
44,
48].
Protein quantification further delineated the formulations, with T3 displaying a notable increase (4.8 g/100 g) relative to the other samples (3.2–3.3 g/100 g), corroborating the elevated protein content characteristic of UHT-treated dairy creams vis-à-vis butter or plant-based fats [
18,
49]. The nutritional implications of heightened protein levels extend beyond caloric contribution, potentially enhancing satiety responses and supporting anabolic processes, making them particularly relevant for demographic groups with elevated protein requirements such as the elderly, athletes, or clinical populations [
50,
51]. A pronounced divergence was observed in saturated fat content, which is a critical determinant in cardiovascular risk modulation [
47,
52]. The low-trans vegetable-fat-based formulation (T1) exhibited the minimal saturated fat fraction (5.2 g/100 g), being consistent with lipid profiles dominated by unsaturated fatty acids prevalent in plant-derived fats [
50,
52]. Conversely, T4 (fresh cream) manifested the highest saturated fat concentration (27.5 g/100 g), reflective of the intrinsic fatty acid composition of full-fat dairy lipids, which include significant proportions of palmitic and myristic acids that are both implicated in lipid metabolism and cardiovascular health outcomes [
1,
53]. The intermediate saturated fat levels in T2 (butter) and T3 (UHT cream) (7.3 and 8.3 g/100 g, respectively) reflect compositional nuances influenced by moisture content, processing modalities, and fat crystallization behavior [
28].
The sodium concentration also varied significantly, with T3 presenting the highest level (77.6 mg/100 g), which is attributable to the incorporation of salt-containing stabilizers and preservatives often added to UHT cream formulations to ensure microbiological stability and texture retention [
53]. In contrast, T4′s sodium content (49.7 mg/100 g) was the lowest, aligning with its composition of minimally processed dairy devoid of additive inclusion [
1,
22]. Given the global emphasis on sodium reduction strategies to mitigate hypertension and cardiovascular disease risk [
46,
47], these findings emphasize the criticality of ingredient and process selection in product formulation.
The utilization of advanced nutritional software tools facilitated precise compositional analysis and enabled iterative formulation adjustments to meet stringent regulatory and health-oriented targets [
19,
54]. This methodological integration exemplifies the intersection of food science and nutritional epidemiology, addressing the contemporary challenge of balancing sensory attributes, product functionality, and health promotion [
14,
51]. Collectively, the comparative nutritional insights derived from these formulations substantiate the pivotal role of fat source selection in optimizing ice cream formulations for targeted consumer groups, aligning product innovation with evolving nutritional paradigms and market demands [
39,
55].
4.8. Correlation Analysis Among Indicators
The integration of physicochemical, rheological, and textural parameters allowed for the identification of significant correlations among the evaluated variables. A strong negative correlation was observed between overrun and hardness (r = −0.87;
p < 0.05), indicating that formulations with higher air incorporation, such as T1 (31.3%) and T3 (29.7%), exhibited lighter and less resistant matrices, while T2, with lower overrun (18.5%), resulted in the highest structural firmness [
3,
12]. This behavior reinforces the role of aeration as a determinant of texture, since the presence of air cells reduces density and increases softness [
1,
2].
In addition, a positive correlation was verified between storage modulus (G′) after maturation and melting resistance (r = 0.81;
p < 0.05). The T3 formulation, produced with UHT cream, presented the highest G′ (>0.5 Hz) and simultaneously the lowest mass loss during the melting test (13 g in 45 min), confirming that more elastic systems provide greater thermal resistance and structural stability [
16,
20]. The protein denaturation effect induced by UHT treatment seems to have contributed to the coupling between whey proteins and fat globules, favoring the formation of a cohesive and stable network [
10,
19]. Furthermore, tan δ values greater than 1, observed in T1, were associated with a predominance of viscous behavior and with faster melting (63 g in 45 min), suggesting reduced ability to retain unfrozen water and lower continuity of the fat–protein network [
22,
24]. This finding is consistent with the hypothesis that vegetable fats with lower solid fat content at −18 °C and less stable polymorphism reduce the ability to form robust three-dimensional structures [
11,
28].
The correlations also indicate a functional coupling between texture and thermal stability: The T2 formulation, with the highest hardness (extrusion force), showed the lowest overrun and intermediate melting (~50 g), suggesting that the rigid crystalline network formed by butter increases mechanical resistance but does not guarantee maximum stability against heating [
7,
21]. This result highlights the need to consider the interaction among lipid crystallinity, partial coalescence, and system elasticity as determinants of overall ice cream performance [
18,
26].
Therefore, the correlation analysis reveals consistent structural relationships: (i) higher overrun → lower hardness; (ii) higher G′ → greater melting resistance; (iii) tan δ > 1 → higher susceptibility to melting. These findings support a clear causal logic of the type “fat source → structure → functional quality,” as suggested by both classical and recent studies on structural stabilization in ice cream systems [
1,
20,
27,
53].