Novel Formulation of Low-Fat Milk Chocolate: Impact on Physicochemical and Sensory Properties

Abstract

This study proposes a dual-approach strategy to formulate reduced-fat milk chocolate by combining cocoa butter (CB) substitution with emulsifier-based rheological optimization. CB was partially replaced at 20%, 30%, and 40% using whey protein isolate (WPI) and inulin (IN) blends (70:30, 50:50, 30:70 w/w). CB reduction increased plastic viscosity and yield stress, particularly in WPI-rich systems. The 50:50 WPI:IN ratio consistently minimized rheological drawbacks while maintaining melting, texture, and sensory quality. Caloric content was reduced by up to 9% (~50 kcal/100 g), most notably in IN-dominant samples. To overcome flow challenges at high substitution levels, emulsifiers—lecithin, ammonium phosphatide (AMP), and polyglycerol polyricinoleate (PGPR)—were assessed. AMP (≤0.5% w/w) and PGPR (0.15–0.3% w/w) effectively reduced viscosity and yield stress; lecithin showed limited effect above 0.6%. The optimized system (0.5% AMP + 0.15% PGPR) applied to 40% CB-reduced chocolate with 50:50 WPI:IN restored desirable rheology (3.42 Pa·s viscosity; 7.91 Pa yield stress) and improved mouthfeel and acceptability. This integrated formulation enables significant fat and calorie reduction without compromising product quality, supporting the development of healthier chocolate products.

1. Introduction

Obesity has surged globally, more than doubling in prevalence since the 1980s when it was first identified as a critical health issue [1]. As of 2023, the World Health Organization (WHO) estimates that over 2 billion adults—more than 40% of the global population—are overweight, with over 750 million classified as obese [2,3,4]. Obesity is a key risk factor for a wide range of chronic conditions, including cardiovascular disease, type 2 diabetes, musculoskeletal disorders, and several forms of cancer. At the individual level, one of the most effective and accessible interventions is reducing daily caloric intake from fats and sugars [4]. In response, health-conscious consumers increasingly seek out food options that align with preventive nutrition and disease risk reduction [5].

Chocolate, a globally consumed indulgent food, is particularly rich in sugar and fat, notably cocoa butter (CB). CB plays a vital role in determining chocolate’s smooth texture, gloss, snap, and unique melting behavior, as it forms a continuous fat matrix that encapsulates sugar and cocoa particles [6,7]. However, CB’s high saturated fat content and caloric density have raised nutritional concerns, especially as consumer awareness of diet-related chronic diseases grows [8,9]. Unlike sugar reduction, which has been widely adopted in commercial chocolate products, reducing CB content presents significant technological challenges due to its structural and rheological roles in chocolate matrices. Consequently, innovative strategies are required to partially replace CB with alternative ingredients that maintain chocolate’s physicochemical integrity and sensory acceptability [10,11]. Innovative fat replacers such as oleogels, fiber–protein composites, and structured emulsions are now being explored as viable CB alternatives that may retain desired properties while reducing overall fat content [12,13].

Efforts to reduce fat in chocolate have focused on incorporating alternative structuring agents such as dietary fibers and proteins. Inulin (IN), a soluble dietary fiber with prebiotic properties, and whey protein isolate (WPI), known for its emulsifying and network-forming abilities, have each shown potential as partial fat replacers. Previous studies demonstrated that IN and β-glucan could partially replace CB while maintaining acceptable textural and sensory attributes, although substitution levels rarely exceeded 20% without significant changes to product quality [14]. Similarly, combining WPI with various carbohydrate-based fat mimetics improved sensory and physical performance in reduced-fat chocolate ice creams [15].

Despite progress in the development of low-fat chocolate formulations, most studies have either limited CB replacement to relatively modest levels (typically under 30%) or focused predominantly on lipid-based substitutes such as oleogels, vegetable oils, and CB equivalents [11,12]. These approaches, while effective to some extent, often fail to preserve critical processing attributes such as viscosity and yield stress, which are fundamental to manufacturability and product quality [9,10]. Furthermore, the impact of high levels of CB substitution on texture, mouthfeel, and consumer acceptability has not been thoroughly investigated in protein–fiber composite systems.

Additionally, most existing research emphasizes fat alternatives rather than integrated, multi-component approaches involving both macronutrient replacement and emulsifier optimization. This leaves a knowledge gap regarding the synergistic potential of protein–fiber blends—such as WPI and IN—in solid chocolate matrices. Meanwhile, the integration of functional bioactives into chocolate formulations is an emerging area, with recent studies showing that microencapsulation techniques can enable the inclusion of health-promoting ingredients without compromising product quality [16].

The present study introduces a novel dual-strategy approach to address both nutritional and processing challenges associated with CB reduction in milk chocolate. The first component involves partially replacing CB with a structured blend of WPI and IN, applied at three substitution levels (20%, 30%, and 40%) and three different WPI:IN ratios (70:30, 50:50, and 30:70 w/w). This strategy aims to retain the essential physicochemical and sensory qualities of traditional milk chocolate, including texture, melting behavior, and palatability.

To complement this approach, the study further investigates the use of advanced emulsifiers—ammonium phosphatide (AMP) and polyglycerol polyricinoleate (PGPR)—to improve the rheological performance of high-viscosity formulations. These emulsifiers are evaluated independently and in combination with conventional lecithin to mitigate the increases in plastic viscosity and yield stress that often accompany high protein- and fiber-based substitutions. Their role is critical in ensuring the processability and structural stability of the reformulated chocolate without compromising consumer acceptability.

This formulation strategy, supported by comprehensive preliminary optimization and validated analytical methods, enables up to 40% CB reduction while preserving essential quality attributes such as texture, flow behavior, and sensory appeal. By combining nutritional improvements with targeted formulation techniques, the study provides a practical and scalable solution for the development of reduced-fat milk chocolate suitable for commercial production.

2. Materials and Methods

2.1. Materials

The ingredients used for chocolate sample preparation included cocoa mass (liquor), skimmed milk powder (0.2% fat), cocoa butter (CB), granulated sugar, vanilla, and emulsifiers (lecithin and PGPR), all kindly provided by the food industry JOTIS S.A. (Athens, Greece), Inulin (IN), used as a CB replacer, was also supplied by JOTIS S.A., while whey protein isolate (WPI) was sourced as Simplesse^® 100 Whey Protein (CP Kelco, Atlanta, GA, USA). The PGPR provided by JOTIS S.A. was used for both the conventional chocolate and the formulations investigated in Section 2.3. Meanwhile, polyglycerol polyricinoleate (Palsgaard^® PGPR 4150), generously provided free of charge by Palsgaard A/S (Juelsminde, Denmark), was used exclusively in the experiments as described in Section 2.10 and Section 2.11. Additional emulsifiers—ammonium phosphatide (Palsgaard^® AMP 4455)—were also generously provided by Palsgaard A/S, whom we gratefully acknowledge.

2.2. Chocolate Production

Chocolate sample production was conducted using the Cocoa T Deluxe Melanger ECGC-12SLTA/12SL (CocoaTown LLC., Alpharetta, GA, USA), which performs simultaneous mixing, particle size reduction, and conching. The recipe of the reference milk chocolate formulation is presented in

Table 1. Composition of the reference chocolate formulation without CB substitution (values expressed as g per 100 g of finished chocolate).

Ingredient	Amount (g/100 g Chocolate)
Cocoa mass	17.50
Sugar	38.35
Skimmed milk powder	23.55
CB *	20.00
Lecithin	0.30
PGPR **	0.22
Vanilla	0.08

* CB: Cocoa Butter. ** PGPR: Polyglycerol Polyricinoleate.

.

The process began by melting CB and cocoa mass—comprising the fat and non-fat cocoa solids—separately at 50 °C. Once fully liquefied, both were added to the preheated Melanger [16]. Subsequently, the dry ingredients, including granulated sugar, skimmed milk powder, and any CB replacers such as IN and WPI (in substituted formulations), were gradually incorporated. The conching and refining stages were carried out continuously for 24 h, enabling uniform distribution of ingredients and reduction in particle size to below 30 μm, as verified by prior standardization. After this initial processing, the emulsifiers (e.g., lecithin, AMP, or PGPR) and flavoring agents such as vanilla extract were added. The mixture was then conched for an additional 2 h to ensure homogeneity and proper emulsifier integration.

The tempered chocolate was prepared through a controlled crystallization sequence. The mass was transferred onto a marble slab and cooled to 18 °C, promoting the formation of stable CB crystals (β-polymorphs). This was followed by reheating the mixture to approximately 30 °C, selectively melting any unstable forms. The tempered mass was cast into polycarbonate molds and subjected to final solidification by cooling at 10 °C.

2.3. Cocoa Butter Substitution Using WPI:IN

Fat replacement was carried out by reducing the CB content at specific substitution levels and replacing it with varying ratios of IN and WPI as presented in

Table 2. Experimental design for CB substitution using IN and WPI: proportions of CB, IN, and WPI as g per 100 g of chocolate sample.

Substitution Level	WPI:IN Ratio	CB *	WPI **	IN ***
0%	-	20	0	0.0
20%	70:30	16	2.8	1.2
	50:50	16	2.0	2.0
	30:70	16	1.2	2.8
30%	70:30	14	4.2	1.8
	50:50	14	3.0	3.0
	30:70	14	1.8	4.2
40%	70:30	12	5.6	2.4
	50:50	12	4.0	4.0
	30:70	12	2.4	5.6

* CB: Cocoa Butter. ** WPI: Whey Protein Isolate. *** IN: Inulin.

.

2.4. Rheological Analysis

Rheological measurements were conducted using a Physica Rheometer MCR Series (51/101/301/501, Anton Paar, Ostfildern, Germany) with a PP25/P2 measuring system. Samples were preheated to 40 °C and placed on the rheometer plate. After equilibration for 2 min, the shear rate was increased from 0.1 to 100 s⁻¹. The resulting flow curve was constructed by plotting shear stress (τ, Pa) against shear rate (γ, 1/s). Plastic viscosity (k, Pa s) and yield stress (τ₀, Pa) were calculated by fitting the data to the Casson model, described by Equation (1) [17].

$$\sqrt{\tau }=\sqrt{{\tau }_0}+\sqrt{k}\times \sqrt{\gamma }$$

(1)

Rotational rheological measurements of the molten white chocolates were performed using a Physica MCR 301 rheometer (Anton Paar, Ostfildern, Germany) with plate-plate geometry (2 mm gap). Chocolate samples were prepared by heating in an oven at 45 °C for one hour to achieve a molten state. Approximately 2–4 g of the melted chocolate was then applied to the plate, and the samples were tempered at 40 °C for 2 min. Shear stress was recorded while varying the shear rate from 0.1 to 100 1/s. All the tests were performed at 40 °C. Shear rate (1/s) and shear stress (Pa) were recorded electronically using Rheoplus/32 V3.40 software. The data were fitted to the Casson model, with the Casson yield stress (τ₀) calculated as the square of the intercept and the Casson plastic viscosity as the square of the slope (γ̇). All measurements were conducted in triplicate to ensure accuracy and reproducibility.

2.5. Color Measurement

Color measurements were performed according to the CIELAB color system using a HunterLab MiniScan XE colorimeter (Hunter Associates Laboratory Inc., Reston, VA, USA) equipped with a 4 mm measuring aperture. Prior to analysis, the instrument was calibrated using standard white and black reference tiles. Each chocolate sample was measured at three different surface locations to account for variability, and the results were recorded as the average of these measurements, along with the corresponding standard deviations. The CIELAB parameters L*, a*, and b* were used to describe the color characteristics of the chocolate. The L* value represents lightness (ranging from 0 = black to 100 = white), a* indicates color along the red-green axis, and b* along the yellow-blue axis. In addition, chroma (C*), which reflects color saturation or intensity, was calculated using Equation (2) [18].

$$C^{*}=\sqrt{({a^{*}}^2+{b^{*}}^2})$$

(2)

All measurements were conducted in triplicate to ensure reliability and reproducibility of the results.

2.6. Differential Scanning Calorimetry (DSC)

The thermal behavior of the chocolate samples was evaluated using a differential scanning calorimeter (DSC-6, Perkin Elmer Ltd., Shelton, CT, USA) operated with Pyris 6 software (version 4.01). Approximately 45 mg of each chocolate sample was weighed into standard aluminum pans, which were then hermetically sealed. An empty, sealed pan was used as the reference. The samples were first held at 0 °C for 1 min to stabilize the thermal environment and were subsequently heated to 55 °C at a constant rate of 5 °C/min under a nitrogen flow of 20 mL/min to ensure an inert atmosphere. Thermograms were recorded for each sample, from which peak temperature (T_peak: melting temperature) was determined. Each analysis was conducted in triplicate to ensure accuracy and repeatability of the measurements.

2.7. Hardness Measurement

Prior to testing, all chocolate samples were equilibrated at room temperature for five hours to ensure uniform thermal conditions. The mechanical hardness was evaluated through uniaxial compression tests using a universal testing machine (Zwick, model Z2.5/TN1S, Ulm, Germany). Each cylindrical sample was measured for its initial height (L₀) and diameter using a Vernier caliper to calculate the cross-sectional area (A). The tests were conducted at ambient temperature with a constant deformation rate of 5 mm/min and a maximum applied load of 2000 N.

Force (F, Ν) and deformation (ΔL, m) were continuously recorded using Zwick PC Software (Version 3.1). These values were used to calculate stress (σ, Pa) and strain (ε), and from their ratio, the modulus of elasticity (E_mod, N/mm²) was determined as an index of the sample’s resistance to deformation (i.e., its hardness). All measurements were performed in triplicate to ensure reproducibility.

The mechanical calculations are described by Equations (3)–(5) [19].

$$\sigma ={\displaystyle \frac{F}{A}}$$

(3)

$${\epsilon }_n={\displaystyle \frac{\Delta L}{L_o}}$$

(4)

$$E_{mod}={\displaystyle \frac{\sigma }{\epsilon }}$$

(5)

To assess the stability of textural properties over time, hardness measurements were repeated after one month of storage. Chocolate samples were stored at 25 °C under controlled conditions in sealed containers to prevent moisture uptake and oxidation. After the storage period, the same uniaxial compression protocol was applied, maintaining identical testing parameters (temperature, load, and deformation rate).

2.8. Caloric Value Determination

The caloric content of each chocolate formulation was theoretically calculated based on the known energy values of the individual ingredients, as provided by their respective manufacturers. By applying proportional calculations to the quantities used per 100 g of chocolate sample, the total energy content (in kcal/100 g) for each formulation was determined, as described in

Table 3. Energy content of chocolate ingredients.

Ingredient	Energy Value (kcal/g)
Cocoa mass	6.40
CB *	8.84
Sucrose	3.87
Skimmed milk powder	3.57
IN **	2.50
WPI ***	3.70
Lecithin	6.00
Vanilla	2.88

* CB: Cocoa Butter. ** IN: Inulin. *** WPI: Whey Protein Isolate.

.

2.9. Sensory Evaluation Protocol

The chocolate samples were randomly assigned three-digit codes and placed on white plates before being presented to the sensory panel, accompanied by a glass of water to cleanse the palate between tastings. Evaluation was conducted using the traditional 9-point scale, where 1 represents the lowest and 9 the highest level of intensity. The sensory attributes assessed included appearance, taste, texture, aroma, and overall acceptability. A total of ten trained panelists, both men and women, participated in the evaluation. The first sensory trial aimed to compare samples with three different levels of CB substitution: 20%, 30%, and 40%. In all cases, the fat replacers were used at a fixed ratio of WPI:IN 50:50.

The second sensory trial examined samples with a constant CB substitution level of 40% but varying ratios of the replacers: WPI:IN (70:30, 50:50, 30:70). This highest substitution level was selected to maximize the differences in ingredient content, thereby enhancing the panelists’ ability to detect potential variations in sensory properties.

2.10. Effect of Emulsifiers on Chocolate Rheology

The influence of emulsifiers—including lecithin, sunflower-derived AMP, and PGPR—on the rheological behavior of reference chocolate samples was systematically evaluated. Lecithin and AMP were tested individually at concentrations ranging from 0.1% to 0.7% w/w, while PGPR was assessed at concentrations between 0.05% and 0.3% w/w.

In the experiments investigating the effects of lecithin and AMP, PGPR was deliberately excluded to allow for independent evaluation and reliable interpretation of each emulsifier’s contribution. The selected concentration ranges reflect typical industrial application levels for these compounds in chocolate manufacturing. For the PGPR evaluation trials, a constant lecithin concentration of 0.3% was maintained across all samples. This concentration facilitated adequate processing while minimizing its confounding influence on the rheological measurements. Only rheological parameters were measured in these tests to isolate the impact of the individual emulsifiers on flow behavior following the procedure described in Section 2.4.

2.11. Emulsifier Optimization for High-Level Cocoa Butter Replacement

To improve the processing properties of reduced-fat chocolate formulations, further optimization was carried out on the sample with 40% CB replacement using a 50:50 ratio of WPI:IN. This formulation had previously demonstrated elevated viscosity, necessitating emulsifier adjustment.

A series of trials were conducted using various concentrations of AMP and PGPR, either alone or in combination, to evaluate their effects on both rheological behavior and sensory quality. AMP was tested at concentrations of 0.5%, 0.6%, and 0.7%, while PGPR was introduced at incremental levels up to 0.15%, in formulations where AMP was fixed at 0.6%.

Both viscosity and yield stress were measured to assess flow characteristics, while sensory evaluations were conducted to determine the impact of emulsifier combinations on mouthfeel and overall acceptability (

Table 4. AMP and PGPR concentration combinations used to optimize rheology and sensory properties in chocolate with 40% CB replacement (WPI:IN = 50:50).

AMP * Concentration (%)	PGPR ** Concentration (%)
0.5	0
0.6	0
0.7	0
0.6	0
0.6	0.05
0.6	0.15

* AMP: Ammonium Phosphatide. ** PGPR: Polyglycerol Polyricinoleate.

).

2.12. Statistical Analysis

One-way and factorial analysis of variance (ANOVA) were applied in order to analyze the differences. Tukey’s range test (a = 0.05) was applied, and all the statistical tests were performed with SPSS software (Version 21).

3. Results and Discussion

3.1. Effects of Cocoa Butter Substitution on Chocolate Properties

3.1.1. Rheological Properties

To evaluate the applicability of the Casson model to chocolate samples with CB substitutions, shear stress vs. shear rate data were transformed using square roots of both axes by the Rheoplus software (Version 2.61) (Figure 1). This transformation linearizes the flow curve and allows for estimation of Casson plastic viscosity and yield stress, which are key parameters in chocolate rheology.

Figure 1 compares the shear behavior of a control chocolate (no fat replacement) to a sample with 40% CB substitution using WPI:IN 50:50. The control sample exhibits a strong linear correlation (R² = 99.27%) to the Casson model, indicating that its flow behavior is well described by the classic Casson-type rheology—a model widely used for molten chocolate characterization due to its ability to predict both yield stress and plastic viscosity. However, the sample with the 40% substitution deviates from linearity, suggesting a disruption in matrix uniformity and flow predictability. This deviation implies that at high substitution levels, the alternative ingredients interfere with the structural and flow dynamics of chocolate, making the Casson model less applicable [20,21].

Similar results have been observed by researchers who found that sugar replacements like inulin or polydextrose significantly increased apparent viscosity and yield stress, reducing the model’s predictive fit [14]. Moreover, chocolate containing fat substitutes such as xanthan gum, corn starch, or glycerin blends also failed to follow the Casson model due to structural interactions disrupting homogeneity [21]. Finally, ref. [22] observed that emulsifier and fat concentrations modulate the Casson model fit in chocolate systems, with high emulsifier levels improving flow prediction.

Figure 2 presents the Casson plastic viscosity across different substitution levels (20%, 30%, 40%) and WPI:IN ratios (70:30, 50:50, 30:70). Viscosity, which quantifies the resistance of a fluid to deformation under shear stress, serves as a direct indicator of the energy required to maintain flow.

The results reveal a progressive increase in viscosity with higher substitution levels, regardless of the WPI:IN ratio. This indicates that the addition of WPI and IN alters the chocolate structure by strengthening intermolecular interactions or restricting particle mobility, which increases resistance to flow [23,24]. The control chocolate (no CB substitution) had the lowest viscosity (2.24 ± 0.33 Pa·s), typical of a smooth fat-rich CB matrix. In contrast, viscosity reached 6.07 ± 0.82 Pa·s at 40% substitution with the 70:30 WPI:IN ratio [24].

These results agree with rheological theory, which predicts that higher solid content increases flow resistance. Supporting studies confirm that inulin strongly contributes to this effect: Aidoo et al. (2017) reported that its fibrous and water-binding properties elevate viscosity and yield stress [24], while Rezende et al. (2015) observed similar effects with inulin and β-glucan in sugar-free chocolate [14]. More recent work by Lim et al. (2021) and Longoria-García et al. (2020) also showed that inulin increases both viscosity and yield stress due to its bulk properties [23,25].

At 20% substitution, the effect of the WPI:IN ratio was minor, with no significant rheological changes. At 30% substitution, however, the role of composition became evident: higher IN levels led to much greater viscosity, whereas increasing WPI (up to 70%) caused only moderate changes. This suggests that WPI has a limited effect on viscosity compared to IN. At 40% substitution, a non-monotonic (V-shaped) pattern was observed. The 50:50 WPI:IN blend produced the lowest viscosity, while ratios skewed toward either WPI (70:30) or IN (30:70) resulted in sharp increases. This indicates a synergistic effect at intermediate ratios, whereas extremes in either component disrupted the balance and increased flow resistance.

Figure 3 shows the Casson yield stress, which reflects the minimum force required to initiate flow—an indicator of the internal structural strength of the chocolate.

In the control sample, yield stress was lowest (49.9 ± 6.31 Pa), highlighting a flowable matrix with minimal internal resistance. At 20% substitution, yield stress remained low across all WPI:IN ratios, indicating minimal structural change at this level. Ref. [26] similarly noted that inulin’s impact on yield stress only becomes prominent at higher concentrations (>10%), as its low solubility limits its structuring ability at lower levels. Especially, at the 20% substitution level, when WPI constitutes 70% of the substitution ratio (WPI:IN = 70:30), the Casson yield stress remains almost identical to that of the control sample. Given that 20% of the CB—an important lipid phase responsible for lubrication and flow—has been replaced, a noticeable increase in yield stress would be theoretically expected. The absence of this increase suggests that whey protein may exert a stabilizing effect on the rheological system, possibly by forming weak protein-based dispersion networks that prevent destabilization despite the partial loss of the fat matrix. This behavior aligns with, Shah et al. (2010) who demonstrated that formulations containing inulin and whey protein showed little to no increase in yield stress at low substitution levels, due to protein hydration counteracting solid-phase interactions [27].

However, at 30% and 40% substitution levels, the yield stress increased substantially, particularly for samples high in WPI. At 40% substitution and WPI:IN = 70:30 (i.e., high WPI content), the highest yield stress was observed (414.26 ± 6.97 Pa). This implies that WPI increasingly acts as a network former, creating internal resistance that delays flow onset. This trend aligns with findings by Zarić et al. (2024), who demonstrated that WPI enhances structural rigidity in chocolate systems [28].

In contrast, high IN samples (e.g., 70% IN at 40% substitution) increased yield stress, but not to the same extent. IN appears to increase matrix thickness without substantially enhancing internal bonding. Interestingly, samples containing a balanced WPI:IN ratio (50:50) consistently showed the lowest increase in yield stress across substitution levels. This suggests a compositional synergy, where the opposing rheological contributions of WPI (network-forming) and IN (dispersed-phase thickener) may counterbalance each other.

In summary, the rheological behavior of chocolate is significantly influenced by both the level of CB substitution and the ratio of IN to WPI. As substitution levels increase, particularly at 40%, the linearity of the shear stress–shear rate relationship declines, indicating that the Casson model becomes less applicable due to changes in the chocolate matrix structure. Plastic viscosity also increases with higher substitution levels, with the most pronounced viscosities observed at extreme WPI:IN ratios (either WPI- or IN-rich), while a balanced 50:50 ratio results in comparatively lower viscosity. In terms of yield stress, WPI demonstrates a stronger capacity for network formation, especially at higher substitution levels, whereas IN mainly enhances viscosity through thickening without significantly reinforcing internal structure. Notably, the 50:50 WPI:IN ratio consistently moderates both viscosity and yield stress across all substitution levels, suggesting it offers a synergistic balance that preserves desirable flow properties in reduced-fat chocolate formulations.

3.1.2. Color Attributes

Color is a pivotal quality attribute in chocolate, shaping consumer expectations and product perception prior to consumption. Typically, darker hues are interpreted as indicative of higher cocoa content, while lighter tones are associated with milkier formulations. In this study, the impact of CB substitution using WPI and IN on chocolate color was assessed, with emphasis on chroma (C*), which represents color intensity (Figure 4), and lightness (L*), denoting brightness (Figure 5).

Although IN has been previously linked to darker coloration in chocolate due to its Maillard reaction potential and its ability to increase color saturation [29], there is limited prior data on the optical effects of WPI when used in fat substitution. The combined impact of these two ingredients—particularly as fat replacers—remains largely unexplored.

In the present study, chroma values (C*) showed minimal variation among the samples and did not exhibit a consistent pattern associated with either the substitution level or the relative proportions of WPI and IN. Therefore, L* values were evaluated more closely, as changes in brightness are more likely to be perceived visually and can significantly influence consumer preference.

Analysis of L* values revealed that none of the substituted samples differed significantly in lightness from the reference chocolate, a finding corroborated by both visual inspection during production and sensory evaluation results, which indicated no perceptible differences in appearance across formulations.

Thus, within the substitution levels tested, the use of WPI:IN as fat replacers does not significantly affect the color characteristics of chocolate. This finding enhances the formulation flexibility for reduced-fat or functional chocolate development without compromising visual quality.

3.1.3. Melting Behavior (T_m, °C)

Figure 6 presents the melting temperature (T_m) values of the chocolate samples as a function of both the substitution level and the WPI:IN ratio. Melting behavior is a critical quality attribute in chocolate, strongly influenced by CB’s polymorphic crystalline forms. The T_m reflects the onset of phase transition and is indicative of both tempering quality and fat composition. These values are especially important in assessing how alternative ingredients impact the thermal stability and sensorial properties of chocolate.

The melting points of many tested chocolate samples ranged from 24.3 °C to 25.8 °C which is slightly lower than that of optimally tempered commercial chocolate, typically reported to melt between 30 °C and 34 °C. However, these values are comparable to some industrial chocolates processed under less rigorous tempering conditions, which can melt in the 25–30 °C range. The narrow melting range (ΔT ≈ 1.5 °C) across all samples indicates relatively consistent thermal transitions, suggesting that the processing method was sufficiently controlled to allow for meaningful comparisons. Despite this, some samples exhibited statistically significant differences in melting points, particularly at higher substitution levels, which may be attributed to variations in crystal formation.

These differences point to the presence of less stable polymorphic forms—such as form II or form III—rather than the preferred form V or VI of CB. The formation of unstable crystals is often a result of suboptimal tempering, a common challenge in laboratory-scale chocolate production where precision in temperature and shear control is limited.

When examining the influence of the WPI:IN ratio, results varied across groups. In the 50:50 and 30:70 WPI:IN formulations, melting points remained consistent regardless of substitution level, indicating that the type and amount of substitute had minimal impact on the thermal behavior of the chocolate matrix. However, in the 70:30 WPI:IN group isolated differences were observed, especially between the 30% and 40% substitution levels, where a slight decrease in T_m was recorded. This drop may be due to batch-specific crystallization inconsistencies or localized disruptions in fat crystallization caused by higher concentrations of WPI, which has been reported to interfere with fat structuring due to its hydrophilic nature.

At lower substitution level (20 and 30%), the melting point remained largely unaffected by variations in WPI:IN ratio, reinforcing the idea that minor replacements of CB do not significantly alter the thermal transition behavior. These findings align with studies by Kiumarsi et al. (2021), who reported that partial substitution of sucrose with modified inulin in chocolate had minimal impact on melting behavior, highlighting the dominant thermal profile of cocoa butter [30]. Similarly, Rodriguez Furlán et al. (2017) demonstrated that incorporating inulin as a stabilizer in compound chocolate improved thermal stability and did not significantly alter melting characteristics compared to control samples [31]. Additional support comes from Aidoo et al. (2015), who observed that sugar-free chocolates made with inulin/polydextrose blends retained melting behavior similar to conventional chocolate, with only minor differences in onset temperature and peak width [32].

In conclusion, neither the CB substitution level nor the WPI:IN ratio had a consistent or significant effect on the melting point of chocolate. The observed differences are likely attributable to experimental variability in crystallization, rather than to the specific thermal effects of the substitution ingredients. CB remains the primary determinant of chocolate’s melting behavior, and its unique polymorphic nature governs thermal transitions even in the presence of functional fat replacers such as IN and WPI.

3.1.4. Textural Properties: Hardness

Figure 7 presents the modulus of elasticity (N/mm²) values of the chocolate samples across varying levels of CB substitution and different WPI:IN ratios. Generally, a higher elasticity parameter value indicates greater hardness of the product. Hardness is a key indicator of the structural integrity and bite resistance of chocolate and are influenced by the interplay between fat content, crystalline structure, and dispersed solid components.

At 20% substitution, all samples exhibit the highest overall hardness, regardless of substitute ratio. This is attributed to the partial retention of the CB network, which is still capable of stabilizing the solid matrix. The addition of powdered ingredients such as IN and WPI further increases the solid phase, thereby enhancing mechanical resistance. As a result, in several cases, hardness even exceeds that of the control sample. Similar outcomes were reported by [33], who showed that adding fiber and polyol-based substitutes increased chocolate hardness due to the concentration of solid particles within a partially intact lipid phase.

As substitution increases to 30% and 40%, the CB matrix is progressively disrupted, leading to a partial collapse of the fat network, and resulting in reduced cohesion. Nonetheless, hardness values remain comparable to the reference in many formulations, particularly where WPI is present at ≥50%, suggesting a structural reinforcement effect.

A more detailed pattern emerges when examining substitute ratio. Samples with higher WPI content (50% and 70%) consistently display greater hardness and modulus of elasticity than their IN-dominant counterparts. At 20% substitution, WPI enhances the structural rigidity of the system, as expected from its known capacity to form intermolecular protein networks.

At higher substitution levels (30–40%), the presence of WPI—particularly at 50% or 70% ratios—helps maintain elasticity and firmness near the control levels. This supports findings by Selvasekaran & Chidambaram (2024), who reported that WPI-based emulsifiers preserve textural characteristics like firmness and elasticity, even when cocoa butter is reduced [34].

Although IN has also been associated with increased hardness in chocolate systems, its effect appears less pronounced here. At 70% IN ratios, particularly under 30–40% substitution, both hardness and elasticity decline, likely due to the inability of IN to form cohesive structural networks at low moisture levels. These observations are supported by Rezende et al. (2015), who noted reduced elasticity and only modest increases in hardness with higher IN levels in sugar-free chocolate formulations, and Kiumarsi et al. (2021), who observed increased hardness but decreased elasticity in inulin-rich chocolate systems [14,35]. Notably, the 50:50 WPI:IN ratio consistently yields mechanical properties closest to the reference sample, especially at 30% and 40% substitution levels. This suggests a synergistic or balancing effect between the fibrous thickening action of IN and the network-forming properties of WPI, similar to the findings by Zarić et al. (2024) in comparative studies of mixed-protein chocolate systems [28].

Figure 8 presents the elasticity parameter (N/mm²) of chocolate samples with varying CB substitution levels and WPI to IN ratios over a storage period of 30 days.

Based on the results, the elasticity of chocolate samples generally decreased, particularly at higher substitution levels (30% and 40%), regardless of the WPI:IN ratio. This trend aligns with recent findings by Tatar et al. (2024), who observed an increase in the modulus of elasticity of stored chocolate pralines under high-temperature storage conditions, indicating matrix aggregation and reduced mechanical flexibility [36]. Similarly, Kiumarsi et al. (2020) reported that low levels of hydrophobically modified IN substitution in chocolate increased the modulus of elasticity during storage, while higher substitution levels led to a more consistent, solid-like behavior, suggesting that high IN contents stabilize the elastic properties despite overall reduction in flexibility [35]. Furthermore, Zhao et al. (2018) highlighted that cycling temperatures during storage promoted an increase in the Young’s modulus of elasticity (indicating higher stiffness) while simultaneously decreasing fracture stress, reflecting microstructural changes that compromise mechanical stability despite apparent elasticity gain [37]. Collectively, these studies corroborate our findings that storage time and higher substitution levels generally lead to diminished chocolate elasticity and mechanical performance, reinforcing the role of fat crystal network integrity and moisture migration as key factors in structural degradation.

3.1.5. Caloric Content

In

Table 5. Calories analysis of chocolate samples with varying levels of CB substitution—20%, 30%, 40% and varying WPI:IN ratios: 70:30, 50:50, 30:70.

Fat Substitution	Ratio	Calories (kcal) Per 100 g Chocolate	Caloric Value Reduction Percentage
0%	Control	523.32	-
20%	70:30	499.4	4.57%
20%	50:50	500.36	4.39%
20%	30:70	501.32	4.20%
30%	70:30	487.44	6.86%
30%	50:50	488.88	6.58%
30%	30:70	490.32	6.31%
40%	70:30	475.48	9.14%
40%	50:50	477.4	8.77%
40%	30:70	479.32	8.41%

the calories analysis of the chocolate samples is presented.

As shown in Table 5, the incorporation of fat substitutes in chocolate formulations led to a measurable decrease in caloric content, consistent with the lower energy values of IN (~1.5 kcal/g) and WPI (~4.0 kcal/g) compared to CB (~9.0 kcal/g). As expected, increasing the degree of CB substitution yielded progressive reductions in energy values. The most substantial caloric reduction observed was approximately 9% (≈50 kcal/100 g), achieved in the sample with 40% CB replacement using a 70% IN—30% WPI blend.

This result is consistent with the literature, where similar formulations using inulin and protein-based fat replacers have demonstrated significant energy reductions. Joseph et al. (2022) emphasized that the synergistic use of β-glucan, IN, and CB alternatives can lower caloric density in compound chocolates by up to 12%, particularly when CB is partially removed or diluted with structurally similar but less caloric components [6].

The dominant role of CB in determining chocolate’s energy value is underscored by the fact that the base formulation in this study contained only 20% CB, a relatively low-fat design compared to commercial milk chocolate, which may contain up to 35–40%. Consequently, if a similar substitution strategy were applied to higher-fat chocolate bases, the caloric reduction would likely be even more pronounced. This potential has been noted by Konar et al. (2022), who reported that in high-fat milk chocolates, substituting up to 50% of CB with IN and stevia-based blends could reduce caloric content by as much as 90 kcal/100 g [38].

The type of substitute also plays a crucial role. Due to its significantly lower caloric density, higher proportions of IN consistently resulted in lower total energy values across all substitution levels. In contrast, blends with greater whey protein content still contributed to energy reduction but to a lesser extent. This is in line with the findings of [15] who evaluated reduced-fat chocolates using various carbohydrate–protein systems and found that energy values dropped most dramatically in high- IN formulations. Moreover, inulin’s multifunctional properties—as both a bulking agent and dietary fiber—offer added nutritional value beyond caloric reduction.

In summary, the study demonstrates that fat substitution with IN and WPI is an effective strategy for caloric reduction in milk chocolate. The highest reductions are achieved with higher substitution levels and IN-rich blends. While the absolute energy decrease may appear modest due to the already low-fat base formulation, the implications for full-fat chocolates are significant. These findings support the growing trend toward functional and reduced-calorie chocolate formulations that maintain acceptable sensory and structural characteristics.

3.1.6. Sensory Evaluation Results

The first sensory evaluation aimed to compare chocolate samples with different levels of CB substitution—20%, 30%, and 40%—all prepared using a fixed 50:50 ratio of IN and WPI as fat replacers. The results of this assessment are summarized in Figure 9.

According to the panelists, key attributes such as milk aroma, bitterness, sweetness, and overall acceptability did not exhibit statistically significant differences across substitution levels. Although a slight increase in sweetness and milk aroma was anticipated with higher substitution—due to elevated concentrations of IN and WPI—these changes were not perceptible, suggesting that the levels used may have remained below sensory threshold limits in the chocolate matrix.

A slight variation was noted in the perceived color of the samples. While both the 20% and 40% substitution samples received a score of 6, indicating a slightly darker tone, the 30% substitution sample was rated 5, which corresponds to the benchmark for standard milk chocolate. Although this one-point difference is statistically significant, it remains within the acceptable visual range for milk chocolate, as illustrated in Figure 9.

In terms of hardness, panel scores ranged from 6.5 to 8.5, with the 40% substitution sample rated the highest. This result contrasts with the instrumental texture measurements, where higher CB content (i.e., lower substitution levels) was associated with greater hardness. The discrepancy may be attributed to the influence of graininess on tactile perception—samples with higher substitution had less refined matrices, which may have created a slightly sandy mouthfeel perceived by the panel as increased hardness.

Other mouthfeel parameters, including melting behavior and stickiness, showed no major differences. Melting scores of 5–6 across all samples indicated acceptable melt-in-mouth performance, while stickiness ratings between 3 and 5.5 confirmed that no sample exhibited excessive residual adhesion.

The second sensory evaluation compared chocolate samples with a fixed CB substitution level of 40%, but varying WPI:IN ratios: 30:70, 50:50, and 70:30. This setup was designed to maximize the concentrations of fat replacers and enhance the potential for detecting sensory differences. Results from this test are presented in Figure 10.

Interestingly, no statistically significant differences were found across the three samples for most attributes—including appearance (color), mouthfeel (hardness, stickiness, melting), milk aroma, bitterness, sweetness, and overall acceptability. These parameters are grouped in Figure 10, as individual breakdowns were unnecessary due to the uniformity of results.

The only sensory attribute showing significant variation was graininess. As shown in Figure 10, the samples with higher WPI proportions (i.e., 50:50 and 70:30 blends) were perceived as more granular compared to the 30:70 blend. This was expected, as higher protein levels can lead to a slightly sandy texture due to incomplete dispersion.

Although the 70% WPI sample was expected to present a more pronounced milk aroma due to its higher protein content, no statistically significant difference was observed. This could be due to the dominant cocoa and vanilla aromas masking the subtler WPI notes, as they likely remained below the sensory detection threshold in the context of the complex chocolate matrix.

3.2. Emulsifier Effect on Chocolate Rheology

The effect of different emulsifiers—lecithin, AMP, and PGPR—on the rheological behavior of the reference chocolate formulation was systematically investigated. For each emulsifier, plots of the square root of shear stress versus the square root of shear rate were used to determine Casson plastic viscosity and yield stress, as suggested by established rheological models [39,40].

Yield stress is a key indicator of chocolate’s machinability and its ability to maintain structural integrity during processes such as molding and enrobing. Conventional emulsifiers like lecithin have been widely used in chocolate production, primarily to reduce viscosity. However, lecithin’s influence on yield stress is limited, often requiring the addition of PGPR to achieve desired flow properties [39,41]. In contrast, AMP has been shown to effectively reduce both plastic viscosity and yield stress, offering a broader functional range even when used alone [40].

Figure 11 shows that lecithin reduced viscosity up to approximately 0.6% concentration, beyond which a reversal occurred, causing increased viscosity likely due to micelle formation or phase destabilization [39,42]. Notably, yield stress did not exhibit a clear trend with lecithin concentration, aligning with earlier studies that highlight lecithin’s primary role in viscosity reduction rather than yield stress modification [22].

Figure 12 illustrates that AMP consistently reduced both viscosity and yield stress, with the effect stabilizing around 0.5% concentration. This behavior is consistent with the surface saturation concept—additional emulsifier beyond this point provides limited improvements [40,43]. In contrast to lecithin, AMP provided a smoother, more predictable reduction in viscosity across tested concentrations, confirming its superior performance [41].

Figure 13 demonstrates the significant impact of PGPR on yield stress, with reductions observed across all tested concentrations while maintaining relatively stable viscosity. This outcome aligns with PGPR’s known mechanism of disrupting particle aggregation through steric hindrance, resulting in smoother flow properties [44]. These findings reinforce its value as a yield stress modifier, particularly in formulations where lecithin or AMP alone may not suffice [43].

3.3. Optimization of Rheology in High Cocoa Butter Replacement Samples

3.3.1. Rheological Optimization of the 40% CB-Substituted Chocolate Sample

Following the individual evaluation of emulsifiers, focused optimization was conducted on the most challenging formulation: the sample with 40% CB replacement using a 50:50 WPI:IN ratio. Based on prior findings where 0.5% AMP was shown to minimize viscosity in the reference system, this concentration was selected as the starting point for optimization. AMP concentrations up to 0.7% were also tested, as this range represents the typical upper threshold for industrial chocolate applications [41].

As shown in Figure 14, AMP effectively decreased viscosity up to 0.5%, beyond which further increases had no substantial effect. This plateau behavior reflects the saturation of particle surfaces, where additional emulsifier no longer influences flow properties [40]. These findings are consistent with those of Simoes et al. (2021), who demonstrated AMP’s ability to slightly alter the crystallization behavior of CB, enhancing its suitability for reduced-fat chocolate systems [45].

Subsequently, PGPR was evaluated in combination with AMP to further optimize rheological performance. PGPR was introduced at concentrations beginning at 0.05% in a matrix containing 0.6% AMP. As illustrated in Figure 15, increasing PGPR concentrations resulted in a substantial reduction in yield stress while having minimal impact on plastic viscosity. This observation aligns with the consensus in the literature that PGPR effectively reduces interparticle friction and yield stress without significantly affecting plastic viscosity [40,44].

The combination of 0.5% AMP and 0.15% PGPR was identified from experimental data, yielding Casson viscosity and yield stress values of 3.42 Pa·s and 7.91 Pa, respectively. These closely match those of conventional full-fat milk chocolate, despite a 40% reduction in CB content. Previous studies have achieved similar rheological behavior only with higher fat content or greater emulsifier loads. For example, N. Hussain et al. (2024) reported comparable properties in a chocolate glaze with 35% CB, 0.49% lecithin, and 0.21% PGPR, highlighting the efficiency of the present formulation [44].

This dual-emulsifier approach has been validated in other chocolate applications as well. Ashkezary and Yeganehzad found that lecithin–PGPR combinations were effective in reducing viscosity, though optimal results required specific refining conditions [46]. Likewise, Ibadullah et al. (2024) demonstrated enhanced flow and stability in dark chocolate through lecithin–glycerol monostearate blends, emphasizing the importance of emulsifier synergy [47].

Notably, the current study achieved desirable rheological behavior in a high-protein, high-fiber, reduced-fat matrix using lower emulsifier doses than those commonly reported. This finding aligns with, Toczek et al. (2022) who highlighted the pivotal role of emulsifier selection in WPI:IN systems for modulating flow, spreadability, and structural uniformity [48].

Overall, these results confirm that the 0.5% AMP and 0.15% PGPR combination is an effective, scalable, and industrially viable strategy for restoring optimal rheological properties in high CB-substituted milk chocolate. This dual-emulsifier system is consistent with previously reported approaches in terms of performance and dosage efficiency, and it integrates seamlessly with the functional and nutritional improvements enabled by WPI and IN [39,40,41].

3.3.2. Sensory Profile of Optimized Formulations

Sensory analysis was performed on chocolate samples with a 40% CB substitution, formulated either with AMP and PGPR or with lecithin as the primary emulsifier. This substitution level was chosen because it represents the highest level of fat reduction tested, offering a clear opportunity to assess how functional modifications can restore or enhance sensory quality. These evaluations provided valuable insights into the ability of different emulsifiers to preserve or improve consumer-relevant attributes in reduced-fat chocolate.

As shown in Figure 16, no significant differences were detected between AMP–PGPR, and lecithin-containing samples for flavor attributes such as sweetness and bitterness, which is consistent with prior research, as emulsifiers do not directly affect chocolate flavor perception. Both formulations contained identical levels of cocoa mass and sugar—the primary contributors to bitterness and sweetness—further supporting that these flavor attributes remain unchanged when emulsifier type varies [41,49].

However, color perception differed notably. The AMP–PGPR-containing sample was rated lower in darkness (≈5), suggesting a more typical milk chocolate appearance, while the lecithin sample appeared darker (≈7.5). This could be attributed to improved flow behavior and finer particle distribution achieved through AMP–PGPR, which enhances optical properties via uniform fat crystallization and reduced particle size during refining and conching [39,50].

Texture-related parameters showed the most significant differences. The AMP–PGPR-based sample was rated as softer and more rapidly melting, while the lecithin sample was perceived as firmer. This may reflect differences in matrix structuring during cooling; AMP–PGPR’s superior flow behavior can lead to less rigid fat networks, resulting in faster oral melt and softer texture [11,45]. Stickiness was also higher in the AMP–PGPR -based chocolate, likely due to its finer granulometry and enhanced melt profile, both of which contribute to greater adherence to the palate—a desirable trait in milk chocolate [39,51]. This observation is in line with prior sensory work in cocoa hazelnut and dark chocolate systems using AMP, which reported enhanced spreadability and mouthfeel [49].

Granulometry perception, assessed through mouthfeel and oral disintegration, favored the AMP–PGPR sample. It was described as smoother and more cohesive, consistent with the particle dispersion benefits observed in rheological analyses. This is supported by earlier studies that identified improved dispersion and texture uniformity in emulsifier systems containing AMP, especially in fat-reduced chocolates [41,52].

Interestingly, milk aroma scores were higher for the AMP–PGPR -based formulation, despite identical levels of WPI and milk solids. This could be due to more efficient volatile release from a homogenous fat matrix, a mechanism described in research linking microstructure to flavor compound diffusion and perception [53].

Overall acceptability was significantly higher for the AMP–PGPR -containing sample. While the lecithin-based chocolate had a slightly slower melt—an attribute sometimes associated with premium perception—it could not compensate for its reduced cohesion and less favorable mouthfeel [11,54]. These findings highlight AMP–PGPR’s potential in maintaining desirable sensory characteristics even at high CB replacement levels.

4. Conclusions

This study successfully developed and evaluated a dual-formulation strategy to reduce fat content in milk chocolate without compromising critical physicochemical, rheological, or sensory properties. The approach combined partial CB substitution (20%, 30%, and 40%) with structured WPI:IN blends (70:30, 50:50, 30:70 w/w) and targeted emulsifier optimization to overcome flow resistance and structural challenges.

Among the tested combinations, the 50:50 WPI:IN ratio consistently delivered balanced rheological behavior and good sensory attributes across all substitution levels (20–40%), with both viscosity and yield stress remaining within acceptable ranges. The most inulin-rich formulations achieved up to 9% caloric reduction (~50 kcal/100 g), highlighting the nutritional potential of this method.

To address the increased viscosity and structural resistance observed at higher substitution levels, the study evaluated three emulsifiers in a full-fat reference chocolate. AMP and PGPR outperformed lecithin in rheological control. AMP effectively reduced both plastic viscosity and yield stress up to 0.5%, while PGPR consistently lowered yield stress across all tested levels (0.05–0.3%) without impacting viscosity. In contrast, lecithin showed limited effectiveness beyond 0.6%, with diminishing returns and signs of phase destabilization.

Based on these results, a dual-emulsifier system (0.5% AMP + 0.15% PGPR) was applied to the most challenging formulation—40% CB substitution with a 50:50 WPI:IN blend. This combination restored rheological values (3.42 Pa·s viscosity, 7.91 Pa yield stress) to levels comparable with full-fat chocolate. Sensory evaluation confirmed the optimized formulation maintained excellent melt-in-mouth feel, smoothness, and overall acceptability, with panelists preferring it over lecithin-only samples.

The Casson model was applied to interpret flow behavior; while effective for conventional formulations, it showed limitations for structured systems. Future work may benefit from incorporating alternative models like Herschel–Bulkley for improved rheological accuracy.

In conclusion, this study presents a viable strategy for formulating reduced-fat milk chocolate that meets current nutritional trends and consumer expectations. The combined use of WPI and inulin as fat replacers, together with optimized emulsifier technology, allows for significant reductions in fat and energy content without compromising key quality attributes. This approach holds strong potential for the confectionery industry, particularly in the development of healthier, functionally enhanced chocolate products ready for commercial-scale production.