Preparation and controlled release properties of heterocyclic flavor CS/Zein microcapsules

2.6.1. Quantification of surface-adsorbed flavor compounds on microcapsules

10mg of microcapsules were extracted with 10 mL of anhydrous ethanol for 5 min, then centrifuged at 4000 rpm for 15 min. The supernatant was collected, and the volume was recorded. After dilution, absorbance was measured at the maximum absorption wavelength, and the surface flavor content was calculated using the regression equation [20].

2.6.2. Determination of total flavor content in microcapsules

For this, 10 mg microcapsules were weighed and extracted with 10 mL of anhydrous ethanol under ultrasonic conditions at 60°C for 20 min, followed by centrifugation at 4000 rpm for 15 min. The supernatant was then allowed to stand, and the volume of the clear upper layer was recorded. After dilution, the absorbance was measured at the maximum absorption wavelength, and the total flavor content within the microcapsules was calculated using a regression equation [20].

The encapsulation efficiency was calculated according to Eq. (1):

3.2.1. Single-factor optimization analysis

As shown in Figure S2(a), the encapsulation efficiency initially increased and then decreased with the wall-to-core ratio, reaching a maximum encapsulation efficiency of 60.33% at a ratio of 5:1. This trend indicates that as wall material is added, encapsulation efficiency increases until the core material is fully encapsulated. Beyond this point, further addition of wall material results in a decline in efficiency, likely because the core is already completely encapsulated, and excess wall material does not contribute to further encapsulation. Therefore, the optimal wall-to-core ratio is 5:1.

Figure S2(b) demonstrates that as the zein-to-CS mass ratio increases, encapsulation efficiency initially rises and then declines, with a maximum of 60.33% at a 3:1 ratio. This may be attributed to the high viscosity of CS, which, when used in excess, causes the microcapsule walls to become prone to cracking during freeze-drying, exposing the core material and reducing encapsulation efficiency. Increasing zein while decreasing CS results in fragile walls that cannot fully enclose the core, thereby decreasing overall efficiency. Consequently, a 3:1 mass ratio of zein to CS is optimal for microcapsule formation.

From Figure S2(c), it can be observed that with increasing reaction temperature, the encapsulation efficiency initially increases and then decreases, reaching a maximum of 82.41% at 60°C. This phenomenon is likely due to the progressive crosslinking between the core material and the wall material as temperature rises, resulting in increased encapsulation efficiency, with complete crosslinking at 60°C yielding the highest efficiency. Beyond this temperature, excessive heat may damage the wall structure, leading to a decline in encapsulation efficiency. Therefore, 60°C is identified as the optimal reaction temperature for the wall-to-core ratio. The smoking temperature of heated cigarettes is between 250°C and 350°C [25], while lower combustion temperatures will affect the release of tobacco aromas, and adding flavors will cause fragrance loss if left for a long time. Under this experimental condition, the encapsulation rate reached 82.41%, which effectively seals DEMP and achieves slow release. This offers a new approach to flavoring heated cigarettes.

3.2.2. Response surface optimization

Based on the results of single-factor experiments, three variables, wall-to-core ratio (A), wall material ratio (B), and temperature (C), which significantly influence encapsulation efficiency, were selected as independent variables. The microcapsule encapsulation rate (Y) served as the response variable. A response surface analysis was conducted using a three-factor, three-level design comprising 17 experimental points to optimize the preparation process. The results have been summarized in Table 2. It is evident that under conditions of a wall-to-core ratio of 5:1, a wall material ratio (corn gluten protein to CS) of 3:1, and a temperature of 50°C, the maximum microcapsule encapsulation efficiency reached 82.5%. Conversely, the lowest efficiency of 57.1% was observed at a wall-to-core ratio of 3:1, a wall material ratio of 3:1, and a temperature of 40°C. Regression analysis of the response surface data using Design-Expert software yielded the following quadratic regression model: Y = 81.3 + 0.675A + 0.5375B + 5.19C - 0.825AB - 1.32AC + 1.4BC - 9.68A² - 5.35B² - 7.05C².

Table 3 presents the ANOVA results for the model. As indicated in the table, the F-value of the established model was 42.24, with a P-value<0.01, demonstrating a high level of significance. The quadratic model effectively fits the experimental data, with minimal residual error, confirming its suitability for analyzing the effects of wall-to-core ratio (A), wall material ratio (B), and temperature (C) on encapsulation efficiency. The influence of each factor varies, with quadratic terms A², B², and C², and the linear term C being highly significant (P < 0.01). The relative impact on microcapsule encapsulation efficiency follows the order: temperature (C) > wall-to-core ratio (A) > wall material ratio (B).

Analysis of the interaction effects, based on the regression equation, reveals significant interactions among the factors, as illustrated by the response surface contour plots in Figure 1, which display elliptical shapes indicative of notable synergistic effects between wall-to-core ratio, wall material ratio, and temperature.

Close

Figure 1.

Response surface diagram of pairwise interaction of various factors, (a, b) material ratio and wall to core and time; (c, d) material ratio and wall to core and temperature; (e, f) time and temperature.

Export to PPT

The first derivatives of the model equation were calculated to identify the optimal conditions, resulting in a predicted maximum encapsulation efficiency of 82.31% at a wall-to-core ratio of 5.01:1, a wall material ratio of 3.10:1, and a temperature of 53.77°C. Considering practical operational constraints, the final preparation parameters were set to a wall-to-core ratio of 5:1, a wall material ratio of 3:1, and a temperature of 54°C, with three validation experiments, yielding an average actual encapsulation efficiency of 82.27%. This confirms that the response surface methodology effectively optimized the preparation process for corn gluten protein-CS microcapsules.

3.3.1. SEM analysis

To evaluate and examine the morphological distinctions among single-wall material microcapsules, empty microcapsules, and composite wall material microcapsules, SEM analysis was performed. As shown in Figure 2, microcapsules encapsulated with DEMP using only zein exhibited an irregular shape characterized by adhesion in Figure 2(a). In contrast, microcapsules encapsulated with DEMP with only CS displayed a porous network structure (Figure 2b). The morphology of the empty microcapsules, which did not contain DEMP, differed significantly from that of the single-wall material microcapsules, appearing as spherical aggregates with uneven sizes (Figure 2c). Microcapsules prepared with a composite wall material of zein and CS exhibited a more uniform spherical structure with a clustered distribution (Figure 2d). The morphology differed between composite wall material microcapsules and those made with single-wall materials or empty microcapsules. This difference indicated the successful preparation of zein-CS composite wall material microcapsules.

Close

Figure 2.

SEM images of (a) Zein + DEMP, (b) CS + DEMP, (c) empty microcapsule, (d) microcapsule.

Export to PPT

3.3.2. FTIR spectroscopy

The FTIR spectra of the samples (zein, CS, DEMP, physical mixtures, and microcapsules) have been illustrated in Figure 3. As shown in Figure 3(a), zein exhibited a prominent absorption peak at 3356 cm^-1, attributed to the stretching vibrations of N-H. The characteristic peak at 2958 cm^-1 was due to the stretching vibrations of -CH. The peaks at 1657, 1539, and 1449 cm^-1 corresponded to the stretching and bending vibrations of O-C-O, -NH, and CH₂, respectively [26]. Concurrently, CS displayed a characteristic peak for -OH stretching vibrations at 3430 cm^-1. The peak at 2876 cm^-1 was associated with -CH stretching vibrations. The peaks at 1601, 1426, 1381, and 1090 cm^-1 were attributed to -NH bending, the amide II band of -NH₂, symmetric deformation of -CH₂ and -CH₃, asymmetric stretching of -C-O-C, and -CN stretching along with O-C-O stretching [27]. In DEMP, the absorption peak for C=C stretching vibrations was observed at 1457 cm^-1, while the C-H stretching vibration peak appeared at 2981 cm^-1, and the aromatic ring skeletal vibration peaks were noted between 1550-900 cm^-1 [28]. The infrared spectrum of the physical mixture revealed characteristic peaks of zein, CS, and DEMP, with the -OH stretching vibration peak of CS appearing at 3429 cm^-1 and the -CH stretching vibration peaks of zein and DEMP at 2973 cm^-1. In contrast, the infrared spectrum of the microcapsules differed from those of zein, CS, DEMP, and the mixtures, showing -OH and N-H stretching vibration characteristic peaks at 3347 cm^-1 and an O-C-O stretching vibration peak at 1084 cm^-1. The shifts in the -OH, N-H, and O-C-O characteristic peaks suggested that hydrogen bonding might play a role in the formation of microcapsules, resulting in a blue shift of the infrared characteristic peaks [29]. Additionally, the significant reduction in the characteristic absorption peaks of DEMP indicated that it was encapsulated within the wall material, rendering the characteristic peak stretching vibrations less pronounced [30].

Close

Figure 3.

(a) Infrared spectra of CS, Zein, DEMP, mix, and microcapsule; (b) TG of DEMP, CS + DEMP, Zein + DEMP, Zein + CS, and microcapsule.

Export to PPT

3.4.1. Thermogravimetric analysis (TGA)

The thermogravimetric results of the wall material, core material, single-wall microcapsules, unloaded microcapsules, and composite wall microcapsules are presented in Figure 3. As shown in Figure 3(b), DEMP begins to lose weight at 62°C and is completely degraded by 149°C, indicating poor thermal stability and underscoring the necessity of embedding DEMP to delay thermal decomposition. Zein and CS both exhibit biphasic weight loss: the first stage occurs between 30°C and 100°C, primarily due to the evaporation of water molecules associated with zein and CS [31]. The second stage, from 275°C to 700°C, was mainly attributed to the decomposition of zein and CS themselves. When zein was incorporated with DEMP, the overall weight loss trend resembled that of zein. However, the weight loss rate during the first stage was 11.3%, which was higher than the 9.6% weight loss rate of zein alone, likely due to the weight loss of DEMP within the microcapsules. The weight loss curve of CS microcapsules showed some differences compared to CS, with a first-stage weight loss rate of 30.6%. This might be due to a significant amount of DEMP adsorbed on the surface of CS, leading to weight loss from the thermal degradation of DEMP. The weight loss of unloaded microcapsules shifted to the right compared to zein and CS. The thermal weight loss of composite wall microcapsules was divided into two stages: the first stage, from 30-269°C, primarily involved the evaporation of moisture and DEMP from the microcapsule surface, with a weight loss rate of 18.6%. The second stage, from 270°C to 647°C, was mainly due to the breaking of chemical bonds in the wall material of the microcapsules, resulting in the heating and evaporation of DEMP [32]. Notably, at 149°C, the weight loss was only 8.7%, indicating effective protection of DEMP within the wall matrix. The weight loss rate of the composite wall microcapsules fell between those of single-wall microcapsules, demonstrating favorable thermal stability. The TGA results confirm that zein-CS composite wall microcapsules possess excellent thermal stability, effectively safeguarding DEMP and slowing its release at elevated temperatures [33].

3.4.2. Thermal release kinetics

Figures 4 depict the TG and DTG curves of zein-CS microcapsules at different heating rates. As shown in Figure 4, prior to 280°C, the mass of the microcapsules decreases gradually, likely due to the volatilization of surface DEMP and water molecules, as well as internal structural rearrangements [34]. With increasing temperature, a pronounced weight loss phase occurs during thermal release. Once complete decomposition is achieved, the microcapsules’ mass stabilizes. Furthermore, higher heating rates cause the TG curves to shift toward higher temperatures, which may result from increased internal and external temperature gradients within the sample, reducing the time required to reach reaction temperatures [34]. Concurrently, the maximum weight loss rate observed in the DTG curves increases significantly with higher heating rates, likely due to accelerated reaction kinetics between the microcapsules and oxygen, leading to greater maximum weight loss rates [34].

Close

Figure 4.

TG and DTG curves of the microcapsule at different heating rates.

Export to PPT

Table 4 presents the characteristic parameters associated with the thermal release process of microcapsules. Here, T_i and T_f represented the initial and final release temperatures of the sample as determined by TG analysis, while DTG_max refers to the reaction temperature corresponding to the maximum weight loss rate. The comprehensive heat release index (CRI), calculated according to Eq. (S1), served as an indicator reflecting the thermal release characteristics of the microcapsules. It was evident from the table that T_i, T_max, and T_f all increased with the rising heating rate. However, the magnitude of increase for these three parameters was less than that of the heating rate itself, indicating that the exacerbation of thermal lag did not exhibit a linear relationship with the heating rate. Similarly, both DTG_max and CRI increased with the heating rate, with DTG_max rising significantly from 2.55%/min to 41.66%/min, and CRI increasing from 0.00805% to 0.16028%. This suggested that a faster heating rate enhanced the rate of thermal release of the microcapsules, facilitating the thermal release reaction. Moreover, there existed a good linear relationship between DTG_max, CRI, and the heating rate, with the specific linear equations being y = 0.0397x + 0.1198 (R²=0.9274) and y = 1.8361x - 0.4378 (R²=0.9813), respectively.

Analysis of TG/DTG data indicates that the heating rate significantly influences the thermal release behavior of the samples. To develop a kinetic reaction model describing the thermal release process of microcapsules, the Coats-Redfern method was employed, utilizing Eqs (S1 and S2) to determine the most appropriate reaction model and kinetic parameters based on the coefficient of determination. The enthalpy change (ΔH), Gibbs free energy (ΔG), and entropy change (ΔS) of the reaction system were calculated, with the fitting results summarized in Table 5. As shown in Table 5 and Table S1, the secondary reaction model F2 effectively describes the thermal release process at various heating rates, with apparent activation energies (E) ranging from 27.66 to 29.92 kJ/mol. Figure 5(a) presents the fitting curves of the F2 model under different heating rates, demonstrating a good linear relationship within the 20-90% release conversion range, with R² values exceeding 0.97.

Close

Figure 5.

(a) Coats-Redfern fitting curves and (b) KAS fitting curves.

Export to PPT

Activation energies were further calculated using the Kissinger-Akahira-Sunose (KAS) method (S3) across heating rates corresponding to release rates of 0.2-0.9. Figure 5(b) shows the linear fits of the KAS method at different release rates, with R² values between 0.9537 and 0.9912, and the fitted equations and activation energies listed in Table 6. The overall activation energies ranged from 29.13 to 58.19 kJ/mol. Activation energy represents the minimum energy required for a chemical reaction to occur [35]. The Coats-Redfern method is a single temperature rate model fitting technique that assumes that the entire reaction process follows a single reaction mechanism. The KAS method solves the activation energy by analyzing the temperature dependence of the same conversion rate at multiple heating rates. This method can independently calculate the activation energy corresponding to each conversion rate without presetting the reaction mechanism model, and is suitable for complex reaction processes [36]. During the heat release of microcapsules, the activation energy of the two methods differed greatly, confirming the multi-stage reaction nature.

3.4.3. Storage stability

The storage stability of microcapsules and essential oils at room temperature over 30 days has been depicted in Figure 6. As shown in Figure 6(a), under identical storage conditions and duration, there are significant differences in the release rates between unencapsulated DEMP and microencapsulated DEMP. With prolonged storage, the release rate of DEMP from microcapsules increases. Notably, after 30 days at room temperature, the release rate of DEMP from microcapsules was 9.23%, compared to 58.73% for unencapsulated DEMP. Kinetic analysis of DEMP release from microcapsules at room temperature (Figure 6b and Table S2) indicates that the release behavior conforms more closely to the Peppas equation. These findings demonstrate that microencapsulation significantly reduces DEMP volatility and achieves a sustained-release effect, consistent with previous studies by Wang et al., which showed that microencapsulation enhances storage stability and reduces release rates of fragrances [37]. The smoking temperature of heated cigarettes is about 350°C, and the lower burning temperature will affect the release of tobacco aroma, and the addition of flavors will cause fragrance loss [38]. The encapsulation rate of the microcapsules reached 82.41% under these experimental conditions, and the DEMP release was less than 9.23% after 30 days of placement. It can better solve the problem of serious fragrance loss after heated cigarettes are stored for too long. With the increase in the number of suction puffs, the puff-by-puff release of monomer fragrance in the aerosol of heated cigarettes shows a trend of increasing first and then decreasing, resulting in inconsistency in taste and affecting the sensory experience, as can be seen from the TG curve that at about 350°C, the weight loss of microcapsules occurs at a high rate, and the addition of microcapsules to heated cigarettes can not only enable the spices to be stored for a long time, but also modify the aroma of the heated cigarettes, so that the suction uniformity is consistent. Microencapsulated flavoring provides a new idea for the flavoring of heated cigarettes.

Close

Figure 6.

(a) Release rates and (b) fitting equations of microcapsules and DEMP at room temperature. *The identical R² value of 0.9967 for both the Peppas and Higuchi kinetic equations, compared to 0.9775 for the first-order model, caused their fitted curves to overlap, thus explaining the absence of the green line in the figure.

Export to PPT

3.5.1. Conformational analysis

Simulations of conformational changes in DEMP, Zein, and CS were conducted. Conformational changes directly reflect interactions between host and guest molecules. Since the structure of zein remains unresolved, and it is known that proteins are composed of amino acids linked via peptide bonds, forming polypeptides and peptide chains that fold and twist into specific 3D structures, the basic structural unit of zein was represented by a dipeptide containing a single peptide bond. During the simulation, a single guest molecule was centered, and conformational snapshots of the DEMP molecule were captured at 0, 5, 10, 15, 20, 25, 30, 35, 40, 45, and 50 nanoseconds within a 3 Å radius. The conformational images of each guest molecule were derived from the total trajectory. As shown in Figure 7, due to hydrogen bonding within the molecules, the amino sugar units in CS undergo rotation and twisting, affecting their mobility and resulting in irregular folding [39]. The binding sites and conformations between host and guest molecules are continuously changing, likely due to intermolecular interactions [40].

Close

Figure 7.

Conformation diagram of DEMP and Zein+CS in 0-50ns.

Export to PPT

3.5.2. RMSD and Rg analysis

The RMSD values of DEMP (main body), Zein+CS (guest molecules), and the entire system within 0-50 ns have been shown in Figure 8(a). The data indicate that during molecular dynamics (MD) simulations, the RMSD of the guest molecules exhibits minimal fluctuations, remaining within a narrow range of 2-2.5 nm, suggesting that the guest molecules maintain stable conformations when isolated [39]. Post-equilibration, the RMSD of DEMP fluctuates between 3.2 and 3.6 nm, while the RMSD of the DEMP/Zein+CS complex varies between 2.4 and 2.9 nm. The RMSD values of all microcapsule systems exceed those of the guest molecules, likely due to intermolecular interactions between the flexible aromatic guest molecules and the main body, leading to structural modifications. As shown in Figure 8(b), during the simulation, the average radius of gyration (Rg) of the guest molecules is approximately 2.68 nm, whereas the Rg of the microcapsule system averages around 2.64 nm. The Rg values of the microcapsules are consistently lower than those of the free guest molecules, with limited variation, indicating that the structure of the microcapsule molecules tends to contract and stabilize upon binding of DEMP to the main body [40].

Close

Figure 8.

(a) RMDS, (b) Rg, RDF (c), and (d) hydrogen bonding results of host and guest molecules.

Export to PPT

3.5.3. RDF and hydrogen bond analysis

To investigate the strength of hydrogen bonds formed between the main body and guest molecules, the nitrogen atom on the pyridine ring of DEMP was selected as the hydrogen bond acceptor, and the RDF values were calculated with respect to amino groups (CNH₂), carboxyl groups (COOH) in Zein, and amino groups (NH₂) in CS. As shown in Figure 8(c), the hydrogen bonding strength of carboxyl groups exceeds that of amino groups, likely due to the higher reactivity of the hydroxyl group (-OH) in the carboxyl, facilitating hydrogen bond formation with the main body. Hydrogen bonds are non-covalent interactions formed when small, highly electronegative atoms approach each other, with hydrogen acting as a mediating atom. The polyhydroxy, amino, and carboxyl groups in corn gluten protein and CS can form hydrogen bonds with heterocyclic aromatic compounds [41]. Figure 8(d) reveals that, because DEMP molecules contain only hydrogen bond acceptors and lack donors, the number of hydrogen bonds between DEMP and the guest molecules is approximately 6.5. Conversely, the amino and carboxyl groups in Zein and CS serve as both hydrogen bond donors and acceptors, resulting in a significantly higher number of hydrogen bonds among the guest molecules, reaching approximately 606.8.

3.5.4. MMPBSA analysis

Previous studies have demonstrated that the formation of microcapsule systems results from the synergistic effect of multiple non-covalent interactions, with thermodynamic parameters providing critical insights into the molecular interaction mechanisms [42]. Quantitative analysis via the MMPBSA method indicates (Table S3) that the positive value of the solvation free energy (ΔG_sol) reflects the non-polar character of DEMP molecules during solvation, confirming the presence of hydrophobic interactions [39,42]. Notably, the van der Waals contribution (ΔG_vdW) to the binding free energy is substantially higher than the Coulombic contribution (ΔG_Coul), consistent with the prevailing understanding of microcapsule formation mechanisms [43,44]. The negative value of the binding free energy (ΔG_Bind) suggests that the encapsulation process is thermodynamically spontaneous [40]. The primary driving forces for microcapsule formation are ranked as: van der Waals interactions > hydrophobic effects > Coulombic forces. Although hydrogen bonds and Coulomb interactions are not dominant, molecular conformation analyses confirm their significant role in modulating the spatial structure of the complex [45].

3.5.5. IGMH analysis

To visually characterize the intermolecular interactions, the IGMH visualization technique was employed [46]. This method maps the interaction regions and their strength distributions through 3D isosurfaces: blue indicates hydrogen bonding attractions, red reflects steric hindrance effects, and green corresponds to van der Waals interactions of moderate strength [47]. As shown in Figure 9, the interaction regions within the DEMP/Zein+CS microcapsule system are predominantly green, indicating that van der Waals forces primarily drive the molecular binding, consistent with the MMPBSA results. Additionally, a small number of blue regions are observed, representing minor hydrogen bonding interactions during complex formation. These findings further confirm that van der Waals interactions are the principal force governing the system.

Close

Figure 9.

IGMH results of the subject and the object.

Export to PPT