Development of a Water-in-Oil Microemulsion Template for Chitosan Nanogel Fabrication via Genipin Crosslinking

Abstract

This study presents a promising strategy for the fabrication of a novel chitosan-based nanogel-in-oil system by integrating the development of a water-in-oil (W/O) microemulsion containing chitosan as a template, followed by crosslinking with genipin, a natural crosslinking agent, via emulsion crosslinking. To develop the W/O microemulsion template, nanometer-sized internal aqueous droplets were successfully formed in cottonseed oil, a vegetable oil, using a blend of nonionic surfactants, polysorbate 80 and sorbitan monooleate. A pseudoternary phase diagram was constructed to investigate the phase behavior of systems composed of chitosan solution, mixed surfactant, and cottonseed oil. Compositions falling within the monophasic region were selected for further formulation optimization. The microemulsions were characterized for droplet size, size distribution, electrical conductivity, and viscosity. The optimal microemulsion exhibited W/O characteristics with the lowest viscosity. Dynamic light scattering (DLS) analysis confirmed the presence of uniformly distributed nanometer-sized droplets, as evidenced by a Z-average diameter of 92.9 ± 2.3 nm and a PDI of 0.100 ± 0.072. The microemulsion system demonstrated physical stability, as confirmed by centrifugal testing. Crosslinking of chitosan with genipin was monitored by fluorescence intensity measurements of the crosslinking products. Fourier transform infrared spectroscopy further confirmed the formation of genipin-crosslinked chitosan structure. DLS and transmission electron microscopy revealed that the nanogels possessed nanoscale dimensions and discrete spherical morphologies. Overall, this approach demonstrates a viable route for producing a nanogel-in-oil system by combining microemulsion templating with emulsion crosslinking.

1. Introduction

Gel-in-oil dispersions consist of gel particles distributed within an oil phase. In water-in-oil systems with internal aqueous phase gelation, the structural components or structurants responsible for forming the internal gel particles are generally biopolymers, including proteins, polysaccharides, and protein–polysaccharide complexes [1]. Although polymeric gel particles can range in size from millimeters to nanometers, nanogels with monodisperse distributions at the nanoscale are particularly desirable due to their high surface-to-volume ratio, enhanced dispersibility, and versatility for various functional applications [2,3]. Nanogels are nanoscale gel particles in which polymer chains are assembled into networks via physical or chemical crosslinking [4]. Their soft, deformable, and permeable structure enables diverse applications across several domains, such as surfactants, emulsifiers, biocatalysts, sensors, food applications, and drug delivery systems [4,5]. The development of nanogel-in-oil systems has attracted considerable attention in recent years due to their potential applications [6,7,8,9].

Chitosan is a copolymeric polysaccharide composed of D-glucosamine and N-acetyl-D-glucosamine units, obtained through the partial deacetylation of chitin, a naturally occurring homopolysaccharide. Typically, chitosan is characterized by a degree of deacetylation ranging from 70% to 95% [10]. Among biopolymers used in the development of biomaterials, chitosan has received considerable research attention as a gel structurant owing to its biocompatibility, biodegradability, and low toxicity, as well as its abundance of amino and hydroxyl functional groups that enable structural modification and crosslinking [10]. Both physical and chemical interactions can drive the three-dimensional structural evolution of chitosan precursor chains. Three-dimensional networks of chitosan chains have been widely fabricated over the past decade through chemical crosslinking using synthetic crosslinking agents such as glutaraldehyde, glyoxal, and epichlorohydrin [4,10]. However, growing concerns regarding safety and sustainability have been raised about the suitability of these synthetic agents. Consequently, increasing attention has been directed toward the development of chitosan-based gels using natural crosslinking agents [4,10,11].

Genipin is a plant-derived iridoid compound extracted from the fruit of Gardenia jasminoides [12]. According to pharmacological studies and existing reviews, genipin exhibits multiple biological activities, including anti-inflammatory, antioxidant, and antibacterial effects [12,13]. In addition, it is widely recognized as a natural and bio-safe crosslinking agent [11]. Compared with synthetic crosslinkers, genipin is considered to possess favorable biocompatibility and eco-friendly properties [14] and has therefore attracted increasing attention in food and medical research. Genipin has been reported to effectively crosslink chitosan chains, forming a stable chitosan network that serves as the structural backbone of chitosan-based gel matrices and films [11,15]. In addition to bulk gel systems, genipin also shows potential as a crosslinking agent for the fabrication and development of chitosan-based particles [16]. Among available fabrication techniques, emulsion crosslinking is particularly appealing for producing chitosan particles. This approach involves generating dispersed-phase liquid droplets as reaction templates, followed by crosslinking to form particles. However, emulsion-templated crosslinking techniques typically yield chitosan particles with micron-scale sizes and broad size distributions [17]. To produce spherical nanomaterials, strategies that enable nanoscale control are required. In this context, water-in-oil (W/O) microemulsions have emerged as a promising platform for tailoring nanomaterials with well-defined size and morphology [18,19]. The development of aqueous droplet templates dispersed in vegetable oil media represents a promising approach, as vegetable oils are natural, renewable, and widely recognized for their biocompatibility [20]. Many vegetable oils have a long history of pharmaceutical use and are considered safe for various routes of administration, including injectable and topical applications [20,21]. However, the use of vegetable oils in W/O microemulsion systems remains challenging due to their limited miscibility with water [22]. Moreover, studies on the development of W/O microemulsion systems using vegetable oils as the continuous phase for the fabrication of chitosan nanogels are limited. A previous study investigated the fabrication of chitosan–genipin nanomaterials using ultrasound-assisted processing in a water-in-palm oil emulsion system coupled with emulsion crosslinking, yielding nanoparticles with a relatively large mean size and broad size distribution [17]. Orifice-induced hydrodynamic cavitation, using a specifically designed orifice plate, has also been employed to facilitate the formation of chitosan–genipin nanoparticles with reduced particle size and narrower size distribution [17]. From the perspective of nanomaterial development, microemulsion-based technologies offer an alternative approach by providing nanoscale reaction environments that can be achieved through low-energy emulsification, with potential advantages in simplicity and cost efficiency [19,23]. In this context, the development of internal aqueous droplet templates within vegetable oil-based W/O microemulsions represents a promising strategy for the fabrication of nanogel-in-oil systems via emulsion crosslinking.

In this study, we propose a promising strategy for the fabrication of a chitosan nanogel-in-oil system by integrating the development of a W/O microemulsion containing chitosan as a template, followed by crosslinking with genipin. The oil phase selected for microemulsion formation was cottonseed oil, a vegetable-based, pharmaceutically acceptable, water-insoluble oil with low viscosity [21,24]. First, the W/O microemulsion templates were prepared and characterized to determine the optimal formulation conditions prior to crosslinking. Subsequently, chitosan nanogels were fabricated via emulsion crosslinking within the optimized microemulsion system. The physicochemical properties of the resulting chitosan nanogels were also characterized.

2. Materials and Methods

2.1. Materials

Chitosan with a viscosity of 5 mPa·s, measured at 0.5% in 0.5% acetic acid at 20 °C, and a degree of deacetylation of 85.9% was obtained from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan). Genipin (≥98% purity) and cottonseed oil (≥98% purity) were purchased from Chanjao Longevity Co., Ltd. (Bangkok, Thailand). Sorbitan monooleate (Span 80) and polysorbate 80 (Tween 80) were obtained from PC Drug Center Co., Ltd. (Bangkok, Thailand). The hydrophilic–lipophilic balance (HLB) values of Span 80 and Tween 80 are 4.3 and 15, respectively. Phosphotungstic acid, glacial acetic acid, and Sudan III were supplied by Loba Chemie Pvt. Ltd. (Mumbai, India). Methylene blue was purchased from Kemaus (Cherrybrook, Australia).

2.2. Development and Characterization of Microemulsion Template

2.2.1. Screening of Surfactant Composition

The chitosan solution served as the aqueous phase, while the cottonseed oil served as the oil phase. To prepare the aqueous phase, a 1% (w/w) chitosan solution was obtained by dissolving chitosan in 1% (v/v) aqueous acetic acid under continuous stirring for 12 h, as previously described [25]. To identify a suitable surfactant composition, pure nonionic surfactants (Span 80 and Tween 80) and their mixed surfactants were tested by conducting a screening evaluation, as previously described [26,27]. The mixed surfactants were prepared at Span 80:Tween 80 weight ratios of 1:9, 2:8, 3:7, 4:6, 5:5, 6:4, 7:3, 8:2, and 9:1. For each mixed surfactant, the required amounts of Span 80 and Tween 80 were placed in flat-bottom bottles and mixed using a mechanical stirrer. These corresponded to Span 80 weight fractions of 0.1–0.9, respectively. In addition, Span 80:Tween 80 ratios of 0:1 and 1:0 were included to represent pure Tween 80 (Span 80 weight fraction = 0) and pure Span 80 (Span 80 weight fraction = 1), respectively.

For the determination of the optimal surfactant system, 1 g of the oil phase was mixed thoroughly with 1 g of either the mixed surfactant or the pure surfactant. The aqueous phase was then added dropwise under continuous stirring until the mixture turned from clear to turbid. The surfactant composition that enabled the maximum proportion of the loaded aqueous phase in the mixtures while maintaining a clear, monophasic appearance was considered optimal [26,27] and selected for further construction of a pseudoternary phase diagram.

2.2.2. Construction of Pseudoternary Phase Diagram

The pseudoternary phase diagram comprising the oil phase, the optimized surfactant composition, and the aqueous phase was constructed using a titration method, as previously described [26,27]. Mixtures of the cottonseed oil and the optimized surfactant composition were prepared at varying weight ratios: 10:0, 9:1, 8:2, 7:3, 6:4, 5:5, 4:6, 3:7, 2:8, 1:9, and 0:10. Each mixture was titrated with the aqueous phase under constant magnetic stirring until a transition point—defined as the point at which the sample appearance changed from clear to turbid—was observed. All experiments were conducted in triplicate, and the average values were used to construct the phase diagram. The boundary was plotted by connecting the transition points, and the diagram was constructed using ProSim software (version 1.0.3), highlighting the region corresponding to the single-phase clear solution, typically considered the microemulsion domain. Following the identification of the monophasic region, selected mixture compositions were used for further characterization and screening of microemulsion templates.

2.2.3. Determination of Droplet Size Features

The size distribution, Z-average diameter, and polydispersity index (PDI) of the microemulsions were determined using dynamic light scattering (DLS) with a Zetasizer Nano ZS (Malvern Instruments, Malvern, Worcestershire, UK) at a scattering angle of 90°. Samples were loaded into square sizing cuvettes and measured at a controlled temperature of 25 °C. All measurements were performed in triplicate.

2.2.4. Electrical Conductivity Measurement

The electrical conductivity of the microemulsions was determined using a DEC1 conductivity tester (DLAB Scientific, Beijing, China). Prior to measurement, each sample was allowed to equilibrate at 25 °C. All conductivity measurements were carried out in triplicate.

2.2.5. Viscosity Measurement

Viscosity measurements were conducted over a shear rate range of 0.1–1000 s⁻¹ at 25 °C using a HAAKE MARS 40 rheometer (Thermo Fisher Scientific, Dreieich, Germany) equipped with a cone-plate system (C60 1°/Ti). All viscosity measurements were conducted in triplicate.

2.2.6. Tyndall Effect and Microemulsion Type Determination

The Tyndall effect of the microemulsion sample was assessed by directing a laser pointer through the dispersion to observe the occurrence of a visible light path, indicating light scattering by the colloidal dispersion [28]. The microemulsion type was determined using a dye solubility test. In this procedure, 10 mg of either a water-soluble dye (methylene blue) or an oil-soluble dye (Sudan III) was separately added to the samples. The diffusion behavior of each dye was visually observed. A faster diffusion of the oil-soluble red dye than that of the water-soluble blue dye indicated that the continuous phase was oil, revealing the system to be a W/O microemulsion [23]. Conversely, a faster diffusion of the water-soluble dye would suggest an oil-in-water (O/W) microemulsion [23].

2.2.7. Transmittance Measurement

The transmittance of the microemulsion sample was measured using a UV–visible spectrophotometer (UV-1800, Shimadzu, Kyoto, Japan). Transparency was assessed by recording the percentage transmittance at 650 nm [29]. To evaluate emulsion stability, the sample was centrifuged at 5000 rpm for 10 min at 25 °C [30]. After centrifugation, the sample was visually inspected for any signs of phase separation [30]. The transmittance of the post-centrifugation sample was subsequently measured. This combination of centrifugation and transmittance analysis enabled a comprehensive evaluation of emulsion stability [31].

2.3. Preparation and Characterization of Nanogels

2.3.1. Fabrication of Genipin-Crosslinked Chitosan Nanogel-in-Oil

The 1:1 and 1:2 molar ratios of genipin to glucosamine units in chitosan were calculated based on the conceptual framework of a previously reported method [32], adapted for application to genipin. Briefly, the average molecular weight of the chitosan monomer was determined based on the molecular weights of D-glucosamine and N-acetyl-D-glucosamine, along with the degree of deacetylation reported by the chitosan supplier for the specific batch used. Using the molecular weight of D-glucosamine, the calculated average molecular weight of the chitosan monomer, and the degree of deacetylation, the molar amount of glucosamine units in 100 g of the prepared 1% w/w chitosan solution was estimated. The required genipin concentrations (expressed as % w/w) to achieve the desired molar ratios were then calculated, assuming the final mixture was prepared by combining equal weights of the aqueous 1% w/w chitosan and genipin solutions. The calculated genipin concentrations were 1.16% w/w for the 1:1 ratio and 0.58% w/w for the 1:2 ratio (genipin:glucosamine units).

Aqueous solutions of 1% w/w chitosan and genipin (1.16% w/w or 0.58% w/w) were separately prepared by dissolving each component in 1% v/v aqueous acetic acid. Based on the optimized microemulsion composition obtained from Section 2.2, the W/O microemulsion containing chitosan was prepared by adding the chitosan solution to the cottonseed oil–surfactant mixture, followed by homogenization under mechanical stirring at 200 rpm. A separate W/O microemulsion containing genipin was prepared using the same procedure, substituting the aqueous chitosan solution with the genipin solution. The genipin-containing microemulsion was then added to the chitosan-containing microemulsion under continuous stirring. Stirring was maintained to promote collisions between internal aqueous droplets containing the crosslinking agent and those containing polymer chains, thereby facilitating the formation of crosslinked nanoparticles [33].

2.3.2. Fluorescence Emission Analysis

The time-dependent evolution of the crosslinked gelation resulting from the reaction between chitosan and genipin was monitored at 25 °C by measuring the fluorescence intensity of the samples using a Clariostar Plus spectrophotometer (BMG Labtech, Ortenberg, Germany) operated in fluorescence (FI) spectrum mode. Each sample (200 µL) was transferred into the wells of a black 96-well flat-bottom microplate, which was then placed in the instrument. For an excitation wavelength of 590 nm, fluorescence emission spectra were recorded over a 615–700 nm wavelength range with a 1 nm step width in top-optic mode. For an excitation wavelength of 550 nm, emission spectra were collected over the range of 575–700 nm with the same step width. All measurements were conducted in triplicate.

2.3.3. Particle Size and Morphology Characterization

The size distribution, Z-average diameter, and PDI of the nanogel dispersion were determined by DLS using the Zetasizer Nano ZS instrument at 25 °C. The morphology of the gel nanoparticles was examined using transmission electron microscopy (TEM) with a Hitachi HT7800 instrument (Hitachi, Tokyo, Japan). For TEM observation, 10 µL of the sample was deposited onto a carbon-coated copper grid, and the excess liquid was gently removed using Whatman filter paper. The sample was then stained with 10 µL of a 1% (w/v) phosphotungstic acid solution. Excess liquid was removed with filter paper, and the grid was dried overnight prior to TEM imaging using the Hitachi HT7800 operated at an accelerating voltage of 80 kV.

2.3.4. Infrared Spectroscopy Characterization

Attenuated total reflectance–Fourier transform infrared spectroscopy (ATR-FTIR) was performed to investigate intermolecular interactions in the genipin-crosslinked chitosan network in comparison with the plain solid components. Spectra were acquired using a PerkinElmer Spectrum Two FTIR spectrometer (PerkinElmer, Waltham, MA, USA) equipped with a PerkinElmer UATR Two accessory. ATR-FTIR spectra were collected over the wavenumber range of 4000–600 cm⁻¹, at a resolution of 4 cm⁻¹, with 32 scans. To eliminate spectral interference from residual liquid components, the nanogels were lyophilized prior to analysis. The genipin-crosslinked chitosan nanogels were separated from the liquid medium by centrifugation using a Kubota 6000 centrifuge (Kubota Corporation, Tokyo, Japan) at 10,000 rpm. The resulting sediment was sequentially washed with n-hexane, acetone, and distilled water to remove residual oil and surfactant. The nanogel was then frozen and freeze-dried using a Christ Alpha 2-4 LDplus freeze dryer (Martin Christ, Osterode, Germany).

2.3.5. Batch-to-Batch Variability

To evaluate the reproducibility of the nanogel fabrication process, two independent batches were prepared [34]. The Z-average particle diameter and PDI values of both batches were statistically compared using Student’s t-test. Statistical analysis was performed using IBM SPSS Statistics 25 for Windows.

3. Results and Discussion

3.1. Microemulsion Template Development and Characterization

The optimum Span 80: Tween 80 ratio was evaluated with respect to the maximum loading of the dispersed aqueous phase in microemulsion. The loading percentage of dispersed aqueous phase as a function of the weight fraction of Span 80 in the total surfactant weight is presented in Figure 1. The loading percentage of the dispersed aqueous phase was first increased with increasing Span 80 composition in the surfactant mixture and then decreased. The highest loading percentage of dispersed water occurred at the weight fraction of Span 80 of 0.5 (Span 80: Tween 80 ratio of 5:5). A certain mixture of Span 80 and Tween 80 provided higher water solubilization in microemulsions than either pure surfactant alone. The results demonstrate a synergistic effect, in which a specific mixture exhibits superior properties compared with those of the individual components [35].

Figure 1.

Effect of the weight ratio of Span 80 to total surfactant on the maximum loading capacity of the dispersed aqueous phase in the microemulsion.

The HLB value of the mixed surfactant (HLB_m) was estimated using the following equation [36]:

HLB_m = 4.3X + 15(1 − X) (1)

where X represents the weight fraction of Span 80 in the total surfactant. The Span 80–Tween 80 mixture at a 5:5 weight ratio yielded an HLB value of 9.65. Although a general recommendation of HLB 3–6 has been proposed for W/O emulsions [37], it has also been suggested that the HLB requirements for microemulsions can differ from those of macroemulsions [35]. Wang and Pal (2015) stated that surfactants with HLB values between 7 and 11 are favorable for W/O microemulsion formation [38]. Mehta and Kaur (2011) noted that an HLB < 10 promotes negative interfacial curvature, favoring W/O microemulsion systems [39]. It has also been found that optimal W/O microemulsions can form within an HLB range of 8.5 to 11, deviating from the HLB 3–6 range associated with macroemulsions [35]. According to Huibers and Shah (1997), synergism in maximum water solubilization for W/O microemulsion occurred at an HLB of 9 in the mixed surfactant [35,40]. Previous studies have also reported that W/O microemulsions can form even when surfactant systems with HLB values greater than 8 are used [41,42]. Additionally, a mixture of Tween 80 and Span 80 at a 1:1 ratio, with an HLB of 9.65, has been reported to favor the formation of W/O emulsions or nanoemulsions [43]. The HLB of 9.65 for the mixed surfactant, which was favorable for the formation of W/O microemulsions with maximum aqueous phase loading, is consistent with literature reports on microemulsions. The effective solubilization of the internal phase in microemulsions occurs when the surfactant’s HLB lies within a specific range [35]. The suitable HLB value obtained from the mixed surfactants is expected to promote favorable conditions for surfactant partitioning at the interface, thereby facilitating the formation of a W/O microemulsion [35]. This specific surfactant mixture ratio was used for subsequent evaluation and preparation of the microemulsion composition.

Phase diagrams serve as a useful tool for analyzing the relationship between component composition and phase behavior. Figure 2 shows a pseudoternary phase diagram for the Span 80–Tween 80 blend (Span 80: Tween 80 = 1:1), chitosan solution, and cottonseed oil at various component compositions. As shown in the pseudoternary phase diagram, the shaded region on the right side of the triangle denotes the formation of a clear, transparent, and homogeneous liquid, corresponding to the characteristic region of the microemulsion system [44]. This finding is comparable to earlier studies, which reported that the W/O microemulsion region appeared as an arc-shaped region at low water fractions [44,45] or was formed near the surfactant–oil binary axis [46]. In this study, the surfactant fraction around 0.5–0.7 appeared to accommodate higher levels of the aqueous phase than both lower and higher surfactant fractions, as indicated by the bulge in the shaded region. Four mixtures (indicated by the triangular points in Figure 2) that formed microemulsions near the phase boundary within this surfactant fraction range were chosen for further characterization and discussion. The aqueous phase–surfactant–oil phase ratios of the four mixtures were 0.08:0.52:0.40, 0.08:0.57:0.35, 0.08:0.62:0.30, and 0.08:0.67:0.25, respectively (Table 1).

Figure 2.

Pseudoternary phase diagram of mixtures composed of the oil phase (cottonseed oil), aqueous phase (chitosan solution), and surfactant (a mixture of Span 80 and Tween 80 at a weight ratio of 1:1). The shaded area represents the monophasic region. Triangle symbols within the monophasic region mark the selected mixtures.

Table 1.

Formulation composition of four selected mixtures, expressed based on a total sample weight of 1 g.

Sample Code	Composition (g)
	Aqueous Phase (Chitosan Solution)	Surfactant 1	Oil Phase (Cottonseed Oil)
M1	0.08	0.52	0.40
M2	0.08	0.57	0.35
M3	0.08	0.62	0.30
M4	0.08	0.67	0.25

¹ Mixture of Span 80 and Tween 80 at a weight ratio of 1:1.

The droplet size and polydispersity index of the selected four mixtures are shown in Figure 3a. The Z-average droplet diameter tends to increase with increasing surfactant ratio, showing only slight variation at lower ratios but a marked rise at higher ratios. When considering the mean PDI values, those at higher surfactant weight fractions were noticeably higher than those at the lowest surfactant weight fraction (M1). A low PDI value (below 0.3) indicates the formation of homogeneous microemulsions with a fairly uniform dispersed phase [47], whereas higher PDI values reflect polydisperse samples with non-uniform size distributions, often comprising bimodal or multimodal populations [48]. Therefore, M1, which exhibited a mean droplet diameter of 92.9 ± 2.3 nm and a PDI of 0.100 ± 0.072, was considered a homogeneous microemulsion. The PDI data were corroborated by the dynamic light scattering (DLS) traces (Figure 3b), which revealed a monomodal distribution for M1 and bimodal size distributions for M2, M3, and M4. In previous reports, high PDI values in some microemulsion systems have been attributed to polydispersity and aggregation of the internal droplets [49,50]. High PDI values at elevated surfactant concentrations have also been reported, attributed to the presence of excess surfactant [51,52]. It has been proposed that once the internal droplets are fully covered with surfactant, unadsorbed surfactant molecules may destabilize the system by promoting droplet coalescence, thereby increasing the PDI and leading to a bimodal size distribution [51,52]. The ‘optimal amount’ theory offers an explanation for the observed variations in droplet characteristics based on different surfactant contents or compositions [53,54]. An appropriate composition of surfactant can cover the surface of the droplets, maximizing their stability against coalescence and leading to the formation of microemulsions with uniformly small droplet sizes. If the concentration of surfactant is insufficient to fully cover the formed droplets, the uncovered droplets are likely to aggregate to reduce interfacial energy. In contrast, when there is an excess of surfactant in the oil phase, unadsorbed surfactant molecules may attach to the droplets, leading to the presence of larger droplets. At the optimal surfactant concentration, the system achieves uniform coverage at the interface, facilitating the creation of monodisperse droplets [55,56].

Figure 3.

(a) Z-average diameter and polydispersity index (PDI) of internal droplets in microemulsions with varying surfactant weight fractions. (b) Intensity-weighted droplet size distributions of microemulsions with varying surfactant weight fractions.

The electroconductive behavior has been reported to correlate with the structural characteristics of microemulsions. Electrical conductivity is a useful parameter that can be applied to differentiate between different types of microemulsions [48]. The electrical conductivities of the microemulsions with different compositions are presented in Figure 4a. All four formulations exhibited low electrical conductivity, and the conductivity values increased with increasing surfactant content. W/O microemulsions, which consist of dispersed aqueous droplets within a continuous oil phase, typically show low electrical conductivity, whereas O/W and bicontinuous systems display considerably higher values [57]. In water–oil–surfactant systems, electrical conductivity values below 5 µS/cm may indicate a W/O microemulsion structure [48,57]. Several studies have noted that conductivities above 1 µS/cm are indicative of the evolution of the bicontinuous or solution-type microemulsions [58,59], whereas values exceeding 10 µS/cm are characteristic of O/W systems, in which water serves as the continuous phase [38]. Accordingly, the low electrical conductivities observed for all four formulations indicate that these systems can be classified as W/O microemulsions. It has also been reported that, at a constant water content, increasing the surfactant concentration results in higher electrical conductivity [48,60]. In the W/O microemulsions, conductivity arises from collisions between dispersed aqueous droplets [48]. The growth, collision, and coalescence of these droplets may therefore contribute to the observed increase in electrical conductivity.

Figure 4.

(a) Electrical conductivity of microemulsions with varying surfactant weight fractions. (b) Viscosity curves of microemulsions with varying surfactant weight fractions.

Figure 4b presents the flow curves of apparent viscosity as a function of shear rate for the microemulsion samples. As shown, the viscosity of all formulations exhibits only a weak dependence on shear rate within the measured range. To further characterize the rheological behavior, the experimental data were fitted to the Ostwald–de Waele (power law) model as follows [61]:

$$\eta =k{\dot{\gamma }}^{n-1}$$ (2)

where $\eta$ and $\dot{\gamma }$ represent the apparent viscosity (Pa·s) and the shear rate (s⁻¹), respectively; k is the consistency coefficient (Pa·sⁿ), and n is the dimensionless flow behavior index.

The fitting parameters obtained from the Ostwald–de Waele model are summarized in Table 2. For all microemulsion samples, the model fitting yielded R² values above 0.9, suggesting a satisfactory agreement with the experimental viscosity data [61,62]. The n values of all samples were close to unity, suggesting that the microemulsions exhibit near-Newtonian behavior [63]. The consistency coefficient (k) reflects the viscous nature of the system [62,63]. In this study, the increase in k with higher surfactant content is consistent with previous reports, which indicate that excess or unadsorbed surfactant in the continuous oil phase can lead to thickening in W/O emulsions and microemulsions [64,65].

Table 2.

Viscosity curve-fitting parameters for the Ostwald-de Waele model.

Sample Code	Ostwald–De Waele Model Parameters
	k (Pa·sn)	n	R2
M1	0.37	0.98	0.98
M2	0.39	0.99	0.92
M3	0.46	0.98	0.93
M4	0.52	0.99	0.93

The low viscosity and Newtonian behavior of microemulsions have been reported to facilitate reactant exchange, making them advantageous as reaction media for nanoparticle synthesis [66,67]. Moreover, the monodispersity of small droplets is a relevant characteristic that contributes to their suitability as nanotemplates [53]. According to the present results, the component composition of the M1 microemulsion system, which contained uniform nanodroplets and exhibited suitable rheological characteristics, was selected for further investigation.

The M1 microemulsion formulation exhibited a percentage transmittance of 92.0 ± 0.5%. This level of transmittance may be considered consistent with microemulsion formation, as suggested by a previous study reporting that transmittance values exceeding 90% reflect optical transparency in such systems [29]. Following centrifugation, no visible phase separation was observed. Additionally, the transmittance of the post-centrifugation sample was 91.5 ± 0.4%, which was comparable to that measured prior to centrifugation. The absence of noticeable change in transmittance, together with the lack of phase separation, indicated that the formulation remained physically stable [30,68,69].

Visual observation of the laser beam passing through the liquids, as shown in Figure 5a, revealed that the path of the laser beam was not visible when it passed through pure water or pure cottonseed oil. This observation is consistent with previous reports for pure liquid components [28]. For the microemulsion system with the optimized composition (M1), its colloidal nature, rather than that of a simple solution, was confirmed by the visible path of the laser beam, indicating the presence of the Tyndall effect [23,28]. This phenomenon arises from the scattering of light by the colloidal droplets in the microemulsion, which possess a large interfacial surface area [28]. To determine the classification of the chosen microemulsion system as W/O or O/W, the oil-soluble dye Sudan III and the water-soluble dye methylene blue were incorporated into the samples, allowing for simultaneous observation of the diffusion characteristics of both dyes. As shown in Figure 5b, the diffusion rate of Sudan III was higher than that of methylene blue, indicating that the continuous phase was oil [70]. The faster diffusion of the oil-soluble dye in the microemulsion confirmed that it was a W/O-type microemulsion [23,70], consistent with the electrical conductivity results.

Figure 5.

(a) Photographs for the observation of the Tyndall effect in pure water, microemulsion M1, and pure oil. (b) Photographs showing the diffusion of Sudan III and methylene blue dyes in microemulsion M1.

3.2. Preparation and Characterization of Genipin-Crosslinked Chitosan Nanogels

The formation of nanogels via the crosslinking of chitosan with genipin was monitored through changes in fluorescence intensity associated with the fluorescent attributes generated during the reaction. Previous studies have reported that the products formed by the crosslinking reaction between amino-containing biopolymers and genipin fluoresce in the 380–700 nm region [71] and that their fluorescence emission maximum depends on the excitation wavelength. Therefore, two excitation wavelengths, 550 and 590 nm, which have been used in the literature [11,71], were initially screened to determine their suitability for monitoring the crosslinking formation of the nanogels. The fluorescence emission spectra of cottonseed oil, the microemulsion containing chitosan, the microemulsion containing genipin, and the mixed microemulsion containing both chitosan and genipin (at a 1:1 molar ratio of genipin to glucosamine units of chitosan) after 24 h of crosslinking are shown in Figure 6. As shown in Figure 6a, although excitation at 550 nm produced the broad fluorescence emission peak in the spectrum of the mixed microemulsion containing both chitosan and genipin, the microemulsion containing chitosan alone also exhibited an upward emission pattern, which could interfere with the interpretation of changes in emission intensity originating from the crosslinking product. Chitosan generally exhibits low intrinsic fluorescence [72]. A prior study reported that chitosan shows weaker fluorescence compared to genipin-reacted chitosan [73]. The red-shifted maximum emission peak observed in genipin-crosslinked chitosan has been attributed to the reaction between genipin and the primary amine groups of chitosan [73]. According to Figure 6b, excitation at 590 nm also produced a broad emission peak in the spectrum of the microemulsion containing both chitosan and genipin, without interference from the oil, the microemulsion containing chitosan, or the microemulsion containing genipin. The maximum emission peak (Figure 6b) appeared at approximately 622 nm, which is close to the reported maximum emission wavelength of about 630 nm for excitation at 590 nm in previous studies [71]. Therefore, an excitation wavelength of 590 nm was chosen for monitoring the progress of the crosslinking reaction between chitosan and genipin.

Figure 6.

Fluorescence emission spectra of cottonseed oil, microemulsion containing chitosan, microemulsion containing genipin, and microemulsion containing both chitosan and genipin (at a 1:1 molar ratio of genipin to glucosamine units of chitosan), measured 24 h after preparation using excitation wavelengths of (a) 550 nm and (b) 590 nm.

To investigate the fluorogenic process, the reaction was monitored over time at 25 °C, and the resulting fluorescence spectra are shown in Figure 7a. Initially, the fluorescence peak at 622 nm was barely detectable. As the reaction progressed, a broad emission band appeared with a maximum at 622 nm. As seen in Figure 7b, the fluorescence intensity increased rapidly during the first 24 h. The increase in fluorescence intensity during the early stage of the investigation likely reflects the progress of the spontaneous reaction between genipin and the primary amine, which generates the conjugated products. The intensity continued to increase slightly beyond 24 h and then flattened out, as indicated by the absence of significant differences in fluorescence intensity between 48 and 72 h. Consequently, the nanogel product from the 48 h reaction at room temperature was used in subsequent experiments.

Figure 7.

Fluorescence monitoring of genipin-crosslinked chitosan nanogel-in-oil upon excitation at 590 nm for the microemulsion system containing chitosan and genipin (at a 1:1 molar ratio of genipin to glucosamine units of chitosan): (a) time-dependent emission spectra and (b) emission intensity at 622 nm.

The fluorescence spectra of reaction systems composed of varied reactant ratios have previously been examined to assess the effect of reactant ratios on fluorescence intensity [74]. In the present study, the fluorescence profiles of microemulsion systems containing chitosan and genipin at two different molar ratios (1:1 and 1:2, genipin to glucosamine units of chitosan) were examined to investigate the effect of the reactant ratio. As shown in Figure 8, the emission spectra at 48 h indicated that the system with a 1:1 molar ratio exhibited higher fluorescence intensity than the 1:2 system. Based on this result, the 1:1 molar ratio was considered suitable for nanogel synthesis under the experimental conditions employed (48 h reaction at room temperature) and was used in subsequent experiments.

Figure 8.

Fluorescence emission spectra of genipin-crosslinked chitosan nanogel-in-oil systems at 48 h post-preparation, obtained upon excitation at 590 nm, for microemulsion formulations containing genipin and chitosan at 1:1 and 1:2 molar ratios (genipin to glucosamine units of chitosan).

The ATR-FTIR absorbance spectra of chitosan, genipin, and chitosan–genipin nanogels over the full wavenumber range of 4000–600 cm⁻¹ are presented in Figure 9a.

Figure 9.

(a) ATR-FTIR spectra of genipin, chitosan, and genipin-crosslinked chitosan in the range of 4000–600 cm⁻¹. (b) Expanded view of the ATR-FTIR spectra of chitosan and genipin-crosslinked chitosan in the region of 1800–600 cm⁻¹.

In the spectrum of genipin, the double peak between 3000 and 3600 cm⁻¹ is associated with the overlapping O-H and C=C-H stretching vibration bands [11,75,76]. The bands observed in the region between 2800 and 3000 cm⁻¹ are attributed to C-H stretching vibrations [75]. The sharp characteristic peaks at 1679 and 1620 cm⁻¹ observed in the functional-group region are associated with the C=O stretching vibration of the carboxymethyl group and the C=C stretching vibration of the olefinic ring in the structure of genipin [75,76]. In the spectrum of chitosan, a broad band centered at 3353 cm⁻¹ corresponded to the overlapping O-H and N-H stretching vibrations [11]. The spectrum of the genipin-crosslinked chitosan nanogels also exhibits a broad peak between 3000 and 3600 cm⁻¹; however, the peak is observed at 3360 cm⁻¹, which is higher than that in the spectrum of chitosan. This shift to a higher wavenumber in the chitosan–genipin particles has been attributed to the formation of a secondary amide through the nucleophilic substitution reaction between chitosan and genipin [77]. The absorption bands between 2800 and 3000 cm⁻¹ observed in the spectra of chitosan and the nanogels arise from the symmetric and asymmetric C-H stretching vibrations of the polysaccharide backbone [11,78].

The changes in the infrared bands below 1750 cm⁻¹ in the ATR-FTIR spectrum of the native chitosan compared with the genipin-crosslinked chitosan were examined to identify possible spectral features associated with the genipin crosslinking of the polymeric structure. To characterize the deviation of the genipin-crosslinked chitosan structure from pristine chitosan, the spectra in the wavenumber range of 1800–600 cm⁻¹ are shown in Figure 9b. A summary of the selected band assignments for the samples is presented in Table 3. In the spectrum of chitosan, the absorption peak at 1652 cm⁻¹ represents the C=O stretching of the remaining acetyl groups [78,79,80]. The broad peak centered at 1591 cm⁻¹ corresponds to N-H bending in primary amines [79,80], while the peak near 1375 cm⁻¹ is attributed to the C-N stretching of N-acetyl groups [80]. Compared with pristine chitosan, the genipin-crosslinked polymer displays a new peak at 1634 cm⁻¹ with a shoulder at 1601 cm⁻¹, replacing the bands at 1652 and 1591 cm⁻¹ observed in the spectrum of chitosan. This change may reflect a spectral pattern arising from the covalent interaction of amino groups with small molecules, leading to amide formation [80]. In addition, the peak near 1633 cm⁻¹ has been assigned to the C=O stretching vibration of amide linkages formed between chitosan and genipin [78]. Meanwhile, the decreased intensity of the N-H bending band is attributed to the formation of a heterocyclic amine structure in the genipin-crosslinked chitosan, resulting from nucleophilic substitution by the amino groups of chitosan on the olefinic carbon atom in genipin [81]. In addition, the small absorption band observed in the spectrum of chitosan near 660 cm⁻¹, which is related to the crystalline region of chitosan [82], disappeared after crosslinking. This implies that the formation of the chitosan–crosslinker complex constrains the regular arrangement of the chitosan chains and diminishes the intensity of this crystalline-sensitive band, as previously described [82].

Table 3.

Selected characteristic ATR–FTIR bands in the wavenumber range of 1800–600 cm⁻¹ of chitosan and genipin-crosslinked chitosan.

Band Assignment	Wavenumber (cm−1)
	Chitosan	Genipin-Crosslinked Chitosan
C=O stretching	1652	1634
N-H bending	1591	1601
C-N stretching	1375	1376
Crystalline sensitive band	660	-

The particle size distribution of the nanogels, as determined by DLS, is presented in Figure 10a. The prepared nanogels exhibited a monomodal size distribution, with particle sizes ranging from 59 to 255 nm. The Z-average particle diameter was 108.3 ± 2.4 nm, accompanied by a PDI of 0.185 ± 0.048. The slight increase in the Z-average diameter of the genipin-crosslinked chitosan nanogels compared with the microemulsion template is consistent with reports in the literature, where the incorporation of crosslinkers among polymer chains leads to intermolecular linking and, consequently, an increase in particle size [83]. The development of chitosan–genipin nanomaterials using techniques such as ultrasound treatment or orifice-induced hydrodynamic cavitation (HC) has previously been explored through water-in-palm oil nanoemulsion systems, followed by emulsion crosslinking [17]. The ultrasound-based approach produced particles with irregular agglomerates and reported a mean size of 725.8 nm (PDI = 0.762), while the HC method, employing a specifically designed orifice plate, yielded more spherical particles with a mean size of 312.6 nm and a PDI of 0.359 [17]. In the present study, a microemulsion-based templating strategy employing a low-energy emulsification method was applied. From the perspective of microemulsion-based technology, this approach emphasizes simplicity and energy efficiency while providing favorable physicochemical characteristics such as nanoscale size and relatively narrow size distribution [23]. Accordingly, this strategy may represent an alternative pathway for the fabrication of chitosan-based nanogels with desirable properties.

Figure 10.

(a) Intensity-weighted particle size distribution of genipin crosslinked chitosan nanogel-in-oil. (b) TEM image of genipin crosslinked chitosan nanogel-in-oil. Scale bar: 500 nm. (c) Particle size distribution histogram.

TEM serves as a useful tool for visualizing the nanoscale morphology of nanodispersions. The particle morphology of the nanogels is shown in the TEM image in Figure 10b, which depicts nanogels with discrete spherical shapes. The particle size distribution histogram (Figure 10c), derived from TEM inspection, showed reasonable agreement with the size range reported by DLS.

To assess the reproducibility of the fabrication process, a second batch of nanogels was prepared. This batch yielded nanogels with a Z-average particle diameter of 111.8 ± 12.6 nm and a PDI of 0.217 ± 0.073. No statistically significant differences were observed between the two batches in either Z-average particle size or PDI (p > 0.05), indicating that the fabrication method was reproducible.

This study presents a potentially useful approach for developing a nanogel-in-oil system by employing a microemulsion template in combination with an emulsion crosslinking technique. The findings may serve as a foundation for future investigations into chitosan-based nanogel-in-oil systems, particularly for their possible use as oil-based nanodispersions for delivering drugs and bioactive compounds in pharmaceutical, cosmetic, and food-related applications. Further exploration of the functional properties relevant to practical applications may help bridge the understanding between the physicochemical characteristics of the fabricated nanogels and their potential utilization as nanocarriers or active ingredients across various product sectors.

4. Conclusions

This study demonstrates a viable strategy for fabricating chitosan-based nanogel-in-oil by integrating the formation of W/O microemulsion templates with genipin crosslinking. The microemulsion system was successfully developed using cottonseed oil and a blend of Tween 80 and Span 80, resulting in nanometer-sized internal aqueous droplets. Guided by the monophasic region of the pseudoternary phase diagram, suitable formulations were selected and optimized. The resulting W/O microemulsion formulation exhibited desirable characteristics, including low viscosity and uniform droplet size. Fluorescence intensity measurements provided evidence for the progression of crosslinking between chitosan and genipin, while FTIR analysis confirmed the associated intermolecular interactions. Additional characterization by DLS and transmission electron microscopy revealed nanogels with nanoscale dimensions and discrete spherical morphology. Together, these results underscore the potential of the combined microemulsion templating and emulsion crosslinking approach for the development of a nanogel-in-oil system.

Author Contributions

Conceptualization, N.H.; methodology, N.H., S.K., P.K. and P.P.; formal analysis, N.H. and S.S.; investigation, N.H. and S.S.; resources, N.H., P.K. and P.P.; writing—original draft preparation, N.H.; writing—review and editing, N.H. and P.K.; visualization, N.H.; supervision, N.H.; project administration, N.H.; funding acquisition, N.H. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not applicable.

Data Availability Statement

The original contributions presented in this study are included in the article; further inquiries can be directed to the corresponding author.

Conflicts of Interest

The authors declare no conflicts of interest.

Funding Statement

This research is supported by Thailand Science Research and Innovation (TSRI) Fundamental Fund, fiscal year 2025, and by Thammasat University Research Unit in Smart Materials and Innovative Technology for Pharmaceutical Applications (SMIT-Pharm) [Project ID 6305016].