Stabilization of ginger essential oil Pickering emulsions by pineapple cellulose nanocrystals

Abstract

Pickering emulsions (PE) are systems made up of two incompatible fluids, these are stabilized by solid organic or inorganic particles located on their interface. Cellulose nanocrystals (CNCs) are sustainable and biocompatible value-added naturally occurring biomolecules which are being investigated as PE stabilizers in the cosmetic, food, and pharmaceutical industries. The objective of this research was to investigate the efficacy of pineapple cellulose nanocrystals as stabilizers for a ginger essential oil-in-water Pickering emulsion. Anionic pineapple cellulose nanocrystals were prepared by acid hydrolysis. Ginger essential oil-in-water emulsions were prepared by ultrasonication. Pineapple CNC produced stable Pickering emulsions with surface average droplet size of 4.3 μm–6.2 μm, high negative zeta potential, high viscosity, and high adsorption at the interface. Pickering emulsions by ultrasonication were stable against droplet coalescence, phase separation, and droplet flocculation for at least 8 weeks at 25 °C or 40 °C at various droplet sizes. The emulsion droplet size and volume density (droplet size distribution) were evaluated by varying the particle concentration (CNC 0.25 g/100 ml or 0.50 g/100 ml) and/or oil fraction (10–20 g/100 ml). At constant oil fraction, the emulsion viscosity increased as the nanocrystal concentration increased. The cellulose nanocrystal-stabilized ginger oil-Pickering emulsions exhibited shear-thinning characteristics of a pseudo-plastic fluid. Pineapple nanocellulose crystal -stabilized ginger oil-Pickering emulsions exhibited high stability with a creaming index of zero. CNC was found to be an effective Pickering stabilizer for oil-in-water emulsions in various food applications.

Graphical abstract

Highlights

• •Cellulose nanocrystals (CNC) were produced from pineapple peel waste.
• •The CNC were used to make Pickering emulsions of ginger essential oil.
• •These Pickering emulsions were stable for 8 weeks at 25 °C and 40 °C.
• •The concentration of CNC influenced the size of the emulsified oil droplets.
• •CNC stabilized the PE emulsion by adsorbing at interface and increasing viscosity.

1. Introduction

Pineapple is a tropical fruit that can be eaten fresh or processed into several food products. It ranks third in terms of tropical fruit production worldwide, behind bananas and citrus. The leading producers of pineapples are Costa Rica, Philippines, Brazil, Thailand, and India (Aili Hamzah et al., 2021). The pineapple business has expanded significantly in recent years due to the fruit's attractive flavor and health benefits, as well as its high demand and competitive retail prices. As a result of increased global demand and the outer surface of pineapples being inedible, waste from pineapple production and consumption has increased (Aili Hamzah et al., 2021). Both fresh consumption and processing of pineapple generate a substantial number of by-products, including leaves, stems, and peels, which regularly account for 50% (w/w) of the pineapple's total weight (Roda and Lambri, 2019). Recently, pineapple peel has been used in many fields, such as bioenergy and for generating bioactive compounds, food ingredients, and biomaterials (Chen et al., 2021). There are several studies on pineapple peel cellulose, such as those involving the production of hydrogels, packaging materials, biopolymers, and emulsions. Nanocellulose is usually used to describe cellulose that is nanoscale in at least one dimension. In recent years, cellulose nanocrystals (CNCs) have garnered a significant research interest due to their unique nanoscale size, ability to be produced from broad variety of sources, amphiphilicity, and surface that is simple to functionalize. Additionally, CNCs possess good physical and chemical properties (Cheng et al., 2022).

Essential oils (EO) are usually distilled liquid extracts from plants that are lipophilic and volatile. Plants develop the components found in EO as secondary metabolites to aid in their survival against environmental stressors like pathogens. Plant EO are recognized to possess antibacterial, antifungal, antiviral, antioxidant, and biological modulating properties. Essential oils from ginger (Zingiber officinale) or GEO are well-known for their antioxidant, anti-inflammatory effects, fungal and antibacterial properties, and their application is increasing for food preservation (Wang et al., 2020). The presence of zingiberene and α-curcumene is most likely responsible for the benefits associated with GEO (Wang et al., 2020). In recent years, GEO has found widespread applications in food, pharmaceutical, cosmetic, and animal feed industries (Radice et al., 2022). Due to the bioactive qualities of EO, their use is increasing in a number of applications, such as alternative medicine, cosmetics, cleaning agents, food items, green antimicrobials, and packaging. However, EO have high volatility, hydrophobicity, photosensitivity, high susceptibility to oxidation, a strong odor, low stability, and poor solubility in water due to their chemical structure and intrinsic features. Numerous encapsulation methods have been devised to make EO less volatile and more stable during processing and storage (Pontes-Quero et al., 2021; Mahdi et al., 2021; Granata et al., 2021).

Pickering emulsions (PE) are stabilized by solid particles or other colloidal surfactants interacting on the interface of two incompatible liquids. PE have many advantages over conventional molecular surfactant-stabilized emulsions, including high resistance to coalescence, long-term stability, excellent biocompatibility, and tunable properties (Sun et al., 2022). The use of nanoparticles to form a colloidal system with increased stability and application ease is a potential strategy for stabilizing EO. In this system, solid particles are adsorbed at the oil/water interface (Souza et al., 2021). In addition, these particles inhibit phase separation and promote emulsion stability by preventing droplet coalescence through mechanical and steric properties (Gestranius et al., 2017). Numerous organic and inorganic particles, such as silica, alumina, titanium oxides, starch, zein, proteins, cellulose nanofibrils and nanocrystals, have been studied to stabilize oil-in-water emulsions. PE from inorganic particles have demonstrated good performance as stabilizers, but it is difficult to control the surface properties of such particles (Zhang et al., 2018). Inorganic-particle-stabilized PE have relatively low biodegradability and biocompatibility, which limit their application in a broad range of industries (Yang et al., 2017). PE from organic particles have more potential applications than those from inorganic particles, as they are amenable for surface modification by chemical reagents (Hossain et al., 2021). In addition, the application of natural and biodegradable organic solid particles in PE formulations has increased gradually throughout the years, adapting to the growing need for safer and more sustainable materials. In recent years, many polysaccharides and natural proteins have been intensively examined under a variety concentrations, temperatures, pH, and ionic strengths for PEs, including high internal phase PEs (Feng et al., 2021; Hossain et al., 2021). Utilizing oregano essential oil PE, Zhou et al. (2018) produced an efficient antibacterial treatment. The authors stabilized the PEs using CNC and the results indicated that the emulsions were stable at higher CNC concentrations and pH values, as well as at lower oil-to-water ratios and salt concentrations. In addition, the antibacterial activity of oregano essential oil-based PE was investigated against four food-related pathogens including S. aureus ATCC 6538, B. subtilis ATCC 6633, E. coli ATCC 8739, and S. cerevisiae ATCC 9763. The results demonstrated that prepared oregano essential oil-based PE inhibited the growth of these bacterial strains by degrading their membrane integrity (Zhou et al., 2018). The same research group also prepared a clove essential oil Pickering emulsion (CO-PE) which was stabilized by sodium-modified carboxymethyl CNC produced through homogenization. CO-PE demonstrated excellent antimicrobial activity against E. coli and S. aureus. The production and stability of CO-PE were examined with respect to CNC concentration, homogenization pressure, CO concentration, ionic concentration, and pH. The findings demonstrated that 1% CNC stabilized CO-PE had small droplet sizes, a rough surface, and it was stable at high pH levels (from 2 to 10) or salt concentrations because CNC occupied the droplet interface. At the same CO concentration, the CNC-stabilized CO-PE displayed more antibacterial activity, which may be ascribed to its ability to effectively attach to bacterial membranes. Consequently, this research provided new insights on the application of PE and EO as antimicrobial materials in the food and other industries (Yu et al., 2021). Han et al. (2022) developed emulsions of cellulose nanofibrils (CNF) and nanocrystals (CNC) in acidic lithium bromide trihydrate. It was observed that, at a constant CNC content (0.3 wt%), increasing the CNF concentration from 0 to 0.9 wt% significantly altered the stability and microstructure of PEs. The authors demonstrated that as CNF loading increased, the oil droplet size decreased and stabilized the emulsion. This emulsification stability was associated with the irreversible attachment of CNCs to the oil droplet surface and the creation of a dense CNF network in the water phase, which made the emulsion more stable (Han et al., 2022).

To the best of our knowledge, no research has been conducted to develop and evaluate GEO-PE stabilized by CNC from pineapple peel. Thus, the focus of this research was the extraction of CNC from pineapple peels, characterization, and formulation of CNC-stabilized GEO-PE (a particle-stabilized emulsion). The emulsions were tested for droplet size, zeta-potential, morphology, rheology, microstructure, and stability.

2. Materials and methods

2.1. Materials and reagents

Pineapples (Chiquita®) were purchased from a local supermarket in Baton Rouge, LA (USA) and GEO (Zingiber officinale, HIQILI®) is 100% pure and natural from Huiqili Supply Chain Manufacture. It was purchased from Amazon (USA), and were steam distilled from roots or rhizomes. Sodium chlorite, sodium hydroxide, glacial acetic acid, and sulfuric acid were purchased from Thermo Fisher Scientific (Waltham, MA, USA); A dialysis trial kit (12-14 KDa) was purchased from Avantor (Radnor, PA, USA). MilliQ water (Synergy® UV systems) was used in all experiments. All chemicals or solvents used in these experiments were of analytical grade and used as received.

2.2. Preparation of Pineapple CNC

2.2.1. Purification of pineapple peel powder

The pineapple peel was removed by slicing off the ends and running a sharp knife along the skin at a thickness of approximately 1.5 cm and washed with running tap water followed by soaking in MilliQ water for 1 h to remove dirt and dust. All the pineapple peels were cut into 2 × 2 cm pieces and chopped in a blender (Oster® 14-speed all-metal drive) on the chop mode for 5 min. The chopped pineapple peels were lyophilized (SP VirTisTM Genesis Pilot Lyophilizer, Gardiner, NY, USA) for 48 h. The dried pineapple peels were ground and sieved through a 100-mesh screen to separate pineapple peel fibers (P1). To remove hemicellulose, pineapple peel fibers were treated with 2% (w/v) aqueous sodium hydroxide (pineapple peel fibers 50 g and 2% soduim hydroxide 800 mL) in a hot air oven at 100 °C for 4 h, with manual stirring every 15 min. The fibers were recovered by vacuum filtration and washed several times with MilliQ water until the pH became neutral (P2). The resultant fibers were then bleached with a solution containing equal parts (v:v) of acetate buffer (27 g NaOH and 75 mL glacial acetic acid, diluted to 1 L with deionized water) and 750 mL of 1.7% NaClO₂ sodium chlorite in water. This bleaching procedure was carried out in a hot air oven at 80 °C for 6 h with manual shaking every 15 min. The bleached fibers were washed repeatedly in MilliQ water until the pH of the washing solution became neutral and the fibers were subsequently lyophilized. The samples which resulted after this purification were designated as the bleached fibers (P3).

2.2.2. Extraction of CNC

Following the above-mentioned method of purification of peel powder, CNC were extracted using acid hydrolysis. The hydrolysis was conducted for 60 min at 50 °C with constant stirring (100 rpm). For each gram of P3, 20 mL of an H₂SO₄ solution (64% by weight) was used. The suspension was diluted 10-fold with cold water to stop the hydrolysis reaction and centrifuged twice for 10 min at 7000 rpm and 12 °C to remove excess acid. The precipitate was then dialyzed against MilliQ water for 7 days to eliminate unreacted sulfate groups, salts, and soluble sugars. The dialyzed colloidal suspension was lyophilized. The CNCs were labelled as P4. The extraction yield of CNC from pineapple peel was about 15% dry basis. The digital images of material obtained after each of the above treatments are shown in Fig. 1.

Fig. 1.

Digital images of the extraction of cellulose nanocrystals from pineapple peel, a: pineapple peel fiber (P1), b: alkali treated fibers (P2), c: bleached fibers (P3) and d: cellulose nanocrystals (P4).

2.3. Characterization of pineapple peel fibers and CNC

2.3.1. Fourier transform infrared spectroscopy (FTIR)

FTIR spectra were recorded using an IR spectrometer (Tensor-27, Bruker Optics Inc., Billerica, MA, USA) in transmittance mode in the wavenumber range from 400 to 4000 cm⁻¹ with a resolution of 4 cm⁻¹. A total of 64 scans were carried out for each sample and reported data is the accumulation of these scans.

2.3.2. Morphology of pineapple peel fibers and CNC

The microstructure of the pineapple peel fiber at each stage of processing was acquired using a scanning electron microscope (SEM, Quanta 3D FEG FIB/SEM, FEI Company, Oregon, USA). The samples were affixed on silicon wafer pieces attached to aluminum stubs using double-sided adhesive conductive carbon tape. Specimens were coated with a thin layer of iridium using a sputter coater to increase conductivity and avoid the formation of electrical charge during imaging. All testing was conducted at ambient temperature. The morphology of CNCs isolated from pineapple peel was acquired using a transmission electron microscope (TEM, JEM-1400, Japan) with 100 kV accelerating voltage. During sample preparation, the CNC were diluted by 0.1% by weight. The negative staining with uranyl acetate (1.0% by weight) was performed to address contrast problems.

2.3.3. Determination of crystallinity

The crystallinity index ($CI$) of the CNCs was determined using a PANalytical Empyrean X-Ray Diffractometer (Malvern Panalytical Inc, MA, USA). The X-ray generator operated at 45 kV and 40 mA with Cu Kα radiation in 2°/min scan rate. The powder samples were placed on nickel coated steel holder to carry out the diffraction procedure. The $CI$ of samples was calculated from the diffracted peak intensity data using Eq (1) (Mondragon et al., 2014).

$$CI=\frac{I_{002}{-I}_{am}}{I_{002}}x100$$ (1)

where, $I_{002}$ is the maximum intensity of the (002) lattice diffraction peak (diffraction angle at around 2θ of 23° representing the crystalline material) and $I_{am}$ is the intensity of the amorphous part of the sample (the location of the amorphous material intensity considered in this work was at around 2θ of 18°).

2.4. Preparation and characterization of CNC-stabilized GEO-PE

2.4.1. Preparation of CNC-stabilized GEO-PE

In this research, when the term Pickering emulsion (PE) was used, the authors meant a particle-stabilized emulsion. The CNC-stabilized GEO-PE were produced using different concentrations of GEO and CNC (Supplementary Material Table S1). GEO-PE stabilized by CNC was prepared as follows. The coarse emulsion was produced by mixing CNC with water and GEO for 5 min using a high-speed homogenizing disperser (T25 digital Ultra-Turrax, IKA, Germany) at 10,000 rpm. The PE was obtained by further emulsifying the coarse emulsion by ultrasonicating (750 W, 20 kHz, and 75% amplitude) for 5 min using a standard 13 mm diameter probe (VC750, USA). Ultrasonication was performed for 5s on and 5s in an ice bath.

2.4. 2characterization of CNC stabilized GEO-PE

2.4.2.1. Droplet size measurement and zeta-potential

The droplet size distribution of freshly formed emulsions was determined making use of light scattering and laser diffraction functions of a Mastersizer 2000 (Hydro 3000 MU, Malvern Instruments, Malvern, USA). The emulsions were diluted in recirculating deionized water (3000 rpm) to a 10% opacity. The refractive indices of GEO and water used were 1.48 and 1.33, respectively. The average diameter of each emulsion sample was expressed as volume mean diameter [d (4,3)]. The zeta potential of the samples was determined through dynamic light scattering (Zetasizer Nano, Malvern, UK) at 25 °C.

2.4.2.2. The morphology of CNC-stabilized GEO-PE

The microscopic morphology of CNC-stabilized-PE was observed using a confocal laser scanning microscope (Leica SP8 Confocal with White Light Laser System, Germany). Nile Red and Nile Blue A were used having excitation wavelengths of 488 and 633 nm, respectively. Each sample was colored with a mixed dye created by dissolving 0.1% Nile Red and 0.1% Nile Blue A in dimethyl sulfoxide prior measurement. Then, 40 μL of the dye mixture was added to 1 mL of emulsions. All the photographs were captured at a 20 × magnification.

2.4.2.3. Rheological properties

A Discovery HR-2 rheometer (TA Instruments, US) was used to measure the rheological characteristics (apparent viscosity) of CNC-stabilized GEO-PE produced using a 20 mm parallel steel plate. Tests were performed at 25 °C, and the gap between the two plates was adjusted to 1 mm. The apparent viscosity was measured as a function of shear rates ranging from 1 s⁻¹–100 s⁻¹.

2.4.2.4. Stability under storage

Fresh emulsion samples were added to glass vials and stored at 25 °C or 40 °C for 8 weeks. The emulsion stability (%) was measured every week. The creaming index of emulsion was calculated by using equation (2) (Souza et al., 2021), where $H_{cream}$ is the height (mm) of cream layer, and $H_{total}$ is the total height (mm).

$$Creamingindex(\%)=\frac{H_{cream}}{H_{total}}\times 100$$ (2)

2.5. Statistical analysis

In this study, we used statistical analysis to compare the size of CNC-stabilized GEO-PE at different concentrations of CNC and GEO and the effect of the combination of CNC and GEO. The null hypothesis of this study was that pineapple CNCs were not an effective stabilizer of GEO emulsions. The alternative hypothesis was that pineapple CNCs were an effective stabilizer of GEO emusions. The experimental design was a Factorial Completely Randomized Design, factor A was the concentration of CNCs (two concentrations: 0.25 g/100 mL or 0.50 g/100 mL), factor B was the concentration of GEO (three concentrations: 10, 15, and 20 g/100 mL), and the combination between factors A and B was (CNCs*GEO, 2*3 = six samples). Each individual measurement was performed five times, and the experiment was carried out in duplicate. Results are reported as the mean ± standard deviation. IBM's statistical software (SPSS®, version 26, IBM Corp.) was used to analyze the data. General Linear Model, Univariate Analysis of variance were conducted to determine the significant difference between the subject's effects (CNC, GEO, and CNC*GEO). Significant groups were identified by post-hoc comparison tests (Duncan's Multiple Range Test, DMRT) at a 95% confidence level (P < 0.05)

3. Results and discussion

3.1. Confirmation of generation of peel fiber and CNC

The FTIR spectra of fibers at different stages of processing and CNC were analyzed to understand the change in their chemical composition and are shown in Fig. 2. Two main transmittance bands in two wave number regions of 3500 to 2800 cm⁻¹, and 1740 to 600 cm⁻¹. The presence of peaks on the spectra of cellulose samples from pineapple peel fiber corresponds to bands of microcrystalline cellulose. Peaks at wavenumber about 3320 cm⁻¹ were attributed to the O–H stretching vibration of the hydrogen-bonded hydroxy groups in the cellulose molecules (Dhali et al., 2021). The band at 2900 cm⁻¹ corresponded to the CH stretching vibration of all hydrocarbon constituents in cellulose. Typical cellulose bands were observed in the range of 1740 to 600 cm⁻¹. The peak observed at 1735 cm⁻¹ could be assigned to acetyl groups of hemicellulose, as well as ester bonds of the carboxylic groups of ferulic and p-coumaric acid in lignin and hemicellulose (Dai et al., 2018). The effectiveness of bleaching agents used during the bleaching process can be determined by the absence of peak at 1735 cm⁻¹ (P3). It was observed that this peak was absent when pineapple peel fiber was treated with sodium chlorite (Fig. 2, P3). The peak at 1635 cm⁻¹ was caused by water absorption. The C–O–C asymmetrical stretching vibration of −1,4 -glycosidic linkages was represented by peaks at 1159 cm⁻¹, whereas the C–OH stretching vibration of D-glucose units was observed at 1105 cm⁻¹. These bands are characteristics of the cellulosic component and are improved in the bleached fibers. The aromatic skeletal vibration of lignin corresponded to the characteristic peak (P1) between 1200 and 1300 cm⁻¹ in the spectra of untreated fiber. The band observed at 1240 cm⁻¹ could be attributed to C–O–C stretching from the ether linkage of P1, indicating the existence of lignin. The main components of pineapple peel fiber were, thus, cellulose, hemicellulose, and lignin with functional groups such as alkanes, esters, aromatics, ketones, and alcohols. The absence of certain functional groups that can be attributed to non-cellulosic components in the spectrum of bleached fibers indicated a high degree of purification accomplished during the bleaching treatment. The hemicellulose and lignin peaks detected in P1 (pineapple peel fiber) were missing in chemically treated fibers (P2 and P3) and CNC (P4). The peak observed at 1163 cm⁻¹ in the CNC spectrum (P4) suggested the presence of a sulfated group (SO₂), which was likely caused by the sulfonation of cellulose during sulfuric acid hydrolysis. The peaks observed at 1429, 1315, 1035, and 896 cm⁻¹ represented typical transmittance peaks associated with cellulose. Each peak could be categorized as CH₂ bending, CH₂ rocking, C–O stretching, C–H bending, or CH₂ bending. In addition, the peak observed at 896 cm⁻¹ indicated the presence of −1,4-glycosidic linkages of the glucose ring in the cellulose chain. These linkages bind the anomeric carbon atoms of saccharides to create polysaccharides. Thus, the presence of glycosidic linkages implies the presence of cellulose. Alkaline treatment and bleaching of pineapple peel revealed that pure cellulose was obtained (lignin and hemicellulose were removed). The glycosidic bonds of the long chain of cellulose were further broken down into nanocrystalline particles (Chieng et al., 2017).

Fig. 2.

FTIR spectra of pineapple peel fiber (P1), alkali treated fiber (P2), bleached fiber (P3) and CNC (P4).

3.2. Morphology of fibers at different stage of treatment

The surface morphology of the pineapple peel, from the initial fibers to those obtained after each treatment stage was acquired through SEM. The elimination of non-cellulosic components (hemicelluloses, lignin, pectin, wax, and other impurities) from the fiber, which causes fibrillations of fibers, alters the fiber's shape, exposing its surface (Chieng et al., 2017). The SEM micrographs of pineapple peel fiber (P1), alkali treated fiber (P2), bleached fiber (P3) and CNC (P4) are shown in Supplementary Material Fig. S1. These micrographs depicted the steady impact of treatment on the fiber surface. Due to the partial delignification and removal of the impure components from the fiber surface, the surface of chemically treated fiber (P2 and P3) appeared uneven and rugged compared to the surface of untreated fiber (P1). This could be attributed to the breaking of the ether bonds as the concentration of hemicellulose and lignin gets reduced during alkali treatment (Dhali et al., 2021). During alkali treatment with sodium chloride, impurities such as wax and cuticle were removed from the fiber surface. SEM micrograph showed a change in morphology of the pineapple fibers following bleaching. As increased proportion of non-cellulose components and contaminants were eliminated during the bleaching process and the breakdown of fibers into microfibers occurred, it led to the greater exposure of fibrils and rendered the fiber surface rougher (Faria et al., 2020). The alkali treatment and bleaching degraded the lignocellulose complex, dissolved the lignin, pectin, and hemicellulose, and increased the porosity and surface area of the hidden cellulose. This led to the alignment and distribution of cellulosic fibers, making them more accessible for the extraction of cellulose nanocrystals (Dai et al., 2018; Chieng et al., 2017). Furthermore, hydrolysis of pineapple peel fiber with sulfuric acid removes the amorphous regions of cellulose. The rod-like cellulose nanocrystals were depicted after extraction of CNC (Supplementary Material Fig. S2).

The morphology and nanoscale of CNC were investigated using transmission electron microscopy. It was anticipated that the acid hydrolysis would eliminate the amorphous region of the cellulose fibers while leaving the crystalline region intact; this would ultimately reduce the size of the cellulose fibers to nanoscale size. It can be seen in Fig. S2 that the CNCs had rod-like shape. In addition, TEM micrographs revealed a fiber network approximately 20 nm in diameter and more than 200 nm in length. The CNCs appeared agglomerated, most likely due to the surface electrostatic charge. The removal of the dispersion medium during TEM sample preparation might also have facilitated the formation of aggregates.

3.3. X-ray diffraction pattern of fibers and CNC

XRD patterns were obtained at the end of each treatment stage to determine the crystalline index (CI) of the pineapple peel fiber. The acquired X-ray diffractograms of the fiber and CNC showed well-defined crystalline peaks at 22^o and 35^o (Supplementary Material Fig. S3) which are characteristics of cellulose I, the most common crystalline form of cellulose polymorphs (Aprilia and Arahman, 2020). The largest peak, which is characteristic of crystalline structure of cellulose I, was observed for all samples at 22.3^o. A much weaker peak was observed at 35.0^o; indicating a poorly crystalline, nanocrystal-based material.

Aside from the presence of a broad peak at around 2θ=16^o associated with the amorphous structure, a major diffraction peak for cellulose I at 22^o and the lowest intensity peak at an angle of 2θ=18^o were also observed (Chieng et al., 2017). The CI of untreated pineapple fiber (P1) had the lowest value (12.6 %) due to its high number of amorphous regions. Following alkaline and bleaching treatments, the CI of the fibers (P2 and P3) increased to 49.4% and 51.0%, respectively. This increase in CI was due to the removal of lignin and hemicelluloses from the fibers. The CI of CNC (obtained after bleaching) increased to 60.1 % (Supplementary Material Fig. S3). The increase in CI caused by acid hydrolysis suggested the dissolution of an amorphous component of the cellulose fiber (Aprilia and Arahman 2020; Chieng et al., 2017). Also, growth and realignment of monocrystals can occur at the same time, which gets reflected in the diffractogram by a rise in cellulose crystallinity and a narrowing of the diffraction peaks (Chieng et al., 2017).

3.4. Droplet size of CNC-stabilized GEO-PE

The droplet size of the CNC stabilized GEO-PE at different concentrations of CNC (0.25 g/100 mL or 0.50 g/100 mL) and GEO (10, 15, and 20 g/100 mL) is shown in Fig. 3a. The CNC stabilized GEO-PE with 10, 15, or 20 g/100 mL of GEO at CNC 0.25 g/100 mL were designated PE1, PE2, and PE3, respectively. The CNC stabilized GEO-PE containing 10, 15, or 20 g/100 mL of GEO at CNC 0.50 g/100 mL were named PE4, PE5, and PE6, respectively. It has been reported that nano- or micro-particles strongly adsorb at the oil-water interface in the form of a densely packed layer that, by a steric and/or electrostatic process, protects droplets from flocculation and coalescence (Duffus et al., 2016).

Fig. 3.

(a) Emulsion droplet size (d [4,3]) and (b) Droplet size distribution of CNC stabilized GEO Pickering emulsion at different GEO and CNC concentrations. The CNC stabilized GEO-PE with 10, 15, and 20 g/100 mL of GEO at CNC 0.25 g/100 mL were designated PE1, PE2, and PE3, respectively. The CNC stabilized GEO-PE containing 10, 15, and 20 g/100 mL of GEO at CNC 0.50 g/100 mL were designated PE4, PE5, and PE6, respectively. Different letters (a–f) in the bar column indicate significant difference at p < 0.05.

The emulsion of PE1, PE2, PE3, PE4, PE5, and PE6 had mean droplet size of 4.4, 6.2, 6.2, 4.3, 5.5, and 6.0 μm, respectively (Fig. 3a). In both concentrations of CNC (0.25 g/100 mL or 0.50 g/100 mL), the GEO concentration of PE increased (as the concentration of GEO increased from 10 to 15 and 20 g/100 mL), resulting in bigger droplet size. There was a tendency for the oil droplet size to increase with the increase of oil content in the emulsions which could be attributed to the decrease of stabilizer (CNC) availability to bind to the oil surface. The oil-to-water ratio had a significant effect on the size of emulsified oil droplets. The CNC had a substantial effect on the average oil droplet size (Campelo et al., 2017; Zhou et al., 2018; Bai et al., 2019). When the concentration of particles in the emulsion is fixed the addition of additional oil could expand the interfacial area, which the particles will not be able to cover. As a result, the droplets coalesced (small droplets coalescing faster) which reduced the interfacial area. At higher oil/water ratios, the CNC concentration had a stronger effect on droplet size. The higher CNC concentration (0.50 mg/100 mL) resulted in smaller droplet sizes at GEO concentrations of 15 or 20 mg/100 mL than the lower CNC concentration (0.25 mg/100 mL). When the CNC concentration was higher at a fixed GEO concentration, there were more solid particles available to stabilize more emulsion droplets. Bai et al. (2019) showed that the average droplet diameter decreased with the increase of CNC concentration from 0.05 to 0.75 wt%, but subsequently plateaued at ∼1 μm between 0.75 and 2.0 wt%. This indicated that the presence of CNC might have prevented droplet disruption within the homogenizer, or that some droplet coalescence happened before the oil droplets were completely coated with CNC. Depending on the concentration of CNC, the viscosity of the aqueous phase would be altered which would affect droplet formation and coalescence (Bai et al., 2019). As shown in Fig. 3b, all CNC stabilized GEO-PE had a monomodal size distribution.

3.5. Microstructure of CNC-stabilized GEO-PE

The microstructure of the CNC stabilized of GEO-PE was obtained using a confocal laser scanning fluorescence microscope (Fig. 4). In this case, the CNC (blue color) and spherical oil droplets (pink color) were stained with Nile blue and Nile Red, respectively. The aqueous phase containing CNC was distributed around oil droplets, which indicated that PE stabilized with CNC were oil-in-water emulsions. Similar results were reported by Nie et al. (2022) who showed that the PE stabilized by CNC and peanut protein isolate were oil-in-water emulsions. The images revealed that most of the CNC was surrounding the oil droplets, i.e., at oil-water interface which was in accordance with the literature (Yu et al., 2021; Bai et al., 2019). The presence of bluish areas in the aqueous phase surrounding the lipid droplets with high CNC loading (0.50 mg/100 mL), as seen in Fig. 4 (PE4, PE5, and PE6), suggested that either multilayers were formed on the droplet surfaces or that excess CNC was present in the aqueous phase.

Fig. 4.

Confocal laser scanning fluorescence micrographs of CNC stabilized GEO-PE at different GEO and CNC concentrations. The CNC stabilized GEO-PE with 10, 15, and 20 g/100 mL of GEO at CNC 0.25 g/100 mL were designated PE1, PE2, and PE3, respectively. The CNC stabilized GEO-PE containing 10, 15, and 20 g/100 mL of GEO at CNC 0.50 g/100 mL were designated PE4, PE5, and PE6, respectively.

3.6. Rheological behavior of CNC-stabilized GEO-PE

Understanding of rheological properties is important from processing and ensuring quality of food materials. The shear viscosity curves of the Pickering emulsions stabilized by CNC are shown in Fig. 5. The CNC-stabilized GEO-PE exhibited shear-thinning characteristic of a pseudo - plastic fluid. The shear-thinning characteristic of PE may be attributable to the weak attractive contacts (hydrogen bonds) between particle surfaces (Wong et al., 2021). Due to insufficient hydrodynamic forces, oil droplets are retained together at low shear rates, resulting in high viscosity. Hydrodynamic forces break the interactions and deform the CNC network and oil droplets as the shear rate increases, resulting in a decrease in viscosity (de Lima et al., 2020; Souza et al., 2021). This behavior is directly related to the chemical bonding or interaction present between the emulsion components (Souza et al., 2021). Here, when the oil concentration was constant, the viscosity increased as the CNC concentration increased. Similar findings were reported in CNC-stabilized GEO-PE containing curcumin (Aw et al., 2022) and oleo gel-in-water PE stabilized by CNC (Qi et al., 2021). When the presence of rigid particles disrupts normal flow in a fluid, it causes higher energy loss due to friction (Meirelles et al., 2020). Consequently, the continuous phase may indirectly stabilize the droplets through entrapment. In the continuous phase, the CNC particles could potentially form a network, resulting in an increase in viscosity (Ng et al., 2022). Cellulose crystals seem to influence both the area at the interface, which affects the stability of emulsion (Meirelles et al., 2020).

Fig. 5.

Apparent shear viscosity of CNC stabilized GEO Pickering emulsion at different GEO and CNC concentrations. The CNC stabilized GEO-PE with 10, 15, and 20 g/100 mL of GEO at CNC 0.25 g/100 mL were designated PE1, PE2, and PE3, respectively. The CNC stabilized GEO-PE containing 10, 15, and 20 g/100 mL of GEO at CNC 0.50 g/100 mL were designated PE4, PE5, and PE6, respectively.

3.7. Storage stability of CNC-stabilized GEO-PE

An emulsion system is thermodynamically unstable and ultimately separates into immiscible oil and aqueous phases. Fig. 6 depicts the digital image (physical stability) of all prepared emulsions stabilized by CNC at week 0 and 8. The white color and homogeneity of the CNC-stabilized emulsions indicated that the oil was fully emulsified. All samples demonstrated resistance to coalescence, sedimentation, and creaming. No phase separation was observed in the emulsion samples over 8 weeks at 25 °C or 40 °C. Thus, the CNC-stabilized GEO-PE exhibited appreciably high stability (creaming index = 0). Previous research has also shown that CNC-stabilized emulsions showed low creaming index for 14 days at a wide range of pH indicating emulsion stability (Ng et al., 2022). CNC increased the stability of the emulsion system by raising the viscosity of the continuous phase which is in accordance with Stoke's law (Winuprasith and Suphantharika, 2015). CNC decreased interfacial tension and created a stable interfacial layer and at the same time increased the continuous phase viscosity both of which restrained the movement of oil droplets and prevented their coalescence (Kasiri and Fathi, 2018). The increased stability of emulsion stabilized by a higher concentration of CNC may be due to the formation of depletion-flocculated network in which a three-dimensional network of droplets and unabsorbed (surplus) CNC particles is expected to form (Winuprasith and Suphantharika, 2015). This viscoelastic network immobilizes droplets, inhibits the coalescence of the oil phase, and stabilizes the emulsion system (Ng et al., 2022). CNC can adsorb at the oil-water interface and helps minimize the coalescence and, thus, enhances emulsion stability (Varanasi et al., 2018). The zeta potentials for PE1, PE2, PE3, PE4, PE5, and PE6 were −47.80, −53.80, −55.23, −56.37, −57.53 and −60.00 mV, respectively. CNC is an anionic polysaccharide with abundant hydroxyl groups on its surface, providing highly negative charges. The absolute zeta potential of CNC-stabilized GEO-PE increased as the CNC and GEO contents increased. The higher the absolute value of zeta potential, the stronger the electrostatic interaction, which is favorable for the stability of emulsions. The greater negative zeta potential is associated with a greater tendency for droplet repulsion from each other and reduced coalescence (Hossain et al., 2021). The CNC increased GEO-PE stability in this study through the Pickering effect and incorporated with a three-dimensional network of CNC to entrap oil droplets with strong electrostatic repulsion between the oil droplets.

Fig. 6.

Visual appearance of CNC stabilized GEO-PE at different GEO and CNC concentrations during storage at 25 °C or 40 °C for 8 weeks. The CNC stabilized GEO-PE with 10, 15, and 20 g/100 mL of GEO at CNC 0.25 g/100 mL were designated PE1, PE2, and PE3, respectively. The CNC stabilized GEO-PE containing 10, 15, and 20 g/100 mL of GEO at CNC 0.50 g/100 mL were designated PE4, PE5, and PE6, respectively.

This study did not investigate the chemical stability of the CNC-stabilized GEO-PE; however, other studies have suggested that the oxidative stability of PE is greater than that of conventional emulsions. In PE, particles with good properties at the interface can be used, such as those that create a thicker interface, have intrinsic antioxidant ability, and have a positive droplet surface load; thus, by carefully selecting the particles (which repel positively charged pro-oxidants such as iron), greater oxidative stability can be achieved. By modulating the properties of interfacial layers (such as by increasing the thickness, antioxidant activity, charge, and selective surfactants), it is possible to enhance the oxidative stability of PE (Keramat et al., 2022). Factors influencing the rate of lipid oxidation in PE o/w emulsions include interfacial layers properties (physical properties, chemical properties, and electric charge), droplet size and pH. The electrical charge of emulsion droplets can have an impact on the removal and absorption of transition metal ions. A negative charge can increase lipid oxidation via attracting the transition metal ions to the interface (Keramat et al., 2022). Our zeta potential results reveal a strong negative zeta potential (−47 to −60 mV). This negative charge at the interface may attract transient metal cations such as ferrous iron, resulting in metal-mediated autooxidation of ginger oil's unsaturated terpenes. In addition, according to a study by Schröder et al. (2021), adding pineapple fibers to o/w emulsions causes them to become oxidatively unstable. Authors stated that pineapple typically contains a sizeable amount of vitamin C, which has the ability to scavenge free radicals and regenerate hydrogen-donating antioxidants. Elevated vitamin C content can also reduce ferric ions to ferrous iron, which is a more potent oxidation catalyst. This in turn may account for the rapid oxidation in the comparable emulsion evaluated in the authors' study.

4. Conclusion

CNC was extracted from pineapple peel and successfully used to prepare oil-in-water (O/W) Pickering emulsion (PE). The CNC primarily occupied the oil-water interface, covered the oil droplets, and produced a stable emulsion. The PEs of GEO stabilized by CNC were creamy white, homogenous, and remained stable for 8 weeks at ambient (25 °C) or elevated (40 °C) temperatures. The CNC produced from pineapple peel could be used to produce surfactant-free and stable O/W emulsions. The CNC improved the stability of the PE emulsion by increasing absolute zeta potential, emulsion viscosity and also tightly covering the oil-water interface. Formation of three-dimensional depletion-flocculated network non-adsorbing excess CNC particles were found to be responsible for the high stability of CNC stabilized PE emulsions. The viscoelastic network CNC particles immobilized the oil droplets and inhibited their coalescence. The potential applications of the CNC-stabilized O/W emulsion could be to fabricate edible egg-free mayonnaise-like Pickering emulsion, dairy products (ice cream, yogurt, and drinking yogurt), salad dressing, etc.

CRediT authorship contribution statement

Arissara Phosanam: Conceptualization, Methodology, Investigation, Formal analysis, Writing – original draft, Software, Data curation, Writing – review & editing. Juan Moreira: Validation, Investigation. Benu Adhikari: Conceptualization, Data curation. Achyut Adhikari: Resources, Project administration, Methodology, Validation, Draft review, editing, and final manuscript approval. Jack N. Losso: Resources, Methodology, Validation, Writing and Draft review, editing, and final manuscript approval.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following is the Supplementary data to this article:

Data availability

Data will be made available on request.