Ultrasound-assisted encapsulation of Pandan (Pandanus amaryllifolius) extract.

Pandan (Pandanus amaryllifolius) is commonly used as a food ingredient in Southeast Asia due to its delicious flavor, appetizing aroma and bright green colour. Pandan plant is uniquely found only in certain parts of the world. Despite its increasing popularity worldwide, its export market is limited by practical issues. One of the main problems for exporting Pandan to global market is its stability during transport. Due to the volatility of its active constituent, the functional properties of Pandan are lost during storage and shipment. In this study, we explored the ability of ultrasound processing technology to encapsulate the aromatic Pandan extract using lysozyme or chitosan as a shell material. 20 kHz ultrasonicator was used to encapsulate the pandan extract at 150 W of applied power. Two parameters, the ultrasonic probe tip and the core-to-shell ratio were varied to control the properties of the encapsulates. The diameters of the probe tip used were 0.3 and 1.0 cm. The core-to-shell volume ratios used were 1:160 and 1:40. The size distribution and the stability of the synthesized microspheres were characterized to understand and explore the possible parameters variation impact. Both size and size distribution of the microspheres were found to be influenced by the parameters varied to certain extent. The results showed that the mean size of the microspheres was generally smallest when using 1 cm probe tip with lower core-to-shell volume ratio but largest when using the 3 mm tip with higher core-to-shell volume ratio. This indicates that the sonication parameters could be fine-tuned to achieve the encapsulation of Pandan extract for storage and export. The pandan-encapsulated microspheres were also found to be stable during storage at least for one month.

Introduction

Encapsulation is a technique to entrap substance or mixture of materials within another material [1], [2], [3]. Inspired by cell model, the concept of encapsulation is invented as a protection of labile functional materials from undesired conditions [4]. The encapsulated substance is termed as core material, active agent, or internal phase, whilst the encapsulating substance is referred to as shell, carrier substance, coating material or external phase [5]. The end-product is referred to as core–shell microcapsules or microspheres. They are also known as microspheres due to their characteristic spherical shape and useful size range at about 0.1–10 µm. Other than core–shell type, the encapsulation might also exist in matrix type of encapsulates [6], [7]. Amongst the materials that is generally recognized as safely edible and widely used for encapsulation in food applications is biomacromolecules [8] such as proteins, polysaccharides, lipids, and carbohydrate polymers [8], [9], [10]. The core–shell microspheres can be used to encapsulate and deliver value-added food ingredients such as flavours and vitamins.

Several techniques are available to encapsulate food compounds such as spray drying, fluid-bed coating, spray chilling, melt injection/extrusion, freeze-drying and emulsification [1], [6]. Spray drying is one of the most used encapsulation techniques especially for large scale production in food industry. Spray-drying technique in general is carried out by dispersing the core material in an aqueous solution containing the coating material, followed by atomization/spraying through a nozzle into a hot chamber for dehydration [3], [6]. The spray drying technique provides the most economical method due to its short time required for drying, thus less energy is used, lowering the production cost, and simplifying the operation [11]. On the other hand, there are several disadvantages of this technique such as the broad size distribution of the particles due to non-uniform conditions in drying chamber depending on the type of the atomizer used [12]. This compromises the ability to control the particle size and the storage stability could be affected by the loss of some food compound during or after spray-drying process [1], [13]. In addition, the core material is subjected to heating that may affect their functional properties. Freeze-drying is another technique used for encapsulation which can be performed by dissolving the core and the coating materials in water and followed by freezing the sample at low temperatures. The sample is then dried by sublimation of the water under vacuum. After drying, it is optional to grind the sample obtained into fine powder [6]. The disadvantages of the freeze-drying are the need of high energy and longer time required for processing [1], [6]. In conventional methods, the procedure involved in generating core–shell materials is generally more complex. In most cases, external reagents such as emulsifiers and chemical crosslinkers are required. In many cases, several time-consuming routines with repetitive reaction steps are involved. The stability of the core–shell materials, synthesized by conventional methods, were also reported to be poor. The synthetic procedures often result in the denaturation of the precursor materials.

Ultrasonic technology is found to be effective in creating core–shell structures with specific physical and functional properties. Several studies are available on the ultrasonic encapsulation to synthesis biofunctional core–shell materials [9], [14], [15], [16]. The sonochemical synthesis of stable lysozyme-shelled microspheres was first reported in 2008 [17]. Zhou et al. reported on ultrasonic synthesis of lysozyme-shelled microspheres for the encapsulation of four different liquids, which are perfluorohexane, tetradecane, dodecane, and sunflower oil [18]. The properties of the encapsulated materials, such as molecular weight, boiling point, surface tension and density were found to impact on the physical characteristics such as the size and stability of the microspheres [18]. In sonochemical approaches, the method offers great versatility and potential, where both emulsification and cross-linking are achieved in a single step. The sonochemical experimental technique is relatively simpler, and stable core–shell materials can be generated within less than a minute of sonication. In addition, there is a greater flexibility in choosing different core and shell materials. The ultrasonic synthetic procedure does not affect the functional properties of core or shell materials and the stability of these materials is very high [9]. Ultrasound emerges to be one of promising technique that can overcome the difficulties in producing stable core–shell materials with controlled physical and biofunctional properties by conventional methods.

This study was an attempt to sono-encapsulate the extract from Pandan (Pandanus amaryllifolilius). Pandan is a tropical plant that belong to Pandanaceae in the screw pine family, native to Southeast Asia. There is a growing interest for Pandan in food industry especially in western countries due to its unique and pleasant taste and aroma, as well as its attractive bright green colour. The Pandan plant and some Pandan-based delicacies are shown in Fig. 1. However, the commercial use of this plant is far from its true potential because of the instability of its active compounds. The aromatic component of Pandan extract easily diffuses into air, thus losing its signature. There have been many attempts to entrap the volatile compound, so that the presence of Pandan aroma can be prolonged in food products. Recently, a group of researchers were able to encapsulate a major compound from Pandan extract, the 2-acetopyrolline (2-AP) through spray-chilling using microcapsules of 2-AP zinc chloride complex (2AP-ZnCh) coated with paraffin. The coated compound gives a pleasant aroma of “pop-corn like” in aromatic rice when it is exposed to heat during cooking [19].

Fig. 1

Pandan plant commonly found around houses, and some Pandan-based delicacies.

Despite the attempts to encapsulate Pandan extract due to its commercialization value as a food ingredient, synthetic Pandan is still being sold widely instead of the pure Pandan extract. This implies the lack of a successful (effective, stable, and cost-worthy) method in encapsulating Pandan. Such realization motivated us to explore a known, simple, and effective technique of encapsulation on the freshly extracted Pandan. In the current study, we have successfully used ultrasonic encapsulation process to encapsulate the Pandan extract. The main objective of this study was to synthesize core–shell microspheres of Pandan extract using lysozyme or chitosan as the shell material. The size and size distribution of the lysozyme and chitosan shelled Pandan microspheres were controlled by manipulating different experimental parameters, namely Pandan to lysozyme or chitosan ratio and the tip size of the ultrasonic probe. The study is also aimed at exploring the suitability of the Pandan extract compounds to be stable in the lysozyme or chitosan shell structure over time. Therefore, we have also evaluated the stability of the Pandan extract core-lysozyme/chitosan-shelled microspheres over a month of storage.

Materials and methods

Extraction of Pandan

Pandan plant as shown in Fig. 1 was obtained from Sitiawan, Perak, Malaysia. 1 kg of Pandan leaves were cleaned, cut into uniform size of 0.5 cm2 and infused in 2 L of vegetable oil for 24 h.

Pandan encapsulated microspheres

Lysozyme from chicken egg white, DL-dithiothreitol (≥98.0 %), tris(hydroxymethyl)aminomethane (≥99.8) and chitosan (low molecular weight) were purchased from Sigma-Aldrich Chemie GmbH. Sodium hydroxide and acetic acid (glacial) anhydrous (100 %) for analysis were purchased from Merck KGaA. In this study, all aqueous stock solutions were prepared using deionized water with resistivity of 10–15 MΩ/cm.

The encapsulation of Pandan extract using lysozyme was carried out as follows. Lysozyme solution was prepared by dissolving 0.2 g of lysozyme in 8 mL Tris-HCl buffer solution at pH 8.3. The partial denaturation of lysozyme was then performed by the addition of 100 mM DL-dithiothreitol (DTT). The partial denaturation was carried out for 2 min. The Pandan extract was layered on the surface of the lysozyme solution. The volume ratio of Pandan core to lysozyme solution was either 1:160 or 1:40. This was done by varying the Pandan core volume as either 50 or 200 µL in 8 mL of lysozyme solution. The tip of the ultrasonic probe (Qsonica sonicator Q500, Qsonica LLC, USA) was then positioned at the interface of Pandan extract-Lysozyme solution. The ultrasonic probes used were either with diameters of 0.3 or 1.0 cm. The microspheres were formed when the solution was sonicated for 30 s at 20 kHz with applied ultrasonic power set at 150 W. The power delivered were determined calorimetrically and used to calculate the power densities. The microspheres were collected, washed with deionized water to remove excess DTT and residual protein for at least five times. The microspheres were then collected after the mixture was left to settle for 4 to 6 h.

The encapsulation of Pandan extract using Chitosan was carried out as follows. Chitosan is known to be sparingly soluble, both in water and organic solvent. However, due to the existence of 1° amine functional group, it would undergo protonation in acidic solutions. Hence, chitosan could be made soluble as a polymer with amphiphilic properties in acidic conditions. Chitosan solution was prepared by dissolving 0.4 g of chitosan in 100 mM acetic acid solution. The ratio of Pandan as the core, to chitosan as the shell of the microsphere was 1:160 and 1:40, similar as described above using lysozyme. The Pandan was first layered on the surface of the chitosan solution with the total solution volume fixed at 8 mL. The tip of the ultrasonic probe was then positioned at the interface of Pandan-Chitosan solution. The sonication and washing procedures to form the microspheres are similar to the procedure described above.

Characterization

Pandan extract analysis

The crude Pandan extract was isolated and analysed using HPLC. The extract was dissolved in 15 mL of methanol (HPLC grade) and filtered through by 0.45 µm syringe filter. HPLC analysis was performed on Agilent’s HPLC system equipped with 1290 Infinity Binary pump and autosampler injector, Agilent Poroshell 120 C18 column (4.6 X 50 mm, 2.7 µm) and PDA detector. The sample volume was 20 µL and eluted using mobile phase of methanol: MiliQ water (50:50, v/v), and delivered at a flow rate of 0.7 mL/min. The UV detection was carried out at 300 nm and at temperature 25 °C. Other characterization and determination of the components in the extract is to be reported in another manuscript. The structures of the plant materials were observed under field emission scanning electron microscope (FESEM, Hitachi SU8220). Samples were uncoated and observed under 1 kV.

Pandan-shell encapsulates analysis

To characterize the lysozyme and chitosan microspheres loaded with Pandan, optical microscope (Leica, DMIL-LED) was used. For each of the experimental set, more than 400 microspheres from approximately 100 images were taken, measured, tabulated, and reported in this study. The size and size distribution were processed and analyzed by using ImageJ, the scientific image analysis software by NIH [20].

Results and discussion

Pandan is commonly found around residential areas in certain parts of Southeast Asia. A common practice is to pluck it fresh before cooking, and not to store it for days before use. This is to avoid the loss of its aromatic active ingredient, thus making it flavourful and aromatic to the food product it is infused into [21], [22]. In this study, the extraction process was mimicked according to the common practice for cooking. The fresh Pandan leaves were immediately cut into uniform pieces and soaked in vegetable oil to allow the solvent extraction process. The extraction of Pandan in other solvents were also studied for comparison purpose, and to be reported in another manuscript. For the encapsulation experiment reported in this manuscript, only the oil extracted Pandan was used.

The active ingredients of Pandan are usually found in and around the oil glands on its leaves. The oil glands can be clearly seen on the SEM images of Pandan leaf in Fig. 2 (a) below. Upon solvent extraction, most of the oil glands collapse by the release of the active components. This can be visibly seen in Fig. 2 (b).

Fig. 2

The image of Pandan leaf under SEM with 1000 times magnification (a) before and (b) after solvent extraction. The oil glands can clearly be seen on both images. The destruction of these oil glands during pretreatment will affect the extraction rate.

The majority of the essential oil is present in oil glands (Fig. 2) and be extracted using solvent extraction process. These oil glands may be physically damaged during the pretreatment of the plant sample such as cutting and grinding. In this study, the Pandan leaves were cut into uniform size of 0.5 cm2 for comparison purpose when other solvents were used. The detailed discussion on the extraction will be reported in a future communication.

The solvent extraction curve of Pandan is shown in Fig. 3. The trend observed in this study agrees with the typical solvent extraction process. The plot of the extraction curve represents the washing and diffusion stages by the solvent action [23]. The active compounds present in natural products are located at various sites of the plant matrices [24]. As the samples are soaked in the medium, the solvent pervades into the plant matrices and solubilized the easily accessible solutes. This is known as the washing stage. The affected solutes are those located near to the surface or originally adsorbed on the surface of the matrices. On the other hand, the solutes located in the interior of the plant matrices will only be diffused out by the solvent diffusion process [24]. It must be made clear that the two stages happen simultaneously. However, washing process dominates during the early stage of the extraction, evidenced by the significant increase of extract yield. The rate of extraction decreases as the washing process slows down, and diffusion process dominates. The pandan extract HPLC characteristic is shown in Fig. 4. 2-acetyl-1-pyrolline was found as the main compound responsible for the scent of the pandan [25], [26]. However, to maintain the traditional usage of pandan extract as a whole in food, the whole Pandan extract will be used in the encapsulation stage of this study. The HPLC of the Pandan extract after encapsulation also showed that there is no major chemical compounds modification by the effect sonication.

Fig. 3

The extraction curve plot for Pandan with washing process dominating at the beginning, and diffusion process dominating after. Inset: The washing process involves the action of solvent washing the compounds around the oil glands, and the diffusion process involves the diffusing action of solvent into the oil glands.

Fig. 4

The HPLC chromatogram of Pandan extract.

The synthesis of Pandan encapsulated microspheres

As stated earlier, the active compounds in Pandan extract, responsible for its high value as food ingredient, is very volatile, thus can easily be lost during storage. This is the main hurdle for its export potential, as well as the exploration of more food products that could be derived from it, especially where it is not freshly available. Therefore, in this study, we explore the encapsulation ability of Pandan extract using lysozyme and chitosan as the shell material, with the aid of ultrasound. Such specific investigation on the ultrasound-aided encapsulation of Pandan extract has never been reported before, therefore, we aim to (i) successfully encapsulate the Pandan extract, (ii) characterize the size and size distribution of the encapsulates, and (iii) investigate the stability of the encapsulates. Achieving these three objectives will help the potential use of ultrasonic method to improve the storage and delivery of such valuable, yet volatile plant products such as Pandan.

The formation of pandan encapsulated microspheres involves two important processes, namely ultrasonic emulsification, and shell formation [27]. In the emulsification process, the propagation of ultrasound will create an interfacial wave where the dispersed phase will be erupted into the continuous phase as large droplets. These droplets then undergo gradual breakdown due to the intense physical sheering effect generated by acoustic cavitation. The radical formation during acoustic cavitation plays a major role to crosslink the biopolymer molecules adsorbed at the droplet interface resulting in the formation of stable microspheres [28].

The formation of pandan-lysozyme microspheres starts with partial breakdown of the intramolecular disulphide bonds in the lysozyme compound by DTT. This process is needed to set off the adsorption of partially denatured lysozyme at the interface of oil. The reducing agent chemically breaks the disulphide bonds, resulting in free thiol groups, where intermolecular crosslinking is possible to create a microsphere. Upon sonication, new intermolecular disulphide linkage is formed between the lysozyme molecules with the aid of HO2 radicals generated during acoustic cavitation process [28]. The images of pandan-lysozyme microspheres observed under an optical microscope are shown in Fig. 5 (a) and (b).

Fig. 5

The images of pandan extract in (a) and (b) lysozyme microspheres and (c) and (d) chitosan microspheres taken by optical microscopy with 100 times magnification.

On the other hand, the formation of pandan-chitosan microspheres is supported by a three-steps mechanism. Prior to sonication, chitosan was dissolved in acidic medium such as acetic acid, introducing amphiphilicity in chitosan molecules. The amino group is protonated into ammonium ion and the acetate ion will counteract the electrostatic repulsive forces. Sonication results in emulsification and the microspheres will be initially stabilized by adsorbed chitosan molecules. Mettu et. al [29] reported that the microspheres are also further stabilized by covalent crosslinking and the existence of intra-intermolecular hydrogen bonds. The images of pandan-chitosan microspheres are shown in Fig. 5 (c) and (d).

Size distribution of the Pandan-lysozyme and Pandan-chitosan microspheres

The design of microspheres loaded with specific valuable materials vary according to the particular purpose of its application, aim of storage (stability) and means of delivery [30]. Therefore, the properties of interest to be characterized are commonly the size of the microspheres, their size distribution, and their stability as well as the release response [27], [31]. Such characteristics could be tailored by the variation of ultrasonic operating parameters such as the frequency [32], ultrasonic power [32], sonication time [33], and the core to shell ratio [34]. To possibly investigate the potential of Pandan-lysozyme and Pandan-chitosan microsphere as stable ingredient for food products, we explored the effect of (i) ultrasonic probe diameter of 0.3 and 1.0 cm and (ii) the volume ratio of core-to-shell materials of 1:160 and 1:40; for both shell materials, lysozyme and chitosan. Other parameters were kept constant throughout the experiments. The sonication frequency was fixed by using 20 kHz sonicator unit with power set at 150 W, and the sonication time was fixed at 30 s. The core material used is the oil extracted Pandan as elaborated in the section above.

The effect of tip diameter

The two probe tips used in this study were 0.3 cm and 1.0 cm in diameter. Though the power set on the sonicator was fixed at 150 W, the power delivered to the 8 mL sample was quantified calorimetrically. Based on the calorimetric experiment, 1.0 cm probe tip diameter was found to deliver a higher power to the system (897.49 W mL−1) as compared to the 0.3 cm probe tip diameter (792.03 W mL−1).

Using the same experimental set up as above, the Pandan:lysozyme and Pandan:chitosan microspheres were synthesized at 2 vol ratios of 1:160 and 1:40. The size of the microspheres were measured from the microscopic images and tabulated using ImageJ. The plots to compare the effect of tip diameter width are shown in Fig. 6. Fig. 6 (a) and (b) represent the Pandan in lysozyme microspheres at 1:160 and 1:40 vol ratio, respectively. Fig. 6 (c) and (d) represents the Pandan in chitosan microspheres at 1:160 and 1:40 vol ratio, respectively. For all 4 subfigures, the filled markers (● and ■) and solid line (▬) are for the experiments carried out with 0.3 cm diameter tip; and the hollow markers (○ and □) and the dashed lines ( − − −) are for experiments carried out with 1.0 cm diameter tip.

Fig. 6

The size distribution of Pandan:lysozyme microspheres with (a) volume ratio of 1:160 using 0.3 cm tip probe (●,▬) and 1.0 cm tip probe (○, - - -); (b) volume ratio of 1:40 using 0.3 cm tip probe (■, ▬) and 1.0 cm tip probe (□, - - -); and Pandan:chitosan microspheres with (c) volume ratio of 1:160 using 0.3 cm tip probe (●,▬) and 1.0 cm tip probe (○, - - -); and (d) volume ratio of 1:40 using 0.3 cm tip probe (■, ▬) and 1.0 cm tip probe (□, - - -).

Analyzing all plots in Fig. 6, it can be clearly seen that the size of microspheres synthesized from experiment using 0.3 cm tip diameter probe are in general larger than the microspheres synthesized from experiments using 1.0 cm tip diameter probe. The change is the most significant in the following order: Pandan:chitosan (1:40) > Pandan:chitosan (1:160) > Pandan: lysozyme (1:40) > Pandan: lysozyme (1:160) with changes of the microsphere diameter peak from 2.7 to 1.6 µm, 1.8 to 1.2 µm, 1.5 to 1.3 µm and 1.5 to 1.2 µm, respectively. Though the difference between each of experimental sets are not much, but the trend can still be clearly seen, and may serve as a guide for the expansion of the study in the future. It is also worthy to note that the impact of changing the probe tip diameter is more pronounced when chitosan was used as the shell material (Fig. 6 (c) and (d)) when compared to lysozyme. In both cases, the microspheres sizes decreased by 33.7% and 43.9% when changing the probe tip diameter from 0.3 to 1.0 cm at 1:160 and 1:40 vol ratio, respectively. At 1:160 Pandan:chitosan ratio, the mean values for the microsphere’s sizes are 1.93 ± 0.04 µm and 1.28 ± 0.02 µm for 0.3 and 1.0 cm probe tip diameter, respectively. At 1:40 Pandan:chitosan ratio, the mean values for the microsphere’s sizes are 2.6 ± 0.07 µm and 1.49 ± 0.04 µm for 0.3 and 1.0 cm probe tip diameter, respectively.

The decreasing size of most of the microspheres when the tip diameter is changed from 0.3 to 1 cm can be explained by the higher power over volume unit delivered as quantified by our calorimetric study. It is known that higher power generates more intense shear, which in results produces smaller particle size upon ultrasonic emulsification process [33]. Though in some cases, “over-processing” sonication may occur at high power. In this situation the higher sonication power density will result in the higher coalescence of cavitation bubble [35]. However, this was not observed with the range of power delivered within the experimental parameters of this study. The size distribution is also observed to be narrower when 1.0 cm tip probe is used. An example is at Pandan:chitosan ratio 1:40 (Fig. 6 (d)), the peak width changes from ∼ 0.5–7 µm to ∼ 1–3.5 µm when the probe tip diameter was changed from 0.3 cm to 1.0 cm. This indicates a better control of the microsphere sizes when larger probe tip was used. Analyzing each of the plots, again we can conclude that the change on the peak width is more significant for Pandan in chitosan samples. The further discussion on the impact of using different shell material will be presented later.

The effect of core-to-shell ratio

The second variation factor to be discussed in this study is the core-to-shell volume ratio. As mentioned before, the core material is the oil extracted Pandan as discussed in Section 3.1 and the shell material is either lysozyme or chitosan. The volume of the shell material was fixed at 8 mL. However, the Pandan volume used were varied at 50 µL or 200 µL, making the two core-to-shell ratios between Pandan and the shell material as 1:160 and 1:40. The size of the microspheres were measured from the microscopy images and tabulated as described before. The plots to compare the effect of core-to-shell ratio are shown in Fig. 7. For all 4 subfigures, the filled markers (● and ■) and solid line (▬) are for the experiments carried out at 1:160 core-to-shell volume ratio; and the hollow markers (○ and □) and the dashed lines ( − − −) are for experiments carried out at 1:40 core-to-shell volume ratio.

Fig. 7

The size distribution of Pandan:lysozyme microspheres using (a) 0.3 cm probe tip with core-to-shell ratio 1:160 (●,▬) and 1:40 (○, - - -); (b) 1.0 cm probe tip with core-to-shell ratio 1:160 (■, ▬) and 1:40 ((□, - - -); and Pandan:chitosan microspheres using (c) 0.3 cm probe tip with core-to-shell ratio 1:160 (●,▬) and 1:40 (○, - - -); and (d) 1.0 cm probe tip with core-to-shell ratio 1:160 (■, ▬) and 1:40 ((□, - - -).

Referring to Fig. 7 (a) and (b), the size distribution of Pandan-to-lysozyme microspheres show the primary peak ranging at 0.1–5.0 µm. Changing the Pandan-to-lysozyme ratio did not change neither the diameter size of the maximum number of microspheres nor the width of the plot distribution. For the microspheres created with 0.3 cm probe diameter, the mean size for pandan extract in lysozyme microspheres are 1.46 ± 0.05 µm and 1.58 ± 0.06 µm for the volume ratio of 1:160 and 1:40, respectively. Meanwhile the mean sizes for microspheres created with 1.0 cm probe diameter are 1.22 ± 0.04 µm and 1.28 ± 0.03 µm for volume ratio of 1:160 and 1:40, respectively. The result show that, there is no significant effect of different volume ratio difference as the size distributions for both sets of volume ratio were relatively similar with small differences of about 0.06–0.12 µm only, when it was sonicated with the same size of probe diameter.

However, changing the Pandan-to-chitosan ratio from 1:160 to 1:40 results in the size of majority of the microspheres to change from 1.9 µm to 2.7 µm when 0.3 tip prob was used (Fig. 7 (c)); and from 1.3 µm to 1.5 µm when 1.0 cm tip probe was used (Fig. 7 (d)). Such observation has been reported before where the ratio of dispersed phase (core material) to continuous phase (shell material) affected the efficacy of the encapsulation [34]. During sonication, the core material will be dispersed as microdroplet, and undergo continuous breaking and coalescence due to the strong force. The smaller the volume ratio, the more space of dispersion in the aqueous medium prior to the encapsulation process take effect, which will create the smaller size of microspheres. In both Figures (c) and (d) involving chitosan, it can also be seen that the peak width increases from 1:160 to 1:40. This indicates better control of most microsphere sizes when more Pandan is incorporated within the chitosan.

The effect of using different shell structures

Referring to Fig. 6, Fig. 7, the Pandan:lysozyme and Pandan:chitosan microspheres exhibit different responses towards the two parameters varied – the tip probe diameter and the core-to-shell ratios. From Fig. 6 (a) and (b); and 7 (a) and (b), the size distribution of the Pandan:lysozyme microspheres exhibited a peak around 1.5 µm, and remain almost unchanged upon variation. This indicates that changing the probe tip size as well as the Pandan:lysozyme volume ratio do not have a significant effect on the size and size distribution of the microspheres. This could be a remarkable advantage when it comes to commercial industries. In such situation where the food product is processed in bulk, it is very common for the processing factories to face issues with precision in maintaining certain parameters to ensure product quality control. Some examples of the parameters are the two presented in this study – the power delivered to the system (by the variation of probe tip diameter) and the exact core-to-shell ratio. However, the need to precisely control the variation in the case where lysozyme is the shell structure is not critical, thus results in a more feasible process. The unchanged size and size distribution shown in Fig. 7 (a) and (b) also indicates that more Pandan core material could be processed using lower amount of lysozyme. This will be cost and energy saving. On the other hand, a different observation brings to a different analysis for microspheres involving chitosan shell. From Fig. 6 (c) and (d); and 7 (c) and (d), it can be seen that changing any of the two parameters changed the size of the majority of the microspheres; as well as changing the width of the size distribution. This will allow a more specific control when certain microsphere size is required for a specific product or application. This opens up the possibility for a better control, and the optimization of Pandan product. However, understanding this further will require an in-depth study for this specific Pandan-chitosan system.

The impact of changing the probe tip and core-to-shell ratios is found to be more in chitosan than lysozyme possibly due the lower cross-linking degree and thinner shell formation of chitosan microspheres. Similar observation was reported by Vong et al [36] where they found the size difference when using BSA and lysozyme as shell materials. The mechanism of microsphere formation involves the action of ultrasonic emulsification and shell formation. During emulsification process, one of the factors that will influence the efficiency of the process is the viscosity of the continuous phase [37]. The difference in the viscosity of the lysozyme and chitosan as continuous phase might affect the size of dispersed droplet during the onset of ultrasonic emulsification process. Then, the process of shell formation might be influenced by the nature of the shell material. In both cases of lysozyme and chitosan as the shell, the microspheres are strengthened by crosslinked shell formation. Typically, lysozyme microspheres are linked by stronger intermolecular crosslinked shell, whereby chitosan microsphere with a relatively weaker covalent crosslinked shell. The chitosan microspheres without additional crosslinker was found to be thinner than the one with external crosslinker [38]. Therefore, the loosely crosslinked system may cause the final microspheres to become more susceptible towards the changing parameter.

Stability of the microspheres

Stability is one of the most valued characteristics in a commercialized food product. This is critical when it involves a highly volatile ingredient such as Pandan extract. Therefore, we were interested in understanding the shell ability of the microspheres to retain and/or release its core material upon storage. The information of structural changes of the microspheres as a function of time is essential to see the extent of effects of the operating parameters on the stability of the newly created microspheres. In addition, it allows further analysis on the benefits as well as to find out the possibility on how to improve the encapsulation as well as the release in the future. For example, microspheres which are prone to structural changes can be used as temporary storage, and the encapsulated materials such as aroma will be released at certain rate upon desired application. Another example is when the microspheres is intended to retain its structural characteristic for longer storage and protection to the highly valued core materials. This is especially critical for exportation. Therefore, with the success in encapsulating valuable material such as pandan, it is equally important to study the stability of the encapsulates.

In this section, the shelf-life of the microspheres as a function of storage time is evaluated. About 100 images were taken to examine the stability of pandan microspheres after a month of storage. Some images are shown in Fig. 8. The size distribution of pandan extract in lysozyme and chitosan microspheres are represented in Fig. 9 and Fig. 10, respectively.

Fig. 8

The images of Pandan in lysozyme encapsulates (a) upon encapsulation, (b) after a month; and Pandan in chitosan (c) upon encapsulation and (d) after a month.

Fig. 9

Pandan-lysozyme microspheres size distribution as a function of storage time. a) volume ratio of 1:160 sonicated with 0.3 cm probe diameter; b) volume ratio of 1:160 sonicated with 1.0 cm probe diameter; c) volume ratio of 1:40 sonicated with 0.3 cm probe diameter; d) volume ratio of 1:40 sonicated with 1.0 cm probe diameter.

Fig. 10

Pandan:chitosan microspheres size distribution as a function of storage time. a) volume ratio of 1:160 sonicated with 0.3 cm probe diameter; b) volume ratio of 1:160 sonicated with 1.0 cm probe diameter; c) volume ratio of 1:40 sonicated with 0.3 cm probe diameter; d) volume ratio of 1:40 sonicated with 1.0 cm probe diameter.

From Fig. 9, it can clearly be seen that the Pandan in lysozyme microspheres were stable after a month. The same goes for Pandan in chitosan microspheres synthesized using 1.0 cm probe tip diameter as shown in Fig. 10 (b) and (d). However, some of the Pandan in chitosan microspheres synthesized using 0.3 cm probe tip diameter showed some instability as seen by the minor change of microspheres to smaller sizes. This results in a decrease of the number of larger size microspheres, and increase of the number of smaller size microspheres, especially at around 1.5 and 2.0 µm for 1:160 and 1:40 Pandan-to-chitosan ratio, respectively (Fig. 10 (a) and (c)).

Conclusions

In this study, Pandan extract-encapsulated microspheres were successfully synthesized by a sonochemical method. The size distribution and stability of microspheres were investigated by the variation of two experimental parameters, namely the ultrasonic probe tip diameter and the core-to-shell volume ratio. Both parameters influenced the size and size distribution of the pandan extract-encapsulated microspheres to certain extent. Smaller pandan extract-encapsulated microspheres ranging between 1.2 and 1.6 µm were obtained when lysozyme was used as the shell material; and slightly wider range between 1.3 and 2.7 µm was obtained when chitosan was used as the shell material. The mean sizes of the microspheres were generally smaller when 1.0 cm probe tip and 1:160 core-to-shell volume ratio were used, but larger when 0.3 cm probe tip and 1:40 core-to-shell volume ratio were used. In terms of stability, all pandan-encapsulated microspheres showed great stability after a month, except for a minor size reduction for Pandan in chitosan microspheres synthesized using 0.3 cm probe tip diameter. Understanding the responses of the size and size distribution of both Pandan-lysozyme and Pandan-chitosan microspheres upon sonication parameters is important, especially for their use in food industry.

CRediT authorship contribution statement

Noridayu Omer: Investigation, Methodology, Formal analysis, Writing – original draft. Choo Yeun-Mun: Conceptualization, Resources, Supervision, Funding acquisition. Noraini Ahmad: Resources, Funding acquisition. Nor Saadah Mohd Yusof: Conceptualization, Resources, Project administration, Supervision, Funding acquisition, Writing – review & editing.

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.