Optimization of drying parameters in the microencapsulation of volatile oil from Spiranthera odoratissima leaves Otimização dos parâmetros de secagem na microencapsulação do óleo volátil de folhas de Spiranthera odoratissima Optimización de los parámetros de secado en la microencapsulación de aceite volátil de hojas de Spiranthera odoratissima

Spiranthera odoratissima A. St.-Hil. (Rutaceae), "manacá", is popularly used for head, muscle aches, rheumatism and, stomach, uterine, renal and liver disorders. The aims of this study were to investigate the physico-chemical and morphological properties of microencapsulated powder of volatile oil from S. odoratissima leaves, optimize the drying process and verify the influence of drying parameters on microencapsulation by spray-drying. The volatile oils from leaves were extracting by hydrodistillation in a Clevenger type apparatus and analyzed by GC/MS. The emulsions were prepared and spray-dried. Box-Behnken experimental model was used for optimize the effects of drying parameters on the encapsulation responses. The β-caryophyllene content in the microcapsules was determined by HPLC. The results suggest that the best operational conditions for the atomization drying of S. odoratissima volatile oil were inlet temperature of 158°C, feed flow of 0.25L/h and drying nozzle diameter of 0.7mm. These results reveal the technological potential of the microcapsules obtained from S. odoratissima volatile oils.

Volatile oil may undergo degradation and/or volatilize when exposed to environmental agents such as heat, light, moisture and oxygen (Bakry, et al., 2016;Turasan, et al., 2015). Due to the increasing demand for volatile oils in the pharmaceutical, food (Subtil, et al., 2014) and cosmetics industries, it is interesting to develop technological strategies such as microencapsulation (ME), which are able to maintain their chemical properties in face of the mentioned factors previously (Bakry, et al., 2016;Rosas, et al., 2017).
The microencapsulation by spray-drying has the advantages of low cost, good volatiles retention and stability of the finished product, besides the flexibility and fastness of the process (Reineccius, 2004). This technique guarantees a satisfactory yield in the drying process, high efficiency encapsulation (EE), high content of microencapsulated material, regardless of the capacity of the equipment and heat sensitivity (Marques, et al., 2016) and biological and pharmacological potential of volatile oils (Asbahani, et al., 2015;Asensio, et al., 2017).
The properties of wall materials, formulation parameters and drying conditions involving the ME process can affect the final characteristics of microencapsulated products (Ribeiro, et al., 2015). These are important conditions in the spray-drying drying process: the air inlet temperature (Frascareli, et al., 2012), the feed flow (Pires & Pena, 2017) and the diameter of the spray nozzle (Alves, et al., 2014). The use of response surface methodology (RSM) allows to verify the effect of the drying parameters on the properties of the obtained material (Baş & Boyaci, 2007).
The aims of this study were to investigate the physico-chemical and morphological properties of microencapsulated powder of volatile oil from S. odoratissima leaves, optimize the drying process and verify the influence of drying parameters on microencapsulation by spray-drying.

Pant material
The Spiranthera odoratissima A. St.-Hil. (Rutaceae) leaves were collected in 2016 (November and December) from approximately 50 individuals, in Aparecida de Goiânia, Goiás, Brazil (16º45'45.2'' S, 49º07'06.8'' W and at an elevation of 762 m above sea level). Prof. Dr. José Realino de Paula identified the botanical material. A voucher specimen was deposited at the Herbarium of the Federal University of Goiás/Brazil under registration 60010. The leaves were air dried at room temperature for three days, triturated using commercial crusher (Skymsen, LS-08MB-N) immediately prior to the extractionof the volatile oil.

Extraction and analysis of volatiles oils by GC/MS
115g of the powdered material (leaves collected in the two months) was subjected to hydro-distillation for three hours in a Clevenger type apparatus. The VO obtained was dried with anhydrous Na2SO4 and kept at -18 °C until analysis. The successive extractions of VO were collected in a single sample and then analyzed by gas chromatography coupled to mass spectrometry (GC/MS) in a QP2010 Plus (Shimadzu Corporation, Japão), equipment with a linear quadrupole mass detector, DB-5MS chromatography column (30 m x 0.25 mm, 0.25 μm); 5% phenylmethylpolysiloxane (Agilent J & W GC columns, USA). Helium was used as carrier gas at a flow rate of 1ml min. The injector and interface temperatures were 225 °C and 240 °C, respectively, with a 1:20 split ratio. The injection volume was 1 μL sample in hexane (1:5) and the oven temperature program consisted of a ramp of 60-240 ºC/3 ºC/min; 240-280 °C/10 °C/min, and final temperature set at 280 °C for 8 min. The temperature of the ion source 240 °C and electron impact ionization, 70 eV; the acquired mass range was 40-350 m/z. The chemical constituents of VO were identified by comparison of its retention index (RI), as well as of its mass spectra described in the literature. Retention indices were obtained with reference to a linear sequence of C7-C30 n-alkanes (Sigma, USA) (Adams, 2007). Research, Society andDevelopment, v. 10, n. 4, e57510414322, 2021 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v10i4.14322 4

Preparation of emulsion
16 emulsions were prepared with VO, arabic gum (AG) (Sigma-Aldrich) and maltodextrin dextrose (MD) equivalent 4.0-7.0 (Sigma-Aldrich) under constant magnetic stirring until complete solubilization. The 16 emulsions were prepared in a proportion VO: AG: MD (w/w) (1:2.7:1.5). The ratio of volatile oil /wall material was 0.25, the solids content ranged 32.8 to 34.59% and the VO percentage ranged 7.75 to 9.14%. After incorporation of the wall material, the blend was homogenized in ULTRATURRAX, IKA® brand and digital T25 model, operating at 15.000 rpm for 10min and spray-drying was performed.

Spray-drying process
Spray drying was performed (Laboratory of Agricultural Engineering -CCET/UEG, Anápolis/GO) using spray-dryer LABMAQ model MSD 1.0, equipped with cylindrical camera of 50 cm of height by 9 cm of diameter; double fluid type conical sprinkler and feed system made by a peristaltic pump. The emulsions were atomized, dried and the microcapsules obtained were packed in an amber bottle at room temperature in a desiccator until further analysis.

Box-Behnken experimental design
The Box-Behnken experimental design was employed to optimize and evaluate main effects, interation effects and quadratic effects of the process variables on the quality of microcapsules. A design 33 was used to investigating quadratic response surfaces and obtaining second order polynomial models for optimization. The parameters investigated and their levels were: drying temperature (DT; 145-175 ºC), emulsion feed rate (FF; 0.25-0.35 L/h) and nozzle diameter (ND; 0.7-1.2 mm).
Yield of the drying process (YP), volatile oil retention on the surface of the microcapsules, encapsulation efficiency (EE), moisture, water activity (Aw), apparently density, β-caryophyllene (CY) content and relative percentage of the major chemical compounds in VO recovered from microcapsules were evaluated as response variables.
To optimize the spray-drying drying process, the desirability function method was used to find the optimum conditions of the analyzed responses (Derringer & Suich, 1980;Battista, et al., 2017).
The individual desirability functions (Equation 2) from the considered responses were combined to obtain the overall desirability D, defined as the geometric average of the individual desirability: The software Statistica 7 was used for analyses. Only the factors with p< 0.05 were considered.

Physico-chemical analysis
The water activity (Aw) of the microcapsules was performed by direct reading in a water activity analyzer of the AQUALAB brand, Water Activity Analyzer model. The apparently density (da) (g/cm) was determined by the test tube method according to.
[40] The moisture content of the microcapsules was performed by the gravimetric method in scale with OHASUS MB-35 halogen lamp. In petri dish, 1.0g of the powder was heated to 105 ºC until constant weight. The percentage of moisture was calculated on a wet basis. All analyzes were performed in triplicate.

Encapsulation Efficiency
The EE was determined by the fraction of oil encapsulated on the amount of total oil (Equation 3) (Tan, et al., 2005).
Where EE is the encapsulation efficiency, VOT is the amount of total oil and VOS is the amount of surface (unencapsulated) oil on the surface of the microcapsules. The non-encapsulated VO present on the particle surface was determined with 5.0 g of microcapsules in a 125 ml capped flask and 20 ml of hexane added, the mixture stirred for 10 min at room temperature. The mixture was filtered on Whatman filter paper number 1 and the retained microcapsules were washed with 10 ml of hexane three times. The microcapsules were then oven dried at 50 ° C, weighed to a constant mass. The mass of unencapsulated VO was calculated by the difference in mass of the microcapsules before and after washing with hexane. The amount of total oil in the encapsulated products was determined by hydrodistillation of 5.0 g of the encapsulated powder for 3 h in Clevenger type apparatus. After extraction, the volume of VO was measured in the graduated tube of the apparatus itself. The total mass of the oil was calculated from the specific mass and volume (Tan, et al., 2005;Bae & Lee, 2008).

Volatile oil retention
The percent retention of VO in the microcapsules was determined in relation to the blend, prior to emulsification. The calculation was obtained by the ratio of the total oil quantified in the particles after drying by spray drying with the total oil initially added to the emulsion preparation, according to Equation 4 (Frascareli et al., 2012;Costa, et al., 2013).

Determination of the β-caryophyllene content in the VO of the of S. odoratissima leaves and of the microcapsules by HPLC
The quantification of β-caryophyllene in the VO of the microcapsules was performed according to the method developed 6 methanol to the final volume. The mixture was subjected to ultrasonic extraction for 45 min. The solution was filtered through Whatman number 1 Filter Paper and Millex1 syringe filters (0.45 mm pore size) and then injected into the HPLC.

Morphological analysis of the microcapsules
For the morphological analysis, the microcapsule samples were fixed in metallic specimen holders (stubs) in carbon double-sided adhesive tape. It was then covered with a thin layer of gold metallic alloy in a Denton Vacuum sputter, Desck V and the time of gold deposition on the sample was 2 min., generating a layer of approximately 250 A. After metallization, the samples were observed on a JEOL® scanning electron microscope (JVM) model JSM-6610, equipped with EDS, Thermo scientific NSS Spectral Imaging. The analysis was performed at the High Resolution Microscopy Multiuser Laboratory (LabMic) at the Institute of Physics of the Federal University of Goiás (IF/UFG).
The particle size distribution analysis was performed using the ImageJ 1.50 ie software represented by the frequency of the intervals using Equation 5, where d is the diameter of the microcapsules (μm) and A is the area of the microcapsules (μm).

Statistical analysis of experimental design
The effects of the drying temperature, feed flow (FF) and nozzle diameter (ND) variables on the dry matter yield (YP), encapsulation efficiency (EE), retention of total volatile oil (RO), and β-caryophyllene content (CY) responses were tested for adequacy of analysis of variance (ANOVA).

Scanning Electron Microscopy (SEM)
The effects of the independent variables drying temperature, feed flow (FF), nozzle diameter (ND) and their interactions on the dry matter yield (YP), retention of total VO (RO) and β-caryophyllene content (CY) responses can be visualized on the surface response graphs. In relation drying temperature and feed flow ( Figure 2A) the best dry matter yield was found in the central point. In the relationship between drying temperature and nozzle diameter ( Figure 2B) the best dry matter yield was obtained with lower values of nozzle diameter over a wide temperature range and with higher nozzle diameter and higher temperatures. There was no statistical relationship between feed flow and nozzle diameter. Source: Authors.
The increase in retention of total VO was observed for smaller values of nozzle diameter in a wide range of the drying temperature domain, with emphasis on the region of the graph ( Figure 3A) with intermediate temperature, obtaining microcapsules with 61.57% retention of total VO (sample 9, Table 2). However, the association of higher values of drying temperature and nozzle diameter produced microcapsules with a high retention of total VO of 55.02% (sample 8, Table 2). For the ratio of feed flow and nozzle diameter ( Figure 3B), the variable that had a significant effect on the retention of total VO was the nozzle diameter, with lower levels of nozzle diameter associated with intermediate feed flow values, we obtained rates highest retention of total VO of 52.49% (sample 5, Table 2). No significant interaction was observed between the feed flow and the nozzle diameter.
In the interaction of drying temperature and feed flow ( Figure 4A), the highest levels of β-karyophylene were observed for lower values of drying temperature and higher values of feed flow. In the interaction between drying temperature and nozzle diameter ( Figure 4B), the lower levels of drying temperature and nozzle diameter resulted in higher values of β-karyophylene content and in the interaction between the feed flow and the nozzle diameter, at the lower levels of these two factors, the higher the value of β-karyophylene content ( Figure 4C).

Optimization of microencapsulation
The optimal conditions of the drying process for the VO microencapsulation of S. odoratissima leaves were obtained based on the Desirable function D represented best solutions for the system of the dry matter yield, retention of total VO and βcaryophyllene content responses (Table 3)

Experimental verification
The verification experiments (in triplicate) had been performed for the optimum point mathematically determined in the optimization process (D = 0.96): dry matter yield (20.3±0.23%), retention of total VO (31.27±3.75%) and β-caryophyllene content (7.46±0.31%). Thus, the second order models were adequate to describe the influence of the operational variables employed in the mathematical model. The others parameters analyzed were disregarded because they were out of point mathematically determined in the optimization process: EE (64.03±10.14), humidity (4.01±0.70), Aw (0.15±0.01) and da (0.23±0.03).

Morphological of the microcapsules
The scanning electron microscopy (

Diameter of the microcapsules(µm)
The α-muurolol and α-cadinol compounds were identified only in the VO of S. odoratissima. These differences can be attributed to the spray-drying process (Asensio, et al., 2017), suggesting that the drying parameters applied to the ME of volatile oil of S. odoratissima leaves caused loss by volatilization, thermal degradation of these constituents or formation of unidentified artifacts (Alves, et al., 2014). According to Gharsallaoui, et al. (2007) and Ribeiro, et al. (2015), the loss of volatiles during ME, occurs predominantly in the initial drying stage.
In the present work a previous screening was done in search of the best proportions of wall material (GA:MD), wall and core material (WM: VO), VO percentage and solids content. The microcapsules that presented the best results for the encapsulation efficiency, retention of total volatile oil, dry matter yield, Aw, particle size distribution and morphological aspect were selected for the experimental design in the drying process.
For microencapsulation, a mixture of gum arabic (GA) and maltodextrin (MD) was selected because of the pleasant taste, low cost, besides being a polymer matrix that produces microcapsules with high encapsulation efficiency (Bora, et al., 2018;Mahdavi, et al., 2016).
Studies have shown the importance of drying factors to obtain microcapsules with physicochemical, chemical and morphological properties (Singh & Van, 2016) within quality standards for use in the food and pharmaceutical industries. The ranges of factors selected for experimental design (temperature, feed flow and spray nozzle diameter), were based on preliminary investigations in the literature, considering studies with VO of Copaifera multijuga (Dias, et al., 2012) and Cinnamomum zeylanicum (Felix, et al., 2017;Hermanto, et al., 2016), Cymbopogon citratus (Ribeiro, et al., 2015).
Following completion of the experiment, it was found that the water activity (Aw) in the microcapsules ranged from 0.17±0.03 to 0.28±0.02. The value of Aw <0.3 is ideal for stability, as this low amount of water on the surface of the microcapsules becomes unavailable and consequently increases the shelf life of the product. Low Aw values are a prerequisite for the commercial production of powders with good characteristics such as high flow capacity, low viscosity and particle agglomeration (Behboudi, et al., 2013). Alves, et al. (2014) obtained VO microcapsules from fruits of Pterodon emarginatus with Aw values close to 0.2.
The moisture content for VO the S. odoratissima microcapsules varied from 1.74±0.22 to 4.40±0.28%. Close values were found in VO microcapsules from Rosmarinus officinalis, with levels varying from 0.26 to 3.16% (Fernandes, et al., 2014) and 1.10 a 3.56% . The moisture contained in the microcapsules represents low efficiency in the removal of solvents during drying and is directly related to DT (Oliveira & Petrovick, 2010). The microcapsules with lower moisture content (1.74%) were obtained with higher drying temperature (175ºC) and justified by the greater heat transfer to the particle, favoring the evaporation of water (Daza, et al., 2016). There was higher moisture content in the microcapsules when the feed flow was higher. This occurred because the contact time of the emulsion with the drying air was not sufficient to cause evaporation of the water by virtue of the high amount of product to be dried (Fernandes, et al., 2014).
The encapsulation efficiency ranged from 63.24±12.73 to 80.21±3.76%, considering that of these values, superficial VO of the microcapsules was discarded. The VO present in the microcapsule indicates the efficiency of the spray-drying encapsulation process Martins, et al., 2014). The surface VO oxidizes and may compromise the quality of microcapsules (Edris, et al., 2016). The encapsulation efficiency of sample 4 (80.21%) of S. odoratissima is in accordance with other studies such as in the microencapsulation of coffee oil (82.57%) (Frascareli, et al., 2012), Zingiber officinale Rosco oil (31.19%)  and Juglans regia L. oil (91.01%) (Shamaei, et al., 2017). Edris, et al. (2016) obtained 96.2% EE in Nigella sativa resin oil with 3.8% oil on the surface of the microcapsules.
After statistical evaluation, only dry matter yield, retention of total volatile oil and β-caryophyllene content presented significant responses as parameters to evaluate the quality of the microcapsules. This is relevant for phytopharmaceutical technology, which is challenging to produce microcapsules with desired content of available active compounds.
The β-caryophyllene content (CY) obtained in the VO of S. odoratissima leaves was 14.57±0.89% and the CY of the VO recovered from the microcapsules ranged from 4.18±0.47 to 9.23±0.63%. These values present index of degradation of the chemical marker varying from 36.65 to 71.31%. In the drying process, the volatility or degradation of some heat-sensitive chemical components may occur (Bakry, et al., 2016).
The yield of dry matter expresses the amount of powder recovered spray-dry by cyclone separation (Behboudi, et al., 2013). Both the feed flow and drying temperature impacted the recovery of microcapsules in the cyclone. The dry matter yield was maximized when the spray-dry was operated at intermediate temperature (160°C) and lower feed flow (0.25 L/h).
The higher concentration of AG in relation to MD and the solids content of the emulsion of 35% contributed to the good morphological appearance and encapsulation efficiency, producing smooth surface microcapsules with few cavities (depressions). According to Koç, et al. (2015) the vitreous properties of MD oppose the plastic properties of GA and together form spherical structures of smooth surfaces. Like the encapsulating material, feed flow can also affect the morphology of the microcapsules along with the airflow. These balanced factors guarantee the effective drying with elimination of the water, avoiding the roughness of the surface of the microcapsule.
The formation of rough surfaces in the microcapsules, according to Reineccius (2004) and Santana, et al. (2016) may occur depending on the drying rate and the cooling process of the microcapsules after swelling with the water vapor outlet.
In the present study, microcapsules with eventual breakage of the walls with an empty center (nucleus) were found, where the VO was trapped. According to Bakry, et al. (2016) and Martins, et al. (2014) this is a typical microcapsule structure of the reservoir type system. In addition, the small vacuoles on the walls suggest that droplets of the volatile oil were also trapped in the walls of the particles (Ramos, 2006).
In the drying process there is strong interaction between the parameters used in the microencapsulation. The size of the microcapsules is determined by the interaction of VO concentration, droplet size, drying temperature and feed flow (Singh & Van, 2016). A study by Littringer, et al. (2012) showed a decrease in microparticle size when an increase in flow rate was employed. The distribution of the VO microcapsule diameter of S. odoratissima leaves in the present study showed a higher percentage contained in the range of 0-3 μm (77.6%). According to Alves, et al. (2014), to be considered microcapsule, the particles in the form of powder should have diameters less than 30 μm. The size of the microcapsules is important to aid in their application as they may interfere with the texture, flavor and stability of the active principle (Comunian & Favaro, 2016), appearance and fluidity (Alves, et al., 2014;Fernandes, et al., 2014).
In the response surface methodology, the predicted model was adequate to describe the experimental domain and the reliability of the adjusted model. The optimum conditions found simultaneously for the response variables from the global Desirability function (D = 0.96) were: 20.31±0.23 for dry matter yield, 31.27±3.75 for RO and 7.46±0.31 for β-caryophyllene content, when drying temperature (158ºC), feed flow (0,25 L h -1 ) and nozzle diameter (0,7mm) were used.

Conclusion
In conclusion, the best operational conditions for the atomization drying of S. odoratissima volatile oil were inlet temperature of 158°C, feed flow of 0.25L/h and drying nozzle diameter of 0.7mm. The Box-Behnken experimental model was able to identify the best operational conditions for the atomization drying of the volatile oil of S. odoratissima, guaranteeing good dry matter yield, retention of total volatile oil and β-caryophyllene. These results reveal the technological potential of the microcapsules obtained from S. odoratissima volatile oils. In addition, the process developed adds value to the raw material studied. For future studies, it is suggested to test the analgesic, anti-inflammatory, and antimicrobial activities from the microcapsules with the volatile oil of S. odoratíssima in order to analyze whether there is potentiation of their actions.