Solvent-Free Microwave Extraction of Essential Oil and Hydrosol From Piper Longum Lour. Fruit and Evaluation of Antioxidant Activities

In this study, essential oil (EO) of Piper longum L. (PL) fruit was extracted by solvent-free microwave assisted extraction (SFME) and hydroditillation (HD) with hydrosol obtained as a by-product. Operational conditions for SFME were maximized by response surface methodology (RSM) combined with a Box-Behnken design (BBD). Results indicated that the maximum extraction rate of EO by SFME with the optimal parameters was 0.747%, which was signicantly higher than 0.496% of HD method. The EOs analyzed by GC-MS, presented 50 and 36 compounds constituting 95.99% and 98.52% of the total peak area obtained by SFME and HD, respectively. The principal components of the both oils were sesquiterpenes and aliphatics while SFME EO contained more bioactive compounds of terpenoids. The predominant components of SFME and HD hydrosols were eugenol (25.44% SFME; 22.82% HD) and linalool (14.10% SFME; 22.37% HD). Trans-isoeugenol, observed for the rst time in PL fruit, was presented in SFME hydrosol and took up 14.33% of all the detected compounds. Both the EO and hydrosol extracted by SFME possessed better antioxidant properties than those from HD method, which indicated that SFME was a superior approach to simultaneously obtain the two products. the Folic-Ciocalteu colorimetric method, as proposed by 22 with a slight modication. Sample solution (0.5 mL) added to 2.5 mL of 10-fold diluted Folin-Ciocalteu’s reagent. After an of 2.0 of 7.5% (w/w) sodium carbonate were added. The mixture reacted at room temperature for 30 min, and the absorbance was recorded at 765 nm attenuated with methanol to an absorbance of 0.700 ± 0.02 at 734 nm. After 100 μL of sample solution reacted with 1.0 mL of ABTS ∙ + solution for 30 min at room temperature, the absorbance at 734 nm would be recorded. The radical solutions added with the equal amount of the corresponding sample diluents were used as control and tests were carried out in triplicate. The percentage inhibition was calculated as (I%) = (A 0 -A)/A 0 ×100, where A 0 is the absorbance of control without adding antioxidants, and A represents the absorbance of the samples. The IC 50 value was calculated as the concentration of the sample needed to scavenge 50% of free radicals by graphing the linear curve of I% versus sample concentration. BHT (butylated hydroxytoluene) was used as a positive control compared with the essential oils and ascorbic acid as a positive control for the hydrosols.

2.2 Extraction of essential oils and hydrosols 2.2.1 Solvent-free microwave extraction SFME was performed in a laboratory microwave-accelerated reaction system as shown in Fig. 1 (MCR-3, KERUI, China). This is a (2450±50) MHz multimode microwave reactor with a maximum output power of 750 W delivered in 1 W increments. This instrument is equipped with an infrared temperature sensor, an electromagnetic stirrer, and a circulating water-cooling system. The microwave irradiation time, and irradiation power can be controlled by an electronic control panel. For SFME procedure, an aliquot of 100 g PL fruit powder was wetted before extraction by splashing certain percentage of water evenly and the wetted material was put into a ask to place into the reactor. Essential oil and hydrosol were simply separated with a built-in separatory funnel of the instrument. The isolated oil was dried by anhydrous sodium sulphate and weighed. The extraction yield of essential oil was calculated according to the equation given:

Single factor study of SFME
The single-factor study was carried out to evaluate the preliminary range of extraction variables including microwave irradiation power (X 1 ), microwave irradiation time (X 2 ), and percentage humidity of the matrix (X 3 ) in SFME.
Hundred grams dried sample was moistened prior extraction by splashing 60 mL water evenly to the material prior extraction to 60 percentage humidity of the matrix prior extraction. And then moistened fruit powder was placed in the reactor of different microwave power (150~750W) for 40 min.
Hundred grams dried sample was moistened by splashing 60 mL water evenly to the material prior extraction to 60 percentage humidity of the matrix prior extraction. And then moistened fruit powder was placed in the reactor of 600 W microwave irradiation power with different microwave irradiation time (20~60 min).
Hundred grams dried sample was moistened by splashing water evenly to the material prior extraction to get different percentages humidity of the matrix (40~90%). And then moistened fruit powder was placed in the reactor of 600 W microwave powers for 30 min.

Optimization of factors by BBD
On the basis of the single-factor test results, three factors at three levels Box-Behnken design (Design Expert software, Trial Version 8.0.6, Stat-Ease Inc., Minneapolis, MN) was employed to optimize the extraction conditions and to explore the effects of individual and the interaction between the factors on the essential oil yield. Table 1 listed the Box-Behnken design matrix, coded and real forms of each independent variable at three levels (+1, 0 and -l), the experimental value from the SFME performed, and the anticipated response obtained through calculation.
In order to determine the maximum point of response, experimental data was tted to a quadratic polynomial model and regression coe cients obtained. The non-linear quadratic model used in the response surface presented below: Where Y is the response variable, X i and X j were the independent values of variables, and β 0 , β i , β ii , β ij , n and ε were the model coe cients of the intercept, linear, quadratic, and interaction coe cients, the number of the independent variables and the experimental error, respectively.

Hydrodistillation (HD)
Conventional hydrodistillation apparatus (Clevenger-type apparatus) according to the European Pharmacopeia was used in this procedure. An aliquot of 300 g materials was hydrodistilled with 1.5 L of water in a glass reaction ask for 6 h (until no more essential oil obtained). The essential oil was collected, dried under anhydrous sodium sulphate, and stored at 4 °C until further analysis. Each extraction was done at least three times, and a standard deviation was calculated.

Chemical analysis
GC-MS analysis of the essential oil were carried out on an Aglient HP 7890A-5975C GC-MS, equipped with a 7693A auto injector (Agilent Technologies Inc., PA, USA) and a capillary column (Agilent HP-5 MS, cross-linked 5% phenyl polysilphenylene-siloxane, 30 m × 0.25 mm × 0.25 μm lm thickness). The column oven temperature was programmed from an initial temperature of 100 °C (2 min) to 250 °C (20min) at 4 °C/min. The column temperature was programmed at 50 °C for 5 min and at 250 °C for 10 min. The injection temperature and ion source temperature were 250 and 230 °C, respectively. A 2 μL sample was injected in the split mode with a split ratio of 25:1. The ionizing energy was 70 eV. Alkanes (C 8 -C 40 ) were used as references in the calculation of retention indices (RI). The constituents of essential oils and hydrosols were identi ed by comparison of their retention indices and mass spectra with the NIST Mass Spectral Library (Agilent Technologies Inc., PA, USA).

Total phenolic content of essential oils and hydrosols
The total phenolic contents of EOs and hydrosols were determined according to the Folic-Ciocalteu colorimetric method, as proposed by 22 with a slight modi cation. Sample solution (0.5 mL) was added to 2.5 mL of 10-fold diluted Folin-Ciocalteu's reagent. After an interval of 5 min, 2.0 mL of 7.5% (w/w) sodium carbonate were added. The mixture reacted at room temperature for 30 min, and the absorbance was recorded at 765 nm (Libra S35 UV/Vis-spectrophotometer, Biochrom Ltd., Cambridge, UK). Distilled water was invoked as a blank and gallic acid was used as the standard. The same procedure was repeated for all standard gallic acid solutions (0~50 μg/mL) to get a standard curve and the results were expressed as milligrams of gallic acid equivalents (GAE) per gram of essential oil (mg GAE/g) or microgram per milliliter of hydrosol (μg/mL). The analyses were repeated in triplicate and standard deviation was calculated. A standard curve was depicted by the following equation: Gallic acid concentration (μg/mL) = 82.37×A 765 nm -2.0242 , R 2 =0.9977.
2.5 Analysis of antioxidant activities 2.5.1 DPPH and ABTS + radicals scavenging A Free radical scavenging activities of 2,2-diphenyl-1-picrylhydrazyl (DPPH) and azino-bis (ethylbenzothiazoline-6-sulfonic acid, ABTS •+ ) was determined by classical spectrophotometric methods with some corrections 23,24 . For the DPPH assay, different concentrations of the EOs in methanol (0.1~10 mg/mL) and hydrosols in distilled water (2~100%) were added to 2 mL of a 0.15 mM solution of DPPH. The mixture was shaken vigorously to be homogenous. After incubated in a dark place for 30 min at room temperature, the absorbance decrease at 517 nm of the reacting solution was recorded. In the ABTS •+ test, the prepared radical solution was attenuated with methanol to an absorbance of 0.700 ± 0.02 at 734 nm. After 100 μL of sample solution reacted with 1.0 mL of ABTS •+ solution for 30 min at room temperature, the absorbance at 734 nm would be recorded. The radical solutions added with the equal amount of the corresponding sample diluents were used as control and tests were carried out in triplicate. The percentage inhibition was calculated as (I%) = (A 0 -A)/A 0 ×100, where A 0 is the absorbance of control without adding antioxidants, and A represents the absorbance of the samples. The IC 50 value was calculated as the concentration of the sample needed to scavenge 50% of free radicals by graphing the linear curve of I% versus sample concentration. BHT (butylated hydroxytoluene) was used as a positive control compared with the essential oils and ascorbic acid as a positive control for the hydrosols.

Reducing power assay
The reducing power of EOs and hydrosols was performed following the standard method 25 with a little modi cation. 0.2 mL sample at different concentrations was added 1.0 mL of 0.2 M phosphate buffer (pH 6.6) and 1.0 mL of 1% potassium ferricyanide. The mixtures were incubated in a water bath at 50 °C for 20 min. Then, 1.0 mL of 10% trichloracetic acid was added into the mixture to terminate the reaction. The resulted solution was mixed with an equal amount of distilled water and 0.1 mL of 0.1% ferric chloride (FeCl 3 ) solution for 10 min. Then, the absorbance of mixture was measured at 700 nm against blank solution 2.6 Statistical analysis Statistical analysis was performed using the SPSS 16.0 software program (Statistical Package for Social Sciences, Chicago, IL, USA). All experimental data were presented as the mean ± SEM of the mean from at least three independent experiments.
Statistical signi cance was assessed by one-way analysis of variance and Turkey test to examine differences between the two groups using SPSS version 16.0 (SPNN Inc., Chicago, IL, USA). p values of less than 0.05 were considered statistically signi cant.

Results And Discussion
3.1. Single factor study of SFME Solvent-free microwave extraction (SFME), an innovative technology combining microwaves and distillation, has been rapidly developed as one of the popular techniques to extract EO from plant materials. In this SFME process, microwave irradiation provides energy to heat in situ water generating pressured atmosphere. This tremendous pressure tends and ruptures the cell walls, which facilitates the aromatic compounds spreading over and subsequent dissolution. The increase of the microwave power improves the yield and leads to a shorter extraction time.

Effect of microwave irradiation power on EO yield
Microwave irradiation power can lead aromatic compounds to rapidly rotate and to leach out from the plant cells, while too high power will degrade volatile components and destroy the material. Thus, an appropriate microwave irradiation power setting is important in EO extraction. It can be observed from Fig. 1 that the EO yield increases with the increasing of microwave irradiation power (from 150 to 750 W), and reaches a well-de ned limit when the irradiation power of more than 750 W resulted in a slight decrease in the yield. Similar results were also obtained in previous studies 26 , 27. Whereas, there was no signi cant difference (p < 0.05) of EO extraction yield at 600 W compared with the yield at 750 W, which were (0.733 ± 0.035)% and (0.731 ± 0.045)%, respectively. To lower energy consumption, microwave irradiation power in the range of 300 ~ 600 W was chosen for the subsequent optimization.

Effect of microwave irradiation time on EO yield
To choose proper microwave irradiation time is important for the SFME process because long irradiation time can issue in excessive energy accumulation, which will lead to degradation of plant material or loss of some volatile constituents26. As presented in Fig. 1, the EO yield gradually increased with the rise in microwave irradiation time from 10 min to 40 min. After 40 min, the yield reached a plateau denoting the maximum EO extracted. However, the EO yield at 50 min was (0.728 ± 0.020)%, which was not signi cantly different (p < 0.05) with the yield of (0.733 ± 0.035)% at 40 min. To promote e ciency of operations, 20 ~ 40 min was selected for the subsequent optimization by BBD.

Effect of percentage humidity of the matrix on EO yield
To obtain the suitable range of percentage humidity of the matrix for further optimization of the SFME procedure, the effect of this single factor on the PL fruit EO extraction was evaluated at a xed microwave irradiation time (30 min) and power (600 W) as shown in Fig. 2. An increase in percentage humidity of the matrix from 40-60% dramatically promoted extraction e ciency, which was likely due to the higher percentage humidity of the matrix generated higher evaporation pressure leading to more e cient extraction. However, further increase of percentage humidity of the matrix decreased the EO extracting e ciency after 60%. This might because much higher humidity needed more heat to evaporate, and could result in the hydrolysis of some components and dissolution in water. Therefore, 50%~70% was chosen as the suitable range of percentage humidity of the matrix for following optimization.

Extraction model tting and statistical analysis
Based on the results of single factor experiments and design principle of BBD, response surface test of three factors and three horizontals was perforemed with microwave irradiation power (X 1 ), microwave irradiation time (X 2 ), percentage humidity of the matrix (X 3 ) as independent variables and the essential oil yield (Y) as the response value. The test results are shown in Table 1. In the experimental range, maximum value of Y was 0.733% recorded under the experimental conditions of X 1 (600 W), X 2 (40 min), and X 3 (60%). By analyzing the exprimental data, the mathematical model equations of Y and three independent variables are obtained as follows: Results of the analysis of variance (ANOVA) for the tted model were shown in Table 2. The regression model is extremely signi cant (p < 0.0001). Smaller value of lack of t with greater p value meant better tness of the equation. For this caculated equation, mean square and p-value of the lack of t was 0.0006 and 0.22, respectively, which indicated that the unknown factors had little effect on the experimental results, and the predicted regression model tted well with the test data. The value of R 2 for the model was 0.9982, which indicated that only 0.18% of the total variations were not explained. The value of the adjusted determination coe cient (R2 adj = 0.9985), also con rmed that the model was signi cantly tted and the experimental value of PL EO yield was highly consistent with the predicted value. Therefore, the regression equation could essentially predict and analyze the yield of PL fruit EO under most combination of the extraction variable values in the experimental range. The model terms having signi cant effects on essential oil yield (p < 0.05) were linear factors of X 1 , X 2, and X 3 , and quadratic ones of X2 1, X2 2, and X2 3, while only one interaction term of X 1 X 3 was also signi cant (p < 0.05). Based on the established model, the maximum predicted essential oil yield was 0.747%. The optimal parameters were: microwave irradiation power 572.12 W, microwave irradiation time 37.37 min, and percentage humidity of the matrix 59.00 %, respectively.

Analysis of response surface
The graphical representations of the polynomial regression equation called three-dimensional response surfaces and contour plots were generated by Design Expert package as shown in Fig. 3.a-c. They illustrated the relationship between the essential oil yield and three extraction variables and the interactions between each two experimental variables. Elliptical and saddle-backed contour plot shows that the corresponding variables signi cantly interrelated ecah other, while circular contour plot explains that the interaction effects between the two investigated variables are negligible 28. In all the presented gures, the other one variable was kept at level zero.
At xed percentage humidity of the matrix (60%), both microwave irradiation power (X 1 ) and time (X 2 ) signi cantly in uenced the PL fruit EO yield in a quadratic manner, scilicet, increasing of the two variables to 594.81 W and 36.05 min could work out the highest essential oil yield (0.745%) as shown in Fig. 3a. However, the contour shape of plot describing of the interaction effect of X 1 and X 2 on the yield was nearly circle, which indicated that the reciprocal action of microwave irradiation power and time on yield of PL fruit EO was not signi cant (Fig. 3a), corresponding to the ANOVA results. Likewise, at a constant irradiation time (30 min), the augment in microwave power (up to 533.08 W) and percentage humidity of the matrix (up to 65.80%) brought out a corresponding increase of the EO yield, though further increase of the two variables lead to an obvious decrease of the response value as presented in Fig. 3b. Nevertheless, the contour line displaying the mutual interaction between microwave irradiation power and percentage humidity of the matrix is a sharp ellipse, which could prove that the interaction effect of the two independent factors on the dependent variable was signi cant. Although the mutual interaction between microwave irradiation time and percentage humidity of the matrix is an ellipse (Fig. 3c), it still can't be deduced that the interaction effect of X 2 and X 3 on Y value was signi cant as the p value was 0.8693 from the ANOVA for the model-tting process.

Optimization and validation of the models
Based on the optimal conditions predicted by the Design-Expert software, con rmatory experiment of the adequacy of the model was performend with following modi ed parameters for operational feasibility: microwave irradiation power (X 1 ) of 572 W, microwave extraction time (X 2 ) of 37.50 min, and percentage humidity of the matrix (X 3 ) of 59.00%. The experimental yield of EO was 0.747 ± 0.005% (n = 3), being not signi cantly different with the predicted maximum value, but much higher than the yield of the HD method, which was (0.496 ± 0.021)%. Thus, those adoptive conditions were recommended for future SFME of EO from PL fruit.

Chemical compositions of EOs
According the reported defenition, terpenes, phenylpropanes and their oxygenated substances, comprising of phenolic compounds, sometimes accounted among the majority of molecules in several EOs' composition 19. From the most studies in being, EO of PL fruit possessed abundant amounts of sesquiterpenes, though there were variances in the contents of each ingredients according to differnet extraciton methods and extracting time 29-31. The chemical pro les of the EOs from PL fruit extracted by SFME and HD were listed in Table 3. 50 and 36 components were identi ed, accounted for 95.99% and 98.52% of the two products, respectively.
Generally, the amount of oxygenated compounds was directly related to the economic value of the aromatice products as these substances were considered more in uential on the aroma of EOs than other compenents 35. However, there are more and more evidence proving that natural terpenes especially sesquiterpenes displayed excellently as antimicrobial 36, antitumor, antiin ammatory, antiviral and antioxidant effects 37. Maybe economic valuation of the EOs should becomes to consider bioactivities as an important evaluation index. In conclusion, SFME could be an appropriate method to extract PL fruit EO, as the obtained sample possessed higher contents of terpenes and oxidated compounds.

Chemical compositon of hydrosols
Hydrosols, which have been commercially used in many industries, were dominated by a wide range of functional compounds, including aldehydes, alcohols 38, phenols 39 , 40 and esters. GC-MS analysis displayed qualitative and quantitative discrepancies in the components of the two aqueous samples isolated from PL fruit of SFME and HD process. 59 components in SFME hydrosol, more than 32 in HD hydrosol were characterized, representing 95.65% and 93.97% of the total sample, respectively. Among these identi ed compounds of SFME and HD hydrosols, eugenol was found to be the most aboundant (25.44% and 22.82%, respectively), followed by linalool (14.10% and 22.37%, respectively). Trans-isoeugenol only consisted in the SFME hydrosol and took up 14.33% of the total chromatogram area. Additionally, there were other primary components presented in both hydrosols obtained by SFME and HD, including piperonal (7.28% and 9.79%), methyl-3-phenylpropanoate (6.73% and 3.77%), phenylethyl alcohol (3.36% and 0.70%), α-terpineol (2.25% and 1.97%), trans-linalool oxide (1.39% and 1.53%), 3,4-dimethyl-3-cyclohexenecarbaldehyde (1.27% and 2.04%), and benzaldehyde (0.67% and 1.23%). Ethyl-3-phenylpropionate (1.05%) was detected in SFME hydrosol and was reported as a volatile avor compound occurring in muskmelon and red fermented soybean curds. Eugenol and trans-isoeugenol were two important aromatic compounds with a spicy and clove-like scent. High amount of these two compounds are bene cial because of their antioxidant, antibacterial, antiantifungal, anticholinergic, and antidiabetic effects 41. Eugenol (64. 6 120 mg/kg) was rstly detected in PL fruit by a new technology based on gas chromatography-high resolution mass spectrometry (GC-HRMS-Q-Orbitrap) 21. In our study, trans-isoeugenol was observed for the rst time in PL fruit but it was not sure that wether it was naturally biosynthesized in vivo or transformed from eugenol with the energy provided by microwave irradiation as eugenol was an intermediate to chemically synthesize trans-isoeugenol. Linalool, a diterpenoid, also can enhance the applying value of the products as it demonstrates to hold a broad spectrum of bioactive properties that can be exploited by the aromatherapies and comestic, pharmaceutical and food industries in the production of herbal-based health products 42. Hydrosol collected during HD process contained furfural (4.62%), 2-ethylhexanol (1.78%), and acetophenone (1.06%) as its primary constituents. In conclusion, phenols and alcohols were two predominate compounds of the hydrosols isolated from PL fruit. Wheras, piperitone and linalool oxide were two dominant constituents in the hydrosols of leaves and in orescence from Piper chimonanthifolium Kunth, which also belonged to the Piperaceae family 43.

Antioxidant activities of EOs and hydrosols
Antioxidant properties play a critical role in many of EOs' commercial applications, which is due to their extension of products shelf life and protection of human health against oxidative stress involved disorders 19. So it was indispensable to evulate antioxidant potential of the EOs and hydrosols obtained from SFME and HD methods to guide their practical applications. The results were summarized in Table 4 and Table 5, respectively.
Free radicals with high ractivities can induce a series of chain reactions harmful to organisms and all raw materials. Radicalscavenging methods are well known as classical and reliable chemical assays for the determination of antioxidant potential of natural products or synthetic compounds. DPPH and ABTS + radicals were employed in this part. In the DPPH test, the EO from SFME method exhibited signi cantly higher scavenging capacities (p < 0.05) than the sample from the HD process at each investigated concentration except 2 mg/mL ( Table 4). The calculated IC 50 value of the SFME EO to scavenge DPPH radical was 3.30 mg/mL, and that value of HD EO was 5.90 mg/mL, which was about 1.79-fold greater. For ABTS •+ scavenging, there were signi cant differences between four tested concentrations (0.2, 0.4, 0.6, 1.0 mg/mL) of SFME and HD EOs. The IC 50 values for the two samples were 0.48 mg/mL and 0.56 mg/mL, respectively. However, scavenging capacities of SFME and HD EOs were moderate when compared with the positive control BHT with 56.22 and 4.86 µg/mL as the IC 50 values for DPPH and ABTS + radicals scavenging, respectively (Table 6).
In reducing antioxidant power assessment, bioactive compounds reduce ferric ions (Fe 3+ ) to the ferrous (Fe 2+ ), which then reacts with ferric chloride to form a colored complex having a maximum absorption at 700nm. More of the absorbing complex produced, higher absorbance values were presented, which indicated higher reductive capabilities and antioxidant potential of the tested samples. As shown in Table 4, the reducing abilities of the tested samples signi cantly increased with concentration increasing and the absorbance of the two EOs exhibited at the same concentration were signi cantly different (p < 0.05). When the concentration of SFME and HD EOs researched 10 mg/mL, the absorbance were 0.5530 ± 0.0141 and 0.4610 ± 0.0127, respectively. Total phenolic content of SFME EO was 3.14 ± 0.12 mg GAE/g, which was a little higher than 2.99 ± 0.02 of HD EO. Absorbance value of the positive control BHT was 0.8740 ± 0.0325 at a concentration of 60 µg/mL, signi cantly higher than those former two. It had been proved that the reducing properties was positively related with the content of phenolic compounds due to their reactive hydroxyl, which can provide hydrogen ions to neutralize perxoyl radicals 44. Furthermore, modest content in unsaturated terpenes, especially the components containing a carbon skeleton of cyclohexane-1,4-diene, could also contribute to good antioxidant behavior of EOs 19.
Total phenolic content of SFME hydrosol was 82.66 ± 0.147 µg GAE/mL, which was signi cantly higher than 36.40 ± 0.06 µg GAE/mL of HD hydrosol. According to the results, both hydrosols exhibited dose-dependent antioxidant activities and SFME hydrosol demonstrated to possess signi cantly stronger scavenging capacities of DPPH and ABTS + radicals, compared with the sample from HD method ( Table 5, p < 0.05). SFME hydrosol eliminated 93.93% of total DPPH radicals at the concentration of 40% and after that the scavenging capacity got a plateau, but the HD did 70.50% at the concentration of 100%, relative to ascorbic acid with 93.23% inhibition at 8 µg/mL. For ABTS + radical assay, the IC 50 value of SFME hydrosol was 5.53%, which was signi cantly lower than 12.43% of HD hydrosol. The half maximal inhibitory concentration of ABTS radical for ascorbic acid was 3.14 µg/mL ( Table 7). The reducing capabilities of the both hydrosols were observed to be signi cantly higher when compared with that of ascorbic acid at the concentration of 10 µg/mL while SFME sample demonstrated a wonderfully higher reducing ability (A 700 nm =1.3881) than ascorbic acid acted at the concentration of 80 µg/ml (A 700 nm = 0.8530). The lower antioxidant activities by HD hydrosol displayed may be attributed to its lower phenolic content in keeping with the aforementioned opinon that phenolic compounds, including both terpenoids and phenylpropanoids, play an important role in antioxidant activities. However, all of the six hydrosols from different parts of arecanut and coconut showed weaker reducing power and ABTS + radical scavenging capacities compared with ascorbic acid at 5 µg/mL 22.

Conclusion
In the present study, solvent-free microwave extraction (SFME) of EO from PL fruit was optimized with BBD of RSM for the rst time. And hydrosols collected during SFME operation as a by-product was also explored for the rst time. A much higher EO yield 0.747% was obtained under the optimal conditions, compared to that of HD approach 0.496%. PL fruit EO obtained by SFME revealed higher contents of terpenes and oxygenated components. Additionally, it possessed stronger radical scavenging activities and reducing power compared with that obtained by traditional HD. Hydrosol collected during SFME process contained more constituents and exhibited more considerable antioxidant capacities than the hydrosol sample from HD. These results proposed that SFME was more effective and environment-friendly approach for the simultaneous extraction of EO and hydrosol from PL fruit. Furthermore, in view of the potential as antioxidant, EO and hydrosol of PL fruit can be recommended as natural supplements in daily chemical products and foods. However, further research is needed to establish these aromatic products as inhibitory of microorganism growth to develop their applied value.