Optimization of Microwave-Assisted Extraction of Polyphenols from Lemon Myrtle: Comparison of Modern and Conventional Extraction Techniques Based on Bioactivity and Total Polyphenols in Dry Extracts

The aromatic herb lemon myrtle is a good source of polyphenols, with high antioxidant and antimicrobial capacity. In this study, the green extraction technique microwave-assisted extraction (MAE) was applied and the extraction parameters were optimized using response surface methodology (RSM) to maximize the extraction yield of phenolic compound and antioxidant properties. Then, it was compared with other popular novel and conventional extraction techniques including ultrasound-assisted extraction (UAE) and shaking water bath (SWB) to identify the most effective technique for extraction of phenolic compounds from lemon myrtle. The results showed that the MAE parameters including radiation time, power, and sample to solvent ratio had a significant influence on the extraction yield of phenolic compounds and antioxidant capacity. The optimal MAE conditions were radiation time of 6 min, microwave power of 630 W, and sample to solvent ratio of 6 g/100 mL. Under optimal conditions, MAE dry extract had similar levels of total phenolic compounds (406.67 ± 8.57 mg GAE/g DW), flavonoids (384.57 ± 2.74 mg CE/g DW), proanthocyanidins (336.54 ± 7.09 mg CE/g DW), antioxidant properties, and antibacterial properties against (Staphylococcus lugdunensis and Bacillus cereus) with the other two methods. However, MAE is eight-times quicker and requires six-times less solvent volume as compared to UAE and SWB. Therefore, MAE is recommended for the extraction of polyphenols from lemon myrtle leaf.


Introduction
Phenolic compounds are non-nutritive phytochemical groups synthesized in the plant as secondary metabolites. Despite the lack of nutritional attributes, they have been widely used as natural antioxidants and antimicrobial agents in foods [1,2]. In addition, they have been used as supplements or therapeutic agents for prevention and/or delay several health complications such as cancer, diabetes, neurodegeneration, early aging, inflammation, hypertension, allergies, arthritis, and cardiovascular diseases [3][4][5]. Lemon myrtle leaves are rich in polyphenols with strong antioxidant properties [6,7]. Polyphenols derived from lemon myrtle leaves also have various bioactivities such as anti-inflammation [8,9], antimicrobial [10,11], cytoprotective, and pro-apoptotic properties [12].
Lemon myrtle leaves have been used as traditional medicine and in cuisine for centuries [13]. The leaves also contain a refreshing sweet lemon-flavored essential oil, which also has numerous applications in various sectors. More applications of lemon myrtle are being explored through the advancement of research [13] and its demand is rising with time. To meet the demand it has been growing commercially in different parts of

Experimental Design
In this study, three extraction techniques were applied including MAE, UAE, and SWB. A 50% aqueous acetone was applied as the extraction solvent as it is the most effective solvent for the extraction from lemon myrtle leaves [28]. Optimal MAE conditions including radiation time, radiation power, and sample to solvent ratio were established as described in Section 2.3.1. Optimal conditions for the UAE extraction technique were applied as described in our previous study [28] and SWB extraction was performed as mentions in the following section. After lemon myrtle leaves were extracted under optimal MAE, UAE, and SWB, the extracts were concentrated to around 30% of initial volume using a rotary evaporator (Buchi Rotavapor R-114, Buchi Australia, Noble Park, VIC, Australia) set at 40 • C and pressure of 150 mbar, which was gradually decreased to 50 mbar and further freeze dried to obtain powdered extracts. Finally, the powdered extract (500 µg) from each extraction technique was diluted in 50% aqueous acetone (1 mL) for further analysis.

Optimization of MAE Conditions Using Response Surface Methodology (RSM)
MAE was performed using a partially modified domestic microwave oven (Pana-sonic NN-7855, Japan) as described previously [29]. The sample and solvent were put into the oven in a round bottom flask then it was connected to a condenser through a glass tube ( Figure 1). The MAE extraction parameter's values were selected according to RSM design (Table 1). After extraction, the extract was cooled down to room temperature in an ice bath followed by filtering through a 0.45 µm cellulose syringe filter (Phenomenex Australia Pty. Ltd., Lane Cove, NSW, Australia). The filtrate was stored at −18 • C for further analysis and also for freeze drying to obtain powdered extract for comparison. order polynomial equation represents a functional relationship between the predicted yield of phenolic or antioxidant properties and extraction parameters, which was as follows:  A three-level three-factor Box-Behnken design was used to observe the individual, interaction, and quadratic effect of extraction parameters on the extraction yield of phenolic compounds and antioxidant capacities and to find out the optimal extraction conditions for the maximum yield of phenolic compounds and antioxidant properties. The extraction parameters were radiation time (X1): 2-6 min, microwave power level (X2): 10% (90 W)-100% (900 W), sample to solvent ratio X3: 2-8 g/100 mL. The parameters range was selected based on initial experiments (data not presented in this manuscript). A secondorder polynomial equation represents a functional relationship between the predicted yield of phenolic or antioxidant properties and extraction parameters, which was as follows:

Ultrasound-Assisted Extraction (UAE)
UAE was performed under optimal conditions as previously described [28] using an ultrasonic bath (Soniclean, 220 V, 50 Hz, and 250 W, Soniclean Pty LTD, Dudley Park, SA, Australia). A preheated (at 60 • C) volume of 25 mL of 50% acetone in water was added with 0.25 g of dried ground lemon myrtle leaves in a 50-mL centrifuge tube with a screw cap. The sonication bath was preheated at 60 • C and the UAE conditions were time (50 min), temperature (60 • C), and sonication power (250 W). During extraction, the tubes were vortexed for 2-3 s every 5 min. After finishing extraction the tubes were cooled down Processes 2021, 9, 2212 6 of 19 to room temperature in an ice bath followed by filtering using a 0.45-µm cellulose syringe filter (Phenomenex Australia Pty. Ltd., NSW, Australia) and stored at −18 • C for further freeze drying to obtain powdered extract for comparison.

Conventional Extraction Using Shaking Water Bath Extraction (SWB)
The conventional extraction was performed by following Saifullah, et al. [30] with modifications. Shaking water bath extraction was performed in a pre-heated shaking water bath at 60 • C and under the extraction conditions of time 50 min, temperature 60 • C, and sample to a solvent ratio 1 g/100 mL. Then, 0.25 g of freeze-dried ground lemon myrtle leaves and 25 mL of solvent (50% aqueous acetone) were put into a 50-mL centrifuge tube. Then the tube was closed with a lid before being put into the shaking water bath. During extraction, the samples were agitated at a constant speed. At the end of extraction, the tubes were immediately cooled to room temperature in an ice bath and then the extract was filtered using a 0.45-µm cellulose syringe filter (Phenomenex Australia Pty. Ltd., Australia) and stored at −18 • C for further freeze drying to obtain powdered extract for comparison. TPC, TFC, and Pro. A were measured according to the method described by Škerget et al. [31], Zhishen, Mengcheng, and Jianming [32], and Li et al. [33] respectively. A UV spectrophotometr (Cary 60 Bio, UV-Vis, Penang, Malaysia) was used to measure the absorbance of light (wavelength 765, 510, and 500 nm for TPC, TFC, and Pro. A, respectively). TPC results were expressed as milligrams of gallic acid equivalent per gram dry weight of the sample. TFC and Pro. A results were stated as milligrams of catechin equivalent per gram dry weight of the sample.

Measurement of Antioxidant Properties
Antioxidant properties of the extracts were measured using four antioxidant assays, including FRAP, ABTS, and DPPH assays Thaipong et al. [34], and CUPRAC assay Apak et al. [35]. The absorbance was taken at 593, 450, 734, and 515 nm for FRAP, CUPRAC, ABTS, and DPPH, respectively. Trolox was used to prepare the standard curves and the results were expressed as mM Trolox equivalent per gram dry weight of the sample.

Total Chlorophyll Content
Total chlorophyll content in leaf extract powder was measured according to a method reported by Sudhakar et al. [36] with some modifications. Then, 0.1 g of leaf extract was dissolved in 10 mL of 80% acetone in water followed by vortex to mix well. Then the optical density of the leaf extract solution was measured using a UV spectrophotometer (Cary 60 Bio, UV-Vis, Malaysia) at 663 and 645 nm. Finally, the amount of chlorophyll present in the dry extract was calculated using following equations: Here, A = absorbance at specific wavelengths, V = final volume of chlorophyll extract, and W = weight of dry powder Australia). The bacterial inoculums were prepared according to a method described by Bhuyan et al. [37]. The bacterial cultures were prepared by growing the selected bacteria in nutrient broth at 37 • C for 24 h. Then sterile physiological saline was used to dilute the nutrient broth containing bacterial culture to adjust its turbidity to MacFarland turbidity standard of 0.5 (Oxoid™) (10 6 colony forming units per mL).

Disc Diffusion Bioassay
This assay was performed according to a method described by Chandrasekaran and Venkatesalu [38] and Bhuyan et al. [37]. In brief, the freeze dried extracts were dissolved in 5% dimethylsulfoxide (DMSO) to prepare concentrations of 15 and 20 mg/mL. A volume of 0.1 mL of bacterial inoculum was spread uniformly on Mueller Hinton agar (MHA) and allowed to dry for 5 min at room temperature. Antimicrobial susceptibility disks (6 mm) Thermo Scientific™ Oxoid™ were impregnated with 20 µL of extracts (concentration of 15 or 20 mg/mL) and left them drying at room temperature. A blank disc soaked with 5% DMSO commercial discs (Thermo Scientific™ Oxoid™) with ciprofloxacin (5 µg) was set as positive and negative control reference, respectively. The discs impregnated with extracts, positive and negative reference were placed on bacterial inoculated MHA plates using sterile forceps. The plates were incubated at 37 • C temperature for 24 h followed by the bacterial growth inhibition zone were measured.

Statistical Analysis
The RSM design was performed using a statistical software JMP (version 14), which was applied to predict the impact and optimize the extraction parameters for MAE. JMP was also applied to formulate an estimation model equation, create 2D counterplots, and 3D surface profiles. To compare the predicted values with experimental results paired sample t-test was performed using SPSS (IBM, Version 27). Different extraction methods were compared using all pairs Tukey-Kramer HSD. The difference between the mean values is considered statistically significant at p < 0.05. The experimental run and data accusation was performed at least three times (n = 3). For antimicrobial studies, experimental run performed at least three times and data accusation was performed at least nine times. The data are represented as mean ± standard deviation.

Accuracy of the Polynomial Prediction Models
The RSM develops polynomial model equations, which predict the value of the dependent variable in the response of independent variables [39]. In this study, the models predict the extraction yield value of phenolics and antioxidant levels in the lemon myrtle extract in response to the MAE parameters. Table 2 represents the analysis of variance (ANOVA) results, which shows how well the prediction models are fit to predict the extraction yield of phenolics and antioxidant capacity values in response to different extraction parameters. Significantly different at * p< 0.05, ** p < 0.01, and *** p < 0.001; β 0 : intercept; β 1 , β 2 , and β 3 : linear regression coefficients for time, power, and ratio; β 12 , β 13 , and β 23 : regression coefficients for interaction between time × power, time × ratio, power× ratio; β 11 , β 22 , and β 33 : quadratic regression coefficients for time × time, power × power, and ratio × ratio.
A high coefficient of determination R 2 value, significant p-value, insignificant (p > 0.5) lack of fit, and low root mean square error are commonly used as indicators to determine the suitability of the prediction models [30,40,41]. The R 2 value is an index of the goodness of fit of the predicting models [42]. The ANOVA results show that the R 2 values for phenolics and antioxidant capacities range from 0.91-0.97, which represents that above 91% of the experimental data can be explained by the predicting models. The adjusted R 2 values ranged from 0.74-0.92, which also shows reliability of the models; as the R 2 value above 0.70 represents a strong relationship. The significant p-values of the model and insignificant lack of fit values (at p < 0.05) for the phenolic compounds and antioxidant capacities further confirm the accuracy of the models. Low RMSE values further support the model fitting. The fitted polynomial model equation for polyphenols (TPC, TFC, and Pro. A), and antioxidants capacity (FRAP, CUPRAC, DPPH, and ABTS) are represented in Equations (3)-(9) as follows: Y TFC = 20.28 + 12.28X 1 + 0.

Influence of MAE Parameters on Polyphenols and Antioxidant Properties Yield
The regression analysis results in Table 2 show the individual, interaction, and quadratic effects of MAE parameters on the extraction yield of polyphenols (TPC, TFC, and Pro. A) and antioxidant capacity (measured by FRAP, CUPRAC, DPPH, and ABTS). Figure 2 also depicts the linear effects of extraction parameters on the extraction yields. The extraction parameters linearly, positively, and significantly (p < 0.05) affected the extraction yield of TPC. Conversely, the quadratic effects of all three parameters on TPC yield were negative, but only the sample to solvent ratio's influence was significant (p < 0.05). The interaction effects of the extraction parameters were insignificant and negative for TPC, however, the effect of the interaction of power × ratio was positive. Figure 3a-c depicts the 3D surface profiles for interaction and quadratic effects of extraction parameters on phenolics. For TFC and Pro. A, the radiation time and power positively and significantly influenced the extraction yield. However, the influence of sample to solvent ratio was insignificant for both TFC and Pro. A. There were also no significant interaction effects between MAE tested parameters on both TFC and Pro. A yield. The quadratic effect of power was not only significantly but negatively influential to the TFC and Pro. A content in the extract. The regression analysis results in Table 2 show the individual, interaction, and quad ratic effects of MAE parameters on the extraction yield of polyphenols (TPC, TFC, and Pro. A) and antioxidant capacity (measured by FRAP, CUPRAC, DPPH, and ABTS). Fig  ure 2 also depicts the linear effects of extraction parameters on the extraction yields. The extraction parameters linearly, positively, and significantly (p < 0.05) affected the extrac tion yield of TPC. Conversely, the quadratic effects of all three parameters on TPC yield were negative, but only the sample to solvent ratio's influence was significant (p < 0.05) The interaction effects of the extraction parameters were insignificant and negative for TPC, however, the effect of the interaction of power × ratio was positive. Figure 3a-c de picts the 3D surface profiles for interaction and quadratic effects of extraction parameters on phenolics. For TFC and Pro. A, the radiation time and power positively and signifi cantly influenced the extraction yield. However, the influence of sample to solvent ratio was insignificant for both TFC and Pro. A. There were also no significant interaction ef fects between MAE tested parameters on both TFC and Pro. A yield. The quadratic effec of power was not only significantly but negatively influential to the TFC and Pro. A con tent in the extract.     The positive linear effect and negative quadratic term indicate that by increasing extraction parameters values, higher extraction yield of phenolics will result to a certain level, then the extraction yield will decrease when the MAE conditions exceed the optimal points.
The higher yield can be explained by the increasing solubility of polyphenols at higher temperatures and increased microwave power and/or extraction time/microwave irradiation time [43]. High temperature can soften the plant tissue and increase penetration of solvent into the sample, increasing mass transfer rate [44], it can also break down chemical bonds or reduce intermolecular interaction that increases the solubility of certain compounds [45,46], resulting in higher extractable compounds in the solution.
In addition, microwaves can create a pressure gradient between the inside and outside of the plant tissues or cells and thus rupture the sample tissue and cell wall, consequently releasing more phenolic compounds [47]. However, high temperatures can lead to the degradation of phenolic compounds, especially heat-sensitive compounds. Therefore, exceeding optimal conditions, a lower level of phenolic compounds was observed [48,49].
Our results reveal that higher sample to solvent ratios have lower extraction yields of phenolic compounds. This could be explained by the saturation of phenolics in the solvent. The higher quantity of sample requires more solvent to be immersed, thus reduces the contact between the sample and solvent [50]. Our findings also show that the linear effect of radiation time, microwave power, and sample to solvent ratio is positive; how- The positive linear effect and negative quadratic term indicate that by increasing extraction parameters values, higher extraction yield of phenolics will result to a certain level, then the extraction yield will decrease when the MAE conditions exceed the optimal points.
The higher yield can be explained by the increasing solubility of polyphenols at higher temperatures and increased microwave power and/or extraction time/microwave irradiation time [43]. High temperature can soften the plant tissue and increase penetration of solvent into the sample, increasing mass transfer rate [44], it can also break down chemical bonds or reduce intermolecular interaction that increases the solubility of certain compounds [45,46], resulting in higher extractable compounds in the solution.
In addition, microwaves can create a pressure gradient between the inside and outside of the plant tissues or cells and thus rupture the sample tissue and cell wall, consequently releasing more phenolic compounds [47]. However, high temperatures can lead to the degradation of phenolic compounds, especially heat-sensitive compounds. Therefore, exceeding optimal conditions, a lower level of phenolic compounds was observed [48,49].
Our results reveal that higher sample to solvent ratios have lower extraction yields of phenolic compounds. This could be explained by the saturation of phenolics in the solvent. The higher quantity of sample requires more solvent to be immersed, thus reduces the contact between the sample and solvent [50]. Our findings also show that the linear effect of radiation time, microwave power, and sample to solvent ratio is positive; however, the quadratic action of the same parameters is negative. The outcomes of this study are in agreement with previous studies on MAE of phenolics from eggplant peel and algae reported by Doulabi et al. [47] and Dang et al. [51] respectively; who found that the extraction yields of various phenolic groups and their antioxidant properties increase with increasing MAE parameters (radiation time, microwave power, and sample to solvent ratio), then gradually declines when these parameters exceeded the optimal levels. Belwal et al. [26] also reported similar findings of a significant positive linear effect of microwave power level and sample to solvent ratio on the extraction yield of total phenolics and antioxidant capacities in MAE derived extract from Berberis asiatica leaves. Their study showed that in quadratic terms, sample to solvent ratio influences ABTS and FRAP significantly and negatively. Various effects on antioxidant capacity and the patterns of effects can also be seen in Table 2. A 2D counter plot ( Figure 2) and 3D surface plot (Figure 3d-g) for antioxidant capacity further depicts the effects of the extraction parameter on antioxidant capacity of the extract. In this study, the antioxidant capacity of the extract was measured using two reducing capacity (FRAP and CUPRAC) assays and two free radical scavenging capacity (DPPH and ABTS) assays.
Similar to phenolic compounds, the antioxidant capacities are also influenced by individual parameters. It is likely because there is a strong correlation between phenolic compounds and antioxidant capacities in lemon myrtle extract [6]. A positive, linear, and significant effect of all extraction parameters on antioxidant capacity were observed; except the ratio had a negative and not significant effect on ABTS. All the interaction effects of the extraction parameters for FRAP, CUPRAC, ABTS, and DPPH were insignificant. The extraction parameters adversely affected the antioxidant capacities of the extract in quadratic terms. Except for ABTS, all of the antioxidant property measures were significantly (p < 0.05) influenced by the quadratic action of sample to solvent ratio. Quadratic (microwave power × microwave power) also showed significant influence on FRAP and ABTS values of the extract.

Optimization and Validation
Based on the RSM analysis, the predicted optimal conditions for the maximum yield of polyphenols and antioxidant capacities are radiation time of 6 min, power of 70% (630 W), and sample to solvent ratio 6 g/100 mL. Under these predicted optimal conditions, the extraction yields of TPC, TFC, Pro. A, FRAP, CUPRAC, DPPH, and ABTS was 106.51 ± 7.32 mg GAE/g DW, 98.85 ± 7.88 mg CE/g DW, 110.18 ± 6.43 mg CE/g DW, 852.30 ± 70.03 mM TE/g DW, 6194.44 ± 601.82 mM TE/g DW, 1067.14 ± 133.93 mM TE/g DW, and 1594.73 ± 161.7 mM TE/g DW respectively. The predicted optimal yields of total phenolic compounds and antioxidants were compared with the experimental yield to validate the predicted optimal MAE conditions and justify the accuracy of the predicted models. An insignificant difference between predicted and experimental values indicates the accuracy of the model and validity of the optimal conditions [48,52]. Table 3 presents the outcomes of paired sample comparison between the predicted and experimental values. The results show that there was no significant difference between the predicted and experimental values at a significant level of p < 0.05; which confirms the predicted optimal extraction conditions were validated successfully and it can be applied for further extraction of phenolic compounds and antioxidant-enriched extract from lemon myrtle leaves. The optimal yield of phenolic and antioxidant from MAE was comparable with the extraction yields from lemon myrtle leaves using UAE reported by Saifullah et al. [6]. The TPC and TFC yield using optimal MAE conditions was considerably high compared to the reported UAE method; however, the Pro. A yield and antioxidant capacity values were close for both methods. The extraction yield of phenolic compounds and antioxidants can vary according to extraction conditions and methods.

Comparison of the Effects of Conventional and Novel Extraction Techniques on Phenolics, Chlorophyll, Antioxidants, and Antimicrobial Properties
The cost-effective extraction of the maximum possible concentration of compounds with the highest bioactivity is important for natural product development industries [53]. The extraction yield and efficiency can vary depending on the extraction methods [54]. The yield of phenolics and their various properties (i.e., antioxidant, antimicrobial, and anti-inflammatory) are important factors for the selection of the extraction method. In this study, we compared the extraction efficiency of two novel and common extraction techniques, including MAE, UAE with a common conventional extraction technique shaking SWB under their optimal conditions on the extraction efficiency of TPC, TFC, and Pro. A, antioxidant capacity, and total chlorophyll content (TCC) ( Table 4). The results show that there was no significant difference among the three tested methods on the extraction efficiency of TPC, TFC, and Pro. A. However, MAE has approximately 11% and 14% less extraction efficiency for chlorophyll in comparison with UAE and SWB respectively. The difference in chlorophyll can be explained by the influence of heat on chlorophyll [55]. In terms of antioxidant capacity, there was no significant impact on FRAP, but SWB yields better CUPRAC, ABTS, and DPPH than MAE and UAE. We further compared the effectiveness of the extracts prepared from MAE, UAE, and SWB techniques on the selected bacteria ( Table 5). The extracts do not possess inhibitory actions against tested gram-negative bacteria Escherichia coli and Enterobacter aerogenes. All the extracts at the concentrations of 15 and 20 mg/mL show antibacterial activities against Staphylococcus lugdunensis and Bacillus cereus. The bacterial growth inhibition zone size was increased with increasing extract concentration for both bacteria. The inhibition zone for both concentrations was significantly lower compared to the control ciprofloxacin (5 µg/disc). A similar finding was reported by Pham, et al. [56], where they found with increasing the applied phenolic extract concentration the growth inhibition zone also increased and the growth inhibition zone for Ciprofloxacin (5 µg/disc) was significantly bigger than that of the extract.
The S. lugdunensis growth inhibition zone ranged from 8.83 ± 0.81-9.42 ± 0.49 mm and 10.17 ± 0.26-10.58 ± 0.38 mm for the sample concentrations of 15 and 20 mg/mL respectively. On the other hand, B. cereus growth inhibition zone ranged from 8.75 ± 0.61-8.58 ± 0.38 mm and 9.83 ± 0.26-9.33 ± 0.26 mm for the sample concentrations of 15 and 20 mg/mL respectively. The bacterial growth inhibition activities of phenolic compounds are influenced by the molecular structure of the polyphenol, the position of the active group, and the tested dose of the compound [57]. Polyphenols prevent bacterial growth by disrupting the function and structural integrity of membranes, denaturing proteins, and altering the intercellular pH of bacteria [58,59]. Our findings are in agreement with a previous study reported by Cock [60], who found antibacterial activities of methanol extract of lemon myrtle against B. cereus (7.6 ± 0.3 mm) upon applying a concentration of extract of 15 mg/mL. He also noticed 6.0 ± 0 mm growth inhibition zone for E. coli at the same extract concentration. Alderees et al. [61] applied lemon myrtle extracts obtained using various solvents and they found that only hexane-derived extract exhibits antimicrobial activities against E. coli. Mehmood and Murtaza [62] reported for the same bacterial strain the growth inhibition capacity of an extract varies with extraction solvent. The total phytochemical yield, its composition, and properties vary with extraction solvent and conditions [28,63,64]. The antibacterial properties of the extracts prepared from MAE, UAE, and SWB are not significantly different. This could be due to the similarity of phenolics and antioxidant properties in the extracts obtained from these extraction techniques. In general, the two novel extraction techniques UAE and MAE do not have significantly higher extraction efficiency for phenolics, and chlorophyll as compared to the conventional technique using the SWB. Their antioxidant capacity and antibacterial properties are also not significantly different. It should be noted that MAE can be conducted with a much shorter time (6 min) than UAE and SWB (both with 50 min). Especially, the volume of solvent for MAE is six-times less (6:100 g/mL) than that of UAE and SWB (ratio 1:100 (g/mL). This is the advantage of MAE; however, the cost of set up remains a major challenge for this technique.

Conclusions
The optimization tool RSM was applied to maximize the extraction yield of phenolics and antioxidant properties from lemon myrtle leaves. The optimal MAE conditions generated by RSM models were validated by the actual experiment and are: radiation time of 6 min, power of 70% (630 W), and sample to solvent ratio 6 g/100 mL. In comparison with UAE and SWB, MAE extract has a similar content of phenolics, antioxidant properties, and antibacterial activities. However, the extraction time of MAE is eight-times less and the volume of the solvent used is six-times less than those of UAE and SWB. Therefore, MAE has the economic advantage of short extraction time with less solvent as compared to UAE and SWB. However, the cost of setting up a microwave extraction system could be an issue on an industrial scale. Future studies are recommended to further study the impact of MAE on extractions of major individual compounds from lemon myrtle leaves, and compare the extraction yield with other eco-friendly novel extraction methods such as supercritical fluid extraction and enzyme-assisted extraction.