Comparison of heat and mass transfer of different microwave‐assisted extraction methods of essential oil from Citrus limon (Lisbon variety) peel

Abstract Dried and fresh peels of Citrus limon were subjected to microwave‐assisted hydrodistillation (MAHD) and solvent‐free microwave extraction (SFME), respectively. A comparison was made between MAHD and SFME with the conventional hydrodistillation (HD) method in terms of extraction kinetic, chemical composition, and antioxidant activity. Higher yield results from higher extraction rates by microwaves and could be due to a synergy of two transfer phenomena: mass and heat acting in the same way. Gas chromatography/mass spectrometry (GC/MS) analysis did not indicate any noticeable differences between the constituents of essential oils obtained by MAHD and SFME, in comparison with HD. Antioxidant analysis of the extracted essential oils indicated that microwave irradiation did not have adverse effects on the radical scavenging activity of the extracted essential oils. The results of this study suggest that MAHD and SFME can be termed as green technologies because of their less energy requirements per ml of essential oil extraction.


Introduction
Citrus is a genus of the Rutaceae family, and it is one of the most common tree fruit crops in the world, with an annual production of approximately 123 million tons in 2010. Citrus is a genus comprised of several important species, the most relevant of which is orange. On a worldwide basis, oranges constitute 56% of the total Citrus crops. Tangerines, mandarins, and clementines make up 17%. Lemons (Citrus limon) and limes comprise 11%, whereas grapefruit (with pomelo) comprises only 6% of the total Citrus crops produced worldwide (Abbate et al. 2012). C. limon is the third most important cultivated citrus species, after orange and mandarin, with a global production of 4,200,000 metric tons each year (Boluda-Aguilar and López-Gómez 2013). Among C. limon cultivars, "Lisbon" is a major variety (Ladaniya 2008).
Today's health care issues and its pertinent concerns have driven man to utilize natural substances in a wide array of industrial fields. Essential Oils (EOs) are one of the most renown of those natural substances. There are hundreds of EOs and complex aromatic substances which are widely used in foods, drugs, and cosmetics (Luque de Castro et al. 1999). The main advantage of EOs is that they are generally recognized as safe (GRAS) and can be used in any foods, as long as their maximum effects are attained with the minimum change to the

ORIGINAL RESEARCH
Comparison of heat and mass transfer of different microwave-assisted extraction methods of essential oil from Citrus limon (Lisbon variety) peel organoleptic properties of the food (Viuda-Martos et al. 2008). EOs obtained from C. limon have been industrially applied in many foods and beverages so far. Each lemon fruit gives 30-40% peel with respect to the whole fruit. Since C. limon EOs are mainly located in the fruit peel, their extraction is economically affordable, because the fruit peel is considered as a waste for the fruit juice industry (Lucker et al. 2002;Settanni et al. 2012).
Cold-pressed EOs from the peels are the first by-products to be obtained during the processing of citrus fruits and any improvement in this process would be of great interest for the citrus processing industry (Coll et al. 1995). Another classical production method of citrus EO is hydrodistillation (HD) which yields a volatile oil from citrus, mainly consisting of monoterpenes and a few sesquiterpenes plus oxygenated derivatives (alcohols, aldehydes, ketones, acids, phenols, lactones, acetals, ethers, and esters) (Luque de Castro et al. 1999;Pourmortazavi and Hajimirsadeghi 2007). However, this method is accompanied by several disadvantages such as losses in the volatile compounds, longer extraction times, degradation of some components through thermal and hydrolytic effects, and the questionable quality of the final product (Bayramoglu et al. 2008;Rezvanpanah et al. 2008). Therefore, in the last few years there has been an increasing demand for novel process technologies. Extraction is a new concept to meet the challenges of the 21st century, to protect both the environment and consumers, and in the meantime enhance competition of industries to be more ecologic, economic, and innovative (Chemat et al. 2012). Pioneering ideas led to the understanding that pollution and hazards have to be eliminated right off when released from the pollution source, thus reducing environmental impacts and costs. Novel microwave-assisted extraction (MAE) (Kaufmann and Christen 2002) methods such as microwave-assisted hydrodistillation (MAHD) (Golmakani and Rezaei 2008a,b;Wang et al. 2010) and solvent-free microwave extraction (SFME) (Filly et al. 2014) have proven to be fast and efficient methods for extracting EOs from medicinal plants.
Microwave-assisted hydrodistillation is an advanced HD method which makes use of a microwave oven. The efficiency of MAHD strongly depends on the dielectric constant value of water and the matrix (Rezvanpanah et al. 2008). MAHD has been used for extracting EOs from Zataria multiflora Boiss. (Golmakani and Rezaei 2008b) and Mango (Mangifera indica L.) flower (Wang et al. 2010). Solvent-free microwave extraction is based on the combination of microwave heating and dry distillation. In contrast to MAHD, SFME does not use a large quantity of water (Lucchesi et al. 2004). Solvent-free microwave extraction has been employed to obtain EOs from ajowan, cumin, and star anise (Lucchesi et al. 2004). Other plant materials that have had their EOs extracted via SFME are cardamom seeds (Lucchesi et al. 2007) and laurel (Bayramoglu et al. 2009;Uysal et al. 2010). MAE (which includes MAHD and SFME) is now recognized as an efficient technique that can reduce the extraction time dramatically. It can increase the yield and the quality of EOs, reduce solvent consumption, lessen pollution, and reduce sample preparation costs (Farhat et al. 2011).
Even though many studies have reported the extraction of EOs from C. limon, none of them are based on the use of microwave energy for the extraction of EOs from fresh and dried specimens. Therefore, the aim of this work was to use the MAHD and SFME techniques for the extraction of EOs from dried and fresh C. limon peels, respectively. Another attempt was to compare their extraction time, extraction yield, physical constants, chemical compositions, antioxidant activities, and energy consumptions with those of the conventional HD method.

Plant materials
The lemons used in this study were C. limon which were gathered in March 2012 from an experimental plantation located in the Jahrom region, Fars province, Southern Iran (28°50´ N latitude, 53°56´ E longitude). The genus and cultivar of the experiment's C. limon were further approved by a plant taxonomy expert in the Department of Horticultural Science, College of Agriculture, Shiraz University, Shiraz, Iran. The "Lisbon" cultivar of C. limon was peeled by hand so as to separate the external part of the C. limon (flavedo), rearing a peel yield of 34.58 ± 3.14% (w/w) with respect to the whole fruit. The C. limon peel was then placed on a large screen tray for 3 days and was left to dry under ambient conditions (30-40°C). The dried peels were then packed in polyethylene (PE) bags and kept in a dark and cool place until extraction was due. Moisture contents of fresh and dry C. limon peels were measured in triplicates according to AACC (1983) method 44-19, using a laboratory oven at 105°C until constant weight was achieved. The moisture contents of fresh and dry peels were 84.78 ± 0.79 and 5.33 ± 0.96% (w/w), respectively. All values are reported on a moisture-free basis.

Microwave-assisted hydrodistillation
In employing MAHD, we used a domestic microwave oven (ME3410W, Samsung Malaysia Electronics, Kuala Lumpur, Malaysia) with a wave frequency of 2450 MHz. The input power consumption of microwave oven was monitored using a separate Wattmeter at the entrance of the electrical power supply and the energy consumption was monitored using the designed software. The average input power of microwave oven was 1200 W. The dimensions of the PTFE-coated cavity of the microwave oven were 23.9 × 37.5 × 38.6 cm. The microwave oven was modified by drilling a hole at the top. A flask with a capacity of 1000 mL was placed inside the oven and was connected to the Clevenger-type apparatus through the hole. Then, the hole was closed with PTFE to prevent any loss of the heat inside.
Fifty grams of dried C. limon peel and 450 mL of distilled water (peel-to-water ratio of 1:9) were placed in the reactor and heated by microwave irradiation with 1200 W (100% power) for 15 min. The different densities and their immiscibility required that the water and EO be separated from each other and the excess water be refluxed to the extraction vessel (Rezvanpanah et al. 2008) in order to provide uniform conditions of solid-to-liquid ratios for extraction. Every 5 min (i.e., 5, 10, and 15 min), the collected EO was decanted from the condensate. Yields are reported in grams of EOs per 100 g of dried C. limon peel.

Solvent-free microwave extraction
Solvent-free microwave extraction was employed via a similar oven as the one described for MAHD. The SFME is based on the combination of microwave heating and distillation without any water being added. In this method, fresh C. limon peels were placed in the microwave oven. The internal heating of the in situ water within the C. limon peel distends the EO glands and makes them burst, thus freeing the EOs, which are subsequently evaporated along with the in situ water content of C. limon peels. The EO then passes through a condenser outside the microwave cavity where it is condensed. The distillate is collected continuously and forms a two-phased liquid content (i.e., EO and in situ water) (Uysal et al. 2010).
As for SFME, fresh C. limon peels and distilled water, which were of a peel-to-water ratio of 1:1, were placed in the microwave reactor using a fixed power of 1200 W. Even though SFME is considered to be a solvent-free extraction method, we used an equal ratio of water to fresh C. limon peels in order to prevent the peels from burning and to ensure that sufficient amounts of water would subsist during the refluxing. The extraction time was fixed at 15 min until no more EO could be obtained further. During this period, the gathered EO was decanted from the condensate in 5 min intervals (i.e., 5, 10, and 15 min).

Hydrodistillation
Hydrodistillation was employed more or less like MAHD, but an Electromantle heater (EM2000/C, 335 W, Electrothermal Engineering Ltd., Rochford, UK) was used instead of the microwave oven. C. limon peels and distilled water were placed into the HD with a Clevenger-type apparatus, and EOs were extracted for 120 min, in 30 min intervals (i.e., 30, 60, 90, and 120 min). The EOs obtained from the three extraction methods were collected, dried with anhydrous sodium sulfate, and stored in amber vials at 4°C until further analysis.

CO 2 Emission
The measurements of CO 2 emitted were carried out based on the procedures mentioned in the previous studies: to obtain 1 kWh of energy from coal or fossil fuels, 800 g of CO 2 will be released into the atmosphere during their combustion (Ferhat et al. 2006).

Physical constants
The usual physical constants (specific gravity, refractive index, color, and visual appearance) of the EOs from the C. limon peels extracted by the three extraction methods (HD, MAHD, and SFME) were analyzed according to the methods outlined by the Food Chemical Codex (FCC) (Burdock 2010). The specific gravity and the refractive index were measured at 25 and 20°C, respectively. Color of the EOs was defined by a number of parameters: L* (lightness), a* (redness-greenness), and b* (bluenessyellowness). These were determined according to the method described by Afshari-Jouybari and Farahnaky (2011). In the L*a*b* color space, color difference can be expressed as a single numerical value, ΔE* ab , which indicates the size of the color difference but not in what way the colors are different. ΔE* ab of the EOs was calculated according to the following equation (Nielsen 2010): (1)

Scanning electron microscopy
In order to elucidate each MAE procedure and to understand the extraction mechanism, samples were ready to be scanned by electron microscopy when the extraction procedure had finished. For the MAHD and SFME, the scanning was done after a time of 15 min had elapsed from the start of the extraction process. The C. limon peels were freeze-dried using a freeze drier (Armfield, UK) and fixed on the specimen holder with aluminum tape and then sputtered with gold in a Polaron SC 7640 Sputter Coater (Quorum Technologies Ltd., Newhaven, UK), All the specimens were examined with a SEM (Cambridge, UK) under high-vacuum conditions at an accelerating voltage of 20.0 kV and a working distance of 15 mm (i.e., the distance between the surface of the sample and the microscope lens).

Gas chromatography/mass spectrometry (GC/MS) identification
The EO constituents obtained via the different extraction methods (HD, MAHD, and SFME) were identified by a gas chromatography (7890A, Agilent Technologies, Santa Clara, CA) coupled with a mass spectrometer (5975C, Agilent Technologies, Santa Clara, CA) operating at 70 eV ionization energy, 0.5 s/scan, and a mass range of 35-400 atomic mass unit (amu), equipped with a HP-5MS capillary column (5% Phenyl Polysilphenylene-siloxane; 30 m length; 0.25 mm internal diameter; 0.25 μm film thickness). The injector and detector temperatures were maintained at 280°C. The temperature timeline of the oven was programmed as follows: the initial temperature was set to be 60°C and was then allowed to increase to 210°C (at a rate of 3°C/min). Finally, the temperature increased to 240°C (at a rate of 20°C/min) whereupon 240°C was held constant for 8.5 min. One microlitre of the sample was injected into the GC/MS with the injector being in the split mode (split ratio of 1/100). Helium was used as the carrier gas with a flow rate of 0.9 mL/min. Data in relative percentages were obtained via the electronic integration of peak areas without the use of correction factors. Then, the MSD ChemStation Software (G1701EA, E.02.01.1177, Agilent Technologies, Santa Clara, CA) was used to handle mass spectra and chromatograms. Most constituents were tentatively identified by comparison of their retention indices (RIs), established in accordance with reference to an homologous series of C 5 -C 28 n-alkanes that had been injected following the injection of the EOs under the same chromatographic conditions. Identifications were confirmed when possible by comparisons of their mass spectral fragmentation patterns with those stored in the database bank (Wiley libraries/NBS) and with the mass spectra of relevant literature data (Adams 1995).

Antioxidant activity: DPPH radical scavenging activity
The antioxidant activities of the EOs from C. limon peels were measured by considering the hydrogen donating molecules or the EO's radical scavenging ability, using the stable DPPH free radical (Brand-Williams et al. 1995). The effect of an antioxidant on DPPH radical manifests itself in the antioxidant's hydrogen donating ability or radical scavenging activity. Mixing the DPPH radical solution with a substrate that donates hydrogen atoms can turn DPPH radical into its reduced (nonradical) form and causes the simultaneous transformation of its color from violet to pale yellow. DPPH radical scavenging activity is shown by the IC 50 value, defined as the concentration of the antioxidant required for the DPPH radical activity to diminish by 50% (Mazidi et al. 2012). Four ml of various concentrations of the EOs (2, 4, 6, 10, and 50 mg/mL), dissolved in methanol, were added to 2 mL of the 0.2 mmol/L methanolic solution of DPPH radical. The mixtures were shaken vigorously and were then left in the dark at room temperature for 60 min. The absorbance was measured at 517 nm against the blank (methanol). The DPPH radical discoloration of the samples was calculated in percentages according to the following equation (Kulisic et al. 2004): (2) The EO concentration that provides 50% inhibition (IC 50 ) was determined through the inhibition percentages plotted against EO concentrations. All the determinations were carried out in triplicates. BHT and vitamin C were used as a positive control against antioxidant activity.

Statistical analysis
Statistical analyses of data (differences among HD, MAHD, and SFME) were determined via SAS (Statistical Analysis Software, Version 9.1; SAS Institute Inc., Cary, NC). Evaluations of chemical and physical constants of the extracted EOs were done in triplicates. A general linear model (GLM) procedure was performed to determine significant differences among the three extraction methods (P < 0.05).

Results and Discussion
Temperature profile Figure 1 shows the temperature profile during extractions by HD, MAHD, and SFME from EOs of C. limon peels. In all extraction methods, the initial temperature of samples was 20°C. The extraction temperature was equal to the boiling point of water (100°C) at atmospheric pressure as regards HD, MAHD, and SFME. The first EO droplets were observed after 23.0 min in HD, 3.5 min in MAHD, and 3.0 min in SFME. The most important reason for this difference is that MAHD and SFME apply three ways of heat transfer within the samples, namely irradiation, conduction, and convection. On the other hand, heat transfer in HD takes place through conduction and convection only (Golmakani and Rezaei 2008a,b). The rapid temperature rise in MAHD and SFME is the reason behind the time reduction accordingly.
The rate of temperature rise was measured by determining the slope of the linear part of the temperature profile (Fig. 1). The details in Table 1 show that the rates of temperature elevations in the MAHD and SFME methods were greater by sevenfold compared to that of the HD method. This phenomenon can also be attributed to the high dielectric constant of water which absorbs the irradiation from the microwaves and causes a more rapid rise in temperature, compared to the case of HD (Kaufmann et al. 2001). These results are in good agreement with the findings of Mazidi et al. (2012). They found that in comparison to the conventional HD, the MAHD method can accelerate the rate of extraction by increasing the temperature rapidly and by causing the quicker rupturing of EO glands in Black Zira. Also, Ferhat et al. (2006) showed that heating the samples for 3 min was enough for the SFME to reach the extraction temperature (100°C) and to obtain the first EO droplets from orange peels. This is in comparison with the 30 min required by the HD. Table 1 shows the effect of different extraction methods on total extraction time, extraction duration (i.e., the difference between the total extraction time and the time when the first droplets of EOs begin to appear), yield and rate of EO accumulation. MAHD and SFME were clearly quicker than the conventional HD. Full recovery of EOs was achieved within the first 15 min of operation in MAHD and SFME, whereas it took at least 120 min for the HD to fulfill the extraction operation. Farhat et al. (2011) found that less time was needed for EOs to be extracted thoroughly from orange peels via microwave extraction (12 min) than via the HD (40 min). Also, Bousbia et al. (2009) presented similar findings in their study on EOs extracted from lime wherein it took 15 min for the microwave hydrodiffusion and gravity method to accomplish the extraction process, in comparison with HD which took 180 min.

Extraction kinetics
As is shown in Table 1, there were no significant differences in the final yields obtained by HD (1.22 ± 0.14% w/w) after 120 min, MAHD (1.18 ± 0.08% w/w) and SFME (1.36 ± 0.06% w/w) after 15 min. Since the EO quantity of the samples was constant and since we had no EO loss caused by evaporation, the final yields in all the three extraction methods (HD, MAHD, and SFME) were equal. Bayramoglu et al. (2009) extracted EO from Laurel and reported that there were no significant differences in the maximum EO yields extracted by HD and SFME. Furthermore, Golmakani and Rezaei (2008a) investigated the effect of different extraction methods on EO yields of Zataria multiflora Boiss. They found that the final yields in HD and MAHD were 3.44 and 3.66%, respectively, and that there were no significant differences among the yields obtained through HD and MAHD.
The rates of EO accumulation observed in HD, MAHD, and SFME are shown in Table 1. This index was obtained  by dividing the amount of extracted EOs (g) by the corresponding total extraction time (min), which equaled the average rates of EO accumulation (g/min). The results show that the average rates of EO accumulation by MAHD and SFME were at least eight times greater than that of HD. Additionally, the findings revealed that the shorter extraction time in MAHD and SFME was not only caused by their earlier onset of extraction, but also by their higher extraction rates which is mainly due to the more efficient heat transfer conducted by the microwave (Golmakani and Rezaei 2008a). According to Figure 2, the extraction yields by MAHD and SFME were 1.14 ± 0.08 and 1.31 ± 0.04% w/w, respectively, after 5.0 min. These yields were significantly similar to the amount of yield obtained by the traditional HD after 60.0 min. Therefore, the extraction time in HD was 12 times lengthier than that of the MAE (MAHD and SFME). Microwave extraction offers a rapid delivery of energy to a total volume of water and also to the C. limon peel matrix with a subsequent heating of the water and the C. limon peel matrix. This delivery of energy via the microwave occurs efficiently and homogeneously. Since the water within the C. limon peel matrix absorbs microwave energy, cells are ruptured by internal superheating which facilitates diffusion of chemicals from the matrix, thus improving the recovery of EO.
As it is shown in Figure 2, the extraction patterns for the three methods were similar, and two phases were observed in the process of extraction kinetics. The first part (I and I') was presented by an ascending line, which denotes the rapid increase in the yield and which represents approximately 85.24, 96.61, and 96.32% of the total yield in HD, MAHD, and SFME, respectively. In the second part, II and II' correspond to a horizontal line which marks the end of the extraction process. The rapid increase in the yield during the first step suggested that the EO was easily accessible by the steam. Indeed, the microwave irradiations distended the C. limon peel and lead to the rupture of the glands. However, one of the most striking differences observed between the MAE (MAHD and SFME) and HD methods is the ability of the MAE process to raise the extraction yield of the sample quickly and notably, within a short time. This higher rate of yield is a result of the higher extraction potential of microwaves and could be due to a synergy combination of the two transfer phenomena -mass and heat -acting in the same way. This could be explained by the fact that the mass transfer occurs from the inside to outside in the HD, MAHD, and SFME methods (Fig. 3G-I). The rate of heat transfer differed among the MAE (MAHD and SFME) and the HD. In the case of the HD, heat transfer occurred from the outside to the inside, exclusively because of conduction and convection happening through the water surrounding the C. limon peels (Fig. 3D). However, we can suggest that the extraction mechanism of EO obtained by MAHD is partly due to internal heating of in situ water under microwaves irradiation from the inside to the outside of C. limon peels, and also mostly due to heat transfer from the outside to inside, similar to the case of HD (Fig. 3E). In SFME, heat transfer partly occurred from the outside to the inside and mostly from the inside to the outside of the C. limon peel, which facilitates oil diffusion from the inside of the peel via steam by an increase in the extraction yield due to the synergy combination of the two transfer phenomena -mass and heat -acting in the same direction (i.e., from the inside to the outside) (Fig. 3F). SFME resulted in significant internal heating, thus creating significantly higher internal pressures which promote the bursting of glands and EO extraction from C. limon peel (Bayramoglu et al. 2008). This is in agreement with the findings of Farhat et al. (2010) who extracted EOs from caraway seeds using microwave dry-diffusion and gravity. They reported that the extraction mechanism of EO by microwave dry-diffusion and gravity is partly due to the internal heating of EO molecules under microwave irradiation from the inside to the outside of caraway seeds without the condition of in situ water and also partly due to a synergy combination of the two transfer phenomena -mass and heat -acting in the same direction.

Structural changes after extraction
The images from the surfaces of C. limon peels obtained by SEM after MAHD and SFME are shown in Figure 4. Both MAHD and SFME resulted in apparent physical changes in the glands of C. limon peels after 15 min. While the EO glands of MAHD and SFME were destroyed totally and extractions were done thoroughly, water temperature in HD (72°C) did not reach the boiling point when the extraction had not started yet. These observations confirm that microwave irradiation has a stronger destructive effect on the oil-bearing structures of C. limon peel. Also, microwave irradiation causes the glandular walls to crumble or rupture more rapidly and more efficiently. These results are in good agreement with the findings of Rezvanpanah et al. (2011) in the extraction of EOs from Satureja hortensis when using MAHD and HD as extraction methods. They showed that MAHD operates faster in destroying EO-bearing glands and, as a result, EO diffuses faster into the medium than HD. Lucchesi et al. (2007) extracted EO from cardamom through HD and SFME and presented similar findings. They reported that microwaves seem to cause the rupture of EO glands more rapidly than the conventional HD. In the case of SFME, when the glands were under great thermal stress and when localized high pressures were induced by microwave heating, the pressure build-up within the glands could have exceeded their capacity for expansion, and hence their rapid rupture compared to the control experiment. This is the mechanism postulated by Chen and Spiro (1995) for the MAE of rosemary leaves. Table 2 lists the mean values for the physical constants (refractive index, specific gravity, visual appearance, and color) of EOs extracted from C. limon peels by HD, MAHD, and SFME. There were no significant differences between the EOs extracted through HD, MAHD, and SFME methods in terms of their specific gravities and refractive indices. However, the visual appearance of the EOs extracted by MAHD and SFME were somewhat brighter than that obtained by HD. A brighter visual appearance in the case of MAHD and SFME was also confirmed by their higher L* and lower b*. Also, specific gravity, refractive index, and visual appearance perceptions of all the samples were within the range indicated by FCC standards (Burdock 2010). Bayramoglu et al. (2008) observed no significant difference between HD and SFME in terms of specific gravities and refractive indices of EOs extracted from oregano. Golmakani and Rezaei (2008a,b) found that specific gravities and refractive indices of thyme EOs extracted by HD and MAHD were similar. Also, they reported that the color of EOs extracted by MAHD was brighter than that of HD. ΔE ab of EOs extracted by MAHD and SFME were 7.65 and 3.54, respectively, relative to the HD as reference, which indicates a clear similarity in the color of EOs extracted by HD and MAE. Therefore, the examination of MAHD and SFME as novel and fast extraction methods do not bear any remarkable standouts in terms of the measured physical constants of the extracted EOs.

Evaluation of physical constants
Antioxidant activity: DPPH radical scavenging activity Figure 5 shows the DPPH radical scavenging activities of different concentrations of EO obtained from C. limon peel, extracted via the various extraction methods applied in this study. The radical scavenging activity increased as the concentration of EOs increased. According to Table 2, IC 50 values of the EOs extracted by HD, MAHD, and SFME were 44.06, 42.03, and 97.23 mg/mL, respectively. In comparison with HD and MAHD, the lower scavenging activity (higher IC 50 value) of SFME is related to its lower concentrations of antioxidant constituents such as limonene. In contrast to our results, Mazidi et al. (2012) showed that there were no significant differences between the IC 50 values of Black Zira EOs extracted by Electromantle and the microwave oven. Also, the antioxidant properties relating to the EOs of C. limon peels, extracted by HD, MAHD, and SFME were insubstantially compared with that of known synthetic antioxidants, BHT, and vitamin C (Table 2). Furthermore, the IC 50 of EOs extracted by the three methods measured significantly higher than the IC 50 of BHT and vitamin C. A lower IC 50 value reflects a better protective action (Ayoughi et al. 2011). Therefore, radical scavenging activity of C. limon EOs was significantly lower than those of BHT and vitamin C. Table 3 lists the retention indices, molecular weights, molecular formulas, IUPAC (International Union of Pure and Applied Chemistry) names, CAS (Chemical Abstract Table 2. Physical constants and IC 50 of essential oils extracted from Citrus limon peels by hydrodistillation (HD), microwave-assisted hydrodistillation (MAHD), and solvent-free microwave extraction (SFME). In each row means with different letters are significantly different (P < 0.05). 3 L*: lightness; a*: redness-greenness; b*: blueness-yellowness.  Retention index (RI) relative to C 5 -C 28 n-alkanes on the HP-5MS column. 5 Mean ± SD (n = 3). 6 Hydrodistillation. 7 Microwave-assisted hydrodistillation. 8 Solvent-free microwave extraction. 9 Not detected. GC/MS results indicated that the EOs extracted by HD, MAHD, and SFME were quite similar in their composition. Based on the values obtained by GC/MS, it was concluded that there was no significant difference in the quantity of the constituents with relative peak areas more than 2%, with a total percentage of 86.36-88.93% in the different extraction methods. A monoterpene called limonene was the most abundant constituent of the EOs extracted from C. limon peels by HD (63.15 ± 4.47), MAHD (61.62 ± 4.36%) and SFME (58.58 ± 4.14%). The EOs also bore γterpinene (11.19-12.71%), βpinene (8.56-9.01%), β-Myrcene (2.66-3.43%), and α-Pinene (2.64-3.11%), regardless of the three extraction methods. Michaelakis et al. (2009) reported similar findings for the constituents in EOs from Citrus. They showed that limonene was the most abundant constituent in the EOs extracted from the fruit peel of orange (96.2%), lemon (74.3%), and bitter orange (96.7%) when extracted via HD. Also, Ferhat et al. (2006) showed that limonene was the most abundant constituent in the EOs extracted from orange peels, with equivalent relative amounts in both extraction methods: 78.5% for HD and 76.7% for SFME.

GC/MS
EOs extracted by HD showed lower amounts of acetate constituents (1.76%) compared to those isolated by MAHD (1.99%) and SFME (2.14%). This result can be attributed to the fact that the higher extraction time of HD had hydrolytic effects on acetate constituents. Results obtained by GC/MS show in this current study that MAE did not cause the deterioration or loss of volatile components in comparison with HD. Therefore, MAE can be recommended as superior method for the extraction of EO.

Electric consumption and environmental considerations
The reduced extraction time is clearly advantageous for the proposed MAHD and SFME methods in terms of cost and energy. The energy requirement needed to perform the extraction methods, based on the maximum power consumptions of the Electromantle (in HD) and the microwave oven (in both MAHD and SFME), considering the total periods of full extractions, was 0.67 kWh for HD and 0.30 kWh for MAHD and SFME (Fig. 6). Li et al. (2013) showed that for the energy requirements, SFME needs less than 0.5 kWh for normal performance but conventional methods expend more than 4.5 kWh. Relative electric consumption for the production of 1 g EO in HD, MAHD, and SFME was 0.55, 0.25, and 0.22 kWh/g EO, respectively (Fig. 6). This indicates a substantial saving in the extraction cost when using MAHD and SFME instead of HD. Filly et al. (2014) reported that the energy required for the extraction of EOs from rosemary was 4.50 kWh per gram of EO in HD, but 0.25 kWh per gram of EO in SFME.
Regarding the environmental impact of pollution, the calculated quantity of CO 2 emitted in the atmosphere was higher in the case of HD (0.533 kg CO 2 ) than those of MAHD and SFME (0.240 kg CO 2 ) (Fig. 6). Golmakani and Rezaei (2008a) found that the amount of CO 2 which was released into the atmosphere was higher in HD (1600 g CO 2 ) than that in MAHD (990 g CO 2 ). Relative amounts of CO 2 emissions that result from the production of 1 g EO were higher in HD (0.437 kg CO 2 /g EO) than those in MAHD (0.203 kg CO 2 /g EO) and SFME (0.176 kg CO 2 /g EO) (Fig. 6). This finding further indicated that there was a significant difference between MAHD and SFME in terms of the amount of CO 2 released into the atmosphere for the production of 1 g EO. Also, waste water was lower in SFME than HD and MAHD. However, there was a significant difference between MAE and HD in terms of the CO 2 released into the atmosphere for the production of 1 g EO. Filly et al. (2014) showed that the amount of CO 2 released into the atmosphere was dramatically higher in HD (3600 g CO 2 /g EO) than that in SFME (200 g CO 2 /g EO). Also, Li et al. (2013) reported that the quantity of CO 2 calculated to be rejected in the atmosphere is higher in the case of HD (3600 g CO 2 /g EO) than for SFME (200 g CO 2 /g EO).

Conclusion
The aim of this work was to confirm the efficiency of MAE methods (SFME and MAHD) and to explain how they speed up the extraction process, without causing considerable changes in the EO composition. For this purpose, the experiment's environmentally friendly processes operated rapidly for the extraction of EOs from dry and fresh C. limon peels. Both MAHD and SFME resulted in a reduced extraction time compared to the conventional HD technique. No significant differences were found in the physical constants (refractive index, specific gravity, visual appearance, and color) of EOs extracted by HD, MAHD, and SFME. Furthermore, the EOs extracted by the three extraction methods had the same compositions. The antioxidant activities of EOs from C. limon peel that had been extracted by HD, MAHD, and SFME were lower than those of BHT and vitamin C. SEM images of C. limon peels that had undergone MAHD and SFME indicated that microwave heating causes a quick rupture of EO glands and their surrounding areas, resulting in a shorter extraction process. The amount of CO 2 emission -a result of the EO extraction process -was dramatically higher in HD than those of MAHD and SFME. Based on our results, MAHD and SFME can be termed as "green" extraction methods (from an energy consumption point of view). In addition to that, MAHD and SFME can also be proposed to be utilized for largescale productions of EOs by commercializing the equipment instead of the conventional HD apparatus.