Boosting the essential oil yield from the rhizomes of cassumunar ginger by an eco-friendly solvent-free microwave extraction combined with central composite design

ABSTRACT

This study aimed to apply the solvent-free microwave extraction (SFME) for isolating the essential oil from the rhizomes of cassumunar ginger and study the kinetic mechanism during the extraction process by light microscopy. A central composite design was employed to evaluate the influence of the key variables including microwave power and irradiation time on the oil recovery. The optimal process conditions of SFME were irradiation of the sample at 700 W for 70 min. Microwave power had the most significant effect on the oil recovery (p < 0.001), followed by irradiation time (p < 0.01). The major volatile constituents in the oil obtained under the optimal conditions were sabinene, terpinen-4-ol, and (E)-1-(3,4-dimethoxyphenyl) butadiene with a higher amount of the oxygenated compounds compared to those isolated by a conventional hydrodistillation (HD). Isolating the essential oil by SFME showed a complete extraction with higher plant cell rupture and was quicker than HD.

KEYWORDS:

• Cassumunar ginger
• central composite design
• essential oil
• green chemistry
• solvent-free microwave extraction

1. Introduction

The search of green extraction technology with superior efficiency (ie cost-effective, sustainable and capable of producing the essential oil with the same characteristics) and more environment friendly (ie reduction of CO₂ emission and waste water) is very important in all countries worldwide to meet the challenges of the 21^st century (1–3). The solvent-free microwave extraction (SFME), an innovative clean and efficient technique, has recently gained popularity for isolating the essential oils of different plant parts (bark, flower, leaf) of many scented plants (2–8). The principle in the extraction of this technique combines both microwave heating and dry-distillation for extraction the volatiles at the atmospheric pressure (7). In SFME processes, the in-situ water present inside the tissue of the fresh plant material or a small amount of water added to the dry samples is heated by microwave energy due to dipole rotation, followed by increasing the localized high pressure, which leads to the bursting of the oil cells and the evaporation of the essential oil by azeotropic distillation (8). Classical extraction techniques like hydrodistillation (HD), steam distillation and solvent extraction are high energy-consuming methods with long processing times, solvent consumption or toxic solvent residues (3), which consequently induce further costs (7).

One of the important traditionally used aromatic plants in tropical Asian countries especially Bangladesh, Cambodia, India, Malaysia, Myanmar, Laos and Thailand with increasing demand at international markets for medicinal, cosmetic and aromatherapy utilization, is cassumunar ginger [Zingiber montanum (J.Koenig) Link ex A.Dietr., Zingiberaceae family] (9–11). The fresh rhizomes where the numerous oil globules are located, contain 0.57–2.10% v/w essential oil, mainly consisting of sabinene (25–45%), γ-terpinene (5–10%), α-terpinene (2–5%), terpinen-4-ol (25–45%), (E)-1-(3,4-dimethoxyphenyl) butadiene (DMPBD) (1–10%) (12–14) that exhibit the strong anti-inflammatory activity (9,15–19). Because of the potential use of the oil in pharmaceutical industry as well as its well-documented clinical efficacy, the Thai National Drug Committee has approved pharmaceutical products containing cassumunar ginger oil as an alternative source of non-steroidal anti-inflammatory drugs for sprains and muscular aches (20). Many commercial products containing cassumunar ginger oil for muscle pain relief are available on the market. The characteristic pleasant odor with earthy, camphoraceous features and relaxation property also makes the cassumunar ginger oil being of interest among aroma-therapists for Thai massage and spa applications (21). However, the cassumunar ginger oil sold in the market is generally extracted by using conventional methods (hydro- and steam distillation) (information obtained from factory visits). Therefore, it is very important to find out a fast and economic method capable of acquiring a highly pure essential oil from the cassumunar ginger rhizomes. The isolation of cassumunar ginger oil using the SFME has not been reported yet.

In continuation of an on-going work of the authors’ group related to the extraction of the potential natural products for health promotion, this study mainly aimed to apply the SFME for the isolation of the essential oil from cassumunar ginger rhizomes. The two key operating factors (microwave power and irradiation time) of the selected method influencing the essential oil yield were optimized by application of the central composite design (CCD) to obtain the highest possible yield of the target oil. This technique has given a large amount of information without the need of numerous experimental trials and provided the data to evaluate the interaction effects of the test factors and the optimum conditions for the desired responses (22–24). Moreover, the extraction mechanism of the method used was investigated by a microscopic technique to understand the action of the microwave irradiation in more details. Additionally, other quality properties of the isolated essential oil (ie volatile compositions, specific gravity, refractive index and solubility) were determined and compared to those of an essential oil extracted by a conventional HD.

2. Materials and methods

2.1. Materials

The fresh 2-year-old rhizomes of cassumunar ginger were supplied by the traditional Thai Medicine Center, Phana Government Hospital, Amnaj Charoen Province, Thailand. The plant was authenticated by Dr. Bancha Yingngam with the voucher specimen (BCY UBU 2,016,031) deposited at the Herbarium of Ubon Ratchathani University. All chemicals and solvents used were of either analytical or gas chromatographic grade and purchased from Sigma Chemical Co. (St. Louis, MO, USA).

2.2. Sample preparations

The fresh cassumunar ginger rhizomes were washed with tap water and allowed to air-dry at 25 ± 1°C overnight. Prior to perform the essential oil extraction, the rhizomes were sliced and their quality was checked by drying some plant sample in a hot air oven at 50°C for 24 h to obtain moisture content of 10% or below, followed by grinding into a coarse powder. The volatile content was determined according to the method described in the Thai Herbal Pharmacopoeia (THP) (25). In this study, the yield of volatile oil in the test sample was higher than the minimum quantity mentioned in THP (not less than 2.00% v/w). Consequently, the fresh rhizomes containing about 80% moisture were sliced into small pieces and the extraction procedure of the volatiles was performed immediately to minimize the loss of essential oil.

2.3. SFME

The essential oil was isolated by the SFME using a microwave equipped with a Clevenger apparatus (2.45 GHz, MAS-II, Sineo Microwave Chemistry Technology Co., Ltd, Shanghai, China). Briefly, slices (200 g) of fresh plant material were put into a glass flask inside the microwave chamber, which was connected to a Clevenger apparatus and condenser. The microwave power and irradiation time were performed at atmospheric pressure according to the design of experiment (see section: Experimental design). The essential oils obtained were dehydrated with anhydrous sodium sulfate and kept at 4°C for further analysis. Three separate measurements were performed in all experiments.

2.4. Experimental design

Two operational variables influencing the yield of cassumunar ginger oil, including microwave power (X₁) and irradiation time (X₂), were investigated. The response variable was essential oil yield (Y). A 4–factorial, 4–axial 5–centered CCD based response surface methodology was employed in the thirteen randomized experiments to develop a response surface and the optimum condition, study their effects on the desired dependent variable, calculate the experimental error and prove the suitability of the model. Five levels of the independent variables were coded at – 1.41, – 1, 0, + 1 and + 1.41 for statistical analysis. The experimental values were analyzed by the multiple regressions through the least square method and the results were fitted with a polynomial model following equation (Eq. 1)

(1)

Y is the yield of essential oil expressed as mL/100 g of initial fresh matter. β₀, β_i, β_ii and β_ji represent the constant coefficients of intercept, linear, quadratic and interactive terms, respectively. X_i and X_j are the independent variables (microwave power and irradiation time). The adequacy of the models was also tested by the predicted R², adjusted R², and coefficient of variance. The relationship between the independent and response variables was constructed as the three- and two-dimensional plots by employing the fitted model obtained.

2.5. Process optimization and validation

The optimum condition of SFME should provide the highest yield of the target oil with minimizing the extraction time. Thus, a desirability function (d) for the simultaneous optimization of the response was applied to calculate the desired conditions of the model as follows (Eq. 2):

(2)

N, r_i and m are the numbers of response, the importance of a particular response and partial desirability function for a specific response, respectively. The value of d varies in the range of 0 ≤ d ≥ 1. The average experimental response obtained was compared to the predicted value.

2.6. Conventional HD

The conventional HD of the essential oil was also conducted using a Clevenger-type apparatus according to the THP (25) with some modifications. A quantity of 200 g sliced plant material was mixed with 700 mL distilled water and introduced to HD process for 5 h until the volume of essential oil kept constant. The essential oil was collected in the same way as in SFME for further analyses.

2.7. Analyses

2.7.1. Essential oil yield

The yield of cassumunar ginger essential oil isolated from the rhizomes by the SFME and HD method was expressed in term of the volume of the oil collected (mL) per 100 g of initial fresh plant material (% v/w).

2.7.2. Identification of the volatile components

The volatile components in the essential oils obtained from both the SFME and HD were analyzed by gas chromatography-mass spectrometry (GC–MS) (Agilent 7890A and 5975C Mass Selective Detector; Agilent Technologies, Palo Alto, CA, USA) according to the previously reported method (23). A fused silica capillary column 5% phenylmethylsiloxane (HP-5MS, 30 m long × 250 μm ID, 0.25 μm film thickness, Agilent J&W, Lawrence, Kansas, USAAQ: Please supply the missing state and city names here.) was applied as stationary phase. The following chromatographic conditions were used: 60°C to 270°C at 3°C/min; carrier gas helium; flow rate 1 mL/min; injection volume of the sample (cassumunar ginger essential oil in hexane) 1 μL; split ratio 1:50; injection temperature 240°C; detection temperature 280°C. The volatile composition of the samples was identified by comparison of their retention indices relative to a homologous series of C₈–C₂₀ alkanes on HP-5MS capillary column and those values reported in the literature and the mass spectra database (NIST 08 MS and Wiley Libraries). The percentage composition of the essential oil was calculated using the normalization method from peak areas of the total oil.

2.7.3. Light/fluorescent microscope image analysis

The microstructure changes in the cassumunar ginger sample (control and spent after extraction) were studied with a Nikon Eclipse E200 light/fluorescent microscope connected with NIS-Elements BR 4.50 software (Nikon Instruments Inc., Japan). A transverse section of samples was stained with a mixture solution of alcian blue and safranin O (3:2, v/v) and examined for their microstructure alterations under a light/fluorescent microscope. The micrographs were carried out at 40× magnification.

2.7.4. Physical properties of the essential oil

The physical properties of the essential oils isolated from the SFME and a conventional HD were measured. Specific gravity was determined by weighing the samples on a highly sensitive balance (accuracy of ± 0.00001 g) (Mettler Toledo, Im Langacher, Greifensee, SwitzerlandAQ: Please supply the missing city and state names here.). The specific gravity value was then calculated by dividing the weight of 10 µL essential oil by that of 10 µL distilled water. Refractive index measured by the RFM 330 refractometer (Bellingham Stanley Ltd., Kent, England). The solubility of the samples was measured by repeated dilution of 1 mL essential oil with 0.1 mL 80% ethanol until a clear mixture was obtained. All measurements were performed at 20 ± 1°C.

2.8. Statistical analyses

The results are expressed as the mean ± standard deviation of three measurements. The data obtained from CCD were analyzed by analysis of variance for the polynomial model and the accuracy of the regression model. P < 0.05 was considered to be a minimum of statistically significant differences among groups.

3. Results and discussion

3.1. Fitting the model

Determining the accurate optimum conditions of the experimental factors and the interrelation among the multi-factors is limited using the one-factor-at-a-time methodology (22–24,26). Thus, the CCD was applied to the key operating factors for achieving the maximum yield of essential oil and the obtained results are presented in Table 1. The essential oil yields varied between 0 and 1.31% (v/w) depending on the operated SFME conditions. Experiment number 6 had the highest essential oil yield; whereas, the lowest yield was found in the experiment numbers 1 and 5. These observations were in conformity with an earlier report for cassumunar ginger oil isolated from the fresh plant material (12) and their variation was due to the extraction conditions applied.

Table 1. The matrix of CCD applied for isolation of the essential oil from cassumunar ginger rhizomes.

		True value (coded value)		Response
Experimental number	Run order	Microwave power (X1, W)	Irradiation time (X2, min)	Essential oil yield (Y, %)
1	4	200 (–1)	20 (–1)	0.00 ± 0.00
2	7	700 (+ 1)	20 (–1)	0.72 ± 0.02
3	9	200 (–1)	80 (+ 1)	0.23 ± 0.01
4	5	700 (+ 1)	80 (+ 1)	1.12 ± 0.01
5	12	96 (–α)	50 (0)	0.00 ± 0.00
6	13	804 (+ α)	50 (0)	1.31 ± 0.01
7	8	450 (0)	8 (–α)	0.43 ± 0.04
8	1	450 (0)	92 (+ α)	0.85 ± 0.02
9^c	10	450 (0)	50 (0)	0.81 ± 0.01
10^c	2	450 (0)	50 (0)	0.90 ± 0.01
11^c	11	450 (0)	50 (0)	0.81 ± 0.02
12^c	3	450 (0)	50 (0)	0.85 ± 0.02
13^c	6	450 (0)	50 (0)	0.91 ± 0.08

^c Center point.

α = 1.41.

Data expressed as means ± SD, n = 3.

Prior to evaluating the key factors influencing the yields of essential oil, the deleted studentized residuals after model refinement were analyzed to check the normal distribution of the residuals. The results are shown in Figure 1. The values of residuals fit on a straight line to a major extent with a slight scattering of some points of the experimental numbers. This result confirmed that the experimental data were normally distributed and capable to give accurate data. A regression analysis was then applied to the obtained experimental data to fit the mathematical model. The quadratic model was suitable to determine the optimum conditions for extraction of the cassumunar ginger oil and the obtained mathematical equation that predicted the essential oil yield was represented as follows (Eq. 3):

Figure 1. Normal probability plot showing the value of studentized residuals after model refinement. The numbers in the graph indicate the experimental numbers in Table 1.

Yield of essential oil

(3)

Figure 2 was plotted to illustrate the relative importance of the effects and rank the factors influencing the oil yield. The linear terms of microwave power and irradiation time (and ) exhibited a positive effect on the oil yield that is contrary to a negative effect of quadratic terms for both independent factors ( and ). There was no interaction between the microwave power and the irradiation time () on the oil extracting yield (p > 0.05). Thus, changing the microwave power and irradiation time had highly influence on the isolated oil yield.

Figure 2. The scaled and centered coefficient plot after model refinement showing the influence of the test factors ( microwave power, irradiation time) on the yield of cassumunar ginger oil.

The goodness of fit and adequacy of the developed model was assessed by application of the analysis of variance. Ideally, the model with good data prediction should have the absolute F-value higher than the tabulated value of F-distribution at a level of significance α (2,23,24). As summarized in Table 2, the Fisher variation ratio (the ratio between the mean square of the model and the residual error of the experimental values, F-value) was 48.87, indicating the significant model (p < 0.0001). There was only a 0.01% chance that a model F-value could occur due to noise. In this case, this result confirmed model reliability.

Table 2. ANOVA statistics showing the effect of microwave power and irradiation time on the yield of essential oil.

Source	Sum of squares	Degree of freedom	Mean square	F-value	P-value Probability > F
Model	1.92	5	0.38	48.87	< 0.0001***
X1	1.48	1	1.48	187.80	< 0.0001***
X2	0.19	1	0.19	24.28	0.0017**
X1X2	0.01	1	0.01	1.27	0.2968
X1^2	0.13	1	0.13	15.96	0.0052**
X2^2	0.15	1	0.15	19.07	0.0033**
Residual	0.06	7	0.01	–	–
Lack-of-fit	0.05	3	0.02	6.10	0.0579
Pure error	0.01	4	0.03	–	–
Corrected total	1.98	12	–	–	–
R^2	0.9722	–	–	–	–

** p < 0.01.

*** p < 0.001.

Moreover, the correlation coefficient () and the lack-of-fit value were also evaluated to check the quality of the proposed model. As depicted in Figure 3, the experimental values of the essential oil yields obtained close to a straight line of the predicted values with the and adjusted-R² values were 0.9722 and 0.9523, respectively. This means that the presented model had the capability to estimate the oil yield because the resulting R² values are higher than 0.75 (23,26). The p-value of lack-of-fit is higher than 0.05 (p = 0.0579, non-significant) and could imply that the assumption of the constant variance was satisfied. The value of an adequate precision that measures the signal-to-noise ratio was 20.40, reinforcing an adequate signal due to the fact that a ratio higher than 4 is desirable. The values of standard deviation and coefficient of variation (CV) were 0.09 and 13.10%, respectively. These results indicate a good precision of the experimental and predicted values. Therefore, it could be concluded that the constructed model could be used to navigate the design space.

Figure 3. The parity plot showing the correlation between the distribution of experiment values and predicted values of the recovery of essential oil from cassumunar ginger rhizomes. The numbers in the graph indicate the experimental numbers in Table 1.

3.2. Response surface analyses

The response surface three-dimensional and contour profiles for the oil yield as a function of microwave power and irradiation time are illustrated in Figure 4. The oil yield tremendously changed with increasing microwave power at a given irradiation time (Figure 4a). The recovery of oil yield reached a peak value [1.20% (v/w)] at 700 W microwave power and 70 min irradiation time. No further rise in the oil yield was observed even if the microwave power was increased to a value higher than 700 W. Moreover, more details were visualized in the two-dimensional contour plot in Figure 4b. The oil yield increased when the microwave power was increased from 100 W to 700 W, as well as the irradiation time from 8 min to 70 min. A similar tendency was reported in literature for Dryopteris fragrans: the quantity of the plant oil mainly depends on microwave power and irradiation time (27). This result can be explained by the in-situ water in the plant cells, which absorb microwave energy, favorable to plant cell swelling and increasing the driving force of mass transfer. However, further irradiating of the plant sample after 90 min tended to diminish the oil yield, probably due to either partial thermal degradation or hydrolysis of some volatile molecules and volatilization (2,23,28). Thus, a moderate microwave power (≤ 700 W) and shorter irradiation time (≤ 90 min) should be chosen as an appropriate condition to increase the extraction yield.

Figure 4. Response surface plot between microwave power and irradiation time on yield of essential oil: (a) three dimensional plot and (b) contour plot.

3.3. Optimization of the SFME for the essential oil isolation and model validation

The test factors were simultaneously considered in the optimization process by solving Eq. (2) and Eq. (3). The results are shown in Figure 5. Based on the desirability function technique, the SFME conditions for obtaining the maximum oil yield were 700 W microwave power and 70 min irradiation time with a high desirability function value of 0.94. This optimal point located within the operating parameter ranges and additional three experiments were conducted to verify the accuracy of the model. Under the SFME conditions obtained by this process, the experiment value [1.21 ± 0.10% (v/w)] of the oil yield was consistent with that of the predicted value [1.22 ± 0.11% (v/w)] (p > 0.05). The high correlation between the experimental and the predicted results confirmed that the resulting model had the accuracy to predict the response value.

Figure 5. Plots showing the profile of the desirability function value for essential oil yield.

3.4. Comparison of SFME and HD

The SFME was applied for the isolation of cassumunar ginger oil to minimize the waste water of the process and reduce processing time. In the present work, the oil isolated from the matrices of cassumunar ginger using SFME showed a reduced extraction time (70 min) compared to conventional HD (5 h). This was due to the different fundamentals of the SFME and HD. In SFME, both, the mass and the heat transfer mostly occur from inside to outside under microwave irradiation. Moreover, heat transfer partly occurs also from outside to inside of the plant cells. In the HD, the heat transfer occurs from outside to inside the plant matrices, which is opposite to the direction of the mass transfer of the essential oil (2,3). Additionally, the temperature inside the extraction flask of the HD process increased much slower than the SFME. Therefore, a synergistic combination of the heat and mass gradients occurring in the SFME process resulted in enhancing the overall rate of essential oil extraction. Literature data reported of other aromatic plants extracted by SFME supported our findings (4,6).

3.5. Chemical composition of essential oil

The volatile compounds in the essential oil obtained by the SFME at the optimal conditions were qualitatively analyzed by GC-MS. A total of 29 volatile substances were identified from cassumunar ginger essential oil (Figure 6). Sabinene, terpinen-4-ol and DMPBD were the major components in the oil extracted by SFME, accounting for 17.80, 20.81 and 28.62%, respectively (Table 3). Some differences were clearly found based on the peak area obtained from GC analysis. SFME produced a higher concentration of oxygenated compounds compared to the HD. Similar findings have been reported for the extraction of essential oil from Amomum tsao-ko (2). This leads to the conclusion that the presence of oxygenated compounds in the oil exhibits different behavior compared to the non-oxygenated groups of compounds when the oil is extracted from plant material by SFME. The oxygenated compounds have a higher dipolar moment with a higher microwave power absorption and easier extraction than monoterpenes or sesquiterpenes with a lower dipolar moment (29).

Table 3. Chemical composition of essential oil extracted from cassumunar ginger rhizomes using the optimized SFME condition and HD.

				Relative peak area (%)
No.	Constituent	RI^a	RI^b	SFME	HD	Identification method	Reference
1	α-Thujene	927	924	0.78 ± 0.35	0.87 ± 0.49	RI, MS	(12,32)
2	α-Pinene	934	932	0.85 ± 0.32	1.08 ± 0.14	RI, MS, CO	(12,32–34)
3	Camphene	949	946	0.86 ± 0.27	1.40 ± 0.34	RI, MS	(33)
4	Sabinene	964	969	17.80 ± 0.25	9.15 ± 0.41	RI, MS	(12,13,32,34)
5	β-Pinene	970	974	1.61 ± 0.40	0.96 ± 0.10	RI, MS	(32,34)
6	β-Myrcene	987	988	1.52 ± 0.19	2.79 ± 0.96	RI, MS	(32,34)
7	α-Phellandrene	1005	1002	1.36 ± 1.10	3.12 ± 1.81	RI, MS	(33)
8	α-Terpinene	1016	1014	3.65 ± 0.26	5.88 ± 0.31	RI, MS, CO	(12,32,34)
9	p-Cymene	1023	1020	1.44 ± 0.02	2.25 ± 0.02	RI, MS	(12,32)
10	β-Phellandrene	1028	1025	3.12 ± 0.17	4.67 ± 0.95	RI, MS	(12,32)
11	1,8-Cineole	1029	1026	tr	tr	RI, MS	(33)
12	γ-Terpinene	1057	1054	3.72 ± 0.17	6.69 ± 0.43	RI, MS	(12,32,34)
13	(Z)-Sabinene hydrate	1065	1065	0.39 ± 0.13	1.02 ± 0.05	RI, MS	(12)
14	Terpinolene	1088	1086	0.38 ± 0.10	0.32 ± 0.07	RI, MS	(12,32,34)
15	(E)-Sabinene hydrate	1098	1098	0.22 ± 0.08	0.20 ± 0.01	RI, MS	(33)
16	(Z)-p-Menth-2-en-1-ol	1120	1118	tr	tr	RI, MS	(33)
17	(E)-p-Menth-2-en-1-ol	1138	1135	0.08 ± 0.04	0.25 ± 0.05	RI, MS	(33)
18	Borneol	1164	1165	tr	tr	RI, MS	(33,34)
19	Terpinen-4-ol	1178	1174	20.81 ± 2.68	24.51 ± 0.61	RI, MS, CO	(12,13,32,34)
20	α-Terpineol	1190	1190	tr	tr	RI, MS	(32,34)
21	(Z)-Piperitol	1195	1195	0.03 ± 0.03	0.20 ± 0.01	RI, MS	(33)
22	(E)-Piperitol	1207	1207	0.02 ± 0.01	0.04 ± 0.02	RI, MS	(33,34)
23	α-Terpinyl acetate	1349	1346	0.18 ± 0.02	0.60 ± 0.06	RI, MS	(12)
24	α-Zingiberene	1495	1493	0.02 ± 0.01	0.05 ± 0.02	RI, MS	(33,34)
25	β-Bisabolene	1508	1509	tr	0.09 ± 0.08	RI, MS	(33,34)
26	β-Sesquiphellandrene	1523	1521	0.04 ± 0.02	0.04 ± 0.01	RI, MS	(32,34)
27	Germacrene B	1546	1543	0.22 ± 0.05	1.23 ± 0.18	RI, MS	(33,34)
28	(E)-1-(3,4-Dimethoxyphenyl)but-1-ene	1555	1558	0.17 ± 0.02	tr	RI, MS	(34)
29	DMPBD	1598	1595	28.62 ± 0.54	17.84 ± 0.50	RI, MS	(12,13,34)
	Total			87.89 ± 2.55	85.25 ± 2.20
	Monoterpene hydrocarbon			37.08 ± 0.95	39.19 ± 2.18
	Sesquiterpenes			0.28 ± 0.02	2.02 ± 0.25
	Oxygenated compounds			50.53 ± 3.03	44.65 ± 0.26

^a RI, retention indices determined on HP-5MS column using C₈-C₂₀ alkane series in this experiment; ^b RI on HP-5MS column taken from literature data; MS, mass spectra; CO, co-injection with authentic substance; tr, trace amount (peak area < 0.02%).

Figure 6. GC chromatogram of essential oil isolated from the rhizomes of cassumunar ginger by SFME with peak identification of the major constituents.

3.6. Mechanism of SFME on boosting the essential oil yield

More details of the extraction mechanism of SFME were studied by microscopic technique. Figure 7a-c shows the normal tissue of the untreated cassumunar ginger rhizomes. The untreated plant material consisted of epidermal cells having rectangular and elongated cells, followed by thin walled, irregular elongated, 6–10 layered cork cells. The cortex compartment had several layers of parenchymatous cells with intercellular air spaces. Oil globules were abundant in the parenchymatous cortex and pith. This observation clearly showed that the essential oil is deposited inside the plant cells.

Figure 7. Microscopic images showing the normal plant tissue and oil globule of the cassumunar ginger rhizomes: (a) light microscope, (b) fluorescent microscope FITC filter and (c) fluorescent microscope Tx Red filter (bar = 250 μm).

The microstructural views of the transverse section of cassumunar ginger rhizomes before and after extraction by SFME and HD are given in Figure 8. The unprocessed sample (Figure 8a) showed a regular structure of the oil globules in parenchyma with normal cellulosic walls. The cell walls of parenchyma were still intact with some ruptures, atrophic and wrinkles appearing in the sample after HD treatment for 5 h (Figure 8b). The losses of oil cells from the plant matrices were clearly observed with completely crimped and broken plant cell walls after SFME for 70 min (Figure 8c).

Figure 8. Light micrographs of cassumunar ginger rhizome samples: (a) untreated sample, (b) sample treated with HD for 5 h and (c) sample treated with SFME for 70 min (bar = 100 μm).

A plausible mechanism of the oil isolation by SFME was illustrated in detail by light microscopy. As shown in Figure 9a, the microstructure of the plant sample before extraction contained oil globules located inside the cells. An obvious effect of the microwave irradiation on the microstructure changes of the samples was found after extraction by SFME with different irradiation times (10, 25, 50, 60 and 70 min) (Figure 9b–f). In this case, the extraction mechanism could be described in that way that the microwaves interacted selectively with the in-situ water (the free water molecules) located inside the plant cells to give the ‘microwave heating effect’ during electromagnetic radiation. The heated areas inside the plant matrices near or above the boiling point of water resulted in a dramatic expansion of plant cells due to higher internal pressures and subsequent to the cell wall rupture and the bursting of oil globules [3]. Thus, the essential oil migrated quickly from the plant cells flowing toward the surrounding system. Moreover, many volatile substances having dipolar moment properties present in the essential oil can also strongly absorb the microwave energy. These properties resulted in an easier essential oil extraction if the sample contained aromatic substances with high dipolar moments in contrast to those having low dipolar moment values. In term of the eco-friendly procedure, SFME would be a promising process requiring 4.3-fold less time than HD.

Figure 9. Microscopic structure of cassumunar ginger samples (a) before and after extraction with 700 W microwave power at (b) 10 min, (c) 25 min, (d) 50 min, (e) 60 min and (f) 70 min (bar = 100 μm).

3.7. Physical properties of the essential oil

The physical properties of the essential oil isolated by SFME and HD were compiled in Table 4. The essential oil obtained from SFME had a pale yellow color comparable to the conventional HD. The oil quality was then subjected to sensory analysis in comparison to the oil resulting from HD. In this case, the aroma of cassumunar ginger oil was a result of a complex mixture of volatile constituents (monoterpenes, sesquiterpenes and oxygenated compounds). A similar smell was found between the essential oils of the two extraction methods showing a strong aroma initially with top to middle notes of cassumunar ginger flavor. Moreover, the scent profile of the tested oils was classified to be cool, earthy, herbaceous and woody-smelling. Several studies suggested that the oxygenated compounds play a key role in the characteristic of the strong odor of the essential oils (23,30). Because the cassumunar ginger essential oil extracted by SFME and HD contained the oxygenated groups of volatile compounds as major components. These predominant compounds including terpinen-4-ol and DMPBD should be important substances in relation to such aromatic profile. Other properties of the oil isolated by SFME including relative density and refractive index were slightly higher than HD (p < 0.05). However, all the tested physical properties corresponded to the specification according to Thai Industrial Standard (31).

Table 4. Physical characteristics of the essential oil isolated from the fresh cassumunar ginger rhizomes by SFME and HD..

Physicochemical property	SFME	HD	Specifications according to Thai Industrial Standard (31)
Appearance	Corresponds	Corresponds	Colorless to yellow and clear liquid
Odor	Corresponds	Corresponds	Special cassumunar ginger characteristic herbaceous odor
Relative density (d20, 20°C, g/mL)	0.9246 ± 0.0023	0.8839 ± 0.0015	0.8800–0.9250
Refractive index (, 20°C)	1.5030 ± 0.0001	1.4767 ± 0.0001	1.4770–1.5090
Solubility (volume parts of 80% ethanol per 1 mL of the essential oil, 20°C)	3	3	3

Means ± standard deviation (SD) (n = 3).

SFME: Solvent-free microwave extraction, HD: hydrodistillation

4. Conclusions

The essential oil of cassumunar ginger rhizomes could be successfully isolated by application of the SFME combined with CCD. The second-order polynomial equation was an adequate mathematic model for calculating the most effective parameters of the extraction process. The optimal conditions were microwave power of 700 W and irradiation time of 70 min that provided the highest yield of essential oil (1.12 ± 0.10% v/w). Microwave power was the main factor underlying the extraction efficiency of oil yield, followed by microwave exposure time. The major components of the essential oil obtained from SFME and HD were slightly different, although SFME gave a higher concentration of oxygenated compounds. Meanwhile, isolation of the essential oil by SFME also provided better extractability than a conventional HD as confirmed by a more plant cell rupture following the microwave treatment. Therefore, application of the SFME with CCD for isolating the essential oil from cassumunar ginger was clearly advantageous in terms of simplicity, rapidity and environmental friendly technique. A large-scale industrial production required to be further studied for commercialization.

Acknowledgments

Research reported in this study was supported by the ASEAN-European Academic University Network (ASEA-UNINET). Special thanks to Mr. Kurt Plöschberger for his technical support on GC-MS analysis.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

Research reported in this study was supported by the ASEAN-European Academic University Network (ASEA-UNINET), the Austrian Federal Ministry of Science, Research and Economy, and the Austrian Agency for International Cooperation in Education and Research (OeAD-GmbH).