Optimization Of Supercritical Carbon Dioxide Extraction Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Optimization using response surface methodology for the extractions of phenolics from Citrus hystrix leaf was carried out by supercritical fluid extraction. The effects of CO2 rate, extraction pressure and extraction temperature on yield, total phenolic content and Diphenyl-picrylhydrazyl-IC50 were evaluated and compared with ethanol extraction. Ethanol extracts and optimum SFE conditions were analysed with HPLC. Among the three variables studied, extraction pressure had the most significant influence on the yield, TPC and DPPH-IC50 of the extracts, followed by CO2 rate and extraction temperature. The optimum conditions of pressure, CO2 rate and temperature were at 267 bars, 18 g/min and 50oC, respectively. The yield, TPC and DPPH-IC50 obtained were 5.06 %, 116.53 mg GAE/g extract and IC50 of 0.063 mg/ml, respectively. These values were reasonably close to their counterpart of predicted (p>0.05). Better inhibition and TPC were obtained using SFE method whereas higher yield and phenolic acids were observed with ethanol extraction.

The stressful life style and less balanced food intake globally partially due to high concentrations of free lipid radicals, both in food (in vitro) and in vivo after food ingestion has given to the need to look at antioxidants as a functional ingredient in food. Synthetic antioxidants such as, butylated hydroxytoluene (BHT), butylated hydroxyanisole (BHA), tertiary- butyl hydro-quanone (TBHQ) and propyl gallate (PG), are conventional food antioxidants. Due to safety issues, consumer concerns and increasing regulatory scrutiny ((Jamilah et al., 2009; Shahidi., 1997) concerning synthetic antioxidants, the possibility of natural antioxidants as alternatives is aggressively researched. The leaves of Citrus hystrix, known locally as, Limau purut, is used in many Malaysian and South-East Asian region local dishes and medicinal preparations. C.hystrix as a potential new source of natural antioxidant was reported by Jamilah et al. (1998), Ching and Mohamed (2001), Jaswir et al. (2004), Idris et al. (2008), Chan et al. (2009) and Butryee et al. (2009). All extracts were extracted using the conventional solvents such as, ethanol, methanol, acetone and water. To produce extracts of high phenolic content and rich in antioxidants from C. hystrix leaves, requires high extraction efficiency influenced by factors such as particle size, extraction methods, solvent type, solvent concentration, solvent-to-solid ratio, extraction temperature, pressure and time (Banik et al, 2007; Lang et al., 2001; Pinelo et al., 2005; Silva et al., 2007).

Steam distillation and organic solvent extraction using percolation, maceration and Soxhlet techniques are conventionally used for the extraction of bioactive compounds from plant sources. They are not efficient and economical and this can be overcome by using the supercritical carbon dioxide (SC-CO2) process (Bimakr et al., 2009). Carbon dioxide (critical temperature, pressure and density ~ 31.18 oC, 72.0 bar; 0.47 gcm-3, respectively) is safe, residue free, non-flammable, in expensive and environmentally- friendly (Pyo and Oo, 2007).

The optimization of supercritical fluids for the extraction of natural antioxidants and phenolic compounds from the leaves of C.hystrix has not been reported. Hence, this study was carried out with the objective of optimizing the extraction of the antioxidant and phenolic acids from the leaves of C. hystrix using supercritical carbon dioxide (SC-CO2) fluid extraction by varying and/or fixing known variables associated with the extraction techniques.

2 Materials and Methods

2.1 Reagents used

Folin-Ciocalteu Reagent (FCR) and 1,1-Diphenyl-2-picryl-hydrazyl (DPPH) were purchased from Sigma (St Louis MO USA). Carbone dioxide, (purity 99.99%), containing in a Carbone dioxide dip tube cylinder, was purchased from Malaysian Oxygen (MOX), Malaysia. Absolute ethanol (99.4%, analytical grade), the modifier for SC-CO2 process, acetonitrile and methanol (HPLC grade) as the mobile phase for HPLC and phenolic acids standards (vanillic acid, syringic acid, p-coumaric acid, M-cumeric, trans cinnamic acid, benzoic acid, gallic acid and sinapic acid) were purchased from Fisher Scientific Chemical (Loughborough, England). All other chemicals used were either analytical or HPLC grade.

2.2 Preparation of Sample

The leaves of C. hystrix were obtained from Pasar Borong, a whole sale market at Puchong, Selangor, Malaysia. Upon arrival at the laboratory, leaves were sorted, washed under running tap water, oven dried at 40°C for 24h and stored at ambient temperature away from the light.

The dried leaves were ground just before extraction in a blender (MX-335, Panasonic, Malaysia) for 10s to produce a powder with an approximate particle size of 0.5mm (Bimak et al., 2009).

2.3 Solvent Extraction

The phenolic compounds in the C. hystrix leaves powder were extracted according to Jamilah et al. (1998) with slight modifications. The first step involved soaking the powder in 95% ethanol for 24h at 50oC at an ethanol to leaf ratio of 10:1 (v/w). The crude extract was then filtered and concentrated by evaporating at 40oC in the rotary evaporator (Eyela, A-1000S, Japan).When the ethanol was evaporated off the concentrated extract was transferred into brown glass bottles, flushed with nitrogen and kept at - 25oC until use. The extraction was carried out in triplicate

2.4 Supercritical Carbon Dioxide (SC-CO2) Extraction

Supercritical carbon dioxide (SC-CO2) fluid extraction using the supercritical fluid extractor (ABRP200, Pittsburgh, PA, USA), with a 500 mL extractor vessel attached, was carried out according to Bimark et al. (2009) with slight modifications. The flow rate of CO2 and modifier, extraction temperature, pressure and time were adjusted using ICE software coupled with the supercritical fluid extractor. The liquid CO2 was pressurized and heated to the desired pressure and temperature with the aid of pressure pump (P-50, Pittsburg, PA, USA) to reach the supercritical state prior to passing it into the extraction vessel. Absolute ethanol was used as the modifier to improve the extraction of phenolics from C.hystrix leaves and fixed at a flow rate of 3 mL / min for all experimental procedures. The duration of the static extraction time was fixed at 30 min, while the dynamic extraction time was constant at 90 min.

Fifty grams of C. hystrix leaves (powder) was mixed with 150g glass beads (2.0 mm in diameters) to systemize the flow rate and the mixture was placed in the extractor vessel. The extraction was then performed under various experimental conditions as generated by the response surface methodology (RSM) design. EtOH was removed from the extracts by vacuum evaporation using a rotary evaporator (Eyela, A-1000S, Japan) at 40 °C. The extracts were collected in the round bottle flask (warped with aluminium foil to minimize light exposure and thus oxidation) and then placed in the oven at 40°C for 30 min before being transferred into desiccators for final constant weight. Extracts were transferred into brown glass bottles, flashed with nitrogen and stored in a freezer of -25°C until further analysis. The extractions were carried out in duplicates.

2.5 Determination of Total Phenolic Content (TPC)

The total phenolic content of C.hystrix leaf extracts was determined using the Folin-Ciocalteu reagent according to the method described by Singletone et al. (1999). An aliquot of the extract (0.5mL) was put in 0.5mL of Folin reagent, under dim light before 10mL (7%) of sodium carbonate was added. The mixture was then left in the dark for 60 min. The absorbance of the mixture was measured against EtOH (blank) at 725 nm by using a UV-Visible spectrophotometer (UV-1650PC, Shimadzu, Kyoto, Japan). The calibration equation for gallic acid, expressed as gallic acid equivalent (GAE) in mg/g extract, was y = 0.0064x + 0.0093 (R2 = 0.9972).

2.6 Determination of Free Radical Scavenging Activity

Free radical scavenging activity of C.hystrix leaf extracts was measured according to the procedure described by Ramadan et al. (2006) with slight modifications. A 0.1 mL aliquot of toluenic sample solution at different concentrations was added with 0.39 mL of fresh toluenic 1,1-Diphenyl-2-picryl-hydrazyl (DPPH) solution (0.1 mM). Triplicates were carried out for each concentration. The mixtures were shaken vigorously and left in the dark for 60 min and absorbance was read against pure toluene (blank) at 515 nm using a UV-Visible spectrophotometer (UV-1650PC, Shimadzu, Kyoto, Japan). The free radical scavenging activity of extracts was calculated as follows:

% Inhibition = ([Acontrol-Asample]/Acontrol)*100

Where Acontrol = absorbance of the control reaction (containing all reagents except samples); Asample = absorbance of the test compound.

Determination of IC50 in this test was defined as the concentration of the extract that was able to inhibit 50% of the total DPPH radicals. IC50 of the sample was expressed in mg/mL and calculated through the interpolation of linear regression analysis (Brand-Williams et al., 1995)

2.7 Determination of Phenolic acids

The phenolic acids of the C.hystrix leaf extracts that were obtained from the optimum SC-CO2 conditions (optimum of yield, TPC and DPPH-IC50) were analysed by a high-performance liquid chromatography (HPLC), [Agilent Technologies 1200 series model, 76337 Waldbronn, Germany] equipped with Diode Array Detector (DAD), and detection at 254nm.

The HPLC parameters were modified from Anderson et al. (1983). The column temperature used was 30°C at a maximum temperature 35°C and the column used was Crespak RP C18S RP C18 (150mm L* 4.6mm ID, JASCO). The solvents were of HPLC grade (Fisher Scientific Chemical, Loughborough, England). All solvents were filtered through a cellulose nitrate membrane filters (0.45 µm). Flow rate of mobile phases used were 1.5ml/min for 25% acetonitrile in formic acid-water (0.5:99.5), run isocratically.

The extracts were first filtered through 0.2 µm nylon (NYL) filter, (Whatman) for the removal of impurities and unwanted compounds. The injection volume used was 20µL with duplicates for each of the SC-CO2 optimum conditions and ethanol extracts.

The standards used were vanillic acid, syringic acid, p-coumaric acid, M-cumeric, trans cinnamic acid, benzoic acid and sinapic acid (Fisher Scientific Chemical Loughborough, England). Identification and quantification of phenolic acids in the extracts were based on the standard curves of the standards as well as their peaks retention times.

2.8 Experimental design and statistical analysis

Response surface methodology (RSM) was used to determine the optimum conditions for the yield, TPC and DPPH-IC50 in C.hystrix leaf extracts. The experimental design and statistical analysis were carried out using the statistical software (MINITAB release 14). Central composite design was chosen to evaluate the joint effect of three independent variables CO2 rate, extraction temperature and pressure, coded as X1, X2 and X3, respectively. The minimum and maximum values for CO2 rate were set at 15 and 25 g/min, extraction temperature between 40 and 60 oC and pressure between 100 and 300 bars. The dependent values were yield, TPC and DPPH-IC50. For optimization, yield and TPC were maximized to achieve highest values and loswest value for DPPH-IC50.

The whole design consisted of 20 combinations including six replicates of the centre point (Table 1) (Myers & Montgomery, 2002). The ANOVA tables were generated and the effect and regression coefficients of individual linear, quadratic and interaction terms were determined. The significances of all terms in the polynomial were analyzed statistically by computing the F-value at a probability (p) of 0.001, 0.01 or 0.05. The statistically found non-significant (p > 0.05) terms were removed from the initial models and only significant (p < 0.05) factors were involved in the final reduced model. It should be noted that non-significant linear terms were kept in the reduced model in cases where their quadratic or interaction terms were significant (p < 0.05) (Mirhosseini et al., 2009). Experimental data were fitted to the following second order polynomial model and regression coefficients were obtained according to the generalized second-order polynomial model proposed for the response surface analysis, given as follows

Where β0, βi, βii, βij were regression coefficients for intercept, linear, quadratic and interaction terms, respectively. Xi and Xj were coded values of the independent variables, while k equaled to the number of the tested factors (k=3).

3. Results and Discussion

3.1 Response Surface Methodology (RSM) Analyses

3.1.1 Model Fitness

Based on the ranges set for the identified parameters, 20 trails of each parameter, including six replicates of the centre points that influence Yield, TPC and DPPH-IC50 were selected. In this study, the lower and upper values for the variables were set at +alpha (+α=1.633) and -alpha (-α=1.633) and so all the factor levels were chosen within the limits that were practical with SFE (above critical tempHYPERLINK "http://en.wikipedia.org/wiki/Critical_temperature"erature of 31°C and critical pressure of 72 bar) and desirable. The experimental and predicted values for responses under the different combinations of extraction conditions via SC-CO2 extractions were as in Table (1). The results indicated that yield, TPC and DPPH-IC50 obtained, ranged from 0.4- 5%, 15 - 128.9 mg GAE/g extract and 0.065 - 0.300 mg/ml, respectively.

By utilizing multiple regression analysis, relationships between the tested parameters and the responses were explained in equations 2, 3, and 4 for yield, TPC and DPPH-IC50, respectively. The fitness of response function and experimental data was evaluated from the linearity, quadratic and regression coefficients of independent variables as shown in Table 2. The ANOVA of regression model showed that the models were noticeably significant due to the extremely low probability value (P<0.001). The coefficient of determination (R2) and significance of lack of fitness was further evaluated to check the fitness and model adequacy. The R2 equal to the unity or ≥ 0.8, is desirable. R2 values for the regression model of yield, TPC, and DPPH-IC50, were 0.935, 0.95, and 0.96, respectively, which were close to 1 (Table 2). Thus, indicating that the predicted second order polynomial models fitted well with the system. The values of adjusted R2 (corrected value for R2 after the elimination of the unnecessary model terms) of yield, TPC and DPPH-IC50 were also very high, hence suggesting the high significance of the model (0.897, 0.92 and 0.93). The simultaneous increase of both R2 and adjusted R2 plus the absence of any lack of fit (p>0.05) in our data has proved its credibility and model adequacy. The multiple regression results and the significance of regression coefficients yield, TPC and DPPH-IC50 models were as shown in Table 3. It could be observed that both the linear and quadratic term of all parameters significantly ( p<0.05) effected the yield, TPC and DPPH -IC50, however, CO2 rate did not significantly affect the DPPH-IC50 where temperature effect on TPC was only significant in quadratic manner to remain in the model (Table 3).

The following regression equations showed the final reduced models fitted for the parameters and their responses.

Yield= - 3.33 + 0.142 X1 + 0.164X2 + 0.00735X3- 0.00669X12 - 0.00218 X22 - 0.000025 X32 Eq(2)

TPC = - 909 + 25.4 X1 + 25.6 X2 + 1.54 X3 - 0.668 X12 - 0.250 X22- 0.00278 X32 Eq(3)

DPPH-IC50= - 0.604 X2 - 0.0177 X3 + 0.00559 X22 + 0.000031 X32


3.1.2 Verification of models

The appropriateness of the response surface equation was tested by the evaluation of experimental and predicted values from the reduced response regression models. A close agreement between the experimental and predicted values (Table1) was noted. No significant (p > 0.05) difference was observed between those values, suggesting the adequate fitness of the response equations.

3.2 Influence of Pressure, CO2 Rate and Temperature on SC-CO2 Extraction efficiency

Figure 1(a) showed the three-dimensional response surface plots by presenting the response in the function of two factors and keeping the temperature at its middle level (50oC).

It showed a higher yield in the region of extraction pressure between 190 to 300 bars and at CO2 rate of 12 to 17g/min. Both extraction pressure and CO2 rate exhibited significant linear and quadratic effects on yield as shown Table (3). The yield was optimum at about 14.8g / min CO2 flow rate and at pressure of 320 bars. Eextraction pressure was more influential than CO2 rate as reflected by its higher linear and quadratic coefficients (β3=0.65819; β33 = -0.25168) compared to the latter (β1= -0.35060; β11=-0.16731). In supercritical fluid extraction (SFE), increased pressures result in, increased solvent density and solvent power of fluid which may lead to higher extraction yields, on the other hand, increased pure CO2 rate under SFE is a good solvent for lipophilic compounds (non- polar) but is poor for phenolics (polar) (Martinez, 2007). Therefore, modifier (ethanol) was used to improve the extraction of phenolics from C.hystrix leaves.

Figure 1(b) showed the effects of extraction pressure and extraction temperature on yield at constant CO2 rate of 20 g/min. Extraction pressure displayed a very significant (p<0.001) on the yield in linear and quadratic manner as also shown in Table (3). At pressure of ≥140 and temperature not exceeding 47oC yield increased, however with further increase in the temperature the yield showed a decrease which is most probably due to the reduced density of CO2.

The relationship of CO2 rate and extraction temperature with yield was plotted in Figure 1(c). Both the parameters exhibited significant linear and quadratic effect (p<0.05) on yield. The yield increased rapidly with decreasing CO2 rate up to 13 g/m and this followed by a slight decrease thereafter.

By combining all the results presented in Figure 1, it was obvious that extraction pressure had the most critical impact on yield of the extract followed by CO2 rate and extraction temperature.

3.3 Total Phenolic Content (TPC)

The TPC of the extract was as shown in Figure 2. Depending on the pressure, temperature and CO2 rate, the TPC of the extract ranged from 15.0 to 128.9 mg GAE/g extract. No available literature report could be be used for comparison for the SC-CO2 extraction method; however, Idris et al. (2008) reported that TPC of the extracts was about 103.2 mg GAE/g extract which was slightly lower than our EtOH extracted TPC (112.7 mg GAE per g extract). Moderate levels of the selected independent variables of SC-CO2 extracts (run order 7, 10, 12, and 17, Table 1) reflected higher TPC of the C.hystrix leaf extracts than our EtOH extraction as well as Idris's; this may have something to do with possible partial degradation of the extracted compounds due to long extraction time when conventional extraction methods are to be used. With SC-CO2 method the extraction time (90 min) was remarkably shorter than that of EtOH extraction (>20 h).

3.4 Free Radical Scavenging Activity

Figure 3 demonstrated the effect of temperature and pressure on the scavenging property of the C.hystrix leave extracts. The antioxidant activity of the extracts, determined by the IC50 of radical scavenging properties of diphenylpicrylhydrazyl (DPPH-IC50), was found to be high at average level of temperature and relatively increased stages of pressure i.e. DPPH-IC50 of the extracts gradually decreased with the increase of extraction temperature and pressure up to 50 °C and 314 bars, respectively to achieve optimum value of IC50 at 0.0585 before it began to increase. The lesser the IC50, the stronger activity is the corresponding matter (Mariod et al., 2010). Under the assay conditions employed here, the IC50 of BHA and α-tocopherol as a positive controls were 0.023mg/ml and 0.031mg/ml, respectively, among the extract run order 12, 9, and 16 (table 1) possessed greater DPPH radical scavenging activities with the lower IC50 values of 0.065, 0.08 and 0.085mg/ml, respectively. This was in agreement to the findings of Idriss et al. (2008), where the activity of BHA was found to be higher than the sample. Compared to conventional solvent extraction method with the IC50 of 0.250 mg/ml (Table 1), it can be observed that SC-CO2 extracts demonstrated notable DPPH radical-scavenging activity remarkably greater than that of traditional extraction method. The IC50 values for CLE extracted by SC-CO2 ranged from 0.065 - 0.300 mg/ml depending on pressure and temperature where an increase in the pressure relatively resulted in an increase in its antioxidant capacity.

3.5 Identification and Quantification of Phenolic Acids of the extracts.

Out of seven standard phenolic acid solutions mixed, six have been detected in solvent and supercritical carbon dioxide extraction of the extracts (table 4). By quantifying the amount of phenolic acids in the extract, a considerable variation between EtOH and SC-CO2 extraction was observed. Higher recovery of phenolic acids than that of SC-CO2 extraction was found using 95% EtOH as shown in Table (4). The number of polar function groups, e.g. hydroxyl groups, may have influenced volatility of the solutes thus determining their optimum extractability with SC-CO2 (Lang and Wai, 2001). For example, (Stahl and Glatz, 1984)successfully extracted steroids with three hydroxyl groups below 300 bars but failed to extract those steroids consisting of four hydroxyl groups, or three hydroxyls and one acid group, or one phenolic hydroxyl with two other hydroxyl groups. Despite the difference in quantity, the type of phenolic acids existing in the extracts for both EtOH and SC-CO2 extraction methods remained constant. Trans-cinnamic, M-coumeric and Vanillic acids represented as the predominant phenolic acids, while P-coumeric, Benzoic and sinapic acids fairly existed in the extracts (Table 4).

3.6 Conclusion

The optimum conditions of pressure at 265 bars, temperature at 50oC and CO2 rate at 18 g/min was needed for higher SC-CO2 extraction of yield, TPC and DPPH-IC50 of C.hystrix leave extracts. Of the three independent variables studied, extraction pressure was the most crucial factor influencing on yield, TPC and DPPH-IC50, flowed by CO2 rate and extraction temperature. Higher amounts of yield and phenolic acids than SC-CO2, was found in solvent extraction. Nevertheless, SC-CO2 extracts exhibited superiority in antioxidant activity measured by IC50 of 1,1-Diphenyl-picrylhydrazyl (DPPH) and total phenolic content (TPC). Even though some good results was achieved with the traditional EtOH extraction, supercritical CO2 extraction showed faster and better extraction of C.hystrix leaves. Therefore, the green technology, recyclable CO2 could be an alternative method of extraction for superior antioxidants from C.hystrix leaves.


The authors appreciate and grateful for the financial support received from the RMC, the University Putra Malaysia for this study.