J. of Supercritical Fluids 47 (2008) 209–214
Contents lists available at ScienceDirect
The Journal of Supercritical Fluids journal homepage: www.elsevier.com/locate/supflu
Supercritical CO2 extraction of phenolic compounds from Baccharis dracunculifolia Carla R. Piantino a , Francisco W.B. Aquino b , Luis A. Follegatti-Romero a , Fernando A. Cabral a,∗ a b
Department of Food Engineering, Faculty of Food Engineering, State University of Campinas, 13083-862 Campinas, SP, Brazil Department of Chemistry & Molecular Physics, Institute of Chemistry of São Carlos, University of São Paulo, 13560-970 São Carlos, SP, Brazil
a r t i c l e
i n f o
Article history: Received 25 October 2007 Received in revised form 4 July 2008 Accepted 18 July 2008 Keywords: Baccharis dracunculifolia Artepillin C Phenolic compounds Supercritical extraction
a b s t r a c t Extracts from Baccharis dracunculifolia leaves were obtained using the following solvents: supercritical carbon dioxide (SC-CO2 ), ethanol and methanol. Supercritical extraction was carried out at temperatures of 40, 50 and 60 ◦ C and pressures of 20, 30 and 40 MPa. Four phenolic compounds were analysed in the extracts by high-performance liquid chromatography: 3,5-diprenyl-4-hydroxycinnamic acid (DHCA or artepillin C); 3-prenyl-4-hydroxycinnamic acid (PHCA); 4-hydroxycinnamic acid (p-coumaric acid) and 4-methoxy-3,5,7-trihydroxyflavone (kaempferide). The global extraction yields (X0 ) obtained by the conventional methods with ethanol and methanol were higher than those obtained by SC-CO2 . However on analysing the components of interest extracted at 60 ◦ C and 40 MPa, the extraction yields of kaempferide, DHCA and PHCA were 156%, 98% and 64% higher, respectively, than in the ethanolic extracts. Only the p-coumaric acid extraction yield was better when extracted using the conventional method. © 2008 Elsevier B.V. All rights reserved.
1. Introduction Baccharis dracunculifolia (local name in Brazil: alecrim-docampo) is a shrub from the Asteraceae family that grows to a height of 2–3 m, growing wild in southern and southeastern Brazil, Uruguay, Paraguay, Argentina and Bolivia. In Brazil, leaf extracts have been used as stomachic and antipyretic medicines and as a health tonic [1]. An essential oil is also obtained from the leaves [2–5]. It has been shown [6,7] that this species is the main vegetable source of the Brazilian propolis collected in the states of São Paulo and Minas Gerais, being rich in phenolic derivatives of cinnamic acids. The various biological activities of propolis have mainly been attributed to the presence of phenolic compounds, principally the flavonoids and phenolic acids [8], which are also encountered in B. dracunculifolia. Phenolic compounds are secondary metabolites and active substances found in many plants, but their antioxidant properties and several other specific biological actions are not yet well known. They are employed as natural antioxidants in foods, pharmaceuticals and cosmetics [9].
∗ Corresponding author at: Department of Food Engineering, Faculty of Food Engineering, State University of Campinas, R. Monteiro Lobato 80, Cx. Postal 6121, 13083-862 Campinas, SP, Brazil. Tel.: +55 19 3521 4030; fax: +55 19 3521 4027. E-mail address:
[email protected] (F.A. Cabral). 0896-8446/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.supflu.2008.07.012
An important phenolic acid, present in Brazilian propolis and also in B. dracunculifolia, is 3,5-diprenyl-4-hydroxycinnamic acid (DHCA), known as artepillin C [10]. Studies have shown that DHCA inhibited lipid peroxidation and the development of pulmonary cancer in mice [11], prevented colon cancer through the induction of cell-cycle arrest, was a useful chemopreventing factor in colon carcinogenesis [12] and had anti-leukemic effects, with limited inhibitory effects on normal lymphocytes [13]. According to the literature, the DHCA concentrations in dry alcoholic extracts obtained from the bud and from the unexpanded and expanded leaves of B. dracunculifolia were 40.54, 13.75 and 1.68 mg/g, respectively [6] as against 38.58 mg/g obtained in dry alcoholic propolis extracts. The flavonoids in propolis have been widely researched [14,15] and their intake may have an effect on many physiologic processes: they aid vitamin absorption and action and act in cicatrization processes as antioxidants, besides presenting antimicrobial activity and modulating the immune system [16]. Various natural extracts containing bioactive compounds can be obtained using supercritical extraction processes [17–22]. This process shows advantages over conventional ones, such as the possibility of continuous modulation of the solvent power/selectivity, elimination of polluting organic solvents and the reduction of post-processing costs since there is no longer the need to eliminate solvents from the extracts [20]. Carbon dioxide is the most frequently used solvent because it is non-toxic, non-flammable,
210
C.R. Piantino et al. / J. of Supercritical Fluids 47 (2008) 209–214
odourless, environment-friendly and easily separated from the extract. It also has a low critical temperature, which allows it to be used to extract thermally labile and reactive compounds. SCCO2 is generally efficient in the purification and fractionation of hydrophobic compounds, such as flavonoids and cinnamic acid derivatives from plant matrixes [10]. Only one paper on the SC-CO2 extraction from B. dracunculifolia leaves was found in the literature, aimed at extracting the essential oil [2]. The objective of the present study was to obtain SC-CO2 extracts from B. dracunculifolia leaves aiming to discover the best conditions for extracting phenolic compounds, and analyse the extracts for the following components: DHCA, PHCA, p-coumaric acid and kaempferide. The SC-CO2 extracts were compared with alcoholic ones (ethanol and methanol) in terms of yield and the concentration of these compounds. 2. Materials and methods 2.1. Raw material B. dracunculifolia leaves were collected from the experimental farm of the Chemical, Biological and Agricultural Pluridisciplinary Research Centre (CPQBA/UNICAMP). The leaves (bud + unexpanded + expanded leaf) were dried in a tray drier with air circulation at 40 ◦ C, ground in a mill (model MA-340, Marconi, Piracicaba, Brazil) and classified according to size using a vibratory sieve shaker (Bertel, São Paulo, Brazil) for 15 min. The mesh sizes 16, 24, 32 and 48 (30%, 27%, 22% and 21% of the particles, respectively) were selected for the assays. After sieving, the samples were packed in plastic bags and stored in a domestic freezer (model 220, Cônsul, São Paulo, Brazil) at −10 ◦ C. The mean geometric particle diameter (dmg ) was calculated using the ASAE method [23]. 2.2. Chemicals Ethanol (HPLC-grade) and methanol (P.A.) were obtained from the Merck Co. (Darmstadt, Germany). The CO2 employed in the extraction was 99.5% pure and supplied by White Martins Gases Industriais (Campinas, Brazil). 3,5-Diprenyl-4-hydroxycinnamic (DHCA) and 3-prenyl-4-hydroxycinnamic acid (PHCA) were isolated during a previous study [24]. 4-Hydroxycinnamic acid (p-coumaric acid) was purchased from Sigma Chemical Co. (St. Louis, MO, USA).
2.3. Experimental procedure 2.3.1. Apparatus and supercritical extraction The experimental apparatus used in this work is shown in Fig. 1. It consisted of a 50 mL equilibrium cell (stainless steel AISI 316, Suprilab, Campinas, Brazil) (7) immersed in a water bath controlled by a heater (Suprilab, Campinas, Brazil) to within ±0.1 ◦ C. The CO2 from the supply tank (1) was cooled to the liquid state (refrigerated bath model 12101-30, Cole Parmer Instrument Company, Vernon Hills, IL, USA) (2) and compressed into the extractor using a high-pressure pump (Model AA100S, Eldex Laboratories Inc., Napa, CA, USA) (3). The other equipments used were: CO2 cylinder (4), manometers (5 and 6), glass flasks (8) and a peristaltic pump (Model L/S 77910, Cole Parmer Instrument Company) (9) to clean the system. About 7 × 10−3 kg of dried leaves were loaded into the extraction vessel and mesh 6 glass beads used to fill the empty spaces in the vessel. A static period of 20 min was used to allow contact between the samples and the supercritical solvent. The amount of SC-CO2 was kept constant at 0.624 kg for all the experiments and the CO2 mass flow rate was maintained at 4 × 10−5 kg/s. After extraction, all the tubing in the process line was washed with ethanol to recover the extract deposited in it. The global extraction yields (X0 ) were calculated as the ratio of the total mass of extract (extraction + cleaning process) to the initial mass of raw material (dry basis). The B. dracunculifolia leaf extracts were obtained in triplicate at pressures of 20, 30 and 40 MPa and temperatures of 40, 50 and 60 ◦ C using carbon dioxide. 2.3.2. Alcoholic extracts Alcoholic extracts (ethanol and methanol) of B. dracunculifolia leaves were obtained using modified methodologies of Sawaya et al. [15]. The samples were extracted for 1 day at room temperature using 10 mL of solvent per 3 g of leaves. The insoluble portion was then separated by filtration, the filtrates kept in a freezer at −10 ◦ C overnight and then filtered again to reduce the wax content of the extracts. The solvent was evaporated off in a vacuum oven at 60 ◦ C to obtain dry B. dracunculifolia extracts. 2.4. Analytical method The chromatographic analyses were carried out using HPLC equipment (Merck-Hitachi, Darmstadt, Germany), equipped with
Fig. 1. Schematic diagram of the SC-CO2 extractor.
66.69 64.48 71.77 67.57 36.82 77.37 57.67 94.36 86.99 33.36 19.06 1.077 1.184 1.490 1.417 0.588 2.110 1.458 2.768 2.825 1.105 1.087 44.87 43.07 48.53 46.93 15.59 53.42 39.51 67.84 59.99 18.08 9.94 0.037 0.051 0.049 0.039 0.081 0.075 0.047 0.107 0.134 0.338 0.317 1.55 1.86 1.60 1.28 2.15 1.91 1.27 2.62 2.84 5.54 2.90 0.051 0.067 0.107 0.053 0.018 0.137 0.049 0.149 0.190 0.116 0.108 a
b
Concentration = mg of solute per gram of extract. Yield = mg of solute extracted per gram of B. dracunculifolia leaves.
2.11 2.45 3.50 1.74 0.48 3.47 1.32 3.65 4.04 1.9 0.99 0.435 0.470 0.541 0.532 0.701 0.734 0.575 0.827 0.948 0.479 0.571 18.15 17.10 17.62 17.62 18.60 18.57 15.57 20.26 20.13 7.84 5.22 0.12 0.11 0.16 0.21 0.05 0.03 0.71 0.18 0.31 0.28 0.49 ± ± ± ± ± ± ± ± ± ± ± 2.40 2.73 3.07 3.02 3.77 3.95 3.69 4.08 4.71 6.11 10.94
Yield (mg/g)b Concentration (mg/g)a Yield (mg/g)b Concentration (mg/g)a Yield (mg/g)b Concentration (mg/g)a Global yield X0 (%) Extraction Methods
Fig. 2. Extraction curve for Baccharis dracunculifolia leaves obtained at 30 MPa and 50 ◦ C.
Table 1 Concentrations in the extracts and extraction yields for DHCA, PHCA, p-coumaric acid and kaempferide
The extraction curves for SC-CO2 at 30 MPa and 50 ◦ C using B. dracunculifolia leaves with four different mesh values were represented in terms of the total mass extract as a function of the total CO2 mass (Fig. 2). The maximum extraction yield (4.38 wt.%) was obtained for the ground material with a mean particle diameter of 5.95 × 10−4 m, and the minimum (3.17 wt.%) with a mean diameter of 1.18 × 10−3 m. The global yield was shown to increase with decrease in mean particle diameter due to the greater amount of material liberated by the rupture of a greater fraction of the particle surface and by the reduction in resistance to internal mass transference by diffusion. In general, extraction kinetics are characterized by three periods. The first period is characterized by the extractions of the solute more accessible to the solvent, found in the superficial layer of the particles. In the intermediate period, the amount of solute available on the surface decreases and part of the solute inside the particle starts diffusing to the surface. Finally, in the third period, exhaustion of the solute on the surface limits the extraction. In this work, the first period was almost inexistent, about 50% of the extract was obtained with approximately 20 g of CO2 (first period) and the other 50% with 400 g of CO2 . It should also be noted that there was no tendency to match the global yields at the different mean particle diameters after long periods of time, indicating that diffusion is not the only phenomenon which limits extraction during the period of decreasing extraction rates.
Concentration (mg/g)a
3.1. Global extraction yield
CO2 , 40 ◦ C/20 MPa CO2 , 40 ◦ C/30 MPa CO2 , 40 ◦ C/40 MPa CO2, 50 ◦ C/20 MPa CO2 , 50 ◦ C/30 MPa CO2 , 50 ◦ C/40 MPa CO2 , 60 ◦ C/20 MPa CO2 , 60 ◦ C/30 MPa CO2 , 60 ◦ C/40 MPa Ethanolic extract Methanolic extract
Yield (mg/g)b
3. Results and discussion
Concentration (mg/g)a
Sum total
Yield (mg/g)b
a pump (Merck-Hitachi, model D-7100) and a diode array detector. Separation was achieved using a Lichrochart column (MerckHitachi, Darmstadt, Germany) (RP-18, 125 × 4 mm i.d., 5 m particle size) using water, plus formic acid (95:5, v/v) (solvent A) and methanol (solvent B). Elution was carried out with a linear gradient and a flow rate of 1 mL min−1 . The analysis time was 60 min and detection monitored at 280 nm. Identification of the substances was based on a comparison of the retention times, the UV spectrum (200–400 nm) and the purity peak, with the values of standards acquired commercially and isolated in the laboratory. Quantification was done using external standards. The data were analysed by the Chromatography Data Station-DAD Manager software.
211
1.601 1.773 2.203 2.041 1.388 3.056 2.128 3.850 4.097 2.038 2.085
C.R. Piantino et al. / J. of Supercritical Fluids 47 (2008) 209–214
212
C.R. Piantino et al. / J. of Supercritical Fluids 47 (2008) 209–214
Fig. 3. Global yield for the B. dracunculifolia extracts obtained at 20, 30 and 40 MPa.
Table 1 presents the values for global extraction yield (X0 ) with SC-CO2 , ethanol and methanol using B. dracunculifolia leaves with a mean geometric particle diameter of 7.20 × 10−4 m and Fig. 3 shows the values for global extraction yield (X0 ) with SC-CO2 as a function of temperature and pressure. The global extraction yield (X0 ) increased with increases in both temperature and pressure. The best global yield (4.71 ± 0.3%) was obtained with the highest temperature and pressure used, and the lowest global yield (2.40 ± 0.1%) at 40 ◦ C and 20 MPa. The standard deviations were higher at 20 MPa and 60 ◦ C and at 20 MPa and 50 ◦ C, which are operating conditions where the density of CO2 varies more with the variations of temperature and pressure. The global yields (X0 ) obtained with the alcoholic extracts were much higher than those obtained by SC-CO2 , resulting in yields of 10.9 wt.% for the methanolic extract and of 6.1 wt.% for the ethanolic extract. Since the greatest global yield was obtained with methanol, this indicated a predominance of substances with a polar nature. 3.2. Yield and selective extraction of phenolic compounds Table 1 shows the chemical structures, the concentrations in the extracts and extraction yields of the four components analysed. The concentration was measured at random in two of three sample extracts, and given as milligram of solute per gram of extract. The concentration values for different phenolic compounds differed between 1% and 3%. The extraction yield was calculated considering the global extraction yield (X0 ) and the concentration in the extract, obtaining a value for the extraction yield of the phenolic compounds, represented in mg of each solute extracted per gram of B. dracunculifolia leaves used in the extraction. The SC-CO2 extracts presented a concentration between 15.57 and 20.26 mg DHCA per gram of extract, values much higher than the values of 7.84 and 5.22 mg DHCA per gram of extract obtained from the dried ethanolic and methanolic extracts, respectively. With respect to DHCA extraction yield, the extracts obtained with SC-CO2 were similar to the alcoholic extracts, except using the conditions 50 ◦ C/30–40 MPa and 60 ◦ C/30–40 MPa, which gave higher values. Under the best condition (60 ◦ C and 40 MPa) an extraction yield of 0.948 mg of DHCA extracted per gram of leaves was obtained, whereas ethanol only extracted half of this amount and methanol, 60%. The influence of temperature and pressure on the SC-CO2 extraction of DHCA can be seen in Fig. 4. At the higher pres-
Fig. 4. DHCA concentration in the supercritical extracts (a) and extraction yield in mg/g of leaves (b).
sures, the concentration of this compound in the extracts increased with increase in temperature, but at low pressures (20 MPa), the inverse behaviour could be seen, the concentration decreasing with increasing temperature at constant pressure. This behaviour can be seen in the data for solubility in SC-CO2 and is explained by the effect of temperature and pressure on the variation in solvent density, and of the temperature on the solute vapour pressure. The crossover of the isotherms occurred between 20 and 25 MPa, and in this region the effect of the temperature on the increase in vapour pressure compensated the effect of the temperature on the decrease in solvent density. Considering the existence of the crossover pressure, the lowest and highest concentrations occurred on the same isotherm of 60 ◦ C. With respect to the extraction yield shown in Fig. 4(b), no inversion of behaviour was seen, due to the contribution of the global extraction yield. Similar behaviour was seen with respect to the concentrations in the SC-CO2 extracts (Table 1) and extraction yields for PHCA and kaempferide. Using the best supercritical condition, the PHCA concentration of 4.04 mg/g was much higher than that obtained in the alcoholic extracts, and the extraction yield (0.190 mg/g) was 64% and 76% higher than the extraction yields obtained with
C.R. Piantino et al. / J. of Supercritical Fluids 47 (2008) 209–214
Fig. 5. Effects of temperature and pressure on the extraction of p-coumaric acid: concentration in the extract (a) and extraction yield in mg/g of leaves (b).
ethanol and methanol, respectively. Similarly for kaempferide at 60 ◦ C and 40 MPa, the concentration obtained was much higher than that obtained in the alcoholic extracts, and the extraction yield of 2.825 mg/g was 156% and 160% higher than the yields obtained with the alcohols. Fig. 5 shows the influence of temperature and pressure on the SC-CO2 extraction of p-coumaric acid, and the behaviour with respect to the variation in concentration and extraction yield of this component was similar to that of the other components analysed. However, as shown in Table 1, the values for the alcoholic extracts were higher than those for the SC-CO2 extracts in terms of both concentration and yield, showing a yield 150% higher than that obtained at 60 ◦ C and 40 MPa in SC-CO2 . Considering the sum of the four components, a maximum of 4 mg was extracted per gram of dried leaves, with a concentration of 87 mg/g (8.7 wt.%), whilst the alcoholic solvents extracted approximately 2–3 wt.%. As compared to the ethanolic extract, the SC-CO2 extracted preferentially, in decreasing order kaempferide, DHCA, PHCA and lastly p-coumaric acid. In the latter case, ethanolic extraction was more efficient. Considering the physical properties, such as melting point and boiling point, which can be estimated from group contribu-
213
tion methods [25,26], these do not explain the selective extraction behaviour or the solubility of the phenolic compounds, since the addition of a prenyl group increased the values of these properties. The chemical structure of a molecule rather than its physical constants plays a large role in such systems. Methyl substitution at any position in a phenol, generally increases its solubility [27]. Analysing the chemical structure of hydroxycinnamic acid and its two prenylated derivatives (DHCA and PHCA), it can be seen that the addition of a prenyl group into the molecule improved extraction by SC-CO2 . The extraction of PHCA (1 prenyl group) with SC-CO2 was much better than the extraction of p-coumaric acid (0 prenyl groups), and the extraction of DHCA (2 prenyl groups) was better than that of PHCA (1 prenyl group). This indicates that the addition of a prenyl group decreased the polarity of the molecule, thus favouring supercritical extraction. Solubility data for pure solutes in SC-CO2 are useful for a prior analysis of the relative extraction efficiency between the different components of a mixture. Murga et al. [28], measured the solubilities of three hydroxycinnamic acids, 4-hydroxycinnamic acid (p-coumaric acid), 3,4-hydroxycinnamic acid (caffeic acid) and 4hydroxy-3-methoxycinnamic acid (ferulic acid) in SC-CO2 with pressures of up to 50 MPa and temperatures from 40 to 60 ◦ C. Under the same conditions of temperature and pressure, ferulic acid showed greater solubility in SC-CO2 than p-coumaric acid, the latter showing greater solubility than caffeic acid. This behaviour could be expected, since the addition of a hydroxyl group generally results in a decrease in solubility of the new molecule (caffeic acid is formed from p-coumaric acid by the addition of a hydroxyl group), and partial etherification (ferulic acid is formed by the etherification of caffeic acid) generally causes a great increase in solubility. The solubility of caffeic acid was about 40 times lower than that of p-coumaric acid and that of ferulic acid 20 times higher than that of p-coumaric acid. Uchiyama et al. [29] measured the solubility of flavone and 3-hydroxyflavone in SC-CO2 . The chemical structure of 3hydroxyflavone is similar to that of kaempferide, and on a mass basis, its solubility was about 300 times greater than that of pcoumaric acid. The sample of B. dracunculifolia leaves used in this study resulted from the sum of the bud and the unexpanded and expanded leaves, and showed a maximum DHCA concentration of 7.84 mg/g in the alcoholic extracts and of 20.13 mg/g under the best supercritical conditions. Discarding the expanded leaves from the sample would result in a considerable increase in the concentration of this component in the extracts, as shown by Park et al. [6]. 4. Conclusions Global extraction yields of 6.11%, 10.94% and 4.71% were obtained for the ethanolic, methanolic and supercritical (at 60 ◦ C and 40 MPa) extractions, respectively. However despite this behaviour, under the best supercritical conditions (60 ◦ C and 40 MPa), the concentrations in the extracts and the yield (or recovery) of three of the four components tested, were higher in the supercritical extracts than in the ethanolic extracts, the exception being p-coumaric acid. The extractions yields for each compound were 156% (kaempferide), 98% (DHCA) and 64% (PHCA) higher for the SC-CO2 extraction when compared to ethanolic extraction, while for p-coumaric acid, the SC-CO2 extraction yield was about 40% of the ethanolic extraction. SC-CO2 was more selective than alcohol to extract phenolic compounds, these four compounds representing about 9% of the SC-CO2 extract, whilst in the ethanolic and methanolic extracts, they only represented 3% and 2%, respectively. The chemical structure of the molecules was determinant for this behaviour.
214
C.R. Piantino et al. / J. of Supercritical Fluids 47 (2008) 209–214
Acknowledgements The authors wish to thank CNPq for its financial support and CPQBA-UNICAMP for supplying the B. dracunculifolia leaves used in this study. References [1] M. Fukuda, E. Ohkoshi, M. Makino, Y. Fujimoto, Studies on the constituents of the leaves of Baccharis dracunculifolia (Asteraceae) and their cytotoxic activity, Chem. Pharm. Bull. 54 (2006) 1465–1468. [2] E. Cassel, C.D. Frizzo, R. Vanderlinde, L. Atti-Serafini, D. Lorenzo, E. Dellacassa, Extraction of Baccharis oil by supercritical CO2 , Ind. Eng. Chem. Res. 39 (2000) 4803–4805. [3] R. Weyerstahl, C. Christiansen, H. Marschall, Constituents of Brazilian Vassoura oil, Flavour Frag. J. 11 (1996) 15–23. [4] V.L. Ferracini, L.C. Paraiba, H.F. Leitão Filho, A.G. Silva, L.R. Nascimento, A.J. Marsaioli, Essential oils of seven Brazilian Baccharis species, J. Essent. Res. 7 (1995) 355–367. [5] I. Loayza, D. Abujder, R. Aranda, J. Jakupovic, G. Collin, H. Deslauriers, F.-I. Jean, Essential oils of Baccharis salicifolia, B. latifolia and B. dracunculifolia, Phytochemistry 38 (1995) 381–389. [6] Y.K. Park, J.F. Paredes-Guzman, C.L. Aguiar, S.M. Alencar, F.Y. Fujiwara, Chemical constituents in Baccharis dracunculifolia as the main botanical origin of southeastern Brazilian propolis, J. Agric. Food Chem. 52 (2004) 1100– 1103. [7] S.M. Alencar, C.L. Aguiar, J. Paredes-Guzman, Y.K. Park, Chemical composition of Baccharis dracunculifolia, the botanical source of propolis from the states of São Paulo and Minas Gerais, Cienc. Rural 35 (2005) 909–915. [8] C.S. Funari, V.O. Ferro, Propolis analysis, Ciência e Tecnologia de Alimentos 26 (2006) 171–178. [9] N. Balasundram, K. Sundram, S. Samman, Phenolic compounds in plants and agri-industrial by-products: antioxidant activity, occurrence, and potential uses, Food Chem. 99 (2006) 191–203. [10] Y.-N. Lee, C.-R. Chen, H.-L. Yang, C.-C. Lin, C.-M.J. Chang, Isolation and purification of 3,5-diprenyl-4-hydroxycinnamic acid (artepillin C) in Brazilian propolis by supercritical fluid extractions, Sep. Purif. Technol. 54 (2007) 130–138. [11] T. Kimoto, S. Koya-Miyata, K. Hino, M.J. Micallef, T. Hanaya, S. Arai, M. Ikeda, M. Kurimoto, Pulmonary carcinogenesis induced by ferric nitrilotriacetate in mice and protection from it by Brazilian propolis and artepillin C, Virchows Arch. 438 (2001) 259–270. [12] K. Shimizu, K.D. Swadesh, T. Hashimoto, Y. Sowa, T. Yoshida, T. Sakai, Y. Matsuura, K. Kanazawa, Artepillin C in Brazilian propolis induces G0 /G1 arrest via stimulation of Cip1/p21 expression in human colon cancer cells, Mol. Carcinog. 44 (2005) 293–299.
[13] T. Kimoto, M. Aga, K. Hino, S. Koya-Miyata, Y. Yamamoto, M.J. Micallef, T. Hanaya, S. Arai, M. Ikeda, M. Kurimoto, Apoptosis of human leukemia cells induced by Artepillin C, an active ingredient of Brazilian propolis, Anticancer Res. 21 (2001) 221–228. [14] N. Volpi, G. Bergonzini, Analysis of flavonoids from propolis by on-line HPLC–electrospray mass spectrometry, J. Pharm. Biomed. Analysis 42 (2006) 354–361. [15] A.C.H.F. Sawaya, D.M. Tomazela, I.B.S. Cunha, V.S. Bankova, M.C. Marcucci, A.R. Custodio, M.N. Eberlin, Electrospray ionization mass spectrometry fingerprinting of propolis, Analyst 129 (2004) 739–744. [16] R.J. Williams, J.P. Spencer, C. Rice-Evans, Flavonoids: antioxidants or signalling molecules? Free Rad. Biol. Med. 36 (2004) 838–849. [17] H. Sovová, L. Opletal, M. Bártlová, M. Sajfrtová, M. Kˇrenková, Supercritical fluid extraction of lignans and cinnamic acid from Schisandra chinensis, J. Supercrit. Fluids 42 (2007) 88–95. [18] R.A. Jacques, J.G. Santos, C. Dariva, J. Vladimir Oliveira, E.B. Caramão, GC/MS characterization of mate tea leaves extracts obtained from high-pressure CO2 extraction, J. Supercrit. Fluids 40 (2007) 354–359. ˜ [19] S. Cavero, M.R. García-Risco, F.R. Marín, L. Jaime, S. Santoyo, F.J. Senoráns, G. ˜ Reglero, E. Ibanez, Supercritical fluid extraction of antioxidant compounds from oregano: chemical and functional characterization via LC–MS and in vitro assays, J. Supercrit. Fluids 38 (2006) 62–69. [20] E. Reverchon, I. de Marco, Supercritical fluid extraction and fractionation of natural matter, J. Supercrit. Fluids 38 (2006) 146–166. [21] R.F. Rodrigues, A.K. Tashima, R.M.S. Pereira, R.S. Mohamed, F.A. Cabral, Coumarin solubility and extraction from Emburana (Torresea cearensis) seeds with supercritical carbon dioxide, J. Supercrit. Fluids 43 (2008) 375–382. [22] G.L. Filho, V.V. De Rosso, M.A.A. Meireles, P.T.V. Rosa, A.L. Oliveira, A.Z. Mercadante, F.A. Cabral, Supercritical CO2 extraction of carotenoids from pitanga fruits (Eugenia uniflora L.), J. Supercrit. Fluids 46 (2008) 33–39. [23] A.S.A.E. Standards, Method of determining and expressing fineness of feed materials by sieving, ASAE S319.3 (1997) 547. [24] M.C. Marcucci, F. Ferreres, C. García-Viguera, V.S. Bankova, S.L. De Castro, A.P. Dantas, P.H. Valente, N. Paulino, Phenolic compounds from Brazilian propolis with pharmacological activities, J. Ethnopharmacol. 74 (2001) 105–112. [25] A. Jain, G. Yang, S.H. Yalkowsky, Estimation of melting points of organic compounds, Ind. Eng. Chem. Res. 43 (2004) 7618–7621. [26] I.N. Tsibanogiannis, N.S. Kalospiros, D.P. Tassios, Prediction of normal boiling point temperature of medium/high molecular weight compounds, Ind. Eng. Chem. Res. 34 (1995) 997–1002. [27] D.K. Dundge, J.P. Heller, K.V. Wilson, Structure solubility correlations: organic compounds and dense carbon dioxide binary systems, Ind. Eng. Chem. Prod. Res. Dev. 24 (1985) 162–166. [28] R. Murga, M.T. Sanz, S. Beltrán, J.L. Cabezas, Solubility of three hydroxycinnamic acids in supercritical carbon dioxide, J. Supercrit. Fluids 27 (2003) 239–245. [29] H. Uchiyama, K. Mishima, S. Oka, M. Ezawa, M. Ide, Solubilities of flavone and 3-hydroxyflavone in supercritical carbon dioxide, J. Chem. Eng. Data 42 (1997) 570–573.
