- Email: [email protected]
Extraction of polyphenols from black tea – Conventional and ultrasound assisted extraction
Ultrasonics Sonochemistry xxx (2013) xxx–xxx
Contents lists available at ScienceDirect
Ultrasonics Sonochemistry journal homepage: www.elsevier.com/locate/ultson
Extraction of polyphenols from black tea – Conventional and ultrasound assisted extraction Simon Both a, Farid Chemat b, Jochen Strube a,⇑ a b
Institute for Separation and Process Technology, Clausthal University of Technology, D-38678 Clausthal-Zellerfeld, Germany Université d’Avignon et des Pays de Vaucluse, INRA, UMR 408, GREEN Extraction Team, F-84000 Avignon, France
a r t i c l e
i n f o
Article history: Received 6 August 2013 Received in revised form 4 November 2013 Accepted 8 November 2013 Available online xxxx Keywords: Extraction Black tea Maceration Ultrasound SEM
a b s t r a c t Products from plant raw materials gain increasing importance in food-, cosmetics and pharmaceutical industry. By way of contrast, due to lack of detailed physico-chemical fundamentals, existing production processes are economically not optimal designed. This leads to a need for deeper understanding of the processes and furthermore a systematic process and equipment design for the potentially applicable extraction techniques. Using the example of polyphenol extraction from black tea (Kenya), the conventional and ultrasound assisted extractions are investigated. Here, the state of the art as well as a comparison between the two techniques is in focus. Especially, resulting quasi-equilibria and mass transport kinetics serves as a criteria. The physico-chemical background is discussed taking particle size distributions and scanning electron microscope (SEM) measurements into account. Conclusively, process alternatives are projected and discussed. Hence, the present study makes influences of ultrasound technique on physico-chemical characteristics during extraction a subject of discussion. Ó 2013 Elsevier B.V. All rights reserved.
1. Introduction Plant-based products gain increasing importance in fields of nutrition, cosmetics and pharmaceutical industries. In recent years, the FAO (Food and Agriculture Organization of the United Nations) predicted annual growth rates of 6–8% for plant-based medical foods and phyto pharmaceuticals. Actually, in the USA, growth rates for these products lie about 15% and hence are higher than initially predicted [1–3]. Besides plant-based pharmaceuticals, for cosmetics and wellness products as well as plant-based flavors and perfumes double-digit annual growth rates are predicted [4– 9]. Here, not only the primal extract can be named as a product, but as well the purified substances out of these extracts. Pretreatment of the raw material and the extraction of valuable components are first steps in processing. Actually, due to a lack of physico-chemical fundamentals these unit operations are economically not optimal designed. First methodical approaches to design optimal processes are published for the extraction [2,10–12] as well as for further purifications of plant-based extracts [13]. For design and optimization of solid–liquid extraction processes, currently statistical experiment planning and physicochemical modeling are mostly investigated. First and foremost, equilibria as well as mass transport kinetics for target and side components are in fo-
⇑ Corresponding author. Tel.: +49 5323 72 2200; fax: +49 5323 72 3570. E-mail address: [email protected] (J. Strube).
cus. Considering technical and economical constraints, these parameters have to be enhanced. Extraction characteristics and hence the equipment are depending on the processed raw material and the solvent system [14]. For example, the location of target and side components as well as the structure of the raw material or the moisture content can be named as crucial factors on equilibrium and mass transport kinetics. Process intensifications and hence enhanced mass transport kinetics and quasi equilibrium behavior can be achieved by using ultrasound techniques [15]. Using the example of polyphenolextraction from black tea, the influence of ultrasound on mass transport and equilibrium are analyzed and discussed. Differences between conventional and ultrasound assisted maceration are investigated considering resulting particle size distributions and scanning electron microscope (SEM) analyzes.
2. Material and methods For the analysis and discussion of different process concepts, especially the equilibrium and mass transport kinetic is considered. The fundamentals of enhancements are discussed employing particle size distribution and SEM measurements. The laboratory experiments are carried out with black tea from Kenya. As extracting solvent pure ethanol as well as a mixture consisting of ethanol/water 90/10 (m/m) are used. The total content of polyphenols is quantified by five stage maceration (each 24 h), fol-
1350-4177/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ultsonch.2013.11.005
Please cite this article in press as: S. Both et al., Extraction of polyphenols from black tea – Conventional and ultrasound assisted extraction, Ultrason. Sonochem. (2013), http://dx.doi.org/10.1016/j.ultsonch.2013.11.005
2
S. Both et al. / Ultrasonics Sonochemistry xxx (2013) xxx–xxx
lowed by 96 h control percolation with 0.1 BV/h to ensure total extraction. This procedure is state-of-the-art [11]. For the total amount of polyphenols and hence a yield of Yi = 100% a loading of cpolyphenols = 211.4 g/kgDry substance tea is determined. The dry substance of the extract is defined as a further side component. Ultrasound-assisted extractions (UAE) were performed in an ultrasonic extraction reactor PEX1 (R.E.U.S., Contes, France) with 14 10 cm internal dimensions and maximal capacity of 1 L, equipped with a transducer at the base of jug operating at a frequency of 25 kHz with maximum input power (output power of the generator) of 150 W. The double-layered mantle (with water circulation) allowed the control of extraction temperature by cooling/heating systems. Considering the actual input power from the device is converted to heat which is dissipated in the medium, calorimetric measurements were performed to assess actual ultrasound power, calculated as shown in the Eq. (1) below.
P ¼ m cP
dT dt
ð1Þ
where cP is the heat capacity of the solvent at constant pressure (J/g/ °C), m is the mass of solvent (g) and dT/dt is temperature rise per second. Fig. 1 depicts the used equipment for the comparison of mass transport kinetics as well as for the determination of the equilibrium line. Mass transport kinetics and phase equilibria are determined at a constant temperature of 40 °C and a solid/liquid ratio of 1/3 (m/m). The solid–liquid equilibrium line is determined in multi stage maceration experiments. Each maceration step has to reach equilibrium for this. Due to additional total extractions and closing mass balances for each component in each phase, the equilibrium concentrations in solid and liquid phase can be measured. To close all mass balances, in addition to polyphenol concentration, the amount of water, dry substance and further side components as well as for the solvent has to be measured in both phases. Table 1 shows the necessary methods. The total amount of polyphenols is determined using the Folin– Ciocalteu test [16] from Seppal. Here, the adsorption Absi is detected at a specific wave length of 620 nm. By using reference solutions with a defined concentration of cRef = 3 g/L, the concentrations of polyphenols cPP,i in each fraction can be determined via Eq. (2).
cPP;i ¼
Absi cRef AbsRef
ð2Þ
Table 1 Materials and methods.
Total polyphenol Dry substance Water Solvent mass
In raw material
In liquid phase
Multi stage maceration Dry balance Dry balance, toluene distillation Dry balance
Folin–Ciocalteu test Dry balance Karl–Fischer titration Calculation
The dry substance as well as the residual humidity in the raw material is determined using the moisture analyzer Sartorius MA 150. SEM measurements are necessary for analyzing the raw material matrix and hence the quantification of the ultrasonic influence. The measurements are carried out using FEI/Philips XL30 FEG ESEM. Each sample is compared with a reference sample of fresh tea. 3. Therory 3.1. Physico-chemical properties One way to consider the physico-chemical fundamentals of extraction is to build up models, describing the different phenomena. For model-based process design the determination of associated model parameters is necessary. Models considering equilibrium behavior [17] can be named as well as models considering equilibrium, mass transport kinetic and fluid dynamic [18]. With an increasing amount of physico-chemical data, the experimental effort increases as well as the prediction accuracy of these models. Further models include the named physico-chemical properties and, in addition, the structure of the raw material as well as particle size and loading distributions of the ingredients [11]. The parameters named here are used to identify and discuss the differences between the two types of extraction: Conventional maceration versus ultrasound assisted extraction. Equilibrium and mass transport kinetics are in focus. Nevertheless, the basics of ultrasound extraction are not discussed in detail. In literature the fundamentals as well as fields of application are given for laboratory and industrial scale [15]. 3.2. Equipment concepts The optimal equipment for the extraction of valuable components from plant material depends on the raw material as well as on the processed volume rates significantly [19]. In general, a differentiation between raw material characteristics, physico-chemical parameters and economic factors is made. As raw material characteristics - Accessibility of target and side components. - Structure of the raw material matrix. - Moisture content and swelling behavior of the processed raw material, respectively can be named [14]. Physico-chemical parameters depending on the characteristics are [19]. - Equilibrium. - Mass transport kinetic. - Fluid dynamic (depending on the equipment). - The design of an economically optimized process requires to take further economic factors into account [13]. - Apparative efforts and invetiment cost, respectively. - Operating cost.
Fig. 1. R.E.U.S. ultrasound equipment.
The solvent and energy consumption influences the operating conditions significantly. These costs have to be considered regarding the process design. An optimum between solvent consumption
Please cite this article in press as: S. Both et al., Extraction of polyphenols from black tea – Conventional and ultrasound assisted extraction, Ultrason. Sonochem. (2013), http://dx.doi.org/10.1016/j.ultsonch.2013.11.005
S. Both et al. / Ultrasonics Sonochemistry xxx (2013) xxx–xxx
3
Fig. 3. Quasi-equilibrium lines for conventional and ultrasound assisted extraction of tea polyphenols with pure ethanol and ethanol/water 90/10 (m/m).
Fig. 2. (a) Multi stage maceration, conventional; (b) multi stage maceration, ultrasound assisted; (c) multi stage maceration, combination of (a) and (b).
and the target components yield has to be determined. For low value products, such as sucrose from sugar beets, the solvent consumption and hence the operating conditions have to be kept low. Only for high value products, such as vanillin from vanilla beans, the solvent consumption is subsidiary. Yields of 100% will bear the operating costs [19]. Hence, as further constraints in process design the yield and purity of the extract and, respectively, the quality of the product can be named. Considering the named aspects, different extraction equipment concepts can be investigated. In this study, a first comparison between conventional and ultrasound assisted maceration is focused on. Characteristics, like the accessibility and hence quasi-equilibria and mass transport kinetics are analyzed and discussed in detail. Initially, three concepts for investigation (Fig. 2) are defined. First of all, the conventional multi stage maceration (Fig. 2a) is compared to the ultrasound assisted multi stage maceration (Fig. 2b). Additionally, the influence of ultrasound is quantified by use of a pretreatment with ultrasound at the first stage and further conventional processing in maceration (Fig. 2c).
4. First results and discussion 4.1. Equilibrium line by multi stage maceration and total extraction Equilibrium lines can be determined by using multi stage maceration followed by control percolation [11,18]. The equilibrium will shift towards the extract phase, if the slope of the line decreases. This results in high extract phase concentrations and low solid phase loadings. In general, three mayor types of equilibrium lines can arise: The linear type [11,19], the Langmuir type [11] and the anti-Langmuir type. The anti-Langmuir type results from limiting capacities in the liquid phase. Equilibrium relation leads to high solid phase loadings and low concentrations in liquid phase. Hence, this is more of theoretical nature. [18]
Fig. 3 depicts the quasi-equilibrium lines for polyphenols from black tea at different solvent ratios (pure ethanol and ethanol/ water) and two maceration techniques (with and without ultrasound assistance). Each equilibrium line shows the Langmuir type. The solvent consisting of ethanol and water shows the best behavior. For a single stage maceration with a solid/liquid ratio of 1/3 an extract concentration of approx. cPolyphenols = 21 g/L can be reached. In the solid phase a residual loading of about 14 g/L results. For pure ethanol, the concentration in the liquid phase ranges between 2 and 4 g/L and for solid phase about 22 g/L. Hence, the equilibrium is increased towards the solid phase. Different maximal residual polyphenol loadings qmax in the solid phase are depending on used technique and solvent. The maximal residual loading for the extraction with pure ethanol lies at about 22 g/gsolid, whereas the loading for the solvent ratio ethanol/water 90/10 (m/m) lies at around 15–17 g/gsolid. Lower maximal residual loadings lead to higher concentrations in the liquid phase and hence lower solvent consumption and operating cost, respectively. Solid phase loadings below maximal residual loadings are attainable with high solvent consumption. Each maceration step results in a low liquid phase concentration and therefore in a low decrease of residual loading. Only with constant high concentration gradients, a total extraction is economically feasible. This can be realized using percolation equipment. A total extraction is achievable with minimal solvent consumption. Nevertheless, maceration equipment is useful for raw material-solvent systems with a linear equilibrium line having a low slope. For one equilibrium stage, the major part of the ingredients is extracted. Low residual concentrations and high extract concentrations can be recognized. Hence, a lower dilution than for the percolation equipment can be realized with low efforts. As described above, Fig. 3 depicts the quasi-equilibrium lines, determined via multi stage maceration. To gain equilibrium values for higher liquid phase concentrations, the liquid phase has to be preloaded. Hence, the concentration gradient between the two phases is reduced. Water can increase the solubility of polar polyphenols in the solvent mixture. In addition, water leads to swelling of the raw material matrix. A higher accessibility for the ingredients is resulting. Hence, better extraction behavior is achieved by higher accessibility and a higher solubility. Furthermore, the ultrasound technique attacks the surface of the raw material particles. Basically, either the whole surface is removed or occasional punctures leading to an increased accessibility can occur [15]. Fig. 4 depicts the two principles. It has to be kept in mind that the surface structure after affection depends on the treated raw material as well as on the exposure time.
Please cite this article in press as: S. Both et al., Extraction of polyphenols from black tea – Conventional and ultrasound assisted extraction, Ultrason. Sonochem. (2013), http://dx.doi.org/10.1016/j.ultsonch.2013.11.005
4
S. Both et al. / Ultrasonics Sonochemistry xxx (2013) xxx–xxx
Fig. 4. Schematic principles of ultrasonic attacks.
4.2. Mass transport kinetic The comparison of the mass transfer kinetics for the maceration with ethanol and ethanol/water mixture and both regarded techniques is depicted in Fig. 5. Due to higher equilibrium concentration with use of the ultrasound technique, an absolute higher mass transport occurs. The kinetic is enhanced, whereas the differences of the relative mass transport with and without ultrasound lies in the range ±5%. Hence, the differences lie in the range of the error of measurement. For both solvents the same extraction kinetic characteristics can be recognized. Nevertheless, differences occur in equilibrium behavior. Using the mixture of ethanol and water instead of pure ethanol, the equilibrium extract concentration can be increased by about 500%. Differences between the two solvents lie around 13% for each technique. Hence, the choice of solvent is more important than the choice of technique. Nevertheless, the use of ultrasound can improve the extraction for a predefined solvent. For the solvent mixture ethanol/water the influence of ultrasound as pretreatment (concept: Fig. 1c) in comparison to multistage ultrasound assisted maceration can be investigated using the mass transport kinetics and equilibrium behavior as objectives. Fig. 6 depicts the extraction kinetics for the first and second maceration stage. Here the first stage is assisted by ultrasound. The second stage is either with ultrasound assistance or without. Even in the second stage, ultrasound leads to higher concentrations in liquid phase (approx. 13%). The extraction kinetic, as seen as well in Fig. 5, is not enhanced by the use of ultrasound. The quasi-equilibrium is reached between 20 and 30 min. for both cases. This raises the question, whether the surface is consistently removed by ultrasound or if an attack occurs more partially and hence if only the specific surface for mass transport is increased. By employment of particle size and SEM analyzes, the principle of attack by ultrasound is investigated. The results are depicted and discussed in the following sections. The advantages and disadvantages in equilibrium behavior and mass transport enhancement have been discussed. Chances for industrial scale applications have to be determined by use of specific separation costs and hence analysis of operational costs [20–
Fig. 5. Comparison: conventional and ultrasound assisted maceration with ethanol and ethanol/water 90/10 (m/m) mixture.
Fig. 6. Mass transport kinetic in two stage maceration: first stage ultrasound assisted, second stage with and without employment of ultrasound.
Fig. 7. Comparison in particle size distribution.
22]. At industrial level one or two more steps in the extraction process can represent an unacceptable additional cost. 4.3. Solid particles – differences by using different techniques 4.3.1. Particle size distribution Ultrasonic employment leads to higher decrease in particle size than using conventional extraction (Fig. 7). The mean particle diameter is reduced from initially 660 lm to 640 lm using conventional maceration and 610 lm using ultrasound assisted maceration. Hence, the particle diameter is about 5% lower with ultrasonic employment. In each case, the extraction time is 30 min. Analyzing the content of dry substance in the extract shows same bias. The content is increased from 2.9% to 4.3% with use of ultrasound and hence by about 30–35%. Thereby, the dry substance is defined as a multi component mixture, containing a high amount of target substances, such as polyphenols. Nevertheless, the raw material matrix is disrupted as well. The extract becomes dull by use of ultrasound. 4.3.2. SEM measurements – cell disruption Fig. 8 shows the raw material, the ginded black tea leaf. Because of a high degree of grinding (mean particle diameter of about 660 lm) even without further treatment the different botanic parts can be seen. Exemplary, epidermis, stoma and vascular bundle can be named. 30 min extraction leads to a removal of the surface and hence further botanic elements appearing. Significant differences between the two methods, conventional and ultrasound assisted maceration, cannot be identified. Here, equal parts of the matrix can be seen. Due to the variablility of the leaf particles, no quantification is possible. Each leaf particle is extracted from overall surface.
Please cite this article in press as: S. Both et al., Extraction of polyphenols from black tea – Conventional and ultrasound assisted extraction, Ultrason. Sonochem. (2013), http://dx.doi.org/10.1016/j.ultsonch.2013.11.005
S. Both et al. / Ultrasonics Sonochemistry xxx (2013) xxx–xxx
5
Three different equipment concepts are discussed on basis of the named results. Here, the conventional multi stage maceration is compared to ultrasound assisted multi stage maceration. In addition, the pre-treatment with ultrasound as a first maceration step followed by further conventional maceration steps is discussed. Concluding, ultrasound leads to higher concentrations in liquid phase, even if a pretreatment with ultrasound has been proceeding. The yield in the second maceration stage is about 13% higher for the two stage ultrasound maceration than for single stage ultrasound assisted extraction followed by conventional maceration. Acknowledgements The authors like to thank their colleagues at the ‘‘Groupe de Recherche en Eco Extraction des produits Naturels’’ at the Université d’Avignon et des Pays de Vaucluse as well as Isabelle Bornard from the National Institute of Agronomic Research (INRA, Avignon) for her kind contribution concerning SEM measurements and further botanic discussions. References
Fig. 8. Tea leaf analyzes via SEM (1: Initially, 2,3: after 30 min maceration with and without ultrasound).
5. Conclusions First results for the extraction of polyphenols from black tea are presented. Here, the two types of maceration, conventional and ultrasound assisted, as well as the influence of water in the extracting solvent are discussed concerning equilibrium and mass transport kinetic. Differences are analyzed employing the dry substance and polyphenol content in the extract as well as particle size distribution and SEM measurements. The use of ultrasonic intensification leads to higher quasi-equilibrium concentrations in the liquid phase. The content of polyphenols is increased by approx. 15%. Thereby, the solubility of the ingredients, here polyphenols, as well as the accessibility for the solvent are discussed. Ultrasound employment does not compensate the content of water in the solvent. The water leads to higher solubility and swelling of the raw material and hence higher yields. Therefore, ultrasound can assist the extraction for a pre-chosen and optimized solvent. Via particle size measurements a diminution of the mean particle diameter during extraction is detected. After an extraction time of 30 min, the particle diameter is 5% lower using ultrasound compared to conventional extraction. In addition to polyphenol extraction an extraction of further leaf elements is detected via analyzing the content of the dry substance in the extract phase. This content is about 30–35% higher by use of ultrasound. SEM analyzes do not show significant differences for tea leaf particles between the two techniques. Because of the high degree of grinding (mean particle diameter of about 600-700 lm) various botanic elements, such as epidermis stoma and vascular bundle can be identified. During extraction, an increasingly number of leaf structures are appearing. Nevertheless, differences between the two maceration types cannot be quantified. The variability of the leaf particles and hence the leaf elements observed via SEM does not allow any quantification.
[1] www.fao.de, FAO, Food and Agricultural Organisation der Vereinten Nationen, 2012. [2] H.J. Bart et al., Positionspapier Phytoextrakte – Produkte und Prozesse, DECHEMA e.V, 2012. [3] M. Tegtmeier, Pflanzenextraktion: Schlüsseltechnologie zur nachhaltigen Nutzung von Bio-Ressourcen, Chemie Ingenieur Technik 84 (6) (2012) 880– 882. [4] A.E. Sloan, Getting Ahead of the Curve: Phytochemicals, Sloan Trends Inc., 2009. [5] T. De Silva, Industrial utilization of medicinal plants in developing countries, www.fao.org, Vienna, Austria, 2012. [6] www.researchandmarkets.com, access 2012. [7] www.slideshare.net, A study of the European cosmetic industry, Executive summary, Global Insight, Mailand, 2007. [8] S. Kumar, Exploratory analysis of global cosmetic industry: major players, technology and market trends, Technovation 25 (11) (2005) 1263–1272. [9] www.aginskyconsuting.com, Cosmetic market research summary, Aginsky Consulting Group, Portland, US, 2007. [10] M. Kaßing, U. Jenelten, J. Schenk, J. Strube, A new approach for process development for plant-based extraction processes, Chem. Eng. Technol. 33 (3) (2010) 377–387. [11] M. Kaßing, U. Jenelten, J. Schenk, R. Hänsch, J. Strube, Combination of rigorous and statistical modeling for process development of plant-based extractions based on mass balances and botanical aspects, Chem. Eng. Technol. 35 (1) (2012) 109–132. [12] J. Strube et al., Kapitel 5: Process Engineering and Mini-Plant Technology, in: H.J. Bart, S. Pilz (Eds.), Industrial Scale Natural Products, Wiley-VCH, 2010. [13] J.P. Josch, S. Both, J. Strube, Characterization of feed properties for conceptual process design involving complex mixtures such as natural extracts, Food Nutr. Sci. 3 (2012) 836–859. [14] M. Kaßing, Process Development for Plant-based Extract Production, Shaker Verlag, 2012. [15] F. Chemat, Éco-Extraction du Végétal, Dunod, Paris, 2011. [16] M. Obanda, P.O. Owuor, Flavanol composition and caffeine content of green leaf as quality potential indicators of Kenyan black teas, J. Sci. Food Agric. 74 (1997) 209–215. [17] A. Mersmann, M. Kind, J. Stichlmair, Thermische Verfahrenstechnik – Grundlagen und Methoden, Springer Verlag, Berlin Heidelberg New York, 2005. [18] S. Both et al., Model-based equipment-design for plant-based extraction processes –considering botanic and thermodynamic aspects. C. R. Chimie (2013), http://dx.doi.org/10.1016/j.crci.2013.11.004 [19] S. Both et al., Optimizing established processes like sugar extraction from sugar beets – design of experiments versus physicochemical modeling, Chem. Eng. Technol. 36 (2013) 12. [20] E. Blass, Methoden, Zielsuche, Lösungssuche, Lösungsauswahl, Springer, Berlin, 1997. [21] M. Hurmw, M. Jarvelainen, Combined process synthesis and simulation system for feasiilty studies, Comput. Chem. Eng. 19 (1995) 663–668. [22] S. Both et al., Resource efficient process technology for energy plants, Chemie Ingenieur Technik 85 (8) (2013) 1282–1289.
Please cite this article in press as: S. Both et al., Extraction of polyphenols from black tea – Conventional and ultrasound assisted extraction, Ultrason. Sonochem. (2013), http://dx.doi.org/10.1016/j.ultsonch.2013.11.005
Contents lists available at ScienceDirect
Ultrasonics Sonochemistry journal homepage: www.elsevier.com/locate/ultson
Extraction of polyphenols from black tea – Conventional and ultrasound assisted extraction Simon Both a, Farid Chemat b, Jochen Strube a,⇑ a b
Institute for Separation and Process Technology, Clausthal University of Technology, D-38678 Clausthal-Zellerfeld, Germany Université d’Avignon et des Pays de Vaucluse, INRA, UMR 408, GREEN Extraction Team, F-84000 Avignon, France
a r t i c l e
i n f o
Article history: Received 6 August 2013 Received in revised form 4 November 2013 Accepted 8 November 2013 Available online xxxx Keywords: Extraction Black tea Maceration Ultrasound SEM
a b s t r a c t Products from plant raw materials gain increasing importance in food-, cosmetics and pharmaceutical industry. By way of contrast, due to lack of detailed physico-chemical fundamentals, existing production processes are economically not optimal designed. This leads to a need for deeper understanding of the processes and furthermore a systematic process and equipment design for the potentially applicable extraction techniques. Using the example of polyphenol extraction from black tea (Kenya), the conventional and ultrasound assisted extractions are investigated. Here, the state of the art as well as a comparison between the two techniques is in focus. Especially, resulting quasi-equilibria and mass transport kinetics serves as a criteria. The physico-chemical background is discussed taking particle size distributions and scanning electron microscope (SEM) measurements into account. Conclusively, process alternatives are projected and discussed. Hence, the present study makes influences of ultrasound technique on physico-chemical characteristics during extraction a subject of discussion. Ó 2013 Elsevier B.V. All rights reserved.
1. Introduction Plant-based products gain increasing importance in fields of nutrition, cosmetics and pharmaceutical industries. In recent years, the FAO (Food and Agriculture Organization of the United Nations) predicted annual growth rates of 6–8% for plant-based medical foods and phyto pharmaceuticals. Actually, in the USA, growth rates for these products lie about 15% and hence are higher than initially predicted [1–3]. Besides plant-based pharmaceuticals, for cosmetics and wellness products as well as plant-based flavors and perfumes double-digit annual growth rates are predicted [4– 9]. Here, not only the primal extract can be named as a product, but as well the purified substances out of these extracts. Pretreatment of the raw material and the extraction of valuable components are first steps in processing. Actually, due to a lack of physico-chemical fundamentals these unit operations are economically not optimal designed. First methodical approaches to design optimal processes are published for the extraction [2,10–12] as well as for further purifications of plant-based extracts [13]. For design and optimization of solid–liquid extraction processes, currently statistical experiment planning and physicochemical modeling are mostly investigated. First and foremost, equilibria as well as mass transport kinetics for target and side components are in fo-
⇑ Corresponding author. Tel.: +49 5323 72 2200; fax: +49 5323 72 3570. E-mail address: [email protected] (J. Strube).
cus. Considering technical and economical constraints, these parameters have to be enhanced. Extraction characteristics and hence the equipment are depending on the processed raw material and the solvent system [14]. For example, the location of target and side components as well as the structure of the raw material or the moisture content can be named as crucial factors on equilibrium and mass transport kinetics. Process intensifications and hence enhanced mass transport kinetics and quasi equilibrium behavior can be achieved by using ultrasound techniques [15]. Using the example of polyphenolextraction from black tea, the influence of ultrasound on mass transport and equilibrium are analyzed and discussed. Differences between conventional and ultrasound assisted maceration are investigated considering resulting particle size distributions and scanning electron microscope (SEM) analyzes.
2. Material and methods For the analysis and discussion of different process concepts, especially the equilibrium and mass transport kinetic is considered. The fundamentals of enhancements are discussed employing particle size distribution and SEM measurements. The laboratory experiments are carried out with black tea from Kenya. As extracting solvent pure ethanol as well as a mixture consisting of ethanol/water 90/10 (m/m) are used. The total content of polyphenols is quantified by five stage maceration (each 24 h), fol-
1350-4177/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ultsonch.2013.11.005
Please cite this article in press as: S. Both et al., Extraction of polyphenols from black tea – Conventional and ultrasound assisted extraction, Ultrason. Sonochem. (2013), http://dx.doi.org/10.1016/j.ultsonch.2013.11.005
2
S. Both et al. / Ultrasonics Sonochemistry xxx (2013) xxx–xxx
lowed by 96 h control percolation with 0.1 BV/h to ensure total extraction. This procedure is state-of-the-art [11]. For the total amount of polyphenols and hence a yield of Yi = 100% a loading of cpolyphenols = 211.4 g/kgDry substance tea is determined. The dry substance of the extract is defined as a further side component. Ultrasound-assisted extractions (UAE) were performed in an ultrasonic extraction reactor PEX1 (R.E.U.S., Contes, France) with 14 10 cm internal dimensions and maximal capacity of 1 L, equipped with a transducer at the base of jug operating at a frequency of 25 kHz with maximum input power (output power of the generator) of 150 W. The double-layered mantle (with water circulation) allowed the control of extraction temperature by cooling/heating systems. Considering the actual input power from the device is converted to heat which is dissipated in the medium, calorimetric measurements were performed to assess actual ultrasound power, calculated as shown in the Eq. (1) below.
P ¼ m cP
dT dt
ð1Þ
where cP is the heat capacity of the solvent at constant pressure (J/g/ °C), m is the mass of solvent (g) and dT/dt is temperature rise per second. Fig. 1 depicts the used equipment for the comparison of mass transport kinetics as well as for the determination of the equilibrium line. Mass transport kinetics and phase equilibria are determined at a constant temperature of 40 °C and a solid/liquid ratio of 1/3 (m/m). The solid–liquid equilibrium line is determined in multi stage maceration experiments. Each maceration step has to reach equilibrium for this. Due to additional total extractions and closing mass balances for each component in each phase, the equilibrium concentrations in solid and liquid phase can be measured. To close all mass balances, in addition to polyphenol concentration, the amount of water, dry substance and further side components as well as for the solvent has to be measured in both phases. Table 1 shows the necessary methods. The total amount of polyphenols is determined using the Folin– Ciocalteu test [16] from Seppal. Here, the adsorption Absi is detected at a specific wave length of 620 nm. By using reference solutions with a defined concentration of cRef = 3 g/L, the concentrations of polyphenols cPP,i in each fraction can be determined via Eq. (2).
cPP;i ¼
Absi cRef AbsRef
ð2Þ
Table 1 Materials and methods.
Total polyphenol Dry substance Water Solvent mass
In raw material
In liquid phase
Multi stage maceration Dry balance Dry balance, toluene distillation Dry balance
Folin–Ciocalteu test Dry balance Karl–Fischer titration Calculation
The dry substance as well as the residual humidity in the raw material is determined using the moisture analyzer Sartorius MA 150. SEM measurements are necessary for analyzing the raw material matrix and hence the quantification of the ultrasonic influence. The measurements are carried out using FEI/Philips XL30 FEG ESEM. Each sample is compared with a reference sample of fresh tea. 3. Therory 3.1. Physico-chemical properties One way to consider the physico-chemical fundamentals of extraction is to build up models, describing the different phenomena. For model-based process design the determination of associated model parameters is necessary. Models considering equilibrium behavior [17] can be named as well as models considering equilibrium, mass transport kinetic and fluid dynamic [18]. With an increasing amount of physico-chemical data, the experimental effort increases as well as the prediction accuracy of these models. Further models include the named physico-chemical properties and, in addition, the structure of the raw material as well as particle size and loading distributions of the ingredients [11]. The parameters named here are used to identify and discuss the differences between the two types of extraction: Conventional maceration versus ultrasound assisted extraction. Equilibrium and mass transport kinetics are in focus. Nevertheless, the basics of ultrasound extraction are not discussed in detail. In literature the fundamentals as well as fields of application are given for laboratory and industrial scale [15]. 3.2. Equipment concepts The optimal equipment for the extraction of valuable components from plant material depends on the raw material as well as on the processed volume rates significantly [19]. In general, a differentiation between raw material characteristics, physico-chemical parameters and economic factors is made. As raw material characteristics - Accessibility of target and side components. - Structure of the raw material matrix. - Moisture content and swelling behavior of the processed raw material, respectively can be named [14]. Physico-chemical parameters depending on the characteristics are [19]. - Equilibrium. - Mass transport kinetic. - Fluid dynamic (depending on the equipment). - The design of an economically optimized process requires to take further economic factors into account [13]. - Apparative efforts and invetiment cost, respectively. - Operating cost.
Fig. 1. R.E.U.S. ultrasound equipment.
The solvent and energy consumption influences the operating conditions significantly. These costs have to be considered regarding the process design. An optimum between solvent consumption
Please cite this article in press as: S. Both et al., Extraction of polyphenols from black tea – Conventional and ultrasound assisted extraction, Ultrason. Sonochem. (2013), http://dx.doi.org/10.1016/j.ultsonch.2013.11.005
S. Both et al. / Ultrasonics Sonochemistry xxx (2013) xxx–xxx
3
Fig. 3. Quasi-equilibrium lines for conventional and ultrasound assisted extraction of tea polyphenols with pure ethanol and ethanol/water 90/10 (m/m).
Fig. 2. (a) Multi stage maceration, conventional; (b) multi stage maceration, ultrasound assisted; (c) multi stage maceration, combination of (a) and (b).
and the target components yield has to be determined. For low value products, such as sucrose from sugar beets, the solvent consumption and hence the operating conditions have to be kept low. Only for high value products, such as vanillin from vanilla beans, the solvent consumption is subsidiary. Yields of 100% will bear the operating costs [19]. Hence, as further constraints in process design the yield and purity of the extract and, respectively, the quality of the product can be named. Considering the named aspects, different extraction equipment concepts can be investigated. In this study, a first comparison between conventional and ultrasound assisted maceration is focused on. Characteristics, like the accessibility and hence quasi-equilibria and mass transport kinetics are analyzed and discussed in detail. Initially, three concepts for investigation (Fig. 2) are defined. First of all, the conventional multi stage maceration (Fig. 2a) is compared to the ultrasound assisted multi stage maceration (Fig. 2b). Additionally, the influence of ultrasound is quantified by use of a pretreatment with ultrasound at the first stage and further conventional processing in maceration (Fig. 2c).
4. First results and discussion 4.1. Equilibrium line by multi stage maceration and total extraction Equilibrium lines can be determined by using multi stage maceration followed by control percolation [11,18]. The equilibrium will shift towards the extract phase, if the slope of the line decreases. This results in high extract phase concentrations and low solid phase loadings. In general, three mayor types of equilibrium lines can arise: The linear type [11,19], the Langmuir type [11] and the anti-Langmuir type. The anti-Langmuir type results from limiting capacities in the liquid phase. Equilibrium relation leads to high solid phase loadings and low concentrations in liquid phase. Hence, this is more of theoretical nature. [18]
Fig. 3 depicts the quasi-equilibrium lines for polyphenols from black tea at different solvent ratios (pure ethanol and ethanol/ water) and two maceration techniques (with and without ultrasound assistance). Each equilibrium line shows the Langmuir type. The solvent consisting of ethanol and water shows the best behavior. For a single stage maceration with a solid/liquid ratio of 1/3 an extract concentration of approx. cPolyphenols = 21 g/L can be reached. In the solid phase a residual loading of about 14 g/L results. For pure ethanol, the concentration in the liquid phase ranges between 2 and 4 g/L and for solid phase about 22 g/L. Hence, the equilibrium is increased towards the solid phase. Different maximal residual polyphenol loadings qmax in the solid phase are depending on used technique and solvent. The maximal residual loading for the extraction with pure ethanol lies at about 22 g/gsolid, whereas the loading for the solvent ratio ethanol/water 90/10 (m/m) lies at around 15–17 g/gsolid. Lower maximal residual loadings lead to higher concentrations in the liquid phase and hence lower solvent consumption and operating cost, respectively. Solid phase loadings below maximal residual loadings are attainable with high solvent consumption. Each maceration step results in a low liquid phase concentration and therefore in a low decrease of residual loading. Only with constant high concentration gradients, a total extraction is economically feasible. This can be realized using percolation equipment. A total extraction is achievable with minimal solvent consumption. Nevertheless, maceration equipment is useful for raw material-solvent systems with a linear equilibrium line having a low slope. For one equilibrium stage, the major part of the ingredients is extracted. Low residual concentrations and high extract concentrations can be recognized. Hence, a lower dilution than for the percolation equipment can be realized with low efforts. As described above, Fig. 3 depicts the quasi-equilibrium lines, determined via multi stage maceration. To gain equilibrium values for higher liquid phase concentrations, the liquid phase has to be preloaded. Hence, the concentration gradient between the two phases is reduced. Water can increase the solubility of polar polyphenols in the solvent mixture. In addition, water leads to swelling of the raw material matrix. A higher accessibility for the ingredients is resulting. Hence, better extraction behavior is achieved by higher accessibility and a higher solubility. Furthermore, the ultrasound technique attacks the surface of the raw material particles. Basically, either the whole surface is removed or occasional punctures leading to an increased accessibility can occur [15]. Fig. 4 depicts the two principles. It has to be kept in mind that the surface structure after affection depends on the treated raw material as well as on the exposure time.
Please cite this article in press as: S. Both et al., Extraction of polyphenols from black tea – Conventional and ultrasound assisted extraction, Ultrason. Sonochem. (2013), http://dx.doi.org/10.1016/j.ultsonch.2013.11.005
4
S. Both et al. / Ultrasonics Sonochemistry xxx (2013) xxx–xxx
Fig. 4. Schematic principles of ultrasonic attacks.
4.2. Mass transport kinetic The comparison of the mass transfer kinetics for the maceration with ethanol and ethanol/water mixture and both regarded techniques is depicted in Fig. 5. Due to higher equilibrium concentration with use of the ultrasound technique, an absolute higher mass transport occurs. The kinetic is enhanced, whereas the differences of the relative mass transport with and without ultrasound lies in the range ±5%. Hence, the differences lie in the range of the error of measurement. For both solvents the same extraction kinetic characteristics can be recognized. Nevertheless, differences occur in equilibrium behavior. Using the mixture of ethanol and water instead of pure ethanol, the equilibrium extract concentration can be increased by about 500%. Differences between the two solvents lie around 13% for each technique. Hence, the choice of solvent is more important than the choice of technique. Nevertheless, the use of ultrasound can improve the extraction for a predefined solvent. For the solvent mixture ethanol/water the influence of ultrasound as pretreatment (concept: Fig. 1c) in comparison to multistage ultrasound assisted maceration can be investigated using the mass transport kinetics and equilibrium behavior as objectives. Fig. 6 depicts the extraction kinetics for the first and second maceration stage. Here the first stage is assisted by ultrasound. The second stage is either with ultrasound assistance or without. Even in the second stage, ultrasound leads to higher concentrations in liquid phase (approx. 13%). The extraction kinetic, as seen as well in Fig. 5, is not enhanced by the use of ultrasound. The quasi-equilibrium is reached between 20 and 30 min. for both cases. This raises the question, whether the surface is consistently removed by ultrasound or if an attack occurs more partially and hence if only the specific surface for mass transport is increased. By employment of particle size and SEM analyzes, the principle of attack by ultrasound is investigated. The results are depicted and discussed in the following sections. The advantages and disadvantages in equilibrium behavior and mass transport enhancement have been discussed. Chances for industrial scale applications have to be determined by use of specific separation costs and hence analysis of operational costs [20–
Fig. 5. Comparison: conventional and ultrasound assisted maceration with ethanol and ethanol/water 90/10 (m/m) mixture.
Fig. 6. Mass transport kinetic in two stage maceration: first stage ultrasound assisted, second stage with and without employment of ultrasound.
Fig. 7. Comparison in particle size distribution.
22]. At industrial level one or two more steps in the extraction process can represent an unacceptable additional cost. 4.3. Solid particles – differences by using different techniques 4.3.1. Particle size distribution Ultrasonic employment leads to higher decrease in particle size than using conventional extraction (Fig. 7). The mean particle diameter is reduced from initially 660 lm to 640 lm using conventional maceration and 610 lm using ultrasound assisted maceration. Hence, the particle diameter is about 5% lower with ultrasonic employment. In each case, the extraction time is 30 min. Analyzing the content of dry substance in the extract shows same bias. The content is increased from 2.9% to 4.3% with use of ultrasound and hence by about 30–35%. Thereby, the dry substance is defined as a multi component mixture, containing a high amount of target substances, such as polyphenols. Nevertheless, the raw material matrix is disrupted as well. The extract becomes dull by use of ultrasound. 4.3.2. SEM measurements – cell disruption Fig. 8 shows the raw material, the ginded black tea leaf. Because of a high degree of grinding (mean particle diameter of about 660 lm) even without further treatment the different botanic parts can be seen. Exemplary, epidermis, stoma and vascular bundle can be named. 30 min extraction leads to a removal of the surface and hence further botanic elements appearing. Significant differences between the two methods, conventional and ultrasound assisted maceration, cannot be identified. Here, equal parts of the matrix can be seen. Due to the variablility of the leaf particles, no quantification is possible. Each leaf particle is extracted from overall surface.
Please cite this article in press as: S. Both et al., Extraction of polyphenols from black tea – Conventional and ultrasound assisted extraction, Ultrason. Sonochem. (2013), http://dx.doi.org/10.1016/j.ultsonch.2013.11.005
S. Both et al. / Ultrasonics Sonochemistry xxx (2013) xxx–xxx
5
Three different equipment concepts are discussed on basis of the named results. Here, the conventional multi stage maceration is compared to ultrasound assisted multi stage maceration. In addition, the pre-treatment with ultrasound as a first maceration step followed by further conventional maceration steps is discussed. Concluding, ultrasound leads to higher concentrations in liquid phase, even if a pretreatment with ultrasound has been proceeding. The yield in the second maceration stage is about 13% higher for the two stage ultrasound maceration than for single stage ultrasound assisted extraction followed by conventional maceration. Acknowledgements The authors like to thank their colleagues at the ‘‘Groupe de Recherche en Eco Extraction des produits Naturels’’ at the Université d’Avignon et des Pays de Vaucluse as well as Isabelle Bornard from the National Institute of Agronomic Research (INRA, Avignon) for her kind contribution concerning SEM measurements and further botanic discussions. References
Fig. 8. Tea leaf analyzes via SEM (1: Initially, 2,3: after 30 min maceration with and without ultrasound).
5. Conclusions First results for the extraction of polyphenols from black tea are presented. Here, the two types of maceration, conventional and ultrasound assisted, as well as the influence of water in the extracting solvent are discussed concerning equilibrium and mass transport kinetic. Differences are analyzed employing the dry substance and polyphenol content in the extract as well as particle size distribution and SEM measurements. The use of ultrasonic intensification leads to higher quasi-equilibrium concentrations in the liquid phase. The content of polyphenols is increased by approx. 15%. Thereby, the solubility of the ingredients, here polyphenols, as well as the accessibility for the solvent are discussed. Ultrasound employment does not compensate the content of water in the solvent. The water leads to higher solubility and swelling of the raw material and hence higher yields. Therefore, ultrasound can assist the extraction for a pre-chosen and optimized solvent. Via particle size measurements a diminution of the mean particle diameter during extraction is detected. After an extraction time of 30 min, the particle diameter is 5% lower using ultrasound compared to conventional extraction. In addition to polyphenol extraction an extraction of further leaf elements is detected via analyzing the content of the dry substance in the extract phase. This content is about 30–35% higher by use of ultrasound. SEM analyzes do not show significant differences for tea leaf particles between the two techniques. Because of the high degree of grinding (mean particle diameter of about 600-700 lm) various botanic elements, such as epidermis stoma and vascular bundle can be identified. During extraction, an increasingly number of leaf structures are appearing. Nevertheless, differences between the two maceration types cannot be quantified. The variability of the leaf particles and hence the leaf elements observed via SEM does not allow any quantification.
[1] www.fao.de, FAO, Food and Agricultural Organisation der Vereinten Nationen, 2012. [2] H.J. Bart et al., Positionspapier Phytoextrakte – Produkte und Prozesse, DECHEMA e.V, 2012. [3] M. Tegtmeier, Pflanzenextraktion: Schlüsseltechnologie zur nachhaltigen Nutzung von Bio-Ressourcen, Chemie Ingenieur Technik 84 (6) (2012) 880– 882. [4] A.E. Sloan, Getting Ahead of the Curve: Phytochemicals, Sloan Trends Inc., 2009. [5] T. De Silva, Industrial utilization of medicinal plants in developing countries, www.fao.org, Vienna, Austria, 2012. [6] www.researchandmarkets.com, access 2012. [7] www.slideshare.net, A study of the European cosmetic industry, Executive summary, Global Insight, Mailand, 2007. [8] S. Kumar, Exploratory analysis of global cosmetic industry: major players, technology and market trends, Technovation 25 (11) (2005) 1263–1272. [9] www.aginskyconsuting.com, Cosmetic market research summary, Aginsky Consulting Group, Portland, US, 2007. [10] M. Kaßing, U. Jenelten, J. Schenk, J. Strube, A new approach for process development for plant-based extraction processes, Chem. Eng. Technol. 33 (3) (2010) 377–387. [11] M. Kaßing, U. Jenelten, J. Schenk, R. Hänsch, J. Strube, Combination of rigorous and statistical modeling for process development of plant-based extractions based on mass balances and botanical aspects, Chem. Eng. Technol. 35 (1) (2012) 109–132. [12] J. Strube et al., Kapitel 5: Process Engineering and Mini-Plant Technology, in: H.J. Bart, S. Pilz (Eds.), Industrial Scale Natural Products, Wiley-VCH, 2010. [13] J.P. Josch, S. Both, J. Strube, Characterization of feed properties for conceptual process design involving complex mixtures such as natural extracts, Food Nutr. Sci. 3 (2012) 836–859. [14] M. Kaßing, Process Development for Plant-based Extract Production, Shaker Verlag, 2012. [15] F. Chemat, Éco-Extraction du Végétal, Dunod, Paris, 2011. [16] M. Obanda, P.O. Owuor, Flavanol composition and caffeine content of green leaf as quality potential indicators of Kenyan black teas, J. Sci. Food Agric. 74 (1997) 209–215. [17] A. Mersmann, M. Kind, J. Stichlmair, Thermische Verfahrenstechnik – Grundlagen und Methoden, Springer Verlag, Berlin Heidelberg New York, 2005. [18] S. Both et al., Model-based equipment-design for plant-based extraction processes –considering botanic and thermodynamic aspects. C. R. Chimie (2013), http://dx.doi.org/10.1016/j.crci.2013.11.004 [19] S. Both et al., Optimizing established processes like sugar extraction from sugar beets – design of experiments versus physicochemical modeling, Chem. Eng. Technol. 36 (2013) 12. [20] E. Blass, Methoden, Zielsuche, Lösungssuche, Lösungsauswahl, Springer, Berlin, 1997. [21] M. Hurmw, M. Jarvelainen, Combined process synthesis and simulation system for feasiilty studies, Comput. Chem. Eng. 19 (1995) 663–668. [22] S. Both et al., Resource efficient process technology for energy plants, Chemie Ingenieur Technik 85 (8) (2013) 1282–1289.
Please cite this article in press as: S. Both et al., Extraction of polyphenols from black tea – Conventional and ultrasound assisted extraction, Ultrason. Sonochem. (2013), http://dx.doi.org/10.1016/j.ultsonch.2013.11.005