Supercritical
Fluid Extraction
of Nutraceuticals
and Bioactive
Compounds
Supercritical
Fluid Extraction
of Nutraceuticals
and Bioactive
Compounds
Edited by
Jose L. Martínez
CRC Press is an imprint of the
Taylor & Francis Group, an informa business Boca Raton London New York
CRC Press is an imprint of Taylor & Francis Group, an Informa business No claim to original U.S. Government works
Printed in the United States of America on acid-free paper 10 9 8 7 6 5 4 3 2 1
International Standard Book Number-13: 978-0-8493-7089-2 (Hardcover)
This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with permission, and sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to publish reliable data and information, but the author and the publisher cannot assume responsibility for the validity of all materials or for the conse-quences of their use.
No part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers.
For permission to photocopy or use material electronically from this work, please access www. copyright.com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC) 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe.
Library of Congress Cataloging-in-Publication Data
Supercritical fluid extraction of nutraceuticals and bioactive compounds / [edited by] Jose L. Martinez.
p. cm.
Includes bibliographical references and index. ISBN 978-0-8493-7089-2 (alk. paper)
1. Supercritical fluid extraction. 2. Functional foods. 3. Bioactive compounds. I. Martinez, José L. (José Luis), 1966-
TP156.E8S835 2007
660.6’3--dc22 2007025441
Visit the Taylor & Francis Web site at http://www.taylorandfrancis.com and the CRC Press Web site at http://www.crcpress.com
Dedication
vii
Contents
Preface...ix Acknowledgments...xi Contributors... xiii Editor...xv Chapter 1 Fundamentals.of.Supercritical.Fluid.Technology...1Selva Pereda, Susana B. Bottini, and Esteban A. Brignole Chapter 2 Supercritical.Extraction.Plants:.Equipment,.Process,.and.Costs...25
Jose L. Martínez and Samuel W. Vance Chapter 3 Supercritical.Fluid.Extraction.of.Specialty.Oils... 51
Feral Temelli, Marleny D. A. Saldaña, Paul H. L. Moquin, and Mei Sun Chapter 4 Extraction.and.Purification.of.Natural.Tocopherols.by. Supercritical.CO2... 103
Tao Fang, Motonobu Goto, Mitsuru Sasaki, and Dalang Yang Chapter 5 Processing.of.Fish.Oils.by.Supercritical.Fluids... 141
Wayne Eltringham and Owen Catchpole Chapter 6 Supercritical.Fluid.Extraction.of.Active.Compounds.from.Algae... 189 Rui L. Mendes Chapter 7 Application.of.Supercritical.Fluids.in.Traditional.Chinese. Medicines.and.Natural.Products... 215 Shufen Li Chapter 8 Extraction.of.Bioactive.Compounds.from.Latin.American.Plants.... 243 M. Angela A. Meireles Chapter 9 Antioxidant.Extraction.by.Supercritical.Fluids... 275
Beatriz Díaz-Reinoso, Andrés Moure, Herminia Domínguez, and Juan Carlos Parajó
Ernesto Reverchon and Iolanda De Marco
Chapter 11 Processing.of.Spices.Using.Supercritical.Fluids... 337 Mamata Mukhopadhyay
Chapter 12 Preparation.and.Processing.of.Micro-.and.Nano-Scale.Materials.
by.Supercritical.Fluid.Technology... 367
Eckhard Weidner and Marcus Petermann
ix
Preface
In.the.last.decade.new.trends.in.the.food.industry.have.emerged,.enhanced.concern. over. the. quality. and. safety. of. food. products,. increased. preference. for. natural. .products,. and. stricter. regulations. related. to. the. residual. levels. of. solvents.. Addi-tionally,.the.nutraceutical.and.functional.food.sector.represents.one.of.the.fastest. growing.areas.in.a.consumer-driven.trend.market..These.trends.have.driven.super-critical.fluid.(SCF).technology.to.be.a.primary.alternative.to.traditional.solvents.for. extraction,.fractionation,.and.isolation.of.active.ingredients..The.aim.of.this.book. is. to. present. the. current. state. of. the. art. in. extracting. and. fractionating. bioactive. ingredients.by.SCFs.
This.book.contains.twelve.chapters.that.primarily.focus.on.implemented.indus-trial.processes.and.trends.of.the.technology..The.content.of.the.chapters.includes.a. review.of.the.major.active.components.in.the.target.material,.including.chemical,. physical,. nutritional,. and. pharmaceutical. properties;. an. analysis. of. the. specific. SCF. process. used;. a. comparison. of. traditional. processing. methods. versus. SCF. technology;.and.a.set.of.conclusions.with.supporting.data.and.insight..A.review.of. the.fundamentals.of.the.technology.and.an.examination.of.SCF.extraction.systems. and.process.economics.are.also.included.
The. contributing. authors. are. international. experts. on. the. topics. covered,. and. I.would.like.to.thank.them.for.their.thoughtful.and.well-written.contributions..This. book.is.addressed.to.food.scientists,.technologists,.and.engineers.as.well.as.other. professionals.interested.in.the.nutraceutical.and.functional.food.sector..Additionally,. I.hope.that.this.book.will.serve.to.stimulate.academia.and.industry.to.search.for.new. process.and.product.developments.as.well.as.their.industrial.implementation.
xi
Acknowledgments
The.authors.of.the.chapters.of Supercritical Fluid Extraction of Nutraceuticals and
Bioactive Compounds wish.to.acknowledge.the.following.funding.agencies.for.their.
support.and.assistance.
Dr.. Feral. Temelli. would. like. to. acknowledge. the. financial. support. from. the. Natural.Sciences.and.Engineering.Research.Council.of.Canada.(NSERC). Dr..Fang.et.al..gratefully.acknowledge.the.21st.COE.program.“Pulsed.Power. Sicence”.and.Wuhan.Kaidi.Fine.Chemical.Industries.Co.,.Ltd.,.for.their.support. Dr..Shufen.Li.would.like.to.thank.Dr..Can.Quan,.Dr..Shaokun.Tang,.Dr..Wenqiang. Guan,.Dr..Yongyue.Sun,.Ms..Luan.Xiao,.and.Ms..Ying.Zhang.for.their.contributions. to.the.research.work.as.well.as.their.assistance.in.the.preparation.of.Chapter.7. Dr..Maria.Angela.Meireles.thanks.CNPq,.CAPES,.and.FAPESP.for.supporting. the.research.done.at.LASEFI.–.DEA/.FEA.–.UNICAMP.
Dr.. Eckhard. Weidner. and. Dr.. Marcus. Petermann. would. like. to. thank. their. coworkers.and.students.from.the.University.Bochum.as.well.as.Prof..Knez.and.his. coworkers.from.the.University.of.Maribor.and.Adalbert-Raps.Research.Center..They. would.also.like.to.thank.Adalbert-Raps.Stiftung,.the.European.Union,.the.Ewald. Doerken.AG,.and.Yara.Industrial.GmbH.for.their.support.
xiii
Contributors
Susana B. Bottini, Ph.D. Planta.Piloto.de.Ingeniería.Química Universidad.Nacional.del.Sur Bahía.Blanca,.Argentina Esteban A. Brignole, Ph.D. Planta.Piloto.de.Ingeniería.Química Universidad.Nacional.del.Sur Bahía.Blanca,.Argentina Owen J. Catchpole, Ph.D. Industrial.Research.Limited Lower.Hutt,.New.Zealand Iolanda De Marco, Ph.D. Dipartimento.di.Ingegneria. Chimica.ed.Alimentare Universita.di.Salerno Salerno,.Italy Beatriz Díaz-Reinoso, M.Sc. Department.of.Chemical.Engineering Facultade.de.Ciencias.de.Ourense Universidade.de.Vigo Ourense,.Spain Herminia Domínguez, Ph.D. Department.of.Chemical.Engineering Facultade.de.Ciencias.de.Ourense Universidade.de.Vigo Ourense,.Spain Wayne Eltringham, Ph.D. Industrial.Research.Limited Lower.Hutt,.New.Zealand Tao Fang, Ph.D. Department.of.Applied.Chemistry.and. Biochemistry Kumamoto.University Kumamoto,.Japan Motonobu Goto, Ph.D. Department.of.Applied.Chemistry.and. Biochemistry Kumamoto.University Kumamoto,.Japan Shufen Li, Ph.D. School.of.Chemical.Engineering.&. Technology Tianjin.University Tianjin,.China Jose L. Martínez, Ph.D. Thar.Technologies,.Inc. Pittsburgh,.Pennsylvania M. Angela A. Meireles, Ph.D. LASEFI-DEAFEA.–.UNICAMP Sao.Paulo,.Brazil Rui L. Mendes, Ph.D. Departamento.de.Energias.Renovaveis INETI Lisboa,.Portugal Paul H.L. Moquin, B.Sc. Department.of.Agricultural,.Food,.and. Nutritional.Science University.of.Alberta Edmonton,.Canada Andrés Moure, Ph.D. Department.of.Chemical.Engineering Facultade.de.Ciencias.de.Ourense Universidade.de.Vigo Ourense,.Spain Mamata Mukhopadhyay, Ph.D. Chemical.Engineering.Department Indian.Institute.of.Technology Bombay,.IndiaUniversidade.de.Vigo Ourense,.Spain Selva Pereda, Ph.D. Planta.Piloto.de.Ingeniería.Química Universidad.Nacional.del.Sur Bahía.Blanca,.Argentina Marcus Petermann, Ph.D. University.Bochum Particle.Technology Bochum,.Germany Ernesto Reverchon, Ph.D. Dipartimento.di.Ingegneria. Chimica.ed.Alimentare Universita.di.Salerno Salerno,.Italy
Marleny D.A. Saldana, Ph.D.
Department.of.Agricultural,.Food,.and. Nutritional.Science University.of.Alberta Edmonton,.Canada Mitsuru Sasaki, Ph.D. Department.of.Applied.Chemistry.and. Biochemistry Kumamoto.University Kumamoto,.Japan University.of.Alberta Edmonton,.Canada Feral Temelli, Ph.D. Department.of.Agricultural,.Food,.and. Nutritional.Science University.of.Alberta Edmonton,.Canada
Samuel W. Vance, P.E.
Thar.Technologies,.Inc. Pittsburgh,.Pennsylvania Eckhard Weidner, Ph.D. University.Bochum Process.Technology Bochum,.Germany Dalang Yang, M.Sc. Wuhan.Kaidi.Fine.Chemical.Industries. Co..Ltd. Wuhan,.Hubei,.China
xv
Editor
Dr. Jose L. Martínez,.a.native.of.León,.Spain,.received.his.B.S..and.Ph.D..degrees.
from.the.University.of.Oviedo.(Spain)..He.is.currently.General.Manager.of.Thar. Technologies,. Inc.,. Process. Division. (Pittsburgh,. USA),. a. company. dedicated. exclusively.to.supercritical.fluid.technology..He.has.nearly.two.decades.of.experi-ence.in.conducting.R&D.and.implementing.industrial.processes.in.supercritical. fluid.technology,.including.applications.in.extraction,.fractionation,.chromatogra-phy,.particle.formation,.coating,.and.impregnation.for.the.food,.nutraceutical,.and. pharmaceutical.industries.
1
Fundamentals
of Supercritical
Fluid Technology
Selva Pereda, Susana B. Bottini, and
Esteban A. Brignole
Contents 1.1 Introduction ...1 1.2 Supercritical Fluids ...2 1.2.1 Physical Properties of Supercritical Fluids ...4 1.3 Phase Equilibrium with Supercritical Fluids ...4 1.3.1 Solid Solubilities ...4 1.3.2 Multiple Fluid Phase Equilibrium ...6 1.4 Phase Equilibrium Engineering of Supercritical Processes ...8 1.4.1 Understanding Phase Behavior ...9 1.5 Conceptual Supercritical Process Design ... 11 1.5.1 Oxychemical Extraction and Dehydration ... 11 1.5.2 Particle Micronization with Supercritical Fluids ... 15 1.5.3 Extraction, Purification, or Fractionation of Natural Products with Supercritical Fluids ... 17 1.5.3.1 Fractionation of Oils... 17 1.5.3.2 Extraction from Vegetable Matrices ... 18 1.5.4 Supercritical Reactions ... 19 References ... 21 . IntroduCtIon Solvents are used in large amounts in the chemical, pharmaceutical, food, and natural-product industries. In the search for environmentally friendly solvents, increasing attention is being paid to supercritical fluids (SCFs) for a wide variety of applica-tions. For instance, supercritical solvents are used in extractions, material processing, micronization, chemical reactions, cleaning, and drying, among other applications. SCFs and near-critical fluids add a new dimension to conventional (liquid) solvents:their density-dependent solvent power. The density of SCFs can be easily tuned to
the process needs, with changes in temperature, pressure, and/or composition. Other important properties of SCFs are their very low surface tensions, low viscosities, and moderately high diffusion coefficients.
The design of processes using supercritical solvents is strongly dependent on the phase equilibrium scenario, which is highly sensitive to changes in operating condi-tions. Therefore, phase equilibrium engineering plays a key role in the synthesis and design of these processes.
. superCrItICal FluIds
The different physical states of a pure substance can be visualized in a three- dimensional pressure–volume–temperature (PVT) diagram, as shown in Figure 1.1. The surfaces represent the different states—solid, liquid, or vapor—that correspond to particular values of pressure and temperature. According to the phase rule, the two-phase (solid–liquid, solid–vapor, and liquid–vapor) regions of a pure substance have only one degree of freedom. Therefore, the equilibrium pressure in each case is a function of temperature. The PT projections of the solid–liquid, solid–vapor, and liquid–vapor equilibrium lines are shown on the left of Figure 1.1. In particular, the vapor–liquid line represents the vapor pressure curve that starts at the triple point (TP) of solid–liquid–vapor coexistence and ends at the critical point (CP). The nature of the CP can be understood following the changes of the fluid properties along the vapor pressure curve. With increasing values of temperature, the density of the liquid phase diminishes and the vapor density increases due to the higher vapor pressure. Eventually, both densities converge at the CP and differentiating the liquid or the vapor state is no longer possible above the critical temperature. When both tempera-ture and pressure are above the critical values (Figure 1.1), the system is considered to be in the supercritical region. Within a region close to the critical conditions, the system properties are highly sensitive to pressure and temperature; this region is considered near-critical. Usually, the SCF solvent is applied at a temperature close to its critical value and at a pres-sure high enough for its density to become greater than the fluid critical density. A Volume Pc Tc Vc Vapor So lid Liquid sv lv sl slv TP sl lv sv Pr es su re Temperature Supercritical Region CP CP FIgure . PVT diagram of a pure substance and its projection on the PT plane.
list of fluids that have been proposed as SCF solvents is shown in Table 1.1. These fluids can be classified as a) low-critical temperature (low-Tc) and b) high-critical temperature (high-Tc) solvents. Some condensable gases, like carbon dioxide (CO2), ethane, and propane, are considered low-Tc solvents, whereas the higher alkanes, methanol, and water can be considered high-Tc solvents. Strong differences in solvent power and selectivity characterize the low-Tc and high-Tc solvents. Francis [1] made a significant contribution on the subject of CO2 solvent proper-ties by studying its behavior with a large number of solutes. Liquid CO2 is miscible with alkanes up to approximately carbon number 10, while the range of miscibil-ity increases for ethane up to 20, and propane up to 35. Therefore, these solvents show selectivity for relatively low-molecular-weight material. Stahl and Quirin [2] have reported the extractability of a wide range of natural products using CO2; they showed that: “1) hydrocarbons and other lipophilic organic compounds of relatively low molecular mass and polarity are easily extractable; 2) the introduction of polar functional groups, hydroxyl or carboxyl groups render the extraction more difficult or impossible; 3) sugars and amino acids cannot be extracted; 4) fractionation effects are possible if there are marked differences in mass, vapor pressure or polarity of the constituents of a mixture.” Regarding the use of high-Tc solvents, such as toluene or water, the extraction is carried out at temperatures from 500 to 700 K, where even a mild pyrolysis of high-molecular-weight material takes place. The solvent power of high-Tc fluids is much higher than that of low-Tc solvents, and high-Tc solvents are proper solvents for high molecular weight materials. However, they have low selectivity and the severe operating conditions, on the other hand, degrade thermally labile materials. A good feature of low-Tc solvents, as compared with conventional liquid solvents, is that they operate at moderate temperature and have low solvent power. There-fore, by carefully choosing the pressure and temperature of operation, selective fractions can be extracted from vegetable matrices, such as essential oils, alkaloids, lipids, or oleoresins. These are the preferred solvents for the pharmaceutical and natural-product industries. A key advantage of low-Tc solvents is that they are easily separated from the extract.
table .
Critical properties of Fluids of Interest in supercritical processes Fluid Critical temperature tc/K Critical pressure pc/bar Critical Volume Vc/cm·mol– CO2 304.12 73.7 94.07 Ethane 305.3 48.7 145.5 Propane 369.8 42.5 200.0 Water 647.1 220.6 55.95 Ammonia 405.4 113.5 72.47 n-Hexane 507.5 30.2 368.0 Methanol 512.6 80.9 118.0
SCF-solute interactions in the liquid phase may originate a second liquid phase (gas salting out effect), improving process selectivity and making it possible, for instance, to separate chemical reaction products in situ [3]. A better understand-ing of supercritical solvent properties will be obtained after considering the phase equilibrium behavior of binary systems that show a different degree of asymmetry in size or intermolecular interactions.
1.2.1 Physical ProPertiesof suPercritical fluids
The physical properties of SCFs are in-between those of a gaseous and liquid states. Typical values of different physical properties for each fluid state are listed in Table 1.2. Density and viscosity of SCFs are lower than those of liquids; however, diffusivi-ties are higher. Thermal conductivities are relatively high in the supercritical state and have very large values near the CP because, in principle, the heat capacity of a fluid tends to infinity at the CP. Interfacial tension is close to zero in the critical region. In general, the physical properties in the critical region enhance mass and heat transfer processes.
. phase equIlIbrIum wIth superCrItICal FluIds
1.3.1 solid solubilities The conditions of phase equilibrium between a SCF (1) and a solid component (2) are formulated on the basis of the isofugacity criterion. If the solid phase is assumed to be a pure component (2), the solubility in the gas phase can be directly obtained as: y E p P s 2= 2 (1.1) supercritical Fluids
physical property gas (tambient) sCF (tc, pc) liquid (tambient)
Density r (kg m–3) 0.6–2 200–500 600–1600 Dynamic viscosity m (mPa.s) 0.01–0.3 0.01–0.03 0.2–3 Kinematic viscosity ha (106 m2s–1) 5–500 0.2–0.1 0.1–5 Thermal conductivity λ (W/mK) 0.01–0.025 Maximumb 0.1–0.2 Diffusion coefficient D (106 m2s–1) 10–40 0.07 0.0002–0.002 Surface tension σ (dyn/cm2) — — 20–40 a Kinematic viscosity defined as η = µ/ρ b Thermal conductivity presents maximum values in the near-critical region, highly dependent on temperature
where E is the enhancement factor over the ideal solubility and ps 2 is the sublima-tion pressure of the solute (2). For a low-volatility, incompressible solid solute, the enhancement factor can be calculated as follows: E P p v RT S sol = − exp ( 2) 2 2 Φ (1.2)
where Φ2 is the fugacity coefficient of the solid solute in the gas phase and v2sol is the
solid molar volume. Φ2 is strongly dependent on the SCF density. Figure 1.2 shows the region of SCF extraction. This region is characterized by a strong variation of fluid density with pressure, at temperatures close to the SCF critical temperature. For a given isotherm, the increase in solubility closely follows the increase in density, as indicated in Figure 1.2. The drastic increase in solubility in the vicinity of the critical region can be of several orders of magnitude and is mainly due to a sharp decrease of the solute fugacity coefficient Φ2 in the fluid phase. This is the classical enhancement effect at the near-critical region. The influence of temperature on the solid solubility is the result of two compet-ing effects: the increase of solid volatility and the decrease of solvent density with temperature rise. Near the critical pressure, the effect of fluid density is predominant. Therefore, a moderate increase in temperature leads to a large decrease in fluid density and a consequent reduction in solute solubility. However, at higher pressures, the increase of solid sublimation pressure with temperature exceeds the density reduction effect, and the solubility increases with temperature. This behavior leads to a region of retrograde behavior of the solid solubility, as illustrated in Figure 1.3. At pressures well above the SCF critical pressure, the isotherms exhibit a maximum in solubility. This maximum is usually observed in the range of 30 to 100 MPa.
Pressure
D
en
sit
y (δ)
Near Critical Region
Solubility ( y2 ) T1 T1 < Tc1 < T2 T2 Pc1
Kurnik and Reid [4] have shown that the maximum is achieved when the partial molar volume of the solute in the fluid phase is equal to its solid molar volume. A quantitative correlation and prediction of the solubility of a pure solid in a SCF is possible if the fugacity coefficient of the solid in the fluid phase is computed using an equation of state. Cubic equations of state, with conventional mixing rules and adjustable binary interaction parameters, have been widely used since the early works of Deiters and Swaid [5] and Kurnik and Reid [4]. However, equations of state that use classical mixing rules, even with energy and size binary interaction parameters, may fail to predict or correlate the solubility of solids with polar or hydrogen-bonding interactions. For instance, Kurnik and Reid [4] found that this approach is not able to model the solubility of stearic acid or n-octanol in CO2. The limitations of cubic equations of state to model the solubility of polar solids can be tackled by using cubic equations of state with local composition mixing rules [6].
When a nonpolar supercritical solvent is used, the separation process does not present specific selectivities; in this case, the addition of a proper cosolvent can enhance solubility and selectivity. Nonpolar cosolvents increase the solubility of solid aromatics several times, whereas polar cosolvents enhance the solubility of solutes that present specific interactions with the cosolvent. For example, Brenecke and Eckert [7] showed a dramatic effect of the cosolvent tributilphosphate on the solubil-ity of hydroquinone in CO2. The cosolvent selection follows the general rules applied for classic solvent selection in solid or liquid-liquid extraction. Brunner [8] studied the effects of cosolvents on the extraction of low-volatility liquids and showed that the use of acetone or methanol, for instance, improves selectivity and solvent power in the extraction of hexadecanol from octadecane.
1.3.2 MultiPle fluid Phase equilibriuM
Equilibrium predictions in systems having two or more fluid phases are more com-plex than those in cases of solid solubilities due to the need to compute fugacities T2 Pressure T1 > T2 > Tc1 y2 Pc1 Supercritical Region FIgure . Typical isotherms of solid solubility in SCF.
in several phases of different compositions. The use of the same equation of state to compute fugacity coefficients in all phases gives the required continuity in the pre-diction of phase equilibrium at the critical region. Cubic equations of state of the van der Waals family have been successfully applied in the correlation and prediction of phase equilibria in mixtures of subcritical and supercritical nonpolar components in the natural gas and petrochemical industries. However, their application to size and energy asymmetric systems, typical of the supercritical extraction of natural subtracts, has found little success. De la Fuente et al. [9] tried to correlate both vapor-liquid equilibrium (VLE) and liquid-liquid equilibrium (LLE) of the system sunflower oil + propane using the Soave [10] equation of state with quadratic mixing rules and binary interaction parameters for both, the attractive energy parameter and the covolume. It was not possible to quantitatively describe both VLE or LLE using only one set of parameters for the attractive energy parameter and the covolume. This indicates the limitations of the van der Waals repulsive term to describe these asym-metric mixtures. The failure of cubic equations of state to model phase equilibria in size asymmetric mixtures can be attributed to the large differences in the pure-component covolumes [11]. Espinosa et al. [12] and Ferreira et al. [13] extensively discussed the application of equations of state to model the supercritical processing of natural products. A group contribution approach is particularly useful when dealing with natural products because a large number of compounds, such as triglycerides, fatty acids, esters, and alcohols, can be represented with a small number of functional groups. Group con-tribution equations of state, such as Modified Huron-Vidal 2 (MHV2) [14, 15] and group contribution equation of state (GC-EOS) [16, 17], are particularly useful to model the complex phase behavior observed in asymmetric mixtures at near-critical conditions. Bottini et al. [18] extended the GC-EOS model to describe both VLE and LLE in mixtures of supercritical gases + vegetable oil mixtures using the same set of parameters. Gros et al. [19] and Ferreira et al. [20] extended this model to represent associating mixtures (GCA-EOS), using a group contribution approach for dealing with self- and cross-associations. The GCA-EOS equation can be derived from a three-term (repulsive, attractive, and associating) Helmholtz residual energy:
A =Arep +Aatt + Aassoc (1.3)
The repulsive (rep) term is given by the Carnahan-Starling equation for hard spheres, the attractive (att) term is a group contribution version of a density- dependent local composition Non-Random Two Liquids (NRTL) model, and the association (assoc) term is a group contribution expression based on Wertheim’s statistical association fluid theory [21]. The hard sphere term performs better than the van der Waals repulsive term when dealing with highly size-asymmetric systems and the other two terms are able to handle strong nonideal specific interactions. The GC-EOS model was compared to MHV2 and PSRK [22] by Espinosa et al. [23]. All three models perform similarly for moderately polar systems of low molecular weight compounds. However, the MHV2 and PSRK models present some limitations when they are applied to very asymmetric systems.
. phase equIlIbrIum engIneerIng oF superCrItICal proCesses
Phase equilibrium engineering is the systematic application of phase equilibrium knowledge to process development. This knowledge comprises data banks, experi- mental data, phenomenological phase behavior, thermodynamic analysis, and math-ematical modeling procedures for phase equilibrium process calculations. Each SCF application has a set of specifications and physical restrictions. In supercritical reactions, for instance, homogeneous phase conditions may be required at the reaction temperature. The solution to this problem is given by the selection of the proper solvent and the determination of a feasible operating pressure range and feed composition to achieve homogeneity in the reaction mixture. On the other hand, a heterogeneous two-phase system may be required to develop supercritical extraction or fractionation processes. Additional phase equilibrium restrictions may include no solid phase pre-cipitation, azeotrope formation, specific solvent solubilities, or saturation conditions. A multicomponent fluid can be a supercritical mixture, a subcooled liquid, a super-heated vapor, or a heterogeneous liquid-liquid, liquid-vapor, or liquid-liquid-vapor mixture. A useful plot to identify each region is a pressure vs. temperature diagram showing the bubble and dew point phase transitions curves, as well as the CP of a given global composition. These lines determine the mixture phase envelope. Different phase scenarios can be selected from this phase envelope (Figure 1.4): a) homogeneous con-ditions for a supercritical reaction, b) homogenous and heterogeneous conditions for a tunable phase split reactor, or c) phase split for a separation process. Certainly, different phase envelopes are obtained during the course of the reaction or separation process. However, the process trajectory should always remain at the required phase scenario. General conditions can also be set from this plot; for instance, above the maximum pressure of the phase envelope there will be a single phase at any temperature. Rigorous simulations of equilibrium stage separations at near-critical conditions are needed for the design and optimization of supercritical processes. However, equilibrium calculations in the near-critical region can present serious convergence
Sometimes We Look for Both b) Tunable Reactors
Sometimes We Look for Phase Split c) Separation Processes
Temperature
Pressure
a) Supercritical Reactions
Heterogenous Region
Bubble Point Curve
Liquid
Vapor
Dew Point Curve
difficulties. In that respect, Michelsen’s [24] phase stability criterion, multiple-phase flash algorithms, and global phase computations are of particular interest for super-critical extraction applications.
Solvent recycle is a major issue in the economic optimization of these processes, because it is the main factor in determining capital and operating costs. Design and synthesis problems have been increasingly solved by formulating mathematical models, which involve continuous and integer variables to represent operating con- ditions and alternative process topologies [25]. With regard to supercritical extrac-tions, Gros et al. [26] have addressed the synthesis of optimum processes for the extraction and dehydration of oxychemicals as a mixed integer nonlinear program-ming problem. Espinosa et al. [23] and Diaz et al. [27] have applied these procedures for the synthesis and optimization of citrus oil deterpenation processes.
1.4.1 understanding Phase behavior
Van Konynenburg and Scott [28] have shown that the fluid phase behavior observed in binary mixtures can be classified into five main types. In type I phase behavior, com-plete liquid miscibility is observed at all temperatures. When partial liquid miscibility occurs at low temperatures, the system is of type II. Type I phase behavior is usually found in systems with components of similar chemical nature and molecular size, like mixtures of hydrocarbons, noble gases, or systems that do not deviate greatly from ideal behavior. Type II is typical of nonideal mixtures of similar size compounds, in which nonideality leads to liquid phase split at subcritical conditions. When the liquid immiscibility persists even at high pressures and temperatures, the systems are of type III. This behavior is characteristic, for example, of mixtures of CO2 with high-molecular-weight alkanes or vegetable oils. When the difference in molecu-lar size becomes significant, in almost ideal systems, liquid-liquid immiscibility is observed near the light-component critical temperature (solvent Tc in supercritical processes). However, complete miscibility is recovered at lower temperatures; this corresponds to type V phase behavior. Type IV, on the other hand, shows discon-tinued liquid-liquid immiscibility (i.e., liquid immiscibility occurs at low and high temperatures but not at intermediate temperatures). Figure 1.5 is a master chart of the different types of binary fluid phase diagrams [29]. The arrows in Figure 1.5 qualitatively indicate the type of fluid phase behavior that can be expected when the system components exhibit greater molecular interactions, size differences, or both.
Figure 1.6 illustrates, in more detail, a Type V phase diagram. The lines in this diagram indicate the boundaries of phase transitions and the critical locus. The three-phase equilibrium line (l1l2g) starts at the lower critical end point (LCEP) and finishes at the upper critical end point (UCEP). This behavior is typical of mixtures of propane with triglycerides, such as sunflower oil or tripalmitin [30]. When the process operating temperatures are above the critical temperature of the solvent, pressures should be higher than the critical pressure of the mixture in order to ensure complete miscibility (i.e., the pressure should be above the l1 = l2 line).
In the search for an adequate supercritical solvent to achieve homogenous or heterogeneous conditions, two different approaches can be followed: 1) to compare the phase behavior of a given substrate with different solvents or 2) to follow the
change in the phase behavior of a given solvent with different families of chemi-cal compounds. In the more general case, when the components of the mixture are of different chemical nature, the second approach should be followed to take into account any possible change in the phase behavior during process evolution.
The liquid-liquid immiscibility of type V phase behavior appears in many binary mixtures between supercritical solvents and organic substrates beyond a certain carbon number. Figure 1.7 shows the regions of liquid-liquid immiscibility for binary mixtures of supercritical solvents (ethane and propane) with hydrocarbons of different chain length [31]. Peters [31] also presented similar data on the liquid-liquid T P lg(2) lg(1) l1 = l2 l1 = g l1l2g l2 = g UCEP LCEP FIgure . Type V phase behavior according to Van Konynenburg and Scott classification. CL CL CL CL CH CH CH CH T T T P P P Type II Type IV Type I T Type V Type III Interaction Molecular Interaction Molecular Interaction + Size Size
Size Three Phase Region (LLV)
P T Molecular Interaction + Size FIgure . Modifications of binary phase behavior with size and energy asymmetries. CL and CH are the critical points of the light and heavy compounds, respectively.
immiscibility domains of the systems ethane + alcohols, ethane + aromatic hydro-carbons, and ethane + alkanes. It becomes clear from these data that ethane is not an adequate supercritical solvent for normal alcohols because it presents liquid-liquid immiscibility even with methanol. However, ethane seems to be a better solvent for aromatic hydrocarbons or paraffins because the liquid-liquid immiscibility appears at carbon numbers greater than 15 or 18, respectively.
CO2 has been the most studied solvent for supercritical processes. However, it exhibits strong liquid-liquid and gas-liquid immiscibility for hydrocarbons with carbon numbers greater than 13. In addition, CO2 presents a rather low critical temperature to be used as a solvent for reactions carried out at moderate or high temperatures. Figure 1.8 shows data on the type of phase transition for the families of CO2 + alkanes compiled by Peters [32], who also showed the behavior of CO2 + alkanol systems. Unfortunately, the type of data shown in Figures 1.7 and 1.8 is only known for a limited number of families of organic compounds with some supercritical solvents. Therefore, reliable thermodynamic models are needed to explore the possible phase scenarios found in mixtures between process components and supercritical solvents.
The phase equilibrium engineering approach will be illustrated with several examples, where thermodynamic and modeling tools are applied for supercritical process development. The examples to be covered are alcohol extraction and dehy-dration, gas antisolvent crystallization, purification of vegetable oils, supercritical fractionation, extraction with near critical fluids, and supercritical reactions.
. ConCeptual superCrItICal proCess desIgn
1.5.1 oxycheMical extractionand dehydration
The supercritical extraction of organic oxygenated compounds from aqueous solu-tions is of great interest in biotechnological processes. Oxygenated compounds and 280 300 320 340 360 380 400 10 15 20 25 30 35 40 45 50 55 60 65 70 UCEP UCEP LCEP LCEP Ethane Propane SL1L2V SL1L2V
Number of Carbon Atoms
Temperature, (K)
FIgure .
Phase transitions for the binaries of ethane and propane with paraffins of dif- ferent chain length. UCEP and LCEP points are upper and lower critical end points, respec-tively. SL1L2V stands for solid–liquid–liquid–vapor equilibria.
water have strong hydrogen bonding interactions that complicate their separation with conventional solvents. Moreover, an oxygenated compound dissolved in a non-polar near-critical solvent will have a rather high activity coefficient (γoxySCF), leading
to a low value of the distribution coefficient: moxy oxy H O oxy SCF = γ γ 2 (1.4)
This is even more pronounced in the case of alcohols or acids that exhibit self-association. A strategy to overcome this problem may be based on the Koenen and Gaube [33] diagram that classifies binary mixtures in an excess Gibbs function (GE) versus excess enthalpy (HE) diagram (Figure 1.9). We can derive the effect of temperature on the activity coefficients directly from this diagram. The aqueous solutions of organic oxygenated compounds are located on the second quadrant of the diagram with negative HE values, whereas the supercritical solutions that correspond to positive HE values are located on the first quadrant. In both cases, there are positive deviations to nonideality (positive GE ). From this diagram, we can see that the activ-ity coefficients in the aqueous phase increase with temperature; however, the reverse occurs with the activity coefficients in the SCF phase. Therefore, extracting at high temperatures leads to more attractive values of the distribution coefficients. This behavior is found in the extraction of isopropanol or ethanol from aqueous solutions using CO2, ethane, or propane as near-critical solvents. However, we should consider another fact to make a proper solvent selection: At optimum extraction temperatures (around 380 to 400K), the solvent power of CO2 or ethane is drastically reduced due to fluid density decrease at temperatures well above the critical temperature of both fluids (around 304 K). To recover the solvent power, relatively high pressures should be used for the extraction process. This makes propane a better candidate as 200 220 240 260 280 300 5 7 9 11 13 15 17 19 21 23 25 UCEP LCEP
Number of Carbon Atoms
Temperature, (K)
FIgure . Phase transitions for the binaries of CO2 with paraffins of different chain length. UCEP and LCEP points are upper and lower critical end points, respectively. Dashed line (open squares): SL1L2V equilibria.
an extraction solvent because its critical temperature is close to 370 K and it has a lower critical pressure than CO2 or ethane. Horizoe et al. [34] and Brignole et al. [35] verified the potential of propane as an extracting supercritical solvent. Dehydration by near-critical solvents finds important applications, for example, in the extraction of solutes from aqueous solutions and in the drying of solid particles after micronization. We will consider first the dehydration of extracted solutes. In low-pressure separations, entrainment agents like cyclohexane or solvents like ethylene glycol have been used to separate water by azeotropic or extractive distillations. In con-nection with supercritical processes, it is of interest to study the equilibrium between water and a near-critical fluid as a function of temperature and pressure. In the case of CO2, the data of Wiebbe [36] and Coan et al. [37] show the solubility of water in CO2 as a function of pressure at subcritical and supercritical temperatures. These data indicate that water follows the classical supercritical effect: the concentration of water in the CO2 phase increases once the supercritical pressure is exceeded (Figure 1.10). At the CO2 saturation pressure, at subcritical conditions, we would have a three-phase VLL equilibrium condition, where the concentration of water in the con-densed CO2 phase exceeds the concentration of water in the vapor phase. Hence, in a CO2–water separation process, the relative volatility of water with respect to CO2 is lower than one. This behavior has important consequences for the separation of water from CO2 extracts. Water, as expected, is less volatile than CO2; therefore, the extract cannot be obtained free from water in the solvent recovery operation. When the same phase equilibria analysis is made for water and light alkanes, such as ethane and propane, a different picture is obtained. The data of Kobayashi and Katz [38] for the solubility of water in propane are plotted against pressure at different temperatures (Figure 1.11). For near-critical propane, the solubility of water decreases when the critical pressure is exceeded (see Table 1.1 for the critical proper-ties of propane). This phenomenon can be called a nonclassical supercritical effect.
< 0 GE HE γ > 1 Regular Solution SE= 0 > 0 δγ δT γ < 1 > 0 δγ δT γ < 1 < 0 δγ δT γ > 1 < 0 δγ δT
FIgure . Value and temperature derivative of activity coefficients, according to the
A very attractive property can be derived from this effect. When working at subcriti-cal temperatures, at the propane saturation pressure, we again have VLL equilibria. In this case, the composition of water in the vapor phase is greater than that in the condensed propane phase, leading to a water-propane relative volatility greater than one. This makes it possible to obtain dehydrated organic oxygenated products during the process of solvent recovery from the extract [39]. 0 0.002 0.004 0.006 0.008 0.01 0 200 400 600 800 Pressure, (bar)
Grams of Water Per Liter of Expanded Gas at s.t.p.
298 K 323 K 348 K
FIgure .0 Composition of the CO2 -rich phase as a function of pressure and tempera-ture. Experimental data from Wiebbe [34]. 0.1 1 10 100 0 20 40 60 80 100 P(bar)
Mole Fraction % of Water
327.6 K 377.6 K
369.6 K
FIgure . Experimental water composition in liquid and vapor propane. Data from
On the basis of the phase equilibrium engineering concepts presented above, a process for the production of bioethanol or for the dehydration of isopropanol with a near-critical solvent (propane) can be developed. The key features of these processes are: a) High temperatures and pressures of extraction favor the solubility of alcohol in propane. b) Liquid-liquid equilibrium at low temperatures is beneficial for reducing the water content in the extract. c) The alcohol product is obtained dehydrated because the relative volatility of water with respect to propane is greater than one over a certain concen-tration range of ethanol in the extract mixture.
All these properties were first predicted by group-contribution thermodynamic modeling and thereafter verified by experimental and pilot plant information. 1.5.2 Particle Micronizationwith suPercritical fluids
Supercritical micronization processes are based on creating a high degree of solution supersaturation that leads to the formation of a great number of nucleation sites and very small crystals. These processes have found many applications in the last decade [40, 41], mainly in the micronization of pharmaceutical solid compounds. Usually, several components may participate in the process: the solute to be crystallized, the solvent, a supercritical fluid, and a cosolvent. The phase equilibrium between these components plays a key role in the selection of the proper technology for the micronization processes. A better understanding of process selection can be made on the basis of the binaries behavior. First, we shall consider the solute + super-critical fluid binary. If the solute solubility under supercritical conditions is high, then only these components participate in the process and micronization is obtained directly by a drastic reduction in the solute solubility by the rapid expansion of the supercritical solution (RESS process) through a nozzle or other convenient device. The main limitation of this RESS process is that it can only be applied to solutes with high solubilities in the supercritical fluid. The low solvent power of supercritical CO2 for polar or medium- to high-molecular-weight material makes this approach uneconomical for these mixtures. When the solute cannot be dissolved in significant amounts in the supercritical fluid, we can look for a good liquid solvent for both the solute and the supercritical gas. In this case, a concentrated solution of the solute in the solvent is prepared and a high degree of supersaturation is obtained by dissolving the supercritical fluid in the liquid phase at high pressure. This technology is called the gas antisolvent (GAS) process and it can be carried out in a batch or semicontinuous process. These processes can be applied to a variety of solutes, but in this case, the ternary phase equilibria should also be evaluated to assure a high degree of supersaturation at the operating pressure and temperature. In the semicontinuous process, both the solution and the supercritical fluid enter together in the precipitation vessel through
a mixing device. Very good precipitation conditions are achieved if the operating conditions are above the CP of the solvent + supercritical fluid mixture. Under these conditions, both feeds are completely miscible and no interfacial resistance is offered to mass transfer [42]. Another possible phase scenario appears when the solid solute is not soluble in the SCF, but the solubility of the SCF in the melted solid is high at elevated pres-sures. Therefore, if the solution is expanded to atmospheric pressure, a large cooling effect occurs that gives rise to the precipitation of micronized solute particles. A different situation arises with solutes that are only soluble in water, such as some organic salts and proteins [41]. Typical nonpolar supercritical fluids like CO2 and ethane are not soluble in aqueous solutions, even at high pressures. Therefore, no antisolvent effect can be obtained in a typical GAS process [43]. In this case, a cosolvent that shows complete miscibility with both the SCF and water can be intro-duced. For example, ethanol was used as a proper cosolvent for the precipitation of an organic salt from aqueous solution [43]. In this application, the aqueous solution is fed as a spray or mist into a precipitation vessel already filled up with a mixture of ethanol + CO2 at the required composition. To obtain a feasible process, the operat-ing conditions of the precipitation chamber should lie inside the homogeneous region of the triangular phase diagram for water + CO2 + ethanol at a given pressure and temperature, as shown in Figure 1.12. In this way, the fine water droplets become quickly supersaturated by the ethanol + CO2 dissolution in the drops and the simulta-neous fast evaporation of water. As a result of this process, highly micronized dried salt particles are obtained [43]. All these examples illustrate that a phase equilibrium engineering analysis is a prerequisite for proper technology selection and successful adequate choice of micronization operating conditions. CO2 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 Water + Lobenzarit 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Feasible Region A B
FIgure . Feasible operating region for Lobenzarit precipitation using supercritical
1.5.3 extraction, Purification, or fractionationof
natural Productswith suPercritical fluids
... Fractionation of oils In the processing of vegetable oils, it is possible to take advantage of the low solu-bility of triglycerides in CO2. For instance, both palm oil and sunflower oil give liquid-liquid or liquid-SCF immiscibility with CO2, even at high pressures, typical of type III systems. In these cases, either supercritical or liquid CO2 can be used as a solvent to remove undesirable components from the oil—for instance, removal of oleic acid from olive oil [44]. Likewise, liquid or near-critical CO2 can be applied to recover valuable components like tocopherols or squalene from fish oil [45]. When dealing with these separation processes, it is possible to find optimum extraction operating conditions that minimize the solvent-feed ratio and, at the same time, keep the coextraction of oil at a low value. Other solvents that have regions of liquid-liquid immiscibility with fatty oils, such as ethane and propane, may be used as alternative solvents. SCF solvents can also be used as fractionating agents. This is of interest in the separation of low-volatile substances of close relative volatility. For instance, CO2 and ethane have been proposed as dense gas extractants to remove the terpene frac-tion from citrus essential oils[27] and also for the fractionation of highly unsaturated fish oil methyl esters to obtain rich eicosapentaenoic acid and docosahexaenoic acid fractions [46]. The binary systems between CO2 and these families of compounds are generally of type II, so complete miscibility for all compositions is obtained above the maximum pressure of the vapor-liquid critical locus. A single dense-gas fractionation column scheme is shown in Figure 1.13. The mixture to be fractionated is fed at an intermediate point in the column. A dense gas N=40 N=40 N=40 Feed Separator CO2 Heat Exchanger Extractor Fresh CO2 Raffinate N=40 Extract FIgure . Dense-gas fractionation scheme process.
and CO2 is recycled to the bottom of the column. A compressor or condenser-pump cycle can be selected for this purpose. These types of separation processes follow the principles of a stripping operation. One of the main differences with ordinary gas stripping is that the dense gas is very soluble in the feed. Therefore, the liquid phase flow rate in the column is much larger than the feed flow rate. On the other hand, the low volatility of the substrates being fractionated leads to a relatively high gas-feed stripping ratio. Both effects contribute to give a fairly constant molar overflow for both phases in a simple counter-current column. The design of these separation processes is highly dependent on the relative volatility between the key components of the oils in each separation stage. It can be shown that a simple countercurrent separation is limited by the recovery of each key component in the bottom and top products. In this case, the limiting recoveries of the key components (φ1, φ2) in the top and bottom products are determined by the relative volatility (α12) between both components under process conditions: α12=φ1/ (1−φ2) (1.5)
In most simple countercurrent extraction columns, this constraint limits the recovery and purity of the products in the separation of components of close relative volatility. Therefore, the use of recycle (reflux) of the top product is required: 1) to increase recovery and purity and 2) to assure that the trajectory of the separation process lies inside the two-phase region. Thus, the column and separator operating conditions (pressure, temperature, and compositions) should always be checked in order to verify a heterogeneous operation.
... extraction from Vegetable matrices
The extraction of lipids and oils from vegetable matrices has been extensively covered in the monograph edited by King and List [47]. In the extraction of fatty oils from grounded seeds, it is advantageous to select a solvent that presents complete miscibility with the oil. CO2 is a cheap, nontoxic solvent; however, the oil solubility in this SCF is very low even at pressures of the order of 30 MPa (type III binary). On the other hand, liquid propane is completely miscible with vegetable oils below the LCEP of this binary. Propane has type II or type IV global phase diagrams with vegetable oils. The main drawback of using propane as a solvent for the extraction of oils from grounded seeds is that it is flammable. Recently, Hegel et al. [48] studied the use of propane + CO2 solvent mixtures for oil extraction, looking for efficient and safe solvent mixtures. Peter [45] has studied these types of mixtures to improve selectivities in the separation of lecithin from vegetable oils. In the work of Hegel et al. [48], the selected phase scenario was to operate in a region of complete liquid miscibility of the oil + solvent mixture, with a nonflammable vapor phase. The selec-tion of operating conditions was based on experimental data on the LLV region at constant temperature, for the system sunflower oil + propane + CO2. At constant
temperature, for a three-component system, the LLV equilibrium is only a function of pressure. Therefore, a binodal curve can be drawn on a triangular diagram, with tie lines linking the two liquid phase compositions at specified pressures (see trian-gular diagram on Figure 1.14). The binodal curve gives the boundary of the LLV region. The diagram also shows the minimum pressure for which LL immiscibility arises at a given temperature. At lower pressures (i.e., lower CO2 composition), the solvent has complete miscibility with the oil. However, there is also a minimum operating pressure to avoid vapor phase flammability because, at pressures lower than this, the propane content of the vapor phase is too high. The feasible operating region can be easily determined with the help of Figure 1.14.
1.5.4 suPercritical reactions
In general, gas-liquid catalyzed reactions are diffusion controlled. The use of an adequate supercritical SCF can bring the reactive mixture into homogeneous
0 10 20 30 40 50 60 70 80 0 0.2 0.4 0.6 0.8 1.0 0 0.2 0.4 0.6 0.8 1.0 CO2 Weight Fraction CO2 Weight Fraction Pr es su re , ( ba r) 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Propane Weight Fraction
Minimum CO2 Content in the Gaseous Phase Liquid-Liquid-Vapor Region Oil CO2 + Propane at 308 K CO2 + Oil + Propane at 308 K Maximum Operating Pressure Minimum Operating Pressure
Safe Operating Region
FIgure . Safe operating extraction region at 308 K. Experimental data from Hegel
studied experimentally the hydrogenation of heavy substrates such as vegetable oils and fatty esters under supercritical conditions.
The use of batch reactors is a common practice in bench scale experimental studies on supercritical reactions. However, the control of homogeneous conditions in these reactors is quite difficult. Baiker and coworkers [50] recommend the use of windows in the reaction vessels in order to control the phase conditions. Even though it is possible to have an independent control of process variables in continuous reac-tors, the selection of pressure, temperature, and composition should be carefully done to obtain the desired homogeneous state. Knowledge of the phase behavior of a reaction process can help to understand the results of experimental studies and to plan and design experimental runs.
The solvent to be used in a supercritical reaction should be nonreactive under process conditions. The critical temperature of the solvent should be lower than the reaction temperature to assure complete miscibility of all gaseous reactants in the supercritical solvent. However, the critical temperature should not be far from the reaction temperature to maintain the favorable properties of the near-critical state. To show the importance of making an adequate phase equilibrium engineering analysis, we select a supercritical reaction carried out by Chouchi et al. [51] as an example. Chouchi et al. have studied the hydrogenation of α-pinene under supercritical CO2 in a batch reactor operating at 323 K and 14 MPa with a Pd/C catalyst. The authors showed that the reaction rate and conversion are low when the reactor oper-ates under homogenous conditions. On the contrary, better conversions were achieved when the CO2 pressure was reduced, although the system became heterogeneous. A phase equilibrium engineering analysis of the reactor operating conditions can give an explanation to these seemingly contradictory results. The batch reactor was first fed with the catalyst, together with a known amount of α-pinene. Then, the system was pressurized with CO2 up to the desired pressure (8, 9, 10, or 12 MPa), and, finally, H2 was fed until a total pressure of 14 MPa was reached. The actual molar composition of the reacting mixture was unknown. This composition may be obtained by using an equation of state suitable for density predictions under the reaction conditions. One possibility is to use the MHV2 [15, 48] equation of state. The computation of the actual mixture compositions requires an iterative procedure for estimating the system compressibility factor, the amounts of each component charged into the cell, and the evolution of the reactor composition with conversion. This analysis indicates that, at the higher CO2 partial pressure, an important reduction in hydrogen concentration occurs, which is likely the reason for the observed decrease in the reaction rates. Phase equilibrium engineering analysis of supercritical processes is of the utmost importance in developing new technologies that replace conventional solvents by high-pressure gases to obtain environmentally friendly chemical processes. Several examples of process development clearly demonstrate that a good understanding of phase behavior and application of rigorous modeling tools are essential to process syntheses in which the fluid properties are extremely dependent on pressure, tem-perature, and composition.
reFerenCes
1. Francis, A.W., Ternary systems of liquid carbon dioxide, J. Phys. Chem, 58, 1099, 1954. 2. Stahl, E. and Quirin, K.W., Dense gas extraction on a laboratory scale: A survey of
recent results, Fluid Phase Equilibria, 8, 93–105, 1983.
3. Eckert, C.A. and Chandler, K., Tuning fluid solvents for chemical reaction, J. Supercrit. Fluids, 13, 187–195, 1998.
4. Kurnik, R.T. and Reid, R.C., Solubility extreme in solid-fluid equilibria, AIChE J., 27, 861–863, 1981.
5. Deiters, U.K. and Swaid, I., Calculation of fluid-fluid and solid-fluid phase equilibria in binary mixtures at high pressures, Ber. Bunsenges. Phys. Chem., 88, 791–796, 1984. 6. Vidal, J., Phase equilibria and density calculations for mixture in the critical range with
simple equation of states, Ber. Bunsenges. Phys. Chem., 88, 784–791, 1984.
7. Brennecke, J. and Eckert, C., Phase equilibria for supercritical fluid process design, AIChE J., 35, 1409–1427, 1989.
8. Brunner, G., Selectivity of supercritical compounds and entrainers with respect to model substances, Fluid Phase Equilibria, 10, 289–298, 1983.
9. de la Fuente, J.C.B., Mabe, G.D., Brignole, E.A. and Bottini, S.B., Phase equilibria in binary mixtures of ethane and propane with sunflower oil, Fluid Phase Equilibria, 101, 247–257, 1994.
10. Soave, G., Equilibrium constants from a modified Redlich-Kwong equation of state, Chem. Eng. Sci., 27, 1197–1203, 1972.
11. Heidemann, R.A. and Kokal, S.L., Combined excess free energy models and equations of state, Fluid Phase Equilibria, 56, 17–37, 1990.
12. Espinosa, S., Fornari, T., Bottini, S. and Brignole, E., Phase equilibria in mixtures of fatty oils and derivatives with near critical fluids using the GC-EOS model, J. Supercrit. Fluids, 23, 91–102, 2002.
13. Ferreira, O., Modelling of association effects by group contribution: Application to natural products, Ph.D. Thesis, Univ. de Porto, Portugal, 2003.
14. Michelsen, M.L., A modified Huron-Vidal mixing rule for cubic equations of state, Fluid Phase Equilibria, 60, 213–219, 1990.
15. Dahl, S. and Michelsen, M.L., High-pressure vapor-liquid equilibrium with a UNIFAC-based equation of state, AIChE J., 36, 1829–1836, 1990.
16. Skjold-Jørgensen, S., Gas solubility calculations II. Application of a new group- contribution equation of state, Fluid Phase Equilibria, 16, 317–351, 1984.
17. Skjold-Jørgensen, S., Group contribution equation of state (GC-EOS): A predictive method for phase equilibrium computations over wide ranges of temperature and pressures up to 30 MPa, Ind. Eng. Chem. Res., 27, 110–118, 1988.
18. Bottini, S.B., Fornari, T. and Brignole, E., Phase equilibrium modeling of triglycerides with near critical solvents, Fluid Phase Equilibria, 158–160, 211–218, 1999.
19. Gros, H.P., Bottini, S.B. and Brignole, E., A group contribution equation of state for associating mixtures, Fluid Phase Equilibria, 116, 537–544, 1996.
20. Ferreira, O., Brignole, E.A. and Macedo, E.A., Modeling of phase equilibria for asso-ciating mixtures using an equation of state, J. Chem. Thermodynamics, 36, 1105–1117, 2004.
21. Chapman, W.G., Gubbins, K.E., Jackson, G. and Radosz, M., New reference equation of state for associating liquids, Ind. Eng. Chem. Res., 29, 1709–1721, 1990.
22. Holderbaum, T. and Gmehling, J., PSRK: A Group Contribution Equation of State Based on UNIFAC, Fluid Phase Equilibria, 70, 251–270, 1991.
23. Espinosa, S., Foco, G., Bermudez, A. and Fornari, T., Revision and extension of the group contribution equation of state to new solvent groups and higher molecular weight alkanes, Fluid Phase Equilibria, 172, 129–143, 2000.
flowsheets, Comput. & Chem. Eng., 21, S421–S426, 1997.
26. Gros, H.P., Díaz, S. and Brignole, E.A., Near-critical separation of aqueous azeotropic mixtures: Process synthesis and optimization, J. Supercrit. Fluids, 12, 69–84, 1998. 27. Diaz, S., Espinosa, S. and Brignole, E.A., Citrus peel oil deterpenation with supercritical
fluids: Optimal process and solvent cycle design, J. Supercrit. Fluids, 35, 49–61, 2005.
28. van Konynenburg, P.H. and Scott, R.L., Critical lines and phase equilibria in binary van der Waals mixtures, Phil. Trans., 298, 495–540, 1980.
29. Lucks, K.D., The occurrence and measurement of multiphase equilibria behavior, Fluid Phase Equilibria, 29, 209–224, 1986.
30. Coorens, H.G.A., Peters, C.J. and De Swaan Arons, J., Phase equilibria in binary mixtures of propane and tripalmitin, Fluid Phase Equilibria, 40, 135–151, 1988. 31. Peters, C.J., Supercritical fluids: Fundamentals for application. Multiphase equilibria
in near-critical solvents, Kluwer Academic Publisher. Editors: Kiran, E., and Levelt Sengers, M.H., 1994.
32. Peters, C.J. and Gauter, K., Occurrence of holes in ternary fluid multiphase systems of near-critical carbon dioxide and certain solutes, Chem. Rev., 99, 419–431, 1999. 33.
Koenen, H-E. and Gaube, J., Temperature dependence of excess thermodynamic proper-ties of binary mixtures of organic compounds, Ber. Bunsenges. Phys. Chem., 86, 31–36, 1982.
34. Horizoe, H., Tanimoto, T., Yamamoto, I. and Kano, Y., Phase equilibrium study for the separation of ethanol-water solution using subcritical and supercritical hydrocarbon solvent extraction, Fluid Phase Equilibria, 84, 297–320, 1993.
35. Brignole, E.A., Andersen, P.M. and Fredenslund, A., Supercritical fluid extraction of alcohols from water, Ind. Eng. Chem. Res., 26, 254–261, 1987.
36. Wiebe, R., The binary system carbon dioxide-water under pressure, Chem. Rev., 29, 475–481, 1941.
37. Coan, C.R. and King, A.D., Jr., Solubility of water in compressed carbon dioxide, nitrous oxide, and ethane., J. Am. Chem. Soc., 93, 1857–1862, 1971.
38. Kobayashi, R. and Katz, D., Vapor-liquid equilibria for binary hydrocarbon-water systems, Ind. and Eng. Chem., 45, 440–446, 1953.
39. Zabaloy, M., Mabe, G., Bottini, S.B. and Brignole, E.A., The application of high water-volatilities over some liquefied near-critical solvents as a means of dehydrating oxychemicals, Fluid Phase Equilibria, 5, 186–191, 1992.
40. Reverchon, E. and Adami, R., Nanomaterials and supercritical fluids, J. Supercrit. Fluids, 37, 1–22, 2005.
41. Martin, A., Precipitation processes with supercritical carbon dioxide: mathematical modeling and experimental validation, Ph.D. Thesis, Universidad de Valladolid, Spain, 2005.
42. Martin, A. and Cocero, M.J., Numerical modeling of jet hydrodynamics, mass transfer, and crystallization kinetics in the SAS process, J. Supercrit. Fluids, 32, 203–219, 2004.
43. Amaro-González, D., Mabe, G., Zabaloy, M. and Brignole, E.A., Gas antisolvent crystallization of organic salts from aqueous solutions, J. Supercrit. Fluids, 17, 249–258, 2000.
44. Simoes, P.C. and Brunner, G., Multicomponent phase equilibria of an extra-virgin olive oil in supercritical carbon dioxide, J. Supercrit. Fluids, 9, 75–81, 1996.
45. Peter, S., Supercritical Fluid Technology in Oil and Lipid Chemistry. Chapter VI: Supercritical fractionation of lipids, Editors: King, J.W. and List, G.R., AOCS Press, Illinois, 65–100, 1996.
46. Espinosa, S., Díaz, S. and Brignole, E.A., Thermodynamic modeling and process optimization of supercritical fluid fractionation of fish oil fatty acid ethyl esters. Ind. Eng. Chem. Res., 41, 1516–1527, 2002.
47. King, J.W. and List, G.R., Supercritical fluid technology in oil and lipid chemistry, Editors: King, J.W. and List, G.R., AOCS Press, Illinois, 1996.
48. Hegel, P.E., Mabe, G.D.B., Pereda, S., Zabaloy, M.S. and Brignole, E.A., Phase equilibria of near critical CO2 + propane mixtures with fixed oils in the LV, LL, and LLV region, J. Supercrit. Fluids, 37, 316–322, 2006.
49. Härröd, M., van den Hark, S., Holmqvist, A. and Moller, P., Hydrogenation at supercritical single-phase conditions, ISSAF - 4th International Symposium On High Pressure Process Technology And Chemical Engineering, Venice, Italy, 2002. 50. Baiker, A., Supercritical fluids in heterogeneous catalysis, Chem. Rev., 99, 453–473,
1999.
51. Chouchi, D., Gourgouillon, D., Courel, M., Vital, J. and Nunes da Ponte, M., The influ-ence of phase behavior on reactions at supercritical conditions: The hydrogenation of alfa-pinene, Ind. Eng. Chem. Res., 40, 2551–2554, 2001.
25
2
Supercritical
Extraction Plants
Equipment, Process,
and Costs
Jose L. Martínez and Samuel W. Vance
Contents 2.1 Introduction ...25 2.2 Supercritical Fluid Extraction: Process Description ...26 2.2.1 Supercritical Fluid Extraction of Compounds from a Solid Matrix ...28 2.2.1.1 Processing Parameters in the Supercritical Extraction of Solids ...30 2.2.2 Supercritical Fluid Extraction of Compounds from a Liquid Feed ... 31 2.3 Supercritical Fluid Processing Plants: Equipment Design ...34 2.3.1 Overview ...34 2.3.2 Vessels ... 35 2.3.3 Pumps and Compressors ... 37 2.3.4 Heat Exchangers ... 38 2.3.5 Piping and Valves... 39 2.3.6 Control Systems ... 41 2.4 Industrial Process Implementation ... 42 2.5 Conclusions ...48 References ...48 2.1 IntroduCtIonIn the last decade, supercritical fluid technology has reemerged, mainly due to a dramatic rise in the research and development activities focused on innovative approaches as well as new trends in the pharmaceutical, food, and chemical sectors. In the food industry, these new trends include an increased preference for natural products over synthetic ones and regulations related to nutritional and toxicity levels of the active ingredients. On the other hand, consumers are taking a more proactive role in maintaining their health, which has driven a new generation of products on the market addressing disease prevention. These trends have made supercritical fluid technology a primary alternative to traditional solvent extraction for the extraction and fractionation of active ingredients.
