Journal of Supercritical Fluids The

Published by Elsevier
Print ISSN: 0896-8446
Bacterial endotoxins have strong affinity for metallic biomaterials because of surface energy effects. Conventional depyrogenation methods may not eradicate endotoxins and may compromise biological properties and functionality of metallic instruments and implants. We evaluated the solubilization and removal of E. coli endotoxin from smooth and porous titanium (Ti) surfaces and stainless steel lumens using compressed CO(2)-based mixtures having water and/or surfactant Ls-54. The CO(2)/water/Ls-54 ternary mixture in the liquid CO(2) region (25 °C and 27.6 MPa) with strong mixing removed endotoxin below detection levels. This suggests that the ternary mixture penetrates and dissolves endotoxins from all the tested substrates. The successful removal of endotoxins from metallic biomaterials with compressed CO(2) is a promising cleaning technology for biomaterials and reusable medical devices.
This study reports the effect of exposure to liquid carbon dioxide on the mechanical properties of selected medical polymers. The tensile strengths and moduli of fourteen polymers are reported. Materials were exposed to liquid CO(2), or CO(2) + trace amounts of aqueous H(2)O(2), at 6.5 MPa and ambient temperature. Carbon dioxide uptake, swelling, and distortion were observed for the more amorphous polymers while polymers with higher crystallinity showed little effect from CO(2) exposure. Changes in tensile strength were not statistically significant for most plastics, and most indicated good tolerance to liquid CO(2). These results are relevant to evaluating the potential of liquid CO(2)-based sterilization technology.
The molecular diffusion coefficients of 2-nitroanisole, 1,2-dichlorobenzene and tert-butylbenzene in supercritical carbon dioxide and carbon dioxide containing modifiers were determined by using the Taylor–Aris dispersion technique. Experimental values are reported for temperatures ranging from 313 to 333 K and pressures between 15.0 and 35.0 MPa. The influence of pressure, temperature, density and viscosity on the binary diffusion coefficients was examined. The addition of low proportions of methanol and n-hexane as modifiers had an important effect on the diffusion coefficient of the three solutes.
High-pressure reactive extraction of Eucalyptus globulus wood was undertaken. Extractions were carried out with CO2 + 1,4-dioxane mixtures at 170 bar pressure at temperatures of 160–180 °C. The influence of temperature, extraction time, composition of the fluid mixture, and flow rate on the extraction were investigated. In the range of temperatures studied, only a small increase in lignin and hemicellulose extraction is observed when temperature is increased. At 180 °C, the degradative extraction of cellulose starts, but at lower temperatures, cellulose losses are small. The selectivity of the extraction depends upon the composition of the extracting mixture. CO2-rich fluid mixtures extract hemicellulose preferentially and total hemicellulose removal can be achieved. Maximum delignification was obtained by extraction with pure dioxane with 75% of the lignin initially present in wood being removed. Selectivity towards lignin vs. hemicellulose extraction increases with increasing flow rate until a plateau is reached.
The thermodynamics and kinetics of adsorption of bis(2,2,6,6-tetramethyl-3,5-heptanedionato) (1,5-cyclooctadiene) ruthenium (II) (Ru(cod)(tmhd)2) on carbon aerogel particles from supercritical carbon dioxide was investigated. The particles had an approximate radius of 1 mm and average pore size of 22 nm. The adsorption isotherms were measured at different temperatures and pressures, thus at different supercritical fluid densities. It was observed that at constant temperature, adsorbed amount (q) decreased with the increasing scCO2 density or pressure at a particular concentration in the fluid phase. The adsorption isotherms were best represented by the Modified Langmuir Model. The maximum uptakes were reached at concentrations considerably less than the solubility of Ru(cod)(tmhd)2 in scCO2 and were found to correspond to monolayer coverage of all the accessible surface of the carbon aerogels as determined by BET measurements. The kinetics of adsorption could be modeled using a model consisting of coupled ODEs based on diffusion in the pore volume and assuming local equilibrium at the adsorbent—fluid interface within the pores. Various simulations of the model were made in order to investigate the effect of isotherm parameters, particle size and pore size of carbon aerogel particles on the kinetics of adsorption. It was observed that in order to prepare Ru(cod)(tmhd)2 impregnated carbon aerogel particles larger than 5 mm with a radially uniform metal distribution which are used as catalysts in industrial applications, one has to take into consideration the long diffusion time. The pore size of the carbon aerogel particles is also very important for the adsorption process. As the pore size starts to approach the solute size, the time to reach equilibrium starts to increase significantly.
The supercritical melt micronization (ScMM) process, also known as particles from gas saturated solutions (PGSS) was applied, in a continuous operated pilot plant, for the particle formation of the edible fat, rapeseed 70 (RP70). The effect of variables like the CO2 concentration, the melt temperature and the atomization pressure were studied in order to investigate particle morphology, density and the particle size distribution. The experiments were performed at CO2 concentrations between 0 and 50 wt%, atomization pressure between 70 and 180 bar and melt temperature between 60 and 100 °C. Particles obtained as a function of the CO2 concentration, showed completely solid, spherical-hollow and aggregated particles with a decrease in particle mean size as the concentration of CO2 was increased. The results obtained as a function of atomization pressure showed no significant influence on particle morphology and size distribution. Experiments carried out as a function of the melt temperature showed distorted, spherical-hollow and aggregated particles. Furthermore a theory was developed to explain the mechanism for particle formation as a function of the CO2 concentration and the melt temperature. The crystallinety of the final product of RP70, showed an alpha polymorph with a crystallinety of 84%.
Extraction of disperse blue 79, red 153, and yellow 119 with supercritical carbon dioxide was investigated, respectively, at 393.2 K and 30 MPa over a wide range of contact times. The dissolution rates of these three dyestuffs were compared. Equilibrium solubilities were also measured over the temperature and pressure ranges of 353.2–393.2 K and 15–30 MPa. As evidenced from the experimental results, the solubilities increase dramatically with increasing pressure and are not as sensitive to temperature. The magnitudes of the solubilities are in the sequence of disperse blue 79>disperse yellow 119>disperse red 153 as pressures higher than 20 MPa. The equilibrium solubility data were correlated with the Chrastil equation. The association numbers of the Chrastil model are approximately 10, 8, and 7 for disperse blue 79/CO2, disperse red 153/CO2, and disperse yellow 119/CO2, respectively.
The solubility of solid 1,4-dimethylaminoanthraquinone (Disperse Blue 14) in supercritical carbon dioxide has been determined in the pressure range of 100–350 bar and in the temperature range of 313–353 K. The values obtained have been correlated with two types of model: the first is based on empirical and semiempirical equations and the second is based on thermodynamic aspects and the use of equations of state, namely Redlich–Kwong (RK), Soave–Redlich–Kwong (SRK) and Peng–Robinson (PR) equations. The thermodynamic model, based on fitting the solid sublimation pressure and binary interaction parameter, shows better agreement with the experimental data than the empirical and semiempirical equations.
The excess molar enthalpies HmE of nitrous + cyclohexane were measured in the vicinity of the critical locus and in the supercritical region by means of an isothermal flow calorimeter. The changes observed in the excess enthalpy with temperature and pressure are discussed in terms of liquid-vapor equilibrium and densities of nitrous oxide and cyclohexane at the conditions of temperature and pressure of the experiments and critical constants for the mixture. Excess enthalpies for the nitrous oxide + cyclohexane system are also calculated using the Peng-Robinson equation of state and several mixing rules, and the resulting HmE values are compared with experiment.
Flow experiments simulating the rapid precipitation of salts during the supercritical water oxidation (SCWO) waste treatment process were performed. Aqueous salt solutions were injected into a coaxially flowing supercritical water stream at a constant pressure of 250 bar. Jet concentrations ranged from 0.1 to 10.0 wt % salt with a typical flow rate of 0.5 g min-1 and temperature of 150 °C. The flow rate of the pure supercritical water stream was typically 10.2 g min-1 with an initial temperature of 550 °C. Results from scanning electron microscopy of collected solids, in situ laser transmission measurements, and low-magnification microscopic or visual observation of the jets indicated that, at 250 bar, sodium chloride solutions first pass through a two-phase, vapor-liquid state before solid salt is formed, while sodium sulfate solutions nucleate solids directly from a homogeneous supercritical-fluid phase. Sodium sulfate solids appeared much finer and also more aggregated than sodium chloride solids. Primary sodium sulfate particle diameters were typically between 1 and 3 μm, while some aggregates reached diameters up to about 20 μm. In contrast, sodium chloride solids ranged from 5− to 25-μm shell-like particles for a 0.5 wt %NaCl jet and 20− to 100-μm semispherical particles for a 10.0 wt % NaCl jet. At a subcritical pressure of 200 bar, the average particle size increased dramatically for both salts. In mixed NaCI/Na2SO4 solutions at 250 bar, the extent of small particle nucleation of sodium sulfate decreased with increasing sodium chloride concentration in the jet feed. Both the observed morphology and mixture effects were explained in terms of different isobaric phase behavior.
Special high-temperature and -pressure multinuclear NMR equipments were constructed, and used for the measurements of 17O-NMR chemical shift and spin-lattice relaxation time (T1) in water over the range from liquid to supercritical (SC) states. The chemical shift could be interpreted in terms of the extent of hydrogen bonding. Although the cleavage of hydrogen bonding of water proceeds continuously from liquid to SC conditions, the hydrogen bonding was found to still remain even under SC conditions. It was confirmed that the spin-lattice relaxation of 17O is mainly controlled by the quadrupole interaction and the T1 values of 17O are related with the values of the molecular reorientational correlation time (τc) over the range from liquid to SC states. It was found that the τc values decrease drastically with increasing temperature and decreasing density in lower temperature and higher density regions (25⩽T<250 °C and 1.1⩾ρ>0.6 g/cm3) and are close to the τc corresponding to a free-rotor in sub-critical and SC ones (250⩽T⩽425 °C and 0.6⩾ρ⩾0.1 g/cm3). It was considered from this result that a decrease of viscosity with increasing temperature and decreasing density leads to acceleration of molecular rotation in lower temperature and higher density regions, and on the other hand the rotational motions of water in sub-critical and SC regions are comparable to those of monometric water molecule. Accordingly, it was concluded that the principal contributions to the molecular rotational motions of water are the variation of the intermolecular hydrogen bonding interactions in lower temperature and higher density regions, and are that of the dipole moment of a water molecule in sub-critical and SC regions, respectively.
In this work, a review of the Brazilian scientific production in the past 10 years in the field of supercritical fluid is presented. The historical facts associated with the previous edition of the Brazilian Meeting on Supercritical Fluids are discussed. Because of the importance of the other South America countries contribution to the filed, a summary of the research published in the past 5 years by all South America countries is also presented.In the past 5 years (1999–2003) 82 papers from South American countries were published in journals indexed in the Web of Science data base. Of these, 26 papers were related to the use of supercritical fluids as an analytical tool. Supercritical extraction from a variety of vegetable raw material contributed with 38 papers and the petroleum industry added 2 papers to the field. Reactions contributed with 3 publications while thermodynamics and fundamental studies were responsible for 13 publications. The Brazilian contribution represented 53–84% of the publications in the above areas.
The oxidation of high concentrations of phenol and 2,4-dinitrophenol (DNP) was investigated in a pilot-scale supercritical water oxidation (SCWO) system. Treatment for approximately 40 s at a pressure of 25 MPa, temperatures of 666–778 K and oxygen excess of 0–34%, resulted in phenol destruction from 94 to 99.98%, consistent with extrapolations of some global rate laws proposed in the literature. Destruction of total organic carbon (TOC) varied from 75 to 99.77%. Two different solutions that contained DNP were studied following the phenol experiments. The first solution contains 2.4 wt.% of 2,4-DNP with 2.1 wt.% of ammonium sulphate. Treatment at under 43 s at 25 MPa, 780 K with a large oxygen excess, resulted in destruction efficiencies of over 99.9996% for DNP and 99.92% for TOC. Mono-nitrophenols were detected as intermediates, but not in the final effluent, where residuals of ammonium bicarbonate and sulphates were detected. This solution was extremely corrosive to the Alloy 625 preheaters at temperatures of approximately 370 °C.The second solution contained 2.26 wt.% of 2,4-DNP, with ammonia but no sulphates and was treated at 24.5 MPa, 742–813 K and oxygen concentrations ranging from sub-stoichiometric to 67% excess. Destruction efficiencies for 2,4-dinitrophenol were over 99.9996% in all cases. TOC destruction efficiencies ranged from 98.98 to 99.98%, while ammonia destruction ranged from 15 to 50%. Picric acid and mono-nitrophenols were detected as intermediates, but not in the liquid effluent. No CO or NOx was present in the effluent gas samples, except in cases with less than stoichiometric oxygen.
Phase behavior and reaction of polyethylene (PE) in supercritical water were studied with a diamond anvil cell (DAC) technique with visual and Raman spectroscopy. When PE+water (12–30% PE ) mixtures were rapidly heated at initial pressures ranging from 110 to 690 MPa, PE first melted and formed a liquid spherule PE phase. The spherule began to expand at above 450°C and underwent a color change to red at about 570°C. At higher temperatures, the red color disappeared and the PE molten phase turned transparent. Upon further heating, the red color returned and other material underwent homogenous reaction as evidenced by a dark color which appeared throughout the cell. Volatile liquids were formed on the surface of the liquid PE phase spherule. For reactions run at higher temperature (645–671°C) at pressures ranging from 1.9 to 2.6 GPa, thin films formed on the anvils after quenching which had CC, OH, and CC Raman bands, which indicated that hydrolysis products formed even though the reaction times were relatively short (290–475 s). Reactions performed at a constant temperature of 423°C and at an initial pressure of 850 MPa showed only a slight decrease (0.03 MPa/s) in pressure with time. The results of this study show, conclusively, that PE and water remain as a heterogeneous system over the polymer (12–30% PE) compositions studied during heating and reaction in supercritical water. Only after PE decomposes to lower molecular weight hydrocarbons, above about 565°C, can homogeneous reaction conditions result.
Experimental data and theoretical predictions of hydrolysis reaction kinetics of model halocarbons and phase equilibria of their associated neutralized salt reaction products are reported for a range of hydrothermal conditions. Specifically, the results of a study of hydrolysis and oxidation of methylene chloride (CH2Cl2) to produce CO2, H2O, and HCl as final mineralized products are presented. An analysis of hydrolysis kinetics and heat transfer was used to generate kinetic parameters for comparison with theoretical predictions based on a modified form of the Kirkwood solution model to show density dependent effects as a function of the dielectric strength of the reaction solvent medium. Phase equilibria data for the ternary NaCl–Na2SO4–H2O, system at 200 and 250 bar were measured and correlated. A modified form of the Anderko–Pitzer semi-empirical PVTxi model was used to predict equilibrium phase boundaries for the NaCl–H2O system at 210, 250, and 300 bar. Molecular simulation results for the NaCl–H2O binary was used to illustrate changes in the solvation power of water as a function of density.
The paper describes a novel extraction procedure for lipopolysaccharides (LPS) from Salmonella enterica subsp. enterica (PCM 2266). Process parameters for the extraction of LPS from bacterial mass were optimized by carrying out a two-level fractional design experiment. Four parameters, namely temperature, CO2 flow rate, pressure and co-solvent composition were analyzed. The best crude extract yields were achieved when the CO2 flow rate and temperature were kept high (10 g/min, 90 °C) and pure water was used as a co-solvent. Pressure had no statistically significant effect within the range of the study performed, whereas the other factors were relevant. The recovery of the extracted LPS by scCO2 was about 3.3% of the biomass used, while in the classical extraction method yield was less than 2%. All isolates were characterized by SDS-PAGE, by the spectra of the thiobarbituric acid reaction products and GLC–MS analysis.
Hydrothermal oxidation is an efficient and clean way for the treatment of wastewater containing organic matter. The purpose of this work is to develop a mathematical model of a reactor for hydrothermal oxidation. This reactor is of the tank type and it is designed for the oxidation of solid particles of waste or biomass. According to some previous work, the experimental device developed by the Institut de Chimie et de la Matiere condensee de Bordeaux is known to behave as a battery of three completely stirred tank reactors (CSTR). To reach our goal, governing equations are written within each of the three CSTR. This set of equations is composed of the mass, species and energy balances for the fluid phase as well as equations allowing for the evaluation of the rate of shrinkage of a particle (shrinking core model) and a population balance. Thanks to this model, the particle size distribution (PSD) of the output of the reactor is computed as a function of the incoming one and of the operating parameters (temperature, residence time, pressure, …). The numerical predictions of the model are compared to experimental profiles obtained in the case of hydrothermal oxidation treatment of black carbon. These comparisons show very good agreement.
Supercritical water flow-through test facility (SCW-TF) for the study of hydrothermal fluids is described. The hydrodynamic behavior of the flow-through reactor is examined from ambient to supercritical water conditions by performing residence time distribution measurements. The results indicate that at 25 MPa, the employed reactor configuration exhibits plug flow behavior with a small extent of dispersion over the temperature range from 298 to 773 K. The experimentally determined effective volume of the reactor was used for the calculation of mean residence times of the fluid in the “hot zone” of the flow-through system. The thermal stability of hydrazine in aqueous solution was examined along the 25 MPa isobar from 473 to 725 K. The obtained first-order rate constant for the thermal decomposition of hydrazine increases from 3.73 × 10−4 s−1 at 473 K to about 0.31 s−1 at 725 K.Graphical abstractResearch highlights▶ The experimental apparatus for the study of hydrothermal fluids is described. ▶ The employed flow-through reactor system exhibits nearly plug flow behavior. ▶ The thermal decomposition of hydrazine follows Arrhenius behavior.
This paper presents the behavior of Alloy C-276 and T60 titanium, submitted to anodic polarization under steady-state and quasi-stationary conditions in oxidative chlorinated aqueous media. The measurements were carried out at temperatures and pressures from usual conditions (20 °C, 0.1 MPa) to supercritical conditions (400 °C, 28 MPa). The influence of various parameters such as pH, chloride ions activity, nature of materials, sub or supercritical state of the medium, was investigated. The effect of the temperature gave rise to the determination of activation energies of the dissolution–passivation processes. The hierarchy of the electrochemical behavior of the materials confirmed the results already obtained in a prevoius work using the determination of corrosion rates.
Hexane is shown to undergo isotopic hydrogen exchange with in supercritical deuterium oxide at 380 and 400°C. The deuteration rate follows pseudo first order kinetics at both temperatures with the methylene reaction rate being about 1.6 times that of methyl. The isotopic exchange reaction is analyzed as a two step acid/base mechanism, with hexane acting as a base analogous to its behavior in `magic acid' solution. Measured Kb's for the methyl group are 3.5×10−28 and 9.2×10−28, while the methylene groups have Kb's of 6.0×10−28 and 1.5×10−27 at 380 and 400°C, respectively. No evidence is seen for hydride abstraction, such as formation of carbocation rearrangement species or hydrogen gas evolution as in `magic acid'. Hydride abstraction to form carbocations either does not occur or occurs at a rate too slow to be observed in the time scale of the experiments reported here.
A new apparatus is described for easy and quick determination of partition coefficients in a system containing an aqueous and a supercritical phase, and an organic substance. The partition coefficient in this case is defined as the ratio of molar fractions of a substance in two different phases in equilibrium. The apparatus consists of a high-pressure cell of 200-mL volume. Equilibrium is attained by recirculation of the fluid phases. Samples can be taken from either phase by using six-way sampling valves. Quantitative analysis is carried out either by UV-spectroscopy or by gas chromatography. In the measurements, carbon dioxide was used as the supercritical fluid in studies of partitioning behavior of the organic compounds, phenol, benzoic acid, benzyl alcohol, 2-hexanone, vanillin, and caffeine. The experiments were carried out at temperatures ranging from 313 to 333 K and pressure of 8 to 30 MPa. Partition coefficients between 0.2 and 1.5 were found for phenol which roughly match the data previously reported by other authors. Partition coefficients of benzyl alcohol and benzoic acid were found to be in a similar range, whereas those of 2-hexanone turned out to be between 10 and 140. The partition coefficients obtained ranged from 0.02 to 0.25 for caffeine and 0.2 to 3 for vanillin.
CO2 promotes penetration and removal of aqueous surfactant cleaning solutions in methylsilsesquioxane (MSQ) low dielectric constant (k) films with 3 nm hydrophobic open pores. The films were characterized by mercury probe dielectric constant (k value) measurements and FTIR spectroscopy. Penetration of a solution of 2 wt.% polyoxyethylene 2,6,8-trimethyl-4-nonyl ether, 5b-C12E8, in H2O at ambient pressure increased the k value of etched and N2/H2 ashed JSR 5109 pMSQ from 2.5 to 7.6, indicating 68% of the total pore volume was filled with the solution. This level of penetration was corroborated by the OH peak at 3150–3560 cm−1 and the CH3 peak (surfactant) at 2800–3000 cm−1. Rapid removal of the surfactant solution was achieved by rinsing and drying with 10 mL/min supercritical carbon dioxide (scCO2) at 45 °C and 10 MPa for 2 min. Both water and surfactant are dissolved and emulsified into CO2. Nearly complete removal of the surfactant and water was observed in the k value, which dropped to 2.5, and in the OH and CH3 peak areas. In addition, the cleaning and drying steps may be integrated with silylation in CO2 to remove silanol groups and to add carbon to further reduce the k value. After rinsing and drying with CO2, silylation with 1 wt.% hexamethyldisilazane (HMDS) in CO2 at 45 °C and 10 MPa followed by annealing at 380 °C for 60 min led to a k value of 2.3, near the original value of 2.15. The ability of CO2 to lower the magnitude of the capillary pressure well below the total pressure facilitates removal of the surfactant solution during rinsing and drying. These results suggest that aqueous surfactant solutions, mixed with CO2, may offer significant advantages for cleaning and drying patterned low-k dielectric insulators, particularly as the feature size shrinks below 50 nm and capillary forces become significant.
In order to analyse the effect of supercritical fluid (SCF) addition on polymer viscosity we have used, for the first time, a high precision vibrating wire instrument, for the simultaneous measurement of viscosities and densities of a SCF-saturated polymer. For the density measurements, this technique makes use of the buoyancy force exerted by a fluid on a solid sinker, detected by means of a vibrating wire sensor placed inside the measuring cell. The viscosity and density of the fluid are obtained by analysis of the resonance curve of the vibrating-wire, using a rigorous model for describing the hydrodynamic effects of the fluid on the wire motion. Results of the preliminary measurements carried out on a commercial linear thermoplastic oligomer [Poly(ethyleneglycol) 400] saturated with CO2 are presented, in a range of temperatures from 313 to 348 K (3 isotherms) and at pressures up to 25 MPa. A sharp decrease of the viscosity was observed for all temperatures. The highest viscosity reduction was reached at the lowest temperature.
Solubility of C.I. Acid Red 57 (AR57) in supercritical carbon dioxide was measured by ion-pairing with hexadecyltrimethylammonium (HDTMA) bromide. The solubility measurements of AR57 and AR57-HDTMA in supercritical carbon dioxide without/with methanol as a modifier solvent were carried out at the temperature range from 35 to 75 °C and for pressures from 250 to 325 bar. The solubility of AR57 and AR57-HDTMA was examined in terms of pressure and temperature of supercritical carbon dioxide. Even though Acid Red 57 is insoluble both in supercritical carbon dioxide and methanol-modified supercritical carbon dioxide, AR57-HDTMA can easily dissolve in methanol-modified supercritical carbon dioxide. The hydrophobic ion-pairing of HDTMA provides a possibility to dissolve a hydrophilic dye in supercritical carbon dioxide. A semi-empirical equation was used to correlate the obtained experimental solubilities of AR57-HDTMA by means of the density of carbon dioxide in methanol-modified supercritical carbon dioxide.
Synchrotron radiation small-angle X-ray scattering (SAXS) technique was used to quantitatively derive structural information of the Dynol-604 (a surfactant) based water-in-CO2 reverse micelles. By using the Guinier plot (ln I(q) vs. q2) on the data sets in a defined small q range (0.025–0.040 Å−1), the radii of the reverse micelles at different pressures and loading water were estimated, which were in the range of 73.8 to ∼78.1 Å.
The supercritical melt micronization (ScMM) process, also known as particles from gas saturated solutions (PGSS) was applied, in a continuous operated pilot plant, for the particle formation of the edible fat, rapeseed 70 (RP70). The effect of variables like the CO2 concentration, the melt temperature and the atomization pressure were studied in order to investigate particle morphology, density and the particle size distribution. The experiments were performed at CO2 concentrations between 0 and 50 wt%, atomization pressure between 70 and 180 bar and melt temperature between 60 and 100 °C. Particles obtained as a function of the CO2 concentration, showed completely solid, spherical–hollow and aggregated particles with a decrease in particle mean size as the concentration of CO2 was increased. The results obtained as a function of atomization pressure showed no significant influence on particle morphology and size distribution. Experiments carried out as a function of the melt temperature showed distorted, spherical–hollow and aggregated particles. Furthermore, a theory was developed to explain the mechanism for particle formation as a function of the CO2 concentration and the melt temperature. The crystallinity of the final product of RP70, showed an alpha polymorph with a crystallinity of 84%.
Measurements of the phase equilibria for the system carbon dioxide–limonene were performed at 50 and 70°C in the pressure range 8.54–13.34MPa. Both the liquid and the gas phase composition were measured. The experiments were carried out by means of a two-chamber recirculation apparatus. According to the procedure, in one of the chambers the gas–liquid equilibrium is established, whereas the second one contains only the gas phase which is continuously recirculated through the chambers to reduce time for equilibration. Before sampling, the chambers are separated to avoid disturbances of the equilibrium when reducing pressure during the withdrawal of the gas phase. Experimental data at 50°C were used for the validation of the experimental method, through comparison with literature data, whereas those at 70°C are mostly original. The experimental data were successfully correlated by means of a thermodynamic model based on the Peng–Robinson equation of state.
The solubilities of seven recently synthesized 1-hydroxy-9,10-anthraquinone (AQ1) derivatives in supercritical carbon dioxide have been measured using a static method. The measurements were performed over the pressure range of 12.2–35.5 MPa at 308, 318, 338 and 348 K. The solubility data have been fitted using the model proposed by Bartle. The calculated results show satisfactory agreement with the experimental data.
In the present work, the technical and economical analysis of extraction and isolation of indole alkaloids from Tabernaemontana catharinensis is presented. The extraction was carried out using supercritical CO2 as solvent and ethanol as cosolvent (5%, v/v). The global yield isotherms were determined at 35 and 45 °C for pressures of 150–350 bar. The mass transfer rate for the constant extraction rate period (CER), the duration of the CER period, and the mass ratio of solute in the fluid phase at the bed outlet were calculated. The extraction curves were adjusted by Crank, Goto et al. and Esquível et al. models. The economical analysis was carried out considering that the cost of manufacturing can be obtained in terms of the costs of investment, operational labor, raw material, waste treatment and utilities. The higher global yields were obtained at 350 bar (1.30 × 10−2 and 1.54 × 10−2 kg/kg, at temperatures of 35 and 45 °C, respectively). The Goto's model was able to quantitatively describe the experimental data. The cost of manufacturing the extracts obtained at 350 bar, 45 °C, using 5% (v/v) of ethanol was US$ 79.35 kg−1 of extract. Using previous experimental data obtained at 300 bar, 55 °C, using 10% of ethanol (v/v), the cost of manufacturing for the fractionation process to obtain a rich alkaloidal fraction (AF) was US$ 440.31 kg−1 of alkaloids.
The optimized structures of several complexes of phosphorus containing compounds and CO2 in the gas phase were determined theoretically. The Gaussian 03 computer software package was used to perform the ab initio computations, including Hartree–Fock (HF/6-31+G(d)) and density functional theory (B3LYP/6-31+G(d)). At the same level of calculation, the binding energy between CO2 and phosphorus containing compounds was calculated. The binding energies for CO2–trialkyl phosphate complexes increased with the number of alkyl groups attached to the phosphorus. Harmonic infrared frequencies were calculated for the CO2 bending mode (ν2) in the various complexes. The appearance of a second peak in the CO2 bending mode (ν2) at lower wavenumber verified the strong interaction between CO2 and phosphorus containing compounds. The results of the ab initio computations were in good agreement with data previously reported.
Abajeru (Chrysobalanus icaco) is a plant that has hypoglycemic properties and is often used in Brazilian popular medicine. In order to identify the flavoring and hypoglycemic substances present in this plant, this work has an objective for the extraction of the essential oil presented in the leaves of abajeru using the supercritical fluid extraction (SFE). The supercritical solvent used is CO2, because of its moderate critical temperature and pressure, atoxicity, low cost and volatility. The experiments were conducted using dried leaves in an apparatus containing a high-pressure pump, a stainless steel extractor of 42mL of volume and a micrometric valve for sampling. Different operational conditions were tested, varying mainly the temperature (313.15–353.15K) and the pressure (10.5–20kPa) in order to investigate the efficiency of the process. The results showed that the best operational condition was at 20kPa and 353.15K. To compare the supercritical carbon dioxide results, the essential oil was also extracted by hydrodistillation and soxhlet, using ethanol as solvent. The chromatographic analysis showed that the different technologies studied extracted the same classes of compounds but the SFE obtained the extract with potential hypoglycemic activity with the presence of lupenol.
The absolute free energy of solvation of methane (CH4) and its fluorinated forms (CH3F, CH2F2, CHF3 and CF4) have been computed via statistical perturbation theory (SPT) in the NPT ensemble at four thermodynamical states (whitin liquid and supercritical regions), in the context of Monte Carlo Simulations. Thermodynamical interpretation of the observed trend in the absolute free energy of solvation in different states reveals an exothermic solvation with ΔSslv < 0 (entropically unfavorable solvation) that the intermolecular interactions play an important role in the solvation process. The fluorinated methanes are confirmed to control the mutual arrangement of neighboring CO2 molecules and the axis of CO2 molecules near F atom of fluorinated solutes make a right angle with F(solute)–C(CO2) axis. Moreover, the energetic distribution along with structural and orientational distributions confirm the existence of a direct interaction between CO2 and F atom, although the extent of this interaction is not very large.
High-pressure CO2 absorption into polymers was measured by a gravimetric method. The amount of CO2 absorbed in the polymer was extrapolated using the CO2 release profile from polymeric samples measured at standard conditions by an external balance. The polymers studied were poly-(methyl methacrylate) (PMMA) and poly(lactic-co-glycolic acid) (PLGA) with different molecular weights (MW) and lactide–glycolide copolymer ratios. Experimental results for PMMA well agree with literature data of Wissinger and Paulaitis [J. Polym. Sci.: Part B: Polym. Phys. 25 (1987) 2497] and Chang et al. [J. Supercrit. Fluids 13 (1998) 113]. For PLGA, CO2 absorption measurements were performed at 313 K and in the pressure range 0.6–4.0 MPa. The method gave reproducible results within the accuracy of ±0.002 for mass fraction, ±0.1 MPa for pressure and ±0.15 K for temperature. A perturbed-hard-sphere-chain equation of state (PHSC EOS) was used to correlate CO2–PLGA absorption isotherms. Pure EOS parameters were estimated by a group contribution method, whereas only one interaction parameter was necessary to fit the measured data. The model fairly describes the experimental data points for all PLGAs considered in this work. These results may be very useful to examine the possibility of successfully using a biodegradable amorphous PLGA polymer for developing effective pharmaceutical products by high-pressure CO2 techniques.
This paper reveals an ultrasonic sensor assembly for accurate determination of the speed of sound and sound attenuation (acoustic energy losses) in dense fluids. It was constructed by means of commercially available 200 kHz emitting/receiving piezoelectric transducers equipped with a special acoustic matching layer. The construction allows changing the distance between transducers without dismantling a high-pressure vessel. This experimental technique makes it possible to determine the sound absorption coefficient, indispensable information for the acoustic determination of gas composition and other properties related to the sound absorption coefficient. The assembly was successfully applied for measuring the speed of sound and sound amplitude in CO2 between 298 and 343 K and 6–14 MPa. The measurements of sound amplitude resulted in accurate location of the critical point of CO2 because in the vicinity of the critical point a decrease of sound signal amplitude is far more significant than a decrease of sound speed.
Organometallic chemistry, chemistry of compounds containing metal–carbon bonds or compounds in which an organic molecule (sometimes with a net negative charge) is bonded to a metal atom through an oxygen or nitrogen atom, is one of the most rapidly growing areas of chemical research. Organometallic compounds are being extensively utilized as reagents in preparation and processing of advanced inorganic materials, as catalysts in production of a wide variety of chemicals and as chemotherapy drugs. Supercritical fluid (SCF) science and technology is another rapidly growing field due to the interesting and desirable properties of SCFs as solvents. The combination of organometallic chemistry and SCFs is a relatively new research area with significant potential. Some applications include (1) use of organotransition metal complexes as homogeneous catalysts for reactions in SCFs [D.A. Morgenstern, R.M. LeLacheur, D.K. Morita, S.L. Borkowsky, S. Feng, G.H. Brown, L. Luan, M.F. Gross, M.J. Burk, W. Tumas, Supercritical carbon dioxide as a substitute solvent for chemical synthesis and catalysis, in: P.T. Anastas, T.C. Williamson (Eds.), Green Chemistry: Designing Chemistry for the Environment, American Chemical Society, Washington, DC, 1996, p. 132 and G.P. Jessop, T. Ikariya, R. Noyori, Homogeneous catalysis in supercritical fluids, Science 269 (1995) 1065], (2) impregnation of polymers with various organometallic complexes from SCF solutions for property enhancement or for subsequent in-situ chemical transformations within such matrices [J.J. Watkins, T.J. McCarthy, Polymer/metal nanocomposite synthesis in supercritical CO2, Chem. Mater. 7 (1995) 1991, and A.I. Cooper, S.G. Kazarian, M. Poliakoff, Supercritical fluid impregnation of polyethylene films, a new approach to studying equilibria in matrices; the hydrogen bonding of fluoroalcohols to (η5-C5Me5)Ir(CO) and the effect on CH activation, Chem. Phys. Lett. 206 (1993) 175], (3) decomposition of organometallic complexes in SCFs for formation of inorganic powders with controlled size distribution [M. Barj, J.F. Bocquet, K. Chhor, C. Pommier, Submicronic MgAl2O4 powder synthesis in supercritical ethanol, J. Mater. Sci. 27 (1992) 2187], (4) SCF extraction of heavy metals from various matrices by formation of organometallic complexes [K.E. Laintz, C.M. Wai, C.R. Yonker, R.D. Smith, Extraction of metal ions from liquid and solid materials by supercritical carbon dioxide, Anal. Chem. 64 (1992) 2875]. At the University of Connecticut, our research efforts are concentrated on evaluation of technical and economical feasibility of some of these applications. The three primary research thrusts in our group have been the utilization of supercritical carbon dioxide (scCO2) as a solvent in rhodium catalyzed homogeneous hydroformylation reactions [D.R. Palo, C. Erkey, Homogeneous catalytic hydroformylation of 1-octene in supercritical carbon dioxide using a novel rhodium catalyst with fluorinated aryl phosphine ligands, Ind. Eng. Chem. Res. 37 (1998) 4203], impregnation of polyurethane foams with organometallic oxidants from scCO2 solutions and subsequent vapor phase polymerization in these foams for production of electrically conductive composite foams [Y. Fu, D.R. Palo, C. Erkey, R.A. Weiss, Synthesis of conductive polypyrrole/polyurethane foams via a supercritical fluid process, Macromolecules 30 (1997) 7611], and investigation of extraction of heavy metals from aqueous solutions by compound formation using scCO2 [J. Murphy, C. Erkey, Copper(II) removal from aqueous solutions by chelation in supercritical carbon dioxide using fluorinated β-diketones, Ind. Eng. Chem. Res. 36 (1997) 5371]. Advances in these areas greatly depend on our understanding the interactions of SCFs and organometallic complexes and how these interactions affect a particular application. The subject matter of this review is extraction of heavy metals from aqueous solutions in the presence of SCFs. Since solvent extraction of heavy metals is utilized on a commercial scale, the replacement of organic solvents by SCFs has been the major driving force behind our research efforts. Therefore, this review was prepared to highlight the areas important for commercial scale application of the technology. In Section 1, an introduction to solvent extraction of metals is given. A brief introduction to the possible advantages of using SCFs is also presented in the same section. The fundamentals of extraction with different types of extractants (cation exchangers, solvating extractants and ion-pair extractants) are given in Sections 2, 3 and 4, together with the studies in the literature on metal extraction using SCFs for each type of extractant. Thermodynamics of extraction is particularly emphasized due to its governing role in the economical feasibility of a large scale process. The experimental methods that are utilized in evaluation of thermodynamic behavior of such systems are provided in Section 5. The current methods to recycle the extractants are presented in Section 6. The kinetics of extraction is described in Section 7 where no studies using SCFs have been reported to date and a brief conclusion is provided in Section 8.
The transesterification between isoamyl acetate and ethanol in supercritical (SC) CO2, 1-n-butyl-3-methylimidazolium hexafluorophosphate ([bmim][PF6], a room temperature ionic liquid), and in CO2+[bmim][PF6] mixed solvent was studied at 65.0 °C and different pressures. p-Xylenesulfonic acid (p-TSA) was used as the catalyst. Results showed that the phase behavior of the reaction system affects the equilibrium conversion of the reaction in SC CO2 and CO2+[bmim][PF6] considerably. The mixtures of SC CO2 and ionic liquids, which are environmental benign solvents, may have potential applications in tuning the equilibrium conversions.
A new apparatus has been developed for the experimental determination of vapour-liquid equilibria in systems containing low-volatility compounds and near-critical carbon dioxide. The apparatus combines the advantages of the gas saturation method with those of a circulation method. An air-lift-functioning draft tube is used in the centre of the equilibrium cell to increase contact time between solute and solvent. The apparatus is tested by measuring the vapour-liquid equilibrium (VLE) of the system CO2+1-butanol at 313.2 and 333.2 K and at pressures up to 11 MPa. For the systems CO2+2-butanol, CO2+vinyl acetate and CO2+2-butyl acetate, VLE data were measured at pressures up to 10 MPa and at the temperatures of 303.2, 313.2 and 333.2 K. The measured mole fractions of the low-volatility compound in carbon dioxide have an accuracy better than 3%. The solubility isotherms were modelled using the PRSV EOS with Wong-Sandler mixing rules and the LCVM model. It is shown that the solubility of the low-volatility compounds in carbon dioxide can be predicted well with the PRSV-WS EOS (NRTL-GE-model) using the fit parameters obtained from bubble point measurements. The PRSV-WS predicts the solubilities better than the LCVM model (t-mPR EOS with an adapted UNIFAC-GE-model).
Solubilities of carbon dioxide in poly(vinyl acetate) (PVAc) were measured at temperatures from 313.15 to 373.15 K and pressures up to 17.5 MPa. Diffusion coefficients of carbon dioxide in PVAc were also measured at 313.15 K and pressures up to 7 MPa. Solubilities and diffusion coefficients of carbon dioxide in molten polystyrene (PS) were studied at temperatures from 373.15 to 473.15 K and pressures up to 20 MPa. An apparatus using a magnetic suspension balance (MSB) was constructed for the measurements. The solubilities in the PVAc and the PS were in good agreement with literature data. The solubility in both polymers were correlated with the Sanchez and Lacombe equation of state to within an average relative deviation of 3.6 and 1.6% for PVAc and PS systems, respectively. The diffusion coefficients in PS were correlated with free volume theory of Kulkarni and Stern to within 10% of relative average deviation.
Pressure-temperature coordinates of critical points for binary mixtures of CO2 and five co-solvents are estimated using a peak-shape-sensitive flow-injection procedure. CO2 is pumped continuously under pressure control into a thermostated capillary tube which is restricted at its outlet and interfaced to a flame ionization detector. The co-solvent is injected into the capillary tube inlet at ambient temperature, where the CO2 is a liquid at the test pressures and transported to the thermostated section of the capillary tube. The shape of the co-solvent peak as it exits the capillary tube and enters the detector is dependent on mass transport rates and phase behavior occurring earlier in the system.
The cubic equation of state proposed by Sako, Wu and Prausnitz (SWP-EOS) is used to calculate the phase equilibrium for quasibinary solutions of poly(ethylene-co-vinyl acetate) copolymers (EVA) in supercritical ethylene. The copolymer is considered to be polydisperse with respect to molecular weight and, additionally, with respect to chemical composition. Equations for the cloud-point curve and for the shadow curve are derived with the aid of continuous thermodynamics incorporating both kinds of polydispersity by the use of a bivariate distribution function. The calculations are performed for EVA+ethylene systems with different averages of molecular weight and chemical composition of the copolymer at five temperatures (393–473 K). The binary interaction parameter is found to be a function of the temperature and of the chemical composition of the copolymer but to be independent of the molecular weight. Predictions of cloud-point curves and shadow curves of EVA+ethylene systems that were not included in parameter fit are very successful.
Isoamyl acetate was successfully synthesized from isoamyl alcohol in supercritical carbon dioxide by enzymatic catalysis. First, the impact of the acyl donor was investigated. Among several reactants, including acetic acid and two different acetates, acetic anhydride gave best yields. Then, two different immobilized lipases (Novozym 435 from Candida antarctica and Lipozyme RM-IM from Rhizomucor miehei) as biocatalysts for the above-mentioned reaction were compared. An esterification extent of 100% was obtained in continuous operation using acetic anhydride as acyl donor and Novozym 435 as enzyme. The amount of enzyme preparation was optimised to 6.25 g/mol alcohol. The effect of substrates load in the solvent was investigated. Operating at a CO2/substrates molar ratio below 7.0, the conversion of alcohol decreased, probably due to an inhibitory effect on enzyme by high concentration of acetic anydride or by produced acetic acid. Pressure in the range of 8–30 MPa showed no effect on this reaction, while an increase in temperature (over 313 K) led to lower production of isoamyl acetate. Novozym 435 was very stable not finding any loss of activity during one month of continuous operation. Finally, carbon dioxide was compared to a conventional organic solvent (n-hexane). Initial reaction rate was higher in SC-CO2, although final esterification extent was similar in both media.
A polymer composite comprised of polyethylene (PE) and poly(vinyl acetate) (PVAc) with a biocompatible surface was developed for use in medical devices by the use of the following two-step processes: (i) the first step was to prepare polymer composite using supercritical carbon dioxide (scCO2) methods and (ii) the second step was to introduce phosphorylcholine groups onto the surface of polymer composite by surface reaction. This article especially reports a detailed study of the syntheses, and mechanical properties and the microstructure of obtained polymer composite by scCO2 methods. The formation mechanism of PE/PVAc composites was that supercritical impregnation of a monomer and initiator into polymer substrate followed by in situ polymerization within the polymer matrix. Differential scanning calorimetry, wide-angle and small-angle X-ray scattering measurement showed that PE and PVAc were blended at the nanometer level. PVAc generated in the amorphous regions at nanometer level did not affect the crystalline of PE. This microstructure affected mechanical properties such as dynamic viscoelasticity, Young's modulus, fracture stress and strain of the PE/PVAc composite. These properties depended on the composition of the composite that can be controlled with polymerization time. The synthesis method of polymer composite using scCO2 can be applied to create polymer biomaterials with suitable mechanical properties for the soft tissue, hard tissue, and organs.
The scaling-up of hydrothermal oxidation process is performed by the use of simulation tools, which require an important experimental support. In this way, we have proposed a new experimental approach to determine the global reaction heats of hydrothermal oxidation reactions. A quasi-adiabatic tubular reactor was developed in order to follow the temperature profiles along the reactor during the oxidation reaction. From the experimental temperature profiles, a model of the reactor thermal behavior was proposed and gives access to the global reaction heats. The experimental setup and model have been validated for acetic acid oxidation in supercritical water with hydrogen peroxide as oxidant. The global reaction heat of acetic acid hydrothermal oxidation (P=25 MPa, 400≤T≤570 °C) has been found equal to −925 kJ mol−1.
Fourier transform infrared (FTIR) spectra and hydrogen bonding of acetic acid in CO2 (1)+n-pentane (2) mixture were studied at 308.2 K up to 11 MPa. The experiments were conducted in the mixed supercritical (SC) and subcritical fluids near critical region and far from the critical region. The results demonstrated that the frequency of CO stretching vibration of the monomer and dimer, the molar absorptivities of monomer and dimer, and the monomer–dimer equilibrium were very sensitive to pressure in the subcritical fluids and supercritical fluids in the critical region, especially as the pressure approached the phase boundary, while the effect of pressure on these properties outside the critical region was very limited.
Acid-base titrations for the KOH-acetic acid or NH3-acetic acid systems were monitored with the optical indicator 2-naphthoic acid at 350 °C and 34 MPa, and those for the HClCl− system with acridine at 380 °C and up to 34 MPa (5000 psi). KOH remains a much stronger base than NH4OH at high temperature. From 298 K to the critical temperature of water, the dissociation constant for HCl decreases by 13 orders of magnitude, and thus, the basicity of Cl− becomes significant. Consequently, the addition of NaCl to HCl raises the pH. The pH titration curves may be predicted with reasonable accuracy from the relevant equilibrium constants and Pitzer's formulation of the Debye-Hückel equation for the activity coefficients.
A synthetic method was used for measuring the solubility of griseofulvin in acetone–CO2 and ethanol–CO2 mixtures. In this method, CO2 is added gradually to a liquid solution previously introduced in a sapphire cell of variable volume. Resulting mixtures may have compositions richer in organic solvents than in CO2, close to compositions found in the batch anti-solvent process. Measurements with the griseofulvin–CO2–acetone system were made at 312.15 K at 60 and 100 bar, and at 326.15 K and 100 bar. Concerning the griseofulvin–CO2–ethanol system, investigations were carried out at 100 bar at temperatures of 312.15 and 326.15 K. The solubility of griseofulvin in acetone decreases at all investigated conditions when the CO2 is added. In this case, CO2 is acting as an anti-solvent. In contrast to this, griseofulvin solubility in ethanol–CO2 mixtures is higher than the solubility in either pure solvents within a certain range of CO2 molar fractions. In this case, CO2 is acting as a co-solvent and promotes the solubility of griseofulvin.
High-pressure viscosity and density of solutions of poly(methyl methacrylate) (PMMA) in acetone and in acetone + CO2 mixtures have been determined in the temperature range from 50 to 125 °C in steps of 25 °C over the pressure range from 7 to 35 MPa in steps of 7 MPa using a falling cylinder type viscometer.Measurements were conducted with PMMA samples of two different molecular weights (Mw = 15,000, Mw/Mn = 1.8 and Mw = 540,000, Mw/Mn = 2.8) at concentrations of 2, 5, 10 and 20 wt.% with Mw=540,000, and at concentrations of 10 and 20 wt.% with Mw = 15,000 sample. The effect of adding CO2 on viscosity was investigated for the 5 wt.% solutions with the high molecular weight polymer sample.The viscosities were observed to be relatively low, with values in the range from 0.2 to 1.6 mPa s for these solutions. The flow activation energies were around 5–10 kJ/mol. Flow activation volumes were in the range from 5 to 40 cm3/mol. Close-packed volumes determined from density correlations were in the range of 0.76–0.98 cm3/g. The overlap concentrations, c*, were estimated to be in the range 0.045–0.075 g/cm3.In the presence of CO2, densities of the solution show an increase, reflecting the higher density of compressed CO2 compared to that of acetone, but viscosities were significantly lowered, with a reduction of about 30% at 4 wt.% CO2 additions. The density dependence of viscosity is used to visually illustrate the need for higher pressures in the presence of CO2 to sustain a given viscosity level.
The composition of the liquid phase of cholesterol–acetone–CO2 system was determined at 308.15 and 318.15 K in the pressure range from 0.1 to 7.5 MPa. The solubility of cholesterol in the acetone–CO2 mixture decreases with increasing pressure and increases with temperature. Recrystallization of cholesterol was performed in acetone solution by using CO2 as an antisolvent. The size of the particles obtained at different conditions was examined by scanning electron microscopy (SEM).
Experimental data for the high-pressure vapor-liquid equilibrium in binary mixtures of carbon dioxide and benzene derivatives (acetophenone, 1-chloronaphthalene, methyl benzoate, and n-propylbenzene) are reported. Measurements were performed at temperatures between 313 and 393 K and pressures up to 18 MPa using a flow-type apparatus. The experimental phase compositions are compared with literature values and are successfully correlated applying a generalized Bender equation of state. Furthermore, the new results for the mixtures containing acetophenone and n-propylbenzene are compared with predictions from the group-contribution equation of state (GC-EOS) by Skjold-Jørgensen.
A method has been developed for measuring the solubility of noble metal acetylacetonates by the direct injection of the supercritical carbon dioxide (scCO2) solution into a high pressure liquid chromatograph. We found the use of (1) reverse phase high pressure liquid chromatography (RP-HPLC) to separate the sample peak from the noise peak on chromatogram and (2) a scCO2-philic eluent to eliminate the noise peak, to be effective. Using this improved method, we measured the solubilities of noble metal complexes of Pt, Pd, Ru, Rh and Ag acetylacetonates in scCO2 from 10 to 30 MPa at 313 K. Measurements were conducted with smaller amounts of samples (less than 0.1 g) and with shorter experimental times than by the standard dynamic flow method. The molar fraction y2 was of the order 10−5 to 10−4 for Ru and Rh acetylacetonate, 10−5 for Pd and Pt acetylacetonate, and 10−7 for Ag acetylacetonate. The solubility data for Pd, Pt, Ru, and Rh acetylacetonates were successfully correlated using the Chrastil model.
Top-cited authors
E. Reverchon
  • Università degli Studi di Salerno
M. Angela A. Meireles
  • University of Campinas
Iolanda de marco
  • Università degli Studi di Salerno
Andrea Kruse
  • University of Hohenheim
Željko Knez
  • University of Maribor