Roumiana P. Stateva

Solubility of β-carotene in supercritical CO2 and the effect of entrainers

Abstract Solubilities of trans - β -carotene in supercritical CO 2 , both pure and modified, were measured at temperatures 313.2, 323.2 and 333.2 K and pressures up to 28 MPa using a one-pass flow apparatus. The solubility data measured in pure CO 2 were correlated by the density-based equation proposed by Chrastil, and modelled using the Soave-Redlich-Kwong cubic equation of state. Most of the literature data were in good agreement with our results; smaller discrepancies were ascribed mostly to β -carotene degradation. While ethanol entrainer increased the solubility of β -carotene in supercritical CO 2 by one order of magnitude, vegetable oil was less efficient. In both cases, the increase in solubility was proportional to square root of entrainer concentration.

Supercritical fluid extraction from vegetable materials

Abstract In the 21st century, the mission of chemical engineering is to promote innovative technologies that reduce or eliminate the use or generation of hazardous materials in the design and manufacture of chemical products. The sustainable use of renewable resources, complying with consumer health and environmental requirements, motivates the design, optimisation, and application of green benign processes. Supercritical fluid extraction is a typical example of a novel technology for the ecologically compatible production of natural substances of high industrial potential from renewable resources such as vegetable matrices that finds extended industrial application. The present review is devoted to the stage of development of supercritical fluid extraction from vegetable material in the last 20 years. Without the ambition to be exhaustive, it offers an extended, in comparison with previous reviews, enumeration of extracted plant materials, discusses the mathematical modelling of the process, and advocates a choice for the appropriate model that is based on characteristic times of individual extraction steps. Finally, the attention is focussed on the elements of a thermodynamic modelling framework designed to predict and model robustly and efficiently the complex phase equilibria of the systems solute+supercritical fluid.

Essential oils from seeds: solubility of limonene in supercritical CO2 and how it is affected by fatty oil

Abstract Fatty oil influence on the solubility of limonene in CO 2 was investigated under pressures 8–12 MPa at 313.2 K, a temperature typically applied in supercritical fluid extraction of essential oils. Solubility in CO 2 was measured using the dynamic method both for limonene and for the mixture of limonene and blackcurrant seed oil. In the whole range of pressures applied, the concentration of fatty oil in the vapour phase is negligible in comparison with the concentration of limonene. Limonene is distributed between the liquid phase rich in fatty oil and the vapour phase rich in CO 2 , and its equilibrium concentration in the latter decreases with the diminishing limonene-to-oil ratio in the saturator. There is a steep increase of the limonene partition coefficient with pressure between 8 and 10 MPa, near the critical pressure of the binary mixture of limonene and CO 2 . The observed behaviour of the three-component system was confirmed and explained by thermodynamic modelling. The thermodynamic model applied was the Soave–Redlich–Kwong cubic equation of state with either the one fluid linear van der Waals mixing rule or with the MHV2 mixing rule. Appropriate conditions for an efficient supercritical fluid extraction of essential oils from seeds follow from the results obtained. Extraction pressure should be approx. 20% larger than the critical pressure of the essential oil+CO 2 binary mixture and rather tight packing of the ground seed in the bed should be applied.

Application of a stochastic method to the solution of the phase stability problem: Cubic equations of state

Abstract Phase equilibrium calculations and phase stability analysis are of fundamental importance in various chemical engineering applications, such as azeotropic and three-phase distillation, supercritical extraction, petroleum and reservoir engineering, etc. Phase stability is often tested using the tangent plane criterion, and a practical implementation of this criterion is to minimise the tangent plane distance function (TPDF), defined as the vertical distance between the molar Gibbs energy surface and the tangent plane for given phase composition. In the present work, we use a modified TPDF and an equation of state as the thermodynamic model. We advocate a stochastic sampling and clustering method to locate the minima of the TPDF and compare its reliability with some of the most promising global optimisation methods. Our method is user-friendly and not computationally demanding regarding the number of function evaluations, and CPU time.

Estimation of normal boiling points of hydrocarbons from descriptors of molecular structure

Correlations for estimation of thermophysical properties are needed for the design of processes and equipment related to phase equilibria. The normal boiling point (NBP) is a fundamental characteristic of chemical compounds, involved in many correlations used to estimate important properties. Modern simulation packages usually require the NBP and a standard liquid density from which they can estimate all other necessary properties and begin the design of particular processes, installations and flowsheets. The present work contributes a correlation between the molecular structure and the normal boiling point of hydrocarbons. Its main features are the relative simplicity, sound predictions, and applicability to diversified industrially important structures, whose boiling points and numbers of carbon atoms span a wide range. An achievement of particular interest is the opportunity revealed, for reducing the number of the compounds required for the derivation (the learning set), through multivariate analysis and molecular design. The high accuracy achieved by the correlation opens up a possibility for systematic studies of chemical engineering applications in which the effects of small changes are important. This also defines a path towards the more general problem of the influence of uncertainties in calculated thermophysical parameters on the final outcome of computer aided simulation and design.

Numerical solution of the isothermal, isobaric phase equilibrium problem

Thermodynamic modelling and phase equilibria are at the heart of chemical process design. For example, 70 to 90% of the equipment and energy costs in modern chemical plants are related to separation and purification processes and they are designed largely on the basis of thermodynamic equilibrium. Furthermore, the occurrence of an azeotrope or a liquid-liquid phase split may be the determining factor in developing a flow sheet for a new process. Thermodynamic modelling is playing an increasing role in such diverse areas as environmental engineering and down-stream processing in biotechnology. The phase behaviour of multicomponent systems is a complex topic from at least five points of view. First, the number of thermodynamic state variables required to describe thermodynamic equilibrium in an ^-component system is Nc + 2, and so can be quite large in industrial practice. Secondly, the experimental study of phase equilibrium is very time consuming and difficult especially for multicomponent systems away from ambient conditions. Thirdly, there exists no exact, practicable theory that relates the phase behaviour of a dense fluid system to the properties of its molecules. Fourthly, the description of the phase behaviour of materials based upon simple heuristic models is prone to substantial error. Finally, even when such simplified thermodynamic models are applied, the determination of the phase behaviour of a multicomponent fluid system (number of phases, their identity

Extraction of thymol from different varieties of thyme plants using green solvents

BACKGROUND Thymol (2-isopropyl-5-methylphenol) is the main monoterpene phenol found in thyme essential oil. This compound has revealed several biological properties, including antibacterial, anti-inflammatory and antioxidant activities. In this work, a comparison was made between the performance of different green solvents (ethanol, limonene and ethyl lactate), by pressurized liquid extraction (PLE) and supercritical fluid extraction (SFE) at different conditions, to extract thymol from three different varieties of thyme (Thymus vulgaris, Thymus zygis and Thymus citriodorus). Additionally, new solubility data of thymol in limonene and ethanol at ambient pressure and temperatures in the range 30-43 °C are reported. RESULTS The highest thymol recoveries were attained with T. vulgaris (7-11 mg g(-1)). No thymol could be quantified in the PLE samples of T. citriodorus. The highest concentrations of thymol in the extracts were obtained with limonene. Thymol is very soluble in both solvents, particularly in ethanol (∼900 mg g(-1) at ∼40 °C), and is the main compound (in terms of peak area) present in the essential oil extracts obtained. CONCLUSION The three solvents show good capacity to extract thymol from T. vulgaris and T. zygis by PLE. Although PLE proved to be a suitable technology to extract thymol from thyme plants, the highest concentrations of thymol were obtained by SFE with supercritical CO2 .

A powerful algorithm for liquid–liquid–liquid equilibria predictions and calculations

Abstract The paper describes a new robust and efficient algorithm for liquid–liquid–liquid equilibria (LLLE) predictions and calculation. The architecture of the algorithm is a two-level one, and a judicious combination of a stability analysis, carried out in the first stage, with phase identification routines and liquid flash calculations, carried out in the second stage, is realised. The numerical routines require modest computational efforts and demonstrate excellent convergence characteristics. The thermodynamic model applied to a system does not influence the robustness of the new algorithm, which can treat any model and handle any number of components and phases. The usefulness and efficiency of the new algorithm is exemplified by using two model systems, a water+oil+non-ionic surfactant system and the system ethylene glycol+lauryl alcohol+nitromethane.

Analysis of predictive thermodynamic models for estimation of polycyclic aromatic solid solubility in hot pressurized water

The ability of two thermodynamic approaches to predict the solubility of solid compounds in hot pressurized water is studied and compared. The Regular Solution Theory, based on the solubility parameter concept, and UNIFAC- based models were applied to calculate the solute activity coefficient and then, solubility predictions were compared with experimental data reported in the literature. The analysis was carried out considering polycyclic aromatic hydrocarbons as model substances, i.e. substances which contain only the aromatic AC and ACH groups, and for which reliable pure physical properties such as melting point, fusion enthalpy and molar volume are available in the literature. The solubility values predicted with the UNIFAC-based models were considerably better than those obtained with the solubility parameter approach. Particularly, the modified Dortmund UNIFAC model presented an appropriate functionality of solubility with temperature, and the extension of this model to other type of aromatic compounds also provided a satisfactory prediction of solubility data.

Identifying equations that represent properties in homologous series using structure-structure relations

Abstract The Quantitative-Structure-Structure-Property Relationships (QS2PR) technique, which we introduced recently, is adapted to the prediction of properties of pure compounds in homologous series. The QS2PR method involves calculation of the molecular descriptors of a target compound of unknown properties, followed by regression of this vector of molecular descriptors versus a database of compounds with known descriptors and measured properties. The regression model, obtained for the target descriptor in terms of predictive compounds and their coefficients, is then used for predicting properties of the target compound. A structure-structure relationship is derived from the number of carbon atoms and one additional, nonlinearly dependent molecular descriptor of the target compound and three predictive compounds. It is shown that such a relationship can provide predictions of satisfactory precision for many properties of the members of the homologous series. This enables users, without access to libraries of molecular descriptors, to employ the QS2PR technique for property prediction.