High-pressure phase equilibria for product and process design

Mestrado Integrado em Engenharia Química High-pressure phase equilibria for product and process design Tese de Mestrado de Bárbara Muge de Almeida Libório Desenvolvida no âmbito da unidade curricular de Dissertação realizado em Orientador na FEUP: Prof. Dr.Domingos Barbosa Orientador na BTS GmbH: Dr. José Fonseca Departamento de Engenharia Química Leverkusen, Alemanha, Julho de 2013 High-pressure phase equilibria for product and process design i "If I have seen further it is only by standing on the shoulders of Giants." Isaac Newton (1643-1727) Philosophiae Naturalis Principia Mathematica, 1687 High-pressure phase equilibria for product and process design ii Acknowledgments I would like to express my thanks to some people, who made a valuable contribution during this period. My first and sincere appreciation goes to my supervisor, Dr. José Fonseca, for all the support, suggestions and comments given, in all stages of this master thesis. I am grateful to him for all the knowledge sharing and continuous help. I would like to express my gratitude to Prof. Dr. Ralf Dohrn for welcoming me in his research group and giving me the opportunity to complete my studies in this demanding and successful group. I am also thankful to all technicians, in special, Mr. Thomas Nicke and Mr. Stefan David, for all the help and patience while performing the experiments. In addition, I would like to thanks my colleague Ximena Rodriguez, for all the friendship and pleasant environment since her first day at the laboratory. My gratitude goes also to Prof. Dr. Domingos Barbosa, from Porto University, for all willingness to help and guidance, not only during those 6 months of internship, but also during the previous period. My greatest appreciation and friendship goes to my friends, Nuno Garrido and Miguel Teixeira, who have always been a great support during these years. I also would like to thank, André Calatré, Eduardo Basto and Tiago Costa, for all the friendship and fruitful discussions/contributions in different areas of engineering. I want to express my gratitude to my closest friends, Joana Vieira, Marlene Faria and Telma Luis, who were always a great support in all my struggles and frustrations. Thanks for all the love, good moments and helping me thinking rationally. A big “thank you”, Flávia Figueiredo, Filipa Coelho and Marcelo Costa, for your help and fellowship during the entire course. To my family, for all the support through these years, and specially my mother, who always encouraged and supported me in my decisions, thanks for believing in me and for being always there for me. I also would like to thank Abdessamad Rafia for all the constant support, companionship and patience during those months. Finally I want to express all my gratitude to the people that in one way or another have contributed to the success of this thesis. Unfortunately I am not able to put all your names here, once only 50 pages are allowed, but I will always remember you for motivation and experience. High-pressure phase equilibria for product and process design iii Sumário Foi desenvolvido um novo equipamento para o estudo do equilíbrio de fases a alta pressão e a temperaturas elevadas, utilizando um método sintético com mudança de fase. O equipamento foi especialmente desenvolvido para o estudo de sistemas com misturas gasosas, a pressões até 20 MPa. A componente principal do equipamento desenvolvido é uma célula de equilíbrio de volume variável, equipado com duas janelas de safira de 360º. Foram ainda re-desenhadas duas células de equilíbrio já existentes no Laboratório de Propriedades Físicas e Termodinâmica, até agora aplicáveis para pressões até 0.6 MPa. Os componentes de vidro foram substituidos por elementos de safira, e as partes de aço redesenhadas, para suportar pressões até 10 MPa. O presente trabalho inclui ainda a participação, através de estudos experimentais de equilibrio de fases, num projeto de grande importância para a Bayer, que visa a utilização de CO2 como matéria-prima na produção de poliuretanos. Como parte da otimização do processo de produção, foram medidas solubilidades de dióxido de carbono e de óxido de propileno no polímero a pressões até 10 MPa e até 0.6 MPa, respectivamente, e verificada a influência da presença de carbonato de propileno nestes valores de solubilidade. Por último, a solubilidade de azoto no polímero à temperatura ambiente e pressão atmosférica foi também determinada. Keywords Equilíbrio de fases, Dióxido de carbono, Solubilidade, alta pressão, design de processos High-pressure phase equilibria for product and process design iv Abstract A new apparatus for the study of high-pressure phase equilibrium at high temperatures using a synthetic visual method with phase change was designed. The apparatus was specially developed for the study of systems with gas mixtures, at pressures up to 20 MPa. The core of the apparatus is a variable-volume equilibrium cell, equipped with two 360◦ sapphire windows. Two equilibrium cells already existent on the laboratory of Property Data and Thermodynamics of BTS, applicable for pressures up to 0.6 MPa, were re-designed and developed. The glass parts were exchanged for sapphire and the stainless steel parts redesigned in order to cope with pressures up to 10 MPa. This work also includes the collaboration, through experimental studies of phase equilibria, in a project of great importance for Bayer, focusing on the use of CO2 as raw material for the production of polyurethanes. As part of the process optimization, solubilites of carbon dioxide and of propylene oxide in the polymer were measured at pressures up to 10 MPa and 0.6 MPa, respectively. The influence of propylene carbonate (also present in the process) on these solubilities was also determined. Finally, the solubility of nitrogen in the polymer at room temperature and ambient pressure was also determined. Keywords Phase equilibria, Carbon dioxide, Solubility, Highpressure, Process design High-pressure phase equilibria for product and process design v Declaração Bárbara Muge de Almeida Libório declara, sob compromisso de honra, que este trabalho é original e que todas as contribuições não originais foram devidamente referenciadas com identificação da fonte. Bárbara Muge de Almeida Liborio states, on honor, that this work is original and that all non-original contributions were properly referenced with source identification. High-pressure phase equilibria for product and process design vi Contents Acknowledgments .............................................................................................................................. ii Sumário ............................................................................................................................................... iii Abstract............................................................................................................................................... iv Declaração ........................................................................................................................................... v Contents .............................................................................................................................................. vi List of Figures ..................................................................................................................................... 1 List of Tables....................................................................................................................................... 3 Nomenclature ..................................................................................................................................... 4 Preface ................................................................................................................................................. 6 1 Introduction ..................................................................................................................................... 7 2 Phase equilibrium measurements – A review ........................................................................... 10 2.1. Analytical Methods................................................................................................................... 12 2.2. Synthetic Methods .................................................................................................................... 14 2.2.1. Synthetic Methods with phase transition ......................................................................... 14 2.2. Synthetic methods without phase transition ...................................................................... 17 3 Design and development of a new apparatus - Synthetic visual method with phase change ................................................................................................................................................ 19 3.1. Preliminary steps ..................................................................................................................... 19 3.2. New apparatus .......................................................................................................................... 20 4 New equilibrium cells - Synthetic isothermal method without phase change ................... 26 5 Solubility of CO2 and propylene oxide in polymers derived from CO2 ................................. 28 5.1. Experimental procedure ......................................................................................................... 29 5.2. Experimental Results ............................................................................................................... 32 6 Conclusions and Further work .................................................................................................... 38 7 References ..................................................................................................................................... 39 Appendixes 1 ..................................................................................................................................... 42 High-pressure phase equilibria for product and process design 1 List of Figures Figure 1 - Different stages of a process that are dependent on phase equilibria data. 7 Figure 2 - Classification of the experimental methods available for the measurement of high-pressure phase equilibrium [14]. 12 Figure 3 - Determination of a boiling point with a synthetic method with phase transition. Left: pressure vs. volume diagram; Right: mole fraction diagram [2]. 15 Figure 4 - Equilibrium cell for visual determination of phase transitions from TU Delft [31]. 16 Figure 5 – Experimental procedure, synthetic isothermal method [2]. 18 Figure 6 – Schematic representation of the new experimental set-up for the measurement of multi-phase equilibria at high-pressures. 20 Figure 7 - 3D computer generated images of the new high-pressure equilibrium cell. – On the left: overview of the cell. On the right: cut of the cell showing the interior. 22 Figure 8 – Fiber optic pressure sensor. Left: schematic pressure sensor; Right: photo [37]. 23 Figure 9 - Pressure sensor’s configuration. Fabry-Perot Cavity in MEMS Pressure Sensor [37]. 24 Figure 10 - Stirring mechanism specially designed for this specific equilibrium cell. 25 Figure 11 - Existing glass cell for phase equilibria measurements, synthetic visual method. 26 Figure 12 - 3D computer generated images of the new high-pressure equilibrium cells. Left: overview of the cell. Right: cut of the cell showing the interior. 27 Figure 13 - CO2 as raw material for the production of polyurethanes [39]. 29 Figure 14 - Apparatus where the experiments for the study of solubility of CO2 and PO were performed [35] 30 High-pressure phase equilibria for product and process design 2 Figure 15 - Schematic representation of the apparatus used for the solubility measurements of N2 [39] 32 Figure 16 - Solubility of CO2 in PPP type1 at two different temperatures. 33 Figure 17 - Solubility of CO2 in PPP type2 at two different temperatures. 34 Figure 18 - Solubility of CO2 in a mixture of PPP (type1) and propylene carbonate at 373 K. 35 Figure 20 - Solubility of PO in a mixture of PPP (type 1) and propylene carbonate at 373 K. 36 Figure 21 - Solubility of PO in a mixture of PPP (type 1) and propylene carbonate at 373 K, repetition of the experiment. 37 Figure 22 - Solubility of N2 in PPP (type 1) at room temperature, 298 K. 37 High-pressure phase equilibria for product and process design 9 Another part of this work included the participation in a project of great importance for Bayer. Around 10% of the production of crude oil is used in the production of chemicals, such as polymers. This highly innovative project consists in the use of CO2 as raw material for the production of polyurethanes. On one hand, this allows decreasing the current dependency of crude oil and thus reduces the influence of oil prices in production costs. Simultaneously, it gives value to CO2, providing an alternative to the storage possibilities after carbon capture in flue gases in pilot plants, bringing added value to the chemical industry. The project is a partnership of Bayer Material Science AG and Bayer Technology Services GmbH, and the Property Data and Thermodynamics group has had an important participation in the study of relevant systems, from phase equilibrium at high pressure for the optimization of the reaction process, to phase equilibrium at low pressure for the separation process, where the final product needs to meet the demanded specifications, including the full characterization of pure substances and relevant mixtures in terms of thermodynamic and transport properties. The subsequent chapters are organized as following: In Chapter 2, a smooth approach focusing on the experimental methods for the measurement of phase equilibrium is presented, with special attention given to the methods of interest for this thesis. In Chapter 3, the design of a new experimental set-up for the measurement of solubilities by using the visual method with phase change is presented. Chapter 4 focuses on the re-design of existing view-cells, which were further developed for application at higher pressures. In Chapter 5, the results obtained in experimental studies of phase equilibria, performed as part of a project of wide importance for Bayer, are presented. Finally, in Chapter 6, conclusions were made and considerations concerning future prospections are presented. High-pressure phase equilibria for product and process design 10 2 Phase equilibrium measurements – A review Phase equilibrium is a subject of wide importance in many industrial and scientific areas as introduced in the previous chapter. Nowadays, it is possible to find a large number of databases, books and reviews in the literature published over the years by several authors, containing important information on this area. Examples of databases include DECHEMA Chemistry Data Series, a German collection of 15 volumes, each one dedicated to a specific set of data (VLE, SLE, LLE, critical data, etc.), or the IUPAC Solubility Data Series, an American collection of 98 volumes published since 1979. Since volume 66 their data is also appearing in the Journal of Physical Chemistry and Reference Data. In 1967 Hála et al [6] published an extensive record of systems for which experimental VLE data at moderate pressures had been published up to May of 1965. The work of Bartle et al [7] focused on solids and liquids of low volatility. Sawamura [8] published a work which focuses on solubility of organic and inorganic compounds in liquids at high pressures and Esperança et al. [9] presented a complete review between 2003 and 2008 about the volatility of aprotic ionic liquids. Another two interesting works that must be mentioned are the publications of Keskin et al [10] on phase behavior of ionic liquids with supercritical fluids and of Marsh et al [11] focusing the critical properties of elements and compounds. Examples of reviews of phase equilibrium measurements for one particular system are the works of Diamond and Akenfiev [12] or Spycher [13], who studied the binary system of carbon dioxide + water. Dohrn and co-workers gathered the systems published between 2000-2004 [14] and 2005-2008 [15] and presented a classification for experimental methods. These works belong to a series of reviews that have been done focusing on the data published between 1978-1987 [16] 1988-1993 [17] and 1994-1999 [18]. The interest in studying the phase behavior at high-pressures is not recent. In 1822, Charles Cagniard de la Tour [19] discovered the critical point of a compound by using a synthetic nonvisual method. The experiments were performed in a sealed cannon filled with fluids at different temperatures. Inside the cannon there was a rolling flint ball, by listening to the discontinuities in the sound in it as the cannon was tilted, he observed the critical temperatures. Over time, certain systems have been studied several times by different authors. But the existence of experimental data from different sources for one same system does not necessarily imply a lower uncertainty in the phase equilibrium, since there are often High-pressure phase equilibria for product and process design 11 discrepancies between authors. Due to this, the need of reliable and accurate data in extended ranges of experimental conditions is still a reality, even for some systems that have been the object of previous studies, since wrong data may erroneously influence industries and processes. A reason for many of these discrepancies is related to the fact that reliable data and accurate measurements are difficult to achieve and often the quality of the operators and the supervising of the experiment can have higher impact on the results than the equipment to be tested [15]. If some reviews focus on certain systems, or on the data published for certain types of compounds, others provide a more instrumental perspective. The work of Smith and Fang [20] for example, focused on the application of diamond anvil cells. The review presented by Richon [5] focus on equipment developed in his laboratory using analytical methods with sampling. In 2012, Fonseca [21] presented the development and test results of new experimental set-ups for the measurement of phase equilibrium at high-pressure and low temperatures using different methods, and then comparing the results obtained by the different apparatus. In experimental studies of phase equilibrium at high pressures a wide variety of methods can be performed. The main reason of this diversity is because there is no way to define which method is the best suitable for each measurement, as this depends on both the properties of the components and the phenomena to be investigated. For the selection of the best method, or for the critical evaluation of experimental literature data, a detailed knowledge and understanding of the different methods is necessary, since each method has different advantages and disadvantages as well as specific error sources. A possible categorization will be presented on the figure bellow [2]. This nomenclature was previous presented by Dohrn and co-workers in recent reviews [14 -15]. As shown on Figure 2, there are two main classes of methods for measurement of highpressure phase equilibrium: the analytical methods, where the compositions of the equilibrium phases are determined, and the synthetic methods, where the mixture has to be prepared with precisely known composition. According to some authors [22] the methods can also be classified in agreement with the working regime, such as, static or dynamic methods. High-pressure phase equilibria for product and process design 12 Figure 2 - Classification of the experimental methods available for the measurement of high-pressure phase equilibrium [14]. 2.1. Analytical Methods In analytical methods, the equilibrium cell is charged with the components of the system to be studied without requiring precise information of the total composition of the system. The experimental conditions such as temperature and pressure are then set. Once the equilibrium is achieved, there are two typical ways to determine the composition of the different phases: withdrawing a sample for subsequent analysis, or applying a suitable physicochemical technique for an in situ analysis. Apparatus for the study of phase equilibrium by analytical methods can be significantly complex since they must include an analytical part, and often it requires calibration and previous optimization of the analytical technique(s). Analytical methods can be distinguished by having samplings withdraw or not. These methods are based on different principles and have different requirements in terms of experimental procedure and error minimization. Methods in which samples are withdrawn from the equilibrium cell can be classified, depending on the procedure used to achieve the equilibrium, into isothermal, isobaric or isobaric-isothermal methods [2]. The main advantage measurement of high-pressure phase equilibria analytical methods with sampling isothermal isobaric isobaricisothermal without sampling spectroscopic gravimetric others synthetic methods with a phase change visual non-visual without a phase change isothermal isobaric others High-pressure phase equilibria for product and process design 13 of the analytical methods with sampling is the possibility of application in multi-component systems which allow the study of systems of industrial interest. However, these methods present some disadvantages, such as the possibility of the occurrence of significant pressure drops when samples are withdrawn [23]. Nevertheless, many techniques can be performed in order to avoid or reduce this problem. Probably the most basic solution for this is the use of cells with larger volumes, so that the volume of the sample withdrawn has a lower influence on the pressure of the system. Apparatus using analytical method with sampling where the volume of the equilibrium cell is small may, nevertheless, be found in literature. However their applications are scarce. A considerable advantage for most of the methods, whilst it can increase the price of the experimental set-up, is the use of windows for visual observation of the contents of the equilibrium cell, interfaces between phases, among other phenomena. Depending on the pressure range for the application, among other factors, these windows can be made of glass or quartz, but most commonly sapphire is used. Although more expensive, sapphire is much strong than glass and it is characterized by a good chemical resistance, thermal conductivity and thermal stability. Another possible approach for circumventing the problem of the pressure drop associated with sampling processes is to avoid sampling, performing the analysis in situ through the use of a convenient physicochemical method. Several spectroscopy techniques can be used, as shown by Andersen et al. [24] where the sapphire windows assumes a critical importance, due to a good transmission characteristics of the visible, near IR and UV spectrum of this material, such as Raman spectroscopy as used by Khanna et al. [25] A disadvantage is that often these methods of analysis only permit the determination of the concentration of a particular compound in different phases, which may be a limiting factor in the study of multi-component systems [26]. Calibrations under different experimental conditions are time-consuming, which could be a negative aspect to take in consideration. Among analytical methods without sampling, there are also gravimetric methods as well as other types of methods. High-pressure phase equilibria for product and process design 14 2.2. Synthetic Methods While in analytical methods phase equilibrium is determined by analysis of the equilibrium phases, as previously explained, the basics of synthetic methods consists on the preparation of a mixture of precisely known composition and subsequently observe its phase behavior in an equilibrium cell, measuring only properties such as pressure and temperature in the equilibrium state. Synthetic methods can be applied in situations in which analytical methods are not convenient. Nevertheless, the problem of analyzing the fluid mixtures is replaced by the problem of carefully preparing (synthesizing) them [3]. Since no sampling is necessary, experimental set-ups using synthetic methods are usually smaller and require few components than the equilibrium apparatus used in analytical methods. However, as well as in analytical methods, a larger cell volume can be advantageous. For example, Nieuwoudt and co-workers [27] increased the volume of a highpressure equilibrium cell to the double, which lead to an increased accuracy on the composition determinations. One possible classification for synthetic methods is based on the existence, or not, of a phase transition. 2.2.1. Synthetic Methods with phase transition In synthetic methods with phase transition, variations on pressure or temperature can be used to promote the appearance or disappearance of a new phase. The moment when a second (gaseous) phase appears, the composition of the condensed phase can be considered to be equal to the global composition of the system. Each experiment yields one point of the pTx phase envelope. Alternatively to varying the pressure or temperature of the system, the overall composition can also be changed in order to promote the appearance / disappearance of a new phase. Figure 3 shows the determination of a boiling point with a phase transition methods. As showed on the left side, the pressure is measured as the volume of the cell changes. The phase transition is detected visually by observing at what pressure the vapor phase appears or disappears. The phase transition can also be detected non-visually by observing the change in the slope of the pV curve. On the right side of the figure is it possible to see the results of a boiling-point determination in a pressure-composition diagram: pressure versus mole fraction High-pressure phase equilibria for product and process design 15 of component 1 in the liquid phase (x1) and in the vapor phase (y1). The points represented on the left side of the figure have correspondence to the points represented on the right side. Figure 3 - Determination of a boiling point with a synthetic method with phase transition. Left: pressure vs. volume diagram; Right: mole fraction diagram [2]. One example of the application of this method, sometimes also called “vanishing-point method”, can be found in Wubbolts et al. [28] who used it in the study of solid-liquid equilibrium. The procedure used by the authors consisted in adding a known solute concentration to a known amount of anti-solvent, in order to make the crystals disappear. The composition of the mixture at this vanishing point is the solubility of the mixture. By performing analogue processes with solutions of different concentrations, additional points for the solubility curve will be found. Synthetic Visual Methods As mentioned, the detection of the phase transition can be done by visual or non-visual means. The first type is most common [2,14-15] and the appearance of a new phase is detected by visual observation of the resulting turbidity or the appearance of a meniscus in a view cell. However, visual synthetic methods present some limitations, for examples, in cases of isooptic systems, where the coexisting phases have approximately the same refractive index, making visual observation unfeasible [26]. An advantage of this method is that it can be applied not only for the determination of simple VLE, but also in the study of more High-pressure phase equilibria for product and process design 16 complex phase behavior, such as SLE, critical curves mixtures, gas hydrate formation, cloudpoint determinations or phase equilibria in polymer solvent systems [26]. One of the most used experimental set-ups using synthetic visual methods is the Cailletet apparatus of TU Delft [29]. The apparatus is shown on Figure 4 and it consists of a thickwalled Pyrex glass tube with the open end placed in an autoclave and immersed in mercury. The mercury confines the sample in the Cailletet tube and a stainless steel ball driven by reciprocating magnets provides the stirring of the sample. Figure 4 - Equilibrium cell for visual determination of phase transitions from TU Delft [30]. To improve the detection of phase transitions is possible to find in literature special techniques such as laser light scattering. Jager and Sloan [31] used Raman spectroscopy in order to detect the appearance of gas hydrates. Dong and co-workers [32] used additional small angle X-Ray scattering measurements in the determination of the median micelle size of the water in CO2 micro emulsions. Non-Visual Synthetic Methods As an alternative to visual detection, other physical properties can be monitored in order to find the occurrence of phase transitions. In applications where the volume of a variablevolume cell can be known accurately at any instant, the appearance of a new phase can be determined from the abrupt change in slop on the pressure-volume plot more accurately than by visual observation [33] (Figure 3). The experimental phase equilibrium set-up used by Kodama et al. [33] has the particularity of being equipped with two density meters in High-pressure phase equilibria for product and process design 17 recirculation loops, for measuring the density of different phases. It also employs a very simple system for the variation of the volume of the cell, in which a piston is manually and directly actuated, without the use of a pressure transmitter medium. This allows the exact position of the piston to be determined with higher precision, and consequently the total volume of the cell, at any instant during the experiment. 2.2. Synthetic methods without phase transition The application of synthetic methods without phase transition is based on the measurements of a number of equilibrium properties, such as, pressure, temperature, phase volumes and densities, which are subsequently used in calculations involving material balances in order to determine the phase compositions [26]. These methods have been often used in the determination of the solubilities of gases in condensed phases according to what is shown in Figure 5. Figure (a) represents the experimental apparatus. Figure (b) shows the pressure change with time during the experiment, which includes several additions of the light compound. Figure (c) shows a pressure composition diagram. Figure 5 – Experimental procedure, synthetic isothermal method [2]. At the beginning of the experiment, the equilibrium cell is charged with a precisely known amount of the first component, normally the less volatile, and then evacuated, before High-pressure phase equilibria for product and process design 18 applying to the system the desired temperature. Then, an exactly known amount of the more volatile component is added to the cell with the help of a syringe pump (Figure 5a), leading to an increase of the pressure in the system. As the gas dissolves into the liquid phase, the pressure inside the equilibrium cell decreases, eventually reaching an equilibrium value (Figure 5b). For this reason, this method is also called “pressure-decay method” in the literature. After the equilibrium is achieved, the pressure and temperature are registered, as well as the volumes of both phases when available, and the composition of the vapor phase is calculated with a phase equilibrium model. The composition of the liquid phase is then calculated by material balance from the known total composition, the composition of the vapor phase and the phase densities and volumes [34], yielding a point in the boiling point curve of the phase diagram. Successive additions are performed, repeating the procedure, and allowing the determination of more points in the boiling point line (Figure 5b and 5c). This method is particularly practical in cases where the high-boiling compound has a negligible vapor pressure, and the composition of the gas phase can be assumed as being the pure light-boiling compound. High-pressure phase equilibria for product and process design 25 The main requirement of the system is to be able to perturb the interface between gas and liquid. A magnetic stirring system was specially designed for this apparatus, consisting of two parallel magnets, aligned and placed on opposite sides of the cell, oscillating vertically at low speed. The generated magnetic field will promote the vertical movement of a magnetic sphere placed inside the equilibrium cell, moving from the upper surface of the piston to the very top of the inner cavity of the equilibrium cell. The stirring is operated by an electrical motor, whose movement is simply adapted to the oscillatory pattern needed for this application. High-pressure phase equilibria for product and process design 26 4 New equilibrium cells - Synthetic isothermal method without phase change The laboratory of Property Data and Thermodynamics of BTS has several equilibrium cells for VLE studies based on the synthetic isothermal method. Most of the high-pressure cells available are made of stainless steel, for economical and safety reasons. Given the already mentioned advantages of view-cells, it was decided to develop two new equilibrium cells, inspired in the view-cells existent in the laboratory, but which have limitations concerning the pressure, as they are made of glass. One of this glass cells is shown in Figure 11, being basically constituted by a glass cylinder between two stainless steel flanges. Figure 11 - Existing glass cell for phase equilibria measurements, synthetic visual method. With the glass cell as starting point, the glass was exchanged for sapphire, of higher resistance, with a thickness dimensioned to support pressures of 10 MPa in its interior. Similar to the already described equilibrium cell, the sapphire was acquired from GoodFellow Cambridge Ltd., UK; with a wall thickness of 150 mm and a surface with a polishing better than 0.1µm Ra. High-pressure phase equilibria for product and process design 27 The two stainless steel segments were also redesigned, in order to cope with the higher pressures, and to promote a better contact with the sapphire, facilitating the sealing, through the use of O-rings. Figure 12 shows the body of the cell. As before, the 3D files were generated in SolidWorks 3D CAD Design Software. Figure 12 - 3D computer generated images of the new high-pressure equilibrium cells. Left: overview of the cell. Right: cut of the cell showing the interior. High-pressure phase equilibria for product and process design 28 5 Solubility of carbon dioxide and propylene oxide in polymers derived from carbon dioxide Crude oil is a raw material widely used in the chemical industry as source of carbon. More than half a million tons of crude oil are processed for the production of polymers every day [36]. Ideally, carbon dioxide could be integrated into the molecular structure of polymers, reducing the consumption of crude oil to produce these indispensable materials. This would lower the dependence of the chemical industry on this fossil raw material, making it also less susceptible to oscillations in the oil prices, which present a long term tendency to increase. For a long time, scientists considered the use of carbon dioxide (CO2) as raw material in the production of polymers, but the reaction of CO2 revealed to be extremely unfavorable from an energetic point of view. However, in combination with “high-energy” monomers like epoxides, the energy balance of the CO2 reaction could be turned into a favorable direction. Additionally, the development of a new appropriate catalyst was crucial in making this copolymerization possible [37]. Scientists at Bayer managed to successfully bring to reality this “Dream Reaction”. In essence, CO2, propylene oxide (PO) and a starter are brought together. As result, polyether polycarbonate polyols (PPP) are formed, which can serve as raw materials for the production of polyurethanes, as depicted graphically in Figure 13. As part of the optimization of this innovative process, where polyurethanes are produced using this pollutant as a raw material, there was a need to know the precise phase equilibrium in the reaction mixture, namely the solubilities of CO2 and propylene oxide (PO) in polyether polycarbonate polyols (PPP), an intermediary in the production of polyurethanes. Measurements were performed at pressures up to 10 MPa at temperatures of 100°C and 120°C, using the synthetic isothermal method. Propylene carbonate (PC) is also present in the process, and due to this, experimental studies of the solubility of CO2 and PO in mixtures of PPP and PC at 100°C were also performed. The solubility of nitrogen (N2) in PPP was studied as well, as required for safety analysis of the process, as when the polymer is introduced in the reactor it contains air. High-pressure phase equilibria for product and process design 29 Figure 13 - CO2 as raw material for the production of polyurethanes [37]. In 2012, the first experimental results of this project were publish by Fonseca et al. [37], but further measurements are needed, since the process is being developed and there is a permanent evolution in the reaction process, with new polymers with improved characteristics and different properties being prepared. 5.1. Experimental procedure The experiments involving PO and CO2 were performed in an apparatus equipped with a highpressure view-cell and which is show in Figure 14. This cell enables the visual determination of the volumes of both phases for every equilibrium stage. The high-pressure cell is made of Hastelloy C4, with sapphire windows, and has an internal volume of approximately 250 cm3, being suitable for pressures up to 35 MPa. A fast equilibration is promoted by a specially designed magnetically driven stirrer. Four electromagnets are positioned on the circumference of the cell, promoting a rotating magnetic field which drives the magnetic stirring rod inside the cell [38]. An adequate circulation of thermal fluid around the cell allows a uniform temperature distribution, which is measured through calibrated Pt100 High-pressure phase equilibria for product and process design 30 resistance thermometers. The pressure is measured with a precision pressure transducer, directly coupled to the cell and calibrated before each experiment. A more detailed description of this apparatus can be found in the literature published by Peper et al [39]. Figure 14 – High-pressure equilibrium view-cell where the experiments for the study of solubility of CO2 and PO in the polymer were performed [39]. The experimental procedure used corresponds to the synthetic isothermal method, also known as “pressure decay method” already presented. A precisely determined amount of the polymer is introduced into a previously evacuated equilibrium cell. Using a high-pressure syringe pump at constant temperature, a small, well known amount of the volatile component (C02 or PO) is added, leading to a pressure increase in the cell [2]. For every addition, the difference in the volume of the syringe is recorded, and the amount of compound injected in the cell is calculated from its density at the temperature of the syringe. Contrary to N2 or to propylene oxide, CO2 was kept at low temperature in the syringe pump, further away from its critical point, in order to reduce the uncertainty in the amounts injected. Closer to its critical point, the density of the liquid carbon dioxide varies significantly with temperature, and High-pressure phase equilibria for product and process design 31 oscillations in the temperature of the syringe pump would lead to a larger error in the calculation of the amount added to the cell. The density of gaseous PO, at pressure and temperature conditions of the equilibrium measurements, were calculated using the SRK equation of state, which at the moderate pressures in question usually provides results with an accuracy between 1% and 2%. The densities of CO2 were calculated through a reference equation of state [40]. After adding the gas to the cell, the pressure inside the cell decreases as the gas dissolves into the polymer, eventually reaching an equilibrium value. Once equilibrium is achieved, the solubility of the volatile component in the liquid phase is calculated through a mass balance, considering the equilibrium pressure, the total composition in the cell, and the volumes of the two phases. Thereafter, a new addition of gas is done and the process must be repeated until reached the desired gas concentration. Due to the low volatility of PPP and propylene carbonate, the vapor phase is considered to consist of pure gas, CO2 or PO. Using equations of state, the density of the gases at equilibrium pressures can be determined. Knowing the volume of the gas phase, it is possible to calculate the amount of gas in the gas phase, and by difference the amount of gas dissolved in the polymer. In the measurements in question, where a high solubility is expected, due to the nature of the compounds involved, the use of a view-cell becomes very important. High solubility can result in variations on volume of the condensed phase, which will influence the material balance used in the calculation of the solubility, which is based on the volume and density of the gas phase. In the study of the solubility of CO2 for example, the volume of the condensed phase increased on average 1.5% per every 1 wt% of CO2 dissolved in the polymer. Some experiments were repeated in order to verify the repeatability of the method and procedure. A very good agreement among the results was found. The measurements with N2 were performed in the phase equilibrium apparatus represented schematically in Figure 15. The stainless steel equilibrium cell, displayed in the figure, has an internal volume of 312 cm3 and it can be used for pressures up to 6 MPa. A Pt100 resistance thermometer and a high-precision pressure transducer were used in the apparatus, calibrated before each experiment for a specific temperature and for the entire relevant pressure range using a pressure balance calibrated in regular intervals. The equilibrium cell is placed in a High-pressure phase equilibria for product and process design 32 thermal bath, with a temperature stability of ±0.01K. The pressure transducer and the Pt100 are connected to a computer in order to monitor and record the data. Figure 15 - Schematic representation of the apparatus used for the solubility measurements of N2 [37]. 5.2. Experimental Results The solubility of the CO2 was measured in two different types of pure PPP at 373 K and 393 K, at pressures up to 10 MPa. These results are given in Figure 16 and Figure 17, as well on Table 1 and Table 2 available on Appendix 1. Comparing both polymers is it possible to observe that CO2 has a higher solubility in polymer type 1 and this was predictable since this polymer has a lower molecular weight. The solubility of CO2 in pure PPP (when expressed in weight fraction) increases almost linearly with the pressure in entire pressure range investigated and it decreases with temperature. For example, considering the polymer type1, at 373 K and pressure 5 MPa, is it possible to observe that the solubility is approximately 4.7 wt% while at 393 K is approximately 6 wt%. At the same conditions for the polymer type 2, at 373 K and 5 MPa, the solubility of CO2 in polymer is approximately 5.6 wt% while at 393 K is 6.9wt %. High-pressure phase equilibria for product and process design 33 Figure 16 - Solubility of CO2 in PPP type1 at two different temperatures. Figure 17 - Solubility of CO2 in PPP type2 at two different temperatures. 00 02 04 06 08 10 00 02 04 06 08 10 p / MPa xCO2 / wt % 373 K 393 K 00 02 04 06 08 10 00 02 04 06 08 10 p / MPa xCO2 / wt% 373 K 393 K High-pressure phase equilibria for product and process design 34 In order to study the influence of propylene carbonate on the gas solubility, a measurement was also performed, at 373 K, using a mixture of cPC:PPP, with a mass ratio of 1:9 (initial concentration of 10 wt% of propylene carbonate in the polymer). The results are shown in Figure 18. Figure 18 - Solubility of CO2 in a mixture of PPP (type1) and propylene carbonate at 373 K. The solubility of PO in pure PPP is significantly higher than that of CO2 under the same conditions. This effect was predictable due to the nature of the gas and since PO has a higher critical temperature than CO2. The results are perfectly in agreement with data already published in previous year [37] and this is shown in Figure 19. As for the results of the solubility of PO in mixtures of PPP and propylene carbonate, the presence of propylene carbonate leads to an increase of the solubility of the gas, this time more remarkable than the increase observed for CO2. The results are presented in the Figure 20 as well on Table 4. 00 02 04 06 08 10 00 02 04 06 08 10 p / MPa xCO2 / wt% Pure PPP cPC:PPP=1:9 High-pressure phase equilibria for product and process design 41 [29] A. Shariati and C. J. Peters, J. Supercrit. Fluids, 23 (2002) 195-208. [30] "Thermodynamic Laboratory TU Delft," [Online]. Available: http://homepage.tudelft.nl/v9k6y/ETh/Website/thermolab.html. [Accessed Março 2013]. [31] M. D. Jager and E. D. J. SLoan, FLuid Phase Equilib., 185 (2001) 88-89. [32] X. Dong, C. erkey, H. J. Dai, H. C. Li, H. D. Cochran and J. S. Lin, Ind. Eng. Chem. Res., 41 (2002) 1038-1042. [33] D. Kodama, J. Miyazaki, M. Kato and T. Sako, Fluid Phase Equilib, 219 (2004) 19-23. [34] K. Fischer and M. Wilken, "Experimental determination of oxygen and nitrogen solubility in organic solvents up to 10 MPa at temperatures between 298 K and 398 K.," J. Chem. Thermodyn.,33 (2001) 285-1308. [35] Opsens, "White-light polarization Interferometry technology". [36] T. Solutions, "The raw material carbon dioxide," (2010). [37] J. M. S. Fonseca, R. Dohrn, A. Wolf and R. Bachmann, "The solubility of carbon dioxide and propylene oxide in polymers derived from carbon dioxide," Fluid Phase Equilibria, 318 (2012) 83-88. [38] C. Braden and R. Dohrn, "Magnetrührer im Hochdruckreaktor, Patent LE A 31 453 , Bayer AG," 1996. [39] H. V. D. R. Peper S., "Measurement of phase equilibria of the systems CO2 + styrene and CO2 + vinyl acetate using different experimental methods," J. Supercrit. Fluids, 55 (2010) 537-544. [40] R. W. W. Span, "A new equation of state for carbon dioxide covering the fluid region from the triple-point temperature to 1100 K at pressures up to 800 MPa.," J. Phys. Chem. Ref. Data,25 (1996) 1509-1596. High-pressure phase equilibria for product and process design 42 Appendix 1 Table 1 - Solubility of CO2 in PPP type1 at two different temperatures. Pure PPP type1 373 K 393 K XCO2 / (wt %) p / (MPa) XCO2 / (wt %) p / (MPa) 0 0.000 0.000 0.000 1.317 0.932 3.121 3.537 3.196 2.940 4.693 5.482 5.207 4.763 5.670 6.009 6.321 5.752 5.935 6.559 9.173 6.940 6.472 7.004 Table 2 - Solubility of CO2 in PPP type2 at two different temperatures. Pure PPP type2 373 K 393 K XCO2 / (wt %) p / (MPa) XCO2 / (wt %) p / (MPa) 0.000 0.000 0.000 0.000 1.377 1.234 1.354 2.396 2.035 1.952 1.767 4.275 3.002 3.177 2.750 5.258 4.279 4.573 3.783 5.966 5.203 5.579 4.027 6.465 4.600 6.887 High-pressure phase equilibria for product and process design 43 Table 3 - Solubility of CO2 in a mixture of PPP and propylene carbonate at 373 K. PPP+10%cPC, Ratio 1:9 373 K XCO2 / (wt %) p / (MPa) 0 0 1.290 1.111 2.375 1.949 2.921 2.4697 5.431 4.3858 6.710 5.3984 9.349 6.4078 9.818 6.7524 High-pressure phase equilibria for product and process design 44 Table 4 - Solubility of PO in PPP at two different temperatures. Pure PPP type1 373 K 393 K XPo / (wt %) p (MPa) XPO / (wt %) p / (MPa) 0 0.000 0.000 0.000 4.092 0.111 2.567 0.107 6.395 0.174 5.301 0.229 8.484 0.229 8.040 0.340 10.410 0.284 10.257 0.420 12.396 0.319 11.743 0.470 14.347 0.357 13.459 0.521 16.219 0.390 15.025 0.565 18.017 0.418 16.574 0.607 19.869 0.447 High-pressure phase equilibria for product and process design 45 Table 5 - Solubility of PO in a mixture of PPP and propylene carbonate at 373 K, 2 trials. PPP+10%cPC, Ratio 1:9 373 K Trial 1 Trial 2 XPO / (wt %) p / (MPa) XPO / (wt %) p / (MPa) 0.000 0.000 0.000 0.000 1.336 0.050 0.070 2.273 2.524 0.090 0.126 4.726 6.655 0.160 0.182 6.965 7.723 0.211 0.233 9.123 9.195 0.244 0.279 11.325 10.896 0.279 0.316 13.289 0.370 15.721 0.382 17.397 0.416 19.772 High-pressure phase equilibria for product and process design 46 Table 6 - Solubility of N2 in PPP at room temperature, 298 K. Pure PPP type1 298 K XN2 / (wt %) p / (MPa) 0.0000 0.000 0.0033 0.030 0.0073 0.069 0.0104 0.115 0.0122 0.147 0.0241 0.330 0.0328 0.440 0.0349 0.508