Experimental study of high pressure phase equilibrium of (CO2 + NO2/N2O4) mixtures
Highlights
► Experimental bubble pressure and liquid density of (CO2 + NO2/N2O4) mixtures at T ranging from (298 to 328.45) K are reported. ► A high pressure variable volume equilibrium cell with the static method have been used. ► (Liquid + vapour) equilibrium was found to be well predicted using Peng–Robinson equation of state. ► Calculation of liquid density values was unsatisfactory with this approach.
Introduction
Supercritical carbon dioxide is now commonly recognized as a very promising compound to be used as a green solvent for chemical reactions as replacement for polluting organic solvents. This is due not only to the interesting physical properties of supercritical compounds, but also to the chemical inertness, low cost, non-toxicity of CO2 and the fact that this compound can be easily recycled. Although a weak solvent for polar interest molecules, scCO2 has been used as a solvent in a wide range of chemical reactions such as hydrogenations, hydroformylations, oxidations, or polymerizations. The use of scCO2 for chemical synthesis has been extensively reviewed by Beckman [1]. Because it is completely miscible with gases such as O2, CO, or H2, high pressure CO2 is very useful to enhance solubility of reactants into liquid phases or even to alleviate liquid–gas mass transfer limitations by solubilizing all the reactants to give a single-phase system. Moreover, carbon dioxide being the result of complete oxidation of organic compounds, it cannot be oxidized, and thus it is the ideal solvent for oxidation reactions. For this reason, CO2 has often been used as a solvent for oxidation reactions, employing mainly oxygen as the oxidant. Recently, another example of oxidation reaction in high pressure CO2, in which nitrogen dioxide is the oxidant, has been described [2] and the process of oxidation patented [3]. This involves oxidation of polysaccharides, and more specifically cellulose, resulting in oxidized cellulose, a very attractive material for bio-medical applications. Indeed, when cellulose is partially oxidized it becomes degradable in the human body (a property termed “bioresorbability”) and in addition possesses haemostatic properties (i.e. it halts bleeding), which can be advantageous in biomedical devices, like surgical compresses for instance. The suitable oxidant for preparing this material with both high carboxyl content and targeted physical properties is nitrogen dioxide (NO2). This compound ensures selective oxidation of the primary hydroxyl groups of cellulose, leading to partially oxidized cellulose, with the already mentioned properties.
As an alternative to the present use of traditional halogenated solvents in the cellulose oxidation process [4], [5], [6], supercritical carbon dioxide (scCO2) has been shown to be an attractive solvent to perform this oxidation [2]. Its major advantage in this case lies in its complete inertness regarding the oxidant, so, preventing a possible degradation of the solvent. Moreover, it ensures the biocompatibility of the processed material, because the latter is free from any solvent residue. Finally, because nitrogen dioxide is soluble in high pressure CO2, it allows operation with a significant concentration of this reactant in the solvent. In a previous paper [2], efficiency of the cellulose oxidation was shown to depend on operating parameters such as pressure, temperature, moisture content, and CO2 vs NO2 ratio. Moreover, in this process, the knowledge of the number of phases and their composition is also a key parameter to obtain an oxidized product with tightly specified properties and for the development of the process on an industrial scale.
In the oxidation process, the reacting mixture, composed of the oxidant NO2 and the solvent CO2, is characterized by the existence of a chemical equilibrium between nitrogen dioxide and its dimer, nitrogen tetroxide (N2O4), as follows:Below T = 262.15 K, the mixture is completely dimerised (N2O4), and this dimer dissociates as temperature increases, the proportion of each compound depending on conditions of temperature and pressure, the equilibrium being governed by the mass action law, with an equilibrium constant K depending only on the temperature. Under atmospheric pressure, the mixture boils at T = 294.25 K, the liquid being yellowish brown and the vapour reddish brown. If moisture is present, the mixture decomposes readily into nitrous and nitric acid, and becomes very aggressive to numerous metals.
Physical properties of the NO2/N2O4 system, and values of the equilibrium constant, with respect to the temperature, have been the subject of several experimental and theoretical studies. For example, Reamer and Sage [7] have performed density measurements in the liquid–vapour coexistence region. Values of equilibrium constants in the vapour phase as a function of temperature can be found in works by Verhoek and Daniels [8] or Chao et al. [9]. These results describe a vapour phase containing mostly dissociated NO2 (around 90 mol% of NO2 at T = 373.15 K and a complete dissociation at 413.15 K). James and Marshall [10] have measured equilibrium constants in liquid phase and have shown that nitrogen dioxide is strongly associated in that physical state. Redmond and Wayland [11] have measured equilibrium constant data for nitrogen dioxide dissolved in some organic solvents.
Some authors studied the reaction of oxidation with NO2 of cellulose in different organic solvents [12] and showed that the degree of dissociation of NO2 increases in non-polar solvents, leading to an increase in the degree of oxidation of cellulose. So, although it is not yet clearly demonstrated, it is very probable that only the NO2 molecule, i.e., the monomer form, is the active oxidant molecule. Therefore, knowledge of the distribution between monomeric and dimeric species, especially when the oxidant NO2 is solubilized in a solvent, appears to be an important parameter of the reaction.
This work reports experimental data of high-pressure equilibrium between CO2 and NO2 in conditions close to operating conditions of the cellulose oxidation process, which have already been described in the literature [2]. As a result of a collaborative study, an attempt to model the thermodynamic behaviour of this mixture has been published by Belkadi et al. [13], who used the crossover soft-SAFT equation of state to predict (vapour + liquid) equilibrium of this mixture under pressure. The soft-SAFT equation of state is a modification of the original Statistical Associating Fluid Theory (SAFT) molecular-based equation of state, which is in nature able to describe the thermodynamic behaviour of associating compounds [14]. Belkadi et al. have shown that predictions matched correctly experimental data of the (CO2 + NO2/N2O4) mixture provided that a binary interaction coefficient fits the experimental results. Although this kind of equation of state is a powerful tool to predict the thermodynamic behaviour of such a system, soft-SAFT equation is not implemented in most commercial thermodynamics software, and thus, from now, they cannot be easily used to compute fluid phase equilibrium of this mixture. Similarly, Bourasseau et al. [15] used the experimental results described in the present study, to show that (vapour + liquid) equilibria of the (CO2 + NO2/N2O4) mixture could be predicted using a Monte-Carlo molecular simulation approach. Their interest in this mixture is justified by the necessity to predict atmospheric (CO2 + contaminants) mixtures. Interestingly, their simulations at T = 300 K predict a very low amount of non associated NO2 molecules in both phases, but they could not compare this result with experimental ones. However, Monte-Carlo simulation is not a common tool that can be routinely used to predict fluid phase equilibria.
In this paper, the experimental set-up and conditions in which (vapour + liquid) equilibria and volumetric properties of the (CO2 + NO2/N2O4) mixture have been obtained will be described, and an attempt will be made to describe a way to predict the experimental data of this mixture using the well-known Peng–Robinson equation of state, in order to use this model in the specific context of the industrial application of cellulose oxidation.
Section snippets
Materials
Liquid NO2 (water content max 0.5 wt.%) and high purity CO2 TP (N45 mass fraction purity 0.99999) were supplied by Air Liquide.
Specific hazards due to nitrogen dioxide handling
Nitrogen dioxide is a non-flammable and highly toxic gas (deadly poison). By consequence, a threshold limit value of 9 · 10−6 by volume, i.e., 9 mg · m−3, is recommended as the maximum concentration allowable in industrial premises [16]. For the purpose of this study, operators are protected thanks to a polypropylene transparent barrier guard placed around the experimental
Evaluation of measurement uncertainties
The experimental set-up previously described allows determination of transition pressure, i.e. bubble or dew pressure, for a fixed composition and temperature of the mixture. This is a very specific system and it has been found useful to assess the composition, pressure and temperature uncertainties associated with such experimental devices.
Measurements are taken after isothermal conditions have been reached, temperature being measured by a thermocouple whose accuracy is ±0.1 K.
In a first
Modelling and discussion
The NO2 and N2O4 form a reacting binary system with only one degree of freedom at (liquid + vapour) equilibrium. That is why in the (P, T, x) coordinates, the two-phase region of this reacting system is a curve which ends at the critical point of these molecules. Thus, the critical coordinates Tc, Pc and the acentric factor ω are the same for NO2 and N2O4. The fugacities calculated with a cubic equation of state are the same for NO2 and N2O4 too. To calculate the equilibrium of such a ternary
Conclusions
This work has provided experimental data about the high pressure (CO2 + NO2/N2O4) phase equilibria. Although this first approach for modelling the liquid–vapour experimental study did not allow a complete description of the actual behaviour of the system, i.e., the dissociation degree of N2O4 was not measured and modelled, the modelling, using a simple Peng–Robinson equation of state with kij = 0 was shown, nevertheless, to be very suitable for predicting the physical state of the (CO2 + NO2/N2O4)
References (20)
J. Supercrit. Fluids
(2004)- et al.
J. Supercrit. Fluids
(2009) - et al.
Thermochim. Acta
(1974) - et al.
Fluid Phase Equilib.
(2008) - et al.
J. Chem. Thermodyn.
(1989) - M. Vignon, S. Montanari, D. Samain, J.S. Condoret, WO 2006/018552,...
- W.O. Kenyon, E.C. Yackel, US 2 448 892,...
- W.H. Ashton, C.E. Moser, US 3 364 200,...
- F. Boardman, L. Saferstein, EP 0492990 A1,...
- et al.
Ind. Eng. Chem.
(1952)
Cited by (17)
Scaling towards the critical point in the combined reaction/Gibbs ensemble
2024, Fluid Phase EquilibriaThe impressive impact of including enthalpy and heat capacity of mixing data when parameterising equations of state. Application to the development of the E-PPR78 (Enhanced-Predictive-Peng-Robinson-78) model.
2022, Fluid Phase EquilibriaCitation Excerpt :Table S1 in the Supporting Information presents the list of the 204 pure components involved in the development of the E-PPR78 model. For each binary system considered in the development of the E-PPR78 model, Table S2 contains the main features of collected experimental data, extracted from references [38–2068]. Bibliographic references are reported in Table S3.
On the phase behaviour of the CO<inf>2</inf> + N<inf>2</inf>O<inf>4</inf> system at low temperatures
2022, Chemical Engineering ScienceCitation Excerpt :The experimental and modelling results for each different isothermal condition are plotted in Fig. 6. The only available literature data for comparison are from Camy et al. (2011). They have studied this binary system using a similar experimental technique.
Thermodynamic challenges for CO<inf>2</inf> pipelines design: A critical review on the effects of impurities, water content, and low temperature
2022, International Journal of Greenhouse Gas ControlCitation Excerpt :Moreover, for the transient transport processes, other properties, including the speed of sound, Joule-Thomson effect (J-T), and diffusivity, should be considered (Tan et al., 2015). In literature, there are experimental studies on some typical interactions such as CO2/Ar, CO2/CO, CO2/H2O, CO2/SO2, CO2/N2O4, CO2/COS, CO2/NH3, and others including more than one substance in the addiction of the main component (CO2) (Al-Siyabi, 2013, M Nazeri et al., 2017, Tenorio et al., 2015, Nazeri et al., 2018, Chapoy et al., 2013, Ahmad and Gersen, 2014, Lasala et al., 2016, Løvseth et al., 2018, Coquelet et al., 2014, M Nazeri et al., 2017, Camy et al., 2011). The experimental datasets available in the literature, however, do not include all possible interactions.
Thermo-physical properties of CO<inf>2</inf> mixtures and their impacts on CO<inf>2</inf> capture, transport and storage: Progress since 2011
2019, Applied EnergyCitation Excerpt :Table 3 presents the progress in the experimental data on density. New data have been published for CO2/O2 [21,22], CO2/COS [23], CO2/CO [22,24,25], CO2/N2O4 [19], CO2/H2 [22,26,27], CO2/Ar/N2 [28,29] and CO2/N2/O2/Ar [30]. But some clear gaps still exist.
High-pressure fluid-phase equilibria: Trends, recent developments, and systems investigated (2009–2012)
2019, Fluid Phase Equilibria