Production of Hydrobromic Acid from Bromine and Methane for Hydrogen Production

A cooperative program between Sandia National Laboratories (SNL) and the SRT Group was recently begun to advance the development of a chemical process to produce hydrobromic acid from bromine and methane. The purpose of the SNL program is to provide fundamental reaction kinetics information that will constitute the basis for system engineering by SRT. The SNL activities consist of computational chemistry and laboratory experimentation. Preliminary calculations of chemical equilibrium behavior and reaction kinetics were performed using the CHEMKIN computer program and its associated thermodynamic and kinetic databases. Calculations were performed for reactions involving the bromine-methane and bromine-methanesteam systems. The equilibrium conversion of bromine to HBr in the two processes is excellent, especially at temperatures below 900K. If equilibrium is achieved in the bromine-methane process, then the other principal product is solid carbon in some form, e.g., coke or soot. If solid carbon is not produced, then the reactor effluent may contain significant amounts of brominated hydrocarbons. From a purely thermodynamic point of view, the bromine-methane-steam process appears to be more attractive because the carbon in methane is converted almost completely into carbon dioxide. Introduction A hydrogen production process that requires the formation of hydrobromic acid as one of the primary steps has been proposed by the SRT Group, Inc., Miami, FL. SRT has described the process and preliminary development studies in detail elsewhere, so only the basic features are given here [Schleif 1997, SRT Group 1998]. The initial step in the process is the formation of hydrobromic acid from bromine and a hydrocarbon, e.g., methane. This reaction is conducted at elevated temperature, as high as 800°C. An alternative reaction scheme includes steam among the reactants to convert the carbon in the feedstock to carbon dioxide rather than solid carbon. The HBr generated in the reactor is collected with water to produce a concentrated solution of HBr that is electrolyzed in a separate process step, yielding hydrogen and bromine. The latter constituent is recycled to the reactor to perpetuate the HBr production cycle. The hydrogen is available for generating energy by various means. Alternatively, both constituents may be used to produce electrical power via a regenerative hydrogen-bromine fuel cell of a proprietary design. A cooperative program between Sandia National Laboratories (SNL) and the SRT Group was recently begun to advance the development of the chemical process to produce hydrobromic acid. The purpose of the SNL participation is to determine fundamental reaction rate equations that will constitute the basis for reactor engineering by SRT. The SNL work consists of several activities. The primary SNL activity in the initial phase of this project is to employ computer modeling of chemical reaction kinetics to determine the reaction rate equations. Computation will be used to evaluate the equilibria of the various reactions studied in order to predict the extent of the desired reaction and the formation of by-products. Another activity is to conduct laboratory experiments to determine the fundamental characteristics of a number of chemical reactions that produce hydrobromic acid from hydrocarbon precursors. Calculations of both chemical equilibrium behavior and reaction kinetics are required to interpret the laboratory reactor data. These calculations will be performed using the CHEMKIN computer program and its associated thermodynamic and kinetic databases. Additional information may need to be input to complete the database for computations. The results of the laboratory experiments will be compared with the predictions of the computations to refine the reaction rate equations. The equilibrium calculations will enable the experimental apparatus to be designed with full regard for all of the reaction products that may be formed. Such calculations also provide useful information regarding the energetics of the reactions which is useful for engineering design. This report summarizes the status of the Sandia National Laboratories segment of this project during the startup period of February 2001 through April 2001. The planned future work is also discussed. Computational Chemistry of Bromine-Methane Reactions This section summarizes the calculations that have been carried out to determine the feasibility of producing hydrogen bromide from bromine, methane, and (optionally) steam. As is customary in such investigations, the process is first examined from a thermodynamic point of view to determine if it is viable. If it is, then kinetic simulations are carried out to estimate whether the time scales are reasonable. It must be emphasized that the results presented here are tentative because the chemical mechanisms are almost certainly incomplete, some of the data are of uncertain accuracy, and a detailed parameter study has not been performed at this stage of the project. The results presented here are useful mainly as a guide for companion experiments that are to be conducted at SNL in the upcoming months. The results of these experiments should allow the models to be refined to the point that they can be used with confidence in detailed reactor simulations. Data Collection Two primary sources were used to assemble the gas-phase mechanism and data needed for the calculations. The species and reactions involved in methane pyrolysis and combustion were taken from GRI-Mech Version 1.2, and the corresponding thermodynamic data file was used as the default. Reactions and thermodynamic data for bromine-containing species were obtained from a NIST Web site [NIST 2001]. Several reactions tabulated by Babushok [1996] but not included in the NIST website (perhaps because of uncertain accuracy) were added to our compilation. About a dozen miscellaneous reactions from various sources were also added, partly to provide a pathway for complete bromination of methane. Thermodynamic data for all bromomethanes and bromomethyl radicals were obtained by fitting the values reported by Paddison and Tschuikow-Roux [1998] to complement the NIST database. An approximate entry for carbonyl bromide (COBr2) was constructed by using handbook values of the enthalpy, entropy, and heat capacity at 298K [Dean 1973]. This value was used only for the equilibrium calculations, as a means of demonstrating that this species would not be present in significant amounts. Thus, a more accurate entry was not required. Equilibrium Calculations The first step is to perform a purely thermodynamic analysis to determine the species present in the system at equilibrium. The results are a function of the prevailing temperature and pressure, which is taken to be 1 atm. The computations have been carried out using the Sandia EQUIL code, which is basically a CHEMKIN interface to the well-known STANJAN code. The results have been checked against a spreadsheet-based calculation at 298K and the predictions of the online solver EQUILIB-Web [Ecole Polytechnique Montreal 2001]. Both STANJAN and EQUILIB-Web have certain advantages: the latter automatically, and almost effortlessly, gives the global equilibrium composition by including every species in its database, while the former allows one to perform partial equilibrium calculations by including only those species that are of interest. We have used STANJAN here in order to take advantage of its flexibility. For the system in which bromine and methane are the reactants, the primary reaction is CH4 + 2 Br2 ! C(s) + 4 HBr (1) where C(s) is solid carbon (graphite). This stoichiometry was input to STANJAN, and the results are shown in Figure 1. Note that the ordinate is the overall mole fraction, on a logarithmic scale, with both gaseous and solid constituents included. Clearly, the conversion of methane to carbon is essentially complete; the formation of HBr is also nearly complete at low temperatures, but at higher temperatures HBr begins to dissociate into H2 and Br2, and eventually Br2 itself begins to dissociate. No brominated hydrocarbons are formed under these circumstances. 1 0 4 1 0 3 1 0 2 1 0 1 1 0 0 3 0 0 5 0 0 7 0 0 9 0 0 1 1 0 0 1 3 0 0 1 5 0 0 M o le f ra c ti o n