J. Phys. Chem. A 2010, 114, 4531–4534
4531
Studies of the Kinetics of Ammonia Decomposition on Promoted Nanocrystalline Iron Using Gas Phases of Different Nitriding Degree Karolina Kiełbasa,* Rafał Pelka, and Walerian Arabczyk Szczecin UniVersity of Technology, Institute of Chemical and EnVironment Engineering, Pułaskiego 10, 70-322 Szczecin, Poland ReceiVed: October 16, 2009; ReVised Manuscript ReceiVed: January 28, 2010
Promoted nanocrystalline iron was nitrided in a differential reactor equipped with systems that made it possible to conduct both thermogravimetric measurements and hydrogen concentration analyses in the reacting gas mixture. The nitriding process, particularly catalytic ammonia decomposition reaction, was investigated under an atmosphere of ammonia-hydrogen mixtures, under atmospheric pressure. Ammonia concentrations, and so nitriding potentials, were changed gradually from 0 to 100% at the inlet of reactor. The temperature was changed in the range of 475-500 °C. While values of nitriding potential were increasing, the rate of catalytic ammonia decomposition on R-Fe(N) was increasing too, but on mixture of R-Fe(N) with γ′-Fe4N nitride the rate was decreasing. The obtained results were interpreted on the basis of the adsorption range model. New equations describing the catalytic ammonia decomposition reaction rate as a function of the logarithm of the nitriding potential of the gas phase, temperature, and nitriding degree of solid samples were proposed. Introduction Catalytic ammonia decomposition in the industrial practice, viz., industrial scale production of a hydrogen and nitrogen mixture or removing of pollutants from exhaust gases, is realized in the presence of nickel catalyst (Ni/Al2O3).1 This reaction also takes place during the nitriding process of nanocrystalline iron. The ammonia decomposition reaction is of great importance in the chemical industry; thus new substances were sought that might be more active in the decomposition reaction. Various materials were investigated, e.g., polycrystalline platinum,2 nickel wire,3 bulk iron,4 iron thin films,5 and industrial catalysts.6 The rate of the catalytic ammonia decomposition reaction over iron depends on many factors such as, for example, composition and purity of the gas phase, a specific surface area, and composition of catalyst on which the reaction proceeds. The influence of water vapor in the gas phase7 and sulfur adsorbed on solid surface8,9 on the rate of ammonia decomposition process was analyzed. The catalytic ammonia decomposition reaction unfolds with different rates depending on the solid-state phase present in the nitriding process.10 During the process in question, ammonia diffuses from the gas phase to the surface of the catalyst. Then ammonia adsorbs on the surface of catalyst’s iron crystallites and then ammonia dissociation to atomic hydrogen and nitrogen occurs (the dissociative adsorption process). Atoms of hydrogen form particles which then desorb. Nitrogen atoms may combine to form dinitrogen molecules, which then desorb or react with iron and create solid solution of nitrogen in iron or iron nitrides. Previous works show that the rate of catalytic ammonia decomposition reaction is a function of ammonia and hydrogen partial pressure.11,12 One of the most popular equations describing the rate of the catalytic ammonia decomposition reaction is the Temkin-Pyzhev equation13,14 * Corresponding author: tel, +48 91 449 47 30; fax, +48 91 449 46 86; e-mail,
[email protected] (Karolina Kiełbasa).
( )
r)k
pNH32 pH23
β
,
β ) 0.5
(1)
The subject of the present considerations is the analysis of the kinetics of the catalytic ammonia decomposition reaction on R-Fe and in the area where two phases exist (R-Fe(N) and γ′-nitride) as well as in the region of saturation of γ′-nitride. The nanocrystalline iron nitriding process which takes place in a differential tubular reactor with mixing in a gas phase was studied. Gas phase of different nitriding potential, P ) pNH3/pH23/2, was applied. Experimental Methods The kinetics of a catalytic ammonia decomposition reaction was investigated making use of a tubular differential reactor equipped with a gas phase composition analyzer and a system that made it possible to conduct thermogravimetric measurments (Figure 1). The samples of the gas phase were collected from sampling points which were located under and over the platinum sample holder. The hydrogen concentration was measured directly. There were no concentration gradients observed in the reaction volume of the reactor. Flow rate of the gas reactants was regulated making use of electronic mass flow controllers. The gas mixture load was constant and equaled 5.5 cm3 s-1 g-1. The mass and temperature of the analyzed sample as well as hydrogen concentrations were recorded digitally. Nanocrystalline iron, promoted with aluminum, calcium, and potassium oxides, was studied. The surface of iron was wetted by the oxides of promoters (3.3Al2O3, 2.8CaO, 0.65K2O in weighed fraction, determined by the inductively coupled plasma and atomic emission spectroscopy method). The structural promoters prevent the iron nanocrystallites from sintering; thus the material has an abiding average size of iron nanocrystallites of 20 nm. The analyzed material in the form of grains was sifted on sieves with 1.0-1.2 mm
10.1021/jp9099286 2010 American Chemical Society Published on Web 03/17/2010
4532
J. Phys. Chem. A, Vol. 114, No. 13, 2010
Kiełbasa et al.
Figure 1. Reactor with thermogravimetric measurement: 1, platinum sample holder with single layer of grains; 2, thermocouple; 3, reactor furnace; 4, quartz reactor; 5, electronic mass flow controllers.
diameter. Grains in this range were selected for analysis. In the platinum basket hanging on the arm of a thermobalance, a 1 g sample of grains was placed in a single layer. It was checked to ensure that the studied reactions were not influenced by platinum from sample holder. The applied reactor can be considered as the differential one, because of the process conditions and used substances. Nitriding process was preceded by a reduction of a passive layer from the analyzed solid samples. The reduction process was performed at 500 °C, under atmospheric pressure. The load of pure hydrogen was 2.5 cm3 s-1 g-1. The nitriding process was investigated in the temperature range of 475-500 °C, under atmospheric pressure. Ammoniahydrogen mixtures, containing various quantities of ammonia (in the range of 0-100%), were allowed into the reactor. Gas mixture composition was varying, but total gas load was constant. Inlet ratio of ammonia to hydrogen was changed when the stationary state was achieved (the promoted iron samples mass and gas phase composition were stable). The values of partial pressure of gas reactants were determined making use of material balance of the reactor and results of the hydrogen content measurements. On the basis of the inlet and outlet composition of the gas phase, a conversion degree of ammonia was calculated (RNH3)
RNH3 )
XH2F0 - FH20 FNH30(1.5 - XH2)
(2)
where F0 is the total molar flow rate of the inlet stream in mol · s-1, FH20 and FNH30 are hydrogen and ammonia molar flow rate in the inlet stream, respectively, in mol · s-1, and XH2 is the molar concentration of hydrogen in the reactor in mol · mol-1.
The rate of the catalytic ammonia decomposition reaction, rdecomp, was calculated with the equation
rdecomp ) RNH3FNH30
(3)
The changes in gas reactants concentrations and so the partial pressure changes resulted in changes of values of the nitriding potential defined as
P)
pNH3 pH23/2
(4)
Results In this paper, the nitriding degree of solid samples, RN [mol of N/mol of Fe] (g of N/g of Fe eventually), is defined as a ratio of moles of nitrogen to moles of iron in the nitrogen-iron system. Thermogravimetric measurements and changes of hydrogen concentrations in the gas phase as functions of time are shown in Figure 2. These are some exemplary results of experiments that were carried out when 100% of the ammonia was allowed into the reactor at 500 °C. Concentration of hydrogen (which left in the reactor after reduction process) was decreasing. During the first 300 s, the process was performed at varying gas phase compositions, which was a concequence of both the mixing process in a gas phase in the reactor and the chemical reaction, until the steady state is reached. After ca. 300 s, changes of the gas phase composition were no longer observed, which means a stationary state was achieved. In the stationary states, the nitriding reaction rate is zero at the stable mass of solid sample (XN ) constant; an equilibrium between
Studies of the Kinetics of Ammonia Decomposition
J. Phys. Chem. A, Vol. 114, No. 13, 2010 4533
Figure 2. The nitriding degree and hydrogen concentration changes versus time (100% ammonia at the reactor inlet, T ) 500 °C). Figure 5. The dependence of the nitriding degree of solid sample and the rate of ammonia decomposition reaction on the nitriding potential (ammonia-hydrogen mixtures at the reactor inlet, T ) 475 °C).
Figure 3. Changes of the nitriding degree of solid samples and concentrations of ammonia, hydrogen, and nitrogen at stationary states (ammonia-hydrogen mixtures at the reactor inlet, T ) 475 °C). Figure 6. The dependence of the nitriding degree of solid sample and the rate of ammonia decomposition reaction on the logarithm of nitriding potential (ammonia-hydrogen mixtures at the reactor inlet, T ) 475 °C).
Figure 4. The dependence of the nitriding degree of solid sample on the logarithm of nitriding potential (ammonia-hydrogen mixtures at the reactor inlet, T ) 475 °C).
gas and solid phases exists) and catalytic ammonia decomposition reaction runs with a constant rate (rdecomp ) constant). In Figure 3 the exemplary results concerning the nitriding degrees of solid sample that were observed in the stationary states during experiments performed at a temperature of 475 °C and under ammonia-hydrogen atmosphere of varying nitriding potential are presented. When values of hydrogen concentrations were lower than 0.43 mol mol-1, the nitriding degree of solid samples became to be constant and was equal to 0.248 mol of N/ mol of Fe, which approximately corresponds to the stoichiometric composition of γ′-Fe4N phase. On the basis of the material balance and the hydrogen concentration measurement results, the concentrations of the other components were calculated. The experimental values are plotted with solid lines, and calculated values are shown with dashed lines. From data on partial pressures of gas reactants, the nitriding potential values were calculated (eq 4). The nitriding potential was in the range from 2.9 × 10-5 to 0.06 Pa-0.5. In Figure 4 the dependence of the nitriding degree of solid sample on the logarithm of the nitriding potential, for a sample which was
nitrided at 475 °C, is presented. The nitriding degree is a function of the nitriding potential (the gas phase composition influences on the nitriding degree of the solid samples). For ln P < -6.3 the observed nitriding degree is relatively low, which corresponds to the nitrogen dissolving and ammonia adsorption on the surface of iron. The nitriding reaction did not occur. At ln P ) -6.3, the nitriding reaction started up. This means the smallest crystallites reacted and underwent the phase transition (according to the adsorption range model). From this moment two phases were present in the solid sample (unreacted iron phase, bigger crystallines, and the new phase, γ′-nitride). When ln P > -4.9, then approximately stoichiometric composition of γ′ phase was observed and the saturating of the nitride by nitrogen started. Discussion According to the Temkin-Pyzhev equation, the rate of ammonia decomposition reaction increases along with increase of the nitriding potential. Such dependence is valid only for some part of the conducted experiment (Figure 5). The inset of Figure 5 shows the dependence of the nitriding degree of solid sample and the rate of ammonia decomposition reaction on the nitriding potential in the nitriding potential range of 0-1 Pa-1/2. It can be observed on this figure that there are two different phases, R-Fe(N) and R-Fe(N) + γ′-Fe4N with various reaction rates. That implied the new equation should be proposed to describe the reaction rate. From the plot in Figure 6, it can be seen that the ammonia decomposition reaction rate linearly depends on the logarithm of the nitriding potential. When values of the nitriding potential
4534
J. Phys. Chem. A, Vol. 114, No. 13, 2010
Kiełbasa et al.
were increasing, the rate of catalytic ammonia decomposition on R-Fe(N) phase was increasing as well. However, on mixture of R-Fe(N) and γ′-nitride the rate of decomposition reaction was decreasing along with increase of the nitriding potential. On the basis of Figure 6, new equations reflecting the logarithmic dependence were proposed. These equations describe the performed experiments in a better way than the Temkin-Pyzhev equation. Thus the rate of the ammonia decomposition reaction over the R phase, as a function of the logarithm of nitriding potential, can be described by the expression
and on the logarithm of the nitriding potential. When values of nitriding potential were increasing, the rate of catalytic ammonia decomposition on R-Fe(N) was increasing as well. On mixture of R-Fe(N) and γ′-nitride, the rate of decomposition reaction was decreasing along with increase of the nitriding potential. New equations describing the dependence of the decomposition rate on logarithm of nitriding potential were proposed.
rdecomp ) 3.5 × 10-5 + 2.4 × 10-6 ln P
(1) Catalytic Ammonia Synthesis; Jennings, J. R., Ed.; Plenum Press: New York, 1991. (2) Lo¨ffler, D. G.; Schmidt, L. D. J. Catal. 1976, 41, 440. (3) Tsai, W.; Vajo, J. J.; Weinberg, W. H. J. Phys. Chem. 1985, 89, 4926. (4) Inokuti, Y.; Nishida, N.; Ohashi, N. Met. Trans A 1975, 6A, 773. (5) Logan, S. R.; Moss, R. L.; Kemball, C. Trans Faraday Soc. 1958, 54, 922. (6) Love, K. S.; Emmett, P. H. J. Am. Chem. Soc. 1941, 63, 3297. (7) Arabczyk, W.; Zamłynny, J.; Moszyn´ski, D.; Kałucki, K. Pol. J. Chem. 2005, 79, 1495. (8) Arabczyk, W.; Zamłynny, J.; Moszyn´ski, D. Pol. J. Chem. 2006, 80, 345. (9) Arabczyk, W.; Moszyn´ski, D.; Narkiewicz, U.; Pelka, R.; Podsiadły, M. Catal. Today 2007, 124, 43. (10) Pelka, R.; Moszyn´ska, I.; Arabczyk, W. Catal. Lett. 2009, 128, 72. (11) Oyama, S. T. J. Catal. 1992, 133, 358. (12) Dje´ga-Mariadassou, G.; Shin, C. H.; Bugli, G.; Mol, J. Catal. A: Chem. 1999, 141, 263. (13) Temkin, M. I.; Pyzhev, V. J. Phys. Chem. 1039, 13. (14) Temkin, M. I.; Pyzhev, V. Acta Physiochem. 1940, 12.
(5)
In the range where two phases (R-Fe(N) and γ′-Fe4N) exist simultaneously and in the region where the γ′ phase is saturated with nitrogen, the rate of the ammonia decomposition reaction is described by the equation
rdecomp ) 6.2 × 10-6 - 2.3 × 10-6 ln P
(6)
The rate of ammonia decomposition depends linearly on the logarithm of the nitriding potential. Conclusions The rate of the ammonia decomposition reaction in the stationary states (XN ) constant, equilibrium between gas and solid phase) depends on the nitriding degree of solid samples
Acknowledgment. The work has been financed as Research Project No. NN 209336837. References and Notes
JP9099286
