Theoretical Foundations of Chemical Engineering, Vol. 37, No. 4, 2003, pp. 416–420. From Teoreticheskie Osnovy Khimicheskoi Tekhnologii, Vol. 37, No. 4, 2003, pp. 445–448. Original English Text Copyright © 2003 by Ceyhun.
Kinetic Studies on Karlıova Coal1 I˙ . Ceyhun Education Faculty, Department of Chemistry, Atatürk University, 25240 Erzurum, Turkey Received July 5, 2001
Abstract—In this study, reaction kinetics of the liquefaction of Karlıova coal in a process development unit having three reactors in series have been studied at temperatures of 530–570°C and pressures of 15–25 MPa. It is shown that the rate of hydrogen consumption can be expressed as a function of the concentrations of coal and catalyst, hydrogen partial pressure, reaction temperature, and residence time, and is controlled by the rates of hydrogenation of polynuclear aromatic components and the rates of formation and stabilization of radicals. The relative contribution of these reactions, at any temperature, determines the influence of the hydrogen partial pressure on the rate of the hydrogen consumption. The kinetics of the decomposition reactions of coal to preasphaltene, asphaltene, and oil also have been studied. The apparent activation energies determined are 20 kJ/mol for coal to preasphaltene, 40 kJ/mol for preasphaltene, 66 kJ/mol for asphaltene to oil, and 174 kJ/mol for oil to gases. 1
1. INTRODUCTION
Turkey, with a population 65 million, is a bridge between Europe and Asia. The country was established in 1923, after the Ottoman Empire collapsed at the end of World War I. The coal samples used were selected from Karlıova in Turkey. The hydroliquefaction of coal involves the thermal decomposition of the complex molecular structure of the coal in the presence of a hydrogen-donor solvent and hydrogen at high pressure. Relative to bituminous coals, coals are richer in oxygen and the variety of functional groups and respond more rapidly to heating [1]. Consequently, the hydroliquefaction reactions of coals can also be expected to differ. In hydroliquefaction of coals, rapid thermal decomposition will be followed by rapid condensation reactions of the primary products unless the availability and transferability of reactive hydrogen is sufficiently high. This paper reports the results of a study of the hydroliquefaction of coal with regard to the influence of catalyst concentration, reaction, reaction temperature, and hydrogen partial pressure on the rates of hydrogen consumption and thermal decomposition of the coal, and of the intermediate products. 2. EXPERIMENTAL In the experiments, a coal from Karlıova (Turkey) was used; the analyses of the coal are given in Table 1. The solvent was the steady-state recycle solvent formed in the same continuous facility as used in this study; during the hydrogenation of coal under these conditions, the first two series were to determine the effect of temperature and the hydrogen partial pressure, respec1 This
article was submitted by the author in English.
tively, on the rate of hydrogen consumption at a constant catalyst concentration and residence time. In the third series, the effect of reducing the catalyst concentration was investigated. In the fourth series, temperature, pressure, etc., were similar to those used in these experiments. Iron oxide and sulfur were used as a catalyst. The coal was mixed completely, with the tank reactors in series. Reaction temperatures were between 530 and 570°C, total pressure between 15 and 25 MPa, residence time between 40 and 90 min, and relative catalyst concentration between 0.2 and 1.0. The products from each experiment were fractionated by distillation into naphtha, solvent, and bottoms. For each experiment, the coal, feed solvent, naphtha, product solvent, and the bottoms were analyzed and the mass balance for carbon, hydrogen, nitrogen, sulfur, and oxygen was checked. The hydrogen consumption was calculated from the material and element balances. The kinetics of the decomposition reactions of the coal and the intermediate products was studied using the yield data for the solvent fractionations of the distillation bottoms. The distillation bottom fractions were separated into hexane-soluble materials, hexane-insoluble–benzene-soluble materials, benzene-insoluble– pyridine-soluble materials, and pyridine-insoluble materials. Table 1. Representative analysis of coal Elemental composition, wt % dry basis
Ash, wt % daf
C
H
N
S
O
18.2
67.0
5.2
0.8
1.04
26.2
0040-5795/03/3704-0416$25.00 © 2003 MAIK “Nauka /Interperiodica”
KINETIC STUDIES ON KARLıOVA COAL
3. RESULTS
Average hydrogen partial pressure
Relative catalyst concentration
Residence time, min
Relative hydrogen consumption
1 2 3 4 5 6 7 8 9 10 11 12 13 14
570 560 550 540 530 570 560 550 530 550 550 530 530 570
146 156 162 165 166 185 196 196 198 165 165 128 124 176
1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 0.5 0.2 0.2 0.2 1.0
60 60 60 60 60 60 60 60 60 60 60 30 52 90
1.26 1.02 1.00 0.92 0.86 1.22 1.06 1.10 1.02 0.92 0.80 0.30 0.46 1.40
4. DISCUSSION A simplified model of the liquefaction of coal [2–6] is used to correlate the rate of hydrogen consumption with the reaction temperature and the hydrogen partial pressure. The liquefaction of coal is believed to involve many reactions, such as the formation and stabilization of radicals and the hydrogenation of the hydrogendonor components, which proceed simultaneously, and the kinetic equilibrium condition established for these reactions, which controls the rate of hydrogen consumption. The radicals are formed by the thermal decomposition of the coal (A), the intermediate products (RR, RH), and the solvent (S): k Ä + RR + RH + S 1 bR·. T
(radical formation reaction)
2
Relative hydrogen consumption
Relative hydrogen consumption
The average values of the hydrogen partial pressures at the inlet and the outlet of the three-reactor system are given in Table 2. The effect of the residence time on the hydrogen consumption obtained under the mildest and the severest reaction conditions with regard to temperature and pressure in Table 2 are given in Fig. 1. For the mildest reaction condition (530°C and 15 MPa), the increase in hydrogen consumption is proportional to residence time. However, for the severest reaction condition (570°C and 25 MPa), the increase in hydrogen consumption with residence time is initially rapid but then decreases, following the thermal decomposition reactions of the coal and the intermediate products, which are rapid and approach completion in a short residence time. The effect of temperature on the rate of hydrogen consumption at total pressures of 20 and 25 MPa is shown in Fig. 2. The increase in the hydrogen consumption with increase in temperature (530– 560°C) is small at both pressures, with the hydrogen consumption being higher for the higher pressure, and when the temperature approaches 570°C, the hydrogen consumption exceeds that obtained at 25 MPa.
Reaction temperature, °C
Four series of experiments were carried out. The first two experimental series (experiments nos. 1–5 and 6–9) were to determine the effects of temperature and the hydrogen partial pressure on the rate of hydrogen consumption at a constant catalyst concentration and residence time. In the third experimental series (experiments nos. 10 and 11), the effect of reducing the catalyst concentration was investigated. In the fourth experimental series (experiments nos. 12–14), different residence times were used, with two experiments (12 and 13) carried out under the mildest reaction conditions with regard to hydrogen partial pressure and catalyst concentration, and the third experiment (14) carried out under severe reaction conditions with regard to all three variables.
Experiment no.
Table 2. Reaction conditions and hydrogen consumption
The results obtained from the 14 experiments are summarized in Table 2, where the data are expressed relative to the catalyst concentration and the calculated hydrogen consumption.
– G 1.5 1.0 1
0.5 0
0.5
417
1.0 1.5 Residence time, h
t, h
Fig. 1. Effect of residence time on hydrogen consumption: (1) 530°C and 15 MPa; (2) 570°C and 25 MPa.
(1)
– G 1.2 0.8
2
0.4
1
0 520
540
560
T, °C
Reaction temperature, °C Fig. 2. Effect of reaction temperatures on hydrogen consumption: (1) 20; (2) 25 MPa; residence time, 60 min.
THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING
Vol. 37
No. 4
2003
418
CEYHUN log(KCa) × 10–3 5 4 3 0
0.5 1.0 logC Log relative catalyst concentration
Fig. 3. Effect of catalyst concentration on rate of reaction at 550°C and 20 MPa for 60 min.
Some of the radicals produced by reaction (1) are stabilized by extracting hydrogen from coal and intermediate products, but most are stabilized by the reactions with donor molecules (DH) and with the radical (R·) itself: k R· + DH 2 RH + D, (2) (radical stabilization reaction) R· + R·
k3
RR. (3) Polynuclear aromatic molecules, derived from coal during thermal decomposition or present in the solvent, are hydrogenated and converted to donor molecules (DH) by reaction (4) and, in part, to naphthenic molecules (DHH) by reaction (5): D + H2 DH + H2
k4 k–4 k5 k–5
DH,
(4)
DHH.
(5)
(formation of hydrogen donors) Hydrogen is consumed in the hydrogenation of the polynuclear aromatic molecules (D) by reaction (4) and donor molecules (DH) by reaction (5). If the rate of decomposition of naphthenic molecules is very small, reaction (5) attains equilibrium rapidly and hydrogen consumption by the reaction can be neglected. When the coal liquefaction reactions are carried out in the complete-mixing tank reactor for residence time ∆θ, the hydrogen consumption (∆H) is expressed in the form (6) G = k4[D][H2]t – k–4[DH]t, where k4 and k–4 are, respectively, the rate constants of hydrogenation of the polynuclear aromatic component and the reverse reaction (4). The concentration of donor components is expressed in the form k 4 [ D ] [ H 2 ]t + [ DH ] 0 -, (7) [ DH ] = ------------------------------------------------1 + ( k –4 + k 2 [ R ] )t where k2 is the rate constant in the radical stabilization reaction (2) of radical with donor molecule, and [DH]0
and [R] are the concentrations of donor components in the feed and the radicals (reaction (1)) R·, respectively. From Eqs. (6) and (7), the hydrogen consumption is expressed in the form k 4 t ( 1 + k 2 [ R ]t ) – ( k 4 /K 1 ) [ DH ] 0 t -, (8) G = ------------------------------------------------------------------------------1 + ( ( k 4 /K 1 ) + k 2 [ R ] )t where K1 = k4/k–4. If (k4/K1)t Ⰷ k2[R]t + 1, Eq. (8) becomes G = [DH](1 + k2[R]t) – [DH]0. (9) The concentration of donor component [DH] approaches [DH]* at equilibrium and is expressed in the form ( [ DH ] + [ DHH ] )K 1 [ H 2 ] [ DH ]* = ------------------------------------------------------------, (10) 2 1 + K 1 [ H2 ] + K 1 K 2 [ H2 ] where K1 = k4/k–4 and K2 = k5/k–5. Equations (9) and (10) predict that the rate of hydrogen consumption decreases with increases in hydrogen partial pressure when (k4/K1)t Ⰷ k2[R]t + 1. However, if (k4/K1)t Ⰶ k2[R]t + 1, Eq. (8) becomes k –4 [ DH ] 0 (11) G = k 4 [ D ] [ H 2 ]t – ----------------------. k2[R ] Equation (11) predicts that the rate of hydrogen consumption increases with increases in hydrogen partial pressure. Thus, experimental observation of the dependence of the rate of hydrogen consumption on the hydrogen partial pressure can be used to indicate which reaction step controls the rate of hydrogen transfer in the coal liquefaction reaction. The rate of hydrogen consumption as expressed by Eq. (8), which was derived on the basis of the proposed model for coal liquefaction, is complicated and difficult to apply in particle. Thus, expressions (12) and (13) are experimentally derived as a function of K, coal concentration (B), catalyst concentration (C), and hydrogen partial pressure ( P H2 ). K incorporates the rate constants kDH, kRDH, kR, and kRR, relating to the radical concentration and the chemical equilibrium constant (K1), and is a function of temperature: b
G = KCaB P H2 t,
(12)
where K, a constant at constant temperature, can be expressed as K = Aexp(–E/(RT)), (13) where A and E are constants, R is the gas constant, and T is the reaction temperature [7]. Figure 3 shows the relation between KCa and C obtained from the results of the experiments carried out with various catalyst concentrations at 550°C and 20 MPa total pressure for 60 min (B and P H2 constant). The relation is rectilinear, and
THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING
Vol. 37
No. 4
2003
KINETIC STUDIES ON KARLıOVA COAL log(KPçb 2) × 10–4 10 8
1 2 3 4
6 4 2 0 10
15
20
logPç2
Log hydrogen partial pressure, MPa
Exponent of hydrogen partial pressure
Fig. 4. Effect of hydrogen partial pressure on rate of reaction: (1) 530°C; (2) 550°C; (3) 560°C; (4) 570°C; residence time, 60 min.
b 1.2 0.8 0.4 0 1.30 1.35 1.40 1.45 Reciprocal reaction temperature, T–1 × 103, K–1
Fig. 5. Relation between exponent b of hydrogen partial pressure and reaction temperature for 60 min of residence time.
419
from the slope a value of 0.18 for the exponent a is obtained, which reflects the fact that the effect of catalyst concentration on the rate of hydrogen consumption at 550°C is small (Fig. 2). Figure 4 shows the relation b between K P H2 and the hydrogen partial pressure PH indicated by the experiments at four reaction temperatures for 60 min with C and B constant. The exponent b calculated at each temperature from Fig. 4 is a rectilinear function of the reciprocal of the reaction temperature in Fig. 5. The value of b becomes unity at 530°C and zero at 570°C. These values of b indicate that, at lower temperatures of ~530°C, the rate-determining step in the hydrogen transfer reaction is the hydrogenation of polynuclear aromatic components (reaction (4)). At higher temperatures of ~570°C, radical formation and stabilization appear to be rate controlling. Thus, on the basis of these results, it seems that the thermal decomposition of most of the coal and the intermediate products of low thermal stability is completed in the early stages of reaction at temperatures of ~570°C and, hence, that the rate of radical formation has decreased to a low level. From this discussion, it can be expected that, even at 570°C, the shorter residence time gives a larger value of b. A plot of logk calculated at each experimental temperature using Eq. (12) versus the reciprocal of the temperature is shown in Fig. 6. The inverse rectilinear dependence substantiates the validity of Eq. (13) and shows that the value of E/R is 103.5 × 103. This study has suggested that, in kinetic studies of coal liquefaction reactions, the solvent should be in equilibrium with the coal. The equilibrated solvent was produced by repeated recycling, using the same coal as
logK
logki
10–2
100 Rate constant k, l/min
10–3 10–4 10–5
1 10–1
2 4
3
10–2
10–6 10–4 1.35 1.40 1.45 Reciprocal reaction temperature, T–1 × 103, K–1
Fig. 6. Relation between K and reaction temperature for 60 min of residence time.
5
1.35 1.40 1.45 Reciprocal reaction temperature, T–1 × 103, K–1
Fig. 7. Arrhenius plots for hydroliquefaction reactions of Karlıova coal. 1, kcp; 2, kcg; 3, kpa; 4, kao; 5, kog.
THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING
Vol. 37
No. 4
2003
420
CEYHUN
that in the experiments, at 530°C and 15 MPa total pressure for a residence time of 60 min. To facilitate the analysis of the reaction kinetics, the following simplified reaction sequence occurring during the hydroliquefaction of the coal was assumed: Coal
kcg
Gases ( CO, CO 2, H 2 O, CH 4 )
kcp
polynuclear aromatic components in the lower temperature region at ~530°C and the radical formation and stabilizations in the higher temperature region at ~570°C. The thermal decomposition reactions of the coal to asphaltene proceed at a high rate even in the lower temperature region (~530°C), because their activation energies are in the relatively low range of 20–40 kJ/mol.
Preasphaltene REFERENCES
kpa
Asphaltene ao Oil og Gases ( C 1 –C 4 ) An Arrhenius plot for each reaction in this sequence is given in Fig. 7. k
k
THEORETICAL FOUNDATIONS OF CHEMICAL ENGINEERING
~
5. CONCLUSIONS In this study of the hydroliquefaction of coal, it has been shown that the rate of hydrogen consumption can be expressed mathematically as a function of the concentration of the coal and the hydrogen partial pressure, the catalyst, and the reaction temperature. An analysis of a model for coal hydroliquefaction, the experimental results for the rate of hydrogen consumption, and the values determined for the exponent b on the hydrogen partial pressure PH in the expression relating the rate of hydrogen consumption to experimental variables indicate that the rate-determining step in hydrogen transfer for 60 min of residence time is the hydrogenation of
1. Schafer, H.N.S., Pyrolysis of Brown Coal. Fuel, 1979, vol. 58, p. 667. 2. Han, K.W. and Weh, C.Y., Initial Stage (Short Residence Time) Coal Dissolution, Fuel, 1979, vol. 58, p. 779. 3. Radi, A.H. Kinetics of Tetralin Extraction of Jordan Oil Shale, Fuel, 1980, vol. 59, p. 535. 4. Wiser, W.H., Kinetic Study of the Thermal Dissolution of High-Volatile Bituminous Coal, Fuel, 1968, vol. 47, p. 475. 5. Cronauer, D.C., Shah, Y.T., and Ruberto, R.G., Investigation of Mechanisms of Hydrogen Transfer in Coal Hydrogenation, Ind. Eng. Chem. Proc. Des. Dev., 1978, vol. 17, p. 281. 6. Ceyhun, I., Kocakerim, M.M., Saraç, H., C olak, S., Dissolution Kinetics of Colemanite in Chlorine Saturated Water, Teor. Osn. Khim. Tekhnol., 1999, vol. 33, no. 3, p. 253. 7. Levenspiel, O., Chemical Reaction Engineering, New York: Wiley, 1972, p. 77.
Vol. 37
No. 4
2003