New Catalytic Conversion of Hydrocarbons in a Heated Tube at a Nearly Stoichiometric Vapor to Gas Ratio

V. S. Igumnov

Joint Institute of High Temperatures, Russian Academy of Sciences, Moscow, 125412 Russia

Moscow, 125412,  Izhorskaya st., 13/19,  e-mail: vsigumnov&mail.ru

(Ðîññèÿ 125412, Ìîñêâà, óë. Èæîðñêàÿ, ä.13/19, Îáúåäèíåííûé èíñòèòóò âûñîêèõ òåìïåðàòóð ÐÀÍ, Êàíöåëÿðèÿ)

Abstract—A method for the catalytic vapor and carbon dioxide conversion of hydrocarbons is suggested. The con­version was performed in heated tubes at a nearly stoichiometric vapor-to-gas ratio. The energy intensity of the pro­cess and the rate of oxidants (H20 and/or C02) thus decreased, while the lifetime of the catalyst increases. An inert Zr02 packing (inhibitor) was used. It was the first to be loaded into the tube along the gas pathway and then alternated with the catalyst. In a layer of inert inhibitor, the gas reaction mixture was heated to a temperature close to that of the tube wall. The catalyst was then fed in, providing additional heat to the catalytic packing layer for endothermal conversion. The catalyst carbonization was thus reduced. The number of layer pairs was calculated with allowance for the consumption of the processed products and the tube size in industrial pipe furnaces. The proposed conversion procedure was tested in the laboratory and on pilot plants and then in the industrial pipe fur­naces of the coke gasworks of PO Angarsknefteorgsintez, the Novocherkassk Plant of Synthetic Products (NPSP), and the Oskol Electrometallurgical Industrial Complex (OEIC).

Keywords: methane, natural gas, hydrocarbons, hydrogen, carbon, carbon oxide, nickel catalyst, inhibitor, heated tube, endothermal reaction, catalytic conversion.

 

INTRODUCTION

The catalytic vapor and carbon dioxide conversion of hydrocarbons in heated tubes is the main procedure for producing the 0/H2 gas mixture used as industrial and reductive gases and as syngas. Conversion of hydrocarbons is the main and at the same time most energy intensive process in nitrogen chemistry, poly­mer production, and direct iron reduction in metal­lurgy. It is, therefore, one of the major challenges in upgrading these industries and analyzing the neck stages in technology remodeling. The catalytic vapor and carbon dioxide conversions of hydrocarbons are represented by the following chemical reactions:

Cn Hm + nÍ20 = nÑÎ + (n + ò/2)H2,    (1)

Cn Hm + nC02 = 2nCO + (w/2)H2.     (2)

The industrial processes are generally performed in reaction mixtures with a large (fourfold or higher) excess of an oxidant at 900—1200 Ê and at pressures from 0.2 to 4 MPa to avoid catalyst carbonization. A stoichiometric catalytic conversion is difficult to per­form in industry. This problem can be solved, for example, by modifying the structure of the tube reac­tor, supplying additional heat in the catalyst bed [ 1 ] for endothermal reactions (1) and (2), and reducing car­bon deposition in the catalyst pores.

The industrial processes are generally performed in reaction mixtures with a large (fourfold or higher) excess of an oxidant at 900—1200 Ê and at pressures from 0.2 to 4 MPa to avoid catalyst carbonization. A stoichiometric catalytic conversion is difficult to per­form in industry. This problem can be solved, for example, by modifying the structure of the tube reac­tor, supplying additional heat in the catalyst bed [ 1 ] for endothermal reactions (1) and (2), and reducing car­bon deposition in the catalyst pores.

A reaction mixture is fed into a heated tube at T ~ 700 Kat the inlet (the temperature at the outlet is 1200 K).

According to our calculations of the thermodynamic probability of carbon deposition, carbon is not depos­ited in vapor conversion under these conditions when the H20/CH4 molar ratio is higher than 1.5 [2]. Reac­tions (1) and (2) with methane occur according to the schemes [3]

CH4 + H20 + 204 kJ/mol = CO + 3H2, (3)

CH4 + C02 + 248 kJ/mol = 2CO + 2H2. (4)

These reactions are highly endothermal (for pur­poses of comparison, water evaporation requires 41 kJ/mol). Under catalytic conversion conditions, carbon is mainly formed by the reactions

CH4 + 78 kJ/mol = Ñ + 2H2,                 (5)

2CO — Ñ + C02.                        (6)

Carbon is further oxidized by the reactions

Ñ + H20 + 118 kJ/mol = CO + H2. (7) Ñ + CO, + 161 kJ/mol = 2CO.          (8)

All these reactions are energy intensive. Direct reac­tions (3) and (4) are unlikely to occur at T < 1200 Ê because the CH4, H20, and C02 molecules must have an energy of more than 2.5 eV. There are up to 0.001% such molecules in the energy distribution. The esti­mated probability of their meeting is up to 0.01, and carbon will not be isolated in the catalyst if the condi­tions of heat transfer from the tube wall to the catalyst bed provide heat for reactions (7) and (8). If C02 and

 

Fig. 1. Dependence of the effective latent heat of reaction on the excess of oxidants n.

 

H2O are added in excess over stoichiometry in (3) and (4), the heat effect of the reactions will decrease expo­nentially [4—6]. This dependence was called the effec­tive latent heat of reaction [6]:

                     (9)

Here, ∆H is the standard thermal effect of the reac­tion under study, and n is the number of moles of all gases in the reaction gas mixture in the left part of the equations, ignoring the moles of gases that provide the stoichiometry;  k - coefficient f ( C i ). Figure 1 shows dependences (9) for the reactions

 

CH4 + H2O + n H2O = CO + 3H2 + nH2O, (10)

CH4 + CO2 + n CO2 = 2CO + 2H2 + n CO2, (11)

 where n = 1 and Qef = 102 kJ/mol if H2O/CH4 = 2.

At n > 1.5 the reaction

CH4 + 2H2O = CO2 + 4H2             (12)

will be pronounced. The total excess of H2O reacts with the produced CO:

H2O + CO = H2 + CO2.                 (13)

When over 50% of CH4 has reacted, parallel reac­tions start that are the reverse of (7) and (8) and liber­ate carbon:

                             CO + H2 = C + H2O,           (14)

                                             2CO = C + CO2.          (15)

The conversion is catalyzed almost completely in the catalyst pores, where carbon nanostructures appear in the form of fullerenes and carbon filaments [5, 6]. The carbon nanostructures increase the active contact surface manyfold. Reactions (7) and (8) with carbon require that the CO2 and H2O oxidant mole­cules have an energy of at least 1.67 and 1.22 eV, respectively. The higher the H2O(CO2)/CH4 ratio, the larger the number of oxidant molecules having these amounts of energy. This is numerically expressed as a decrease in the effective latent heat of the reaction. Equation (1) defines (in general form) the dependence of the energy distribution of oxidant molecules, the probabilities of intermolecular collisions, and the den­sities of molecules and the molecular diffusion in the catalyst pores.

 

EXPERIMENTAL

A unit for studying the conversion of hydrocarbons (including those with stoichiometric ratios of compo­nents) on a catalyst in a heated tube was created at the Joint Institute for High Temperatures, Russian Acad­emy of Sciences [2]. In the experiments described in this work, we used catalysts with known physicochem- ical and hydrodynamic properties of the catalyst bed (GIAP-16 and GIAP-8). Similar results can be obtained for any modern catalyst. The test tube of cor­rosion-proof steel was 60 mm in diameter and 1.5 m in length; the wall thickness was 3 mm. For GIAP-16, we used Raschig rings 15 x 12 x 7 mm in size; the NiO concentration was 26 wt %, porosity 35.5 vol %; resis­tance 70 MPa. ZrO2 balls with a diameter of 21 mm and a strength of 140 MPa were used as an inhibitor. Uniform heating of the wall to 1400 K was possible due to the direct electrical heating of the tube. The wall temperature (Tw) was measured along with the average gas temperature at the inlet and outlet, the tempera­ture along the tube and its radius (T), and the reaction gas concentrations on and along the tube axis and near the wall. The adjusted parameters were the rate and composition of gases at the inlet of the tube, the pres­sure in the reactor, the heat flux from the tube wall, the temperature of reaction gases at the outlet of the tube, and the H2O(CO2)/CH4 ratio. A detailed description of the measurement procedure and the scheme of the experimental unit are presented in [2].

       Figure 2 shows a heated tube filled with a continuous layer of GIAP-16 catalyst, as in the case of industrial conditions. The GIAP-16 charge was 3.6 x 10-3 m3. In the experiments, the methane flow rate was (0.1—0.32 x 10-3) nm3/s; the heat flow rate was 10—35 kW/m2; and the pressure in the reactor was 50—100 kPa.

           

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Fig. 2. Variation in the reactant gas temperature Tf, wall temperature Tw, and methane concentration CCH4             along a tube without carbon deposits.

 

Steady conversion was observed at H2O/CH4 > 4 (at H2O/CH4 < 4, steady conversion could not be achieved without carbon liberation). Figure 2 shows the characteristic experimental data. The active reac­tion zone lies on a segment of ±0.1 m in the vicinity of the section Lr. The radial temperature distribution on a segment of section Lr in the tube is presented in Fig. 3. According to these parameters, the converted gas contained at least 40 vol % oxidants at the outlet of the tube, which is not. acceptable for the reduction of iron oxides.

  A scheme for the catalytic conversion of methane in a heated tube was proposed based on the results and using the calculations in [2]. The liberation of carbon was attributed to a lack of heat in the heated tube. We used overheating of the reaction mixture on an inhibit­ing packing separated into two layers for an additional heat supply. A layer of inert inhibitor on which there was no catalytic conversion was placed at the inlet of the tube. According to Fig. 4, the layer was 0.46 m in length and consisted of ZrO2 balls with a diameter of 21 mm; the the charge was 1.2 dm3. The experimental data on the inhibiting properties of ZrO2 are given in [8].

The next layer was the GIAP-16 catalyst (charge, 2.4 dm3). In the first layer, methane did not react and did not decompose with the liberation of carbon, but was only heated to a temperature close to that of the wall. The inert ZrO2 support was chosen based on comparative experiments with substances used as cat­alyst supports [7—9]. The experiments with the inhib­itor layer were performed under conditions similar to those of the experiment illustrated in Fig. 2. They started at H2O(CO2)/CH4 = 3 and continued until the H2O(CO2)/CH4 ratio gradually approached a stoichi­ometric ratio of 1. The appearance of carbon deposits and, as a consequence, catalyst decomposition were determined from the pressure differential at the inlet and outlet of the heated tube and along the catalyst bed. The increase in the pressure differential on a spe­cific segment corresponded to the zone of the most active liberation of carbon. The pressure differential increased as carbon was deposited and the catalyst decomposed.

 

 

 

Fig. 3. Temperature distribution within the radius of a tube on segment L, calculated using the quasihomogeneous model (A,ef = 3 W (m K)). For parameters, see Fig. 2.

 

 

A steady technological mode of catalytic conver­sion was attained at an oxidant to methane ratio of 1. Figure 4 shows the characteristic change in the tube wall and reactant gas temperatures. The reaction mix­ture on ZrO2 balls at the inlet was heated to a temper­ature close to that of the wall. As can be seen in Fig. 4, the methane concentration remained constant. According to the chromatographic analysis data, hydrogen and carbon oxide did not form in the ZrO2 packing. An active reaction occurred when the gases reached the catalyst bed. The energy stored in the course of heating on ZrO2 was sufficient for reactions (3) and (4) to occur without carbon liberation. As the heat was spent on the reaction, the reactant gas tem­perature in the catalyst bed reached its minimum (Tmin), which is the function of all main parameters of the conversion. The segment between the inhibitor- catalyst (IN-CA) boundary and the extreme temper­ature of reactant gases in the catalytic packing layer corresponded to the most active reaction zone (2) (MARZ-2). The segment from IN—CA to the cross section where up to 50% of the initial amount of meth­ane reacted was the most active reaction zone (1) (MARZ-1). If Tmin falls to a certain value (TminCR), MARZ-1 and MARZ-2 do not coincide, and the dif­ference between their linear sizes equals the active car­bon deposition zone (ACDZ).

Fig. 4. Distributions of the wall (Tw) and reactant gas (Tf) temperatures and methane concentration along the tube. MARZ-1, MARZ-2, and ACDZ are the most active zones of reactions (1) and (2) and the active carbon deposition zone, respectively. For the parameters of the experiment, see Fig. 2.

 

 

 

 

      Based on the laboratory experiments and calcula­tions using the quasihomogeneous model

with the ideal ousting [10] (the notion of effective heat conduc­tivity was used for calculating the heat transfer), we proposed a scheme for the catalytic conversion of hydrocarbons in a heated tube with intermediate over­heating on inhibiting packings, which was first consid­ered in a different formulation in [12]. In industry, this technique was first tested on the pilot unit of the Novocherkassk Synthetic Products Plant (NSPP). The scheme for the charging of the tube being tested is presented in Fig. 5. The length of the catalyst bed was calculated under the condition that the reduction in the temperature of the catalyst bed be no more than 150 K. The length of the heated tube of the industrial furnace was 11 m, the outer diameter 120 mm, and the length of the illuminated part of the furnace 8.7 m.

The temperature range of reactant gas heating in the pilot plant was set based on the heating conditions and technical specifications for tube operation. The working rate of methane for conversion was 80 nm3/h at H2O/CH4 = 1.5. The sizes of the inhibitor layer and catalyst bed were calculated from these conditions using the heat transfer conditions and the ideal ousting model. The calculations gave eight inhibitor—catalyst layer pairs in the given mode at a methane rate of 130 nm3/h and H2O/CH4 = 1.5. The average pressure in the tube was 1.7 MPa; the residual methane content was 11 vol %, which is close to the equilibrium meth­ane content under the given test conditions. The space velocity of methane was 2050 h-1 under the normal conditions. The operation time of the tube under the chosen test conditions was 4 days, after which the ini­tial properties of the catalyst according the technical specifications did not change.

Industrial trials of the catalytic conversion of Cj—C5 refinery gases in the intermediate heating mode. The trials were carried out in the coke gasworks of PO Angarsknefteorgsintez. The refinery gas was a mixture of hydrogen (up to 30 vol %) and a wide fraction of C1-C5 hydrocarbons (mainly propane and butane). Conversion was performed with a water vapor at H2O/C = 5 in a pipe furnace containing 72 pipes with an illuminated part of 6.8 m and an inner diameter of 158 mm. The GIAP-8 catalyst in the form of cylinders was used; the cylinder height was 16 mm, diameter 14mm, 7.5 wt % NiO, porosity 54.5 vol %, stability 25 MPa. The inhibitor were ZrO2 balls 21 mm in diameter used as a packing for high-temperature heat regenerators. The run duration of the furnace before the catalyst reloading was 10-15 months. After months, the catalyst was carbonized and half decomposed in the furnace. It was discharged and for­warded for recycling.

Fig. 6. Location of packings and variation in reactant gas temperature along a tube with an inner diameter of 158 mm for vapor conversion of C1—C5 hydrocarbons (ZrO2 balls with a diameter of 21 mm; GIAP-8 catalyst; cylinders with a diameter of 14 mm and a height of 6 mm).

Oil gas hydrocarbons generally quickly carbonize the catalyst in the temperature mode of the catalytic conver­sion of natural gas. The temperature mode of the pipe furnace, therefore, had a reaction gas temperature 670 K at the inlet and 1070 K at the outlet. Higher hydrocar­bons relative to methane decompose to carbon by a num­ber of chemical schemes [11]. An estimation of the ther­modynamic probability of carbon formation shows that the majority of hydrocarbons decompose according to a conventional scheme [12]

C5H12 → C2H6 → C2H4 → C2H2 → CH4 → C     (16)

            The scheme can be more complex. The thermody- namic analysis of the equilibrium states was rough. The reaction rate of a chemical transformation to car­bon depends on the transfer conditions in the catalyst bed. The rates of hydrocarbon transformations were adjusted using intermediate overheating of reactant gases with an inhibitor. The use of the inhibitor allowed us to set the requited temperature mode and increase heat supply to the reaction zone.    

Figure 6 shows the arrangement of packings and the variation in the temperature of reactant gases along the tube (the packing layer sizes are indicated). The ZrO2 inhibitor balls were placed at the inlet of the tube, and the reaction gas mixture was heated in a relatively short layer of 1.5 size. A long (12-size, as in Fig. 5) inhibitor layer cannot be placed here because higher hydrocarbon mole­cules decompose to carbon and hydrogen at T > 750 K according to scheme (11). In a small (up to double sized) layer, the decomposition of heavy hydrocarbons occurs according to schemes similar to (12). The hydrocarbons then react only slightly with H2O in the catalyst bed. Hydrogen is formed in this layer and retards the formation of carbon [3]. At the outlet of the catalyst in the first pair of inhibitor—catalyst layers, the reaction gas no longer contains the C5 and C4 fractions and can therefore be heated to 950 K. In the second pair of layers, full conver­sion ofC3 and C2 hydrocarbons occurs. This is why the gas is heated to 1100 K in the third pair of layers [8], the most active conversion stage starts [7], and the temperature falls to a dangerous level. In the fourth pair oflayers, the hydro­carbons are fully converted to compounds whose compo­sition meets the conditions and input parameters of the general process [7]. The inlet and outlet temperatures (Fig. 6) and the wall temperature were measured, and the intermediate temperatures were calculated.

The pipe furnace was stopped after 15 months of operation because of the catalyst decomposition in the majority of tubes. The catalyst was half destroyed in the main pipes but remained in its initial state in the tested pipes; the inhibitor did not change and could be recycled. The main results of full-scale industrial trials were described in [13, 14].    

CONCLUSIONS

We have presented the results from our laboratory experiments and industrial trials of hydrocarbon conver­sion in tubes with intermediate overheating of reactant gases in an inhibitor—catalyst system and the calculated data. All of the trials (carried out at the NSPP, PO Angar- sknefteorgsintez coke gasworks, and OEIC) were recog­nized as successful and promising. The laboratory trials showed that stoichiometric conversion of methane with water vapor, carbon dioxide, and their mixtures was pos­sible without carbon isolation and hence catalyst decom­position. The inhibiting packing in industrial pipe fur­naces was found to prolong the catalyst's service life. For precise calculations of the size of the inhibitor—catalyst layers and their number in the reactor pipe, we recom­mend using the effective latent heat ofreaction Qef, which is a precise characteristic of the heat flows responsible for the endothermal reactions in the catalyst bed. Our main conclusion: The intermediate overheating of a reaction mixture on an inhibitor in the catalytic conversion of hydrocarbons in a pipe reactor allows the temperature profile to be smoothed along the pipe radius in the cata­lyst bed, increasing the heat supply to the reaction zone and hence the reactor output and the catalyst service life. This is especially important if tubes with larger diameters and hydrocarbons heavier than methane are used. Tubes with larger diameters and natural gas with lower methane contents are planned for use in pipe reactors (reformers) at OEIC. A program of trials has been developed for this purpose under the supervision of the Joint Institute for High Temperatures, Russian Academy of Sciences.

 

REFERENCES

1.   Igumnov, VS. and Vizel', Ya.M., Pretsizionnaya katali- ticheskaya konversiya uglevodorodov dlya pitaniya tverdo- oksidnykh toplivnykh elementov: Mater. VIIMezhdunarod- nogo foruma "Vysokie tekhnologii XXI veka" (Precision Catalytic Conversion of Hydrocarbons for Supply of Solid Oxide Fuel Cells, Proc. 7th Int. Forum "High Technolo­gies in the 21st Century"), Moscow, 2007.

2.    Igumnov, VS., Cand. Sc. (Eng.) Dissertation: MIKhM, 1989.

3.   Spravochnik azotchika: Fiziko-khimicheskie svoistva gazov i zhidkostei. Proizvodstvo tekhnologicheskikh gazov. Sintez ammiaka. (Nitrogen Chemist's Hand­book: Physicochemical Properties of Gases and Liq­uids. Production of Industrial Gases. Synthesis of Ammonia), Moscow: Khimiya, 1986, vol. 1.

4.    Igumnov, V.S., Technical and Technological Methods of Realization of Steam Catalytic Conversion of Natural Gas with a Methane—Water Proportion Close to Stoichiomet- ric Ratio, in Hydrogen Materials Science and Chemis­try of Carbon Nanomaterial, NATO Security Science Series A: Chemistry and Biology, The NATO Program for Security Through Science, 2007, p. 555.

5.   Igumnov, VS., Statistical Weight of Formation Fullerene in Conditions Catalytic Interactions of Methane with Water the Ferry, Conference ICHMS, 2005.

6.    Igumnov, V.S., Carbon Nanostructure: An Intermedi­ate Stage in Catalytic Conversion of Methane, 3rd Int. Symp. "Fullerene and Fullerenoid Structures in the Con­densed Environments", Minsk, 2004.

7.   Vizel', Ya.M., Doctoral (Eng.) Dissertation, St. Peters­burg: Tekhnol. Institute, 1993.

8.   Vizel', Ya.M. and Igumnov, V.S., Teplofiz. Vys. Temp., 1983, vol. 21, no. 3.

9.    USSR Inventor's Certificate 1308648, 1985.

10.   Mostinskii, I.L., Igumnov, VS., Vizel', Ya.M., and Zyry- anov, S.I., in Atomno-vodorodnaya energetika i tekh- nologiya. Sb. st. Vyp. 8 (Atomic Hydrogen Energetics, ACollection of Papers), no. 8, Moscow: Atomizdat, 1988.

11.   Tesner, P.A., Obrazovanie ugleroda iz uglevodorodoi gazovoi fazy (Carbon Formation from a Hydrocarbon Gas Phase), Moscow: Khimiya, 1972.

12.   US Patent 3617227, 1971.

13.   USSR Inventor's Certificate 1829181, 1989.