Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

The Effect of Fuel Mass Fraction on the Combustion and Fluid Flow in a Sulfur Recovery Unit Thermal Reactor

The Effect of Fuel Mass Fraction on the Combustion and Fluid Flow in a Sulfur Recovery Unit... applied sciences Article The Effect of Fuel Mass Fraction on the Combustion and Fluid Flow in a Sulfur Recovery Unit Thermal Reactor Chun-Lang Yeh Department of Aeronautical Engineering, National Formosa University, Huwei, Yunlin 632, Taiwan; clyeh@nfu.edu.tw; Tel.: +886-5-6315527 Academic Editor: Yuyuan Zhao Received: 7 September 2016; Accepted: 26 October 2016; Published: 2 November 2016 Abstract: Sulfur recovery unit (SRU) thermal reactors are negatively affected by high temperature operation. In this paper, the effect of the fuel mass fraction on the combustion and fluid flow in a SRU thermal reactor is investigated numerically. Practical operating conditions for a petrochemical corporation in Taiwan are used as the design conditions for the discussion. The simulation results show that the present design condition is a fuel-rich (or air-lean) condition and gives acceptable sulfur recovery, hydrogen sulfide (H S) destruction, sulfur dioxide (SO ) emissions and thermal 2 2 reactor temperature for an oxygen-normal operation. However, for an oxygen-rich operation, the local maximum temperature exceeds the suggested maximum service temperature, although the average temperature is acceptable. The high temperature region must be inspected very carefully during the annual maintenance period if there are oxygen-rich operations. If the fuel mass fraction to the zone ahead of the choke ring (zone 1) is 0.0625 or 0.125, the average temperature in the zone behind the choke ring (zone 2) is higher than the zone 1 average temperature, which can damage the downstream heat exchanger tubes. If the zone 1 fuel mass fraction is reduced to ensure a lower zone 1 temperature, the temperature in zone 2 and the heat exchanger section must be monitored closely and the zone 2 wall and heat exchanger tubes must be inspected very carefully during the annual maintenance period. To determine a suitable fuel mass fraction for operation, a detailed numerical simulation should be performed first to find the stoichiometric fuel mass fraction which produces the most complete combustion and the highest temperature. This stoichiometric fuel mass fraction should be avoided because the high temperature could damage the zone 1 corner or the choke ring. A higher fuel mass fraction (i.e., fuel-rich or air-lean condition) is more suitable because it can avoid deteriorations of both zone 1 and heat exchanger tubes. Although a lower fuel mass fraction (i.e., fuel-lean or air-rich condition) can avoid deterioration of zone 1, the heat exchanger tubes may be damaged. This paper provides a guideline for adjusting the fuel mass fraction to reduce the high temperature inside the thermal reactor and to ensure an acceptable sulfur recovery. Keywords: SRU thermal reactor; fuel mass fraction; sulfur recovery; H S destruction; SO emission; 2 2 thermal reactor temperature 1. Introduction Desulfurization is very important in the petroleum refining process because oxysulfides from the petroleum refining process are one of the major sources of air pollution. The most frequently used desulfurization process is the Claus process, which converts the H S in natural gas or crude oil into sulfur elements, which reduces the formation of oxysulfides. A sulfur recovery unit (SRU) thermal reactor is the most important equipment in a sulfur plant. It converts the NH , H S and hydrocarbons 3 2 in the reactants into sulfur. Most of the sulfur elements are recovered from the SRU thermal reactor. Appl. Sci. 2016, 6, 331; doi:10.3390/app6110331 www.mdpi.com/journal/applsci Appl. Sci. 2016, 6, 331 2 of 13 The first section of a SRU that uses the Claus process is composed of a burner, a thermal reactor and a waste heat exchanger. The configuration and dimensions of the first section of a SRU for a typical petroleum refinery are shown in Figure 1. The main chemical reactions in a SRU thermal reactor include a thermal step and a catalytic step. Appl. Sci. 2016, 6, 331  2 of 12  2H S + 3O ! 2SO + 2H O 2 2 2 2 hydrocarbons in the reactants into sulfur. Most of the sulfur elements are recovered from the SRU  thermal  reactor.  The first section of a SRU that uses the Claus process is composed of a burner, a  2H S + SO ! 3S + 2H O 2 2 2 thermal reactor and a waste heat exchanger. The configuration and dimensions of the first section of  a SRU for a typical petroleum refinery are shown in Figure 1.  In addition to the above reactions and the combustion of hydrocarbon fuels, other chemical The main chemical reactions in a SRU thermal reactor include a thermal step and a catalytic  reactions taking place in a SRU thermal reactor are step.  2H2S + 3O2 →  2SO2 + 2H2O  2NH + 1.5O ! N + 3H O 3 2 2 2 2H2S + SO2 →  3S + 2H2O  In  addition  to  the  abov 2NH e reacti+ ons SO   and ! the N combustion + H S  + of 2H hydro Ocarbon fuels, other  chemical  3 2 2 2 2 reactions taking place in a SRU thermal reactor are  The interior of a SRU thermal reactor is divided into two zones by a choke ring, in order to increase 2NH3 + 1.5O2  →  N2 + 3H2O  the residence time and enhance the chemical reaction. The zone ahead of the choke ring is called zone 2NH3 + SO2 →  N2 + H2S + 2H2O  1 and the zone behind the choke ring is called zone 2. The inner surface of a SRU thermal reactor The interior of a SRU thermal reactor is divided into two zones by a choke ring, in order to  is fabricated using a refractory to protect its walls because a SRU thermal reactor operates at very increase the residence time and enhance the chemical reaction. The zone ahead of the choke ring is  high temperatures. SRU thermal reactors are negatively affected by high temperature operations called  zone  1  and  the  zone  behind  the  choke  ring  is  called  zone  2.  The  inner  surface  of  a  SRU  thermal reactor is fabricated using a refractory to protect its walls because a SRU thermal reactor  because high temperature can damage the refractory and the heat exchanger tubes. Therefore, operates  at  very  high  temperatures.  SRU  thermal  reactors  are  negatively  affected  by  high  the operating temperature range that is suggested by the manufacturers for the operation of a SRU temperature  operations  because  high  temperature  can  damage  the  refractory  and  the  heat  thermal reactor exchanger must  be tube strictly s.  Thereadher fore,  the ed  to. operating  temperature  range  that  is  suggested  by  the  manufacturers for the operation of a SRU thermal reactor must be strictly adhered to.  (a)  (b)  Figure 1. The configuration and dimensions of the first section of the SRU for a typical petroleum refinery: (a) The overall view; (b) Enlarged view for the burner section. There have been theoretical and experimental studies of SRU thermal reactors. Adewale et al. [1] studied the thermal decomposition of H S into hydrogen and sulfur using a process simulator. Using 2 Appl. Sci. 2016, 6, 331 3 of 13 the net fraction of the acid gas feed to the cracking coils as the controlling parameter, its effect on hydrogen production, the thermal reactor ’s energy requirement, the stability of the burner flame, steam production, the temperature of a Claus reactor and sulfur recovery of the primary SRU was studied. Chardonneaua et al. [2] presented experimental and simulation results for the addition of various amounts of toluene or carbon dioxide/toluene mixtures into the H S gas stream. The results show that there is a decrease in the conversion efficiency when the amount of toluene or carbon dioxide/toluene added to the H S gas stream increases. The role of the reactor ’s operating temperature was also studied. The addition of toluene increases the optimum reactor temperature for enhanced sulfur recovery, but the presence of CO reduces the optimum operating temperature. Selim et al. [3] examined the quality of sulfur deposits that were collected from H S combustion. Sulfur deposits from H S combustion under various conditions were captured and analyzed using X-ray powder diffraction and laser-induced breakdown spectroscopy diagnostics. Monnery et al. [4] experimentally studied the reaction between H S and SO using practical Claus thermal reactor temperatures between 850 and 2 2 1150 C and residence times between 0.05 and 1.2 s. The kinetic data obtained were used to develop a new reaction rate expression. Our experience of operating a practical SRU thermal reactor in Taiwan shows that the refractories at the zone 1 corner and the choke ring are the parts of the thermal reactor that experience the greatest deterioration. The zone 1 corner has a suddenly expanded geometry and a recirculation zone forms behind it. The temperature at the zone 1 corner can exceed the maximum service temperature of the refractory and cause collapse or deformation. The choke ring is subjected to a bending moment from the rapid combustion gas stream and can collapse or deform. Several methods to alleviate this problem have been assessed, including (1) changing the choke ring dimensions [5]; (2) changing the choke ring position [6]; (3) modifying the geometry of the zone 1 corner to a streamlined geometry [7,8]; and (4) replacing the choke ring by a vector wall [9]. However, these methods are either expensive or require an overhaul of the SRU thermal reactor. This paper presents a new, easier and more economical method. The effect of the fuel mass fraction on the combustion and fluid flow in a SRU thermal reactor is investigated numerically. This paper provides a guideline for adjusting the fuel mass fraction to reduce the high temperature inside a thermal reactor and to ensure an acceptable sulfur recovery. Practical operating conditions from a petrochemical corporation in Taiwan were used as the design conditions for the discussion. 2. Numerical Methods and Physical Models In this study, the FLUENT commercial code is used to simulate the reaction and fluid flow in a SRU thermal reactor. The SIMPLE algorithm by Patankar [10] is used to solve the governing equations. The discretizations of convection terms and diffusion terms are respectively performed using the power-law scheme and the central difference scheme. In terms of physical models, considering the accuracy and stability of the models and the evaluations of other researchers, the standard k-" model [11], the P-1 radiation model [12] and the non-premixed combustion model with a -type probability density function [13] are respectively used for the turbulence, radiation and combustion simulations. The standard wall functions [14] are used to resolve the flow quantities (velocity, the temperature, and the turbulence quantities) at the near-wall regions. For the steady-state three-dimensional flow field with the chemical reaction in this study, the governing equations for the continuity equation, momentum equation and turbulence equation (k-" model) are described below. The governing equations for the energy equation, radiation equation, combustion equation and convergence criterion are described in a previous study by the author [15]. Continuity equation: r v = 0 (1) Appl. Sci. 2016, 6, 331 4 of 13 Momentum equation: In an inertial reference frame, the momentum equation [16] may be expressed as !! ! r ($ v v ) = rp +r  + $ g + F (2) where F is the possible source term of the momentum equation which may include the body force caused by the interaction with dispersed phases or the extra force terms caused by physical models or numerical schemes. In Equation (2), the stress tensor t may be expressed as ! ! 2 ! =  r v +r v r v I (3) where I is the unit tensor, the superscript T represents the transpose operator, and the third term on the right-hand side of the equal sign represents the volume dilation effect. By the long-time averaged method, Equation (2) may be converted to Reynolds-averaged Navier-Stokes (RANS) equations as follows: " !# ¶u ¶ ¶p ¶ ¶u j 2 ¶u ¶ i l $u u = +  +  + $u 0u 0 (4) i j ij i j ¶x ¶x ¶x ¶x ¶x 3 ¶x ¶x j i j j i l j In Equation (4), the turbulent stresses (Reynolds stresses) may be related to the mean velocity gradient by the Boussinesq hypothesis as follows: ¶u ¶u j 2 ¶u i k $u 0u 0 =  + $k +   (5) i j ij t t ¶x ¶x 3 ¶x j i k In Equation (5), the turbulent (or eddy) viscosity may be expressed in terms of the turbulence kinetic energy (k) and turbulence dissipation rate (") as: = C $ (6) As mentioned above, k and # are modeled by the k-" model, as described below. k-" model equation: The turbulence kinetic energy (k) and turbulence dissipation rate (") may be solved from their transport equations [11] as follows: " # ¶ ¶  ¶k ($u k) =  + + G + G $" Y + S (7) i k b k ¶x ¶x  ¶x i j j " # ¶ ¶  ¶" " " ($u #) =  + + C (G + C G ) C $ + S (8) 3" 2" " i 1" k b ¶x ¶x  ¶x k k i j j In Equations (7) and (8), G represents the turbulence kinetic energy production and may be expressed as: G =  S (9) k t where S is the modulus of the mean rate-of-strain tensor defined as S  2S S (10) ij ij ¶u ¶u j In Equation (10), S = + . ij ¶x ¶x j i Appl. Sci. 2016, 6, 331  5 of 12  æö ¶u ¶u 1 j Appl. Sci. 2016, 6, 331 5 of 13 ç÷ i In Equation (10),  .  S=+ ij ç÷ ç÷ 2 ¶¶xx ji èø Gb is the generation of turbulence kinetic energy due to buoyancy, calculated as  G is the generation of turbulence kinetic energy due to buoyancy, calculated as μ ¶T ¶T Gg =β   t (11) bi G = g (11) b Pr ¶x ti Pr ¶x t i where  Prt  =  0.85  is  the  turbulent  Prandtl  number  for  energy  and  gi  is  the  component  of  the  where Pr = 0.85 is the turbulent Prandtl number for energy and g is the component of the gravitational gravitational vector in the ith direction. The coefficient of thermal expansion, β, is defined as  vector in the ith direction. The coefficient of thermal expansion, b, is defined as æö 1 ¶ρ ç÷ 1 ¶$ β=-   (12) ç÷ = (12) ρ ¶T èø $ ¶T YM represents the contribution of the fluctuating dilatation in compressible turbulence to the  Y represents the contribution of the fluctuating dilatation in compressible turbulence to the overall dissipation rate and can be calculated as follows:  overall dissipation rate and can be calculated as follows: YM = 2ρε   (13) Mt Y = 2$"M (13) M t where Mt is the turbulent Mach number, defined as  where M is the turbulent Mach number, defined as M =   (14) M = (14) In Equation (14), a (= γRT ) is the speed of sound.  In Equation (14), a (= gRT) is the speed of sound. The values of turbulence model constants C1ε, C2ε, Cμ, σk and σε are as follows:  The values of turbulence model constants C , C , C ,  and  are as follows: 1" 2"  " C1ε = 1.44, C2ε = 1.92, Cμ = 0.09, σk = 1.0, σε = 1.3  C = 1.44, C = 1.92, C = 0.09,  = 1.0,  = 1.3 1" 2"  k " The value of turbulence model constant C3ε is calculated from the following equation  The value of turbulence model constant C is calculated from the following equation 3" C = tanh   (15) 3ε C = tanh (15) 3" where  v  is  the  component  of  the  flow  velocity  parallel  to  the  gravitational  vector  and  u  is  the  component of the flow velocity perpendicular to the gravitational vector.  where v is the component of the flow velocity parallel to the gravitational vector and u is the component of the flow velocity perpendicular to the gravitational vector. 3. Results and Discussion  3. Results and Discussion In  this  study,  the  numerical  model  of  a  SRU  thermal  reactor  is  constructed  using  an  In this study, the numerical model of a SRU thermal reactor is constructed using an unstructured unstructured  grid.  Five  cell  densities—10,826  cells,  187,354  cells,  342,856  cells,  683,672  cells  and  grid. Five cell densities—10,826 cells, 187,354 cells, 342,856 cells, 683,672 cells and 1,124,627 cells—are 1,124,627 cells—are tested to ensure a grid‐independent solution. The computational results show  tested to ensure a grid-independent solution. The computational results show that the sizes of the that the sizes of the corner recirculation zone in zone 1 and zone 2 and the cross‐sectional average  corner recirculation zone in zone 1 and zone 2 and the cross-sectional average temperature profiles temperature profiles obtained using the last two meshes almost coincide, with a deviation of less  obtained using the last two meshes almost coincide, with a deviation of less than 0.5%. Therefore, than 0.5%. Therefore, the mesh of 683,672 cells is used for the subsequent discussion. Figure 2 shows  the mesh of 683,672 cells is used for the subsequent discussion. Figure 2 shows the numerical model of the  numerical  model  of  the  SRU  thermal  reactor  investigated.  In  Figure  2,  the  heat  exchanger  the SRU thermal reactor investigated. In Figure 2, the heat exchanger section consists of 19 cooling section  consists  of  19  cooling  tubes,  which  have  a  diameter  of  0.5  m,  as  shown  schematically  in  tubes, which have a diameter of 0.5 m, as shown schematically in Figure 3. The heat absorption rate Figure 3. The heat absorption rate for each heat exchanger tube is 40,000 W/m  and the other walls  for each heat exchanger tube is 40,000 W/m and the other walls are adiabatic. No slip condition is are adiabatic. No slip condition is applied on all solid walls. The exit of the heat exchanger section is  applied on all solid walls. The exit of the heat exchanger section is connected to other equipment at connected to other equipment at 300 K and 1 atm by a pipe that is 1.372 m in diameter and 11.5 m in  300 K and 1 atm by a pipe that is 1.372 m in diameter and 11.5 m in length. length.  Figure 2. The numerical model for the SRU thermal reactor investigated. Appl. Sci. 2016, 6, 331  6 of 12  Appl. Sci. 2016, 6, 331 6 of 13 Figure 2. The numerical model for the SRU thermal reactor investigated.  Figure 3. An illustration of the arrangement of heat exchanger tubes. Figure 3. An illustration of the arrangement of heat exchanger tubes.  In this study, two types of oxygen supplies are investigated: an oxygen‐normal supply and an  In this study, two types of oxygen supplies are investigated: an oxygen-normal supply and oxygen‐rich  supply.  An  oxygen‐rich  supply  increases  sulfur  recovery.  The  design  conditions  an oxygen-rich supply. An oxygen-rich supply increases sulfur recovery. The design conditions (including the species compositions, the temperature, the pressure and the velocity) at the acid gas  (including the species compositions, the temperature, the pressure and the velocity) at the acid gas inlet inlet holes of zone 1 and zone 2 and at the air inlet hole are listed in Table 1. These conditions are  holes of zone 1 and zone 2 and at the air inlet hole are listed in Table 1. These conditions are practical practical  operating  conditions  that  are  used  by  a  petrochemical  corporation  in  Taiwan.  The  operating conditions that are used by a petrochemical corporation in Taiwan. The turbulence kinetic turbulence kinetic energy is 10% of the inlet mean flow kinetic energy and the turbulence dissipation  energy is 10% of the inlet mean flow kinetic energy and the turbulence dissipation rate is computed rate is computed using Equation (16).  using Equation (16). 3/2 3/4 3/2 ε= C k (16) 3μ/4 " = C l (16) where l = 0.07L and L is the hydraulic diameter.  where l = 0.07L and L is the hydraulic diameter. To validate the numerical methods used in this study, the simulation results were compared  with  available  practical  data.  The  exit  sulfur  mole  fractions  from  the  SRU  thermal  reactor  of  a  Table 1. The design conditions at the acid gas inlet holes and the air inlet hole. petrochemical corporation in Taiwan were measured with the design conditions listed in Table 1.  Oxygen-Normal Supply Oxygen-Rich Supply The  results  are  shown  in  Table  2.  It  can  be  seen  that  the  respective  deviations  are  1.3%  for  an  Location Acid Gas to Acid Gas to Acid Gas to Acid Gas to oxygen‐normal supply and 8.3% for an oxygen‐rich supply, which are acceptable from a viewpoint  Air Inlet Air Inlet of engineeringZone  appl1ications.Zone   2 Zone 1 Zone 2 Species x (%) Table 1. The design conditions at the acid gas inlet holes and the air inlet hole.  O 0 0 19.87 0 0 23.85 N 0 0 74.98 0 0 71.26 Oxygen‐Normal Supply  Oxygen‐Rich Supply  H O 7.83 4.12 5.15 4.12 27.97 4.89 Location  Acid Gas to  Acid Gas to  Acid Gas to Acid Gas to  CO 1.27 1.5 0 1.48 0 0 Air Inlet  Air Inlet  H S 82.06 Zone 1  89.88 Zone 2  0 89.9 Zone 1  39.61 Zone 2  0 CH 2.28 2.7 0 2.7 0 0 Species  x (%) C H 1.52 1.8 0 1.8 0 0 2 6 O2  0  0  19.87 0 0 23.85  NH 5.04 0 0 0 32.42 0 N2  0  0  74.98  0  0  71.26  T (K) 319.92 316.15 403.15 313.15 316.15 397.15 H2O  7.83  4.12  5.15 4.12 27.97 4.89  CO2  1.27  1.5  0  1.48  0  0  76,920 75,068 74,382 75,068 75,068 89,572 P (N/m ) H2S  82.06  89.88  0  89.9  39.61  0  V (m/s) 11.62 2.08 12.4 (Radial) 11.46 1.88 10.8 (Radial) CH4  2.28  2.7  0  2.7  0  0  34.1 (Tangential) 29.8 (Tangential) C2H6  1.52  1.8  0  1.8  0  0  Appl. Sci. 2016, 6, 331 7 of 13 To validate the numerical methods used in this study, the simulation results were compared with available practical data. The exit sulfur mole fractions from the SRU thermal reactor of a petrochemical corporation in Taiwan were measured with the design conditions listed in Table 1. The results are shown in Table 2. It can be seen that the respective deviations are 1.3% for an oxygen-normal supply and 8.3% for an oxygen-rich supply, which are acceptable from a viewpoint of engineering applications. Table 2. Comparison of the measured and simulation results. Appl. Sci. 2016, 6, 331  7 of 12  Case Practical Data Simulation Results Deviations NH3  5.04  0  0  0  32.42  0  Oxygen-normal T (K)  319.92  316.15  403.15  313.15  316.15  397.15  0.078 0.079 1.3% operation P (N/m )  76,920  75,068  74,382  75,068  75,068  89,572  V (m/s)  11.62  2.08  12.4 (Radial)  11.46  1.88  10.8 (Radial)  Oxygen-rich operation 0.084 0.091 8.3%      34.1 (Tangential)    29.8 (Tangential) Table 2. Comparison of the measured and simulation results.  To determine the effect of the fuel mass fraction, the zone 1 fuel mass fraction is defined as the mass ratio of the zone 1 acid gas to the total acid gas (acid gas to zone 1 plus acid gas to zone 2). Case  Practical Data Simulation Results Deviations  Oxygen‐normal operation  0.078  0.079  1.3%  The design zone 1 fuel mass fractions, calculated from Table 1, are 0.875 for an oxygen-normal supply Oxygen‐rich operation  0.084  0.091  8.3%  and 0.895 for an oxygen-rich supply. Figure 4 shows a comparison of the maximum and the average temperatures for a SRU thermal To determine the effect of the fuel mass fraction, the zone 1 fuel mass fraction is defined as the  reactor with different zone 1 fuel mass fractions. The air is kept at the design condition. Eight zone mass ratio of the zone 1 acid gas to the total acid gas (acid gas to zone 1 plus acid gas to zone 2). The  design zone 1 fuel mass fractions, calculated from Table 1, are 0.875 for an oxygen‐normal supply  1 fuel mass fractions are used, including 0.0625, 0.125, 0.25, 0.375, 0.5, 0.625, 0.75 and 0.875 (for an and 0.895 for an oxygen‐rich supply.  oxygen-normal case) or 0.895 (for an oxygen-rich case). The zone 1 fuel mass fractions ranging from Figure 4 shows a comparison of the maximum and the average temperatures for a SRU thermal  0.0625 to 0.895 are calculated to determine the stoichiometric fuel mass fraction which produces reactor with different zone 1 fuel mass fractions. The air is kept at the design condition. Eight zone 1  the most complete combustion and the highest temperature. Figure 4a–c show that the maximum fuel mass fractions are used, including 0.0625, 0.125, 0.25, 0.375, 0.5, 0.625, 0.75 and 0.875 (for an  oxygen‐normal case) or 0.895 (for an oxygen‐rich case). The zone 1 fuel mass fractions ranging from  temperature and the average temperature reach maximum values at a zone 1 fuel mass fraction of 0.0625 to 0.895 are calculated to determine the stoichiometric fuel mass fraction which produces the  about 0.375, which implies that the stoichiometric fuel mass fraction is around 0.375. The temperature most  complete  combustion  and  the  highest  temperature.  Figure  4a–c  show  that  the  maximum  decreases when the zone 1 fuel mass fraction deviates from the stoichiometric fuel mass fraction. temperature and the average temperature reach maximum values at a zone 1 fuel mass fraction of  In Figure 4c, the average temperature of the thermal reactor represents the average temperature about  0.375,  which  implies  that  the  stoichiometric  fuel  mass  fraction  is  around  0.375.  The  temperature  decreases  when  the  zone 1  fuel  mass  fraction  deviates  from  the  stoichiometric  fuel  for both zone 1 and zone 2. In a practical SRU thermal reactor, high temperature can damage the mass fraction.  In Figure 4c, the average temperature of the thermal reactor represents the average  refractory. For example, the zone 1 corner has a suddenly expanded geometry and a recirculation zone temperature for both zone 1 and zone 2. In a practical SRU thermal reactor, high temperature can  forms in this region. The temperature in this region can exceed the maximum service temperature damage  the  refractory.  For  example,  the zone 1 corner has a suddenly expanded geometry and a  for the refractory and cause collapse or deformation of the zone 1 corner. The suggested maximum recirculation zone forms in this region. The temperature in this region can exceed the maximum  service temperature for the refractory and cause collapse or deformation of the zone 1 corner. The  service temperature for the refractory varies for different manufacturers. A typical value is 1700 C suggested  maximum  service  temperature  for  the  refractory  varies  for  different  manufacturers.  A  (1973 K). Figure 4a shows that the maximum temperatures for all oxygen-rich cases exceed the typical  value  is  1700  °C  (≈1973  K).  Figure  4a  shows  that  the  maximum  temperatures  for  all  suggested maximum service temperature. However, for oxygen-normal cases, when the zone 1 fuel oxygen‐rich  cases  exceed  the  suggested  maximum  service  temperature.  However,  for  mass fraction is less than 0.3 or greater than 0.6, the maximum temperature is lower than the suggested oxygen‐normal  cases,  when  the  zone 1 fuel mass fraction is less than 0.3 or greater than 0.6, the  maximum  temperature  is  lower  than  the  suggested  maximum  service  temperature.  The  current  maximum service temperature. The current design condition for an oxygen-normal operation (zone 1 design condition for an oxygen‐normal operation (zone 1 fuel mass fraction of 0.875), listed in Table  fuel mass fraction of 0.875), listed in Table 1, allows an acceptable temperature field. 1, allows an acceptable temperature field.  normal oxygen rich oxygen normal oxygen rich oxygen 0.2 0.4 0.6 0.8 0.2 0.4 0.6 0.8 zone 1 fuel mass fraction zone 1fuel massfraction (a)  (b) Figure 4. Cont. Tmax (K) Tavg (K) 1 Appl. Sci. 2016, 6, 331  8 of 12  Appl. Sci. 2016, 6, 331 8 of 13 Appl. Sci. 2016, 6, 331  8 of 12  normal oxygen rich oxygen normal oxygen rich oxygen 0.2 0.4 0.6 0.8 0.z 2one 1 fu 0.e 4l mass f 0r .6action 0.8 zone 1 fuel mass fraction (c) (c) Figure 4. A comparison of the maximum and average temperatures for a SRU thermal reactor with  Figure 4. A comparison of the maximum and average temperatures for a SRU thermal reactor with  Figure 4. A comparison of the maximum and average temperatures for a SRU thermal reactor with different zone 1 fuel mass fractions: (a) Maximum temperature; (b) The average temperature in zone  different zone 1 fuel mass fractions: (a) Maximum temperature; (b) The average temperature in zone  different zone 1; (c) 1The fuel  ave mass rage temperature fractions: in (a the ) Maximum  thermal reactor. temperatur   e; (b) The average temperature in zone 1; 1; (c) The average temperature in the thermal reactor.  (c) The average temperature in the thermal reactor. Figure  5 shows a  comparison  of  the  cross‐sectional  average  temperature for a SRU  thermal  Figure  5 shows a  comparison  of  the  cross‐sectional  average  temperature for a SRU  thermal  reactor with different zone 1 fuel mass fractions. It is seen that the average temperature in zone 2 is  Figurehigher 5 shows  than a the comparison  average tempera of the turecr  inoss-sectional  zone 1 for zoneaverage  1 fuel matemperatur ss fractions ofe 0. for 062a 5 and SRU 0.12 thermal 5. A  reactor reactor with different zone 1 fuel mass fractions. It is seen that the average temperature in zone 2 is  zone  1  fuel  mass  fraction  lower  than  the  stoichiometric  fuel  mass  fraction  (e.g.,  0.0625  or  0.125)  with different zone 1 fuel mass fractions. It is seen that the average temperature in zone 2 is higher higher than the average temperature in zone 1 for zone 1 fuel mass fractions of 0.0625 and 0.125. A  produces  a  lower  zone  1  temperature  and  this  helps  protect  zone  1;  however,  the  zone  2  than the average temperature in zone 1 for zone 1 fuel mass fractions of 0.0625 and 0.125. A zone zone  1  fuel  mass  fraction  lower  than  the  stoichiometric  fuel  mass  fraction  (e.g.,  0.0625  or  0.125)  temperature is increased because a lower zone 1 fuel mass fraction represents a lower zone 1 fuel  1 fuel mass fraction lower than the stoichiometric fuel mass fraction (e.g., 0.0625 or 0.125) produces produces  a  lower  zone  1  temperature  and  this  helps  protect  zone  1;  however,  the  zone  2  supply and a higher zone 2 fuel supply. It is worth noting that a high zone 2 average temperature  a lower zone can 1da temperatur mage the heat e exch andanger this tubes. helps For pr otect larger zone zone 11; fuel however  mass fract , the ions, zone  the av 2er temperatur age temperateure is increased temperature is increased because a lower zone 1 fuel mass fraction represents a lower zone 1 fuel  in  zone  1  is  higher.  An  oxygen‐rich  supply  also  enhances  the  chemical  reaction  and  produces  a  because a lower zone 1 fuel mass fraction represents a lower zone 1 fuel supply and a higher zone 2 fuel supply and a higher zone 2 fuel supply. It is worth noting that a high zone 2 average temperature  higher  temperature.  The  temperature  decreases  abruptly  across  the  choke  ring  because  thermal  supply. It is worth noting that a high zone 2 average temperature can damage the heat exchanger tubes. can damage the heat exchanger tubes. For larger zone 1 fuel mass fractions, the average temperature  energy  is  converted  into  kinetic  energy  when  the  flow  is  accelerated.  It  is  also  seen  that  the  For larger zone 1 fuel mass fractions, the average temperature in zone 1 is higher. An oxygen-rich in  zone  1  is  higher.  An  oxygen‐rich  supply  also  enhances  the  chemical  reaction  and  produces  a  temperature is more uniform for the design zone 1 fuel mass fractions (0.875 for the oxygen‐normal  supply also case enhances  and 0.895 for the the chemical  oxygen‐rich reaction  case). This and  canpr  aloduces so be observed a higher  from temperatur Figure 6 whice. h sh The ows temperatur the  e higher  temperature.  The  temperature  decreases  abruptly  across  the  choke  ring  because  thermal  temperature profile on the symmetric plane for a SRU  thermal  reactor  with  different  zone 1 fuel  decreases abruptly across the choke ring because thermal energy is converted into kinetic energy when energy  is  converted  into  kinetic  energy  when  the  flow  is  accelerated.  It  is  also  seen  that  the  mass fractions.  the flow is accelerated. It is also seen that the temperature is more uniform for the design zone 1 fuel temperature is more uniform for the design zone 1 fuel mass fractions (0.875 for the oxygen‐normal  mass fractions (0.875 for the oxygen-normal case and 0.895 for the oxygen-rich case). This can also case and 0.895 for the oxygen‐rich case). This can also be observed from Figure 6 which shows the  be observed from Figure 6 which shows the temperature profile on the symmetric plane for a SRU temperature profile on the symmetric plane for a SRU  thermal  reactor  with  different  zone 1 fuel  thermal reactor with different zone 1 fuel mass fractions. mass fractions.  0.0625 2000 X= 1fuel 0.0625 X= 1fuel 0.125 X= 1fuel 1200 0.125 X= 1fuel 0.25 X= 1fuel 0.25 X= 1fuel 0.375 X= 1fuel 0.375 X= 1fuel 0.5 X= 1fuel 1000 0.5 X= 1fuel 0.625 1000 1600 X= 1fuel 0.625 X= 1fuel 0.75 X= 1fuel 1600 0.75 X= 1fuel 0.875 1fuel X= 800 0.875 X= 1fuel 0 246 8 10 12 02468 10 12 X(m) 0.0625 X(m) X= 1fuel 0.0625 X= 1fuel 0.125 X= 1fuel 1200 0.125 X= 1fuel 0.25 X= 1fuel (a)  (b)  0.25 X= 1fuel 0.375 X= 1fuel 0.375 X= 1fuel 0.5 Figure 5. A comparison of the cross‐sectional average temperature for a SRU thermal reactor with  X= 1fuel 1000 0.5 X= 1fuel 0.625 1000 X= 1fuel different zone 1 fuel mass fractions: (a) Oxygen‐normal operation; (b) Oxygen‐rich operation.  0.625 X= 1fuel 0.75 X= 1fuel 0.75 X= 1fuel 0.875 1fuel X= 800 0.875 X= 1fuel 0 246 8 10 12 02468 10 12 X(m) X(m) (a)  (b)  Figure 5. A comparison of the cross‐sectional average temperature for a SRU thermal reactor with  Figure 5. A comparison of the cross-sectional average temperature for a SRU thermal reactor with different zone 1 fuel mass fractions: (a) Oxygen‐normal operation; (b) Oxygen‐rich operation.  different zone 1 fuel mass fractions: (a) Oxygen-normal operation; (b) Oxygen-rich operation. T(K) T(K) Tavg (K) Tavg (K) T(K) T(K) Appl. Sci. 2016, 6, 331 9 of 13 Appl. Sci. 2016, 6, 331  9 of 12  Figure 6 shows that the maximum temperature occurs at the heat exchanger tubes for zone 1 fuel Figure 6 shows that the maximum temperature occurs at the heat exchanger tubes for zone 1  mass fractions of 0.0625 and 0.125. Because the heat exchanger tubes are made of carbon steel, which fuel mass fractions of 0.0625 and 0.125. Because the heat exchanger tubes are made of carbon steel,  has awhi maximum ch has a ma service ximum temperatur  service tempera e thatture is lower that is than lower that  than for tha the t for refractory  the refracof tory the ofthermal  the therm reactor al  , reactor, the tubes can be damaged and care must be taken when a low zone 1 fuel mass fraction is  the tubes can be damaged and care must be taken when a low zone 1 fuel mass fraction is used. used.  (a)      (b)  (c)  (d)  (e)  Figure 6. Cont. Appl. Sci. 2016, 6, 331 10 of 13 Appl. Sci. 2016, 6, 331  10 of 12  (f)  (g)  (h)  Figure 6. The temperature profile on the symmetric plane for a SRU thermal reactor with different  Figure 6. The temperature profile on the symmetric plane for a SRU thermal reactor with different zone 1 fuel mass fractions: (a) Zone 1 fuel mass fraction of 0.0625; (b) Zone 1 fuel mass fraction of  zone 1 fuel mass fractions: (a) Zone 1 fuel mass fraction of 0.0625; (b) Zone 1 fuel mass fraction of 0.125; (c) Zone 1 fuel mass fraction of 0.375; (d) Zone 1 fuel mass fraction of 0.875; (e) Zone 1 fuel  0.125; (c) Zone 1 fuel mass fraction of 0.375; (d) Zone 1 fuel mass fraction of 0.875; (e) Zone 1 fuel mass fraction of 0.0625; (f) Zone 1 fuel mass fraction of 0.125; (g) Zone 1 fuel mass fraction of 0.375; (h)  mass fraction of 0.0625; (f) Zone 1 fuel mass fraction of 0.125; (g) Zone 1 fuel mass fraction of 0.375; Zone 1 fuel mass fraction of 0.895  (note:(a–d) are  for  oxygen‐normal  operations  and  (e–h) are  for  (h) Zone 1 fuel mass fraction of 0.895 (note: (a–d) are for oxygen-normal operations and (e–h) are for oxygen‐rich operations).  oxygen-rich operations). Figure 7 shows a comparison of the sulfur and sulfur dioxide concentrations at the exit for a  SRU thermal reactor with different zone 1 fuel mass fractions for an oxygen‐normal operation. It is  Figure 7 shows a comparison of the sulfur and sulfur dioxide concentrations at the exit for a SRU seen  that  the  zone  1  fuel  mass  fraction  has  a  minor  influence  on  the  sulfur  and  sulfur  dioxide  thermal reactor with different zone 1 fuel mass fractions for an oxygen-normal operation. It is seen concentrations  at  the  exit.  The  sulfur  concentration  at  the  exit  is  about  0.08.  An  oxygen‐rich  that the zone 1 fuel mass fraction has a minor influence on the sulfur and sulfur dioxide concentrations operation shows a similar tendency with the sulfur concentration at the exit being around 0.09.  at the exit. The sulfur concentration at the exit is about 0.08. An oxygen-rich operation shows a similar From the above discussion, it is seen that the major influence of the zone 1 fuel mass fraction is  tendency with the sulfur concentration at the exit being around 0.09. on the temperature. The current design conditions listed in Table 1 allow an acceptable temperature  From the above discussion, it is seen that the major influence of the zone 1 fuel mass fraction is field for an oxygen‐normal operation. However, for an oxygen‐rich operation, the local maximum  on the temperature. The current design conditions listed in Table 1 allow an acceptable temperature temperature  exceeds  the  suggested  maximum  service  temperature,  although  the  average  field for an oxygen-normal operation. However, for an oxygen-rich operation, the local maximum temperature is acceptable. Therefore, the high temperature regions, such as the zone 1 corner, must  temperatur be  inspeect exceeds ed  very the   casuggested refully  dur maximum ing  the  ann service ual  ma temperatur intenance  e,peri although od  if  there the average   are  oxygen temperatur ‐rich  e operations.  is acceptable. Therefore, the high temperature regions, such as the zone 1 corner, must be inspected very carefully during the annual maintenance period if there are oxygen-rich operations. Appl. Sci. 2016, 6, 331 11 of 13 Appl. Sci. 2016, 6, 331  11 of 12  0.1 0.1 0.08 0.08 0.06 0.06 0.04 0.04 0.02 0.02 0.2 0.4 0.6 0.8 0.2 0.4 0.6 0.8 zone 1 fuel mass fraction zone 1 fuel mass fraction (a)  (b)  Figure 7. A comparison of the (a) sulfur and (b) sulfur dioxide concentrations at the exit for a SRU  Figure 7. A comparison of the (a) sulfur and (b) sulfur dioxide concentrations at the exit for a SRU thermal reactor with different zone 1 fuel mass fractions.  thermal reactor with different zone 1 fuel mass fractions. 4. Conclusions  4. Conclusions This paper numerically investigates the effect of the fuel mass fraction on the combustion and  This paper numerically investigates the effect of the fuel mass fraction on the combustion and fluid flow in a SRU thermal reactor to provide a guideline for adjusting the fuel mass fraction to  fluid flow in a SRU thermal reactor to provide a guideline for adjusting the fuel mass fraction to reduce  the  temperature  inside  the  thermal  reactor  and  to  allow  an  acceptable  sulfur  recovery.  reduce the temperature inside the thermal reactor and to allow an acceptable sulfur recovery. Practical Practical  operating  conditions  for  a  petrochemical  corporation  in  Taiwan  are  used  as  the  design  operating conditions for a petrochemical corporation in Taiwan are used as the design conditions. conditions. To determine a suitable fuel mass fraction for operation, a detailed numerical simulation  To determine a suitable fuel mass fraction for operation, a detailed numerical simulation should should  be  performed  first  to  find  the  stoichiometric  fuel  mass  fraction  which  produces  the  most  be performed first to find the stoichiometric fuel mass fraction which produces the most complete complete combustion and the highest temperature. This stoichiometric fuel mass fraction should be  combustion and the highest temperature. This stoichiometric fuel mass fraction should be avoided avoided because the high temperature could damage the zone 1 corner or the choke ring. A higher  because the high temperature could damage the zone 1 corner or the choke ring. A higher fuel mass fuel  mass  fraction  (i.e.,  fuel‐rich  or  air‐lean  condition)  is  more  suitable  because  it  can  avoid  fraction (i.e., fuel-rich or air-lean condition) is more suitable because it can avoid deteriorations of deteriorations of both zone 1 and the heat exchanger tubes. Although a lower fuel mass fraction (i.e.,  both zone 1 and the heat exchanger tubes. Although a lower fuel mass fraction (i.e., fuel-lean or fuel‐lean or air‐rich condition) can avoid deterioration of zone 1, the heat exchanger tubes may be  air-rich condition) can avoid deterioration of zone 1, the heat exchanger tubes may be damaged. damaged. The present simulation results show that the current design condition is a fuel‐rich (or  The present simulation results show that the current design condition is a fuel-rich (or air-lean) air‐lean)  condition  and  allows  an  acceptable  thermal  reactor  temperature  for  an  oxygen‐normal  condition and allows an acceptable thermal reactor temperature for an oxygen-normal operation. operation.  However,  for  an  oxygen‐rich  operation,  the  local  maximum  temperature  exceeds  the  However, for an oxygen-rich operation, the local maximum temperature exceeds the suggested suggested maximum service temperature, although the average temperature is acceptable. The high  maximum service temperature, although the average temperature is acceptable. The high temperature temperature regions, such as the zone 1 corner, must be inspected very carefully during the annual  regions, such as the zone 1 corner, must be inspected very carefully during the annual maintenance maintenance period if there are oxygen‐rich operations. For lower zone 1 fuel mass fractions such as  period if there are oxygen-rich operations. For lower zone 1 fuel mass fractions such as 0.0625 and 0.0625 and 0.125, the average temperature in zone 2 is higher than the average temperature in zone 1,  0.125, the average temperature in zone 2 is higher than the average temperature in zone 1, which could which  could  cause  damage  to  the  downstream  heat  exchanger  tubes.  If  a  low  zone  1  fuel  mass  cause damage to the downstream heat exchanger tubes. If a low zone 1 fuel mass fraction is used to fraction is used to produce a lower zone 1 temperature, the temperatures in zone 2 and the heat  produce a lower zone 1 temperature, the temperatures in zone 2 and the heat exchanger section must exchanger section must be monitored closely and the zone 2 wall and heat exchanger tubes must be  be monitored closely and the zone 2 wall and heat exchanger tubes must be inspected very carefully inspected very carefully during the annual maintenance period.  during the annual maintenance period. Acknowledgments: The author gratefully acknowledges the grant support from the Ministry of Science and  Acknowledgments: The author gratefully acknowledges the grant support from the Ministry of Science and Technology, Taiwan, R.O.C., under contract MOST 104‐2221‐E‐150‐032. The author would also like to express  Technology, Taiwan, R.O.C., under contract MOST 104-2221-E-150-032. The author would also like to express thanks for the useful data and constructive suggestions provided by the Formosa Petrochemical Corporation in  thanks for the useful data and constructive suggestions provided by the Formosa Petrochemical Corporation in Taiwan. Taiwan.  Conflicts of Interest: The author declares no conflict of interest. Conflicts of Interest: The author declares no conflict of interest.  Abbreviations  The following abbreviations are used in this manuscript:  Cμ  turbulence model constant (=0.09)  2 2 k  turbulence kinetic energy (m /s )  Sulfur mole fraction SO mole fraction 2 Appl. Sci. 2016, 6, 331 12 of 13 Abbreviations The following abbreviations are used in this manuscript: C turbulence model constant (=0.09) 2 2 k turbulence kinetic energy (m /s ) L hydraulic diameter (m) l characteristic length (m) P pressure (N/m ) T temperature (K) V velocity (m/s) XYZ cartesian coordinates with origin at the centroid of the burner inlet (m) X zone 1 fuel mass fraction 1fuel x mole fraction (%) Greek symbols 2 3 " turbulence dissipation rate (m /s ) Subscripts avg average avg1 zone 1 average avg2 zone 2 average max maximum References 1. Adewale, R.; Salem, D.J.; Berrouk, A.S.; Dara, S. Simulation of hydrogen production from thermal decomposition of hydrogen sulfide in sulfur recovery units. J. Clean. Prod. 2016, 112, 4815–4825. [CrossRef] 2. Chardonneaua, M.; Ibrahim, S.; Gupta, A.K.; AlShoaibi, A. Role of toluene and carbon dioxide on sulfur recovery efficiency in a Claus process. Energy Procedia 2015, 75, 3071–3075. [CrossRef] 3. Selim, H.; Gupta, A.K.; AlShoaibi, A. Effect of reaction parameters on the quality of captured sulfur in Claus process. Appl. Energ. 2013, 104, 772–776. [CrossRef] 4. Monnery, W.D.; Hawboldt, K.A.; Pollock, A.; Svrcek, W.Y. New experimental data and kinetic rate expression for the Claus reaction. Chem. Eng. Sci. 2000, 55, 5141–5148. [CrossRef] 5. Yeh, C.L. Effect of choke ring dimension on thermal and fluid flow in a SRU thermal reactor. Trans. Can. Soc. Mech. Eng. submitted for publication, 2016. 6. Yeh, C.L. Effect of choke ring position on thermal and fluid flow in a SRU thermal reactor. Int. J. Mech. Eng. Robot. Res. 2015, 4, 273–277. 7. Yeh, C.L. Effect of streamlining geometry on thermal and fluid flow in a SRU thermal reactor. In Proceedings of International Multi-Conference on Engineering and Technology Innovation 2015 (IMETI2015), Kaohsiung, Taiwan, 30 October–3 November 2015; J5043. 8. Yeh, C.L. Numerical analysis of the effects of streamlining geometry and a vector wall on the thermal and fluid flow in a SRU thermal reactor. Trans. Can. Soc. Mech. Eng. submitted for publication, 2016, 40. 9. Yeh, C.L. Effect of a vector wall on the thermal and fluid flow in a SRU thermal reactor. In Proceedings of International Multi-Conference on Engineering and Technology Innovation 2015 (IMETI2015), Kaohsiung, Taiwan, 30 October–3 November 2015; J5042. 10. Patankar, S.V. Numerical Heat Transfer and Fluid Flows; McGraw-Hill: New York, NY, USA, 1980. 11. Launder, B.E.; Spalding, D.B. Lectures in Mathematical Models of Turbulence; Academic Press: London, UK, 1972. 12. Siegel, R.; Howell, J.R. Thermal Radiation Heat Transfer; Hemisphere Publishing Corporation: Washington, DC, USA, 1992. 13. Sivathanu, Y.R.; Faeth, G.M. Generalized state relationships for scalar properties in non-premixed hydrocarbon/air flames. Combust. Flame 1990, 82, 211–230. [CrossRef] 14. Launder, B.E.; Spalding, D.B. The numerical computation of turbulent flows. Comput. Method Appl. Mech. Eng. 1974, 3, 269–289. [CrossRef] Appl. Sci. 2016, 6, 331 13 of 13 15. Yeh, C.L. Numerical analysis of the combustion and fluid flow in a carbon monoxide boiler. Int. J. Heat Mass Trans. 2013, 59, 172–190. [CrossRef] 16. Batchelor, G.K. An Introduction to Fluid Dynamics; Cambridge University Press: Cambridge, UK, 1967. © 2016 by the author; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/). http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Applied Sciences Multidisciplinary Digital Publishing Institute

The Effect of Fuel Mass Fraction on the Combustion and Fluid Flow in a Sulfur Recovery Unit Thermal Reactor

Yeh, Chun-Lang
Applied Sciences , Volume 6 (11) – Nov 2, 2016
Download PDF
13 pages

Loading next page...
 
/lp/multidisciplinary-digital-publishing-institute/the-effect-of-fuel-mass-fraction-on-the-combustion-and-fluid-flow-in-a-Hh4bc3tufZ
Publisher
Multidisciplinary Digital Publishing Institute
Copyright
© 1996-2019 MDPI (Basel, Switzerland) unless otherwise stated
ISSN
2076-3417
DOI
10.3390/app6110331
Publisher site
See Article on Publisher Site

Abstract

applied sciences Article The Effect of Fuel Mass Fraction on the Combustion and Fluid Flow in a Sulfur Recovery Unit Thermal Reactor Chun-Lang Yeh Department of Aeronautical Engineering, National Formosa University, Huwei, Yunlin 632, Taiwan; clyeh@nfu.edu.tw; Tel.: +886-5-6315527 Academic Editor: Yuyuan Zhao Received: 7 September 2016; Accepted: 26 October 2016; Published: 2 November 2016 Abstract: Sulfur recovery unit (SRU) thermal reactors are negatively affected by high temperature operation. In this paper, the effect of the fuel mass fraction on the combustion and fluid flow in a SRU thermal reactor is investigated numerically. Practical operating conditions for a petrochemical corporation in Taiwan are used as the design conditions for the discussion. The simulation results show that the present design condition is a fuel-rich (or air-lean) condition and gives acceptable sulfur recovery, hydrogen sulfide (H S) destruction, sulfur dioxide (SO ) emissions and thermal 2 2 reactor temperature for an oxygen-normal operation. However, for an oxygen-rich operation, the local maximum temperature exceeds the suggested maximum service temperature, although the average temperature is acceptable. The high temperature region must be inspected very carefully during the annual maintenance period if there are oxygen-rich operations. If the fuel mass fraction to the zone ahead of the choke ring (zone 1) is 0.0625 or 0.125, the average temperature in the zone behind the choke ring (zone 2) is higher than the zone 1 average temperature, which can damage the downstream heat exchanger tubes. If the zone 1 fuel mass fraction is reduced to ensure a lower zone 1 temperature, the temperature in zone 2 and the heat exchanger section must be monitored closely and the zone 2 wall and heat exchanger tubes must be inspected very carefully during the annual maintenance period. To determine a suitable fuel mass fraction for operation, a detailed numerical simulation should be performed first to find the stoichiometric fuel mass fraction which produces the most complete combustion and the highest temperature. This stoichiometric fuel mass fraction should be avoided because the high temperature could damage the zone 1 corner or the choke ring. A higher fuel mass fraction (i.e., fuel-rich or air-lean condition) is more suitable because it can avoid deteriorations of both zone 1 and heat exchanger tubes. Although a lower fuel mass fraction (i.e., fuel-lean or air-rich condition) can avoid deterioration of zone 1, the heat exchanger tubes may be damaged. This paper provides a guideline for adjusting the fuel mass fraction to reduce the high temperature inside the thermal reactor and to ensure an acceptable sulfur recovery. Keywords: SRU thermal reactor; fuel mass fraction; sulfur recovery; H S destruction; SO emission; 2 2 thermal reactor temperature 1. Introduction Desulfurization is very important in the petroleum refining process because oxysulfides from the petroleum refining process are one of the major sources of air pollution. The most frequently used desulfurization process is the Claus process, which converts the H S in natural gas or crude oil into sulfur elements, which reduces the formation of oxysulfides. A sulfur recovery unit (SRU) thermal reactor is the most important equipment in a sulfur plant. It converts the NH , H S and hydrocarbons 3 2 in the reactants into sulfur. Most of the sulfur elements are recovered from the SRU thermal reactor. Appl. Sci. 2016, 6, 331; doi:10.3390/app6110331 www.mdpi.com/journal/applsci Appl. Sci. 2016, 6, 331 2 of 13 The first section of a SRU that uses the Claus process is composed of a burner, a thermal reactor and a waste heat exchanger. The configuration and dimensions of the first section of a SRU for a typical petroleum refinery are shown in Figure 1. The main chemical reactions in a SRU thermal reactor include a thermal step and a catalytic step. Appl. Sci. 2016, 6, 331  2 of 12  2H S + 3O ! 2SO + 2H O 2 2 2 2 hydrocarbons in the reactants into sulfur. Most of the sulfur elements are recovered from the SRU  thermal  reactor.  The first section of a SRU that uses the Claus process is composed of a burner, a  2H S + SO ! 3S + 2H O 2 2 2 thermal reactor and a waste heat exchanger. The configuration and dimensions of the first section of  a SRU for a typical petroleum refinery are shown in Figure 1.  In addition to the above reactions and the combustion of hydrocarbon fuels, other chemical The main chemical reactions in a SRU thermal reactor include a thermal step and a catalytic  reactions taking place in a SRU thermal reactor are step.  2H2S + 3O2 →  2SO2 + 2H2O  2NH + 1.5O ! N + 3H O 3 2 2 2 2H2S + SO2 →  3S + 2H2O  In  addition  to  the  abov 2NH e reacti+ ons SO   and ! the N combustion + H S  + of 2H hydro Ocarbon fuels, other  chemical  3 2 2 2 2 reactions taking place in a SRU thermal reactor are  The interior of a SRU thermal reactor is divided into two zones by a choke ring, in order to increase 2NH3 + 1.5O2  →  N2 + 3H2O  the residence time and enhance the chemical reaction. The zone ahead of the choke ring is called zone 2NH3 + SO2 →  N2 + H2S + 2H2O  1 and the zone behind the choke ring is called zone 2. The inner surface of a SRU thermal reactor The interior of a SRU thermal reactor is divided into two zones by a choke ring, in order to  is fabricated using a refractory to protect its walls because a SRU thermal reactor operates at very increase the residence time and enhance the chemical reaction. The zone ahead of the choke ring is  high temperatures. SRU thermal reactors are negatively affected by high temperature operations called  zone  1  and  the  zone  behind  the  choke  ring  is  called  zone  2.  The  inner  surface  of  a  SRU  thermal reactor is fabricated using a refractory to protect its walls because a SRU thermal reactor  because high temperature can damage the refractory and the heat exchanger tubes. Therefore, operates  at  very  high  temperatures.  SRU  thermal  reactors  are  negatively  affected  by  high  the operating temperature range that is suggested by the manufacturers for the operation of a SRU temperature  operations  because  high  temperature  can  damage  the  refractory  and  the  heat  thermal reactor exchanger must  be tube strictly s.  Thereadher fore,  the ed  to. operating  temperature  range  that  is  suggested  by  the  manufacturers for the operation of a SRU thermal reactor must be strictly adhered to.  (a)  (b)  Figure 1. The configuration and dimensions of the first section of the SRU for a typical petroleum refinery: (a) The overall view; (b) Enlarged view for the burner section. There have been theoretical and experimental studies of SRU thermal reactors. Adewale et al. [1] studied the thermal decomposition of H S into hydrogen and sulfur using a process simulator. Using 2 Appl. Sci. 2016, 6, 331 3 of 13 the net fraction of the acid gas feed to the cracking coils as the controlling parameter, its effect on hydrogen production, the thermal reactor ’s energy requirement, the stability of the burner flame, steam production, the temperature of a Claus reactor and sulfur recovery of the primary SRU was studied. Chardonneaua et al. [2] presented experimental and simulation results for the addition of various amounts of toluene or carbon dioxide/toluene mixtures into the H S gas stream. The results show that there is a decrease in the conversion efficiency when the amount of toluene or carbon dioxide/toluene added to the H S gas stream increases. The role of the reactor ’s operating temperature was also studied. The addition of toluene increases the optimum reactor temperature for enhanced sulfur recovery, but the presence of CO reduces the optimum operating temperature. Selim et al. [3] examined the quality of sulfur deposits that were collected from H S combustion. Sulfur deposits from H S combustion under various conditions were captured and analyzed using X-ray powder diffraction and laser-induced breakdown spectroscopy diagnostics. Monnery et al. [4] experimentally studied the reaction between H S and SO using practical Claus thermal reactor temperatures between 850 and 2 2 1150 C and residence times between 0.05 and 1.2 s. The kinetic data obtained were used to develop a new reaction rate expression. Our experience of operating a practical SRU thermal reactor in Taiwan shows that the refractories at the zone 1 corner and the choke ring are the parts of the thermal reactor that experience the greatest deterioration. The zone 1 corner has a suddenly expanded geometry and a recirculation zone forms behind it. The temperature at the zone 1 corner can exceed the maximum service temperature of the refractory and cause collapse or deformation. The choke ring is subjected to a bending moment from the rapid combustion gas stream and can collapse or deform. Several methods to alleviate this problem have been assessed, including (1) changing the choke ring dimensions [5]; (2) changing the choke ring position [6]; (3) modifying the geometry of the zone 1 corner to a streamlined geometry [7,8]; and (4) replacing the choke ring by a vector wall [9]. However, these methods are either expensive or require an overhaul of the SRU thermal reactor. This paper presents a new, easier and more economical method. The effect of the fuel mass fraction on the combustion and fluid flow in a SRU thermal reactor is investigated numerically. This paper provides a guideline for adjusting the fuel mass fraction to reduce the high temperature inside a thermal reactor and to ensure an acceptable sulfur recovery. Practical operating conditions from a petrochemical corporation in Taiwan were used as the design conditions for the discussion. 2. Numerical Methods and Physical Models In this study, the FLUENT commercial code is used to simulate the reaction and fluid flow in a SRU thermal reactor. The SIMPLE algorithm by Patankar [10] is used to solve the governing equations. The discretizations of convection terms and diffusion terms are respectively performed using the power-law scheme and the central difference scheme. In terms of physical models, considering the accuracy and stability of the models and the evaluations of other researchers, the standard k-" model [11], the P-1 radiation model [12] and the non-premixed combustion model with a -type probability density function [13] are respectively used for the turbulence, radiation and combustion simulations. The standard wall functions [14] are used to resolve the flow quantities (velocity, the temperature, and the turbulence quantities) at the near-wall regions. For the steady-state three-dimensional flow field with the chemical reaction in this study, the governing equations for the continuity equation, momentum equation and turbulence equation (k-" model) are described below. The governing equations for the energy equation, radiation equation, combustion equation and convergence criterion are described in a previous study by the author [15]. Continuity equation: r v = 0 (1) Appl. Sci. 2016, 6, 331 4 of 13 Momentum equation: In an inertial reference frame, the momentum equation [16] may be expressed as !! ! r ($ v v ) = rp +r  + $ g + F (2) where F is the possible source term of the momentum equation which may include the body force caused by the interaction with dispersed phases or the extra force terms caused by physical models or numerical schemes. In Equation (2), the stress tensor t may be expressed as ! ! 2 ! =  r v +r v r v I (3) where I is the unit tensor, the superscript T represents the transpose operator, and the third term on the right-hand side of the equal sign represents the volume dilation effect. By the long-time averaged method, Equation (2) may be converted to Reynolds-averaged Navier-Stokes (RANS) equations as follows: " !# ¶u ¶ ¶p ¶ ¶u j 2 ¶u ¶ i l $u u = +  +  + $u 0u 0 (4) i j ij i j ¶x ¶x ¶x ¶x ¶x 3 ¶x ¶x j i j j i l j In Equation (4), the turbulent stresses (Reynolds stresses) may be related to the mean velocity gradient by the Boussinesq hypothesis as follows: ¶u ¶u j 2 ¶u i k $u 0u 0 =  + $k +   (5) i j ij t t ¶x ¶x 3 ¶x j i k In Equation (5), the turbulent (or eddy) viscosity may be expressed in terms of the turbulence kinetic energy (k) and turbulence dissipation rate (") as: = C $ (6) As mentioned above, k and # are modeled by the k-" model, as described below. k-" model equation: The turbulence kinetic energy (k) and turbulence dissipation rate (") may be solved from their transport equations [11] as follows: " # ¶ ¶  ¶k ($u k) =  + + G + G $" Y + S (7) i k b k ¶x ¶x  ¶x i j j " # ¶ ¶  ¶" " " ($u #) =  + + C (G + C G ) C $ + S (8) 3" 2" " i 1" k b ¶x ¶x  ¶x k k i j j In Equations (7) and (8), G represents the turbulence kinetic energy production and may be expressed as: G =  S (9) k t where S is the modulus of the mean rate-of-strain tensor defined as S  2S S (10) ij ij ¶u ¶u j In Equation (10), S = + . ij ¶x ¶x j i Appl. Sci. 2016, 6, 331  5 of 12  æö ¶u ¶u 1 j Appl. Sci. 2016, 6, 331 5 of 13 ç÷ i In Equation (10),  .  S=+ ij ç÷ ç÷ 2 ¶¶xx ji èø Gb is the generation of turbulence kinetic energy due to buoyancy, calculated as  G is the generation of turbulence kinetic energy due to buoyancy, calculated as μ ¶T ¶T Gg =β   t (11) bi G = g (11) b Pr ¶x ti Pr ¶x t i where  Prt  =  0.85  is  the  turbulent  Prandtl  number  for  energy  and  gi  is  the  component  of  the  where Pr = 0.85 is the turbulent Prandtl number for energy and g is the component of the gravitational gravitational vector in the ith direction. The coefficient of thermal expansion, β, is defined as  vector in the ith direction. The coefficient of thermal expansion, b, is defined as æö 1 ¶ρ ç÷ 1 ¶$ β=-   (12) ç÷ = (12) ρ ¶T èø $ ¶T YM represents the contribution of the fluctuating dilatation in compressible turbulence to the  Y represents the contribution of the fluctuating dilatation in compressible turbulence to the overall dissipation rate and can be calculated as follows:  overall dissipation rate and can be calculated as follows: YM = 2ρε   (13) Mt Y = 2$"M (13) M t where Mt is the turbulent Mach number, defined as  where M is the turbulent Mach number, defined as M =   (14) M = (14) In Equation (14), a (= γRT ) is the speed of sound.  In Equation (14), a (= gRT) is the speed of sound. The values of turbulence model constants C1ε, C2ε, Cμ, σk and σε are as follows:  The values of turbulence model constants C , C , C ,  and  are as follows: 1" 2"  " C1ε = 1.44, C2ε = 1.92, Cμ = 0.09, σk = 1.0, σε = 1.3  C = 1.44, C = 1.92, C = 0.09,  = 1.0,  = 1.3 1" 2"  k " The value of turbulence model constant C3ε is calculated from the following equation  The value of turbulence model constant C is calculated from the following equation 3" C = tanh   (15) 3ε C = tanh (15) 3" where  v  is  the  component  of  the  flow  velocity  parallel  to  the  gravitational  vector  and  u  is  the  component of the flow velocity perpendicular to the gravitational vector.  where v is the component of the flow velocity parallel to the gravitational vector and u is the component of the flow velocity perpendicular to the gravitational vector. 3. Results and Discussion  3. Results and Discussion In  this  study,  the  numerical  model  of  a  SRU  thermal  reactor  is  constructed  using  an  In this study, the numerical model of a SRU thermal reactor is constructed using an unstructured unstructured  grid.  Five  cell  densities—10,826  cells,  187,354  cells,  342,856  cells,  683,672  cells  and  grid. Five cell densities—10,826 cells, 187,354 cells, 342,856 cells, 683,672 cells and 1,124,627 cells—are 1,124,627 cells—are tested to ensure a grid‐independent solution. The computational results show  tested to ensure a grid-independent solution. The computational results show that the sizes of the that the sizes of the corner recirculation zone in zone 1 and zone 2 and the cross‐sectional average  corner recirculation zone in zone 1 and zone 2 and the cross-sectional average temperature profiles temperature profiles obtained using the last two meshes almost coincide, with a deviation of less  obtained using the last two meshes almost coincide, with a deviation of less than 0.5%. Therefore, than 0.5%. Therefore, the mesh of 683,672 cells is used for the subsequent discussion. Figure 2 shows  the mesh of 683,672 cells is used for the subsequent discussion. Figure 2 shows the numerical model of the  numerical  model  of  the  SRU  thermal  reactor  investigated.  In  Figure  2,  the  heat  exchanger  the SRU thermal reactor investigated. In Figure 2, the heat exchanger section consists of 19 cooling section  consists  of  19  cooling  tubes,  which  have  a  diameter  of  0.5  m,  as  shown  schematically  in  tubes, which have a diameter of 0.5 m, as shown schematically in Figure 3. The heat absorption rate Figure 3. The heat absorption rate for each heat exchanger tube is 40,000 W/m  and the other walls  for each heat exchanger tube is 40,000 W/m and the other walls are adiabatic. No slip condition is are adiabatic. No slip condition is applied on all solid walls. The exit of the heat exchanger section is  applied on all solid walls. The exit of the heat exchanger section is connected to other equipment at connected to other equipment at 300 K and 1 atm by a pipe that is 1.372 m in diameter and 11.5 m in  300 K and 1 atm by a pipe that is 1.372 m in diameter and 11.5 m in length. length.  Figure 2. The numerical model for the SRU thermal reactor investigated. Appl. Sci. 2016, 6, 331  6 of 12  Appl. Sci. 2016, 6, 331 6 of 13 Figure 2. The numerical model for the SRU thermal reactor investigated.  Figure 3. An illustration of the arrangement of heat exchanger tubes. Figure 3. An illustration of the arrangement of heat exchanger tubes.  In this study, two types of oxygen supplies are investigated: an oxygen‐normal supply and an  In this study, two types of oxygen supplies are investigated: an oxygen-normal supply and oxygen‐rich  supply.  An  oxygen‐rich  supply  increases  sulfur  recovery.  The  design  conditions  an oxygen-rich supply. An oxygen-rich supply increases sulfur recovery. The design conditions (including the species compositions, the temperature, the pressure and the velocity) at the acid gas  (including the species compositions, the temperature, the pressure and the velocity) at the acid gas inlet inlet holes of zone 1 and zone 2 and at the air inlet hole are listed in Table 1. These conditions are  holes of zone 1 and zone 2 and at the air inlet hole are listed in Table 1. These conditions are practical practical  operating  conditions  that  are  used  by  a  petrochemical  corporation  in  Taiwan.  The  operating conditions that are used by a petrochemical corporation in Taiwan. The turbulence kinetic turbulence kinetic energy is 10% of the inlet mean flow kinetic energy and the turbulence dissipation  energy is 10% of the inlet mean flow kinetic energy and the turbulence dissipation rate is computed rate is computed using Equation (16).  using Equation (16). 3/2 3/4 3/2 ε= C k (16) 3μ/4 " = C l (16) where l = 0.07L and L is the hydraulic diameter.  where l = 0.07L and L is the hydraulic diameter. To validate the numerical methods used in this study, the simulation results were compared  with  available  practical  data.  The  exit  sulfur  mole  fractions  from  the  SRU  thermal  reactor  of  a  Table 1. The design conditions at the acid gas inlet holes and the air inlet hole. petrochemical corporation in Taiwan were measured with the design conditions listed in Table 1.  Oxygen-Normal Supply Oxygen-Rich Supply The  results  are  shown  in  Table  2.  It  can  be  seen  that  the  respective  deviations  are  1.3%  for  an  Location Acid Gas to Acid Gas to Acid Gas to Acid Gas to oxygen‐normal supply and 8.3% for an oxygen‐rich supply, which are acceptable from a viewpoint  Air Inlet Air Inlet of engineeringZone  appl1ications.Zone   2 Zone 1 Zone 2 Species x (%) Table 1. The design conditions at the acid gas inlet holes and the air inlet hole.  O 0 0 19.87 0 0 23.85 N 0 0 74.98 0 0 71.26 Oxygen‐Normal Supply  Oxygen‐Rich Supply  H O 7.83 4.12 5.15 4.12 27.97 4.89 Location  Acid Gas to  Acid Gas to  Acid Gas to Acid Gas to  CO 1.27 1.5 0 1.48 0 0 Air Inlet  Air Inlet  H S 82.06 Zone 1  89.88 Zone 2  0 89.9 Zone 1  39.61 Zone 2  0 CH 2.28 2.7 0 2.7 0 0 Species  x (%) C H 1.52 1.8 0 1.8 0 0 2 6 O2  0  0  19.87 0 0 23.85  NH 5.04 0 0 0 32.42 0 N2  0  0  74.98  0  0  71.26  T (K) 319.92 316.15 403.15 313.15 316.15 397.15 H2O  7.83  4.12  5.15 4.12 27.97 4.89  CO2  1.27  1.5  0  1.48  0  0  76,920 75,068 74,382 75,068 75,068 89,572 P (N/m ) H2S  82.06  89.88  0  89.9  39.61  0  V (m/s) 11.62 2.08 12.4 (Radial) 11.46 1.88 10.8 (Radial) CH4  2.28  2.7  0  2.7  0  0  34.1 (Tangential) 29.8 (Tangential) C2H6  1.52  1.8  0  1.8  0  0  Appl. Sci. 2016, 6, 331 7 of 13 To validate the numerical methods used in this study, the simulation results were compared with available practical data. The exit sulfur mole fractions from the SRU thermal reactor of a petrochemical corporation in Taiwan were measured with the design conditions listed in Table 1. The results are shown in Table 2. It can be seen that the respective deviations are 1.3% for an oxygen-normal supply and 8.3% for an oxygen-rich supply, which are acceptable from a viewpoint of engineering applications. Table 2. Comparison of the measured and simulation results. Appl. Sci. 2016, 6, 331  7 of 12  Case Practical Data Simulation Results Deviations NH3  5.04  0  0  0  32.42  0  Oxygen-normal T (K)  319.92  316.15  403.15  313.15  316.15  397.15  0.078 0.079 1.3% operation P (N/m )  76,920  75,068  74,382  75,068  75,068  89,572  V (m/s)  11.62  2.08  12.4 (Radial)  11.46  1.88  10.8 (Radial)  Oxygen-rich operation 0.084 0.091 8.3%      34.1 (Tangential)    29.8 (Tangential) Table 2. Comparison of the measured and simulation results.  To determine the effect of the fuel mass fraction, the zone 1 fuel mass fraction is defined as the mass ratio of the zone 1 acid gas to the total acid gas (acid gas to zone 1 plus acid gas to zone 2). Case  Practical Data Simulation Results Deviations  Oxygen‐normal operation  0.078  0.079  1.3%  The design zone 1 fuel mass fractions, calculated from Table 1, are 0.875 for an oxygen-normal supply Oxygen‐rich operation  0.084  0.091  8.3%  and 0.895 for an oxygen-rich supply. Figure 4 shows a comparison of the maximum and the average temperatures for a SRU thermal To determine the effect of the fuel mass fraction, the zone 1 fuel mass fraction is defined as the  reactor with different zone 1 fuel mass fractions. The air is kept at the design condition. Eight zone mass ratio of the zone 1 acid gas to the total acid gas (acid gas to zone 1 plus acid gas to zone 2). The  design zone 1 fuel mass fractions, calculated from Table 1, are 0.875 for an oxygen‐normal supply  1 fuel mass fractions are used, including 0.0625, 0.125, 0.25, 0.375, 0.5, 0.625, 0.75 and 0.875 (for an and 0.895 for an oxygen‐rich supply.  oxygen-normal case) or 0.895 (for an oxygen-rich case). The zone 1 fuel mass fractions ranging from Figure 4 shows a comparison of the maximum and the average temperatures for a SRU thermal  0.0625 to 0.895 are calculated to determine the stoichiometric fuel mass fraction which produces reactor with different zone 1 fuel mass fractions. The air is kept at the design condition. Eight zone 1  the most complete combustion and the highest temperature. Figure 4a–c show that the maximum fuel mass fractions are used, including 0.0625, 0.125, 0.25, 0.375, 0.5, 0.625, 0.75 and 0.875 (for an  oxygen‐normal case) or 0.895 (for an oxygen‐rich case). The zone 1 fuel mass fractions ranging from  temperature and the average temperature reach maximum values at a zone 1 fuel mass fraction of 0.0625 to 0.895 are calculated to determine the stoichiometric fuel mass fraction which produces the  about 0.375, which implies that the stoichiometric fuel mass fraction is around 0.375. The temperature most  complete  combustion  and  the  highest  temperature.  Figure  4a–c  show  that  the  maximum  decreases when the zone 1 fuel mass fraction deviates from the stoichiometric fuel mass fraction. temperature and the average temperature reach maximum values at a zone 1 fuel mass fraction of  In Figure 4c, the average temperature of the thermal reactor represents the average temperature about  0.375,  which  implies  that  the  stoichiometric  fuel  mass  fraction  is  around  0.375.  The  temperature  decreases  when  the  zone 1  fuel  mass  fraction  deviates  from  the  stoichiometric  fuel  for both zone 1 and zone 2. In a practical SRU thermal reactor, high temperature can damage the mass fraction.  In Figure 4c, the average temperature of the thermal reactor represents the average  refractory. For example, the zone 1 corner has a suddenly expanded geometry and a recirculation zone temperature for both zone 1 and zone 2. In a practical SRU thermal reactor, high temperature can  forms in this region. The temperature in this region can exceed the maximum service temperature damage  the  refractory.  For  example,  the zone 1 corner has a suddenly expanded geometry and a  for the refractory and cause collapse or deformation of the zone 1 corner. The suggested maximum recirculation zone forms in this region. The temperature in this region can exceed the maximum  service temperature for the refractory and cause collapse or deformation of the zone 1 corner. The  service temperature for the refractory varies for different manufacturers. A typical value is 1700 C suggested  maximum  service  temperature  for  the  refractory  varies  for  different  manufacturers.  A  (1973 K). Figure 4a shows that the maximum temperatures for all oxygen-rich cases exceed the typical  value  is  1700  °C  (≈1973  K).  Figure  4a  shows  that  the  maximum  temperatures  for  all  suggested maximum service temperature. However, for oxygen-normal cases, when the zone 1 fuel oxygen‐rich  cases  exceed  the  suggested  maximum  service  temperature.  However,  for  mass fraction is less than 0.3 or greater than 0.6, the maximum temperature is lower than the suggested oxygen‐normal  cases,  when  the  zone 1 fuel mass fraction is less than 0.3 or greater than 0.6, the  maximum  temperature  is  lower  than  the  suggested  maximum  service  temperature.  The  current  maximum service temperature. The current design condition for an oxygen-normal operation (zone 1 design condition for an oxygen‐normal operation (zone 1 fuel mass fraction of 0.875), listed in Table  fuel mass fraction of 0.875), listed in Table 1, allows an acceptable temperature field. 1, allows an acceptable temperature field.  normal oxygen rich oxygen normal oxygen rich oxygen 0.2 0.4 0.6 0.8 0.2 0.4 0.6 0.8 zone 1 fuel mass fraction zone 1fuel massfraction (a)  (b) Figure 4. Cont. Tmax (K) Tavg (K) 1 Appl. Sci. 2016, 6, 331  8 of 12  Appl. Sci. 2016, 6, 331 8 of 13 Appl. Sci. 2016, 6, 331  8 of 12  normal oxygen rich oxygen normal oxygen rich oxygen 0.2 0.4 0.6 0.8 0.z 2one 1 fu 0.e 4l mass f 0r .6action 0.8 zone 1 fuel mass fraction (c) (c) Figure 4. A comparison of the maximum and average temperatures for a SRU thermal reactor with  Figure 4. A comparison of the maximum and average temperatures for a SRU thermal reactor with  Figure 4. A comparison of the maximum and average temperatures for a SRU thermal reactor with different zone 1 fuel mass fractions: (a) Maximum temperature; (b) The average temperature in zone  different zone 1 fuel mass fractions: (a) Maximum temperature; (b) The average temperature in zone  different zone 1; (c) 1The fuel  ave mass rage temperature fractions: in (a the ) Maximum  thermal reactor. temperatur   e; (b) The average temperature in zone 1; 1; (c) The average temperature in the thermal reactor.  (c) The average temperature in the thermal reactor. Figure  5 shows a  comparison  of  the  cross‐sectional  average  temperature for a SRU  thermal  Figure  5 shows a  comparison  of  the  cross‐sectional  average  temperature for a SRU  thermal  reactor with different zone 1 fuel mass fractions. It is seen that the average temperature in zone 2 is  Figurehigher 5 shows  than a the comparison  average tempera of the turecr  inoss-sectional  zone 1 for zoneaverage  1 fuel matemperatur ss fractions ofe 0. for 062a 5 and SRU 0.12 thermal 5. A  reactor reactor with different zone 1 fuel mass fractions. It is seen that the average temperature in zone 2 is  zone  1  fuel  mass  fraction  lower  than  the  stoichiometric  fuel  mass  fraction  (e.g.,  0.0625  or  0.125)  with different zone 1 fuel mass fractions. It is seen that the average temperature in zone 2 is higher higher than the average temperature in zone 1 for zone 1 fuel mass fractions of 0.0625 and 0.125. A  produces  a  lower  zone  1  temperature  and  this  helps  protect  zone  1;  however,  the  zone  2  than the average temperature in zone 1 for zone 1 fuel mass fractions of 0.0625 and 0.125. A zone zone  1  fuel  mass  fraction  lower  than  the  stoichiometric  fuel  mass  fraction  (e.g.,  0.0625  or  0.125)  temperature is increased because a lower zone 1 fuel mass fraction represents a lower zone 1 fuel  1 fuel mass fraction lower than the stoichiometric fuel mass fraction (e.g., 0.0625 or 0.125) produces produces  a  lower  zone  1  temperature  and  this  helps  protect  zone  1;  however,  the  zone  2  supply and a higher zone 2 fuel supply. It is worth noting that a high zone 2 average temperature  a lower zone can 1da temperatur mage the heat e exch andanger this tubes. helps For pr otect larger zone zone 11; fuel however  mass fract , the ions, zone  the av 2er temperatur age temperateure is increased temperature is increased because a lower zone 1 fuel mass fraction represents a lower zone 1 fuel  in  zone  1  is  higher.  An  oxygen‐rich  supply  also  enhances  the  chemical  reaction  and  produces  a  because a lower zone 1 fuel mass fraction represents a lower zone 1 fuel supply and a higher zone 2 fuel supply and a higher zone 2 fuel supply. It is worth noting that a high zone 2 average temperature  higher  temperature.  The  temperature  decreases  abruptly  across  the  choke  ring  because  thermal  supply. It is worth noting that a high zone 2 average temperature can damage the heat exchanger tubes. can damage the heat exchanger tubes. For larger zone 1 fuel mass fractions, the average temperature  energy  is  converted  into  kinetic  energy  when  the  flow  is  accelerated.  It  is  also  seen  that  the  For larger zone 1 fuel mass fractions, the average temperature in zone 1 is higher. An oxygen-rich in  zone  1  is  higher.  An  oxygen‐rich  supply  also  enhances  the  chemical  reaction  and  produces  a  temperature is more uniform for the design zone 1 fuel mass fractions (0.875 for the oxygen‐normal  supply also case enhances  and 0.895 for the the chemical  oxygen‐rich reaction  case). This and  canpr  aloduces so be observed a higher  from temperatur Figure 6 whice. h sh The ows temperatur the  e higher  temperature.  The  temperature  decreases  abruptly  across  the  choke  ring  because  thermal  temperature profile on the symmetric plane for a SRU  thermal  reactor  with  different  zone 1 fuel  decreases abruptly across the choke ring because thermal energy is converted into kinetic energy when energy  is  converted  into  kinetic  energy  when  the  flow  is  accelerated.  It  is  also  seen  that  the  mass fractions.  the flow is accelerated. It is also seen that the temperature is more uniform for the design zone 1 fuel temperature is more uniform for the design zone 1 fuel mass fractions (0.875 for the oxygen‐normal  mass fractions (0.875 for the oxygen-normal case and 0.895 for the oxygen-rich case). This can also case and 0.895 for the oxygen‐rich case). This can also be observed from Figure 6 which shows the  be observed from Figure 6 which shows the temperature profile on the symmetric plane for a SRU temperature profile on the symmetric plane for a SRU  thermal  reactor  with  different  zone 1 fuel  thermal reactor with different zone 1 fuel mass fractions. mass fractions.  0.0625 2000 X= 1fuel 0.0625 X= 1fuel 0.125 X= 1fuel 1200 0.125 X= 1fuel 0.25 X= 1fuel 0.25 X= 1fuel 0.375 X= 1fuel 0.375 X= 1fuel 0.5 X= 1fuel 1000 0.5 X= 1fuel 0.625 1000 1600 X= 1fuel 0.625 X= 1fuel 0.75 X= 1fuel 1600 0.75 X= 1fuel 0.875 1fuel X= 800 0.875 X= 1fuel 0 246 8 10 12 02468 10 12 X(m) 0.0625 X(m) X= 1fuel 0.0625 X= 1fuel 0.125 X= 1fuel 1200 0.125 X= 1fuel 0.25 X= 1fuel (a)  (b)  0.25 X= 1fuel 0.375 X= 1fuel 0.375 X= 1fuel 0.5 Figure 5. A comparison of the cross‐sectional average temperature for a SRU thermal reactor with  X= 1fuel 1000 0.5 X= 1fuel 0.625 1000 X= 1fuel different zone 1 fuel mass fractions: (a) Oxygen‐normal operation; (b) Oxygen‐rich operation.  0.625 X= 1fuel 0.75 X= 1fuel 0.75 X= 1fuel 0.875 1fuel X= 800 0.875 X= 1fuel 0 246 8 10 12 02468 10 12 X(m) X(m) (a)  (b)  Figure 5. A comparison of the cross‐sectional average temperature for a SRU thermal reactor with  Figure 5. A comparison of the cross-sectional average temperature for a SRU thermal reactor with different zone 1 fuel mass fractions: (a) Oxygen‐normal operation; (b) Oxygen‐rich operation.  different zone 1 fuel mass fractions: (a) Oxygen-normal operation; (b) Oxygen-rich operation. T(K) T(K) Tavg (K) Tavg (K) T(K) T(K) Appl. Sci. 2016, 6, 331 9 of 13 Appl. Sci. 2016, 6, 331  9 of 12  Figure 6 shows that the maximum temperature occurs at the heat exchanger tubes for zone 1 fuel Figure 6 shows that the maximum temperature occurs at the heat exchanger tubes for zone 1  mass fractions of 0.0625 and 0.125. Because the heat exchanger tubes are made of carbon steel, which fuel mass fractions of 0.0625 and 0.125. Because the heat exchanger tubes are made of carbon steel,  has awhi maximum ch has a ma service ximum temperatur  service tempera e thatture is lower that is than lower that  than for tha the t for refractory  the refracof tory the ofthermal  the therm reactor al  , reactor, the tubes can be damaged and care must be taken when a low zone 1 fuel mass fraction is  the tubes can be damaged and care must be taken when a low zone 1 fuel mass fraction is used. used.  (a)      (b)  (c)  (d)  (e)  Figure 6. Cont. Appl. Sci. 2016, 6, 331 10 of 13 Appl. Sci. 2016, 6, 331  10 of 12  (f)  (g)  (h)  Figure 6. The temperature profile on the symmetric plane for a SRU thermal reactor with different  Figure 6. The temperature profile on the symmetric plane for a SRU thermal reactor with different zone 1 fuel mass fractions: (a) Zone 1 fuel mass fraction of 0.0625; (b) Zone 1 fuel mass fraction of  zone 1 fuel mass fractions: (a) Zone 1 fuel mass fraction of 0.0625; (b) Zone 1 fuel mass fraction of 0.125; (c) Zone 1 fuel mass fraction of 0.375; (d) Zone 1 fuel mass fraction of 0.875; (e) Zone 1 fuel  0.125; (c) Zone 1 fuel mass fraction of 0.375; (d) Zone 1 fuel mass fraction of 0.875; (e) Zone 1 fuel mass fraction of 0.0625; (f) Zone 1 fuel mass fraction of 0.125; (g) Zone 1 fuel mass fraction of 0.375; (h)  mass fraction of 0.0625; (f) Zone 1 fuel mass fraction of 0.125; (g) Zone 1 fuel mass fraction of 0.375; Zone 1 fuel mass fraction of 0.895  (note:(a–d) are  for  oxygen‐normal  operations  and  (e–h) are  for  (h) Zone 1 fuel mass fraction of 0.895 (note: (a–d) are for oxygen-normal operations and (e–h) are for oxygen‐rich operations).  oxygen-rich operations). Figure 7 shows a comparison of the sulfur and sulfur dioxide concentrations at the exit for a  SRU thermal reactor with different zone 1 fuel mass fractions for an oxygen‐normal operation. It is  Figure 7 shows a comparison of the sulfur and sulfur dioxide concentrations at the exit for a SRU seen  that  the  zone  1  fuel  mass  fraction  has  a  minor  influence  on  the  sulfur  and  sulfur  dioxide  thermal reactor with different zone 1 fuel mass fractions for an oxygen-normal operation. It is seen concentrations  at  the  exit.  The  sulfur  concentration  at  the  exit  is  about  0.08.  An  oxygen‐rich  that the zone 1 fuel mass fraction has a minor influence on the sulfur and sulfur dioxide concentrations operation shows a similar tendency with the sulfur concentration at the exit being around 0.09.  at the exit. The sulfur concentration at the exit is about 0.08. An oxygen-rich operation shows a similar From the above discussion, it is seen that the major influence of the zone 1 fuel mass fraction is  tendency with the sulfur concentration at the exit being around 0.09. on the temperature. The current design conditions listed in Table 1 allow an acceptable temperature  From the above discussion, it is seen that the major influence of the zone 1 fuel mass fraction is field for an oxygen‐normal operation. However, for an oxygen‐rich operation, the local maximum  on the temperature. The current design conditions listed in Table 1 allow an acceptable temperature temperature  exceeds  the  suggested  maximum  service  temperature,  although  the  average  field for an oxygen-normal operation. However, for an oxygen-rich operation, the local maximum temperature is acceptable. Therefore, the high temperature regions, such as the zone 1 corner, must  temperatur be  inspeect exceeds ed  very the   casuggested refully  dur maximum ing  the  ann service ual  ma temperatur intenance  e,peri although od  if  there the average   are  oxygen temperatur ‐rich  e operations.  is acceptable. Therefore, the high temperature regions, such as the zone 1 corner, must be inspected very carefully during the annual maintenance period if there are oxygen-rich operations. Appl. Sci. 2016, 6, 331 11 of 13 Appl. Sci. 2016, 6, 331  11 of 12  0.1 0.1 0.08 0.08 0.06 0.06 0.04 0.04 0.02 0.02 0.2 0.4 0.6 0.8 0.2 0.4 0.6 0.8 zone 1 fuel mass fraction zone 1 fuel mass fraction (a)  (b)  Figure 7. A comparison of the (a) sulfur and (b) sulfur dioxide concentrations at the exit for a SRU  Figure 7. A comparison of the (a) sulfur and (b) sulfur dioxide concentrations at the exit for a SRU thermal reactor with different zone 1 fuel mass fractions.  thermal reactor with different zone 1 fuel mass fractions. 4. Conclusions  4. Conclusions This paper numerically investigates the effect of the fuel mass fraction on the combustion and  This paper numerically investigates the effect of the fuel mass fraction on the combustion and fluid flow in a SRU thermal reactor to provide a guideline for adjusting the fuel mass fraction to  fluid flow in a SRU thermal reactor to provide a guideline for adjusting the fuel mass fraction to reduce  the  temperature  inside  the  thermal  reactor  and  to  allow  an  acceptable  sulfur  recovery.  reduce the temperature inside the thermal reactor and to allow an acceptable sulfur recovery. Practical Practical  operating  conditions  for  a  petrochemical  corporation  in  Taiwan  are  used  as  the  design  operating conditions for a petrochemical corporation in Taiwan are used as the design conditions. conditions. To determine a suitable fuel mass fraction for operation, a detailed numerical simulation  To determine a suitable fuel mass fraction for operation, a detailed numerical simulation should should  be  performed  first  to  find  the  stoichiometric  fuel  mass  fraction  which  produces  the  most  be performed first to find the stoichiometric fuel mass fraction which produces the most complete complete combustion and the highest temperature. This stoichiometric fuel mass fraction should be  combustion and the highest temperature. This stoichiometric fuel mass fraction should be avoided avoided because the high temperature could damage the zone 1 corner or the choke ring. A higher  because the high temperature could damage the zone 1 corner or the choke ring. A higher fuel mass fuel  mass  fraction  (i.e.,  fuel‐rich  or  air‐lean  condition)  is  more  suitable  because  it  can  avoid  fraction (i.e., fuel-rich or air-lean condition) is more suitable because it can avoid deteriorations of deteriorations of both zone 1 and the heat exchanger tubes. Although a lower fuel mass fraction (i.e.,  both zone 1 and the heat exchanger tubes. Although a lower fuel mass fraction (i.e., fuel-lean or fuel‐lean or air‐rich condition) can avoid deterioration of zone 1, the heat exchanger tubes may be  air-rich condition) can avoid deterioration of zone 1, the heat exchanger tubes may be damaged. damaged. The present simulation results show that the current design condition is a fuel‐rich (or  The present simulation results show that the current design condition is a fuel-rich (or air-lean) air‐lean)  condition  and  allows  an  acceptable  thermal  reactor  temperature  for  an  oxygen‐normal  condition and allows an acceptable thermal reactor temperature for an oxygen-normal operation. operation.  However,  for  an  oxygen‐rich  operation,  the  local  maximum  temperature  exceeds  the  However, for an oxygen-rich operation, the local maximum temperature exceeds the suggested suggested maximum service temperature, although the average temperature is acceptable. The high  maximum service temperature, although the average temperature is acceptable. The high temperature temperature regions, such as the zone 1 corner, must be inspected very carefully during the annual  regions, such as the zone 1 corner, must be inspected very carefully during the annual maintenance maintenance period if there are oxygen‐rich operations. For lower zone 1 fuel mass fractions such as  period if there are oxygen-rich operations. For lower zone 1 fuel mass fractions such as 0.0625 and 0.0625 and 0.125, the average temperature in zone 2 is higher than the average temperature in zone 1,  0.125, the average temperature in zone 2 is higher than the average temperature in zone 1, which could which  could  cause  damage  to  the  downstream  heat  exchanger  tubes.  If  a  low  zone  1  fuel  mass  cause damage to the downstream heat exchanger tubes. If a low zone 1 fuel mass fraction is used to fraction is used to produce a lower zone 1 temperature, the temperatures in zone 2 and the heat  produce a lower zone 1 temperature, the temperatures in zone 2 and the heat exchanger section must exchanger section must be monitored closely and the zone 2 wall and heat exchanger tubes must be  be monitored closely and the zone 2 wall and heat exchanger tubes must be inspected very carefully inspected very carefully during the annual maintenance period.  during the annual maintenance period. Acknowledgments: The author gratefully acknowledges the grant support from the Ministry of Science and  Acknowledgments: The author gratefully acknowledges the grant support from the Ministry of Science and Technology, Taiwan, R.O.C., under contract MOST 104‐2221‐E‐150‐032. The author would also like to express  Technology, Taiwan, R.O.C., under contract MOST 104-2221-E-150-032. The author would also like to express thanks for the useful data and constructive suggestions provided by the Formosa Petrochemical Corporation in  thanks for the useful data and constructive suggestions provided by the Formosa Petrochemical Corporation in Taiwan. Taiwan.  Conflicts of Interest: The author declares no conflict of interest. Conflicts of Interest: The author declares no conflict of interest.  Abbreviations  The following abbreviations are used in this manuscript:  Cμ  turbulence model constant (=0.09)  2 2 k  turbulence kinetic energy (m /s )  Sulfur mole fraction SO mole fraction 2 Appl. Sci. 2016, 6, 331 12 of 13 Abbreviations The following abbreviations are used in this manuscript: C turbulence model constant (=0.09) 2 2 k turbulence kinetic energy (m /s ) L hydraulic diameter (m) l characteristic length (m) P pressure (N/m ) T temperature (K) V velocity (m/s) XYZ cartesian coordinates with origin at the centroid of the burner inlet (m) X zone 1 fuel mass fraction 1fuel x mole fraction (%) Greek symbols 2 3 " turbulence dissipation rate (m /s ) Subscripts avg average avg1 zone 1 average avg2 zone 2 average max maximum References 1. Adewale, R.; Salem, D.J.; Berrouk, A.S.; Dara, S. Simulation of hydrogen production from thermal decomposition of hydrogen sulfide in sulfur recovery units. J. Clean. Prod. 2016, 112, 4815–4825. [CrossRef] 2. Chardonneaua, M.; Ibrahim, S.; Gupta, A.K.; AlShoaibi, A. Role of toluene and carbon dioxide on sulfur recovery efficiency in a Claus process. Energy Procedia 2015, 75, 3071–3075. [CrossRef] 3. Selim, H.; Gupta, A.K.; AlShoaibi, A. Effect of reaction parameters on the quality of captured sulfur in Claus process. Appl. Energ. 2013, 104, 772–776. [CrossRef] 4. Monnery, W.D.; Hawboldt, K.A.; Pollock, A.; Svrcek, W.Y. New experimental data and kinetic rate expression for the Claus reaction. Chem. Eng. Sci. 2000, 55, 5141–5148. [CrossRef] 5. Yeh, C.L. Effect of choke ring dimension on thermal and fluid flow in a SRU thermal reactor. Trans. Can. Soc. Mech. Eng. submitted for publication, 2016. 6. Yeh, C.L. Effect of choke ring position on thermal and fluid flow in a SRU thermal reactor. Int. J. Mech. Eng. Robot. Res. 2015, 4, 273–277. 7. Yeh, C.L. Effect of streamlining geometry on thermal and fluid flow in a SRU thermal reactor. In Proceedings of International Multi-Conference on Engineering and Technology Innovation 2015 (IMETI2015), Kaohsiung, Taiwan, 30 October–3 November 2015; J5043. 8. Yeh, C.L. Numerical analysis of the effects of streamlining geometry and a vector wall on the thermal and fluid flow in a SRU thermal reactor. Trans. Can. Soc. Mech. Eng. submitted for publication, 2016, 40. 9. Yeh, C.L. Effect of a vector wall on the thermal and fluid flow in a SRU thermal reactor. In Proceedings of International Multi-Conference on Engineering and Technology Innovation 2015 (IMETI2015), Kaohsiung, Taiwan, 30 October–3 November 2015; J5042. 10. Patankar, S.V. Numerical Heat Transfer and Fluid Flows; McGraw-Hill: New York, NY, USA, 1980. 11. Launder, B.E.; Spalding, D.B. Lectures in Mathematical Models of Turbulence; Academic Press: London, UK, 1972. 12. Siegel, R.; Howell, J.R. Thermal Radiation Heat Transfer; Hemisphere Publishing Corporation: Washington, DC, USA, 1992. 13. Sivathanu, Y.R.; Faeth, G.M. Generalized state relationships for scalar properties in non-premixed hydrocarbon/air flames. Combust. Flame 1990, 82, 211–230. [CrossRef] 14. Launder, B.E.; Spalding, D.B. The numerical computation of turbulent flows. Comput. Method Appl. Mech. Eng. 1974, 3, 269–289. [CrossRef] Appl. Sci. 2016, 6, 331 13 of 13 15. Yeh, C.L. Numerical analysis of the combustion and fluid flow in a carbon monoxide boiler. Int. J. Heat Mass Trans. 2013, 59, 172–190. [CrossRef] 16. Batchelor, G.K. An Introduction to Fluid Dynamics; Cambridge University Press: Cambridge, UK, 1967. © 2016 by the author; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).

Journal

Applied Sciences – Multidisciplinary Digital Publishing Institute

Published: Nov 2, 2016

There are no references for this article.