- Email: [email protected]
Thermal damage of sulfur processed chamber under Claus operating reaction conditions—A case study
Case Studies in Construction Materials 8 (2018) 517–529
Contents lists available at ScienceDirect
Case Studies in Construction Materials journal homepage: www.elsevier.com/locate/cscm
Case study
Thermal damage of sulfur processed chamber under Claus operating reaction conditions—A case study S. Chidambaram Numaligarh Refinery Limited (A Group Company of Bharat Petroleum Corporation Limited), India
A R T I C L E I N F O
Article history: Received 19 September 2017 Received in revised form 11 April 2018 Accepted 12 April 2018 Available online 20 April 2018 Keywords: Thermal damage Refractory Numerical formulation Combustion chamber Industrial failure
A B S T R A C T
The sulfur from acid gas and sour gas is recovered to produce useful sulfur products. The claus reaction is industrially important process to recover sulfur from corrosive acid and sour gases. To achieve claus reaction, high temperature reactor is utilized under controlled conditions. The Claus combustion reaction heater made of 94% alumina refractory bricks were damaged under ideal equilibrium operating conditions. The first principle mathematical model was developed to measure the temperature profiles at hot face bricks and its interfaces to predict outer steel surface temperature. The maximum outer steel wall surface temperature was measured as 548.6 C using thermograph experiments and compared against the predicted temperature. Large variations in temperature differences have confirmed that the refractory wall was damaged due to abnormal reactions. This thermal damage is discussed and presented with various evidences from visual inspection. The hot face wall bricks, matrix blocks and orifice throat were damaged and presented by macroscopic visual inspection. However, hardness of the steel shell is within specified limit. Therefore, the refractory repair inside the reactor and thermal insulation of external shell has been proposed to prevent steel shell from creep, graphitization, and high temperature oxidization and corrosion damages. © 2018 The Author. Published by Elsevier Ltd. This is an open access article under the CC BYNC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
1. Introduction The refractory materials are designed to resist against heat, corrosion from industrial corrosive gases, mechanical and thermal stress, strains, and abrasion at higher temperatures. The performance of refractory directly depends upon its chemical composition, manufacturing method and its implementation methods [1]. Mostly, the refractories are used in steel industries, petroleum and petrochemical industries. The right choice of refractory selection at design stage is economically beneficial. In steel industry, the basic refractories are applied to the wall of basic oxygen furnace (BOF) [2]. The quality and productivity of the liquid steel is determined by the performance of applied refractory in addition to various other parameters [3]. Similarly, petroleum industries also line various basic refractories to inner wall of the reactor and pressure vessel on its hot face sides. The service life of the reactor and pressure vessel for chemical processing directly depends on the lined refractory. Alumina refractories known for its good performance is highly essential for steel processing and hydrocarbon industries. Alumina refractories ranges from 60% Al2 O3 to 99% Al2 O3 and are used in various applications depending upon the type of product. For example 60% Al2 O3 are used in general furnace and incinerator bricks, 94% Al2 O3 are used in internals of combustion chamber reactor walls and 99% Al2 O3 are used in certain specialty products [4]. The reduction of mullite in alumina increases the service temperature. Therefore, the performance of alumina refractory is better than fire clay refractory, particularly in better creep resistance and high temperature corrosion. However, the refractory performance will come down if it is operated beyond design temperatures. https://doi.org/10.1016/j.cscm.2018.04.005 2214-5095/© 2018 The Author. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/).
518
C. S. / Case Studies in Construction Materials 8 (2018) 517–529
One such case study from petroleum refinery is discussed in this paper. In petroleum industry, sulfur will be extracted from sour gas and acid gas which majorly contains H2 S. The schematic view of sulfur recovery process flow is shown in Fig. 1. The sour water generated from high pressure cracker and coker unit supplies sour gas feed to sulfur recovery unit. These sulfur recovery unit produces sulfur as a final product from crude processed hydrocarbon. The fundamental process is that partial conversion of H2 S to sulfur-di-oxide as stated in Eq. (1) and further reaction between remaining H2 S gas and SO2 gas will form sulfur and water as shown in Eq. (2) [6]. The first stage partial oxidation and second stage remaining combustion is called overall claus reaction as shown in Eq. (3) [7]. The amount of air feed to the process will define the amount of sulfur-di-oxide formation. Therefore, for maximum sulfur formation the ratio of
H2 S SO2
in the process gas was
maintained at the ratio of 2:1 [8] and accordingly air to gas ratio was maintained in-between 1.4 to 1.5 for clauss operating reaction [9]. The overall Claus reaction is controlled by air injection and this endothermic reaction approximately recovers 99% of sulfur from sour rich acid gas [10]. 1 1 1 1 H2 S þ O2 ! SO2 þ H2 O þ 41:3 kcal per mole 3 2 3 3
ð1Þ
2 1 2 1 H2 S þ SO2 ! H2 O þ S2 þ 7:4 kcal per mole 3 3 3 2
ð2Þ
H2 S þ
1 O2 ! S þ H2 O 2
Fig. 1. Schematic view of sulfur recovery process flow and sulfur reactor thermal damage encircled within unit.
ð3Þ
C. S. / Case Studies in Construction Materials 8 (2018) 517–529
519
The refractory used in mixing chamber was designed to resist against creep damage, material instability at high temperature and high temperature corrosion, particularly in sour environment. At high velocity flow regime, the particles present in hydrocarbon feed or in sour water results in abrasion damage to refractory. Numerous literatures are available for chemical reaction kinetics and process modeling of sulfur reactor for improving its performance [6–11]. However, no literature is available for thermal modeling and thermal damage especially for sulfur reactor. This paper discusses refractory degradation in sulfur reactor chamber due to high temperature operation. The temperature calculations were performed for ideal operating Claus reactions and root cause for detected damage is presented with temperature simulations. 2. Materials and methods The sulfur reactor was constructed with composite layer of steel chamber, insulating castable and refractory bricks as shown in Fig. 2. The nominal wall thickness of main burner chamber and combustion chamber steel shell made up of SA 516 Grade 70 steel (refer steel chemical composition in Table 4 is 10 mm and 14 mm respectively [5]. The corrosion allowance is specified as 3 mm for both the chambers. The main burner chamber was internally lined with 94% Al2 O3 alumina. However, the combustion chamber has been internally lined by insulating material and refractory bricks. The material of construction of insulating castable and refractory brick is 94% Al2 O3 alumina (refer Table 1). The physical and thermal property of 94% Al2 O3 alumina material is shown in Table 3 and Table 5 respectively. In main burner chamber, the refractory lined was 225 mm in thickness and additional 7 mm expansion allowance was provided in-between steel shell and refractory lining (refer Fig. 2a and b). In combustion chamber, the refractory brick wall is 230 mm thickness at hot face internal layer and 110 mm thickness at the subsequent second insulating layer ensuring that sulfur reactor is protected by double layer thermal insulation (refer Fig. 2c). Mortar made up of 94% Al2 O3 was applied to obtain grip and strength between each refractory brick of 2.5 mm approximately. The orifice throat has twelve tiles and matrix block has twenty blocks constructed within the combustion chamber (refer Fig. 2d and e). The chemical composition of acid gas stream feed is shown in Table 2. The grain size and water required for casting 94% Al2 O3 refractory is specified as 5 mm and 9–11% respectively [12]. The typical values mentioned in Table 1, Table 3 and Table 5 are average measured values. The physical and thermal property of 94% Al2 O3 refractory depends upon the proportion of SiO2 , Fe2 O3 and TiO2 constituents. Even small changes in the proportions of these oxide elements will drastically alter the performance of alumina refractory intended for high temperature applications [13], especially in sulfur reactor (Table 3 and Table 5). The main burner chamber with single castable lining and combustion chamber with thirty four rings were bolted to each other. Before bolting both the chambers, the linear expansion of 7 mm ceramic blanket was fixed on face side of main burner chamber. This insertion of ceramic blanket will provide linear expansion allowance at specified temperature [12,13]. The 3 nos. of V-lug anchor was dissimilarly welded with carbon steel main burner chamber inner wall surface at 1208 apart from each other. Similarly, large numbers of V-lug anchors were dissimilarly welded with carbon steel shell combustion chamber. These anchors made up of austenitic stainless steel (SS316L) holds the cast refractory in main burner chamber, and insulating castable and refractory brick layer in combustion chamber. These anchors were coated with bituminous coating and wrapped with plastic tape. The Minimum Metal Design Temperature (MMDT) is specified as 343 C for SA 516 Grade 70 steel plate. Both the main burner chamber and combustion chamber steel plates were initially gas tungsten arc welded (GTAW) using ER70S2 filler and subsequently weld joints were subjected to post weld heat treatment to reduce internal stress induced by welding process. Subsequently weld joints were radiographed and hydro tested before refractory lining, and the welded steel plates of both the chambers passed all structural integrity tests. The orifice throat between refractory brick on ring no: 10 and ring no: 11 was packed with 3 mm ceramic fiber paper of 1485 C grade for expansion allowance. The design pressure and operating pressure of sulfur reactor is 7 kg per sq cm and 0.45 kg per sq cm respectively. The minimum working pressure is 0.003 kg per sq cm. The design temperature is 1760 C and operated working temperature is in range of 900 C to 1400 C. One thermocouple has been inserted within thermo well at 300 mm from inner refractory for online monitoring of actual process temperature. 3. Experimental methods The thermograph was conducted by thermal analyzer using Testo make model 881 version V1.16. The thermograph measured with pre-defined emissivity of 0.92 in measurement range 0 C to 1500 C. The refractory failure at main burner chamber, combustion chamber, and orifice throat and matrix block sections were inspected visually. The damaged portions of refractory wall were subjected to photo documentation and presented macroscopically. The thermal heat transfer analysis of main burner chamber is simulated and thermal analysis simulations is performed under ideal Claus operating reaction conditions. The hardness of both main burner chamber and combustion chamber steel cover shell was measured in room temperature using MIC 201A GE model with diamond pyramid indenter supplied by general electric. The chemical composition of both chamber shell material was analyzed in room temperature conditions using x-ray fluorescence (XRF) supplied by oxford instrumentation X-Met 7500 model.
520
C. S. / Case Studies in Construction Materials 8 (2018) 517–529
Fig. 2. a) front view of main burner chamber b) side view of main burner chamber c) side view of combustion chamber d) front view of orifice throat and e) front view of matrix block.
C. S. / Case Studies in Construction Materials 8 (2018) 517–529
521
Table 1 Chemical composition of alumina refractory. Chemical Analysis
Typical value
Specification [1]
Al2 O3 SiO2 Fe2 O3 TiO2 CaO LOI
94.40 0.20 0.16 0.18 4.85 0.20
94.0 min 0.30 max 0.30 max – 5.00 max –
Table 2 Chemical composition of Acid gas stream. Weight percentage (%) Acid gas feed
H2 S 97.6
NH3 0.00
H2 0.0067
H2 O 2.2
Hydrocarbon 0.1
Table 3 Physical properties of Alumina Refractory. Parameters
Typical value
Specification [1,2]
Dry density in g/cc, after drying at 110 C for 24 h Cold crushing strength kg/cm2 After drying at 110 C for 24 h After heating at 800 C for 3 h After heating at 1100 C for 3 h After heating at 1550 C for 3 h
2.84
2.80 min
660 480 450 750
600 min 450 min 400 min 700 min
100 75 50 100 1
– – – – 5 max
Modulus of Rigidity kg/cm2 After drying at 110 C for 24 After heating at 800 C for 3 h After heating at 1100 C for 3 h After heating at 1550 C for 3 h % retained above maximum size
Table 4 Chemical composition of ASTM A516 Grade 70 [5]. Elements weight percentage (%)
Carbon
Manganese
Silicon
Phosphorous
Sulfur
Iron
ASTM Specifications
0.28 max
0.85–1.20
0.13–0.45
0.035 max
0.035 max
balance
Table 5 Thermal properties of Alumina Refractory. Parameters
Typical value
Specification [2]
Refractoriness Orton/ C
+38/+1835
37/1820 min
Permanent Linear Change % After heating at 800 C for 3 h After heating at 1100 C for 3 h After heating at 1550 C for 3 h
0.06 0.12 0.24
0.20 max 0.20 max 0.50 max
Thermal Conductivity, kcal/m/hour/8C at 400 C HF at 600 C HF at 800 C HF
1.37 1.40 1.42
– – –
522
C. S. / Case Studies in Construction Materials 8 (2018) 517–529
Fig. 3. Cylindrical coordinate system for main burner chamber of sulfur reactor with (r, u) expressed in polar coordinate.
4. Results and discussions 4.1. Numerical formulation The numerical heat conduction and convection equation were solved using Eq. (4) [14] and input data for solving these equations were incorporated from design data of main burner chamber. The heat transfer constitutive equation in radial coordinate system for main burner chamber of sulfur reactor (refer Fig. 3) has solved using time dependent boundary conditions.
2 2 @T 1@ @T 1@ T @ T ¼ a½ r þ 2 2 þ 2 r @r @r @t r @u @z
ð4Þ
a is thermal diffusivity
Under steady state condition, the heat flow in radial direction is re-written as in form of Eq. (5) [14] @ @T r ¼0 @r @r
ð5Þ
The Fourier constitutive heat conduction equation is stated as in Eq. (6) [14] Q ¼ kA
@T @r
ð6Þ
k is thermal conductivity or inverse of thermal resistance in watt per kelvin A is cross sectional area of sulfur reactor in mm, T is temperature in C and t is time in seconds The hot acid gas stream was fed in gas pipe and simultaneously air was injected to air pipe i.e. adjacent to gas burner to produce a claus combustion reaction at 1350 C in main burner chamber subject to initial time (to ) on internal diameter (ID) refractory wall surface. An ambient condition prevails at outer diameter (OD) metal wall surface at 50 C temperature subject to final time (tf ). The 7 mm expansion allowance between inner diameter of steel shell and outer diameter of refractory lining at claus reaction conditions resulted in increased inner diameter of refractory wall from 225 mm to 232 mm after attaining 1350 C. Therefore, increased diameter has been substituted at an initial boundary condition for main burner chamber of sulfur reactor. All input data required for solving equations were shown in Table 6. By integrating Eq. (5) and by applying Table 6 Input data used in numerical modeling. Input data
Unit
Value
Outer diameter of furnace wall Inner diameter of furnace wall Metal wall thickness of furnace wall Thickness of refractory wall Inner diameter of refractory wall Temperature of hot gas inside furnace Temperature of ambient air outside furnace Coefficient of thermal convection of air Coefficient of thermal convection of sour gas Coefficient of thermal conduction of steel Coefficient of thermal conduction of refractory
mm mm mm mm mm C C Wm2/ C Wm2/ C Wm2/ C Wm2/ C
904 884 10 232 420 1350 50 50 60 50.2 1.65
C. S. / Case Studies in Construction Materials 8 (2018) 517–529
523
initial boundary conditions, the constants from integration of Eq. (5) were obtained. Subsequently, back substituting an integration constant to temperature distribution (Eq. (4)) using Eq. (6) will result in computed thermal resistance. Thermal resistance at various layers, equivalent thermal resistance, heat flow, temperature distribution at hot face bricks, temperature at refractory steel shell plate interfaces and temperature at steel shell outer surface of main burner chamber were computed using numerical formulation as shown in Table 7. The temperature profile prediction is shown in Fig. 4 and temperature cooling profile was plotted across radial direction of main burner chamber exhibited parabolic curve (refer Fig. 5). From uni-dimensional model results, it shall be concluded that maximum surface temperature in outer diameter (OD) wall surface was 150.02 C. Table 7 Numerical modeling output value. Output results obtained from model
Unit
Thermal resistance (R1) Thermal resistance (R2) Thermal resistance (R3) Thermal resistance (R4 Equivalent thermal resistance (R) Heat Flow Temperature at hot face refractory wall Temperature at interface between refractory wall and steel wall Temperature at outer diameter (OD) wall of metal surface
kelvin kelvin kelvin kelvin kelvin watt C C C
Computed value watt1 watt1 watt1 watt1 watt1
Fig. 4. Predicted temperature profile at various layer of main burner chamber.
Fig. 5. Temperature plot of main burner chamber of sulfur reactor in radial direction.
0.01263775 0.07182034 7.0965e-05 0.00704583 0.09157488 14196.0325 1170.59 151.03 150.02
524
C. S. / Case Studies in Construction Materials 8 (2018) 517–529
4.2. Thermography and visual inspection The thermography was conducted on steel surface of main burner chamber outer diameter (OD) wall and it was noted that steel surface experiences high temperature than the design limit. On observing higher surface temperature on both main burner chamber and combustion chamber, the sour gas fluid flow was terminated within the reactor system and then reactor was subjected to equilibrium cooling. After attaining room temperature, this sulfur reactor was subjected to detailed inspection. The root cause for thermal damage is analyzed and corrective actions for thermal damage prevention are discussed in detail. The surface temperature of 548.6 C was noted at main burner chamber shell and combustion chamber shell as shown in thermograph Fig. 6a and b. The corresponding physical location of both main burner chamber and combustion chamber is shown in Fig. 6. Similarly, higher temperature was observed exactly at 12 o’clock position of the flange connections between both the chambers along with detection of gas leakage (refer Fig. 6c). It was observed that the high temperature paint coating was peeled off from both the chamber shells due to high temperature. The paint peeled off partially on chamber surface exposed the bare metal. These high temperature paints were designed to protect the bare shell surface against atmospheric corrosion, rain water and dew point corrosion. However, the paint was still intact at certain location. It is suspected that the
Fig. 6. Thermograph at corresponding locations a) 1 o’clock position b) 11 o’clock position and c) 6 o’clock position of sulfur reactor while operating at Claus operation reaction conditions.
C. S. / Case Studies in Construction Materials 8 (2018) 517–529
525
Fig. 7. a) Circumferential crack in main burner chamber front view b) air injection nozzle damage c) air injection nozzle melt out partially and oxidized and d) gas nose cone geometry features lost. a)-d) in as received condition after thermal damage.
internal refractory might be damaged only at that paint peeled off locations. The bolt locations between both the chambers were suspected for leakage in addition to high temperature damage. The refractory bricks and mortar lining may dislodge from its locations due to internal thermal stress induced by linear thermal expansion resulting to leakage of sour gas. The absence of ceramic blanket on face side of main burner chamber restraints the refractory casting. 4.3. Main burner chamber The main burner chamber and combustion chamber were dismantled and inspected in detail. The main burner chamber was observed with circumferential crack as shown in Fig. 7a. The circumferential crack propagated in brittle fashion perpendicular to induced thermal stress. The refractory lining was completely dislodged from inner steel shell in main burner chamber which confirms that internal refractory lining was completely under high temperature service. In addition to observed crack at refractory mid wall and dis-bonding of refractory lining with steel shell, hydrocarbon deposits were also found at main burner chamber ID as black and slag debris (refer Fig. 7a, b and c). This black debris and slag debris are suspected as hydrocarbon and refractory or metal meltdown respectively. The sour gas obtained from sour water after chemical refining and amine treating contains hydrocarbon and this residual hydrocarbon might have increased the temperature within reactor during Claus reactions. The air nose cone on main burner chamber was found damaged and nearly one fourth of the cone section was found melted (refer Fig. 7d). The material of construction of nose cone was verified by x-ray analyzer and it was confirmed as AISI 310 austenitic stainless steel material as specified in original design. The slag of melted nose cone was suspected for slag debris on ID of main burner chamber. Therefore, the temperature within main burner chamber has reached above melting point of nose cone leading to severe thermal damage. All the purge connections in main burner chamber were found in good structural condition. 4.4. Combustion chamber The combustion chamber was visually inspected in detail and it confirms that sulfur reactor was severely damaged and degraded as the reactor was operated beyond design temperature. The 94% Al2 O3 alumina bricks were found degraded up to
526
C. S. / Case Studies in Construction Materials 8 (2018) 517–529
Fig. 8. a) and b) Dislodged bricks fallen on ID refractory wall, c) and d) loose bricks hanging at 12 o’clock, e) mortar spalling and brick dislodge at 12 o’clock position and f) bricks crack at 9 o’clock position. a)-f) in as received condition after thermal damage.
10 mm on hot face out of 230 mm of total thickness in ring no:1 and ring no: 2 at 12’o’ clock position of reactor (refer Fig. 8a and b). Similarly, the bricks were degraded at a maximum up to 8 mm on hot face side from ring no: 4 to ring no: 8. These damages were observed from 11’o’ clock to 1’o’ clock position of the reactor (refer Fig. 8c, d and e). The clock position of reactor is represented with respect to front side of the combustion chamber. The dislodged refractory pieces of alumina bricks were found lying on downstream of the chamber. The alumina bricks degraded up to 8 mm on hot face out of total 230 mm thick brick from ring no: 9 to ring no.16. These damages were observed exactly from 10’o’ clock to 3’o’ clock reactor positions. The cracks were observed in refractory bricks on ring no.17 at 9 o’clock position (refer Fig. 8f) 4.5. Orifice throat and matrix block The orifice throat and matrix block inside the reactor was also observed with severe thermal damages. Each brick of orifice throat and matrix block is called as tile and block respectively. The orifice throat is constructed by ten numbers of tiles
C. S. / Case Studies in Construction Materials 8 (2018) 517–529
527
counting tile no.1 to tile no.10 in anticlockwise direction of reactor. Each individual tile from tile no.1 to tile no.6 were dislodged from its original position and were found lying in downstream of orifice throat (refer Fig. 9a, b and c). However, the remaining tiles i.e. from tile no. 7 to tile no. 10 were intact with reactor wall. The matrix block was completely dislodged from its original position and fell down to ring no. 18 and ring no.19 i.e. towards main burner chamber burner side (refer Fig. 9d). In addition, several loose alumina refractory bricks and mortars were found lying on ring no: 17 and ring no. 18 at 12 ‘o’ clock reactor position. The bricks from ring no.19 and ring no.20 was degraded and damaged up to 10 mm on hot face out of total 230 mm thick brick at 12 ‘o’ clock reactor position. Similarly, the loose refractory lining fell upon ring no: 27 to ring no. 29 at 12 ‘o’ clock reactor position. However, alumina bricks from certain rings i.e. from ring no: 21 to ring no. 26 and from ring no.30 to ring no.35 was intact and no damages were observed. 4.6. Relining after thermal damage The black and slag debris is to be chipped out completely from the ID of refractory wall of main burner chamber and to be relined with 94% alumina mortar material. The refractory bricks damaged on various abovementioned rings are proposed for relining with 94% alumina mortar for required thickness to maintain the radius of curvature and profile at various sections in internal diameter of combustion chamber. The dislodged tiles from tile no.1 to tile no.6 are proposed to be reconstructed by new 94% alumina bricks in the orifice throat and further expansion allowance of 3 mm is to be provided at an interface between orifice throat and refractory bricks. Similarly, ceramic fiber of 1425 C grade is proposed to be packed between ring no.2 and ring no.3, between ring no.18 and ring no.19, between ring no.23 and ring no.24, and between ring no.28 and ring no.29 with 5 mm expansion allowance. The new burner nose cone (refer Fig. 10) is to be fixed and assembled in main burner chamber. 4.7. Characterization of metallic shell surface after refractory thermal damage The chemical composition of metal shell wall was measured (refer Table 8) and its chemical composition matches with SA516 Grade 70 material as per international codes and standard ASTM/ASME limit [5]. However, x-ray analyzer was not able
Fig. 9. Orifice throat collapsed a) at 9 o’clock ‘position b) magnified A location of figure a, c) amplified B location at 2 o’clock position of figure a and d) matrix block collapsed. a)-d) in as received condition after thermal damage.
528
C. S. / Case Studies in Construction Materials 8 (2018) 517–529
Fig. 10. Gas nose cone a) back view and b) front view after repair.
Table 8 Chemical composition of thermal damaged shell metal. Elements weight percentage (%)
Carbon
Manganese
Silicon
Phosphorous
Sulfur
Iron
Measured value
–
0.88 0.02
0.16 0.03
0.035 0.02
0.04 0.01
99.0 0.01
to detect carbon percent in the material due to its limitations. The hardness of metallic shell was measured (refer Fig. 11) and it was found within the limit as prescribed by the codes and standards [5]. The measured hardness value have confirmed the absence of unwanted hard or any brittle phases in the micro structure of steel shell. The shell material must resist against creep, graphitization, high temperature corrosion and oxidation. 5. Mechanism of refractory degradation The main causes for refractory degradation are mechanical, thermal and corrosion damage or any of these combinations. The thermo mechanical stress-strain effect will lead to spalling, thermal shock and crack. The mechanical stress-strain effect will lead to erosion and abrasion and the chemical attack will lead to corrosion as a result of oxidation or sulfidation. The restraint force induced between various bricks, mortar, insulating castable and steel shell due to temperature difference, material heterogeneity and anisotropic on reactor wall generates single or multiple cracks in refractory wall, especially on
Fig. 11. Shell surface hardness in as received thermal damaged conditions.
C. S. / Case Studies in Construction Materials 8 (2018) 517–529
529
bricks, mortar and castable depending upon type of localized stress and interface conditions [15]. The brittle crack initiated in refractory material will lead to fracture depending upon chemical composition, manufacturing method of refractory material, crack size, geometry, critical stress intensity factor and temperature. The stress intensity factor (KI ) of flaw in the refractory reaches or exceeds the critical stress intensity factor (KI Kc ) propagating the crack in refractory. Anchor used for industrial refractory applications should be resistant to oxidation, sulfidation and condensing dew point corrosion. Further, refractory would degrade if thermal coefficient expansion of anchor is not compatible with steel shell. The pre-existing crack present in the refractory will lead crack surface to close up as a result of compressive stress. Contrastingly, the pre-existing crack in the refractory will lead to open and propagate crack surface further as a result of tensile stress. Therefore, after cooling from firing temperature, the cracks in the refractory brick apparently appears in naked eyes. The thermal impact on corrosion behavior of carbon steel is high [16]. An increase in temperature will increase the rate of corrosion in steel shell, if hot corrosive gases pass through these refractory cracks. 6. Conclusions The temperature measured using thermograph experiments on outer surface of main burner chamber was much higher than the predicted temperature. The refractory wall is severely damaged as the reactor was operated beyond design temperature and it was confirmed by appearance of damages. The hydrocarbon carryover in the feed gas stream favored an un-controlled reaction proportionally increases the operating temperature beyond the design temperature limit which might be the probable cause of failure. The outer steel shell surface temperature exceeds the creep design temperature. The termination of claus condition beyond 1400 C resulted to damage. The proposed sour water treatment in surge drum to remove hydrocarbon traces will prevent thermal damages in claus condition operating reactors. The hydrocarbon in the sour water should be chemically separated such that no hydrocarbon will remain in the sour gases. Insulating the external shell of reactor is also proposed to avoid dew point corrosion of the reactor steel shell. Conflict of interest The author has no conflict of interest to declare. References [1] Chaouki Sadik, El Amrani Iz-Eddine, Abderrahman Albizane, Recent advances in silica-alumina refractory: a review, J. Asian Ceram. Soc. 2 (2014) 83–96, doi:http://dx.doi.org/10.1016/j.jascer.2014.03.001. [2] D.A. Brosnan, Alumina-silica brick, in: C.A. Schacht (Ed.), Refractories Handbook, 2004, pp. 80–107. [3] V.L.K. Lou, A.H. Heuer, High-Temperature Corrosion of Technical Ceramics, Elsevier Applied Science, 1990, pp. 33–52. [4] A.G.M. Othman, N.M. Khalil, Sintering of magnesia refractories through the formation of periclase–forsterite–spinel phases, Ceram. Int. 31 (2005) 1117–1121, doi:http://dx.doi.org/10.1016/j.ceramint.2004.11.011. [5] ASME boiler and pressure vessel code, rules for construction of pressure vessel, Section II Part A Ferrous Metal Specifications, (2015) , pp. 928–933. [6] B. ZareNezhad, N. Hosseinpour, Evaluation of different alternatives for increasing the reaction furnace temperature of clauss SRU by chemical equilibrium calculations, Appl. Therm. Eng. 28 (2007) 738–744, doi:http://dx.doi.org/10.1016/j.applthermaleng.2007.06.014. [7] Samane Zarei, Hamid Ganji, Maryam Sadi, Mehdi Rashidzadeh, Thermo-kinetic modeling and optimization of the sulfur recovery unit thermal stage, Appl. Therm. Eng. 103 (2016) 1095–1104, doi:http://dx.doi.org/10.1016/j.applthermaleng.2016.05.012. [8] A. Garmroodi Asil, A. Shahsavand, Sh. Mirzaei, Maximization of sulfur recovery efficiency via coupled modification of GTU and SRU processes, Egypt. J. Pet. 26 (2017) 579–592, doi:http://dx.doi.org/10.1016/j.ejpe.2016.08.003. [9] Flavio Manenti, Davide Papasidero, Giulia Bozzano, Eliseo Ranzi, Model-based optimization of sulfur recovery units, Comput. Chem. Eng. 66 (2014) 244–251, doi:http://dx.doi.org/10.1016/j.compchemeng.2014.01.019. [10] H. Kazempour, F. Pourfayaz, M. Mehrpooya, Modeling and multi-optimization of thermal section of claus process based on kinetic model, J. Nat. Gas Sci. Eng. 38 (2017) 235–244, doi:http://dx.doi.org/10.1016/j.jngse.2016.12.038. [11] Samane Zarei, Hamid Ganji, Maryam Sadi, Mehdi Rashidzadeh, Kinetic modeling and optimization of Claus reaction furnace, J. Nat. Gas Sci. Eng. 31 (2016) 747–757, doi:http://dx.doi.org/10.1016/j.jngse.2016.03.086. [12] API recommended practice 936, Refractory Installation Quality Control Guidelines Inspection and Testing Monolithic Refractory Lining and Materials, American Petroleum Institute, Washington DC, 2004, pp. 1–23. [13] ASTM C 71, annual book of ASTM standards, Standard Terminology Relating to Refractories, (2012) , pp. 1–6. [14] Jiansheng Pan, Jianfeng Gu, in: Cemil Hakan Gur, Jiansheng Pan (Eds.), Handbook of Thermal Process Modelling of Steels, CRC Press Taylor & Francis Group, 2009, pp. 2–59. [15] API 571, Damage Mechanisms Affecting Fixed Equipment in the Refining Industry, American Petroleum Institute, 2011 pp 62. [16] Hussein Jwad Habeeb, Hasan Mohammed Luaibi, Thamer Adnan Abdullah, Rifaat Mohammed Dakhil, Abdul Amir H. Kadhum, Ahmed A. Al-Amiery, Case study on thermal impact of novel corrosion inhibitor on mild steel, Case Stud. Therm. Eng. 12 (2018) 64–68, doi:http://dx.doi.org/10.1016/j. csite.2018.03.005.
Contents lists available at ScienceDirect
Case Studies in Construction Materials journal homepage: www.elsevier.com/locate/cscm
Case study
Thermal damage of sulfur processed chamber under Claus operating reaction conditions—A case study S. Chidambaram Numaligarh Refinery Limited (A Group Company of Bharat Petroleum Corporation Limited), India
A R T I C L E I N F O
Article history: Received 19 September 2017 Received in revised form 11 April 2018 Accepted 12 April 2018 Available online 20 April 2018 Keywords: Thermal damage Refractory Numerical formulation Combustion chamber Industrial failure
A B S T R A C T
The sulfur from acid gas and sour gas is recovered to produce useful sulfur products. The claus reaction is industrially important process to recover sulfur from corrosive acid and sour gases. To achieve claus reaction, high temperature reactor is utilized under controlled conditions. The Claus combustion reaction heater made of 94% alumina refractory bricks were damaged under ideal equilibrium operating conditions. The first principle mathematical model was developed to measure the temperature profiles at hot face bricks and its interfaces to predict outer steel surface temperature. The maximum outer steel wall surface temperature was measured as 548.6 C using thermograph experiments and compared against the predicted temperature. Large variations in temperature differences have confirmed that the refractory wall was damaged due to abnormal reactions. This thermal damage is discussed and presented with various evidences from visual inspection. The hot face wall bricks, matrix blocks and orifice throat were damaged and presented by macroscopic visual inspection. However, hardness of the steel shell is within specified limit. Therefore, the refractory repair inside the reactor and thermal insulation of external shell has been proposed to prevent steel shell from creep, graphitization, and high temperature oxidization and corrosion damages. © 2018 The Author. Published by Elsevier Ltd. This is an open access article under the CC BYNC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
1. Introduction The refractory materials are designed to resist against heat, corrosion from industrial corrosive gases, mechanical and thermal stress, strains, and abrasion at higher temperatures. The performance of refractory directly depends upon its chemical composition, manufacturing method and its implementation methods [1]. Mostly, the refractories are used in steel industries, petroleum and petrochemical industries. The right choice of refractory selection at design stage is economically beneficial. In steel industry, the basic refractories are applied to the wall of basic oxygen furnace (BOF) [2]. The quality and productivity of the liquid steel is determined by the performance of applied refractory in addition to various other parameters [3]. Similarly, petroleum industries also line various basic refractories to inner wall of the reactor and pressure vessel on its hot face sides. The service life of the reactor and pressure vessel for chemical processing directly depends on the lined refractory. Alumina refractories known for its good performance is highly essential for steel processing and hydrocarbon industries. Alumina refractories ranges from 60% Al2 O3 to 99% Al2 O3 and are used in various applications depending upon the type of product. For example 60% Al2 O3 are used in general furnace and incinerator bricks, 94% Al2 O3 are used in internals of combustion chamber reactor walls and 99% Al2 O3 are used in certain specialty products [4]. The reduction of mullite in alumina increases the service temperature. Therefore, the performance of alumina refractory is better than fire clay refractory, particularly in better creep resistance and high temperature corrosion. However, the refractory performance will come down if it is operated beyond design temperatures. https://doi.org/10.1016/j.cscm.2018.04.005 2214-5095/© 2018 The Author. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/).
518
C. S. / Case Studies in Construction Materials 8 (2018) 517–529
One such case study from petroleum refinery is discussed in this paper. In petroleum industry, sulfur will be extracted from sour gas and acid gas which majorly contains H2 S. The schematic view of sulfur recovery process flow is shown in Fig. 1. The sour water generated from high pressure cracker and coker unit supplies sour gas feed to sulfur recovery unit. These sulfur recovery unit produces sulfur as a final product from crude processed hydrocarbon. The fundamental process is that partial conversion of H2 S to sulfur-di-oxide as stated in Eq. (1) and further reaction between remaining H2 S gas and SO2 gas will form sulfur and water as shown in Eq. (2) [6]. The first stage partial oxidation and second stage remaining combustion is called overall claus reaction as shown in Eq. (3) [7]. The amount of air feed to the process will define the amount of sulfur-di-oxide formation. Therefore, for maximum sulfur formation the ratio of
H2 S SO2
in the process gas was
maintained at the ratio of 2:1 [8] and accordingly air to gas ratio was maintained in-between 1.4 to 1.5 for clauss operating reaction [9]. The overall Claus reaction is controlled by air injection and this endothermic reaction approximately recovers 99% of sulfur from sour rich acid gas [10]. 1 1 1 1 H2 S þ O2 ! SO2 þ H2 O þ 41:3 kcal per mole 3 2 3 3
ð1Þ
2 1 2 1 H2 S þ SO2 ! H2 O þ S2 þ 7:4 kcal per mole 3 3 3 2
ð2Þ
H2 S þ
1 O2 ! S þ H2 O 2
Fig. 1. Schematic view of sulfur recovery process flow and sulfur reactor thermal damage encircled within unit.
ð3Þ
C. S. / Case Studies in Construction Materials 8 (2018) 517–529
519
The refractory used in mixing chamber was designed to resist against creep damage, material instability at high temperature and high temperature corrosion, particularly in sour environment. At high velocity flow regime, the particles present in hydrocarbon feed or in sour water results in abrasion damage to refractory. Numerous literatures are available for chemical reaction kinetics and process modeling of sulfur reactor for improving its performance [6–11]. However, no literature is available for thermal modeling and thermal damage especially for sulfur reactor. This paper discusses refractory degradation in sulfur reactor chamber due to high temperature operation. The temperature calculations were performed for ideal operating Claus reactions and root cause for detected damage is presented with temperature simulations. 2. Materials and methods The sulfur reactor was constructed with composite layer of steel chamber, insulating castable and refractory bricks as shown in Fig. 2. The nominal wall thickness of main burner chamber and combustion chamber steel shell made up of SA 516 Grade 70 steel (refer steel chemical composition in Table 4 is 10 mm and 14 mm respectively [5]. The corrosion allowance is specified as 3 mm for both the chambers. The main burner chamber was internally lined with 94% Al2 O3 alumina. However, the combustion chamber has been internally lined by insulating material and refractory bricks. The material of construction of insulating castable and refractory brick is 94% Al2 O3 alumina (refer Table 1). The physical and thermal property of 94% Al2 O3 alumina material is shown in Table 3 and Table 5 respectively. In main burner chamber, the refractory lined was 225 mm in thickness and additional 7 mm expansion allowance was provided in-between steel shell and refractory lining (refer Fig. 2a and b). In combustion chamber, the refractory brick wall is 230 mm thickness at hot face internal layer and 110 mm thickness at the subsequent second insulating layer ensuring that sulfur reactor is protected by double layer thermal insulation (refer Fig. 2c). Mortar made up of 94% Al2 O3 was applied to obtain grip and strength between each refractory brick of 2.5 mm approximately. The orifice throat has twelve tiles and matrix block has twenty blocks constructed within the combustion chamber (refer Fig. 2d and e). The chemical composition of acid gas stream feed is shown in Table 2. The grain size and water required for casting 94% Al2 O3 refractory is specified as 5 mm and 9–11% respectively [12]. The typical values mentioned in Table 1, Table 3 and Table 5 are average measured values. The physical and thermal property of 94% Al2 O3 refractory depends upon the proportion of SiO2 , Fe2 O3 and TiO2 constituents. Even small changes in the proportions of these oxide elements will drastically alter the performance of alumina refractory intended for high temperature applications [13], especially in sulfur reactor (Table 3 and Table 5). The main burner chamber with single castable lining and combustion chamber with thirty four rings were bolted to each other. Before bolting both the chambers, the linear expansion of 7 mm ceramic blanket was fixed on face side of main burner chamber. This insertion of ceramic blanket will provide linear expansion allowance at specified temperature [12,13]. The 3 nos. of V-lug anchor was dissimilarly welded with carbon steel main burner chamber inner wall surface at 1208 apart from each other. Similarly, large numbers of V-lug anchors were dissimilarly welded with carbon steel shell combustion chamber. These anchors made up of austenitic stainless steel (SS316L) holds the cast refractory in main burner chamber, and insulating castable and refractory brick layer in combustion chamber. These anchors were coated with bituminous coating and wrapped with plastic tape. The Minimum Metal Design Temperature (MMDT) is specified as 343 C for SA 516 Grade 70 steel plate. Both the main burner chamber and combustion chamber steel plates were initially gas tungsten arc welded (GTAW) using ER70S2 filler and subsequently weld joints were subjected to post weld heat treatment to reduce internal stress induced by welding process. Subsequently weld joints were radiographed and hydro tested before refractory lining, and the welded steel plates of both the chambers passed all structural integrity tests. The orifice throat between refractory brick on ring no: 10 and ring no: 11 was packed with 3 mm ceramic fiber paper of 1485 C grade for expansion allowance. The design pressure and operating pressure of sulfur reactor is 7 kg per sq cm and 0.45 kg per sq cm respectively. The minimum working pressure is 0.003 kg per sq cm. The design temperature is 1760 C and operated working temperature is in range of 900 C to 1400 C. One thermocouple has been inserted within thermo well at 300 mm from inner refractory for online monitoring of actual process temperature. 3. Experimental methods The thermograph was conducted by thermal analyzer using Testo make model 881 version V1.16. The thermograph measured with pre-defined emissivity of 0.92 in measurement range 0 C to 1500 C. The refractory failure at main burner chamber, combustion chamber, and orifice throat and matrix block sections were inspected visually. The damaged portions of refractory wall were subjected to photo documentation and presented macroscopically. The thermal heat transfer analysis of main burner chamber is simulated and thermal analysis simulations is performed under ideal Claus operating reaction conditions. The hardness of both main burner chamber and combustion chamber steel cover shell was measured in room temperature using MIC 201A GE model with diamond pyramid indenter supplied by general electric. The chemical composition of both chamber shell material was analyzed in room temperature conditions using x-ray fluorescence (XRF) supplied by oxford instrumentation X-Met 7500 model.
520
C. S. / Case Studies in Construction Materials 8 (2018) 517–529
Fig. 2. a) front view of main burner chamber b) side view of main burner chamber c) side view of combustion chamber d) front view of orifice throat and e) front view of matrix block.
C. S. / Case Studies in Construction Materials 8 (2018) 517–529
521
Table 1 Chemical composition of alumina refractory. Chemical Analysis
Typical value
Specification [1]
Al2 O3 SiO2 Fe2 O3 TiO2 CaO LOI
94.40 0.20 0.16 0.18 4.85 0.20
94.0 min 0.30 max 0.30 max – 5.00 max –
Table 2 Chemical composition of Acid gas stream. Weight percentage (%) Acid gas feed
H2 S 97.6
NH3 0.00
H2 0.0067
H2 O 2.2
Hydrocarbon 0.1
Table 3 Physical properties of Alumina Refractory. Parameters
Typical value
Specification [1,2]
Dry density in g/cc, after drying at 110 C for 24 h Cold crushing strength kg/cm2 After drying at 110 C for 24 h After heating at 800 C for 3 h After heating at 1100 C for 3 h After heating at 1550 C for 3 h
2.84
2.80 min
660 480 450 750
600 min 450 min 400 min 700 min
100 75 50 100 1
– – – – 5 max
Modulus of Rigidity kg/cm2 After drying at 110 C for 24 After heating at 800 C for 3 h After heating at 1100 C for 3 h After heating at 1550 C for 3 h % retained above maximum size
Table 4 Chemical composition of ASTM A516 Grade 70 [5]. Elements weight percentage (%)
Carbon
Manganese
Silicon
Phosphorous
Sulfur
Iron
ASTM Specifications
0.28 max
0.85–1.20
0.13–0.45
0.035 max
0.035 max
balance
Table 5 Thermal properties of Alumina Refractory. Parameters
Typical value
Specification [2]
Refractoriness Orton/ C
+38/+1835
37/1820 min
Permanent Linear Change % After heating at 800 C for 3 h After heating at 1100 C for 3 h After heating at 1550 C for 3 h
0.06 0.12 0.24
0.20 max 0.20 max 0.50 max
Thermal Conductivity, kcal/m/hour/8C at 400 C HF at 600 C HF at 800 C HF
1.37 1.40 1.42
– – –
522
C. S. / Case Studies in Construction Materials 8 (2018) 517–529
Fig. 3. Cylindrical coordinate system for main burner chamber of sulfur reactor with (r, u) expressed in polar coordinate.
4. Results and discussions 4.1. Numerical formulation The numerical heat conduction and convection equation were solved using Eq. (4) [14] and input data for solving these equations were incorporated from design data of main burner chamber. The heat transfer constitutive equation in radial coordinate system for main burner chamber of sulfur reactor (refer Fig. 3) has solved using time dependent boundary conditions.
2 2 @T 1@ @T 1@ T @ T ¼ a½ r þ 2 2 þ 2 r @r @r @t r @u @z
ð4Þ
a is thermal diffusivity
Under steady state condition, the heat flow in radial direction is re-written as in form of Eq. (5) [14] @ @T r ¼0 @r @r
ð5Þ
The Fourier constitutive heat conduction equation is stated as in Eq. (6) [14] Q ¼ kA
@T @r
ð6Þ
k is thermal conductivity or inverse of thermal resistance in watt per kelvin A is cross sectional area of sulfur reactor in mm, T is temperature in C and t is time in seconds The hot acid gas stream was fed in gas pipe and simultaneously air was injected to air pipe i.e. adjacent to gas burner to produce a claus combustion reaction at 1350 C in main burner chamber subject to initial time (to ) on internal diameter (ID) refractory wall surface. An ambient condition prevails at outer diameter (OD) metal wall surface at 50 C temperature subject to final time (tf ). The 7 mm expansion allowance between inner diameter of steel shell and outer diameter of refractory lining at claus reaction conditions resulted in increased inner diameter of refractory wall from 225 mm to 232 mm after attaining 1350 C. Therefore, increased diameter has been substituted at an initial boundary condition for main burner chamber of sulfur reactor. All input data required for solving equations were shown in Table 6. By integrating Eq. (5) and by applying Table 6 Input data used in numerical modeling. Input data
Unit
Value
Outer diameter of furnace wall Inner diameter of furnace wall Metal wall thickness of furnace wall Thickness of refractory wall Inner diameter of refractory wall Temperature of hot gas inside furnace Temperature of ambient air outside furnace Coefficient of thermal convection of air Coefficient of thermal convection of sour gas Coefficient of thermal conduction of steel Coefficient of thermal conduction of refractory
mm mm mm mm mm C C Wm2/ C Wm2/ C Wm2/ C Wm2/ C
904 884 10 232 420 1350 50 50 60 50.2 1.65
C. S. / Case Studies in Construction Materials 8 (2018) 517–529
523
initial boundary conditions, the constants from integration of Eq. (5) were obtained. Subsequently, back substituting an integration constant to temperature distribution (Eq. (4)) using Eq. (6) will result in computed thermal resistance. Thermal resistance at various layers, equivalent thermal resistance, heat flow, temperature distribution at hot face bricks, temperature at refractory steel shell plate interfaces and temperature at steel shell outer surface of main burner chamber were computed using numerical formulation as shown in Table 7. The temperature profile prediction is shown in Fig. 4 and temperature cooling profile was plotted across radial direction of main burner chamber exhibited parabolic curve (refer Fig. 5). From uni-dimensional model results, it shall be concluded that maximum surface temperature in outer diameter (OD) wall surface was 150.02 C. Table 7 Numerical modeling output value. Output results obtained from model
Unit
Thermal resistance (R1) Thermal resistance (R2) Thermal resistance (R3) Thermal resistance (R4 Equivalent thermal resistance (R) Heat Flow Temperature at hot face refractory wall Temperature at interface between refractory wall and steel wall Temperature at outer diameter (OD) wall of metal surface
kelvin kelvin kelvin kelvin kelvin watt C C C
Computed value watt1 watt1 watt1 watt1 watt1
Fig. 4. Predicted temperature profile at various layer of main burner chamber.
Fig. 5. Temperature plot of main burner chamber of sulfur reactor in radial direction.
0.01263775 0.07182034 7.0965e-05 0.00704583 0.09157488 14196.0325 1170.59 151.03 150.02
524
C. S. / Case Studies in Construction Materials 8 (2018) 517–529
4.2. Thermography and visual inspection The thermography was conducted on steel surface of main burner chamber outer diameter (OD) wall and it was noted that steel surface experiences high temperature than the design limit. On observing higher surface temperature on both main burner chamber and combustion chamber, the sour gas fluid flow was terminated within the reactor system and then reactor was subjected to equilibrium cooling. After attaining room temperature, this sulfur reactor was subjected to detailed inspection. The root cause for thermal damage is analyzed and corrective actions for thermal damage prevention are discussed in detail. The surface temperature of 548.6 C was noted at main burner chamber shell and combustion chamber shell as shown in thermograph Fig. 6a and b. The corresponding physical location of both main burner chamber and combustion chamber is shown in Fig. 6. Similarly, higher temperature was observed exactly at 12 o’clock position of the flange connections between both the chambers along with detection of gas leakage (refer Fig. 6c). It was observed that the high temperature paint coating was peeled off from both the chamber shells due to high temperature. The paint peeled off partially on chamber surface exposed the bare metal. These high temperature paints were designed to protect the bare shell surface against atmospheric corrosion, rain water and dew point corrosion. However, the paint was still intact at certain location. It is suspected that the
Fig. 6. Thermograph at corresponding locations a) 1 o’clock position b) 11 o’clock position and c) 6 o’clock position of sulfur reactor while operating at Claus operation reaction conditions.
C. S. / Case Studies in Construction Materials 8 (2018) 517–529
525
Fig. 7. a) Circumferential crack in main burner chamber front view b) air injection nozzle damage c) air injection nozzle melt out partially and oxidized and d) gas nose cone geometry features lost. a)-d) in as received condition after thermal damage.
internal refractory might be damaged only at that paint peeled off locations. The bolt locations between both the chambers were suspected for leakage in addition to high temperature damage. The refractory bricks and mortar lining may dislodge from its locations due to internal thermal stress induced by linear thermal expansion resulting to leakage of sour gas. The absence of ceramic blanket on face side of main burner chamber restraints the refractory casting. 4.3. Main burner chamber The main burner chamber and combustion chamber were dismantled and inspected in detail. The main burner chamber was observed with circumferential crack as shown in Fig. 7a. The circumferential crack propagated in brittle fashion perpendicular to induced thermal stress. The refractory lining was completely dislodged from inner steel shell in main burner chamber which confirms that internal refractory lining was completely under high temperature service. In addition to observed crack at refractory mid wall and dis-bonding of refractory lining with steel shell, hydrocarbon deposits were also found at main burner chamber ID as black and slag debris (refer Fig. 7a, b and c). This black debris and slag debris are suspected as hydrocarbon and refractory or metal meltdown respectively. The sour gas obtained from sour water after chemical refining and amine treating contains hydrocarbon and this residual hydrocarbon might have increased the temperature within reactor during Claus reactions. The air nose cone on main burner chamber was found damaged and nearly one fourth of the cone section was found melted (refer Fig. 7d). The material of construction of nose cone was verified by x-ray analyzer and it was confirmed as AISI 310 austenitic stainless steel material as specified in original design. The slag of melted nose cone was suspected for slag debris on ID of main burner chamber. Therefore, the temperature within main burner chamber has reached above melting point of nose cone leading to severe thermal damage. All the purge connections in main burner chamber were found in good structural condition. 4.4. Combustion chamber The combustion chamber was visually inspected in detail and it confirms that sulfur reactor was severely damaged and degraded as the reactor was operated beyond design temperature. The 94% Al2 O3 alumina bricks were found degraded up to
526
C. S. / Case Studies in Construction Materials 8 (2018) 517–529
Fig. 8. a) and b) Dislodged bricks fallen on ID refractory wall, c) and d) loose bricks hanging at 12 o’clock, e) mortar spalling and brick dislodge at 12 o’clock position and f) bricks crack at 9 o’clock position. a)-f) in as received condition after thermal damage.
10 mm on hot face out of 230 mm of total thickness in ring no:1 and ring no: 2 at 12’o’ clock position of reactor (refer Fig. 8a and b). Similarly, the bricks were degraded at a maximum up to 8 mm on hot face side from ring no: 4 to ring no: 8. These damages were observed from 11’o’ clock to 1’o’ clock position of the reactor (refer Fig. 8c, d and e). The clock position of reactor is represented with respect to front side of the combustion chamber. The dislodged refractory pieces of alumina bricks were found lying on downstream of the chamber. The alumina bricks degraded up to 8 mm on hot face out of total 230 mm thick brick from ring no: 9 to ring no.16. These damages were observed exactly from 10’o’ clock to 3’o’ clock reactor positions. The cracks were observed in refractory bricks on ring no.17 at 9 o’clock position (refer Fig. 8f) 4.5. Orifice throat and matrix block The orifice throat and matrix block inside the reactor was also observed with severe thermal damages. Each brick of orifice throat and matrix block is called as tile and block respectively. The orifice throat is constructed by ten numbers of tiles
C. S. / Case Studies in Construction Materials 8 (2018) 517–529
527
counting tile no.1 to tile no.10 in anticlockwise direction of reactor. Each individual tile from tile no.1 to tile no.6 were dislodged from its original position and were found lying in downstream of orifice throat (refer Fig. 9a, b and c). However, the remaining tiles i.e. from tile no. 7 to tile no. 10 were intact with reactor wall. The matrix block was completely dislodged from its original position and fell down to ring no. 18 and ring no.19 i.e. towards main burner chamber burner side (refer Fig. 9d). In addition, several loose alumina refractory bricks and mortars were found lying on ring no: 17 and ring no. 18 at 12 ‘o’ clock reactor position. The bricks from ring no.19 and ring no.20 was degraded and damaged up to 10 mm on hot face out of total 230 mm thick brick at 12 ‘o’ clock reactor position. Similarly, the loose refractory lining fell upon ring no: 27 to ring no. 29 at 12 ‘o’ clock reactor position. However, alumina bricks from certain rings i.e. from ring no: 21 to ring no. 26 and from ring no.30 to ring no.35 was intact and no damages were observed. 4.6. Relining after thermal damage The black and slag debris is to be chipped out completely from the ID of refractory wall of main burner chamber and to be relined with 94% alumina mortar material. The refractory bricks damaged on various abovementioned rings are proposed for relining with 94% alumina mortar for required thickness to maintain the radius of curvature and profile at various sections in internal diameter of combustion chamber. The dislodged tiles from tile no.1 to tile no.6 are proposed to be reconstructed by new 94% alumina bricks in the orifice throat and further expansion allowance of 3 mm is to be provided at an interface between orifice throat and refractory bricks. Similarly, ceramic fiber of 1425 C grade is proposed to be packed between ring no.2 and ring no.3, between ring no.18 and ring no.19, between ring no.23 and ring no.24, and between ring no.28 and ring no.29 with 5 mm expansion allowance. The new burner nose cone (refer Fig. 10) is to be fixed and assembled in main burner chamber. 4.7. Characterization of metallic shell surface after refractory thermal damage The chemical composition of metal shell wall was measured (refer Table 8) and its chemical composition matches with SA516 Grade 70 material as per international codes and standard ASTM/ASME limit [5]. However, x-ray analyzer was not able
Fig. 9. Orifice throat collapsed a) at 9 o’clock ‘position b) magnified A location of figure a, c) amplified B location at 2 o’clock position of figure a and d) matrix block collapsed. a)-d) in as received condition after thermal damage.
528
C. S. / Case Studies in Construction Materials 8 (2018) 517–529
Fig. 10. Gas nose cone a) back view and b) front view after repair.
Table 8 Chemical composition of thermal damaged shell metal. Elements weight percentage (%)
Carbon
Manganese
Silicon
Phosphorous
Sulfur
Iron
Measured value
–
0.88 0.02
0.16 0.03
0.035 0.02
0.04 0.01
99.0 0.01
to detect carbon percent in the material due to its limitations. The hardness of metallic shell was measured (refer Fig. 11) and it was found within the limit as prescribed by the codes and standards [5]. The measured hardness value have confirmed the absence of unwanted hard or any brittle phases in the micro structure of steel shell. The shell material must resist against creep, graphitization, high temperature corrosion and oxidation. 5. Mechanism of refractory degradation The main causes for refractory degradation are mechanical, thermal and corrosion damage or any of these combinations. The thermo mechanical stress-strain effect will lead to spalling, thermal shock and crack. The mechanical stress-strain effect will lead to erosion and abrasion and the chemical attack will lead to corrosion as a result of oxidation or sulfidation. The restraint force induced between various bricks, mortar, insulating castable and steel shell due to temperature difference, material heterogeneity and anisotropic on reactor wall generates single or multiple cracks in refractory wall, especially on
Fig. 11. Shell surface hardness in as received thermal damaged conditions.
C. S. / Case Studies in Construction Materials 8 (2018) 517–529
529
bricks, mortar and castable depending upon type of localized stress and interface conditions [15]. The brittle crack initiated in refractory material will lead to fracture depending upon chemical composition, manufacturing method of refractory material, crack size, geometry, critical stress intensity factor and temperature. The stress intensity factor (KI ) of flaw in the refractory reaches or exceeds the critical stress intensity factor (KI Kc ) propagating the crack in refractory. Anchor used for industrial refractory applications should be resistant to oxidation, sulfidation and condensing dew point corrosion. Further, refractory would degrade if thermal coefficient expansion of anchor is not compatible with steel shell. The pre-existing crack present in the refractory will lead crack surface to close up as a result of compressive stress. Contrastingly, the pre-existing crack in the refractory will lead to open and propagate crack surface further as a result of tensile stress. Therefore, after cooling from firing temperature, the cracks in the refractory brick apparently appears in naked eyes. The thermal impact on corrosion behavior of carbon steel is high [16]. An increase in temperature will increase the rate of corrosion in steel shell, if hot corrosive gases pass through these refractory cracks. 6. Conclusions The temperature measured using thermograph experiments on outer surface of main burner chamber was much higher than the predicted temperature. The refractory wall is severely damaged as the reactor was operated beyond design temperature and it was confirmed by appearance of damages. The hydrocarbon carryover in the feed gas stream favored an un-controlled reaction proportionally increases the operating temperature beyond the design temperature limit which might be the probable cause of failure. The outer steel shell surface temperature exceeds the creep design temperature. The termination of claus condition beyond 1400 C resulted to damage. The proposed sour water treatment in surge drum to remove hydrocarbon traces will prevent thermal damages in claus condition operating reactors. The hydrocarbon in the sour water should be chemically separated such that no hydrocarbon will remain in the sour gases. Insulating the external shell of reactor is also proposed to avoid dew point corrosion of the reactor steel shell. Conflict of interest The author has no conflict of interest to declare. References [1] Chaouki Sadik, El Amrani Iz-Eddine, Abderrahman Albizane, Recent advances in silica-alumina refractory: a review, J. Asian Ceram. Soc. 2 (2014) 83–96, doi:http://dx.doi.org/10.1016/j.jascer.2014.03.001. [2] D.A. Brosnan, Alumina-silica brick, in: C.A. Schacht (Ed.), Refractories Handbook, 2004, pp. 80–107. [3] V.L.K. Lou, A.H. Heuer, High-Temperature Corrosion of Technical Ceramics, Elsevier Applied Science, 1990, pp. 33–52. [4] A.G.M. Othman, N.M. Khalil, Sintering of magnesia refractories through the formation of periclase–forsterite–spinel phases, Ceram. Int. 31 (2005) 1117–1121, doi:http://dx.doi.org/10.1016/j.ceramint.2004.11.011. [5] ASME boiler and pressure vessel code, rules for construction of pressure vessel, Section II Part A Ferrous Metal Specifications, (2015) , pp. 928–933. [6] B. ZareNezhad, N. Hosseinpour, Evaluation of different alternatives for increasing the reaction furnace temperature of clauss SRU by chemical equilibrium calculations, Appl. Therm. Eng. 28 (2007) 738–744, doi:http://dx.doi.org/10.1016/j.applthermaleng.2007.06.014. [7] Samane Zarei, Hamid Ganji, Maryam Sadi, Mehdi Rashidzadeh, Thermo-kinetic modeling and optimization of the sulfur recovery unit thermal stage, Appl. Therm. Eng. 103 (2016) 1095–1104, doi:http://dx.doi.org/10.1016/j.applthermaleng.2016.05.012. [8] A. Garmroodi Asil, A. Shahsavand, Sh. Mirzaei, Maximization of sulfur recovery efficiency via coupled modification of GTU and SRU processes, Egypt. J. Pet. 26 (2017) 579–592, doi:http://dx.doi.org/10.1016/j.ejpe.2016.08.003. [9] Flavio Manenti, Davide Papasidero, Giulia Bozzano, Eliseo Ranzi, Model-based optimization of sulfur recovery units, Comput. Chem. Eng. 66 (2014) 244–251, doi:http://dx.doi.org/10.1016/j.compchemeng.2014.01.019. [10] H. Kazempour, F. Pourfayaz, M. Mehrpooya, Modeling and multi-optimization of thermal section of claus process based on kinetic model, J. Nat. Gas Sci. Eng. 38 (2017) 235–244, doi:http://dx.doi.org/10.1016/j.jngse.2016.12.038. [11] Samane Zarei, Hamid Ganji, Maryam Sadi, Mehdi Rashidzadeh, Kinetic modeling and optimization of Claus reaction furnace, J. Nat. Gas Sci. Eng. 31 (2016) 747–757, doi:http://dx.doi.org/10.1016/j.jngse.2016.03.086. [12] API recommended practice 936, Refractory Installation Quality Control Guidelines Inspection and Testing Monolithic Refractory Lining and Materials, American Petroleum Institute, Washington DC, 2004, pp. 1–23. [13] ASTM C 71, annual book of ASTM standards, Standard Terminology Relating to Refractories, (2012) , pp. 1–6. [14] Jiansheng Pan, Jianfeng Gu, in: Cemil Hakan Gur, Jiansheng Pan (Eds.), Handbook of Thermal Process Modelling of Steels, CRC Press Taylor & Francis Group, 2009, pp. 2–59. [15] API 571, Damage Mechanisms Affecting Fixed Equipment in the Refining Industry, American Petroleum Institute, 2011 pp 62. [16] Hussein Jwad Habeeb, Hasan Mohammed Luaibi, Thamer Adnan Abdullah, Rifaat Mohammed Dakhil, Abdul Amir H. Kadhum, Ahmed A. Al-Amiery, Case study on thermal impact of novel corrosion inhibitor on mild steel, Case Stud. Therm. Eng. 12 (2018) 64–68, doi:http://dx.doi.org/10.1016/j. csite.2018.03.005.