Engineering Failure Analysis xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Engineering Failure Analysis

journal homepage: www.elsevier.com/locate/engfailanal

Fitness for service assessment of a pressure vessel subjected to fire

damage in a refinery unit
R. Bakhtiaria,⁎, S. Zangeneha, M. Bakhtiari Fotouhb, S.M. Jamshidib, A. Shafeieb
a
Department of Materials and Textile Engineering, Faculty of Engineering, Razi University, Kermanshah, Iran
b
Kermanshah Oil Refinery Company (KORC), Kermanshah, Iran

AR TI CLE I NF O AB S T R A CT

Keywords: The fire-damaged pressure vessel reactor, evaluated in this study, was a cylindrical vessel with
Fitness for service design conditions of 460 psig (3.17 MPa) at 413 °C. The fitness for service (FFS) of the fire-
Pressure vessel damaged pressure vessel was assessed according to API 579-1/ASME FFS-1. First, the maximum
ASTM A204 temperature during the fire accident was determined based on deterioration of outer protective
layer. Then, the possible damage was examined by hardness test, in-situ field metallography and
metallographic replicas. Grain growth and spheroidized pearlite were observed in the base metal
at the inner surface of the vessel. Outer surface of the vessel showed ferritic microstructure with
smaller grain size as compared to the inner surface. Decarburizing and carbide formation were
also visible at the outer surface. The lower hardness of the base metal at both sides of the vessel
was compared to the standard value and related to the microstructure evolution during high
temperature/fire exposure condition. The results of the FFS evaluation indicated that the vessel is
not suitable for the current design conditions, and therefore a new maximum allowable working
pressure was determined.

1. Introduction

Some parts of refineries and petrochemical industries may be damaged and need to be assessed to ensure continued safe op-
eration. Replacement is frequently not a good option because of high capital costs, and the much lower cost of continuing the
operation of the older plant. However, reliability and safety are issues that have become much more important in recent years, so the
assessment of damage and the failure risk have become increasingly important [1], especially for pressure vessels used in petroleum
industries [2].
Design methods and code structure generally have specific damage tolerances and their application for damage assessment during
the operation life is likely to produce improperly conservative assessments. Therefore, design codes do not provide rules to evaluate
equipment that degrades during service, and deficiencies due to degradation or original fabrication that may be found during
subsequent inspections [3]. Fitness-for-service (FFS) assessment method has been developed in recent years to deal with this chal-
lenge by (I) assessment of the current state of the (damaged) structure, (II) extrapolation from the current state to estimate the
remaining safe and serviceable life, and (III) providing guidelines to make run, rerate, repair, or replace decisions about aging
pressure components and defect-containing structures. API 579-1/ASME FFS-1 defines FFS as quantitative engineering evaluations
that are performed to demonstrate the structural integrity of an in-service component that may contain a flaw or damage.

⁎
Corresponding author.
undefinedr.bakhtiari@razi.ac.ir",0,0,2
>r.bakhtiari@razi.ac.ir (R. Bakhtiari).

http://dx.doi.org/10.1016/j.engfailanal.2017.07.020
Received 15 July 2016; Received in revised form 7 July 2017; Accepted 12 July 2017
1350-6307/ © 2017 Published by Elsevier Ltd.

Please cite this article as: Bakhtiari, R., Engineering Failure Analysis (2017), http://dx.doi.org/10.1016/j.engfailanal.2017.07.020
R. Bakhtiari et al. Engineering Failure Analysis xxx (xxxx) xxx–xxx

In general, FFS assessments are quantitative engineering evaluations, which are especially used in oil and gas industry. Crack-like
flaws is one of the major failure modes of pressure vessels. Furthermore, pressure vessels have been assessed using FFS according to
pitting corrosion and thermal hot spots [4–6].
If the results of a FFS assessment indicate that the equipment is suitable for the current operating conditions, then the equipment
can continue to be operated at designed condition provided that suitable monitoring/inspection programs are established. If the
results of the FFS assessment indicate that the equipment is not suitable for the current operating conditions, then the equipment can
be rerated using the calculation methods based on API 579-1/ASME FFS-1. These calculation methods can be used to find a reduced
maximum allowable working pressure (MAWP) [7].
Low alloy steels, used in pressure vessels and piping at high temperatures, typically have a microstructure of pearlite colonies in a
matrix of ferrite grains of a given average size. With extended exposure to high temperatures, the pearlite decomposes to form a
dispersion of carbides in a ferrite matrix. Similar effects occur in carbon–molybdenum and carbon–chromium–molybdenum steels.
Over-aging promotes the coalescence of the small carbides, formed during the pearlite decomposition, and prolonged aging leads to a
ferritic structure including spheroidized carbides with an almost uniform distribution. This occurs after prolonged exposure to
temperatures higher than 400 °C for carbon steels, and of around 420 °C for low alloy steels. For low-carbon steel pressure vessels, the
short-time heat exposure higher than 850 °C, the temperature level for the complete transformation to austenite, could produce grain
refinement rather than grain growth. Therefore, estimating the temperatures of heat exposure based on grain size is difficult [8].
Furthermore, decarburizing could occur for these steels during exposure to high temperatures which is important for quality control
of the related parts as can affect the mechanical properties such as hardness, wear and fatigue resistance. Destructive methods of
determining the depth of decarburized layer include metallographic and hardness tests [9].

2. Background of the damaged pressure vessel reactor

The fire-damaged pressure vessel reactor (named V-201), evaluated in this study, was a cylindrical vessel which was used in
Kermanshah Oil Refinery Company (Fig. 1). The reactor had a 5969 mm height, 1067 mm inside diameter and 20.2 mm (including
3.2 mm thick cladding material) thick wall. The reactor was fabricated from SA-204 Grade 70 steel with an A-240 corrosion-re-
sistance cladding. The vessel design conditions are 460 psig (3.17 MPa) at 413 °C, and the weld joint efficiency (E) is 1. A future
corrosion allowance of 1/16 in. (1.59 mm) is required for operation.
The pressure vessel reactor was subjected to a fire damage due to spontaneous combustion. According to observations of the fire
marshal, the flames expanded upward and covered half of the reactor. Since an accidental fire is a random event, extensive data
collection during an accidental fire is seldom possible. Unfortunately, in this case, there was no evidence such as videotape to
determine the nature and extent of the fire. Whereas videotape evidence is available, it could be possible to deduce the nature of the
fuel, the fire's progression from its ignition source, and temperature extremes from visual evidence on the tape. Therefore, the data of
the fire damage was collected after the fire was extinguished.

3. Experimental/FFS procedure

To investigate the damage of V-201 vessel reactor, three different zones on the inner surface and two ones on the outer surface
were studied to determine the extent of damage. The base metal of the reactor is a low alloy steel (ASTM A204) with chemical
composition given in Table 1. For the selected zones, surface preparation and replicating were conducted according to ASTM E 1351
standard. On the inner surface, the cladding was removed and preparation was carried out on the base metal surface. Then, me-
tallographic preparation of the replicas was conducted using ASTM E 3-01(2007):2010 standard. Micro-etching of the Au-coated
replicas was performed through ASTM E 407-07:2010 standard using Nital-2% etchant. Olympus-DP12 optical microscope was used
to study the microstructures, using ASTM E 883-02(2007): 2010 standard. An image analyzer software was used to measure phase

Fig. 1. Image of V-201 vessel reactor in Kermanshah Oil Refinery Company (KORC), showing different studied zones.

2
R. Bakhtiari et al. Engineering Failure Analysis xxx (xxxx) xxx–xxx

Table 1
Chemical composition of ASTM A204 steel (wt%).

Fe C Mn Ni Mo P S Si

Bal 0.25 0.9 0.45–0.6 0.45–0.6 0.035 0.04 0.15–0.3

volume fraction and grain size, according to ASTM E 112-96(2004): 2010 and ASTM E 562-11: 2013 standards, respectively. A
portable Brinell hardness tester was also utilized to measure the hardness value of various studied zones using ASTM E110-14
standard. Five measurements were conducted on each zone and the mean hardness values were reported.
Fitness for service (FFS) assessment was also performed to investigate the extent of damage and to estimate the efficiency of
reactor. This analysis was performed according to API-579 standard at levels I, II and III.

4. Microstructure studies

In this investigation, attempts were made to achieve the extent of the damage during the fire event; one aspect to accomplish this
goal was through microstructural observations. The followings are the related results of the degraded microstructures in the selected
zones. Fig. 2 shows schematic illustration of the reactor vessel to indicate the selected zones for microstructural studies.

4.1. Zone A

This zone (Fig. 2.a) was selected in the damage region of the inner surface of the reactor (at a distance of 200 cm from the lower
weld line). Fig. 3 shows the microstructure of zone A which is dominantly ferritic-pearlitic with spheroidized pearlite. During heating
the steels below the austenitizing temperature, the microstructure remains the same except that the cementite within the pearlite
phase tend to be spheroidized [8].
The amount of pearlite volume fraction and ferrite grain size were 30% and 5ASTM, respectively. Spheroidizing occurs generally

Fig. 2. Schematic illustration of the reactor vessel showing different selected zones in a) inner and b) outer surface.

3
R. Bakhtiari et al. Engineering Failure Analysis xxx (xxxx) xxx–xxx

Fig. 3. Optical micrograph of zone A at different magnifications.

at high temperatures and accompanies increase of grain size. Heating and cooling of steels from beyond the austenitizing temperature
usually causes a microstructure with lower grain size, if the heating is applied for a short period of time. During a fire, this heating
could not necessarily produce grain growth. The higher the overheating temperature, the faster and the larger the grains will grow
[10].

4.2. Zone B

Fig. 2.a shows Zone B which was selected in the damage region of the inner surface of the reactor (at a distance of 250 cm from the

Fig. 4. Optical micrograph of zone B at different magnifications.

4
R. Bakhtiari et al. Engineering Failure Analysis xxx (xxxx) xxx–xxx

Fig. 5. Optical micrograph of zone C at different magnifications.

lower circumferential weld line). Microstructure of this zone, shown in Fig. 4 at different magnifications, reveals ferrite and
spheroidized pearlite. For low-carbon steels, long-time exposure to temperatures in the range of 600–727 °C may cause some degree
of spheroidization within the pearlite phase, but this microstructural evolution could be unlikely for short exposure time [8].
Pearlite volume fraction and ferrite grain size were calculated as 30% and 5ASTM, respectively. Spheroidization could be due to
increase of temperature during ignition of the reactor. The microstructure of zones A and B, in the damage region of the inner surface,
are close.

4.3. Zone C

This zone (Fig. 2.a) was selected in the damage region of the inner surface of the reactor at heat affected zone (HAZ) of the upper
circumferential weld line. Fig. 5 shows the microstructure of zone C at different magnifications. Fig. 5.a, related to base metal/HAZ
interface, shows ferritic-pearlitic microstructure. Fig. 5b–d are related to HAZ and also show a microstructure including ferrite and
spheroidized pearlite. Pearlite volume fraction was calculated of 10% which is lower than those of zones A and B. The reason could be
the heat input of HAZ during welding which caused decomposition of pearlite and reduction of its volume fraction. A non-equilibrium
cooling process is occurred after welding which could not result in pearlitic transformation. So the pearlitic decomposition during
welding could be remained and resulted in carbide formation and increase of ferrite volume fraction. Furthermore, the grain size was
calculated as 8ASTM, lower than those of previous studied zones. The distance of this zone from the damage region is higher than the
zones A and B. Therefore, the generated heat during the damage caused increase of grain size in these zones. As grain growth needs
exposure to a high temperature for a long time, service condition of the reactor (417 °C) for several years could be a complementary
reason for grain growth.

4.4. Zone D

In addition to the studied zones at the inner surface, two zones at the outer surface of the reactor were also studied. The objective
of this investigation was to study the effect of damage on the base metal where cladding not to be considered. In this way, zone D
(Fig. 2.b) was selected in the damage region at the outer surface of the reactor (at a distance of 200 cm from the lower circumferential
weld line).
Fig. 6 shows the microstructure of zone D including ferrite with low amount of grain boundary carbides. Considerable reduction
of pearlite volume fraction and formation of dominantly ferritic microstructure could be due to partial decarburizing at the outer
surface of the reactor. But this decarburizing has no certain reason and could be probable. The reason of not observing decarburizing
in the inner surface could be cladding and its effect on carbon diffusion. The final microstructure of low-carbon steels after heating
and cooling could be finer ferrite-pearlite with lower volume fraction of ferrite. Since bulk decarburization of the steel was not
considered, distribution of cementite within the microstructure after the thermal cycle caused reduction of pearlite volume fraction
[10].

5
R. Bakhtiari et al. Engineering Failure Analysis xxx (xxxx) xxx–xxx

Fig. 6. Optical micrograph of zone D at different magnifications.

During service condition of the reactor, an oxide layer could be formed on the outer surface due to exposure in air. The thickness
of oxide layer at the outer surface was measured as a mean of about 3 mm. This thickness could not inhibit heat transfer during
ignition at a high extent but could be act as an insulation layer. This insulation layer could be effective on pearlite decomposition to
carbides. These grain boundary carbides are visible in Fig. 6. Formation of carbides increased the probability of partial decarburizing.
Grain size was calculated as 8ASTM, similar to that of zone C in HAZ of circumferential weld line at the inner surface. It could be
concluded that initial grain size of the base metal was about 8ASTM and the microstructure evolution during ignition caused its
increase at the inner surface. If not much time is allowed for grain growth and heating proceeds to complete the transformation of the
steel microstructure to austenite, the final resulting microstructure after cooling will be a finer ferritic-pearlitic structure compared to
the original one. Noticeable grain growth could be evident when the steel is heated to much higher temperature levels, for example,
to 1000 °C [4]. Therefore, the observed grain growth at the inner surface could reveal the temperature raise to higher than 1000 °C,
during the ignition.
Residual stresses between the cladding and base metal at the inner surface contributed to recrystallization and grain growth
during the ignition. Recovery, recrystallization and grain growth are the consequent stages of annealing the steels at high tem-
peratures. A degree of prior cold working as residual stresses is mandatory for nucleation of new grains during recrystallization and
consequent grain growth [11].
In general, the grain growth process is highly dependant on the amount of prior cold working [10]. The manufacturing process of
the reactor included stages with plastic deformation as cold working. Furthermore, the studied reactor had a repair process in which
some of the parts were metal- formed including cold working.
However, at the outer surface, due to no residual stresses as well as decreasing temperature gradient from the inner to the outer
surface, considerable grain growth could not occur.

4.5. Zone E

Zone E (Fig. 2.b) was selected out of the damage region at the outer surface of the reactor (close to the lower circumferential weld
line). Fig. 7 shows the microstructure of this zone at different magnifications which is dominantly ferritic, similar to zone D. However,
the carbide phase is visible in the grains as well as at the grain boundaries and its content increased in comparison with that of zone
D. Closeness of microstructural features between zones D and E, which are in the damage region and far from this zone, respectively,
shows that the source of decarburizing was not necessarily the ignition and long-time servicing of the reactor could be the another
source. Furthermore, lower content of carbides in the microstructure of zone D could reveal higher degree of decarburizing. Indeed,
the ignition accelerated the decarburizing.
Observing spheroidization of the cementite or even partial austenitization could not reject a pressure vessel which was subjected
to a fire damage. These microstructural changes influence the creep properties of a low-carbon steel but have less effect on its room
temperature properties [8]. Therefore, hardness test could be conducted to determine the degree of softening.

6
R. Bakhtiari et al. Engineering Failure Analysis xxx (xxxx) xxx–xxx

Fig. 7. Optical micrograph of zone E at different magnifications.

5. Hardness study

Table 2 shows variation of hardness at different studied zones. According to ASTM A204/A204M-12 standard and converting the
ultimate tensile strength of the reactor base metal to hardness, the allowed range of hardness values was determined as 133-171HB.
According to Table 2, zones A and B, at the inner surface of the reactor, have close hardness values but lower than the allowed range.
This could be due to increase of grain size and pearlite spheroidizing in the damage region. The hardness value of zone C, in HAZ of
circumferential weld line at the inner surface, is higher than those of zones A and B and is equal to the lower limit of the allowed
range. Since, distance of zone C from the damage region is higher and its grain size is lower. Also, the mean hardness value of zones D
and E, at the outer surface, is higher than those of the zones at the inner surface but still lower than the allowed limit. Higher relative
hardness of these zones in spite of decarburizing and ferritic microstructure could be due to carbides formation and lower grain size.
Therefore, hardness values of the base metal at both sides of the vessel are lower than the standard limit, due to increase of grain size
and partial decarburizing, respectively. Considerable softening, detected by hardness testing, might be a good reason to reject a
vessel, but softening may not occur in short-term heating of low carbon steels, unless heated to very high temperatures [8]. Fur-
thermore, a reduction in hardness may correspond to a reduction in yield strength but not necessarily to a reduction in ultimate
tensile strength. Low-carbon steels only turn hard and brittle if to be austenitized during the fire and cooled very rapidly by fire-
fighting water. This will be discussed using FFS.

6. Fitness-For-Service (FFS) assessment

Fitness-For-Service (FFS) assessment procedures can be used to evaluate pressurized components containing flaws or damage.
API-571 in section 11 provides FFS procedures by three levels of assessment for pressurized components potentially damaged by
exposure to a fire. Generally, each assessment level provides a balance between conservatism, the amount of information required for
the evaluation, the skill of the personnel performing the assessment and the complexity of analysis being performed.
The assessment procedures in level 1 are intended to provide conservative screening criteria that can be utilized with a minimum
amount of inspection or component information. If the pressurized components are rejected in this level, level 2 assessment should be

Table 2
Hardness and estimated ultimate tensile strength (UTS) at different studied zones in comparison with ASTM-A204 standard.

ASTM-A204 Zone A Zone B Zone C Zone D Zone E

Hardness (HBN) 133–171 97–105 100–104 127–133 104–110 110–120

Mean value (HBN) 152 101 102 130 107 115
UTS, ksi (MPa) 75 (517) 46 (317) 47 (324) 65 (448) 49 (338) 58 (400)

7
R. Bakhtiari et al. Engineering Failure Analysis xxx (xxxx) xxx–xxx

performed. The assessment procedures in level 2 are intended to provide a more detailed evaluation that produces results that are
more precise than those from a Level 1 assessment. In a level 2 assessment, inspection information similar to that required for a Level
1 assessment are needed; however, more detailed calculations are utilized in the evaluation. Similar to level 1, if the components are
rejected, level 3 should be performed. The assessment procedures in level 3 are intended to provide the most detailed evaluation
which produces results that are more precise than those from a Level 2 assessment. In a level 3 assessment, the most detailed
inspection and component information are typically required, and the recommended analysis is based on numerical techniques such
as the finite element method or experimental techniques when appropriate.
If the results of the FFS evaluation indicate that the equipment is not suitable for current design conditions, then one of the
following methods may be used. a) A new maximum allowable working pressure, MAWP, maximum design temperature, and/or
minimum design metal temperature may be established using the appropriate evaluation procedures. b) Defective sections of the
equipment may be repaired or replaced. c) The equipment may be retired from the service.

6.1. Level 1 assessment procedure

The acceptability for continued service of the pressurized components in the level 1 assessment procedure is based only on heat
exposure zone. The objective of this Level is to document the observations and data used to assign a heat exposure zone for the fire-
damaged component. The heat exposure zone, based on field observations, is typically established after the fire is out. In this case,
observations after the fire indicated that external protective fiberglass has deteriorated during a process upset. In addition, the
internal thickness readings showed that 3% metal loss occurred due to the past operation.
The deterioration of the fiberglass at the outer surface of V-201 reactor (Fig. 1) was used as a temperature indicator to assign the
appropriate heat exposure zone. It seems that V-201 reactor was subjected to a higher temperature than 425 °C. According to API-
579, when a low alloy steel exposes to a temperature higher than 425 °C, level 1 assessment requirements doesn't meet, and then the
level 2 assessment should be considered.

6.2. Level 2 assessment procedure

Pressurized components that do not pass the level 1 assessment may be evaluated for continued service using the level 2 as-
sessment. The level 2 assessment procedure determines the structural integrity of a fire-damaged component by evaluating the
material strength. The first step in the assessment is to conduct dimensional checks on the pressure component. Dimensional checks
on V-201 reactor indicated no vertical deviation and bulging. In addition, to understand any microstructure degradations occurred
during damage and the related effect on mechanical properties, hardness testing is a helpful aid in assessing the loss of tensile
strength in carbon and low alloy steels, and loss or extent changes in other mechanical properties such as toughness and ductility.
Hardness can be measured in areas of the reactor known to be damaged during fire. Table 2 shows the results of hardness testing in
the damaged zones, and also the converted hardness values to estimated ultimate tensile strength based on API-579.
Allowable stress for the fire-damaged reactor (Safd) based on the estimated ultimate tensile stress was determined by Eq. (1). In
this equation, Cism is the in-service margin. The in-service margin may be taken equal to the design margin used to estimate the
ultimate tensile strength in the original construction code. However, this value was not known here and a value of Cism = 4.0 was
utilized regarding the recommended value of API-579.

⎡ ⎧ S ht S ⎫ ⎤
Safd = min ⎢ ⎜⎛ uts ⎟⎞. ⎛ aT ⎞ , {SaT} ⎥
⎜ ⎟

⎨ ⎝ C ism ⎠ ⎝ SaA ⎠ ⎬ (1)

⎣⎩ ⎭ ⎦
where Sutsht is the estimated ultimate tensile strength (ksi) based on the results from a hardness test. The hardness value of the
damaged side of the reactor was obtained of 100 HB, corresponds to a tensile strength of 45.8 ksi (315.8 MPa). SaT and SaA, re-
spectively, are allowable stress (ksi) at the specified design temperature and at the ambient temperature, both of which were ex-
tracted from ASME section II part D.
Maximum allowable working pressure (MAWP) was calculated using allowable stress for the fire-damaged reactor (Safd), as given
in Eq. (1). MAWP is the minimum value of maximum allowable working pressure in circumferential (MAWPC) and longitudinal
(MAWPL) direction of the reactor. Since P > 0.385SafdE, MAWPC and MAWPL were calculated using Eq. (2) and Eq. (3), respec-
tively.
Safd Et c
MAWPC =
R c + 0.6t c (2)

2Safd E (t c − t sl )
MAWPL =
R c − 0.4(t c − t sl ) (3)

where E is the weld joint efficiency (according to original construction code, all welds were fully radiographed, and therefore E was
taken to be equal to 1), tc is real reactor thickness calculated through tc = t-LOSS-FCA in which LOSS is metal loss in the shell prior to
the assessment equal to the nominal thickness (or furnished thickness if available) minus the measured minimum thickness at the
time of the inspection, FCA is future corrosion allowance (equal to 1/16 in., 1.59 mm), Rc is inside radius corrected for LOSS and FCA
and tsl is supplemental thickness for mechanical loads other than pressure that results in longitudinal stress.

8
R. Bakhtiari et al. Engineering Failure Analysis xxx (xxxx) xxx–xxx

MAWPC and MAWPL were calculated of 321 psi (2.21 MPa) and 661 psi (4.56 MPa) according to Eq. (2) and Eq. (3), respectively.
Regarding Eq. (4), MAWP was calculated of 321 psi (2.21 MPa), lower than that of the original reactor design condition (460 psi,
3.17 MPa). Therefore, based on level 2 assessment the reactor will be rejected to work under the design condition.
MAWP = min [MAWPC, MAWPL] (4)
The results of the FFS evaluation indicated that the equipment was not suitable for the current design condition and a new
maximum allowable working pressure, MAWP, was established based on Eq. (4). In general, if hardness testing indicated softening
below an acceptable limit, the required action would be to remove and prepare samples for tensile testing to verify compliance with
minimum tensile requirements. However, sample removing due to some limitations didn't perform.

7. Conclusions

For V-201 reactor, portable metallography and portable hardness test at different zones of inner and outer surfaces and also FFS
according to API-579 standard was conducted. The results showed that:
–The microstructure of two different zones of the damage region at the inner surface was ferritic/pearlitic with spheroidized
pearlite. The volume fraction of pearlite and grain size of ferrite were measured as 30% and 5ASTM, respectively.
–The microstructure of HAZ in the damage region at the inner surface was ferritic/pearlitic with spheroidized pearlite. The
volume fraction of pearlite and grain size of ferrite were measured as 10% and 8ASTM, respectively, which both were lower than the
other two zones at the inner surface.
–The microstructure of two zones at the outer surface (in damage region and away from that) was ferritic including carbide phases
which showed partial decarburizing. The grain size was measured as 8ASTM. Close microstructure features of these two zones
showed that decarburizing occurred during service of the reactor and ignition just made it to be accelerated. Also, cladding could
inhibit carburizing at the inner surface.
–The Brinell hardness test showed lower hardness values of different zones compared to the allowed range (133–171 HB) for the
reactor base metal according to ASTM A204/A204M – 12 standard. The lower hardness of base metal at the inner and outer surface
compared to the standard values could be due to grain growth and partial decarburizing, respectively.
–Observation of glass wool melting at the outer surface showed that the temperature increased higher than 425 °C during the
ignition. The reactor was rejected at Level I of FFS.
–The hardness of damage region of the reactor was measured of 100 HB which was converted to ultimate tensile strength as
45.8 ksi (315.8 MPa). Accordingly, the allowed pressure was obtained of 321 psi (2.21 MPa), lower than the design pressure (460 psi,
3.17 MPa). The reactor was rejected at level II of FFS.
–Level III of FFS needed destructive tensile test to obtain accurate ultimate tensile stress and to determine the allowed pressure.

Acknowledgment

The author would like to acknowledge Kermanshah Oil Refinery Company (KORC) (4pt) for financial supports of this research.

References

[1] L. Garverick, Corrosion in the Petrochemical Industry, ASM International, 1994.

[2] N. Roy, et al., Damage assessment of service exposed reactor and distillation column materials of a petrochemical industry, Eng. Fail. Anal. 28 (2013) 289–301.
[3] C. Matthews, Handbook of Mechanical In-Service Inspection: Pressure Systems and Mechanical Plant, John Wiley & Sons, 2004.
[4] X.J. Zhou, et al., Crack-Like Flaws Pressure Vessel Fitness-For-Service Assessments and Software Programming Based on API RP579, Procedia Engineering, 130
(2015), pp. 1359–1370.
[5] P. Tantichattanont, et al., Fitness-for-service assessment of spherical pressure vessels with hot spots, Int. J. Press. Vessel. Pip. 84 (12) (2007) 762–772.
[6] E. Shekari, et al., A predictive approach to fitness-for-service assessment of pitting corrosion, Int. J. Press. Vessel. Pip. 137 (2016) 13–21.
[7] U. Zerbst, et al., Fitness-for-Service Fracture Assessment of Structures Containing Cracks: A Workbook based on the European SINTAP/FITNET Procedure,
Elsevier Science, 2013.
[8] J.L. Hau, Assessment of fire damage to pressure vessels in a refinery unit, Corrosion 49 (5) (1993) 420–437.
[9] S. Kahrobaee, et al., Magnetic NDT technology for characterization of decarburizing depth, Surf. Coat. Technol. 205 (16) (2011) 4083–4088.
[10] J.L. Hau, Understanding the microstructure of overheated carbon steel, Mater. Perform. 43 (11) (2004) 46.
[11] I. Toda-Caraballo, et al., Effect of residual stress on recrystallization behavior of mechanically alloyed steels, Scr. Mater. 62 (1) (2010) 41–44.