Received: 2 June 2020 Revised: 30 September 2020 Accepted: 30 September 2020

DOI: 10.1002/prs.12213

ORIGINAL ARTICLE

Corrosion prevention of crude and vacuum distillation column

overheads in a petroleum refinery: A field monitoring study

Chidambaram Subramanian

NDT & Metallurgy Group, Central Mechanical

Engineering Research Institute, CSIR, Abstract
Durgapur, West Bengal, India The aim of the present work is to monitor dew point corrosion of crude and vacuum
Correspondence distillation column overhead section in a petroleum refinery. The material of con-
Chidambaram Subramanian, NDT & Metallurgy struction of overhead line pipe was carbon steel. The state of art of crude unit pro-
Group, Central Mechanical Engineering
Research Institute, CSIR, M.G. Avenue, cess was introduced. The corrosion and associated hydrolyzing mechanisms were
Durgapur 713209, West Bengal, India. briefly described. The crude processing in a petroleum refinery was extremely
Email: [email protected]
dynamic and thus field monitoring attempts were made, which provides some useful
information on structural integrity of column downstream equipments. The
demulsifier, neutralizer, and corrosion inhibitor were mixed in crude unit at specific
locations for corrosion prevention. Field monitoring includes evaluation of pH and
chloride content of desalter extracted water and overhead condensed water, Base
Sediment & Water (BS&W) and chloride content of as-received & desalted crude,
and iron content of overhead condensed water. Furthermore, corrosion coupon was
installed in overhead section to obtain the actual corrosion rate. All the field study
was conducted for a specified period. The acidity, chloride concentration, and iron
loss of condensed water confirmed that acidic corrosion was under control. The cor-
rosion rate of the installed coupons confirmed that corrosion rate was within the
specified limit. The results revealed in this paper intended for refinery operators and
corrosion engineers involved in understanding of actual process parameters in pre-
vention of overhead acidic corrosion.

KEYWORDS

acidic HCl, aqueous corrosion, crude oil refining, naphthenic acid corrosion, refinery overhead
corrosion

1 | I N T RO DU CT I O N A N D B A CK GR O U N D CO2) collectively referred as naphthenic acids, and measured in Total

INFORMATION Acidic Number (TAN, mg KOH/g). The main types of corrosion occurs
in crude unit of petroleum refinery are naphthenic acid corrosion
Corrosion always occurs in petroleum refining process especially in (NAC), sulfidation, hydrochloric acid (HCl) corrosion, ammonium chlo-
crude unit which process crude oil that constitutes different boiling ride (NH4Cl) corrosion, ammonium bi-sulfide (NH4HS) corrosion, poly-
point and molecular weight of long chain hydrocarbon. Major impuri- thonic acid stress corrosion cracking (PASCC) and oxidation corrosion
ties in crude oil are chloride salts, sulfur compounds, and carbon-di- (rust formation).2,4-6 The size and structure of naphthenic acid decides
oxide gases.1-3 The chloride salts could be in form of 75% NaCl, 15% corrosion severity between 180
C and 400
C, which could be operat-
MgCl, and 10% CaCl approximately. Further crude contain various ing temperature of heat exchanger, heater tube outlets, transfer lines,
acidic components including organic acids (R[CH2]nCOOH, R is distillation column especially at flash zones and reflux lines of crude
cyclopentane or cyclohexane, n > 12) and mineral acids (H2S, HCN, & processing unit. The presence of elemental sulfur, mercaptans (RSH),

Proc Safety Prog. 2020;e12213. wileyonlinelibrary.com/journal/prs © 2020 American Institute of Chemical Engineers 1 of 10
https://doi.org/10.1002/prs.12213
2 of 10 SUBRAMANIAN

and sulfides including polysulfides (RSnR) in crude convert to H2S to increase in molybdenum concentration that will increase passive
called sulfidation, and its corrosion severity depends on partial pres- layer stability.6
sure of H2S at high temperature. Steam injected into crude distillation The schematic view of typical crude distillation process of a
column for improved fractionation and H2S in stream dissolves in petroleum refinery is shown in Figure 1. The water in crude feedstock
steam condensate can cause wet H2S SCC in overhead exchanger was removed in crude tank by natural gravity and if any improper
tubes during condensation. The sulfide reacts with oxygen and mois- water settlement within the tank will result in bottom plate corrosion.
ture (oxidation) to form sulfurous acid especially during shutdown, Crude feedstock was pumped from tank to preheat exchangers for
resulting in polythonic stress corrosion cracking (PASCC). The distilla- heating up to 122
C that is slightly closer to the operating tempera-
tion column internals constructed by stainless steel are often exposed ture of desalter. Demulsifier was injected into desalter for separating
to sulfidation, so the treatment of column internals with 2% soda ash water from crude oil. On heating crude above 120
C in preheat
prior to shutdown will avoid PASCC and intergranular stress corrosion exchanger leads to HCl formation from magnesium and calcium chlo-
cracking.4 High nitrogen in crude reacts with hydrogen to form ammo- rides. The corrosion of equipments such as preheat exchanger, fur-
nia, which in turn reacts with Cl and H2S to form NH4Cl & NH4HS nace tube, distillation column internal and overhead pipeline can also
causing under deposit corrosion. NAC combined with sulfidic corro- lead to fouling and choking in overhead condenser tubes. Sour water



sion can occur at higher temperature 180 C to 400 C. NAC combined of 6% to 7% volume was mixed with crude feedstock using mixing
with phenol can corrode at both lower (65
C) and higher tempera- valve located at desalter upstream. The removal of dissolved salts pre-
tures. Although NAC occurs in combination with sulfidation, nature of sent in form of tiny water droplets are referred as water-in oil emul-
corrosion like pitting & impingement, and its severity due to high sion and by application of electric field, salt water droplets are
velocity can differentiate NAC from sulfidation in crude distillation coalesced and migrated to aqueous phase. The electric field speeds up
unit. Typical NAC exhibits isolated deep pits in partial film area and the emulsion break to separate oil and water phases. Further desalted
impingement attack in essentially film free area.7 SS316 sometimes crude was heated up to 288
C in product heat exchanger by recov-
resist NAC but Type 317 alloy suitable for resistance against NAC due ered process heat. Subsequently feedstock was heated up to 382
C in

F I G U R E 1 Schematic view of petroleum refinery crude unit and its process monitoring locations [Color figure can be viewed at
wileyonlinelibrary.com]
SUBRAMANIAN 3 of 10

fired heater to obtain partially vaporized crude and transferred to col- vapor velocity to prevent thermal cracking. The vacuum column top
umn flash zone. Organic compounds can be separated through distilla- and bottom design pressure were 3.5 kg/cm2 and full vacuum, respec-
tion column with different range of boiling point. The column top and tively. The vacuum was obtained by using number of ejectors. The
bottom working temperature were 116
C and 374
C, respectively. vacuum column top and bottom design temperatures were 325
C and
The column top and bottom working pressure were 2 kg/cm and 2
425
C, respectively; however, vacuum bottom was operated at 350
C
2.5 kg/cm2, respectively. The atmospheric distillation column contains to avoid cracking during hold up. The vaporized portion rises in col-
50 trays, while 6 trays were provided below the flash zone where umn through wash zone pass and demister bed for asphaltene
reduced crude was steam stripped for diesel and lighter constituent's removal. As similar to atmosphere distillation column, the side draw
removal. The stripping section vapor together with vaporized portion products were also yielded from vacuum distillation column. The
of crude was fractionated on the trays above flash zone to yield liquid crude and vacuum column during 100% plant operating conditions are
side draw products and overhead vapor stream. Gas oil, kerosene, and shown in Figure 2A and inner view of crude column during previous
heavy naphtha were side draw withdrawn and stripped in respective shutdown is shown in Figure 2B,C. The distillation column overhead
side strippers. section of crude unit is one of the vulnerable locations in a petroleum
On one hand the striping steam can move up from column bot- refinery, prone to corrosion which results in plant accidents and
tom until overhead, while on the other hand condensate fraction of outrages.8
cold reflux can move down, therefore stripping steam combined with The material of construction of both the distillation columns was
condensate fraction were responsible for column temperature gradi- SA515 Grade60 with 3 mm inner clad of SA240 Type 410S martens-
ent. The condensation of various petroleum products was possible itic stainless steel. The material of construction of crude and vacuum
due to temperature gradient. The heat was removed from column by overhead line pipe was SA672 Grade B60 CL12 and SA672 Grade
various top circulating refluxes for maximum heat recovery that bal- C70 CL12, respectively.9 The dimensions of overhead crude and vac-
ances the column loading. Re-circulating refluxes were withdrawn at uum line pipes were φ712 × t10mm2 and φ1220 × t10 mm2, respec-
respective product draw off trays and routed for heat recovery tively. The corrosion allowances of former and later ones are 3 mm
through various pumps. The overhead vapors were partially con- and 1.5 mm, respectively. The crude and vacuum line pipes were fab-
densed in overhead air and trim condensers. Overhead ricated from SA515 Gr60 coarse grain killed steel and SA516 Gr70
section includes distillation column top dish end, overhead line pipe, fine grain killed steel, respectively.9 The chlorine concentration of
overhead condenser, and overhead vessel. overhead process stream was designed for 0.00056 mol/L and often


Hot reduced crude of 364 C from crude (atmospheric) column lower chloride concentration could be favorable to protect the steel
bottom was again mixed with slope recycle and heated/vaporized in against corrosion. The wall thickness measurements were conducted
vacuum fired heater and transferred into vacuum column flash zone. on entire overhead line pipe of the both units up to condenser and no
The vacuum column was operated at reduced pressure of comparable wall loss was observed. The caustic injected into desalter and

F I G U R E 2 A, Crude and vacuum

distillation column during 100% plant
loading conditions. B, Top tray inside
crude distillation column during
previous turnaround. C, Column
overhead dish end inner view during
previous turnaround [Color figure can
be viewed at wileyonlinelibrary.com]
4 of 10 SUBRAMANIAN

corrosion inhibitor dosed into overhead process stream are common The presence of sulfur tri-oxide in heated crude can lead to formation
practice in petroleum refineries for corrosion prevention in overhead of sulfuric acid and thus leads to corrosion attack. In order to increase
locations. Some corrosion inhibitor technology patents were reported crude oil recovery in oil wells, carbon-di-oxide will be injected in well
for prevention of overhead corrosion but in present study, inhibitor and upon crude heating in refinery, CO2 carry over in crude dissolves
composition was unknown due to its proprietary nature. Many inhibi- within water and lead to carbonic acid near pH = 4.5 at column oper-
tors were developed for various concentrations of hydrochloric acids ating temperature.
10-20
and inhibitor efficiencies were evaluated. Many refineries have
reported accidents in overhead locations of crude and vacuum distilla- O2 + H2 S ! H2 O + S ð1Þ
tion units, yet only very few actual plant studies related to monitoring
overhead corrosion are available.21-27 This paper reports one such MgCl2 + 2H2 O ! MgðOHÞ2 + 2HCl ð2Þ
actual corrosion monitoring on distillation column over head locations
in crude refining units with brief introduction of corrosion mecha- CaCl2 + 2H2 O ! CaðOHÞ2 + 2HCl ð3Þ

nisms at vulnerable locations.

NaCl + H2 O ! NaOH + HCl ð4Þ

2 | O V E R H E A D C O R R O S I O N M EC H A N I S M HCl ! H + + Cl − ð5Þ

Fe + 2HCl ! FeCl2 + H2 ð6Þ

Free oxygen and hydrogen sulfide H2S present in crude feedstock can
react to form elemental sulfur according to Equation (1). This sulfur
FeCl2 + H2 S ! FeS + 2HCl ð7Þ
can be obtained in fractionator column bottom as well as in overhead
line pipe due to oxygen entry in crude feedstock. Further, water-in oil
NH3 + HCl $ NH4 Cl ð8Þ
can react with various salts in crude containing sodium chloride, NaCl,
calcium chloride, CaCl2, and magnesium chloride; MgCl2 to form HCl.
NH4 Cl ! NH4+ + Cl − ð9Þ
The condensation of HCl acid at overhead operating temperature is
referred as “dew point” according to Equations (2) to (4). These hydro-
NH + + Cl − ! NH3 + H + Cl − ð10Þ
lysis reactions are accelerated in opportunity crude due to higher
amount of sodium, calcium, magnesium salts in addition to organic
NH3 + H2 S $ NH4 HS ð11Þ
acids and inorganic clays. During initial condensation of water in over-
head condenser, available hydrogen chloride can react with steel sur-
MgCl2 + 2NaOH ! MgðOHÞ2 + 2NaCl ð12Þ
face to form pitting corrosion. The corrosion product scale contains
iron chloride (FeCl2), that is, often present within pits according to
CaCl2 + 2NaOH ! CaðOHÞ2 + 2NaCl ð13Þ
Equation (6). These reactions sufficiently lowers the pH in overhead
section as stated in Equation (5) and if pH is lower than 5, it even dis-
HCl + 2NaOH ! H2 O + NaCl ð14Þ
solves the available corrosion products due to HCl corrosion, which
tends to be carried away in downstream.
FeCl3 + 3NaOH ! FeðOHÞ3 + 3NaCl ð15Þ
The H2S in crude can react with iron chloride to form iron sulfide
and lead to HCl formation again according to Equation (7) which
destructs the construction material. Ammonia gases present in crude Caustic NaOH can be injected in desalter to avoid overhead HCl
feedstock can react with HCl acid vapor to form acidic ammonium corrosion according to Equations (12) to (15). The stable sodium chlo-
chloride (NH4Cl) according to Equation (8). This acid salt is hygro- ride and remaining products obtained from above reactions can be
scopic and can absorb moisture present in vaporized stream above separated as brine water. Improper separation can lead to choking,
the water dew point to exhibit pH as low as 3.3. NH4Cl salt often fouling, and foaming in desalter downstream. Equations (1) to (15) are
sticks on the construction equipment metal surface. However, with important reactions in crude refining process.3
small amount of water and heat, these ammonium chlorides dissociate
again into ammonia and HCl acid according to Equations (9) and (10).
The formation of HCl acid according to Equation (10) can exhibit pH 3 | EX PE RI MENT AL M ET H ODS
of 5.0, and destructs the metal surfaces again in form of pitting corro-
sion due to presence of moisture (refer Equation (6)). Excess H2S in The pH of extracted water from crude oil in desalter and condensed
crude reacts with ammonia vapor to form ammonium bi-sulfide water from overhead stream of crude and vacuum section were mea-
(NH4HS) according to Equation (11), which results in under deposit sured using digital pH meter (DBK make, Model: 10PHM01) with
corrosion. The presence of oxygen or oxidizers like ferric chloride accuracy of ±0.01 pH.
increases the rate of corrosion attack. It is widely believed that control The boot water condensed from overhead stream was collected
of free oxygen in crude feedstock can directly control the corrosion. from overhead receiver vessel of atmospheric and vacuum section for
SUBRAMANIAN 5 of 10

determination of chloride in the condensed water. Initially three drops installed coupons were removed and cleaned by acetone in ultrasonic
of cadmium acetate was added to 50 mL of collected water sample. cleaner. Subsequently the coupons were dried and weighed to mea-


For homogeneous mixing, the sample was heated up to 70 C and sub- sure weight loss.
sequently cooled. Using graduated cylinder the sample was filtered in The chemical compositions of the overhead line pipe material of
filter paper. Further two to three drops of potassium chromate was crude and vacuum distillation section were measured by x-ray fluores-
added to 20 mL of water sample in a conical flask, where potassium cence, XRF (Oxford instrumentation make, Model: X-Met 7500).
chromate acts as an indicator. Finally, the water sample was titrated
with 0.02 N of silver nitrate until obtaining a brick reddish color from
yellow and burette reading was noted. The chloride content in the 4 | R E S U L T S A N D D I S C U S S I O NS
water sample that contain H2S was calculated according to
Equation (16). The chloride salt in incoming crude oil was approximately less than
10 ptb (per thousand barrels), which can be in form of magnesium
Chloride content in the sample ðppmÞ = chloride, calcium chloride and sodium chloride. However, chloride in
Burette reading × equivalent weight of chloride × normality × 1000 desalted crude salt appeared to be 1 ptb (refer Figure 3A) and even
Volume of sample in ml 1 ptb chloride concentration in desalted crude can react with con-
ð16Þ densed water in overhead section at dew point temperature to initi-
ate corrosion. The chloride in crude oil is a potential source of HCl in
To obtain useful information of desalter efficiency, crude oil and crude distillation unit (CDU) and vacuum distillation unit (VDU). The
Base Sediment & Water (BS&W) samples were collected from inlet pH of extracted water from raw crude feedstock measured as 7.5
and outlet section of desalter for examination. The crude sample was while pH of condensed water from both overhead streams were
vigorously shaken in tight packed separator flask container before slightly acidic measuring 6.5 as shown in Figure 3B. Other
analysis. Sample of 100 mL was taken in the flask and mixed with researchers have reported that pH of desalter brine water at the
100 mL xylene, 25 mL acetone and 100 mL hot de-mineralized water. range of 5 to 6.5 for sour crudes.10,28 The caustic injection into
Subsequently the mixed sample was thoroughly shaken and pressure desalter increased the pH of desalted crude. The demulsifier was
was then slowly released by knobble attached in the flask. Again sepa- mixed with incoming crude in desalter upstream to break oil and
rator flask was shaken for two to three minutes and firmly placed in a water emulsion. The water-in oil reacts with demulsifier to form mud
stand. The oil and water was clearly separated and the water was col- sediments which settles at the bottom of desalter. The Base Sedi-
lected in a beaker. Further two to three drops of potassium chromate ment and Water (BS&W) was examined at desalter inlet and outlet
was added to the water sample in the beaker and titrated with sections as salt content in desalted crude directly depends on BS&W
0.019 N of silver nitrate solution to obtain reddish color from yellow. in desalter outlet. BS&W in inlet location was measured to be higher
The burette reading was then measured and chloride salt content of than 1%vol but at outlet location measured to be less than 1%vol as
as-received and desalted crude oil was calculated according to shown in Figure 3C. BS&W appeared to be consistent with crude
Equation (16). feedstock chloride concentration. The chloride content in desalter
BS&W was measured using oil testing centrifuge (ELTEK make, brine water was measured in between 112 ppm to 128 ppm as
Model: OC2F). The representative sample of 50 mL was mixed shown in Figure 3D and maximum chloride in as-received crude was
with 50 mL toluene and transferred to centrifuge tube and oper- removed through brine water had attributed to higher chloride in
ated for 10 minutes at 800 rpm. BS&W reading was measured brine water.
directly from oil testing centrifuge with minimum 0.025% detection Neutralizing amine was injected in overhead stream of crude and
limit. vacuum distillation to increase condensed sour water pH for preven-
The chloride content from desalter brine water (extracted water) tion of corrosion. The chloride concentration in condensed water from
was also measured according to Equation (16). The iron content in both the overhead sections was measured to be less than 25 ppm as
condensed water was measured by DR1800 spectrometer. Initially shown in Figure 3E. Other researchers have reported that crude and
spectrometer was standardized by demineralized water to set zero vacuum overhead sections can have better corrosion resistance, if
ppm. Subsequently 1 mL hydrochloric acid and 1 mL hydroxyl ammo- chloride concentration of condensed sour water lies at a range of
nium hydrochloride were added to 50 mL water sample. Further 40 ppm to 50 ppm.10,19,27 It was widely believed that control of dew
10 mL of ammonium acetate and 5 mL of phenopthaline indicator was point corrosion beyond 100 ppm chloride concentration is extremely
mixed with 25 mL of aforesaid sample mixture and iron content was difficult. The chlorine concentration of condensed water in vacuum
measured from spectrometer. distillation overhead measured to be 2 ppm higher than atmospheric
The C1020 grade carbon steel corrosion coupons of 7.5 cm distillation overhead. However, overall chloride concentration was
(length) × 1.5 cm (width) × 0.25 cm (thick) were prepared and measured to be less than 25 ppm in both the overhead sections. Mea-
installed in overhead line pipe of crude distillation and vacuum distilla- sured chloride concentration cannot be sufficient enough for HCl dew
tion section under actual service conditions for 55 days. After that point formation at operating temperature. This study reports
6 of 10 SUBRAMANIAN

F I G U R E 3 A, Chloride salt content in crude oil. B, pH test results. C, Base Sediments & Water (BS&W) from desalter. D, Chloride content in
brine water. E, Chloride content from overhead section. F, Iron content from overhead section [Color figure can be viewed at
wileyonlinelibrary.com]

minimum chloride concentration in overhead condensed water and can further raise the dew point to 125
C which will result in over-
previous experience suggests 50 ppm of chloride can be threshold head HCl corrosion. Therefore, water separation and optimization of
limit of overhead operations. as-received crude is considered as most important factor in preven-
The concentration of iron in condensed sour water of crude and tion of aqueous corrosion.29
vacuum distillation overhead sections are shown in Figure 3F. The Acidic corrosion can be prevented if overhead operating tempera-
average iron loss in condensed sour water measured to be 0.5 ptb ture is above the dew point. Often condensed sour water generated
and it can be of dissolution of construction material from equipment from other units (coker, hydrotreater, hydrocracker, and catalytic
internal surfaces. Aforesaid parameters were monitored for specific reforming) was also injected together with overhead condensed sour
period (52 days) which shows similar trend for number of monitored water of crude unit into desalter. In addition to this, de-mineralized
samples under actual operating conditions of overhead section. The water was also used in re-circulation. NH3 & H2S concentration of
salt content in desalted crude primarily depends on brine water salt sour water should be monitored before desalter injection. Otherwise
content and BS&W content. The separator in desalter not only as stated earlier it can lead to complex corrosion involving NH4Cl &
degasses the initial crude feedstock but also removes the formed NH4HS under deposit corrosion. Controlling overhead corrosion is
water. In case of crude distillation overhead section, it can be seen extremely difficult due to formation of stable NH4Cl & NH4HS salts.
that operated temperature was above the dew point of water phase These compounds are difficult to be removed during preheating. In
and thus, water condensation was avoided. The settling time of as- present study both ammonia and sulfur were monitored and mea-
received crude feedstock in crude tank is directly proportional to sured to be very low in sour water from crude and other units. There-
initial water separation. Therefore, water content in crude shall be fore injection of various sources of sour water into desalter was
optimized before and after desalting process. Sometimes water con- optimized and controlled.
tent in desalted crude was higher which lead to an additional cost of Oil based organic corrosion inhibitor will be adsorbed in overhead
vaporizing the water to steam and steam to water by additional equipment metal surfaces by injecting continuous inhibitor in over-
heating and cooling, respectively. The extra steam can increase the head section. The point of injection of neutralizing amine and corro-
pressure as steam occupies more volume than hydrocarbon which sion inhibitor shall be ahead of overhead condenser header box, so
reduces the yield of light end products. Further extra steam increases that the corrosion is prevented at the first location of condensation.
the dew point temperature. For example overhead stream operated Further, corrosion coupons were installed in overhead location prior
at 120
C and dew point of water was 110
C then that extra steam to condenser and corrosion rate was calculated according to
SUBRAMANIAN 7 of 10

continuously monitored and its average value obtained from digital

control system (DCS) was significantly higher than the calculated safe
overhead temperature.
The presence of high amount of naphthenic acids such as
TAN > 0.5 mg KOH g−1 is referred as high acid crude (HAC) or oppor-
tunity crude. In actual process conditions, NAC often lacks in evidence
that is, accumulative corrosion products especially in transfer lines are
not available due to high velocity, which is major drawback associated
with NAC. Also oil soluble iron naphthenates reacts with H2S to form
sulfidic scale (sulfidic corrosion) on metal surface to have semi protec-
tion against NAC, especially in side cut piping. At high temperature
various oxidizing gases reacting with metal is known as oxidation and
it acts as a thermodynamic stable film protecting the metallic surface
for example chromium alloyed steel. But presence of sulfide destructs
the oxide scale to form porous non adhering film and results in corro-
F I G U R E 4 Corrosion coupon before and after installation from sion.32 In presence of moisture corrosion rate will be enhanced. High
crude distillation overhead section [Color figure can be viewed at TAN crude always promotes hydrolysis of salts resulting to HCl corro-
wileyonlinelibrary.com]
sion in CDU and VDU.
Opportunity crude contains large number of corrosive species,
which can increase the dew point temperature. In order to control
Equation (17).20 The coupon was visually inspected before installation overhead acidic corrosion refinery follows certain strategies. Before
and it was free from dirt and grease as shown in Figure 4. Initial and refining, blending of opportunity crude with lower or nonacid crude
final weight of coupon was 11.99 g and 11.9613 g, respectively, and oil called dilution is the foremost strategy followed in industries. After
weight loss of coupon can be measured as 28.7 mg after exposing it various stream analyses, correct blending of opportunity crude with
for 55 days. The density of material and area exposed was low acid crude will be done. Incompatible crude blending and iron sul-
7.86 g/cm3 and 22.618 cm2, respectively. On substituting in Equa- fide scales evolving from NAC can lead to fouling in various equip-
tion (17), corrosion rate can be calculated as 0.06 mpy (miles per year) ments such as desalter, exchanger, and furnace tube. The visbreaking
against 3 mpy of allowable corrosion rate limit. The corrosion rate is treatment removes high sulfur, nitrogen, and aromatic components in
measured to be within the specified limit. The coupon shows absence opportunity crudes.32 Poor solid wetting and high solid wetting can
of pitting corrosion and uniform general thinning as shown in lead to oil carryover in brine water and water carry over in crude
Figure 4. stream, respectively. Opportunity crudes contain many filterable
solids and high asphaltene constituents which stabilizes water-in oil
Corrosion rate ðmpyÞ = emulsion. Subsequently it leads to BS&W carry over in desalted crude
22:3 × Weight loss ðmgÞ causing fouling and foaming. Excessive foaming in desalter leads to
ð17Þ
Densityðgm=cm3 Þ × Area exposed ðcm2 Þ × Days exposed ðdaysÞ poor desalting and oil & water removal which impacts the product
specification. However, filterable solids in sweet crude can stabilize
The column top pressure can be controlled by back pressure of the desalter emulsions. HAC also contains non-water extractable chlo-
overhead reflux vessel. As stated earlier overhead operating pressure rides that can increase the overhead neutralizer dosing load making
2
is 196.13 kPa (2 kg/cm ) and overhead stream constitutes 6.67 mol% the overhead vulnerable to corrosion. On one hand increased wash
water, therefore the safe overhead operating temperature can be esti- water rate in desalter can dilute stabilizing molecule concentration at
mated by water saturation temperature at partial pressure of water in oil water interfaces to reduce droplet coalescence while on the other
overhead vapor. The partial pressure of water can be calculated as hand increased wash water rate can also form water droplets in closer
196.13 × 0.0667 = 13.08 kPa and from steam table the saturated and thus facilitating the coalescence of small droplet into larger ones.
steam temperature for 13.08 kPa is 54
C.30 Now the safe tempera- Also higher mixing energy (mixing valve) will result in formation of
ture at which overhead section can be operated is 54
C + 17
C = 71
C. small water droplets that can lead to poor salt removal along with high
An increase in concentration of both moisture and HCl in process BS&W in desalted crude. Similarly, insufficient mixing also leads to
31
stream will increases acid dew point temperature. Hence, lower poor salt removal in desalter. Opportunity crude with high density has
concentration of water and HCl in the overhead stream had avoided high asphaltene with low viscosity. For blended opportunity crude, a
HCl dew point at overhead operating temperature in investigated sys- slight increase in desalting temperature can make good oil water sepa-
tem. The average operated temperature from attached thermocouple ration (emulsion break) due to low viscosity, and further containment
at crude distillation overhead section adjacent to inlet condenser dissolves at oil and water phases due to improved temperature.
tubes was measured to be 119.71
C for past 1450 hours at 100% High naphthenic concentration in opportunity crude contains
plant operated conditions. Actual operating temperature was sodium naphthenates, which drives the stable emulsion if pH of water
8 of 10 SUBRAMANIAN

increases while calcium naphthenates leads to poor salt removal.33 after water treatment.35,36 Opportunity crude containing toxic com-
However, unstable asphaltene constituents at oil water interface can ponents pollutes the refinery waste water and H2S stripped in sour
increase water solubility in oil which leads to low oil and high BS&W water unit to produce elemental sulfur.37
carry over in desalted crude. HAC reported sludge and gum blocking
in the pipelines.32,33 Washing of sludge and mud in desalter bottom
at-least once in a day for HAC/opportunity crude is a recommended 5 | CONC LU SIONS
practice for greater working volume of desalter. HAC causes water
soluble soap formation leading to tight emulsion and poor quality of The organic chloride present in crude oil is responsible for crude unit
oil in water. Proper optimization is required between overhead corro- distillation column overhead corrosion and this study suggests the
sion and fouling. The cumulative effects of neutralizing amine and cor- importance of chloride management and water content in crude. Vari-
rosion inhibitor injection in overhead stream section resulted in film ous field monitoring techniques were conducted for specified period
formation on overhead metal surface, and thus preventing direct con- for evaluation of prevention of acidic corrosion in crude and vacuum
tact of metallic surface with overhead stream. Caustic injection in distillation unit overhead sections. The investigated field parameter
desalter in addition to controlling the pH of crude feedstock, it also results revealed lower chloride concentration in desalted crude, lower
converts magnesium chloride to sodium chloride. The advantage of base sediment & water (BS&W) content at desalter outlet, lower chlo-
sodium chloride is that, it cannot hydrolyze readily to hydrochloric ride and iron concentration of overhead condensed water. The pH of
acid. Various factors that increase chloride concentration in overhead extracted water from as-received crude and pH of condensed water
stream are improper mixing of injected sour water with crude, from overhead streams were optimum. The column overhead was
improper crude blending, presence of large number of filterable solids, operated significantly higher than dew point temperature. The results
poor brine water separation from crude and high BS&W at desalter concluded that no significant corrosion occurred and verified by the
outlet that leads to poor desalting. Improper tank settling and high installed corrosion coupons.
solid wetting are key factors that increase water concentration in
overhead stream. Initial HCl condensation must be avoided in over-
head sections by lower concentration of chloride and water; other- 6 | RECOMMENDATIONS
wise it will lead to pitting corrosion that liable for acidic corrosion in
overhead section due to chloride deposits. The chemical composition • Double desalter may be installed in crude unit to remove more
of crude and vacuum distillation overhead line pipe was measured than 95% of chloride constituents in as-received crude oil.
using XRF positive material identification (PMI) prior to coupon instal- • Water in as-received crude feedstock shall be optimally separated
lation (refer Table 1). in crude tank by tank settling mechanism. It may reduce an extra
Attractive combination among correct crude blending, corrosion steam formation on overhead section.
prediction at high risk location, catalytic decarboxylation using copper • The overhead sections should be operated 14
C to 17
C above
as catalyst in low temperature process zone, acidic neutralization dew point temperature to prevent water condensation.
through chemical injection such as organic amine, ammonia & film for- • Electrical resistance (ER) probe and permanent ultrasonic testing
ming inhibitors, control of overhead temperature, limited retrofitting & (UT) sensors shall be implemented at maximum corrosion monitor-
up-gradation of existing construction material with anticorrosive ing locations (CML) on overhead sections for determination of
material (Type 317 L stainless steel or higher Mo alloyed steel6 or actual corrosion rates. ER and UT technique are intrusive and non-
internal Al2O3 ceramic coating34) and inspection/monitoring of high intrusive, while both measures localized corrosion rates. Frequent
risk location with verification of control methods are the effective wall thickness measurements shall be conducted in overhead pip-
strategies to control overhead corrosion. Neutralizer also prevents ing systems.
corrosion due to condensation of weak acids. The leakage and acci- • The carbon steel shall be used with caution for overhead operating
dents in overhead locations can be prevented with combined effect of conditions in crude unit. Up-gradation of existing material of con-
metallurgy up-gradation, efficient desalting process, improvements in struction with 3 mm internal cladding to Alloy C276 (Hastelloy)
corrosion inhibitor and process design. The refinery waste water can shall be substituted in over head line pipe. Existing condenser tube
be treated to remove Cl & H2S, and the extracted water (refer metallurgy of SA179 shall be upgraded with admiralty brass or tita-
Figure 1) will be reused in downstream refinery units and utility plant nium tubes.

T A B L E 1 Chemical composition of
Location C Mn P S Si Fe
overhead line pipe
CDU overhead line pipe — 0.55 0.01 0.01 0.18 99.25
9
ASTM A 672 GRB60 CL12 0.24 0.98 <0.035 <0.035 0.13-0.45 Balance
VDU overhead line pipe — 0.58 0.01 0.01 0.17 99.23
9
ASTM A 672 GRC70 CL12 0.21 0.55-0.98 <0.035 <0.035 0.13–0.45 Balance
SUBRAMANIAN 9 of 10

ACKNOWLEDGMENTS 10. Nathan CC, Perugini JJ. Prevention of corrosion in refinery overhead
This work was performed during author previous employment in streams by use of neutralizing and film forming inhibitors. Mater Per-
form. 1974;13(9):29-33.
Numaligarh Refinery (A Group Company of Bharat Petroleum Corp.
11. Quraishi MA, Sardar R. Dithiazolidines—A new class of heterocyclic
Ltd), India and acknowledges help received from technical staff of inhibitors for prevention of mild steel corrosion in hydrochloric acid
Crude/Vacuum Distillation Unit. Author also acknowledges Director, solution. Corrosion. 2002;58(2):103-107. https://doi.org/10.5006/1.
Technical, NRL-BPCL for providing fund to conduct this study and 3277308.
12. Ita BI, Offiong OE. The study of the inhibitory properties of
acknowledges Director, CSIR-CMERI for constant encouragement to
benzoin, benzil, benzoin-(4-phenylthiosemicarbazone) and benzil-(4-
publish this work. The author gratefully acknowledges the anonymous phenylthiosemicarbazone) on the corrosion of mild steel in
reviewers for many valuable suggestions that helped substantially to hydrochloric acid. Mater Chem Phys. 2001;70(3):330-335. https://doi.
improve this article. org/10.1016/S0254-0584(00)00476-4.
13. Khaled KF, Babic-Samardžija K, Hackerman N. Piperidines as corro-
sion inhibitors for iron in hydrochloric acid. J Appl Electrochem.
AUTHOR CONTRIBUTIONS 2004;34(7):697-704. https://doi.org/10.1023/B:JACH.0000031160.
Chidambaram Subramanian: Conceptualization; data curation; formal 88906.03.
analysis; investigation; methodology; validation; visualization; writing- 14. Ait Chikh Z, Chebabe D, Dermaj A, et al. Electrochemical and analyti-
cal study of corrosion inhibition on carbon steel in HCl medium by
original draft; writing-review and editing.
1, 12-bis (1, 2, 4-triazolyl) dodecane. Corros Sci. 2005;47(2):447-459.
https://doi.org/10.1016/j.corsci.2004.05.028.
CONF LICT OF IN TE RE ST 15. Bentiss F, Lebrini M, Lagrenée M. Thermodynamic characterization of
The author declares no conflict of interest to disclose and have no metal dissolution and inhibitor adsorption processes in mild steel/2,
known competing financial interests or personal relationship that 5-bis (n-thienyl)-1, 3, 4-thiadiazoles/ hydrochloric acid system. Corros
Sci. 2005;47(12):2915-2931. https://doi.org/10.1016/j.corsci.2005.
would have appeared to influence the work reported in this work.
05.034.
16. Qiu LG, Xie AJ, Shen YH. The adsorption and corrosion inhibition of
DATA AVAI LAB ILITY S TATEMENT some cationic gemini surfactants on carbon steel surface in
Data can be made available on request hydrochloric acid. Corros Sci. 2005;47(1):273-278. https://doi.org/10.
1016/j.corsci.2004.05.009.
17. Ropital F. Current and future corrosion challenges for a reliable and
sustainable development of the chemical, refinery, and petrochemical
ORCID
industries. Mater Corros. 2009;60(7):495-500. https://doi.org/10.
Chidambaram Subramanian https://orcid.org/0000-0002-8188- 1002/maco.200805171.
8958 18. Finšgar M, Jackson J. Application of corrosion inhibitors for steels in
acidic media for the oil and gas industry: a review. Corros Sci. 2014;
86:17-41. https://doi.org/10.1016/j.corsci.2014.04.044.
RE FE R ENC E S
19. Kapusta S, Van Den Berg F, Daane R, Place MC. The impact of oil
1. Lahiri AK. Material Selection and Performance in Refining Industry. field chemicals on refinery corrosion problems. NACE Paper
Singapore: Springer; 2017:349-418. https://doi.org/10.1007/978- No. 03649; 2003.
981-10-4684-1_10 20. Papavinasam S. Evaluation and selection of corrosion inhibitors.
2. Papavinasam S. The Main Environmental Factors Influencing Corrosion. Uhlig's Corrosion Handbook. Hoboken, New Jersey: Wiley; 2011:
In Corrosion Control in the Oil and Gas Industry. Amsterdam, 1121-1127. https://doi.org/10.1002/9780470872864.ch84.
Netherlands: Elsevier; 2014:179-247. https://doi.org/10.1016/B978- 21. Wu W, Cheng G, Hu H, Zhou Q. Risk analysis of corrosion failures of
0-12-397022-0.00004-2 equipment in refining and petrochemical plants based on fuzzy set
3. Timmins PF. Failure Control in Process Operations. Fatigue and Frac- theory. Eng Fail Anal. 2013;32:23-34. https://doi.org/10.1016/j.
ture. Cleveland, OH: ASM International; 2018:468-482. https://doi. engfailanal.2013.03.003.
org/10.31399/asm.hb.v19.a0002387 22. Saxena RC, Jayaraman A, Chauhan RK, Goyal HB. Studies on corro-
4. Groysman A. Corrosion problems and solutions at oil refinery and pet- sion control of naphtha fractions in overhead condensing system
rochemical units. Top Saf Risk Reliabil Quality. 2017;32:37-99. https:// using laboratory distillation device. Fuel Process Technol. 2010;91(10):
doi.org/10.1007/978-3-319-45256-2. 1235-1240. https://doi.org/10.1016/j.fuproc.2010.04.003.
5. Huang BS, Yin WF, Sang DH, Jiang ZY. Synergy effect of naphthenic 23. Schempp P, Preuß K, Tröger M. About the correlation between crude
acid corrosion and sulfur corrosion in crude oil distillation unit. Appl oil corrosiveness and results from corrosion monitoring in an oil
Surf Sci. 2012;259:664-670. https://doi.org/10.1016/j.apsusc.2012. refinery. Corrosion. 2016;72(6):843-855. https://doi.org/10.5006/
07.094. 1940.
6. Alvisi PP, Lins VFC. An overview of naphthenic acid corrosion in a 24. Valenzuela DP, Dewan AK. Refinery crude column overhead corrosion
vacuum distillation plant. Eng Fail Anal. 2011;18(5):1403-1406. control, amine neutralizer electrolyte thermodynamics, thermo chemi-
https://doi.org/10.1016/j.engfailanal.2011.03.019. cal properties and phase equilibria. Fluid Phase Equilib. 1999;158-160:
7. Jenbali Jahromi SA, Janghorban A. Assessment of corrosion in 829-834. https://doi.org/10.1016/S0378-3812(99)00067-9.
low carbon steel tubes of Shiraz refinery air coolers. Eng Fail Anal. 25. Kim J, Tak K, Moon I. Optimization of procurement and production
2005;12(4):569-577. https://doi.org/10.1016/j.engfailanal.2004. planning model in refinery processes considering corrosion effect. Ind
09.006. Eng Chem Res. 2012;51(30):10191-10200. https://doi.org/10.1021/
8. American Petroleum Institute. Risk-Based Inspection, API RP 580. 3rd ie300270s.
ed. Washington, DC: American Petroleum Institute; 2016:1-96. 26. Tak K, Kim J. Corrosion effect on inspection and replacement plan-
9. ASME. Boiler and Pressure Vessel Code Rules for Construction of Pres- ning for a refinery plant. Comput Chem Eng. 2018;117:97-104.
sure Vessel Section II Part A. New York: ASME; 2015:1201-1208. https://doi.org/10.1016/j.compchemeng.2018.05.027.
10 of 10 SUBRAMANIAN

27. Hassanudin SN, Aziz IA, Jaafar J, Qaiyum S, Zubir WMAM. Predictive 34. Fadhil, A. A., Khadom, A. A., Fu, C., Liu, H., Mahood, H. B.,
analytic dashboard for desalter and crude distillation unit. 2017 IEEE Mahmoud, A. K., … Karim, A. M. em A. (2020). Ceramics coating mate-
Conference on Big Data and Analytics. Vol 2017. Piscataway, rials for corrosion control of crude oil distillation column: experimen-
New Jersey: IEEE; 2018:55-60. https://doi.org/10.1109/ICBDAA. tal and theoretical studies. Corros Sci. 162:108220. https://doi.org/
2017.8284107. 10.1016/j.corsci.2019.108220
28. Chambers B, Yap KM, Srinivasan S, Yunovich M. Corrosion in crude 35. Subramanian C. Localized pitting corrosion of API 5L grade A pipe
distillation unit overhead operations: A comprehensive review. NACE used in industrial fire water piping applications. Eng Fail Anal. 2018;
Paper No: 11360; 2011. 92:405-417. https://doi.org/10.1016/j.engfailanal.2018.06.008.
29. Iannuzzi M, Barnoush A, Johnsen R. Materials and corrosion trends in 36. Subramanian C. Deterioration of economizer tube in utility boiler. Eng
offshore and subsea oil and gas production. NPJ Mater Degrad. 2017; Fail Anal. 2020;115:104643. https://doi.org/10.1016/j.engfailanal.
1:2. https://doi.org/10.1038/s41529-017-0003-4. 2020.104643.
30. Narayanan KV. Chemical Engineering Thermodynamics. New Delhi, 37. Chidambaram S. Thermal damage of sulfur processed chamber
India: PHI Learning Private Ltd; 2013:483-484. under Claus operating reaction conditions—a case study. Case Stud Constr
31. Huijbregts WMM, Leferink RGI. Latest advances in the understanding Mater. 2018;8:517-529. https://doi.org/10.1016/j.cscm.2018.04.005.
of acid dewpoint corrosion: corrosion and stress corrosion cracking in
combustion gas condensates. Anti-Corros Methods Mater. 2004;51(3):
173-188. https://doi.org/10.1108/00035590410533129.
32. Speight JG. Mechanism of Acid Corrosion. High Acid Crudes. How to cite this article: Subramanian C. Corrosion prevention
Waltham, MA: Gulf Professional Publishing; 2014:31-55. https://doi. of crude and vacuum distillation column overheads in a
org/10.1016/b978-0-12-800630-6.00002-2 petroleum refinery: A field monitoring study. Proc Safety Prog.
33. Speight JG. Corrosion in Refinery Units. Oil and Gas Corrosion Preven-
2020;e12213. https://doi.org/10.1002/prs.12213
tion. Waltham, MA: Gulf Professional Publishing; 2014:39-66.
https://doi.org/10.1016/c2013-0-18548-0