Managing the Threat of Naphthenic Acid Corrosion in Refining Assets

Introduction and Context

??????????? The highly competitive environment of the refining industry requires high availability and reliability of the refining hardware in the sense to maximize the operational lifecycle of the process units avoiding unnecessary shutdowns and production losses. One of the great threats to the availability and integrity of equipment in the refining industry is the corrosion phenomena that can lead to a reduction in the operation lifecycle of process equipment and, in extreme cases, serious accidents.

??????????? Among the corrosion mechanisms find the crude oil refining industry, the naphthenic corrosion. This corrosion mechanism occurs in hot sections of the process units like crude oil distillation and delayed coking units, naphthenic corrosion leads to quickly material loss, representing a dangerous threat to the integrity of these process units.

Crude Oil Distillation Unit – General Overview

The crude oil distillation unit defines the processing capacity of the refinery and, normally the others process units are sized on the basis of his yields. ?Figure 1 shows a basic process flow diagram for a typical atmospheric crude distillation unit.

Figure 1 – Process Flow Diagram for a Typical Atmospheric Crude Oil Distillation Unit

The crude oil is pumped from the storage tanks and preheated by hot products that leaving the unit in heat exchangers battery, then the crude oil stream receives an injection of water aim to assist the desalting process, this process is necessary to remove the salts dissolved in the petroleum to avoid severe corrosion problems in the process equipment. The desalting process involves the application of an electrical field to the mixture crude oil-water aim to raise the water droplets dispersed in the oil phase and accelerate the decanting, as the salts solubility is higher in the aqueous phase a major part of the salts is removed in the aqueous phase effluent from the desalter, called brine.? Normally the petroleum desalting process is carried out at temperatures among 120 and 160 oC, higher temperatures raise the conductivity of oil phase and prejudice the phase separation, and this can lead to drag oil to the brine and result in process inefficiency.

??????????? The desalting process involves the mixture of crude oil with water aiming the dissolution of the salts considering the higher solubility of these compounds in the aqueous phase. Figure 2 depicts a typical desalting process with two separation stages, the salt content in de desalted crude is normally controlled bellow of 5,0 ppm (as NaCl).?

Figure 2 – Crude Oil Desalting Process with Two Separation Stage

In the desalter exit, the desalted oil is heated again by hot products or pump around and fed into a flash drum, in this equipment the lighter fractions are separated and sent directly to the atmospheric tower, the main role of this vessel is reducing the thermal duty needed in the furnace. ?Following, the stream from the bottom of the flash vessel is heated in the fired heater to temperatures close to 350 to 400 oC (depending on the crude oil to be processed) and is fed the atmospheric tower where the crude oil is fractionated according to the distillation range, like example presented in Table 1.

At the exit of the atmospheric tower, the products are rectified with steam aim to remove the lighter components.?

The gaseous fraction is normally directed to the LPG (C3-C4) pool of the refinery and the fuel gas system (C1-C2) where will feed the furnaces and boilers.? The light naphtha is normally commercialized as petrochemical intermediate or is directed to the gasoline pool of the refining complex, the heavy naphtha can be sent to the gasoline pool and in some cases, this stream can be added to the diesel pool since not compromise the specification requirements of this product (Cetane number, density, and flash point). ?Kerosene is normally commercialized as jet-fuel while the atmospheric residue is sent to the vacuum distillation tower, in some refining schemes it’s possible sent this stream directly to the residue fluid catalytic process unit (RFCC), in this case, the contaminants content (mainly metals) of the residue needs to be very low to protect the catalyst of the cracking unit. ?

Nowadays, face to the necessity to reduce the environmental impact of the fossil fuels associated with the restrictive legislations, difficultly the straight run products can be commercialized directly. The streams are normally directed to the hydrotreating units aim to reduce the contaminants content (sulfur, nitrogen, etc.) before being marketed.

In distillation units with higher processing capacity, normally the flash drum upstream of the atmospheric tower is substituted by a pre-fractionation tower. In these cases, the main advantage is the possibility of reduction of the atmospheric tower dimensions that implies in cost reductions associated with the unit implementation and improve the hydraulic behavior in the distillation tower, consequently with better fractionation. This arrangement is shown in Figure 3.

Figure 3 – Typical arrangement to Atmospheric Distillation with Pre-Fractionation Tower.

Vacuum Distillation Section

The bottom stream of the atmospheric column (Atmospheric Residue) still contains recoverable products capable to be converted into high added value derivatives, however, under the process conditions of the atmospheric unit, the additional heating led to thermal cracking and coke deposition.

??????????? Aiming to minimize this effect, the atmospheric residue is pumped to the vacuum distillation column where the pressure reduction leads to a reduction in the boiling point of the heavy fractions allowing the recovery while minimizing the thermal cracking process.

??????????? The vacuum generated in the column can be humid, semi-humid and dry. Humid vacuum occurs when is applied steam injection in the fired heater and in the column aiming to reduce the partial pressure of the hydrocarbons improving the recovery while in the semi-humid vacuum the steam is injected only in the fired heater minimizing the residence time reducing the coke deposition. The dry vacuum does not involve the steam injection, in this case, is possible to achieve pressures between 20 to 8 mmHg while in the humid vacuum the column operates under pressures varying between 40 to 80 mmHg, however, it’s possible to achieve comparable yields through the injection of stripping steam. Figure 4 presents a process arrangement for a typical vacuum generation system in a vacuum crude oil distillation unit.

Figure 4 – Process Arrangement for a Typical Vacuum Generation System for a Vacuum Crude Oil Distillation

??????????? As shown in Figure 5, the traditional arrangement of vacuum units presents two side drawn, heavy and light gasoil.? These streams are normally directed to conversion units like hydrocracking or fluid catalytic cracking (FCC), according to the adopted refining scheme. The fractionating quality achieved in the crude oil vacuum distillation column has a direct impact upon the reliability and conversion units operation lifecycle, once which in this step is controlled the metals content and the residual carbon (CCR) concentration in the feedstock to these processes, high values of these parameters lead to a quick catalyst deactivation raising operational costs and reducing profitability. ?

Figure 5 – Schematic Process Flow Diagram for Vacuum Distillation

Some refiners include additional side withdraws in the vacuum distillation column. When the objective is to maximize the diesel production, it’s possible to add a withdraw of a stream lighter than light vacuum gasoil that can be directly added to the diesel pool or after hydrotreating, according to the sulfur content in the processed crude oil. When the crude oil presents high metals content, it’s possible to include a withdraw of fraction heavier than the heavy gasoil called residual gasoil or slop cut, this additional cut concentrates the metals in this stream and reduce the residual carbon in the heavy gasoil, minimizing the deactivation process of the conversion processes catalysts as aforementioned. The vacuum residue is normally directed to produce asphalt and fuel oils, however, in most modern refineries this stream is sent to bottom barrel units as delayed coking and solvent deasphalting to produce higher-value products. ?

According to the refining scheme, the installation of vacuum distillation units can be dispensed. Refiners that rely on residue fluid catalytic cracking units (RFCC) can sent the atmospheric residue directly to feed stream of these units, however, it’s necessary to control the contaminants content (metals, sulfur, nitrogen, etc.) and residual carbon (CCR) aiming to protect the catalyst, this fact restricts the crude oil slate that can be processed, reducing the refiner operational flexibility. On the other hand, in refineries that process extra heavy crudes, normally the crude oil distillation unit is restricted to the vacuum unit once the yields of the atmospheric column would be very low and the coking risk very high.??

In refineries optimized to produce lubricants, the distillation process is modified face to the paraffinic characteristics of the crude oil processed, mainly the vacuum distillation step. The necessity to separate the lubricants fractions requires higher fractionation quality in the column and some configurations rely on two columns, as presented in Figure 6.

Figure 6 – Vacuum Distillation Process to Produce Lubricants

??????????? The distillation unit design is strongly dependent by the characteristics of crude oil that will be processed by the refinery, for extra-heavy oils normally the crude is fed directly to the vacuum column. The design is generally defined based on a limited crude oil range that can be processed in the hardware (Contaminant content, API grade, etc.).???????????

Delayed Coking Technologies – General Overview

Delayed coking employs the thermal cracking concept under controlled conditions to produce light and middle streams (LPG, naphtha and gas oils) from residual streams which would normally be used as diluents in fuel oils production. ?

The typical feed stream for delayed coking units is the residue from vacuum distillation process that contains the heavier fractions of processed crude oil, however, streams like decanted oil from FCC unit and asphaltic residue produced in solvent deasphalting can compose the feed stream to the delayed coking unit, depending upon the refining scheme adopted by the refiner. Another possibility is sent the residue from atmospheric distillation directly to the delayed coking unit, in this case, the unit design is quite modified demanding greater robustness of the fractionating and gas compression section.

Due to the thermal cracking characteristics (low availability of hydrogen during the reactions), the streams produced by delayed coking unit have a high concentration in olefinic compounds, which are chemically unstable. ?Furthermore, due to the processing of residual streams that have high contaminants content like nitrogen, sulfur, and metals, therefore the refiners that apply delayed coking units need high hydrotreating capacity to convert these streams into added value products and that meets the contaminants level according to the environmental regulation. Figure 7 presents the process flow scheme for a typical delayed coking unit. ?

Figure 7 – Typical Arrangement for Delayed Coking Unit (SILVA & CLARK, 2021)

The feed stream is fed into the bottom of the main fractionating tower where is mixed with the heavier fraction of the thermal cracking products and then sent to the fired heater where thermal cracking reactions are initiated, the reaction conditions are controlled so that the reactions are completed in the coke drums, the residence time in the fired heater must be the lowest possible to minimize the coke precipitation in the fired heater tubes.? A manner of minimizing the coke formation in the walls of tubes is the steam injection that raises the velocity and consequently reduces the residence time. ?

After the fired heater the feed stream is sent to the coke drum or reactor, where the thermal reactions are completed, and the coke is deposited. The thermal cracking products are removed from the top of the reactor and receive an injection of quench with a cold process stream (normally heavy or middle gas oil) and directed to the main fractionators where the products are separated. The coke deposited in the reactor is removed through a cut with water under high pressure (about 250 bars).

Delayed coking is a process that occurs in batch, in order to make a semi-continuous process are always employed pairs numbers of reactors and each two reactors is applied one fired heater when one reactor is under reaction the other is in decoking step and so on. The delayed coking process occurs in cycles that can vary from 14 to 24 hours.

The main operational variables of the delayed coking unit are recycling ratio which is the quantity of the total feed stream which corresponds to the heavier fraction of the reaction products that are mixed with the fresh feed, reactor temperature, normally considered in the top of the coke drum, pressure in the top of reactor and the time of the reactor cycle. ?

The recycle ratio vary normally between 5 to 10% (to units dedicated to producing fuels) and the refiners seeks to operate the unit with de lower recycle ratio possible in order to maximize the capacity of the plant in processing residual streams.? The reactor temperature is close to 430 oC and is linked with the fired heater temperature, throughout the thermal cracking reactions the temperature falls due to the endothermic characteristics of the reactions.

The pressure in the reactor can vary between 1 to 3,5 bars, in units optimized to producing fuels the variable is maintained at lower levels, on the other hand, when the unit is dedicated to producing high-quality coke, the unit is operated under higher pressures.

Reactor cycle time is linked to the function performed by the delayed coking in the refining scheme. Units dedicated to producing fuels operate at shorter cycles and units optimized to producing high-quality coke operate under longer cycles.

The coke produced normally is seen as a by-product of the delayed coking unit, however, in some cases, the delayed coking process is optimized to producing high-quality coke and the coke becomes to the principal product of the process.

Depending on the feedstock quality that will be processed, three types of the coke can be produced:

·?????? Shot coke – Poor quality coke produced from feedstock with high asphaltenes and contaminants (sulfur, nitrogen and metals) content, normally this type of coke is commercialized as fuel;

·?????? Sponge coke – In this case, the feedstock have a lower asphaltenes and contaminants content and the coke can be directed to raw material to anodes production process to the aluminum industry;?

·?????? Needle Coke – The production of this type of coke require the processing of feedstock with high aromatics content (decanted oil from FCC, for example) and these products are sent as raw material to producing anodes to the steel industry;?

As mentioned above, production of high-quality coke requires a quality control of the feed stream that will be processed, in the most of the cases the refiners choose to install delayed coking units focusing in the production of middle and light distillates. Therefore, the unit optimization to produce needle coke occurs only in specific cases. ?

??????????? The heavy gas oil stream is normally directed to the fluid catalytic cracking unit or can be utilized as fuel oil, in refining schemes that have deep hydrocracking units this stream can be used like feedstock to the unit. The sending of this stream to the fluid catalytic cracking unit needs be controlled to avoid the premature deactivation of catalyst, face of the high level of contaminants, mainly nitrogen and metals.?

??????????? Middle and light gas oils are normally sent to severe hydrotreating units to compose the diesel pool of the refinery. The heavy coker naphtha can be directed like feed stream to FCC units. When the flash point specification of diesel is not restricted this stream can be sent to the diesel pool, after deep hydrotreating process.

??????????? The lighter fraction of naphtha can be sent to the gasoline pool of the refinery after hydrotreatment or directed to FCC units, in this case this stream contributes to raise de LPG production in the FCC unit. In some cases, the light coker naphtha can be sent to catalytic reforming units aiming to produce high octane gasoline or petrochemical precursors (benzene, toluene and xylenes).

??????????? The overhead products from main fractionator are still in gaseous phase and are sent to the gas separation section. The fuel gas is sent to the refinery fuel gas ring, after treatment to remove H2S, where will be burned in fired heaters while the LPG is directed to treatment and further commercialization.

??????????? Delayed coking technology becomes especially attractive for refiners installed in countries with large heavy and extra-heavy crude oil reserves, like Brazil, Mexico and Venezuela. The use of delayed coking in the refining scheme can minimize the production of low added value products like fuel oils and guarantees higher flexibility to the refinery in a relation of processed crude oil, minimizing the necessity to acquire light oils.

??????????? On the other hand, the delayed coking technology obliges the refiners the necessity of high hydrotreatment capacity once the streams produced by the unity needs severe treating process before being sent to the commercialization, this fact can raise the operational and installations costs.

Naphthenic Corrosion – An Overview

??????????? The naphthenic corrosion occurs in processing units that process bottom barrel streams that tends to present high concentration of naphthenic acids and operates under high temperatures. Due to these characteristics, naphthenic corrosion phenomenon is observed in crude oil distillation units and residue upgrading units like delayed coking.

??????????? The characteristics of the processed crude oil slate are a determinant factor in the naphthenic corrosion. A very relevant characteristic of oils for refining hardware is naphthenic acidity. Naphthenic acidity is determined based on the amount of KOH required to neutralize 1.0 gram of crude oil, normally a mixture of crude oils is sought in the refinery load so that it does not exceed 0,5 mg KOH/g, above this reference, the bottom sections of the distillation units can undergo a severe corrosive process, leading to shorter periods of operational campaign and higher operating costs in addition to problems associated with integrity and safety. Naphthenic acidity is directly linked to the concentration of oxygenated compounds in the crude oil that tend to be concentrated in the heavier fractions, giving instability and odor to the intermediate currents.

??????????? The sulfur content in the crude oil is another key factor in the naphthenic corrosion phenomena. In crude oils with sulfur content higher than 2,0 %, a protective layer of iron sulfide (FeS) is formed in the metal surface that is insoluble, avoiding then the attack by naphthenic acids as presented in Figure 8. In this sense, refiners processing very low sulfur crude oils with high Total Acid Number (TAN) can face severe issues with naphthenic corrosion in their refining hardware.

Figure 8 – Naphthenic Corrosion Process

The equation 1, presents the chemical representation of naphthenic corrosion.

Fe + 2RCOOH → Fe(RCOO)2 + H2? (1)

??????????? According to the literature, above of 400 oC the iron naphthenate (corrosion product) is soluble in hydrocarbons and is attacked by hydrogen sulfide (H2S) leading to the regeneration of the naphthenic acid, as represented in equation 2.

Fe(RCOO)2 + H2S → FeS + 2RCOOH (2)

??????????? In crude oil distillation units, the bottom section of atmospheric column and the vacuum distillation column are the most common regions where is observed naphthenic corrosion while in delayed coking units the phenomenon is observed in bottom section of main fractionators column.

??????????? The carbon steel, series 300 and 400 stainless steel, and nickel alloys tend to suffer naphthenic corrosion.?

??????????? Among the actions to control the naphthenic corrosion in the refining hardware is the blending of crude oils aiming to keep the Total Acid Number (TAN) and sulfur content inside the adequate limits. Other alternatives are the injection of neutralizers or corrosion inhibitors in the processing streams and the selection of materials with higher resistance to naphthenic acid attack.

??????????? Refiners processing crudes with high acidity and low sulfur normally applies chromium and molybdenum alloys in the bottom sections of crude oil distillation columns and transfer pipes as well as in residue upgrading units. In extreme conditions, it’s possible to consider apply stainless steel 317 L that presents high resistance to naphthenic corrosion, this decision needs to consider the higher capital investment due to the high cost of this material.

??????????? As inspection and monitoring strategies it’s normally applied ultrasound and X-ray assay to identify localized corrosion and thickness loss as well as sacrifice metals coupons to determine the corrosion rates.

Conclusion

??????????? The availability of the refining hardware is a key parameter to ensure the economic sustainability of the refiners, especially those inserted in high competitive markets.

????????? Like described above the naphthenic corrosion can compromise the reliability and availability of the key units to the refining hardware, by this reason, the naphthenic corrosion issues can need to take into account in the crude oil selection aiming to minimize integrity risks and shorter operational campaigns. In this sense, adequate monitoring and control of corrosion process in these units is fundamental to ensure the competitiveness of the players in the downstream sector.

References

GARY, J. H.; HANDWERK, G. E. Petroleum Refining – Technology and Economics.4th ed. Marcel Dekker., 2001.

RAMANATHAN, L.V. Corrosion and his Control. 1st ed. Hemus Press, 1978.

ROBINSON, P.R.; HSU, C.S. Handbook of Petroleum Technology. 1st ed. Springer, 2017.

SILVA, M. W.; CLARK, J. – Flexible Upgrading of Heavy Feedstocks. PTQ Magazine, 2021.

Dr. Marcio Wagner da Silva is Process Engineer and Project Manager focusing on Crude Oil Refining Industry based in S?o José dos Campos, Brazil. Bachelor’s in Chemical Engineering from University of Maringa (UEM), Brazil and PhD. in Chemical Engineering from University of Campinas (UNICAMP), Brazil. Has extensive experience in research, design and construction to oil and gas industry including developing and coordinating projects to operational improvements and debottlenecking to bottom barrel units, moreover Dr. Marcio Wagner have MBA in Project Management from Federal University of Rio de Janeiro (UFRJ) and is certified in Business from Getulio Vargas Foundation (FGV).