W. Fazackerley, Emerson, London, England
The production of carbon-neutral and renewable
transportation fuels is growing rapidly, driven by the need to find feedstocks beyond
fossil resources. Environmental, social and governance (ESG) initiatives—along
with accelerated consumer demand for sustainable and carbon-negative fuel
products—are also driving demand. Renewable fuels for aviation and road
transportation are at the forefront of this trend, and low-carbon and renewable
fuel standards have been implemented by many nations worldwide to accelerate progress.
One area considered critical to profitable and
successful process adoption is plant integrity to deliver continuous production
with limited downtime. This requires corrosion management, which can be
difficult because most refiners have a limited operational track record or a
lack of long-term documentation of occurring corrosion mechanisms. These stores
of information serve as a knowledge basis for traditional integrity programs
for fossil fuel refineries, but they are not widely available for biofuel refining.
This article provides a holistic overview of
renewable fuel processing equipment and associated corrosion mechanisms, and it
shows how corrosion monitoring can be used to mitigate issues.
Renewable feedstocks. There are multiple feedstock options for renewable
fuels, with many varying by region and season. ESG pressures can influence
feedstock selections and may lead to variations throughout a plant’s service
lifecycle. Retrofits of existing processing units are often required to match unit
designs and materials to changing process conditions and corrosive
Fresh vegetable oils were a common feedstock
in first-generation biofuels, with refiners converting them into fatty acid
methyl esters. Fueled by incentive programs, this well-researched
transesterification process required relatively low capital expenditures,
leading to the conversion of many traditional crude oil refineries into biodiesel
plants throughout the early 2000s. Recently, concerns around fuel blending and
engine compatibility have led to the development of hydrotreated processes instead
These types of processes produce drop-in fuels
that are chemically nearly identical to fossil-based fuels, and their use is less
limited by restrictive blending limits. For example, hydrotreated vegetable
oils are straight-chain paraffinic hydrocarbons that are free of aromatics,
oxygen and sulfur. They also have high cetane numbers and do not exhibit any of
biodiesel’s concerning effects at end-use points, such as increased nitrogen
oxide emissions, poor cold-operation properties, deposit formation, storage
stability issues and more rapid aging of engine oil. However, societal questions
regarding the agricultural use of energy crops in competition with food crops
remain an issue.
These concerns have led to second-generation
biofuels, referred to as green or renewable diesel and sustainable aviation
fuel (SAF). Novel catalysts allow advanced, sustainable feedstock (e.g., tallow
oil from bacon production) to be hydroprocessed into drop-in fuels. Used and rendered
fat and grease, collected in traps as waste products, are another primary area
of investigation for sustainable feedstocks. Tax credits and avoided landfill
costs can create a lucrative revenue stream, boosting the profitability of a
renewable fuel production unit.
However, as with all waste products, there are
challenges of consistent feedstock quality and reliable supply chains. Polyethylene
packing impurities are typically found in animal fats, resulting in process
challenges with catalyst deactivation, fouling concerns with heat exchangers, and
flow constraints and catalyst bed pressure drops triggered by plugging. In
general, these types of waste oils require heat-traced feed piping due to their
lowered cloud point.
Compared to vegetable oils, animal fats (such
as tallow oil) tend to have larger amounts of free fatty acids (FFAs) (TABLE 1). While
organic sulfur components in renewables are decreased compared to petroleum
feedstocks, the increased influx of organic chlorides and nitrates creates
challenges with process equipment and piping integrity downstream of
In addition to these and other corrosion-accelerating
substances, the presence of water facilitates corrosion mechanisms by providing
an electrolytic pathway. Water solubilizes organic and inorganic acids, creating
carbon dioxide (CO2) via hydrodecarboxylation, which turns organic
chlorides into hydrochloric acid (HCl) and promotes microbiologically influenced
corrosion (MIC). Water forms in the hydroprocessing reactor section, and it is
a decomposition product of renewable feedstocks in tanks and pretreatment
sections. In addition, biofuel feedstocks are often hygroscopic and will
attract humidity from the environment into process streams.
Biofuel refining facilities around the world
are addressing these processing issues by proactively deploying permanently
installed and continuous wall-thickness monitors to track corrosion in critical
locations. Tighter monitoring enables cost-effective tracking of corrosion in
areas of concern, and enables refiners to pinpoint specific feedstocks or
process operations that result in accelerated corrosion rates. This information
facilitates the optimization of corrosion mitigation strategies online, along with
the validation of the effectiveness of these mitigation strategies, so that
timely and evidence-based integrity management decisions can be made most
Renewable fuel production
processes. With wide-ranging recent developments, various
novel processes have been researched, engineered and patented. As a general
overview, this article will define a generic process scheme to assess dominant
concerns of corrosion and corresponding risk to plant integrity. Drop-in fuel-grade
production processes require advanced phases of chemical conversion involving
purification, deoxygenation and novel catalysts.
FIG. 1 depicts
and identifies key units, defining specific corrosion mechanisms in each. Like
traditional refinery hydroprocessing units, base feedstocks and intermediates are
first pretreated, blended and stored in separate pretreatment units and tank farms.
However, hydrogen consumption with alternative feedstocks has significantly
Storage and tank farms may experience
corrosion from feedstock degradation products and as low-temperature
environments promote MIC. Pretreatment steps may include the separation of
solid particles, pH adjustments or a liquefaction step for solid feedstocks
into bio-oils. Bio-oils usually have an aqueous and lipid phase, and aqueous
phase processing and storage equipment may experience acidic corrosion from FFAs.
The pretreated feed is passed through a bio-oil
heater or heat exchanger to raise its temperature to 315°C–400°C (600°F –750°F),
and it is subsequently injected with hydrogen, which is typically a major shifting
point for increased corrosion and degradation mechanisms.
Depending on licensors, renewable feedstocks
and final fuel products, the degree of process cracking severity will differ,
and hydrogen may be combined with the renewable feedstock in a central mixing
point or dispersed over multiple reactor quenches. Nonetheless, when hydrogen
is added, corrosion mechanisms will change with the presence of competing
Process designs may incorporate one or
multiple reactors, depending on feedstock preparation, contaminations and
process product output requirements.
The first is commonly referred to as the hydrodeoxygenation
reactor. This stage is used for feedstock cleanup (capturing contaminants such
as metals, particles and scales) to mitigate catalyst deactivation and
clogging. Simultaneously, hydrocarbon hetero atoms (such as oxygen, sulfur and
nitrogen) are eliminated to create hydrocarbons suitable as drop-in fuels. Unsaturated
hydrocarbons with double bonds—commonly referred to as olefins—are converted into
For corrosion mechanism assessment, it is
pertinent to note the formation of ammonia from organic nitrogen components,
water from oxygen and hydrosulfuric acid from sulfur components in renewable
HCl is another corrosive hydrotreating
byproduct, and it is best controlled by feedstock monitoring and the
implementation of potential corrosion mitigation strategies. Additionally, the
presence of chlorides and amines will lead to salting issues.
The next unit is a hydro-isomerization or
dewaxing reactor, or a hydrocracker phase, depending on processed feedstocks
and desired end products. These units produce branched hydrocarbons with improved
cold flow properties for renewable diesel products—whereas, more severe hydrocracking
stages are required to produce SAF.
While these reactor stages may experience
decreased corrosion after the initial removal of contaminants, they may also experience
other severe degradation mechanisms due to increased temperatures and to system
and hydrogen partial pressures.
A series of high- and low-pressure separators
and strippers typically connects the two reactor stages, usually with acid-removal
stages like traditional amine units and sour water strippers. These systems are
challenged by strong acid loads that may fluctuate dramatically with feedstock
variations and inconsistent input qualities.
Major integrity concerns arise in the separation
section due to the presence of hydrogenation reaction side products mentioned
in the preceding paragraphs. The safe operation of this section depends on
monitoring systems to trigger appropriate corrosion mitigation measures.
Sour water byproducts are further treated in
wastewater treatment facilities that are often challenged by novel corrosion
severity. In traditional refining, material grades chosen for this type of service
experience higher corrosion due to very low pH water phases and MIC.
corrosion monitoring technologies. For the past 60 yr,
ultrasound has been used to measure metal wall thickness. This involves manually
placing a transducer directly onto the metal surface to generate ultrasound,
which then travels through the metal until it reflects off the inner metal
surface. By recording the reflected ultrasound signal and calculating the time
difference between the sending and reflected signals, the wall thickness can be
However, completing a full set of
measurements for a medium-sized refinery with > 80,000 corrosion measurement points can
be extremely time consuming and labor intensive. As a result, the wall
thickness at low-level to medium-level risk points may only be measured every 2
yr–3 yr. This makes it challenging to obtain data frequently enough to
determine corrosion rates with confidence, or to link periods of high wall loss
to specific feedstocks or process operations, with the latter requiring daily measurements.
being relatively simple, manual ultrasound methods have two other main disadvantages.
Repeatability and reproducibility errors are common, as it is highly unlikely
that consecutive measurements will be taken in precisely the same location by
the same technician. In addition, the equipment used and the skill level of the
technician can vary between measurements, thus introducing high variability. FIG. 2 shows manual
measurements at a single location over an 11-yr period. Different conclusions
regarding wall thickness and corrosion rates can be drawn over time. From such
data, it could be inferred that the accuracy of manual ultrasound is ± 0.5 mm–1
The technician needs access
to the equipment at the measurement locations of interest; therefore,
scaffolding (possibly permanently installed) and the stripping of insulation to
expose the metal work to make the manual measurements are required.
installed ultrasonic wireless wall thickness monitoring sensors (FIG. 3) address these
and other issues, making them the best option for most high-temperature corrosion
These sensors can measure small changes in
wall thickness and exhibit robustness to extreme plant conditions, while also
having extended battery life to provide reliable operation over the entire
cycle between turnarounds. Additionally, they are simple and cost effective to
install at scale and have been field proven in thousands of applications
worldwide over the past few decades.
The following will
examine how these and other types of sensors can be used to manage corrosion in
the different processes required for biofuel refining.
Pretreatment, storage and reactor feed. FFAs
can cause severe corrosion to the piping and equipment upstream of the hydrogen
mixing point. Similar to past experiences with opportunity crudes and
naphthenic acid corrosion, this mechanism is sparsely documented due to the varying
feedstock scenarios and the limited availability of research or case studies.
A biofuel feedstock’s acidity is described by
its total acid number (TAN), which is a measure of acidity but not direct corrosivity.
Beyond the hydrogen mixing point in the reactors, the FFAs are converted into
water and CO2, aided by the hydrotreating catalyst.
The following processes typically experience
the following types of corrosion.
Acidic corrosion. Renewable
feedstocks contain a mix of fatty acids, long carbon chains with one or
multiple double bonds, and more branched acidic components, such as resin
TANs in biofuel processing are very high in
comparison to fossil fuel feedstocks, varying between 0 mg KOH/g and 200 mg
KOH/g feed. In addition, the decomposition of triglycerides will lead to an
additional acid load in processing units, which is not detected with TAN
testing in feedstocks.
The acidic corrosion mechanism causes
localized metal thinning, which increases with process temperatures and flow
velocities. While temperatures > 230°C (446°F) are considered susceptible
for fossil feeds, this threshold is reduced to 150°C (302°F) with renewable
feedstocks due to high levels of acidity. The formed iron salts are highly
soluble in oil and do not form protective scales and deposits in the catalyst
bed, causing pressure drop concerns and catalyst clogging.
FFAs are no longer a concern after entering
the hydrotreating section, but this process results in the formation of CO2.
Carbonic acid may be present in the storage section through feedstocks or their
degradation processes, or it can be formed in the reactor effluent due to the
presence of water and CO2. Carbonic acid corrosion will be described
in the reactor effluent section of Part 2 of this article.
Renewable diesel feedstocks may also contain
contaminants such as metals and chlorides, leading to known corrosion
MIC. Metabolic products of living organisms (e.g., bacteria, algae,
fungi) cause localized corrosion in the form of distinctive pitting and
tubercles, or crevice corrosion, often under bio-film deposits. All
aqueous environments are affected, even when water is only intermittently
present. All low-temperature areas are affected, from feedstock handling and
tank farms to air coolers in processing units to final wastewater treatment
Low-flow and stagnant conditions
promote the growth of these microorganisms and corresponding MIC. This
multitude of organisms will allow populations to be present in all conditions
and nutrients (such as inorganic substances like sulfur, ammonia, iron and
sulfate compounds), as well as hydrogen sulfide, organic hydrocarbons and
acids. The presence of carbon, nitrogen and phosphorus are required and
available in higher concentrations with biofuel processes—there are also aerobic
and anaerobic organisms, depending on the presence of oxygen. Organisms can be
found in pH ranges of 0–12 and in temperatures from –17°C–110°C (0°F–230°F).
As MIC damage is very localized, it is often detected
with corrosion coupons and timely sampling. Frequent flushing and biocide
application may control organism colonies. Established colonies may build up
protective sludge layers, complicating organisms and MIC control.
Part 2 will be featured in the October issue of Hydrocarbon
a Rosemount™ Wireless Permasense WT210 sensors
WILLIAM FAZACKERLEY is a Global Product Manager in Emerson’s Corrosion
and Erosion business unit. He has more than 10 yr of experience in information
technology (IT) and software development, and specializes in digital
transformation, working closely with customers to guide the strategic direction
of Emerson’s product portfolio. Fazackerley studied computing
and applied information and communications technology (ICT) at Central Sussex
College in the UK.