Chapter 9

Crude Storage, Blending, Desalting,

Distillation and Treating

Large quantities of crude oils are transported through pipelines or tankers. At the
marine terminal, a cargo of crude oil may be routed through a pipeline directly to a
storage tank in the refinery tank farm, or transferred to holding tank, where it is kept
temporarily before going to the refinery. The marine terminal also has berths to load
refined products. Storage tanks are containers that hold the crude oil or compressed
gases (gas tanks), as well as intermediate stocks (partially refined), finished products
and chemicals for the short- or long-term storage.

9.1 Crude Storage and Blending

Crude oils shipped to the refinery are stored in storage tanks. Liquid storage tanks
are often cylindrical in shape, perpendicular to the ground with flat bottoms, with
a fixed frangible or floating roof. Gas tanks for compressed or liquefied gas are in
spherical shape.
Different grades of crude oils are stored in different tanks. Care must be taken
to avoid mixing incompatible crudes, which can cause precipitation (due to deas-
phalting)., and fouling. Above ground storage tanks can also be used to hold blended
crudes, refined products, water, waste matter, and hazardous materials, while meeting
strict industry standards and regulations.
A refinery seldom refines a single crude. Prior to being charged to the refinery
units, crude oils and imported stocks are unloaded to storage tanks, then blended and
transferred to surge tanks for individual units. Proper blending improves distillation
unit throughput and the performance of downstream units. It also can improve product
quality and reduce energy cost [1]. Crude blending is based on computer models,
which are used in conjunction with scheduling models and operations plans. The
models and plans include timing, desired volumes, etc. They are unique for every
refinery, because they depend on the refinery configuration, logistics constraints,
tank inventory, feedstock composition, and local (and global) market forces. For
operations planning, some refiners still use comparatively simple stand-alone LP
© Springer Nature Switzerland AG 2019 159
C. S. Hsu and P. R. Robinson, Petroleum Science and Technology,
https://doi.org/10.1007/978-3-030-16275-7_9
160 9 Crude Storage, Blending, Desalting, Distillation and Treating

(linear program) models, which include just a limited number of parameters (5 or

6) for important process units. But the industry is moving toward sophisticated non-
linear optimizers, which are linked to fully rigorous process models. The non-linear
optimizers include site-specific scheduling and some are linked to corporate models
which include multiple refineries and chemical plants. Important advanced features
include back-casting (a comparison of previous plans with actual operation) and
trader-assistance applications.
Several crudes with similar molecular characteristics are usually blended together.
Incompatibilities between crude oils can occur, for example, between paraffinic crude
oil and heavy asphaltic crude oil. Blending is used to maintain, as nearly as possible,
constant feed quality for the crude distillation unit. For example, high acid crudes
are mixed with low acid crudes, and high-sulfur crudes are mixed with low sulfur
crudes to bring the acid or sulfur contents to a tolerable level for refinery operations.

9.2 Desalting

Crude oil introduced to refinery processing contains many undesirable impurities,

such as sand, inorganic salts, drilling mud by-products, polymers, corrosion byprod-
ucts, etc. The salt content in the crude oil varies depending on source of the crude oil
and how it is transported. When a mixture from many crude oil sources is processed
in refinery, the salt content can vary greatly.
These undesirable impurities, especially salts and water, need to be removed
prior to distillation. Two processes are used: dewatering and desalting. Dewatering
removes water and constituents of brine that accompany the crude oil. Desalting is a
water-washing operation that removes water-soluble minerals and entrained solids.
Both operations can be performed at the production field, at the refinery, or both
places.
The impurities of most concern in crude oil include:
• The inorganic salts, which can be decomposed to form acids or alkali in the crude
oil pre-heat exchangers and heaters. Salts are most frequently present in crude oil
are CaCl2 , NaCl and MgCl2 , which can form hydrogen chloride when heated in
the presence of water. Hydrogen chloride gas condenses to aqueous hydrochloric
acid at the overhead systems of distillation columns, causing serious corrosion of
equipment.
• Naphthenic and carboxylic acids, as measured by the total acid number (TAN)
test, also induce corrosion.
• The sand, silt, or salt cause deposits and foul heat exchangers. They can also cause
significant damage to pumps, pipelines, etc. due to abrasion or erosion.
• Inorganic and organic compounds of sodium, calcium nickel, vanadium, iron,
arsenic, and other metals in the crude can poison and deactivate catalysts.
9.2 Desalting 161

Fig. 9.1 Flow diagram of desalting involving an electrostatic desalter

• Organo-silicon compounds are added to crudes to improve flow rates through

pipelines and pumps. These Si compounds degrade catalysts by weakening alu-
mina or aluminosilicate supports. They are only partly removed by desalting.
At the refinery, the crude is treated with hot water in one or more desalters.
Desalters employ either chemical or electrostatic precipitators to remove dissolved
salts and collect remaining solids. In chemical desalting, water and surfactants are
added to the crude, heated to dissolve salts and other impurities, then sent to a settling
tank, where the water and oil separate.
Figure 9.1 shows a flow diagram of a typical desalting unit. The blended crude
is mixed with 3–10 vol.% water at 200–300 °F to dissolve salts in the crude oil and
to reduce viscosity and surface tension for easier mixing and subsequent separation
of the water. The oil/water mixture is homogenously emulsified. Then the emul-
sion enters the desalter vessel, where it separates into oil and water phases under
the influence of electrostatic coalescence, in which a high-potential field (typically
12–35 kV), is imposed across a pair of electrodes, shown in Fig. 9.2. The electro-
static field promotes coalescence with polarization. Polarization of water droplets
pulls them out from the oil-water emulsion phase. The salt dissolved in these water
droplets is also separated along the way. Coalescence is aided by passage through a
tower packed with sand, gravel, and the like.
An emulsion between oil and water can also be broken by adding chemicals or
treating agents, such as soaps, fatty acids, sulfonates, long-chain alcohols or other
de-emulsifiers.
A desalting unit can be designed with single stage or two stages, shown in Fig. 9.3.
Two-stage desalting system is normally applied, that consists of 2 electrostatic coa-
lescers (desalter). For resid processing, 3-stage desalting is used for some crudes.
162 9 Crude Storage, Blending, Desalting, Distillation and Treating

Fig. 9.2 Cross-section view of a desalter

If the crude isn’t desalted, residual solids can clog downstream equipment and
deposit on heat exchanger surfaces, thereby reducing heat-transfer efficiency. Salts
can induce corrosion in major equipment and deactivate catalysts.

9.3 Distillation

Distillation can be considered as the heart of any refinery. Crude oils are made
of numerous components of different boiling points. The simplest way to separate
them is with continuous distillation into different fractions (distillates or cuts) of
various boiling ranges. At just above atmospheric pressure, heavy molecules in most
crude oils decompose above 650 °F (~350 °C). To achieve additional separation of
heavy fractions, continuous distillation is carried out under vacuum (i.e., reduced
pressure-typically at 40 mmHg) to boil out additional components. Vacuum distil-
lation increases the yield of total distillates. The relationship between boing points
under atmospheric pressure and under 40 mmHg vacuum is shown in Fig. 2.1. For
9.3 Distillation 163

Fig. 9.3 Single- and Two-Stage electrostatic desalting systems

example, at 500 °F under 40 mmHg vacuum, the compounds with boiling point
at 750 °F under atmospheric pressure can be boiled out. Hence, the atmospheric
equivalent boiling point (AEBP) at 500 °F under 40 mmHg vacuum is 750 °F. The
AEBP of 650 °F is ~900 °F. Hence, additional components that boil between 650
and 900 °F can be distilled out of the crude oil under 40 mmHg vacuum without
severe decomposition. The upper temperature for vacuum distillation in refineries
can be slightly higher, up to 1050 °F. At this temperature, there is a greater tendency
for thermal decomposition, but the decomposition does not occur immediately; the
feed flows out of the column and undergoes cooling before any damage is done.
Figure 9.4 shows a simplified diagram for crude oil distillation. There are many
trays in a distillation tower (also called a pipestill or column), which will be discussed
later. The desalted crude is introduced near the bottom of the atmospheric distillation
tower. The lightest fractions, gases and naphtha, flow out of the tower at the top as
an overhead stream, which can be further fractionated into separate gas and naphtha
streams. The effluents from middle trays are heavy naphtha, kerosene, light gas oil and
heavy gas oil. Kerosene and light gas oil are often referred also as middle distillates
or distillate fuels that include kerosene, jet fuel and diesel. Steam is introduced to
164 9 Crude Storage, Blending, Desalting, Distillation and Treating

LPG, Propane
Condenser
Reflux Drum
Naphtha
Naphtha
Reflux Splitter
Atmospheric
Distillation Unit

Light Gas Oil

Desalters
Heavy Gas Oil

Vacuum Gas Oil

Vacuum
Atmospheric Distillation Unit
Reboiler Residue
Vacuum Residue

Fig. 9.4 Simplified diagram of crude distillation

Table 9.1 Boiling ranges of distillation fractions

Fraction Boiling range (to the nearest 5°)
°C °F
Liquefied petroleum gas −40 to 0 −40 to 30
(LPG)
Light naphtha 30–85 80–185
Gasoline 30–200 90–400
Heavy naphtha 85–200 185–390
Kerosene 170–270 340–520
Light gas oil 180–340 350–650
Heavy gas oil 315–425 600–800
Lubricant oil >400 >750
Vacuum gas oil 340–565 650–1050
Residuum >540 >1000

improve separation by reducing the partial pressures of hydrocarbons. The residue, or

bottom, of the atmospheric pressure distillation tower, also called as reduced crude,
is introduced into a vacuum distillation unit (tower) to be fractionated into vacuum
gas oil and vacuum residue.
The boiling ranges of the distillation fractions are listed in Table 9.1. These num-
bers are only used as references. The actual cut points may vary at various refineries,
depending on the season and market demands.
The distillation cuts obtained in refineries are not well-defined. There are always
overlaps between adjacent cuts, as shown in Fig. 9.5 for a desired cut point of 315 °F
9.3 Distillation 165

overlaps

Tail ends

Fig. 9.5 Overlaps and tail ends of distillation cuts [2]

between naphtha and kerosene. The leading edge of the kerosene distillation curve
overlaps with the tailing edge of naphtha. The overlap between cuts can be “sharp”
or “sloppy,” depending on several factors, especially oil flow rates, steam flow rate,
and heat balance. The designations are arbitrary, and often nothing can be done to
decrease overlap at maximum flow rate without making hardware changes. Overlaps
exist for all distillation cuts, in some commercial units, the overlaps are very large.
Such “sloppy cuts” are more common in units running far above (or far below) their
design feed rate.
Figure 9.6 demonstrates the application of boiling point overlaps. The desired
cutpoints for light naphtha, heavy naphtha, kerosene, and heavy diesel are 90 °F
(32 °C), 190 °F (88 °C), 300 °F (149 °C), and 525 °F (274 °C), respectively. Due to
operational constraints, the observed effective cutpoints are 99 °F (37 °C), 188 °F
(87 °C), 302 °F (150 °C), and 523 °F (273 °C), respectively. The overlap between
kerosene and heavy diesel is considerable. If we move the initial boiling point (IBP)
for the heavy diesel at 360 °F (182 °C) instead of 523 °F (274 °C), it would result
in considerable entrainment of valuable kerosene into the far-less-valuable bottom
product.
A petroleum refinery can adjust distillation yields to meet market demands, in
part, just by adjusting cut points. The swing cut between 150 and 205 °C can go into
any of the three products—naphtha (gasoline), kerosene (jet fuel) and gas oil (diesel
and heating oil), shown in Fig. 9.7, depending on the seasonal and market demands.
166 9 Crude Storage, Blending, Desalting, Distillation and Treating

Heavy Diesel
Cutpoint
523°F (273°C)
Temperature, ºF

Kerosene
Heavy Naphtha Cutpoint
Cutpoint 302°F (150°C)
188°F (87°C)

Light Naphtha
Cutpoint
99°F (37°C)

CumulaƟve Volume-Percent Recovery

Component DisƟllaƟons Composite Curve

Fig. 9.6 Overlap in distillation curves from a commercial hydrocracker

Fig. 9.7 Swing cut region for naphtha, kerosene and gas oil

In summer “driving season,” the cutpoint is moved to higher temperatures to increase

the yield of transportation fuels. In winter, the cutpoint is set at lower temperatures
to increase the yield of heating oil (gas oil heavier than diesel).

9.3.1 Atmospheric Distillation

Figure 9.8 shows a flow diagram of an atmospheric distillation unit. The crude oil
enters a desalter at 250 °F to remove salt and water, as described before. The desalted
oil goes through a network of pre-heat exchangers to a fired heater, which brings the
temperature up to 657–725 °F (347–385 °C). If the oil gets much hotter than this,
9.3 Distillation 167

Fig. 9.8 Flow diagram of an atmospheric distillation tower

it starts to crack, generating carbon. The carbon would deposit inside the pipes and
equipment through which the oil flows.
The hot crude enters the tower just above the bottom, as shown in Fig. 9.8. Steam is
added at the bottom to enhance separation; it does so largely by decreasing the vapor
pressure of hydrocarbons in the column. When it enters the tower, most of the oil
vaporizes. The steam flows upward with vaporized crude while the condensed liquid
flows downward as in a countercurrent fashion. The hottest trays are in the bottom
section with the coolest at the top section. Unvaporized oil drops to the bottom of
the tower, where it is drawn off.
Products are collected from the top, bottom and side of the column. Side-draw
products are taken from trays where the temperature corresponds to the cutpoints for
a desired product (naphtha, kerosene, light gas oil and heavy gas oil). Some of the
side-draws can be returned to the tower as a pump-around or pump-back stream to
control tower temperatures and improve separation efficiency.
Two side cut strippers for naphtha and gas oil are shown as examples. There can be
additional strippers for kerosene (jet fuel) and diesel (light gas oil). Also not shown
is the reboiler at the bottom of the tower; this will be discussed later. An atmospheric
distillation tower usually contains 30–50 fractionation trays, with 5–8 trays in each
section. Product strippers for cut streams also have 5–8 trays. The strippers remove
entrained light components from liquids. Stripper bottom streams can be drawn off
as products of a specific boiling range or returned to the distillation tower.
Inside the distillation tower (also called pipestill or column), the vapors rise
through the distillation trays, which contain perforations, bubble caps, downcom-
ers, and/or modifications thereof, shown in Fig. 9.9 (perforations on the trays not
168 9 Crude Storage, Blending, Desalting, Distillation and Treating

Fig. 9.9 Bubble cap and downcomer on a distillation tray. Intermediate products are removed
through side-draw trays

shown). Vapors and liquids flow counter-currently. Each tray permits vapors from
below to bubble up through the relatively cool condensed liquid on top of the tray.
This vapor/liquid contact knocks heavy material out of the vapor. Condensed liquid
flows down through a pipe (downcomer) to the hotter tray below, where the higher
temperature causes re-evaporation. A given molecule evaporates and condenses many
times before finally leaving the tower.
Figure 9.10 is another drawing for an atmospheric distillation tower. The section
above the feed tray is called enriching, or rectification, section and the section below
the feed trap is stripping section. Gas and naphtha are withdrawn from the top tray
as overhead. After condensation, a portion of the liquid is introduced back to the
top tray as reflux. Reflux also controls temperature in the enriching section. It also
controls entrainment of heavier components in the naphtha (the lightest distillate).
Reflux ratio is the amount of reflux liquid returning to distillation tower divided by
the amount of liquid withdrawn as product per unit time. With a higher reflux ratio,
fewer theoretical plates are required.
At the bottom of the tower, the liquid (atmospheric residue) passes through a
reboiler to recover light components from the heavy liquid. The reboiler helps control
temperatures in the stripping section.
The bottom stream from the main fractionator (atmospheric distillation tower) is
called atmospheric bottoms, atmospheric residue, reduced crude, topped crude, or
long resid.
9.3 Distillation 169

Fig. 9.10 Atmospheric distillation (Fractionation) tower with a reflux for overhead and a reboiler
for bottoms

9.3.2 Vacuum Distillation

Figure 9.11 shows a flow diagram of a vacuum distillation unit. The atmospheric
residue goes to a fired heater, where the typical outlet temperature is about 730–850 °F
(390–450 °C). From the heater, the atmospheric residue goes to a vacuum distillation
tower. Steam ejectors reduce the absolute pressure to 25–50 mmHg vacuum, or about
7.0 psia (0.5 bara). Under vacuum, hydrocarbons vaporize at lower temperatures than
atmospheric boiling points. For example, the equivalent atmospheric boiling point of
800 °F under 40 mmHg vacuum is ~1050 °F. Thus, molecules with normal boiling
points above 650 °F (343 °C) are less likely to undergo thermal cracking and can
be vaporized at lower temperatures. There are fewer trays than the atmospheric
distillation tower to fractionate the topped crude into light vacuum gas oil, heavy
vacuum gas oil and vacuum residuum at the bottom. As in atmospheric distillation,
some gas and light components entrained or decomposed during heating in furnace
are carried out at the top of the tower.
170 9 Crude Storage, Blending, Desalting, Distillation and Treating

Fig. 9.11 Flow diagram of a vacuum distillation Unit

9.3.3 Distillation Yields of Straight Run Fractions

The products from distillation prior to upgrading are called straight-run products. At
a given set of cutpoints, the yields of different fractions depend on the crude oil being
processed. Figure 9.12 shows TBP of a light and a heavy crude for kerosene yield
(cut points between 315 and 450 °F). The light crude yields more kerosene than the
heavy crude, and hence has a higher value.
Table 9.2 lists a few crude oils and their typical straight-run yields. Total naphtha
includes light, medium and heavy naphtha, and the middle distillates include kerosene
and atmospheric gas oil. Naphtha is used for making gasoline and aromatics, kerosene
for jet fuel and atmospheric gas oil for diesel. Table 9.3 shows that the demand
for transportation fuels exceeds the straight-run yields for the crudes in Table 9.2.
Obviously, crudes containing less heavy material—VGO and vacuum residue are
more valuable.
The higher-valued crude oils, such as Brent and Bonny Light, have higher API
gravity with higher naphtha and middle distillate yields. They tend to have less sulfur.
The oil having high sulfur content increases processing costs because the sulfur must
be removed. Hence, the oils, such as Green Canyon and Ratawi, have lower values.
Since sulfur is not removed during distillation, the straight-run distillation products
have to be treated for sulfur removal.
Products from the crude distillation unit, i.e., the straight-run distillates, go to other
process units, as shown in Table 9.4. The lightest cuts are gas and light naphtha. The
gas goes to a gas processing plant or is liquefied into liquefied petroleum gas (LPG).
9.3 Distillation 171

Fig. 9.12 Comparison of kerosene yields from a light and a heavy crude oil [2]

Table 9.2 Typical straight-run yields from various crudes

Source field Brent Bo liny Lt. Green Canyon Ratawl
Country Norway Nigeria USA Mid East
API gravity 38.3 35.4 30.1 24.6
Specific gravity 0.8333 0.8478 0.8752 0.9065
Sulfur, wt% 0.37 0.14 2.00 3.90
Yields, wt% feed
Light ends 2.3 1.5 1.5 1.1
Light naphtha 6.3 3.9 2.8 2.8
Medium naphtha 14.4 14.4 8.5 8.0
Heavy naphtha 9.4 9.4 5.6 5.0
Kerosene 9.9 12.5 8.5 7.4
Atmospheric gas oil 15.1 21.6 14.1 10.6
Light VGO 17.6 20.7 18.3 17.2
Heavy VGO 12.7 10.5 14.6 15.0
Vacuum residue 12.3 5.5 26.1 32.9
Total naphtha 30.1 27.7 16.9 15.8
Total middle distillate 25.0 34.1 22.6 18.0
172 9 Crude Storage, Blending, Desalting, Distillation and Treating

Table 9.3 Average U.S. consumption of petroleum products, 1991–2003

Product Consumption (barrels/dav) Percent of total (%)
Gasoline 8,032 43.6
Jet Fuel 1,576 8.6
Total distillates 3,440 18.7
Residual fuel oil 867 4.7
Other oils 4,501 24.4
Total consumption 18,416 100
Sum of gasoline, Jet and distillates 13,048 70.8

The light naphtha can be hydrotreated and sent to the motor gasoline blending pool.
Heavy naphtha is a feed for catalytic reforming units.
Kerosene can be used for lighting, heating, and for making jet fuel. In either case,
it must first undergo hydrotreating. Light gas oil can go to diesel fuel (distillate fuel
oil) blending.
Heavy gas oil can become fuel oil, diesel, lube base stock, or a light component
of feed for fluid catalytic cracking (FCC) or hydrocracking.
Vacuum gas oil (VGO) and vacuum resid (residuum or residue, VR) are low
valued. They are normally converted into higher-value products through various
upgrading processes, as in a conversion refinery. VGO can become fuel oil or lube
base stock, but its primary destinations are FCC and hydrocracking units, which
are discussed in subsequent chapters. Figure 8.12 gives more details of the possible
destinations of vacuum resid which is also known as “bottom of the barrel”.

9.4 Hydrogenation, Dehydrogenation and Condensation

The saturation of aromatics and polyaromatics (Fig. 9.13) is common to sev-

eral petroleum refining processes. It is especially important in catalytic reforming,
hydrotreating, hydrocracking, and cyclohexane production. Aromatics saturation is
reversible, i.e., hydrogenation is accompanied by dehydrogenation, except when
saturated C-C bonds are cracked or when polyaromatics condense to form larger
polyaromatics. Carbon-carbon single bonds in polynaphthenes and naphthenoaro-
matics can be broken irreversibly, but bonds within aromatic rings are more stable
and difficult to break. Condensation is a reaction in which two ring molecules are
combined into one. Aromatic ring condensation is accompanied by dehydrogenation.
Figure 9.14 summarizes thermodynamic calculations for the competition between
the saturation of naphthalene and the condensation of naphthalene with o-xylene to
form chrysene. At high pressures and low temperatures, equilibrium favors satura-
tion. At low pressures and high temperatures, equilibrium favors dehydrogenation.
At high-enough temperatures, equilibrium favors condensation. Fixed-bed catalytic
9.4 Hydrogenation, Dehydrogenation and Condensation 173

Table 9.4 Destinations for straight-run distillates

Approx. Boiling Next Ultimate product(s)
range or
Fraction °C °F Destination Subsequent
destination
LPG −40 to 0 −40 to 31 • Sweetener • Propane fuel
Light 39–85 80–185 • Hydrotreater • Gasoline
naphtha
Heavy 85–200 185–390 • Cat. reformer • Gasoline,
naphtha aromatics
Kerosene 170–270 340–515 • Hydrotreater • Jet fuel, No. 1
diesel
Gas oil 180–340 350–650 • Hydrotreater • Heating Oil, No.
2 diesel
Atmospheric 340+ 650+ • Visbreaker • FCC or
resid hydrocracker
feed,
low-viscosity
resid
• Resid • Resid FCC
hydrotreater
• Ebullated bed • Naphtha, gas oils,
hydrocracker FCC
Vacuum gas 340–566 650–1050 • FCC • Gasoline, LCO,
Oil gases including
C3 /C4 olefins
• Hydrotreater • Fuel oil, FCC,
lubes
• Hydrocracker • Naphtha, jet,
diesel, FCC,
olefins, lubes
• Solvent refining • DAO, asphalt
Vacuum 540+ 1000+ • Coker • Coke, coker gas
resid oil, coker
naphtha, gases
• Solvent refining • DAO, asphalt
• Slurry-phase • Traditional
hydrocracker hydrotreater or
hydrocracker
Resid is an abbreviation for residua, and 340+ (etc.) means everything that boils above 340 °C
(etc.)
LCO: light cycle oil; FCC: fluidized catalytic cracking
174 9 Crude Storage, Blending, Desalting, Distillation and Treating

CH3 3 H2 CH3

toluene methylcyclohexane

+ 4 H2

phenanthrene sym-dinaphthenobenzene

+ 2 H2

naphthenonaphthalene

+ H2

dihydrophenanthrene

+ 3 H2

dinaphthenocyclohexane

Fig. 9.13 Examples of saturation reactions of aromatics and polyaromatics

hydrotreating and hydrocracking units operate in the aromatics crossover region

between 315 and 425 °C (600 and 800 °F).
Figure 9.15 shows the so-called zig-zag mechanism for the production of large
polyaromatics (aromatic ring growth) by adding 2-carbon (cata-condensation) and
4-carbon (peri-condensation) species [3]. The condensation of large polyaromatics
via the Scholl reaction can lead eventually to coke formation and deactivation of the
catalysts used in FCC, hydrotreating and hydrocracking.
Figure 9.16 shows a mechanism for the one-at-time buildup of rings on a nucleus
of coke, or ring growth of aromatics. The mechanism includes the following steps,
all of which are to some extent reversible:
• Hydrogen abstraction by a radical
• Alkylation (olefin addition)
• Cyclization
• Dehydrogenation
9.4 Hydrogenation, Dehydrogenation and Condensation 175

Fig. 9.14 Thermodynamic calculations illustrating the competition between the saturation (hydro-
genation) and the condensation (dehydrogenation) of polyaromatics. Data for the graphs were
generated by Aspen Plus for a six component system comprising naphthalene (C10 H8 ), tetralin
(C10 H12 ), decalin (C10 H18 ), o-xylene (C8 H10 ), chrysene (C18 H12 ) and hydrogen (not shown)
176 9 Crude Storage, Blending, Desalting, Distillation and Treating

A
peri-condensaƟon

C
cata-condensaƟon

Fig. 9.15 Zig-zag mechanism for the condensation of polyaromatics by sequential addition of
2-carbon and 4-carbon units. The isomers shown are a naphthalene, C10 H8 ; b phenanthrene,
C14 H10 ; c pyrene, C16 H10 ; d benzo[e]pyrene, C20 H12 ; e benzo[ghi]perylene, C22 H12 ; f coronene,
C24 H12 ; g dibenzo[b,pqr]perylene, C26 H14 ; h benzo(pqr)naphtho(8,1,2-bcd)perylene, C28 H14 ;
i naphtho[2’.8’,2.4]coronene, C30 H14 ; and j ovalene, C32 H14 . Note how the H/C ratio goes down
as condensation increases, from 0.8 for naphthalene to 0.4375 for ovalene

Hydrogen abstraction Dehydrogenation

CH3 CH4 H 2 H2
Alkylkation
(olefin addition) Cyclization
H

Fig. 9.16 Mechanism for the addition of rings to an existing layer of coke
9.5 Treating/Sweetening 177

9.5 Treating/Sweetening

Gases and straight-run distillates need to be treated for removal of undesirable impu-
rities before they can be used as products.
Several treating processes entail seemingly simple acid-base reactions, such as
alkanolamine treating and caustic scrubbing. Alkanolamine treating removes acid
gases—H2 S and CO2 —from fuel gas and off-gas streams. In some hydrotreaters and
hydrocrackers, high-pressure amine units remove H2 S from the recycle gas. Caustic
scrubbers are used in several ways, including removing the last traces of H2 S from
the hydrogen used for processes in which the catalysts are highly-sulfur sensitive.

9.6 Hydrotreating

Hydrotreating is used in the pretreatment of process streams or the finishing of

products in hydrofining or hydrofinishing. It removes sulfur, nitrogen, oxygen and
trace elements which are detrimental to subsequent processes or affect the quality
and appearance of the products. If the contaminants are not removed, products will
not meet specifications.
Hydrotreater feeds range from naphtha to vacuum residues. Generally, each
fraction is treated separately. Materials with higher boiling points require more
severe treatment conditions. Table 9.5 lists feeds and products for hydrotreating
and hydroprocessing units.
The main hydrotreating reactions are summarized below:
Olefin saturation: olefin + H2 → paraffin + heat
Aromatic saturation: aromatics + H2 → naphthenes + heat
Dehydrogenation: naphthenes + heat → aromatics + H2
HDS: sulfur compounds + H2 → hydrocarbons + H2 S + heat
HDN: nitrogen compounds + H2 → hydrocarbons + NH3 + heat
HDO: oxygen compounds + H2 → hydrocarbons + H2 O + heat
HDM: organometallics + H2 → hydrocarbons + metal + adsorbed
contaminants
As mentioned in Sect. 9.4, the saturation of aromatics and the dehydrogenation of
the corresponding naphthenes are in equilibrium. It is important to note that, except
for dehydrogenation, all reactions are exothermic.
HDS is hydrodesulfurization. HDN is hydrodenitrogenation. HDO is hydrodeoxy-
genation, and HDM is hydrodemetallation [4]. HDM is especially relevant to the
hydrotreating of heavier feeds, such as VGO and residue. From such streams, HDM
removes trace elements, such as nickel, vanadium, iron, silicon, arsenic, etc., which
are poisons to downstream catalysts.
Arsenic and mercury are especially challenging. They form volatile alkyl com-
pounds, which can appear in all distillate fractions.
178 9 Crude Storage, Blending, Desalting, Distillation and Treating

Table 9.5 Feeds and products for hydroprocessing units

Feeds Products from hydrotreating Products from hydrocracking
Heavy naphtha Catalytic reforming feed LPG
Straight-run light gas oil Kerosene, jet fuel Naphtha
Straight-run heavy gas oil Diesel fuel Naphtha
Atmospheric residue Lubricant base stock, low Naphtha, middle distillates,
sulfur fuel oil, RFCC feed FCC feed, lubricant base
stocks.
Vacuum gas oil FCC feed, lubricant base stock Naphtha, middle distillates,
FCC feed, lubricant base
stocks, olefin plant feed
Vacuum residue RFCC feed See note (b)
FCC light cycle oil Diesel blend stock, fuel oil Naphtha
FCC heavy cycle oil Fuel oil blend stock Naphtha, middle distillates
Visbreaker gas oil Diesel blend stocks, fuel oil Naphtha, middle distillates
Coker gas oil FCC feed Naphtha, middle distillates,
FCC feed, lubricant base stock,
olefin plant feed
Deasphalted oil Lubricant base stock, FCC feed Naphtha, middle distillates,
FCC feed, lubricant base stock
Note (a) FCC = fluid catalytic cracking, RFCC = residue FCC
(b) Traditional fixed-bed hydrocrackers cannot process vacuum residue. The use of ebullated-bed
and slurry-phase hydrocrackers produces naphtha, middle distillates and FCC feed

Organosilicon compounds can be present in naphtha-range cuts from delayed

coking units. Heavier fractions may be contaminated by silicon-containing flow
improvers.
All reactions occur at the same time, to one extent or another. Obviously, HDO
and HDM occur only when oxygen and trace elements are present.

9.6.1 Hydrotreating Process Flow

Figure 9.17 presents a process flow scheme for a one-reactor fixed-bed hydrotreater
with four catalyst beds. Reaction conditions depend on feed quality and process
objectives.
The feed is warmed with heat exchange, mixed with hydrogen-rich gas and passed
through a furnace, where it is heated to the desired reactor inlet temperature, typically
600–780 °F. The temperature depends on process objectives and catalyst activity.
The heated mixture flows down through reactors loaded with catalysts. In many
diesel hydrotreaters and almost all VGO hydrotreaters, the reactors have multiple
beds, separated by quench decks. As the hydrotreating reactions occur, they con-
9.6 Hydrotreating 179

Treat
Gas
Recycle Gas
Furnace Compressor

Makeup H2
Quench
Gas Makeup Gas
Recycle Compressor
Gas
TC

Purge
TC
Gas
Reactor

TC
Amine Unit

H2S
(ads)

CHPS
F/E Heat LP Flash
Exchange Gas
Cooler

Wash Water CLPS

Feed Pump NH4SH (aq)
To Stripper,
Fractionator

Fig. 9.17 Representative hydrotreating unit with one reactor, four catalyst beds, and two flash
drums. Naphtha hydrotreaters have one bed in one reactor, and may employ hydrogen once-through
(no recycle). F/E exchanger is the feed/effluent heat exchanger. Temperature controllers (TC) control
temperature by manipulating the flow of quench gas. CHPS is the cold high-pressure separator.
CLPS is the cold low-pressure separator. Wash water is injected to remove ammonia as aqueous
ammonium bisulfide, which otherwise would precipitate in cold spots in or downstream from the
CHPS, blocking flow and inducing corrosion

sume hydrogen and generate heat. The heat is controlled by bringing relatively
cool recycle gas into the quench decks, where it mixes with reaction fluids from
the bed above. Makeup gas comes into replace consumed hydrogen. Gas flow can
be once-through in naphtha hydrotreaters, but in distillate and VGO hydrotreaters,
unconsumed hydrogen is recycled.
Fluids exiting the reactor are cooled with heat exchange before going to an arrange-
ment of separation towers, which include a stripper and, for high-pressure units, two
or more flash drums.
Hydrotreating produces both H2 S and NH3 . Under reaction conditions, these
remain in the gas phase. But at lower temperatures, they combine to form solid
ammonium bisulfide (NH4 SH). Ammonia also reacts with chlorides to form NH4 Cl;
chloride can come with makeup gas, feed, or wash water. These salts can deposit in air
coolers and heat exchangers, blocking flow and—even worse—inducing corrosion.
180 9 Crude Storage, Blending, Desalting, Distillation and Treating

Fortunately, they are water-soluble, so they can be controlled by injecting wash water
into the reactor effluent for removal.
The CHPS flash gas is recycled. An optional amine unit removes H2 S from the
recycled gas. Some of the recycle gas might be purged to control hydrogen purity.
Depending on the unit and the feed, oils from the separation towers can go various
places. Stripped naphtha, kerosene or diesel might go directly to downstream units
or product blenders. In VGO or residue hydrotreaters, liquids from the separation
section go to a fractionator. Other off-gases, along with sour high-pressure purge gas,
are treated with amine before going to other units or the refinery fuel-gas system.

9.6.2 Hydrotreating Chemistry

A summary of hydrotreating reactions appeared in the introduction to this section.

Here, we discuss the reactions in more detail.

9.6.2.1 Saturation of Olefins

Olefins are rare in straight-run feeds, but they are relatively abundant in cracked stocks
from coking or FCC units. Saturation of olefins occurs rapidly. If not controlled, it
can lead to polymerization and consequent plugging of catalyst beds. The best ways
to control olefins are (a) to design hydrotreaters with small low-temperature beds up
front and/or (b) to employ activity grading. In activity grading, catalysts are layered,
with low activity catalysts on top, followed by successively more-active catalysts.

9.6.2.2 Saturation of Aromatics/Dehydrogenation of Naphthenes

Saturation of aromatics and dehydrogenation of naphthenes are described in Sect. 9.4.

Saturation reactions play a significant role in hydrotreating and hydrocracking. In
hydrotreating, the aromatics crossover effect increases the difficulty of removing
nitrogen compounds and large sulfur compounds. The condensation of aromatics
leads to coke formation and consequent catalyst deactivation.

9.6.2.3 Hydrodesulfurization (HDS) for Sulfur Removal

Figure 9.18 shows representative hydrodesulfurization (HDS) reactions. For sulfur

removal from the first four reactants, the mechanism is straightforward. That is,
the sulfur-containing molecule interacts with an active site on the catalyst, which
removes the sulfur atom and replaces it with two hydrogen atoms. Additional hydro-
gen converts the sulfur atom into H2 S, which desorbs from the catalyst, leaving
behind a regenerated active site.
9.6 Hydrotreating 181

2 H2
CH3CH2-SH CH3CH3 + H2S
ethyl mercaptan Direct ethane

2 H2
CH3-S-C4H9 CH4 + C4H10 + H2S
t-butylmethylsulfide Direct methane isobutane

3 H2
CH3-S-S-CH3 2 CH4 + H 2S
Dimethyldisulfide (DMDS) Direct methane

2 H2
+ H 2S
Direct
S
dibenzothiophene biphenyl

2 H2
+ H2S
S Direct
H 3C CH3 H3C CH3

4,6-Dimethyldibenzothiophene

Indirect 2 H2

2 H2 + H2S
S
H 3C CH3 H3C CH3

Fig. 9.18 Representative hydrodesulfurization (HDS) reactions

For the fifth reactant—4,6-dimethyldibenzothiophene (4,6-DMDBT) which has

two methyl groups near the vicinity of the sulfur atom—HDS proceeds via both a
direct and an indirect route. Overall, the indirect route is considerably faster. The
4,6-DMDBT molecule is planar, and the two methyl groups block the access of the
sulfur atom to the catalyst surface as “hindered sulfur”, thereby inhibiting direct
HDS. In the indirect route, saturating one of the aromatic rings that flank the sulfur
atom converts the planar structure into a puckered configuration with tetrahedral C-
C bonds. This puckering rotates one of the inhibiting methyl groups away from the
sulfur atom, giving its better access to the catalyst.
Making ultra-low-sulfur diesel (ULSD), in which the sulfur content is less than
10–15 wppm, requires severe hydrotreating, after which the only remaining sulfur
compounds are the above-mentioned 4,6-DMDBT and other di- and trimethyl diben-
zothiophenes. Because the removal of sulfur from these compounds is more facile
182 9 Crude Storage, Blending, Desalting, Distillation and Treating

Fig. 9.19 Representative hydrodenitrogenation (HDN) reactions

after prior saturation, the crossover phenomenon affects the production of ULSD
significantly, so much so that it governs the design of commercial units.

9.6.2.4 Hydrodenitrogenation (HDN) for Nitrogen Removal

Figure 9.19 shows representative HDN reactions, and Fig. 9.20 presents the mech-
anism for the HDN of quinoline. As with sulfur removal from hindered DMDBTs,
the aromatics crossover phenomenon is important for deep HDN because nitrogen
removal requires prior saturation of an aromatic ring adjacent to the nitrogen atom.

9.6.2.5 Hydrodeoxygenation (HDO) for Oxygen Removal

In feeds to industrial hydroprocessing units, oxygen is contained in furans, organic

acids, ethers, peroxides, and other compounds. Some are formed by reaction with
air during transportation and storage of crude oil, distillates, and cracked stocks
generated by delayed coking and FCC units (see below). Phenols and cresols and
quite stable, but some hydrodeoxygenation (HDO) reactions proceed so rapidly that
they can cause problems with excessive heat release. Also, oxygen compounds can
form gums and polymers, which inhibit flow and increase pressure drop. Feedstocks
from biomass contain considerably more oxygen than conventional crude oil and
petroleum distillates. While a conventional vacuum gas oil may contain 0.5 wt%
9.6 Hydrotreating 183

Extent of Reaction Controlled by

Kinetics at Low T

2 H2 C 3H 7
H2

N N NH2
fast slow
quinoline H

slow 2 H2 fast 3 H2 3 H2

3 H2 H2 C3H7

N N Hydrogenolysis NH2
H Rate limiting
Not reversible
H2

C 3H7

+ NH3

Fig. 9.20 Mechanism for the HDN of quinolone

oxygen, a bio-derived oil in the same boiling range might contain >40% oxygen.
Transportation fuels must meet tight specifications, or they can’t be used—at least
not for very long. Producing fuels from bio-derived oils in conventional oil refineries
requires extra hydrogen and presents processing and storage challenges. Large-scale
hydrogen production generates large quantities of CO2 . When assessing the environ-
mental impact of replacing conventional crude oil fractions with bio-derived oils, it
is crucial to consider the hydrogen required for upgrading.

9.6.2.6 Hydrodemetallation (HDM) for Trace Element Removal

Metals and other trace elements poison catalysts in hydrotreaters and downstream
process units. The following are the most troublesome:
• Nickel and vanadium are present in high-boiling fractions, mostly in asphaltenes.
Asphaltenes are mixtures of waxy solids with porphyrins.
• Corrosion generates soluble iron.
• Entrained salt brings in alkali and alkaline earth salts, primarily carbonates and
bicarbonates of sodium and calcium.
• Arsenic is present in crudes from West Africa, the Ukraine, Canada, and elsewhere.
Synthetic crudes from oil sands and oil shale tend to contain significant amounts.
Arsenic is one of the worst catalyst poisons—hundreds of times worse, on a weight
basis, than Ni, V, or Fe. Arsenic forms organo arsines, which are relatively volatile.
They tend to be most abundant in middle distillates.
184 9 Crude Storage, Blending, Desalting, Distillation and Treating

• Mercury components, including elemental mercury, mercuric chloride, mercuric

sulfide, mercuric selenide, dimethylmercury, diethylmercury, etc. are present in
all oil and gas reservoirs in trace amounts, usually in a few ppb levels or less.
However, several hundred ppb mercury can be found in crudes and natural gas
in southeast Asia, Australia and South America, such as those produced offshore
Thailand. Mercury is nearly as bad as arsenic, both for volatility and poisoning
potential.
• Silicon comes in as silicones, which are added to crude oil to facilitate flow though
pipelines. Silicones are also used to control foaming in delayed coking units. Most
end up in light fractions, such as heavy naphtha.
In hydrotreating, trace elements are removed with graded guard material (adsor-
bents), including special wide-pore guard catalysts. Guard materials remove some
contaminants by chemisorption. Other contaminants, such and Ni, V, and soluble Fe,
are removed by HDM. HDM reactions convert the metals into sulfides, which adhere
to the guard material.

9.6.3 Hydrotreating Catalysts

The earliest hydrodesulfurization catalysts were bauxite and fuller’s earth. Later,
catalysts containing cobalt molybdate on alumina and nickel tungstate on alumina
substantially replaced the earlier catalyst and these catalysts are still used very exten-
sively.
Most commonly used hydrotreating catalysts are MoS2 promoted by Cox Sy and/or
NiS on gamma alumina. WS2 has also been used commercially in place of MoS2 .
Improved compositions and manufacturing methods have increased activity more
than ten-fold since the process was invented in the 1950s, but despite intense ongoing
efforts to find better catalysts, the original raw materials remain unsurpassed. Other
metal sulfides, such as RuS2 , IrS2 , OsS2 , and RhS2 , are more active than MoS2 and
WS2 , but they either deactivate too quickly or they are too expensive.
When the catalysts are manufactured, the metals are oxides, which have relatively
low activity. Before use, they must be activated by reductive sulfidation, a process
commonly known as sulfiding.
For the low-pressure HDS of light feeds, CoMo catalysts are preferred. For high-
pressure deep desulfurization and HDN, NiMo catalysts are used, either alone or in
combination with CoMo. Some suppliers claim that “sandwich” configurations, with
alternating layers of CoMo and NiMo catalysts, provide superior HDS performance.
In the past, catalysts based on Ni-promoted WS2 were employed for HDN, due to
the high saturation activity of the WS2 . However, WS2 catalysts are seldom if ever
used now.
The catalyst support is just as important as the active metals. The support must be
strong enough to endure considerable pressure and hydraulic stress, and it must have
a high surface area. The supports are microporous, with 99+ percent of their surface
9.6 Hydrotreating 185

areas inside the pores. The pore diameters must be wide enough to admit reactants
but narrow enough to exclude very large residue molecules.
In fixed-bed units, hydrotreating catalysts last for 1–5 years, typically 2–3 years.
During a catalyst cycle, the catalysts slowly deactivate and temperatures are raised to
compensate. Start-of-run reactor average temperatures can range from 600 to 700 °F
(315–370 °C), depending on hydrogen partial pressure, feed rate, feed quality, and
desired product quality. When the required temperature reaches a limit set by metal-
lurgy, the feed rate must be reduced, the product quality objective must be relaxed, or
the catalyst must be changed. Typical end-of-run average temperatures range from
780 to 800 °F (416–427 °C), and typically end-of-run peak temperatures range from
800 to 825 °F (427–440 °C). A catalyst cycle can end for other reasons, such as
a predetermined schedule, unacceptable pressure drop, or excessive production of
low-value light gases.
Spent catalysts are fouled with coke. They are either regenerating offsite by careful
combustion of the coke, or sold to catalyst reclamation companies. In-place regen-
eration is now very rare. Catalysts heavily contaminated with trace elements are not
regenerated. It is not unusual to see uncontaminated catalysts restored to 85–95% of
their initial activity.

9.7 Mercaptan Oxidation

Mercaptan oxidation processes, such as the two-step UOP Merox process, convert
foul-smelling mercaptans into disulfides. For Merox, the overall reaction is as fol-
lows:

4 R - SH + O2 → 2 R - S - S - R + 2 H2 O

In the equation, R represents an alkyl group, such as methyl (–CH3 ), ethyl

(–CH2 CH3 ), and so on. Methyl mercaptan is CH3 SH. Feed mercaptans can include
methyl mercaptan (methane thiol), ethyl mercaptan (ethanethiol), 1-propyl mercap-
tan, 2-propyl mercaptan, t-butyl mercaptan, pentyl mercaptan, etc. C6 + mercaptans
end up in heavy naphtha cuts and are moved, together with other sulfur compounds,
with conventional hydrotreating. The alkyl groups in the disulfide product can be
different. For example, one of the products could be methylethyldisulfide.
In the conventional Merox process, the mercaptan-containing feedstock is pre-
washed with dilute NaOH to remove hydrogen sulfide (H2 S). The prewashed feed
reacts with a different sodium hydroxide solution, one which contains a proprietary
catalytic additive:

2 R - SH + 2 NaOH → 2 NaSR + 2 H2 O

The next step regenerates the sodium hydroxide while producing a disulfide:
186 9 Crude Storage, Blending, Desalting, Distillation and Treating

Fig. 9.21 Flow diagram of MEROX process

4 NaSR + 2 O2 + H2 O → 2 R - S - S - R + 4 NaOH

In the flow diagram, shown in Fig. 9.21, the mercaptan-containing feedstock enters
the prewash vessel, which removes H2 S. A coalescer prevents caustic from being
carried out of the vessel. The feed then enters the extractor, flowing up through several
contact trays, which enhance the mixing of feedstock with caustic. The feed then goes
to a settler, in which the gas separates from the caustic. The gas is water-washed to
remove any residual caustic, and sent through a bed of rock salt to remove entrained
water. The dry, sweet LPG exits the Merox unit. The hydrocarbon-rich caustic from
the bottom of the extractor is mixed with a proprietary liquid catalyst, heated in
an exchanger, then mixed with compressed air. The mixture goes to the oxidizer,
where the extracted mercaptans are converted to into disulfides. The caustic/disulfide
mixture flows to a separator, in which the mixture separates into a disulfide layer
on top of an aqueous layer of “lean” caustic. The disulfides go to storage or to a
hydrotreater. The caustic is recycled back to the extractor.
References 187

References

1. Foxboro: Crude oil blending. http://www.process-nmr.com/fox-app/pdf/crude-blend.pdf.

Retrieved 6 Aug 2016
2. Hsu CS, Lobodin V, Rodgers RP, McKenna AM, Marshall AG (2011) Compositional boundaries
for fossil hydrocarbons. Energ Fuels 25:2174–2178
3. Park J-I, Mochida I, Marafi AMJ, Al-Mutairi A (2017) Modern approaches to hydrotreating
catalyst. Chapter 21 In: Hsu S, Robinson PR (eds) Springer handbook of petroleum technology.
Springer, New York
4. Leffler WL (2008) Petroleum refining in nontechnical language, 4th Edition, PennWell