Originally appeared in: Clean fuels SpecialReport

February 2010, pgs 35-38.

Used with permission.

Consider new methods for bottom

of the barrel processing—Part 1
Advanced methods use molecule management to upgrade heavy ends
M. Motaghi, KBR, Houston, Texas; and K. Shree and S. Krishnamurthy, KBR,
New Delhi, India

L
arge price differentials between light, sweet crudes vs. heavy, to carbon dioxide cap-and-trade and carbon footprint limitations.
sour crudes have created strong incentives for refiners to lower As the world moves towards cleaner bunker fuels, finding alterna-
costs by incorporating as much heavy crude blends into the tive means to upgrade bottom of the barrel streams will become
refinery processing in scheme as it can tolerate. Several refineries increasingly important.
have little or no bottoms processing capabilities, thus yielding large In parallel, the demand for jet fuel and diesel is expected to
volumes of high-sulfur fuel oil (FO). New bunker-fuel legislation grow, and many refiners are actively focused on shifting demand
and pending carbon-footprint initiatives create the need to further from motor gasoline to diesel, while still analyzing all available
upgrade refinery resid products, both for expansion and enhancement options. The price differential between diesel and gasoline is likely
reasons. Several case histories will cover the staggered investment to be sustained over the long-term and is validated by the billions
options to produce premium road asphalt and solid fuels (Fig. 1). of dollars of investment announcements by the major interna-
tional and national oil companies toward dieselization. In markets
Changing crude diet. Recent economic and geopolitical global dominated by fluid catalytic cracking (FCC)-based refineries, this
uncertainties are factors in the steep rise in crude prices, thus affect- need to increase distillate production has taken on a new dimen-
ing refinery operations worldwide. Refiners are finding themselves sion that will impact long-term refining margins.
faced with the difficult quest to search for crude blends to maxi-
mize margins. While refinery margins will continue to be dictated Changes in C/H ratios. In its simplest form, refining is
by processing heavier, more sour crudes, the dramatic increase in the process of changing the carbon-to-hydrogen (C/H) ratio of
residuum content from 10% in light sweet crudes to 50% in extra naturally occurring crude oils. Thus, at the molecular level, the
heavy crudes poses interesting challenges, while presenting some operation of all 650 refineries in the world is essentially targeted at
unique opportunities. The bulk of the global operating refineries converting high C/H ratio feedstocks into high hydrogen to car-
have little or no residuum processing capabilities and produce large bon ratio for transportation fuels. This ratio change between the
volumes of high-sulfur FO (HSFO) and bunker fuel. crudes and products can only be accomplished through two broad
As demand shifts to natural gas, FO demand is expected to processing routes: carbon rejection and hydrogen addition.
drop adversely, thus affecting FO prices in the future (Fig. 2). This While profitability can only be sustained by economically con-
situation is only expected to worsen as refiners face regulatory verting the large volumes of residuum into high-value transport
pressures ranging from new maritime bunker fuel specifications fuels, these objectives must be accomplished in a difficult business
climate that mandates applying low-investment solutions while
optimizing usage of existing refinery resources.

0
FO crack margin, $/bbl

-10

-20
2000 2004 2008
Source: Singapore margins w.r.t. Dubai crude from Platts

Fig. 1 The residuum oil supercritical extraction unit at the Fig. 2 FO cracking margins 2000 to 2008.
Navaho Refinery.
SpecialReport Clean fuels

Carbon rejection. Carbon rejection is favored by low crude effect for immediate widespread adaptation.
prices and high hydrogen prices, when it is economical to reject the
residuum as petroleum coke, while producing the required trans- Principles of molecule management. In the background
port fuel volumes by incremental crude oil processing. During these of high volatility within the markets, which is expected to con-
processes, the bulk of feed contaminants are rejected with the car- tinue, the optimal solution may require most refiners to adopt a
bon into the coke, rather than into the liquid products. Traditional combination of carbon rejection and hydrogen addition processes.
carbon rejection techniques include FCC, resid fluidized catalytic This is especially true in an environment of uncertain refining mar-
cracking (RFCC), delayed coking (DC) and visbreaking (VB). gins where the size of capital investment can come under a higher
FCC and RFCC in combination are widely used carbon rejec- scrutiny over traditional project “return on investment” criteria. It
tion technologies to convert the high-boiling, high-molecular weight is in this context that a staggered investment option involving the
hydrocarbon fractions into more valuable gasoline, olefinic gases ability to achieve partial benefits at lower initial investments, while
and other products. However, due to the nature of the process, they preserving options for incremental benefits with higher investment
are limited in processing lighter, low-metals, low-sulfur residues. in the future, gains increasing importance.
Visbreakers are essentially a means of improving viscosity of the Refiners are often constrained by the need to convert a defined
residuum so as to minimize the addition of valuable distillate boiling crude slate to a set product slate without realizing the change required
range cutter stock to meet FO specifications. In addition, with the at the molecule level and the cost associated with such conversion.
increasing shift in product demand from motor gasoline to diesel Refiners can also be blinded by the compelling need to produce tra-
and jet fuels, these technologies will no longer be the first choice. ditional refinery products such as transport fuels and petrochemical
Residues from heavy crude oils contain high concentrations of feedstock from every barrel of crude, often expending substantial
sulfur, complex hydrocarbons and heavy metals such as nickel and capital in the process, while ignoring economic synergistic opportuni-
vanadium. Due to the nature of these residues, delayed coking ties that may exist with other nontraditional industrial applications.
technology is the most commonly used carbon-rejection technology. The principles of molecule management dictate that the best eco-
However, this process produces highly hydrogen-deficient, unstable nomics are derived by capturing the highest value of every molecule
products that require further processing and yields coke residue—a present in naturally occurring crude oils at every point in the process.
high C/H ratio molecule, as a byproduct. When viewed in this context, it is evident that it will be prudent to
Although, coking is a mature technology with low implemen- analyze the residuum fraction not only by the traditional barometers
tation risks, in a carbon footprint future, this technology is likely of boiling range and gravity, but by molecular speciation.
to face stiff environmental and regulatory resistance even in the While distillation based separation schemes for the virgin
face of lower crude prices. High-sulfur petroleum coke prices are crude fractions are economical and adopted almost universally, in
distressed and as is evident in the Canadian inland environment, almost all cases, the volume and quality of the residuum is essen-
coke is just being piled up in large quantities with no real eco- tially determined by the quality of the vacuum gasoil (VGO) frac-
nomic outlet. This trend cannot be sustained in the long run. tion and the ability to process this fraction through conventional
hydroprocessing or catalytic cracking conversion units.
Hydrogen addition. Conversely, hydrogen addition is favored In most cases, the limiting factor is the metals content or the
by high crude prices and low hydrogen prices when it is more Conradson Carbon Residue (CCR) in the GO. The residuum vol-
economical to upgrade the residuum to transport fuels, while ume and quality is, by balance, a reject defined by GO quality, and
maximizing the transport fuels production from the base crude is characterized as black oil. By conventional wisdom, this stream is
capacity. Thus far, residue-hydrogen addition technologies have either removed as FO or asphalt, or is subject to thermal conversion
focused on fixed-bed hydrocracking as against ebullated-bed processes for upgrading.
hydrocracking and slurry-phase hydrocracking, the former requir- While it is a well-established fact that hydro or catalytic conver-
ing periodic shutdowns to regenerate catalyst. sion of the heavy gasoil (HGO) fractions will result in substan-
Ebullated-bed processes are continuous and produce higher tially better yields and qualities of transport fuels (gasoline, jet fuel
levels of liquid fuels (no coke). But they are unable to achieve com- and diesel) than thermal conversion processes, and the incentive to
plete resid conversion and still produce 20%–30% of heavy-resid maximize this fraction of the crude exists, operating economics are
product. Ebullated beds have also been prone to high operating substantially influenced by the incremental concentration of impuri-
costs, and have sometimes been plagued with poor operability. The ties (metals and CCR) in the feed due to their impact on conversion
quality of liquid products, although improved over coking, still unit catalysts.
requires secondary processing to produce clean fuels. The inability When analyzing the residuum at a molecular level, it will be
to achieve near complete conversion requires further processing evident that a substantial volume of higher boiling range white oil
of unconverted resid. As a result, ebullated-bed technologies have molecules worthy of effective catalytic upgrading are present in
not achieved huge deployment, which, when coupled with the this fraction, which by conventional methods, are rejected as black
high capital cost, makes them the least robust at low-oil price oil products or subjected to thermal conversion processes. This
scenarios. phenomenon is essentially caused by limitations in distillation-
Slurry-phase hydrocracking. The recent flurry of activities based separation processes where the lowest boiling point species
indicates the advent of slurry-phase hydrocracking into the market of the undesirable impurity is the determining factor in the vol-
place. This technology adopts high operating pressures and can umes of the GO and residuum derived. This impacts the refinery
achieve near complete conversion of the residuum while produc- product slate and economics negatively.
ing finished saleable products. While the projected economic It is, therefore, essential to look at supplemental alternate
conditions of high crude prices and low gas prices are ideal for separation technologies to effectively extract these higher boiling
investment in these high-pressure/high conversion technologies, white oil molecules from the residuum. The options become obvi-
the capital investment requirements may have some dampening ous when analyzing these molecular species, and it is clear that
HYDROCARBON PROCESSING february 2010
 Clean fuels SpecialReport

these undesirable impurities are essentially asphaltenic or resinic

in nature and can be separated by solubility driven processes. Easier to process downstream

White oil molecule management. The solution involves

applying the solubility-based physical separation process, solvent
deasphalting (SDA), wherein the saturates can be effectively separated

Bottom-of-the-barrel
from the resinic and asphaltenic molecules contained in the residuum.
Feed
SDA uses a paraffinic solvent, which by molecular structure (like dis- DAO
solves like) preferentially dissolves paraffinic and naphthenic molecules
while rejecting the aromatic-rich molecules in the pitch (Fig. 3).
Although light-paraffin solvent-based deasphalting is often referred
to as a metals or CCR rejection process, in essence, it is an aromatics- Asphaltene
rejection technology. The reject contains complex aromatic molecules
that are the least soluble in paraffinic solvents, are highly hydrogen
deficient and contain a majority of polars (metals and CCR) that are Asphaltenes w/CCR, metals,
least desirable when fed to a hydro or catalytic conversion unit. nitrogen, etc.
Conversely, the extract or the deasphalted oil (DAO) contains
essentially the saturates, with very low metals and CCR, making Fig. 3 Solvent de-asphalting process and product streams.
this a white oil ideal for conversion processes. Also, the top bar-
rel of DAO derived from the residue via solvent extraction by all
Table 1. Watson K of FCC feed fractions
measures, will always represent a superior conversion feedstock
to any secondary unit when compared to the last barrel of VGO Feedstock Atmospheric Propane Butane
derived from a distillation process. source resid VGO DAO DAO Coker GO
When examining the suitability of feedstocks to a conversion Arabian light 11.60 11.68 11.81 11.74 11.4
process, the inherent molecular content (as opposed to boiling Arabian heavy 11.44 11.62 11.86 11.78 11.4
range) and their impact on conversion unit performance will
become obvious. For example, a close examination of the Watson Table 2. FCC yields from Mid-Continent VGO/DAO
K factors of feeds derived from GO fractions and resids from and CGO
staple Arabian crudes to an FCC unit will reveal that the DAO
derived from a solvent extraction process brings in resid boiling 100% DAO 100% VGO 100% CGO
range white oil molecules that are not otherwise achievable by API 19.2 24.7 19.0
conventional distillation based processes (Table 1). S, wt% 0.79 0.75
The DAO, rich in paraffins and saturates, is an excellent feed CCR, wt% 3.9 0.39 Less than 1
to the FCC. Also, due to the inherent nature of the aromatic rejec-
Ni + V, PPM 16 1 Less than 1
tion associated with solvent-based separation processes, the DAO
will always have a higher Watson K than the corresponding VGO FCC yields, wt%
from the same crude. Conversion 80.3 81.05 63.2
As listed in Table 2, the FCC yields and product qualities derived C2– 4.86 3.65 1.49
from a DAO based feed (Mid Continent) is very similar to the Total C3’s 6.37 6.80 4.60
VGO yields from the same crude in spite of processing heavier Total C4’s 10.30 11.76 8.87
molecules derived from the resid boiling range. The alternate option
Total gasoline 48.98 52.12 40.16
would be to reject these valuable resid white oil molecules to lower
value black oil products, or to process through a lower yield thermal Total cycle oil 19.70 18.95 35.78
conversion process while reprocessing the derived coker GOs in Coke 9.79 6.72 6.00
the FCC. The understanding of the value of these high hydrogen
content resid molecules and their disposition is a critical data point
in determining refinery yields and economics. exposure. While the temptation to use CCR as a parameter to
An understanding of the metals and CCR content in the DAO establish conversion unit performance exists, a molecular level
stream and their impact on conversion unit catalysis is critical to examination will highlight significant differences. In short, all
determining the extractable volume of white oil molecules con- CCRs are not the same. While the heptane insolubles have an
tained in the residuum. While metals are inherently addressed almost direct permanent deactivation effect on downstream pro-
by demetalization catalysis in a hydrocracker, or by FCC catalyst cess catalysts, the other CCR molecules have a far less telling
consumption, and the economics are easy to calculate, the under- effect.
standing of the CCR is more complicated (Table 2). The CCR derived from a distillation-based separation pro-
Notice that the CCR content of DAO is nearly 10 times higher cess will contain substantial C7 insolubles, while that in a DAO
than the VGO, yet the yields and product qualities and coke make derived from a solvent extraction, by definition, should be non-
are still comparable. The obvious inference is that the CCR in detectable. This distinction can be clearly demonstrated by operat-
this case does not impact the conversion simply by virtue of the ing performance differences between units processing atmospheric
molecules that contribute to it. or vacuum resids and those processing DAOs with identical CCR
In its simplest form, CCR is the residue derived from a test content. This difference is often not well understood by refiners
wherein a heavy hydrocarbon is subject to a temperature-time while correlating the impact of CCR on their unit performance.
HYDROCARBON PROCESSING february 2010
SpecialReport Clean fuels

Table 3. Commercial data for conversion Table 3 lists commercial data published by a US refiner that
shows the before and after conversion of a non-supercritical SDA
After conversion to residuum unit to an advanced residuum oil supercritical extraction technology
Supercritical SDA oil supercritical extraction
with advanced internals. As evident from the data, the addition of
Feed rate, bpsd 7,000 10,000 (up to 15,000) new internals and adoption of residuum oil supercritical extraction
Solvent ratio vol/vol 4.5–6 5–6 methods achieved a higher throughput, higher yield and a better
Energy MMBtu/bbl 99 69 DAO product quality. All this was achieved at similar operating
Deasphalted oil quality conditions with significantly lower specific energy consumption.
Yield, LV% 65–75 70–85 Note, the decrease in asphaltene content of the DAO to nearly
below detectable limits, conforming the assertion that the presence
Asphaltene, (C7 insols) ppmw 200–800 < 25
of C7 insolubles in the DAO is an artifact of entrainment that can
CCR, wt% 12–13 9–11
be controlled by good technology and internal design features. HP
High pressure vessels
Extractor With trays With advanced intervals Next month. In Part 2, the authors discuss new supercritical
DAO separator With mesh pad With advanced intervals extraction methods that can be applied to optimize molecule
management of residuum.

While the DAO must theoretically contain no C7 insolubles, Mitra Motaghi is an associate with the KBR refining technology business unit
an examination of the DAO derived from conventional SDA in Houston Texas, with specific focus on resid and hydroprocessing technologies. She
processes would often indicate levels ranging from 300–1,000 holds an MS degree in chemical engineering from Texas A&M, Kingsville, Texas.
ppmw. This can only be explained by the phenomenon of entrain-
ment. With the advent of state of the art structured-packing based Kanu Shree is an associate with the KBR refining technology business unit in
internals into the solvent deasphalter separator vessels in 1995, the New Delhi, India, with specific focus on resid and hydroprocessing technologies. She
level of C7 insolubles is now controlled at or below 100 ppmw holds a BS degree in chemical engineering from the Indian Institute of Technology,
in most deasphalting units and below 25 ppmw in several high New Delhi, India.

performing units. As an added benefit, the process unit can now

operate at about twice the phase rates of conventional separators Sujatha Krishnamurthy is an associate with the KBR refining technology
business unit in New Delhi, India, with specific focus on resid and hydroprocessing
and provides about twice the mass transfer efficiency of conven- technologies. She holds a BS degree in chemical engineering from Anna University,
tional extraction contacting devices. Chennai, India.

Article copyright © 2010 by Gulf Publishing Company. All rights reserved. Printed in U.S.A.
Not to be distributed in electronic or printed form, or posted on a Website, without express written permission of copyright holder.
Originally appeared in: Clean fuels SpecialReport
March 2010, pgs 55-58.
Used with permission.

Consider new methods for bottom

of the barrel processing—Part 2
New principles of molecule management dictate
the best economics when upgrading residuum
M. Motaghi, KBR, Houston, Texas; and K. Shree and S. Krishnamurthy, KBR,
New Delhi, India

G
lobally, refiners are often constrained by the need to con- lift of the white molecules from the residuum to be very low, mak-
vert a defined crude feed slate into a desired finished prod- ing this entire solution uneconomical.
uct slate without realizing the reaction changes required at Road asphalt is often driven by seasonal demand changes, and
the molecule level and associated costs for such conversions. Too refiners address this through direct air blowing of the vacuum
often, refiners can be blinded by the compelling need to produce resids (VRs). While air blowing is a solution of convenience, when
traditional refinery products such as transportation fuels and viewed through the prism of molecule management, it involves
petrochemical feedstocks from every barrel of crude. Such actions the downgrading of valuable white oil molecules contained in the
involve expending substantial capital in the process, while ignor- VR to a black oil product—an economic negative, not withstand-
ing economic synergistic opportunities that may exist with other ing the implications of the associated environmental control issues
nontraditional processing applications. For example, cost-effective that come with it.
upgrading of the bottom of the barrel can involve applying new An interesting refinery hydrogen balance can be established
supercritical extraction technologies that can manage the “black” through gasification of the pitch. The liquid feed system low-
molecules or residuum oil and produce end products other than ers the net capital investment requirements, although it can
transportation fuels. still be very high when viewed in absolute terms. In addition,
commercial experience suggests a substantial improvement in
Black oil molecule management. In Part 1, the merits reliability—in some cases, twice as that of traditional solid-fuel
of DAO as a valuable feedstock to conversion units were pre- based gasifiers. However, with the projection of depressed natural
sented. However, the limited outlets for the effective disposition gas prices, a cheaper hydrogen source availability in the long-
of the pitch has historically had some dampening effects on the term, refiners may find it difficult to justify large investments in
widespread acceptance of this molecular solution. The com- gasification technologies.
mercially practiced solutions, so far, involve using the pitch as
a high-sulfur FO (HSFO) blending component, a road-asphalt Pitch molecule management. Against this background,
blending component, thermal or delayed coke feedstock, or as a eliminating the pitch at the molecular level will be required to
liquid feed to a gasifier. explore all available options to maximize the benefits of the solu-
Although the blend out to FO will lower the refinery HSFO bility driven process. The pitch consists of essentially asphaltenic
production, with the diminishing outlets for FO, the ultimate goal and resinic molecules. For this exercise, resinic molecules are
of a refiner to eliminate this product is not achieved. In addition,
using high-value distillate boiling range cutter stock will have a
negative impact on refinery economics. C84 H98 N2S2O3 O
As an alternate option, the most obvious solution would be to O O

direct these black oil molecules to a delayed coker. As the CCR 1248 MW H S
40.4 % Aromatic carbons N
content of the pitch is essentially a product of a concentration 80.85 wt % C
S
effect, its inherent limitation in the delayed coker feed and its 7.92 wt % H N
2.24 wt % N
impact on coker furnace run length must be considered. In com- 5.14 wt % S
mercial applications, this parameter is closely monitored and is 3.85 wt % O
often limited to about 34 wt%, with the option to go up to 38 wt%
in a grassroots design. This, in turn, means that limitations in coker
feed CCR can sometimes be the controlling factor in the amount Asphaltene materials with complex structures
Precipitated from crude oils by aliphatic solvents. Soluble in benzene.
of white oil molecules that can be lifted from the residuum. Mol. wt. 1,000–3,000. High in S, N, O and metals ( V + Ni).
In many cases, the asphaltene molecules directly produced
from a traditional SDA process are too hard to be sold as road Fig. 4 Asphaltene materials with complex structures.
asphalt. The production of road asphalt would require the overall
SpecialReport Clean fuels

those composed of aromatic rings with long side chains, while

Cooler Exchanger asphaltenic molecules are complex aromatic compounds with
relatively shorter side chains (Fig. 4). This observation leads to the
Resin obvious questions associated with the ability to separate the pitch
heater into two distinct molecular fractions and the refinery benefits that
can be derived by finding economic outlets for these fractions.
Steam Resin
Resin separator
stripper Exchanger New supercritical extraction methods. In a solubility
driven process, the type of molecular structure is a critical param-
Resin
Condenser eter in determining the distribution coefficient of that molecule
DAO between the two liquid phases. Therefore, the expectation of sepa-
separator
Mixer Asphaltene Solvent ration between structure-based resinic and asphaltenic molecules
START separator surge in this process will not be unreasonable (Fig. 5).
Resid DAO In reality, this is practiced in three operating commercial
Asphaltene DAO heater
heater stripper residuum oil supercritical extraction units where the operating
Solvent conditions of the asphaltene separator are adjusted to lift the
Solvent recycle
circulation Steam resinic molecules in the DAO. The resinic molecules are then
Steam Asphaltene recovered from the DAO by partially expanding the solvent under
stripper
supercritical conditions.
Asphaltene DAO
The intent is to fractionate the DAO exiting the asphaltene
Fig. 5 Flow diagram of a residuum oil supercritical extraction separator (that contains saturates and resins) into a light DAO
process. (predominantly saturates) and a resin product. This is accom-
plished by adjusting the operating conditions of the resin sepa-
rator so that sufficient expansion of the solvent is realized to
selectively drop out the heavy part of the oil, resulting in solubility
based separation at supercritical conditions. The yield and quality
of the resin may be varied to influence the quality and quantity
of the DAO.
This arrangement will also provide the flexibility to balance the
streams to the downstream processing needs, while consistently
meeting the required DAO quality and exercising other disposi-
tion options for the intermediate resin stream.

Resins and road asphalt. Examining the properties of the

resin fraction will reveal that this stream can now manifest into
multiple product outlets. One economic outlet is the direct road
asphalt production—solely from black oil molecules without
further processing. The resin molecules are lighter than road
Fig. 6 Solid asphaltene pellets are easier to transport and store.
asphalt and the required grade of road asphalt can be produced
by selectively blending the asphaltenic molecules in the required
proportion. Due to the dewaxed nature of
these molecules, they exhibit excellent duc-
Proprietary tility properties at the same penetration not
high-capacity otherwise achievable by traditional technol-
feed distributor ogy options.
Pelletizer Inherent in the process is the ability to
vessel adjust the volume and quality of the resin.
Surge Seasonal demand variations in road asphalt
drum Trim heat Cooling water may be addressed by adjusting the resin
exchanger volumes derived and by selectively direct-
Pellets
Hot liquid and water ing the excess to either the DAO or the
asphaltene asphaltene fraction.

Feed pump Asphaltenic rejects. While this solu-

tion provides an attractive economic option
Solid pellets Conveyor belt Screen for Water to produce premium-grade road asphalt, the
to storage for pellet dewatering dump Solids Cooling Fin-fan or
filter water water cooler
challenges associated with dealing with the
transportation drying
pump residual asphaltenic molecules remain.
The asphaltenic molecules are high in
Fig. 7 Flow diagram of asphaltene solid pelletizing unit. CCR, metals and sulfur content and are
basically not worthy of upgrading. This
HYDROCARBON PROCESSING march 2010
 Clean fuels SpecialReport

represents the lowest-value true black oil content of the residues, tion characteristics. The heating value, organic carbon, and chemi-
and the best economics lies in solutions that divert these molecules cal properties such as sulfur, nickel and vanadium are governed
away from the refinery to industries or end users outside the refin- by the crude properties. The asphaltene pellets have 20%–50%
ing business that have an incentive to process these streams. higher heating value than petroleum coke. In view of the superior
The major challenge here is in the handling and transporta- heating value, combustion characteristics and ease of grinding, the
tion of these molecules. The asphaltene product is produced as asphaltene pellets should demand a higher value per ton when
a high-viscosity liquid that solidifies at ambient temperature. A compared to fuel-grade coke and coal.
low-cost, high-capacity solid-pelletization technology is the obvi-
ous transport solution and will help refiners to economically store Refinery case illustration. In this Base Case, a 200,000
and move these rejects to a more desirable end use (Fig. 6). bpsd refinery processes heavy, sour, Middle Eastern crude (API
Solid fuel. This represents a simple and cost-effective option 31.4) producing 6,230 tpd of VR and 8,847 tpd VGO (Fig. 8).
for asphaltene disposition for the cement and steel market where The refinery has no bottoms processing units and is currently
a large demand for high Btu solid fuel exists. using high-value distillates to cut its VRs to produce large volumes
There are existing commercial technologies that produce solid of low-value HSFO (Tables 4 and 5).
fuel from the asphaltene rejects. However, these processes are Interim investment option. In this interim option, the
generally capacity limited, high in maintenance, low in reliability, refiner invests in a residuum oil supercritical extraction unit and
and are manpower intensive. A new solid-pelletizing technology asphaltene solid pelletizing unit, essentially eliminating the pro-
is an ideal solution to solidify asphaltenes and other heavy hydro- duction of HSFO. The investment in the ISBL portion of the two
carbons. This method is a low-cost process, easy to operate, and units will amount to about US$ 65–75 million (Fig. 9).
has a high expected onstream factor. This combination represents the most ideal short-term solu-
The produced pellets are resistant to dusting and can be easily tion to allow all the DAO produced to be processed in the existing
handled, stored, and transported. These pellets are near spherical FCC units offsetting the equivalent VGO that may be sold. The
with an expected size distribution between 1 mm and 3 mm, and change in FCC feed will have a small effect on the FCC yields.
they have good grindability, storage and transportation charac- The lower the DAO make, the better quality of the DAO, but
teristics as indicated by the high Hargrove Grindability Index a higher, lighter resin volume must be handled. The higher vol-
(HGI), storage test temperature and low friability. The high angle ume, lighter resin will lead to a higher road asphalt volume make,
of repose provides high capacity on conveyors. The small amount and a resultant lower solid fuel production (Table 6).
of residual moisture on the pellets helps to minimize dust forma- Final investment option. In this ultimate option, the refiner
tion during transport (Fig. 7). invests in a product residuum oil supercritical extraction unit and
The asphaltene pellets can be used as solid fuel in the cement
kilns, steel industry and in utility industries. The pellets can be
added to fuel-grade coke or coal as additive to enhance combus- Vacuum distillation Cutter
3,581 tpd
Table 4. Interim solution: unit yields, (resin yield 20%,
DAO yield 30%, asphaltene yield 50 wt%)
VR 6,230 tpd FO 9,811 tpd
Feed VR Asphaltene Products resin DAO
Yield on VR, tpd 6,230 3,115 1,246 1869 Base Case refinery
S.G.@ 60°F 1.033 1.112 0.984 0.952
Fig. 8 Base Case refinery processes heavy crude with no bottoms
Nitrogen, wt% 0.4 0.6 0.3 0.2 processing capability.
Sulfur, wt% 5.5 7 4.5 3.7
CCR, wt% 24 40 12.5 5
Nickel, wppm 29 56 4 0.9 Export VGO
3,847 tpd
Vanadium, wppm 110 216 7.4 1.7
VGO VGO FCC feed
R&B softening pt, °F 250 8,847 tpd 5,000 tpd 6,869 tpd

Table 5. Overall material balance, interim solution DAO

Vacuum 1,869 tpd
distillation
FCC Resin
Virgin Road Solid feed VGO 1,246 tpd
VR VGO asphalt fuel blend export VR 6,230
tpd Residuum oil
Yield on VR, tpd 6,230 8,847 1,908 2,453 6,869 3,847 supercritical
extraction
S.G.@ 60°F 1.033 0.922 1.025 1.112 0.930 0.922
Nitrogen, wt% 0.4 0.1 0.4 0.6 0.1 0.1 Solid fuel Asphaltenes Asphaltenes Road asphalt
2,453 tpd Pelletizing 2,453 tpd 662 tpd 1,908 tpd
Sulfur, wt% 5.5 3.3 5.3 7 3.4 3.3 technology
CCR, wt% 24 0.9 22.1 40 2 0.9 Interim solution
Nickel, wppm 29 0.1 21.8 56 0.3 0.1
Fig. 9 Base Case refinery processes heavy crude with residuum
Vanadium, wppm 110 0.4 79.8 216 0.7 0.4
oil supercritical extraction unit and solid asphaltene
R&B softening pt, °F 250 pelletizing unit.

HYDROCARBON PROCESSING march 2010

SpecialReport Clean fuels

Table 6. Overall material balance, final solution

VGO Hydrocracker feed
8,847 tpd 10,716 tpd Virgin Road Solid HC
VR VGO asphalt fuel feed
Yield on VR, tpd 6,230 8,847 1,908 2,453 10,716
DAO
Vacuum 1,869 tpd S.G.@ 60°F 1.033 0.922 1.025 1.112 0.928
distillation
Resin Nitrogen, wt% 0.4 0.1 0.4 0.6 0.1
VR 6,230 1,246 tpd Sulfur, wt% 5.5 3.3 5.3 7 3.3
tpd Residuum oil
supercritical CCR, wt% 24 0.9 22.1 40 1.6
extraction Nickel, wppm 29 0.1 21.8 56 0.3
Solid fuel Asphaltenes Asphaltenes Road asphalt Vanadium, wppm 110 0.4 79.8 216 0.6
2,453 tpd Pelletizing 2,453 tpd 662 tpd 1,908 tpd R&B softening pt, °F 250
technology
Final solution
Table 7. Hydrocracker yields
Fig. 10 Base Case refinery processes heavy crude with residuum
oil supercritical extraction unit and solid asphaltene Case Low conversion High conversion
pelletizing unit with hydrocracking unit to upgrade extra Pressure, psig 1,960 1,960
DAO product.
H2 circulation, scf/bbl 6,000 6,000
Temperature °F Base Base +25
an asphaltene solid pelletizing unit, essentially eliminating HSFO 650°F+ conversion 15 53
production. The higher volume of conversion feedstock available
H2 consumption 825 1,530
will have to be processed in a hydroprocessing unit.
The required conversion can vary from a nominal 20% that Yields, vol%
is associated with a traditional catfeed hydrotreater to a nominal Naphtha, C5-180°F 0.2 7.2
65% conversion achievable in a once-through conversion unit. Kerosine, 180°F–330°F 1.8 17
The design and operating pressure of the hydrocracker can Diesel, 330°F–650°F 14.6 39.9
be adjusted to vary the conversion from 15%–55% by varying FCC feed 85.2 47.5
the operating conditions of the hydrocracker beds. This, in turn,
will provide the refiner with the additional flexibility to vary the
gasoline-to-diesel ratio derived from the two conversion units, appropriate end user. The optimal solution will involve adopting
while eliminating the production of fuel oil altogether. The unique means of integrating the principles of carbon rejection and
cost of this investment for a large hydrocracker can vary from hydrogen addition while maintaining a keen recognition of the
$400–600 million and can be differed until economic conditions prevalent investment climate.
warrant (Fig. 10). The combination of SDA and asphaltene pelletizer technolo-
In the low-conversion mode, the FCC will be at its maximum gies represents an economic low-cost solution to eliminate FO
capacity, and the yields and quality are expected to be better than production and can be implemented at a fraction of the cost of all
current operations (Table 7). In the high conversion mode, the other resid processing options. The DAO is an excellent feedstock
FCC will be at low throughput and out of heat balance, prompt- and can be easily processed in the refinery FCC or hydrocracker
ing the need for a small bypass DAO or resin feed to be sent units. Premium road asphalt can be derived from the pitch, and
directly to the FCC. the reject solid asphaltene pellets can be sold to the cement, steel
and power industries. HP
Molecule management summary. When adopting the
principles of molecule management, it is clear that the best value End of series: Part 1. February 2010.
is derived by directing the selected residuum molecule to the

www.kbr.com/refining

Article copyright © 2010 by Gulf Publishing Company. All rights reserved. Printed in U.S.A.
Not to be distributed in electronic or printed form, or posted on a Website, without express written permission of copyright holder.