Ben Ziesmer, Jacobs Consultancy
The recession caused demand for petroleum coke (petcoke) in
traditional markets to plummet. Asia, especially China and India,
stepped in to fill the void. While Japan has been the largest
single destination for US petcoke exports for over a decade,
recent trends indicate that the remainder of Asia is likely to be a
significant market for seaborne petcoke.
Petcoke is produced as a by-product in many, though not a
majority of, oil refineries. Traditionally, cokers are installed in oil
refineries to convert vacuum tower bottoms (residuum) and
other heavy residual oils into higher-value light transportation
products (e.g., gasoline, jet fuel, and distillate). A coker almost
invariably increases refinery profitability because the yield of
high-value transportation fuels is maximized and production of
low- value residual fuel oil (RFO) is minimized. While the coking
process has been in use since the 1930s, petcoke production has
seen its largest growth since 1990 because worldwide light
petroleum products demand has grown faster than residual fuel
oil demand. Cokers have been and continue to be the preferred
refining technology that allows the refining industry to reduce its
production of residual fuel oil per barrel of crude oil processed
and bridge the gap between rising light product demand and
stagnant RFO demand.
Additionally, beginning in the late 1990s, two new factors have
been driving construction of cokers:
  reducing crude oil purchase cost — coking units allow a
refinery to process lower cost, heavy, sour crude oils. This was
the driving force for the nine new or expanded cokers installed
on the US Gulf Coast over the 1996–2004 period, and continues
to be the driving force for current US coking capacity additions
(new or expansions) from 2008–2012. The installation of many
of these units has been supported by long-term crude oil supply
  Increasing crude oil production — cokers are used in
upgraders that produce various grades of synthetic crude oil
(SCO) from bitumen or ultra-heavy crude oils. This type of
upgrader exists in Venezuela where ultra-heavy Orinoco Belt
crude oil is upgraded and exported as lighter crude oils, and in
Canada where upgraders are used to produce SCO from the oil
derived from Alberta oil sands.
There are two general applications for petcoke: one as a
carbon source and the other as a heat source. The former
requires better quality (i.e., low sulphur and metals) and
commands higher prices. Petcoke is usually upgraded by
calcination when it is used as a carbon source. Petcoke that has
been calcined is referred to as calcined petroleum coke (CPC).
The largest market for CPC is in the production of anodes for
aluminium smelting. Other significant uses for CPC are in the
production of carbon electrodes for electric arc furnaces,
titanium dioxide (TiO2), and as a recarburizer in the steel
industry. While there are a variety of value-added markets for
higher-quality petcoke, about 75% of petcoke is sold into the fuel
market, where it competes with coal.
For the first time since we began tracking worldwide petcoke
production, production decreased from approximately 101
million metric tonnes (MT)1 in 2008 to 98 million MT in 2009.
Petcoke production at existing cokers is typically not impacted
by recessions because refineries equipped with cokers are
generally more profitable than those without coker(s). Thus,
refineries without cokers generally reduce their production
much more than refineries with cokers to balance refined
product supply with demand.
However, 2009 and the first quarter of 2010 have proved to
be the time when ‘the exception proves the rule’, as coking
refineries were impacted by the recession the same as noncoking
refineries. Several factors combined to make coking
economically unattractive. First, the recession decreased
worldwide refined product demand. Second, the price spread
between light, sweet and heavy, sour crude oil collapsed, which
caused refiners to migrate to higher-quality crude oils; this
resulted in less production of residual fuel oil (RFO) and/or
coker feedstock. Third, bunker fuel demand remained relatively
stable, driven by China’s strong economic growth. The
combination of less RFO production per barrel of crude oil, less
crude oil throughput, and stable bunker fuel oil demand caused
RFO pricing to be unusually strong compared to light product
pricing. The combination of weak refined product demand —
especially diesel — and relatively strong bunker demand negated
the traditional advantage of coking refineries.
Initially deteriorating coking economics was manifested by
lower operating rates and decreased petcoke production. By
the second quarter the issue of poor coking economics became
much more visible. In June 2009, Valero announced the idling of
its Aruba refinery and Corpus Christi, Texas coker. In November
2009, Valero announced it was permanently closing its Delaware
City refinery. These were the highest profile examples of how
weak refining and coking economics were impacting petcoke
production. Additionally, coking capacity additions did not start
up as soon as projected. For example, Reliance’s 585,000 bbl/day
expansion of it Jamnagar, Gujarat, India refinery, which includes a
very large coker, did not approach full production until the third
quarter instead of the first quarter of 2009.
Petcoke prices have staged a significant recovery since
bottoming in March/April of 2009. Petcoke prices plummeted in
late 2008 and the first quarter of 2009 as demand — especially
demand by the key European cement industry — fell sharply due
to the worldwide recession. Cement demand is particularly
important to the petcoke market because the cement industry is
the largest market for fuel-grade petcoke. Just as despair was
setting in as to where to market petcoke, Indian buyers began,
for the first time, to purchase significant quantities of US Gulf
Coast (USGC) petcoke. Shortly thereafter, China also began to
buy USGC petcoke and increased purchases of US West Coast
(USWC) petcoke. This new demand from Asia put a ‘bottom’ in
petcoke prices.
The economies of the world began to stabilize in the second
quarter of 2009. For example, the US went from –6.4% GDP
growth in the first quarter to –0.7% GDP growth in the second
quarter. China and India saw strong GDP growth of 7.9% and
6.8%, respectively, in the second quarter. This increased
economic activity, especially in China and India, boosted demand
for commodities, and steam coal prices began a gradual rally.
Increased economic activity also led to stronger RFO pricing,
weakening coking economics and less petcoke production. As
2009 progressed, ship loading problems at Venezuelan ports
reduced petcoke availability further.
There are four projects — PetroMonagas (formerly Cerro
Negro), PetroAnzoátegui (formerly Petrozuata), PetroCedeño
(formerly Sincor), and PetroPiar (formerly Hamaca) — that
produce synthetic crude oil (SCO) from super-heavy Orinoco
belt crude oil/bitumen. Each project has an upgrading plant
located in the Port of Jose, where SCO is produced from
Orinoco bitumen. All four upgraders utilize coking technology
and can supply 25% of the USGC/Caribbean petcoke market
seaborne trade. Moreover, Venezuelan petcoke tends to be
lower sulphur (i.e., 4.0–4.5% S, dry basis) material, so Venezuelan
exports are especially important to the lower sulphur portion of
the USGC/Caribbean petcoke market. There are two terminals
located at the Port of Jose, each of which was plagued with slow
loading rates and undependable operations due to breakdowns.
Then the PetroCedeño terminal was shut down for a planned
maintenance outage and overhaul during almost all of the fourth
quarter of 2009 and first quarter of 2010. The net result was
that supply of Venezuelan petcoke to the seaborne market was
severely restricted.
The combination of new demand for petcoke from India and
Japan, less USGC petcoke production, and restricted supply out
of Venezuela drove the strong bull run in petcoke prices that
began in March/April of 2009 and has continued since.
One might think that coal and petcoke prices are
closely correlated because fuel-grade petcoke is typically used as
a substitute for coal. Moreover, coal and petcoke prices
bottomed in March/April of 2009, followed by stronger pricing in
both commodities. While there is an apparent correlation
between steam coal cost and the price of petcoke, petcoke
prices do not move in lock step with coal prices. Petcoke is not
fungible with coal due to its higher sulphur content, inferior
combustion characteristics, different ash characteristics, and
various peculiarities of environmental regulations/permits. If
petcoke prices were entirely determined by coal prices, then the
discount petcoke to coal would be constant (i.e., a straight line);
however, this is not the case2 (see ‘Fuel-Grade Petcoke Discount
vs. Coal Alternative’ chart).
When the fuel-grade petroleum market is weak, then petcoke
prices tend to fall relative to coal and the discount for petcoke
compared to coal increases. One can see that when the
USGC/Caribbean market was at its nadir in March/April of 2009,
the discount relative to coal in the Mediterranean ($/MMBtu
basis) was close to 70%, providing a strong incentive to utilize
petcoke instead of coal. Then, as the petcoke market recovered,
the discount to coal tumbled. Historically, Colombian and South
African delivered coal cost into the Mediterranean have been
very close, but this changed in the later part of 2009 as South
African coal prices increased while Colombian coal prices were
stable. This was because Colombia lost one of its key markets,
the US power market, as power plants in the eastern and
southeastern US switched back to domestic coal.3
In the fourth quarter of 2009, the European power industry
stopped buying petcoke for blending with coal as the petcoke
discount was not large enough to offset increased costs
associated with burning petcoke (e.g., higher limestone
consumption for SO2 scrubbers, increased CO2 emissions cost,
more handling costs). USGC/Caribbean petcoke prices
continued to increase and the petcoke discount vis-à-vis
Colombian coal dropped to less than 10%. This small economic
incentive to burn petcoke led to some European/Mediterranean
market cement kilns purchasing petcoke instead of coal. Then
Colombian producers were able to expand their sales to Asia,
and Colombian coal went from being $20–25/MT to $10–12/MT
less expensive than South African coal. As Colombian coal came
closer in price to South African coal, the discount for using
Colombian coal once again became comparable to South African
coal, and there was less incentive for cement kilns to switch
from petcoke to Colombian coal.
Water is the primary transportation mode for petcoke, given
its need to be transported significant distances to reach
customers, combined with the cost advantages of water
transportation. US and Caribbean cokers4 produce virtually all
of the petcoke that moves by seaborne trade, even though they
accounted for only ~50% of the world’s production in 2009.
This is because petcoke produced in other parts of the world
(such as Europe, India, etc.) is almost always used domestically.
In 2009, weak US petcoke demand caused petcoke exports to
increase by 10% while production decreased by 7 percent.
Consequently, 63% (26+ million metric tonnes of petcoke) of US
petcoke production was exported. Additionally, 60+% (~2.6
million metric tonnes) of US calcined petcoke production was
exported in 2009. Virtually all Caribbean petcoke production is
For decades, Europe was the primary market for USGC
petcoke production and Japan was the primary market for
USWC petcoke production. Then Latin America— especially
Mexico and Brazil—became important markets for USGC
production. Most recently, it has been India and China that have
become significant markets for US petcoke. This migration of
markets is illustrated in the charts comparing the distribution of
US exports in 2001 and in 2009.
Transportation costs become more important as petroleum
moves to more distant markets. For example, ocean freight cost
can equal, or even exceed, the FOB Load Port price of USGC
petcoke into China, India, or other distant locations. As ocean
freight rates increase, it becomes pricier for USGC petcoke to
compete against coal in more distant markets like India or
In the second quarter of 2010 we saw signs that coking is
becoming profitable. Many US cokers are operating at close to
normal rates. The Valero Corpus Christi coker is once again
operating, and Valero is considering re-starting its Aruba refinery
later this year. PBF Energy Partners Company LLC closed on
the purchase of Valero’s Delaware City refinery June 1 and plans
to re-start it, along with its fluid coker, sometime during the
second quarter of next year.
Looking beyond 2010, we expect that new cokers and coker
expansions under construction will be completed. We forecast
worldwide petcoke production will increase by 51% by 2013 .
Several observations can be made by looking at planned coking
capacity additions:
  significant coking capacity additions are proceeding, especially
in Brazil, India, and China, and we expect coking capacity
additions to continue into the foreseeable future.
  significant coking capacity additions currently under
construction in the US will be completed, coming on line in
2011 and 2012. Consequently, Jacobs Consultancy projects that
US petcoke production will increase by 24% by 2013.
  two new 400,000 bbl/day refineries, which include new
delayed cokers in the configurations, are proceeding in Saudi
Arabia to process heavy, sour Arabian crude oil. These two
refineries could well be the harbinger of more export-orientated
refineries equipped with cokers being constructed in the Middle
Chinese and Indian coking capacity additions are driven by
rapidly growing light product demand (gasoline, jet fuel, diesel,
etc.), whereas US coking capacity is driven by refinery upgrades
to handle less expensive, heavy crude oils. As discussed earlier,
coking is the preferred route to convert heavy, tar-like bitumen
into a crude oil that can be processed by the refining industry,
which drives the Canadian coker additions.
Sharply lower crude oil prices in 2008 caused many Alberta
oil sands projects to be delayed or indefinitely deferred.
However, the recent announcement by Total that it is resuming
development work on its Fort Saskatchewan, Alberta-area oil
sands project indicates that interest in Alberta oil sands is
resuming as oil prices move toward $80/bbl. Many Alberta oil
sands projects will blend the bitumen they produce with diluents
such as natural gas liquids (resulting in a crude stream known as
dilbit) or with SCO (producing a crude stream known as synbit)
to produce a blended product that can meet pipeline viscosity
and gravity specifications. The dilbit or synbit will be very heavy,
requiring refineries to have substantial coking capacity to
process the crude oil. This Canadian heavy oil is driving coking
capacity additions in the northern US—BP (Whiting, IL);
ConocoPhillips (Wood River, IL); and Marathon (Detroit, MI).
Even with these substantial coking capacity additions in the
northern portion of the US, the US Gulf Coast will continue to
be the center of US petcoke production, with production
increasing by 25+% by 2013.
Petcoke production grew rapidly from 1994 through 2004, but
production growth slowed markedly (except for China) from
2005 through 2008. The decline in petcoke production in 2009
was unprecedented and we believe an anomaly. We opine
petcoke production will experience rapid growth from 2010 —
2013. Reduced petcoke demand in traditional European,
Canadian, and US markets due to the recession has presented
new opportunities to move North American petcoke into China
and India. However, the continued viability of China and India as
markets for USGC petcoke depends on a number of factors,
including the cost of locally available coal, domestic petcoke
competition, and petcoke ocean freight rates. New coker
construction will be concentrated in the US, Canada, Brazil,
China, and India. Additionally, the Middle East will become a
significant new production area. Water transportation is the
dominant mode used to move petcoke to market, and
transportation costs will continue to be a key driver for petcoke
marketing. Coking capacity additions in the US, Canada, and the
Middle East will likely increase demand for seaborne petcoke
transportation services.
Ben Ziesmer (Senior Consultant, Jacobs Consultancy Inc.)
Contributing editor to Jacobs Consultancy's Pace Petroleum Coke Quarterly, with indepth background
in the power sector, including experience in procurement, operations, environmental
compliance, and engineering. He has been the project manager
for numerous studies involving the fuel-grade petroleum coke
market, environmental issues, and power generation.
Jacobs Consultancy Inc. has been publishing the Pace
Petroleum Coke Quarterly© (PCQ) since 1983. The PCQ has been
published monthly since 1984 and is considered the worldwide
authoritative source for petroleum coke market information.
The author would like to acknowledge the assistance of Ms.
Alisa Allen, Consultant in the Carbon Group of Jacobs
Consultancy, for her assistance.

1. Production was 93 million dry metric tonnes (DMT) ~ 101 million MT @8%moisture (typical value).
2. The Mediterranean cement market has typically been the clearing market for USGC/Caribbean petroleum coke, so the delivered cost of South African coal into
the Mediterranean is of particular importance to the USGC/Caribbean petroleum coke market.
3. Many power plants completed retrofits of SO2 scrubbers and SCR NOx control equipment during 2007–2009 as a result of the Clean Air Interstate Rule (CAIR),
obviating the need to purchase low-sulphur Colombian coal instead of highersulphur, but often lower-cost, domestic coal. The effective compliance date for CAIR
was 1 January 2009 for NOx and 1 January 2010 for SO2.
4. Aruba and Venezuela; Hovensa located in St. Croix, US Virgin Islands is included in the US total.