Fuel Program Feasibility 

Overview of Refining Operations 

  REF _Ref307313482  Figure 4-1  shows a process flow diagram for a
typical complex refinery, capable of making a wide product slate (shown
on the right side of the figure) from crude oil (input on the left). 
Following the figure is a brief description of key units and streams
focusing more on the gasoline producing units.  It’s important to note
that not all refineries have all of these units, which is a key factor
in both the variation in their baseline sulfur levels as well as their
cost of sulfur control.

 

Figure   STYLEREF 1 \s  4 -  SEQ Figure \* ARABIC \s 1  1  Process Flow
Diagram for a Typical Complex Refinery

Crude Tower

The purpose of the crude tower is to perform a distillation separation
of crude oil into different streams for additional processing in the
refinery and for the production of specific products.  Crude oil is
shipped to the refinery via pipeline, ship, barge, rail, or truck,
whereupon it is sampled, tested, and approved for processing.  The crude
oil is heated to between 650 °F and 700 °F and fed to crude
distillation tower.  Crude components vaporize and flow upward through
the tower.  Draw trays are installed at specific locations up the tower
from which desired side cuts or fractions are withdrawn.  The first
side-cut above the flash zone is usually atmospheric gasoil (AGO), then
diesel and kerosene/jet fuel are the next side-cuts, in that order.  The
lightest components, referred to here as straight run naphtha, remain in
the vapor phase until they exit the tower overhead, following which they
are condensed and cooled and sent to the naphtha splitter.  

Naphtha Splitter

The purpose of the naphtha splitter is to perform a distillation
separation of straight run naphtha into light straight run naphtha and
heavy straight run naphtha. The feed can be split between the C5’s and
C6’s in order to assure the C6’s and heavier are fed to the
reformer.  

Naphtha Hydrotreater

The purpose of the naphtha hydrotreater is to reduce the sulfur of light
and heavy straight run streams before those streams are refined further
by the isomerization and reformer units.

Isomerization Unit

The purpose for the isomerization unit is to convert the light naphtha
from straight chain hydrocarbons to branched chain hydrocarbons,
increasing the octane of this stream. The isomerate is sent to gasoline
blending.   

Reformer

The purpose of the reformer unit is to convert heavy straight run (C6 to
C8 or C9 hydrocarbons) into aromatic and other higher octane compounds
(benzene is one of the aromatic compounds produced), typically necessary
to produce gasoline with sufficient octane.  To protect the very
expensive, precious metal catalyst used in reformers, heavy straight run
naphtha must be hydrotreated first before it is fed to the reformer.  As
the reformer converts the feed hydrocarbons to aromatics, hydrogen and
light gases are produced as byproducts. The liquid product, known as
reformate, is sent directly to gasoline blending, or to aromatics
extraction.   

Aromatics Extraction Unit

The purpose of aromatics extraction is to separate the aromatic
compounds from the rest of the hydrocarbons in reformate using chemical
extraction with a solvent to concentrate the individual aromatic
compounds, (mainly xylene and benzene) for sale to the chemicals market.


Vacuum Tower

The purpose of the vacuum distillation tower unit is to enable a
refinery to produce more gasoline and diesel fuel out of a barrel of
crude oil.  It separates the vacuum gasoil (VGO), which is fed to the
FCC unit, from the vacuum tower bottoms (VTB) which is sent to the
coker, or in other refineries is made into asphalt.  Because most sulfur
contained in crude oil is contained in the heaviest part of crude oil,
the VGO and VTB are very high in sulfur.

Fluidized Catalytic Cracker

The purpose of the fluidized catalytic cracker is to convert heavy
hydrocarbons, which have very low value, to higher value lighter
hydrocarbons.  AGO and VGO are the usual feeds to a fluid catalytic
cracker (FCC).  The full boiling range cracked product leaves the
reactor and is sent to a fractionator.  The overhead includes propane,
propylene, butane, butylene, fuel gas and FCC naphtha, which contains a
significant amount of sulfur.  There are two heavy streams; light cycle
oil (LCO), which can be hydrotreated and blended into diesel fuel or
hydrocracked into gasoline; and heavy cycle oil, sometimes called slurry
oil, which can be used for refinery fuel.  Very simple refineries do not
have FCC units, and therefore, their gasoline is very low in sulfur.

FCC Feed Hydrotreater or Mild Hydrocracker “A”

FCC feed hydrotreaters and mild hydrocrackers hydrotreat or mildly
hydrocrack the feed to the FCC unit which provides two distinct
benefits.  First, by increasing the amount of hydrogen in the feed to
the FCC unit, the FCC unit increases the conversion of the feed to high
value light products, particularly FCC naphtha which increases the
gasoline yield.  Second, hydrotreating the feed removes some
contaminants in the feed such as nitrogen and sulfur.  Nitrogen in the
feed negatively affects the FCC catalyst.  Removing the sulfur in the
feed helps in two ways.  Some of the sulfur in the feed is released by
the cracking process and results in high SOX emissions that would
otherwise have to be controlled by scrubbers – the FCC feed
hydrotreaters may prevent the need to add a scrubber.  Also, FCC feed
hydrotreaters remove sulfur which can allow a refinery to comply with
gasoline sulfur standards.  

FCC Postreat Hydtrotreater “B”

Postreat hydrotreaters solely hydrotreat the naphtha that is produced by
the FCC unit to reduce its sulfur level which enables compliance with
gasoline sulfur standards.  The FCC naphtha is high in olefins which can
be saturated by postreat hydrotreaters resulting in lower octane of the
FCC naphtha.  Vendor companies have developed postreat hydrotreating
technologies which minimize this octane loss.  

Distillate Hydrotreater

The purpose of the distillate hydrotreater is to reduce the sulfur of
distillate, which is also called diesel fuel.

Gas Plant

The purpose of the gas plant is to use a series of distillation towers
to separate various light hydrocarbons for further processing in the
alkylation or polymerization units or for sale.   

Alkylation Unit

The purpose of the alkylation unit is to chemically react light
hydrocarbons together to produce a high quality, heavy gasoline product.
 Alkylation uses sulfuric or hydrofluoric acid as catalysts to react
butylene or propylene together with isobutane.  Following the main
reaction and product separation, the finished alkylate is sent to
gasoline blending.  Alkylate is low in RVP.   

Polymerization Unit

The purpose of the polymerization unit is to react light hydrocarbons
together to form a gasoline blendstock.  A polymerization unit, often
referred to as a “cat poly” is somewhat similar to an alkylation
unit, in that both use light olefins to produce gasoline blendstocks. 
The feed is generally propylene and/or butylene from the gas plant.  The
product, called polygas is sent to gasoline blending.

Coker Unit

The purpose of the coker unit is to process vacuum tower bottoms (VTB)
to coke and to crack a portion to various lighter hydrocarbons.  The
hydrocarbons produced by the coker include cracked gases, coker naphtha,
coker distillate and gas oil.  The gas is fed to the gas plant, the
naphtha to the naphtha hydrotreater after which the heavy coker naphtha
is typically fed to the reformer, and the distillate either to
distillate hydrotreating or to the hydrocracker.  

Hydrocracker

The purpose of the hydrocracker is to crack and “upgrade” the
feedstock into higher value products.  The feedstock to the hydrocracker
is usually light cycle oil (LCO) and coker distillate, poor quality
distillate blendstocks, which are upgraded to diesel fuel, or cracked to
gasoline.  Heavier hydrocarbons such as AGO and HVGO can be feedstocks
as well.  

A more complete description for naphtha hydrotreating is contained in
Section   REF _Ref307313504 \w  4.2 .

Feasibility of Removing Sulfur from Gasoline 

The case can be made in two ways that it is feasible to comply with the
proposed 10-ppm gasoline sulfur standard.  First, feasibility can be
demonstrated by understanding the technologies currently available which
can achieve the necessary reductions in sulfur, and that these
technologies are currently being used to achieve significant reductions
in gasoline sulfur.  The second way to make the case that it is feasible
to comply with the proposed 10-ppm gasoline sulfur standard is to
highlight that refiners in certain countries or other regions are
currently complying with a 10-ppm gasoline sulfur cap standard.  These
two cases will be made below, but first we will review the source of
sulfur in gasoline to understand how sulfur levels can be further
reduced.

Source of Gasoline Sulfur 

Sulfur is in gasoline because it naturally occurs in crude oil.  Crude
oil contains anywhere from fractions of a percent of sulfur, such as
less than 500 ppm (0.05 weight percent) to as much as 30,000 ppm (3
percent).  The average amount of sulfur in crude oil refined in the U.S.
is about 10,000 14,000 ppm.  Most of the sulfur in crude oil is in the
heaviest part, or in the heaviest petroleum compounds, of the crude oil
(outside of the gasoline boiling range).  In the process of refining
crude oil into finished products, such as gasoline, some of the heavy
compounds are broken up, or cracked, into smaller compounds and the
embedded sulfur can end up in gasoline.  Thus, the refinery units which
convert the heavy parts of crude oil into gasoline are the units most
responsible for putting sulfur into gasoline.  

The fluidized catalytic cracker (FCC) unit is a refinery processing unit
that creates a high sulfur content gasoline blendstock.  FCC naphtha
contains from hundreds to several thousand parts per million of sulfur. 
The FCC unit cracks large carbon molecules into smaller ones and
produces anywhere from 25 to 50 percent of the gasoline in those
refineries with FCC units.  Because the FCC unit makes a gasoline
blendstock out of the heavier, higher sulfur-containing compounds, more
than 95 percent of sulfur in gasoline blendstocks comes from streams
produced in that unit.  When complying with the 30-ppm Tier 2 gasoline
sulfur standard, refiners reduced the sulfur content of the FCC naphtha.
 The impact of this action is described below in subsection   REF
_Ref309121749 \w  4.2.2 .  

Straight run naphtha is a gasoline blendstock which contains a moderate
amount of sulfur.  Straight run naphtha is the part of crude oil which
after distillation in the atmospheric crude oil tower falls in the
gasoline boiling range.  The heaviest portion of straight run, which
would have more sulfur, is normally desulfurized and reformed in the
reformer (to improve its octane), so its contribution to the gasoline
pool is virtually nil.  The light straight run which contains the
five-carbon hydrocarbons contains on the order of 100 ppm sulfur and if
this material is not hydrotreated and processed in an isomerizaition
unit, it is blended directly into gasoline.  

Another refinery unit which produces naphtha with a significant amount
of sulfur is the coker unit.  These units produce coke from the heaviest
part of the crude oil.  In the process of producing coke, a naphtha is
produced that contains more than 3,000 ppm sulfur and many very unstable
olefins.  Because this stream is highly olefinic and unstable, refiners
tend to hydrotreat coker naphtha.  Coker naphtha is normally split into
two different streams.  The six- to nine-carbon hydrocarbons are
hydrotreated along with the rest of the heavy naphtha and fed to the
reformer.  The five-carbon hydrocarbon part of coker naphtha is called
light coker naphtha and usually contains on the order of several hundred
parts per million sulfur.  Light coker naphtha is usually hydrotreated
along with the light straight run, and refined further in an
isomerization unit if the refinery has one.

Other gasoline blendstocks contain little or no sulfur.  Alkylate, which
is produced from isobutene and butylenes that contain a small amount of
sulfur, can end up with a small amount of sulfur.  Most refineries have
less than 15 ppm sulfur in this pool, however, some refineries which
feed coker naphtha to the alkylate plant can have much more.  On
average, alkylate probably has about 10 ppm sulfur.  One more gasoline
blendstock with either very low or no sulfur is hydrocrackate, which is
the naphtha produced by hydrocrackers.  It is low in sulfur because the
hydrocracking process removes the sulfur.  Ethanol which is eventually
blended into gasoline usually has very little or no sulfur.  However,
the hydrocarbon used as a denaturant and blended with ethanol at 2
percent is usually natural gasoline, a C5 to C7 naphtha from natural gas
processing, and it contains anywhere from a few parts per million to a
couple hundred parts per million sulfur.  After the denaturant is
blended in, the denatured ethanol contains somewhere between 0 and 10
ppm sulfur.  To meet current pipeline and California specifications,
denatured ethanol must be less than 10 ppm sulfur.

Complying with the Current Tier 2 Gasoline Sulfur Standard

It is important to understand the steps that refiners took to comply
with the 30-ppm Tier 2 gasoline sulfur standard because those capital
investments and operational changes will play a major role in
determining the steps that refiners take to comply with a more stringent
gasoline sulfur standard.  

The Tier 2 sulfur standard was promulgated February 10, 2000.  The
sulfur standard requires that refiners reduce their annual average
gasoline sulfur levels down to 30 ppm and each gallon of gasoline cannot
exceed a per-gallon standard of 80 ppm.  The sulfur standards were
phased in from 2004 to 2006.  The only exceptions were western refiners
(GPA) and small refiners who were given until 2008.  Some small refiners
had their gasoline sulfur deadlines extended through 2010 in the highway
diesel fuel sulfur rule in exchange for on-time compliance there.  As of
January 1, 2011, all refineries are complying with the Tier 2 30-ppm
sulfur standard.  	

A refinery’s previous average gasoline sulfur level is an important
factor which determined whether a refiner would need to make a
substantial capital investment to meet the Tier 2 gasoline sulfur
standards.  We believe that those refiners with low gasoline sulfur
levels to begin with (i.e., gasoline sulfur levels lower than, perhaps,
50 ppm) probably did not invest in expensive capital.  These refineries
have very low sulfur levels due to one or more of a number of possible
reasons.  For example, some of these refiners may not have certain
refining units, such as a fluidized catalytic cracker (FCC) unit, or a
coker, which convert heavy boiling stocks to gasoline.  As stated above,
these units push more sulfur into gasoline and their absence means much
less sulfur in gasoline.  Alternatively, refiners may use a very low
sulfur (sweet) crude oil which can result in a low sulfur gasoline.  Or,
these refiners may have already installed an FCC feed hydrotreater to
improve the operations of their refinery which uses a heavier, higher
sulfur (more sour) crude oil.  As described above, this unit removes
much of the sulfur from the heaviest portion of the heavy gas oil before
it is converted into gasoline. 

Of the refiners in this first category, the refineries with average
sulfur levels below 30 ppm may not have had to do anything to meet the
Tier 2 standards.  On the other hand, those refineries which had sulfur
levels above 30 ppm but below some level, such as 50 ppm, probably are
meeting the 30-ppm sulfur standard by employing operational changes only
and avoided making capital investments.  Most of the refineries with
gasoline sulfur levels below 50 ppm prior to the Tier 2 investments
either do not have a FCC unit, or if they do, they probably have an FCC
feed hydrotreating unit.  

The vast majority of gasoline which was being produced was by refineries
with higher sulfur levels, and these refiners had to either adapt some
existing hydrotreating capital or install new capital equipment in these
refineries to meet the Tier 2 gasoline sulfur standards.  As stated
above, the FCC unit is responsible for most of the sulfur in gasoline. 
Thus, investments for desulfurizing gasoline involved the FCC unit to
maximize the sulfur reduction, and to minimize the cost.  This
desulfurization capital investment can be installed to treat the gas oil
feed to the FCC unit, or treat the gasoline blendstock which is produced
by the FCC unit.  Each method has advantages and disadvantages.

Using FCC Feed Pretreat Hydrotreating to Comply with Tier 2

Some refiners installed FCC feed hydrotreaters (also known as
pretreaters) at their refineries to comply with the Tier 2 gasoline
sulfur standards.  FCC pretreaters treat the vacuum gas oil, heavy coker
gas oil and, in some cases, atmospheric residual feed to the FCC unit
using a hydrotreater or a mild hydrocracker.   These units are designed
to operate at high pressures and temperatures to treat a number of
contaminants in the feed.  Besides sulfur, FCC pretreaters also reduce
nitrogen and certain metals such as vanadium and nickel.  These
nonsulfur contaminants adversely affect the FCC catalyst, so the
addition of this unit would improve the functioning of the unit.  Also,
the hydrotreating which occurs in the FCC pretreater reacts hydrogen in
the feedstock which increases the yield of the FCC unit by increasing
the highest profit-making products produced by refineries, such as
gasoline and light olefins.  While FCC pretreaters provide yield
benefits that offset the capital costs of adding this type of
desulfurization, the costs are still high enough that many refiners
would have a hard time justifying the installation of this sort of unit.
 For a medium to large refinery (i.e., 150,000-200,000 BPCD), the
capital costs may exceed $250 million.  Because of the higher
temperatures and pressures involved, utility costs are expensive
relative to postreat hydrotreating as explained below.  Another
justification for this approach is that it allows refiners to switch to
a heavier, more sour crude oil.  These crude oils are less expensive
per-barrel and can offset the increased utility cost of the FCC
pretreater, providing that the combination of reduced crude oil costs
and higher product revenues justify the switch.  Another benefit for
using FCC pretreaters is that the portion of the distillate pool which
comes from the FCC unit would be partially hydrotreated as well.  This
distillate blendstock, termed light cycle oil, comprises a relatively
small portion of the total distillate produced in the refinery (about 20
percent of on-road diesel comes from light cycle oil), and like FCC
naphtha, light cycle oil contributes a larger portion of the total
sulfur which ends up in distillate.  Thus, FCC pretreaters would also
help a refiner meet the 15-ppm highway and nonroad diesel fuel
standards.

In terms of desulfurization capability, FCC preateaters have different
abilities to remove sulfur from the gas oil feed depending on the unit
pressure.  FCC pretreaters can be subdivided into high pressure units
(1400 psi and above), medium pressure units (900 to 1400 psi), and low
pressure units (under 900 psi).  High pressure FCC pretreaters are
capable of removing about 90 percent of the sulfur contained in the gas
oil feedstock to the FCC unit, while low and medium pressure units are
capable of removing 65 to 80 percent of the feed sulfur.  Because there
is no postreating at many of the refineries with FCC pretreaters,
control of the feed to these units is a critical determining factor for
how well the FCC pretreater will function as desulfurizers.  If the feed
becomes too heavy, the concentration of contaminants increases and the
catalyst may lose its effectiveness.

FCC pretreaters improve desulfurization indirectly by improving the
desulfurization performance of the FCC unit itself.  When FCC units
crack the vacuum gas oil into naphtha,  about 90 percent of the sulfur
is typically cracked out of the hydrocarbons converted to FCC naphtha
(or the FCC naphtha contains only about 10 percent of the sulfur present
in the feed) and is removed as hydrogen sulfide.  When FCC units are
preceded by FCC pretreaters, the amount of sulfur in the feed which ends
up in the FCC naphtha is only about 5 percent, which means that about 95
percent of the sulfur in the feed is removed from the FCC feed when it
is cracked into FCC naphtha.  This effect is caused by the additional
hydrogen which reacts with the feed hydrocarbons.  With more hydrogen
molecules available in the feedstock after hydrotreatment, the FCC
cracking reactions can react more hydrogen with the sulfur contained in
the feed to produce more hydrogen sulfide.  

For complying with Tier 2, refiners which already had or which installed
high pressure FCC pretreaters were able to comply with the 30-ppm sulfur
standard without the need to install a FCC naphtha hydrotreater. 
However, if a refinery had a low pressure, or perhaps even a medium
pressure FCC feed hydrotreater, they were generally less able to comply
with the 30-ppm gasoline sulfur standard with the FCC hydrotreater by
itself, and these refineries were more likely to also install an FCC
postreater.

Using FCC Naphtha Postreat Hydrotreating to Comply with Tier 2

A less capital intensive alternative for reducing FCC naphtha sulfur
levels to comply with Tier 2 is FCC naphtha hydrotreating (also known as
postreaters).  FCC postreaters only treat the gasoline blendstock
produced by the FCC unit.  Understandably, this unit is much smaller
because only about 50 to 60 percent of the feed to the FCC unit ends up
as FCC naphtha, a gasoline blendstock.  The unit is sometimes smaller
than that as some refiners which choose to use a fixed bed hydrotreater
may only treat the heavier, higher sulfur portion of that stream with
hydrotreating, and then treat the lighter fraction with another lower
desulfurization cost technology.  FCC postreaters operate at lower
temperatures and pressures than FCC pretreaters, which further reduces
the capital and operating costs associated with this type of
desulfurization equipment.  Furthermore, the feed to the FCC unit has
corrosive properties which require that FCC pretreaters use more
corrosion-expensive metallurgy, which is not needed for postreaters. 
For a medium to large-sized refinery, the capital costs are on the order
of $70 million for a conventional FCC pretreaters – less than half
that of an FCC pretreater.  

One drawback of this desulfurization methodology is that the octane
value and/or some of the gasoline yield may be lost depending on the
process used for desulfurization.  Octane loss occurs by the saturation
of high octane olefins which are produced by the FCC unit.  Most of the
olefins are contained in the lighter fraction of FCC naphtha.  Increased
olefin saturation usually means higher hydrogen consumption.  There can
also be a loss in the gasoline yield caused by mild cracking which
breaks some of the gasoline components into smaller fractions which are
too light for blending into gasoline.  If there is octane loss, two of
the ways that the octane loss can be made up is by blending in more
ethanol, or by increasing the feed to or the severity of the reformer,
the aromatics production unit of the refinery.  Sometimes vendors of FCC
pretreater technologies design octane increasing capability into their
designs, which is discussed below in the section about the individual
postreater technologies.

The loss of octane and gasoline yield caused by FCC postreating is lower
with technologies which were developed prior to the implementation of
the Tier 2 program.  These processes are termed selective because they
achieve the lower sulfur while preserving much of the octane and
gasoline yield (they were designed specifically for treating FCC
naphtha).  Octane is preserved because the hydrotreating units and their
catalysts are specially designed to avoid saturating olefins.  These
selective processes, or parts of these processes, usually operate at
less severe conditions which result in less cracking preserving yield
compared to conventional hydrotreating processes.  The less severe
conditions also lower the capital and operating costs for this process. 
The lower operating costs arise out of the reduced utility requirements
(e.g., lower pressure).  For example, because these processes are less
severe, there is less saturation of olefins, which means that there is
less hydrogen used.  Less olefin saturation also translates into less
octane loss which would otherwise have to be made up by octane boosting
processing units in the refinery.  The lower capital and operating costs
of these newer FCC postreaters are important incentives for refiners to
choose this desulfurization methodology over FCC pretreaters.  For this
reason, refiners chose to use the more recently developed FCC
postreaters technologies for meeting the 30-ppm Tier 2 gasoline sulfur
standard.  

Not saturating the olefins to preserve octane and limit hydrogen
consumption provides a different challenge.  When the hydrogen sulfide
is formed and there is a significant concentration of olefins present,
the hydrogen sulfide compounds tend to react with the olefinic
hydrocarbon compounds forming mercaptan sulfur compounds.  This reaction
is called “recombination” because the removed sulfur recombines with
the olefinic hydrocarbons contained in the naphtha. This is particularly
a problem if the light cat naphtha is present in the hydrotreater
because the highest concentration of olefins is in the light cat
naphtha.  The recombination reactions occur more readily if the
hydrotreater is operated more severely (at a higher temperature) to
increase the sulfur removal, and the feed to the hydrotreater is high in
sulfur.  However, while operating this type of hydrotreater more
severely can result in the further removal of the original sulfur
present in the hydrocarbons, it also can result in the formation of more
recombination mercaptans that results in a “floor” reached for the
amount of sulfur that can be removed from the hydrocarbons.  This cycle
of increased sulfur removal and simultaneous increase in recombination
results in the saturation of more olefins and increases the consumption
of hydrogen.  There are a number of different vendor-specific
technologies that each vendor may use to avoid or address recombination
reactions as discussed below.  It is important to note that the
technologies employed to reduce recombination may require the addition
of some capital costs which offsets some or perhaps all the capital cost
savings due to the milder operating conditions of these selective
hydrotreater technologies compared to nonselective hydrotreating.  

One means to achieve high levels of desulfurization while avoiding much
of the problem with recombination reactions is by using a two-stage
hydrodesulfurization methodology.  A two-stage unit has two
desulfurization reactors, but instead of just adding additional reactor
volume, the hydrocarbons exiting the first reactor are stripped of
gaseous compounds (most importantly, the hydrogen sulfide is removed),
injected with fresh hydrogen, and then hydrodesulfurized again in the
second stage.  Both reactors undergo modest desulfurization and hydrogen
sulfide concentrations remain sufficiently low to avoid recombination
reactions.  The downside of this approach is that the second stage
incurs greater capital costs compared to single-stage configurations. 
Because Tier 2 was not too constraining, we believe that refiners
installed few, if any, two-stage desulfurization units to comply with
those gasoline sulfur standards.    

Whatever strategy chosen by the refiner to comply with Tier 2, a
critical criterion is that the postreater be capable of cycle lengths
that match that of the FCC unit, which typically is 5 years.  If the
postreater were to require a catalyst changeout before the FCC unit
requires a shutdown, either the refiner would have to shutdown the FCC
unit early to mirror that of the postreater, or need to store up the
high sulfur FCC naphtha (this stream would be too high in sulfur to
blend directly to gasoline under the Tier 2 80-ppm cap standard) until
the postreater was started up again and is able to hydrotreat the stored
up high sulfur FCC naphtha.

We know of six FCC postreater technologies that refiners used to comply
with the Tier 2 gasoline sulfur standards.  These are Axens (was IFP)
Prime G and Prime G+, Exxon Scanfining, CDTech’s CDHydro and HDS,
Sinopec’s (was Phillips)S-Zorb and UOP’s ISAL and Selectfining.  

Of the list of FCC postreaters , Axens Prime G+, Exxon Scanfining and
UOP’s ISAL and Selectfining are fixed bed desulfurization
technologies.  These processes are called fixed bed because the catalyst
resides in a fixed bed reactor.  The high sulfur gasoline blendstock is
heated to a high temperature (on the order of 600 degrees Fahrenheit)
and pumped to a high pressure to maintain the stream as a liquid.  It is
then combined with hydrogen before it enters the reactor.  The reactions
occur within the bed of the catalyst.  While the petroleum is in contact
with the catalyst in the reaction vessel, the sulfur reacts with
hydrogen and is converted to hydrogen sulfide.  Also, depending on the
process, some of the olefin compounds which are present in the cracked
stream are saturated which increases the amount of octane lost and
hydrogen consumed.  After the reactor, the gaseous compounds, which
include unreacted hydrogen, hydrogen sulfide, and any light end
petroleum compounds which may have been produced in the reactor by
cracking reactions, are separated from the liquid compounds by a
gas/liquid separator.  The hydrogen sulfide must be stripped out from
the other compounds and then converted to elemental sulfur in a separate
sulfur recovery unit.  The recovered sulfur is then sold.  If enough
hydrogen is present, and it is economical to recover, it is separated
from the remaining hydrocarbon stream and recycled.  Otherwise, it is
burned with light hydrocarbons as fuel gas. 

Despite the similarities, each of these desulfurization technologies has
its differences.  Axens Prime G+ desulfurization process largely
preserves olefins as its strategy for diminishing octane loss.,   The
Axens process employs a selective hydrogenation unit (SHU) as a first
step.  The role of this unit is to saturate the unstable diolefin
hydrocarbons in a hydrogen rich environment, and react the light
mercaptan and sulfide hydrocarbons together.  The SRU also converts
exterior olefins to interior olefins which results in a small increase
in octane.  The mild operating conditions of the SHU tend to avoid the
saturation of monoolefins.  After exiting the SRU, the FCC naphtha is
sent to a distillation column which separates the light FCC naphtha
(typically comprising about one fourth of the total cat naphtha) from
the heavy naphtha.  Because the light sulfur compounds were reacted
together and those compounds no longer fall within the light cat naphtha
boiling range, the light cat naphtha is low in sulfur and can be blended
directly into gasoline.  The heavy cat naphtha which is naturally high
in sulfur and which also contains the self-reacted light mercaptans and
sulfides from the SHU, is sent to a fixed bed hydrotreater.  The fixed
bed hydrotreater contains both cobalt-molybdenum and nickel-molybdenum
catalyst.  An important way that Axens avoids recombination reactions is
by separating the light sulfur compounds from the light naphtha and
keeping the light naphtha out of the fixed bed hydrotreater.  The
desulfurized heavy cat naphtha is blended into the gasoline pool.  

If the feed to the Axens Prime G unit is very low in sulfur, a low
capital investment option was available to the refiner by feeding the
entire FCC naphtha stream to the hydrotreating reactor avoiding the SHU
and splitter.  This option trades lower capital cost with somewhat
higher octane loss and hydrogen consumption.  Because of the low
severity of the hydrotreating reactor (low severity is possible because
the lower amount of desulfurization that is occurring), the amount of
octane loss and hydrogen consumption is modest.  There are more than 180
Prime G+ units operating worldwide.

The first step in Exxon’s fixed bed Scanfining process is to mildly
heat the full FCC naphtha and pass it through a small reaction vessel
which reacts the diolefins to monoolefins.       The full FCC naphtha is
then heated further, injected with hydrogen gas and sent to the fixed
bed hydrotreating reactor which is packed with a catalyst developed
jointly between Exxon and Akzo Nobel (now Albermele).  If the degree of
desulfurization is relatively modest, the amount of recombination is low
and the FCC naphtha is sent to gasoline blending.  If, however, the
degree of desulfurization is higher (due to FCC naphtha with a higher
sulfur content), then there likely would be an excessive number of
recombination reactions.  In this case, Exxon recommends either one of
two different technologies to address the recombination reactions.  One
technology is Zeromer.  Zeromer is a fixed bed reactor vessel installed
after the main fixed bed hydrotreater reactor that specifically designed
to hydrodesulfurize the mercaptan sulfur from the FCC naphtha without
saturating olefins.  Another technology Exxon developed, in conjunction
with Merichem, is an extractive mercaptan removal technology named
Exomer.  The Exomer technology differs from other sulfur extraction
technologies in that it is capable of extracting mercaptans from the
entire FCC naphtha pool.  Like Zeromer, the Exomer technology would be
an add-on technology installed after the Scanfining fixed bed reactor. 
There are 16 Scanfining units operating in the U.S.  

UOP has licensed two FCC naphtha hydrotreating technologies.  When Tier
2 was being phased-in, UOP was licensing a technology named ISAL
developed by INTEVEP S.A.   The ISAL process is different than the other
FCC naphtha hydrotreaters because instead of avoiding the saturation of
olefins as sulfur is being hydrotreated out of FCC naphtha, the ISAL
process completely saturates the olefins.  To avoid a large octane loss,
the ISAL process separates the olefin-rich, light cat naphtha from the
heavy cat naphtha.  The light cat naphtha is treated by an extractive
desulfurization technology such as Merox which does not saturate
olefins.  Only the heavy cat naphtha is sent to the ISAL reactor.  To
offset the octane loss caused by the saturation of the olefins in the
heavy cat naphtha as it is being desulfurized, the ISAL catalyst
isomerizes and conducts some mild cracking and reforming of the heavy
cat naphtha.  One downside of the ISAL process is that, due to the
complete saturation of olefins, the hydrogen consumption is higher
relative to the selective hydrodesulfurization technologies that avoid
saturating olefins.

UOP has since developed and licenses its own FCC naphtha desulfurization
technology named SelectFining.  SelectFining is a selective
hydrodesulfurization technology that seeks to minimize olefin saturation
to minimize both octane loss and hydrogen consumption.  SelectFining
treats the full FCC naphtha.  The full range FCC naphtha is first sent
to a diolefin saturating reactor before being sent to the SelectFining
reactor.  SelectFining relies on its catalyst design to selectively
remove sulfur and prevent recombination reactions.  UOP recommends a
two-stage reactor setup for high levels of desulfurization.  

The next two FCC naphtha desulfurization technologies, CDTech and S-Zorb
do not use fixed bed reactors, but very different technologies which are
also very different from each other.   Each will be discussed
separately.

Although the CDTech process is significantly different from the fixed
bed hydrotreating technologies, it still uses the same type of catalyst.
 The CDTech process utilizes catalytic distillation.,,  Catalytic
distillation is a technology which has been applied for a number of
different purposes.  CDTech is currently licensing the technology to
produce MTBE and selective hydrogenation processes, including FCC
naphtha desulfurization and benzene saturation.  As the name implies,
distillation and desulfurization, via catalyst, take place in the same
vessel.  This design feature saves the need to add a separate
distillation column sometimes used with fixed bed hydrotreating.  All
refineries have a distillation column after the FCC unit (called the
main fractionation column) which separates the FCC naphtha from the most
volatile components (such as liquid petroleum gases), the distillate or
diesel (light cycle oil), and the heavy ends or residual oil.  However,
if a refiner only wishes to treat a portion of the FCC naphtha, then a
second distillation column would need to be added after the main FCC
fractionation column to separate off the portion of the FCC naphtha
which he wishes not to treat.  With the CDTech process, the refiner can
choose to treat the entire pool or a portion of the pool, but choosing
to treat a part of the pool can be an option in how the CDTech hardware
is applied, thus negating any need for an additional distillation
column.

The most important portion of the CDTech desulfurization process is a
set of two distillation columns loaded with desulfurization catalyst in
a packed structure.  The first vessel, called CDHydro, treats the
lighter compounds of FCC gasoline and separates the heavier portion of
the FCC naphtha for treatment in the second column.  The second column,
called CDHDS, removes the sulfur from the heavier compounds of FCC
naphtha.  All of the FCC naphtha is fed to the CDHydro column.  The
five- and six-carbon petroleum compounds boil off and head up through
the catalyst mounted in the column, along with hydrogen which is also
injected in the bottom of the column.  The reactions in this column are
unique in that the sulfur in the column is not hydrotreated to hydrogen
sulfide, but they instead are reacted with dienes in the feed to form
thioethers.  Their higher boiling temperature causes the thioethers to
fall to the bottom of the column.  They join the heavier petroleum
compounds at the bottom of the column and are sent to the CDHDS column. 
Because the pressure and temperature of the first column is much lower
than conventional hydrotreating, saturation of olefins is reduced to
very low levels.  The olefin saturation which does occur is necessary to
eliminate diolefins.  Thus, little excess hydrogen is consumed.  CDTech
offers an option to refiners to put in an additional catalyst section in
the CDHydro column to increase octane.  This octane enhancing catalyst
isomerizes some of the olefins, which increases the octane of this
stream by about three octane numbers, and few of the olefins are
saturated to degrade this octane gain. The seven-carbon and heavier
petroleum compounds leave the bottom of the CDHydro unit and are fed
into the CDHDS column.  There, the heavier compounds head down the
column and the lighter compounds head up.  Both sections of the CDHDS
column have catalyst loaded into them which serve as hydrotreating
reaction zones.  Similar to how hydrogen is fed to the CDHydro column,
hydrogen is fed to the bottom of the CDHDS column.

The temperature and pressure of the CDTech process columns are lower
than fixed bed hydrotreating processes, particularly in the upper
section of the distillation column, which is where most of the olefins
end up.  These operating conditions minimize yield and octane loss. 
While the CDTech process is very different from conventional
hydrotreating, the catalyst used for removing the sulfur compounds is
the same.  One important difference between the CDTech process and
conventional hydrotreating is that CDTech mounts its catalyst in a
unique support system, while conventional catalyst is usually dumped
into the fixed bed reactor.  CDTech has 13 CDHydro/CDHDS desulfurization
units in operation in the U.S.  

Phillips Petroleum Co. commercialized and licensed an adsorption
desulfurization technology called S-Zorb.   In 2007, Phillips sold the
S-Zorb process to SINOPEC.  S-Zorb uses a chemical adsorption process,
instead of hydrotreating, as the principal methodology for the removal
of sulfur from FCC naphtha.  Adsorption has the benefit of operating at
much lower pressure and temperatures, which lowers operating costs. 
S-Zorb, uses two separate columns and is constantly moving an adsorption
catalyst from the reactor vessel to the regeneration column, and back
again.  The untreated FCC naphtha and hydrogen are fed to the reaction
vessel where the catalyst catalytically removes the sulfur from the
petroleum compound facilitated by the hydrogen present in the reactor. 
The catalyst, which begins to accumulate the removed sulfur, is
transferred over to the regeneration column on a continual basis where
the sulfur is removed from the catalyst using hydrogen as the scavenging
compound.  Then the hydrogen disulfude is converted to sulfur dioxide
and sent to the sulfur recovery unit.  Because the process still relies
upon catalytic processing in the presence of hydrogen, there is some
saturation of olefins, with a commensurate reduction in octane.  Through
a literature search, we believe that 7 S-Zorb desulfurization units were
originally licensed for Tier 2, but our information sources have
communicated that only 4 units are actually operating today.  

We conducted a literature search or in some cases asked vendors to name
which refineries installed their FCC naphtha desulfurization technology
to enable compliance with Tier 2.  A summary of the total number of
units by vendor and technology type is summarized in   REF _Ref307313649
 Table 4-1 .

Table   STYLEREF 1 \s  4 -  SEQ Table \* ARABIC \s 1  1  Estimated
Number of FCC Desulfurization Technologies Installed to comply with Tier
2 by Vendor Company or Technology 

Axens Prime G	Exxon

Scanfining	CDTech	Sinopec S-Zorb	UOP ISAL 

UOP Selectfining	FCC Feed HT	No FCC Unit

40	16	15	4	1	17	14

Meeting a 10-ppm Gasoline Sulfur Standard

To meet a 10-ppm average gasoline sulfur standard, we believe that the
primary strategy that refiners would adopt would be to further reduce
the sulfur level of FCC naphtha.  There are three primary reasons why we
settled on this as the primary strategy we chose for analyzing the
compliance costs for Tier 3.  The first reason is that, even after
refiners used hydrotreating to reduce the sulfur in the FCC naphtha to
comply with Tier 2, FCC naphtha is by far the largest contributor of
sulfur to the gasoline pool, by virtue of both its volume and sulfur
content.    REF _Ref308595986 \h  Table 4-2  below summarizes the
estimated average volumes and average sulfur levels for the primary
blendstocks typically blended into gasoline for the current Tier 2
situation.  By using the refinery-by-refinery model to model today’s
situation for the typical refinery, we estimate that the FCC naphtha
contains about 70 ppm for the typical refinery complying with the 30-ppm
Tier 2 sulfur standard and that gasoline blendstock typically
contributes to about 40 percent of a refiner’s gasoline pool.    REF
_Ref308595986 \h  Table 4-2  also summarizes the changes in gasoline
blendstock sulfur levels we believe would occur when complying with the
proposed 10-ppm gasoline sulfur standard.  Using the
refinery-by-refinery model, we project that a 10-ppm gasoline sulfur
standard can be met by a typical refinery by reducing the sulfur level
of FCC naphtha from about 70 ppm to 20 ppm.  In fact, for virtually all
refineries that have an FCC unit, refiners would not be able to comply
with the proposed 10-ppm gasoline sulfur standard without further
desulfurizing the FCC naphtha.  The second reason why we believe that
refiners would address the sulfur in the FCC naphtha is because both
vendors and refiners have told us that this is the gasoline blendstock
stream that they intend to address.  Both vendors and refiners have
explained to us that, for most refineries, FCC naphtha hydrotreaters
will already be in place that can be retrofitted with only a modest
capital cost to realize the sulfur reduction needed.  Third, further
reducing the sulfur of the FCC naphtha as the means to comply with Tier
3 is supported by other cost studies.  When these studies assessed the
costs for further reducing the sulfur levels of gasoline, they also
focused further reducing the sulfur levels of the FCC naphtha.  See the
subsection at the end of Chapter 5 discussing these other cost studies. 


Table   STYLEREF 1 \s  4 -  SEQ Table \* ARABIC \s 1  2  Estimated
Typical Gasoline Blendstock Volumes and Sulfur Levels after Tier 2 and
Complying with a 10-ppm Sulfur Standard

Gasoline Blendstock	30-ppm Tier 2 Gasoline Sulfur Standard	10 -ppm
Gasoline Sulfur Standard

	Volume (Percent)	Sulfur (ppm)	Volume (Percent)	Sulfur (ppm)

FCC Naphtha	37	72	36	22

Reformate	23	0.5	22	0.5

Alkylate	13	5	13	5

Isomerate	3	0.5	3	0.5

Butane	4	10	4	10

Light Straight Run Naphtha and Natural Gas Liquids	5	34	5	1

Hydrocrackate	3	8	3	8

Ethanol	10	5	12.5	5

Coker Naphtha	2	1	2	1

Other Gasoline Blendstocks	1	10	1	1

Total/Sulfur Average	100	30	100	10

Reducing FCC naphtha from 70 ppm to 20 ppm would likely be accomplished
in different ways depending on the desulfurizing technology and
configuration used for Tier 2, and whether the current capital employed
for lowering gasoline sulfur is severely taxed or not severely taxed. 
For purposes of this discussion, we will discuss the likely steps taken
to comply with Tier 3 based on whether a refiner solely used an FCC
pretreater or FCC postreater to comply with Tier 2.  While we provided
an example for a typical refinery needing to reduce its FCC naphtha from
70 ppm to 20 ppm to enable compliance with Tier 2, there are many
refineries which are not typical and so their starting and ending sulfur
levels would be different from this example.  Despite these differences,
we believe that every refinery could technically comply with a 10-ppm
gasoline sulfur standard.  This is because gasoline sulfur is easy to
remove - the challenge is to comply while minimizing the cost of doing
so.  This challenge is further discussed in Section   REF _Ref310414725
\r \h  4.2.3.5  below which discusses the value of the proposed
averaging, banking and trading program.

	The one exception is the case where a refinery does not have an FCC
unit.  Refineries in this situation would likely already be producing
gasoline which is 10 ppm or below.  If the refinery’s gasoline is
above 10 ppm, then the refiner would need to address one or more of
several different gasoline blendstocks, including light straight run,
butane and natural gas liquids.  This is discussed at the end of this
section about other gasoline streams.  

Meeting 10 ppm if Refiners Used an FCC Feed Pretreater to Comply with
Tier 2 

	If a refiner relied on an FCC pretreater to comply with Tier 2 at a
refinery, the refiner would likely only be able to achieve 10-ppm sulfur
gasoline if their FCC pretreater is a high pressure unit.   This is
because most refineries which have FCC pretreaters process sour crude
oils and if the unit is a mid or low-pressure unit, the unit pressure
would likely be too low to sufficiently desulfurize the FCC feed.  This
is likely true even if the refiner added reactor volume to its existing
low or medium pressure FCC pretreater which does cause additional
desulfurization.  The problem with the mid and low pressure FCC
pretreaters is that they just cannot remove enough of the sulfur in the
gas oil feed to the FCC unit to achieve adequately low sulfur levels in
the FCC naphtha.  If a refinery processes moderate to low sulfur crude
oil and has a low to mid-pressure FCC pretreater, however, it may be
able to achieve an adequate degree of desulfurization in the FCC naphtha
to enable the refiner to reduce its gasoline sulfur down to 10 ppm. 
Thus, if a refinery cannot achieve a sufficient level of desulfurization
with its current or revamped FCC pretreater to comply with a 10-ppm
gasoline sulfur standard, then the refiner will have to install a
grassroots FCC postreater.  Alternatively, refiners in this situation
would be in the best situation to take advantage of the averaging aspect
of the averaging, banking and trading program (ABT).  Using the ABT
provisions to its advantage, the refiner would achieve the most
desulfurization that it can with its existing FCC pretreater (perhaps 20
ppm sulfur gasoline), and then would need to purchase credits to
demonstrate the remainder of its compliance with the 10-ppm gasoline
sulfur standard.  This scenario would avoid the need for a refiner to
install an expensive grassroots FCC postreater. 

While they are expensive to install, FCC pretreaters provide important
operating cost advantages over postreaters.  An important advantage of
FCC pretreating is that it occurs upstream of the FCC unit and therefore
does not jeopardize the octane value of the olefins produced in the FCC
unit.  Another advantage of the FCC pretreater is that it tends to
increase the yield of naphtha from the FCC unit which improves operating
margins for the refinery with such a unit.  Thus, refiners which are
able to use FCC pretreaters to comply with the Tier 3 sulfur standard
would likely yield a further return on any investment made, and offset
some or all of the increased operating costs incurred.  Perhaps only 5
refineries have high pressure FCC pretreaters in the U.S.

Meeting 10 ppm if Refiners Used an FCC Postreater to comply with Tier 2 

If a refiner installed an FCC postreater to comply with the Tier 2
gasoline sulfur standard, there are several considerations about the
current configuration of the postreater which would affect how a refiner
would use this unit to comply with a 10-ppm gasoline sulfur standard. 
The first issue is what is the degree of desulfurization the postreater
is currently facing?  It makes sense to work through several examples to
understand the types of revamps and associated investments that might
occur.  

For the first example, if the refinery is refining a very sour (high
sulfur) crude oil and the sulfur of the FCC naphtha exiting the FCC unit
is 2,400 ppm, the postreater is currently removing almost 97 percent of
the feed sulfur assuming that the sulfur level of the FCC naphtha
exiting postreater is 70 ppm, which is a very high level of
desulfurization.  When attempting to achieve further sulfur reduction in
the FCC naphtha, the refiner must be concerned about the increased
occurrence of recombination reactions and the potential for much more
octane loss and hydrogen consumption.  This refiner would strongly
consider adding a second stage, which may actually reduce the level of
recombination reactions and the octane loss currently experienced by the
postreater.  Most all the vendors offer a second stage option.  In the
case of CDTech, they call the second reactor, added as part of its
second stage, a polishing reactor.  We contacted the desulfurization
engineer at Sinopec who explained that these units could just be turned
up and that no additional capital investments would be needed.  A
Conoco-Phillips hydrotreating specialist we spoke to confirmed that this
would be the strategy for their S-Zorb units.  Yet one more option, if
the refiner is interested in improving its operating margins such as
increased gasoline production, and has ample capital dollars to spend,
the refiner could add an FCC feed hydrotreater to increase its yield of
FCC naphtha, or a mild hydrocracker to increase its production of low
sulfur distillate.  

In contrast, if a refiner is processing a very sweet (low sulfur) crude
oil, the sulfur level exiting the FCC unit may be as low as 300 ppm, and
under Tier 2, the level of desulfurization necessary to bring that
stream down to 70 ppm is about 81 percent which is a very modest level
of desulfurization.  Similarly, a refinery processing a moderately sour
crude oil with a medium pressure FCC feed hydrotreater could be in a
similar situation.  The refineries in this situation could have a lot
more capacity in their existing postreaters to achieve lower sulfur
without additional capital cost investments.  However, many refiners in
this situation which invested in an FCC postreater for Tier 2 may have
minimized their capital investments.  For example, a refiner may have
avoided the capital and operating cost of a splitter with its postreater
by hydrotreating the full range FCC naphtha.  Therefore, the increased
severity of the postreater needed to achieve 20 ppm in the FCC naphtha
to meet a 10-ppm gasoline sulfur standard might create a larger octane
loss and higher hydrogen consumption than what the refinery could easily
provide without a significant additional capital investment.  In this
case, the refiner can invest some capital in the postreater to minimize
the increase in octane loss and hydrogen consumption.  For example a
refiner with an Axens unit in this situation could add the SHU and a
splitter.  A refiner with a Scanfining unit in this situation wishing to
minimize the octane loss and hydrogen consumption could add a Zeromer or
an Exomer unit.  Alternatively, if the refiner is processing a
moderately sour crude oil and has a moderate pressure FCC feed
hydrotreater, the refinery may choose instead to revamp the FCC feed
hydrotreater for its operational benefits rather than revamp the
postreater. 

The last example of a postreater deserving some discussion is the case
where the sulfur level exiting the FCC unit is 800 ppm.  This is
probably most typical of a refinery refining a crude oil containing an
average amount of sulfur, or, perhaps a refinery refining a very sour
crude oil but treating the vacuum gas oil with a low pressure FCC feed
hydrotreater.  The current FCC naphtha hydrotreater would be achieving
about 90 percent desulfurization when producing FCC naphtha with 80 ppm
sulfur.  In looking to reduce the FCC naphtha down to 20 ppm to comply
with a 10-ppm sulfur standard, a refiner in this position would not
likely consider adding a second stage.  This is because avoiding the
increased octane loss and increased hydrogen consumption for the
additional increment of sulfur reduction would probably not justify the
capital costs associated with a second stage.  Instead of a second
stage, a refiner could revamp the existing FCC postreater with
additional reactor volume, or add capital for addressing recombination
reactions, both likely to be a lot less capital intensive than a second
stage.  A no investment option is possible for refiners in this
situation, although the increase in octane loss and hydrogen consumption
is likely to be significant.  

Perhaps the most important part of an FCC hydrotreater is the catalyst
used in the unit.  Due to continuing research, catalysts are constantly
being developed which are more active, thus achieving greater
desulfurization at a lower temperature, and minimize octane loss and
hydrogen consumption due to lower olefin saturation.  When the Tier 2
naphtha desulfurizers were being put into service the most recent
catalysts were likely used in those units.  These catalysts can be
changed out when the postreater is being taken down for regular
maintenance, and new and improved catalysts can be used to improve the
desulfurization capacity of the unit.  If refiners indeed need to comply
with a 10-ppm gasoline sulfur standard, they would be expected to
upgrade to the most recent catalyst to minimize their costs.  Using the
most active catalyst available would reduce the capital cost that would
need to be incurred and reduce the hydrogen consumption and octane loss
that would otherwise occur.  We are aware of newer lines of catalysts
being marketed by the various vendors.  We can confirm that Axens and
UOP have introduced more active catalysts since the catalysts were
loaded into the FCC postreaters to comply with Tier 2, although it is
likely that all the vendors now offer improved hydrotreating catalysts. 
 

Desulfurizing Other Blendstocks

A more stringent gasoline sulfur standard could require refiners to have
to address other gasoline streams that are high enough in sulfur to be a
concern to the refiners.  This is because without addressing these
gasoline streams, the refiner would have to reduce their FCC naphtha
even lower in sulfur resulting in high per gallon costs at the lower
sulfur levels.  The gasoline streams that we have identified that could
require additional desulfurization include light straight run naphtha,
natural gas liquids and butane.  

Light straight run naphtha (LSR) is naturally occurring in the crude oil
and is desulfurized at many refineries before it is sent to an
isomerization unit.  However, a number of refineries don’t have
isomerization units and therefore some or perhaps many of these
refineries may not be treating this stream today.  Natural gas liquids
(also termed pentanes plus) are naphtha streams sourced from natural gas
wells which are purchased by refiners and blended into the gasoline
pool.  Depending on the source of the specific naphtha stream being
purchased, these streams could vary widely in gasoline sulfur, ranging
from a few ppm sulfur up to several hundred ppm sulfur.  Butane is
natural occurring in crude oil and butane is also produced by the FCC
unit, and to a lesser degree, hydrocrackers.   Refiners separate the
butane from these various streams and then blend it back into their
gasoline pool depending on the RVP requirements of the gasoline market
that the refiner is selling their gasoline into. 

Refiners have multiple options for addressing the sulfur levels of these
various streams.  The LSR and natural gas liquids can be hydrotreated in
either the FCC postreaters or the naphtha hydrotreaters.  Because these
naphtha streams do not have any olefins, there is essentially no octane
loss and, therefore, hydrogen consumption is lower compared to
hydrotreating FCC naphtha.  Another way of treating these streams would
be to use caustic extraction to extract the mercaptan sulfur from these
streams.  Since only the mercaptans are removed with the extraction
technology, the final sulfur level won’t be as low compared to
desulfurization using hydrotreating.  Finally, the refiner could choose
to simply not purchase the natural gas liquids and sell the LSR on the
open market as opposed to treating these streams.  If a refiner decides
to not treat the LSR or natural gas liquids, other refiners with excess
capacity in their FCC postreaters or naphtha hydrotreaters could treat
these streams.  

Butane is normally treated using mercaptan extraction technologies.  We
are aware that some refiners have installed at least a few of these
units at their refineries to address butane sulfur.  Because butane is
usually relatively low in sulfur to begin with, refiners are likely to
only pursue desulfurizing this gasoline blendstock if their butane is
higher than average in sulfur, or if they are considering producing a
very low sulfur gasoline, such as 5 ppm.  

In summary, to comply with a 10-ppm gasoline sulfur standard, refiners
have a range of options available to them, most of which involve
reducing the sulfur in the FCC naphtha.  If a refinery has a high
pressure FCC pretreater, the refiner may be able to just turn up the
hydrotreating severity of that unit.  If a refinery has a low or medium
pressure FCC pretreater and no postreater, the refinery would likely
need to install a grassroots FCC postreater to comply with a 10-ppm
gasoline sulfur standard, or achieve the most that it can with its
current capital and rely on the ABT program.  Refiners with FCC
postreaters have multiple options.  If a refinery is short on octane and
hydrogen, the refiner is likely to invest in capital to avoid as much
octane loss and hydrogen consumption as possible.  However, if the
refiner has a lot of excess octane and hydrogen, the refiner may choose
to avoid any capital cost investments or only make small capital
investments and tolerate the higher octane loss and hydrogen consumption
by simply turning up the severity of its current FCC postreater. 
Refineries with postreaters could always invest in an FCC pretreater
(hydrotreater or mild hydrocracker) to improve its refinery’s margins
or to produce more low sulfur diesel fuel.  Finally, in blending up
their gasoline, some refiners may still be blending in some produced or
purchased gasoline blendstocks that continue to have high enough sulfur
levels which would be a concern when faced with a more stringent
gasoline sulfur standard, and several options exist for addressing the
sulfur in these gasoline blendstocks.  

Demonstrated Compliance with a 10-ppm Gasoline Sulfur Standard	

There are multiple cases today where refiners are complying with 10-ppm
or lower gasoline sulfur programs.  The State of California required
gasoline sold in the State to meet a 15-ppm gasoline sulfur standard on
average and a 20-ppm cap (California gasoline’s per-gallon sulfur cap
dropped to 20 ppm on January 1, 2012).  Furthermore, refiners can
produce gasoline which varies in composition, provided that the
California Predictive Emissions Model (which, like EPA’s Complex
Model, estimates vehicle emissions from fuels of varying composition)
confirms that the proposed fuel formulation meets or exceeds the
emissions reduction that would occur based on the default fuel
requirements.  California refineries are using the flexibility provided
by the Predictive Model to surpass the prescriptive standards for
gasoline sulfur and are producing gasoline which contains around 10 ppm
sulfur on average.  They are making this very low sulfur gasoline
despite using Californian and Alaskan crude oils which are poorer
quality than most other crude oils being used in the U.S. today.  Thus,
the experience in California demonstrates that commercial technologies
already exist to permit refiners to produce very low sulfur gasoline.  

Japan currently has a 10-ppm gasoline sulfur cap that took effect
January 2008.  Europe also has a 10-ppm sulfur cap that has been adopted
by the 30 Member States that comprise the European Union (EU) and the
European Free Trade Association (EFTA) as well as Albania and
Bosnia-Herzegovina.  Under a 10-ppm cap standard, the gasoline sulfur
level likely averages about 5 ppm.  Although gasoline in Japan and
Europe is made from different crude oil sources and much of the heavier
ends are cut into diesel fuel, these international fuel programs (along
with California) provide evidence that advanced gasoline desulfurization
technologies have been deployed and are readily available enable
compliance with the proposed Tier 3 fuel program.   

 Improved Feasibility with the Proposed ABT Provisions 

The averaging, banking and trading (ABT) and small refiner and small
volume refinery aspects of the proposed Tier 3 gasoline sulfur program
would ease the feasibility of compliance with the program.  To make the
point, it is useful to first understand compliance if the ABT and small
refiner and small volume refinery provisions did not exist.  Without
these provisions, all refineries would have to comply with the 10-ppm
gasoline sulfur standard by January 1, 2017.  In the approximate 4 years
after finalizing this rulemaking, most refiners would have to make
capital investments in their refineries to enable compliance with the
10-ppm gasoline sulfur standard.  These investments include revamped FCC
pretreaters and postreaters, and the installation of grassroots FCC
postreaters.  As described above, reaching 10 ppm sulfur in the gasoline
pool is attainable by each refinery.  However, refiners assess the
economic feasibility of their refineries differently depending on past
and expected future economic performance.  They therefore have different
tolerances for making capital investments and absorbing increased
operating costs.  This is particularly true during a period of time in
which gasoline demand is projected to be flat and renewable fuel
blending is expected to increase.  Refiners who own small refineries are
concerned about the higher per-barrel costs for the capital installed at
those small refineries. 

The small refiner and small volume refinery provisions will delay
compliance for these entities until January 1, 2020.  Small refiners
need more time because they have smaller engineering staffs that they
can dedicate to oversee the necessary refinery changes, thus they are
more likely to complete the necessary changes to their refineries later
than large refiners.  Also, the delay allows the small refiners to
experience improved margins for a couple of years when other larger
refineries are complying with the gasoline sulfur standards. 

The banking provisions of the ABT program effectively phase in the
sulfur standard over six years starting in 2014 through the end of 2019.
 The phase-in allows refiners to stagger their investments to their
economic advantage.  Refineries which are expected to incur the lowest
costs for achieving lower gasoline sulfur levels can comply early and
earn sulfur credits.  These credits can then be used to demonstrate
compliance starting in 2017 by the refineries which are expected to
incur higher costs for reducing their gasoline sulfur levels allowing
those refineries to delay investments for lowering their gasoline
sulfur.  This phasing-in of the gasoline sulfur standard will help
spread out the preliminary design demands on the vendor companies which
license the desulfurization technology to refiners, spread out the
detailed design demands on the engineering companies which provide that
service to refiners, spread out the permitting demands on the states
which must provide environmental permits to refiners, and spread out the
demands on the fabrication shops which construct the reactors and other
major hardware which must be installed at refineries to realize the
gasoline sulfur reductions.  For more on how the proposed ABT provisions
are expected to help with lead time, refer to Section   REF
_Ref309137096 \w  4.3 .

Finally, the averaging provisions of the ABT program will provide
additional flexibility and help to reduce the costs of the gasoline
sulfur program.  The averaging provisions will allow refiners to reduce
the gasoline sulfur levels to under 10 ppm at their lower cost
refineries to show compliance or help to show compliance at higher cost
or financially challenged refineries.  

Implications of an Average Gasoline Sulfur Standard Less than 10 ppm

While there may be emissions motivations for further reducing gasoline
sulfur levels, there are practical reasons for proposing a 10-ppm annual
average sulfur standard instead of a more stringent standard, e.g., 5
ppm.  The lower the sulfur standard, the more costly it is for refiners
to achieve the lower sulfur standard.  We identified several reasons why
the costs increase so much for more deeply desulfurizing the gasoline
pool.  

As desulfurization severity increases, the operating and capital costs
associated with desulfurizing FCC naphtha also increases.  FCC naphtha
is very rich in high-octane olefins.  As the severity of desulfurization
increases, more olefins are saturated, further sacrificing the octane
value of this stream and further increasing hydrogen consumption.  Also,
as desulfurization severity increases, there is an increase in the
amount of the removed sulfur (in the form of hydrogen sulfide) which
recombines with the olefins in the FCC naphtha, thus offsetting the
principal desulfurization reactions.  There are means to deal with the
recombination reactions; however, this probably means even greater
capital investments.  For example, the most expensive capital investment
for an FCC postreater is a two stage desulfurization unit.  A sulfur
standard less than 10 ppm would likely require more refiners to invest
in a second stage for their FCC postreater.

There are several other reasons which further increases the
desulfurization cost for a gasoline sulfur standard less than 10 ppm
beyond the higher FCC postreater cost.  Per   REF _Ref308595986  Table
4-2 , other refinery streams contain a very modest amount of sulfur, yet
a 5-ppm sulfur standard would likely require desulfurization of some of
these streams.  For example, we believe that to comply with a 5-ppm
gasoline sulfur standard, most refiners would need to treat the butane
blended into the gasoline pool.  Because refineries have different
sulfur levels in their non-FCC streams based on their feedstock sulfur
levels and their configurations, those with higher sulfur levels in
other refinery streams may have to desulfurize additional streams
besides butane.  Each additional individual gasoline stream that
requires desulfurization is incrementally a lot more expensive than
addressing the sulfur from the FCC unit because the amount of sulfur
reduction is a lot lower, but the capital costs are higher on a
per-barrel basis for lower volume gasoline blendstock streams. 
Furthermore, desulfurizing gasoline down to 5 ppm essentially removes
the flexibility offered by the 10-ppm gasoline sulfur standard with ABT
program.  Each U.S. refinery is in a different position today, both
technically and financially, relative to the other refineries.  In
general, they are configured to handle the different crude oils they
process and turn their crude oil slate into a widely varying product
slate to match their available markets.  Those processing heavier, sour
crudes would have a more challenging time reducing gasoline sulfur under
the proposed Tier 3 program.  Also, U.S. refineries vary greatly in size
(atmospheric crude capacities range from less than 5,000 to more than
500,000 barrels per day) and thus have different economies of scale for
adding capital to their refineries.  As such, it is much easier for some
refineries to get their sulfur levels below 10 ppm than for others to
reach 10 ppm.  This allows the ABT program to be used to reduce the cost
of the proposed gasoline sulfur standard.  If the gasoline sulfur
standard were to be 5 ppm, this would essentially end the ability of the
refiners to average sulfur reductions across their refineries thus
significantly increasing the costs while significantly reducing the
desulfurization flexibility.  

Going lower than 10 ppm would cause control costs to quickly escalate as
more challenged refineries would be forced into much larger investments.
 Our cost estimates for 5 ppm versus 10 ppm with averaging bears this
out.  We estimate the cost for a 10-ppm gasoline sulfur standard
(assuming intra-company credit trading) to be 0.89 ¢/gal compared to
1.38 ¢/gal for the 5-ppm standard.  The cost per sulfur reduction for
the 10-ppm average standard is 0.89 ¢/gal for the 20 ppm sulfur
reduction from Tier 2, or 0.045 ¢/ppm-gal.  The cost per sulfur
reduction for the 5-ppm standard is 0.49 ¢/gal for the 5-ppm sulfur
difference from the 10-ppm average standard, or 0.098 ¢/ppm-gal, which
is over 2 times higher.  Therefore, we believe that an annual average
standard of 10 ppm at the refinery gate with an ABT program appears to
be the point which properly balances feasibility with costs.  

In much of Europe and Japan, the gasoline sulfur level is capped at 10
ppm.  We, however, are not considering a 10-ppm cap for the U.S.  The
U.S. gasoline distribution system poses contamination challenges that
make it difficult to set and enforce tight downstream per-gallon sulfur
standards.  The U.S. gasoline distribution system poses contamination
challenges that make it difficult to set and enforce tight downstream
sulfur standards.  In Europe, Japan, and California, finished petroleum
products are generally shipped short distances directly from the
refinery to the terminal with limited susceptibility to contamination. 
The U.S. has the longest and most complex gasoline distribution system
in the world, making it harder to control sulfur contamination than in
other countries.  Petroleum products are shipped long distances through
multi-product pipelines.  Further, gasoline goes through the same
pipelines and terminals back-to-back with jet fuel (containing up to
2,000 ppm sulfur).  Products are often in the custody of a number of
separate companies before reaching the terminal.  This system is very
effective at delivering petroleum products to the bulk of the country,
but pipeline transport inevitably introduces the potential for sulfur
contamination of the gasoline being shipped through pipelines.  Gasoline
additives, needed to provide critical fuel performance characteristics
(e.g., corrosion control, demulsifiers), also contain varying levels of
sulfur which contribute to the overall sulfur content of gasoline. 
Therefore, we are proposing a 10-ppm average sulfur standard coupled
with higher per gallon caps at both the refinery gate and at all points
downstream, as currently exists under the Tier 2 program.  We believe
this is the most prudent approach for lowering in-use sulfur while
maintaining flexibility considering cost and other factors.  These
per-gallon caps are important in the context of an average sulfur
standard to provide an upper limit on the sulfur concentration that
vehicles must be designed to tolerate.  Since there are many
opportunities for sulfur to be introduced into gasoline downstream of
the refinery, these caps also limit downstream sulfur contamination and
enable the enforcement of the gasoline sulfur standard in-use.  For more
on our consideration of downstream caps, refer to Section   REF
_Ref307936639 \w  4.2.4.2 .  

Challenges with Lowering Today’s Sulfur Caps

Impacts of Lowering the 80-ppm Refinery Cap

We considered lowering the 80-ppm cap standard that applies to refiners
under the Tier 2 program.  If we were to lower the cap standard, we
analyzed lowering it to two different possible sulfur levels, either 50
ppm or 20 ppm.  If we lowered the refinery cap standard to 20 ppm, then
the averaging aspect of the ABT program would essentially not be
available to refiners.  That is because, under a 20-ppm cap standard, we
estimate that refiners would average about 10 ppm sulfur.  Thus, the
compliance scenario if the cap standard were 20 ppm would essentially be
the same as the non-ABT case we analyzed.  In this case, refiners would
not have much of the flexibility offered by the ABT program.   

If the cap standard were to be lowered to 50 ppm, the final compliance
scenario under the Tier 3 fuels program would be somewhere between the
ABT scenario that we analyzed and the non-ABT scenario that we analyzed
(probably much closer to the ABT case).  According to EPA batch data,
there were 20 refineries that averaged between 40 and 80 ppm sulfur
during 2009.  These refineries are using credits to show compliance with
the Tier 2 30-ppm gasoline sulfur standard.  If the 80-ppm cap were to
be reduced to 50 ppm, those refineries that were averaging over 40 ppm
would be forced to reduce their sulfur levels below the cap even if
their financial situation is more tenable compared to other refineries. 
However, even if the cap standard were to remain at 80 ppm, most of the
20 refineries that averaged between 40 and 80 ppm under Tier 2 would
have to lower their sulfur anyways because of the stringency of the
proposed 10-ppm sulfur standard.  There would not be sufficient credits
available to allow most of those refineries remain at high gasoline
sulfur levels.  Our cost analysis, which assumes intra-company credit
trading, projects that only one refinery would remain just above 40 ppm
when the fuels program is fully phased in.  For more on our cost
analysis, refer to Chapter 5 of the draft RIA.  

Another way that a more stringent cap would affect refiners would be to
restrict the ability of refiners to process high sulfur FCC naphtha when
there is a short term shutdown of the FCC postreater.  If the FCC
postreater goes down, the refinery would likely continue operating the
FCC unit and store up the high sulfur FCC naphtha.  Since the FCC
naphtha is too high in sulfur to blend directly with gasoline, the
refinery would have to either sell the material to other refiners, or
hydrotreat the stored up FCC naphtha along with the ongoing production
of high sulfur FCC naphtha once the FCC postreater was back online.  If
a stringent cap were in place, the refiner would have little room for
short term production of higher sulfur gasoline if it was feeding a
larger than normal quantity (stored and new production) of FCC naphtha
to the FCC postreater.  Without this flexibility, the refiner may have
to oversize the FCC postreater and FCC naphtha storage to ensure that,
regardless of the higher feed volume needed to process the stored
material, the FCC naphtha desulfurization unit could continue to
desulfurize the FCC naphtha down to the required sulfur level that would
result in 10 ppm sulfur in the gasoline pool.  If the cap were to be
lowered, a 50-ppm cap standard would clearly continue to provide
refiners with some flexibility while a 20-ppm cap would not.  Even if
refiners planned to tolerate some higher sulfur batches when
hydrotreating stored FCC naphtha, it could not tolerate much volume of
higher sulfur batched because of the need to average 10 ppm over the
calendar year.  If such outages happen very infrequently, then a small
amount of credits could regularly be banked over time that would allow
for some longer term higher sulfur batches of gasoline as the stored FCC
naphtha was being hydrotreated.  Alternatively, the averaging of sulfur
credits would help refiners with FCC naphtha hydrotreater outages. 
Thus, the flexibility of the ABT program coupled with a higher cap
standard would provide refiners with some flexibility to handle FCC unit
outages. 

Downstream Sulfur Caps  

The feasibility of complying with a downstream sulfur cap is dependent
on the differential between the refinery/importer gate sulfur cap and
the downstream cap.  This differential must provide sufficient
flexibility for worst-case situations when the potential sources of
sulfur addition downstream of the refinery/importer compound in a single
batch of gasoline that was introduced into the system at the
refinery/importer gate sulfur cap.

We are proposing two potential options for the per-gallon downstream
sulfur cap.  Under the first option, we are proposing to maintain the
current 95-ppm downstream sulfur cap.  This option is associated with
the proposed maintenance of the current 80-ppm refinery/importer gate
sulfur cap and is reflected in the draft regulatory text.  Under the
second option, we are proposing that the downstream sulfur cap would be
reduced to 65 ppm.  This option is associated with the proposed
reduction in the refinery/importer gate sulfur cap to 50 ppm.  Under
both of these options, we would be maintaining the current 15-ppm
differential between the refinery/importer gate sulfur cap and the
downstream sulfur cap.

We are also requesting comment on the potential implementation of a
downstream sulfur cap as low as 25 ppm.  This scenario is associated
with a reduction of the refinery/importer gate sulfur cap to as low as
20 ppm.  Under this scenario, the differential between the
refinery/importer gate sulfur cap and the downstream sulfur cap might be
a little as 5 ppm.  

Under all of these potential approaches, the downstream sulfur cap would
apply at all locations downstream of the refinery or importer gate
including the gasoline produced by transmix processors and after the use
of additives.  The potential sources of sulfur addition downstream of
the refinery/importer gate and issues associated with the feasibility of
meeting the downstream sulfur caps under consideration are discussed in
the following subsections.

 Sulfur Addition Downstream of the Refinery and Importer Gate  

The sulfur content of gasoline can increase downstream of the
refinery/importer due to contamination during distribution, the use of
additives, and the disposition of transmix generated during
distribution.  

A small amount of sulfur contamination takes place during distribution
as a result of the shipment of gasoline over long distances by pipeline
and other modes due to the sharing of the same distribution assets with
other higher-sulfur petroleum products, e.g., jet fuel.  Steps can be
taken to limit sulfur contamination.  However, it is an unavoidable
feature of the efficient multi-product distribution system in the U.S. 
We estimate that sulfur contamination of gasoline can be limited to a
worst case maximum of 3 or 4 ppm in the future, even for the most
involved and long-distance distribution pathways.

There are currently no direct regulatory controls on the sulfur content
of gasoline additives.  The contribution to the sulfur content of
finished gasoline from gasoline additives is accommodated in the
differential between the refinery gate and downstream sulfur caps.  The
functional components of some gasoline additives such as silver
corrosion inhibitors and demulsifiers are inherently high in sulfur
content.  However, the contribution to the overall sulfur content of the
finished fuel is very limited.  For example, silver corrosion inhibitors
can contain as much as 30 percent sulfur but because of very low
treatment rates can add only 0.17 ppm to the sulfur content of the
finished fuel.  At seldom used highest treatment rates, the use of
gasoline additives upstream of the consumer has the potential to add ~1
ppm to the sulfur content of the finished fuel.  Aftermarket additives
that are added directly into the vehicle fuel tank also have the
potential to increase gasoline sulfur content.  One particular
aftermarket performance and anti-wear additive can contribute ~2 ppm
sulfur to the treated fuel.

Transmix is a necessary byproduct of the multi-product refined product
pipeline distribution system.  Batches of different products are shipped
in sequence in pipelines without any physical barrier between the
batches.  Transmix is produced when the mixture at the interface between
two adjacent products cannot be cut into either batch.  Transmix
typically accumulates at the end of pipeline systems far from
refineries.  There are two methods of disposing of transmix.  Most
transmix is sent to transmix processing facilities for separation into
salable distillate and gasoline products through use of a simple
distillation tower.  

The other means of transmix disposal is for pipeline operators to blend
small quantities directly into batches of gasoline during shipping. 
This typically takes place at remote pipeline locations where small
volumes of transmix accumulate that would be difficult to consolidate
and ship to transmix processors.  Pipeline operators that blend transmix
into the gasoline in their systems must ensure that the resulting
gasoline meets all fuel quality specifications and the endpoint of the
blended gasoline does not exceed 437 °F.  This practice currently can
add as much as 3 to 5 ppm to the sulfur content of gasoline although we
believe that the contribution is typically less.  

Transmix processing facilities do not handle sufficient volumes to
support the installation of currently-available desulfurization units. 
Therefore, the sulfur content of the products they produce is
predominantly governed by the sulfur content of the transmix they
receive.  In many cases, transmix contains jet fuel which can have a
sulfur content as high as 3,000 ppm.  Due to the overlapping
distillation characteristics of jet fuel and gasoline, it is unavoidable
that some jet fuel in transmix will be present in the gasoline produced
by transmix processors.

Transmix processors produce ~0.1 percent of all gasoline consumed in the
U.S.  The small volume of transmix-derived gasoline along with the fact
that such gasoline is typically mixed with other gasoline before
delivery to the end user, substantially limits the potential impact on
gasoline sulfur levels.  Furthermore, data provided by the largest
operator of transmix processing facilities, shown in   REF _Ref309138805
 Figure 4-2 , indicates that relatively few batches of the gasoline they
produce approach 80 ppm sulfur.  Most batches are approximately 10 ppm
above the current 30-ppm refinery sulfur average.  We anticipate that
this 10-ppm differential would likely continue under the proposed 10-ppm
refinery average sulfur standard.

Figure   STYLEREF 1 \s  4 -  SEQ Figure \* ARABIC \s 1  2  Kinder Morgan
Transmix Gasoline Product Sulfur Levels

Maintaining the Current 15-ppm Differential Between the Refinery
/Importer Gate and Downstream Sulfur Caps

Under both of the co-proposals for a downstream sulfur cap (95 ppm and
65 ppm), we would be maintaining the current 15-ppm differential between
the refinery/importer gate sulfur cap and the downstream sulfur cap.

The current 15-ppm differential was established under the Tier 2 program
to accommodate the sulfur contamination during distribution, the sulfur
contribution from transmix blending by pipeline operators, the sulfur
contribution from the use of additives, and to enable compliant gasoline
to be produced by transmix processors.  Transmix processors need to
produce gasoline sufficiently below the downstream sulfur cap to
accommodate the addition of sulfur from the use of additives and
contamination during further distribution.  Experience under the Tier 2
program has shown that a 15-ppm differential is sufficient for
downstream parties to ensure compliance with the downstream sulfur cap

Our co-proposal to maintain the current 95-ppm downstream sulfur cap
with an 80-ppm refinery/importer gate sulfur cap represents no change
from current requirements.  As a result, there would be no increased
difficulty or additional costs associated with satisfying a 95-ppm
downstream sulfur cap beyond those that were already incurred under the
Tier 2 program. 

Our co-proposal to implement a 65-ppm downstream sulfur cap with a
50-ppm refinery gate sulfur cap would also maintain the current 15-ppm
differential between these sulfur caps under the Tier 2 program.  Since
it is this differential that determines the difficulty in complying with
the downstream sulfur cap, we expect that there would be no operational
changes and additional costs for downstream parties associated with
satisfying a 65-ppm cap downstream sulfur cap beyond those that were
already incurred under the Tier 2 program to comply with the current
95-ppm downstream sulfur cap.

Under both of the co-proposals, the reduction in the refinery average
sulfur standard may make it somewhat easier to comply with the
downstream sulfur cap given that most gasoline produced would be at or
near 10 ppm sulfur.

Potential Reduction in the Differential Between the Refinery/Importer
Gate and Downstream Sulfur Caps

We requested comment on the potential implementation of a refinery gate
sulfur cap as low as 20 ppm and a corresponding downstream sulfur cap as
low as 25 ppm.  This was driven by vehicle manufacturers concerns about
the potential impacts on emissions performance if vehicles are exposed
to gasoline above the proposed 10-ppm refinery average standard.  As
discussed in Sections 1.2 and 5.2 of this draft RIA, the vehicle
emissions benefits associated with today’s proposal are driven by the
proposed reduction in the average sulfur content of gasoline from 30 to
10 ppm.  We believe that the potential benefits from further reductions
in the sulfur caps would be minimal.  However, further reductions in the
sulfur caps could have negative impacts on refiners and downstream
parties.  

The potential impacts on refiners and additional costs associated with a
lower refinery/importer gate sulfur cap discussed in Section   REF
_Ref307467358 \w  4.2.4.1 .  A reduction in the differential between the
refinery gate sulfur standard and the downstream sulfur standard could
also result in negative impacts and additional costs to downstream
entities.  Reducing the current 15-ppm differential between the
refinery/importer gate and downstream sulfur caps could limit the
ability of transmix processors to continue to produce finished gasoline,
limit the ability of pipeline operators to continue to blend transmix in
gasoline, and potentially require the direct regulation of gasoline
additive sulfur content which might cause certain gasoline additives to
be removed from the market. 

Some gasoline additive manufactures relate that it would not be
technically possible to reformulate their additives to meet a lower
sulfur cap.  Hence, the implementation of a sulfur cap for gasoline
additives could result in the withdrawal of some necessary and
cost-effective gasoline additives (e.g., corrosion inhibitors, and
demulsifiers) from the market.  Other additive manufactures related that
there would be significant costs in reformulating their additives to
meet a lower sulfur cap.  Some additive manufactures related that they
could not justify the cost of reformulation and would need to cease
manufacture.

Since gasoline additives may add as much as 3 ppm to the sulfur content
of the finished fuel, allowing for further sulfur contamination during
distribution and for test variability means that transmix processors
must produce gasoline about 5 ppm below the downstream sulfur cap.  The
sulfur levels in the transmix that processors must cope with would be
reduced due to the proposed reduction the gasoline sulfur requirements
for refiners/importers.  However, the continued presence of high-sulfur
jet fuel in transmix would continue to significantly influence the
sulfur content of the gasoline produced by transmix processors.  Given
these considerations, a reduction in the differential between the
refinery gate sulfur cap and the downstream sulfur cap might require
that the majority of the gasoline produced by transmix processors to be
desulfurized, something that is cost-prohibitive to do at transmix
processing facilities today.  Other options for dealing with transmix,
however, are just as impractical, including shipping it back to
refineries for reprocessing.  Refiners are typically averse to accepting
transmix into their facilities for reprocessing due to technical,
logistical, and economic constraints.  In addition, transmix would
typically need to be shipped long distances from the ends of the product
distribution by truck to reach a refinery.

If pipeline operators were further limited in their ability to blend
small amounts of transmix into gasoline due to a reduction in the
differential between the refinery/importer gate and downstream sulfur
caps, they could be compelled to install additional transmix storage and
shipping facilities at numerous remote locations to facilitate the
shipment of small volumes of transmix to transmix processors by truck. 
One major pipeline operator reported that it has over 100 locations on
its system where transmix can be injected into gasoline, some of which
do not have tank truck access.  Thus, a reduction in the differential
between the refinery/importer gate and downstream sulfur caps could
result in substantial additional costs and potential changes to transmix
operating practices for the pipeline operators.

Lead Time Assessment

Engineering and Construction Analysis

Given the complexity of gasoline refining, numerous planning and action
steps would be required  for refiners to complete the refinery changes
needed to comply with the proposed Tier 3 sulfur standards.  The steps
required to implement these changes include: the completion of scoping
studies, financing, process design for new or revamped refinery units or
subunits, permitting, detailed engineering based upon the process
design, field construction of the gasoline sulfur reduction facilities,
and start-up and shakedown of the newly installed desulfurization
equipment.

We conducted a more thorough lead time analysis in which we sequenced
the estimated time to complete scoping studies, process design,
permitting, detailed engineering, field construction, and start-up and
shakedown in advance of production based upon the methodology used in
our recent gasoline and diesel rules.

For the proposed Tier 3 gasoline sulfur program, we estimated refinery
lead times required for two general types of refinery projects: the
construction of new grassroots FCC postreaters and the revamp of
existing pre and postreaters.  For each refinery project, we estimated
lead times for scoping studies, process design, permitting, detailed
engineering, field construction, and start-up and shakedown.  Estimated
required lead times for scoping studies are six months.  Process design
ranged from six months for desulfurization equipment revamping to nine
months for a grassroots postreater.  Based on discussions with refiners,
a review of the permitting experience for Tier 2 and our current
analysis, we conservatively estimated that permitting for
desulfurization equipment revamping and the construction of a grassroots
postreater would to take 9 months.  However, we estimate the overall
lead-times for Tier-3-related revamps to be considerably shorter, as
described below. The estimates for permitting time are consistent with
those of EPA’s Office of Air Quality Planning and Standards (OAQPS)
and our regional offices, both of which have engaged in extensive dialog
with potentially affected parties.  A discussion of the permitting
implications of Tier 3 is contained in Section V.B of the preamble. 
Detailed engineering efforts were estimated to require six months for
desulfurization equipment revamping and nine months for grassroots
postreaters.  Field construction was estimated to require six months for
revamped pre-and postreaters and 12 months for grassroots postreaters. 
Start-up and shakedown processes were estimated to require six months
for revamped FCC treaters and 9 months for grassroots postreaters. 
There is some degree of overlap among each of these steps as shown in
Table 4-3.

To allow refiners to complete all these different steps and comply with
the 10 ppm average gasoline sulfur standard, assuming the Tier 3
proposal were to be finalized by the end of 2013, we would be providing
three years of lead time.  In addition to the three years of lead time,
the proposed rulemaking also provides additional flexibility provided by
the ABT program, small refinery delays, and hardship provisions.  To
support this timeline, we conducted several analyses of the expected
refinery lead time requirements associated with the proposed Tier 3
standards and found that refinery operators would have more than
adequate time to implement the required refinery charges.  A
justification for proposed timeline appears below.

Complying with Tier 3 is expected to involve some grassroots (new) FCC
postreaters, but mostly we believe that refiners will revamp existing
FCC postreaters.  Revamping of existing FCC postreaters can be
accomplished in approximately 2 years, or less (See Table 4-3) 
Grassroots FCC postreaters are expected to require on average about
three-years to install and start-up (See Table 4-3).  In comparision to
FCC pretreaters, hydrocrackers and distillate hydrotreaters, FCC
postreaters are much less costly, low pressure units that take less time
to scope out, require shorter lead times for ordering the equipment, and
less time to install.  Furthermore, the grassroots FCC postreaters to be
installed for Tier 3 are expected to be in a moderate to light
desulfurization service because the refineries they will be installed in
will already be complying with Tier 2 using an FCC pretreater.  FCC
naphtha from a refinery with an FCC pretreater is expected to only
contain about 100 ppm sulfur.  To comply with Tier 3, refiners
installing these grassroots FCC postreaters would only need to
desulfurize the FCC naphtha down to 25 ppm (about a 75% reduction).  In
comparison, a single-stage FCC postreater would have to desulfurze FCC
naphtha from as high as 2400 ppm sulfur down to 25 ppm, a 99% sulfur
reduction.  The more moderate desulfurization service of the grassroots
FCC postreaters needed to comply with Tier 3 would be expected to
streamline the scoping and design work.

Table   STYLEREF 1 \s  4 -  SEQ Table \* ARABIC \s 1  3  Anticipated
Compliance Timelines

	Months

	0-

3	3-

6	6-

9	9-12	12-15	15-18	18-21	21-24	24-27	27-30	30-33	33-36	36-39	39-42	42-45

Revamp Pre- & Postreaters	Scoping Studies

















Process

Design

















Permitting

















Detailed Engineering

















Field Construction

















Start-up / Shakedown















	Grassroots Postreater	Scoping Studies

















Process

Design

















Permitting

















Detailed Engineering

















Field Construction

















Start-up / Shakedown















	

It is useful to compare the proposed lead time for Tier 3 to what was
provided for Tier 2.  In the case of the Tier 2 standard, we provided a
three-year lead time along with an ABT program and other flexibilities
to ease compliance.  Refiners, though, commented that the three year
timeline that we provided was not enough time.  For the Tier 2 analysis,
we assumed that refiners would solely install low-pressure FCC
postreaters, which we believe could be scoped out, designed, installed
and started up within a 3 year time period.  However, many refiners
complied with Tier 2 by installing high-pressure FCC pretreaters which
require long lead times for the procurement of the required equipment. 
Furthermore, those refiners that did not install high-pressure FCC
pretreaters instead installed grassroots FCC postreaters, many of which
were designed for severe desulfurization service.  An additional
difference between Tier 3 and Tier 2 is that for Tier 3 we expect the
installation of only 16 grassroots units, along with many revamps, but
for Tier 2 virtually all refiners installed both grassroots FCC
pretreaters and postreaters.  The demands on the desulfurization vendors
for scoping studies, and on the E & C industry for design and
construction, and on the refiners to train their operations staff and
start up the new units, was a lot greater for Tier 2 than what we would
expect for Tier 3.  The total estimated investment cost for Tier 2
versus Tier 3 also highlights the difference in investment demands 

The total investment for Tier-2 desulfurization processing units was
estimated to be about $6.1 billion, while the total investment for
Tier-3 desulfurization processing units is estimated to be about $2.1
billion.  This simple comparison indicates that the proposed Tier 3 lead
time should be adequate for refineries to obtain necessary permits,
secure engineering and construction (E&C) resources, install new
desulfurization equipment and make all necessary retrofits to meet the
proposed sulfur standards.  

We assessed the permitting situation in more detail working in
conjunction with the Office of Air Quality, Planning and Standards
(OAQPS).  On a refinery-by-refinery basis, we provided OAQPS estimates
of the additional heating demands for the new and revamped units per the
desulfurization vendor submissions.  OAQPS was able to project which
refineries would likely trigger NOx, particulate matter and greenhouse
gas emission permitting limits, which would likely lengthen the
permitting process as refiners would need to offset the projected
emission increases.  As it turns out, only 7 of the 16 refineries which
are projected to install grassroots units were projected to exceed
particular permitting limits, and these solely did so based on the most
conservative assumption that each would produce all the additional
hydrogen on site using hydrogen plants (as opposed to using existing
reforming capacity) and produce the electricity on site, to satisfy the
needs of the new desulfurization equipment.  When we provided a second
heat demand estimate to OAQPS which assumes that refiners purchase their
hydrogen and electricity from third parties, none of the refineries
which we projected would install grassroots units was projected to have
emission increases which would require offsets.  Thus, many of the
grassroots units that we project would be installed may end up with a
streamlined permitting processe.

The various flexibilities that the proposed Tier 3 rule provides to
refiners provide refiners additional time for complying.  These
flexibilities include the ABT program, the small refiner delay
provisions and the hardship provisions.  The ABT program allows a
refiner, either within its own company or by purchasing credits on the
open market, to delay higher investment cost investments, such as the
investments in grassroots FCC postreaters, which would provide
additional lead time for installing these units.  This would occur if
refiners would reduce the sulfur levels of their gasoline through
operational changes or revamps of their existing FCC pretreaters and
postreaters when the ABT Program begins in 2014.  Potentially every
refinery with either an FCC pretreater or an FCC postreater may be
capable of generating early credits.  Furthermore, we project that 66
refineries would revamp their existing FCC postreaters to comply with
Tier 3.  Since revamps can be completed within two years or less, these
refiners could potentially begin generating early credits during 2016,
or before if refiners begin each of these revamps in early 2014.  During
the period between 2014 and 2017, these refineries which reduce their
gasoline sulfur levels below that required by Tier 2 would generate
credits.  Refineries with higher cost capital investments, such as the
grassroots FCC postreaters, could then delay making those investments
through the purchase of credits.  We estimate that sufficient credits
could be generated early to allow many refineries to delay compliance
until as late as 2020.   The quantitative early credit analysis that we
conducted showed that if refiners with an existing pretreater or
postreater would generate early credits by lowering their gasoline
sulfur down to 20 ppm starting in 2014 and if revamps were started up in
2016, one year before the program start date, that almost 6 times more
credits would be available to offset the early credit demand by the
refiners installing grassroots postreater units, assuming that they
start up those units in 2018.  Even if all grassroots postreaters were
assumed to not start up until 2020, there would be almost 4 times more
early credits available to those refiners installing grassroots
postreaters assuming that the same early credit generation scenario
would occur

Additional flexibility is also provided by the small refineries
provisions which delays compliance for the refineries which refine less
than a net of 75,000 barrels of crude oil per day until 2020.  Three of
the 16 FCC postreater grassroots units that we project will be installed
would be by small refineries.  However, small refineries could also
decide to comply early and generate credits starting as early as 2014. 

As in previous fuel programs, we are proposing hardship provisions to
accommodate a refiner’s inability to comply with the proposed standard
at the start of the Tier 3 program, and to deal with unforeseen
circumstances that may occur at any point during the program.  These
provisions would be available to all refiners, small and non-small,
though relief would be granted on a case-by-case basis following a
showing of certain requirements; primarily that compliance through the
use of credits was not feasible.  We are proposing that any hardship
waiver would not be a total waiver of compliance; rather, a hardship
waiver would be short-term relief that would allow a refiner facing a
hardship situation to, for example, receive additional time to comply.  
This hardship provision would allow a refiner to seek a delay in the
case that there was insufficient time to comply. 

Finally, we believe that in reality, less leadtime than shown in Table
4-3would actually be necessary.  We held discussions with many refiners
during most of 2011, and so they have been well aware of Tier 3 and are
familiar with the likely requirements.  During our subsequent
discussions with technology vendors and engineering firms, they
explained to us that many refiners have already initiated, and by now,
likely completed their scoping studies. Thus, actual time needed for
designing, installing and starting of new desulfurization equipment for
Tier 3 times would even be less than what we projected because many
refineries may have already completed required scoping studies in
anticipation of the Tier-3 standards.  Moreover, lead times for those
refineries that have yet to start the scoping process can also be
expected to decrease, since fewer refineries will be competing for the
services of the desulfurization vendors.

Permitting Analysis

Our analysis found that GHG emission increases were the most common
reason that Prevention of Significant Deterioration (PSD) applicability
would be triggered, followed by NOX emissions.  Specifically, 19
refineries appeared likely to have significant emissions for one or more
pollutants and thus would trigger major source New Source Review (NSR).
Of these 19 refineries, 13 refineries would need permits for
NAAQS-related pollutants. 

With respect to NAAQS-related pollutants, 6 of these13 refineries were
predicted to require both PSD and Nonattainment NSR permits.  Of the
remaining 7 refineries, 2 required only a Nonattainment NSR permit while
the remaining  5 refineries required a PSD permit. 

In comparison, for the Tier 2 program, EPA expected the need for
NAAQS-related NSR permits might be widespread among refineries.  For the
proposed Tier 3 gasoline sulfur standard, however, only about 10
refineries would need air permits that address NAAQS pollutants.  

This number could be lower if those refineries apply emission controls,
such as selective catalytic reduction (SCR) for NOX, to reduce the
emission increases below the significance level.  For refineries that do
need a major source NSR permit for NAAQS pollutants, the permitting
process is expected to take 9 - 12 months.  For an in depth assessment
of stationary source implications, refer to Section V.B of the preamble.
 

Employment Constraint Analysis

As in prior rules, we also evaluated the capability of E&C industries to
design and build gasoline hydrotreaters as well as performing routine
maintenance.  This includes an employment analysis.  Two areas where it
is important to consider the impact of the fuel proposed sulfur
standards are: 1) refiners’ ability to procure design and construction
services and 2) refiners’ ability to obtain the capital necessary for
the construction of new equipment required to meet the new quality
specification.  We evaluated the requirement for engineering design, and
construction personnel, in a manner consistent with the Tier 2 analysis,
particularly for three types of workers: front-end designers, detailed
designers and construction workers, needed to implement the refinery
changes.  We developed estimates of the maximum number of each of these
types of workers needed throughout the design and construction process
and compared those figures to the number of personnel currently employed
in these areas.  

The number of job person-hours necessary to design and build individual
pieces of refinery equipment and the job person-hours per piece of
equipment were taken from Moncrief and Ragsdale.  Their paper summarizes
analyses performed in support of a National Petroleum Council study of
gasoline desulfurization, as well as other potential fuel quality
changes.  The design and construction factors for desulfurization
equipment are summarized in   REF _Ref307578495  Table 4-4 .

Table   STYLEREF 1 \s  4 -  SEQ Table \* ARABIC \s 1  4  Design and
Construction Factorsa

Gasoline Refiners

Number of New Pieces of Equipment per Refinery	60

Number of Revamped Pieces of Equipment per Refinery	15

Job Hours Per Piece of New Equipmenta

Front End Design	300

Detailed Design	1,200

Direct and Indirect Construction	9,150

Note:

a Revamped equipment estimated to require half as many hours per piece
of equipment

Refinery projects will differ in complexity and scope.  Even if all
refiners desired to complete their project by the same date, their
projects would inevitably begin over a range of months.  Thus, two
projects scheduled to start up at exactly the same time are not likely
to proceed through each step of the design and construction process at
the same time.  Second, the design and construction industries will
likely provide refiners with economic incentives to avoid temporary
peaks in the demand for personnel.  

Applying the above factors, we projected the maximum number of personnel
needed in any given month for each type of job.  The results are shown
in   REF _Ref307578574  Table 4-5 .  In addition to total personnel
required, the percentage of the U.S. workforce in these areas is also
shown, assuming that half of all projects occur in the Gulf Coast in  
REF _Ref307578574  Table 4-5 .  Very few refineries are expected to
require the full 45-month period to complete scoping studies, process
design, permitting, detailed engineering, field construction, and
start-up/shakedown.

Table   STYLEREF 1 \s  4 -  SEQ Table \* ARABIC \s 1  5  Maximum Monthly
Demand for Personnel

	Front-End Design	Detailed Engineering	Construction

Tier 3 Gasoline Sulfur Program

Number of Workers	202	809	6,012

Percentage of Current Workforcea	11%	9%	4%

Note:

a Based on current employment in the U.S. Gulf Coast assuming half of
all projects occur in the Gulf Coast

To meet the proposed Tier 3 sulfur standards, refiners are expected to
invest $2.2 billion between 2014 and 2019 and utilize approximately
1,000 front-end design and engineering jobs and 6,000 construction jobs.
 The number of estimated jobs required is small relative to overall
number available in the U.S. job market.  As such, we believe that four
years is adequate lead time for refineries to obtain necessary permits,
secure E&C resources, install new desulfurization equipment and make all
necessary retrofits to meet the proposed sulfur standards.  

ABT Impacts

We conducted a refinery-by-refinery analysis to determine the impacts on
refinery E&C demand of implementing the 10-ppm standard without an ABT
program.  The analysis suggests that a greater number of refineries
would need to make investments in refinery apparatus and upgrades than
would have under an ABT program.  This would result in a greater demand
on the E&C industry.  Moreover, the analysis also indicated that the
demand upon the E&C industry would be spread over a shorter period than
with the ABT case.  In particular, our refinery-by-refinery analysis
indicates that without an ABT program, 73 refineries would revamp
existing pre- and postreaters and 21 would install grassroots
postreaters in order to meet the proposed sulfur standards.  The
remaining 17 refineries are either already in compliance with the 10-ppm
standard or expected to comply with simple process changes.  This is
compared to 66 refineries that would revamp existing pre- and
postreaters and 16 refineries that would install grassroots postreaters
in order to meet the proposed sulfur standards under an ABT program.

References

  Sulfur interferes in the function of the precious metal catalyst used
in the reforming process.  As a result, refiners historically have
desulfurized the heavy straight run naphtha feed to the reformer from
several hundred ppm sulfur down to less than 1 ppm.

 Aftermarket additives are defined as additives sold to vehicle
operators for direct addition to vehicle fuel tanks.  

 437 F is the maximum endpoint allowed for gasoline in the ASTM
International specification for gasoline in ASTM D4814. 

Chapter 1 

*** E.O. 12866 Review – Revised Version – Do Not Cite, Quote, or
Release During Review***

Page   PAGE  2 

  PAGE   \* MERGEFORMAT  4-5 

 Meyers, Robert A., Handbook of Petroleum Refining Processes, McGraw
Hill, 1997.

 40 CFR 80 Subpart H

 Shorey, Scott W., AM – 99-55, Exploiting the Synergy Between FCC and
Feed Pretreating Units to Improve Refinery Margins and Produce
Low-Sulfur Fuels, National Petroleum and Refiners Association’s 1999
Annual Meeting.

 Conversation with Woody Shiflett, Advanced Refining Technologies,
October 2011.

 Barletta, Tony, Refiners must optimize FCC feed hydrotreating when
producing low-sulfur gasoline, Oil and Gas Journal, October 14, 2002.

 Conversation with Woody Shiflett, Advanced Refining Technologies,
October 2011.

 Shorey, Scott W., AM – 99-55, Exploiting the Synergy Between FCC and
Feed Pretreating Units to Improve Refinery Margins and Produce
Low-Sulfur Fuels, National Petroleum and Refiners Association’s 1999
Annual Meeting.  

 Brunet, Sylvette, On the hydrodesulfurization of FCC gasoline:  a
review, Applied Catalysis A: General 278 (2005) 143 – 172.

 Leonard, Laura E., Recombination: A Complicating Issue in FCC Naphtha
Desulfurization, Prepared for the AIChE 2006 Spring National Meeting,
April 26, 2006

 Petroleum Refinery Process Economics, Maples, Robert E., PennWell
Books, Tulsa,  Oklahoma, 1993.

 Nocca, J.L., et al, Cost-Effective Attainment of New European Gasoline
Sulfur Specifications within Existing Refineries, November 1998.

 Prime G, A Sweet Little Process for Ultra-Low Sulfur FCC Gasoline
without Heavy Octane Penalty, IFP Industrial Division.

 Debuisschert, Quentin, Prime G+ Update, 12th European FCC Conference
– Grace Davidson Seminar, Seville Spain, May 2004.

 Beck, J.S., Advanced Catalyst Technology and Applications for Higher
Quality Fuels and Fuels, Prepr. Pap. Am Chem Soc., Div. Fuel Chem, 2004
49(2), 507.

 McGihon, Ron, Exxon Mobil, FCC Naphtha Desulfurization – New
Developments, Presentation at the 2009 Technology Conference, October 5
&6, Dubai, United Arab Emirates.

 Ellis, E.S., Meeting the demands of low sulfur gasoline, Petroleum
Technology Quarterly Spring 2002.

 Successful Start-Up of New Scanfining Unit at Statoil’s Mongstad
Refinery, November 19, 2003.

 Greeley, J.P., Zaczepinski, S., Selective Cat Naphtha Hydrofining with
Minimal Octane Loss, NPRA 1999 Annual Meeting (this document available
from docket A-97-10).

 Halbert, Thomas R., Technology Options for Meeting Low Sulfur Mogas
Targets AM-00-11, Presented at the 2000 Annual Meeting of the National
Petrochemical and Refiners Association, March 2000.

 McGihon, Ron, Exxon Mobil, FCC Naphtha Desulfurization – New
Developments, Presentation at the 2009 Technology Conference, October 5
&6, Dubai, United Arab Emirates.

 Refining Processes 2004, Hydrocarbon Processing.

 Upson, Lawrence L., Low-sulfur specifications cause refiners to look at
hydrotreating options, Oil and Gas Journal, December 8, 1997.

 Krenzke, David L., Hydrotreating Technology Improvements for
Low-Emissions Fuels AM-96-67, Presented at the 1996 Annual Meeting of
the National Petrochemical and Refiners Association, March 1996.

 UOP SelectFiningTM Process – New Technology for FCC Naphtha HDS, 2009


 CDTECH, FCC Gasoline Sulfur Reduction, CDTECH, Sulfur 2000, Hart’s
Fuel and Technology Management, Summer 1998.

 Rock, Kerry J., Putman, Hugh, Global Gasoline Reformulation Requires
New Technologies, Presented at Hart’s World Fuels Conference, San
Francisco, March 1998.

 Rock, Kerry L., et al, Improvements in FCC Gasoline Desulfurization via
Catalytic Distillation, Presented at the 1998 NPRA Annual Meeting, March
1998.

 Greenwood, Gil J., Next Generation Sulfur Removal Technology AM-00-12,
Presented at the 2000 NPRA Annual Meeting, March 2000.

 Meier, Paul F., S Zorb Gasoline Sulfur Removal Technology – Optimized
Design AM-04-14, Presented at the 2004 NPRA Annual Meeting, March 2004.

 Printed Literature by Phillips Petroleum Shared with EPA September
1999.

 Patal, Raj, Advanced FCC Feed Pretreatment Technology and Catalysts
Improves FCC Profitability AM-02-58, Presented at the 2002 NPRA Annual
Meeting, March 2002.

 Conversation with Woody Shiflett of Advanced Refining Technologies
October 2011.

 Letter to Margo Oge, EPA, from Mike Ricca, Baker Hughes, July 25, 2011.

 Letter to Caryn Muellerleine, EPA, from Richard Kelly, Marvel Oil
Company, July 13, 2011.

 The requirements for transmix blenders are contained in 40 CFR
80.84(d).

 Graphs of transmix gasoline product sulfur levels at Kinder Morgan
transmix processing facilities e-mail from James Holland, Kinder Morgan,
August 24, 2011.  

 Letter from the Alliance of Automobile Manufacturers to Administrator
Lisa Jackson, October 6, 2011.

 Moncrief, Philip and Ralph Ragsdale, “Can the U.S. E&C Industry Meet
the EPA’s Low Sulfur Timetable,” NPRA 2000 Annual Meeting, March
26-28. 2000, Paper No. AM-00-57.

