103

NUM B E R 1 0 3 S pr i n g 200 8
The GENESISTM Catalyst System
Maximize yield with

Grace Davison’s
GENESISTM FCC catalyst.
Most refiners need flexible catalyst systems that allow them to
take advantage of changing economic scenarios. Grace Davison
delivers this flexibility with the GENESISTM catalyst system. GENESISTM
catalysts provide a means to maximize yield potential through optimization
of discrete cracking catalyst functionality. GENESISTM catalysts are a blend of
two catalyst types. The key component is MIDAS®, which maximizes conver-
sion of bottoms and improves coke selectivity by eliminating coke precursors.
The other GENESISTM component is most often an IMPACT® catalyst. The
inclusion of IMPACT® provides critical zeolite surface area and activity as well
as superior coke and gas selectivity in a broad range of applications, from
severely hydrotreated gasoils to heavy resid feeds.
GENESISTM catalysts demonstrate a true yield synergy with a superior

coke to bottoms relationship than either component alone. The synergy
exists because each component cracks specific feed species.
GENESISTM catalyst provides

the ultimate in formulation
and FCC operational flexibility.
For more information, contact your Grace Davison Technical Sales Manager
www.grace.com 7500 Grace Drive • Columbia, MD 21044 USA
A message from the editor...
This issue of the Catalagram once again finds us discovering the synergies between Grace
Davison and ART products and technologies and applying them to the challenges refiners face.
Our lead article, “New Opportunities for Co-Processing Renewable Feeds in Refinery Process”,
presented at the 2008 NPRA Annual Meeting, is a collaboration between our Refining
Technologies, ART, and Biofuels Technologies groups. The article explores the ramifications of
co-processing renewable fuels in a conventional FCC or DHT unit with traditional straight run
diesel or cat feed. Extensive pilot plant testing over a wide range of feed blends and conditions
resulted in valuable data that can aid refiners in designing the optimum configuration to maxi-
mize their profitability.
More exciting news about the commercial performance of our innovative catalysts continues to
come in from the field. We have introduced two new high activity GENESIS™ catalyst compo-
nents and have made improvements to our MIDAS® catalyst and PINNACLE® catalyst families,
which will deliver higher activity and improved gasoline selectivity at equivalent bottoms and
coke yields. ART’s 420DX™ catalyst, the newest member of the ultra high activity DXTM catalyst
series, is expected to exceed refinery expectations in its ability to tolerate difficult feed blends in
demanding ULSD applications.
Please join us in welcoming our new Vice President and General Manager for Refining
Technologies, Shawn Abrams. Shawn will run the global FCC business, while Bob Bullard con-
tinues as Managing Director of ART and focuses his Davison responsibilities on joint ventures
across the businesses. This added depth to our team is yet another way that Grace Davison
and Advanced Refining Technologies demonstrate our strong commitment to the future of the
refining industry.
Joanne Deady
Vice President
Marketing/R&D
Grace Davison Refining Technologies
IN THIS ISSUE
NU M B E R 1 0 3 Spring 2008
New Opportunities for Co-Processing Renewable Feeds in Refinery Processes
By Brian Watkins, Supervisor, Laboratory Technology, Advanced Refining Technologies; Charles
1
Olsen, Worldwide Technical Services Manager, Advanced Refining Technologies; Kevin Sutovich,
Senior R&D Chemist, Grace Davison Refining Technologies; Natalie Petti, Vandelay Management
In the ever growing market of transportation fuels, we demonstrate the effectiveness of
an ART hydroprocessing catalyst and state-of-the-art FCC catalysts in co-processing
renewable oils (biofeeds) as possible new feedstock opportunities for hydrotreaters and
FCC units.
GENESISTM Catalyst Commercial Update

By Rosann K. Schiller, Product Manager; Doc Kirchgessner, Technical Sales Manager; Kelly
15
Stafford, Technical Sales Coordinator, Grace Davison Refining Technologies
CATALAGRAM 103 GENESISTM catalysts provide flexibility to refiners by offering high activity catalyst
options for maximizing bottoms cracking while allowing refiners to operate within opti-
Spring 2008 mum fluidization characteristics.
Managing Editor: Effect of Hydrocarbon Partial Pressure on Propylene Production in the FCC
By Ruizhong Hu, Sr. Principal Scientist; Gordon Weatherbee, Principal Engineer; Hongbo Ma, 22
Joanne Deady Research Engineer; Terry Roberie, Director FCC Evaluations; Wu-Cheng Cheng, Director R&D,
Grace Davison Refining Technologies
Contributors: We have been able simulate the effect of varying hydrocarbon partial pressure in the
FCC unit in the Grace Davison DCR Unit. The results indicate that increasing the
Mike Beshara
hydrocarbon partial pressure increases the hydrogen transfer activity. The result is a
Alan Birch
decrease in the olefinicity and olefins yield of LPG and gasoline, and a decrease in the
Wu-Cheng Cheng
concentration of gasoline sulfur species.
Ruizhong Hu
Garry Jacobs
Meet Clean Fuels Challenges with Advanced Refining Technologies
Al Jordan
Newest ULSD Catalysts – ART 420DXTM 33
Adam Kasle
By Brian Watkins, Supervisor, Laboratory Technology; Charles Olsen, Worldwide Technical
Doc Kirchgessner Services Manager; Dave Krenzke, Technical Services Manager, Advanced Refining Technologies
Ernst Köhler ART has developed a new technology catalyst that will provide the refiner enhanced
Dave Krenzke sulfur removal activity with greater flexibility in meeting their HDS activity requirements
Hongbo Ma while minimizing hydrogen consumption.
Charles Olsen
Natalie Petti
Ben Prins Successful Implementation of State-of-the-Art ULSD/Dewaxing Technology
Terry Roberie
Greg Rosinski
at Irving Oil, Saint John, NB
By Mike Beshara, Project Manager, Irving Oil; Greg Rosinski, Technical Services Engineer,
36
Rosann K. Schiller Advanced Refining Technologies; Charles Olsen, Worldwide Technical Services Manager,
Kelly Stafford Advanced Refining Technologies; Ben Prins, Senior Process Engineer, Fluor; Garry Jacobs,
Kevin Sutovich Technical Director, Fluor; Alan Birch, Account Manager, Süd Chemie; Ernst Köhler, Global Product
Brian Watkins Manager-Zeolites, Süd Chemie
Gordon Weatherbee Irving Oil decided to convert the existing VGO Hydrocracker/LCO Desulfurizer at the
Saint John refinery in New Brunswick, Canada to an LCO /heavy diesel ULSD unit. An
integrated approach between Grace, Süd-Chemie and Fluor met Irving Oil’s request for
Please address
a process to produce 7 ppm sulfur in diesel within cloud point specifications while
your comments to
allowing them to “turn off” the dewaxing function during the summer months.
betsy.mettee@grace.com
Grace Davison Multi-Loader System
By Al Jordan, Director, Sales Operations; Adam Kasle, Technical Sales Manager, Grace Davison 43
Refining Technologies
Grace Davison’s Multi-Loader System provides an effective reliable means of adding
FCC fresh catalyst and additives on a continuous basis, whether you are adding one or
several materials to the FCC.
W. R. Grace & Co.-Conn. • Advanced Refining Technologies

7500 Grace Drive • Columbia, MD 21044 • 410.531.4000
www.e-catalysts.com
©2008 The information presented herein is derived from our testing and experience. It is offered, free of charge, for your considera-
tion, investigation and verification. Since operating conditions vary significantly, and since they are not under our control, we
W. R. Grace & Co.-Conn.
disclaim any and all warranties on the results which might be obtained from the use of our products. You should make no
assumption that all safety or environmental protection measures are indicated or that other measures may not be required.
New Opportunities for
Co-Processing Renewable
Feeds in Refinery
Processes
Brian Watkins he use of renewable or bio-
T
processed renewable diesel. Some
Supervisor, based sources of feed to pro- common sources of renewable
Laboratory Technology, duce fuels is becoming more feeds are those produced for food
Advanced Refining Technologies widely employed as a means of grade oils such as soybean, rape-
decreasing dependence on non- seed and other vegetable oils. The
renewable fossil fuel sources. There traditional process for introducing
Charles Olsen are typically three common produc- these sources into the diesel pool is
Worldwide Technical Services tion routes for biodiesel. Fuel which is to use the transesterification reac-
Manager, Advanced Refining produced by the FAME (Fatty Acid tion for breaking the glycerol from
Technologies Methyl Ester) process to meet a fuel the fatty acid chains. This reaction
specification of ASTM D6751 is con- requires the use of an alcohol (such
sidered biodiesel. Fuels produced as methanol) and a catalyst (such
Kevin Sutovich from biological material using thermal as sodium or potassium hydroxide,
Senior R&D Chemist, depolymerization to meet ASTM D975 NaOH or KOH) in order to break the
Grace Davison Refining or ASTM D396 are considered renew- long chained fatty acids apart from
Technologies able diesel. Fuels that are produced the glycerin molecule. (Figure 1)
when vegetable oils or animal fats are
processed in traditional refining
Natalie Petti processes are considered co-
Vandelay Management
Catalagram 103 Spring 2008 1

Background
Figure 1
Transesterification reaction Taking a detailed look at the com-
pounds found in typical renewable
oils shows that these oils can be
O O
treated as classic petroleum based
CH2 O C R1 CH2 O C R1 CH2 OH compounds. ART analyzed several
different renewable sources of fuels
O O
in order to better understand the
CH2 O C R2 CH3OH CH2 O C R2 CH OH possible chemistry that would occur
Catalyst
O O if they were processed in a conven-
tional hydrotreater. Soybean, palm
CH2 O C R3 CH2 O C R3 CH2 OH
and rapeseed oils were studied, as
these materials are readily avail-
able. A table listing various oils and
their structural makeup is shown in
These long chained fatty acids blended with conventional diesel. Table I. Bio-based sources of oils
(Fatty Acid Methyl Esters or FAME) Discrete storage and supporting can be of significant value when
are most commonly between 16-20 inventory of the biofuels would be incorporated into the ULS diesel
carbons in length with a few excep- required to guarantee continuous sup- pool due to the low contaminant
tions. The long carbon chains are ply. Using one of ART’s high perform- concentrations and high cetane
similar in structure to the 16-20 car- ance catalysts, we have been able to number of the resulting products.
bon chains found in typical diesel capitalize on renewable sources of
except that FAME compounds con- fuel by bypassing the purchase of Components and Description
tain almost no sulfur or nitrogen and FAME products and instead process-
no aromatics, which make them ing the raw materials through conven- The major saturated and unsaturat-
excellent blending components in tional hydrotreating equipment to pro- ed fatty acids found in these oils
the diesel pool. FAME products, duce a higher quality ULSD product. consist of palmitic acid (C16:0),
however, do have a high percentage Refiners who use the co-processing linolenic acid (C18:3), linoleic acid
of oxygen that enables them to be method would have exact knowledge (C18:2), oleic acid (C18:1),
tracked directly at the pump. FAME of the bio-based fuels that are incor- eicosenoic (C20:1) and erucic
production occurs in separate facil- porated into the diesel pool which acids (C22:1) in varying percent-
ities and requires a distribution helps to ensure the finished blend ages. If each of these fatty acid
infrastructure to transport the biofu- quality. chains were to be separated from
els to a location where they can be the glycerol molecule they can be
included easily into the diesel pool
Table I
Composition of Various Oils and Fats1,2
C20:0 Mono- C20:1 C22:1

C22:0 unsaturated Arachidonic-
C4:0 C6:0 C8:0 C10:0 C12:0 C14:0 C16:0 C16:1 C18:0 C18:1 C18:2 C18:3 Arachydic- acids Erucic &
Fats and Oils Butyric Caproic Caprylic Capric Lauric Myristic Palmitic Palmitoleic Stearic Oleic Linoleic Linolenic Behenic & <C16:1 others
Molecular wt. 88 116 144 172 200 228 256 254 284 282 280 278 326 226 324
Tallow, wt.% 3 27 2 24.1 40.7 2 0.7 0.3
Lard, wt.% 1 26 2 13 45.2 10.3 2.5
Butter, wt.% 3.5 1.5 25 3 11 30 3.5 12 26 3 1.65 1.5 0.85
Coconut, wt.% 8 8 48 16 8.5 2.5 6.5 2 0.5
Palmkernel, wt.% 3 5 48.5 17 7.5 0.5 2 14 1 1.5
Palm, wt.% 3.5 395 3.5 46 7.5
Safflower, wt.% 52 2.2 76.4 16.2
Peanut, wt.% 0.5 7 1.5 4.5 52 27 7.5
Cottonseed, wt.% 1.5 19 2 31 44 2.5
Maize, wt.% 1 9 1.5 2.5 40 45 1
Olive, wt.% 1 13 2 2 68 12 0.5 1
Sunfower, wt.% 6 4.2 18.7 69.4 0.3 1.4
Soy, wt.% 0.3 7.8 0.4 2.5 26 51 5 7
Rapeseed, wt.% 3.5 0.2 2 13.5 17 7.5 0.9 56.3
Mustard, wt.% 4 3 1.5 39.5 12 8 36
Codliver Oil, wt.% 0.2 10 14.5 0.5 28 1 42
Linseed, wt.% 6 5 17.3 16 55 0.5
Tung, wt.% 8 12 80
2 www.e-catalysts.com
as normal paraffin components in Figure 2
the 500-650˚F boiling range. These Simulated Distillation (D2887) of Soybean oil
n-paraffins can be of significant
value for ULSD as they have typi-
cal cetane numbers ranging from Chromatogram: Boiling point (°F)
95 to 110, which can provide a sig- 2.200e+006

100
200300 400 500 600 700 800 900 1000 1100
nificant boost for those refiners 2.000e+006

processing feeds with lower
cetane (i.e. FCC LCO’s). The typi- 1.800e+006
cal diesel hydrotreater has only a 1.600e+006
small effect on cetane with cetane

1.400e+006
upgrade of about 2-4 numbers.
1.200e+006
Signal
In the unbroken, unprocessed form,
1.000e+006
the triglyceride molecules are signif-
icantly outside the diesel pool range 8.000e+005
as they have molecular weights of 6.000e+005

700 or greater, while the typical
diesel pool has a molecular weight 4.000e+005
of less than 400. The simulated GC
Start Time
End Time
2.000e+005
analysis of soybean oil is shown in
FBP
IBP
Figure 2 and indicates that these 0.000e+000
0 5 10 15 20 25 30 35 40
materials have a fairly narrow distil- Retention time (min)
lation showing up in the C50-C60
range. Note that simulated distilla-
tion of these compounds is based
on the carbon content and molecu-
lar weight of the materials and this
can sometimes skew the estimated
boiling points. Biofeed sources typ- Table II
ically have a true boiling point that is Analysis of Different Biofeed Sources
much lower than that reported by
simulated distillation equipment due Soybean Rapeseed *Palm
to molecular weight interference. In Oil Oil Oil
the unconverted state these triglyc- API (°) 21.58 21.98 22.98
eride molecules cannot be blended Specific Gravity (g/cc) 21.6 22.0 23.0
into the diesel pool at the levels Sulfur, ppm 0 3 1
required to meet renewable fuel Oxygen, wt.% 10.5 10.62 11.33
standards. Nitrogen, ppm 3.9 16 1.6
D2887 Distillation, °F
Another concern is that these IBP 702 710 625
renewable feed sources can 5% 1059 1065 941
include various contaminants. An 10% 1069 1077 1026
analysis of several different bio- 30% 1090 1095 1062
feed sources has indicated the 50% 1102 1106 1079
presence of contaminants such as 70% 1111 1115 1090
sodium, calcium and phosphorus. 90% 1183 1188 1146
Table II shows the measured con- 95% 1232 1238 1197
taminant levels of the soybean, FBP 1301 1311 1302
rapeseed oil and palm oils used in Metals Contamination,
this work. The palm oil shows no ppm
trace impurities, which indicates Na 2.0 4.7 0.0
that it has been previously Ca 3.0 13.8 0.0
processed while the soybean and Mg 0.9 0.3 0.0
rapeseed oils have not. In the fore- P 6.5 4.0 0.0
seeable future it is unlikely that the Zn 0.1 0.6 0.0
use of these renewable sources Al 0.1 0.2 0.0
Mn 0.0 0.1 0.0
* Oil was pre-processed to remove impurities

would exceed 20% in conventional Figure 3
hydrotreating applications, which Pathways to Hydrotreating of Renewable Oils
would bring the level of all of these
contaminants down to 2.5 ppm or
H
less. At these levels ART’s high H C -O-C
=
O
+ 6H2
capacity guard materials and Grace O
=
H C -O-C
Davison specialty catalysts are O
H C -O-C
=
capable of protecting the down- H
HEAT
stream high activity catalysts from

these damaging poisons.
H H
H C -O-C
H
=
O
Since these renewable feeds are O
H H
=
H C -O-C
derived from a biological source, O H H
H C -O-C H
=
they also contain a high concentra- H H
H
+ 3 H2O H H H
tion of oxygen. For the materials
listed in Table I the oxygen content
ranges from 10 to 15%, and is H
H C -H H
H
entirely dependent on the length H C -H H HEAT
H
and degree of saturation of the fatty H C -H +
H
H H 4.5H2
acid chains. + 6H2
+ 3 O2
This quantity of oxygen is important,

as under normal hydrotreating con-
ditions the oxygen will react with the per standard cubic foot of hydrogen This reaction is expected to take
hydrogen to form water. This water, consumed. The reaction pathways to place at a rate similar to that of sul-
if generated in a significant enough hydrotreat the bio-oils is shown in fur compounds such as sulfides
quantity, may cause problems such Figure 3. and disulfides. In order to verify that
as weakening the catalyst support this is indeed the reaction that is
or redistribution of the active metals In the first step of the reaction, the taking place, several different feed
and loss of surface area. At the unsaturated fatty acid chains are sources were analyzed for oxygen
expected blending ratios of 10%, quickly converted into fully saturated in order better understand where
the oxygen content is around 1 to n-paraffins. The second reaction that they are likely to be distributed in
1.5 wt.%, and even if all the oxygen must occur in order to ensure that the the feeds. This analysis has a
is converted, this is unlikely to gen- compounds will be of the appropriate detection limit of approximately 2.6
erate enough water to be a signifi- size for the diesel pool is the breaking ppm oxygen. Figure 4 is a Carbon-
cant problem. of the fatty acid chains away from the Sulfur-Oxygen chromatogram for a
glycerin molecule which requires diesel feedstock that contains 263
General Co-Processing Ideas cleavage of a carbon – oxygen bond. ppm oxygen.
Looking at these compounds from a Figure 4

hydrotreating perspective, the bio- Chromatogram of Sulfur, Carbon & Oxygen
feeds can be classified as mono
and di-olefins, since a majority of
in a Diesel Feedstock
these compounds have one or two
double bonds per fatty acid chain
with a few having three. Using the
numbers listed in Table I, palm oil
has an average of 1.1 double bonds
per chain, rapeseed oil has 1.3 and
soybean oil has 1.7. It has been
widely established that the olefin
saturation reactions occur quite
rapidly and tend to happen near the
top of the catalyst bed in a
hydrotreater. The reactions go to
near completion at typical
hydrotreating conditions, and will
generate between 130-150 BTU’s
Figure 5
Chromatogram of Sulfur, Carbon and sources of feedstock at the target-
Oxygen in a Co-Processed Product ed 10% level. Figure 6 summarizes
some of the results of the testing.
The testing showed that soybean
and rapeseed oils behave similarly
when co-processed in a SR diesel.
The feed blends required essential-
ly the same temperature for 10 ppm
product sulfur, and the apparent
activation energy (temperature
response) for the two feed blends is
similar to that of the SR feedstock
alone. The palm based oil, which
had been previously processed,
was apparently easier to treat to low
sulfur diesel levels, but for 10 ppm
product sulfur the temperature was
only slightly lower than that for the
Analytical techniques using GC- sulfur, nitrogen and aromatic contents SR feed. The apparent activation
AED have shown that in normal and decrease the API gravity. energy for this feed blend was lower
ULSD operation, no oxygen is than the SR component indicating
detected in the products at levels ART then conducted testing on blends the temperature response in the unit
below 500 ppm sulfur. This can be containing the various renewable was lower. Comparing the feeds at
seen in Figure 5 which shows the
analysis of one of the co-processed
products which has a total sulfur of Table III
31 ppm and less than 1 ppm nitro-
Straight-Run (SR) and Bio-Blend Analyses
gen.
SR 10% 10% 10% 40% 80%
Pilot Plant Testing of Renewable Oil Soybean Palm Rapeseed Soybean Soybean
Oils
API 34.44 33.03 33.50 33.29 29.24 24.38
In order to understand the process Specific 0.852 0.859 0.857 0.858 0.879 0.907
for co-treating renewable fuel com- Gravity, g/cc
ponents in a hydrotreater, Advanced Sulfur, wt.% 1.123 1.083 1.092 1.042 0.670 0.210
Refining Technologies completed a Ni trogen, 130 82 75 67 47 16
number of pilot plant studies. A ppm
wide range of ULSD operating con- Oxygen, 0.0 1.2 1.2 1.1 4.7 8.9
ditions were investigated to deter- wt.%
mine if there is an optimal operating Aromatics,
window for processing these types wt.%
of feeds. The conditions included Mono 17.76 15.85 15.87 15.86 10.32 3.34
hydrogen pressures from 450 to Di 7.39 6.60 6.60 6.60 4.29 1.39
1100 psia and hydrogen to oil ratios Poly 2.1 1.87 1.88 1.87 1.22 0.39
of 1000 to 3000 SCFB. Total 27.25 24.32 24.34 24.33 15.84 5.12
D2887
The three different renewable Distillation,
sources of oil were blended in sepa- ˚F
rately with a typical straight run (SR) IBP 222 209 209 210 239 329
diesel feedstock. The renewable 10% 477 465 459 465 498 571
component level was varied from
30% 579 559 557 559 579 1009
10% to 80% and hydrotreated over
50% 613 595 592 594 631 1119
the range of processing conditions
70% 643 632 628 630 1108 1130
listed above. The SR component
properties are listed in Table III, along 90% 681 720 688 715 1130 1135
with 5 different blends of the bio FBP 740 1127 1121 1127 1134 1139
components. As can be seen in the Cloud 19.9 21.7 24.1 22.8 22.0 19.1
table, the effects of blending in the Point, ˚F
renewable source are to dilute the Cetane Index 53.8 50.6 51.2 50.9 46.4 NA

Figure 6 ultra low sulfur levels suggests the
Results of Various Renewable Components co-processing of the renewable oils
has only a small effect on the per-
10% Levels formance of the hydrotreater. The
data indicates that the least reactive
630 blend contains rapeseed oil, fol-
lowed by the soybean oil blend and
620
finally the palm oil blend. The differ-
610 ence, however, is only about 10˚F for
WABT, ˚F
<10 ppm product sulfur.

600
590 Looking at other diesel product

properties that are important to
580 refiners reveals that there is a boost
570
in the product cetane index by
0 05 100 150 almost 2 numbers. This increase in
Product Sulfur, ppm cetane was seen over the wide
range of conditions tested and is a
SR 10% Soy 10% palm 10% rapeseed
reflection of the normal paraffins
from the renewable oil discussed
Figure 7 above. Figure 7 summarizes some
of the cetane index results achieved
Cetane Boost when Co-Processing Bio-Feeds at higher pressure for each of the
at High Pressure bio-feed blends.
63 The impact of low pressure opera-

62 tion on cetane improvement is
Product Cetane Index
61
shown in Figure 8 for a 10% renew-
able feed blend. Not surprisingly,
60
lower pressure operation results in a
59
lower cetane index for the SR feed,
58 but the addition of the renewable oil
57 again provides a consistent two
56 number increase in cetane index.
55
570 580 590 600 610 620 630 This is a good indicator that the
WABT, ˚F large fatty acid molecules are being
broken down into the three individ-
SR 10% Soy 10% palm 10% rapeseed ual fatty acid chains via the break-
ing of the C-O bonds. Figure 9
Figure 8 compares the D-2887 distillation
chromatograms of the SR products
Cetane Boost when Co-Processing Bio-Feeds
at 10 ppm sulfur to that of the co-
at Low Pressure processed products, and it is evi-
dent that there is an increase in the
60 concentration of the n-paraffins
between 500°F and 600°F boiling
Product Cetane Index
59 points. It is this increase that yields

the significant boost in cetane.
58
The hydrotreating of the bio-blend-
57
ed oil results in a product that no
longer contains material in the C50
56
to C60 range. This is consistent with
55 the theory that the individual fatty
590 600 610 620 630 640 650 acid chains are being broken apart
WABT, ˚F to hydrocarbons of similar size to
those in the SR diesel.
SR Renewable
Due to the addition of unsaturated
Figure 9
chains from the bio component,
there is expected to be an increase Boiling Point Comparison Between SR and 10% Bio Blends
in hydrogen consumption to satu-
rate these C=C bonds. With this
additional hydrogen usage, it is Chromatogram: Boiling point (°F) Chromatogram: Boiling point (°F)
important to also be aware of any 1.000e+006

100
200300 400 500 600 700 800 900 1000 1100
1.800e+006
100
200300 400 500 600 700 800 900 1000 1100
changes in product aromatics, as 9.000e+005

1.600e+006
reactions to saturate aromatics are

8.000e+005
1.400e+006
7.000e+005
high consumers of hydrogen and 6.000e+005

1.200e+006
would compete with the saturation 1.000e+006
Signal
Signal
5.000e+005
reactions under hydrogen limited 4.000e+005

8.000e+005
conditions. Figure 10 summarizes 3.000e+005

6.000e+005
the product aromatics for one of the 2.000e+005

4.000e+005
IBP Time
bio-blended feeds. The total aro-
Start Time
End Time
End Time
1.000e+005 2.000e+005
Start
FBP
FBP
IBP
matics are consistently two num- 0.000e+000
0 5 10 15 20 25 30 35 40
0.000e+000
0 5 10 15 20 25 30 35 40
bers lower than the SR feed, which Retention time (min)
Straight-run product
Retention time (min)
Co-processed product
is the same as the actual difference @ 10 ppm sulfur @ 10 ppm sulfur
in the total aromatic content of the
two feeds. The lower aromatic con-
tent of bio-blended feeds allows the Figure 10
refiner to achieve lower product aro- Comparison of Total Aromatics of SR Oil with
matic content, which may be valu-
10% Renewable Oil at High Pressure
able as future regulations may
require a lower total aromatic limit
on diesel fuel. 30
Total Product Aromatics, wt.%
At lower pressure and H2/oil ratios, 25
the total aromatic content shows a

similar response, with two numbers 20
lower total aromatics when co-pro-

cessing bio-based feedstock. With 15
reduced operating pressure, there
is a decrease in the aromatic satu- 10
ration ability of the catalyst, and the
possible use of renewable oils may
5
help to offset this. 570 580 590 600 610 620 630
WABT, ˚F
Cloud point specifications vary
SR 10% palm
based on the location of the refinery
and the end user of the fuel.
Although the blending of the bio Figure 11
components yields improvements in Cloud Points of Bio-Blended Feeds Compared to SR Feed
the diesel cetane, there is the con-
cern about the biofeed based 32
diesel cloud point. It is widely
30
known that n-paraffins have a signif-
Product Cloud Point, ˚F
icantly higher cloud point than other 28

same carbon number hydrocar- 26
bons. Since hydrotreating converts
24
the fatty acid chains into long
22
chained n-paraffins, the cloud point
of the mixture will increase. Figure 20
11 summarizes the product cloud 18
points after hydrotreating the three
16
blended feeds. The SR feed is 570 580 590 600 610 620 630
included for comparison.
WABT, ˚F
SR 10% Soy 10% palm 10% rapeseed

Processing the SR feed has essen- Figure 12
tially no impact on the product Conversion vs. Cat to Oil
cloud point and the products are all
within a few degrees of the feed
cloud point shown in Table III. The 112.0
delta Conversion, wt.%

renewable containing feeds all have Base Feed
109.0
slightly higher cloud points com- 7.5% Bio
pared to the SR feed (see Table III), 106.0 15% Bio
and after hydrotreating the cloud
point increases by 6 to 10°F. This 103.0
increase in cloud point can be sig- 100.0
nificant especially in cold weather
climates. By using other technology 97.0
provided by ART, the problem of
94.0
increased cloud point can be
reduced or eliminated. 4 6 8
Cat to Oil
Co-Processing in FCC Units
Another option for refiners could be

to co-process bio components in yields. Characterization of the bio- Pilot Plant Testing of Renewable
fluid catalytic cracking units feedstock and an understanding of Oils
(FCCUs). Unlike hydrotreater units the refiner’s objectives can allow the
where catalyst cannot be changed catalyst supplier to develop a formula- To illustrate the impact on FCC
without taking the unit out of service tion to maximize profits. yields with the incorporation of veg-
to reload the reactor, continuous etable oil feedstocks into FCC feed,
replacement of catalyst in the FCC a pilot plant study was conducted
unit enables the refiner to adjust the by starting with a composite feed of
catalyst formulation to optimize VGO and resid with properties
shown in Table IV, then the soybean,
Table IV palm and rapeseed oils used in the
Feedstock Analysis for FCC Study hydrotreating testing were blended
to 0%, 7.5% and 15% concentra-
tions. The hydrotreated VGO in this
Composite Hydrotreated table is used in a later study to com-
Resid VGO pare the effects of different base
feedstocks.
API 24.4 27.3
Sulfur, wt.% 0.53 0.20 The blended FCC feeds were
Nitrogen Total/Basic, ppm 813/287 800/280 cracked over an FCC catalyst,
Concarbon, wt.% 1.12 0.2 which was deactivated using a
CPS-3 type protocol3,4 to 1000 ppm
K-Factor 11.96 12.23 nickel and 2000 ppm vanadium.
Aromatic Ring, wt.% 22.7 18.9 The catalyst was formulated to pro-
Naphthenic Ring, wt.% 13.6 11.5 vide maximum bottoms upgrading.
Paraffinic Carbons, wt.% 63.8 69.5 Properties of the deactivated cata-
lyst are shown in Table V.
Vanadium, ppm 2.5 1.8
Nickel, ppm 1.1 0.7 The pilot unit was run at a constant
reactor temperature of 1000˚F, and
D6352 Distillation, ºF the deactivated catalyst was tested
IBP 494 460 at three catalyst-to-oil ratios (4,6,8)
10 689 703 for each of the blended feeds in the
30 775 786 pilot unit.
50 834 849
70 899 923 Figure 12 shows that all three of the
90 1018 1034 blended bio-feeds are easier to
95 1110 1088 crack than the base feed. The addi-
FBP 1279 1226 tion of the bio component increases
the cracking activity (lower cat to oil
Figure 13
Conversion vs. Coke potentially allowing for downstream
hydrotreating benefits (less gaso-
Biofeed Pilot Plant Study line octane loss, extended catalyst
81.0
80.0
run length) or for a lower cost base
79.0 FCC feed at constant product sulfur.
78.0
Conversion, wt.%
77.0 Since the increase in the concentra-

76.0 tion of the biofeeds in the feed
75.0 directionally increases the magni-
74.0 tude of the response in the FCC
73.0 yields, to simplify the trends the
72.0 remaining results will be presented
71.0 for only 15% blends. Additionally,
3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5
while commercial FCC units operate
Coke, wt.% to constant coke, if there is a signif-
icant difference in coke yield when
Base Feed 7.5% Rapeseed 7.5% Soy 7.5% Palm 15% Soy 15% Rapeseed 15% Palm comparing individual product
yields, the constant coke compari-
for a given conversion), and age of 1.1 double bonds per fatty acid son will accentuate the differences
increasing the concentration of a chain, this explains the performance in yields. Comparing data on a con-
given bio component yields an of the palm oil versus the soy or rape- stant conversion basis will smooth
increase in conversion at constant seed oils. out the data and allow for a more
cat to oil ratio. representative comparison. Figure
The bio-feeds also have significantly 14 shows the yields for the bio-
Figure 13 shows that the bio-feeds less contaminants (sulfur, nitrogen, feeds at 15% concentration for con-
produce significantly less coke per and potentially metals) than the base stant coke and conversion com-
unit conversion than the base feed. feed, but additional characterization of pared to the base feed.
Increasing the concentration of any these materials is very difficult.
of the three bio components further Traditional analysis methods devel- Beginning with hydrogen, Figure 15
reduces the amount of coke pro- oped for fossil based hydrocarbons confirms that the reduction in hydro-
duced for a given conversion. The will not apply to the bio-based materi- gen yield with the biofeeds is signif-
rapeseed oil is the easiest to crack, als. Measurements such as concar- icant. Hydrogen can come from
followed by the soybean oil with the bon, n-d-M (which is an estimate of three sources: a by-product of
palm oil being the most difficult of the chemical composition of the feed- dehydrogenation with metals, a
the components to crack. stock using refractive index (n), the product of thermal cracking, or a
density (d) and the molecular weight product of catalytic cracking.
The trends observed in Figures 12 (M) of the feed to calculate the
and 13 can be explained by the amount of paraffinic (Cp), napthenic One potential method to determine
individual fatty acid compositions (Cn), and aromatic (Ca) carbon the reason for the reduction in
for each of the bio components species in the feed) and even GC hydrogen with the bio-based materi-
reported previously in Table I. The based distillations can be flawed due als involves a comparison of the C5
reaction pathways for triglycerides to the thermal cracking response of olefins yields in gasoline for the
have been studied extensively by the bio-feeds. With virtually no sulfur feeds tested. During dehydrogena-
Dupain et. al.5, and under FCC con- in each of the bio-feeds, incorporating tion reactions, gasoline range
ditions, were found to occur in two them into FCC feed would dilute key olefins react with the metals on the
distinct steps. First, thermal crack- FCC product stream sulfur levels, catalyst to produce hydrogen7. If
ing of the triglyceride occurs,
releasing the fatty acid chains from Table V
the glycerin backbone. Conversion
of free fatty acids via thermal crack- Deactivated Catalyst Properties
ing is very low, requiring catalytic MAT, wt.%
cracking in the subsequent step to
Total Surface, m2/g 175
break the high molecular weight
fatty acid molecules into smaller, Zeolite Surface, m2/g 99
more valuable products. Free fatty Matrix Surface, m2/g 76
acids with more saturation will be Unit Cell Size, Å 24.29
more difficult to crack than those
Rare Earth, wt.% 2.04
that contain greater amounts of
double bonds6. With the lowest aver- Alumina, wt.% 50.23

Figure 14 the biofeed materials produce less
Comparison of 15% Blended Feeds gasoline range olefins than the base
feed, this could be the cause for the
Constant Coke, 5%
reduction in hydrogen.
Constant Conversion, 76%
Delta Change, wt.% ex-octane
From our analysis, the addition of

biofeeds to the base feed does not
reduce the amount of gasoline
olefins produced, and therefore the
reduction in hydrogen observed
with the biofeeds is not likely due to
reduced dehydrogenation reac-
tions.
Alternatively, the potential for the

biofeeds to produce hydrogen dur-
ing the catalytic cracking process
Hydrogen
Total Dry Gas
Propylene
Total C3's
Total C4='s
Total C4's
C5+ Gasoline
RON
MON
Gaso. Isoparaffins
Gaso. Aromatics
Gaso. Napthenes
Gaso. Olefins
LCO
Bottoms
Coke
Conversion
can be evaluated by comparing the
propane yield for the feeds tested.
In cracking a biofeed, the fatty acid
molecules that were liberated by the
Figure 15 initial thermal cracking step will sub-
Hydrogen Yield vs. Conversion sequently crack along the pathways
Biofeed Pilot Plant Study defined for either paraffin or olefin
0.55
molecules. If the fatty acid mixture
0.50
is more olefinic, it will be very reac-
tive and will easily crack to produce
0.45
smaller gasoline range olefins6. A
0.40
Hydrogen, wt.%
more paraffinic fatty acid mixture

0.35
can react along multiple potential
0.30
pathways to produce a variety of
0.25
products (Figure 16).
0.20
0.15 The initiation step occurs on either

0.10 Bronsted or Lewis acid sites on the
0.05 catalyst, and on the Bronsted site
71.0 72.0 73.0 74.0 75.0 76.0 77.0 78.0 79.0 80.0 81.0
there are two additional potential
Conversion, wt.%
pathways for the paraffin to react to
Base Feed 15% Palm
produce a carbenium ion. Protolytic
15% Soy 15% Rapeseed
cracking can occur with the elimina-
tion of the paraffin, or the carbeni-
Figure 16 um ion can be formed with the elim-
Reaction Pathway Network for Alkane Cracking ination of a hydrogen molecule
(Figure 17.)
INITIAL CHAIN
REACTIONS REACTIONS
If protolytic cracking is prevalent,
the chain terminating product is
PROTOLYTIC ALKANE
CARBONIUM ION CRACKING (Product) propane. Figure 18 shows that the
+ H+
STRONG propane yield for all three of the bio-
ALKANE
BRONSTED
INITIATION feeds is lower than the base feed.
- H+
STRONG Thus there is an indication that the
LEWIS
CARBENIUM ION
ß OLEFIN
reduction in propane and hydrogen
SCISSION
+A
(Product)
is potentially due to reduced pro-
LK
AN
E ALKANE PROPAGATION tolytic cracking of the fatty acids
(Product)
ORM OLEFIN
compared to a typical FCC feed.
RE F
- H+ B RO
NST
ED (Product)
TERMINATION
This also indicates that the double
CARBENIUM ION
+ H+
REF
ORM
ALKANE bonds on each of the free fatty acid
LEW (Product)
IS
molecules, which are more reactive,
are the initiation sites for the crack-
Catalytic Cracking Catalysts, Chemistry and Kinetics, Chemical
Industries 25, reprinted by permission ing reactions.
Figure 17
The fatty acid molecules are crack- Paraffin Reaction Pathway
ing into gasoline and propylene, as
can be seen in Figures 19 and 20.
The gasoline trend indicates over-

H
cracking for all three of the bio-
feeds, and also for the base feed. -H2
R CH2 C
+
R’
H H Z -
Over cracking occurs when the
R CH2 CH2 R’ HZ R CH2 C R’
gasoline molecules produced are +
H
-
subsequently cracked into lighter H Z -RH
R CH2 C+ Z -
molecules. Catalyst activity, cat to
oil ratio, and temperature can all H
drive a feed to over-cracking, and
the easier the feed is to crack, the
higher the tendency to over-crack.
Table IV shows that the base feed
itself is fairly paraffinic, with a
Watson K-Factor that indicates it will Figure 18
produce high conversions. The Propane Yield vs. Conversion
incorporation of the highly paraf- Biofeed Pilot Plant Study
finic fatty acids from the biofeeds
makes the blended feeds even eas- 1.4
ier to crack. Catalyst activity will be 1.3
determined based on specific unit
1.2
constraints and objectives when
Propane, wt.%
commercial processing of the 1.1
blended bio-containing feeds. The 1.0

overcracking trends observed in the
0.9
pilot plant testing can be corrected
0.8
and is not expected in commercial
operations. 0.7
0.6
The influence of the bio-based 71.0 72.0 73.0 74.0 75.0 76.0 77.0 78.0 79.0 80.0 81.0
materials on the gasoline properties Conversion, wt.%
is important, as the oxygen species

in the triglycerides could break Base Feed 15% Palm 15% Soy 15% Rapeseed
down to water, or they could poten-

tially be converted into aldehyde
and furan species which are unde-
sirable from an environmental,
health and safety perspective. Gas Figure 19
Chromatography-Atomic Emission Gasoline Yield vs. Conversion
Detector (GC-AED) was performed Biofeed Pilot Plant Study
on the liquid product, which was
52.0
recovered after each test run
through the pilot unit. The GC-AED 51.5
was run in oxygen mode in order to 51.0
detect oxygen species using the
Gasoline, wt.%
50.5
same technique as in Figure 4 and
5. The only peaks that showed up 50.0
were initial peaks between two and 49.5

four minutes from the syringe wash
49.0
solvents. No discernible oxygen
peaks were present in the liquid 48.5
product thus, no undesirable oxy- 48.0

gen species were present. 71.0 72.0 73.0 74.0 75.0 76.0 77.0 78.0 79.0 80.0 81.0
Conversion, wt.%
Base Feed 15% Palm 15% Soy 15% Rapeseed

The (R+M)/2 octane of the gasoline Figure 20
produced by the biofeeds is lower Propylene Yield vs. Conversion
than the base feed, with contribu-
Biofeed Pilot Plant Study
tions from both RON and MON con-
tributing to the trend. The magni- 7.0
tude of the response in gasoline
RON may depend on the source of 6.5
the biofeed, but in general the loss
Propylene, wt.%
6.0
of RON is fairly small. (Figure 21)
MON is consistently lower with the 5.5
biofeeds, and this is due to the
lower aromatics content of the 5.0
gasoline produced with the bio-

4.5
based materials in the feed.
4.0
The response in C4 yields can be 71.0 72.0 73.0 74.0 75.0 76.0 77.0 78.0 79.0 80.0 81.0
Conversion, wt.%
explained by again reviewing the
individual fatty acid compositions
for the three biofeeds as shown ear- Base Feed 15% Palm 15% Soy 15% Rapeseed
lier. Palm oil contains significant per-

centages of both C16 (palmitic) and
C18 (oleic) acids. The highest % Figure 21
fatty acid in soybean oil is C18 Gasoline RON & MON for Bio-blends
(linoleic), and C20 and C22 (arachi-
94 85
donic and erucic) acids are present
in the largest concentrations in 93.5
84
rapeseed oil. Once these mole-
93
Gasoline MON
cules are liberated from the glycerin
Gasoline RON
83
backbone via thermal cracking, 92.5
they follow a typical FCC rule of 92 82
thumb, which is that the longer the
chain (ie. the higher the carbon 91.5
81
number in the molecule), the more 91
broad the distribution of product 90.5
80
olefins that will result from catalytic
cracking of the molecule. Thus, the 90 79
palm oil produces the largest 71 73 75 77 79 81
amount of C4 olefins, as seen in
Figure 22. Base Feed Palm Soy Rapeseed
Influence of Base Properties

Figure 22
Ultimately the yield response for the Butylene Yield vs. Conversion
addition of a bio-feed material into
Biofeed Pilot Plant Study
FCC feed will depend on the prop-
erties of the base FCC feed. 8.2
Dupain et al.5 observed that the
8.0
addition of rapeseed oil to a
hydrowax feed would yield less 7.8
Total C4='s, wt.%
gasoline and C4 minus products,

7.6
and higher amounts of LCO, slurry
and coke. (Figure 23) 7.4
7.2
While the properties of the specific
hydrowax feed are not known, in 7.0
general these feeds are highly
6.8
paraffinic and the addition of veg- 71.0 72.0 73.0 74.0 75.0 76.0 77.0 78.0 79.0 80.0 81.0
etable oils to that type of feed would Conversion, wt.%
in fact degrade the overall proper-
Base Feed 15% Palm 15% Soy 15% Rapeseed
Figure 23
Quantitative Yields for Different
Hydrowax/Rapeseed Oil Blends References
1. Data derived from Organic

70 Chemistry, W.W. Linstromberg, D.C. Health
(a) HWX/RSO product fractions
and Co., Lexington, MA, 1970
60
2. Data derived from Organic
Gasoline
Product fraction, wt.%
50 Chemistry, Morrison and Boyd, 6th Edition,
525 °C/CTO 4/21.2m 1992
40
30 LCO
3. Wallenstein, D., Roberie, T., and
Bruhin, T., Catalysis Today 127, 2007, pp. 54-
20 69.
HCO
10 Gas 4. Wallenstein, D., Harding, R.H., Nee,

Coke J.R.D., and Boock, L.T., Applied Catalysis A:
0 General 204, 2000, pp. 89-106.
0.0 0.2 0.4 0.6 0.8 1.0
RSO fraction in feed blend (-)
5. Dupain, X., Costa, D.J., Schaverien,
C.J., Makkee, M., Moulijn, J.A., Applied
Catalysis B: Environmental 72, 2007, pp. 44-
Applied Catalysis B, Environmental 72, reprinted by permission 61.
6. W.-C. Cheng , E.T. Habib, Jr., K.

ties of the feed. This is confirmed to catalyst stability or yields, but the Rajagopalan, T.G. Roberie, R.F.
by comparing the response in effect on an individual operation will Wormsbecher, M.S. Ziebarth, Handbook of
yields for the addition of rapeseed depend on the base feed and condi- Heterogeneous Catalysis, 2nd edition, G.
tions. Grace Davison’s Biofuel Ertl, H. Knoezinger, F. Schueth, J. Weitkamp
oil to the two base FCC feeds from
(Editors), Wiley-VCH, Weinheim, 2008,
Table IV. The general trends in Technologies Group utilizes resources Chapter 13.5.
yields for constant conversion are across Grace and its affiliates to evalu-
shown in Figure 24. The addition of ate options for refiners who wish to 7. Zhao, X., J. A. Rudesill, W-C. Cheng,
consider incorporating co-processing Preprints of Symposia - American Chemical
rapeseed oil to the hydrotreated
Society, Division of Fuel Chemistry, 46(1),
VGO feed produced more coke and biofeeds into their operation, but want 2001, pp. 240-244.
hydrogen, at constant conversion, to understand the optimum configura-
but at constant coke the conversion tion to maximize their profitability. 8. B. W. Wojciechowski, and A. Corma,
Cataytic Cracking. Catalysts, Chemistry,
was actually lower with the rape-
and Kinetics. Chemical Industries 25, 1986,
seed oil in the feed. By comparison, The authors of this paper would like to
New York: Marcel Kekker.
the composite feed shows strong thank the following people for their
improvements in yields with the contributions to this project:
addition of rapeseed.
Susan Ehrlich, Business Director
These trends suggest that if a base Biofuel Technologies Group; Rick
FCC feed is paraffinic, yields may in Wormsbecher, Research and
fact worsen with the addition of a Development Fellow, Refining
bio based material into the feed, Technologies
whereas for VGO or resid types of Figure 24
feed, yields may instead improve. Yield Trend Comparison for Feeds at Constant Conversion
Future work will include examining
the role of the base feed on yields Hydrotreated VGO Composite Feed
with the incorporation of biofeed Hydrogen
materials into the FCC feed.
Wet Gas
Conclusion
Gasoline
Based on these results, the use of
ART’s high activity hydroprocessing LCO
catalysts or Grace’s high perform-
ance FCC catalysts can enable Bottoms
refiners to co-process renewable

Coke
oils through conventional refining
equipment. Co-processing can be
incorporated into a refiner’s operat- Red: Undesirable Green: Undesirable
Grey: Depends on the refiners objectives and constraints
ing strategy with minimal detriment
Make biofuels better with Grace
Molecular Sieve Silica for C h r o m a t o g r a p h y To o l s Catalysts for

for Dehydration Biodiesel Purification for Chemical Analysis Biofeedstock Conversion
Grace Davison Biofuel Technologies Group
It’s a natural progression.

Grace, a leading provider of specialty catalysts, adsorbents and analytical tools
to petroleum refiners, is now applying its peerless materials science expertise
to improving the manufacture of renewable fuels.
With Grace innovations customized for your process, biorefining becomes

simpler, faster, and more efficient.
Our technologies ensure that fuels that are environmentally friendly are
manufactured in processes that are economically sound.
Make biofuels with Grace, fueling innovation for sustainable energy worldwide.
www.GraceBiofuels.com
Biofuels@Grace.com
GRACE® is a trademark, registered in the United States and/or other countries, of W. R. Grace & Co.-Conn.
©2008 W. R. Grace & Co.-Conn. All Rights Reserved.
GENESISTM Catalyst
Commercial Update
he marketplace has shown great lyst families which will deliver higher
Rosann K. Schiller
Product Manager,
Grace Davison Refining
Technologies
T enthusiasm for the GENESIS™
catalyst approach to catalyst for-
mulations1. The number of applications
activity and improved gasoline
selectivity at equivalent bottoms
and coke yields.
Doc Kirchgessner
is growing and we expect this trend to
continue in 2008 and beyond. GENESISTM Catalyst Provides
Operating Flexibilty to US
Technical Sales Manager, GENESIS™ catalyst performance has Refiner
Grace Davison Refining met or exceeded expectations in
Technologies applications versus competitive tech- A major US refiner processing
nologies as well as over existing hydrotreated feed was using a
and Grace Davison catalyst technology. Grace Davison LIBRA® catalyst.
Kelly Stafford
We will share some of the results with New operating economics dictated
you here. a change in catalyst formulation,
specifically to maximize LCO with a
Technical Sales Coordinator,
We are also pleased to introduce two further goal of maximum gasoline +
Grace Davison Refining
new high activity GENESIS™ compo- LCO yield (G+D). Since the refinery
Technologies
nents. We have made improvements was constrained by main air blower
to our MIDAS® and PINNACLE® cata- (MAB) and wet gas compressor

Figure 25
Commercial Data from Refiner A
2.5
10
2.0
5
Delta Coke
1.5
Delta RgT, ˚F Delta RxT, ˚F Shift, % Rel
0
C/O Shift % Rel 1.0
-5
0.5
-10 slurry, vol.% LPG, vol.% Dry Gas, wt.%

0.0
G+D, vol.%
-15
-0.5
-20
-1.0
(WGC), the desired shifts should be hydrotreater operation, as favorable caused by circulation instability. In
achieved without increasing LPG, refinery margins shift between gaso- the last example, Refiner E realized
coke, or dry gas. The main opera- line, alkylate, and ULSD production. a dramatic 15% improvement in flow
tions strategy was to lower conver- characteristics with GENESIS™ cat-
sion without sacrificing bottoms GENESISTM Catalysts Improve alyst, and this improvement has
cracking. Laboratory testing pre- Fluidization Properties of been sustained for over one year.
dicted that GENESIS™ catalyst Circulating Inventory
would provide an incremental 0.8 How do GENESIS™ catalyst sys-
wt.% G+D yield, lower bottoms, and GENESIS™ catalysts systems have tems improve fluidization? The
reduced H2 yield at constant coke demonstrated superior yield perform- Umb/Umf is a function of unit oper-
and LPG yield. Based on the posi- ance in field application. Improved ations as well as catalyst properties
tive laboratory results, a GENESIS™ fluidization characteristics have also such as particle size distribution
formulation was selected for use in been observed where GENESIS™ and pore volume. The excellent
this application. catalyst systems are being used. The fines retention of Grace Davison
Umb/Umf is a fluidization factor used alumina-sol catalysts ensure that
GENESIS™ catalyst has exceeded to determine the fluidization capabili- the optimal mix of particles is
expectations at this refinery. In ties of an equilibrium catalyst2. The retained in the circulating inventory.
addition to the selectivity benefits value of a good Umb/Umf is unit MIDAS® catalysts have very high
observed in the laboratory testing, dependent and a value of 2.0 or pore volume1 that when combined
GENESIS™ catalyst has also pro- greater represents an optimal opera- with an optimal particle size distri-
vided a significant reduction in delta tion. Figure 26 shows trends in bution creates an easily fluidized
coke and hydrogen. The improve- Umb/Umf over time in four separate inventory, increasing the Umb/Umf.
ment in delta coke allowed the refin- FCC applications. The higher this ratio, the more for-
er to increase cat-to-oil ratio (C/O) giving the fluidized catalyst is to
at lower operating severity. The The first example is Refiner B, who changes in density, and the more
increase in C/O coupled with the had good fluidization characteristics easily it will tend to circulate in an
selectivity advantages resulted in in the inventory with a Grace catalyst FCC unit3. The unique properties of
an increase in gasoline + LCO of prior to starting GENESIS™ catalyst. the individual components in GENE-
more than 2 lv.% (Figure 25). The As the unit turned over to GENESIS™ SIS™ catalyst systems result in an
reduction in dry gas alleviated the catalyst, Umb/Umf improved further equilibrium catalyst inventory that
WGC constraint, providing room to facilitating smooth circulation. has preferred flow characteristics.
optimize reactor temperature with- Refiners C, D, and E were using com-
out compromising octane. Since petitive catalysts before GENESIS™. High Activity GENESISTM
the introduction of GENESIS™ cat- Refiners C and D were operating with Catalyst Components
alyst, the unit has realized signifi- a fluidization factor below the desired
cant operating flexibility. The optimum of 2.0. The flow characteris- The industry requires high activity
reduced operating severity has tics of GENESIS™ catalysts enabled and stability of FCC catalyst.
allowed the refiner the ability to opti- these units to achieve optimal Stability is needed to combat cata-
mize FCC operations with feed Umb/Umf levels and minimize upsets lyst deactivation at high metals
Figure 26
GENESIS™ Catalyst Improves Equilibrium Umb/Umf Trends
2.30
2.00
Base Grace Davison Competitor 1
GENESISTM GENESISTM
2.25
1.95
UMB/UMF
UMB/UMF
2.20 1.90
2.15 1.85
Refiner C
2.10 1.80
Refiner B
Sep Nov Jan Mar
Nov Jan Mar May
2.1 2.4
Competitor 1 Competitor 2
GENESISTM GENESISTM
2.3
2.0
UMB/UMF
UMB/UMF
2.2
1.9
2.1
2.0
1.8
Refiner D Refiner E
Nov Dec Feb Apr Jul Jan Jul Jan
loadings when processing resid. lyst stability without a gas penalty4. Its version, without affecting particle
High activity and optimal fluidization open particle morphology makes it integrity, at equivalent coke and gas
characteristics can overcome a cir- easy for MIDAS® catalyst to selectively selectivity of MIDAS®-100. Catalyst
culation limit. Stable activity pro- crack the heaviest feed components properties are shown in Table VI.
vides the necessary delta coke for without compromising coke or gas
an ultra-clean hydrotreated feed. selectivity. The balance between As shown in Figure 27, MIDAS®-238
Grace Davison is pleased to intro- properly sized mesopores and opti- catalyst increases conversion by 2.5
duce two new catalyst technologies mized matrix acid strength delivers points over the base MIDAS®-138
that address these specific opera- the selectivity advantages over com- catalyst at constant C/O in Davison
tional challenges. The MIDAS®-200 petitive catalyst systems. Pore volume Circulation Riser (DCR) Pilot Plant
and PINNACLE® series of catalysts in the 100-600 Å range is critical for testing. The increased hydrogen
deliver increased activity relative to the destruction of heavy hydrocar- transfer activity of the zeolite
IMPACT® and MIDAS®-100 catalysts. bons and MIDAS® catalysts contain improves gasoline selectivity and
Both catalysts are excellent singular the highest amount of mesoporosity reduces wet gas yields without
catalyst solutions and exhibit syner- available in the industry today. compromising coke selectivity
gistic benefits when utilized in GEN- (Table VII).
ESIS™ catalyst systems. To answer the industry call for higher
activity, we have commercialized the Commercial testing demonstrates
New High Activity MIDAS® MIDAS®-200 series of catalysts. Each the activity advantage of the
Catalyst Technology is FCC catalyst technology is limited by MIDAS®-200 catalyst series of cata-
Commercialized the amount of active ingredients that lyst. Refiner F switched from GENE-
can be incorporated before attrition SIS™-1 catalyst, containing
MIDAS® catalysts are designed for resistance is compromised. The MIDAS®-138 catalyst, to GENE-
refiners who are interested in maxi- MIDAS®-200 family of catalysts SIS™-2 catalyst with an equivalent
mum bottoms upgrading and cata- achieves our goal of increased con- amount of MIDAS®-238 catalyst
Table VI replacing the MIDAS®-138 catalyst
Properties of MIDAS®-200 Catalysts in the formulation. As shown in
Figure 28, GENESIS™-2 catalyst
increased equilibrium catalyst
Fresh Catalyst Properties MIDAS-238 MIDAS-138 (Ecat) MAT by 1# at equivalent cat-
alyst additions. Commercial data is
Al2O3 wt.% 49.8 50.8
Re2O3 wt.% 2.35 1.97
shown in Figure 29 and confirms the
gasoline selectivity advantages
Surface Area m2/g 280 290 seen in pilot testing.
ZSA m2/g 170 175
MSA m2/g 110 115 MIDAS®-200 series catalysts are
ABD g/cc 0.73 0.75 currently in use at four FCC units
DI 12 12
around the world. The ability to
adjust MIDAS® catalyst activity over
Deactivated Properties, with 3,000 ppm Ni and 3,000 ppm V
a broad range provides additional
Surface Area m2/g 155 160 flexibility to GENESIS™ catalyst
ZSA m2/g 90 85 users who may require higher MAT
MSA m2/g 65 75 but cannot sacrifice coke or bot-
Unit Cell Size Å 24.32 24.30 toms selectivity.
Grace Announces Improvements

Figure 27 to the PINNACLE® Family of
MIDAS®-238 Catalyst is More Active than Catalysts
MIDAS®-138 Catalyst at Constant C/O Ratio
We are pleased to announce an
improvement to the PINNACLE®
9.5
family of catalysts. PINNACLE®
9.0
catalysts are now formulated with
8.5 novel zeolite technology that maxi-
mizes gasoline selectivity at equiva-
Catalyst to Oil Ratio
8.0
7.5
lent coke yield. The new formula-
tions are appropriate for refiners
7.0
interested in maximizing total fuels
6.5
production and reducing the load
6.0 MIDAS-238 on their wet gas compressor
5.5
MIDAS-138
(WGC).
5.0
66 68 70 72 74 76 78 Introduced in 2006, PINNACLE®
Conversion, wt.%
catalysts are alumina-sol catalysts
formulated with Grace Davison’s
nickel resistant matrix (TRM-400)
that provides superior coke and gas
Table VII selectivity in the presence of high
Resid Feed Constant Activity Comparison nickel feedstocks5. PINNACLE® cat-
alyst technology combines TRM-
MIDAS-238 MIDAS-138 400 with a moderate level of the
integral vanadium trap that is a sig-
nature of our IMPACT® catalyst tech-
Conversion, wt.% 74.5 72.0 nology. This combination of premi-
um metals-trapping technologies
Total Dry Gas, wt.% 2.8 3.1 minimizes deactivation and reduces
Total C3, wt.% 10.1 9.9 coke formation from contaminants,
Total C4, wt.% 13.6 13.2 resulting in enhanced zeolite stabili-
ty. And just like IMPACT® catalysts1,
Gasoline, wt.% 41.6 39.5 PINNACLE® catalyst’s unique inte-
LCO, wt.% 18.4 19.9 gral metals traps are effective at
low, moderate, and high metals
Bottoms, wt.% 7.1 8.1
Coke, wt.% 6.2 6.1
loadings, demonstrating a coke and Figure 28
gas advantage over competitive GENESIS™ with MIDAS®-238 Catalyst Increased Ecat MAT
catalysts whether processing
hydrotreated, VGO or resid feed-
stock.
79
The proprietary zeolite modifica-
tions in PINNACLE® catalysts shift
Ecat MAT, wt.%

light ends into gasoline and deliver 78
incremental activity relative to
IMPACT® catalyst formulations. This 77
higher activity should benefit opera-
tors of circulation-constrained FCC
units. The new zeolite system has 76
been optimally formulated with
Grace matrix technologies to pre- Refiner F
75
vent any detriment to physical prop-
erties or coke selectivity, making GENESISTM - 1 GENESISTM - 2
PINNACLE® catalyst a key blend

component for GENESIS™ catalyst
systems. Figure 29
Commercial Data Demonstrates Gasoline
Pilot plant data below demonstrate
the performance differences Selectivity Advantage of GENESIS™-2 Catalyst
between two GENESIS™ catalyst
systems. The base case, GENE- 51
SIS™ catalyst A, is comprised of GENESIS-1
GENESIS-2
IMPACT® catalyst and MIDAS® cat-
50
C5 + Gasoline, wt.%
alysts. GENESIS™ catalyst B is

comprised of PINNACLE ® cata-
49
lyst/MIDAS® catalyst. Note both
blends contain the same formula-
48
tion of MIDAS ® catalyst in the
same proportion. Catalyst proper-
47
ties are presented in Table VIII.
Both blends were deactivated and Refiner F
tested over VGO feed in an ACE 46
70 72 74 76 78 80
unit.
Conversion, wt.%
The proprietary enhanced zeolite

technology in PINNACLE® catalyst
achieves equivalent conversion at
1# lower C/O. The boost in activity Table VIII
increases gasoline yield by 0.5 GENESIS™ Catalyst Properties
wt.% (Figure 30) and reduces wet
gas by 0.3 wt.% at equivalent coke GENESIS A GENESIS B
(Table IX). The yield benefits are IMPACT® PINNACLE®
accomplished at equivalent coke to w/ MIDAS® w/ MIDAS®
bottoms (Figure 31). As with
MIDAS®-200 catalyst, the alleviation Al2O3, wt.% 46.7 48.4
RE2O3, wt.% 4.9 3.6
of a WGC constraint allows room to
SA, m2/g 315 300
increase conversion either through Zeolite, m2/g 240 225
activity or unit severity optimiza- Matrix, m2/g 75 75
tions. Available neat or in GENE-
SIS™ catalyst systems, PINNACLE® Deactivated Properties
SA, m2/g 175 160
catalyst delivers increased prof-
Zeolite, m2/g 125 110
itability to any refinery interested in Matrix, m2/g 50 50
maximizing gasoline production.

Conclusions Figure 30
GENESIS™ Catalyst Systems with PINNACLE®
In this issue we have shown how
GENESIS™ catalyst systems have
Catalyst Boost Gasoline Selectivity
improved the operations at six
refineries. Contact your Grace 53
Davison technical sales representa-
52
tive to hear more about how GENE-
C5+ Gasoline, wt.%

SIS™ catalyst systems can enhance 51
your operation and refinery prof-
50
itability.
GENESISTM A
49
GENESISTM B
48
References
47
1.Schiller, R.K., et.al., “The GENESIS™ 62 64 66 68 70 72 74 76 78 80
Catalyst System,” Catalagram 102, Fall 2007.
Conversion, wt.%
2.Abrahamsen, A.R. & Geldart, D., "Powder

Technology", 26 (1980), pp35-46.
3.Mott, R.W., “Trouble-Shooting FCC

Standpipe Flow Problems,” Catalagram 83,
1992.
Table IX
4.Hunt, L., “Maximize Bottoms Upgrading:
Give Resid the MIDAS® Touch,” Catalagram VGO Feedstock Constant Conversion Comparison
93, Fall 2003.
5.Petti, N., Yaluris, G., Hunt, L., “Recent GENESIS A GENESIS B

Commercial Experience in Improving
Refinery Profitability with Grace Davison Catalyst to Oil Ratio 7.2 6.1
Alumina-sol Catalysts,” Catalagram 99,
Spring 2006. Hydrogen, wt.% 0.13 0.11
Dry Gas, wt.% 1.70 1.63
Total C3's, wt.% 5.7 5.6
Total C4’s, wt.% 11.0 10.9
Gasoline, wt.% 51.4 51.9
LCO, wt.% 21.1 21.0
Coke, wt.% 2.8 2.6
Figure 31
Both GENESIS™ Catalyst Formulations
Have Equivalent Coke to Bottoms
10
GENESIS A
8
Bottoms, wt.%
GENESIS B
4
1.5 2 2.5 3 3.5 4
Coke (wt.% feed)
Grace Davison:
Your proven FCCU emissions solution.
Grace Davison environmental technologies
have helped refiners reduce FCCU emissions
for over a quarter of a century.
The FCCU is often the largest point source within the refinery for SOx
and NOx emissions. The use of innovative catalytic technologies
reduces emissions from the FCCU regenerator, without the need for
capital intensive “end-of-pipe” hardware solutions.
Super DESOX® provides industry leading SOx removal effective-

ness. More and more refiners are finding that Super DESOX can
cost effectively control SOx emissions below 25 vppm from a wide
range of uncontrolled SOx baseline emissions.
Grace Davison researchers have studied the complex forma-

tion of NOx and developed two additives to reduce NOx.
XNOx® is a low NOx combustion promoter designed to
replace conventional promoters, which often cause
increased NOx formation. For units not using a promoter,
or requiring additional NOx reduction, DENOX® is the
additive of choice. NOx reductions in excess of 50%
have been observed commercially with both products.
With more than 50 commercial applications of NOx
additives, Davison has more experience than the rest
of the industry combined and would be happy to
assist you in reducing FCCU NOx emissions.
SO
Ox x Extensive customer-driven research efforts at Grace are providing new insights into improving
Davison
Environmental
N
SOx and NOx removal. Want to find out more about our environmental technologies?
Technologies
CO
Contact us at www.e-catalysts.com or call us at (410) 531-8226.
Let Grace Davison help you meet your FCCU emission challenges.
Effect of
Hydrocarbon Partial Pressure
on Propylene Production
in the FCC
Ruizhong Hu
Introduction pressure. Studies documenting the
effect of hydrocarbon partial pres-
Senior Principal Scientist any refiners have continually sure on FCC yields are scarce.
Gordon Weatherbee
Principal Engineer
M revamped and debottle-
necked their FCC units to
increase feed throughput and improve
It is generally expected that an
increase in hydrocarbon partial
Hongbo Ma
profitability Most FCC units are run- pressure will increase the rate of all
ning at a significantly higher feed rate bimolecular reactions, including
than the original design. With higher hydrogen transfer, relative to crack-
Research Engineer throughput, in order to maintain cata- ing, which is unimolecular. An
Terry Roberie
lyst and vapor velocity in the riser and increase in the rate of hydrogen
cyclones, the unit pressure and con- transfer will result in a reduction of
sequently the hydrocarbon partial olefins in both gasoline and LPG
Director, FCC Evaluations
and an increase in gasoline range
Wu-Cheng Cheng
pressure need to be increased.
Current laboratory methods for evalu- aromatics and paraffins. The
ating FCC catalysts and additives change in the rate of hydrogen
Director, Grace Davison Refining cannot match hydrocarbon partial transfer could also affect gasoline
Technologies R&D pressures in commercial FCC units. sulfur concentration and the effec-
One reason is that available laborato- tiveness of gasoline sulfur reduction
Grace Davison Refining ry testing equipment, such as ACE catalysts and additives. Moreover,
Technologies and MAT typically operate at atmos- the effectiveness of ZSM-5 addi-
pheric pressure. The Davison tives, which are used to produce
Circulating Riser (DCR), a pilot plant- light olefins, especially propylene,
scale testing unit, is regularly operat- could be affected by hydrocarbon
ed under total pressure similar to com- partial pressure. Since ZSM-5
mercial FCC units1. However, due to works by cracking gasoline range
the small diameter of the DCR riser, a olefin molecules, changing the rate
relatively large amount of nitrogen is of hydrogen transfer could have a
needed to lift the catalyst, thus profound impact on propylene yield.
decreasing the hydrocarbon partial
This paper will discuss the results of Schematic Diagram of Grace Davison DCR
a series of cracking experiments in
the DCR, where the hydrocarbon
partial pressure was varied by vary-
Meter
ing the total reactor pressure, the
feed rate and the amount of lift gas. Control
Condenser
Valve
The effect of changing hydrocarbon
Regenerator
partial pressure on hydrocarbon Stabilizer
Column
yields, especially that of light olefin,
Riser Reactor
and gasoline sulfur will be dis- r
Stripper
cussed. Feed
Tanks
Scale Scale
Experimental
Heat
Feed Preheater Exchanger
At the right is a schematic diagram Liquid Product
of the standard DCR setup. The Receivers
range of operating conditions in the Feed Pump
Dispersant Stripping
Steam Steam
DCR is shown in Table X. Operation
of the DCR has been described
previously1. Similar to commercial
FCC units, the DCR is operated in
adiabatic mode. In typical DCR tube heat exchanger. The rate of heat nitrogen lift gas and steam injection
operation, the regenerator tempera- transfer across this exchanger pro- rate constant while increasing the
ture, the riser outlet temperature vides a precise and reliable method to total pressure and feed rate. The
and the feed rate are set. The cata- calculate the catalyst circulation rate. latter case is similar to some com-
lyst circulation rate and thus, the The stabilizer column, also called the mercial FCC unit revamps where
catalyst to oil ratio, is changed by debutanizer column, is operated to the total pressure of a FCC unit is
varying the feed pre-heat tempera- separate C4 minus from the liquid increased to accommodate higher
ture. During operation of the DCR, product, which is condensed and col- feed and catalyst circulation rate.
a metering pump precisely controls lected. The collected liquid is ana-
the feed rate as feed is pumped lyzed by GC (SIMDIS – simulated dis- Table XI shows the three DCR oper-
from the load cell through a pre- tillation) to provide gasoline (ibp - ation conditions. Condition 3 is a
heater. Nitrogen and steam, inject- 430°F), LCO (430-700°F), and 700°F + commonly used DCR operating
ed through a separate pre- bottoms fractions. The gaseous prod- condition, while Conditions 1 and 2
heater/vaporizer, are used as a feed ucts are metered and batch collected are modifications to raise the hydro-
dispersant. Catalyst and product for subsequent analysis by GC. carbon partial pressure closer to
pass from the riser to the stripper the value in commercial FCC opera-
overhead disengager. Products exit We investigated two methods of tions. Since cracking is a molecular
the disengager through a refrigerat- changing hydrocarbon partial pres- weight reduction process, the
ed stabilizer column to a control sure. The first method involved keep- hydrocarbon mole fraction and,
valve which maintains unit pressure ing the total pressure, feed rate, and therefore, partial pressure increase
at the desired level. A section of the steam injection rate constant while along the riser. The molar expan-
stripper-regenerator spent catalyst reducing the nitrogen lift gas. The sec- sion (moles of product/moles of
transfer line consists of a shell and ond method involved keeping the feed) in a typical FCC unit is
between four and five. For the pur-
Table X pose of engineering calculations, it
is common to approximate the
DCR Operating Ranges hydrocarbon mole fraction as equal
Control Parameter Range to 1/3 of the mole fraction at the inlet
and 2/3 of the mole fraction at the
System Pressure < 45 psig
outlet of the riser. The total moles of
Catalyst Charge 1500-4000 g
the hydrocarbon products are cal-
Catalyst Circulation Rate 2500-15000 g/h
culated by using GC analyses of
Feed Rate 350-2000 g/h
the light gases and gasoline PIONA
Feed Types GO, VGO, Resid
and assuming average molecular
Feed Preheater Temperature 120-400°C (250-750°F) weight values of 220 and 350 for
Riser Temperature <590°C (<1100°F) LCO and bottoms, respectively.
Disengager Temperature <746°C (800-1100°F)
Stripper Temperature 427-593°C (800-1100°F)
Stabilizer Column Temperature -34°C (-30°F)

Table XI
Operating Conditions in the DCR
DCR C o m m ercial
C o n d itio n 1 C o n d itio n 2 C o n d itio n 3 AC E FCCU
R eactor T op T em p (°F ) 970 970 970
R eg enerator T em p (°F ) 1300 1300 1300
U nit P ressure (psig ) 40 25 25
R eactor D elta P (in H 2 O ) 5.24 4.11 3.45
F eed T em p (°F ) 575 575 300
F eed R ate (g /h) 1500 1000 1000 180
R eactor W ater R ate (g /h) 30 30 30
R eactor N 2 (L/h) 25 31 131 7.8
C atalyst C ir. R ate (g /H ) 8820 6160 6790
C /O R A T IO 5.9 6.1 6.9 6.0 6 to 9
C onversion w t% 72.0 72.2 72.7
M olar E xpansion 4.3 4.3 4.2
G as R esidence tim e (s) 2.4 2.3 1.7 2.0 2 to 4
S lip F actor 2.0 2.0 1.7 Infinite 1 to 1.3
C atalyst C ontact T im e (s) 4.9 4.6 2.9 30 to 150 2 to 5
C atalyst H old U p (g ) 11.9 7.9 5.4 9
W H S V (h-1) 125 126 183 20 100 to 250
H C P artial P ressure 1/3
inlet + 2/3 outlet (psia) 44 28 20 12 20 to 50
In varying hydrocarbon partial pressure of Condition 3 was 2.3 times to the operating conditions of ACE,
sure, we chose operating conditions lower. However, its WHSV was also the operating conditions of the DCR
so as not to greatly change the somewhat greater, due to the higher are much closer to those of the
weight hourly space velocity level of lift nitrogen used. In this case, commercial unit.
(WHSV), as that in itself could we would need to rationalize the con-
change the selectivity and compli- tribution of hydrocarbon partial pres- Two Davison commercial FCC cata-
cate the interpretation of the results. sure to the selectivity shifts. lysts, labeled Catalyst A and
The slip factor in the riser (ratio of Catalyst B, containing 1.2 and 3.1%
the gas velocity to catalyst velocity), Table XI also compares the current RE2O3, respectively, were used in
estimated by the correlation of DCR operating conditions with that of this study. Both catalysts were
Pugsley and Berruti2, varied from commercial FCC units and ACE. steam deactivated according to
1.7 to 2. These values were consis- Compared to the earlier operating CPS-3 protocol4 at 1480˚F with 500
tent with those reported by Bollas et conditions (Condition 3), Conditions 1 ppm nickel and 500 ppm vanadium.
al.3. Once the slip factor was deter- and 2 are closer to the commercial The chemical and physical proper-
mined, the catalyst holdup (the units, especially in hydrocarbon par- ties of the two catalysts are listed on
amount of catalyst in the riser), cat- tial pressure. Furthermore, compared Table XII. The deactivated unit cell
alyst contact time and WHSV were
readily calculated (Table XI). The
Table XII
catalyst holdup values followed the
trend of pressure drop measure- Properties of Catalysts Deactivated at
ments across the riser. Conditions 1 500ppm Ni/500ppm V CPS-3/1480˚F
and 2 varied in hydrocarbon partial
pressure by a factor of 1.55. Analysis Catalyst A Catalyst B
However, the values of the WHSV, Al2 O 3 , wt.% 40.8 46.9
catalyst-to-oil ratio and conversion RE 2 O 3 , wt.% 1.16 3.05
were essentially identical. Na 2 O, wt.% 0.39 0.28
Therefore, the changes in selectivity Ni, ppm 537 523
could be attributed principally to the V, ppm 520 510
change in hydrocarbon partial pres- Surface Area, m2 /g 219 146
sure. Compared to Conditions 1 ZSA, m2 /g 179 113
and 2, the hydrocarbon partial pres- MSA, m2 /g 40 33
Unit Cell Size, Å 24.24 24.32
Table XIII
Properties of VGO Feedstock
API Gravity 25.5 Average MW 406

Specific Gravity, g/cm3 0.9012 Ni, ppm 0.4
K Factor 11.94 V, ppm 0.2
Refractive Index 1.5026
Sulfur, wt.% 0.369 Similated Distillation
Basic Nitrogen, wt.% 0.05 IBP 307
Total Nitrogen, wt.% 0.12 10% 607
Conradson Carbon, wt.% 0.68 30% 740
ndm Analysis 50% 818
Arom Ring Carbons Ca, wt.% 18.9 70% 904
Naphthenic Ring Carbons Cn, wt.% 17.4 90% 1034
Paraffinic Carbons Cp, wt.% 63.6 End Point 1257
size measurements of the low and feed rate and catalyst circulation the propane/propylene, n-butane/(1-
high RE2O3 catalysts are 24.24Å total pressure of the unit has to be butylene + trans-2-butylene + cis-2-
and 24.32Å, respectively. Catalyst increased to maintain velocity. butylene) and isobutane/isobuty-
A was also blended with 20% lene. These ratios are shown in
OlefinsUltra® additive, a commer- The plots of catalyst to oil ratio, total Table XIV. In this analysis we are
cially available ZSM-5 additive and C3, total C4, gasoline, LCO, and coke assuming that the C3 and C4 alka-
deactivated according to CPS-3 yields against conversion are shown in nes are the product of hydrogen
protocol at 1480˚F with 500 ppm Figure 32. Increasing HC partial pres- transfer from their parent alkenes
nickel and 500 ppm vanadium. A sure increases dry gas and coke at and ignoring the alkanes formed by
Gulf Coast vacuum gas oil feed the expense of gasoline. The yields of thermal or protolytic cracking. The
was used in this study. The prop- total C3, C4 and LCO remain about the hydrogen transfer reaction of
erties of the feedstock are shown same. The higher coke yield may be isobutene proceeds via a tertiary
on Table XIII. attributed to a higher rate of oligomer- carbenium ion intermediate and
ization, which is a bimolecular reac- thus occurs at a much faster rate
Results and Discussion tion and favored at high pressure. The than the hydrogen transfer reactions
higher dry gas could be the result of of propylene and linear butenes,
Case I oligomerization/recracking. which proceed through a less sta-
ble secondary carbenium ion inter-
In this example, Catalyst A was test- Figure 33 shows that increasing the mediate. All of the hydrogen trans-
ed in the DCR under both HC partial pressure decreases the fer indices increase by a factor of
Conditions 1 and 2. Under yields of propylene, butenes, and 1.5, as the HC partial pressure
Condition 1, the unit pressure was gasoline olefins, while increasing the increases almost proportionally by a
40 psig, the feed rate was 1500 g/h, yield of gasoline isoparaffins. factor of 1.6 from 28 to 44 psia.
the dispersing steam was 30 g/h, Increasing HC partial pressure sub- Thus, all the yield shifts are consis-
and 25 l/h nitrogen was injected to stantially lowers the C3 and C4 tent with an increase in the rate of
help disperse the feed as well as to hydrogen transfer with the increase
olefinicities. These yield shifts sug-
lift the catalyst. Based on the above in HC partial pressure.
gest that the rate of hydrogen transfer
discussion, the time-averaged (1/3
increases with HC partial pressure, as
inlet + 2/3 outlet) hydrocarbon par- Case II
would be expected for a bimolecular
tial pressure under this condition
reaction.
was 44 psia. Under Condition 2, the Catalyst A was blended with 20%
unit pressure was 25 psig, the feed OlefinsUltra® additive and tested in
The interpolated yields at 73 wt.%
rate was 1000 g/h, whereas the the DCR under Conditions 1 and 2.
conversion are listed on Table XIV. A
steam and nitrogen flow rates were The catalyst to oil ratio, total C3, total
convenient way to gauge the hydro-
the same as that of Condition 1.
gen transfer rate is to look at the paraf- C4, gasoline, LCO, and coke yields
The hydrocarbon partial pressure
fins to olefins ratio of C3, as well as lin- against conversion plots are shown
under this condition is 28 psia. The
above comparison is very similar to ear and branched C4 compounds5. in Figure 34. The yields of C3=,
a common revamp of a commercial The hydrogen transfer indices are C4=, gasoline olefins and gasoline
FCC unit where in order to increase defined as the ratios of isoparaffins, as well as the olefinici-

Figure 32
Effect of DCR Operating Conditions on the Yields of Catalyst A
70.0 72.5 75.0

C/O Ratio Total C3, wt.% Total C4, wt.%
8 12
6.5
7 11
6.0
6
10
5.5
5
9
5.0
Gasoline, wt.% LCO, wt.% Coke, wt.%
4.5
52
23
4.0
51 22
3.5
21
50
20 3.0
49
19 2.5
48
70.0 72.5 75.0 70.0 72.5 75.0
Conversion, wt.%
28 psia base 44 psia base
Figure 33
Effect of DCR Operating Conditions on the Olefins Yield and Olefinicity of Catalyst A
70.0 72.5 75.0

C3 =, wt.% Total C4 =, wt.% G-Con O, wt.%
5.5
7.6 35.0
5.0 7.2 32.5
6.8 30.0
4.5
6.4 27.5
4.0 6.0 25.0

C3 = / Total C3 C4 = / Total C4 G-Con I, wt.%
0.76 28
0.86
0.72 26
0.84
0.68 24
0.82 22
0.64
0.80 0.60 20
70.0 72.5 75.0 70.0 72.5 75.0

Conversion, wt.%
Table XIV
Interpolated Yields at 73 wt.% Conversion Over Catalyst A
Condition 1 Condition 2
Ratio of HC Pressure
HC Partial Pressure, psia 44 28 1.6
Cat to Oil 6.1 6.5
H2 Yield , wt.% 0.05 0.05
C 1 + C 2 's , wt.% 2.8 2.4
C 2 = , wt.% 0.8 0.8
Total C 3 , wt.% 5.9 5.9
C 3 = , wt.% 4.7 5.0
Total C 4 , wt.% 10.5 10.6
iC 4 , wt.% 3.0 2.5
nC 4 , wt.% 0.8 0.6
Total C 4 = , wt.% 6.7 7.5
iC 4 = , wt.% 1.9 2.4
Gasoline, wt.% 50.1 50.8
G-Con P, wt.% 4.0 3.7
G-Con I, wt.% 24.8 22.2
G-Con A, wt.% 30.5 30.2
G-Con N, wt.% 10.9 11.5
G-Con O, wt.% 29.6 32.2
G-Con RON EST 92.2 92.0
G-Con MON EST 79.8 79.2
LCO, wt.% 20.5 20.7
Coke, wt.% 3.5 3.0
Hydrogen Transfer Index Ratio of HT Index
C 3 /C 3 = 0.24 0.17 1.4
nC 4 /(1C 4 = + t 2 C4 = +c 2 C 4 =) 0.18 0.12 1.5
iC 4 /iC 4 = 1.56 1.05 1.5
Figure 34
Effect of DCR Operating Conditions on the Yields of
Catalyst A with 20% OlefinsUltra® Additive
70.0 72.5 75.0

C/O Ratio Total C3, wt.% Total C4, wt.%
8 12.5
16.5
12.0 16.0
7
15.5
11.5
6 15.0
11.0 14.5
Gasoline, wt.% LCO, wt.% Coke, wt.%
41 22
4.0
21
40 3.5
20
3.0
39 19
2.5
18
38 2.0
70.0 72.5 75.0 70.0 72.5 75.0
Conversion, wt.%
28 psia 20% OlefinsUltra® 44 psia 20% OlefinsUltra®

Figure 35
Effect of DCR Operating Conditions on the Olefins Yield and
Olefinicity of Catalyst A with 20% OlefinsUltra® Additive
70.0 72.5 75.0

C3 =, wt.% Total C4 =, wt.% G-Con O, wt.%
11.5 11.7 40.0
11.0 37.5
11.4
35.0
10.5 11.1
32.5
10.0 10.8
30.0
9.5 10.5
C3 = / Total C3 C4 = / Total C4 G-Con I, wt.%
0.80
0.900
21.0
0.885 0.75
19.5
0.870
0.70 18.0
0.855
16.5
0.840 0.65
15.0
70.0 72.5 75.0 70.0 72.5 75.0
Conversion, wt.%
ties of C3, C4, and gasoline are ZSM-5. The rate of hydrogen transfer, position are shown in Figure 37. For
shown in Figure 35. As in the case as estimated by the hydrogen transfer both the high and low unit cell size
without OlefinsUltra® additive (Case indices, described above, increases catalysts, the response of the LPG
I), increasing HC partial pressure approximately proportionally to HC and gasoline olefin yields to the
increases coke and dry gas and partial pressure. changes in DCR conditions are very
dramatically decreases gasoline similar to that observed in Case 1,
and LPG olefinicity. The C3 olefinic- Case III namely increasing HC partial pres-
ity of ca. 0.84 at the higher HC par- sure decreases LPG and gasoline
tial pressure is much more realistic In this example, Catalysts A and B, olefins and olefinicity. The hydrogen
and close to the commercially having unit cells size values of 24.24Å transfer indices increased by a fac-
observed values. and 24.31Å, respectively, were tested tor of two as HC partial pressure
under DCR Conditions 1 and 3. increased by a factor of 2.3 (Table
The interpolated yields at constant Condition 3 featured a unit pressure of XVI). Thus, as in Case I, the change
conversion of 73 wt.% are shown on 25 psig, 1000 g/h feed rate, 30 g/hour in the hydrogen transfer indices are
Table XV. Remarkably, increasing steam, and 128 l/h nitrogen. The main nearly proportional to the change in
the HC partial pressure from 28 to difference between Condition 1 and HC partial pressure. This suggests
44 psia decreases the propylene Condition 3 was the greater amount of that the change in HC partial pres-
yield by 1 wt.% absolute and nitrogen lift gas used in Condition 3, sure is mainly responsible for the
decreases the butylenes yield by which not only decreased the HC par- yield changes while the shifts in
0.6 wt.% absolute. It is known that tial pressure by a factor of 2.3, from WHSV may be responsible for the
the addition of ZSM-5 increases 44 to 19 psia, but also increased the shifts in conversion at a given cata-
LPG olefins by cracking gasoline WHSV by a factor of 1.4. The effect of lyst to oil ratio.
range olefins6-8. Increasing the rate the change in WHSV will be dis-
of hydrogen transfer, by increasing cussed. The yields shifts due to changing
the HC partial pressure, depletes unit cell size are consistent with
the gasoline range olefins and The main yields are shown in Figure what has been reported in the liter-
decreases the effectiveness of 36, while the LPG and gasoline com- ature9,10, namely that the higher UCS
Table XV
Interpolated Yields at 73 wt.% Conversion Over Catalyst A with 20% OlefinsUltra® Additive
Condition 1 Condition 2
Ratio of HC Pressure
HC Partial Pressure, psia 44 28 1.6
Cat to Oil 6.7 6.7
H2 Yield , wt.% 0.04 0.05
C 1 + C 2 's , wt.% 3.0 2.6
C 2 = , wt.% 1.4 1.3
Total C 3 , wt.% 11.4 12.0
C 3 =, wt.% 9.7 10.7
Total C 4 , wt.% 15.8 15.5
iC 4 , wt.% 3.9 3.3
nC 4 , wt.% 1.0 0.8
Total C 4 = , wt.% 10.9 11.4
iC 4 = , wt.% 4.1 4.3
Gasoline , wt.% 39.2 39.8
G-Con P , wt.% 4.0 3.8
G-Con I , wt.% 19.6 17.1
G-Con A , wt.% 33.7 33.9
G-Con N , wt.% 8.9 9.2
G-Con O , wt.% 33.5 35.8
G-Con RON EST 95.2 94.9
G-Con MON EST 81.5 80.9
LCO , wt.% 19.6 19.8
Bottoms , wt.% 7.3 7.1
Coke , wt.% 3.3 2.9
Hydrogen Transfer Index Ratio of HT Index
C 3 /C 3 = 0.17 0.12 1.5
nC 4 /(1C 4 = + t 2 C4 = +c 2 C 4 =) 0.15 0.11 1.4
iC 4 /iC 4 = 0.95 0.76 1.2
Table XVI
Interpolated Yields at 75 wt.% Conversion Over Catalyst A and B
Catalyst A Catalyst B
Condition 3 Condition 1 Condition 3 Condition 1
HC Partial Pressure, psia 19 44 19 44

Cat to Oil 7 7 5 5
H2 Yield, wt.% 0.04 0.05 0.03 0.03
C1 + C2's, wt.% 2.2 2.7 1.9 2.2
C2=, wt.% 0.7 0.8 0.6 0.7
Total C3, wt.% 5.9 6.2 5.6 5.6
C3=, wt.% 5.2 5.0 4.9 4.4
Total C4, wt.% 10.8 11.2 10.4 10.3
iC4, wt.% 2.4 3.4 2.9 3.7
nC4, wt.% 0.5 0.9 0.6 1.0
Total C4=, wt.% 8.0 6.9 6.8 5.5
iC4=, wt.% 2.7 1.9 1.8 1.4
Gasoline, wt.% 53.6 50.7 53.6 53.4
G-Con P, wt.% 3.4 4.0 4.0 4.6
G-Con I, wt.% 20.8 26.3 26.3 31.8
G-Con A, wt.% 29.1 31.3 30.8 32.3
G-Con N, wt.% 11.4 10.6 12.5 12.1
G-Con O, wt.% 35.4 27.7 26.3 19.5
G-Con RON EST 92.4 92.1 89.9 88.8
G-Con MON EST 79.2 80.0 78.4 78.6
LCO, wt.% 20.0 19.3 19.0 19.5
Bottoms, wt.% 6.0 5.6 5.9 5.6
Coke, wt.% 2.9 3.9 3.2 3.7
Hydrogen Transfer Index
C3/C3= 0.13 0.24 0.14 0.27
nC4/(1C4= + t2C4= +c2C4=) 0.09 0.18 0.12 0.24
iC4/iC4= 0.88 1.79 1.57 2.69

Figure 36
Variaton of Yields with Unit Cell Size and DCR Operating Conditions
64 72 80
C/O Ratio Total C3 wt.% Total C4 wt.%
6.5 12
7
6.0 11
6
5.5 10
5 5.0 9
4.5 8
Gasoline wt.% LCO wt.% Coke wt.%
54.0 25.0
6
52.5 22.5
5
51.0
20.0 4
49.5
17.5 3
48.0
15.0 2
64 72 80 64 72 80
Conversion, wt.%
24.24 19 psia 24.24 44 psia 24.32 19 psia 24.32 44 psia
Figure 37
Variaton of Olefins Yield and Olefinicity with Unit Cell Size and DCR Operating Conditions
64 72 80
C3 = wt.% Total C4 = wt.% G-Con O wt.%
5.6 35
7.5
5.2 30
7.0
4.8 25
6.5
4.4 6.0 20
4.0 5.5 15
C3 = / Total C3 C4 = / Total C4 G-Con I wt.%
0.8 36
0.87
0.84 0.7 32
0.81 0.6 28
0.78 24
0.5
0.75
20
0.4
64 72 80 64 72 80
Conversion, wt.%
24.24 19 psia 2.24 44 psia 2.32 19 psia 2.32 44 psia
Scheme 1 4.D. Wallenstein, R.H. Harding, J.R.D. Nee,
L.T. Boock, “Recent Advances in the
Deactivation of FCC Catalysts by Cyclic
R R
Propylene Steaming (CPS) in the Presence
R -C4 + H2S and Absence of Metals,” Appl.Catal. A:
General 204 (2000) 89.
S S cracking
HT
5.Cheng, W-C., Suarez, W., and Young, G. W;
catalyst makes higher gasoline, nature, increase with increasing HC “The effect of catalyst properties on the
lower octane, lower LPG and gaso- partial pressure. The hydrogen trans- selectivities of isobutene and isoamylene in
FCC,” AIChE Symposium Series, 291
line olefins. These trends are fer index, defined as the paraffin/olefin (1992) 38.
observed at both DCR conditions. ratio of C3, linear C4 and branched C4
The effect of unit cell size and HC species increase almost linearly with 6.K. Rajagopalan, G.W. Young, in Fluid
partial pressure on the rate of Catalytic Cracking − Role in Modern
HC partial pressure. It has been
Refining, M.L. Occelli (Ed.), ACS Symposium
hydrogen transfer appears to be demonstrated that the effectiveness of Series 375 (1988) 34.
simply additive. The rate of bimole- ZSM-5 additives is lessened at high
cular reactions can be increased by HC partial pressure due to the deple- 7.X. Zhao, T.G. Roberie,”ZSM-5 Additive in
increasing acid site density as well Fluid Catalytic Cracking. 1. Effect of Additive
tion of gasoline range olefins via
Level and Temperature on Light Olefins and
as increasing HC partial pressure. hydrogen transfer reactions. The con- Gasoline Olefins,” Ind. Eng. Chem. Res. 38
The ratios of the hydrogen transfer centration of gasoline sulfur species (1999) 3847.
indices of Catalyst B to Catalyst A decreases at higher HC pressure,
are about the same at both low and 8.R.J. Madon, “Role of ZSM-5 and
again due to higher rate of hydrogen
Ultrastable Y Zeolites for Increasing Gasoline
high HC partial pressure (Table transfer. Recent advancements in Octane Number,” J. Catal. 129 (1991) 275.
XVI). DCR operation enable more realistic
simulation of commercial FCCU oper- 9.L.A. Pine, P.J. Maher, W.A. Wachter,
Figure 38 shows the concentration “Prediction of Cracking Catalyst Behavior by
ation.
a Zeolite Unit Cell Size Model,” J. Catal. 85
of gasoline sulfur (including all thio- (1984) 466.
phene species with a boiling point References
below 430°F, tetrahydrothiophene, 10.G.W. Young, W. Suarez, T.G. Roberie, W.C.
and benzothiophene) for both 1.G.W. Young, G.D. Weatherbee, “FCCU Studies Cheng, “Reformulated Gasoline: The Role of
with an Adiabatic Circulating Pilot Unit,” AIChE Current and Future Catalysts,” NPRA Annual
Catalysts A and B under the two Meeting, AM-91-34, 1991.
Annual Meeting, November, 1989.
DCR operation conditions.
Gasoline sulfur concentration 2.S. T. Pugley, F. A. Berruti, “A Predictive 11.R.H. Harding, R. Gatte, J.A. Whitecavage,
decreases with increasing unit cell Hydrodynamic Model for Circulating Fluidized R.F. Wormsbecher, “Reaction Kinetics of
Bed Risers,” Powder Technol., 89 (1996) 57. Gasoline Sulfur Compounds,” in
size and with increasing HC partial Environmental Catalysis, J.N. Armor (Ed.),
pressure. These results suggest 3.G. M. Bollas, I. A. Vasalos, A. A. Lappas, D. American Chemical Society, Symposium
that the reduction of gasoline sulfur Iatridis, “Modeling Small-Diameter FCC Riser Series 552 (1994) 286.
follows the trend of increase hydro- Reactors, A Hydrodynamic and Kinetic
Approach,” Ind. Eng. Chem. Res., 41 (2002) 12.F. Can, A. Travert , V. Ruaux , J.-P. Gilson ,
gen transfer activity and are consis- F. Maugé , R. Hu, R.F. Wormsbecher, “FCC
5410.
tent with the previously proposed Gasoline Sulfur Reduction Additives:
mechanism, shown below11,12. Mechanism and Active Sites,” J. Catal. 249
(2007) 79.
Scheme 1
Figure 38
High rate of hydrogen transfer Effect of UCS and DCR Operating Conditions
speeds up this reaction by promot-
on Gasoline Sulfur Concentration
ing the formation of the reaction
intermediate, tetrahydrothiophene.
230
220
Conclusions
210
Gasoline Sulfur (ppm)
By varying the operating conditions 200

of the DCR, we have been able to 190
24.24 19 psia
24.32 19 psia
conduct cracking experiments over 24.24 44 psia
180
a wide range of hydrocarbon partial 24.32 44 psia
pressure. The results indicate that 170
increasing the hydrocarbon partial 160
pressure decreases the olefinicity 150

and olefins yield of LPG and gaso-
140
line. This observation is consistent 65.0 67.5 70.0 72.5 75.0 77.5 80.0
with the notion that hydrogen trans- Conversion, wt.%
fer reactions, being bimolecular in

Your Clean Fuels Solutions
When it comes to commercially proven technology to meet increasingly stringent
Clean Fuels regulations, make Grace Davison and Advanced Refining Technologies
your one-stop shop for FCC catalysts, additives and hydroprocessing catalysts.
For gasoline sulfur reduction, today's refiners are challenged more

than ever before to blend various refinery streams to meet stringent
product specifications and government regulations for clean air.
Proven in over 80 commercial units, Grace Davison's portfolio of
FCC gasoline sulfur reduction technologies includes the D-PriSM®
and GSR®-5 additives and the SuRCA®, SATURN® and
Neptune™ catalyst families.
>Recently commercialized, NEPTUNE catalyst further expands

our sulfur reduction technologies to continuously improve both
performance and cost effectiveness. This next generation tech-
nology is a step out improvement over earlier technologies, pro-
viding 35-50% full range gasoline sulfur reduction commercially
with full catalyst formulation flexibility.
>For refiners with FCC pretreaters, the ApART Catalyst System™

utilizing combinations of ART AT575, AT775 and AT792 offers the
opportunity to significantly increase sulfur removal in the
hydrotreater while at the same time maximizing FCC feed quality. The
improved performance of the pretreater results in higher gasoline
potential in addition to decreasing FCC gasoline sulfur.
>For ULSD processing, the SmART Catalyst System™ utilizes

state-of-the-art catalyst technology which is staged in the proper
proportions to provide the best performance, while at the same
time meeting individual refiner requirements. The catalyst stag-
ing is designed to selectively take advantage of the different reac-
tion mechanisms for sulfur removal with efficient hydrogen usage.
ART CDXi, our newest generation of high activity CoMo catalyst,
efficiently removes the unhindered, easy sulfur via the direct
abstraction route, while ART NDXi, our high activity NiMo cata-
lyst, then attacks the remaining sterically hindered, hard sulfur.
The SmART system provides higher activity than either the tradi-
tional CoMo or NiMo catalyst alone while effectively helping the
refiner manage hydrogen utilization.
Have questions about our clean fuels technology?

Contact us at www.e-catalysts.com or call us at (410) 531-8226.
Let Grace Davison and Advanced Refining Technologies
give you the custom catalytic solution to your clean fuels challenges.
Meet Clean Fuels Challenges with
Advanced Refining Technologies
Newest ULSD Catalyst -
ART 420DXTM Catalyst
Brian Watkins n keeping up with refiners’ including Co and Ni. It has been
Supervisor, Laboratory Technology I demand for superior technology shown that when applied correctly,
chelates can promote the formation
Charles Olsen
and premium performance,
Advanced Refining Technologies of Type II metal sulfide sites (see
introduced its line of ultra high activity Catalagram® 96, 2004). ART’s pre-
Worldwide Technical Services DX series of catalysts. ART’s series of mium CDXi has proven its perform-
Manager DXTM catalysts has exceeded refinery ance advantage with greater stabili-
David Krenzke
expectations in their ability to tolerate ty and exceptional ability to utilize a
difficult feed blends in demanding minimum amount of hydrogen to
ULSD applications. Key to that effort provide refiners with consistent
Technical Services Manager is maximizing the utilization of the ULSD production.
active metals on the catalyst through
Advanced Refining Technologies ART’s chelate chemistry. This impreg- In keeping with this tradition, ART is
nation technology offers outstanding releasing its newest generation of
potential for significantly improving ultra high activity CoMo DXTM cata-
metals utilization in catalysis due to a lyst, 420DXTM. Figure 39 compares
superior ability to control metal ions the activity of a variety of CoMo cat-

Figure 39 alysts supplied by ART. Through
Advanced Refining Technologies Line further optimization of ART’s
uniquely engineered pore network,
of High Performance Catalysts this newest member to the DXTM
family is capable of significantly
175
reducing required SOR tempera-
tures for 10 ppm diesel. 420DXTM
150 catalyst offers refiners a significant
advantage due to its ability to
extract even the most hindered sul-
RVA, %
125
fur molecules, yet avoid the unnec-
essary additional saturation of the
100 monoaromatic compounds which
can increase hydrogen consump-
tion.
75
AT405 CDXi 420DXTM
ART’s dedicated staff of
HDS HDN researchers has continued to inves-
tigate ways to improve catalytic per-
formance, and surface acidity has
been identified as an important
Figure 40 property. It is generally accepted
IR Spectra of the Support Material for ART 420DXTM Catalyst that higher surface acidity increas-
es reactions controlled through ring
26 saturation such as nitrogen and hin-
1451.67
24
LC243-260-1
dered sulfur removal. This acidity
22 has also been shown to affect the
20 interaction of active metals with the
18 alumina surface. ART was able to
16
1624.43 exploit this in the design of ART
14
1616.20 420DXTM catalyst. This catalyst uti-
12
1494.12 lizes similar impregnation technolo-
10 gy as CDXi, but is built on a modi-
8 fied alumina carrier which results in
6 a dramatic increase in activity. The
1700 1650 1600 1550 1500 1450 IR chart in Figure 40 shows a dou-
Wave numbers (cm -1) ble peak at 1624 and 1616 wave
numbers as well as one at 1451
which are believed to indicate the
presence of Lewis acid sites. This
Figure 41 feature was not as prevalent in the
Comparison of CDXi and ART 420DXTM Catalyst spectra of the CDXi support and
at High Pressure confirms the incorporation of sur-
face acidity in the new support.
(980 hydrogen partial pressure, 2500 H2/Oil) While the acid sites give ART
420DXTM catalyst superior perform-
650 ance for both HDS and HDN activi-
ty, they are not strong enough to ini-
640 tiate any cracking reactions under
WABT, ˚F (10ppm sulfur
and 1ppm nitrogen)
typical hydrotreating conditions.

630
ART has conducted pilot plant test-

620
ing at various conditions in order to
610
demonstrate the superior perform-
ance of how ART 420DXTM catalyst
600 would perform. Figure 41 shows
HDS HDN
the results of side-by-side testing of
CDXi 420DXTM CDXi and ART 420DXTM catalyst at
980 psi hydrogen partial pressure
and 2500 Scfb H2/Oil. At these con-
Figure 42
Comparison of CDXi and ART 420DXTM Catalyst
at Low Pressure
(580 hydrogen partial pressure, 1500 SCFB H2/Oil)
680
WABT, F (10ppm sulfur

and 1ppm nitrogen)
670
660
650
HDS HDN
CDXi 420DXTM
Figure 43
Comparison of CDXi and ART 420DXTM Catalyst
at ULSD Conditions
300 29
240 27 Total aromatic Content, wt.%

Product Sulfur, ppm
180 25
120 23
60 21
0 19
600 620 640 660 680
Temperature, ˚F
420DXTM CDXi
ditions ART 420DXTM catalyst clearly tial pressure and 1500 Scfb H2/Oil This enhanced sulfur removal activ-
outperforms CDXi by over 20°F at using a feedstock containing cracked ity offers refiners greater flexibility
10 ppm sulfur on a difficult feed material. in meeting their HDS activity
containing 30% cracked stocks. requirements while minimizing
The primary benefit of ART 420DXTM hydrogen consumption using ART
The performance gains seen at high catalyst is that the improved HDS and 420DXTM catalyst as a stand alone
pressures are also available to refin- HDN activity does not result in an catalyst or in combination with
ers operating at lower unit pressure increase in aromatic saturation and ART’s premium NDXi catalyst in a
and hydrogen circulation. Figure 42 consequently does not increase SmART System® for producing
shows the benefits of using ART hydrogen consumption. As can be ULSD from difficult feeds.
420DXTM catalyst at 10 ppm sulfur seen in Figure 43, ART 420DXTM cata-
and lower pressure. A clear 15°F lyst shows equal aromatic conversion
advantage over CDXi is apparent to that of CDXi at lower product sulfur.
even at only 580 Scfb hydrogen par-
Successful Implementation of
State-Of-The-Art ULSD/Dewaxing
Technology at Irving Oil,
Saint John, NB
Mike Beshara
Project Manager, Irving Oil
Greg Rosinski ew regulations for Ultra-Low to maximize ULSD yield since low
Technical Services Engineer,
Advanced Refining Technologies N Sulfur Diesel (ULSD) in
Canada and the United States
cloud point is not required at that
time of year. Irving Oil also wanted
Charles Olsen
took effect in June 2006, reducing the to minimize the hydrogen consump-
on-road diesel sulfur content from 500 tion so that feed rate could be max-
Worldwide Technical Services to 15 ppmw. Anticipating the new sul- imized within make-up hydrogen
Manager, Advanced Refining fur regulation, Irving Oil decided to constraints.
Technologies convert the existing VGO
Ben Prins
Hydrocracker/LCO Desulfurizer at the In the summer of 2004, Irving Oil
Saint John refinery in New Brunswick, contacted several catalyst suppliers
Canada to an LCO /heavy diesel including ART and Süd-Chemie Inc.
Senior Process Engineer, ULSD unit. to begin the catalyst selection
Fluor process for the revamped unit. ART
Garry Jacobs
Technical Requirements is a supplier of top-tier hydrotreat-
ing catalysts, but does not have a
Technical Director, Fluor Fluor Corporation provided engineer- dewaxing catalyst in its portfolio.
Alan Birch
ing, procurement and supported tech- Süd-Chemie offers premium
nology selection. The project included dewaxing catalyst technology but
increasing feed capacity from 30,000 does not have hydrotreating prod-
Account Manager,
to 45,000 BPSD while at the same time ucts. ART and Süd-Chemie joined
Süd Chemie
producing ULSD from a feedstock together to offer a complete pack-
Ernst Köhler
containing up to 50% LCO. Irving Oil age.
wanted to make 7 ppm sulfur diesel
Global Product and required at least 30°F improve- The high level of LCO in the feed,
Manager-Zeolites, ment in cloud point for winter diesel. combined with high unit operating
Süd Chemie During the summer months they want- pressure and the need to minimize
ed to “turn off” the dewaxing function hydrogen consumption made it a
challenge to design the appropriate Table XVII
catalyst system to meet the desired Feedstock Properties
product characteristics. This was
further complicated by the high
level of nitrogen in the feed. Type 50% LCO
API 22.8
Hydrodesulfurization and saturation Sulfur, wt.% 1.16
of olefins and aromatics are very Nitrogen, wppm 409
exothermic reactions and bed activ- Aromatics, vol.%
ity must be controlled to avoid Mono- 17.8
excessive temperature rise. The unit Di- 21.4
is equipped with inter-bed quench Poly 11.1
facilities to control the overall tem-
SimDist (D2887)
perature levels. Hydrodewaxing
(HDW) is endothermic and the IBP, °F 305
dewaxing activity is controlled 50% 588
through the bed inlet temperature, FBP 798
using lower temperature to “turn off”
the dewaxing catalyst activity.
a very high temperature rise. These mine the expected hydrogen con-
constraints dictated that the HDW cat- sumption and activity, ART pilot test-
It is easiest to control the tempera-
alyst should be placed below at least ed three catalyst systems: 100%
ture to the first bed which makes it a
one of the hydrotreating catalyst beds. NiMo, 50%/50% NiMo/CoMo
convenient position to place the
SmART System® and 100% CoMo
HDW catalyst. However, there are
The key question became how to using Irving’s feed which is listed in
several problems with installing the
achieve the desired product specifi- Table XVII.
HDW catalyst in that location. Most
cations while minimizing hydrogen
dewaxing catalysts are sensitive to
consumption. NiMo catalysts tend to ART selected CDXi, a premium high
nitrogen compounds so they must
have higher activity for saturating aro- activity CoMo catalyst for ULSD and
be protected by a NiMo catalyst.
matics and removing nitrogen while AT505 which is a high activity con-
LCO contains a high concentration
CoMo catalysts tend to give lower ventional NiMo catalyst and prede-
of olefins, and that can deactivate
hydrogen consumption through less cessor to NDXi.
the HDW catalyst quickly due to
aromatics saturation.
olefin polymerization and related
Figure 44 compares the HDS activi-
coking. The required operating
Hydrotreating Catalyst System ty observed for each catalyst sys-
temperature window for the HDW
tem. Under these conditions the all
catalyst is not compatible with man-
ART’s SmART Catalyst System® Series NiMo system is clearly the most
agement of the overall temperature
offers custom system design to meet active for sulfur removal, followed by
profile, as the feed olefins and
individual refiner constraints and the SmART System® and finally the
‘easy’ sulfur react rapidly, producing
objectives. To help Irving Oil deter- CoMo catalyst, CDXi.
Figure 45 shows the hydrogen con-

Figure 44 sumption for each catalyst system.
Comparison of HDS Activity Not surprisingly, AT505 exhibited
the highest hydrogen consumption
while ART CDXi has the lowest
670 CDXi SmART Catalyst System® AT505 hydrogen consumption. The SmART
665 System® falls between the CoMo
and NiMo catalysts. Comparing
660
Figures 44 and 45 it is apparent that
Temperature, °F
655 the SmART System® Series provides

650 the best combination of activity and
645 hydrogen consumption; HDS activi-
ty was only slightly less than the all
640
NiMo AT505 system and the hydro-
635 gen consumption was significantly
630 lower.
0 10 20 30 40
Product Sulfur, ppm

Figure 45 Irving Oil also wanted to improve
Comparison of H2 Consumption the aromatics content and cetane
index of the ULSD product. These
product attributes are summarized
1500 CDXi SmART Catalyst System® AT505
in Figures 46 and 47. There is a
wide range in aromatics conversion
1400 between the NiMo and CoMo sys-
H2 Consumption, Scfb
tems (about 15 numbers absolute)

1300
which explains the large difference
1200 in hydrogen consumption shown in
Figure 45. The target aromatics
1100
level was <35 vol% which is easily
1000 achieved by AT505, but more of a
challenge for the all CoMo system.
900 The SmART System® resulted in
800 product aromatics which were
630 640 650 660 670 about three numbers (absolute)
Temperature, °F higher than the all NiMo system,
and the target was readily achieved
at a reasonable temperature.
Figure 46 The increase in cetane index tells a

Total Product Aromatics similar story. The cetane index for
the CoMo system is roughly two
numbers lower than achieved by the
50.0 all NiMo catalyst, and the SmART
Feed Aromatic Content is 50
45.0 System® resulted in a cetane
Aromatic Content, vol.%
improvement essentially equal to

40.0 the all NiMo catalyst.
35.0
The pilot data summarized in
30.0 Figures 44-47 clearly show that the
25.0 staged catalyst approach is much
more effective in meeting the objec-
20.0 tives set out by Irving Oil. The
CDXi SmART Catalyst System® AT505
15.0 SmART System® offers nearly the
630 640 650 660 670 same HDS activity compared to the
Temperature, °F NiMo catalyst, but with lower hydro-
gen consumption. It also provided
higher aromatics conversion relative
to the CoMo catalyst, easily meeting
the aromatics and cetane index tar-
Figure 47 gets. Using these data and per-
Comparison of Product Cetane Index forming additional modeling calcu-
lations allowed ART to determine
11.0 CDXi SmART Catalyst System® AT505
the optimum SmART System® pro-
portions which were ultimately used
10.0 in the commercial unit.
Cetane Index Increase
9.0 Feed Cetane Index is 35.5 Dewaxing Catalyst and its Impact
on the Process Design
8.0
7.0
With the CoMo/NiMo ratio resolved,
the design proceeded to the
6.0 amount and placement of the
HYDEX®-G HDW catalyst. This
5.0 required a detailed evaluation of the
620 630 640 650 660 670 overall system with respect to heat
Temperature, °F release, quench capabilities and
catalyst requirements.
Figure 48 ferences in deactivation
ULSD/Dewaxing Process Flow and rates).
• Selected combinations of the
Catalyst Loading Scheme Example above.
Recycle Hydrogen
Amine Unit
The results of this analysis revealed
H2S + NH3 + H2 that Bed 3 was the optimal location
Hydrogen for the HDW catalyst.
Gas Oil Blend

HDS
HDS
Figure 48 shows a process flow
scheme with a two-reactor system
LPG/Fuel Gas
and an example for the location of
the various catalyst layers and
Hydex-G Gasoline
Dewaxing
quench gas pipes to adjust for opti-
Catalyst mum bed temperatures.
The experiments that produced the

Make-up Hydrogen
ULSD data shown in Figure 49 were per-
Diesel Fuel
formed by an independent contrac-
tor on behalf of Irving Oil to under-
• Peak catalyst bed tempera- stand the HDW catalyst operating
HDW does consume some hydro- tures must be limited to pre- temperature required to meet the
gen and will also generate some serve catalyst activity target cloud point reduction. The
light ends which slightly reduce the experiments also quantified the light
diesel yield. The activity of the HDW The sensitivity analysis investigated ends/naphtha production. These
bed can be controlled with the inlet the impact of several parameters on data were used in the sensitivity
temperature so during the summer the ability to manage the overall tem- analysis, discussed above. The
a lower temperature turns down the perature profile (e.g., avoid overtreat- curve for the cloud point as a func-
degree of dewaxing. ing in HDS catalyst and/or undesired tion of the HDW bed temperature
dewaxing in the HDW catalyst). The also provides a means of predicting
As mentioned previously, most HDW parameters investigated were: the extent of dewaxing that will
catalysts are very sensitive to nitro- occur during summer mode opera-
gen poisoning and must be placed • Reduced heat release in the tion. The ability to effectively “turn
downstream of NiMo hydrotreating uppermost catalyst beds off” the HDW catalyst depends
catalyst to protect the activity of the (e.g., due to catalyst deactiva- upon the practical constraints and
catalyst. Süd-Chemie’s HYDEX®-G tion) other parameters mentioned previ-
is very nitrogen tolerant. In conjunc- • Variations in the required ously.
tion with its high activity, this gives operating temperatures for
much more flexibility in terms of its both the HDS and HDW cata-
placement in the reactor. The lysts (e.g., to account for dif-
HYDEX®-G can be placed in an
optimal position in the reactor
Figure 49
where the HDW bed inlet tempera- Pilot Plant Simulation to Confirm
ture is more readily controlled with the Dewaxing Design Temperature
inter-bed quench.
+25
The position of the HYDEX®-G in the
catalyst load was dictated by prac- +20
for Dewax WABT
tical constraints and a sensitivity

Cloud Point (ºF) vs Target
+15
analysis. Practical constraints
+10
included:
+5
• The catalyst bed volumes in 0 Target Range for Cloud Point
Design Range
the existing reactors are fixed

-5
• The total available quench Reactor Effluent
gas is fixed -10 Feed
• Individual quench gas rates -15
must be compatible with the Dewaxing Temperature
existing reactor internals
hardware
With the location of the dewaxing Table XVIII
catalyst set, a second pilot plant Second Test Feed
test was completed by ART using
the proposed SmART System®
including the required volume of Type 50% LCO
HYDEX®-G catalyst. The properties AP I 20.96
of the second feed provided by
Sulfur, wt.% 1.22
Irving Oil for the test are summa-
rized in Table XVIII. Nitrogen, wppm 499
Aromatics, vol.%
Figure 50 shows the sulfur conver- Total 53.5
sion achieved by the SmART
2 ring+ 36.2
System® compared to the SmART
System® + HYDEX®-G observed in Distillation, D2887
the test. The product sulfur is the IBP, °F 314
same for both catalyst systems indi- 50% 607
cating that sulfur conversion was
FBP 763
not affected by the addition of the
HDW catalyst. This was expected
since HYDEX®-G contains only
small amounts of base metals, and Figure 50
the metals function on this catalyst
Product Sulfur Comparison
is to help keep the catalyst clean
and prevent deactivation by exces-
sive coking.
10000
While HDS conversion was demon- SmART System®
SmART System® + HYDEX®-G
strated in the first test, the second 1000
test focused on proving the viability
Sulfur (ppmw)
of the entire design and the dewax-

ing performance. The cloud point 100
was of specific concern. Figure 51
summarizes the Cloud Point 10
Improvement (CPI) as a function of
the operating temperature for the
SmART System® + HYDEX®-G com- 1
pared to the SmART System® alone. Operating Temperature,
650 ºF
The CPI for the HYDEX®-G system
shows a characteristic threshold
temperature. Below the threshold
the CPI is low, while above this initi- Figure 51
ation temperature the CPI increases Cloud Point Improvement
substantially. At the temperature
required to achieve <10 wppm sul-
fur the cloud point improvement is
20 60
low, which is desired for “summer
mode.” Increasing the temperature 0 50
Cloud Point Change, ºF
Product Sulfur, ppm
will give the desired 30ºF improve-

ment for winter ULSD. -20 40
-40 30
The quench capability in the com-
mercial reactors is used to control -60 20
the bed temperatures to avoid over- SmART System®
SmART System®
treating and achieve the ultimate -80 + HYDEX®-G 10
Sulfur ppmw
goal of 7 ppm sulfur with control of
-100 0
the degree of dewaxing. The pilot Operating Temperature, ˚F
plant data indicates that increasing
the temperature by 40°F above the
threshold temperature increases the
CPI with HYDEX®-G to 90°F, much Figure 52
more than necessary. At this high Commercial Unit Normalized Reactor Temperature
level of CPI the yields of naphtha
and light ends also increase which
can have an adverse effect on the 700
economics of the operation. 690
However, this reserve of activity 680
ensures the system is in balance
670
with regards to stability of the HDS
Temperature, °F
and HDW functions guaranteeing 660
that all product specifications are 650

met over the desired three year 640
cycle life. 630
620
Commercial Unit Performance
610
The commercial unit started up in 600

0 100 200 300 400 500 600
the Spring of 2006, and has been
Days on Stream
producing ULSD since the first day
on stream. Figure 52 shows the
normalized temperature for the
operation thus far. It is apparent
Figure 53
from the figure that the operation is
quite stable especially considering Commercial Dewaxing Activity
the amount of LCO included in the
feedstock. The unit came on stream 40
with higher than expected activity, 35

Cloud Point Improvement, °F
and the dewaxing activity has been 30

exceptional.
25
All indications are that Irving Oil will 20
be able to extend the cycle beyond 15

the original 36 month cycle life esti-
10
mate.
5
Figure 53 summarizes the dewaxing 0

activity of the system. The feed HDW Operating Tempeature, °F
varies quite a bit depending on Crude 1 Crude 2

crude source and the amount of
LCO. These data have been sorted
to demonstrate the performance on
two different crude types at approx-
imately equal LCO levels processed
Figure 54
through the unit. As was observed API Upgrade and Product Sulfur
in the pilot plant testing discussed
above, the cloud point improvement 50 Crude 1 Sulfur Crude 2 Sulfur Crude 1 API Crude 2 API 14
is relatively low for low operating 45
12
temperatures, and increases signifi- 40
API Improvement, °API
Product Sulfur, wppm
cantly as temperature is increased. 35 10

Irving Oil has not needed to achieve 30
8
the original CPI target of 30°F on a 25
consistent basis, but as can be 20
6
seen in the figure, the unit has 15 4

achieved CPI’s of 35°F at reason- 10
2
able operating temperatures. 5
0 0
The API increase has varied Operating Temperature, °F
between 4-8 numbers depending
on the crude source and the

amount of LCO being charged to perature at a level for the catalysts to activity, especially when it comes to
the unit. This is shown in Figure 54 work as effectively and reliably as processing heavier and difficult
along with the product sulfur. The expected; the unit produced 7 ppm, feeds with high nitrogen content as
product sulfur has consistently low cloud point diesel from start of in the Irving Oil case.
been around 5 ppm with an average run, thus creating a new benchmark in
API uplift of 6+ numbers. The the industry for low cloud point ULSD Considering the complex require-
cetane uplift mirrors the API technology. ments for Irving Oil’s ULSD unit, it is
change, and thus, the unit is seeing obvious that only an integrated
a large increase in cetane com- Optimal catalyst selection as well as approach to catalysts and unit
pared to most ULSD units. precision design work is a prerequi- design, as demonstrated by ART’s
site when it comes to meeting ultimate and Süd-Chemie’s collaboration
Conclusions specifications and catalyst life time with Fluor, leads to a successful
goals. The catalyst ensemble in the commercial operation. The technol-
Based on the excellent performance Irving Oil ULSD unit comprises ART’s ogy as implemented in Irving Oil’s
of the Irving Oil unit, it can be con- high activity SmART Catalyst System® ULSD unit offers tremendous poten-
cluded that the design work by ART, Series in combination with Süd- tial to improve product quality and
Süd-Chemie, and Fluor described in Chemie’s HYDEX®-G dewaxing cata- economics in many new and
this paper was extremely success- lyst. The stacked-bed configuration of revamped ULSD units.
ful. It prevailed over competitive high performance CoMo and NiMo
offers in Irving Oil’s extended evalu- catalysts provides the high HDS activ-
ation program and the commercial ity required for ULSD while minimizing
operation has exceeded expecta- hydrogen consumption. The HYDEX®-
tions. Not only is the dewaxing tem- G catalyst takes advantage of its high
Please welcome...
Shawn Abrams has joined Grace Davison as Vice President and General
Manager of Refining Technologies. Shawn will have direct responsibility for
the Refining Technologies Global Organization. Shawn joins us with over 20
years of management experience in global industrial markets at Evonik
Industries AG (formerly Degussa); where he held a variety of progressive
leadership positions including Director of Sales in Frankfurt, Germany, and
Vice President and General Manager of Asia Pacific for their Bleaching and
Water Chemicals Business Unit. For the past five years, Shawn has served
as a member of the Degussa Executive Group as Senior Vice President and General
Manager of their Active Oxygens Business Unit.
Shawn graduated from Lehigh University with a B.S. in Mechanical Engineering, and com-
pleted his MBA at the Thunderbird School of Global Management. Reporting to Shawn will
be
Joanne Deady, Vice President Marketing/R&D,

Frank Cunningham, General Manager, North America Sales and Service
Al Jordan, Director, Sales Operations
Ruben Cruz, General Manager, Latin America
Jim Nee, General Manager, Asia Pacific
Bruno Tombolesi, General Manager, EMEA
Please join us in welcoming Shawn to Davison.
Grace Davison
Multi-Loader System
Al Jordan Summary refiner to optimize the unit operation

Director, Sales Operations while meeting product and environ-
Adam Kasle
It has long been understood that opti- mental regulations.
mum catalyst performance and unit
stability in an FCC unit are greatly Additive System Justification
Technical Sales Manager assisted by continuous and steady
Grace Davison Refining injection of FCC catalyst and additives The continuous goal of every FCC
Technologies throughout the course of the day. operator is to optimize unit perform-
However, this goal has been difficult ance and maximize profitability. As
for many refiners to achieve due to physical unit constraints are pushed
inadequate or unreliable addition sys- with mechanical solutions and cata-
tems that have not been up to the task. lyst formulations are tuned, refiners
The challenge has been further com- are looking more at unit control to
plicated in recent years as refiners derive further value from their
must now operate within multiple envi- assets. One such area of control is
ronmental regulations and depend on the maintenance of a continuously
stable operation and product addition optimized catalyst condition in the
rates to do so. One solution that has unit through stable, ongoing addi-
been rapidly gaining acceptance in tion of fresh catalyst. This allows the
the industry is the Grace Davison unit to provide the optimum yield
Multi-Loading System. This four-in- slate throughout the day without the
one catalyst addition system provides degradation associated with bulk
an accurate and reliable method for catalyst loading. Proper cracking
adding both cracking catalyst and catalyst management has begun to
additives to the FCC unit allowing the

Table XIX regenerator emissions, regenerator
Current Multi-Loader Users List temperature excursions, catalyst
circulation changes, required
reduction of feed rate, and/or con-
Refinery Location Number of Installation version loss attributed to loss of cat-
(USA) Products Date alyst activity.
Injecting
1 Mid West 3 Apr -04 To solve this problem for the refiner,
2 Mid West 3 Dec -04 Grace Davison has developed a
Multi-Loader System that allows the
3 Gulf Coast 3 Mar -05
continuous addition of up to four
4 Southwest 4 Mar -05 materials independently in a coordi-
5 East Coast 2 Jun -05 nated manner. This would include
6 Rocky Mountain 2 Aug -05 any combination of fresh catalyst,
7 East Coast 4 Aug -05 equilibrium catalyst, and/or addi-
8 Rocky Mountain 2 Aug -05 tives. For control purposes, this sys-
9 Gulf Coast 1 Nov -05
tem includes weight cells to track
and record the weight of each injec-
10 Rocky Mountain 3 Dec -05
tion, and logic programming that
11 Gulf Coast 2 Apr -06 automatically adjusts to meet the
12 Gulf Coast 3 Aug -06 target daily addition rates of each
13 Gulf Coast 1 Oct -06 catalyst/additive.
14 Latin America 1 Jan -07
15 Rocky Mountain 3 Mar -07 The benefits of a Multi-Loader sys-
16 Rocky Mountain 1 tem can be summarized as follows:
Apr -07
17 Southwest 2 Apr -07
• Consistent control of
18 Gulf Coast 3 Jun -07 catalyst/additive injections
19 Gulf Coast 1 Jun -07 • Reliability with low mainte-
20 Gulf Coast 1 Jun -07 nance requirements
21 West Coast 1 Aug -07 • Robust range of injection
22 Gulf Coast 2 Oct -07 rates
23 Gulf Coast 2 • Accurate injection weight and
Dec -07
record keeping ability
24 Mid West 2 Jan -08
• Advanced Touch Screen con-
25 Mid West 2 Jan -08 trol for user-friendly operation
26 Mid West 2 Jan -08 • Skid mounted self-contained
play an even more critical role in the to meet gasoline sulfur regulations design for simple installation
profit optimization equation, as the while preserving octane, and a NOx or • Fugitive dust containment (no
dynamics in crude oil prices have SOx reducing additive to meet emis- environmental impact)
resulted in ever increasing changes sions targets, all at the same time. • Injection of multiple materials
in FCC feed quality and contami- independently
nant levels. Grace Multi-Loader • Flexibility to add material from
fresh catalyst hopper, tote-
In addition to yield slate optimiza- Historically, the refiner has injected the bins, drums, or catalyst
tion, there is also the added pres- catalyst and additives with separate trucks.
sure of meeting the ever changing loading systems that share a common • Output signal to Distributed
environmental (NOx, SOx, CO) and injection line to the regenerator. Control System (DCS)
product quality (Gasoline Sulfur) However, multiple loaders trying to • Feedback control capability
regulations. In order to meet the inject material in an unsychronized (DCS compatible)
new regulations at maximum profit fashion can result in simultaneous • Reduced plot space required
with minimum capital investment, loading of multiple products. This can vs. multiple loading systems
many refiners are using several create backpressure on one of the
types of additives in the FCC Unit systems or plugging in the common
as supplements to the FCC crack- loading line, each resulting in erratic
ing catalyst. It is not unusual for a additions of all materials in use. As
refiner to be using a combustion additions become erratic, there is the
promoter, a sulfur reducing additive likelihood of unit upsets such as
Commercial Experience The decision was made to install a Multi-Loader allowed the refiner to
Case Study #1 Gulf Coast Refiner

Grace Davison Multi-Loader. The increase daily catalyst additions in a
results were dramatic. They were able controlled manner resulting in high-
to set all product injections and con- er in-unit catalyst activity and
The refiner suffered from daily trol them independently with all injec- reduced contaminant effects, trans-
regenerator temperature excursions tion amounts transmitted to the refin- lating to higher profitability. Since
resulting from once per day batch ery data collection system. In addi- the installation and start up of the
catalyst additions. These tempera- tion, the ease of use and system relia- Grace Davison Multi-Loader
ture excursions forced a reduction bility gained the loader a very high System, the regenerator tempera-
in feedrate of 5% for several hours degree of operator acceptance. ture surges have been greatly
Case Study #3 East Coast Refiner

each day resulting in lost revenue reduced, allowing for improved unit
on the order of $3,000/day. Based operating stability.
on these economics, the continuous
Grace Davison Mulit-Loader was A refiner with an antiquated catalyst This loading system has greatly
easily justified and installed. addition system was having problems reduced the effort required on the
Case Study #2 Gulf Coast Refiner

adding sufficient catalyst to maintain part of the operations and mainte-
in-unit activity and control contami- nance staffs to maintain sufficient
nant levels. In addition, the unreliable catalyst additions while improving
A refiner needed to improve his nature of the catalyst additions result- multiple aspects of the unit opera-
fresh catalyst injection system and ed in wide temperature swings in the tion.
add a SOx additive as well as a regenerator throughout the day. The
ZSM-5 type additive. installation of the Grace Davison
Catalyst or additive can be loaded from a storage hopper or a

shipping vessel (55-gallon drum or tote bin) directly to Multi-Loader
Loading directly from shipment vessel allows refiners to disconnect

one product WITHOUT stranding quantities in the loader
Fresh
Additive Cat
Tote-Bin Hopper
55 gal.
Drum

One Ton Unit
Summary More than Just a Loader itability. As refiners move to meet

stringent product and environmen-
Grace Davison Multi-Loader • Every multi-loader is tested tal regulations, addition systems will
prior to shipment in our test become an essential part of the unit
• Unique loader capable of facility control. Grace Davison’s Multi-
injecting and controlling addi- • Pre-installation training avail- Loader System provides an effec-
tions of up to four different able at our test facility tive and reliable means of adding
FCC materials • Extensive on-site start-up FCC fresh catalyst and additives on
• Pre-assembled, modular con- assistance and operating/ a continuous basis, whether you are
struction for simplified installa- maintenance training adding one or several materials to
tion • On-line loader operations the FCC. Grace has successfully
• Can inject from 1.0 lb to ≥ 24 manual installed Multi-Loaders in 26 units in
tons per day • Interactive web-based trou- the last four years with even more
• Operator friendly touchscreen bleshooting guide currently in the planning stages.
controls with help menus • Critical spare parts inventory Please contact one of our team
• Output signals for each addi- maintained members for further information on
tive/catalyst this unique and proven system.
• Moving parts housed from Conclusion
weather for industry-leading
reliability Stable catalyst and additives addi-
tions to the FCC unit are an important
part of achieving maximum FCC prof-
SAVETHEDATE
Grace Davison/Advanced Refining Technologies

Technical Seminar
Thursday, August 21, 2008
Intercontinental Hotel
Houston, Texas
(immediately following the NPRA Cat Cracker Conference)
Save the date to hear the latest on FCC and

Hydroprocessing Catalysts technologies.
Hear industry experts discuss trends on
catalysts and processes.
For more information or to be put on the mailing list, contact betsy.mettee@grace.com

Davison Refining Technologies Davison Refining Technologies Davison Refining Technologies
Advanced Refining Technologies Advanced Refining Technologies Advanced Refining Technologies
W.R. Grace & Co.-Conn. Asia Pacific Europe
7500 Grace Drive W.R. Grace (Singapore) Pte. Ltd. Grace GmbH & Co. K.G.
Columbia, MD 21044 #07-02 Wheelock Place In der Hollerhecke 1
410.531.4000 Singapore 238880 Postfach 1445
65.6737.5488 D-67545 Worms, Germany
49.6241.40300
catalysts@grace.com artinfo@grace.com www.e-catalysts.com

103

Uploaded by

Copyright:

Available Formats

103

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

103

Uploaded by

Copyright:

Available Formats

What are GENESIS catalysts and their components?

What are GENESIS catalysts and their components?

What are the benefits of GENESIS catalysts?

What are the benefits of GENESIS catalysts?

NUM B E R 1 0 3 S pr i n g 200 8

The GENESISTM Catalyst System

Maximize yield with

GENESISTM catalysts demonstrate a true yield synergy with a superior

GENESISTM catalyst provides

GENESISTM Catalyst Commercial Update

W. R. Grace & Co.-Conn. • Advanced Refining Technologies

Brian Watkins he use of renewable or bio-

Catalagram 103 Spring 2008 1

C20:0 Mono- C20:1 C22:1

95 to 110, which can provide a sig- 2.200e+006

nificant boost for those refiners 2.000e+006

cal diesel hydrotreater has only a 1.600e+006

small effect on cetane with cetane

as they have molecular weights of 6.000e+005

of less than 400. The simulated GC

Catalagram 103 Spring 2008 3

stream high activity catalysts from

This quantity of oxygen is important,

Looking at these compounds from a Figure 4

Catalagram 103 Spring 2008 5

<10 ppm product sulfur.

590 Looking at other diesel product

63 The impact of low pressure opera-

59 points. It is this increase that yields

important to also be aware of any 1.000e+006

changes in product aromatics, as 9.000e+005

reactions to saturate aromatics are

high consumers of hydrogen and 6.000e+005

would compete with the saturation 1.000e+006

reactions under hydrogen limited 4.000e+005

conditions. Figure 10 summarizes 3.000e+005

the product aromatics for one of the 2.000e+005

bers lower than the SR feed, which Retention time (min)

At lower pressure and H2/oil ratios, 25

the total aromatic content shows a

lower total aromatics when co-pro-

icantly higher cloud point than other 28

SR 10% Soy 10% palm 10% rapeseed

Catalagram 103 Spring 2008 7

delta Conversion, wt.%

Another option for refiners could be

77.0 Since the increase in the concentra-

Catalagram 103 Spring 2008 9

From our analysis, the addition of

Alternatively, the potential for the

Total Dry Gas

more paraffinic fatty acid mixture

0.15 The initiation step occurs on either

The gasoline trend indicates over-

commercial processing of the 1.1

blended bio-containing feeds. The 1.0

is important, as the oxygen species

down to water, or they could poten-

were initial peaks between two and 49.5

product thus, no undesirable oxy- 48.0