The Influence of Alumina on the Performance of FCC Catalysts ...

62 Energy & Fuels 2003, 17, 62-68
<strong>The</strong> <strong>Influence</strong> <strong>of</strong> <strong>Alumina</strong> on the Performance <strong>of</strong> FCC
Catalysts during Hydrotreated VGO Catalytic Cracking
S. Al-Khattaf †
Chemical Engineering Department, King Fahd University <strong>of</strong> Petroleum & Minerals,
Dhahran, 31261 Saudi Arabia
Received March 13, 2002
Two categories <strong>of</strong> FCC catalysts were prepared. <strong>The</strong> first one is an FCC catalyst containing
Y-zeolite and amorphous matrix. <strong>The</strong> second category is amorphous catalyst without Y-zeolite.
<strong>Alumina</strong> was systematically added to both catalyst categories. <strong>The</strong> addition <strong>of</strong> alumina to the
FCC catalyst containing Y-zeolite forms an active matrix. <strong>The</strong> inactive matrix <strong>of</strong> the catalyst is
composed <strong>of</strong> kaolin filler and silica sol binder. <strong>The</strong> experimental data clearly show that alumina
addition to both catalyst categories substantially increased both catalyst surface area and acidity.
<strong>The</strong> addition <strong>of</strong> alumina to the amorphous catalyst resulted in a significant increase in
hydrotreated VGO conversion. However, similar alumina addition to Y-zeolite containing
catalysts did not appear to have a significant effect on the hydrotreated VGO conversion.
1. Introduction
Commercial FCC catalysts are manufactured using
1-2-µm zeolites dispersed on an amorphous matrix
forming 60-µm particles. 1,2 Zeolites are extremely active
and provide catalyst activity, selectivity, and stability. 3
<strong>The</strong>ir regular pore dimensions make them selective as
to which molecules are adsorbed or converted. Hence,
Y-zeolite is the active and the most important component
in FCC catalysts. Y-zeolite provides the major
part <strong>of</strong> the surface area and the active sites. 1,2 Thus it
is the key component, which controls catalyst activity
and selectivity. 4-9 <strong>The</strong> catalytic activity <strong>of</strong> Y-zeolite is
mainly controlled by its unit cell size (UCS) and to a
lesser extent by its crystal size. Recently, Al-Khattaf and
de Lasa have studied the effect <strong>of</strong> Y-zeolite crystal size
on the activity and selectivity <strong>of</strong> FCC catalysts. 10-12
Most commercial FCC catalysts contain between 15 and
40 wt % zeolite. 2
<strong>The</strong> activity and selectivity <strong>of</strong> FCC catalysts are
mainly controlled via three processing strategies: 13
† E-mail: skhattaf@kfupm.edu.sa. Phone: 9663-8601429. Fax: 9663-
8604234.
(1) Scherzer, J. In Fluid Catalytic Cracking: Science and Technology;
Magee, J. S., Mitchell, M. M., Eds.; Elsevier: Amsterdam, 1993.
(2) Humphries, D.; Harris, D.; O’Connor, P. In Fluid Catalytic
Cracking: Science and Technology; Magee, J. S., Mitchell, M. M., Eds.;
Elsevier: Amsterdam, 1993; pp 41.
(3) Humphries, A.; Wilcox, J. R. Oil Gas J. Feb 6, 1989.
(4) Pine, L. A.; Maher, F. K.; Wachter, W. A. J. Catal. 1984, 85,
466.
(5) Arribas, J.; Corma, A.; Fornes, V.; Melo, F. J. Catal. 1987, 108,
135.
(6) Ino, T.; Al-Khattaf, S. Appl. Catal. 1996, 142, 5.
(7) Al-Khattaf, S. Appl. Catal. 2002, 231, 293.
(8) Rajagopalan, K.; Peters, A. W. J. Catal. 1987, 106, 410.
(9) Ritter, R. E.;Creighton, J. E.; Roberies, T. G.; Chin, D. S.; Wear,
C. C. NPRA Annual Meeting, Los Angeles, March 1986.
(10) Al-Khattaf, S.; de Lasa, H. I. Ind. Eng. Chem. Res. 1999, 38,
1350.
(11) Al-Khattaf, S.; de Lasa, H. I. Can. J. Chem Eng. 2001, 3, 79,
341.
(12) Al-Khattaf, S.; de Lasa, H. I. Appl. Catal. 2002, 226, 139.
zeolite characteristics such as silica/alumina ratio which
is directly related to unit cell size, the presence <strong>of</strong>
additives such as ZSM-5, and the type <strong>of</strong> matrix in the
the FCC catalysts as the active matrix can contribute
to the cracking reactions.
<strong>The</strong> matrix comprises 60-85 wt % <strong>of</strong> the commercial
FCC catalyst and usually contains synthetic and natural
components. Clay is the natural component and amorphous
silica or silica-alumina is the synthetic portion. 14
Matrix composition can influence catalyst performance
but to a lesser extent than the zeolite component. 14 An
inactive matrix does not contain enough strong acid
sites that can influence the catalyst performance. On
the other hand, an active matrix contains acid sites
associated with aluminum atoms (e.g., alumina). However,
only those sites that have sufficient strength are
able to contribute to the cracking reactions. 15 Among
the different alumina available commercially, boehmite
is the most important one that is used in catalyst
formulations. <strong>The</strong> high acidity <strong>of</strong> boehmite is the major
reason behind its catalytic activity. Its addition to the
catalyst can improve the catalyst performance, improves
attrition resistance and serves as a metal trap. 1
Active matrix is believed to improve bottoms upgrading
and LCO quality by increasing the cetane number.
1,15 <strong>The</strong> large pore <strong>of</strong> the matrix >500 Å facilitates
large heavy oil molecule (10-100 Å) diffusion and may
cause cracking if the matrix contains acid sites with
certain strength. <strong>The</strong> active matrix is characterized by
high surface area and acidity. 1,15
Usually active matrix enhances bottom conversion
and gasoline octane. 1,15-18 However, it also produces
(13) Magee, J. S.; Letzsch, W. S. In Fluid Catalytic Cracking III;
Materials and Processes; Occelli, M. L., O’Connor, P., Eds.; American
Chemical Society: Washington, DC, 1994.
(14) Scherzer, J. In Fluid Catalytic Cracking III; Materials and
Processes; Occelli, M. L., and O’Connor, P., Eds.; American Chemical
Society: Washington, DC, 1994.
(15) Scherzer, J. Catal. Rev.-Sci. Eng. 1989, 31, 3, 215.
10.1021/ef020066a CCC: $25.00 © 2003 American Chemical Society
Published on Web 12/05/2002

Performance <strong>of</strong> FCC Catalysts Energy & Fuels, Vol. 17, No. 1, 2003 63
more coke, dry gas, and less gasoline than inactive
matrix. 16-18 <strong>The</strong> effect <strong>of</strong> alumina addition to FCC
catalyst matrix on processing different kinds <strong>of</strong> VGO
feedstock was studied by Otterstedt et al. 17 <strong>The</strong>y
observed that alumina addition to the catalyst matrix
had increased the conversion <strong>of</strong> the heavier feed oil but
at the same time increased coke selectivity and decreased
gasoline selectivity. However, alumina addition
did not have a significant effect during lighter feed oil
processing. Coke and gasoline selectivities were found
to have similar behavior to those <strong>of</strong> heavier feed oil.
It has been reported that the olefin/paraffin ratio
decreases by increasing zeolitic content and decreasing
matrix activity. 19 Moreover, Scherzer 15 claimed that
increasing matrix activity (by adding alumina) would
increase both olefin/paraffin ratio and gasoline octane.
This was attributed to lower hydrogen transfer reaction
rate versus cracking reaction rate over amorphous
matrix component.
Thus, it is clearly known that alumina addition to a
FCC catalyst matrix can influence the catalyst performance
by increasing catalyst acidity and surface area.
Consequently, the VGO catalytic cracking might be
enhanced by alumina addition to the matrix without
changing the USY-zeolite structure.
In the present study, a thorough investigation for the
role <strong>of</strong> active matrix on catalytic cracking <strong>of</strong> hydrotreated
VGO has been conducted. Systematic addition
<strong>of</strong> alumina to both matrix samples and FCC catalysts
containing USY-zeolite has been studied. Testing both
catalyst systems under standard MAT conditions has
been carried out. <strong>The</strong> activity <strong>of</strong> matrix with alumina
and catalyst containing alumina is determined. Furthermore,
the effect <strong>of</strong> alumina addition on product
selectivity is highlighted.
2. Experimental Section
2.1. Catalyst Preparation. USY-zeolite was mixed with
kaolin clay and stabilized silica sol (SI-550) supplied by
Catalyst and Chemicals Industries Co. <strong>The</strong> resulting slurry
was heated over a sand bath with continuous stirring. <strong>The</strong>
dried product was crushed and sieved to the proper particle
size (520-710 µm). <strong>The</strong> base catalyst, the one with inactive
matrix (CAT0), contained 30% zeolite, 50% kaolin, and 20%
silica sol binder. <strong>The</strong> prepared catalysts were calcined in air
at 600 °C for 2 h. Finally the catalysts were treated with 100%
steam at 810 °C for 6 h.
<strong>The</strong> series <strong>of</strong> catalysts containing the active matrices were
prepared in the same manner as the one containing inactive
matrix with the exception that part <strong>of</strong> the kaolin was replaced
by boehmite. <strong>The</strong> amount <strong>of</strong> boehmite used in these preparations
corresponded to 5 wt % (CAT5), 10 wt % (CAT10), and
20 wt % (CAT20). <strong>The</strong> three catalysts CAT5, CAT10 and
CAT20 all contained 30 wt % USY-zeolite and 20 wt % silica
sol. <strong>The</strong> other component is the kaolin as shown Table 1.
<strong>The</strong> series <strong>of</strong> matrix samples are prepared as follows;
inactive matrix contains only kaolin (M0). Active matrices were
prepared in the same manner as the FCC catalysts with active
matrix except that in these series no zeolite is present. Part
(16) Van de Gander, P.; Penslay, R. M.; Chuang, K. C.; Cormier,
W. E.; Walterman, G. M.; Boldingh, E. P. Advanced Fluid Catalytic
Cracking Technology; AIChE Symposium Series vol. 88, no. 291;
AIChE: New York; pp 21-29.
(17) Otterstedt, J. E.; Zhu, Y.; Sterte, J. Appl. Catal. 1988, 38, 143.
(18) Sterte, J.; Otterstedt, J. E. Appl. Catal. 1988, 38, 131.
(19) de Jong, J. I. Ketjen Catalysis. Symposium 1986, Scheveningen,
<strong>The</strong> Netherlands, Paper F-2.
catalyst
Table 1. Catalyst Composition Content
Y-zeolite
(wt %)
silica sol
(wt %)
kaolin
(wt %)
alumina
(wt %)
CAT0 30 20 50 0
CAT5 30 20 45 5
CAT10 30 20 40 10
CAT20 30 20 30 20
M0 0 0 100 0
M5 0 20 75 5
M10 0 20 70 10
M20 0 20 60 20
Table 2. Treatment and Characterization <strong>of</strong> Catalysts
catalyst treatment
unit cell
size (Å)
surface
area
(m 2 /g)
Al
(wt %)
acidity
(mmol/g)
CAT0 steaming, 810 °C 24.206 123 0 0.0190
CAT5 steaming, 810 °C 24.206 177 5 0.0296
CAT10 steaming, 810 °C 24.206 180 10 0.0311
CAT20 steaming, 810 °C 24.206 196 20 0.0377
M0 steaming, 810 °C - 22 0 0.006
M5 steaming, 810°C - 25 5 0.0088
M10 steaming, 810 °C - 31.7 10 0.0114
M20 steaming, 810 °C - 60 20 0.0227
<strong>of</strong> kaolin was substituted by boehmite. <strong>The</strong> amount <strong>of</strong> boehmite
used in these preparations corresponded to 5 wt % (M5), 10
wt % (M10), and 20 wt % (M20) see Table 1. All matrices were
calcined in air at 600 °C for 2 h. <strong>The</strong>n they were treated with
100% steam at 810 °C for 6 h (see Table 2).
2.2. Catalyst Characterization. <strong>The</strong> unit cell size was
determined by X-ray diffraction following ASTM D-3942-80.
High purity silicon powder (99.999%) was used as the calibration
standard. Surface area <strong>of</strong> the sample catalyst were
measured by nitrogen adsorption at -196 °C. <strong>The</strong> zeolite used
has 75.9 m 2 /g surface area (this surface area is for the zeolite
which is steamed at 810 °C for 8 h) and negligible Na content.
Moreover, the unit cell size is 24.206 Å. Hence its silica/
alumina ratio can be found from this unit cell size.
<strong>The</strong> acid property <strong>of</strong> the catalyst was characterized by
pyridine temperature programmed desorption (TPD). In all
experiments, 50 mg <strong>of</strong> sample was charged in a tubular cell.
Prior to obtaining TPD spectra, the sample was outgassed at
400 °C for 30 min in flowing helium and then cooled to 150
°C. At that temperature, pyridine was adsorbed on the sample
by injecting pulses <strong>of</strong> pyridine (2ul/ pulse). <strong>The</strong> injection <strong>of</strong>
pyridine was repeated until the amount <strong>of</strong> pyridine detected
was the same for the last two injections. After the adsorption
<strong>of</strong> pyridine was saturated, the sample was flushed at 150 °C
for 1 h with helium to remove excess pyridine and then
temperature programmed at 30 °C/min to 1000 °C in flowing
helium at 30 mL/min. An FID detector was used to monitor
the desorbed pyridine. <strong>The</strong> properties <strong>of</strong> the catalysts used in
the present study are presented in Table 2.
2.3. Catalyst Evaluation. Catalytic experiments were
carried out in a microactivity test (MAT) unit (fixed bed), which
had been designed according to the ASTM D-3907 method with
minor modifications. <strong>The</strong> reactor was operated at atmospheric
pressure and 520 °C. For a given catalyst, the conversion was
varied by varying the catalyst-to-oil ratio (C/O) in the range
<strong>of</strong> 0.5-3.0. <strong>The</strong> C/O ratio is defined as the amount <strong>of</strong> catalyst
divided by the total amount <strong>of</strong> oil fed in a given time on stream
and was varied by changing the weight <strong>of</strong> the catalyst while
the total amount <strong>of</strong> oil fed and the time were kept constant,
i.e., 1g<strong>of</strong>theoilwascharged to the reactor in 30 s (see Table
3). <strong>The</strong>rmal effects and changes in the bed volume were
minimized by diluting the catalyst with kaolin particles
(having the same size as the catalyst particle), and the total
weight <strong>of</strong> the catalyst bed was kept at 3 g. <strong>The</strong> distribution <strong>of</strong>
gaseous products was analyzed by gas chromatographies. <strong>The</strong>
boiling point range <strong>of</strong> the liquid products was determined by

64 Energy & Fuels, Vol. 17, No. 1, 2003 Al-Khattaf
Table 3. MAT Operating Conditions
temperature 520 °C
feed rate 1 g/30 s
amount <strong>of</strong> catalyst 0.5-3.0 g
feed type VGO
Table 4. MAT Feed Oil Properties
specific gravity (15/4 °C) 0.8821
sulfur (wt %) 0.18
Conradson carbon (wt %) 0.09
refractive index (15 °C) 1.4719
bromine number 3.2
Ni (ppm)

Performance <strong>of</strong> FCC Catalysts Energy & Fuels, Vol. 17, No. 1, 2003 65
Table 5. MAT Products Yields for VGO Cracking at C/O ) 3 Using Matrices (M0, M5, M10, and M20)
compound name M0 yield (wt %) M5 yield (wt %) M10 yield (wt %) M20 yield (wt %)
H2 0.11 0.08 0.09 0.11
C1 0.5 0.35 0.39 0.31
C2 0.47 0.3 0.32 0.26
C2 ) 0.39 0.29 0.32 0.28
dry gas (H2+C1+C2+C2 ) ) 1.47 1.02 1.13 0.97
C3 0.31 0.22 0.25 0.25
C3 ) 1.3 1.07 1.5 2.47
i-C4 0.25 0.23 0.4 0.79
n-C4 0.12 0.08 0.12 0.17
t2-C4 ) 0.46 0.36 0.55 1.02
1-C4 ) 0.38 0.31 0.45 0.81
i-C4 ) 0.8 0.65 0.92 1.63
c2-C4 ) 0.34 0.27 0.41 0.76
total C4 ) 1.98 1.6 2.32 4.23
total C4 (olefins + paraffins) 2.35 1.91 2.84 5.19
LPG (total C3 + total C4) 3.96 3.2 4.59 7.9
gasoline 14.13 16.65 20.63 28.12
LCO 20.28 19.88 19.02 21.93
HCO 58.66 58.06 52.83 40
coke 1.51 1.19 1.35 1.14
total 100 100 99.55 100
conversion ) 100 - (LCO wt % + HCO wt %) 21.1 22.1 28.1 38.1
Figure 1. Second-order conversion versus C/O ratio for matrix
samples. O 5 wt % alumina (M5), 0 10 wt % alumina (M10),
and 4 20 wt % alumina (M20).
active matrix) show higher conversion at all C/O ratios
in good agreement with literature. 17 However, this
increase in conversion is not significant. For example,
the base catalyst (CAT0) had 70.1% conversion while
CAT20 has 74.5% (both at C/O ) 3). Furthermore, it
was noticed that CAT5 and CAT10 have similar conversion
at the same reaction conditions. <strong>The</strong>refore, a
modest increase in hydrotreated VGO conversion was
obtained by adding alumina to the matrix <strong>of</strong> FCC
catalysts.
<strong>Alumina</strong> addition to FCC catalysts is supposed to
facilitate molecular diffusion to the zeolite pores. An
active matrix is believed to have the capability to
precrack VGO molecules before their transport to the
zeolite pores. Thus matrix precracking can increase the
conversion efficiency since zeolite pores cannot accommodate
many <strong>of</strong> the VGO molecules. However, as
mentioned earlier, only a slight increase in VGO conversion
was seen by adding alumina. Increasing alumina
content up to 20 wt % increased the conversion
only from 70% to 74% (at C/O ) 3). Consequently, the
Figure 2. Second-order conversion versus C/O for catalysts
9 0 wt % alumina (CAT0), 0 5 wt % alumina (CAT5), 4 10 wt
% alumina (CAT10), and × 20 wt % alumina (CAT20).
addition <strong>of</strong> alumina to FCC catalysts did not significantly
alter the transport behavior <strong>of</strong> the hydrotreated
VGO at the present reaction conditions.
Figure 2 shows the C/O ratio versus the second-order
conversion for the different catalysts with different
alumina content. <strong>The</strong> slope <strong>of</strong> such plot represents FCC
catalyst activity. 6,7 As the slope increases, catalyst
activity increases too. For example, the activity <strong>of</strong> the
base catalyst (CAT0) was 0.74, comparing to 0.98 the
activity for catalysts containing alumina. <strong>The</strong> catalyst
activity for CAT5, CAT10, and CAT20 are all in the
same range around 0.98. Thus alumina addition to FCC
catalysts increases catalyst activity.
<strong>The</strong> effect <strong>of</strong> alumina addition to the FCC catalyst
matrix on gasoline selectivity as function <strong>of</strong> C/O (or VGO
conversion) is shown in Figure 3. It is shown that
gasoline selectivity decreases with the C/O ratio for all
catalysts. This behavior is expected due to high overcracking
reaction rate at high C/O ratio. It was also
noticed that alumina addition to FCC catalyst matrix
has resulted in decreasing gasoline selectivity. Gasoline

66 Energy & Fuels, Vol. 17, No. 1, 2003 Al-Khattaf
Figure 3. <strong>The</strong> effect <strong>of</strong> alumina addition on gasoline selectivity.
9 0 wt % alumina (CAT0). 0 5 wt % alumina (CAT5), 4
10 wt % alumina (CAT10), and × 20 wt % alumina (CAT20).
Figure 4. <strong>The</strong> effect <strong>of</strong> alumina addition on LCO selectivity.
9 0 wt % alumina (CAT0). 0 5 wt % alumina (CAT5), 4 10 wt
% alumina (CAT10), and × 20 wt % alumina (CAT20).
selectivity for CAT0 was found to be around 73%, while
for CAT20, 69% gasoline selectivity was obtained.
Gasoline selectivites for CAT5 CAT10 were observed to
be bounded between selectivities <strong>of</strong> CAT0 and CAT20
as shown in Figure 3. Similar effects for alumina on
gasoline selectivity were reported by several other
researchers. 15,17
LCO selectivity was found to decrease with C/O ratio
(or VGO conversion) similar to gasoline trend. This
behavior is due to the increase in LCO cracking rate as
conversion increases (see Figure 4). It is also shown in
Figure 4 that alumina addition decreases the LCO
selectivity similar to that <strong>of</strong> gasoline selectivity. Thus
it can be concluded that alumina addition to FCC
catalyst matrix has a negative role on the liquid
products <strong>of</strong> FCC reactions. This conclusion is in good
agreement with the published literature. 15,17
Figure 5. <strong>The</strong> effect <strong>of</strong> alumina addition on LPG selectivity.
9 0 wt % alumina (CAT0). 0 5 wt % alumina (CAT5), 4 10 wt
% alumina (CAT10), and × 20 wt % alumina (CAT20).
Figure 6. <strong>The</strong> effect <strong>of</strong> alumina addition on coke selectivity.
9 0 wt % alumina (CAT0). 0 5 wt % alumina (CAT5), 4 10 wt
% alumina (CAT10), and × 20 wt % alumina (CAT20).
<strong>The</strong> decrease in liquid product selectivity has to be
compensated by an increase in other products to close
the mass balance. Liquefied petroleum gas (LPG) selectivity
was found to increase with conversion on the
contrary to LCO and gasoline selectivities (see Figure
5). Thus, one reason to the decrease in the liquid product
selectivities during the reaction is their conversion into
gases such as LPG. <strong>Alumina</strong> addtion to the matrix <strong>of</strong>
FCC catalysts was found to slightly increase LPG
selectivity as shown in Figure 5.
Similar to LPG, coke and dry gases (H2-C2) selectivities
were noticed to increase with both conversion (or
C/O ratio) and alumina content. Figures 6 and 7
illustrate the effect <strong>of</strong> alumina on LPG and dry gas
selectivities as function <strong>of</strong> VGO conversion. Catalysts
containing active matrix always produce more coke and
dry gases than those with inactive matrices. This is due
to the presence <strong>of</strong> Lewis acid sites in alumina. <strong>The</strong>se
acid sites have a tendency for coke and gas production

Performance <strong>of</strong> FCC Catalysts Energy & Fuels, Vol. 17, No. 1, 2003 67
Table 6. MAT Products Yields for VGO Cracking at C/O ) 3 for Catalyst CAT0, CAT5, CAT10, and CAT20
compound name CAT0 yield (wt %) CAT5 yield (wt %) CAT10 yield (wt %) CAT20 yield (wt %)
H2 0.02 0.03 0.06 0.09
C1 0.22 0.34 0.36 0.42
C2 0.18 0.27 0.28 0.3
C2 ) 0.34 0.45 0.42 0.45
dry gas (H2+C1+C2+C2 ) ) 0.76 1.09 1.12 1.26
C3 0.51 0.74 0.73 0.76
C3 ) 4.96 5.91 5.63 5.63
i-C4 3.91 5.13 4.77 4.85
n-C4 0.57 0.78 0.75 0.78
t2-C4 ) 2.08 2.34 2.23 2.18
1-C4 ) 1.58 1.77 1.71 1.66
i-C4 ) 2.02 1.94 1.94 1.88
c2-C4 ) 1.57 1.77 1.69 1.65
total C4 ) 7.25 7.82 7.56 7.37
total C4 (olefins + paraffins) 11.73 13.74 13.08 13
LPG (total C3 +total C4) 17.2 20.39 19.44 19.38
gasoline 50.8 52.29 50.82 51.63
LCO 18.57 16.04 17.79 17.09
HCO 11.33 8.2 9.06 8.39
coke 1.34 1.93 1.77 2.24
total 100 99.94 100 100
conversion) 100 - (LCO wt % + HCO wt %) 70.1 75.8 73.2 74.5
Figure 7. <strong>The</strong> effect <strong>of</strong> alumina addition on dry gas selectivity.
9 0 wt % alumina (CAT0). 0 5 wt % alumina (CAT5), 4
10 wt % alumina (CAT10), and × 20 wt % alumina (CAT20).
in processes such as FCC. 17 <strong>The</strong> present results agree
with above arguments (see Figures 6 and 7). Table 6
reports the product distribution for FCC catalysts.
<strong>The</strong> effect <strong>of</strong> alumina addition on light olefin selectivity
is shown in Figure 8. Light olefin selectivity is
defined as the total gas olefin (C2-C4) yield divided by
VGO conversion. It can be seen that alumina addition
has contributed to a mild increase in the product
olefinicity. <strong>The</strong>se results are in agreements with the
argument that alumina addition might increase olefin/
paraffin ratio due to the increased cracking over hydrogen
transfer reaction ratio. 16
Taking the role <strong>of</strong> coke in FCC unit into consideration,
Mott 20 introduced the concept <strong>of</strong> “dynamic activity”
which is defined as the second-order conversion [conversion/(100
- conversion)] divided by coke yield (wt %).
It was found that a plot <strong>of</strong> coke wt % versus the second
order conversion gives a straight line with a slope equal
to the reciprocal <strong>of</strong> catalyst dynamic activity (the specific
coke) as shown in Figure 9. Catalysts with high dynamic
activity have better performance at commercial conditions.
For each catalyst, (with different alumina content)
the graphs were plotted and a slope (reciprocal <strong>of</strong>
Figure 8. <strong>The</strong> effect <strong>of</strong> alumina addition on light olefin (C2-
C4) selectivity. 9 0 wt % alumina (CAT0). 0 5 wt % alumina
(CAT5), 4 10 wt % alumina (CAT10), and × 20 wt % alumina
(CAT20).
dynamic activity) was obtained for each alumina content.
It is extremely important that the relationship
between the VGO second-order conversion and coke
yield is linear at FCC conditions. 20 For the base catalyst
(CAT0), the dynamic activity was 1/(0.391) ) 2.56 while
for those catalysts containing alumina was 1/(0.5771)
) 1.73. Thus although alumina addition to FCC catalyst
matrix has increased the catalyst activity as shown in
Figure 2, it decreased its dynamic activity. Usually
dynamic activity parameter is more important than
activity parameter. 20
Figure 9 shows that CAT0 with inactive matrix has
higher dynamic activity than CAT20 with active matrix.
This result is expected, since alumina increases coke
formation. Thus it can be concluded that FCC catalyst
acidity can be increased by incorporating alumina or by
increasing Y-zeolite unit cell size. 6,7 Increasing these
factors leads to a decrease in the catalyst dynamic
activity. 6,7 However, increasing alumina has better

68 Energy & Fuels, Vol. 17, No. 1, 2003 Al-Khattaf
Figure 9. <strong>The</strong> effect <strong>of</strong> alumina addition on catalyst dynamic
activity. 9 0 wt % alumina (CAT0) and × 20 wt % alumina
(CAT20).
dynamic activity than increasing Y-zeolite unit cell
size.
Finally, it is worth mentioning some aspects <strong>of</strong>
catalysts (with Y-zeolite) and amorphous matrix catalysts
(without Y-zeolite). Referring to Table 2, it can
be seen that the surface area <strong>of</strong> the catalysts is mainly
contributed by the presence <strong>of</strong> Y-zeolite. For example,
CAT10 has a surface area <strong>of</strong> 180 m 2 /g while that <strong>of</strong> M10
has only a 30-m 2 /g surface area or CAT10 has 6 times
larger surface area than that <strong>of</strong> M10. It was reported
that an optimum zeolite to matrix surface area ratio
could maximize the gasoline yield. This ratio was found
to be dependent on the feed and type <strong>of</strong> catalyst used. 21
However, when acidities <strong>of</strong> both M10 and CAT10 are
compared, CAT10 has only 3 times higher acidity than
that <strong>of</strong> M10. CAT20 has around 3.3 times higher surface
area and only 1.7 higher acidity. Consequently, the
zeolite’s effect on the FCC catalyst performance is
mainly attributed to the increase in surface area and
type (strength) <strong>of</strong> acid sites not the amount <strong>of</strong> the acid
sites.
If a comparison is made between the performance <strong>of</strong>
matrix catalyst (M0-M20) and FCC catalysts (CAT0-
CAT20) a large difference can be seen. At C/O ) 3 and
similar reaction conditions, the FCC catalysts have a
VGO conversion in a range <strong>of</strong> 70-75%. Thus no significant
increase in the VGO conversion was obtained by
adding alumina to the catalysts. However, adding
alumina to the kaolin (M0) has clearly increased the
matrix ability to convert the VGO. Kaolin alone (M0)
has conversion <strong>of</strong> 21.1% while M20 (adding 20 wt %
alumina to kaolin) has a 38.1 VGO conversion. Hence,
adding 20 wt % alumina to the kaolin has increased the
VGO conversion by 81%. On the other hand, the VGO
conversion has increased by only 6% by adding 20 wt
% alumina to the FCC catalysts.
This large difference in alumina effect on both cases
is due to the following: kaolin is almost inert material
(21) Andersson, S. I.; Myrstad, T. Oil Gas Eur. Mag. 1997, 23, 19.
and its activity starts to revive as soon as alumina is
added. Alumnia addition to kaolin forms strong active
site and increases the acidity <strong>of</strong> the matrix. Hence the
ability to crack VGO also increases. However, in FCC
catalysts, Y-zeolie forms the main source for the VGO
conversion. Thus, for the feedstock used in the present
study (hydrotreated VGO) which is not heavy feedstock,
no significant influence for alumina addition was found
on the VGO conversion.
4. Conclusions
<strong>The</strong> addition <strong>of</strong> alumina to matrix samples (amorphous)
has influenced its performance and its characteristics.
Active matrix has been shown to have higher
surface area and higher acidity than those without
alumina. Those two characteristics were found to increase
as alumina content in the amorphous matrix
increases. Amorphous matrix sample with 20 wt %
alumina has almost doubled the conversion compared
to the amorphous sample with pure kaolin. It was
shown that the matrix activity was responsible for
cracking the HCO molecules into lighter products.
Amorphous matrix samples containing alumina (active
matrix) were observed to produce more coke and gas
and less gasoline than pure kaolin matrix (inactive
matrix).
Regarding alumina addition to matrix incorporated
in the FCC catalysts, it was noticed that surface area
and acidity <strong>of</strong> the catalysts have increased proportionally
with alumina content. However, surprisingly the
alumina content has influenced the total catalyst acidity
more than the total catalyst surface area.
<strong>The</strong> role <strong>of</strong> alumina addition to FCC catalysts for
processing the hydrotreated VGO was found to be
modest. For example, it increased the FCC catalyst
activity and decreased FCC catalyst dynamic activity.
Thus for hydrotreated VGO, the addition <strong>of</strong> alumina to
the matrix does not appear to make a significant
difference. FCC catalysts with active matrix have more
coke and gas selectivies than the one with inactive
matrix. This increase in gas and coke selectivities was
balanced with a decrease in the total liquid product
selectivity (gasoline and LCO). Thus, it seems the
addition <strong>of</strong> alumina to FCC catalysts has increased the
rate <strong>of</strong> the secondary reactions such as liquid product
overcracking and coking more than the primary VGO
cracking reaction. Finally, the gas olefinicity was found
to increase with alumina content.
Acknowledgment. <strong>The</strong> author acknowledges the
support <strong>of</strong> King Fahd University <strong>of</strong> Petroleum & Minerals.
Financial support was provided by the Petroleum
Energy Center under the organization <strong>of</strong> the Japan
Petroleum Institute. <strong>The</strong> author also thanks Nippon Oil
Company Ltd, for allowing this work to be done in its
Central Technical Laboratory. Special thanks are due
to Dr. T. Ino, Y. Nakatsuka, K. Kato, and T. Okuhara
for their valuable assistance. <strong>The</strong> suggestions <strong>of</strong> Dr. A
Aitani and Dr. K. Loughlin are extremely appreciated.
EF020066A