catalysts
Article
A Comparison of Laboratory Simulation Methods of Iron
Contamination for FCC Catalysts
Yitao Liao 1,2 , Tao Liu 2 , Huihui Zhao 2 and Xionghou Gao 2, *






Citation: Liao, Y.; Liu, T.; Zhao, H.;
Gao, X. A Comparison of Laboratory
Simulation Methods of Iron
Contamination for FCC Catalysts.
Catalysts 2021, 11, 104. https://
doi.org/10.3390/catal11010104
Received: 11 December 2020
Accepted: 7 January 2021
Published: 14 January 2021
Publisher’s Note: MDPI stays neu-
tral with regard to jurisdictional clai-
ms in published maps and institutio-
nal affiliations.
Copyright: © 2021 by the authors. Li-
censee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and con-
ditions of the Creative Commons At-
tribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Catalysts 2021, 11, 104. https://doi.org/10.3390/catal11010104
1 College of Chemistry and Chemical Engineering, Northwest Normal University, Lanzhou 730070, China;
liaoyitao@petrochina.com.cn
2 Lanzhou Petrochemical Research Center, Petrochemical Research Institute, PetroChina, Lanzhou 730060,
China; liutao5@petrochina.com.cn (T.L.); zhaohuihui3@petrochina.com.cn (H.Z.)
* Correspondence: gaoxionghou@petrochina.com.cn; Tel.: +86-010-8016-5606
Abstract: Two different methods of simulating iron contamination in a laboratory were studied.
The catalysts were characterized using X-ray diffraction, N2 adsorption–desorption, and SEM-EDS.
The catalyst performance was evaluated using an advanced cracking evaluation device. It was found
that iron was evenly distributed in the catalyst prepared using the Mitchell impregnation method
and no obvious iron nodules were found on the surface of the catalyst. Iron on the impregnated
catalyst led to a strong dehydrogenation capacity and a slight decrease in the conversion and bottoms
selectivity. The studies also showed that iron was mainly in the range of 1–5 µm from the edge of the
catalyst prepared using the cycle deactivation method. Iron nodules could be easily observed on the
surface of the catalyst. The retention of the surface structure in the alumina-rich areas and the collapse
of the surface structure in the silica-rich areas resulted in a continuous nodule morphology, which was
similar to the highly iron-contaminated equilibrium catalyst. Iron nodules on the cyclic-deactivated
catalyst led to a significant decrease in conversion, an extremely high bottoms yield, and a small
increase in the dehydrogenation capacity. The nodules and distribution of iron on the equilibrium
catalyst could be better simulated by using the cyclic deactivation method.
Keywords: Mitchell impregnation; cyclic deactivation; iron contamination; iron nodules
1. Introduction
How to test a fluid catalytic cracking (FCC) catalyst’s performance and metal tolerance
in the laboratory and calibrate the results to simulate the industrial performance is an
ongoing discussion. Another important aspect of catalyst testing is the preparation of
the catalyst [1]. In evaluating a fluid cracking catalyst’s performance in lab-scale testing,
selecting the proper catalyst deactivation method is just as important as the testing itself [2].
When heavy metals are deposited on the catalyst, the physical and chemical properties
of the catalyst will be affected by the distribution, valence state, and existing forms of the
heavy metals. Therefore, it is difficult to accurately simulate the industrial heavy-metal-
polluted equilibrium catalyst by selecting the laboratory deactivation protocols. Recently,
due to the relatively low supply of light and low-sulfur crude oil, which has a relatively
high market price, refiners are obliged to process increasing quantities of heavy crude
by cracking them into lighter distillates [3,4]. With the increasing trend of heavy and
poor-quality FCC feedstocks, the quantity and species of heavy metal deposition, such as
Ni [5,6], V [7,8], Ca [9,10], Na [11], and especially iron [12], on the catalyst have gradually
increased. Iron deactivation effects can be subdivided into the following two classifications:
(1) iron nodules may affect the catalyst’s apparent bulk density (ABD) and fluidization [13]
and (2) increasing the amount of iron will cause a loss of activity and bottoms cracking,
as well as increased SOx emissions and coke on the regenerated catalyst (CRC) in partial
burn units [14].
https://www.mdpi.com/journal/catalysts
Catalysts 2021, 11, 104 2 of 14

Multiple kinds of heavy metals deposition on a catalyst increase the difficulty of

selecting a deactivation protocol. The detrimental effects of nickel and vanadium that were
studied using deactivation protocol simulations have been described elsewhere. Buurmans
et al. [15] investigated the effect of nickel and vanadium deactivation on the structure
and acidity of FCC catalyst particles by using steaming, cyclic deactivation, and the
Mitchell impregnation method. Nguyen et al. [16] investigated the effect of hydrothermal
conditions on the catalytic deactivation of an FCC catalyst by using a cyclic propylene
steaming (CPS) method. Etim et al. [17,18] investigated the role of nickel on a vanadium-
poisoned FCC catalyst and the effect of vanadium contamination on the framework and
micropore structure by using the Mitchell impregnation method. However, few studies
on the simulation and catalytic performance of iron contamination have been reported.
Nobody has successfully simulated the iron nodules observed on the surface of an iron-
contaminated equilibrium catalyst in the laboratory. If there is no accurate method for
simulating iron deactivation, the research on the impact of iron contamination and the
development of iron tolerance technology in the laboratory will not have real and reliable
theoretical support, and sometimes wrong conclusions may be obtained. Therefore, in order
to understand the real effect of iron on the catalyst, it is necessary to establish an accurate
and effective simulation method of iron deactivation.
Laboratory catalyst deactivation protocols are used in order to simulate long-term com-
mercial catalyst deactivation in an accelerated way [19]. A traditional way was proposed
using the Mitchell impregnation method, which involves the impregnation of the catalyst
with metal solutions, followed by calcination and a steam treatment [20]. An advanced
method was to use the cyclic deactivation (CD) method, which has been described else-
where. By using the CD method, Psarras et al. [21] investigated the accessibility effect on
the irreversible deactivation of FCC catalysts. Almas et al. [22] studied the transformations
of FCC catalysts and carbonaceous deposits during repeated reaction–regeneration cycles.
Rainer et al. [23] investigated the Akzo Accessibility Index (AAI) of iron-contaminated cat-
alysts under different conditions. One CD section consisted of repeated cycles of cracking,
stripping, and regeneration treatment. The deactivation of the FCC catalyst in a CD unit is
a simulation of the deactivation processes in a commercial plant.
In this work, a commercial catalyst (CAT-BASE) was characterized and tested in an
advanced cracking evaluation (ACE) unit following cyclic deactivation (CAT-CD) and
Mitchell impregnation with different iron species (CAT-MM1 using iron chloride and
CAT-MM2 using iron naphthenate) in order to obtain a more direct comparison of the
two laboratory methods regarding an iron-contaminated equilibrium catalyst simulation.
Finally, the nodules, iron distribution, and the performance of the iron contaminated
equilibrium catalyst could be better simulated by using the cyclic deactivation method.

2. Results
2.1. Iron Deposited on the Catalyst Using Different Methods
In order to study the effect of different contamination methods on the properties
and catalytic performance of an iron-contaminated catalyst, three catalysts based on CAT-
BASE were prepared by using iron chloride and iron naphthenate as precursors. An iron-
contaminated equilibrium catalyst (E-cat) was selected as the target catalyst for different
deactivation methods. CAT-BASE and E-cat were collected from the industrial plant of
Dalian West Pacific Petrochemical Co., Ltd. (WEPEC, Dalian, China). The metal contents of
the catalysts were obtained using XRF. The properties of the catalysts are shown in Table 1.
There was a certain amount of iron (1828 µg/g) on CAT-BASE, which came from the
kaolin used to prepare the catalyst but it had no poisoning effect on the performance of
the catalyst [24]. Three different metal-depositing methods were used to load almost the
same amount of iron onto the catalyst. Finally, the target iron deposition was 12,000 µg/g,
where the actual iron deposition on CAT-MM1 was 11,480 µg/g and that on CAT-MM2
was 11,440 µg/g. A similar amount of iron (11,690 µg/g) was deposited onto CAT-CD.
 Catalysts ω(Na22O) (%) ω(Al22O33) (%) S.A.(m22g−1
−1) P.V.total (cm33g−1
total
−1) Unit Cell Size (nm) Fe on Catalyst (μ

CAT-BASE 0.16 48.6 269 0.42 2.462 1828

CAT-MM1 0.16 48.6 149.4 0.21 2.440 13,308
CAT-MM2 0.16 48.6 137.6 0.20 2.439 13,268
Catalysts 2021, 11, 104 3 of 14
CAT-CD 0.16 48.6 122.5 0.16 2.436 13,518
E-cat 0.14 47.4 109.7 0.15 2.437 10,471
S.A.: Surface
S.A.: Surface Area;
Area; P.V.:
P.V.: Pore
Pore Volume;
Volume; CAT-BASE:
CAT-BASE: commercial
commercial catalyst,
catalyst, CAT-MM1:
CAT-MM1: catalyst
catalyst after
after using
using iron
iron chloride,
chloride, CAT
CAT
Table 1. Properties of different catalysts.
catalyst after
catalyst after using
using iron
iron naphthenate,
naphthenate, CAT-CD:
CAT-CD: catalyst
catalyst after
after cyclic
cyclic deactivation,
deactivation, E-cat:
E-cat: iron-contaminated
iron-contaminated equilibrium
equilibrium cataly
cataly
Fe on
ω (Na
There was O) ω (Al
a 2certain 2 O3 )
amount S.A. (1828P.V.
of iron total on Unit
μg/g) Cell
CAT-BASE, which came fro
fr
Catalysts 2 − 1 3 − 1 Catalyst
(%) (%) (m ·g ) (cm ·g ) Size (nm) −1)
(µg · g
kaolin used to prepare the catalyst but it had no poisoning effect on the performa
the catalyst [24].
CAT-BASE 0.16Three different
48.6 metal-depositing
269 0.42 methods were used
2.462 1828to load alm
CAT-MM1 0.16 48.6 149.4 0.21 2.440 13,308
same amount of iron onto the catalyst. Finally, the target iron deposition was
CAT-MM2 0.16 48.6 137.6 0.20 2.439 13,268
μg/g,
CAT-CD where the 0.16 actual 48.6
iron deposition
122.5 on CAT-MM1
0.16 was 11,480
2.436 μg/g and th
13,518 t
E-cat 0.14 47.4 109.7 0.15
CAT-MM2 was 11,440 μg/g. A similar amount of iron (11,690 μg/g) was deposite 2.437 10,471
S.A.: Surface Area; P.V.: Pore Volume; CAT-BASE: commercial catalyst, CAT-MM1: catalyst after using iron
CAT-CD.
chloride, CAT-MM2: catalyst after using iron naphthenate, CAT-CD: catalyst after cyclic deactivation, E-cat:
iron-contaminated equilibrium catalyst.

2.2. Iron-Contaminated Catalysts’ Morphology

2.2. Iron-Contaminated Catalysts’ Morphology
The surface morphology of an industrial iron-contaminated equilibrium cata
The surface morphology of an industrial iron-contaminated equilibrium catalyst is
described
described elsewhere
elsewhere [25–27].
[25–27]. The
The iron iron can
nodules nodules canobserved
be clearly be clearly observed
on the on the E-c
E-cat surface
face
in in Figure
Figure 1a.the
1a. During During theofformation
formation iron nodules,ofthe
iron
ironnodules, the iron
formed a eutectic formed
and releaseda eutect
areleased
lot of heatatolot
melt
ofthe silica-alumina
heat to melt the matrix layer [28]. A dense
silica-alumina layerlayer
matrix was formed
[28]. Athat is
dense lay
laye
typically a few microns in depth, as shown in Figure 2a.
formed that is typically a few microns in depth, as shown in Figure 2a.

Figure1. 1.
Figure
Figure 1.
SEMSEM micrographs
micrographs
SEM of
of (a) of
micrographs (a) (b)
E-cat,
(a) E-cat, (b) CAT-MM1,
CAT-MM1,
E-cat, (b) CAT-MM1, (c) CAT-MM2,
(c) CAT-MM2,
(c) CAT-MM2, and (d)
and (d) CAT-CD.
and (d) CAT-CD.
CAT-CD.

Figure2.2.
Figure
Figure 2.SEM
SEM
SEM micrographs
micrographs
micrographs ofcatalyst
of
of the the catalyst
the catalyst cross-section:
cross-section:
cross-section: (a)
(a) E-cat,(a) E-cat,
(b)E-cat, (b) CAT-MM1,
(b)
CAT-MM1, CAT-MM1, (c) CAT-M
(c)
(c) CAT-MM2, CAT-M
and(d)
and
and (d)
(d) CAT-CD.
CAT-CD.
CAT-CD.
 The contamination sources used for the impregnation of CAT-MM1 and CAT
were iron chloride and iron naphthenate, where their lengths [25] were less th
pore diameter of CAT-BASE. During the impregnation process, iron chloride an
Catalysts 2021, 11, 104 naphthenate may have entered into the pore of the catalyst; therefore, 4 ofno
14 obviou
nodules and boundary can be observed on the catalyst surface and cross-section (
2b,c)
Iron
The nodules
surfaces similar to
of CAT-MM1 andthose in Figure
CAT-MM2 in Figure1a1b,c
canwerebe relatively
observedsmooth.
on the surf
CAT-CD
The in Figure
contamination 1d.used
sources A few microns
for the of theoflayer
impregnation can also
CAT-MM1 be observed
and CAT-MM2 were in Figu
iron chloride and iron naphthenate, where their lengths [25] were less than
Therefore, compared with the Mitchell impregnation method, the cyclic deacti the pore diame-
ter of CAT-BASE.
method During
could truly the impregnation
simulate process,
the surface iron chloride
morphology of and iron naphthenate
an industrial iron-contam
may have entered into the pore of the catalyst; therefore, no obvious iron nodules and
equilibrium catalyst.
boundary can be observed on the catalyst surface and cross-section (Figure 2b,c)
Iron nodules similar to those in Figure 1a can be observed on the surface of CAT-CD
2.3.
in Iron1d.
Figure Distribution on the
A few microns Catalyst
of the layer can also be observed in Figure 2d. Therefore, com-
pared with the Mitchell impregnation method, the cyclic deactivation method could truly
Within the industrial FCC unit, the metals are usually deposited on the equil
simulate the surface morphology of an industrial iron-contaminated equilibrium catalyst.
catalyst with different deposition profiles. For example, nickel is known to depo
theIron
2.3. outer part of on
Distribution thethecatalyst
Catalyst [29]. Vanadium mainly diffuses from the outside to
sideWithin
[30,31].theIron doesFCC
industrial not unit,
migrate to theare
the metals interior
usuallyof the catalyst
deposited on thebut exists in the de
equilibrium
catalyst
1~5 μmwith different
[32,33]. deposition
Because profiles.
the type For example,
of metal deposition nickel is known
dictates thetoactivity
deposit on
and beha
the outer part of the catalyst [29]. Vanadium mainly diffuses
the metal, which may have an influence on the physical and chemical propertiesfrom the outside to the
inside [30,31]. Iron does not migrate to the interior of the catalyst but exists in the depth
catalyst, the deposition characteristics are important to study and quantify [34].
of 1~5 µm [32,33]. Because the type of metal deposition dictates the activity and behavior
of the As shown
metal, whichin Figure
may have an3, influence
all catalysts
on thewere lineand
physical scanned
chemical from centerofto
properties theedge. T
tribution
catalyst, theof iron on characteristics
deposition E-cat gradually decreased
are important to from the quantify
study and outside [34].
to the inside in Fig
TheAs shown
iron in Figure
mainly 3, all catalysts
deposited on the were
outerline scanned
surface offrom center
E-cat. to edge.4b,c,
In Figure The dis-
the iron p
tribution
ed a uniform distribution from the center to the edge. During the process4a.of the M
of iron on E-cat gradually decreased from the outside to the inside in Figure
The iron mainly deposited on the outer surface of E-cat. In Figure 4b,c, the iron presented
impregnation, as the diameters of the iron chloride and iron naphthenate were s
a uniform distribution from the center to the edge. During the process of the Mitchell
than that of as
impregnation, thethecatalyst’s
diameters pores, twochloride
of the iron kinds and of iron-contaminated
iron naphthenate were solution
smaller diffuse
the catalyst
than that of thepore at one
catalyst’s time
pores, tokinds
two formofan equal-volume state.
iron-contaminated solutionAfter calcination
diffused into at
the catalyst pore at one time to form an equal-volume state. After
temperature, the iron was evenly distributed in the catalyst. When using the M calcination at a high
temperature,
impregnation the method
iron was evenly distributed
to simulate ironincontamination,
the catalyst. When dueusing the characteristics
to the Mitchell
impregnation method to simulate iron contamination, due to the characteristics of the
method, it is impossible to simulate the real reaction and regeneration environm
method, it is impossible to simulate the real reaction and regeneration environment of iron
iron
in in an industrial
an industrial plant.iron
plant. Therefore, Therefore,
was evenlyiron was evenly
distributed distributed
in the catalyst obtained in the cataly
using
tained
the using
Mitchell the Mitchell
impregnation impregnation method.
method.

Figure3. 3.
Figure Line
Line scan
scan across
across a particle:
a particle: (A)(B)
(A) E-cat, E-cat, (B) CAT-MM1,
CAT-MM1, (C)and
(C) CAT-MM2, CAT-MM2, and (D) CAT-
(D) CAT-CD.
Catalysts 2021,
Catalysts 11,11,x 104
2021, FOR PEER REVIEW 5 of 14 5 of 14

Catalysts 2021, 11, x FOR PEER REVIEW 5 of 14

Figure 4. Iron
Figure 4.
Figure Irondistribution
distribution
distribution ainparticle:
inin aaparticle:
particle: (a)E-cat,
(a) E-cat,
(a) E-cat, (b) CAT-MM1,
(b) CAT-MM1,
(b) (c) (c)
(c) CAT-MM2,
CAT-MM1, CAT-MM2,
and (d)and
CAT-MM2, and
CAT-CD.(d) CAT-CD.
(d) CAT-CD.
CPS: Counts
CPS:
CPS: Counts Per
Counts PerSecond.
Per Second.
Second.

The iron
The iron content ininthe depth ofof 1~5 μmµm from the the
edge on CAT-CD was was
higher.
The iron content
content inthe thedepth
depth of 1~5 1~5 μm from from the edge on CAT-CD
edge on CAT-CD higher.
was higher.
When
When moving
moving closer
closer to
to the
the center
center of
of the
the catalyst,
catalyst, the
the iron
iron content
content decreased
decreased and
and tended
tended
When moving
to become closer
balanced to the
(Figure4d). center
4d).The of the
Thedistribution catalyst,
distribution of the iron content decreased andtotended
become balanced (Figure of iron
ironon onCAT-CD
CAT-CDwas wassimilar
similar totothat
that of
toof become
E-cat. The balanced
CD method (Figure 4d). Thecycles
usedmultiple
multiple distribution
of reaction of iron on CAT-CD was similar to that
E-cat. The CD method used cycles of reactionand andregeneration
regeneration steps to to
steps real-
realize
of E-cat.
heavy The
metalCD method
deposition onused
the multiple
catalyst’s cycles
surface. of
In a reaction
heavy metal deposition on the catalyst’s surface. In a single cycle, 10 g of vacuumgas
ize single and
cycle, 10 regeneration
g of vacuum steps to real-
gas oil
ize
oil
(VGO; heavy
(VGO;Fe onmetal
Fe VGOdeposition
on VGO = 860
= 860 µg/g) onwas
μg/g) thepumped
was catalyst’s
pumped surface.
into
into the In areactor
quartz
the quartz single to
reactor tocycle,
react10
react with g of150
with 150 vacuum
g g of gas
of the
oil catalyst.
(VGO; Fe Iron
on VGOcompounds
= 860 were was
μg/g) deposited on theinto
pumped catalyst
the surface after
quartz reactorthe to
cracking
react with 150 g
the catalyst. Iron compounds were deposited on the catalyst surface after the cracking step.
ofstep. During
the catalyst. the regeneration
Iron compounds step, the
were iron compounds
depositedformed formed
on theacatalyst a eutectic with silica,
During the regeneration step, the iron compounds eutectic surface after
with silica, the cracking
alumina,
alumina, and alkali metal compounds and released a lot of heat to melt the sili-
step. During
and alkali metalthecompounds
regeneration andstep,
released theairon
lot ofcompounds
heat to melt formed a eutecticmatrix
the silica-alumina with silica,
ca-alumina matrix layer to block and cover the pores of the catalyst. With an increase in
layer to block
alumina, and covermetal
the pores of the catalyst. With an increase in of
theheat
number of cycles,
the number of cycles, the pore closing became more serious. It was difficult for the new- the sili-
and alkali compounds and released a lot to melt
the pore closing became
ca-alumina layermore serious. It was the
difficult for the newly depositedan iron to
ly deposited matrix
iron to penetrate to block and cover
the catalyst, which could pores
only of theand
react catalyst.
exist in With
the range increase in
penetrate
the
of anumber the
few microns catalyst,
of cycles, which
on thethe could only
pore closing
catalyst’s react
surface. became and
Therefore, exist
more in the range of a few
serious. It characteristics
the distribution microns
was difficult of for on
the new-
the
iron catalyst’s
on an surface.equilibrium
industrial Therefore,catalyst
the distribution
can be characteristics
simulated by using of iron
the on deactiva-
cyclic an industrial
ly deposited iron to penetrate the catalyst, which could only react and exist in the range
equilibrium
tion catalyst can be simulated by using the cyclic deactivation method.
of a method.
few microns on the catalyst’s surface. Therefore, the distribution characteristics of
iron
2.4. on an Surface
industrial
Areaequilibrium catalyst can be simulated by using the cyclic deactiva-
2.4. Specific
Specific Surface Area
tion method.
The trend of the zeolite, matrix, and total surface areas measured using nitrogen
The trend of the zeolite, matrix, and total surface areas measured using nitrogen
adsorption forE-cat,
adsorption for E-cat,the
theMitchell
Mitchellimpregnated
impregnated catalysts
catalysts (with
(with different
different ironiron precursors),
precursors),
2.4.
and Specific
the Surface Area
cyclic-deactivated catalysts are shown in Figure
and the cyclic-deactivated catalysts are shown in Figure 5. 5.
The trend of the zeolite, matrix, and total surface areas measured using nitrogen
adsorption for E-cat, the Mitchell impregnated catalysts (with different iron precursors),
and the cyclic-deactivated catalysts are shown in Figure 5.

Figure Zeolite,matrix,
Figure 5. Zeolite, matrix,and
andtotal
totalsurface
surface areas
areas of of
thethe different
different catalysts.
catalysts.
Catalysts 2021, 11, x FOR PEER REVIEW 6 of

Among the three deactivation methods, the total surface area of CAT-CD was t
Catalysts 2021, 11, 104 6 of 14
closest to that of E-cat. The zeolite and matrix surface areas were also studied. The ze
lite surface was not significantly affected by different deactivation methods, while t
matrix surface area changed greatly. Because of the uniform distribution of iron
Among the
CAT-MM1 andthree deactivation
CAT-MM2, the methods, the total
pore closing surface
effect had area of CAT-CD
no serious wason
effect thethe mat
closest to that of E-cat. The zeolite and matrix surface areas were also studied. The zeolite
surface area. As iron is mainly deposited on the outer surface of CAT-CD and form
surface was not significantly affected by different deactivation methods, while the matrix
layer that is a few microns thick, its matrix surface area was the lowest among the thr
surface area changed greatly. Because of the uniform distribution of iron on CAT-MM1 and
methods.
CAT-MM2, The reduction
the pore in thehad
closing effect total surfaceeffect
no serious areaoncaused by surface
the matrix the iron contamination
area. As iron w
mainly due to the influence of the matrix. Therefore, an iron resistance catalyst
is mainly deposited on the outer surface of CAT-CD and forms a layer that is a few microns may
designed fromsurface
thick, its matrix the perspective
area was theoflowest
improving the three
among the ability of the The
methods. matrix to accommoda
reduction in
the total surface area caused by the iron contamination was mainly due to the influence of
iron.
the matrix. Therefore, an iron resistance catalyst may be designed from the perspective of
improving the ability of the matrix to accommodate iron.
2.5. Iron Effects on the Catalyst Product Yields
TheEffects
2.5. Iron product
on theyields
CatalystofProduct
catalysts contaminated using the Mitchell impregnation a
Yields
cyclicThedeactivation
product yields methods were
of catalysts measuredusing
contaminated using
the an ACEimpregnation
Mitchell unit. The product
and yie
cyclic deactivation
curves methods
were obtained bywere measured
changing theusing an ACE unit.ratio
catalyst-to-oil The product yield curves
(C/O) [35]. The conversi
were obtained
curves by changing
as a function thecatalyst-to-oil
of the catalyst-to-oil ratio
ratio(C/O) [35]. The conversion
for different curves
catalysts are as a in Figu
shown
function of the catalyst-to-oil ratio for different catalysts are shown in Figure 6.
6.

Figure
Figure 6.6.Conversion
Conversionas as a function
a function ofcatalyst-to-oil
of the the catalyst-to-oil (C/O)
(C/O) ratio forratio for different
different catalysts. catalysts.

The catalysts’ conversion increased with the increase of the C/O ratio. The adjustment
The catalysts’ conversion increased with the increase of the C/O ratio. The adju
of the C/O ratio could be realized by changing the catalysts’ addition amounts under
ment
a fixedoffeed
theinjection
C/O ratio could An
quantity. be realized
increasedby C/O changing the that
ratio meant catalysts’ additionwere
more catalysts amounts u
der
addeda and
fixedmorefeed injection
active centersquantity.
were able to An increased
function in theC/O ratiowhich
reaction, meant that more
eventually led cataly
were
to the added
increaseand more active centers were able to function in the reaction, which even
in conversion.
The to
ally led average conversion
the increase in of the catalyst that was deactivated using the CD method was
conversion.
lowerThe
thanaverage
that usingconversion
the Mitchell impregnation
of the catalyst method. The conversion
that was deactivated values and the
using trends
CD meth
of CAT-CD were also close to those of E-cat. During the process
was lower than that using the Mitchell impregnation method. The conversion valu of iron contamination
using the Mitchell impregnation method, the effect of pore closing on the catalyst was not
and trends of CAT-CD were also close to those of E-cat. During the process of iron co
serious, which led to the highest conversions for CAT-MM1 and CAT-MM2. A layer that
tamination usingthick
was a few microns the covered
Mitchell theimpregnation
surface of CAT-CD, method,
which the effect the
prevented of accessibility
pore closing on t
catalyst was
of the active sitesnot serious,
of the catalystwhich
may have ledresulted
to thein highest
an at leastconversions
4.84% decreasefor CAT-MM1 a
compared
CAT-MM2.
to CAT-MM1 A and layer that was a few microns thick covered the surface of CAT-CD, whi
CAT-MM2.
The hydrogen
prevented factor curvesof
the accessibility of the
different
activecatalysts arethe
sites of shown in Figure
catalyst may 7. have
The hydrogen
resulted in an
factor is the mole ratio of hydrogen to methane
least 4.84% decrease compared to CAT-MM1 and CAT-MM2. in cracking products, which is an indicator
of dehydrogenation reactions and can reflect the dehydrogenation ability of a catalyst [36].
The hydrogen factor curves of different catalysts are shown in Figure 7. The hyd
gen factor is the mole ratio of hydrogen to methane in cracking products, which is an
dicator of dehydrogenation reactions and can reflect the dehydrogenation ability o
catalyst [36].
Catalysts 2021, 11, x FOR PEER REVIEW 7 of 14

Catalysts 2021, 11, 104 7 of 14

Catalysts 2021, 11, x FOR PEER REVIEW 7 of 14

Figure 7. Hydrogen factor (H2/CH4 ratio) as a function of the catalyst-to-oil ratio for different cata-
lysts.

Figure
Figure The7. dehydrogenation
7. Hydrogen
Hydrogen factor(H
factor (H ability
2/CH
2 /CH ofasas
ratio)
44 ratio) E-cat was of
aafunction
function the
of
the weakest,
the as part
catalyst-to-oil
catalyst-to-oil of different
ratio
ratio for thedifferent
for ironcatalysts.
was pas-
cata-
lysts. after a long period of operation in the FCC unit. Both CAT-MM1 and CAT-MM2
sivated
had theThestrongest
dehydrogenation ability of E-cat
dehydrogenation abilitywasand thetheir
weakest,
hydrogenas part of thevalues
factor iron was werepassi-
close.
vated after
The a long period
dehydrogenation of operation
ability of in
E-catthe FCC
was unit.
the
Iron penetrated through the inner channels of CAT-MM1 and CAT-MM2. Therefore, Both
weakest, CAT-MM1
as part of and
the CAT-MM2
iron was pas-
it
had the
sivated strongest
after a long dehydrogenation
period of operation ability and
in the their
FCCthe hydrogen
unit. factor
Bothcontamination
CAT-MM1 andvalues were close.
CAT-MM2
had a stronger dehydrogenation capacity under same content. The
Iron
had penetrated
the strongest through the inner channels
dehydrogenation abilityof and
CAT-MM1 and CAT-MM2.
their During
hydrogen Therefore, it had
hydrogen factor curves were in the highest position. thefactor
process values were
of cyclic close.
deacti-
aIron
stronger dehydrogenation capacity under the same contamination content. The hydrogen
vation, part of the iron may have gathered in the nodules and had a weak dehydrogena-it
penetrated through the inner channels of CAT-MM1 and CAT-MM2. Therefore,
factor acurves
had capacity.
stronger weredehydrogenation
in the highest position. capacityDuringunder the the
process
same of cyclic deactivation,
contamination part ofThe
content.
tion Therefore, the dehydrogenation capacity
the iron may have gathered in the nodules and had a weak dehydrogenation capacity. of CAT-CD was lower than for
hydrogen
CAT-MM1 factor curves
anddehydrogenation were
CAT-MM2 under in the highest
the same position.
contamination During the
content. process
CAT-CD of cyclic deacti-
couldand truly
Therefore, the capacity of CAT-CD was lower than for CAT-MM1
vation,
simulate part
the of the iron
dehydrogenation may have gathered
capacity of in
an the nodules
industrial and had
iron-contaminateda weak dehydrogena-
equilibrium
CAT-MM2 under the same contamination content. CAT-CD could truly simulate the
tion capacity. Therefore, the dehydrogenation capacity of CAT-CD was lower than for
catalyst.
dehydrogenation capacity of an industrial iron-contaminated equilibrium catalyst.
CAT-MM1
The
The bottoms
and CAT-MM2
bottoms curvesofofthe
curves
under
the the same
different
different
contamination
catalysts
catalysts are are
shown shown content.
in Figure
in Figure
CAT-CD could
8. According
8. According
truly
to the to
simulate
the positions
positions the
of the dehydrogenation
of different
the different
curves, capacity
curves, of an
the bottoms
the bottoms industrial
yield ofyield iron-contaminated
of the catalyst
the catalyst that wasthat equilibrium
was deac-
deactivated
catalyst.
tivated
using using
the the Mitchell
Mitchell impregnation
impregnation method was method lowerwas than lower
that ofthan thatThe
E-cat. of E-cat.
matrixThe matrix
surface
surface The bottoms
area of CAT-MM1
area of CAT-MM1 curves of
and CAT-MM2 the different
and CAT-MM2 catalysts
was lessThe
was less affected. are shown
affected. in Figure
The feedand
feed molecules 8. According
molecules
productsand to
the positions
could move
products of the
relatively
could different
movefreely curves,
in the catalyst
relatively the bottoms
freely inchannel. yield
Therefore,
the catalyst of the catalyst
the average
channel. Therefore, that
bottoms was
the yieldsdeac-
average
tivated
of CAT-MM1
bottoms using
yieldsandtheofMitchell
CAT-MM2
CAT-MM1 impregnation
were
and 4.4% method
lower
CAT-MM2 than was
were thatlower
4.4% than The
of E-cat.
lower that
thanporeofthat
E-cat.
of The
structure matrix
E-cat. ofThe
surface
CAT-CD
pore area
structure ofofCAT-MM1
was seriously CAT-CD affected andby
was CAT-MM2
iron nodules
seriously was
affectedsuch less
by affected.
that its nodules
iron matrix The feed that
surface
such molecules
area itsgreatly and
matrix
products
decreased. could
It was move relatively
difficult for the freely
feed in
moleculesthe catalyst
to diffuse channel.
surface area greatly decreased. It was difficult for the feed molecules to diffuse into the into the Therefore,
catalyst and the average
contact
the activeand
bottoms
catalyst center
yields forCAT-MM1
of
contact cracking.
the active Asandacenter
result,
CAT-MM2the cracking.
for bottoms
wereyield As aoflower
4.4% CAT-CD
result, thegreatly
than that
bottoms increased.
of E-cat.
yieldThe of
Its bottoms
pore structure
CAT-CD curve was
greatlyofincreased. similar
CAT-CD Its to
was that of E-cat.
seriously
bottoms curveaffected
was similarby iron nodules
to that such that its matrix
of E-cat.
surface area greatly decreased. It was difficult for the feed molecules to diffuse into the
catalyst and contact the active center for cracking. As a result, the bottoms yield of
CAT-CD greatly increased. Its bottoms curve was similar to that of E-cat.

Figure
Figure 8. Bottoms as
as aa function
functionof
ofthe
thecatalyst-to-oil
catalyst-to-oilratio
ratioforfor different
different catalysts.
catalysts.

Figure 8. Bottoms as a function of the catalyst-to-oil ratio for different catalysts.

Catalysts 2021, 11, 104 8 of 14

Catalysts 2021, 11, x FOR PEER REVIEW 8 of 14

The products’ selectivities toward the catalysts that were deactivated using different
methods are indicated
The products’ in Tabletoward
selectivities 2 at thethe
same 74 wt.%
catalysts thatconversion.
were deactivated using differ-
ent methods are indicated in Table 2 at the same 74 wt.% conversion.
Table 2. Product yields at 74 wt.% conversion.
Table 2. Product yields at 74 wt.% conversion.
ω(Yields) (%) E-Cat CAT-CD CAT-MM1 CAT-MM2
ω(Yields) (%) E-Cat CAT-CD CAT-MM1 CAT-MM2
Coke 9.18 9.33 10.29 9.87
Dry Coke
gas 3.289.18 9.33
3.22 10.29
3.24 9.873.06
Dry
H2 gas 0.463.28 3.22
0.51 3.24
0.69 3.060.67
H2
Gasoline 44.560.46 0.51
44.87 0.69
43.00 0.6743.19
LCO
Gasoline 14.06
44.56 13.89
44.87 14.21
43.00 14.12
43.19
Bottoms
LCO 11.94
14.06 12.11
13.89 11.79
14.21 11.88
14.12
LPG 16.88 16.58 17.47 17.88
Bottoms 11.94 12.11 11.79 11.88
LCO: Light Cyclic Oil; LPG: Liquefied Petroleum Gas.
LPG 16.88 16.58 17.47 17.88
LCO: Light Cyclic Oil; LPG: Liquefied Petroleum Gas.
The coke and hydrogen selectivities of E-cat and CAT-CD were similar and were
smaller
Thethan
coke those of CAT-MM1
and hydrogen and CAT-MM2.
selectivities of E-cat andThis indicates
CAT-CD were that theand
similar ironwere
on CAT-
MM1
smallerand CAT-MM2
than those ofhad strongerand
CAT-MM1 dehydrogenation
CAT-MM2. This activity underthat
indicates the the
same amount
iron on of
iron contamination,
CAT-MM1 which produced
and CAT-MM2 had stronger more coke and hydrogen.
dehydrogenation Part
activity of thethe
under iron in E-cat
same
amount
and of iron
CAT-CD wascontamination,
embedded and which
henceproduced more coke and activity,
lost dehydrogenation hydrogen. Part of in
resulting theweak
iron in E-cat and CAT-CD was embedded and hence lost dehydrogenation
dehydrogenation activity. The bottoms selectivities of E-cat and CAT-CD were higher activity, re- than
sulting in weak dehydrogenation activity. The bottoms selectivities
those of CAT-MM1 and CAT-MM2, which indicates that the impregnated catalyst had a of E-cat and
CAT-CD
better were higher
cracking than
capacity forthose
heavy ofoil.
CAT-MM1 and CAT-MM2,
The product selectivity of which indicates
CAT-CD that the
was closer to that
impregnated
of E-cat. catalyst had a better cracking capacity for heavy oil. The product selectivity
of CAT-CD was closer to that of E-cat.
2.6. NH3 -TPD Analysis of Different Catalysts
2.6. NH3-TPD Analysis of Different Catalysts
Figure 9 shows the NH3 -TPD (Temperature Programmed Desorption) profiles of E-cat
Figure
and the 9 shows
catalysts the NHusing
prepared 3-TPD (Temperature Programmed Desorption) profiles of
different methods. The NH3 -TPD profiles demonstrate
that the peak area of the weakusing
E-cat and the catalysts prepared different methods.
and medium-strong The NH
acids 3-TPD profiles demon-
of CAT-CD was smaller than
strate that the peak area of the weak and medium-strong
those of CAT-MM1 and CAT-MM2. A pore-closing effect made acids of CAT-CD was for
it harder smaller
the NH3
molecules to get through the pores, which meant that the number of acid sites for
than those of CAT-MM1 and CAT-MM2. A pore-closing effect made it harder the
in CAT-CD
NH3 molecules to get through the pores, which meant that the number of acid sites in
was fewer. The number of acid sites was equal to the number of active sites. The number
CAT-CD was fewer. The number of acid sites was equal to the number of active sites.
of active sites decreased in the following order: CAT-MM1 > CAT-MM2 > E-cat > CAT-CD.
The number of active sites decreased in the following order: CAT-MM1 > CAT-MM2 >
This was consistent with the change in the catalyst conversion and product distribution
E-cat > CAT-CD. This was consistent with the change in the catalyst conversion and
discussed before.
product distribution discussed before.

Figure 9.
Figure 9. NH
NH3-TCD (Thermal Conductivity Detector) profiles of the different catalysts.
3 -TCD (Thermal Conductivity Detector) profiles of the different catalysts.
Catalysts 2021, 11, 104 9 of 14
Catalysts 2021, 11, x FOR PEER REVIEW 9 of 14

2.7. SEM-EDS Analysis of the Iron Nodule Formation Mechanism

2.7. SEM-EDS Analysis of the Iron Nodule Formation Mechanism
The element content on the raised and valley areas of the catalyst deactivated using
The element content on the raised and valley areas of the catalyst deactivated using
the cyclic deactivation method was observed in area scan mode using scanning electron
the cyclic deactivation method was observed in area scan mode using scanning electron
microscopy (SEM)inincombination
microscopy (SEM) combinationwith with X-ray
X-ray energy
energy dispersive
dispersive spectroscopy
spectroscopy (EDS).
(EDS).
The distribution of silicon, aluminum, iron, calcium, and sodium in the selected ar- area
The distribution of silicon, aluminum, iron, calcium, and sodium in the selected
was analyzed
ea was analyzedusing area
using areascan mode.
scan mode.Six
Sixareas
areaswere
wereselected
selectedas
asshown
shown in
in Figure
Figure 10 for area
10 for
scan analyze. The results are shown in Table 3.
area scan analyze. The results are shown in Table 3.

Figure 10.
Figure 10. Area
Areascan
scanofofaaparticle
particleofof
CAT-CD.
CAT-CD.

Table 3.
Table 3. SEM-EDS
SEM-EDS analysis
analysisresults
resultsofofthe element
the contents
element on on
contents CAT-CD.
CAT-CD.
Location Scanning Area Si Al Fe Na Ca
Location Scanning Area Si Al Fe Na Ca
Area scan 1 (%) 34.16 48.89 15.18 1.77 0
Raised area Area scan 1 (%)
Raised Area scan 2 34.16
(%) 48.89
33.55 15.1818.29
43.62 1.77
4.53 0 0
onarea
surface Area scan 2 (%) 33.55 43.62 18.29 4.53 0
on surface Area scan 3 (%) 38.91 43.42 16.36 1.31 0
Area scan 3 (%) 38.91 43.42 16.36 1.31 0
Area scan 4 (%) 45.94 40.94 11.86 1.26 0
Valley area Area scan 4 (%) 45.94 40.94 11.86 1.26 0
Valley area Area scan 5 (%) 43.91 42.69 11.59 1.81 0
on surface Area scan 5 (%) 43.91 42.69 11.59 1.81 0
on surface Area scan 6 43.60
Area scan 6 (%)
(%) 43.60
41.35
41.35
11.41
11.41 3.64
3.64
0 0
The content of aluminum was higher than that of silicon in the raised area, and it
was the
Theopposite
content in ofthe valley area.
aluminum wasThe initialthan
higher melting
thattemperatures
of silicon in of thethe eutectic
raised that
area, and it
consisted of Na 2O, Al2O3, and SiO2 were higher than 770 °C. It was difficult for Fe2O3 to
was the opposite in the valley area. The initial melting temperatures of the eutectic that
combine with
consisted of Na Al2O,
O3Al
, NaO2O,, and
and SiO
SiO2 were
to form low-melting-temperature
higher phases in
than 770 ◦ C. It was difficult forareas
Fe2 O3 to
2 2 3 2
that were rich in Al 2O3; therefore, the structure of this area was preserved.
combine with Al2 O3 , Na2 O, and SiO2 to form low-melting-temperature phases in areas
However, in the areas with a relatively high content of SiO2, it was easy for the
that were rich in Al2 O3 ; therefore, the structure of this area was preserved.
Fe2O3 to form a eutectic with Na2O and SiO2, whose initial melting temperatures were
However, in the areas with a relatively high content of SiO2 , it was easy for the Fe2 O3
lower than 500 °C [14]. During the regeneration process, the SiO2 enrichment zone fur-
to form a eutectic with Na2 O and SiO2 , whose initial melting temperatures were lower
ther intensified the melting and sintering, resulting in a regional collapse and local de-
than 500 ◦ C [14]. During the regeneration process, the SiO2 enrichment zone further
pression. The retention of the surface structure in the Al2O3-rich areas and the collapse of
intensified the melting and sintering, resulting in a regional collapse and local depression.
the surface structure in the SiO2-rich areas resulted in a nodule morphology on the
The
highlyretention of the surface
iron-contaminated structure
catalyst surface.in the Al2 O3 -rich areas and the collapse of the
surface structure in the SiO 2
The formation of iron nodules was -rich areas
theresulted in a nodule
main reason morphology
for the various on the of
performances highly
iron-contaminated catalyst surface.
the catalysts contaminated by iron. During the process of the Mitchell impregnation, the
The formation solution
iron-contaminated of iron nodules was internal
filled the the mainchannels
reason forofthe various
the catalystperformances
to form anof the
catalysts contaminated by iron. During the process of the Mitchell
equal-volume state. After the high-temperature calcining treatment, the iron was evenly impregnation, the iron-
contaminated solution filled the internal channels of the catalyst to
distributed in the catalyst and there was no condition that allowed for the formation of form an equal-volume
state. After the
iron nodules high-temperature
[37]. Therefore, iron nodulescalcining treatment,
could the irononwas
not be observed the evenly distributed
catalyst’s surface in
the
andcatalyst and there
the deactivated was nocould
catalyst condition that allowed
not simulate for the formation
an iron-contaminated of iron nodules
equilibrium cata- [37].
Therefore,
lyst. iron nodules could not be observed on the catalyst’s surface and the deactivated
catalyst could not simulate an iron-contaminated equilibrium catalyst.
Iron deposition on the surface of the cyclic-deactivated catalyst was realized using
long-time repeated cracking–regeneration cycles. The iron deposition in a single cycle was
Catalysts 2021, 11, 104 10 of 14

very low. After the cracking step, iron oxide combined with silicon oxide, aluminum oxide,
and alkali metal oxides to form low-melting-temperature phases before it diffused inside the
catalyst. The formation conditions of iron nodules in industrial plants could be simulated
by using the cyclic deactivation method. The CD method is an effective and accurate
method for simulating iron contamination in a laboratory.

3. Materials and Methods

3.1. Materials
In this research, a commercial catalyst (CAT-BASE) without iron contamination,
which was used to simulate iron deactivation, was collected from the industrial plant
of the Dalian West Pacific Petrochemical Co., Ltd. (WEPEC, Dalian, China). In order to
simulate the actual state of an iron-contaminated equilibrium catalyst in an FCC unit,
the target amount of iron deposition and appropriate conditions were selected to carry out
Mitchell impregnation and the cyclic deactivation method. The equilibrium catalyst (E-cat)
with a high level of iron contamination was also collected from WEPEC.

3.2. Methods
3.2.1. Mitchell Impregnation
Iron chloride and iron naphthenate were used as contamination species to carry out
this method. Iron chloride was added to deionized water (Fe on solution = 14,400 µg/g)
and iron naphthenate was added to a toluene solution (Fe on solution = 12,500 µg/g).
The catalysts (150 g) were impregnated with 120 g of iron chloride solution and 150 g of
iron naphthenate solution, respectively, dried for 2 h at 150 ◦ C, and then calcined for 6 h
at 600 ◦ C before use. During the drying process, the catalyst was stirred every 20 min
to ensure that it was evenly heated and would not agglomerate. The catalysts that were
impregnated with iron chloride and iron naphthenate were labeled as CAT-MM1 and
CAT-MM2, respectively.

3.2.2. Cyclic Deactivation

The deactivated catalyst was treated using 200 cycles of cracking, stripping, regen-
eration, and cooling in the MCD (Multi Cyclic Deactivation) unit produced by 360◦ KAS
Company (Amsterdam, The Netherlands). The cracking step was performed at a reaction
temperature of 530 ◦ C and the regeneration step was performed at 780 ◦ C. During the
cracking step, 10 g of VGO spiked with iron naphthenate (Fe on VGO = 860 µg/g) was
pumped into the quartz reactor to react with 150 g of the catalyst at the reaction temperature.
Finally, the aged catalyst was marked as CAT-CD.

3.2.3. Catalyst Characterization

The metal content was determined using a ZSX PrimusIIX-ray fluorescence instrument
produced by the Japan Science and Technology Company (Tokyo, Japan). The X-ray tube
voltage was 50 kV, the tube current was 50 mA, and the diaphragm aperture was 20 mm.
The catalyst phase structure was determined using a D/max-3C X-ray diffractometer
produced by the Japan Science and Technology Company (Tokyo) with the following
settings: Cu target, Kα radiation, tube voltage: 40 kV, tube current: 20 mA, scanning range:
5 to 50◦ , scanning speed: 4◦ /min.
NH3 -TPD was performed on a Micromeritics AUTOCHEM II 2920 produced by
the Micromeritics Instruments Corporation (Norcross, GA, USA). 0.1 g of sample was
purified with He gas flow at room temperature for 1 h. NH3 adsorption was performed by
adjusting temperature with two-stage method. The desorption curve was obtained after
the adsorption is stable, and the temperature was raised to 500 ◦ C within 50 min.
The specific surface area and pore volume of the catalyst were measured using a
Micromeritics ASAP 3000 automatic physical adsorption instrument produced by the Mi-
cromeritics Instruments Corporation (Norcross, GA, USA). The surface area was deduced
from the adsorption isotherms using the BET (Brunauer, Emmett and Teller) equation.
Catalysts 2021, 11, 104 11 of 14

The catalyst surface morphology was observed using an ULTRA PLUS thermal field
emission scanning electron microscope produced by Zeiss Optical Instruments (Oberkochen,
Germany). Secondary electron resolution: 1.0 nm (15 kV) and 1.9 nm (1 kV); electron gun:
LaB6 thermal field emission electron gun; acceleration voltage: 0.1~30 kV, magnification:
×12~×10000.
The iron distribution on the catalysts’ cross-section was analyzed using line scan mode
during SEM in combination with an X-ray energy dispersive spectroscope produced by
Oxford Instruments (Abingdon, UK). Resolution: better than 127 eV; The MnKα peak-to-
back ratio: 20000:1; stability: 1000~100,000 cps, spectral peak drift <1 eV, resolution change
<1 eV. All four catalysts were measured from the center to the edge. In order to ensure the
accuracy of the measurement, a total of 350 points were measured by using a spot analysis
technique over a length of 35 µm to make up the results of the line scan.

3.2.4. Catalyst Cracking Activity and Selectivity

The cracking performance of different catalysts was tested in an Advanced Crack-
ing Evaluation unit developed by the KTI Technology Company (Huston, TX, USA).
The flowchart of the ACE unit is shown in Figure 11. The reaction conditions were the
following: The amount of catalyst added was 9 g. The reaction and regeneration tem-
peratures were 530 ◦ C and 715 ◦ C, respectively. The variation of the catalyst-to-oil ratio
(C/O = 3.5, 4.0, 5.0, 6.0, 7.0) was performed by varying the catalyst amount. Gaseous
products were analyzed using an GC-3000 on-line chromatograph produced by INFICON
Company (East Syracuse, NY, USA) according to the UOP method 539. GC-3000 uses four
chromatographic modules for detection, the work temperature is 0 ◦ C~50 ◦ C. The carrier
gases used are helium, hydrogen, nitrogen and argon. The simulated distillation of liquid
products was carried out using a 7890B chromatograph produced by Agilent Technolo-
gies Inc. (Santa Clara, CA, USA) according to the SH/T 0558 procedure. The working
environmental temperature of Agilent 7890B is 15 ◦ C~35. Retention time repeatability
<0.0008 min. Carrier and makeup gas settings selectable for helium, hydrogen, nitrogen and
argon/methane. Coke deposited on the catalyst was quantified using a CO2 analyzer pro-
duced by Servomex Group Co., Ltd (Sussex, UK). The detection range of CO2 analyzer is
0~20%. When the detection value is lower than 0.4%, the regeneration step is considered to
be completed. Conversion and yields of dry gas (H2 + C1 + C2), LPG (Liquefied Petroleum
Gas) (C3 + C4), gasoline (C5 < bp < 221 ◦ C), LCO(Light Cyclic Oil) (221 ◦ C < bp12<of343
Catalysts 2021, 11, x FOR PEER REVIEW
◦
14 C),
HCO (Heavy Cyclic Oil)(bp > 343 ◦ C), and coke were calculated.

Figure 11.
Figure 11. The
The flowchart
flowchartofofthe
theAdvanced
AdvancedCracking
CrackingEvaluation (ACE)
Evaluation unit.unit.
(ACE)

4. Conclusions
Iron nodules are the main factor that leads to the poor performance of
iron-contaminated catalysts. The retention of the surface structure in the alumina-rich
areas and the collapse of the surface structure in the silica-rich areas resulted in a con-
tinuous nodule morphology on a highly iron-contaminated equilibrium catalyst.
Iron was evenly distributed in the catalyst prepared using the Mitchell impregna-
tion method, and no obvious iron nodules or boundary were found on the surface of the
Catalysts 2021, 11, 104 12 of 14

4. Conclusions
Iron nodules are the main factor that leads to the poor performance of iron-contaminated
catalysts. The retention of the surface structure in the alumina-rich areas and the collapse
of the surface structure in the silica-rich areas resulted in a continuous nodule morphology
on a highly iron-contaminated equilibrium catalyst.
Iron was evenly distributed in the catalyst prepared using the Mitchell impregnation
method, and no obvious iron nodules or boundary were found on the surface of the catalyst.
The surface area was less affected by iron contamination. Iron deposited on the catalyst
resulted in a small reduction in the conversion and it had little effect on the bottoms
selectivity, but it led to a strong dehydrogenation capacity. The catalytic performance of the
impregnated catalyst was quite different from that of the iron-contaminated equilibrium
catalyst.
The iron was mainly present in the depth of 1~5 µm from the edge on the catalyst
prepared using the cycle deactivation method. As iron nodules formed on the surface of
the cyclic-deactivated catalyst, pore closing limited the feed molecules’ diffusion inside
the particles and the products’ diffusion outward, resulting in a significant decrease in
conversion, extremely poor bottoms selectivity, and a small increase in dehydrogenation
capacity.
We established a laboratory simulation method that can accurately simulate the
industrial iron-contaminated equilibrium catalyst. By using this method, we can recognize
the real mechanism of iron contamination, understand the main factors that affect the
physical and chemical properties of an iron-contaminated catalyst, and predict the actual
reaction performance and product distribution of a catalyst prepared in a laboratory.
This can effectively reduce the time and economic costs caused by the failure of laboratory
simulations.
In our next research work, we can rely on the accurate CD method to develop iron
tolerance technology. On the other hand, we can also investigate the combined effect of iron
and other heavy metals, such as nickel and vanadium, on the catalyst to further improve
the iron deactivation simulation method.

Author Contributions: Y.L. and T.L. conceived and designed the experiment; Y.L. performed the
experiments and analyzed the data. All the authors listed have made contributions to drafting and
writing the manuscript. All authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by the Ministry of Science and Technology Management of
PetroChina (2019D-5006-20).
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Data available in a publicly accessible repository.
Acknowledgments: We acknowledge the financial support from the Development of New Cata-
lysts with Low Emission and High Gasoline Yield project, the Ministry of Science and Technology
Management of PetroChina (2019D-5006-20). We thank L. Zhang for kind assistance in SEM-EDS
experiments, and X.Q. Sun for her kind helps in NH3 -TPD analysis.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Rainer, D.R.; Rautiainen, E.; Imhof, P. Novel lab-scale deactivation method for FCC catalyst: Inducing realistic accessibility
responses to iron poisoning. Appl. Catal. A Gen. 2003, 249, 69–80. [CrossRef]
2. Vincz, C.; Rath, R.; Smith, G.M.; Yilmaz, B.; Mcguire, R. Dendritic nickel porphyrin for mimicking deposition of contaminant
nickel on FCC catalysts. Appl. Catal. A Gen. 2015, 495, 39–44. [CrossRef]
3. Mathieu, Y.; Corma, A.; EchardI, M.; Bories, M. Single and combined fluidized catalytic cracking (FCC) catalyst deactivation by
iron and calcium metal–organic contaminants. Appl. Catal. A Gen. 2014, 469, 451–465. [CrossRef]
4. Mehla, S.; Kukade, S.; Kumar, P.; Rao, P.V.C.; Sriganesh, G.; Ravishankar, R. Fine tuning h-transfer and β-scission reactions in
VGO FCC using metal promoted dual functional ZSM-5. Fuel 2019, 242, 487–495. [CrossRef]
Catalysts 2021, 11, 104 13 of 14

5. Tangstad, E.; Andersen, A.; Myhrvold, E.M.; Myrstad, T. Catalytic behaviour of nickel and iron metal contaminants of an FCC
catalyst after oxidative and reductive thermal treatments. Appl. Catal. A Gen. 2008, 346, 194–199. [CrossRef]
6. Zhang, C.C.; Gao, X.; Yilmaz, B. Development of FTIR Spectroscopy Methodology for Characterization of Boron Species in FCC
Catalysts. Catalysts 2020, 10, 1327. [CrossRef]
7. Souza, N.L.A.; Tkach, I.; Morgado, E., Jr.; Krambrock, K. Vanadium poisoning of FCC catalysts: A quantitative analysis of
impregnated and real equilibrium catalysts. Appl. Catal. A. Gen. 2018, 560, 206–214. [CrossRef]
8. Long, J.; Zhu, Y.X.; Liu, Y.J.; Da, Z.J.; Zhou, H. Effects of vanadium oxidation number on desulfurization performance of FCC
catalyst. Appl. Catal. A Genera 2005, 282, 295–301. [CrossRef]
9. Whitcombe, J.M.; Agranovski, I.E.; Braddock, R.D. Identification of Metal Contaminants on FCC Catalyst. Part. Part. Syst. Charact.
2005, 22, 268–275. [CrossRef]
10. Tian, H.; Huang, C.; Fan, Z. Metals on a novel USY zeolite after hydrothermal aging. Stud. Surf. Sci. Catal. 2001, 139, S0167–S2991.
11. Sandoval-Díaz, L.; Ruíz-Cardona, Y.; Trujillo, C. Amorphization of USY zeolite induced by sodium chloride and high temperature
steaming. Microporous Mesoporous Mater. 2016, 224, 168–175. [CrossRef]
12. Liu, Z.; Zhang, Z.; Liu, P.; Zhai, J.; Yang, C. Iron contamination mechanism and reaction performance research on FCC catalyst.
J. Nanotechnol. 2015, 2015. [CrossRef]
13. Whitcombe, J.M.; Agranovski, I.E.; Braddock, R.D. Impact of metal ridging on the fluidization characteristics of FCC catalyst.
Chem. Eng. Technol. 2002, 25, 981–987. [CrossRef]
14. Yaluris, G.; Cheng, W.C.; Peters, M.; Mcdowell, L.T.; Hunt, L. Mechanism of fluid cracking catalysts deactivation by Fe. Stud. Surf.
Sci. Catal. 2004, 149, 139–163. [CrossRef]
15. Buurmans, I.L.; Soulimani, F.; Ruiz-Martínez, J.; van der Bij, H.E.; Weckhuysen, B.M. Structure and acidity of individual fluid
catalytic cracking catalyst particles studied by synchrotron-based infrared micro-spectroscopy. Microporous Mesoporous Mater.
2013, 166, 86–92. [CrossRef]
16. Nguyen, T.H.; Ngo, P.T.; Tran, T.V.; Nguyen, S.; Vu, D.M.; Ha, Q.L.M.; Dao, X.T.T.; Dang, T.T. Effect of hydrothermal conditions on
the catalytic deactivation of a fluid cracking catalyst. React. Kinet. Mech. Catal. 2013, 109, 563–574. [CrossRef]
17. Etim, U.J.; Xu, B.; Bai, P.; Ullah, R.; Yan, Z. Role of nickel on vanadium poisoned FCC catalyst: A study of physiochemical
properties. J. Energy Chem. 2016, 25, 667–676. [CrossRef]
18. Etim, U.J.; Xu, B.; Ullah, R.; Yan, Z. Effect of vanadium contamination on the framework and micropore structure of ultra stable
Y-zeolite. J. Colloid Interface Sci. 2016, 463, 188–198. [CrossRef]
19. Chiranjeevi, T.; Gokak, D.T.; Ravikumar, V.; Viswanathan, P.S. Development of new deactivation method for simulation of fluid
catalytic cracking equilibrium catalyst. J. Chem. Sci. 2014, 126, 353–360. [CrossRef]
20. Wallenstein, D.; Farmer, D.; Knoell, J.; Fougret, C.M.; Brandt, S. Progress in the deactivation of metals contaminated FCC catalysts
by a novel catalyst metallation method. Appl. Catal. A. Gen. 2013, 462, 91–99. [CrossRef]
21. Psarras, A.C.; Iliopoulou, E.F.; Nalbandian, L.; Lappas, A.A.; Pouwels, C. Study of the accessibility effect on the irreversible
deactivation of FCC catalysts from contaminant feed metals. Catal. Today 2007, 127, 44–53. [CrossRef]
22. Almas, Q.; Naeem, M.A.; Baldanza, M.A.S.; Solomon, J.; Kenvin, J.C.; Christoph, R.; Müller, C.R.; da Silva Teixeira, V.; Jones, C.W.;
Sievers, C. Transformations of FCC catalysts and carbonaceous deposits during repeated reaction-regeneration cycles. Catal.
Technol. 2019, 9, 6977–6992. [CrossRef]
23. Rainer, D.R.; Rautiainen, E.; Nelissen, B.; Imhof, P.; Vadovic, C. Simulating iron-induced FCC accessibility losses in lab-scale
deactivation. Stud. Surf. Sci. Catal. 2004, 149, 165–176. [CrossRef]
24. Brandt, S.; Knöll, J.; Jiang, H.; Saar, J.; Fougret, C. The role of magnetic susceptibility in detecting iron poisoning in FCC
equilibrium catalyst samples and its combination with other macroscopic bulk analysis techniques. Ind. Eng. Chem. Res. 2019, 58,
20528–20535. [CrossRef]
25. Zhu, Y.X.; Du, Q.S.; Lin, W.; Tang, L.W.; Jun, L. Chapter 13 studies of iron effects on fcc catalysts. Stud. Surf. Sci. Catal. 2007, 166,
201–212. [CrossRef]
26. Jiang, H.; Livi, K.J.; Kundu, S.; Cheng, W.C. Characterization of iron contamination on equilibrium fluid catalytic cracking catalyst
particles. J. Catal. 2018, 361, 126–134. [CrossRef]
27. Souza, N.L.; Paniago, R.; Ardisson, J.D.; Morgado, E., Jr.; Krambrock, K. Iron contamination of FCC catalysts: Quantification of
different crystalline phases and valence states. Appl. Catal. A Gen. 2019, 569, 57–65. [CrossRef]
28. Hochheiser, T.; Tang, Y.; Allahverdi, M.; Graaf, B.D. FCC Additive Improves Residue Processing Economics with High Iron Feeds.
In Proceedings of the AFPM Annual Meeting, Orlando, FL, USA, 23–25 March 2014.
29. Meirer, F.; Morris, D.T.; Kalirai, S.; Liu, Y.; Andrews, J.C.; Weckhuysen, B.M. Mapping metals incorporation of a whole single
catalyst particle using element specific X-ray nanotomography. J. Am. Chem. Soc. 2015, 137, 102–105. [CrossRef] [PubMed]
30. Pan, S.S.; Lin, L.T.X.; Komvokis, V.; Spann, A.; Clough, M.; Yilmaz, B. Nanomaterials Fueling the World. In Nanomaterials for
Sustainable Energy; American Chemical Society: Washington, DC, USA, 2015; Volume 1213, pp. 3–18. [CrossRef]
31. Cristiano-Torres, D.V.; Osorio-Pérez, Y.; Palomeque-Forero, L.A.; Sandoval-Díaz, L.E.; Trujillo, C.A. The action of vanadium over
Y zeolite in oxidant and dry atmosphere. Appl. Catal. A. Gen. 2008, 346, 104–111. [CrossRef]
32. Wise, A.M.; Weker, J.N.; Kalirai, S.; Farmand, M.; Shapiro, D.A.; Meirer, F.; Weckhuysen, B.M. Nanoscale chemical imaging of
an individual catalyst particle with soft x-ray ptychography. Acs Catal. 2016, 6, 2178–2181. [CrossRef] [PubMed]
Catalysts 2021, 11, 104 14 of 14

33. Ihli, J.; Ferreira Sanchez, D.; Rosh, J.; Cuartero, V.; Mathon, O.; Krumeich, F.; Borca, C.; Huthwelker, T.; Cheng, W.-C.; Shu, Y.Y.; et al.
Localization and speciation of iron impurities within a fluid catalytic cracking catalyst. Angew. Chem. 2017, 129, 14219–14223.
[CrossRef]
34. Senter, M.; Mastry, M.C.; Mannion, A.M.; McGuire, R.; Houtz, D.; Yilmaz, B. Quantitative visual characterization of contaminant
metals and their mobility in fluid catalytic cracking catalysts. Catalysts 2019, 9, 831. [CrossRef]
35. Alkhlel, A.; de Lasa, H. Catalyst/Feedstock Ratio Effect on FCC Using Different Catalysts Samples. Catalysts 2019, 9, 542.
[CrossRef]
36. Chen, J.W.; Xu, Y.H. Fluid Catalytic Cracking Process and Engineering; Sinopec Press: Beijing, China, 2018; pp. 392–395.
37. Bai, P.; Etim, U.J.; Yan, Z.; Mintova, S.; Zhang, Z.; Zhong, Z.; Gao, X. Fluid catalytic cracking technology: Current status and
recent discoveries on catalyst contamination. Catal. Rev. 2019, 61, 333–405. [CrossRef]