Applied Catalysis A: General 489 (2015) 123–130

Contents lists available at ScienceDirect

Applied Catalysis A: General

journal homepage: www.elsevier.com/locate/apcata

Formation of aromatics in heavy gasoline and light LCO ends in FCC

Richard Pujro, Marisa Falco, Ulises Sedran ∗
Instituto de Investigaciones en Catálisis y Petroquímica INCAPE (FIQ, UNL—CONICET), Santiago del Estero 2654, 3000 Santa Fe, Argentina

a r t i c l e i n f o a b s t r a c t

Article history: Decalin, a naphthenic bicyclic compound, was reacted over a commercial FCC catalyst in its fresh, equi-
Received 9 May 2014 librium and three dealuminated forms in order to analyze the formation of aromatic hydrocarbons in the
Received in revised form gasoline and LCO cuts, heavy and light ends, respectively. A batch, fluidized bed CREC Riser simulator
17 September 2014
laboratory reactor was used at 450 ◦ C with short contact times from 1 to 8 s. The dealuminated catalysts
Accepted 19 September 2014
showed an activity intermediate between those of the fresh and equilibrium catalysts, which can be
Available online 22 October 2014
correlated to zeolite content and acidity. Also those properties, at least in the range of values observed
in this work, seem to be the most important characteristic in controlling the selectivity of the set of
Keywords:
Aromatics
reactions. A reaction network was described, where hydrogen transfer reactions are considered to be the
Decalin most important to form aromatics, a fact which is essentially unavoidable when a bicyclic naphthenic
LCO is converted on these catalysts. Aromatic belonged mainly to the gasoline boiling point range and, to a
FCC lower extent, to LCO.
Y zeolite © 2014 Elsevier B.V. All rights reserved.

1. Introduction LCO yield and quality by either modifying operative conditions

[7] or using new catalysts which would produce less aromatics
The process of catalytic cracking of hydrocarbons (FCC) is one of [8,9].
the most important and profitable to produce different fuels such FCC catalyst particles are composed of Y zeolite supported
as gasoline and middle distillates and petrochemical raw materials on a matrix (active or inactive) together with various additives,
in current refineries [1]. The increasing demand for diesel fuel has fillers and binders. At present, the catalysts are custom made,
induced changes in the operating schemes of the FCC units in order according to the particular conditions in a refinery (feedstock, tech-
to maximize the yield of light cycle oil (LCO, boiling range usually nology, production scheme) [10]. Due to the cyclic process between
between 180 and 350 ◦ C), which can be blended together with other the reducing atmosphere of the reactor at about 500 ◦ C and the
cuts from various processes to produce diesel fuels. oxidizing, steam containing atmosphere at about 700 ◦ C in the
There exists a particular interest in the composition of LCO, the regenerator, the catalyst changes drastically from the fresh state to
fuel quality of which depends on the content of paraffins. Thus, the the so-called equilibrium form, due to a strong dealumination pro-
main problem of LCO as a contributor to diesel fuels is the high con- cess of the zeolite which reduces its unit cell size, thus impacting
centration of aromatic compounds, typically between 50 and 70% on chemical composition and acidic properties; moreover, physical
[2], which induces a very low cetane index of about 24–28, while properties also change significantly [11].
commercial fuels indexes are higher than 50 [3,4]. Moreover, sul- However, if specific catalysts are to be developed to decrease
phur compounds are present at high level [5]. Overall, these issues the concentration of aromatics in the LCO cut, it is necessary
depend on the crude source, the catalyst and the process conditions. to know the mechanisms of formation of these compounds in
Even though technological options exist which can lead to the cut, and the impact which various factors such as mod-
upgrading the quality of LCO, such as partial hydrogenation of aro- ifications in the structure, acidity and zeolite content of the
matics and ring opening of naphthenic compounds [6], these may catalysts, among others, may induce. The formation of aromatic
be expensive. However, the versatility of the FCC in processing s in the gasoline–LCO border could be evaluated through the con-
different types of feedstocks offers opportunities to improve version of naphthenic compounds [12,13], which are present at
significant concentrations in that boiling range and play an impor-
tant role in some of the reactions taking place in the complex FCC
∗ Corresponding author. Tel.: +54 342 452 8062; fax: +54 342 453 1068.
network.
E-mail addresses: rpujro@fiq.unl.edu.ar (R. Pujro), mfalco@fiq.unl.edu.ar
It is the objective of this work to analyze the influence of
(M. Falco), usedran@fiq.unl.edu.ar (U. Sedran). dealumination in FCC catalysts on the formation of aromatic

http://dx.doi.org/10.1016/j.apcata.2014.09.051
0926-860X/© 2014 Elsevier B.V. All rights reserved.
124 R. Pujro et al. / Applied Catalysis A: General 489 (2015) 123–130

Table 1
Properties of the catalysts used.

Catalyst Zeolite RE content Specific surface Acidity (␮mol Py g−1 )

(wt.%) area (m2 g−1 )

UCS (nm) Load(wt.%) Si/Al Matrix Total Bronsted (1545 cm−1 ) Lewis (1450–1460 cm−1 )
◦ ◦ ◦
150 C 300 C 400 C 150 ◦ C 300 ◦ C 400 ◦ C

F-cat 2.456 22.00 3.50 0.94 92 243 128.5 102.1 76.3 118.1 42.4 36.9
Cat-1 2.437 18.07 8.26 0.80 72 191 14.6 8.9 8.3 16.7 12.9 11.6
Cat-3 2.434 17.71 10.10 0.94 64 181 12.0 11.6 8.9 23.4 13.4 11.2
Cat-5 2.432 17.04 11.80 1.17 62 173 13.3 10.5 5.0 25.2 17.7 9.8
E-cat 2.430 9.83 14.20 0.70 102 162 5.0 5.1 5.0 13.6 11.0 10.9

hydrocarbons in the heavy end of gasoline and light end of LCO in catalyst in the reactor, were performed at the same temperature
order to generate more information to assist in the formulation of and the longest residence time.
new, more selective FCC catalysts. Decalin, a bicyclic naphthenic, The reaction products were analyzed by on-line standard capil-
was used as a test reactant in a CREC Riser Simulator laboratory lary gas chromatography, using a 30 m long, 250 mm diameter and
reactor under very short contact times. 0.25 mm film thickness, non-polar, dimethylpolysiloxane column.
Product identification was performed with the help of standards
and GC–MS analysis. The coke content was assessed by means
2. Experimental of a method with temperature-programed oxidation and further
methanation of the carbon oxides over a Ni catalyst, quantified with
The base catalyst was a commercial FCC catalyst designed to the help of a FID detector [20]. Mass balances (recoveries) closed to
maximize the LCO yield in the process; both its fresh (F-cat) and more than 94% in all the cases.
equilibrium (E-cat) forms were used. The fresh catalyst was sub-
jected to hydrothermal treatment with steam at high temperature 3. Results and discussion
in order to induce changes in its physical properties and dealumi-
nate the Y zeolite component. Dealumination was performed on a 3.1. Catalyst properties
fluidized bed of catalyst with steam in a tubular quartz reactor at
788 ◦ C, and three different severities were used: 1, 3 and 5 h. The Table 1 shows the most important catalyst properties. It can
catalyst samples were identified according to the steaming time as be seen that the majority of the properties of the fresh catalyst
Cat-1, Cat-3 and Cat-5, respectively. were modified after dealumination, as long as the Si/Al rela-
The assessment of specific surface areas (BET and t-plot meth- tionship changed, increasing steadily with steaming time. As a
ods) was performed with N2 isotherms at 77 K and the zeolite consequence, the UCS of the zeolite crystals decreased due to
content (wt.%) was determined following the method by Johnson the higher Si content in the zeolite [21]. In this way, the sample
[14]. The zeolite unit cell sizes (UCS) were determined with the which was dealuminated during 5 h showed an UCS close to that
ASTM D-3942-91 X-ray diffraction technique, using a Shimadzu of the equilibrium catalyst. The hydrothermal treatment causes
XD-1 equipment. The content of rare earths was determined by the loss of aluminum atoms from the crystalline zeolite network
the inductively coupled plasma (ICP) method, using a Perkin Elmer due to the high temperature hydrolysis of the Si O Al bonds [22].
Optical Emission Spectrometer OPTIMA 2100 DV. These aluminum atoms migrate to positions out of the zeolite
The nature, amount and strength of acidic sites in the different structure and are usually designed as extraframework aluminum
zeolites were determined by means of the FTIR analysis of adsorbed [23].
pyridine (Merck, 99.5%) as a probe molecule in a Shimadzu FTIR In this type of catalysts, micropores are contributed by the
Prestige-21 equipment. Approximately 100 mg of the zeolite were zeolite component and mesopores by the matrix and binder com-
pressed at 1 t cm−2 in order to produce self supporting wafers with ponents [24,25], but vacancies in the zeolite framework due to
density 440 g m−2 , which were then placed into a cell with CaF2 dealumination can produce breaks or even partial collapsing of
windows. Samples were initially degassed at 450 ◦ C during 2 h and a the crystalline material [22]. Matharu et al. [26], who compared Y
background spectrum was collected at room temperature. Pyridine zeolite dealumination by hydrothermal treatment and ammonium
adsorption was performed at room temperature and after succes- hexafluorosilicate extraction, observed both micro- and meso-
sive desorptions at 150, 300 and 400 ◦ C, spectra were recorded porosity in the hydrothermally treated samples, the mesopore
at room temperature with a resolution of 4 cm−1 at pressure of volume increasing with the severity of the treatment. These new
10−4 Torr. The amounts of Brönsted and Lewis acid sites were cal- pores after dealumination, together with similar consequences on
culated from the integrated absorbance of the bands at 1545 and the matrix, manifest in the increasing average mesopore size (for
1450–1460 cm−1 , respectively, by means of the integrated molar example, 8.5 nm in F-cat to 10.7 nm in Cat-3) and the decreasing
extinction coefficients, which are considered independent from the total specific surface area of the catalysts (Table 1), as also observed
catalyst and site strength [15,16]. by Salman et al. [27].
The experiments of decalin conversion (Sigma-Aldrich C10 H18 , The partial collapse of the crystalline material is also reflected
≥99%, cis- and trans-decahydronaphthalene isomers mixture), boil- in the loss of zeolitic material, with the zeolite content decreasing
ing point 189–191 ◦ C, were performed in a CREC Riser Simulator steadily from the fresh to the equilibrium catalyst through the
reactor, which is a batch, fluidized bed laboratory reactor which steamed samples, and also in the loss of acidity. As it will be dis-
closely mimics the conditions of the commercial FCC process [17]. cussed later, particular attention will be given to Brönsted acidity in
The unit has been described comprehensively elsewhere [18,19]. view of its mechanistic importance. In effect, total acidity (desorp-
Reaction times in the experiments were from 1 to 8 s, temperature tion at 150 ◦ C) decreases significantly from the fresh catalyst to the
was 450 ◦ C and catalyst to oil relationship was 1.12. The mass of dealuminated and equilibrium samples. It is to be noted, however,
catalyst was 0.2 g and the volume of decalin injected was 0.2 ml that the amounts of acid sites in the dealuminated catalysts do not
in all the cases. Experiments of purely thermal cracking, with no differ significantly.
 R. Pujro et al. / Applied Catalysis A: General 489 (2015) 123–130 125

Table 2
Product yields (wt.%) in decalin reaction (average conversion 18%).

Products F-cat Cat-1 Cat-3 Cat-5 E-cat

C1 –C2 0.14 0.05 0.05 0.05 0.03

C3 1.21 0.46 0.50 0.40 0.65
C4 olefins 0.11 0.08 0.11 0.08 0.15
C4 paraffins 3.59 1.58 1.63 1.41 0.85
C5 olefins 0.03 0.02 0.03 0.02 0.06
Cyclopentene 0.01 0.01 0.02 0.01 0.03
C5 paraffins 1.45 0.71 0.75 0.61 0.66
Cyclopentane 0.14 0.11 0.12 0.10 0.06
C6 olefins 0.01 0.01 0.01 0.01 0.03
C6 paraffins 0.73 0.47 0.50 0.43 0.21
C6 naphthenics 1.71 1.31 1.35 1.29 0.81
Benzene 0.24 0.09 0.08 0.06 0.07
Methylcyclohexene 0.02 0.02 0.02 0.02 0.01
C7 paraffins 0.14 0.12 0.12 0.11 0.06
C7 naphthenics 0.80 0.77 0.80 0.77 0.50
Toluene 0.84 0.31 0.30 0.24 0.12
C8 paraffins 0.12 0.12 0.12 0.13 0.08
C8 naphthenics 0.28 0.28 0.28 0.30 0.28
C8 aromatics 0.99 0.40 0.41 0.33 0.15
C9 paraffins 0.06 0.09 0.07 0.08 0.06
C9 naphthenics 0.05 0.07 0.06 0.07 0.08
C9 aromatics 0.85 0.85 0.84 1.05 0.74
C9 naphthenic-aromatics 0.34 0.68 0.69 0.85 0.95
trans-Decalin 41.03 43.62 43.64 43.55 46.52
cis-Decalin 40.04 40.30 40.25 38.30 34.60
Fig. 1. Conversion of decalin as a function of reaction time. Symbols: () F-cat, () Isobutylcyclohexane 0.15 0.23 0.22 0.30 0.32
Cat-1, () Cat-3, () Cat-5, (䊉) E-cat. 3-Butylcyclohexene 0.08 0.13 0.12 0.17 0.19
C10 aromatics 1.30 2.37 2.46 2.61 4.68
Naphthalene 0.49 0.78 0.62 0.72 0.69
C10 naphthenic-aromatics 0.43 0.27 0.49 0.46 0.74
Two periods can be observed during dealumination, UCS, zeolite Tetralin 0.28 0.44 0.49 0.44 0.60
content and amount of acid sites decreasing much faster during the 1,1 -Bicyclopentyl 0.10 0.21 0.19 0.28 0.35
first hour and then showing a much slower variation, in consistency Trimethylbicycloheptane 0.35 0.59 0.56 0.75 0.63
with previous observations [22]. Dimethylbicyclooctane 0.93 1.45 1.04 2.60 2.40
C11 aromatics 0.03 0.02 0.02 0.11 0.03
Methylnaphthalene 0.09 0.05 0.08 0.07 0.06
3.2. Catalyst activity C11 naphthenic-aromatics 0.06 0.04 0.05 0.06 0.06
C12 + aromatics 0.04 0.01 0.02 0.02 0.02
C12 + naphthenic-aromatics 0.01 0.00 0.01 0.01 0.01
The experiments of purely thermal cracking at 450 ◦ C at the
longest contact time showed that conversion was lower than 2%,
which was considered negligible as compared to the conversion true for catalysts Cat-1 and Cat-3, while Cat-5 has less zeolite and
observed in the catalytic experiments. consequently is somewhat less active.
Decalin was indeed a mixture of the cis- (52%) and trans-(48%) The size of the decalin molecules has been calculated to be
configurations. The cis-isomer was more reactive, as also noted slightly smaller than that of the micropores in the zeolite [32].
previously over various FCC catalysts [18]. Different authors also However, a number of publications showed that the conversion
observed this characteristic using different zeolites under a wide of decalin on Y zeolite is not subjected to diffusion restrictions. In
range of experimental conditions [28,29]. The conversion of decalin that sense, Mostad et al. [33] compared ZSM-5, offretite, mordenite,
as a function of reaction time is shown in Fig. 1 for the different different Y zeolites and an amorphous silica-alumina, while Corma
catalysts; it can be seen that catalyst F-cat is the most active, con- et al. [32] reported to ZSM-5, MCM-22, ITQ-2, USY, beta, and MCM-
version being two times the values of catalyst E-cat, which is the 41. Al-Sabawi and de Lasa [34] also compared the performance of
least active one. The activity of the dealuminated catalysts is inter- Y zeolite with different crystal sizes to conclude that pore diffu-
mediate between those of the fresh and equilibrium samples. The sion resistance would not affect the overall rate of conversion, in
conversion profiles of catalysts Cat-1 and Cat-3 are similar, while consistency with previous observations by Sousa-Aguiar et al. [35].
Cat-5 is less active. By comparing the observed conversions at a It is possible to consider decalin isomers as a single reactant
given reaction time, the activities of these catalysts can be corre- species in the forthcoming yield analysis, based on a large body of
lated linearly to their zeolite content (refer to Table 1). Differences evidences about product distributions, kinetics, and isomer reac-
in activity among the various catalysts can be considered to be the tivities (for example, [28,29,36]).
consequence of dealumination in the zeolite component, which is
responsible of activity and selectivity in this type of catalysts [30], 3.3. Reaction products
impacting significantly on catalyst acidity and zeolite (crystalline
material) content, as shown in Table 1. The lowest zeolite content The set of reactions can be initiated through the electrophilic
in catalyst E-cat is the consequence of the reaction–regeneration attack of catalyst surface protons on the tertiary carbons in decalin,
cycles in the commercial unit, and also of the metal contamination as suggested by Sousa-Aguiar et al. [35]; then, activity and selec-
by vanadium, particularly, and nickel, which are always present tivity are expected to be directly correlated to the Brönsted sites in
in VGO feedstocks and play a negative role on the catalyst perfor- the zeolite.
mance [31]. Reaction product hydrocarbons were linear and cyclic olefins
The conversion profiles shown by the dealuminated catalysts and paraffins, bicyclic naphthenics, naphthenic-aromatics, alky-
are similar, a fact that could be due to the small differences in laromatics and diaromatics, comprising the C1 –C12 range (Table 2),
the amount of crystalline material among them. This is particularly and thus defining a very complex mixture. This is consistent
126 R. Pujro et al. / Applied Catalysis A: General 489 (2015) 123–130

Scheme 1. Main reactions in the conversion of decalin.

with the wide distribution of acid strengths found in these cata-

lysts (see Table 1). Based on the detailed product identification,
reactions which were identified in the decalin conversion
network over this type of catalyst were cracking, direct hydro-
gen transfer, isomerization, ring contraction, ring opening and
alkylation–disproportionation [18]. Fig. 2. Product yields of cracking and hydrogen transfer reactions as a function of
The products can be grouped representing the different reac- conversion. () F-cat, () Cat-3, (䊉) E-cat. Closed symbols, solid lines: cracking; open
tions promoted by the catalysts, but it is to be noted that different symbols, dashed lines: hydrogen transfer.
classifications were used by other authors, in response to different
analysis of the experimental observations [29,32,34,36]. Cracking used prototypes of commercial catalysts. The primary character of
reactions are represented by C1 –C9 compounds (olefins, paraffins, the products and reactions can be assigned in view of the linear
naphthenes, naphtheno-aromatics and aromatics); direct hydro- relationship between yield and conversion, while a secondary char-
gen transfer by naphthalene and tetralin; isomerization by bicyclic acter would be revealed by a zero slope at initial conversion.
C10 compounds (mainly dimethyl octabicyclic); ring contraction The yield of products from ring opening and
by indane-type compounds; ring opening by C10 alkylated naph- alkylation–disproportionation reactions were always lower
thenes and cyclic olefins (such as isobutyl cyclohexane and 3-butyl than 1% (results not shown), but it was possible to clearly iden-
cyclohexene, respectively) and alkylation–disproportionation can tify their secondary character. These facts contribute to assume
be particularly represented by the yield of C11 –C12 alkylated aro- alkylation–disproportionation reactions as consecutive to direct
matics or naphthenic-aromatics, such as methylnaphthalene and hydrogen transfer (Scheme 1). Ring opening products are sub-
methyltetralin. jected to other reactions, such as cracking and hydrogen transfer
Most of the products were observed at very low conversions (Scheme 1), due to their high reactivity. Scheme 2 is an example of
and reaction times, thus supporting the concept that reactions such the reaction network which can lead to the production of aromatic
as isomerization, ring opening and hydrogen transfer are simulta- compounds from ring opening and hydrogen transfer reactions, in
neous and occur initially; moreover, reactions that would occur both the light C6 –C9 and intermediate C10 range.
after them in a proposed model [18], such as cracking, ring con- The product yields of cracking, hydrogen transfer and ring con-
traction and alkylation–disproportionation would also take place traction reactions on dealuminated catalysts (Fig. 3A, B and D) are
immediately (Scheme 1). Other authors, who observed small con- between those of the fresh and equilibrium catalysts, in accor-
centrations of aromatics, did not consider that these hydrocarbons dance to the observed activities (Fig. 1), while the product yield
can be formed initially; as a consequence, only ring opening was of isomerization reactions over E-cat is very close to those of the
considered a initial reaction at 450 ◦ C [32] or, in systems were dealuminated catalysts (Fig. 3C), a fact which could be due to the
hydrogen was present, isomerization and ring contraction [28], or similar zeolite load and acid amount in those samples.
ring opening and contraction [29] were considered initial at 400 ◦ C.
According to the product yields as a function of conversion,
cracking reactions were the most important ones over catalyst F-
cat, while the other reactions increased their magnitude as long as
activity in the catalysts decreased. For example, Fig. 2 shows the
product yields of cracking reactions decreasing in the order F-cat,
Cat-3 and E-cat, thus defining a ranking which is opposite to that
of the product yields of hydrogen transfer reactions; particularly,
the product yields of these two reactions are the same over E-cat
catalyst. This behavior points again to the controlling effect of the
catalysts’ zeolite content, and the consequent amount of acid sites,
on catalyst activity and reaction network.
Fig. 3 shows the yield curves of the products from the most
important reactions (cracking, hydrogen transfer, isomerization
and ring contraction) over all the catalysts. It can be seen that
they all show a primary character, following the very early occur-
rence of their products, as also reported by Pujro et al. [18], who Scheme 2. Formation of light C6 –C9 and intermediate C10 aromatics.
 R. Pujro et al. / Applied Catalysis A: General 489 (2015) 123–130 127

Fig. 3. Yield curves for the different reactions. (A) cracking, (B) hydrogen transfer, (C) isomerization, (D) ring contraction. Symbols: () F-cat, () Cat-1, () Cat-3, () Cat-5,
(䊉) E-cat.

A brief analysis in relation to the type of hydrocarbons observed further cracking reactions, are all very similar for the dealumi-
can provide additional information about the mechanisms nated and equilibrium catalysts (Fig. 4C). Fig. 4D shows that even
leading to the formation of aromatics from bicyclic naphthenics though the activity of the catalyst decreases, the yield of aromatics
in the gasoline (heavy end) and LCO (light end) cuts, showing that increases, being the consequence of hydrogen transfer reactions on
aromatic compounds are the most important over all the catalysts, naphthenic compounds (Scheme 2).
with olefins being the less favored and paraffins and naphthen- The previous analysis shows that the dealumination process
ics showing intermediate values which, moreover, are very similar does not impact in a negative way on the formation of aromatics
for the dealuminated catalysts. Fig. 4 shows the yields for each of from bicyclic naphthenics in the range under study, as the change
these groups over the five catalysts. Some evidences are that the in the zeolite unit cell size and the loss of acidic sites might suggest
yield of olefins and paraffins, which are the main consequence of (Table 1) [13,38]. The higher Si/Al relationship and the lower UCS of
cracking reactions [18], decrease together with catalyst activities the dealuminated catalysts are in line with a decrease in the density
(Fig. 4A and B), with the exception of the yield of olefins over E-cat, of paired acid sites in the zeolite [38], which control bimolecular
which is slightly higher than those of the dealuminated catalysts. hydrogen transfer reactions.
On the contrary, the yields of naphthenics, as the consequence of This observation could be associated to the characteristics of
isomerization (leading to bicyclic saturated compounds) and ring the model compound, a bicyclic naphthenic, where one of the rings
opening (leading to alkylated naphthenics) initial reactions, and could be continuously susceptible to act as a donor in hydrogen
128 R. Pujro et al. / Applied Catalysis A: General 489 (2015) 123–130

Fig. 4. Yield curves of hydrocarbon groups. (A) olefins, (B) paraffins, (C) naphthenics, (D) aromatics. Symbols as in Fig. 3.

transfer reactions and the other ring subjected to the other reac- preponderant on the dealuminated and equilibrium catalysts and,
tions (ring opening and contraction). Opening both rings in decalin in all the cases, heavy C11 + aromatics are formed only in small
would not be feasible, or would proceed to a low extension on these amounts (yields lower than 1 wt.%). Fig. 5 shows these facts. Light
catalysts, since no linear C10 paraffins or olefins were observed, and C6 –C9 aromatics yields decrease as long as the catalysts are less
the yields of C7 –C9 paraffins were also very low. Then, the forma- active (Fig. 5A), as well as cracking reactions also decrease (Fig. 3A).
tion of aromatics would be always present, and even increasing, On the contrary, the yields of intermediate C10 aromatics increase
regardless of the dealumination process which decreases the den- if the catalyst activity decreases (Fig. 5B).
sity of paired acid sites in the zeolite. The degree of site isolation, In relation to the formation of aromatics (most important reac-
however, would not be enough to interfere with hydrogen trans- tions are shown in Scheme 2), it can be concluded that a more active
fer reactions; moreover, the catalysts lose activity due to the loss of catalyst would promote the formation of light naphthenic com-
crystalline material and reduce the yield of products from cracking. pounds by way of the upper branch in the scheme, finally leading
Aromatic hydrocarbons can be grouped into light C6 –C9 , inter- to light aromatics through hydrogen transfer reactions (example of
mediate C10 and heavy C11 + molecular weight compounds. The catalyst F-cat). As long as the catalysts are less active, light com-
higher incidence of catalyst activity, particularly on cracking reac- pounds from cracking decrease their yields, and the bottom branch
tions, can be observed on catalyst Cat-F, which favors the formation in the scheme becomes more important, leading to intermediate
of light aromatics (Scheme 2). Intermediate C10 aromatics are C10 aromatics.
 R. Pujro et al. / Applied Catalysis A: General 489 (2015) 123–130 129

Fig. 5. Yield curves of aromatic hydrocarbons. (A) C6 –C9 , (B) C10 . Symbols as in Fig. 3.

All the catalysts showed that the species with a single ring 4. Conclusions
are the preferred ones among the aromatic products (more than
90%) from intermediate bicyclic naphthenic compounds, in coinci- The hydrothermal dealumination of a commercial FCC catalyst
dence with reports by Mostad et al. [36] and Al-Sabawi and de Lasa produces important changes in the crystalline structure of the Y
[34]. zeolite component, which are very fast at short contact times in
Coke yields are not too high. In effect, it can be seen in Fig. 6 the steaming method. These dealuminated catalysts showed an
that the most active catalyst showed a 3 wt.% yield, while for all the intermediate activity between the fresh and equilibrium catalysts,
other catalysts, they were lower than 1 wt.%, a fact also observed by which can be correlated to zeolite content and total acidity; more-
Al-Sabawi and de Lasa [34], who converted decalin over Y zeolite over, their conversion profiles were very similar.
with different crystal sizes at 550 ◦ C, and by Corma et al. (decalin The conversion of decalin as a test reactant allowed defining the
on USY zeolite) at 450 ◦ C [32], and by Corma and Ortega (decalin relative importance of the different reactions occurring on these
over a commercial low rare earth USY FCC catalyst) at 400 ◦ C [37]. FCC catalysts that differ in their most important properties such as
zeolite content, acidity and unit cell size. The reactions that could
be identified were cracking, hydrogen transfer, isomerization, ring
contraction, ring opening and alkylation–disproportionation.
The content of zeolite and the amount of acidic sites, at least
in the range of values observed in this work, seems to be the most
important characteristic controlling the selectivity of the reactions.
Among the most important reactions, the lower the zeolite content
and the acidity, the lower the selectivity to cracking products.
The occurrence of hydrogen transfer reactions is essentially
unavoidable when a bicyclic naphthenic is converted on this type of
catalysts under FCC conditions, and lead to an elevated concentra-
tion of aromatic compounds which belong mainly to the gasoline
boiling point range and, to a lower extent, to LCO. The dealumina-
tion of the catalysts and consequent decrease in the zeolite unit cell
size up to the levels shown in this work is not enough to interfere
seriously with hydrogen transfer.
In order to minimize the yield of aromatics from bicyclic naph-
thenic hydrocarbons, a relatively more active catalyst (higher
zeolite content) is convenient. Alternatively, and considering the
usual performance of commercial catalysts in the FCC units, higher
stability in the zeolite component would be desirable, or a higher
make-up rate needed.

Acknowledgements

This work was performed with the financial assistance of

Fig. 6. Coke yield curves. Symbols as in Fig. 3. National University of Litoral (UNL), Secretary of Science and
130 R. Pujro et al. / Applied Catalysis A: General 489 (2015) 123–130

Technology (Santa Fe, Argentina) (CAID 2011 no. 501 201101 00329 [18] R.A. Pujro, M.G. Falco, A.M. Garrido Pedrosa, M.J.B. Souza, E. Morgado Jr., U.
LI); CONICET (PIP 1257/09) and the National Agency for Scientific Sedran, J. Braz. Chem. Soc. 23 (2012) 1378–1387.
[19] F. Passamonti, G. de la Puente, E. Morgado Jr., W. Gilbert, U. Sedran, Chem. Eng.
and Technological Promotion (PICT 2010/2123). J. 183 (2012) 433–447.
[20] S.C. Fung, C.A. Querini, J. Catal. 138 (1992) 240–254.
References [21] J. Scherzer, Catal. Rev. Sci. Eng. 31 (1989) 215–354.
[22] Q.L. Wang, G. Giannetto, M. Torrealba, G. Perot, C. Kappenstein, M. Guisnet, J.
Catal. 130 (1991) 459–470.
[1] P. O’Connor, Stud. Surf. Sci. Catal. 166 (2007) 227–251. [23] C. Marcilly, Arabian J. Sci. Eng. 21 (1996) 297–312.
[2] UOP LLC, Diesel Fuel. Specifications and Demands for the 21st Century, UOP [24] P. Schneider, Appl. Catal., A: Gen. 129 (1995) 157–165.
LLC, Des Plaines, IL, 1998. [25] Y.Y. Agámez Pertuz, L.A. Oviedo Aguiar, U. Navarro Uribe, M.A. Centeno, J.A.
[3] B.H. Cooper, B.B.L. Donnis, Appl. Catal., A: Gen. 137 (1996) 203–223. Odriozola, Rev. Colomb. Quim. 35 (2006) 7–17.
[4] Directive 2009/30/EC of the European Parliament and of the Council, OJ L140, [26] A.P. Matharu, L.F. Gladden, S.W. Carr, Stud. Surf. Sci. Catal. 94 (1995) 147–154.
2009, 88-113. [27] N. Salman, C.H. Rüscher, J.-Chr. Buhl, W. Lutz, H. Toufar, M. Stöcker, Microporous
[5] Ch. Song, X. Ma, Appl. Catal., B: Environ. 41 (2003) 207–238. Mesoporous Mater. 90 (2006) 339–346.
[6] V. Calemma, R. Giardino, M. Ferrari, Fuel Process. Technol. 91 (2010) 770–776. [28] D. Kubička, N. Kumar, P. Mäki-Arvela, M. Tiitta, V. Niemi, T. Salmi, D. Yu Murzin,
[7] R. Torchio de Oliveira, Anales IV Encuentro Sudamericano de Craqueo Catalítico, J. Catal. 222 (2004) 65–79.
Manaus, Brasil, 2000, pp. 101–108. [29] M. Santikunaporn, J.E. Herrera, S. Jongpatiwut, D.E. Resasco, W.E. Alvarez, E.L.
[8] A. Corma, C. Martínez, L. Sauvanaud, Catal. Today 127 (2007) 3–16. Sughrue, J. Catal. 228 (2004) 100–113.
[9] W.R. Gilbert, E. Morgado Jr., M.A.S. de Abreu, G. de la Puente, F. Passamonti, U. [30] J. Scherzer, Stud. Surf. Sci. Catal. 76 (1993) 145–182.
Sedran, Fuel Process. Technol. 92 (2011) 2235–2240. [31] S-J. Yang, Y-W. Chen, Zeolites 15 (1995) 77–82.
[10] H.L. Hoffman, Hydrocarbon Process. February (1990) 53–54. [32] A. Corma, V. González-Alfaro, A.V. Orchillés, J. Catal. 200 (2001) 34–44.
[11] B. Behera, S.S. Ray, Catal. Today 141 (2009) 195–204. [33] H.B. Mostad, T.U. Riis, O.H. Ellestad, Appl. Catal. 58 (1990) 105–117.
[12] A. Corma, F. Mocholi, V. Orchilles, G.S. Koermer, R.J. Madon, Appl. Catal. 67 [34] M. Al-Sabawi, H. de Lasa, Chem. Eng. Sci. 65 (2010) 626–644.
(1991) 307–324. [35] E.F. Sousa-Aguiar, C.J.A. Mota, M.L.M. Valle, M.P. da Silva, D.F. da Silva, J. Mol.
[13] G. de la Puente, U. Sedran, Appl. Catal., A: Gen. 144 (1996) 147–158. Catal. 104 (1996) 267–271.
[14] M.F.L. Johnson, J. Catal. 52 (1978) 425–431. [36] H.B. Mostad, T.U. Riis, O.H. Ellestad, Appl. Catal. 63 (1990) 345–364.
[15] C.A. Emeis, J. Catal. 141 (1993) 347–354. [37] A. Corma, F.J. Ortega, J. Catal. 233 (2005) 257–265.
[16] M.S. Renzini, U. Sedran, L.B. Pierella, J. Anal. Appl. Pyrolysis 86 (2009) 215–220. [38] G. de la Puente, U. Sedran, Chem. Eng. Sci. 55 (2000) 759–765.
[17] H.I. de Lasa, US Patent 5,102,628, 1992.