FCC catalysts: Tính chất, độ axit, cấu trúc và khả năng phản ứng với kích thước tinh thể zeolite khác nhau

Applied Catalysis A: General 270 (2004) 9–25

FCC catalysts with different zeolite crystallite sizes:

acidity, structural properties and reactivity

G. Tonettoa, J. Atiasb, H. de Lasaa,∗

aChemical Reactor Engineering Centre, Faculty of Engineering Science, University of Western Ontario,

London, Ont., Canada N6A 5B9

bDepartamento de Termodinámica y Fenómenos de Transferencia, Universidad Simón Bol´ıvar, Caracas, Venezuela

Received in revised form 21 March 2004; accepted 25 March 2004

Abstract

Two FCC catalysts with a 60 ␮m average size, prepared with different crystallite sizes (0.4 and 0.9 ␮m), are studied in terms of total

acidity, energy of ammonia desorption, specific surface area, relative amount of Brönsted and Lewis acid sites, crystallite and particle size,

and micropore size distribution.

Characterization of these catalysts is relevant in view of their application to test the extent of intracrystallite diffusive transport influence on

the overall catalytic cracking rates as developed, in separate studies [Ind. Eng. Chem. Res. 42 (2003) 4162] in a novel CREC Riser Simulator.

Keywords: Y zeolite; FCC catalysts; Nitrogen adsorption; Ammonia TPD; Surface area; Porosity

1. Introduction

Catalytic cracking of hydrocarbons combined with fluid

bed processes (FCC) is one of the most important and most

profitable processes in the petroleum refining industry [2].

FCC units are used to upgrade heavy gas oils to gasoline,

diesel fuel, and light gases. Catalytic cracking of hydrocar-

bons is a vapor-phase reaction occurring via a carbonium ion

mechanism. This reaction is catalyzed by materials having

acidic properties [3,4].

The introduction of zeolite cracking catalysts in the early

1960s revolutionized the petroleum refining industry. Y ze-

olites have been used extensively in catalytic cracking since

1962. The currently available commercial FCC catalysts

are manufactured with 1 ␮m HY crystallites, which are

dispersed via spry drying in an amorphous silica-alumina

matrix [5]. The fluidizable pellets are about 60 ␮m

diameter.

Y zeolites are constituted by building blocks of a fauj-

asite structure. Y zeolites display a three-dimensional pore

network with pores running perpendicular to each other.

∗Corresponding author. Tel.: +1-519-661211; fax: +1-519-6613498.

E-mail address: gtonetto@plapiqui.edu.ar (H. de Lasa).

Twelve-member oxygen rings delimit the 7.4 Å opening con-

necting supercages with a 12 Å diameter [6].

These crystallites have catalytic active sites located both

in the internal pore structure and as well as on the outer

surface of the crystallites. The majority of the active sites

is located within the zeolite pore structure [7]. In order for

the reaction to proceed, molecules have to diffuse through

the large matrix pores into the zeolite crystals. Diffusion in

catalysts belongs, in many cases, to the Knudsen regime,

because the zeolite pore wall significantly influences dif-

fusing molecules. Diffusion in zeolites, however, falls more

in a so-called “configurational regime”, since the size of

the hydrocarbon molecules is nearly the size of the ze-

olite inner channels and the diffusing molecules evolve

under strong pore constrains and may not pass each other

[8,9].

The specific characteristics of zeolites that are respon-

sible for their ability to catalyze chemical reactions are

their regular crystalline structure and acidity. In addition,

the presence of molecules blocking the pores of a zeolite

or the partial destruction of its structure may significantly

reduce the activity of zeolites and modify the selectivity by

limiting the accessibility of the internal active sites to the

reactants. For instance, dealumination, a common treatment

doi:10.1016/j.apcata.2004.03.042

10 G. Tonetto et al. / Applied Catalysis A: General 270 (2004) 9–25

Nomenclature

Cconstant related with the experimental

condition

Edes energy of desorption of ammonia (kJ/mol)

kdadsorption constant of ammonia (ml/min)

kd0 pre-exponential factor of adsorption of

ammonia (ml/min)

Runiversal gas constant (8.314 J/mol K)

ttime (min)

Ttemperature (K)

Tptemperature at maximum desorption rate (K)

T0centering temperature (K)

Vdes volume of ammonia desorbed (ml)

Vmvolume of ammonia adsorbed at saturation

(ml)

Greek letters

βrate of temperature increase (K/min)

θcoverage

applied to zeolites, is known to modify the activity as

well as the microporous volume and the external surface

of these catalysts. Then, the accurate determination of the

external surface area, microporous volume and acidity of

zeolite is of major importance for understanding its catalytic

behavior.

The feedstocks for FCC units have been changing

throughtout the years. Heavier feedstocks are commonly

used by refiners for several reasons such as: price, availabil-

ity, etc., pushing the process to its limit. By using heavier

feedstocks, the process tend to be more diffusion controlled,

jeopardizing the characteristics of activity and selectivity

of USY zeolites. Thus, a reduction of the crystallite size in

the FCC catalysts is a valid solution to have a FCC process

more robust, which can meet the challenges of refining

heavier feedstocks.

In this work the effect of HY crystallite size (0.4 and

0.9 ␮m) on several catalyst properties is investigated: (a)

the number, strength, and distribution of acid sites are stud-

ied using temperature-programmed desorption of ammonia,

with acidity evaluated using the higher temperature desorp-

tion peak; (b) FTIR analysis of pre-adsorbed pyridine is used

to determine the relative amount of Brönsted and Lewis acid

sites on the samples; (c) SEM is used to confirm particle

and pellets size; (d) nitrogen and argon isotherms performed

at 77 K are studied to determine the following microstruc-

ture properties: specific surface area, pore volume and pore

size distribution. In this respect, the characterization of these

catalysts is very relevant in view of their value to establish,

in a novel Chemical Reactor Engineering Center (CREC)

Riser Simulator [10,11], the extent of diffusion control in

FCC catalysts and the influence of this phenomenon on the

overall reaction rate.

2. Experimental

2.1. Catalysts preparation

This study was carried out using two commercial USY

zeolites with different crystals size from Tosoh Corpora-

tion (0.4 and 0.9 ␮m). HYZ-SC and HYZ-LC refer to small

and large crystal zeolites, respectively. Both zeolites have

a similar SiO2/Al2O3ratio (5.6 and 5.7 mol/mol). In order

to remove the traces of Na, both zeolites were transformed,

through ion exchange, to the ammonium form with a 1 M

NH4NO3solution at 80 ◦C for 6 h under stirring conditions.

This operation was followed by filtration and three water

washings. The zeolite samples were dried in air at 90 ◦C for

2 h and following this at 120 ◦C for 6 h. The general char-

acteristics of the zeolites used in this study are reported in

Table 1.

These parent zeolites were spray-dried using kaolin, as

filler, and silica sol, as binder. The formed catalyst pel-

lets had a particle size in the range of 60 ␮m. Such a

catalyst particle size was needed for evaluating catalyst

activity in a fluidized CREC Riser Simulator unit. The

catalyst composition was 30 wt.% zeolite, 50 wt.% kaolin,

and 20 wt.% silica sol. The pelletized catalysts were ion

exchanged following the procedure described above. Once

the ion exchange operation completed, the catalysts were

calcined at 600 ◦C for 2 h in an oven to remove all volatile

material. After that, the catalysts were treated with steam,

in order to increase its thermal and hydrothermal stability.

Steaming conditions were 100% steam for 5 h at 760 ◦C

(Scheme 1).

The FCC-steamed catalysts, prepared from small crys-

tal zeolites (HYZ-SC), are referred here to as CAT-SC,

while the catalysts derived from large crystals (HYZ-LC)

are called CAT-LC. CAT-LC-BS and CAT-SC-BS are also

acronyms for the catalyst samples before the steaming

treatment.

The unit cell size for the catalyst samples was determined

using X-ray diffraction following ASTM D-3942-80. Na

content was measured by EDX. Table 2 reports the catalyst

main properties following catalyst pretreatment.

2.2. Catalysts characterization

2.2.1. NH3-TPD

The apparatus used for TPD test was the Micromeritics

AutoChem II analyzer. A catalyst sample (0.1–0.3 g) con-

tained in a quartz container was degassed for 2 h at 500 ◦C.

Table 1

Properties of studied USY

HYZ-SC HYZ-LC

SiO2/Al2O3(mol/mol) 5.6 5.7

Unit cell size (Å) 24.49 24.51

HY crystallite size (␮m) 0.4 0.9

G. Tonetto et al. / Applied Catalysis A: General 270 (2004) 9–25 11

Scheme 1. Diagram of the Riser Simulator experimental setup and accessories.

The sample was brought to saturation using an NH3/He gas

mixture (4.55% ammonia, 95.55% helium). In order to know

the acid strength, ammonia adsorption was carried out for

1 h at various temperatures: 100, 140, 160, 190, 240, and

300 ◦C. The ammonia flow was then switched off, and re-

placed by an inert purge gas (He) at a rate of 50 ml/min for

1 h and this to remove physically adsorbed ammonia. The

samples were purged at the same temperature of ammonia

adsorption. Then, the temperature was raised up at a rate of

15 ◦C/min. As the temperature was increased, the ammonia

desorbed as it gained enough energy to overcome their ac-

tivation energy barrier.

Furthermore, the energy of desorption, associated with the

high-temperature desorption peak, was calculated. With this

aim, NH3-TPD chromatograms were collected at different

heating rates (7, 10, 15, 20, and 30 ◦C/min). During this set

of experiments, adsorption of ammonia and the subsequent

purge of the catalysts were carried out at the set temperature

of 100 ◦C, for a period lasting 1 h. The total flow used was

50 std ml/min STP.

Table 2

Properties of studied catalysts

CAT-SC CAT-LC

Zeolite contenta(%) 31 29

Unit cell size (Å) 24.28 24.27

HY crystallite size (␮m) 0.4 0.9

Relative dealuminationb(%) 83.3 88.2

aTaken in account specific surface area (BET) for zeolites, matrix and

catalysts before steaming.

bAccording to Ref. [13].

2.2.2. Pyridine IR

The zeolite samples were dried, in situ, by heating under

N2flow at 550 ◦C and cooled to 100 ◦C. The samples were

maintained at this temperature and saturated with pyridine

using a N2contains pyridine stream. Adsorption of pyridine

was performed for a 60 min period. Then, the zeolite was

flushed with N2, at the same temperature for 90 min, and

this to remove weakly adsorbed pyridine species.

In order to establish the acid strength, following pyridine

adsorption, zeolites samples were purged at different tem-

peratures: 100, 150, 200, and 300 ◦C.

Zeolites samples were placed later between NaCl win-

dows and infrared spectra were recorded using a Bruker

IFS55 FTIR spectrometer operating at 4 cm−1resolution and

100 scans. FTIR spectra were collected at room temperature.

FCC catalyst pellets (60 ␮m), zeolites dispersed in a

kaolin and silica-sol matrix, as well as reference pellets

constituted by kaolin and silica-sol only, were ground into a

fine powder with an agate mortar. For these samples, diffuse

reflectance infrared spectroscopy (DRIFTS) measurements

were obtained at the same conditions that for the zeolite

samples using the same Bruker IFS55 FTIR spectrometer

and a DRIFTS cell (HVC-DRP, Harrick Scientific). The

sample holder had 50 mg capacity.

2.2.3. N2and Ar isotherms

Nitrogen and argon adsorption measurements were car-

ried out at 77 K on an ASAP 2010 automatic adsorption

analyzer equipped with the micropore options (from Mi-

cromeritics). Before the measurements, samples weighing

from 0.1 to 0.3 g were placed in the degas port of the adsorp-

tion apparatus, degassed at 100 ◦C for 1 h and at 300 ◦C for

12 G. Tonetto et al. / Applied Catalysis A: General 270 (2004) 9–25

10 h. The data from the low-pressure region were obtained

by contacting the sample with successive increments of anal-

ysis gas (doses of 3 cm3/g STP of sample) and waiting un-

til thermal equilibration was reached. Adsorption isotherms

were measured under the relative pressure range, based on

the saturation vapor pressure, from ∼10−6to 1.

2.2.4. SEM-EDX

A Hitachi S-4500 field emission SEM fully equipped with

an EDAXTM EDX system was used in the present study.

The Hitachi S-4500 was employed for obtaining the imaging

of the samples (electron beam voltages: 5 kV, resolution:

2–60 ␮m) with samples being Au coating to reduce charging

problems. The EDX system was also used to measure Na

contents.

2.3. Catalytic experiments

Experiments with CAT-SC and CAT-LC were car-

ried out using the CREC fluidized Riser Simulator in-

vented by de Lasa [10,11]. The model compounds were

1,2,4-trimethylbenzene (1,2,4-TMB) and 1,3,5-triisopro-

pylbenzene (1,3,5-TIPB) with molecular diameter of 7.4 and

9.4 Å, respectively. Selection of the model compounds al-

lows catalytic reaction condition with or without diffusional

constrains. A Hewlett Packard 5890A Gas Chromatograph

quantifies reaction products using a flame ionization detec-

tor (FID) and a HP-1 25 m capillary column of cross-linked

methyl silicone with an outer diameter of 0.20 mm and an

internal diameter of 0.33 ␮m. A more detailed description

of various components of the Riser Simulator and the se-

quence of injection and sampling can be found in Atias

et al. [1]. An average of 10 experimental runs were car-

ried out at different temperatures (350, 400, 450, 500, and

550 ◦C) and residence times (3, 5, and 7 s), given a total of

about 150 experiments for every catalyst.

3. Results and discussion

3.1. Catalysts preparation

Tables 1 and 2 report the catalyst properties of the samples

studied. Na content was found negligible in all samples after

ion exchange, which demonstrates the effectiveness of the

method used for sodium remove. Moreover the values for the

unit cell sizes (UCS), zeolites content and specific surface

area (Table 6) were found to be consistent with typical for

FCC catalysts [12,46].

For both zeolites it was observed (Table 1) that the initial

unit cell size of 24.5 Å was reduced to 24.28 Å (Table 2)

given both samples were subject to close steaming condi-

tions.

In fact, under steaming, dealumination takes place and as

a result, there is a higher abundance of Si–O bonds, which

are shorter than Al–O bonds. Given that the UCS parameter

is related to the abundance of Al–O bonds in the crystalline

structure framework, dealumination has to lead to a UCS re-

duction. The extent of dealumination can be evaluated as the

difference of UCS values between the treated and untreated

samples [13]:

dealumination (%)=UCS(untr)−UCS(tr)

UCS(untr)−24.238 ×100 (1)

where UCS(tr) refers to the steam-treated sample and

UCS(untr) refers to the untreated USY sample. It can be

noticed in Table 2 that both CAT-SC and CAT-LC samples

present a relative dealumination above 80% and can be

considered as direct consequence of the severe steaming

treatment [13].

3.2. NH3-TPD

3.2.1. Total acidity and acid distribution

Acid sites of zeolites were characterized by temperature-

programmed desorption using NH3as basic probe molecule

(NH3-TPD). The total acidity and acid distribution can be

obtained from the peak area, position, and shape, respec-

tively. NH3-TPD can distinguish sites by sorption strength

only and, in fact, it cannot differentiate between Brönsted

from Lewis-type sites [15,16].

At first, it is important to highlight that it was found in

separate experiments, developed with catalyst pellets made

of the silica matrix only, that the NH3adsorption on the

silica matrix was not significant.

NH3-TPD spectra of HYZ-SC and CAT-SC-BS samples

are reported in Figs. 1 and 2. Both samples present close

NH3desorption profiles. This proves that the pelletization

process does not strongly affect catalyst acidity. When total

acidity of zeolite and before-steaming catalyst is compared,

the difference is less than 13% (Table 3, referring to 190 ◦C

NH3adsorption temperature).

Figs. 1 and 2 also show that when the NH3adsorption

was performed at temperatures below 240 ◦C, NH3-TPD

curves display two overlapped peaks. The high-temperature

peak can be related to desorption of ammonia from both

Table 3

Temperature-programmed desorption of studied samples: amount of am-

monia desorbed (in mmol NH3/g HY) for different ammonia adsorption

temperature

Sample Ammonia adsorption/purge temperature (◦C)

100 140 160 190 240 300

HYZ-LC 1.61 1.43 1.33 1.30 1.48 0.98

HYZ-SC 1.70 1.43 1.22 1.37 1.42 0.88

CAT-LC-BS 0.99 0.96 0.95 1.14 1.02 0.66

CAT-SC-BS 1.14 1.13 0.95 1.19 1.20 0.80

CAT-LC 0.57 0.21 0.20 0.17 0.13 0.08

CAT-SC 0.34 0.18 0.15 0.19 0.13 0.08

Experimental condition—Q:50cm

3/m; temperature rate: 15 ◦C/min; gas

mixture: 4.52% NH3(balance in He).

G. Tonetto et al. / Applied Catalysis A: General 270 (2004) 9–25 13

Temperature (˚C)

50 100 150 200 250 300 350 400 450 500 550

TCD Concentration

0.0

0.2

0.4

0.6

0.8

100˚C

140˚C

160˚C

190˚C

240˚C

300˚C

Fig. 1. Thermal desorption chromatograms of ammonia for HYZ-SC sample. Experimental conditions—temperature rate: 15 ◦C/min, helium flow:

50 cm3/min. NH3adsorption temperature are indicated in the graphic.

strong Brönsted and Lewis sites [14,17]. On the other hand,

the low-temperature peak can be attributed to a number of

causes: (a) desorption of weakly adsorbed NH3on weak acid

sites or non-acidic site [13], (b) weak acid Brönsted [17],

Temperature (˚C)

5050 100 150 200 250 300 350 400 450 500 550

TCD Concentration

0.00.0

0.1

0.2

0.3

0.4

0.5

100˚C

140˚C

160˚C

190˚C

240˚C

300˚C

Fig. 2. Thermal desorption chromatograms of ammonia for CAT-SC-BS sample. Experimental conditions—temperature rate: 15 ◦C/min, helium flow:

50 cm3/min. NH3adsorption temperature are indicated in the graphic.

of NH4+(NH3)n(n≥1) groups [19,20].

In catalytic cracking, the active centers are Brönsted acid

sites in the zeolite framework [21]. Therefore, the amounts

FCC catalysts with different zeolite crystallite sizes: acidity, structural properties and reactivity

Chủ đề:

Tài liệu liên quan

Tài liêu mới

AI tóm tắt

Giới thiệu tài liệu

Đối tượng sử dụng

Từ khoá chính

Nội dung tóm tắt

Hỗ trợ

Phương thức thanh toán

Theo dõi chúng tôi