Basicity and basic catalytic properties of zeolites

Basicity and basic catalytic properties of zeolites

Materials Chemistry and Physics, 18 (1988) 553-575 553 BASICITY AND BASIC CATALYTIC PROPERTIES OF ZEOLITES D. BARTHOMEUF Laboratoire de rCactivit& ...

1MB Sizes 4 Downloads 85 Views

Materials Chemistry and Physics, 18 (1988) 553-575

553

BASICITY AND BASIC CATALYTIC PROPERTIES OF ZEOLITES

D. BARTHOMEUF Laboratoire de rCactivit& de surface et structures, Universite Paris VI, F 75252 Paris CQdex 05 (France) G. COUDURIER and J.C. VEDRINE* Institut de Recherches sur la Catalyse, CNRS, 2 avenue Albert Einstein, F 69626 Villeurbanne Cedex (France) Received

April 13, 1987; accepted

August

31, 1987

ABSTRACT In the present review article one describes basicity for zeolites from theoretical and experimental points of view. Zeolites appear to be rather soft bases,their basicity increasing within the series Li < Na < K < Rb < Cs. The physical methqts used_for characterizing basic sites either themselves (anionic) species : 0 OH-) or through adsorption of acidic probes, as r A104,, C02, S02, organic acids, pyrrole, etc., are described.A review of basic type reactions catalyzed by zeolites is also given. The influence of Lewis acidity of exchangeable cations and of basicity of framework oxide ions on adsorption processes

is

discussed.

INI'RODUCTION

Acidity and basicity of catalysts are features which are often mentioned to explain their catalytic

properties.

For

liquids

one

defines

usually

the

pKa

constant by the following relation AH 2 H+ + ABH+=

K = IH+IIA-I/I~I a = I?31 lH+l/@i+l Ka

b + H+

for acidity for basicity

with pKa = - log 10 Ka In the 1920's Brijnsted and Pedersen [l] and further Hammett and Deyrup [Z] quantified the relationship between the acid strength of an acid in solution and the rate of the catalytic reaction. For solids Hammett has defined the H

0

acidity

function

activity

of

a

neutral

the base

: Ho

proton i

= and

and

-

yifi~~+ its

[II

log(aH+. Yi/YiH+) the

ratio

conjugate

of acid

the -

where

activity +

BH .

One

aH+

is

coefficients

writes

the of also

* Author to whom correspondence should be sent. 0254-0584/88/$3.50

0 Elsevier Sequoia/Printed in The Netherlands

554 H = PUBH' - log( IBH+I/ ISI) (2) 0 The ]BH+/]B] ratio is directly observable Prom the Hammett indicator in its two different colored forms. Acid catalysts are characterized by the value of Ho which is negative, the highest negative value corresponding to the strongest acid strength. Often an acid strength distribution is observed. The Ho acidity function evaluates the trend to give a proton to an indicator. Indicators are chosen such as yi/yBH+ is identical for all of them and therefore measures the acidic property of the medium whatever the indicator. After analogy with the acid strength of solids, their basic strength can be defined from the Brijnstedconcept [l], as the proton-accepting solid surface. It is quantitatively

ability of the

expressed by the Hammett and Deyrup H_

function, [2]. Considering an acid indicator BH reacting with a solid base g BH+B

:

B-+BH+

(3)

The basic strength H

of % is given by an equation similar to the well

known Ho equation 121 for acid strength H _ = eKBH + log(]B-]/]BH]) where

pKBH

=

PK,,

]B-] and

(4) ]BH] are

concentrations

of

the

adsorbed

indicators. Equation (4) means that the basic strength of a solid surface increases with an increase in H .

Basic character of zeolites Zeolite materials are now well known in the field of catalysis for their acidic properties. This acidity feature was assigned to the tetracoordinationof Al in a tetrahedral Si02 framework which results in a negatively charged AlO; anion : this

charge is compensated

by cations and particularly by

protons resulting in Briinstedacidity. A question then arises : could we prepare a positively charged zeolitic framework? The idea was to incorporate a pentavalent element rather than the trivalent Al into the Si02 framework and to neutralise the charge with anions of basic character. Phosphorus as a pentavalent element was largely investigated without success to our knowledge. New molecular sieves discovered by Union Carbide

[3] and designated ALP04 (AlO;, PO;), SAP0 ]AlO;, (Si02)x, (PO:),_x],

MePO et ElAPO where Me and El are metals or elements incorporated in the ALP0 framework might have some basic properties. The second idea was to exchange protons by cations in order to neutralize the acidity and hopefully develop basicity. As cations are soft acids one should then expect that the basic character of the zeolite framework should be higher than the cationic Lewis acidity, if one would have any chance to obtain basic 2zeolite. The basicity of the framework should correspond to anions such as 0 , A104- or OH-.

555 Because much

of

important

attention

zeolites.

industrial

applications

has been paid up to now to acidic

However

basic

catalysts

are

mainly

for catalytic

zeolites

known

cracking

and much less to basic

as important

for many reactions

[3, 41. In this review the

role

articles

article we have tried to summarize

of basicity

for

have already

The present baslcity

paper

is divided

in zeolites,

(iii) adsorption

seolites

and separation

reactions

in the field of catalysis

APPROACHES

As described from

the

carried

and

by the oxide

Such calculations method

feature

have been carried

unequally

currently

of anions

for faujasite over

the

ions,

the basic properties

are

in the xeolite

out in order

adsorption (iv) basic

framework.

to determine

distribution

expected

the

Theoretical

negative

all along

for characterizing

type calculation

basicity.

four

oxide

ions.

Unexpectedly

Si/Al

ratio

varied

may

perform

principle

assumption

on the

which

extent

a significant

phosphates,

opposite

way

with

increases

as the zeolite

alkaline

metal

respect

ion series.

dimeric

monovalent

increases

charge

compensating

data of

type cation

atoms

without

were obtained

to zeolites

the basic

[14].

for oxides

for acidity

strength,

Sanderson

any

in the compound

varies

[17,18] in an

electronegativity.

and the cation

It

size goes up in the

was then found to be the most basic

any influence

of the

expect Pauling

electronegativity

on

are only based on the chemical

are based on quantum

clusters

charge

i.e.

CsX faujasite

i.e. do not reflect

simple

on experimental

of bonds

properties

to the intermediate

The calculations

Other calculations taking

charge,

Al content

the

Sanderson

an average

[15,16] and extended

[19]. The negative

material,

the

ionicity-covalency

and basicity

the series.

using

with acid-base

etc.)

from 1 to 6 [lo]. The

either

based

are

[13].

gives

of

using

calculations

calculations

The first data in connection (sulfates,

calculated

and X-ray emission

also

equalization

values

CNDO/Z

IlO,ll]. It was shown that the charges

as the

[12] or more recent

X-ray diffraction One

the

charge

the framework.

(T = Si or Al) were found not to be fully ionic as one could and

on type

to stem

T-O-T

a SiO2 matrix

of

basicity,

was

bonds

observed

by selective oxide

[5-g].

approaches

difference

from

not

mixtures

review

IN ZEOLITES

and data may be useful

distributed

Some

in catalysis

for characterizing

and framework

ions and the charge

has been applied

catalysis.

about

by zeolites.

in the introduction

presence

calculations

cation

OF BASICITY

of

knowledge

: (i) theoretical

parts

used

of gaseous

role of exchangeable

field

on the role of basicity

in four

(ii) methods

zeolites,

THEORETICAL

in the

been published

the present

composition

in

of the

of the structure. chemical

(OH)3

ab initio calculations

SiOKf(OH)3

and T

is Si,

[20]

as model where M is the Al or B. The

effect

of

+

cations as H , Li

+

and Na

+

on the charge distribution and the Si-O-Al bridge

stability has been calculated. It was found that the M-O bond becomes more ionic + when H is replaced by more electropositive cations. The negative charge, originallylargely delocalized in the cluster, gets more localized on the oxygen + of the Si-O-T bond as charge compensating M . It follows that bonds are primarily covalent in character, i.e. the anionic framework has to be considered as a soft base [20].

CHARACTERIZATIOR OF BASICITY The usual way is to use acidic adsorbates of different strengths in order to probe

the basicity

of solid materials.

The difficulty

arises

from the

appropriate choice of acidic molecules. For acidity measurement there exists a large number of basic probes useful for zeolites although the use of large organic molecules as Hammett indicators may be limited for small pore zeolites because of steric hindrance. For basicity measurement [21] the number of acidic probes able to cover a wide range of pKa is rather small. Moreover a difficulty stems from the fact that the acidic probe molecules may interact with cations. For instance CO2 may either be adsorbed on the cations or physisorbed or may react with

hydroxyls and

framework oxide

ions

to give a carbonated species

[22-251. The adsorption of molecules on a catalyst involves several interactions as depicted by Barrer and Gibbons [26]. The energy of adsorption may be written as the summation

p, being the dispersion; aR the short range repulsion, $6Fthe polarization, !i$, the field dipole, $?J;+the field gradient quadrupole and $

the sorbate-sorbate sP The first two terms depend on interaction (sp denotes self potential) energies.

the sorbent while the three following terms depend on both the heteropolarityOf the sorbent and the nature of the sorbate. The fourth and fifth terms are depending on the orientation of the sorbate molecule with respect to the sorption site. The $KD term equals -A/r6 where r is the distance between the centers of the interacting adsorbent-adsorbateand A is a constant calculated 12 from London and Kirkwood-Muller formulae. p, is depending on B/r function (Lennard Jones potential), lap= field strength) and !K)i_Q=

F*/2 (

polarixability of the adsorbate, F

- QF(3cos*@-1)/4r with Q being the quadrupole

moment. In order to calculate some of these terms, fields and field gradients have been evaluated by means of various approximations for the charge distributions on cations and on oxide ions [26-311. Some values as calculated following the principles described above are given in Table I.

Table I. Experimental initial values of sorption energies E and calculateq values H for co 2 adsorbed on X-type zeolite with various cations in kJ.mol from ref. 26b.

Energy values

Li

AE AH

*

51.4 55.2 15.9 0.4 9.6 30.9 1.7

Cationic form K Na

45.2 38.9 13.0 0.8 5.0 21.3 1.3

43.9 28.8 7.1 3.3 2.1 17.6 1.3

Rb

cs

42.2 23.4 4.6 4.6 0.8 14.6 1.3

36.8 22.2 4.6 9.2 0 9.6 1.3

go: estimated zero point energy It is worthwhile

noting that the energy of CO2 sorption in X zeolite

decreases from Li to Cs, which is mainly due to the decrease of the quadrupole energy term while the dispersion energy term 9, value increases for cations and decreases for oxide ions. Probe molecules The ideal

probe

molecule

should

be specific

to basic

sites

and distinguish

interaction with oxide ion and hydroxyls. It should also be stable with time and with temperature. CO2 has widely been chosen to characterize basicity of solids [22-251 but it interacts with several types of basic sites and may give rise to chemical reaction (carbonation).

Carboxylic acids or bensoic acid have

been

used [7,21] and phenol as well [32]. However for oxides and mixed oxides phenol may decompose, which precludes its practical use. Pyrrole has also be used particularly

at room temperature

and

for short contact

time

to

avoid

decomposition [19,33]. Benzene has not be used to titrate basicity but has been shown to interact presumably with oxides ions in the 12-R window of faujasite as a function of their basic strength [34,35]. Colored indicators Few results are available. Attemps for basicity measurements using benzoic acid solution in dry benzene shows the presence of strong basicity in highly exchanged CsX samples [36] Hammett indicators have also been used to determine the change in weak acidity of basic seolite by the addition of acidic cation. For instance Li+ ion was shown to exhibit stronger acidic nature than the other = + 3.3 for LiX and KbLiX against + 4.0 for RbX and RbKX [37].

cations H

0

max

550 Thermal

methods

They

correspond

to

Thermogravimetry of

isotherm

and then

adsorption.

of

CO2

[38-40, Li

adsorption

to

Cs

in the

very

low

extent

of

coverages

for

[UO].

the

Adsorption

of

141,421.

respect

to

Unfortunately basicity

Clausius-Clapeyron

used

both

to

has

coverage

observed

interaction

and

I)

The

between

the with

studied

been

increase

[39].

to

stable

to

increased

data

adsorption equation

widely

(Table

1261 and to as

giving

the

extensively

handle

zeolites

X-zeolites

a maximum was

it

is

Moreover

was

at

and

the

number

experiment

and

then

of

decrease

from

when Na+

ions

[38]

except

effect

of

adsocbates

for

coverage

or

from

site

on

both

one

but sites

in

methods of

anions

UV-visible This

or

H-Y

Na-Y

but

and

in

and

on

of

using the

increase

sample,

nature

it

of

strength

small

adsorption

CO2

[43]. at

pulses

adsorption.

on the

occur.

a decrease

outgassed

heat

of

CO2 this

adsorption

increases that

acidic

4OO’C

In

to

may be suggested

present

are

in

heat , corresponding As the

.

microcalorimetry

their

of

NaY -1

about

did

heat

results

successive heat

acidification,

in

in

for

12 kJ.mol

conclusions

previously

with

sites.

of

by

form

a function

highest

cations

zeolites

The

as

differential

acidic

adsorption

basic

upon

CO2 with

the

differential

total

basic

the

the

temperature, the

amount

presumably

purpose

different

room

the

(Na+)

not

Spectroscopic The

to

measures

cations

strength strong

down

of

of

the

to draw

unambiguously

in contacting

calorimetrically

were

a difference

interaction

measure

calorimetrically.

Eoor CO2 studies

with

of the Na form to the sites

studied as

and entropy (421

possible

temperature

adsorption

cooled

experiment

not

the

consists

and measuring

enthalpy

that

shows that acidification

The

the

also

same trend

simultaneous

noting

room

was

the

KY zeolites

then

since

worth

adsorption

molecules

exhibited

NaX

strength

It

in

rather

HY zeolite.

characterization such basic

is

techniques sites

through

to

try

to

characterize

adsorbed

probes.

transfer

transitions

either

the

spectcsocopy

technique

interest

for

obtained

in

varies

the

observed

decreased

increased

benzene

isotherms

cations

with

on

in was

methods

to determine

easy

molecules.

heterogeneity.

adsorption

in

of It

which

from

cations

used

probe

one

been

gas

of

adsorption

commonly

using

a

adsorption

series

by Ca2+

may arise

energy

heat

alkaline

replaced

most

have

is

2 with

the

allows

enthalpy

CO

interaction The

the

and calorimetry

adsorption.

26,271.

were

of

the

Its

temperature.

are

on

Thermogravimetry

Thermogravimetry

study

based

and calorimetry

on heats

[271.

techniques

(angle,

is

sensitive

characterizing that

field

covalency

but

to

the one

character

charge

oxide

ion

properties.

may reasonably etc.)

the

expect

charge

that

transfer

and could

Few

data

when the bands

be of

have

been

T-O-T

bond

corresponding

to the T-O found

bond could be modified.

to be different

aluminium

oxide

investigation

carefully

in

are

cation

of

the

to framework

mordenite

nature

opinion

the nature of the Al-O bond was

[44].

has

been

that

Al or to extra

Up

to

now

performed

such

no

lattice

systematic

with correlation

features

should

be

with

looked

for

in the future.

Infrared This

it corresponds

compound

versus

We

basicity.

when

For instance

spectroscopy

technique

is

certainly

the study of adsorbates. characterize

molecules

Pyridine

Brijnsted

characterization

and

most

commonly

and ammonia

Lewis-type

several

are probably

the

acidic

probes

acidic

or

have

sites

amphoteric

isotherms

pressure Co-A of

and

zeolites

adsorption

adsorption the

been

widely

in zeolites.

probes

for

used

have been employed.

analyzing

the

[45]. This

is related

can

be evaluated

physisorption

cations

region

heat

giving

latter

CO2

co2 chemical

species

Pyrrole [33,19].

is shifted

H atoms

with

evaluates different

[14]. The

Y)

and

L

be drawn: decreases (ii)

zeolites

substract obtain

with oxide

The

ions.

extent

zeolite with

value

behave

the

of

VNH

charge in

samples

1300-1600 influences

the

in the strongest

on oxide

the

same

for mordenite,

is due to the interaction

of to

mode

values

frequency

ion increased trend

basic strength

zeolites

of this shift with respect

while

are

presented

to the charge

mordenite

UhH

on

equalization was observed

1191. Faujasite

The following

the experimental

the high asymmetry

[zzI.

of

electronegativity

ion charge.

of

of basic sites thevNH

Experimental

vibration

groups

in generation

basicity

and may be compared

Sanderson

the ions -1 cm

in Ca-Y and Mg-Y zeolites

to characterize

low wavenumbers , which

while the calculated

for mordenite

position resulting

and in the presence

VNH shift for the same oxide (i) except

in the

resulting

cations,

the basic strength.

when the negative

important

cations

groups

as a probe

oxide

ions as calculated

decrease

on polyvalent

towards

framework

in

principle

in exchangeable

is amphoteric

liquid pyrrole Fig. 1 for

therefore

from interaction

absorbing

polyvalent

and acidic hydroxyl

has also be used

This compound

vibration

and

heat

law. If the total one may

[22-251. CO2 may also interact with basic hydroxyl

formed by water decomposition carbonated

IR data

arises

species

the smallest

interaction

give

and the corresponding

by calorimetry,

from

energy

carbonate-like

bands,

They

using the Clausius-Clapeyron for instance

[23]. The nature of the cations

infrared

oxide

to physisorption

as calculated

The

heat.

adsorbates.

to

For basicity

be determined by IR with increasing CO2 partial -1 band intensity for Na-A, NaCa-A and = 2350 cm

heat may be measured,

chemisorption and

could

used

particularly

rise in 2+ infrared to a shift due to the adsorption on cations, for instance for Ca and -1 2+ an IR band near 2350 cm . The in NaCaMg-Y zeolites 1231 with Mg adsorption

the most currently

used,

also

gives

conclusions

vibration

to

(X and an may

wavenumber

increases.

of the oxygen

ring modifies

the adsorbed

15

Fig.

1

0.30 0.35 0.40 0.45 Negative oxygen charge

. Variations of VNR values against the negative lattice oxide ion

charge calculated using the Sanderson electronegativityequalization principle [14] for various zeolites

: 2 mordenite, _b L, _c Y and _d X in different

cationic forms (A) Li, (0) Na, (A) K, (0) Rb and (0) Cs [19].

pyrrole

molecules

as much

as the oxide ion charges do by themselves. It follows

that pyrrole turns out to be a reliable probe for basicity characterization except

for

adsorbed

zeolitic

structures

which

also

generate

a

high

asymmetry

to

the

molecule.

Using

acid

probes

(formic,

(pyridine) and IR spectroscopy

acetic

or

Bielanski

benzoi'c acids)

and

a

basic

and Datka have shown

probe

(46) the

amphiprotic properties of NaHY zeolite. An IR study of the methanol decomposition on alkali-metal X zeolites (47) correlates the formation of surface formates to the basicity strength of the material. Benzene is also a rather valuable probe molecule. It has been shown (34,35) that for faujasites the ITelectrons of the benzene ring interact with cations in S

and S sites giving rise to v and v,6 + v,, combination bands of CH II III 5 + "17 out-of-plane vibrations slightly shifted towards higher wavenumbers with respect -1 to liquid benzene absorbing in the 1800-2100 cm region. The interaction of CR with the 0, and O4 oxide ions in the 12-R window, when benzene is located in the plane of this aperture, gives the highest VNH shift for the V5 + V,7 and V 10

+V 17 bands. It was shown that the wavenumber values could be related to

the charge on the oxide ions calculated from the Sanderson

eleCtrOnegatiVity

561 equalization principle

[14]. In addition the amount of benzene molecules

interacting at high loading with the 12-R window increased when one went from Li to Cs in the alkaline series and as the Si/Al ratio decreased (Table II). Table II. Determination by IR studies of the number of benzene molecules adsorbed on various zeolites from ref. L34.351.

Zeolite

NaYDI (c)

Si/Al

C6H6 per S.C.(a)

0

27 6 6

0 0

oxygen charge (b)

- 0.223 - 0.275 - 0.303

NaY RbY

2.4 2.4

0.6 + 0.3 0.6 + 0.1

- 0.350 - 0.380

NaX

1.2 1.2 1.2

0.45 + 0.25 1.2 + 0.2 1.2 + 0.1

- 0.410 - 0.449 - 0.463

RbX

csx

(a) number of benzene molecules per supercage interacting with 0 ions of the 12-R window (b) Calculated from Sanderson electronegativity (c) D means samples dealuminated with ammonium hexafluorosilicate This means that in the more basic zeolites, benzene molecules prefer to be located in the 12-R window where they can interact with basic oxide ions. This suggestion shows that, at least for faujasite where benzene molecules just fit the 12-R window, the amount and energy of adsorption of benzene may be used to characterize the 0 1 and O4 oxide ion basicity.

Raman Spectroscopy Adsorption of benzene on alkali metal-X and -Y zeolites was also studied by Raman laser spectroscopy [48] and W

reflectance spectroscopy [49]. It has been

shown that Raman shifts of the ring breathing mode of adsorbed benzene depend on the electrostatic fields within the supercage and that the excess cations in X versus Y results in higher fields at the aromatic nucleus of initially adsorbed benzene. Further increases in field are caused by crowding in the supercage as the size of cation increases. UV reflectance studies show the n electron interaction with the cation in the supercage : the position of the vibronic maxima with both the degree of surface coverage and the nature of metal ion exchanged.

NMRand neutron diffraction techniques They have been used with benefit for the study of benzene adsorption on Na-X and Na-Y zeolites [50,51].They show that benzene molecules are adsorbed in the 12-R window and on cations in SII sites, which supports the infrared data.

562 ESR technique When a surface exhibits electron donating properties with one electron transfer, radical anions may be formed. Such ions are paramagnetic and may be studied by ESR, this idea gave rise 20 years ago to many works concerning charge transfer complexes on oxides and zeolites [52-541. Molecules anthracene, the strength

benzene,

potential

to this aspect

low electron donating

into radical

of the‘oxidizing'electron

on the ionisation devoted

etc. were ionised

affinity

(reducing)

value

concerned were

as tri-, di-, mono-nitro

on

radical

benzene

sites

of the adsorbates. radical

adsorbed

properties

acceptor

cations

cations. oxides

or

being

evaluated

The majority

In contrast, zeolites

by ESR,

depending

of the works

if molecules

exhibiting

anions were detected

and tetracyanoethylene

as perylene,

as evidenced

by ESR.

of

electron

Adsorbates

were currently

used. For

instance anion radicals were detected for trinitrobenzene adsorbed on H-Y zeolite dehydroxylated tri-nitrobenzene

at 550-C

[55], for tetracyanoethylene

and di-or

adsorbed on dehydroxylated H-Y zeolite [56], for SO2 adsorbed

on dehydroxylated H-Y, H-M [57] and H-L

[58] and for triphenylamine

zeolites

adsorbed on dehydroxylated H-Y zeolite [59]. The structure of the anion sites formed upon dehydroxylation was postulated to correspond to AlO; anion : H+

2

Such a conclusion was supported by the observation by Naccache and Ben Taarit [SO]

that

electron

acceptor

properties

of NH4Y

zeolite

as measured

by

anthracene cation, were decreasing upon dehydroxylation.Note also that it was shown that oxides as alumina, silica-alumina, zeolite exhibit simultaneously both electron donating and electron accepting sites [61,62]. Some studies on alkaline zeolite (Nay -fCsY) showed low electron donating properties as evidenced by ESR, particularly

for

samples

dehydrated

non dehydroxylated. This indicates that the ESR technique

c

for characterizing

basic

sites as 0

2-

or OH- presumably

because

at

400°C

is not reliable of their too low

strength.

XPS techniaue This technique may be used with profit to characterize basicity. It is well established that factors, value

the

one being the charge

is decreasing

in a crude binding

binding

with

approximation

energy carried

the negative of

Madelung

energy value of the Ols peak

value

of

elements

by the element. charge

carried

potential is expected

depends

Usually

on

by the element

[63,64].

It follows

to decrease

different

the binding

when

energy

increasing that

the

the negative

563

on

charge effects

the

due

zeolites,

oxygen

atom

the

ejection

to

it is difficult

chemical

shift

obviously

of

needed

ADSORPTION

electrons

to determine

0 ls binding

has

SEPARATION

CATIONS

BY

with

to

be

effect.

OXIDE

considered

For instance

[65-671.

their 71 electrons

removal

of Hz0

the and

on the

above,

CO 2 from

ions

for directing of the

[681, para-xylene

equalization basicity

four

for

ethylbenzene

the

oxide

molecule

either

or is adsorbed adsorbed contents

acting

the

work

is

ROLE

OF

due to the existence

between

K-Y

aromatic

+

cation

other

it was possible

alkaline

to determine

ion

etc.

Since

the

form as discussed

is very

probably

of

for example

the

benzene

Sanderson

zeolite [70]. The

electronegativity

: Rb-X > K-Y > Na-Y

in

the

opposite

[71]. This

[19]. The direction

suggests

with the less basic aromatics of conjugate

acid-base

that and

pairs the

and Na-Y as the zeolite basicity both with the cations

molecules on

in Rb-X

for Nay-zeolites

can interact

adsorbed

the

olefins-paraffins,

cationic

ethyl

order

more strongly

of benzene

[65].

as for instance

Let us consider

increases

the aromatics

of

: ethyl benzene, para-, ortho- and

from

from Rb-X to

in alkaline

interaction

isomers,

basicity

gets

calculated

sites flat on Na

and

adsorbate

zeolitic

isomers one

limitation

are based on non geometric

aromatic

adsorption.

that

adsorbed

to the

etc.

selective

as Lewis acid sites

framework

zeolite

cations,

direct

or the separations

NaY

and the zeolite

it has

in the hexagonal

in the plane of the 12-R window

on Na-

may


decreases

case

ZEOLITES.

to any diffusion

and patents

on the the

isomers

ions, the interaction In the

of

certainty

accurate

atoms,

is due

[69] and meta-xylene

interact

[19]. Since

complex.

which

are in the following

four

In addition

Lewis acidity

increases

a

< para-xylene

reciprocally.

charging

and non acidic

interaction

in addition

flow

aromatic

ions

the more basic zeolites

cation

that

K-Y

principle of

a gas

shown

oxide

cationic

of one specific

selective C

It has been

the

cations

depends

suggested

importance

of

an

processes

adsorption

separation

charges

Such

ON

with accessible

aromatics-cycloparaffins,

oxide

it may be

meta-xylenes.

of

properties

and

further

ADSORPTION

paraffins,

with the metal

aromatics-olefins, charge

accuracy

and

are known to be preferentially

with

A large number of industrial favoring

because

insulating

IONS

in a mixture

olefins

by comparison

effects

enough

concerns

of adsorbates

of one adsorbate

faujasites

the

value

SELECTIVE

AND FRAMEWORK

The interaction

adsorption

and

energy

Most of the work done on adsorption materials.

Unfortunately

increased. of

in that field

AND

EXCHANGEABLE

the

is

150,511.

faujasites

been

the amount of benzene

is rather

shown that the

face of the supercage

In a study

with

and the

three

of benzene different

adsorbed

Al

on either

564 site and to show that the energy of the corresponding interaction depends on the cation Lewis acidity, the oxide ion basicity and the cation loading

t341 .

The interaction cation-benzene TT ring is smaller for instance for Rb than for Na and the interaction CH-framework oxide ions increases simultaneously with Al content, i.e. with its basicity. Hence the more basic Rb-X interacts more strongly with benzene

located

interaction with the cations

in the

12-R window

than with benzene

in

[35]. These results strongly suggest that the

interaction of any aromatic and more generally any adsorbate with zeolite framework depends not only on the molecule geometry and on the zeolite structure itself but also on the chemical properties of the zeolite. Since the only accessible atoms of the framework are cations and oxide ions, the charge on both i.e. the cation Lewis acidity and the oxygen basicity are factors to consider in adsorption processes.

BASIC REACTIONS IN CATALYSIS BY ZEOLITES The application of zeolites as acidic catalysts in reactions proceeding through

a carbonium

mechanism

(cracking,

isomerization,

alkylation

of

hydrocarbons, etc.) has received widespread attention in contrast with their use as basic catalysts. homogeneous bases

However,

for many industrial reactions catalyzed by

1721, the replacement of liquid bases by heterogeneous

catalysts may be an appreciable amelioration such as limitation of reactor corrosion, easy separation of used catalyst and its possible regeneration, etc. etc. Generally,

the

solid-base

catalyzed

reactions

occurring via

anion

intermediates are characterized by higher activity and selectivity than the solid-acid or metal catalyzed reactions. The mechanism of base catalyzed reactions and the active sites on solid bases have been discussed in some review articles

[5,6] [72-741. More particularly the reactions catalyzed by

basic zeolites have been reviewed by Ono [61. The most important ones are the side-chain alkylation of substituted aromatics, the dehydrogenation of alcohols, the ring transformation ofy

butyrolactone or tetrahydrofuran with hydrogen

sulfide or primary amines. Side-chain alkylation of substituted aromatics The acid-catalyzed alkylation of alkylaromatics with olefins or methanol results in selective alkylation of the benzene ring [74]. H+, di and trivalent cation exchanged zeolites are efficient catalysts for this ring alkylation while alkali exchanged zeolites are inactive in the same temperature range (< 25O'C) [75,76], However, Sidorenko et al. [77]

observed that at higher temperature

(425-475.C) the alkylation oE toluene with methanol over alkali cation exchanged zeolites produces a mixture of xylenes, ethylbenzene and styrene showing that the alkylation may occur both on the side-chain as over catalysts with basic

properties

[78] and

on

Yashima

al.

for the

et

[79]

over alkali exchanged depends

was greater

for X-type

methanol

conditions, low

of

yields

were

by methanol

X or Y type.

The

metal element

products.

atomatics

However,

confirmed

by

or formaldehyde

activity

Formaldehyde

zeolites.

similar

in CS

results

of toluene

of the alkali

than for Y-type

produced

These

ring.

alkylation

on the basicity

but the

aromatic

zeolites

alkylation

than

the

for

side-chain

(Na K Rb Cs) and

was more under

reactive

the

(styrene and ethylbenzene)

optimum

were

rather

Co

180,811

(of the order of 10 8). Appreciable

amelioration

with the addition zeolites.

of a boron or a phosphorus

Recently,

containing

Exxon Co patented

boron and phosphorus

In order

to explain

nature of the cation, reaction nature

methylal

CH30H

C6H5-CH

alkylating with

(77) was developped chemistry.

The

side-chain toluene implied

‘I-

ratio,

promoters, time,

[84].

+

(Fig.

the

for the alkylation

Experimental

have

of

of toluene

with

shown

this

that

(1)

= CH2 + Hz0

(2) (3)

is dehydrogenated

formed

reaction

in reaction (1). This

indicated

of toluene with

Ht-.-.-.-._._.-.-

model.

to

formaldehyde

(2) is hydrogenated

scheme

which

by Sidorenko

investigated

by using

a basic

site is necessary

and suggesreu

a model

for

of

acidic

the

is the

to ethylbenzene

as proposed

that

acidic and basic

configuration 2):

the

:

by Itoh et al. [85] and was

Fig. 2. Configurational

results

and

temperature,

-t CSH5 CH2CH3

formaldehyde

specific

of

toluene

of the

2 CO + Hz

calculations

alkylation

and

interactions

-

process

agent. Styrene by

[82].

exchange

(1) methanol

H 2 produced

exchange

X or Y-zeolite

to methylal

the

+ HCHO+CSH5-CH = CR2 + Hz

alkali

of

HCHO + H

In reaction

metals

gas , etc, were studied

HCHO C6H5-CH3

compounds

consecutively

+

to the

the use of alkali exchanged

the influence

[83] or methanol proceeds

compound

by Monsanto

the role of the alkali metal cation,

: contact

parameters

of the carrier

reaction

of these yields was claimed

et al. quantum to the

interaction

of

sites on the zeolite. This model and

basic

sites

with

steric

The basic site determines the selectivity to the side-chain alkylation whereas the

acidic

site

adsorbs

and

stabilizes

toluene.

This

was

confirmed

: the importance of the basic sites is generally demonstrated

experimentally

[77,80,81,85] and it has been recently shown that addition of an acidic cation to a basic zeolite accelerates side-chain alkylation [37]. For the same reaction Garces et al. [87] found that LiNa-X zeolite gives xylenes

(100

%) while

majoritarily authors

metals

The

constitute

and with

the

(dehydrogenation

and copper

of

metal

metal

that

the

for

the

toluene

for the metal

in

were

the

sample.

alkylation

of

and

may

the

of the alkyl

The

be

pores.

detected

oxides

for the activation

CsNa-X

adsorption

ions

vapors

give ethylbenzene

the

In

above

fact

600

K

tOluene with metal

vapors

groups

in the

of alkylaromatics. appears

of methanol)

such as boron

to be formaldehyde

has to be favored which

[BO, 811, silver

[85] [84b]

[85,87] the reaction is presumably

the role

, cobalt, iron, manganese

[82].

Similarly, with methanol converted

for

temperature

suggested

the alkylating agent

of promoters

yield

sites

first

the basic sites needed

Since

[881

conversion

active

the onset

authors

Na, K, Rb and Cs-X zeolites

by the reduction

CsNaX

side chain alkylation

(1)

the

generated

coincides

methanol.

highest

that

RbNaX

KNaX,

which

the

postulated

alkali over

with

the others

xylenes

[37] and

on the side chain

to

~1 and

and toluene

@

ethylbenzene

[79] and

ethyl

with ethylene

a

over and

RbX

B

and vinyl naphtalenes

zeolite were alkylated

methyl

naphtalenes

were

over KX and RbX zeolites

over RbX zeolite gave cumene

and

~1 methylstyrene

WI. Dehydrogenation of alcohols Transformation

of alcohols is a catalytic probe to determine the relative

importance of catalysis by acidic or basic sites. Dehydration products (olefins

or

ethers)

are

dehydrogenation products The decomposition

essentially

obtained

on

acid

catalysts

and

(ketones or aldehydes), on basic catalysts [5,73].

of isopropanol on acid zeolites proceeds via ionic

intermediates to propene and diisopropyl ether [74,89,90] the selectivity of the reaction over alkali cation.

Li-and

propene

whereas

and acetone

Addition

acetone

depends

on the content

that

or

depending

high selectivity -Y zeolites

(pyridine)

the dehydration

and the decrease show

exhibit

K-, Rb-, Cs-X

of a basic compound

respectively

results

zeolites

on

the

nature

of

for dehydration

convert

the to

isopropanol

to

[91].

poisoned

radius

exchanged

Na-X or -Y zeolites

[91,92]

propene

metal

the

and

and of an acidic

the

dehydrogenation.

of ion and increases

of electron

negativity

compound The

(phenol) yield

of

with the increase of ionic

of the alkali metal cation.

These

K, Rb, Cs zeolites exhibit both acidic and basic sites.

567 The acidic centers are suggested to be decationated sites and possibly the alkali cation and the basic sites are the lattice A10 - paired with alkali 4 cation [911. Similarly, isopropanol is converted to propylene on acidic sites 1931 and to acetone on basic sites [941. On K- and Cs- exchanged ZSM-5 and mordenite

[9511

the selectivity of this reaction depends on the reaction temperature and on the nature of the exchanged cation. At ZOO'C, formation of propylene is essentially observed whatever the zeolite or the exchanged cation. At 3OO'C, formation of acetone is detected in each case and is higher on Cs-ZSM-5 zeolite. These results suggest that acetone formation is favored on the more basic zeolite and at

high

temperature

due

to

the

higher

energy

of

activation

for

the

dehydrogenation reaction. The dehydrogenation of methanol over sodium modified silicalite was reported to give formaldehyde with high selectivity thermodynamically

[961 though this reaction is

unfavorable. Since the catalytic activity does not depend on

the aluminium content (present as impurities) but only on the sodium excess, it was assumed that the active sites are the sodium ions and not the aluminium ion and alkali-metal cation pairs as proposed by Yashima et al. [761 for the decomposition of isopropanol.

Oxygen-sulfur or nitrogen interchange reactions The interchange reactions may be resumed by the following scheme where furan is converted to thiophene or pyrrole by reacting respectively with H2S or NH3 [741.

NH

H2S

3 3OO'C

a, J

O

\

340'c

QG&J

For these reactions, the optimum activities are obtained with seolites of low OK

negligible acidity such as NaX or NaY zeolites. The

r

reaction

of

y was

thiobutyrolactone

butyrolactone

zeolites and acidic zeolites [6,97].

+

H2S

-

and

hydrogen

sulfide

to

give

studied on alkali metal cation exchanged X or Y

+

H20

It was shown that the alkali metal cation exchanged zeolites are much more active than the acidic zeolites, that the catalytic activity depends on the alkali metal cation in the order LiY
and that NaX is more

active than Nay. Kinetic study has allowed to calculate the rate constant k which increases in the order Li c Na < Cs and to determine the activation energy -1 for k to be 163, 130 and 109 kJ.mol for LiY, Na and CsY, respectively.The effect of addition of hydrogen chloride which inhibits the ring conversion and of pyridine which enhances the catalytic activity indicates that the active centers are associated with basic sites. As already proposed

[91], the basic sites are described as oxygen anions

bound to aluminium cations AlO;.

The negative charge of the site AlO; is

neutralized by the alkali metal cation which is more weakly bound to the basic site as its ionic radius increases. Hence the order of catalytic activity LiY < NaY < KY < RbY ( CsY may be correlated to an increasing basic strength of the AlO; sites, which is confirmed by the decrease of the activation energy. The higher activity of NaX as compared to NaY can be explained by the higher number and basic strength of AlO; sites due to its higher Al/(Al + Si) ratio [19]. For the ring transformation of tetrahydrofuran [6,9Sl, the alkali metal

into tetrahydrothiophene

cation exchanged zeolites also exhibited the higher

activity and the addition of hydrogen chloride inhibited the reaction. However, addition of pyridine greatly decreased the activity. Therefore,

for this

reaction, acidic and basic sites appeared essential. Since the ring opening of tetrahydrofuran

[99,100] necessitates Brijnsted acidity, it was assumed that

interaction of basic sites with hydrogen sulfide produces acidic OH groups which react with tetrahydrofuran according to the following scheme

H2S + Na+ + OZ-

n+oz+

Na+SB

-+

C-J -( 0

:

+ H+OZ-

+z-

m=p*) 0

R

A (II (1)

(I) + NaSH

+

C-J

+ H

0

2

+

Na’

S

The

first step was confirmed by infrared studies of hydrogen sulfide

adsorption on NaX which revealed that hydrogen sulfide dissociated on NaX [94] Among the other interchange reactions catalyzed by alkali metal cation exchanged zeolites, one may mention :

569

- the conversion of acetic anhydride to thioacetic acid [74] - the ring transformation of Y butyrolactone into I-alkyl pyrrolidinone [102] - the reduction of nitro compounds with hydrogen sulfide into amines [103] Other basic reactions involving anionic intermediates are mentioned such as the aldol condensation of n-butyraldehyde to 2 ethyl 2 hexanol over Na-, K- or H- Y zgolites [104], the dimerization of cyclopropenes over KA or NaA

[105],

the dialkylation of o-ethylphenol over NaX and the dehydrocyclisation

of

o-ethylphenol with COS over NaX [106]. Hydrocarbon aromatization on platinum alkaline zeolite The

aromatization

Pt/A1203-C1

through

of paraffins

a bifunctional

occurs

mechanism.

usually

on catalysts

such

as

It was shown that a monofunctional

scheme is operating on non acidic zeolites fully exchanged with alkaline cations and that L type zeolite is the most selective for aromatic formation [107]. Table

III. Dehydrocyclization

pressure,

46O'C)

Table to

% % % % %

nC6 conversion (wt%)

benzene

zeolites

that

was

the highest

observed

for

and selectivity

the alkali

metal

dependence

of CO adsorption

that besides

series

any cage

2.9 8.3 2.0 0.5 0.8

(atmospheric

zeolite

catalyst

specific

80 66 58 49

41.2 54.0 23.1 15.1 0.9

cyclodehydrogenation

with

to

respect

sulfur

on cation

P'c on

It was further

other

Na

shown that

when moving

from Li to Cs in

resistance

[110 1 and

and on zeolite

of the L type

by the chemical

(%)

for n-hexane

improved

A high

[108-1101

Benzene

MCP

6

Pt/A1203-Cl.

are greatly

Benzene selectivity

(wt %)

state of platinum

is modified

cation

This may be r-elated to the increase

[108-1101.

Pt catalysts

3.5 5.3 12.4 7.4 5.3 9.8 0.4 15 4 11.2

selectivity

[108,109].

effect

iC

c1-c5

Pt-KL

and to the usual acidic

the conversion

Yields

59 83 41 31 17

Pt/KL Pt/NaX Pt/NaY Pt/Na Pt/Na mordenite

III shows

on various

[107].

Catalyst

0.6 0.6 0.6 0.6 0.6

of n-hexane

framework,

properties

type the

large

suggested electronic

induced by the alkaline

of the basic

character

of

the framework oxygen from Li to Cs zeolites [19]. The importance of the absence of any acidity is also emphasized in more recent results obtained with Pt/BaKL zeolite prepared in such a way to locate the which precludes the formation of acidity

(111).

a

Ba ions in small cages,

The catalysts are very active

for both paraffin and alkyl cycloalkane aromatizations.

570 CONCLUSION Basic type reactions in heterogeneous catalysis by zeolites remain of low practical interest at the present time by comparison to acidic reactions, presumably because there have been too few studies on that subject. Basic type zeolites or molecular sieve type materials with a cationic framework have not yet been synthesized but basicity may be obtained in anionic lattices by alkalineion exchange, the resulting basic strength depending

on

the

nature

and

size of cations. Acidic and basic sites are shown to coexist simultaneously. Basicity is a feature rather difficult to characterize. Physical techniques and the use of acidic probes are useful to try to quantify basic sites with regards to their strength, nature and amount. Lattice oxide ions which bear a more or less large negative charge, AlO; species and hydroxyl groups are the three main basic sites to consider. Their characterizationby physical technique is actually not

straightforward

but

is

useful for the understanding of basic

type reactions. More emphasis on basic type reactions zeolite should be done with promising applications in the future for new catalytic selectivities.

REFERENCES J.N. Briinstedand

K.

Pedersen, J. Phys. Chem., -108 (1924) 185.

L.P Hammett and A.J. Deyrup, J. Am. Chem. Sot., 54 (1932) 2721. E.M. Flanigen, B.M. Lok, R.L. Patton and S.T. Wilson, Pure and Appl. Chem., 58 (1986) 1351 and references therein. 4

K. Tanabe, Solid Acids and Bases, Academic Press, New York, 1970.

5

H. Pines, The Chemistry of Catalytic Hydrocarbon Conversion, Academic Press, New York, 1981, p. 123.

6

Y. Ono in B. Imelik et al. (eds.), Catalysis by Zeolites, Elsevier, Amsterdam, 1980, vol. 5. p. 19.

7

K. Tanabe in J.R. Anderson and M. Boudart (eds.), Catalysis: Science and Technology, Springer-Verlag,Berlin, 1981, vol. 2, chap. 5.

6

K. Tanabe in B. Imelik et al. (eds.), Catalysis by Acids and Bases, Elsevier, Amsterdam, 1981, vol. 20, p. 1.

9

S. Malinowski and J. Kijenski, Specialist Chemical Society, London 1981 p. 130.

10

S. Beran and J. Dubsky, J. Phys. Chem., 83 (1979) 2538.

11

W.J. Mortier, P. Geerlings, C. Van Alsenoy and H.P. Figeys, J. Phys. Chem., 83 (1979) 855. --

12

L. Pauling, The Nature of the Chemical Bond, Cornell Univ. Press, Ithaca, 1939, c. 74.

Periodical

Reports, The

571

13

Structural F. Liebau, 1985, p. 46.

Chemistry

14

R.T. Sanderson, 1976.

15

T.

16

K.I.

Tanaka

17

W.J.

Mortier,

18

W.J. P.A. Jacobs, -40 (19781 1919.

19

D. Barthomeuf,

20

E.G.

21

s. Malinowski and J. London, -4 (1981) 130.

22

C.

Chemical

bonds

Yamanaka and K. Tanabe,

J.

and K. Tamaru, J.

Catal.,

J.

Derouane

P.

Phys.

Bull.

Mortier

Phys.

Chem.,

Fripiat,

J.

and

-SO (1976), Japan,

Press,

New York,

1723.

-37 (1964)

Uytterhoeven,

(1984)

1662.

Chem.,

Inorg.

-91 (1997)

Catalysis,

D.

J.

Nucl.

Chem.

42.

Phys.

Kijenski,

Pichat

Academic

Berlin,

138.

J.B.

8

Springer-Verlag,

energy,

Chem. Sot.

and

Chem.,

Silicates,

and bond

-55 (1978),

and J.G.

Mirodatos,

of

Royal

Barthomeuf,

J.

145.

Society

Phys.

Chemistry,

Chem.,

-80

(1976)

1335.

23

P.A. Jacobs, F.H. Van Cauwelaert, Chem. Sec., -69 (1973) 1056.

24

L. Bertsch

25

J.W.

26

R.M. Barrer and R.M. and -b 61 (1965) 948.

27

R.M. 162.

28

A.V. Kiselev and A.A. London, 1968 p. 252.

29

H.J. Van Spangenberg, K. Fielder, Chem. Leipzig, 248 (1971) 49.

30

E. Dempsey,

31

B. Barrachin (1986) 1953.

32

O.V. Krylov and -8 (19551 246.

33

P.O.

34

A.

35

A. de Mallmann,

36

Y. Barbarin

and H.W. Habgood,

Ward and H.W. Habgc-od,

Scokart

and

Cohen

E.A.

Trans.

H.J.

Chem. Ind., J.

and P.G. Rouxhet,

Bull.

Sot.

Chem.

Kinet.

Chim.

Uytterhoeven,

-J.

1621.

1179.

SOc.,

Academic

Ortlieb

Problemy

Fokina,

(1963)

-70 (1966)

Molecular

Lara,

and J.B.

Faraday

Minerals,

Sot. de

-67

Chem.,

Lopatkin,

Sieves,

Vansant

Chem.,

Phys.

and Clay

Molecular E.

J.

Phys.

Gibbons,

Zeolites

Barrer,

J.

E.F.

a 59 --

Press,

Sieves,

London,

Sot.

and W. Schirmer,

London, Sot.,

Katal.,

(1963)

1968 p.

2569

1978,

Chem.

p.

Ind.,

Zeit. Phys.

293.

Faraday

Tran6.

Akad.

Nauk,

II, -82 Moskva,

Belge, -90 (1981) 983.

De blallmann and D. Barthomeuf in Y. Murakami, A. Iihima and J.W. Ward, teds.), Proceed. of 7th Intern. Zeal. Conf., Kodansha, Tokyo, 1986, p. 609. PhD Thesis,

Paris,

and D. Barthomeuf,

1986.

to be published.

572 37

H. Itoh, T. Hattori, K. Suzuki, A. Miyamoto and Y. Murakami, J. Catal., 72, (1981) 170. r Ii.Itoh, T. Hattori, K. Suzuki and Y. Murakami, J. Catal. -79 (1983) 21.

38

S.S. Khvoshchev, V.E. Skazyvaev and E.A. Vasiljeva, in L.V. Rees (ed.), Proc.

a.

of the 5th. Intern.

Confer.

on Zeol., Naples,

London,

1980, p. 476.

39

P. Cartraud, 8. Chauveau, Analysis, -11 (1977) 51.

40

S.S. Khvoshchev, V.E. Skazyvavev and S.P. Shdanov, Izv. Akad. Nauk SSSR, ser. Khim, -1 (1978) 29.

41

T.R.

42

D. Barthomeuf and B.H. Ha, J. Chem. SoC., Faraday Trans. I, -69 (1973) 2158.

43

M. Bernard

Heyden,

and A. Cointot,

J. Thermal.

Brueva, A.L. Klachko-Gurvich and A.M. Rubinshtein, 1s~. Akad. Nauk. SSSR, ser. Khim, -12 (1972) 2807.

A. Auroux and J.C. Vhdrine, Bases, 311.

Elsevier,

Amsterdam:

in B. Imelik et al. (eds.), Catalysis by Acids and Studies in Surf. Sci. and Catal., series, 20 (1985)

44

E.D. Garbowski and C. Mirodatos, J. Phys. Chem., 86 (1982) 97.

45

Y. Delaval, R. Seloudoux and E. Cohen de Lara, J. Chem. Sot., Faraday Trans. I, 82 (1986) 365. --

46

A. Bielanski and J. Datka, J. Catal., 2

47

M.L. Unland, J. Phys. Chem., 82 (1978) 580.

40

J.J. Freeman and M.L. Unland, J. Catal., -54 (1978) 183.

49

M.L. Unland and J.J. Freeman,

50

H. Lechert and K.P. Wittern, Ber. Busenges Phys. Chem., -82 (1978) 1054.

51

A.N. Fitch, H. Jobic and A. Renouprez, J. Phys. Chem., 90 (1986) 1311.

52

B.D. Flockhart, C. Naccache, J.A.N. Scott and R.C. Pink, (1965) 341.

53

B.D. Flockhart in Surface and Defect Properties of Solids, Special. Period. Report, The British Chemical Society, London, 1 (1973) 69.

54

M.I. Loktev and A.A. Slinkin, Russian Chem. Rev., 45 (1976) 87.

55

J. Turkevich and Y. Dno, Adv. in Catal., -20 (1969) 135.

56

B.D. Flockhart, L. McLaughlin and R.C. Pink, J. Catal., -25 (1972) 305.

57

Y. Ono, H. Tokunaga and T. Keii, J. Phys. Chem., -79 (1975) 752.

58

Y. Ono, M. Kaneko, K. Kogo, 8. Takayanagi and T. Keii J. Chem. Sot., Faraday Trans. I, -72 (1976) 2150.

59

D.N. Stamires and J. Turkevich, J. Am. Chem. Sot., -06 (1964) 749.

60

C. Naccache and Y. Ben Taarit, VI Seminario Latin0 American0 de Quimica, Chile (1974) p. 8.

(1934) 183.

J. Phys. Chem. 82 (1978) 1036.

Chem.

Commun.,

573 61

B.D. Flockhart, I.R. Leith and 66 (1970) 469.

and R.C. Pink, Trans.

62

B.D. Flockhart,

63

W.N. Delgass, J.C. Vedrine,

64

X.M. Minachev, Nauk. SSSR Ser. T.L. Barr in D. and in Practical New York, 1983,

65

A.V. Kiselev,

66

A.V. Kiselev in L.V. Rees (ed.). Proceed. Naples, Heyden, London, 1980, p. 400.

67

J. Sauer,

68

P.R. Geissler,

69

R. Beardent

70

R.W. Neuzil,

71

G.A. Olah, Act. Chem.

72

H.

73

G.V. Krylov 116.

in Catalysis

74

P.A. Venuto

and P.S. Landis,

75

T. Yashima, 273.

H. Ahmad,

76

T. Yashima, (1970) 151.

77

Y.N. Sidorenko, P.N. Galich, V.S. Gutyrya, V.G. Il'in and I.E. Neimark, Dokl. Akad. Nauk. SSSR 173 (1967) 132; O.D. Konoval'Chijov, fi. Galich, U.S. Gutyrya and G.P. Zugowskaya, Kin. i. Kat.,2- (1968) 1146.

78

K. Tanabe, 0. Takahashi (1977) 347.

79

T. Yashima,

K. Sata and N. Hara, J.Catal.,

80

M.L. Unland

and G.E. Barker,

U.S. Patent -4 -115 424 (1978).

81

M.L. Unland

and G.E. Barker,

140 726 (1979). U.S. Patent 4 --

82

H.C. Lin and R.J. Spohn,

83

C. Lacroix, (1984) 473.

L. McLaughlin

Faraday

and R.C. Pink, J. Catal.,

Sot., -65 (1969) 542

25 (1972) 305.

and C.S. Fadley, Catal. Rev., 4 (1970) 179. et Spectrosc. Electron. -1 (197x) 285.

T.R. Hughes J. Microsc.

G.V. Antoschin, E. Schapiro and T.A. Navrousov, Izv. Akad. Khim. 9 (1973) 2131 and 2134. Briggsand M.P. Seat, (eds.), Applic. Surf. Sci., 15 (1983) 1 Surface Anal. by Auger and Photoelectron Spectroscopy, Wiley, p. 283.

Adv. Chem. Ser.,

K. Fiedler,

m

W. Schirmer

U.S. Patent

of the 5th Intern. Conf. on Zeolites,

and R. Zahradnik,

ibid. p. 501.

4 175 099 (1979). ---

Jr and R.J. Defoe, U.S. Patent

(1971) 37.

U.S. Patent

3 686 343 (1972). ---

4 326 092 (1982).

Res., -4 (1971)

240.

Base-catalyzed reactions of hydrocarbons Pines and W.M. Stalick, related compounds, Academic Press, New York, London 1977. by Non-metals,

Adv. on Catal.,

K. Yamazak,

K. Yamasak,

Academic

M. Katsuta

H. Ahmad,

and

M.

Press, New-York,

1970, p.

-18 (1968) 259. and N. Hara, J. Catal.,

Katsuta

H. Hattori,

and

React.

5

and N. Hara, J.Catal.,

Kinet.

Catal.

(1970)

-17

Lett., -7

-26 (1972) 303.

U.S. Patent -4 483 936 (1984).

A. Deluzarch,

A. Kiennemann

and A. Broyer.

J. Chim. Phys., -81

574

04

a. C. Lacroix, A. Deluzarch, A. Kiennemann and A. Br0yer.J. Chim. Phys., 81 (1984) 481. b, C. Lacroix, A. Deluzarch and A. Kiennemann, Zeolites, 4 (1984) 109.

a5

H. Itoh, A. Miyamato and Y. Murakami, J.Catal., g

87

J.M. Garces, G.E. Vricland, S.I. Bates and F.M. Scheidt, in B. Imelik et al. (eds.) Cata1ysis by Acids and Bases, Studies in Surf. Sci. and Catal. Series, l_o (1985) 67.

87

M.L. Unland and G.E. Barker, in W.R.Moser (ed.), Catalysis of Organic Reactions, Chem. Ind. Ser. vol. 5 (1981) p. 51.

a0

O.D. Konoval'Chikov, P.N. Galich, V.S. Gutyrya, G.P. Lugovakay, Kin. i Kat., -9 (1968) 1387.

89

P.A. Jacobs, Carboniogenic Activity of Zeolites, Elsevier, Amsterdam 1977, p. 99.

90

R. Rudham and A. Stockwell, Catalysis vol. I, Specialist Periodical Reports, The British Chemical Society, London, 1977, chap. 3, p. 87.

91

T. Yashima, H. Suzuki and N. Hara, J. Catal., 33 (1974) 486.

92

P.A. Jacobs, M. Tielen and J.B. Uytterhoven,-J. Catal. 50 (1977) 98.

93

K. Kochloffel and H. Knszinger, in J.W. Hightower (ed.) Proc. 5th Int. Congr. Catal., North Holland Publ., Amsterdam, 1973, p. 1171.

94

S. Siddhan and K. Narayaman, J. Catal., &

95

J.B.-Ngy, J.P. Lange, A. Gourgue, P. Bodart and Z. Gabelica in B. Imelik et al. (eds.), Catalysis by Acids and Bases, Elsevier, Amsterdam, 0 (1985) 127.

96

Y. Matsuma, K. Hashimoto and S. Yoshida, J. Catal., 100 (1986) 392.

97

K. Hatada, Y. Takeyma and Y. Ono, Bull. Chem. SOC. Japan, 50 (1977) 2517.

98

Y. Ono, T. Mor and K. Hatada, Acta Phys. Chem., 2

99

K. Fujita, K. Hatada, Y. Ono and T. Keii, J.Catal., 35 (1974) 325.

100

Y. Ono, K. Hatada, K. Funita, A. Halgni and T. Keii, J. Catal., 41 (1976) 322.

(1980) 284.

(1979) 405.

(1978) 233.

101 K.G. Karge and J. Rask, J. Colloid. Interface Sci., z

(1978) 522.

102 K. Hatada and Y. Gno, Bull. Chem. Sot. Japan, 51 (1978) 448. 3 253 038 (1966). 103 J.J. Wise, U.S. Patent --104

Ya.

I. Isakov, Kh. M. Minachev and N. Ya Usachev. Bull. Acad. Sci. USSR, Div. Chem. Sci., (1972) 1124.

105 A.J. Schipperijin and J. Lukas, Rec. Trav.

Chim.

Pays Bas, 92 (1973) 572.

106 D.E. Boswell and P.S. Landis, Bull. H-J. Acad. Sci., -11 35 (1966). 107

J.R. Bernard, in L.V.C. Rees (ed.), Proceed. 5th Intern. COnf. ZeOliteS, Naples, Heyden, London (1980) 686.

575

108

C. Bezuhanova, J. Guidot, D. Barthomeuf, sot., Faraday Trans. I, z (1981) 1595.

M. Breysse

and J.R. Bernard,

J. Chem.

109

C. Bezuhanova, M. Breysse, J.R. Bernard and D. Barthomeuf, in T. Seiyama and K. Tanabe (eds.), Proceed. 7th Int. Cong. Catalysis, Kodansha, Tokyo, (19801, 1410.

110

C. Bezuhanova, M. Breysse, J.R. Bernard and D. Barthomeuf, in B. Delmon Froment (eds.), Catalyst Deactivation, Elsevier, Amsterdam, 1980, 201.

111

T.R. Hughes, W.C. Buss, P.W. Tamm and R.L. Jacobson, in Y. Hurakami, A. Iijima and J.W. Ward (eds.), Proceed. 7th Intern. Zeolite Conf., Kodansha, Tokyo (1986). 725.

and G.F.