comparison of the methods for the determination of surface acidity of solid catalysts

This article was downloaded by: [McGill University Library]
On: 15 December 2012, At: 07:35
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number:
1072954 Registered office: Mortimer House, 37-41 Mortimer Street,
London W1T 3JH, UK
Catalysis Reviews: Science
and Engineering
Publication details, including instructions for
authors and subscription information:
/>Comparison of the Methods
for the Determination of
Surface Acidity of Solid
Catalysts
Lucio Forni
a
a
Istituto di Chimica Fisica Universita di Milano,
Milano, Italy
Version of record first published: 13 Dec 2006.
To cite this article: Lucio Forni (1974): Comparison of the Methods for the
Determination of Surface Acidity of Solid Catalysts, Catalysis Reviews: Science
and Engineering, 8:1, 65-115
To link to this article: />PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: />terms-and-conditions
This article may be used for research, teaching, and private study
purposes. Any substantial or systematic reproduction, redistribution,
reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make
any representation that the contents will be complete or accurate or

up to date. The accuracy of any instructions, formulae, and drug doses
should be independently verified with primary sources. The publisher
shall not be liable for any loss, actions, claims, proceedings, demand, or
costs or damages whatsoever or howsoever caused arising directly or
indirectly in connection with or arising out of the use of this material.
Downloaded by [McGill University Library] at 07:35 15 December 2012
Comparison
of
the Methods
for
the
Determination
of
Surface Acidity
of
Solid
Catalysts
LUCIO
FORNI
Istituto di Chimica
Fisica
Universitii
di
Milano
Milano.
Italy
I
.
INTRODUCTION

66
I1
.
DETERMINATION
OF
THE ACID STRENGTH
OF
THE
CENTERS

67
A
.
Method
of
Adsorption
of
Colored Indicators

67
B
.
Spectrophotometric Method

69
C
.
Adsorption
of
Gaseous Basic Substances

71
D
.
Calorimetric Methods

73
E
.
OtherMethods

78
I11
.
DETERMINATION
OF
THE SURFACE DENSITY
OF
ACID CENTERS

80
A
.
80
B
.
Titration after Ionic Exchange

81
C

.
Titration with Bases in Nonaqueous Solvents

82
D
.
Calorimetric Titration

87
E
.
Adsorption and Desorption
of
Gaseous Bases

88
91
94
96
Direct Titration
of
Aqueous Suspensions

F
.
Method
of
the Poisoning
of
Specific Surface Reactions

G
.
Hydrogen-Deuterium Exchange Reactions

H
.
Indicator Reactions Method

I
.
Spectroscopic Methods

98
J
.
Reaction with Hydrides

100
K
.
OtherMethods

102
65
Copyright
0
1973
by Marcel Dekker. Inc
.

All
Rights Reserved
.
Neither this work nor any part
may
he
reproduced or transmitted in any form or by any means
.
electronic
or
mechanical. includ-
ing phococopying. microfilming
.
and recording.
or
by any information storage and retrieval sys-
tem. without permission in writing from the publisher
.
Downloaded by [McGill University Library] at 07:35 15 December 2012
66
L. FORNI
IV. DETERMINATION
OF
THE NATURE
OF
ACID SITES:
BRBNSTED TYPE AND
LEWIS
TYPE

103
A.
Determination
of
Brhsted Sites Alone

103
B.
Determination
of
Lewis Sites Alone

104
V.
CONCLUSIONS

108
REFERENCES

111
I.
INTRODUCTION
The concept of surface acidity was originally introduced with
the
aim
of
justifying the presence of some substances formed in catalytic
chemical reactions, not
as
a

consequence
of
suppositions about
the
nature of surface-active
sites
of
solid catalysts. The formation of
such substances in some reactions (e.g., cracking, isomerization,
or
polymerization) can be better explained by admitting the forma-
tion of reaction intermediates having the structure of
a
carbonium
ion, which can
be
formed by interaction between
the
reacting sub-
stance (hydrocarbon) and an acid center.
As
an example, in the
cracking of alkylaromatics catalyzed
by
decationated zeolites, the
following reaction mechanism
is
generally accepted:
where the
first

stage can be interpreted
as
an electrophilic substi-
tution of the proton onto
the
alkyl group.
A
complete description of the surface acid properties of
a
solid
must involve the determination
of
the acid strength
of
the sites, their
density (number
of
acid centers per unit surface area of
the
solid),
and their nature (Bransted or Lewis type). Such
a
description
is
not
easy to make, since
the
strength and the density
of
the

sites are
generally strictly connected to each other and, besides, the distribu-
tion of the acid strength
is
usually heterogeneous. Furthermore,
most
of
experimental methods can distinguish the centers only on the
grounds
of
their strength. They cannot distinguish between Bransted
and Lewis centers, but simply give
a
measure of total acidity of both
types.
lished in
the
literature
[I-31.
Nevertheless, none of them gives
a
complete
list
of
the
available methods. The scope of the present
Some excellent reviews on the argument have recently been pub-
Downloaded by [McGill University Library] at 07:35 15 December 2012
SURFACE ACIDITY
OF

SOLID CATALYSTS
67
paper
is
then
a
collection and
a
critical comparison of
all
the methods
actually employed for the determination
of
surface acidity of solids.
Each of them, in fact, taken by
itself,
allows some useful informa-
tions to be collected, but can give
rise
to some criticisms. The com-
bination of the information obtainable from
two
or
more of them can
often be the only way to give
a
complete picture of
the
surface acid
properties of the solid under examination.

11. DETERMINATION
OF
THE ACID STRENGTH
OF
THE CENTERS
According to Walling
[4],
the
acid strength of
a
solid can be de-
fined
as
its
ability to convert
a
neutral base, adsorbed on
its
sur-
face,
into the corresponding conjugated acid.
If
the reaction
takes
place through the transfer of
a
proton from the solid surface to the
adsorbed molecule (Brdnsted acidity) or of an electron pair from
the adsorbed molecule to
the

solid surface (Lewis acidity), the acid
strength can be expressed, respectively, by means of the Hammett
function
H,
in the following way
[5-71:
H,
=
pK, + log ([Bl/[BH+l)
(1)
or
where
K,
is
the equilibrium constant of the dissociation of the acid,
and [B], [BH'], and [AB]
are
the concentrations of neutral base,
its
conjugated acid, and the addition product formed during the adsorp-
tion of
the
base
on the Lewis center, respectively.
A. Method of Adsorotion of Colored Indicators
An immediate application of Walling's analysis, originally adopted
by Walling, Weil-Malherbe, and Weiss
[a,
Ikebe
et

al.
[9], and many
others,
is
the
observation of the color shown
by
suitable indicators
adsorbed on
the
solid surface, If
the
adsorbed indicator assumes
the color of
its
acid form,
the
value of
H,
of the surface
is
lower
or
equal to the pK, of the indicator. The lower the value of
H,
(and
the
lower the pKJ,
the
higher

is
the
acid strength of the solid. Benzene,
isooctane, decalin, or cyclohexane may
be
employed
as
solvents. In
Table
1
the most important indicators are reported. In the last
column of
the
table the wt% of H,SO, in sulfuric acid solution, which
has the acid strength corresponding to the respective pK,,
is
given
for some of the indicators. In Table
2
the acid strength, obtained by
such
a
method by various authors
[lo-131,
is
given.
Downloaded by [McGill University Library] at 07:35 15 December 2012
68
L.
FORNI

TABLE
1
Basic Indicators Used for the Measurement of Acid
Strength of Solids
Indicator
Neutral red
p-Ethoxychrysoidin
Methyl red
Phenylazonaphth ylamine
Aminoazox ylene
p-Dimethylaminoazobenzene
(dimethyl yellow or butter
yellow)
2-Amino-5-azotoluene
1,4-Diisopropylaminoan-
Benzeneazodiphenylamine
4-Dimethylaminoazo-1-naph-
Crystal Violet
p-Nitro benzeneazo-#( p’-nitro)-
diphenylamine
Dicinnamalacetone
Benzalacetophenone
Anthraquinone
thraquinone
thalene
Color
Base form Acid form pK, H,SO,,
5%
Yellow Red +6.8
8

X
Yellow Red
+5.0
Yellow Red +4.8
Yellow Red +3.5
-
-
Yellow Red +4.0 5
x
10-5
-
Yellow Red +3.3
3
x
10-4
Yellow
Red +2.0 5
x
10-3
-
Blue Red
+1.7
Yellow Purple +1.5 2
X
lo-’
t1.2
3
x
lo-’
Yellow Red

Blue
0.1
Yellow
+0.8
Orange
Purple +0.43
-
Yellow
Red
-3.0
48
Colorless
Yellow -5.6
71
Colorless
Yellow
-8.2
90
TABLE 2
Acid Strength of Some Solids
Solid acids
~
Hll
Refs.
Original kaolinite
Hydrogen kaolinite
Original montmorillonite
Hydrogen montmorillonite
Silica-alumina
Silica-magnesia

1.0
mmole/g H3B03/Si0,
1.0 mmole/g H3P04/Si02
1.0 mmole/g H2SO4/Si0,
NiSO,
.
xH,O heat-treated (350°C)
NiSO,. xH,O heat-treated (460°C)
ZnS heat-treated (300°C)
ZnS heat-treated (500°C)
ZnO heat-treated (300°C)
TiO, heat-treated (400°C)
AI,O,-B,O3
-3.0
-
-5.6
-5.6
-
-8.2
+1.5
-
-3.0
-5.6
-
-8.2
<-8.2
<-8.2
+1.5
-
-3.0

t1.5
-
-3.0
-5.6
-
-8.2
<-8.2
+6.8
-
-3.0
+6.8
-
+1.5
+6.8
-
+4.0
+6.8
-
+3.3
+6.8
-
C3.3
+6.8
-
+1.5
10
10
10
10
10

10
10
10
10
10
11
11
12
12
13
13
Downloaded by [McGill University Library] at 07:35 15 December 2012
SURFACE ACIDITY OF SOLID CATALYSTS
69
In the case of black or dark-colored solids, when the observation
of
the
indicators color
is
impossible or very difficult,
a
small amount
of
a
white solid of known acidity may
be
added
to
the
sample and

the
acidity value obtained corrected for such
an
addition.
reproducibility to be obtained, although some difficulty may arise
either in the determination of the exact end point of titration or due
to moisture contamination. Also,
the
acidity values obtained are not
absolute because they are not related to energetic factors, but simply
to
the
pK, values of the indicators employed. Other disadvantages of
the method
are
the
impossibility of making acidity determinations in
the
real
working conditions of the catalyst and sometimes the long
period required for the equilibrium between adsorbed and free base
to be reached.
The adsorption method
is
generally quite accurate and allows good
B.
Spectrophotometric Method
Since
the
visual judgment of

the
color shown by the indicator in
the
preceding method can be uncertain
at
times, some absorption
spectra of dyeing materials, adsorbed on various solids, have been
determined
[13,14].
For example, Leftin and Hobson [13] recorded
the
absorption spectra of phenylazonaphthylamine (pK,
=
+4.0)
on
a
12%
alumina silica-alumina catalyst for both
the
basic and acid form
of
the
indicator. Such spectra were recorded in pure isooctane and
in an ethanolic solution, acidified with HC1, respectively. The
re-
sults
are
reported in
Fig.
1.

One can observe that the spectrum of
the
adsorbate reveals
that
it
is
adsorbed exclusively
in
its
acid form.
In
a
similar way Dzisko and co-workers
[15]
determined
the
sur-
face acid strength of mixtures of oxides. They employed
the
indica-
tors reported in Table
3
and established the following qualitative
scale
of
acid strength: SiO,
.
A1,0,
>
ZrO,

.
SiO,
-
GqO,
.
SiO,
>
Be0
.
SiO,
-
MgO
.
SiO,
>
Y,O,
.
SiO,
>
L+O,.
SiO,
>
SnO
.
SiO,
-
PbO
.
SiO,.
Finally, Kobayashi [16-191 recorded

the
absorption spectra of
dimethyl yellow, dimethyl
red,
and bromophenol blue adsorbed over
partially n-butylamine covered silica-alumina in
a
nonpolar solvent.
He determined not only
the
acid strength, but also the total number
of acid centers present on
the
catalyst surface. Apart from some
discrepancies due to
the
change in the activity of
the
adsorbed base
when surface coverage became
higher
and higher, he confirmed
that
the
values of
H,,
obtained by this method,
are
independent on
the

nature of the indicator employed.
The spectrophotometric method gives good qualitative informa-
tions on the form in which
the
dyeing substance
is
adsorbed onto the
solid surface, but does not eliminate
the
main disadvantages con-
Downloaded by [McGill University Library] at 07:35 15 December 2012
70
L.
FORNI
0.8
-
0,
u
C
0
0.6
-
+
0,
2
0.4
-
0.2
-
C

4500
A
6000
FIG.
1.
Absorption spectra
for
phenylazonaphthylamine. (a) In isooctane solution,
(b) In ethanolic
HCI,
and (c) adsorbed on silica-alumina
[
131.
TABLE
3
Basic Indicators for Spectrophotometric Determination
of
Acid Strength
Indicator PK,
Phenylazonaphthylamine
pDimethylaminoazobenzene
Aminoazobenzene
Benzeneazodiphenylamine
pNitroaniline
o-Nitroaniline
p-Nitrodiphenylamine
2,4-Dichloro-6-nitroaniline
p-Nitroazobenzene
2,4-Dinitroaniline
Benzalacetophenone

p-Benzoyldiphenyl
Anthraquinone
2,4,6-Trinitroaniline
3-Chloro-2,4,6-trinitroaniline
p-Nitrotoluene
Nitrobenzene
2,4-Dinitrotoluene
+4.0
+3.3
t2.8
t1.5
+1.1
-0.2
-2.4
-3.2
-3.3
-4.4
-5.6
-6.2
-8.1
-9.3
-9.7
-10.5
-11.4
-12.8
Downloaded by [McGill University Library] at 07:35 15 December 2012
SURFACE ACIDITY OF SOLID CATALYSTS
71
nected with the adsorption method, e.g., nonabsolute acidity value
determinations and nonreal working conditions of the catalyst.

C. Adsorption
of
Gaseous Basic Substances
The strength with which
a
base adsorbs on the surface acid cen-
ters of
a
solid
is
directly proportional to the acid strength of the
centers. If,
after
the adsorption, the solid
is
heated
at
growing tem-
peratures and the quantity of desorbed base
is
recorded,
a
measure
of
the acid strength of the centers can be obtained. Before the exper-
iment the solid must be pretreated in order to obtain reproducible
results. Such
a
treatment usually consists in the elimination of the
volatile impurities by evacuation and/or heating and flushing in an

inert gas flow. The
less
volatile impurities can often be eliminated
by converting them in more volatile compounds by reaction with
oxygen
or
hydrogen. The adsorption equilibrium of the base can
often
be
reached
at
relatively high temperatures and low pressures.
A
real
chemical reaction of the base with the surface may also occur.
Peri
[20] and Wilmot
[21],
for example, showed
that
an exchange
re-
action between ammonia and OH surface groups, with the formation
of water and
NH,
surface groups, may take place together with the
adsorption of ammonia. Such
a
side reaction, on the other hand,
usually wastes but

a
small fraction of the adsorbed ammonia, so
that
it
is
difficult to determine the amount of ammonia consumed in
this
way by simply analyzing the gaseous phase, particularly
if
the
volume of the gaseous phase
is
large.
A
method
for
reducing the
error due to such
a
side
reaction
is
to make
a
series of cyclic ad-
sorptions and desorptions on the sample by varying the temperature
or
the pressure of the system. In this way
it
is

likely
that
only the
reversibly adsorbed ammonia takes part in such cycles.
The measures of acid strength can
be
performed by determining
the amount of desorbed ammonia obtained by heating in vacuo [22,23],
or in
a
closed system
[24],
or by flash desorption in an inert gas
flow [25-271. Webb [23] worked with HF-A1,0, samples
at
various
HF
percentages. After outgassing and dehydrating
at
500°C and’
lo*
Torr for 16
hr,
the solid was exposed for 30 min
at
175°C to
10 Torrs pressure of gaseous ammonia. The desorption was made
by evacuating the sample up to 500°C and collecting the desorbed
ammonia in
a

liquid-nitrogen cooled trap. From the difference in
weights of adsorbed and desorbed base, the amount of base remained
on the solid surface could be determined. The results are reported
in Fig.
2.
One can observe that the higher the HF percentage in the
solid, the higher the amount of adsorbed ammonia. This means
that
the acid strength of the solid
is
directly proportional to the HF frac-
tion.
Downloaded by [McGill University Library] at 07:35 15 December 2012
72
L.
FORNI
0.8
0.6
0.4
0.2
0.0
200
3
00
400
0~500
FIG.
2.
Fraction
of

ammonia retained on catalyst surface
vs
evacuation temperature.
(A)
0.0%
HF
content,
(0)
3.23%,
(A)
0.65%,
(0)
6.46%
[
231.
The amount of base desorbed by heating
a
previously covered
solid may also
be
determined by differential thermal analysis (DTA).
In
such
a
case, by combining DTA with thermogravimetric analysis
(TGA),
it
is
possible to make
a

simultaneous determination of both
the acid strength and the distribution of the centers
as
a
function of
their
acid strength. One of the best examples of such
a
procedure
is
given in
a
paper by Shirasaki et
al.
[28].
They worked on silica-
alumina covered with pyridine, n-butylamine, or acetone. By plotting
both the changes
of
solid temperature [with respect to the reference
sample (DTA)] and of solid weight (TGA) vs temperature (see Fig.
31,
they simultaneously determined the amount
of
adsorbed base (x) and
adsorption
heat
(S).
By plotting
S

vs x one can get (dS/dx), which
is
directly proportional to the acid strength
of
the centers. By plotting
x
vs (dS/dx), the number
of
centers
of
given acid strength can
be
ob-
tained.
By means of the DTA technique Bremer and Steinberg
[29]
also
observed an inverse dependence of
the
amount of adsorbed base
(pyridine) on
the
pretreating temperature
of
the solid
(MgO
.
SiO,).
In fact, preheating
at

high temperature gave
a
pyridine desorption
peak
at
a
lower temperature, with progressively higher pyridine de-
sorption peak temperatures
as
the preheating temperature was
lowered.
employed, particularly with ammonia. The main advantage of
this
method
is
that the acidity measurements can be made in the real
working conditions of the catalyst. Another advantage, particularly
The gaseous bases adsorption-desorption method
has
been widely
Downloaded by [McGill University Library] at 07:35 15 December 2012
SURFACE ACIDITY OF SOLID CATALYSTS
73
FIG.
3.
Schematic
DTA
and
TGA
curves

[28].
when TGA and DTA techniques
are
employed,
is
the
possibility of
obtaining
a
measure
of
the gaseous
base
desorption activation energy,
which allows an absolute determination of
the
acid strength of
the
surface
sites
to be obtained. However, the results often cannot
be
related to
the
catalytic activity and, when ammonia
is
employed,
its
adsorption on
the

solid
is
so
strong
that
a
careful evaluation of the
acid strength distribution becomes very difficult or impossible. In
addition, the method cannot distinguish between physical and chemi-
cal adsorption of the
base.
Such
a
difficulty
has
been avoided by in-
troducing some standard conditions
(e.g.,
heating up
to
a
given tem-
perature and evacuating down to
a
given low pressure for
a
given
time) with the aim of establishing
a
given limiting point between the

two types of adsorption, but such
a
procedure
is
obviously empirical.
D. Calorimetric Methods
Another method for measuring the acid strength of
a
catalyst
sur-
face
is
based
on
the
determination
of
the
heat
of adsorption of basic
substances. Richardson and Benson
[30]
measured the heat of ad-
sorption of trimethylamine over cracking catalysts calorimetrically.
The values of
AH&,
obtained ranged from
-33
to
-38

kcal/mole.
Zettlemoyer and Chessick
[31]
determined the energy distribution
of the acid centers for kaolin and attapulgite catalysts by means of
a
relationship between the differential
heat
of adsorption and the
amount of adsorbed base (n-butylamine). A procedure suggested by
Harkins
[32]
consists of the rapid immersion of
the
solid in the
liquid base. In such
a
way,
after
suitable corrections, an integral
Downloaded by [McGill University Library] at 07:35 15 December 2012
74
L.
FORNI
heat
of immersion
is
determined. In order to obtain differential
heats, orie must preequilibrate
the

sample with various amounts of
base before
the
immersion. From the slopes of
the
curves obtained
by plotting
the
integral
heats
of immersion
so
determined vs
the
fraction
of
the surface precovered before the immersion, one can
obtain differential heats
as
a
function of coverage. By plotting such
differential heats vs coverage (see Fig.
4),
the distribution of acid
strength of the centers can
be
obtained. The behavior of the curve
for kaolin
has
been interpreted by assuming an interaction among

the adsorbed molecules of
base.
Such an exothermic interaction
gives
a
maximum
heat
evolution for coverages ranging from
0.1
to
0.6,
and the
heat
evolved
adds
to
that
evolved by
the
adsorption
re-
action.
0
v
1111
I1
IIJ
0
0.2
0.4

0.6
0.8
Surface Coverage
FIG.
4.
Net differential enthal y
of
adsorption at
25°C
for
n-butylamine on a kaolin
catalyst
(0)
and on attapulgite
(4
[
321.
Other examples of the application
of
calorimetric methods are
given by Stone and Rase
[33],
who worked with pyridine, and
by
Hsieh
"1,
Kevorkian
et
al.
[35],

and Clark
et
al.
[36,37]
who worked
with ammonia on silica-alumina. Other studies refer
to
the adsorp-
tion of
H
S
on
alumina
[38]
and of pyridine and benzene on silica-
alumina
8391,
but the acid strength was determined by means of
a
Downloaded by [McGill University Library] at 07:35 15 December 2012
SURFACE ACIDITY OF SOLID CATALYSTS
75
chemical
[38]
or
gas chromatographic analysis
[39]
of
the
gases

evolved by heating the sample after
the
adsorption. The work by
Hsieh
[34]
on silica-alumina
is
another example of interaction among
the molecules of adsorbed base (ammonia) with
a
consequent defor-
mation of
the
plot of differential heat of adsorption vs coverage (see
Fig.
5).
The interpretation by Hsieh of the behavior of such
a
curve
is
as follows:
For
low coverages
(8
<
0.1)
the
ammonia adsorbed
neutralizes
all

types of surface acid groups,
first
on stronger cen-
ters and then on weaker ones.
At
9
=
0.1
all
the acid centers have
been neutralized with
the
formation of either
NH,+
ions
or
highly
polarized
NH,
molecules. On further adsorption, ammonia molecules
interact with
NH,'
ions and polarized NH, molecules, to which they
are
bound by Coulombian forces;
the
new ammonia also may interact
by van der Waals forces with
the
catalyst surface. The Coulombic

interaction, due to its stronger force (inverse square law, with re-
spect to inverse sixth power law of van der Waals interaction), pro-
vides for
a
much higher adsorption
heat,
thus explaining the large
interaction
heats
shown for
8
>
0.1.
Hsieh
also discusses the effect
of acid strength of
the
centers on
the
catalytic activity of
the
solid.
7t
f
51
I
I
I
1
I

I
1
I
0
.l
.2
.3
.4
.5
.6
.7
.8
6
FIG.
5.
Differential heat of adsorption
for
NH,
on silica-alumina vs surface coverage
The methods
based
on the determination of the immersion heat of
t341.
the
solid have
the
advantage of allowing an absolute measure of
a
sufficiently accurate acid strength to
be

obtained. Their main dis-
advantage
is
connected to the interactions among the adsorbed mole-
Downloaded by [McGill University Library] at 07:35 15 December 2012
76
L.
FORNI
0.5
0.4
0,
\
U
W
E
0.3
\
-
\
\
\
-
\
8.
\
\%;
\O9,
'
%
%O

\
-
\
\
\
By means of the flash desorption method of preadsorbed bases in
an inert gas flow, plots of the type reported in
Fig.
8
can be obtained
[26].
The apparatus employed by Amenomiya
et
al.
[26]
is
shown in
Fig.
9.
Such
a
method has also been employed, with some modifica-
tions, to
the
study of hydrogenation of ethylene over alumina, with
cules of the base, particularly
at
high coverages, and to
the
difficulty

of calculating
the
heat corresponding to such interactions.
The method of desorption
of
a
preadsorbed base by progressive
heating of
the
solid in
a
closed system, followed for example by
Ballou
[24],
leads to plots of the type reported in Figs.
6
and
7.
From such graphs
it
is
possible to obtain both the acid strength
distribution
and
the density of the acid centers.
Downloaded by [McGill University Library] at 07:35 15 December 2012
SURFACE ACIDITY
OF
SOLID
CATALYSTS

77
175 2
55
335
41
5
495
"C
FIG.
7.
Ammonia adsorption capacity
from
thermal desorption experiments (see also
Fig.
6)
[
241.
the aim of determining
the
fraction of centers active
in
such
a
reac-
tion [40].
The activation energy
of
desorption reactions can
be
calculated

by performing some experiments
at
various heating rates
(p)
and
recording the temperatures
(TJ
corresponding to the maxima of
the peaks of Fig.
8
graphs. The formula employed
for
such calcula-
tions
is
where
Ed
is
the
desorption energy,
R
the gas constant,
V,
the
maxi-
mum volume
of
gas adsorbed on the solid, and
k,
a

constant, inde-
pendent on temperature. When
the
energies of the centers are hetero-
geneous,
the
value
of
T,
must be
for
samples which
have
been
pre-
treated
so
that
they have
the
same degree of surface coverage
at
the
start
of
flash
desorption.
A
similar method was followed by Kubokawa
[22],

who calculated
the
desorption activation energies from desorption rates instead of
doing experiments by
flash
desorption.
Downloaded by [McGill University Library] at 07:35 15 December 2012
78
L.
FORNI
1
f
I I
I
I
I
r
min
10
20
3b
40
50
FIG.
8.
Flash desorption chromatograms. Flashing rate 15.2"C/min. Evacuation
before flashing
for
2.5 hr at the following temperatures ("C): 24
(1);

75 (2); 150 (3); 175
(4); 325 (5); 400
(6);
500
(7)
[26].
E.
Other Methods
The acid strength of
a
solid may
be
determined on
the
basis
of
its
catalytic activity toward some suitably chosen reactions.
For
example, Pines and Haag
[41]
estimated
the
acid strength of some
alumina catalysts by measuring the
rates
of cyclohexane and
di-
methyl-1-butene isomerization and
1

-n-butyl alcohol dehydration.
Another method, followed by Aonuma
et
al.
[42],
starts with the de-
termination
of
the
equilibrium constant for
the
adsorption of am-
monia
on
the solid surface. Such
a
constant
is
obtained by
experi-
ments on the progressive and reversible poisoning of the catalyst
with the base in
the
cumene-cracking reaction. Chapman and
Hair
[43]
determined
the
acid strength from measures of the shifting of
the characteristic absorption bands

of
the base when
it
adsorbs
on
Downloaded by [McGill University Library] at 07:35 15 December 2012
SURFACE ACIDITY OF SOLID CATALYSTS
79
'm
0
/$
/
Downloaded by [McGill University Library] at 07:35 15 December 2012
80
L.
FORNI
the solid. The experiment was
first
performed with benzaldehyde
[43],
but
it
was observed
that
such
a
substance oxidizes too quickly,
so
hexachloroacetone
[44]

was used and the observations were made
following the shifting of the carbonyl group band during the desorp-
tion of the ketone
at
growing temperatures. Other experiments were
also made with various alcohols in CC1,
[45,46].
A
comparison was
made of frequency
shifts
of
the
OH group band
at
various concentra-
tions of
the
alcohol in
the
solvent, in the presence of
the
solid, with
the known acidity constants of
the
alcohols. From such
a
compari-
son
the

authors determined
the
following
pK,
values for
the
solids
examined: magnesia,
15.5;
boria,
8.8;
silica,
7.1;
silica-alumina,
7.1;
phosphorus,
-0.4.
111. DETERMINATION OF THE SURFACE DENSITY
OF ACID CENTERS
The number of acid centers present on
a
solid surface
is
usually
expressed
as
surface density, e.g., as the number of centers,
or
millimoles, per unit weight
or

unit surface
area.
A.
Direct Titration of Aqueous Suspensions
When an acid solid
is
suspended in water,
it
often lowers the pH
of
the
aqueous phase.
A
direct titration of the aqueous suspension
with
a
standard base,
either
in
the
presence of an indicator
or
poten-
tiometrically, can then give a measure of
the
surface acidity
[23,47,
481.
Very often, however,
the

acidity measured by such
a
titration
does not measure
the
acidity of
the
solid.
A
typical case
is
reported
by Oblad et al.
[49]:
By titrating
an
aqueous suspension of silica-
alumina catalysts with NaOH, they observed that
the
first
end point
is
reached very rapidly. But, after some hours,
the
pH
of the sus-
pension decreased, and
a
quantity of titre
had

to be added to reach
the
new end point. They interpreted such a phenomenon by postulat-
ing the formation of the following equilibrium:
2HA10,
.
xSi0, H,O
+
A1,0,
+
ZxSiO,
(4
)
By neutralizing
the
acid alumino-silicate with the titrating
base,
the
equilibrium slowly
shifts
to the left and
a
new quantity of acidity
is
formed. Another typical case
is
given by the water itself, which re-
acts with
the
solid, e.g., by transforming

the
Lewis centers into
Brbnsted centers. Then titration of the catalyst in
the
aqueous phase
gives a quantity of Brdnsted centers greater than
that
obtained in an-
Downloaded by [McGill University Library] at 07:35 15 December 2012
SURFACE ACIDITY
OF
SOLID CATALYSTS
81
hydrous solvents. The error may be very large,
as
reported in Fig.
10,
in
the
case of
two
different types of silica-alumina
[50].
From the experimental point of view, titration methods in aqueous
solution can be considered
as
the simplest ones for the determination
of surface acidity. On
the
other hand,

the
presence of water, which
cannot
be
considered an inert medium, introduces such heavy limita-
tions
that
titration methods may
be
employed only in particular cases.
1.2
E
0.8
a
0
I
1 1
1
01234567
>
I-
m
>
1.2
-
01234567
%Alumina
A
B
0

2 4
6
8
101214
%Alumina
FIG.
10.
Brdnsted acidity
of
synthetic
(A)
and commercial (B) silica-alumina
catalysts;
(X)
moist;
(0)
dry
[50
1.
B. Titration
after
Ionic Exchange
The previously described direct titration method
is,
in fact, based
on
an
ionic exchange between the acid solid and the aqueous phase.
But in
the

case
of
cracking catalysts based on molecular sieves, some
Downloaded by [McGill University Library] at 07:35 15 December 2012
82
L.
FORNI
titration methods involving a direct neutralization of
Ht
cations have
been developed. In fact, in such catalysts the active
sites
are
local-
ized in some regions characterized by
a
strong local electrostatic
field
[51],
generated by the simultaneous presence of basic and acid
centers. The cationic exchange on molecular sieves takes place
very quickly
at
room temperature. The pH of exchanging solutions
must not be too low to avoid damage to the crystal structure
of
the
sieves. Usually
the
pH must not

be
lower than
4,
although in some
cases
it
is
possible to operate down to
pH
=
2.
According to Gren-
hall
[52], the
Ht
ion
of
the
solid can
be
exchanged with
Nat
by means
of
a
5%
NaCl aqueous solution. The exchanged solutions obtained
at
various exchanging times are titrated and the results extrapolated
to t

=
0
in order to eliminate the influence of the reaction between
the solid and water, according to equilibrium reaction
(4).
Plank
[53]
performed
the
ion exchange with a
0.1-8
ammonium acetate solution.
The ion exchange technique, followed by titration, was also employed
by Mahl
[54],
Trambouze et
al.
[55],
Holm et al. [56], and Danforth
[57].
Holm et al. in particular performed
an
accurate study which
demonstrated
the
independence of surface acidity on catalyst particle
size. They also outlined
the
influence of sample quantity on
the

re-
sults of acidity determination. Their results clearly indicate
that
the reaction reaches an equilibrium in which the specific quantity
of acid transferred from
the
solid surface to the solution increases
with
an
increase in
the
amount of employed sample. Since below
0.1
g
of sample
the
increase in acidity became negligible, the authors per-
formed
all
their determinations on samples weighting less than
0.1
g.
They were also
able
to determine the relative acid strength
of
such
centers from the change in the degree of exchange with the quantity
of sample.
Titrimetric methods following ion exchange are interesting in

that they do not need direct contact between the titrating
base
and
the
solid, but they do not avoid the presence of water. This
is
an
important limitation because,
as
with the direct titration method,
all
the centers whose acid strength
is
lower than that of water itself
cannot
be
titrated.
C. Titration with Bases
in
Nonaqueous Solvents
This method, originally introduced by Tamele [58] and Benesi
110,591
and subsequently modified, can be briefly described
as
fol-
lows:
A
small quantity of predehydrated solid
is
covered with an

inert, anhydrous solvent in which
a
predetermined amount of
a
base
is
dissolved. After equilibration an indicator, which adsorbs on the
solid surface,
is
added and assumes the color of
its
acidic
or
basic
form. The quantity
of
base needed to impart to the solid the color
Downloaded by [McGill University Library] at 07:35 15 December 2012
SURFACE ACIDITY OF SOLID CATALYSTS
83
0
4
H
of
the
basic form of the indicator represents
the
measure of the
quantity
of

acid centers present on the catalyst surface. Obviously
only the centers whose strength
is
higher than the
pK, of
the
indi-
cator are titrated. However, by employing indicators of various pK,
it
is
possible to titrate the centers
of
various acid strengths. Ex-
perimental
details
can be found in an example reported by Tanabe
and Katayama
[60].
Some subsequent modifications of the method
concerned only the employment
of
different solvents and/or
bases.
For example, Johnson
[61]
employed CC1, and isooctane
as
solvents
and observed
that,

for some solids such
as
silica-alumina, the solu-
tion equilibration times for drop by drop titrations may
be
very long
(2-3
days). Such
a
difficulty was overcome in
the
Benesi method
[59]
by adding
the
indicator to
the
suspension
aftek
equilibrium had been
reached, and the end point was attained by successive approxima-
tions. In addition, the Benesi method strongly reduces
the
danger of
moisture contamination because the amine
is
added
all
at
once.

Matsuzaki et
al.
[62]
studied the effects of some parameters on
the results
of
titration. One can observe (see Fig,
11A)
that
the
amount of indicator added affects the results only
if
it
is
lower than
a
given minimum (e.g., for dimethyl yellow
0.2-0.3
ml
of
1%
ben-
zene solution), while
the
titration time
has
an influence only
if
it
is

A
0
0.8
a
-
0
0.2
-
I I I
I
I
0
L
1
I
I
FIG.
11.
(A)
Effect
of
added indicator volume
on
measured acid amount. Sample:
0.5
g
of
>lo0
mesh silica-alumina in
10

ml benzene. Indicator:
1%
benzene solution of
dimethyl yellow.
(B)
Effect
of
titration time on measured acid amount. Sample as in
A.
Indicator:
0.3
ml
of
1%
benzene solution of benzeneazodiphenylamine
[
621.
Downloaded by [McGill University Library] at 07:35 15 December 2012
84
L.
FORNI
shorter than
50
hr
(Fig.
11B).
The latter difficulty (too long titration
times) can
be
overcome

if
the solid
is
ground down to fine powder
(2
100
mesh),
as
may be seen
in
Fig.
12A.
The harmful effects
of
exposing the solid to
a
moist atmosphere are shown in Fig.
12B.
The
stronger the acidity of the centers
(H,
+1.5),
the stronger
is
the
latter effect, This
is
probably due to the transformation of
a
part

of
the
stronger centers into weaker ones. In fact, the pK, of water
is
-1.7,
so
that
its
adsorption on the solid surface would poison
only
the
H,
-1.7
centers.
mesh
,Ka
of
indicator
FIG.
12.
(A)
Effect
of
powder size on acid amount. Sample:
0.5
g
silica-alumina in
10
ml benzene. Indicator:
0.3

ml
of
1%
benzene solution
of
dimethyl yellow. Titration
time 2 hr.
(B)
Effect of moisture on acid amount. Sample:
0.5 g
of
>lo0
mesh silica-
alumina in
10
ml
benzene. Indicators:
0.3
ml each of several pKa values
(1%
benzene
solution). Titration time; 2 hr. Predried catalyst left in
90%
humidity at
20°C
for
(a)
0
min,(b)5 min,(c) 10min[62].
Perhaps

the
best
technique was developed by Bertolacini
[63].
The main advantage of such
a
titration method consists of an ultra-
sonic generated stirring by which the equilibrium conditions can be
reached in a very short time (some tens
of
minutes, instead
of
days)
(see Table
4).
As
repeatedly mentioned,
the
most common source of error in
such determinations
is
contamination with water. For example,
on
silica-alumina, even
after
dehydration in
air
at
600°C,
at

least
0.5
wt% of water
still
remains on the solid.
If
all that water was asso-
ciated
with
surface centers, and
if
the
surface area of the solid was
Downloaded by [McGill University Library] at 07:35 15 December 2012
SURFACE ACIDITY OF SOLID CATALYSTS
85
TABLE
4
Titration Time Dependence
of
Titrations by
Bertolacini Method
[63]
Titration time,
Catalyst hr
Low
alumina
(13%)
Semisynthetic
Silica-magnesia

0.5
2
4
8
24
0.5
2
4
8
24
0.5
2
4
8
24
Acid titre,
meqk
0.34
0.35
0.33
0.35
0.33
Av
0.34
0.20
0.21
0.20
0.21
0.21
Av

0.21
0.70
0.68
0.68
0.69
0.70
Av
0.69
about
300
m2/g, this could account for 5
X
1013
sites/cm2,
i.e.,
a
num-
ber
very close to the total number of sites present on the solid sur-
face.
If the solid surface
is
colored, the method can
still
be employed
by adding to the mixture
a
given quantity of a white acid solid. The
end point of titration
is

determined by observing the change in color
of the white solid surface.
A
correction factor must obviously be
introduced for the quantity of base consumed in the neutralization of
such
a
white solid [61].
A
practical example of such
a
procedure
is
given by Tanabe and Watanabe [64] in which the surface acidity of
titanium trichloride in the presence of
a
small amount of silica-
alumina
is
determined. In this case the sharpest change in,color
was observed for mixtures of about 0.02-0.05 g
of
TiCl, and 0.2 g of
Si0,-Al,03. Another example
is
given by Voltz et al. [65] in which
the surface acidity of
a
dark green sample
of

Crz03, previcusly dried
for
4
hr
at
500°C, was determined in the presence of a given amount
of alumina.
strength and the number of surface
sites.
The results
are
reported
in Fig. 13. The three substances on which he worked were Si0,-MgO,
SiO,-Al,O,, and Filtrol. Another example of correlations
of
this type,
Benesi [59] also tried to find
a
correlation between the acid
Downloaded by [McGill University Library] at 07:35 15 December 2012
86
L.
FORNI
0.0
o*2*
+4 +2
0
-2
-4
-6

-8
He
FIG.
13.
Butylamine titres
vs
acid strength
for
catalysts calcined at
500°C.
(1)
Silica-
magnesia,
(2)
silica-alumina (MS-A-l),
(3)
Filtrol
SR
[
591.
reported by Goldstein
[l],
refers to
a
sample of silica-alumina
(Fig.
14)
either freshly prepared
or
steamed at

715°C.
It may be
seen that, in the steamed sample, the sites of strongest acidity still
remain poisoned
by
water.
A
7
A
-I
FRESH
Y
4.0
-
3.0
H
0
-2
-4
-6
-8
0.01
I
I
'
I
'
I
I
I

H*
+4 +2
FIG.
14.
Acid strength distribution
for
silica-alumina on unit area basis
[
11.
Downloaded by [McGill University Library] at 07:35 15 December 2012
SURFACE ACIDITY OF
SOLID
CATALYSTS
87
Another possible source of error in such determinations
is
re-
lated to
the
indicator employed.
A
spectrophotometric study per-
formed by Drushel and Sommers
[66]
showed that some indicators
are not able to measure protonic-type acidity, some show two end
points and some others are improper for the usually assigned pK,
value. Some aromatic alcohols have also been employed
as
indica-

tors.
It
has been reported that such aromatic alcohols, called HR
indicators,
are
specifically employable for protonic-type acidity.
In Table
5
a short
list
of such substances, reported by Hirschler
[67],
is
given. They dissociate according to
ROH
+
H+
R+
+
H,O
so
that
the
definition
of
H,
is
HR
=
PK,

+
log ([ROHI/ER+I)
In Fig.
15
the plot of
H,
and
HR
vs
the
H,SO,
wt% in aqueous solution
is
reported. The double end point observed with some indicators
could then be due to such a different behavior
of
these substances
with respect to the type of acidity present on the solid surface.
TABLE
5
HR
Indicators
[67]
Indicator
4,4’,4”-Trimethoxytriphenylmethanol
4,4’,4“-Trimethyltriphenylmethanol
Triphenylmethanol
3,3‘,3”-Trichlorotriphenylmethanol
Diphenylmethanol
4,4’,4“-Trinitrophenylmethanol

2,4,6-Trimethylbenzy1 alcohol
PKR
+0.82
-4.02
-6.63
-11.03
-13.3
-16.27
-17.38
D, Calorimetric Titration
The calorimetric titration method was
first
introduced by Tram-
bouze and co-workers
[68-701
and subsequently developed by
Top-
chieva et
al.
“711,
and Tanabe and Yamaguchi
“721.
The bases em-
ployed can be n-butylamine, ethyl acetate,
or
dioxane.
As
previously
reported, the method is also useful for the determination
of

acid strength.
Experimental details
are
given by Tanabe and Yamaguchi
[72].
The
Downloaded by [McGill University Library] at 07:35 15 December 2012