.5 . .... .5 - s
.. $5. . . .. . 5....
.. .5 , . 5%.... . .._.......
.v. to JV. . .- . 3H..E5.PQ% $- . -

. . '4 VJ!”
. . . . ‘3.“ l.
v 1- . . . . . .
. . .

I.
n

. 5 . .
$445.2...
4 r
. _ -

..
. . I .14 4\;
. 33.9... . Eﬁ...x.
. . . . . . . .53.... L.
. . . . .r._:....~‘.M.. .
4...... . . .. . 9.3% y . .
5.... . . . .
v -

I _
.. 3:0 .5 x3?

3 .

. a

o d
It. ru
-
5-. m.
. . 915...... ...-. - . .
5 . .
a . . .. . T.
5%”...
$11.1..- .
.
we. a:

232.“ ........ Fur-Ln.
_ zuﬁuﬁﬁmyJWH-L Ava-n
. v. 5 . .
. . . 5.. 53%...- ...u-. .1 $5
. a... 6.132.. .. . . . . - . . . . #1.» ......: n .. . .
. 2.33M“... . .x . . . . . .mmwtwmm. mu dﬁréwmwﬁm.
VS. . . . . . . . . .2 1.. a. L 55 .. . u
a. . . . . . _ a... Em...- . . . . 2...m.m.~1§.m.£wx $3WWHMM7
.3 any”. . . - . . 3%” . . . . . . . . . y . A.) t .
. 1&3. . . . . . .. . . . . . - .....n...w-m...uw.qwmw
.. . Wan-v3. .
$12.3; - . - . :Lvn .
5r... . . .. . . . 5%...
t- Why.- ‘
.._. ﬂaw,

 

5 . a
.Iﬁﬁhm}...
try-i... NM“...

2).“)wwvil.

 

 

. t
.A\._.-

. 7 v .-

«I.»

 

M!

II... by? xéihlﬁvzmg.

 

EWIL:

.FEELr-h

THESIS

Zau

This is to certify that the

dissertation entitled

LAYERED DOUBLE HYDROXIDES:
I. SYNTHESIS BY DESIGN

II. CATALYSIS BY INTERCALATED
POLYOXOMEtféfTES

presente

Elizabeth Ann Gardner

has been accepted towards fulﬁllment
of the requirements for

Ph.D. degree in Chemistry

%///777M ’ ﬁzz/ow“

homas Jv innavaia

Major professor

 

 

Date December 18, 1998

 

MSU is an Afﬁrmative Action/Equal Opportunity lnsrirution 0- 12771

 

LIBRARY
Michigan State
Unluersity

 

 

 

PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 CZICIRCDateDUQpGS-p. 15

LAYERED DOUBLE HYDROXIDES:
I. SYNTHESIS BY DESIGN
II. CATALYSIS BY INTERCALATED
POLYOXOMETALATES
by

Elizabeth Ann Gardner

A DISSERTATION

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Chemistry

1998

ABSTRACT

LAYERED DOUBLE HYDROXIDES:
I. SYNTHESIS BY DESIGN
II. CATALYSIS BY INTERCALATED
POLYOXOMETALATES

by
Elizabeth Ann Gardner

Layered double hydroxides (LDHs) have two predominant properties
which can be exploited for materials applications. Firsr, the hydroxylated
lamellar framework can possess the desired characteristics, as in their use as
antacids. Second, the exchangeable intergallery anions can be chosen for a
speciﬁc function, with the framework serving merely as an inert support. The
oxidation of alkenes by tungState immobilized on an LDH intercalated with
hydrophobic anions is an example which uses bOth the basicity of the LDH to
enhance the seleCtivity for the epoxide over the diol product and the
hydrophobic gallery anions to create a favorable environment for the organic
substrate.I

Taking advantage of the hydroxylated lamellar structure of the LDH
intercalate, a method for making thin LDH ﬁlms for application as sensors,
devices, and ﬂame retardant materials was developed. LDHs were synthesized
in organic solvents, resulting in colloidal sized LDH particles which formed well
oriented ﬁlms less than a micron thick, as judged by powder XRD and SEM
studies. With this method, ﬁlms with varying layer charge, intergallery anions,
and framework cations are easily synthesized.

Catalysis by intercalated polyoxometalates (POMS) is an example of the

second approach. Polyoxometalates are a promising class of anionic catalysts for

intercalation into LDHs and much research has been inveSted in introducing
gallery functionality by intercalation of the POM into LDH layers.

However, there are a few problems which complicate the LDH-POM
chemistry. First, the synthesis of an LDH-POM is always accompanied by an
impurity phase which results from the reaction of the basic LDHs with acidic
polyoxometalates. This problem was to be addressed by using a more acidic
anAl-LDH, which can be synthesized at a pH of 6.2. The second problem
arose from discrepancies in the reported d spacing for otherwise identical
polyoxometalate intercalates, namely 9.9 and 12.2 A in the case of LDH-
M07024“, and 12.2 and 14.6 A for LDH-HZWUOw" intercalates. In this case,
Raman spectroscopy was used to unambiguously identify the molybdenum
polyoxometalate intercalate. Finally, size selective catalysis implies pillaring of
the gallery anions. Nitrogen adsorption was used to measure the pore volume of
the LDH-POM sysrems. The catalytic activity of four syStems; 1\/Ig3Al-1V'IO7O2 4‘”,
Mngl-HZWDOwG', Zny’d-Mo7Ozf‘, and anAl-HZWIZOwé’ was tested for the
peroxide oxygenation of cyclohexene to the corresponding epoxide and diol.
However, our results indicate that only the surface bound anions are catalytically
active and that the selectivity is a result of the basic nature of the LDH
support?"3 In addition, the impurity salt may contribute to the catalytic activity.
The acidity of the LDH support is not the determining factor in the formation

of the impurity salt, but rather the afﬁnity of the POM for the layer cations.

This Dissertation is dedicated to my son, Mathew Miller

iv

TABLE OF CONTENTS

Chapter 1. Thesis ObieCtives: An Overview ................................................... 1
Chapter 2. Layered Double Hydroxides: A Brief Review ............................ 8
History ............................................................................................................. 8
Applications .................................................................................................... 9
Structure ......................................................................................................... 13
Composition ................................................................................................. 15.
The Nature of M” ................................................................................. 15
The Nature of A” ................................................................................... 18
The interlayer water 23
Synthesis ........................................................................................................ 24
Precipitation ............................................................................................ 25
Salt-Oxide method ................................................................................. 28
Thermal Treatments .............................................................................. 31
Physicochemical CharaCterization ............................................................. 31
Chapter 3. Synthesis of ngAl-Layered Double Hydroxides at Acidic
Introduction .................................................................................................. 37
Results and Discussion ................................................................................ 46
Conclusion ..................................................................................................... 52
Experimental Methods ................................................................................ 52
Chapter 4. The Synthesis of Layered Double Hydroxides with
Colloidal-Size Particles ...................................................................................... 54
Introduction .................................................................................................. 54
Results and Discussion ................................................................................ 56

M-LDHs .................................................................................................. 60

The Precipitate from Methanol ........................................................... 75
Conclusions ................................................................................................... 76
Experimental Methods ................................................................................ 78

Chapter 5. Layered Double Hydroxides Pillared with Polyoxometalate
Catalysts ............................................................................................................... 81
Polyoxometalates .......................................................................................... 83

Applications ............................................................................................ 84

Catalysis by polyoxometalates .............................................................. 84

History ..................................................................................................... 88

Composition ........................................................................................... 89

Structure ................................................................................................... 90
Polyoxometalate Intercalated Layered Double Hydroxides ................. 97

Layered Double Hydroxides Intercalated with

Isopolyoxometalates .............................................................................. 99

LDHs Intercalated with Heteropolyoxometalates .......................... 101

Applications .......................................................................................... 106

Chapter 6. Polyoxometalates and Their Intercalation into Layered

Double Hydroxides .......................................................................................... 107
Introduction ................................................................................................ 107
Results and Discussion .............................................................................. 109

CataIySt Synthesis and Characterization ........................................... 109

Catalytic Properties MgrAl— and anAl-POMs ............................... 122

Catalytic Oxidation of Cyclohexene by anAl-PJP ........................ 126

anAl-NaPSWMO, m” .......................................................................... 127

Catalytic Oxidation by anAl-NaPSWwOnON' ................................. 128
Conclusions ................................................................................................. 1 33

Experimental Methods ........................................................................... 138
Refernces ....................................................................................................... 139

LIST OF TABLES

Number Page
Table 2-1. Different Metals Used in the Synthesis of Layered Double
Hydroxides .......................................................................................................... 16

Table 2-2 Inorganic Anions That Have Been Intercalated Into Layered
Double Hydroxides ............................................................................................ 20

Table 2-3. Organic Anions That Have Been Intercalated Into Layered

Double Hydroxidesflfl
Table 2-4. Change in d spacing with Increasing Humidity .................................... 24
Table 2-5. Identifying \X’awnumbers for Common Gallery Anions .................... 34

Table 3-1. Three Merhods Used to Synthesize MgAl Layered Double
Hydroxides .......................................................................................................... 40

Table 3-2. Indices (/Jé/ ) for LDHs Synthesized by the Methods of Miyata
and Reichle .......................................................................................................... 44

Table 5-1. Applications for Polyoxometalates, Excluding Catalysis and

Medicine .............................................................................................................. 85

Table 5-2. Summary of Polyoxometalate Intercalated Layered Double

Hydroxides ........................................................................................................ 1 02

Table 5-3. Oxidation Reactions with Intercalated Polyoxometalate

CatalySts ............................................................................................................. 105
Table 6-1. Raman Bands (cm“) of Isopolymolybdates ......................................... 117

Table 6-2. Peroxide Oxidation of Cyclohexene Over Polyoxometalate

Catalysts ............................................................................................................. 1 23

Table 6—3. Peroxide Oxidation of Cyclohexene in the Presence of

Different Forms of NaP5\\"3OOm,”‘ .............................................................. 1

LIST OF FIGURES

N'umber

Figure 2-1. A schematic of a 3:1 My/M3+ Layered Double Hydroxide .........

Figure 2-2. X-ray diffraction pattern of a Mg3Al-N'OS' double hydroxide

showing Miller indices. .................................................................................

Figure 3-1. Schematic of pH where M(OH)n and Layered Double

Hydroxides form ...........................................................................................

Figure 3-2. X-ray diffraction patterns of MgAl-COf' LDHs made by
Miyata’s coprecipitation at constant pH. The pH is maintained at

10.0 i0.2 and the reaction is digested for 18 hours at room

temperature. a. Mg/Al = 4; b. Mg/Al = 3; c. MgAl/ = 2;. The

peak assignments are given in Table 3-2. ..................................................

Figure 3-3. MgAl-COf‘ LDHs made by Reichle’s coprecipitation at
variable pH. The ﬁnal pH is in the range of 8-11 and the reaction
is digested for 18 hours at room temperature. a. Mg/Al = 4; b.

Mg/Al 1' 3; c. MgAl/ = 2; The Miller indices are given in Table

3-2. ...................................................................................................................

Figure 3-4. a. Transmission Electron Micrograph of Mg3Al-CO32' made
by Reichle’s coprecipitation method. b. TEM of anAl-NO3‘
made by Taylor’s induced hydrolysis method. The particle size is

considerably larger. c. TEM of a single particle of anAl-NOS’ at a

higher magniﬁcation. ....................................................................................

Figure 3-5. The diffraction patterns of products obtained by the reaction
of Mg(NO3)2 (0.03 mol) and Al(N03)3 (0.01 mol) in 200 ml HBO)

Page

..... 12

...... 3 ._

..... 38

..... 42

..... 43

..... 47

in the presence of various amounts of 2.0 M Na(OH). The
volumes of NaOH solution added to the reaction mixrure are
indicated along with the ﬁnal pH value of the reaCtion mixrure a

reaction time of 3 days at 90-95 °C. ................................................................ 49

Figure 3-6. FI‘IR 0?ngqu- synthesized at pH = 6.9. The resonance
at 1633 cm'1 is assigned to coordinated water, 1385 cm‘l to nitrate,

and 447 cm’1 is Al-OH ...................................................................................... 30

Figure 4-1. X ray diffraCtion pattern of products prepared in the reaCtion
of a. Mg(NO3)3, Al(N03)3, and NaOH in acetonitrile and b.
acetone. Synthesis of LDHs in acetone and acetonitrile by
standard coprecipitation methods is unsatisfactory due to the

insolubility of the Mg and Al salts in these solvents ..................................... 57

Figure 4-2. Powder X-ray diffracrion patterns for LDHs and LDH
precursor products precipitated from various organic solvents and
water. The LDHS synthesized in propanol and butanol were
washed with methanol to remove the remaining NaNO3 salt

byproduct ............................................................................................................. 58

Figure 4-3. Infrared Specrra of A'Ig‘Al-NOg, synthesized in ethanol,
methanol, and water. From top to bottom: A Mg3Al-NO3'
synthesized in ethanol; the MgAl-NO3} Mg3Al-NO3’, and MggAl-
NO; product from precipitation in methanol; the MggAl-NOg
precipitation prodUCt after suspension in water and an LDH

formed by traditional coprecipitation. ............................................................ 59

Figure 4-4. The X-ray diffraCtion pattern of an oriented thin ﬁlms of
Mg3Al-NO3‘ a. product prepared by coprecipitation of an in

methanol and subsequent hydrolysis in water, 0.030 mole

Mg(NO3)3 and 0.010 mol Al(NO3)3 in 100 ml methanol are dripped
into a solution of 0.09 mol NaOH in 100 ml methanol and Stirred
for 3 days at 55 °C. b.. LDH product prepared by coprecipitation
in water at 55 °C under conditions analogous to the synthesis in
methanol. The orientation of the LDH layers in pattern a. is
evident from the complete lack of in-plane reflections for the M-
LDH . c. A less oriented powder pattern of the M-LDH product.
The presence of the in-plane reﬂections verify that the M-LDH is

an authentic layered double hydroxide. .......................................................... 61

Figure 4-5 Scanning electron micrographs of; 6. A 20 mm self supporting

Mg3Al-NO3' M-LDH ﬁlm. b. A 0.3 mm MggAl-Cl' ﬁlm supported

on glass ................................................................................................................. 63

Figure 4:6. Scanning electron micrographs of c. The surface of the

Mg3Al-NO3' M-LDH ﬁlm. d. The surface of a 30 mm ﬁlm of the

q

Mg3Al-NO3‘ control synthesized in water. ..................................................... 6/

Figure 4-7. a. X ray diffracrion pattern of an M-LDH calcined at 500 °C.

b. the same material after being reconstituted in water to reform the

LDH. .................................................................................................................... 69

Figure 4-8. Thermogravimetric analysis of ngAl-NO; M-LDH a. M-

LDH with a 2:1 Mg2Al ratio. b. M-LDH with a 3:1 Mngl ratio. c.
Mg3AJ-NO3' made by coprecipitation in water. Cl. M-LDH with 4:1
Mngl ratio ........................................................................................................... 70

Figure 4-9. X-ray diffraction patterns of M-LDHs with theoretical 4:1

(4.0:1.0), 3:1 (3.21.0), and 2:1 (10:10) MgAl ratios. The actual

ratios are in parentheses. ................................................................................... 71

Figure 4—10. X ray diffraction patterns for a. M-LDH where the gallery
anion of intereSt was intercalated via an exchange reaction. b. M-
LDH where Co3+ cation of interest is present in a fractional

subsdtution .......................................................................................................... 73
Figure 4~11. XRD of M-LDH with varying layer cations and anions. ................. 74

Figure 4-12. FTIR of a. Mg3Al-NOS' dried under ambient conditions. b.
Mg3Al-NO3' dried under vacuum at 150 °C. ................................................. 77

Figure 5-1. Schematic of the Keggin.”7 The OL-XMqu Keggin ion is
shown with common derivatives. The left side shows the three
isomers formed by the rotation of 1,2, or 4 M3 triads. The
lacunary ions on the right are formed by the dissociation of neutral
M03 octahedra. (HAVuOJ'). The formula for the Keggin ion and
its derivatives can be written as @09an03,“ where each Keggin
ion (excepting HAYDOJ‘) has an internal XO4 tetrahedra and each
addenda atom has one terminal oxygen. The Keggin anions are

considered to be very weak bases with great softness .................................. 91

Figure 5-2. Examples of Type 2 Structures where each addenda atom has

a

two terminal oxygens.” .................................................................................... 93

Figure 5-3. Schematic of some representative macropolyanions. a. A
Dawson ion made from two MgOM Keggin lacunary ions. b. A
Dawson TSMP complex sandwiching four planar Co ocrahedra. c.
A PJP anion, NaPSW3(,Om,”. The Na cation is displaced from the
center of the disk shaped anion. ...................................................................... 95

Figure 6-1. X-ray diffraction patterns of the mixed phase product formed
by in situ hydrolysis of NaZW/O4 in the presence of a MgZAl-

terephthalate: a. thin ﬁlm sample dried under ambient conditions;
the sharp reflections are (001) lines of
[l\'1g12A16(OI-I)36](Mo7024).with a basal spacing of 12.2 A, whereas
the broad lines are due to a salt-like impurity phase with a spacing
of 10.2 A. b. powder sample, dried under ambient conditions,
showinga marked reduction in the intensity of the LDH phase; c.
powder sample dried overnight at 120 °C showing the loss of the
Mngl-Mo70346‘ phase; cl. thin ﬁlm of sample after use as a catalyst

in the oxidation of cyclohexene. .................................................................... 110

Figure 6-2. Raman spectra of (NH 4)6Mo7O2 4 and the Mngl-Mo703f‘
product formed by in situ hydrolysis of M004} and ion exchange

with a Mngl-terephthalate LDH. ................................................................. 1 12

Figure 6-3 X-ray diffraction patterns of the mixed phase product formed
by in situ hydrolysis of Na3\\"O4 in the presence of Mngl-
terephthalate. Parts a-d are analogous to those given in Figure 6-1

for the corresponding molybdate sysrem. .................................................... 114

Figure 6-4. X-ray diffracuon patterns of [anzAl6(OH)36](MO7034)mHzO
prepared by ion exchange of the corresponding LDH-nitrate with
M07024“; a. thin film sample dried under ambient conditions; b.
powder sample dried under ambient conditions; c. powder sample
dried overnight at 120 "C showing the loss of the LDH phase; d.
thin ﬁlm sample after use as a catalyst in the oxidation of
cyclohexene. The broad weak reflection at 10.8 A in spectrum (b)

is due to a salt-like impurity phase ................................................................. 116

Figure 6-5 Raman spectra of anAl-Mo7024‘” after heating to various

temperatures: a. the air sample, showing bands analogous to those

xiv

in ammonium heptamolybdate; b. After drying at 105 °C, showing

the splitting of the terminal Mo=O band at 950 cm-1; c. After

heating to 220 °C, the terminal Mo=O band at 936 cm-1 and the
Mo-O-Mo polymeric band at 220 cm-1 decreased in intensity

similar to the band for Mo-O-Al in alumina supported molybdate

d. The sample dried at 220 °C and then slurried in distilled water,
showing the further shift in the terminal Mo-O band to 922 cm-1 ......... 119

Figure 6-6 X-ray diffraction patterns of
[ZnuAl6(OI-il)36](I-I3W1ZO4O)'mI-IZO prepared by ion exchange of the
corresponding LDH-nitrate with HZWIZOw": a. thin ﬁlm sample
dried under ambient conditions; b. powder sample dried under
ambient conditions; c. powder sample dried overnight at 120 °C;
d. thin ﬁlm sample after use as a cataIySt in the oxidation of
cyclohexene. The broad line at 11.1 A is due to a salt-like POM
impurity phase ................................................................................................... 121

Figure 6-7 a. XRD pattern of a physical mixture of Zn-HZWDO406' salt
and Al-H3\X’1._,O 406‘ salts b. XRD pattern of a physical mixture of
Zn-Mo7OZ4‘" salt and Al-Mo703f" salts. ....................................................... 125

Figure 6-8. a. Powder XRD pattern of an Aluminum-NaPSWwOno” salt
analogous to the impurity phase in the synthesis of anAl-PJP.
The d spacing of the ﬁrst peak is 15.5 A. b. XRD of a thin ﬁlm
sample of anAl-PJP The d spacing of the LDH is 21.6 A and the
broad reﬂection due to the impurity phase is 15.1 A. c. XRD of a
thin ﬁlm sample of anAJ-PJP with a slightly greater volume of
impurity phase. The d spacing of the LDH is 22.0 A and that of
the impurity phase is 15.1 A. d. XRD of a thin ﬁlm sample of

anAl-PJP with a the greateSt volume of impurity phase. The d
spacing of the LDH is 21.7 A and 14.8 A for the impurity phase. .......... 129

Figure 6-9. a. MAS-NMR of (NH 4)1 4NaP5W3OOm” with a single peak at -
13.9 ppm. The two small peaks are side bands. b. anAl-
NaPSWwOnOH‘, the MAS-NMR of the intercalated anion is the
same as a., verifying that only PJP anion is present. c. MAS-NMR
of the supported catalyst after use in the oxidation of cyclohexene.

The peak at -2.89 ppm is (BuO)3PO and the small peaks are side
bands. ................................................................................................................. 130

Chapter 1. Thesis Objectives: An Overview

My graduate work actually began with the study of oxidative catalysis by
polyoxometalates (POMS) intercalated into layered double hydroxides (LDHS).
While several of my predecessors in the Pinnavaia group had worked on the
synthesis of various LDH-POM systems, I was to be the ﬁrst Student to
concentrate on their catalytic applications

Polyoxometalates were ﬁrst intercalated into LDHs in 1984.4 Since then,
there have been over 45 articles published on the intercalation of POMS into
LDHS and catalysis by the LDH-POM systems. The driving force behind all
this effort has been the catalytic potential of the LDH supported POMS.
Polyoxometalates are the most widely applied class of homogeneous catalysts in
industry today. They serve as both hydration5 and dehydration catalysts, as
polymerization catalysts,6 as co-catalysts with palladium or rhodium, and most
importantly for this work, as oxidation catalysts. While many of these POMS
serve as homogeneous catalysts, supporting the ‘POM in the layers of an LDH
adds the potential value inherent to a heterogeneous catalyst. The LDH-POM
can be easily ﬁltered from the reaction mixture and reused, making the catalyst
less detrimental to the environmental and adding economic value for POMS as '
industrial catalysts. When the POMS are constrained between the layers of an
LDH, selectivity can be added to the reaction scheme. The selectivity can be

conformational, where the substrate can only approach the catalyst in one

orientation, or by size, where access to the interlayer region is restricted by the
dimensions of the catalyst.

But in many ways, POM-LDHS have not lived up to their potential.
There is an impurity byproduct in all POM-LDH intercalation reactions which is
the product of a hydrolysis reaction between the acidic POM and basic LDHS.
And of all the studies pertaining to catalysis by LDH-POMS, only Tatsumi et al.7
have shown a direct correlation between the LDH-POM catalysts and substrate
selectivity. Small molecules had greater access to the catalyst when Mngl-LDH
was intercalated with the catalyst HzWuOu‘O' (d spacing = 12.2 A) than when
M070246 was the intercalated catalyst (d spacing = 9.9 A). The oxidation of
cyclohexene (4.7 A by 6.2 A) was suppressed to a greater extent than that of the
2-hexene (3.9 A by 4.7 A) in the Mngl-MmOzf” relative to the Mngl-
HZWnOwé‘ system. My project was to take the preliminary results of Tatsumi,
add the expertise of our lab in LDHs and the intercalation of POMS, and see
how far I could take the chemistry. Speciﬁcally to intercalate the oxidation
catalysts into a more acidic anAl-LDH in order to eliminate the impurity phase
present in all synthesis of LDH-POMS to date.

The ﬁrst step was to intercalate M07038 and HleszS into Mngl- and
anAl-LDHS and repeat the size selective experiments of Tatsumi et. al. It
immediately became obvious that there were some major problems at the most
basic level of the experiment. The identity of the catalytic species was equivocal,
the catalysts were not stable under the thermal or catalytic conditions of the
experiments, and ﬁnally, there was an impurity phase, present in all LDH-POMS
reported to date,2 but which is generally ignored.

A search of the literature Showed that these problems were the result of
ambiguities that have propagated through the literature on LDH-POMS;

Diffraction patterns with unassigned peaks and no evidence of ordered

reﬂections are the only evidence to support the claim that a POM anion has
been intercalated into an LDH. Discrepancies in the reported d spacing are
discussed in vague terms of ‘grafted pillars’ and dismissed. In discussions of
catalysis by POM intercalated LDHS, any possible catalytic activity by the ever-
present impurity phase is ignored. And ﬁnally, the X-ray diffraction patterns of
oriented ﬁlm samples are used to determine the relative amount of impurity
phase present in an LDH-POM, without acknowledging that impurity phases are
obscured in the X-ray diffraction patterns of oriented samples.8 It was evident
that a method for the synthesis of a pure LDH-POM phase was necessary
before the catalytic properties of the LDH-POM systems could be investigated.
In addition, careful characterization of the LDH-POM system would be
necessary, not only to verify the identity of the intercalated POM species, but to
substantiate the thermal and catalytic Stability of the catalyst.

I was eventually able to untangle the MngI- and anAl-POM intercalate
syStems. Raman Spectroscopy and the known aqueous chemistry of molybdate
and tungstate were used to unambiguously identify the intercalated POMS as the
Mngl—Mo70246‘, anAl-Mo7Ozf, Mngl-W7Oz46’ and anAl-HZWqu". X-ray
diffraction of the LDH-POMS before and after use as catalysts Showed that the
Mngl-Mo7Ozf, Mngl-W7Ozf, and anAl-Mo7Ozf were not stable catalysts.
Finally, Al-POM salts synthesized under conditions analogous to the POM
intercalation reactions were used to Show that the impurity phase has a
signiﬁcant reactivity of it’s own.

. Still, all the previous experiments did not result in any new insight into the
synthesis of impurity-free LDH-POMS. It was time to step back and take a
fresh look at the problem. POMS are formed by a series of hydrolysis reactions
as the pH is lowered. Most POM-LDHS are made at acidic pH where the POM

is stable, but LDHS begin to decompose. However, some POMS are stable at

quite basic pH. Why haven’t these POMS been intercalated? Also, some LDHS
are more acidic; ZnAl-LDH can be synthesized at pH 6.4 and relatively pure
anAl-POMS are easily synthesized. But much work focuses on intercalating
POMS into MgAl-LDHS. Why? Nowhere did I see an analytical discussion of
the relative merits of a MgAl- versus a ZnAl-LDH or any other M2+M’+-LDH as
a support for POM catalysts; nor was there any discussion of why a particular
POM was chosen to be intercalated into an LDH for catalysis or any other

purpose.
The impurity phase is believed to be the product of an acid/ base reaction

between the acidic POMS and the basic LDHS. The synthesis of the POMS, the
LDHS, and the ﬁnal LDH-POM is done in aqueous solution. The acid/ base
reaction is very much a result of the aqueous solvent system. An LDH in 0.1 M
acetic acid will decompose overnight, while after 3 days in glacial acetic acid,
there is very little decomposition of the LDH. How would replacing the water
with an organic solvent affect the acid/ base reaction between the POM and
LDH?

All these questions led me to believe that before I could ﬁgure out the
best way to intercalate POMS into LDHS, I needed to thoroughly understand the
LDH support. It is generally stated that the pH required for LDH precipitation
reactions is equal to or higher than the pH where the more soluble cation
precipitates. However, MgAl-LDHS are routinely synthesized at pH greater than
12, where Mgz’precipitates, but Al” is soluble. In addition, ZnAl-LDHS are
synthesized at pH = 6.2, where Zn2+ is soluble. It seemed that the pH required
for precipitation reactions would be any pH where the least soluble cation
precipitates. If this were true, then it should be possible to synthesize MgAl-
LDI-Is at pH less than 8.

In a similar vein, it is generally accepted that contamination by carbonate
is inevitable because of the preference of LDHS for CO3?” anions. There are two
factors that make CO3} highly favored for intercalation into LDHS. The ﬁrst
facror is the divalent charge; since LDHS have a high layer charge, multivalent
anions are preferentially intercalated. Second, the structure of CO3} is such that
the three oxygens ﬁt into adjacent sites of the LDH layers. But is carbonate
really that favorable, or is it merely ubiquitous? Since CO2 from the air easily
dissolves in basic, aqueous LDH reaction systems and forms CO32’, synthesis of
LDHS is generally done under an inert atmosphere (N9, or at a pH less than 7
where HCO3' is the predominant species. What would be the effect be if the
entire LDH precipitation reaction was done at an elevated temperature where
CO2 is less soluble? What would the effect be if the solvent system was organic,
where again CO2 is less soluble?

These questions generated from the literature did suggest two sets of
experiments on the fundamental chemistry of LDH synthesis. My ﬁrst study
dealt with the synthesis of MgAl-LDHS at pH < 8. Even a slightly acidic pH
might make the intercalation of POMS into MgAl-LDHS result in a more
crystalline LDH-POM product. My second study was the synthesis of LDH in
organic solvents. Not only would contamination by C032“ be minimized, the
organic solvent system might be useful in POM intercalation reactions.

The results of these experiments led me in some unexpected directions.
The synthesis of LDHS in MeOH results in a colloidal LDH which makes very
thin, oriented LDH ﬁlms. There are several possible applications for the
colloidal LDHS in the areas of catalysis, sensors and: devices, and ﬁre retardant
materials.

This is all very far from the initial purpose of ﬁnding better ways to
intercalate POMS cleanly into LDHS. But while I didn’t ﬁnd a direct application

of the two new methods to the synthesis LDH-POM catalysts, I did begin to see
a new interpretation of the LDH-POM systems. While the majority of the
catalytic activity demonstrated by the LDH-POM systems is probably
attributable to the impurity salt surface sorbed to the LDH, the selectivity is
enhanced by the basicity of the LDH support. Through such effects as
increasing access to the catalysts, selective adsorption of substrate on the surface
of the LDH, and enhancing product selectivity, the LDH acts as a co-catalyst in
conjunction with the POM. These arguments are clariﬁed in the ﬁnal catalytic

oxidation experiments using Zn:.Al-(NaP_;,W3OOl m)” as the catalyst.

I. SYNTHESIS BY DESIGN

Chapter 2. Layered Double Hydroxides: A Brief Review

Most simply stated, LDHS consist of positively charged mixed metal
hydroxide sheets with charge balancing interlayer anions. Sometimes designated
anionic clays in reference to the anionic intercalates, they are the only general
class of layered anion exchange materials. Three contemporary review articles
on LDH chemistry are available: one by Vacarri in 1991,9 another by De Roy in
199210 and a 1998 review by Jones,” which focuses mainly on LDHS containing

organic intercalates.

History

LDHS are relatively rare natural materials generally associated with
serpentines in metamorphic formations or in saline deposits. They have been
linked to prebiotic chemistry and it is postulated that they may have played a role
in the origin of life.“ The ﬁrst layered double hydroxide to be identiﬁed was
hydrotalcite, ([Mg6A12(OI-I),6]CO3'4HzO). Discovered in Sweden around 1842, it
is easily crushed into a white powder like talc. Also like talc, it occurs in nature as
a foliated material with contorted plates or in ﬁbrous masses. Pyroaurite, a
MgFe-LDH that is similar in appearance to gold when heated, was discovered at
about the same time. It was later recognized to be isostructural to hydrotalcite

The formula for LDHS was determined in 1915 by E. Manasse12 and in
1930, Aminoff and Broomé” recognized the existence of two polytypes of
hydr0talcite, one with rhombohedral symmetry and a second one with hexagonal
symmetry which was called manasseite after Manasse. The papers by Manasse

8

and Aminoff and Broome went relatively unnoticed, so it was not until 1941 that

I” recognized the interrelations between hydrotalcite, pyroaurite and

Fronde
several other minerals. The ﬁrst laboratory synthesis was done in 1942 when
Feictknect15 reacted metal salt solutions with base.

There was a considerable time gap between the discovery and synthesis of
LDHS and the determination of their Structure due to their nonstoichiometric
nature and the difﬁculty in obtaining crystals large enough for X-ray analysis.
The proposed structure went through several successions before Allman16 and
Taylor17 determined the correct conﬁguration to be a layered structure with
cations located in the same layer with anions and water in the interlayer.

Meanwhile, catalysis by LDHS took a separate but parallel path beginning
in 1924 with coprecipitated NiAl-LDH as a hydrogenation catalyst.” This

continued until 1970 when the two paths merged in a patent for the preparation

of a hydrOtalcite-like material as catalyst precursors.19

Applications

LDHS are an excellent example of materials by design. A variety of
framework cations, layer charges, and intergallery anions can be incorporated
into the LDH structure, and it is this very versatility that allows LDHS to be
utilized in exceedingly diverse areas. Industrial applications of LDHS can be
loosely divided into four areas; pharmaceuticals, anion exchangers and
adsorbents, the improvement of materials properties, and catalysis.

An early application of LDHS in pharmaceuticals was in the treatment of
peptic ulcers. The buffering pH of MgAl-COf’ is around 4, it has a high
antipeptic activity, dissolves completely, and intestinal absorption is low. The

high anion exchange capacity is utilized by replacing the carbonate with salicylic

acid for use as an anti-inﬂammatory medication.20 Isocarbostyril derivatives used
in the treatment of heart disease are stabilized by intercalation into LDHS.”
Many forms of LDHS are not only non-toxic but can also supply essential
nutrients. Iron-containing LDHS are used in treating iron deﬁciencies? Other
uses include ointrnents and poultices9 for the protection of damaged skin.

There are a wide variety of applications of LDHS as anion exchangers and
adsorbents in the area of environmental remediation. The ion exchange
capacity, 2-3meq/ g, is similar to that of anion exchange resins, but because of
their higher thermal Stability, LDHS are used in the treatment of the cooling
water in nuclear reactors.” Mg3Al-CO32' can be used to remove low levels of
acidic impurities in synthetic lubricant oils and plasticizers that can not be
eliminated by traditional adsorbents.24 MgAl-LDHS have been tested for
adsorption of Tc7+ from radioactive wastes.25 Stereoselective isomer separation
can be achieved through anion exchange reactions.26 Perhaps the most
important applications are the adsorption of ecologically undesirable anions
from dilute waste streams and of halogens in polyoleﬁn production.27'28

Flame retardation, puriﬁcation, and Stabilization of polymers are the main
applications of LDHS in the improvement of material properties. Resistance to
combustion in polymers is attained through alteration of the degradation or
combustion processes by the addition of certain chemicals. LDHS act mainly by
dilution and by the generation of non-ﬂammable gases. The thermal
decomposition of LDHS creates large amounts of non-ﬂammable CO2 and H20,
reducing the concentration of O2 at the ﬂame front. The metal oxide products
of LDH combustion also act as a diluent, reducing the concentration of
combustible material.

The greatest number of LDH patents in recent years have been for the

puriﬁcation and Stabilization of polymers. An LDH will adsorb HCl produced

10

as vinyl chlorides degradation by heat or UV light, thus protecting them from
further degradation.28

Another materials improvement applications is compression mouldability.
LDHS are more compressible than the Standard microcrystalline cellulose,
capping does not occur and the ﬂaky crystals are oriented, producing product
with good luster. One to two tons of pressure/cm2 will form a 1 mm thick
transparent plate which can be used in agricultural ﬁlms.29

When added in small amounts LDHS can improve the ﬂexibility and
mouldability of a material. This is useful in producing molded heat resistant
insulation parts such as switchboards, capacitors and insulation wires.30

Additional materials applications include:

1. Spacer sheets in electrolytic capacitors”
2. Corrosion inhibitors in paints and coatings32
3. Protection against bacterial corrosion”

Perhaps the largest area of industrial applications for LDHS is in catalysis.
There are two approaches to using LDHS as catalysts. The ﬁrst concentrates on
the framework cations. Mixed metals oxides produced by calcining LDHS at
500 °C, form a large group of industrial catalysts.9 Very recently, MgAl-LDHS
have been used as co-catalysts in the oxidation of alkenes to epoxides. The
basicity of the LDH aids in oxygen transfer and increases product selectivity?4
The second approach concentrates on the intercalated anion. Many catalytic
species are anionic, and intercalation into LDHS can prevent catalyst
degradation, control access to the catalysts or transform a homogeneous catalyst
to a heterogeneous catalyst. Catalyst degradation can occur through different

routes. Loss of ligands can occur in many

11

 

O M2+ O M3+

OH Groups
. Intergallery Anions

 

Figure 2-1. A schematic of a 3:1 MZ’L/M’+ Layered Double Hydroxide

12

organometallic catalysts as the organic ligands are oxidized. Cannibalism can
occur when the catalyst reacts with itself to form an inactive species

Many catalysts are functional in homogeneous solutions, making them
difﬁcult to remove when the reaction reaches its endpoint. Since the
concentration of catalyst to product is generally very small relative to the product
being formed, the catalyst may just be left in the ﬁnal product. Supporting the
catalyst by intercalating it into an LDH transforms it into a heterogeneous
catalysts that can be ﬁltered from the reacﬁon mixture. This adds economical
value to the catalyst as it has the potential to be reused many times, and
environmental value, especially in the case then the catalyst contains toxic
materials. Miscellaneous applications include: the removal of Al3+ from recovery
systems of a closed kraft pulp mill,’5 novel conducting materials}6 and clay-
modiﬁed electrodes.37

Recent research efforts into applications for LDHS include phenol
hydroxylation by Cu containing LDHS,38 LDHS as molecular containers for the
preparation of microporous carbons,39 clay modiﬁed electrodes,40 and as

1 Myrisate intercalated LiAl-LDH membranes can

photchromic sensors,4
partition styrene from a methanol/ water solution."2 Heating MgAl-
dodecylsulfate resulted in the formation of C60 between the layers indicating that
LDHS are useful in the preparation of new materials.” The interaction of TiOx
with LDH-porphyrins results in a catalyst for the photodegradation of

pollutants.“

Structure

The most detailed investigations into the structure of LDHS were done by

Allmann45 and Taylor.46 The main characteristics are the brucite-like sheets, the

13

positions of interlayer anions and water, and the type of stacking. The layers are
composed of inﬁnite sheets of edge sharing M2+ and M3+ octahedra with 6-fold
coordination to OH' (Figure 1). The interlayer domains are in a quasiliquid47
state with the anions solvated by one or two layers of water.

The divalent and trivalent cations randomly occupy the octahedral holes
in the close packed conﬁguration of OH' ions. The interlayer anions and water
are free to move by breaking and reforming hydrogen bonds with the layer OH'.

LDHS consist of two polytypes,48 a 3-layer polytype with rhombohedral

crystal structure, space group = R2m and a two-layer hexagonal polytype, space
group = P63mmc. These are designated 3R and 2H, respectively. In natural
LDHS the hexagonal polytype may be high temp form of the rhombohedral.
The d spacing, c = 23.4 A, is 3 times the ﬁrst order peak for a 3R LDH, where
c’= 7.80 A. The hexagonal lattice parameters for Mg3Al-CO32’ in the 4,17 plane
are 3.06 A, the average distance between neighboring metal cations in the
hydroxide bilayer. The cation distances are unaffected by the polytype.

As Stated in the previous section, the My/Mi’+ ratio varies continuously,
and integer values of the stoichiometric coefﬁcients are favorable for the
appearance of superstructure, ex. for M27 M3+ = 2:1, a=ao\/3; for M2+/ M3” =
3:1, a=2ao. A superstructure has only been detected in the case of Mngl-LDHS,
verifying the random distribution of the cations.

Heterostructures with regularly alternating interlayer anions, have been
seen when both CO32' and SO 42’ are intercalated into an LDH. The d spacing is
the sum of the two layer heights with a total d spacing of 16.5 A (7.6 A + 8.9
A)”

14

Composition

Anionic clays are two-dimensional solid bases composed of positively
charged sheets of di- and trivalent metal hydroxides with Shared octahedral
edges and contain intergallery anions and water.9 The parent material is brucite
Mg(OH)2 composed of neutral layers held together by H-bonding and Van der
Waals interactions. Substitution by a trivalent anion such as Al3+ results in a net
positive charge compensated by solvated anions in the interlayers with a general
formula of M2+(1_X)Mx’+(OPDZ'An'x/n'mHzO. (Figure 1) The ratio of MH/M3+
ranges from approximately 4 to 2, which translates into an x value of 0.20 to
0.33.

The divalent cations can be Mg”, Zn}, Niz”, or any divalent cation with a
radius less than that of Baz”. The charge is introduced into the layers by
fractional substitution of M2+ with a trivalent cation such as Al“, Cr’I, or Fe“.
The layer charge, while quite high when compared to cationic clays, is easily
manipulated by controlling the MZI/M3+ ratio. Although the charge balancing
anion, A”, can be almost any negatively charged species, multiply charged anions
such as carbonate are preferentially intercalated.” The intergallery anions are
anions are solvated by water molecules. The number of water molecules (m)

varies with x, n, and the size of the anion.51
The Nature of M‘”

A list of the cation combinations used in to synthesis LDHS is shown in

Table 1. Cations with an ionic radius in the range of 0.50-0.74 A allows
them to ﬁt in the Oh holes of close packed OH' groups. Ions such as Be2+ (0.59

A) are too small, and Ba2+ (1.35 A) is too large to occupy an Oh position without

lattice distortion. For bigger cations such as Ca2+ (0.98 A) and

15

Table 2-1. Different Metals Used in the Synthesis of Layered Double

Hydroxides
MgAl‘
CuAl
MgAJ
NiAl
MnAl
NiAl
NngAl

CuZnAl
ZnCr
ZnAL
CuCoAl
MgCr
FeAl
CuCoZnAl

a. From references 9 and 10, unless otherwise annotated

16

CuZnCr
CoAl
LiAl
CuCoCr
MgAlV52

Cd2+(0.97 A), the octahedral environment becomes unStable. While
Ca2+ and Cd” are borderline, LDHS have been synthesized from these
metals.”54 All trivalent anions from 0.5 to 0.8 A form LDHS, though V” and
Ti3+ are notair Stable. In addition, LDHS with more than 2 cations are
possible.

Copper containing LDHS can only be made when another M2+ is present
in the greater concentration. When the Cu2+ cations are separated by another
divalent anion, the Cu2+ forms the undistorted octahedra typical of LDHS.
When the CuZI/MZI ratio is greater then one, adjoining Cu2+ octahedra can form
a copper compound with a jahn Teller distortion energetically favored over the
hydrotalcite structure.

The fractional substitution, x, (x: M’+/ (M2+ + M“) can range from 0.5 to
0.167, but pure phases appear only in the range of 0.33 to 0.2 and the 0.2 value
has been disputed. Another way of expressing the amount of substitution by
M’Iis the My/M3+ ratio. Therefore, when x = 0.33, MEI/M3+ = 2; when x =
0.2, Mz”/M3+ = 5. The theoretical maximum limit M3+ substitution is M2+/M3+
= 2. If the substitution by M” were greater, there would positive charges in
adjacent octahedra. When MZI/M“ = 5, the high density of divalent cations
leads to the formation of brucite-like M(OH)2.

There is a statistical distribution of trivalent cation substitution in LDHS.
This random occupation of all octahedral sites allows the M2I/M3I ratio to vary
continuously throughOut the allowable range of x. The ratio of M2"/M3+ added
to the reaction is the ratio found in the ﬁnal LDH; product, although this is
somewhat dependent on the synthetic method. This ratio can be easily veriﬁed
by the size of the unit parameter a, which is not affected by intergallery anion.

LDHS adhere to Vegard’s law which states that increasing substitution by a

17

smaller anion will cause a proportional decrease in the size of the unit cell. The
in-plane parameters, a=b, have been indexed with hexagonal axes. In an ideal
structure, a=2'/2rm,0) where r= the distance between the M2+ and O centers. For
brucite, where x = 0, this is equal to 3.15. In an LDH, ram) is replaced with the

mean ionic radius, 2, where z = [(1-x)r“2+ + x(r“3,)]. Therefore, 4:21/22. Since a
ji x, they can be plotted to form a line with a slope of -2'/2[r“2+ -(r\l3+)]. This is

valid within the range of x = 0.2 and 0.33. MiyataSS observed an increase in a
when x < 0.2, but in addition to the LDH, brucite and hydromagnesite where
observable by XRD. Above x = 0.33, a becomes constant. Pausch“ claims that
LDHS with x = 0.44 is possible for MgAl LDHS and that the constancy of a is
due to the Al”'-Al’+ repulsion balancing the decrease in 2.

In addition to the most common Mz’/M3+ LDH, other combinations of
metal cations are possible. The M'I/M’+ combination of LiAl-LDH has been
extensively studied and there have been reports of Mz’/M4+ combinations of Co

and Ti.

The Nature of A"'

The anion exchange properties are one of the more spectacular
characteristics of LHDS. There is almost no limit to the nature of A"' and
exchange reactions are extremely simple. The sample to be exchanged is
maintained in an aqueous solution of anion to be inserted. The only problems
are contamination by carbonate and forming pure crystalline material. It is
generally assumed that the interlayer anions are present in the precise amount to
attain charge neutrality, but there are exceptions. Hansen and Koch57 have
found a charge imbalance ranging between +0.77 and -1.09 eq per formula unit
in MgFe-LHDS. The charge imbalance depended linearly on x and was
attributed to the amphoteric properties of the hydroxyl groups.

18

Powder X-ray diffraction and infrared spectroscopy are the two main
methods used study anion exchange reactions in most LDH-A"' systems.
Infrared spectroscopy is often used to verify the presence of the intercalated
anion and powder X-ray diffraction is used to track the change in layer Spacing
caused by the exchange reaction. The diffraction pattern can lend insight into
the number, Size, and orientation of the anions as well as the strength of the
bonds between A"' and the OH' groups of the LDH layer.

The anion affects the thickness of interlayer regions in two ways. The
greatest effect on the layer separation (the d Spacing minus 4.8 A, the thickness
of a brucite layer) is the physical size of the anion. For example, the d spacing
for an LDH intercalated with F‘ is 7.51 A and for n-C18H37SO4' it is 32.6 A.58
Similarly, the d spacing increases linearly with the chain length of an organic acid
and with the ionic radius of halogens.

The orientation of the intergallery anion also has an effect on the layer
spacing. When a-PszOoz" is intercalated into an LDH with the D3}, axis
perpendicular to the layers the d spacing is 2.1 A greater than when the D3}, axis
is parallel to the layers.59 The same argument has been put forth for the
differences in d spacing between SO 42' and C1041

These isostructural anions have a difference in d spacing of 0.3 A when
intercalated into LDHS(See Table 2-2). Tetrahedral X04“ anions can have at
least two orientations with respect to the LDH layers. An orientation with one
apical oxygen pointing to one hydroxyl plane and the other three pointing to the
adjacent hydroxyl plane. This orientation corresponds to the larger d spacing
and would allow more of the monovalent ClOf' anions to be intercalated.
Anorher orientation would have two oxygens coordinating to the opposite
hydroxyl planes and two in the interlayer region. This corresponds to the

minimum layer separation and could be adopted by SO 42'. However, Bish"O

19

Table 2-2 Inorganic Anions That Have Been Intercalated Into Layered

 

 

Double Hydroxides
Anion LDH Framework d spacing/ A Ref'
OH‘ MgAl 7.55
CO3?" MgAl 7.65
F ' MgAl 7.66
Cl' MgAl 7. 86
Br' MgAl 7.95
I' MgAL 8.16
NO3' MgAl 8.79
SO 42‘ MgAl 8.9
ClO,’ MgAl 9.2
103'
8303}
WOf'
CrOf'

FC (CN)63-,4- 61

SiO(OH)3'

Ru(BPs) "H’ MgAl 62
TcOf’ MgAl 2’
ReOf’ “ “ “

SPc MgAl 17.7 (’3
PTSd “ “ “

1C: MgAl 17-22 8‘
NC “ “

From Cavani et. al9 and references within, unless otherwise noted.

BPS = 4,7-diphenyl-1,10-phenanthrolinedisulphonate

SP = sulphonated indolinespirobenzopyran

PTS = p-toluenesulfonate _

IC = disodium 2-(1,3-dihydro-3-oxo-5-sulpho-ZI-l-indol-2-ylidene)-2,3-dihydro-3-oxo-
1 H-indole-S-sulphonate

trisodium 7-hydroxy-8-[(4-sulpho-1-naphthalenyl)azo]-1,3-naphthalene disulphonate

{DP-90"!”

2"“.

20

states this orientation would destroy the rhombohedral symmetry and Miyata64
attributes the difference in d spacing to the greater electrostatic attraction
between the OH' and the divalent SOf‘ as opposed to monovalent 00;. This
is the second way the anion can affect the d spacing of an LDH, through the
strength of the electrostatic interactions between the anion and the layers.

In addition to the great number of inorganic anions which can be
intercalated into LDHS, there are an impressive number of organic anions that
have been intercalated into LDHS (Table 2-2). In many way, the behavior of
LDH-organics mirrors that of cationic clays. Layered double hydroxides
intercalated with alkyl sulfates and alkyl sulphonates undergo substantial swelling
in n-alkylalcohols or n-alkylamines.’8 Dodecylsulfate intercalated into NiAl-,
MgAl-, or ZnCr-LDHS can take on three packing arrangements. The shortest
interlayer spacing, 26 A correlates to a perpendicular monolayer. The
intermediate d spacing, 36 A was attributed to a bilayer with a tilt of
approximately 40 ° or an interleaving vertical bilayer. The largest d spacing, 47 A
was consistent with a vertical bilayer of dodecylsulfate."5 The most thoroughly
studied organic intercalates have been terephthalate (TA) and benzoate. These
studies demonstrate the balance of forces that determine the interlayer
arrangement of anions. In a 2:1 MgAl-TA, the gallery height was 9.2-9.6 A (the
physical length of TA = 9.9 A). The d spacing of MgAl-benzoate was 15.2-15.5
A (benzoate = 8.8) which is proposed to be a bilayer with a tilt angel of 35 °.
The d spacing was the same for TA and benzoate in a 3:1 MgAl-LDH before
drying. But when the 3:1 MgAl-LDHS were dried at 60 °c, both TA and
benzoate oriented with the long axis parallel to the layers. The gallery height of
3.5 A is the thickness of a benzene ring. The change in orientation was

reversible for repeated cycles of rehydration and drying. The ability of

21

Table 2-3. Organic Anions That Have Been Intercalated Into Layered

 

 

Double Hydroxides
Anion LDH d spacing/ A Synthesis Ref.
Framework
CnH2n(COz‘)2 ZnAl 9.4-1 8-1 Exchange 6"
n=0-8
alkyl sulfate ZnCr Exchange (’7
aromatic Exchange (’8
diphosphonates
b-cyclodextrin MgAl Exchange (’9
MgAl Rehydration 7”
MgAl Rehydrationb 7‘
naphthalenecarboxylate Rehydration 73
Co(II)pthalocyanine Rehydration 73
tetrasulfonate
sebacate MgAl Thermal 7‘
exchange
caprate, Thermal 75
phenylphosphate exchange
terephthalate MgAl 14.6 Coprecipitation 7"
benzoate 1 5.2 36
CnH2n+ iSO4‘ ZnCr 21.1 -32.61l 77
napthalenedisulfonate ZnAl 73
9,10-anthrquinone-2,6- MgAl 12, 19C Coprecipitation 7"
disulfonate
BHMBSd ZnAl Exchange 30
TSPPc MgAl 22.4 Exchange 3‘
phosphonates MgAl, CdAl Exchange 68,32
organo polymers MgAl, CaAl Exchange
Ru(BPS)3"' MgAl 22 anion exchange 62

The d Spacing increases linearly with the aliphatic chain length.

Gycerol was added as a swelling agent

c. D spacing was found to be dependent on the orientation of the 9,10-anthrquinone-2,6-
disulfonate gallery anion

d. 5-benzoyl-4-hydroxy-2-methoxybenzenesulfonate 5,10,15,20-tetra(4-sulfonatophenly)

porphyrine

ST!”

22

water to stabilize the vertical packing for 3:1 MgAl-LDH emphasizes the
importance of the hydrophobic/hydrophilic interactions

The interlayer water

There are two types of water associated with LDHS, interlamellar (intrinsic)
water and surface sorbed (extrinsic) water. Recent 1H NMR studies have shown
that in MgAl-C032“ LDHS, the interlayer water is freely spinning with its C2 axis
perpendicular to the LDH layers.83 The number of water molecules (m) varies
with layer charge, anion charge and size and temperature and humidity. A
formula by Miyata used to calculate the maximum amount of water illustrates the

above relationships;
m: 1 -’\lx / n

in terms of My“,,)M’+x(OH):,A"'X/n'mH30, where N is the number of sites
occupied by the anion, x = M”/ M2++M"), and n is the charge on the anion.
When N increases, the number of sites in the close packed conﬁguration
of oxygen atoms that decreases, leaving fewer sites for occupation by water. The
same relationship is true for x, as x increases, more charge balancing anions are
required. There is a reverse relationship with n, as the charge on the anion
increases, fewer are needed leaving more sites for water. For hydrotalcite,
Mgs -CO32', m=1-3(.25) / 2. In this case m = 0.625 or Mg6A12(OH),6CO3'5H20,
which is higher than the measured value of m=4.
The amount of water is measured by thermogravimetric analysis as
measured by weight loss with increasing temperature. The weight loss is
reversible at low temperatures, but at 60-80 °C, it becomes irreversible. A similar

effect is seen in LDH-8042‘ and -ClO4'. Both of these LDHS have an reversible

23

increase in d spacing when placed in an environment with >50°/o humidity. The
d spacing does not change with humidity when anions such as CO3} or NO3’ are

the gallery anion.

Table 2-4. Change in d spacing with Increasing Humidity

Low humidity 50% humidity

d spacing/ A d spacing/ A
LDH-5043‘ 8.9 10.8
LDH-C104' 9.2 11.7

In addition, an LDH in aqueous suspension will not absorb more than
one or two layers of water. This illustrates an important difference between
LDHS and traditional cationic clays, LDHS are non-swelling in aqueous solution
due to the high layer charge, 25 A2 / unit charge. An exception to this is seen
when the gallery anion is a long chain organic. These LDHS will expand in
alcohols and glycols84 in proportion to the size of the solvent molecule.

In addition to interlayer solvation of the guest anions, there is also water
adhering to the surface of LDH crystals. Since this water had been found to
effect the proton and ionic conductivity's, acid/ base activity, and ion exchange

properties, it should be included when considering an LDH intercalated system.

Synthesis

In addition to being easy to synthesis in the lab, LDHS are truly ‘materials
by design.’ The layer charge, layer cations, and gallery anions are remarkably
variable. Still, this does not eliminate the need for a precise chemical foundation
in the choice of synthesis conditions. The nature of the ﬁnal product is very

dependent upon the reaction conditions. The choice of the original MZI/M’I

24

composition, precipitation conditions, hydrothermal treatment, type of reagents,
aging, washing, drying, calcination, and activation all affect the homogeneity,
crystallinity and particle size of the ﬁnal product. There are a few requirements
which are general to any LDH synthesis; the right ratio of cations and anions
must be present, the pH domain of anion and LDH must overlap, and the
desired anions must be present in higher concentration and have greater afﬁnity
for the LDH than any other anions present in the reaction. For many of these
reactions, the ratio of Mz’i/M3+ in the reactants is directly reﬂected in the
product, especially in the range of MZI/M3+ = 2-4. Carbonate from dissolved air
must be rigorously excluded, usually by running the reaction under an inert
atmosphere, and still cannot be completely excluded. Finally, the anions must
have a greater afﬁnity for the LDH than any of the cations. From there, each
synthetic method has its own advantages and requirements. However, in all
these methods, a small amount of CO3?" is present in the gallery, no matter what
precautions are taken.

There are three general methods for the synthesis of LDHS which can be
loosely deﬁned by the starting reagents: coprecipitation from metal salts,

synthesis from metal oxides, and synthesis using a preformed LDH.

Precipitation

Precipitation from metal salts is recognized as the most reliable,
reproducible technique and is used in almost all LDH synthesis. The advantages
of synthesis by precipitation include the preparation of an LDH with a deﬁnite ,
composition, the anion is ﬁxed by the metal salts or in solution, and the
crystallinity and particle size are easily controlled. This method is often used to
make LDHS precursors for subsequent anion exchange reactions as the

crystallinity of the LDH of the original LDH is maintained in subsequent

25

treatments. In precipitation reactions the Structural and physicochemical
properties depend on:

1. method of precipitation
nature of the reagents
temperature

presence of impurities

aging times

999.99!“

washing conditions
7. drying conditions

Precipitation methods are usually done under conditions of
supersaturation. Within this general class there are two types of precipitation
reactions, coprecipitation and sequential precipitation.

Coprecipitation reactions necessitate aqueous mixtures of M3’ and M24" at
a pH equal to or higher than the one at which the more soluble cation
precipitates and can be done at high or low supersaturation. In precipitation
reactions done at high supersaturation, a mixed metal solution containing the
divalent and trivalent cations is added very quickly to a solution of NaOH or
NaHCO3. The reaction is generally aged at a temperature of 60-90 °C. The
product obtained by coprecipitation at high supersaturation is generally less
crystalline than an LDH precipitated at low supersaturation due to the high
number of crystallization nuclei.

Coprecipitation at low supersaturation is accomplished by the slow
addition of the M3"/M3+ solution with the simultaneous addition of base (KOH,

NaOH, or NaHCO3 if a carbonate LDH is desired) in two dilute streams into a

single ﬂask containing a solution of the interlayer anion. This reaction is often
done at a constant pH. These products are generally aged at 60-80 °C and result

in a more crystalline product than coprecipitation as high supersaturation.

26

The second type of precipitation reactions are sequential precipitation or
the increasing pH method. Here supersaturation can be achieved either by
evaporation or more generally by raising the pH. Taylor’s induced hydrolysis is
an example of sequential precipitation. For example, a solution of Al(NO3)3 is
precipitated at pH 7, then Zn(NO3)2 is added, maintaining the pH at 6.4 with
NaOHW. These reactions are an advantage for LDHS containing metals such as
Ni which precipitates at a lower pH or preparations with Cu2+ and Zn:+ which

85 A disadvantage of the sequential

form several mono and binary compounds.
precipitation reaction is that often the MEI/M3+ ratio added to the reaction is not
reﬂected in the product.

The previously described reactions are all done in aqueous solution.
There are a few examples of LDH synthesis done in non-aqueous solvents. In a
1992 patent, Burba et al.86 reported on the synthesis of layered mixed metal
hydroxides from metal alkoxides in organic solvents such as diethylene glycol or
methanol. In these examples, a stoichiometric amount of water was added as
the source of hydroxide.

Malherbe et a1.87 synthesized layered double hydroxides in a mixture of
organic solvent and water in an N2 environment. The solvent systems were
50:50 mixtures of water with methanol, ethanol, propanol, acetone or glycols.
Their method resulted in microtextural modiﬁcation of the surface and porous
properties of the LDHS.

While precipitation conditions are the most widely used method for
attaining a crystalline LDH, they are not always necessary as:

1. aging can rectify improper precipitation

2. sequential precipitation does not occur

3. pure LDH may not be required and synthesis may be followed by

thermal treatments and/ or ion exchange.

27

Salt-Oxide method

Another method for LDH synthesis is called the salt-oxide method.10 It
has also been labeled hydrothermal synthesis, but it becomes hard to
differentiate between hydrothermal synthesis and hydrothermal treatments.
Here the salt oxide method is deﬁned as a synthetic method where at least one
of the cations is present as a solid in the reaction. These reactions can range
from high temperature, high pressure reactions to reactions done under ambient
conditions. Reactions at temperatures above 273 °C are done in an autoclave or
in a sealed tube under pressure. The ﬁrst synthesis of an LDH from a
mechanical mixture of MgO and AlZO3waS reported by Roy et al.88 in 1988. The
mixtures were autoclaved under COZ, at a temperatures to 325 °C, and pressures
to 130 Mpa. The product was a mixture of LDH, magnesite and
hydromagnesite. Synthesis at higher temperatures resulted in the decomposition
of the LDH.

Thermal methods for LDH synthesis generally consist of calcining metal
oxides at temperatures under 600 °C and pressures from 13 to 130 Mpa under
water vapor and C02. The LDH forms very slowly, and the reaction is typically
incomplete.

The ﬁnal methods for the synthesis of an LDH-A“‘ are those that begin
with a preformed LDH. Most of these synthetic methods can be typed as anion
exchange or structure reconstruction. Anion exchange reactions can be

accomplished through several routes including: exchange of an easily replaced
anion such as OH- or NO3',9 synthesis of an LDH with a large anion and
exchanging with a metalate anion,76 and swelling the LDH layers with glycols

prior to the anion exchange reaction.89
Anion exchange reactions of LDHS are very easy considering the high

layer charge. Mica, one of the most highly charged Silicate clays, is a poor cation

28

exchanger. LDHS do not delaminate in solution like clays, so the exchange
reactions are topotactic. The ﬁrst mention of the anion exchange properties of
LDHS was in 1973, where the selectivity coefﬁcients for NO3', ClO4', Cl',
CH3OO‘, CO32', CrOf‘, and SO41 were given.’0 In 1977, Bish and Brindley90
found that the basic LDH structure of takovite treated with HCl was not
destroyed. The only change was in the d spacing which had expanded to
accommodate the intercalation of Cl‘ anions. Since these initial experiments,
LDH exchange reactions have become a very important method for synthesizing
the desired LDH-A“’ system and much research has been done on optimizing
the exchange reactions.
Many of the requirements of exchange reactions are similar to those of
any of the other synthetic methods:
1. matching the anionic charge density with the charge density of the

layers
matching the acid/ base characters of anions and host structure
matching the redox characters of anions and host structure

insuring hydrophilic or hydrophobic sites for organic anions

View?J

the absence of too high energetic barrier in the case of anion
exchange reactons.10

Clearﬁeld" stated that thermodynamically, anion exchange reactions of
LDHS depend on the electrostatic interactions between positively charged sheets
and exchanging ions and to a lessor extent on the free energy of hydration. In
addition, the equilibrium constant increases as ionic radius decreases, favoring
exchange ions with a high charge density.

Selectivity series for anion intercalation have been determined for several
anionic species. Miyata92 in 1983 found the selectivities for the following

monovalent and divalent anions:

29

OH'>F’>CI'>Br’>NO3'>I’,
CO32'>C,0H4NO32’>SO42'

Miyata’s results were veriﬁed by Yamaoka”3 who added these divalent
anions:

HPO,,2'>HAsO,, 2'>CrOf'>SOf‘>MoO42'

Exchange reactions can be facilitated by running the reaction at a pH
where the anion to be exchanged is protonated. In Drezdzon’s method for
intercalating isopolyoxometalates via exchange with terephthallic acid, the
exchange reaction is done at pH = 4.4 where the terephthalate is protonated.
Exchange of carboxylate anions into a LDH-CO3?” should be done at pH 4-6, in
order to protonate the carbonate and expel it in favor of the exchange anion.

Organics anions can be directly exchanged form inorganics. For example,
LDH nitrates have been successfully exchanged with n-CmHZmHSOf',
n='-8,12,14,16,1894 or 5,10,15,20-tetra sulphonato phenly porphyrin.

A relatively new method for preparing LDH-organics has been developed
by Carlino and Hudson.74 A mixture of sebacic acid and preformed MgAl-CO32'
is heated to 20-30 °C above the melting point of the acid. This method has also
been used to intercalate caprate95 and phenylphosphonate, although the reaction
is not complete

The ﬁnal LDH synthesis method takes advantage of the memory effect of
LDHS calcined at 450-500 °C. At these temperatures the LDH loses its hydroxyl
groups to form mixed metal oxides solid solution. While the crystalline structure
is lost, as evidenced by the X-ray diffraction pattern, scanning electron
micrographs showed that calcination does not change the morphology, nor was

96

there exfoliation of the layer Structure. When the solid solution is placed in

water, it rehydrates and combines with anions to reform an LDH. This reaction

30

is so favorable that a synthetic meixnerite (MgAl-OH) will form in the absence
of any other anion.

Thermal/ anion exchange methods recently developed by Chibwe and
Jones"7 begins with the synthesis of a highly crystalline LDH formed by a

controlled pH at low supersaturation. The product is then calcined under N2 at
500° C to form the mixed metal oxides. The metal oxides are added to degassed

water where the LDH restructures with OH- as the interstitial anions. These are
easily replaced with the desired organic or inorganic anion in a subsequent anion

exchange reaction.

Thermal Treatments

Aging is or digestion is generally done at a relatively low temp < 100 °C
for a period of 18 hours to seven days. The time for aging is dependent on the

gallery anion, for example an LDH-NO, needs a longer aging time than CO3}.

Physicochemical Characterization

Characterization of LDHS is done by X-ray diffraction (XRD), Fourier
transform infrared spectroscopy (FTIR), thermogravimetric analysis (T GA),
Inductively Coupled Plasma (ICP), scanning electron microscopy (SEM), and
RAMAN spectroscopy.

XRD

Single crystal analysis of LDHS has been possible with only a few materials
making powder X-ray analysis the main analytical technique for layered double
hydroxides with the presence of both sharp basal peaks and diffuse in plane
peaks. The general features of LDHS are intense sharp basal (001) peaks at low

31

 

(001)

(002)

Relative Intensity

I

 

 

 

 

 

 

 

 

T l I f l l l
10 20 30 40 50 60 70 80
2 theta/ degrees

Figure 2-2. X-ray diffraction pattern of a Mg3Al-NO3' double hydroxide
showing Miller indices.

32

26, corresponding to successive orders of reﬂection of the layer spacing. At

higher 20', the less intense, asymmetric inplane are evident (Figure 2-2).
Turbostratic disordering can make the inplane reﬂections quite broad.

Poor crystallinity, disorder, and the inherent non stoichiometric nature in
the layer stacking of LDHS can lead to broad, asymmetric diffraction lines with
differing relative intensities. The basal (000 reﬂections correspond to successive
orders of the basal Spacing c', and can be used in Bragg’s Law to calculate the

layer distance (d)

d = nA/Z sinO’

The basal reﬂections are designated as the c axis in LDHS. The (001)
reﬂection in Figure 2-2 is actually the (003) reﬂection, or c'. True c is a multiple
of c', 2c' for 2H and 3c' for 3R. The height of LDH layers is known, (in the case
of Mg/ AL 2 4.8 A) and can be subtracted from the d spacing to ﬁnd the gallery

height, verifying intercalation of the desired anion. Increasing My substitution

will cause a decrease in layer separation, due to both smaller ionic radius of the

M3+ valence, and increased electronic attraction caused by the increase in layer
charge.

The in plane reﬂections are indexed on a hexagonal axis, and are
independent of layer stacking and dependent only the cationic species and x.
The crystal parameter a can be calculated from the (110), a=2dmo) dependent
only upon cation, x and the degree of hydration.

LDH samples for XRD tend to be oriented due to the platy morphology of
LDH particles. Orientation will increase the intensity of reﬂections of the c axis

relative to the in plane reﬂections. It will also tend to mask the presence of

33

Table 2-5. Identifying Wavenumbers for CommOn Gallery Anions

 

 

Anion Characteristic Ref-
wavenumber/ cm "
C03?" 1355 9
N03' 1385 93
5043' l 100
carboxylate l 560, 1400
carboxylic acid 1700
Fe (CN) 63‘ 2147 99

 

34

impurities and so unoriented powder samples should be used when looking at

the relative purity of an LDH product.
FTIR

Gallery anions retain much of their IR features after being intercalated
into LDHS, so FTIR is useful, not only for the characteristic absorption features
of the LDH, but also for verifying retention of the guest structure.

In MgAl-CO32' LDHS there is a Strong adsorption band at 3512-3448 cm'1
indicating H-bound layer hydroxyl groups. When x=0 (brucite) the OH band is
3700 cm" and increasing Al3+ substitution causes broadening and a Shift to lower
frequency. The half width decreases with 2:1 LDH indicating cation ordering.
A shoulder at 3000 cm'1 is caused by H-bonding with the gallery anion.
Interstitial water has a bending mode at about 1635 cm".The identifying band
for CO32' is an v, band at 1350-1380 cm‘1 (1355 cm'1 in these spectra). A
Shoulder at 1400 or a doublet at 1399 and 1362 cm'1 from the splitting of the

CO3} V3 suggests a lowering from the ideal D3h symmetry for the interlayer

anion. In addition, a V2 band at 850-880 cm‘1 can be assigned as a vibration of
CO32-. A band at 1624 could be evidence of HCO32'. Nitrate is distinguishable
by a band at 1485 cm'1 and SO42’by a band at 1110 cm'1

TGA

There are two main features in the thermal behavior of LDHS, both
endothermic. The ﬁrst feature is the loss of surface and interlayer water and the

second transition is the loss of hydroxyl groups from the layers and the gallery

. . . 7+ 3+ . .
amons. Many factors affect the two transrtton states; M / M composrtton and

ratio, type of anions, thermal treatments, and aging.

35

. . . 3+ . .
Each transmon can have two stages. The ﬁrst stage in Al containing

LDHS ranges from 97-297° C, with increasing Al3+ substitution leading to
increasing transition temperature. The two stages can be attributed to extrinsic
surface water, and intrinsic water in the gallery. Samples with small crystalline
dimensions may have large amounts of physisorbed water and the dehydration
and dehydroxylation transitions may overlap. The loss of interlayer water is not

accompanied by loss of Structure, and at low temperatures is reversible100

In the second transition state, the two Stages correlate to the loss of OH-

from Al3+ and then OH- loss from Mg2+ and CO: in hydrotalcite

MgGAL2(OH)12(CO3)4HZO, for example. The temperature at which the loss of

OH and anions occurs is dependent on MZIM” ratio, type of anion, low
temperature thermal treatment, and in the case of oxidizable elements such as
Cr“, the gas used in heat treatment. For example, NiAl-LDHS dehydroxylate
and a lower temperature than MgAl-LDHS and Cr containing LDHS at a higher
temperature. If calcined below 500 °C, the LDH will regenerate in water even
though the XRD pattern shows a change in the structure to that of mixed metal

oxides. If calcined above 550 °C the irreversible formation of spinel occurs.
ES R

Electron Spin Resonance has been used to study the decomposition of

Cu2+ containing LDHS.101

SEEM/TEM

Along with transmission electron microscopy (TEM), scanning electron
microscopy (SEM) indicates textural and crystal morphology. TEM can also be
used to calculate interlayer distance and compared to the value obtained through

XRD.

36

SYNTHESIS OF MGzAL-LAYERED DOUBLE
HYDROXIDES AT ACIDIC PH.

INTRODUCTION

It is generally accepted that the synthesis of an LDH should be done at a
pH where the most soluble M(OH)n precipitates.’ However, there are many
exceptions to this rule. A very relevant example is Taylor’si induced hydrolysis
method of LDH synthesis by sequential precipitation. In this method, a fully
hydrolyzed cation causes the rapid hydrolysis of the second cation at 0.2-0.5
pH units below that at which the second cation would normally precipitate. A
modiﬁed form of Taylor’s induced hydrolysis method was used by Kwonii to
form pure anAl-NO3'. Synthesizing the LDH at pH 6.2 eliminated the
byproduct Zn5(OH)8(NO3)2, which forms at basic pH.

Obviously there is some uncertainty as to what pH should be used to
synthesize a speciﬁc LDH. In a 1992 review of LDH, De Roy’oet al. states, “If
one tries to relate individual precipitation pH of metals and the precipitation
pH of mixed hydroxides no clear relation appears,” (Figure 3-1).10 For
example, A1"+ will form a well ordered ZnAl-LDH at pH 6.2, while the
synthesis of ZnCr-LDH must be done at a lower pH (4.4-5.5) even though the
pH for Cr(OI-I)3 formation encompasses that for Al(OH)3. (The ﬁgure from
reference 10 indicates that a well ordered ZnAl-LDH forms at pH 7-9, but that
does not preclude the presence of the Zn impurity which is avoided by

synthesis at pH 6.2.).

37

WWW
012 3 4 5 6 7 8 9 101112

(“W009
(MgAlCl)
(COFGC')

(ZnNC') 'Ill III/II. VIIA

‘°"°'°" W
(NW)

 

 

 

ZZZ LDH existing domain pH
well-ordered LDH
* Hydrothermal treatment

Figure 3-1. Schematic of pH where M(OH),, and Layered Double
Hydroxides form

38

One problem encountered with the synthesis a Cr-LDH is the number of
oligomeric Cr(OH)3 species that form in the pH range 6-10. So the ﬁrst criteria
should be to avoid any pH where undesirable byproducts form. In the case of
ZnAl-LDH, this means avoiding a pH greater than 6.4. But how does the pH
control the formation of an LDH? The most thoroughly studied LDHS are
MgAl-A"' and these have been synthesized at pH values ranging from 8 to above
12. Even Taylor17 did not synthesize MgAl-LDHS at a pH lower than 8.

Three fundamental methods for MgAl-LDH synthesis in the pH range of 8
to 12 and above are methods by Miyata,'0" Reichle, and Drezdzon.76 These
methods, along with Taylor’s induced hydrolysis method for anAl-LDHS are

105

described in Table 3-1. Miyata’s method of coprecipitation at constant pH is the
one generally recommended as the easiest method for producing crystalline
ngAl-LDH with accurate control of the MZ’i/M3+ ratio. At pH 10, both Mg
and Al hydroxides will precipitate as well as the desired LDH product. To insure
that the ratio of cations in the ﬁnal product reﬂects the ratio of the reactants,
very careful pH control is required in order to avoid the formation of M(OH)2
and/ or M(OH)3 byproducts.n The advantage of this method is that the
digestion Step may be very short or completely eliminated.

Reichle’s method is coprecipitation at high supersaturation with variable
pH. A solution of Mg2+ and Al3+ is added rapidly and with good stirring to a
solution of sodium hydroxide and sodium carbonate. The amount of the
hydroxide should be equal to two times the Mg and three times the Al content in
moles. The amount of carbonate Should be three times of greater than the
stoichiometric amount required to balance the layer charge. While the ﬁnal pH
should be in the range of 8-11, when the mixed metal solution is ﬁrst added to
the caustic solution, the pH is greater than 12. At this point only Mg(OI~l)2 and

39

Table 3-1. Three Methods Used to Synthesize MgAl Layered

Double Hydroxides

 

Method used to prevent

 

 

 

 

 

 

Method pH Precipitates Co3z' contamination
Miyata104 10 Mg(OH)2, The reaction is run under
coprecipitation at low Al(OH)3, N2
supersaturation, constant MgrAl-
pH Cl‘,NO3', etc.
Reichle105 8-11 Mg(OH)2, The reaction is run under
coprecipitation at high ngAL-COf‘ N2 in the case of
supersaturation, variable or meixnerite
pH MgrAl-OH’
Drezdzon“ >12 Mg(OH)2, Reaction is run under
coprecipitation at low Mg3Al- inert N2
supersaturation terephthalate
Taylor102 6.2 Al(OH)3, Sparging the reaction with
sequential precipitation at anAl-NO3' N2 is will prevent
constant pH contamination by C033-

 

 

 

 

40

 

MgAl-CO3; will form a precipitate, although the pH decreases over the course
of the reaction. When 2:1, 3:1, and 4:1 MgAl-LDHS shown in (Figure 3-3) were
made, the ﬁnal pH did not go below 11. A longer digestion period (normally 18
hours at an elevated temperature) is required for this reaction. The powder
XRD patterns of 2:1, 3:1,and 4:1 MgAl-LDHS made by Miyata’s and Reichle’s
methods are shown in Figure 3-2 and Figure 3-3, respectively. Except for the
2:1 Mngl-CO32', LDHS synthesized by Miyata’s coprecipitation at constant pH
did not result in a well ordered product, although other groups have found this
to be a very satisfactory method. The key to the synthesis of well ordered LDHS
by Miyats’s method is careful control of the pH.

The two remaining methods are done at a pH outside of the range 8-11,
where both Mg(Ol-I)2 and Al(OH)3 precipitate. In Drezdzon’s coprecipitation at
low supersaturation, a mixed metal solution is added to a basic solution of
terephthalate (TA). The ﬁnal pH is above 12 where Al3+ is present as the soluble
AL(OH)4'. Again, the only precipitates are Mg(OH)2 and Mg3Al-TA. The
reaction is allowed to digest for 18 hr. at 74-75 °C. The MgZI/Al” ratio of the
reactants is maintained in the product.

The ﬁnal method is Taylor’s sequential precipitation of anAl-NO3'
performed at a pH = 6.2, where Zn(OH)"'+ is soluble. This method
demonstrates a method where the pH is below that where the divalent cation
precipitates, even though Taylor did nor synthesize MgAl-LDHS at a low pH.
Using the modiﬁed method by Kwon,’°’ Al(Ol-l)3 is precipitated at pH 7, and
allowed to digest for one hour. A solution of Zn(NOa)2 is added simultaneously
with 2 M NaOH, in order to maintain a pH of 6.2. Careful control of the pH is
not as essential in this method as the ﬁnal pH is below the pH where Zn(OH)2

precipitates. There is a long digestion period, one week at 95 °C, just below the

41

 

c. Mg 7A1-co 3""

L_.JL MA.

b. Mg Al-CO 2‘
3 3

a. Mg Al-CO 2'
4 3

 

 

Releative Intensity

 

 

 

 

 

 

 

8 1 6 24 32 40 48
2 theta

Figure 3-2. X—ray diffraction patterns of MgAl-CO32' LDHS made by Miyata’s
coprecipitation at constant pH. The pH is maintained at 10.0 10.2 and the
reaction is digested for 18 hours at room temperature. a. Mg/Al = 4; b.
Mg/ Al = 3; c. MgAl/ = 2;. The peak assignments are given in Table 3-2.

42

 

 

 

Relative Intensity
l

 

 

 

 

 

 

 

8 16 24 32 40 48
2 theta/ degrees

Figure 3-3. MgAl-COf’ LDHS made by Reichle’s coprecipitation at variable
pH. The ﬁnal pH is in the range of 8-11 and the reaction is digested for 18
hours at room temperature. a. Mg/Al == 4; b. Mg/Al = 3; c. MgAl/ = 2;
The Miller indices are given in Table 3-2.

43

Table 3-2. Indices (bkl ) for LDHS Synthesized by the Methods

 

 

of Miyata and Reichle
Sample d spacing
(001) (002) (003) (01 5) (01 8)

Miyata

2:1 7.57 4.57 3.78 2.58 2.53 2.28 1.94

3:1 7.79 4.52 3.86 2.59 2.32 1.97

4:1 7.96 3.93 2.61 2.32 1.99
Reichle .

2:1 7.62 6.35 3.81 3.67 2.60 2.54 2.32 1.98

3:1 7.78 3.88 2.59 2.33 1.99

4:1 7.94 4.57 3.96 2.62 2.34 2.02

 

boiling point of the LDH. The disadvantage of this method is that only a 2:1
ZnAl-NO; LDHS can be synthesized. No matter what the initial cation ratio is
in the reactants, the product ratio is 2:1. The advantage is a very well ordered,
large particle LDH. Figure 3-4 shows the transmission electron micrographs of
a Mg3Al-CO32' synthesized by Reichle’s method. The anAl-NO; in Figure 3-4
is of a much larger particle LDH. Figure 3-4 is a Single particle of the anAl-
LDH at a higher magniﬁcation. The large particle size is manifested in an effect
called ‘streaming,’ where the sheer forces of the particles ﬂowing across each
other cause a Visible opalescent effect in dilute suspensions (approx. 0.01 g/ ml)of
the LDH.

To summarize, it is clear that LDHS can be synthesized at the pH where at
least one M(OH),, will precipitate. In the middle pH range where both cations
precipitate and at higher pH where only the divalent cation precipitates the
My/M’+ ratio can be made to reﬂect the ratio of the reactants. This is achieved
by careﬁil control of the pH or by a long digestion period, respectively. When
the synthesis is done at low pH, where only the trivalent cation precipitates, only
a 2:1 Mz’i/M3+ ratio will result.

Since a well ordered 2:1 Mg2Al- LDH can be made at any of the pH ranges
in these experiments, it is likely that the 2:1 M‘°’+/M3+ ratio is the most stable
ratio. In the middle range, where both cations precipitate, the cation distribution
is Statistical. In the situations where only one cation precipitates, the driving
force would seem to be the removal of charged species from the solution. At
pH above 12, both Mg(OH)2 and LDH are possible, but only the more stable
LDH product is expected to be present at equilibrium. As long as the ratio of
Mz’i/M3+ reactants is greater than two, the Mg(OH)2 will accommodate all the

aluminum.

45

Under low pH conditions, where only the My” cation precipitates, the
incorporation of the divalent cation will be limited to the stable 2:1 cation ratio.
Finally, the minimization of possible byproducts enhances the formation of a
pure well ordered LDH.

If these conclusions are valid, it should be possible to synthesize a MgAl-
LDH at a pH below 8. The minimum pH for MgAl-LDH synthesis will
probably be above that of the ZnAl-LDH and the Mg/ Al ratio in the product
will be 2, no matter what the ratio of the reactants.

Synthesis of MgAl-LDH at a pH less than 8 would have another advantage;
minimal contamination by carbonate. Carbonate is preferentially intercalated
into LDHS due to its divalent charge and conformation that matches the LDH.
Whatever precautions are taken to eliminate carbonate from the reaction
medium, there is always some carbonate present in the product. But at pH 7 and
below, the predominate species in solution is monovalent HCO3'. Monovalent
bicarbonate should be no more favorable than nitrate for intercalation into
LDHS.

The hypothesis is that a Mngl-LDH can be synthesized at a pH
signiﬁcantly less than 8.4, that the Mg:Al ratio will equal 2 regardless of the initial

cation ratio, and that contamination by carbonate will be minimized.

Results and Discussion

The reaction conditions were as follows; Mg(NOS)2 (0.03 mol) and
Al(N03)3 (0.01 mol) were dissolved in 200 ml water and the temperature was
raised to 90-95 °C. The elevated temperature aided in the elimination of
dissolved C02. The pH of the reaction mixture was controlled by the addition

of predetermined amounts of 2 M NaOH (10-45 ml) to the mixed metal ion

46

Figure 3-4. a. Transmission Electron Micrograph of Mg3Al-COsz' made by
Reichle’s coprecipitation method. b. TEM of anAl-NO; made by Taylor’s
induced hydrolysis method. The particle size is considerably larger. c. TEM
of a Single particle of anAl-NO; at a higher magniﬁcation.

47

.x x. . dawn”...

   
 

48

 

37.8A

pH = 7.6 40ml

pH = 7.9 35ml

pH = 6.9 30ml
AM

 

 

Relative Intensity/ CPS

 

pH = 6.9 25 ml

\ A89 A pH = 6.3 20 ml
\ 3 A pH=5.3 15ml

 

 

 

 

 

I I I F I I I
10 20 30 40 50 60 70 80
2 theta

Figure 3-5. The diffraction patterns of products obtained by the reaction of
Mg(NO3)3 (0.03 mol) and Al(NO3)3 (0.01 mol) in 200 ml H20) in the presence
of various amounts of 2.0 M Na(OH). The volumes of NaOH solution added
to the reaction mixture are indicated along with the ﬁnal pH value of the
reaction mixture a reaction time of 3 days at 90-95 °C.

49

 

1633

 

447

 

 

Transmission / °/o

 

1385

 

 

 

l l I l I l

I
4000 3500 3000 2500 2000 1500 1000 500

Wavenumbers

Figure 3-6. FTIR of Mngl-NO3' synthesized at pH = 6.9. The resonance at
1633 cm‘1 is assigned to coordinated water, 1385 cm'1 to nitrate, and 447 cm'1
is Al-OH

50

solution. The reaction digested at 90-95 °C for three days. The products were
centrifuged and the pH of the decantate was recorded as the pH of the synthesis.

The XRD of the products obtained as the volume of 2 M NaOH added to
the mixed metal solution is increased by 5 ml increments are shown in Figure 3-
5. When 15 ml of base (0.03 mol) is added, the ﬁnal pH is 5.3 and the only
precipitate is pseudo boehmite. The pH increases to 6.3 when 20 ml base (0.04
mol) are added to the mixed metal solution. At this pH, a well ordered LDH is
already beginning to form, although it is accompanied by pseudo boehmite. The
d spacing is 8.9 A, consistent with Mngl-NO3'. The ﬁnal pH is 6.9 when 25 to
30 ml base is added. The product from the addition of 25 ml (0.05 mol) has
intense LDH peaks and some pseudo boehmite. The addition of 30 ml of
NaOH (0.06 mol) has the stoichiometric amount of OH' needed to form a 2:1
MgAl-LDH. (There would be excess Mg3+ in solution as the Mg:Al ratio of the
reactants is 3.) Very little pseudo boehmite or Mngl-CO32' (d spacing = 7.6 A)
is evident in the XRD. When more than 30 ml of 2 N NaOH (greater than 0.03
mol) is added, the pH begins to rise above 7.0. Even though the pH is now
nearly 8 when 35 ml (0.07 mol) base is added, the main product is a MgAl-NO3'
LDH. The Mg:Al ratio of the products obtained in the addition of 30 and 35 ml
2 N NaOH was two, as analyzed by ICP.

Surprisingly, the pH is lower when 40 ml base is added than when 35 ml is
added. The XRD peaks are broad and low in intensity and contamination by
carbonate is evident in the tailing of the (00!) peaks. By the time the pH reaches
9.7 (45 ml, 0.09 mol NaOH) the primary product is Mg2AI-cof; ascertained
from the 7.8 A d spacing. The LDH is poorly ordered because even though
carbonate is favored at this pH, the concentration is limited by the solubility of
C02. Even in a traditional synthesis done a high pH, carbonate is added in a

three times excess relative to the Al“ concentration.

51

Conclusion

These results Show that Mngl-LDHS can be synthesized at a pH
appreciably lower than the literature value of 8.0. The elevated reaction
temperature and low pH, precludes the intercalation of C0321 When the
reaction is run in the pH range of 75-80, the product is poorly crysralline,
probably due to increased competition between nitrate and contamination by
carbonate from the atmosphere. This may explain why the synthesis at pH
lower than 8 has not been attempted. In conclusion, when determining the
optimum pH for LDH synthesis, the following criteria should be considered;

1. Minimization of byproducts
2. At least one metal hydroxide should be a precipitate, bearing in mind
that only a 2:1 M2I/M’I LDHS form when M‘i’+ is the precipitate.

Synthesis of Mngl-LDHS at low pH a practical application where mild

conditions are required or when a minimal volume of caustic waste is desirable.

Experimental Methods

SWIM-Ii: 0f MgeAl-COX' 12y Mama 3 met/20am

A 1.0 molar mixed metal solution of Mg(NO3)2 and A1(NO3)3 was added
to a round bottom ﬂask with the pH maintained at 10.0 i 0.2 by the
simultaneous addition of a solution of 1.0 M NazCO3 and 2 M NaOH. The
mixture was stirred during the reaction. The reaction was stirred for 18 hour
at room temperature. The solid product was rinsed times by centrifuging and
resuspending the pellet in water until the decantate was neutral, about four

times.

52

Synt/Jeri: of M ngl-C O ,2 ' by Rcir/J/ei met/20a"?

An aqueous solution of Mg(l\lO3)2 and Al(NO3)3 (Mg + A1 = 0.04 mol)
were added rapidly and with good Stirring to a solution of NaOH and
NazCOf’. The amount of OH' should equal the sum of two times the
concentration of the magnesium and three times the concentration of the
aluminum. The carbonate should be three or more times the stoichiometric
requirement. The reaction was Stirred for 18 hour at room temperature. The
product was rinsed times by centrifuging and resuspending the pellet in water

until the decantate was neutral, about four times.
Synt/Jeri: of 1Mg2AI-NO 3' at Ion/pH.

Aluminum Nitrate Al(NO3)3 9HZO (0.01 mol) and Mg(NO3)2'6H20 (0.03
mol) were dissolved in 200 ml water and the temperature raised to 95 °C.
Thirty ml of 2 M NaOH was added all at once. The reaction was Stirred for
three days at 95 °C. The product was rinsed two times by centrifuging and

resuspending the pellet in water.

53

Chapter 4. The Synthesis of Layered Double Hydroxides with
Colloidal-Size Particles

Introduction

Layered double hydroxides (LDHS), also called anionic clays or
hydrotalcite-like compounds, are versatile materials with a wide range of
applications including catalysis, environmental remediation, and

0 However, while traditional cationic clays such as

pharmacology."‘1
montmorillionite, ﬂuorohectorite, and vermiculite have been used in ﬁlm
applications since 1983,106 a parallel development has not been seen for anionic
clays. Even though the conductivity, capacitance, and the potentiometric,
calorimetric, and colorimetric properties of LDHS have the potential to be
applied in sensor technologies,107 progress in this area as well as in ﬁre resistant
coatings and barriers has been hampered by the difﬁculty in making thin
continuous ﬁlms of LDHS.

Layered double hydroxides are easily synthesized in the lab. Methods of
preparation include coprecipitation, the salt-oxide method, and induced
hydrolysis. In all cases, the product is a precipitate or gel which is rinsed and
generally dried before use. The dried LDH particles form aggregates which
cannot be easily resuspended into a non-settling dispersion. Films made from
these materials consist of powdery aggregated LDH particles which do not

adhere well to the substrate, typically glass. However, there is an example where

ﬁlms as thin as 100 nm were made from an LDH synthesized by coprecipitation.

54

The air dried product was sonicated for 1 hour to disperse the aggregated
platelets and then centrifuged at 2,000 rpm for 1 hour to remove the largest
particles. Only the small fraction of suspended ﬁne particles which remained in
suspension was used for making ﬁlms, in effect selecting only the colloidal size
particles.37 The remainder of the LDH was discarded.

The problem was to make the colloidal size particles directly, avoiding both
the extra step of sonicating the LDH and the loss of product. The solution is to
synthesize the LDH in an organic solvent where particle interactions are less
favorable. All other conditions are analogous to Standard coprecipitation
methods.

There are a few examples of synthesis of LDHS in organic solvents. In a
1992 patent, Burba et al.86 reported on the synthesis of layered mixed metal
hydroxides from metal alkoxides in organic solvents such as diethylene glycol or
methanol. In these examples, a stoichiometric amount of water was added as
the source of hydroxide. No free base hydroxide (e.g. NaOH) was added to the
reaction mixture and the reactions were performed under nitrogen. These
LDHS can be used in applications where the presence of water is be detrimental,
such as electrorheological ﬂuids for high temperature applications and as
additives to polymers, oil-based paints, and coatings. A similar method using
metal alkoxides in ethanol through sol-gel synthesis was used to synthesize
hydrotalcite with a high speciﬁc area (150 m2/ g) compared to natural
hydrotalcite.108

Malherbe et al.87 synthesized layered double hydroxides from the nitrate
and chloride metal salts in a mixture of organic solvent and water in an N2
environment. The solvent systems were 50:50 mixtures of water with methanol,
ethanol, propanol, acetone or glycols with NaOH as the hydroxide source.

Their method resulted in microtextural modiﬁcation of the surface and porous

55

properties of the LDHS. The LDH synthesized in water and water/ ethylene
glycol had nitrogen adsorption/ desorption isotherms characteristic of plate-like
particles with slit-shaped pores. The LHDs formed in ethylene glycol/ water
system had pores with regular geometry while the glycerol/water system
produced a MgAl-CO32' with a variable geometry pore Structure. This affected
the surface area in that the LDH synthesized in water/ ethylene had a surface
area twice that of the LDH synthesized in water and there was a loss of surface

area in the sample synthesized in a glycerol / water mixture.

Results and Discussion

Magnesium aluminum layered double hydroxides have been made using
primary alcohols, acetone, or acetonitrile as the only solvent without other
alteration to the standard conditions of synthesis by coprecipitation (Figure 4-1).
The reactions in acetone and acetonitrile produced LDHS with a high volume of
undesirable Mn’i(OI-I)n byproducts. Consequently, acetonitrile and acetone are
not considered to be adequate solvents under standard coprecipitation
conditions.

The X-ray diffraction patterns of the products precipitated from methanol,
ethanol, propanol, butanol, and a control reaction in water are Shown in Figure
4-2. The intensity of the diffraction patterns in increases with the chain length of
the alcohol used as the solvent and the LDH formed in water gave the most
crystalline product as judged by x-ray diffraction. The products precipitated
from ethanol, propanol and butanol are equivalent to the LDH synthesized in
water and are authentic LDHS. The gallery anions in a traditional LDH are
solvated by water (d spacing = 7.9 A, [Mg3Al]NO3'mHZO], see Figure 42 water)

so it was possible that LDHS synthesized in alcohols would have the respective

56

 

b. Acetone

 

 

 

 

 

Relative Intensity

  

 

a. Acetonitrile

  

 

 

 

I I I I I I I
10 20 3O 40 50 60 7O 80

2 theta/ degrees

Figure 4-1. X ray diffraction pattern of products prepared in the reaction of a.
Mg(NOs)_,_, Al(NO3)3, and NaOH in acetonitrile and b. acetone. Synthesis of
LDHS in acetone and acetonitrile by standard coprecipitation methods is
unsatisfactory due to the insolubility of the Mg and Al salts in these solvents.

57

 

7.85 A

 

-1

 

 

 

 

 

U
0.)
U)
m
8 Water
\ :
B‘
7;; 8.22 A
8
l
E.“ Butanol
H -.
“5
'3: 8.28 A
.3 Propanol
°‘ I ‘ ‘“
8.12 A Ethanol
L A
9'20 A Methanol
A A

 

 

 

l l l l T I l

10 20 30 40 50 60 70 80

2-theta / degrees

Figure 4-2. Powder X-ray diffraction patterns for LDHS and LDH precursor
products precipitated from various organic solvents and water. The LDHS
synthesized in propanol and butanol were washed with methanol to remove
the remaining NaNO3 salt byproduct

58

4000 3500 3000 2500 2000 1 500 1 000 500

ﬁ I I ' I I I

 

. I T

      

 

       

Mg Al-NO '- EtOH
3 3

 
  

 

I Mg Al-NO ‘- MeOH
4 3

 
 
     
 

g3Al-NO31 MeOI—I

 

 

°/o Transmission

I MgoAl-NOSZ MeOH (a)

 

 

 
 

i (a) suspended in water

' Mg Al-NO '- Control
. 3 3

   

 

2950 cm"
C-H

 

    

 

 

 

I I I I I I '
4000 3500 3000 2500 2000 1500 1000 500

-1
wavenumber/ cm

Figure 4-3. Infrared Spectra of ngAl-NO3', synthesized in ethanol, methanol,
and water. From top to bottom: A Mg3Al-NO3' synthesized in ethanol; the
Mg4Al-NO3', Mg3Al-NO3', and Mg:Al-NO; product from precipitation in
methanol; the Mngl-NO3' precipitation product after suspension in water and
an LDH formed by traditional coprecipitation.

59

ROH in the interlayer region. The a d spacing of 8.2 A is a little higher than the
water control, but suggests that the gallery anions are water solvated. The water
can come from the metal salts used in the reaction or from the air.

The FT-IR spectra for a Mg3Al precipitation product as synthesized in
ethanol and for Mg4Al-, Mg3Al-, Mg:Al-LDHS as synthesized in methanol are
shown in Figure 4-3. The intergallery anion for the LDHS synthesized in ethanol
and water is NO3', (1385 cm"). Even though it was expected that contamination
by CO3} (1355 cm") would be eliminated due to the low solubility of air in
organic solvents, apparently maintaining an elevated temperature throughout the
synthesis is enough to eliminate most dissolved COZ. There is some water
present in all the LDHS synthesized in organic solvents as evidenced by the
absorption at 1640 cm", characteristic of coordinated water. In addition to the
coordinated water, an organic species is present in the ngAl-NO3‘ product
synthesized in methanol. The adsorptions at 2950 cm‘1 are C-H stretching
vibrations and the absorption at 1070 cm‘1 is characteristic of the C-O stretch of

an alcohol. These absorptions are absent in the LDH synthesized ethanol.

M-LDHs

Even though the powder diffraction pattern of the product precipitated
from methanol has low broad reﬂections, there are no crystalline impurity
phases present. The low broad diffraction peaks of the product precipitated
form methanol could be evidence of a polycrystalline material with poor inter
particle interactions due to the limited H-bonding available with MeOH as the

solvent. AS a test, the precipitate was centrifuged and the pellet suspended in

60

 

)1

 

002

 

I: I l
I Lm—é

\ 004 0018 a.

 

 

 

U)
0..
U
\
3‘
"v";
C l
B 003
C
H
9; _.
'2: 002 b.
.53
cu
Of. v; __ A.
091
- 003

 

 

 

 

 

I I I f I I r
10 20 3O 4O 50 6O 7O 8O

2-theta / degrees

Figure 4-4. The X-ray diffraction pattern of an oriented thin ﬁlms of Mg3Al-
NO3' a. product prepared by coprecipitation of an in methanol and
subsequent hydrolysis in water, 0.030 mole Mg(l\lO3)2 and 0.010 mol Al(NO,)3
in 100 ml methanol are dripped into a solution of 0.09 mol NaOH in 100 ml
methanol and stirred for 3 days at 55 °C. b.. LDH product prepared by
coprecipitation in water at 55 °C under conditions analogous to the synthesis
in methanol. The orientation of the LDH layers in pattern a. is evident from
the complete lack of in-plane reﬂections for the M-LDH . c. A less oriented
powder pattern of the M-LDH product. The presence of the in-plane
reﬂections verify that the M-LDH is an authentic layered double hydroxide.

61

water. After standing overnight, the resulting suspension was nearly transparent
and remained in suspension after being centrifuged at 8,000 rpm for 20 minutes.
When the suspension was allowed to dry on a glass slide, a transparent ﬁlm
formed. The X-ray diffraction pattern of the transparent ﬁlm is shown in Figure
4-4a. Figure 4-4c shows the less oriented powder X-ray diffraction pattern of
the same material shown in Figure 4-4a and is the diffraction pattern of an

authentic LDH (hereafter designated M-LDH). The intense (001) peaks and the

absence of any in-plane reﬂections (h,k at 0) in Figure 4~4a are evidence of an
extremely well oriented sample. In addition, the (004) reﬂection is present in the
oriented thin ﬁlm, whereas in the X-ray diffraction pattern of the water control,
the (004) reﬂection is buried under the broad asymmetric (015) reﬂection with
in-plane components (Figure 4—4b). Applying the Scherrer equation (1) to the
(002) reﬂections of Figure 4-4a. and Figure 4-4c. gives a particle size of 80 A in

the c direction.

L 1' AK/Bcosf) (1)

where L is the mean crystallite dimension in A, K is a constant taken to be near
unity, B is the peak width at half height in radians. The mean particle size of the
colloidal M-LDH is 66 nm, measured by light scattering.

The surface area for a small particle LDH Should be very high. The surface
area of the Mg3Al-NO3' precipitate is very high when dried under vacuum at 150
°C is 600 m2/ g. The surface area of the M-LDH or the MgrAl-NO; precipitate
dried under ambient conditions decreases to only 2-69 mz/ g, in the range of a
traditional LDH. The tendency for the LDH particles to form stacked
aggregates is seen in the reduction of the surface area for the M-LDH.

Film thickness measurements were taken from SEM micrographs of the

cross sections of both a self-supported Mg3Al-NO3' and a supported Mg:Al-Cl-

62

ﬁlm of an M-LDH. Figure 4:6a. shows the cross section of a 20 mm thick self
supporting ﬁlm with no substrate. This ﬁlm was made by drying the aqueous
suspension of a Mg:Al-Cl M-LDH on a Sheet of Teﬂon. Figure 4:6b shows the
SEM micrograph of a cross section of a 0.3 mm thick Mg3Al-NO3' ﬁlm
deposited on a glass microscope slide. The ﬁlm was made by diluting the
aqueous suspension of colloidal M-LDH by a factor of 10 and depositing two ml
of the suspension on a standard glass microscope slide. The M-LDH was slowly
dried over a 3-5 day time period. While the magniﬁcation is too low for the
individual particles to be discerned, the horizontal orientation of the plates is
evident.

Figure 4:6 compares the ﬁlm surface of the Mg3Al-NO3’ ﬁlm from Figure
4-5a to a 30 mm ﬁlm of the water control. As can be seen in the micrograph of
the water control (Figure 4:6d), Storing the LDH as a slurry eliminates the
aggregation of particles generally seen in LDH ﬁlms, but Figure 4:6c shows the
further improvement possible when a colloidal M-LDH is used in making the
ﬁlm.

The electron micrographs make it clear that pre-dying an LDH reduces the
quality of the ﬁnal LDH ﬁlm. This is true for the M-LDHS as well. If the as
synthesized or water suspended M-LDH is dried under ambient conditions, it
will not resuspend in water as a colloidal suspension.

The M-LDHS exhibit the same thermal behavior as LDHS synthesized by
traditional methods. Figure 4~7 shows the diffraction pattern of the mixed metal
solution that forms when an LDH is calcined at a temperature at or below 500
°C. When the mixed metal oxide solution is stirred in water, the LDH reforms
as seen in the upper XRD pattern in Figure 4-7. The TGA for 4:1, 3:1 and 2:1
M-LDH and the water control are shown in Figure 4-8. The M-LDHS have
TGA proﬁles with two transition stages. The ﬁrst is the loss of gallery and

63

physisorbed water and occurs between 97 and 300 °C. The second stage is the
dehydroxylation of the Mg‘2+ and Al“. The Al3+ tends to dehydroxylate at a
slightly lower temperature than the Mg3+. The two Steps for Al’+ and Mg2+ are
clearer in the water control than in the M-LDHS. The only other remarkable
difference is that the M-LDHS tend to lose Slightly more (2-50/0) gallery and
sorbed water, which is advantageous for applications as ﬁre retardants.

As discussed earlier, LDHS are very versatile materials. The MZI/M3+ ratio,
the intergallery anions, and the framework cations can all be easily varied in
LDHS. In many of the methods for synthesizing an LDH, the ratio of MZI/M’I
in the LDH product reﬂects the ratio of cations added to the reaction. In
addition to Mg3Al-NO3‘, M-LDHS with Mg:Al reaction ratios of 2:1 and 4:1
were synthesized. The XRD of these LDHS Shown in Figure 4—9 have
essentially the same d spacing (8.0 A) even though the layer spacing should
change with charge.

There are two ways the gallery anion can affect the d spacing when only
the layer charge changes. When the layer charge increases, the attraction
between the opposite charges of the layers and the anions is increased and the d
spacing decreases. This is a well studied phenomenon in MgAl-CO3 LDHS.9
The d spacing for MgAl-carbonates decreases from 8.0 A for a 4:1 MgAl-CO32'
to 7.6 A for a 2:1 MgAl-CO32' LDH. The second effect on layer spacing comes
from the orientation of the gallery anion. In Mngl-CO32' LDHS, the gallery
carbonate lies parallel to the layers. While nitrate is isostructural to carbonate, it
is monovalent and twice as many charge balancing nitrate anions as are required.
In a 2:1 anAl-NO3', the d spacing increases to 9.0 A, indicating that the nitrates
are tilted relative to the layers. The d spacing for the Mg:Al-NO; is 8 A, too
small for a 2:1 LDH nitrate.

64

Figure 4-5 Scanning electron micrographs of; 6. A 20 mm self supporting
Mg3Al-NO3’ M-LDH ﬁlm. b. A 0.3 mm Mg:Al-Cl ﬁlm supported on glass.

65

 

66

Figure 4:6. Scanning electron micrographs of c. The surface of the Mg3Al-
NO3' M-LDH ﬁlm. d. The surface of a 30 mm ﬁlm of the Mg3Al-NO; control

synthesized in water.

67

 

68

 

 

2,000 CPS

b. MgsAl-NOsi- Rehydrated Powder

 

 

 

 

M

Relative Intensity

Kg a. MgsAl-NOsl Calcined

I I I I I I l
10 20 30 40 50 60 70 80

2-Theta / degrees

 

 

 

Figure 4-7. a. X ray diffraction pattern of an M-LDH calcined at 500 °C. b.
the same material after being reconstituted in water to reform the LDH.

69

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

b. Mg3Al-NO3'

8 :2

o o
I—I ,__2 _
E H
.29 "Se

0) "5 ..
B B
o\° c,\" c
I I I I I I I I I I I _ I I I
0 100 200 300 400 500 600.700 800 100 200 300 400 500 600 700 800
Temperature/Celcrus Temperature / Celcius
C- Mgf’d’NO, d. Mg4Al-N03'

3 control in water 3

o o

._I _ >4 _

"‘ :5

"Sn .9.“

8 _ Q) I...

B B

3 _ o\° —

I I I I I I I I I I I I I (I)O
100 200 300 400 500 600 700 800 100 200 300 400 500 600 7 800
Temperature/ Celcius Temperature/ CClClUS

Figure 4-8. Therrnogravimetric analysis of ngAl-NO3' M-LDH a. M-LDH
with a 2:1 Mg:Al ratio. b. M-LDH with a 3:1 MgAl ratio. c. Mg3Al-NO3'
made by coprecipitation in water. (1. M-LDH with 4:1 Mg:Al ratio

70

 

7.98 A

 

 

 

8.
Q MgAl—LDH
g 3.08A
E” ‘2 A
8 _
.5. ; 7.97A
<1) 3
>
on I
.55
a: .
Mg3Al-LDH
3.06A
8.00A

 

 

 

 

 

 

2 theta/ degrees

Figure 4-9. X-ray diffraction patterns of M-LDHS with theoretical 4:1
(40:10), 3:1 (32:10), and 2:1 (30:10) MgAl ratios. The actual ratios are in
parentheses.

71

As discussed in the introduction, the lattice parameters of the a,b plane of
an LDH are not affected by the gallery anion. Again, the effect of varying the
layer charge on the in plane crystal parameters has been thoroughly Studied for
ngAl-CO32' LDHS. The crystal parameter a is equal to two times the d spacing
of the (110) reﬂection. As the fraction of aluminum substitution increases, the
lattice parameter 4 decreases from 3.07 A for a 4:1 to 3.06 A for a 3:1 to ﬁnally,
3.04 for a 2:1 Mngl-COf' LDH. The values of the lattice parameter a for the
M-LDHS with varying Mg/Al ratio are listed in Figure 4-9. A unit cell value of
3.08 A and 3.06 A for the ratios of 4:1 and 3:1 M-LDH respectively, indicate that
the composition of the LDH reﬂects the ratio of the reactants. The 2:1 M-LDH
unit cell dimension of 3.06 A, suggests that the product ratio is 3:1. Analysis by
ICP veriﬁed these results. The Mg/Al ratios were 4.1, 3.2, and 3.0 for the 4:1,
3:1, and 2:1 initial ratios respectively. The M-LDH products are slightly enriched
in magnesium and do not form an LDH with a ratio of 2:1 under these
conditions.

The methanol synthesis was extended to other intergallery anions with the
synthesis of MgAl-Cl' and MgAl-SOf’ M-LDHS. The MgAl-CI M-LDHS
behaved similarly to the MgAl-NO3', however attempts to produce MgAl-8042’
resulted in poor products as judged by X-ray diffraction. The reactants were
magnesium and aluminum sulfates, which are only sparingly soluble in methanol.
Therefore, the solubility of the metal salts reactants is a limiting factor for the
synthesis of LDHS in methanol. This can be circumvented by utilizing anion
exchange reactions. Figure 410 shows the X-ray diffraction pattern of a Mg3Al-
Fe(CN)63“ made by an exchange reaction with a Mngl-Cl '.

Next, M-LDHS were made from cations other than Mg and Al. The trend

here is less obvious. Magnesium, aluminum, nickel, and cobalt produce M-

72

 

b. MgaAl-Fe(CN)63'

UUULL

 

 

Relative Intensity

 

 

 

 

 

 

a. Mg4CoAI-CI
i I I ﬁ' I I I I

 

1 0 20 30 4O 50 6O 7O 80
2-theta/degrees

Figure 4-10. X ray diffraction patterns for a. M-LDH where the gallery anion
of interest was intercalated via an exchange reaction. b. M-LDH where Co’+
cation of interest is present in a fractional substitution

73

17L

°suorue ptm suopeo 19118] BUIAJBA qJIAx Haj-W 30 (DIX '11-17 athd
seeﬁap /219qi-z

08 OL 09 OS 017 09 OZ OI
j=_.=%L;I.:i.I_..V..Iu..h.g.ﬂi.x..1..L.i....I...

i "‘77) .

 

 

 

 

 

 

-D'IVCOZ)

Kitsueiul QADEIQH

 

 

 

 

 

LDHS which form colloidal suspensions in water. Chromium forms a very
poorly crystalline M-LDHS, even when suspended in water, and zinc will not
form an LDH at all. All these metals will result in a reasonable LDH when
synthesized in water. This problem can be circumvented by substituting only a
small fraction of the desired cation. The x-ray diffraction pattern of CoAl-,
MgCr-, and NiAl-M-LDHS are shown in Figure 4-11. Figure 4~10 shows a
MngoAl-Cl- with only a small fraction of Co.

The Precipitate from Methanol

The infrared and X-ray diffraction data indicate that only the precipitate
synthesized in methanol is solvated by or coordinated to the organic solvent.
Although the LDHS synthesized in the remaining alcohols (especially ethanol)
also form oriented ﬁlms when suspended in water, the best ﬁlms are made from
the LDHS formed by the precipitation of a MgAl-NO3' product from methanol,
followed by hydrolysis in water. When the product of the precipitation of
MgAl-A"’ is suspended in water, the result is an authentic LDH and this is a
reasonably general method for the synthesis of LDHS. However, the precipitate
as formed is methanol is not indisputably an. LDH. The ﬁrst two reﬂections in
the X ray diffraction pattern of the MgAl-NO3° product of coprecipitation from
methanol are ordered reﬂections characteristic of a layered material, but the d
spacing is 2.5 A greater than the LDH coprecipitated from water. The third
peak is not a third order reﬂection, although it is so broad that it does
encOmpass the third order reﬂection and the (015) reﬂections of an LDH. The
product could be an LDH with methanol solvating the gallery anions. It is also
possible that the product of precipitation from methanol is a layered magnesium

aluminum alkoxide hydroxide.

75

Analysis for HCN veriﬁes that nitrate is present in the precipitate formed in
methanol supporting the argument for an anionic layered material. In addition,
the nitrate can be substituted by other anions and a variety of metals can be
substituted for the magnesium and aluminum. Hydrotalcite must be the parent
Structure.

Only the intensity of the peaks in the FTIR of the Mg3Al-NO3’ precipitate
changes when it is dried under vacuum at 150 °C (

Figure 4-12). It would be expecred that any methanol solvating the gallery
or coordinated to the surface would be volatilized under these conditions. The
continued presence of a methanol derivative supports the argument that the
ngAl-NO3' precipitate is a MgAl-methoxy hydroxide.

The weight percent of Al, Mg, (ICP) and H, C, N was used to calculate the
molecular formula for a 3:1 MgAl-NO3‘, and Show the percent of methoxy
substitution in the precipitate as formed in methanol. The formula is
[Mg3_6Al(OH)7,2(OCH3)](COQO,3(NO3)O,4. The Mg/Al ratio for this synthesis was
higher than usual, but the results of the elemental analysis do indicate the relative

amount of methoxylation of the precipitate from methanol.

Conclusions

Layered double hydroxides can be synthesized by coprecipitation in
primary alcohols (n=2-4) without signiﬁcant contamination by carbonate. An
LDH precursor synthesized in methanol and followed by hydrolysis in water
results in a crystalline LDH which is distinctly different from that synthesized in
water under the same conditions. When the precipitate formed in methanol is

transferred to water, it becomes a transparent colloidal LDH suspension from

76

 

 

 

% Transmission

  
    

 

  

- _, ,, 1070 cm"
2950 cm 1385 cm C-j-O
CTH NO,’ \ _ :
I I I I l l I I

4000 3500 3000 2500 2000 1500 1000 500

-1
wavenumber/ cm

 

 

 

Figure 4-12. FTIR of a. Mg3Al-NO3' dried under ambient conditions. b.
MgsAl-NO3’ dried under vacuum at 150 °C.

77

which very well oriented transparent thin ﬁlms can be formed. It is important to
note that these M-LDHS were not dried prior to being made into ﬁlms. Similar
to results reported by Pinnavaiaw’ and Clearﬁeld for anion exchange reactions,
the best ﬁlms are made when the LDH is maintained as a suspension prior to
use.

M-LDHS with a variety of layer charges, intergallery anions and framework
cations have been made, demonstrating that this is a reasonably general method.
The gallery anions are limited by the solubility of the starting materials, but this
can be circumvented by anion exchange reactions. The trends for incorporating
cations into the layers have not been determined.

The lack of a good method for making thin ﬁlms of LDHS has been a
limiting factor in their utility as sensors and membranes. The development of a
facile method for the synthesis of oriented continuous thin ﬁlms may well prove
to be the keystone advance enabling the practical application of these materials

as membranes, sensors, and devices and in the area of ﬁre retardant materials.

Experimental Methods

Syntberir of [M g, ,__,.)A /_,.( OH ) Z7N O 3 in met/Jana]

NaOH (36-40 g, 0.09—0.10 mole) was dissolved in 100 mL CH3OH
solvent and heated with stirring to 55 °C. A mixed metal solution of
Mg(NO3)2'6HzO and Al(NO3)3'6HzO (0.04 M in total metal ion concentration,
with ratios of Mg:Al of 2-4) in 100 mL of CH3OH was slowly dripped into the
sodium hydroxide solution and stirred for 3 days at 55° C. The solid product
was rinsed at least 4 times by centrifuging and resuspending in MeOH, until

NaNO3 was not evident in the powder x-ray diffraction pattern and the pH of

78

the decantate was below 8 as judged by Hydrion pH paper. The product was
stored as a slurry in 150 mL MeOH.

MgAl-NO3' LDHS, MgAl-Cl‘ and MgAl-8042' LDHS were made by
substituting the metal chlorides or sulfates for the Mg and Al nitrates.

Ethanol, propanol, butanol, acetone, and acetonitrile were also used to
synthesize Mg3AL-NO3' LDHS. However at 55 °,NaOH reacts with propanol
and butanol, so better results are obtained when solid sodium hydroxide is added
to the mixed metal solution (0.02 M, 200 mL) at 55 °C.

Portions of the LDH slurry were centrifuged and the pellet suspended in
water at a concentration of approximately half the initial concentration. After
standing overnight, the cloudy suspension would generally become clear. When

centrifuged at 8,000 rpm for 20 min., the LDH remained in suspension.

79

II. CATALYSIS BY INTERCALALATED
POLYOXOMETALATES

80

Chapter 5. Layered Double Hydroxides Pillared with

Polyoxometalate Catalysts

From large volume manufacture to laboratory synthesis, oxidative
transformations of organic functional groups are basic to synthetic chemistryno
Due to economic and environmental pressures, there is an increased intereSt in
developing catalysts which can simplify the stoichiometric reagents needed in
liquid phase oxidation reactions, especially those run at ambient temperatures
and pressures. While many of the promising catalytic systems are molecular
homogeneous catalysts, there are several faCtors driving the development of
heterogeneous analogs. Because they are easily removed from the reaction
mixture and recycled, non-labile supported catalysts are environmentally sound
and economically attractive. There are many well documented homogeneous
catalysts that can be supported on a solid surface, such as alumina-supported

111

molybdate. These hybrid catalysts combine the known catalytic properties of
homogeneous catalysts with the advantages of a heterogeneous catalyst. If the
support is a layered or porous material into which the catalyst may be
intercalated, the additional beneﬁts of size and chemical selectivity may be
achieved. Size selectivity is accomplished through manipulation of the pore Size
of the material. Fine tuning of the chemical environment inside the pores
increases substrate selectivity.

Catalysis by anions intercalated into a layered material implies that the

catalysts are also functioning as pillars. In order for a substrate to reach catalyst

centers deep within the layers, intercalation of the catalysts must cause a

81

signiﬁcant layer separation and an equally signiﬁcant lateral separation of the
catalysts.

Many of the catalysts employed in the oxidation of organic substrates are
anionic. Layered double hydroxides are the only general class of inorganic host
materials into which anionic catalysts can be intercalated by ion exchange.
Anionic catalysts that function as pillars upon intercalation into LDHS and
retained their catalytic activity include metal porphyrins and organometallic
complexes. The LDH supported Co(II) pthalocyanine tetrasulphonate is twice
as active and has a greater longevity than the homogeneous catalyst.112 Another
example is the intercalation of Sterically hindered metal complex
[I\/Io(V'I)OZ(OzCC(S)Ph1)]2'.m Constraining the catalyst in an LDH eliminates

the formation of inactive Mov through the comproportionation reaction,

Mo"'o,(OCC(S)Ph,);' + Mo“'O(OCC(S)Ph2)22' + 2H+ —)
2Mo"'o (OCC(S)Ph922‘ + H30

eliminating the need for bulky organic ligands. Another important advantage is
that water can be used as a solvent in this system.

Pillaring is another example of the utility of LDHS as materials by design.
Potential applications of pillared LDHS include:

1. heterogeneous and supported homogeneous catalysis: acid, base,
redox, enzymatic, phorocatalysis

2. processing of selective nanoreactors

3. separation and membrane technology: separation by adsorption, by
optically active isomers, and by adsorption-desorption cycles,

4. membrane technology: membrane materials for ﬁltration, permeation,
ion selectivity and heating and cooling systems,

5. scavenging and controlled release: pollutant scavenging and waste
management, controlled drug or pesticide delivery for environmental,
agricultural, or pharmaceutical applications

82

6. electroactive materials - electrodes, dialectics, electrolytes for batteries,
microelectronic and microionic devices, and sensors

7. photoactive materials - pigments, photocatalysts, luminescent
materials, and optical devices

Polyoxometalates are a large class of anionic catalysts that hold great
promise for intercalation into LDHS. Layered double hydroxides pillared with
POM anions have subStantial potential for applications in industrial and

environmental catalysis.

Polyoxometalates

Polyoxometalates are inorganic metal oxygen clusters which contain
symmetrical core assemblies of MO, units (M = V, W, and Mo).
Polyoxometalates form a class of inorganic compounds that is unmatched in
terms of molecular and electronic structural versatility, reactivity, and relevance
to analytical and clinical chemistry, catalysis, biology, medicine, geochemistry,
materials chemistry, and topology.m Underlying all aspects of the chemistry of
polyoxometalates is their extreme thermal robustness and stability against
oxidation. Signiﬁcant general properties of POMS are their solubility in solvents
with polarity ranging from dimethyl sulfoxide to benzene (with the proper
cation), their extreme stability, their completely inorganic composition, and most
important, their ability to undergo reversible multiple electron reductions.

An extensive and highly regarded review of POMS was written by Pope”"
in 1984, followed by a second review in 1991.”" More recent reviews include an
entire issues of Chem. Rev. 1998, vol. 98 no. 1, journal of Molecular Catalyri:
A:Cbemical, 1998, vol. 144, C. L. Hill, ed. A review by Hill and Prosser-
McCartha115 focuses mainly on mixed addenda and transition metal substituted

polyoxometalates (I'SMP).

83

Applications

The applications of POMS are centered primarily on their redox properties,

photochemical response, ionic charge, conductivity, and ionic weights. About

80 - 85% of the patent and applied literature concentrates on the catalytic activity

of POMS. The remaining 15-20°/o can be grouped in the categories listed in
Table 5-1."6 The divisions are based on the material form of the POM or on a

particular value-adding property of the POM;“6

metal oxide like

large size (6-25 A in diameter)

discrete Size and/ or structure (conﬁned geometric factors)
anionic charge form -3 to -14

high ionic weight (1 03-1 04)

fully oxidized compounds are reducible

variable oxidation number for the addenda atoms (131/2 = 0.5 to -1.0 V
vs. SCE)

color of oxidized forms different from color of reduced forms
. photoreducible

10. Arrhenius acids (pK, < O)

11. processing advantages

8.

9

NP‘P‘PPN?

a.

b.

air and water stable

incorporate over 70 elements and from large number of
structures

acid forms very soluble in water and other oxygen carrying
solvents (ethers, alcohols, and ketones)

soluble or transferable to nonpolar solvents

hydrolyzable to form deﬁcient Structures

low toxicity

Catalysis by polyoxometalates

Polyoxometalates are multi-functional catalysts. They are very strong

Bronsted acids, approaching the superacid region, and more applicable to POMS

84

Table 5-1. Applications for Polyoxometalates, Excluding
Catalysis and Medicine

1. Pharmaceuticals
2. Clinical analysis
3. Coatings
4. Analytical Chemistry
5. Processing radioactive waste
6. Separations
7. Sorbents of gases
8. Membranes
9. Sensors
10. Dyes and pigments
11. Electrooptics
12. Electrochemistry and electrodes
13. Capacitors
14. Dopants in nonconductive polymers
15. Dopants in conductive polymers
16. Dopants in sol-gel matrices
17. Flammability control

18. Food Chemistry

85

supported on LDHS, they are efﬁcient oxidants with fast, reversible multi-
electron redox transformations under mild conditions. Their redox properties
can be varied over a wide range by changing the chemical composition.

Polyoxometalates are remarkable because they function as reversible
oxidants under mild conditions. In liquid phase oxidation by heteropolyanions,
an addenda atom must be substituted by vanadium or one or more of redox
active transition metal such as Fe“, Mn“, Co“, or Cu3+.117 Examples of
oxidative catalysis by Keggin ions include the oxidation of alkanes, phenols and
oleﬁns, C-C bond cleavage of vir- diols, the epoxidation of alkenes, allylic
acetoxylations, forming glycerol from allylic alcohols, gluteraldehyde from
cyclopentene,"8 and oxidative bromination of aromatics.119

Tungsten andmolybdenum catalysts are known to be efﬁcient catalysts for
oxidation by peroxides. Polyoxometalates with tungsten and molybdenum
addenda atoms have been used to catalyze the oxidation of various organic
substrates by hydrogen peroxide. Peroxo polyoxometalates are the active
intermediate in the peroxidation of oleﬁns, the oxidation of alcohols, glycols,

117

and phenols in homogeneous or two-phase reactions. The oxygen transfer
mechanism in alkene epoxidation by transition metal complexes is still under
debate. Sharpless has proposed a mechanism via a three center transition state
where the alkene coordinates to one of the peroxo oxygens on the peroxo
POM.120 Experimental data from mononuclear peroxo complexes support the
Sharpless mechanism,121 but there is no direct information on the peroxo
polyoxometalates.

Venturello designed the ﬁrst polyoxometalate catalyzed peroxide oxidation
in a biphasic CHCl3-HZO system with 2-15% H202 and tungstate and phosphate

ions as the catalytic precursors to {PO4[WO(OZ)Z]4} 32122 The epoxidation of 1-

86

octene with 15% H202 in the presence of a quaternary salt of {PO4[\X/O(O:,)2]4}"
resulted in 89% yield of epoxyoctane based on the initial H202-

In a Similar system with 35% H202 and tris-cetylpyridinium salts of
PMonOw" and PWIZOw’} it was determined that the POM decomposed and
the active catalysts were lower nuclearity peroxo species Similar to the Veturello
catalystsm The anions Bwuogt, Pzwmoézt, and SiWnOw‘ were inactive in
these systems and did not change structure during the reaction, while HZWO4
and HgWquS showed high activity for the epoxidation of 1-octene.124

Transition metal substituted polyoxometalates (T SMP) such as
[WM3(HZO)2(XW9034)2]12' (X, M=Zn2+, Co”, Mn”, Fe”, Fe“, Ni2+, Cu“, sz”,
Pt”, V“, etc.) are stable to hydrolysis and also to degradation to H202. These
anions are promising oxidation catalysts. The solvolytically stable disubstituted
WZnMn2(I-IZO)2(ZntO34)2’3' is a highly active catalyst for the epoxidation of
alkenes and for the oxidation of alcohols with thousands of turnovers and grater
than 99% selectivity.125 These catalySts have also been used to oxidize alkenols,
diols, and aminesm’

Mixed addenda POMS such as PWgV3 and PWszm and the TMSP anions
FePW1,O3.,“,128 and TizPWmO 4073129 catalyze the peroxo oxidation of aromatics
via an HO radical. Catalysis by PWwV2 Shows 92-98% selectivity for the
peroxide oxidation of benzene to phenol and the catalyst structure and activity
was maintained for over 576 hours. The FePWnO394' anion will also oxidize
benzene to phenol although the decomposition of the H202 is the major
reaction.

{PO4[WO(Oz)l]4}3” and WZnMn2(HZO)2(ZntO34)2’2' have been
supported on silicate xerogels modiﬁed with phenyl and quaternary ammonium
groups. The catalysts epoxidize alkenes at room temperature in the absence of

an organic solvent."0

87

POM chemistry also can also be environmentally relevant, particularly
when it is applied to non-polluting 'green technology' or to remediation methods
to clean polluted segments of the environment. An example of an application to

green technology is their use as a highly selective, multifunctional and reusable

1

catalyst in production of wood pulp into paper.” The decontamination of

water by the photocatalytic decomposition of organic pollutants, such as
aromatic and chlorinated hydrocarbons by WwO324‘, PWIZOW}, and SiWnOw‘L, is
132

an-example of their potential use for environmental remediation.

POMS can be used in the efﬁcient oxidation of H25 to sulfurm and thioethers to

A range of

sulfoxides and sulphones.""

Recent industrial applications of POMS in catalysis include the vapor phase
oxidation of methacrolien to methacrylic acid”5 (at a volume of over 80,000
tons/ year), the hydration of oleﬁns, the polymerization tetrahydrofuran”6 and
the oxidation of methane.”7

Approximately two-thirds of the applied chemistry of POMS is based on
H,1>Mo,,o,0, H,Pw,,o,.,, H,SiMo,,o,o, and H,Siw,,o,o. Their popularity is
due to the large volume of literature describing their fundamental chemistry and
their commercial availability. Given the enormous number of available POM
complexes, it is obvious that plenty of opportunity exists for expanding the
application of these materials. However, for these potentially promising
reactions to be used in practice, complete recovery and recycling of POM

catalysts is needed for catalyst design and scale up.“7

History

Berzelius‘"8 is credited for the discovery of polyoxoanions in 1826 when he
mixed ammonium molybdate and phosphoric acid to form 12-molybdophosphate.

Merignac followed with the discovery of tungstosilicic acid and its salts in 1862.139

88

The ﬁeld developed rapidly, by the end of the century, sixty anions and hundreds
of salts were known140 and by the 19305, hundreds of POMS were described in
Gmelin.141 The structures were harder to determine. Several theories on the
Structure of POMS were proposed, but the ﬁrst Structure was not solved until
1934, by Keggin.142 By 1971 only a few structures were understood, but now X-

ray diffraction and multinuclear NMR make structure determination much easier.

Composition

Polyoxometalates (POMS) are inorganic metal-oxygen clusters of quasi-
octahedral-coordinated metal oxides with the formula [MmOY]P (isopolyanions)
and [(XO4),,IVI,,,OY]“' (m>x) (heteropolyanions). Generally M = V", M0“, or WVI
in the highest oxidation State, with only a few POMS formed from Nb", Tav, and
er. The M atoms deﬁning the framework structure of a POM are called
“addenda” atoms. More than one element can be present to form mixed-addenda

POMS. The ability of an element to form a polyoxoanion is determined by ionic

radius, charge, the ability to form d1t-p1t M-O bonds, and the capacity to accept
variable coordination states. The metals must be able to ﬁt in the holes of close-
packed oxygen tetrahedra and octahedra and must be stable in the d° oxidation
state. The large number of polyoxoanions formed by V", Mo“, and WVI is due to
their ability to take on variable coordination numbers, while CrV1 is limited to
POMS with Cr"1 in four coordination, and Nbv and Tav to six coordination. In
addition to the mixed addenda POMS, the heteroatom X (if present) can be from
nearly any group of the periodic table except the noble gases.

In addition to the mixed addenda POMS, a fraction of the addenda atoms
can be substituted by transition metals such as Fey“, Mn”, Zn2+, Co”, or Cu2+.
These POMS are often designated Transition Metal Substituted Polyoxometalates
(T SMP).

89

Structure

Polyoxometalate structures are governed by two general principles:“4

1. Each metal atom occupies an MOx polyhedron (most commonly an
octahedron or square pyramid) in which the metal atoms are displaced,

as a result of M-O 1t bonding, towards those polyhedral vertices that
form the surface of the structure.
2. Structures with MO(, octahedra that contain three or more free vertices
are not generally observedm
The polyoxoanion (MOn) tetrahedral and octahedral building blocks are
connected by comer and edge (octahedra only) linkages so that each polyhedra
has one or two unshared oxygen vertices. The small highly charged addenda

atoms polarize the exterior O atoms, causing the metal atoms to be displaced

from the centers of their ocrahedra toward the external oxygen(s). Short M-O

bonds are due to oxygen 1t-donation into the vacant d-orbitals of the metal
These oxygen atoms are relatively polarized and, therefore, they have a very
weak attraction for HI. There are three types of external oxygen atoms: terminal
M=O, comer sharing M-O-M, and edge sharing M-O-M. The interior oxygens
have cations on all sides and are not polarized in any one direction.

There are two types of polyoxometalates. Type 1 POMS have one terminal
oxygen per addenda atom and can accommodate do, (1’, cl2 metals. Type 1
POMS are able to undergo multiple electron reductions. Type 2 POMS have 2
terminal oxygens per metal center. The two mutually cis terminal oxygens of
Type 2 POMS are restricted to d0 metals and they do not undergo reversible

reductions.

90

   

I3 (Ch) XIVII 4042 (sz)

 

 

 

B'XM9034 (Car)

Figure 5-1. Schematic of the Keggin.”4 The (x-XMwO40 Keggin ion is shown
with common derivatives. The left side shows the three isomers formed by the
rotation of 1,2, or 4 M3 triads. The lacunary ions on the right are formed by
the dissociation of neutral MO3 octahedra. (HZWung’). The formula for the
Keggin ion and its derivatives can be written as (XO,,),,M,,,O,m where each
Keggin ion (excepting Hlesz") has an internal XO4 tetrahedra and each
addenda atom has one terminal oxygen. The Keggin anions are considered to
be very weak bases with great softness.144

91

Most polyoxometalate structures can be derived from two parent Structures.

The a-{(XO4)(M,;,_O3G)} Keggin ion (Figure 5-1) is a common product of the self
assembly of 12 M03 units which will occur even in the absence of a central
tetrahedral XO4 unit.

Four isomers of the Keggin ion are possible. These are distinguished by a

60 ° rotation of 3 edge shared octahedra MO3 units, as illustrated on the left of
Figure 5-1. Of the four possible isomeric structures, the existence of the or and

B have been veriﬁed. The lacunary anions Shown on the right side of Figure 5-1
are formed by the removal of 1 or 3 adjacent MO3 groups. The dissociation of
the neutral MO3 unit is facilitated by the strong trans effect of the terminal M-O
bond. The resulting lacunary species can function as building blocks for multi-
anion structures or as a reactive multi-dentate ligand. Transition metal
substituted Keggin ions are formed when the lacunary ions coordinate the
transition metal into the vacant octahedra. Larger ions form when two lacunary
ions trap the transition metal to form a sandwich complex (Figure 5-3).
Including heteroatoms, mixed addenda, and TSMPS, the number of isomers for
Keggin ions is enormous.

The parent anion {(XOG)(M,2032)} Shown in Figure 5-2 has not been
observed as a discrete polyanion. All the derivatives of this structural type form
Type 2 polyoxometalates with two terminal oxygens. They are based on edge
sharing MO6 octahedra and may be considered fragments of cubic closest
packing of oxygen. The M7O24 POMS illustrated in Figure 5-2 are isomers. The
planar bent structure is generally found for isopolyoxometalates as the central
octahedra in the Anderson structure does not have any terminal M-O bonds.
Internal MO6 are not generally observed for the addenda atoms. The Anderson

structure is favored by heteropolyoxometalates.

92

 

   

miion (Du)
Capped Structure ‘4
P arent Structure

\ XMr02+(CnD
{ML

 

 

 

 

 

mton (933)
Anders on

 

Figure 5-2. Examples of Type 2 structures where each addenda atom has two
terminal oxygens.114

93

The structures {(XOIZXMDOscg}, {MGOW}, and {(XOQZMSOIS} are
examples of less common polyoxometalate structures.

Recently, there has been considerable research involving
macropolyoxometalates. These include the {(XO 029118054"- Dawson and
{X4M3OZ4(HZO)ZO112}“' doubled Dawson, {240420)2(XM9O34)2“' (Figure 5-3)
Finke and polyoxocryptates such as NaPSWmOml‘L (Z = transition metal). The
Dawson ion is formed by the fusion of two Keggin fragments formed
spontaneously from MoVI and WW phosphates and arsenates. Isomers are
formed from rotation of edge shared triads or the rotation of two half groups.
A derivative of the Keggin ion, the Dawson structure incorporates internal XO4
Td attached by bridging oxygens and has single terminal oxygen for each
addenda atom. The doubled Dawson is formed from a ring of four transition
metals in oxide octhedra sandwiched between two lacunary Dawson ions. The
Finke ion is similar to the doubled Dawson ion, with four transition metal
cations sandwiched between lacunary Keggin ions.

The polyoxocryptate, NaPSWwOHO”, (Figure 5-3) is called the PJP anion
after Preyssler,145 Jeanin, and Pope.146 It is composed of ﬁve (PO4)W(,O,8"
fragments with a rare 5-fold symmetry. The DS symmetry is lowered to CS due
to the displacement of the Na atom to the center of one of the rings of ten W

octahedra.
Sjﬂfbefi!

Conversion of aqueous metalates to polyoxoanions is an equilibrium
strongly dependent upon the ionic Strength of the metal, the nature of the
countercation, kinetics, and temperature. A thorough compilation of synthetic

methods has been done by Tytko and Glemse147 in 1990. The most common

method of POM synthesis is through acidiﬁcation of aqueous solutions of the

94

 

      
 
 
 

A [L .1
'4’ V in

VA!”

             

 

Figure 5-3. Schematic of some representative macropolyanions. a. A Dawson

ion made from two MgO31 Keggin lacunary ions. b. A Dawson TSMP

complex sandwiching four planar Co octahdra. c. A PJP anion, N aPSWwOno"
The Na cation is displaced from the center of the disk shaped anion.

95

mononuclear oxoanions and necessary heteroatoms. POMS can often be

crystallized from stoichiometric mixtures of the components.

qMO,” + nH“ -> Mpg"- + n/ZHZO
12WO42' + HPOf‘ + 23H" -) PVC/”0403* + 12HZO

Several species may be in equilibrium and the predominant species may not
be the compound isolated in crystalline form. In some cases, careful control of

temperature or pH, sequence of addition, or speciﬁc catalysts may be necessary.
I ropoﬁam'om

Of Group 5 elements, only Vv forms a signiﬁcant range of isopolyanions.
At pH > 6, tetrahedral [VZOJL is the dominant species. In the acidic region, six-
coordinate decavanadates [HXVIOOZB]6"‘ predominate. Other anions form in non-
aqueous solvents or are metastable aqueous species. Niobium and Tantalum are
limited to hexametalates [MOO,.)]6’.

Chromium from group 6 forms POMS based on tetrahedral CrO4, which
are limited to [C907]; and [Cr3O,0]2'. Six coordinate molybdenum and tungsten
form many isopolyanions. At pH = 5, paramolybdate and paratungstate
[IVI7OZ4]6' are formed. In subsequent hydrolysis Mo and W form different
structures, with tungsten ultimately converted to the Keggin Structure, although
other kinetically stable structures are known. In nonaqueous solvents it is

possible to synthesize other Structures.
H eteropoblaniom

The formation of a heteropoly complex involves the polymerization of
addenda atoms around a heteroatom through a mechanism involving the

attachment of protons to oxygens. The stereochemical requirements of the

96

heteroatom that control the architecture of the heteropolyanions. The Strong
polarization of the exterior layers of oxygens terminates further

polymerization.’48

Polyoxometalate Intercalated Layered Double Hydroxides

There are several important considerations relevant to the intercalation of
POMS into the gallery of an LDH. The pH stability of the POMS and LDHS,
the reaction temperature, the swelling properties of the LDH, and the charge
compensation between the LDH layers and the POM all affect the ﬁnal product.
The uniformity of the layer charge of the LDH effects the purity and crystallinity
of the ﬁnal product. In all examples of LDH-POM synthesis, there is an

accompanying impurity phase present.
Syntberz':

The only method for synthesizing LDHS that has not been used to
synthesize LDH-POMS is the high temperature hydrothermal method (see
Chapter 1). Table 5-2 lists the many polyoxometalates that have been
intercalated into different LDH frameworks. The ﬁrst preparation of an LDH

149

polyoxometalate intercalate was reported by Woltermann. However, the only
evidence of the successful intercalation of the POMS into LDHS was the surface
area of the calcined samples. The ﬁrst authenticated crystalline LDH-POM was
achieved through an anion exchange reaction of decavanadate (V 10028‘”) and
Keggin ions (XMnme) with Zn_?Al-NO;,’.150 Since then several strategies for
the intercalation of POMS have been advanced. In the method developed by
Drezdzon,76 large organic ions were used to separate the layers, facilitating POM
exchange. Chibwe and Jones151 used a calcined Mg:Al-CO3} to form a metal

oxide precursor which was rehydrated to from meixnerite, a Mngl LDH

97

intercalated with OH‘. This was followed by anion exchange with the desired
POM. Narita et al. later showed this to be a general method152 and in 1990,
Dirnotakis71 showed that meixnerite could be swelled with glycols to facilitate
POM anion exchange reactions. Direct synthesis of LDH-POMS153 was
followed by the intercalation of larger POMS such as Dawson and Finke ions
into glycol solvated MgAl-adipate LDHS by Sang and Pinnavaia.154 A direct
coprecipitation method by Clearﬁeld"1 used the hydrolysis of the urea to control
the gradual increase in pH and avoid local changes in pH.

A novel (bimie dome, or soft chemistry method was used by Delmas155 to
prevent ﬂuctuation in chemical composition caused by the reaction of the acidic
POMS with the basic LDH. A solid NazNi(,,,)Co,02 was formed through a high
temperature synthesis to form a solid product with a d Spacing of 5.18 A. An
oxidizing hydrolysis reaction with NaClO and KOH resulted in the insertion of
K", 7.08 A, and was followed by reduction in an HzOz/NH4VO3 solution. The
intercalate was a polymeric metavanadate (V 03),,"1

Serwicka156 intercalated PMolew3' via electrochemical reduction of POM.
The reduced Keggin ion is less acidic which aided in reducing, but not
eliminating, the decomposition which occurs upon reaction with the basic
LDH. Finally, very large macropolyanions such as the doubled-Dawson and PJP
anions were intercalated into Mg3Al-OH' and anAl-NO3' through anion
exchange reactions.’

Ulibarri157 surveyed the various routes to intercalate POMS. The main
effect of the different routes was in the pore size distribution. The use of large
organic ions as pre-swelling agents creates stable pores with a narrow size
distribution and minimizes the formation of impurity phases.

At the present time, anion exchange is the most common method for

synthesizing LDH-POMS. Several techniques can be used to facilitate the most

98

difﬁcult exchange reaction. One approach is to sonicate the LDH during the

158

exchange reaction. The reaction is aided by partial delamination of the LDH.
Kooli and jonesm found that exchange reactions done at pH 4.5 did not need to
be kept under N2. The low pH eliminated carbonate formation and its
subsequent intercalation into the LDH phase. They also identiﬁed the pH

needed for successful anion exchange of different LDH cation compositions.162

Layered Double Hydroxides Intercalated with
Isopolyoxometalates

Mob/Mater

Heptamolybdate (M07036) which is the predominant solution species
between pH 6-4, has been the major isopolymolybdate to be intercalated into
LDHS. Drezdzon76 claimed that the MgAl-M07038 was stable to 400 °C,
however this conclusion was based exclusively on morphology of the calcined
LDH-POM. Twu and Dutta'oO used Raman and XRD to study MgAl-Mo7OZ46’
intercalates and found that the MgAl—LDH layers decomposed at 120 °C, but
that the M07024“ anion was Stable to 350 °C. The XRD was not extended to the
range where the LDH in plane refleCtion would appear and the Raman spectrum
was not recorded in the range where polymeric Mo-O-Mo bonds would appear.
Hibino and Tsumashima saw a loss of the XRD pattern when MgAl-M07036“
was dried at 80 °C, but the heptamolybdate intercalate was reconstituted in
water.161 They also performed anion exchange reactions in a 50:50
MeOH/ water solvent mixture and saw a difference in the physical properties of
the exchanged product, however, the XRD patterns were not signiﬁcantly

different from the product produced by Drezdzon's in .titu hydrolysis method.76

99

Vanadate:

There have been many studies on isopolyvanadate intercalated LDHS

There are several isopolyvanadates species present in the pH range 12-4.

V,OOZG(OH)2‘L, VmOz7(OH)5‘, V1003“) at pH = 1-3
VO(OH)3, VOZ(OH), V309}, V4Olz4', at pH = 4-6
VO3(OH)2‘, HVzO7’“, VzOf, pH = 8-11
V043”, vanadate at pH > 12

The ﬁrst crystalline LDH-POM was a V1002? intercalated into ZnAl-,
NiAl-, and ZnCr-LDHS by anion exchange at pH = 4.5. The d spacing of the
product was 11.9 A.103 Battacharyyamg and Besse169 both used careful regulation
of the reaction pH to control the nuclearity of intercalated Species. On the basis
of Battacharyya’s Studies of MgAl and ManAl LDHS and the work of Besse on
CrCu LDH systems, the following relationship between pH and the vanadate
intercalate has been established:

at pH = 10-11 '9 V2071
6-7 9 V401;
4-5 -) VmOm"

Besse used a large organic anion as swelling agents to facilitate anion
exchange and found that the size of the precursor anion favored the
intercalation of a vanadate with similar Size.

Similar studies with MgCr and NiCr LDHS veriﬁed the importance of using
a hydrated LDH as a precursor, but the layer cation determined the optimal pH
for intercalation. The optimal pH for intercalation of decavanadate into a MgCr

LDH was 6.5 for and 5.5 for a NiCr LDH.162

100

Tungrtater

The isopolytungstate intercalates have not been as thoroughly studied as
the molybdates and vanadates, although they are commonly used for catalytic
activity studies. Along with heptatungstate, two isomers of dodecatungstate,
namely metatungsrate and paratungstate A,“ have been intercalated into MgAl-

and ZnAl-LDHS.

LDHS Intercalated with Heteropolyoxometalates

Kwon and Pinnavaia“ intercalated a variety of Keggin anions with varying
charges into ZnAl-LDHS. The intercalation of a Keggin ion was limited by the
charge on the ion. Keggin ions with a charge less than 6- did not intercalatate
into anAl-LDHS. This is due to the area needed by the Keggin ion, diameter =
9.8 A, area = 83 A2. The charge density of 2:1 LDH is 16.6 AZ/e. The surface

area for anAl-a-[SiV3WgO4O]7' was 155 m2/ g, compared to 26 m2/ g for the
anAl-NO; precursor. Wang et al. intercalated PWqu" into MgAl-LDHS
with a d spacing of 12 A. The low d Spacing (compared to 14.6 for Keggin ions)
was explained as being an effect of the partial dissolution of the divalent cation
in the acidic media. This results in partial removal of hydroxyl groups and leads
to vacancies where the Keggin ions could ﬁt. This would lead to lower d
spacings than expected and also explains the intercalation of a 3- Keggin ion.I73
Studies with NiAl-CO32' LDH precursor showed that exchange only occurred
when the NiAl-LDH was maintained in aqueous suspension and not allowed to
dry prior to anion exchange.’“"65

Keggin ions are treated as spherical anions, whereas Dawson and Finke
anions are cylindrical. The d Spacing for an LDH intercalated with a Keggin ion
has been shown experimentally to be 14.6-14.8 A. Studies involving 3"P and 27Al

MAS-NMR have been used to determine that Keggin ions are oriented with

101

Table 5-2. Summary of Polyoxometalate Intercalated Layered

 

 

Double Hydroxides
Polyoxometalate LDH Method pH d “C‘-
Framework spacing
A
Isopolyoxometalates
M07024“ MgsAl, Mg:Al, anion exchange 4.4- 9.9-12.2 16176
anA1 swell/ exchange amb. 166.2
reconstruction 7
“(70248 Mg2Al swell/ exchange 4.4 12.2 3
anAl
Him/120406 anAl, ngAl anion exchange 4.4- 14.6 1637‘
X: 2,3 reconstruction amb. 178.179
swell / exchange 167.2
H2W12042m' MgsAl anion exchange 4.4- 1 2.2-1 4.8 ‘63
anAl, swell / exchange amb. 7
W207} Mg:Al copreciptation 8.8 9.4 ‘68
VzO74' Mg:Al, coprecipitation 1 0-1 1 1 0. 5 "’8
Mg:anAl CuzCr 169
V401?” Mg:Al, Coprecipitation 6-8 9-10 '68
Mg2Zn3Al CuzCr anion exchange 169
VmOzs‘* Mg:Al, anAl, anion exchange 4-6.5 11.9 76
anCr, NI3A-I, swell/ exchange 161
CuCr, MgCr, reconstrucdon 166
NiCr 150,171
Hetero-
polyoxometalates
(1:5iW120404” Zny‘U anion exchange 14,6 163
a-PW12040} anAl anion exchange 14.6 '63
mﬂ-SiWnOso” MgrAl, anAl reconstrucdon/ 14.6-14.8 73
swell 6.0 ‘53
reconstruction 153
coprecipitation
PW110397' MggAl, ngGa swell/ exchange 5-6 1 4.6-14.8 17“
x=2-4
BW110399‘ MggAl, anAl swell/ exchange amb. 14.5 ‘79
PIX/9034" M831“, anAl swell/ exchange amb. 14.5 179
V2W4Ol9+ Mg:Al, anAl ion exchange <7 12 17‘
coprecipitation
HV «NI/813040“ Mg:Al ion exchange <7 14,7 171
coprecipitation

102

 

 

 

 

Polyoxometalate LDH Method pH d M
Framework spacing
A
or-SiV3W90407- ZnAl anion exchange 1 4.6-14.8 163453
reconstruction ‘79
PVW11O404' an anion exchange 14.5-14.8 ‘73
175
179
PVnW(12.n)an(3+“)' MgeAl, NixAl, anion exchange 4-7 11.4-12.2 ‘73
(n=0-4) anAl, x=2,5 17"
PV3W9040(* CugAl, Ni 5A] coprecipitation 7-11 1 1-12 (’3
BVW11O407' MgsAl, anAl swell/ exchange 6- 14.1 ‘79
amb.
PCu(I-IzO)W110395 ZnAl anion exchange 14.8 ‘63
PCr(I-le)W11O394“ ZnAl anion exchange 14.6 175
P'I'1W11O405 ZnAl anion exchange 14.6 ‘75
SiCo(HzO)W11039‘* ZnAl sonication/exchan 14.7 17"
ge
[XZ(HzO)W11O391”’ anAl anion exchange amb. 14.5 ‘73
(X=Si; 2: Co, Ni,
Cu, Al)
[111(XW11039)2]"' anﬁl anion exchange amb. 14.4 ‘73
(Ln=La, Ce'";
X=Si, P, B)
SiX(HzO)W11039“ MgAl anion exchange 15.2 ‘77
(X=Co, Mn)
a'P2W180626' Mg3Al anion exchange amb. 1 7.4-1 9.3 ‘78
reconstruction
C04(HzO)2(PWoO34 2‘“ ngAl anion exchange amb. 17.7 ‘73
reconstruction
(LEVI/1706310 Mg;Al, anAl anion exchange amb. 17.4 3
reconstruction
NbxW(crx)O19(x+z>° ngAl, anAl anion exchange 5.8- 14.7 ‘7‘
x < 2 coprecipitation 11.6
TizWioOaF‘ MgeAl, anAl anion exchange 7.6 14.7 74
x < 2 coprecipitation
SiFe(III)(SOs)W11O.197‘ MgrAl, Zn 3A] swell / exchange amb. 14.7 ‘79
PM02W90397' ngAl, Zn 3A1 swell / exchange amb. 14.5 ‘79
BCo(II)W11O397' MgrAl, Zn 2A] swell/ exchange amb. 14.3 ‘79
BCu(II)W110397' ngAl, anAl swell/ exchange amb. 14.4 ”9‘79
P4W31)Zn4(HzO)2O112"* MgrAl, anAl anion exchange amp 16.3 3
swell / exchange
NaSb3W21Osom MgrAl, anAl anion exchange amb. 17.6 3
swell / exchange
NaAS4W40014027' Mg:Al, anAl anion exchange amb. 14.7 3
swell / exchange
NaP5W30011014' MgrAl, anAl "amb. 21 -22 3

103

 

their C2 axis perpendicular to the LDH layers.172 Two orientations are possible
for Dawson and Finke transition metal substituted polyoxometalates. The
Dawson ions can be intercalated with either the long C3 axis perpendicular (19.3
A) or parallel (17.4 A) to the LDH layers. The parallel orientation, obtained at
higher temperature, is the most Stable orientation. Intercalation of Finke ions
always results in the more stable orientation with the long axis parallel to the

LDH layers.178

M acropoboxometalater

There are only a few reports of LDH-macropolyoxometalates. The ﬁrst
large polyoxometalates intercalated into LDHS by anion exchange were reported
in a patent by Pinnavaia et al.,179 one of which was the PIP anion. Evans et al.
used extended X-ray absorption ﬁne structure (EXAFS) to characterize a
Zn,3 -P]P. The EXAFS derived structural parameters did not give any
evidence for the formation of mixed salts or for the polyoxometalate ﬁtting into
defects in the LDH layers. Wang et al., intercalated PZWISOGZG' and the PJP
anion into NiSAl-LDHS for the catalytic conversion of isopropanol to propylene
and acetone. Recently, macropolyoxometalates including the Dawson ion, 0(-
PngOg1 10', two F inke ions, Zn4(HzO)2(AngO34)2m and
WZn3(HZO)2(ZnW9O},)3‘2', a double Dawson, P4W3OZn4(HzO)Om"" and the
inorganic cryptates, and NaSb9W3,O%’& have been intercalated into MgAl- and
ZnAl-LDHS.3 These LDH-POMS can be deﬁned as pillared LDHS, even

though the surface area is less than expected from theoretical calculations.

104

I' .

Table 0-3. Oxidation Reactions with Intercalated
Polyoxometalate Catalysts

 

 

Polyoxometalate LDH Oxyge Substrate Product Effect of Re f.
n Intercalatio
Source 11
V100286’ anAl O: isopropanol acetone increased (’3
I photo-
activity
PV3W90411(", NisAl isopropanol acetone, increased 7"
PW1zO4o3' propylene selectivity
for
propylene
M07024“ Mg:Al H 302 2-hexene, oxide, size 7‘3
14207120409 an1 cylohexene, diol selectivity
91/120411“ B-
methylstyrene
SiW11O398* ZnAl H202 benzaldehyde benzoic acid increased 77
SiZ(HzO)W11039‘* activity
Z = Co“, Ni”,
Cu2+
XZ(HzO)W11O39 Oz cyclohexene tuning of 7“
X = P, Si; Z = catalytic site
Mn”, Fe”, Co”,
Cu?"
SiCo(HzO)WnO39 ZnAl H 302 benzaldehyde benzoic acid increased 73
9 activity
SiW1zO42" ZnAl isobutene alkylation bifunctional 7"
with butene acid-base
catalyst
PW10039‘5‘ MgAl propylene 2-ethoxy-1- acid versus 8“
SiW1zO404' oxide + propanol, base
ethanol 1 -ethoxy-2- catalysis
propanol

105

 

Applications
Table 5-3 lists many of the polyoxometalate intercalated LDHS which have

been used in oxidation reactions. One of the ﬁrst examples of catalysis by
LDH-POMS was photocatalysis of isopropanol to acetone by anAl-HBWquG,
BVWHO407‘, and SiV3W9O407'.184 This was followed by Tatsumi's7report of size
selective oxidation of oleﬁns by MgAl-M07038 and -H2WIZO40°'.

The Keggin ions, SiWuO428“ and PW1ZO427' were used for the alkylation of
iso-butene with butene.182 The calcined LDH-POM (300 °C) had a higher
butene conversion as well as a higher C12 and C16 product selectivity. Wang et al.
used TSMP Keggin ions intercalated into a variety of LDH supports as catalysts
for the conversion of isopropanol to acetone and propene.’7’ Lopez-Salinas177
found that the coordinatively unsaturated X centers in LDH-SiX(I-IZO)WHO39(*
(X=Co, Mo) can reversibly absorb small molecules such as water methanol,
ethanol or ammonia, while larger molecules like pyridine or tetrahydrofuran
cannot access the interlayer Keggin anion. The variation acid/ base functionality
of MgAl-HszzOmG’ as determined through the dehydration/ cleavage reaction
of 2-methyl-3-butyn-2-ol as a reactive probe, was attributed to the differences in

167

porosity obtained by the various LDH-POM synthetic methods. Kayuga and
Jones correlated the catalytic activity of MgAl-SiWnOw‘L to the surface area and
basic sites in the aldol condensation of acetaldehyde. The catalytic activity is
directly related to the surface area and the selectivity is related to the number and

strength of the basic sites of the LDH support.

106

 

Chapter 6. Polyoxometalates and Their Intercalation into

Layered Double Hydroxides

Introduction

Considerableeffort has been invested in the design of layered double
hydroxides (LDHS) intercalated with robust POMS for applications in shape-
selective adsorption and catalysis. In practice however, polyoxometalate
intercalated LDHS have not yet reached their anticipated potential as
heterogeneous catalysts. Substrate size and shape selectivity implies a pillared
LDH, which would allow the substrate access to POM catalysts buried deep
within the LDH layers. The intercalation of large, highly-charged
polyoxometalate anions into LDHS fulﬁlls the requirement of a pillared material
as a result of the increase in the layer separation and the lateral separation of the
anions as calculated from the charge per unit area of the POM and the LDH.
However, the observed microporous surface areas are not as high as the
theoretical values calculated from the gallery heights and charge densities of the
intercalated POM anions. Access to the microporous galleries may be limited by
POM salt impurities deposited on the edges of the particles. These impurities
are present in all polyoxometalate intercalated layered double hydroxides
reported to date. As a result, two different approaches to the immobilization of
catalytic anions on LDH supports have emerged. One approach is to

immobilize the catalysts on the external LDH surface without intercalation. In

107

 

this case, selectivity is introduced through the basic nature of the LDH
framework and the hydrophilic/ hydrophobic properties of the gallery anion.‘85
The second approach is to intercalate an anionic catalyst in order to control the
selectivity. Several pillared LDH-POM derivatives have been Shown to be active
catalysts for the oxidation of cyclohexene with molecular oxygen by transition
metal substituted Keggin ions,181 the epoxidation of alkenes by lower nuclearity
peroxotungsten compounds,1 as well as other oxidation reactions.186 But in all
these examples, it has not been possible to separate the catalytic activity of the
intercalated anion from that of the ever present impurity phase. So far, only one
study has been reported concerning shape selectivity for an LDH-POM catalytic
reaction.

Tatsumi et al.,7 were the ﬁrst to suggest substrate Shape selectivity for
LDH catalysts interlayered by polyoxomolybdate and tungstate anions. The
active phases were identiﬁed as Mg:Al-LDHS of heptamolybdate,
[Mg,,A1,(OH),,]Mo7o._,,rnH_._o, (d spacing = 9.9 A) and paratungstate B
(dodecatungstate), [MgzOAIIO(OH)60]H2W,ZO4Z'mHzO (d spacing = 12.2 A).
Both intercalates were better catalyStS for the oxygenation of 2-hexene than the
unsupported POMS. More importantly, for the epoxidation of 2-hexene and
cyclohexene, the intercalated POMS were 4-6 times more selective in favor of
the smaller 2-hexene substrate than the homogeneous catalysts.

In light of the substrate selectivity effects reported by Tatsumi et al., the
present study was initiated with the intent of extending the selective oxidation
catalysis of LDH-POM to include acidic anAl-LDH hosts. In contrast to basic
Mngl-POMS, anAl-LDH derivatives can be formed at acidic pH values where
many POM anions are stable. Consequently, hydrolysis reactions between the
POM and LDH host are less likely to occur. However, the synthetic method

used to prepare the original catalysts affords products that are much more

108

complex that previously realized. The results to be presented here indicate that
the relevant Mg:Al-LDH phases are structurally unstable even under relatively
mild conditions. In contrast, the acidic Znﬂ-LDH phases are more stable,
particularly when the gallery anion is the relatively robust HzWqu‘“ Keggin ion.
Even in this latter case, substrate shape selectivity is precluded by the lack of
intracrystalline microporosity under condensed phase reaction conditions. Thus,
the selectivity reported by Tatsumi cannot be attributed to Shape selectivity
effects.

Next, a stable anAl-PSWMOHON' intercalate LDH-POM with a d spacing
of 21.5 A was selected for Study as a catalyst for the oxidation of cyclohexene.
The expectation was that in the case of anAl-PSWMOHO”, the partial blocking
of the pores by the impurity salt would not be sufﬁcient to restrict access of
organic substrates to the intergallery catalysts. The results of the catalytic studies
Show that while access to the intergallery P5W3OOHOM catalyst remains a
limitation, the LDH support is vital to the catalytic activity and product

selectivity.

Results and Discussion

Catalyst Synthesis and Characterization

The substrate-selective LDH-POM catalysts7 were prepared, according to
Tatsumi, by in 57714 hydrolysis of MOf’ followed by anion exchange, a method
originally developed by Drezdzon.76 A Mg:Al-Terephthalate (TA) was
synthesized by coprecipitation and added to a solution of M0422 The pH of the
solution is lowered with the addition of HNO3, the isopolyoxometalate forms,
and undergoes ion exchange reaction with the LDH-TA. As shown by the X-
ray diffraction patterns in Figure 6-1, the XRD of a thin ﬁlm sample of Mg:Al-

109

Mg Al-Mo O 6' System
2 7 24

 

 

 

    
   
 

°<:
“i
— 2
I
O ' . o
g a- Thin ﬁlm, 25 c
e"
a °< ‘
'8 01
C O o
8 b "‘ Powder, 25 C
.3 .
o
>
'3
T) Powder, 120 °C
Cﬁ
After catalysis, 70 0C

 

 

 

IIIITTTHIIIIIITIIITIUFII

10 20 30 4O 50
2-theta/ degrees

Figure 6-1. X-ray diffraction patterns of the mixed phase product formed by
in situ hydrolysis of NaZWO4 in the presence of a Mg2Al-terephthalate: a. thin
ﬁlm sample dried under ambient conditions; the sharp reﬂections are (001 )
lines of [MgnAl6(OI-I)36] (M07024).W1th a basal spacing of 12.2 A, whereas the
broad lines are due to a salt-like impurity phase with a spacing of 10.2 A. b.
powder sample, dried under ambient conditions, showing a marked reduction
in the intensity of the LDH phase; c. powder sample dried overnight at 120 °C
showing the loss of the Mngl-Mo7Oz46' phase; (1. thin ﬁlm of sample after use
as a catalyst in the oxidation of cyclohexene.

110

 

M07024“ dried at room temperature has sharp 001 reﬂections characteristic of
the 12.2 A phase of Mg2Al-Mo7Oz46', although they are of very low intensity.
The 001 reﬂection is buried under a broad impurity peak at ~10.2 A. The XRD
pattern is consistent with a mixture of an authentic MggAl-Mo7Oz46' LDH phase
and a salt-er Mgzt or Al3+-POM impurity phase. When the XRD pattern of
the reaction product is obtained for an unoriented powder sample (Figure 6-1b)
rather than a ﬁlm, the impurity phase reﬂection increases greatly relative to the
LDH phase. That is, the preferred orientations of the LDH platelets in a thin
ﬁlm sample masks the salt-like impurity. The ratio of Mg:Al:Mo by ICP analysis
is 1.7:1.0:0.82, whereas the expected ratio is 2.00:1.00:1.17. These values do not
give an accurate indication of the Mg:Aleo ratio in the LDH phase, because the
impurity also adds to the total metal content.

In the catalytic oxidation Studies reported by Tatsumi, the MngI-
M07024“ catalyst was dried at 120 °C before use. The X-ray diffraction pattern
of the Mg:Al-M07024“ LDH dried at 120 °C shows an unexpected result (Figure
6-1c). The heptamolybdate intercalated LDH component undergoes
decomposition at this temperature, as indicated by the disappearance of the 002
and 003 reﬂections, whereas the salt-like impurity phase is unaffected by heating
at 120 °C.

Raman spectroscopy was used to identify the intercalated anion and the
impurity salt obtained as the major components of the MgAl-M07024" reaction
products (Figure 6-2). Previous work187 has shown that as the pH of a M0042“
solution is reduced from an initial value of 11.0 to 4.2, the anion is progressively
converted to heptamolybdate as the only other POM species in solution. The
MoO42' anion exhibits a Raman Mo=O, stretching frequency at 898 cm" and a
Mo=O bending frequency at 320 cm". Heptamolybdate is readily distinguished
from molybdate by a Mo=Ot symmetric stretch at 947 cm", a Mo=O, bend at

111

 

Raman Spectra

 

 
 

MgoAl-Mo7Oo46' System

 
 

 

 

Relative Intensity

 

 

 

L

I I I I j
200 400 600 800 1 000 1200

 

 

 

 

Wavelength / cm"1

Figure 6-2. Raman spectra of (NH4)6Mo7O,_4 and the Mngl-Mo7Ozf product
formed by in situ hydrolysis of MoOf’ and ion exchange with a Mg2Al-
terephthalate LDH.

112

340 cm '1, and a Mo-O-Mo deformation at 220 cm ". The Raman spectrum of
the reaction product formed by in Jim MoOf' hydrolysis and exchange reaction
with Mg:Al-TA matched closely the spectrum of solid (NH4)Mo7Oz4. The
presence of heptamolybdate as the only Raman-identiﬁable POM suggests that
the relatively large fraction of the impurity phase formed in the Mg:Al-M07034“
reaction system is a magnesium / aluminum salt of M07038.

The in riﬁr hydrolysis method also was used to prepare a paratungstate B
pillared LDH, (MgaAl-HZWHO 4210) LDH intercalate, for the peroxide oxidation
of oleﬁns. As shown by the XRD patterns in Figure 6-3, however, the product
by this reaction route is a mixture of two phases with XRD properties very
similar to those found for the analogous molybdate product (cf., Figure 6-1).
The d spacing of the minor Mg:Al-LDH component is 12.1 A, and the salt-like
impurity phase gives a peak at 10.1 A. The observed Mg:Ale molar ratio of
1.8:1.0:0.74 is close to the calculated ratio of 2.0:1.0:0.83, but, again, the
elemental ratio is not an accurate indication of the LDH composition due to the
large amount of impurity present. The 12.1 A d spacing for the Mg2Al-
isopolytungstate phase is n0t consistent with the previously reported literature
value of 14.6 A for HlezO 4210 intercalated into an LDH.188 Drying the
isopolytungstate catalysr at 120 °C results in the decomposition of the LDH
component (see Figure 6-3c). This result also is incompatible with a H2W12042m
intercalated LDH phase, because such intercalates have been previously shown
to be Stable to 220 °C.188

The hydrolysis reactions of WO42’ are known to be considerably more
complex than those of MoOf‘.189 At least three species are present in solution
over the pH range 11.0 to 4.2. At pH 9, the predominant species is WO4Z‘. As
the solution is acidiﬁed, W7Oz46' is rapidly formed at pH 5 to 6. The W702,“ ion,
which is in equilibrium with HZWIZOQ’I’ is converted primarily to HngOn'C’ at

113

6-
MgZAI-W7O24 System

 

 

   
 
    
 
  

 

*3
o
o
on
5 Powder, 25 °C
'5
C
B
.5
a)
>
"3
T) Powder, 120 0C
a:
After catalysis, 73 0C

 
 

 

 

I I l l l
10 20 30 40 50

2-theta/ degrees

Figure 6-3 X-ray diffraction patterns of the mixed phase product formed by in
situ hydrolysis of NaZWO4 in the presence of Mg:Al-terephthalate. Parts a-d
are analogous to those given in Figure 6-1 for the corresponding molybdate
system.

114

 

pH 4.4. It was not possible to diStinguish between these isopolytungstate
possibilities by Raman spectroscopy, the X-ray spacing of 12.1 A (cf. Figure 6-3)
suggests that W7Ozf is the intercalated anion in the Mg:Al-tungstate formed by
in Jim hydrolysis. Apparently, the W7Oz4(* anion rapidly exchanges with the
Mg2Al-TA as soon as it is formed in the in Jim hydrolysis reaction and is
stabilized in the LDH galleries at a pH where HZWUOQ’G would normally form.
By further analogy to the molybdate system, the impurity phase was assigned as a
salt of the W7Ozf anion.

6' nor the

From the above results, it is clear that neither the Mo7O24
HZWHOQW are the active catalyst in the size selective epoxidation studies.
Instead, the initial catalysts consist of salt-ﬁke M07038 phases that are present as
an initial impurity and a decomposition product that forms as a result of the
thermal decomposition of the LDH component when dried at 120 °C. In order
to obtain stable LDH systems, authentic anAl-POMS were synthesized,
wherein the layered host is acidic and less likely to hydrolyze the intercalated
POM. Well crystallized anAl-Mo7Ozf and -H2W,ZO4O(* LDH intercalates were
prepared by anion exchange of the ammonium polyoxometalate salt with anAl-
NO3',102 using the method developed by Kwon and Pinnavaia,‘88 with only slight
modiﬁcations. The anAl-NO; suspension was formed at 95 °C and an the
anion exchange reaction was conducted at the ambient pH of the aqueous
polyoxometalate solution. The product was digested for 18 h digestion period at
80 °C. The digestion was especially helpful for promoting the Mo7Oz4(’
exchange. In general, the digestion step is desirable because it helps in
producing consistently well-ordered intercalated products as judged by XRD. A
12.1 A d spacing for anAl-Mo7Oz4" was obtained by averaging the values from
ﬁve orders of (00]) reﬂections. (Figure 6-4a). Analysis by ICP gave the ratio of

anAle0 as 2.0:1.0:1.1, very close to the expected ratio of 2.00:1.00:1.17.

115

Zn Al-Mo O 6' System
2 7 24

 

5,000 cts

Thin {magic

Powder, 25 0C

 

 

 

, A Powder,12‘0i#

 

5.01

Relative Intensity
I

 

 
   

After catalysis, 73 0C

 

 

 

10 20 30 40 50
2-theta / degrees

Figure 6-4. X-ray diffraction patterns of [Zn12A16(OH)36](Mo7OZ4)'mHZO
prepared by ion exchange of the corresponding LDH-nitrate with M070246'; a.
thin ﬁlm sample dried under ambient conditions; b. powder sample dried
under ambient conditions; c. powder sample dried overnight at 120 °C
showing the loss of the LDH phase; d. thin ﬁlm sample after use as a catalyst
in the oxidation of cyclohexene. The broad weak reﬂection at 10.8 A in
spectrum (b) is due to a salt-like impurity phase.

 

116

 

Table 6-1. Raman Bands (cm") of Isopolymolybdates

 

 

Bend
Sample d spacing (A) vs,.m(M:0,) (M=Or) Def (M-O-M)
M00421“)a 898 320 -
(NH4)(.M07Oz4 947 340 208
PM0120403‘(aq)‘ 995 251 -
anAl-MO70246‘ (25)b 12.1 950 360 220
anAl-M07Oz4“ (105) 10.6 950(919) 360(317) 222
anAl-M07Oz46‘ (220) 7.8 936 341 206
anAl-MmOy‘” (25)C, 9.6 922 333 213

rehydrated

a. Values for the aqueous POMS were taken from Payen et al.“
b. Values in parenthesis are the temperatures at which the LDHS were heated prior to the

recording of the Raman spectra.
c. The anAl-M070246- dried at 220 °C, was rehydrated by slurrying in water at 25 °C and

air drying-

117

 

The Raman spectrum of the anAl-Mo7Ozf’ LDH product showed the
presence of bands at 950, 360 and 220 cm'1 (Table 6-1), in good agreement with
the Raman spectrum of crystalline (NHQMo7OZ4. The Raman data, together
with the ICP and X-ray diffraction measurements) are consistent with a well
deﬁned intercalated product. In fact, the X-ray diffraction pattern of the
oriented thin ﬁlm of anAl-Mo7Ozf (Figure 5:4a) is a textbook example of an
LDH diffraction pattern with sharp, intense (001) peaks. However, the XRD
pattern of a less oriented powder sample shows a small impurity peak, though
relatively weak in intensity, at a d spacing of 10.8 A. (Figure 6-4b) Thus, as
emphasized earlier, thin ﬁlm samples can mask the presence of an impurity
phase by enhancing the (001) reﬂections of the LDH phase through preferred
orientation of the platy crystallites. Although the X-ray diffraction pattern, ICP
elemental analyses, and Raman spectrum for the Znﬂ-Mo7Oz46' intercalate
veriﬁed that the initial gallery anion is heptamolybdate, this intercalate also
decomposes when dried at 120 °C (Figure 5:4c).

The change in d spacing for the Mg2Al- and anAl-Mo7Ozf LDHS upon
drying at 120 °C could be due to the loss of water in the gallery or due to a
decomposition of the M07024“. A Raman study was carried out for the thermal
decomposition for the relatively pure anAl-Mo7024" LDH phase (Figure 6-4)
because the Raman features could be attributed to an authentic LDH-POM and
not a artifact of the impurity salt. The important Raman features, given in Table
6-1, are the (M=O,)ng band above 900 cm", the polymeric Mo-O-Mo
deformation band at 220 cm“. The spectrum of ZnAl-M07024“ dried at 105 °C
shows a splitting of the (M=OQVM Stretch (950, 919 cm") and the (M=O,) bend
(360, 317 cm"), respectively. The Mo-O-Mo deformation frequency does not
change signiﬁcantly in wavelength or intensity. By 220 °C, the bands have
shifted to 936, 341, and 206 cm".

118

 

Raman Spectra of Zn Al-Mo O 6' LDH
2 7 24

 

   

220 °C Rehydrated

Relative Intensity

105 °C

  

 

   

 

 

I I I I I
200 400 600 800 1000 1200

-1
Wavenumber/ cm

Figure 0-5 Raman spectra of anAl-Mo7OZf after heating to various
temperatures: a. the air sample, showing bands analogous to those in
ammonium heptamolybdate; b. After drying at 105 °C, showing the splitting
of the terminal Mo=O band at 950 cm-1; c. After heating to 220 °C, the
terminal Mo=O band at 936 cm-1 and the Mo-O-Mo polymeric band at 220
cm-1 decreased in intensity Similar to the band for Mo-O-Al in alumina
supported molybdate d. The sample dried at 220 °C and then slurried in
distilled water, showing the further shift in the terminal Mo-O band to 922

cm-1.

119

 

The polymeric Mo-O-Mo band at 206 cm'1 has broadened and decreased in
intensity. This is similar to the effect seen in alumina supported M07038 and
has been attributed to a Mo-O-Al deformation band. The sample dried at 220
°C was slurried in water to determine if rehydration would return the layer
spacing to that of the original LDH intercalate. Ordered reﬂections for a layered
material with a d spacing of 9.6 A were observed. The Raman spectrum of the
rehydrated product has the most intense band at 922 cm’1 and low intensity,
broad bands near 320 and 200 cm”. The spectrum is similar to those reported
for the Raman bands arising from the Al-O-Mo linkages in rehydrated MoO3
supported on alumina or a protonated molybdate,190 which suggests that the
decomposition product contains similar linkages. There is no evidence for the
survival or reformation of the heptamolybdate anion. Thus, the changes in
XRD properties upon heating to 220 °C are not due simply to the loss of gallery
water, but to the transformation of the LDH- POM Structure itself. In a related
Raman study of LDH heptamolybdates, Twu and Dutta160 concluded that the
LDH layers degraded at 200 °C, while the integrity of the heptamolybdate was
maintained to 400 °C. However, they did not conﬁrm this result by examining
the polymeric Mo-O-Mo 320 cm" Raman band or the in plane reﬂection of an
LDH layer at 60 ° 2-theta. There is not enough evidence to support their claim.

A Keggin ion intercalated anAl-HZWIZOwC" LDH also was prepared by
the same general ion exchange reaction between the POM salt and a anAl-NO3‘
LDH precursor. The molar Zn:Al:W ratio for the reaction product was
2.1:1.0:2.2, close to the expected 2.0:1.0:2.0 ratio, but still with some impurity
present. The d-spacing for the LDH was 14.6 A, in good agreement with the
sum of 4.8 A or an LDH layer and 9.8 A for a Keggin ion (Figure 6-6a).

However, a

120

 

Relative Intensity

 

3,000 cts

 

Thin ﬁlm, 25 0C

b. Powder, 25 0C
J W _ 4*A__.:_Ak H A

 

 

 

 

 

Powder, 120 0C
A;_A-_J\_

 

After catalysis, 73 oC

 

 

 

I I
1 O 20 30 40

2-theta / degrees

Figure 6-6 X-ray diffraction patterns of [Zn12A16(OH)36](l-12WIZO40)‘mHZO
prepared by ion exchange of the corresponding LDH-nitrate with Hlesz“:
a. thin ﬁlm sample dried under ambient conditions; b. powder sample dried
under ambient conditions; c. powder sample dried overnight at 120 °C; d.

thin ﬁlm sample after use as a catalyst in the oxidation of cyclohexene. The
broad line at 11.1 A is due to a salt-like POM impurity phase.

121

 

salt-like impurity also was present, as indicated by the broad peak at 11.1 A. For
this LDH intercalate, there is no change in the diffraction pattern when it is
dried at 120 °C (Figure 6-6b, c). The identiﬁcation of the intercalate as an
Hleszé' anion, was based on the ICP elemental analysis, the d spacing, and
the thermal stability, all of which are in agreement with previous literature
results.‘88

In the present work only the anAl-HzWquc" LDH intercalate was
Stable when dried at 120 °C. Although the LDH structure is retained, the
intragallery POM anions are not accessible for reaction with oleﬁnic substrates in
solution. Not even N2 is capable of accessing the space between the solvated
gallery anions, as judged by the absence of microporosity and low surface area
when the LDH is dried at the catalytically relevant temperature of 70 °C. The
BET surface area for anAl-HZWqu" was 10 m2/ g, compared to Mg:Al-
W702,“ with a surface area of 44 m2/ g. Both materials exhibit isotherms
characteristic of non-porous (solvated) structures. It should be noted, however,
that anAl-HzWqué' is microporous, indeed pillared, when outgassed under

vacuum at 150 °C, in accord with the results of previously reported studies?”188

Catalytic Properties Mg2Al- and anAl-POMS

As shown by the results given in Table 6-2, we have veriﬁed that the
Mg:Al-LDH catalysts precursors as prepared by the in .tz'tu hydrolysis method of
Tatsumi et al. are indeed active for the peroxide oxidation of cyclohexene to the
corresponding epoxide and diol in (BuO)3PO solution. The results for the
anAl-LDH derivatives prepared by direct ion exchange of anAl-NO; the

122

 

Table 6-2. Peroxide Oxidation of Cyclohexene Over
Polyoxometalate Catalysts

 

 

Catalyst Precursor 0/0 Conversiona
epoxide diol
Controls

(N H 4) 6Mo7O2 4 0 21

(N H 4) 6HngO 406' 12 60

Mg:Al-CO3} 0 0

LDH-Heptamolybdatesb
anAl-Mo7Oz46‘ 1.9 3.2
Mngl-Mo70346’ 0 4.0
b

Mate—S-

anAl-HZWqu“ 0 22
anAl-Hlesz", 0 26
recycled,
Mngl-W7Oz46' 17 0
Al-POM SaltC
Al-Hleszo' 0 5.8

 

21. Reaction conditions: 100 mmol of 10% H 202 in (BuO) 3PO was allowed to react with
100 mmol of cyclohexene in the presence of 1 mmol of metal (Mo or W) at 73 °C for 3
hrs. The Mngl-POIVIS were dried at 120 °C and the Zn 2Al-POMS were dried at 25 °C
prior to use as catalysts.

b. The Mg2Al-LDl-I phases were prepared by in situ hydrolysis of M00 42' or W043” in the
presence of MggAl-TA, whereas the Zn 3Al-LDH phases were prepared by ion-exchange
reaction of the POM with Zn 3Al-NO3‘ LDH. All of the heptametallate LDH
intercalates are unstable under reaction conditions and are transformed into lower
nuclearity species on the LDH surface. Only the Zn 2Al- H2W12040‘* intercalate is Stable
under reaction conditions .

C. The Al-POM salt was made by reaction of A1(NO 3)3 and (NH4)(,H2W1zO4o under
conditions analogous to those used to prepare Zn 2Al-H2W1zO406' LDH.

123

 

unsupported catalyst precursors, (NH4)6M07024 and (NH4)(,H2WIZO4O as well
Mg2Al-CO32' are also Shown in Table 6-2. No oleﬁn oxidation products were
detected in the absence of a POM catalyst or in the presence of a Mngl LDH in
which carbonate is the gallery anion.

The results given in Table 6-2 Show that the LDH intercalates are far less
reactive than the unsupported catalysts. This is consistent with the inaccessibility
of the gallery POM centers under reaction conditions. It is highly unlikely that
the activities observed for the LDH catalysts are due to leaching of POM ions
into solution. LDH intercalates have virtually no solubility in (BuO)3PO and no
anions are generated in the oxidation reaction for displacing POM ions from the
LDH surfaces.

On the basis of the X-ray powder diffraction results described above, the
Mg:Al-M07038 and Mg:Al—W702,“ LDH precursors decompose when heated
to 120 °C. The decomposition products are not substantially altered under
reaction conditions (see parts c and d in Figure 6-1 and Figure 6-3). Because the
original catalysts were dried at 120 °C prior to use for oleﬁn oxidation in
(BuO)3PO solution, it was concluded that the LDH decomposition products and
not the initial Mg:Al-POM LDH phases that were responsible for favoring 2-
hexene oxidation over cyclohexene oxidation relative to the unsupported POMS.
Drying the Mngl-LDH precursors at room temperature instead of 120 °C does
not preserve the LDH structure under catalytic reaction conditions. In fact,
when air-dried Mngl-Mo70246' was merely Stirred in cyclohexene at 75 °C for
three hours, the result was a loss of ordered reﬂections for the MggAl-Mo702f
LDH phase. The phase formed under the conditions of catalytic oxidation is
uncertain, but the X-ray powder pattern, Shown in Figure 6-4d, is consistent with

a layered structure, perhaps a M004?" LDH intercalate. However, this layered

124

 

Zn Al-H w o 6and Salts
2 2 12 4O

 

 

 

 

 

 

 

 

 

 

 

A?

g _

a: 14.7 A

T.

>

.5 ._

£3

é) M b
9.8 A

val-kg waf -,~-_:L_LL -4.__
121’ A l l l 1 l

8 16 24 32 40 48

2 theta/ Degrees

Figure 6-7 a. XRD pattern of a physical mixture of Zn-HZWquG' salt and Al-
HngOw‘” salts b. XRD pattern of a physical mixture of Zn-Mo70246' salt and
Al-Mo7Ozf salts.

125

 

phase is no more reactive for cyclohexene oxidation than the
decomposition product obtained from the corresponding Mngl-LDH precursor
(cf., Table 6-2). In the present work only the anAl-HZWqu“ LDH derivative
was stable toward both to drying at 120 °C and to oleﬁn oxidation conditions
(see Figure 6-6c and d). Although most of the intercalated POM anions in this
intercalate are inaccessible to the substrate under catalytic reac1ion conditions,
signiﬁcant activity is observed. This suggests that the accessible POM sites at
the outer LDH surfaces are more reactive than unsupported POMS. But even
this catalyst in as-synthesized form contained a salt-er impurity phase that was
observable by XRD. The salt-er impurity cannot be ignored as a potential
oleﬁn oxidation catalyst. In an effort to demonstrate the possible role this
impurity phase in the catalytic process, aluminum and zinc salts of H2W120406'
were prepared by a precipitation reaction of A1(NO3)3 or Zn(NO3)2 and
(NH4)6H2W,ZO4O" under conditions analogous to the LDH-POM synthesis. The
diffraction pattern of a physical mixture of the Al- And Zn-HZWnOw" is shown
in Figure 6-7. The precipitate of the Al-POM salts gave an x-ray diffraction
pattern Similar to the impurity phase found in anAl-POMS, while the Zn-
HngOmS salt is more crystalline. Table 6-2 Shows that the Al-HZWIZOma salt
is reactive for the oxidation of cyclohexene to the corresponding diol.
Consequently, the role of the salt-like impurity phases as contributors to the
catalytic properties cannot be discounted, particularly for the Mg:Al systems

where these impurity phases are an abundant component.

Catalytic Oxidation of Cyclohexene by anAl-PJP

Recently, large macropolyoxometalates have been intercalated into LDHS.’
A Dawson ion, a-PZWWOmm', two Finke ions, Zn,,(I-IZO)2(ASW9O34)210 and
WZn3(HzO)2(ZntO34)3’2', a double Dawson, P4W3OZn4(HzO)Om"* and the

126

 

4' and NaSb9WZIOgéw were intercalated into

inorganic cryptates, NaPSWwOHO1
MgAl- and ZnAl-LDHS. A stable LDH-POM with a high surface area,
NaPszOno” was selected to repeat the oxidation of cyclohexene experiments
in the expectation that the partial blocking of the pores by the impurity salt
would not be sufﬁcient to restrict the access of organic substrates to the
intergallery catalysts. The results of the catalytic studies Show that while access
to the intergallery NaPSWwOnO” catalyst is still a limitation, the presence of the

LDH support is vital to the selectivity and catalytic rate.

anAl-N aPSWwOmu‘
The catalytic results for LHD-Mo702f and -H2W,ZO40‘* Show that there

are two factors that need to be taken into consideration when discussing the
catalytic activity of LDH-POM catalysts, namely, the Stability of the intercalated
polyoxometalate anion and the impurity phase present in all syntheses of LDH
supported polyoxometalates. The PJP anion, NaPSWwOHOM", named after
Preyssler, Jeanin, and Pope, is a good candidate for catalytic oxidation Studies
where the stability of the intercalate is important. The remarkable robustness of
NaPSWwOm” has been utilized in other studies of catalysis by

’9' the most notable of which is the aerobic oxidation of

polyoxometalates,
hydrogen sulﬁde to sulfur.192 After 14 days under reaction conditions there was
little or no degradation of the homogeneous NaPSWmOm'+ catalyst, as judged
by “P NMR. Another aspect of the PIP anion which makes it very compatible
with LDH chemistry is its Stability to basic pH. The PJP anion has been
reported stable in a pH range of 0 to 11 at 25 °C.‘”

The PJP anion was ﬁrst intercalated into anAl-LDHS by Pinnavaia et al. in
1992 by anion exchange. The XRD pattern exhibits six orders of (001 )

reﬂections corresponding to a gallery height of 16.3 A (Figure 6-8b). This POM

127

 

anion lies with its CS axis perpendicular to the c axis of the LDH. This
orientation is stabilized by H-bonding interactions with the host layer hydroxyls.
The anAl-NaPSWwOnOH‘ was dried at 200 °C without change in the XRD
pattern.

The ﬁve equivalent phosphorous atoms in the PIP anion have a single
MAS 31P NMR resonance at -13.9 ppm versus NazHPO4 (Figure 6-9). When the
PIP ion was exchanged into a anAl-NO; LDH, the “P MAS-NMR was
unchanged, verifying that authentic NaP5W3OOnol4' has been intercalated in he
LDH layers (Figure 6-9a,b). The absence of any other peak in the NMR
spectrum of anAl-NaPSWwO,10"L indicated that any impurity salts present in
the LDH-POM catalyst must also contain the PIP anion.

However, the observed microporous surface area 65 m2/ g is lower than
one may expect from the gallery height and charge density of the PIP ion, even
when the high molecular weight of the PIP intercalate is taken into
consideration. Limited access to the microporous gallery may be caused by the
deposition of the POM salt impurity phase at the LDH edges or by structural
defects that result in a higher concentration of POM pillars near the gallery
edges.

Catalytic Oxidation by Zn,Al-NaP5W3oOm“‘
The catalytic transfer of oxygen from H202 to cyclohexene was used as a
probe of the relative reactivities of the non-supported NaPSWwOHOM' catalyst

versus the anAl-LDH supported catalyst. Cyclohexene, 10% H202 in
(BuO)3PO, and catalyst in a molar ratio of 20:30:0.002 were added to a 100 ml

128

 

 

 

Relative Intensity
II
I

 

 

 

 

 

 

I I I I I I I
10 20 30 40 50 60 70
2 theta/ degrees

 

Figure 6-8. a. Powder XRD pattern of an Aluminum-NaPSWwOHOM' salt
analogous to the impurity phase in the synthesis of anAl-PIP. The d spacing
of the ﬁrst peak is 15.5 A. b. XRD of a thin ﬁlm sample of anAl-PIP The d
spacing of the LDH iS 21.6 A and the broad reﬂection due to the impurity
phase is 15.1 A. c. XRD of a thin ﬁlm sample of anAl-PIP with a slightly
greater volume of impurity phase. The d spacing of the LDH is 22.0 A and
that of the impurity phase is 15.1 A. d. XRD of a thin ﬁlm sample of anAl-
PIP with a the greatest volume of impurity phase. The d spacing of the LDH

is 21.7 A and 14.8 A for the impurity phase.

129

 

 
   

anAl-NaP 5W O

30 110
after reaction

Zn Al-NaP W O
2 5 30 110

b. - I I I I I‘ before reaction

 

 

 

 

W (NI-IQMNaPSWMOH”

j I I T T I I I I [ TI I I I I I I I T I I I I I

100 50 0 -50 -100 -150

 

 

 

Figure 6-9. a. MAS-NMR of (NH4)1 4NaP5W3OOno, with a single peak at -13.9
ppm. The two small peaks are side bands. b. anAl-NaPSWwOm”, the MAS-
NMR of the intercalated anion is the same as a., verifying that only PIP anion
is present. c. MAS-NMR of the supported catalyst after use in the oxidation of

cyclohexene. The peak at -2.89 ppm is (BuO)3PO and the small peaks are side
bands.

130

 

ﬂask and Stirred at 70 °C for 4 hours. Although anAl-NO; is unreactive as a
catalyst, the corresponding PIP anion intercalates are active epoxidation catalysts
(See Table 6-3). In general, the anAl-PIP LDH intercalates are selective for
cyclohexene oxide while the non-supported (NH4)1 4NaP5W3OOHOH’ catalyst
produces exclusively cyclohexanediol. The basis for this product selectivity is
probably due to the less acidic nature of the LDH. The thermal stability of the
PIP intercalate has already been demonstrated by the XRD and FTIR.2 The 3‘P
MAS-NMR (Figure 6-9c) and the powder X-ray diffraction pattern of anAL-
PIP after use as a catalyst veriﬁes its stability under the reaction conditions.

The turnover numbers reported in Table 6-3 are calculated based on the
moles of POM present, as determined by ICP analysis, and include any
contribution from the salt-like impurity. This impurity cannot be discounted as a
catalyst component. Three samples of the ZnZAl-PIP intercalate made by the
same method, but differing in the amount of impurity present as judged by
XRD, were compared for catalytic activity. On comparing the XRD patterns in
Figure 6-8 to the turnover numbers in Table 6-3, one ﬁnds that the catalytic
activity directly correlates to the amount of impurity present. Sample anAl-
LDH-B with the least amount of impurity phase shows only a trace of
cyclohexene oxide, while anAl-LDH-C exhibits 107 turnovers and anAl-
LDH-D with the greatest amount of impurity gave 182 turnovers in 4 hours.
But attributing the catalytic activity to the PIP-salt impurity may not be the
complete answer. Figure 6-8a shows the diffraction pattern of an AIM-PIP salt
made from Al(NO3)3 and (N H 4)I 4NaPSWwO110 under conditions similar to those
used to carry out the anAl-NO; exchange reaction. The XRD powder pattern
is similar to that of the impurity phase in anAl-PIP intercalate. As can be seen
in Table 3, the A13+ salt of PIP is basically inactive under the conditions of the

131

 

Table 6-3. Peroxide Oxidation of Cyclohexene in the Presence
of Different Forms of NaP5W3oOuol4‘

 

 

 

Catalyst Major Product Turnovers
(N H 4) I Salt Diol 150
anAl-LDH-b Oxide <10
anAI-LDH-c Oxide 107
anAl-LDH-d Oxide 182
Al“ Salt Oxide <10

 

a. Cyclohexene (2 ml) and 10% by wt. H 202 (10 ml) were allowed to react with 0.200 gram

catalyst at 70 °C in (BuO) 3PO for 4 h.
b. The anAl-LDH supported catalysts correspond to the samples with XRD patterns

shown in Figure 6-8.

132

oxygen transfer reaction. A trace of cyclohexene oxide is observed as a reaction
product, indicating the same selectivity as anAl-PIP catalyst. As discussed
previously, the impurity salt is believed to be dispersed on the surface of the
LDH particles and this may result in a better access to the catalytically active
anion centers of the Salt, while at the time limiting access to the intergallery
anions of the intercalate. Even though the NaPSWmOm” intercalate has the
greatest surface area among the macro polyoxocryptates, access to the gallery

region by substrates larger than N2 is still restricted.

Conclusions

The results of the present study clarify the nature of the catalyst precursors
used in the earlier work of Tatsumi et al. in terms of mixed phases of salt-like
POM impurities and Mg:Al-LDH intercalates interlayered by M07038 and
W703“. Moreover, the Mg:Al-POM LDH components of the as-synthesized
catalysts decompose at the temperatures used to dry the intercalates prior to
catalytic reaction. Despite the thermal instability of Mngl-POM intercalates, the
substrate selectivity effects observed by Tatsumi for the decomposition products
are nonetheless impressive. For instance, the preference of the Mg:Al-POM
catalyst precursor systems for the epoxidation of 2-hexene was enhanced 4 to 6
fold relative to the unsupported catalysts. The origin of the enhanced substrate
selectivity cannot be attributed to molecular sieving based on substrate size or
Shape as originally suggested because the intercalated structure of the LDH is
lost after drying and use as a catalyst. Moreover, even if the LDH-POM
structure was retained, shape selectivity effects based on sieving would not be
expected because of limited access to the solvated gallery anions under reaction

conditions. Factors other than pillaring and substrate size selectivity, such as

133

 

selective adsorption, may be responsible for the previously observed preference
of LDH - POM catalysts toward smaller oleﬁns.

The largest gallery height (~16.6 A) is observed for the NaPSWmOno14~
intercalate. Because the gallery height is more than three times as large as the
layer thickness (4.8 A ), this derivative may be regarded as being a supergallery
intercalate and is the most likely to have accessible intergallery anions. Catalytic
oxidation of cyclohexene with hydrogen peroxide as the oxygen source and the
PIP anion as the catalyst resulted in a difference in cyclohexanediol and
cyclohexene oxide product selectivity between the non-supported catalyst and
the anAl-NaPSWwOHOM' LDH. When the LDH supported PIP catalysts
containing varying amounts of impurity phase were compared for catalytic
activity, the turnover numbers increased with the presence of the impurity phase.
One explanation for this effect, supported by TEM results, is that the impurity is
a Mn+ salt of the polyoxometalate dispersed on the surface and edges of the
LDH particles. Dispersion of the salt phase on the LDH external surfaces may
increase access to the catalytically active impurity salt, while decreasing access to

the intercalated NaP5W3O0110" anions.
Experimental Methods

P/yriral M earurementr

Raman Spectra in the region 1200-200 cm'1 were recorded on a BIO-RAD
FT Raman spectrometer equipped with a Spectra-Physics Topaz T10-106c 1.064
nm YAG laser and a Ge detector. The samples were ground into a ﬁne powder
and loaded into glass sampling tubes.

X-ray diffraction patterns were obtained with a Rigaku x-ray

diffractometer equipped with DMAXB software, a curved graphite

134

 

monochromator and Cu KOI radiation. Samples were either thin ﬁlms on glass
slides formed by evaporation of an LDH slurry or dried samples ground into a
ﬁne powder.

Proton NMR spectra of the cyclohexene oxidation products were
recorded on a VXR 300 or VXR 500$ spectrometer. The delay time between
pulses was long relative to the proton relaxation time, so that the integral
intensities of the cyclohexene, epoxide and diol resonance’s were quantitative.
31P MAS-NMR was used to verify retention of the PIP anion Structure
following catalysis.

Elemental analyses were performed by ICP emission at the Toxicology
Laboratory at MSU. Solutions for analysis were prepared by dissolving 15 mg of
solid in 100 mL of 15% HNO3.

Mngl-Mo70246’ and [VIM/41710:," LDH:

A Mg:Al-terephthalate LDH was prepared according to literature
methods"76 and Stored as an suspension in degassed deionized water. The
terephthalate-intercalated LDH was converted to the desired POM intercalate
according to the procedure used by Tatsumi et al.7 N22M004 (25.4 g, 0.105
mole) or NaZWO4 (27.7 g, 0.84 mole) was added to 200 mL of the Mngl-TA
Slurry and then the pH was lowered to 4.4 with 2 M NaOH. After 10-15
minutes of additional Stirring, the product was washed 4 times by centrifuging
and resuspending the pellet in deionized water. A thin ﬁlm sample was made,
and the remainder was dried at 120 °C. MgaAl-Mo7Oz46' - XRD (A): 10.2,
impurity; 6.08,(002); 4.03, (003); 3.03, (004); 2.44, (005); and 1.99 A, (006). ICP
(wt. %): Mg = 11.0, A1 = 7.20 and Mo = 25.3. Mngl-szf - XRD (A): 12.2,
(001); 10.1, impurity; 6.06, (002); 4.04, (003); 3.03, (004); 2.44 (005) and 1.99
(006). ICP (wt %): Mg 2 10.8, A] = 6.73, and W = 32.9.

135

 

S Jill/185i: of Zn2Al-N03' LDH

Taylor's induced hydrolysis method,102 as modiﬁed by Kwon et al.,188 was
used to prepare a Stable, highly ordered nitrate - intercalated anAl layered
double hydroxide. Even though the reaction is not very sensitive to carbonate
contamination through exposure to CO;Z in the air, N2 was bubbled into the
reaction mixture, but no other steps were taken to prevent contact with air.

The pH of a 200 mL solution of 0.1 M A1(NO3)3'9HZO in a 1-liter three
neck round bOttom ﬂask was adjusted to 7.0 with 2 M NaOH. The white slurry
was stirred for 1 hour, and then 200 mL of 0.3 M Zn(NOQZ'6HZO was added
dropwise. As the Zn” solution is added, the pH was allowed to fall to 6.4 and
then maintained at that level with the simultaneous addition of 2 M NaOH.
Upon complete addition of the zinc nitrate solution, the temperature was raised
to 95 °C and digested with gentle Stirring for one week. The product was rinsed
four times by centrifuging. After the ﬁnal rinse, the pellet was resuspended in
500 mL water.

In order to maintain a supply of anAl-NO3' for subsequent anion
exchange experiments, the above slurry was purged with N2 and stored in a
stoppered 2-liter single neck round bottom ﬂask. There was no degradation of
the LDH over a several month period. XRD (A): 8.93, (001); 4.46, (002); 2.97,
(003); 2.82; 2.61;, and 2.48.

ZnZAl-Mo70246' and anAl-H2 ”7,20406' LDH:

One hundred mL of anAl-NO3’ LDH Slurry was heated to 95 °C and
dripped into 100 mL of a 1.2X10-2 M solution of (NH4)6Mo7Oz4'4HzO or
(N HJGHZWIZOw'4HzO at room temperature and pH = 5.1 and 4.3, respectively.

The pH of the solution was too low for carbonate contamination to be a

136

signiﬁcant problem, so there was no attempt to prevent exposure to air. The
solution was allowed to stir for 90 min. and rinsed three times by centrifugation
and resuspended in 100 mL deionized water. After the ﬁnal rinse, the pellet was
resuspended in 100 mL deionized water and digested for 18 h at 80 °C. A thin
ﬁlm sample was made for X-ray diffraction analysis, and the remainder was dried
at room temperature. Powder samples were made from the dried product.
zn,AI-H,w,,o,o‘* - XRD (A): 14.7, (001); 10.8, impurity; 7.31, (002); 4.86, (003);
3.65, (004); 2.91, (005); 2.43, (006); 2.03 (007) and 1.85 (008). ICP (wt. %): Zn =
16.5, A1 = 3.08 and w = 43.4. anAl-Mo7024" - XRD (A): 12.1, (001); 6.02,
(002); 4.01, (003); 2.60; 2.44; 2.41, (005); and 2.00, (006). ICP (wt. %): Zn =
26.7, A1 = 5.43 and Mo = 21.8.

Zﬂ2A/‘ZNaP5W300170j’4- I‘DH

A hot (95 °C) 100-mL portion of the LDH nitrate suspension (2.6 meq) at
pH 6.0 was added dropwise into 200 mL of 0.02 M POM salt solution (4.08
meq, 22 times the LDH AEC). After a reaction time of 2 h, the product was
washed several times by suspending in water and centrifuging, and then dried in
air at room temperature. anAl-NaPSWwOHO” - XRD (A) 21.7 (001); 14.8,
impurity salt; 10.7, (002); 7.1, (003); 5.34 (004); 4.47, (005); 3.55, (006). ICP (wt
%) Zn = 15%, Al = 2.7%, and W = 47.9%.

Oxidation of gololoexeue a} supported and uou-ruppoﬁed mobba’ate: and tungstate:

Oxidation reactions of cyclohexene were carried out according to
previously described methods7. The catalyst (1 mmol of W or Mo) and 100
mmol of cyclohexene were stirred at 73 i 2° in an oil bath. One hundred mmol
of 10% H202 in (BuO)3PO was slowly dripped into the mixture of substrate and

catalyst and the reaction was stirred at 73 °C for a total of three hours. The

137

products were identiﬁed by Total Correlation Spectroscopy (T OCSY) NMR, ,
also known as homonuclear Harman-Hahn (HOHAHA) NMR, using purchased

standards as reference samples. Yields were calculated from NMR.
C drab/tic Oxidation of Cyrlobexono a)! the rzgopoﬁed and non-.mppon‘ed P] P anion.

Hydrogen peroxide in (BuO)3PO (10% by wt., 10 ml) was added to 0.200 g
of anAl-NaPszOnO” (1.87x10'5 mol of PIP anion) or an equivalent amount
of the control (a potassium or ammonium salt of NaPSWwOno”) in a 100 ml
three neck ﬂask. Cyclohexene (0.02 mole) was added, and the reaction placed in
an oil bath at 70 °C for 4 hr. A sample was taken, and the yield of cyclohexene
oxide and diol products determined by 1H NMR. The non-supported catalyst
was recovered in D30 and ”P NMR used to determine that the catalyst was still
intact. The supported catalyst was ﬁltered and rinsed with acetone. 31P MAS-

NMR was used to verify that the catalysts were still intact.

138

9.

. Sels B. F., de Vos D. E.,Iacobs P. A., Tet. Lett, 1996, 37(47), 8557.

Gardner E. A. and Pinnavaia T. I., Appl Catal. A: General, 1998, 7 67, 65.
Gardner E. A., Yun S. , and Pinnavaia T. I., Appl. Clay Sci, in press
Wolterrnann, US Pat. 4454244 1984.

Yamada, T., Petmteeb., 1990, 13 627.

Aoshima,, A., Tonomurs, S., and Yamamatsu, S., Pobrrners Ado. Tee/7., 1991,
2, 127.

(a) Tatsumi, T., K. Yamamoto, H. Tajirna, and H. Tominga, Chem. Lett.,
1992, 815; b. Tatsumi, T., H Tajirna, K. Yamamoto and H. Tominga, New
Frontiers in Catabrsis, Guzi et al. (ed), Elselvier Science Publ. B. V. 1993.

Moore D. M. and Reynolds R. C. Ir., X-Ray Dﬂ'aerion and Anablsis of Clay
Minerals 2end ed, Oxford University Press, N.Y., 1998, 311.

Cavani, F., Triﬁro, F., and Vaccari, A., Catabrsis Today, 1991, 77, 173.

10. De Roy A., Forano C., El Malki K., and Besse I.-P., in Occelli M. L. And

Robson H. E. (eds.) Expanded Clays and Other Mieroporous Solids, 1992, 108.

11. Newman S. and Jones W., New]C/1ern., 1998, 22(2), 105.

12. Manasse, E., Atti Soc. Toseana .S' e. Nat. Pme. Verb., 1915, 24, 92.

13.Aminoff, G. and B. Bromé, Kungl Sven. Vetensk Handl. 9, 1930, 3 (no. 5), 23.

14. Frondel, C., Amer. Min, 1941, 26, 295.

15. Feitknecht C., Helu. C/Jirn. Aeta, 1942, 25, 131.

16. Allrnann, Aeta Cyst, 1968, 324, 972.

17.Taylor, H. F. W., Miner: Mag, 1969, 37, 338.

18. Zelinski, N. and Kommarewsky, W., C/Jern. Ber, 1924, 5 7, 667.

139

 

19. Brocker, F. I. and L. Kainer, German Patent 2,024, 282 1970, BASF AG,
and UK Patent 1,342,020 1971, BASF AG.

20. Miyata, S., Kagaku Gijutsushi MOL, 1977, 75 ( 7 0), 32.
21. Ueno, M. and Kubota, US Patent 4,666,919 1987.

22. Kyowa Chem. Ind. Co., jpn. Kokai Tolekyo Kobo IP 60 06,619 1985; CA 1985,
102, 226,039f.

23. French Patent 2,091,785 1971, to BASF AG.
24. Gardner E. A. and Pinnavaia T. I., Appl Catal. A: General, 1998, 167, 65.

25.Kang M. I., Rhee S. W., Moon H., Neck V., and Fanghanel T., Radioebirn.
Aeta, 1996, 75, 169.

 

26.1keda T., Amoh H., and Yasunaga T., ]. Arn. Cbern. Joe, 1984, 106, 5772.
27. Reichle, W. T., Cbernteeb., 1986, 58.

28. Miyata, S., and Kuroda, M., US Patent 4,299,759 1981, to Toyo Soda Man.
Ltd.

29.Miyata, S. And M. Kuroda M., US Patent 4,299,759 1981 to Kyowa Chem.
Ind. Ltd.

30. Miyata, S., Eur. Patent 142,773 1985 to Kyowa Chem. Ind. Co.
31. Miyata S., Kagaku Gijutsushi MOL, 1977, 75 (10), 10.

32.Marcon Elictronics Co., jpn. Kokai Tokkyo Kobo IP 60 106,122 1985; CA.
1986, 104, 13785.

33. Suzuki S., Araga T., Tsuji R., Fukushima Y., Banno K., and Hiruta 0., Eur.
Patent 150,741 1984, to Toyota Central R&D Lab.

34. Toyota Central R&D Lab., jpn. Kokai Tokkyo Ko/Jo IP 59 166,568 1984; CA.
1985, 102, 80,421.

 

140

35. Ulrngren, P., Nordic Pup and Paper Researc/J ]., 1987, 7, 4.

36. Dutta, P. K., and Puri, M. ]. P/Jys. Client, 1989, 93, 376.

37. Itaya, K., Chang, H.. C., and Uchida, 1., Inorg Cbern., 1987, 26, 624.

38. Zhu K., Liu c., Ye X., and Wu Y., Applied Catal. A: General, 1998, 768, 365.

39. Putyera K., Bandosz T. I., Iagiello I., Schwarz A., Applied Clay Science, 1995,
70, 177.

40. Mousty C. Therias S. Forano C. and Besse I. P., ] Electroanal. Clam, 1994,
347, 435.

41.Tagaya H., Sato S., Kuwahara I., Masa K., and Chiba K., ]. Mater. Clara,
1994, 4, 1907.

42. Dutta P. K., and Robins D. S., Iangrnuir, 1994, 70, 1851.

43.Tseng W. Y., Lin]. T., Mou C. Y., Cheng S. F., Liu S. B., Chu P. P., and Liu
H. W.,]. Am. Cbern. 50a, 1996, 778, 4411.

44. Robbins D. S. and Dutta P. K., Langmuir, 1996, 72, 402.
45.Allmann, R.. and Lohse, H. H., N. fab. Miner. Mix, 1966, 6, 161.

46. Ingram, L. and Taylor, H. F. W., Miner. Mag, 1967, 7 2, 544.

47.Lal, M., and Howe, A. T., ]. Cbern. Soc. Chem. Commun, 1980, 75, 737.
48. Bookin A. S. and Drits V. A., Clays C19! Miner, 1993, 47, 551, 558.

49. Drits, V. A., Sokolova, T. N ., Sokolova, G. V., and Cherkashim, V. I., Clays
and Clay Miner, 1987, 35, 401.

50. Miyata, S., Clays Clay Mineral, 1983, 3 7, 305.

51. Miyata, Gigs and Clay Minerals, 1977, 25, 14.

141

 

52. Kooli F., Crespo I., Barriga C., Ulibarri M. A., and Rives V., ]. Mater. Chem,
1996, 6(7), 1199.

53.Allmann R., N. jhh. Miner. Mh., 1977, 3, 136.
54. Miyata S. and Kumura T., Chem. Lett., 1979, 23, 843.
55. Miyata, 8., Clay Clays Miner., 1980, 28, 50.

56. Pausch, L., Lohse, H. H.., Schiirmann, K., and Allrnann, K., Clays Clay
Miner., 1986, 34, 507.

57. Hansen H. C. B. and Koch C. B.,Appl Clay S ci., 1995, 70, 5.

58. Boehm H. P., Steinle,I., an dVieweger, C. Angew. Chem. Int. Ed. Engl, 1977,
76, 265.

59. Yun S. K. and Pinnavaia Y. I, Inorg Chem, 1996, 35, 6853.
60. Bish D. L., Bull Mateiz, 1980, 703, 170.

61.Karrado, K. A., Kostapapas, A., and Suib, S. L., Solid State Ionics, 1988, 26,
77.

62. Giannelis, E. P., Nocera, D. G., and Pinnavaia, T. I., Inorg Chem, 1987, 26,
203.

63. Tagaya H., Kkuwahara T., Sato S., Kadokawa I., Karasu M., and Chiba K.,
]. Mater. Chem, 1993, 3, 317.

64. Miyata S., Clays and Clcy Minerals, 1975, 23, 377.

65.Clearﬁeld A., Kieke M., Kwan I., Colon I. L., and Wnag R. C., ]. Inclusion
Phenom. Mol. Recognit. Chem, 1991, 77, 361.

66. Miyata S., and Kurnara T., Chem. Let, 1973, 843.

67.Boehm H.-P., Steinle I., and Vieweger C., Angezu.Chem Int. Ed. Engl, 1977,
7 6, 265.

142

 

68.WangI. D., Serrette G., Tian Y., Clearﬁeld A., Appl. C197 S ci., 1995, 70, 103.
69. Zhao H., and Vance G. F., ]. Chem. S oc., Dalton Trans, 1997, 1961.
70. Chibwe K., and Iones W., ]. Chem. S oc., Chem. Commun, 1989, 926.
71. Dirnotakis E. D., and Pinnavaia T. j., Inog. Chem, 1990, 29, 2393.

72. Tagaya H., Sato S., Morioka H., Kadokawa I., Karasu M., and Chiba K.,
Chem Mater, 1993, 5, 1431.

73.Ukrainczyk L., Chibwe M., Pinnavaia T. I., and Boyd S. A., ]. Phys. Chem,
1994, 98, 926.

74. Carlino S., and Hudson M. I., ]. Mater: Chem, 1994, 4, 99.

75. Carlino 8., Hudson M. I., Husain S. W., and Knowles I. A., Solid State Ionics,
1996, 84, 117.

76. Drezdzon M. A., Inorg Chem, 1988, 27, 4628.

77.Boehm H.-P., Steinle I., and Vieweger C., Angew. Chem. Int. Ed. Engl, 1977,
7 6, 265.

78. Kanezaki E., Kinugawa K., and Ishikawa T., Chem. Phys. Lett., 1994, 226,
325.

79. Kanezaki E., Sugiyarna S., and Ishikawa Y., ]. Mater. Chem, 1995, 5, 1969.
80. Franklin K. R., Lee E., and Numm C. C., ]. Mater. Chem, 1995, 5, 565.

81.Wang I. D., Serrette G., Tian Y., and Clearﬁeld A., Appl Clay S ci., 1995 70,
103.

82. Vichi M. F., and Alver O. L., ]. Mater. Chem, 1997, 7, 1631.

83.Van der Pol A., Mojet B. L., van de Ven E., and de Boer E., ]. Phys. Chem,
1994, 98, 4050.

84. Hudson, D. R. and Bussell, M., Miner. Mag, 1981, 44, 345.

143

 

85. Courtney, PH., and Marcilly, C. H., in Poncelet, G., Grange, P., and Jacobs,
P. A. (Eds) Preparation of Catalysts III, Elsevier, Amsterdam, 1983, 485.

86. Burba, Iohn L. III; Steve A. Sims, and Thomas M. Knobel, United States
Patent 5,084,209,Ian. 28, 1992.

87.Malherbe, F.; Forano C.; and Besse,I. P.; Micropor: Materz, 1997, 10( 7 -3), 67-
84.

88. Roy, D. M., Roy, R., and Soburn, E. F., Amer. ]. of Science, 1988, 27, 77.
89. Yun, S., thesis, Michigan State University, U. S., 1994, 151.

90. Bish, D. L., and Brindley, G. W., Amer. Miner., 1977, 62, 458.
91.Clearﬁeld A., Chem. Rev., 1988, 88, 125.

92. Miyata, S., Clays Clay Miner., 1983, 37(4), 304.

93. Yamoaka, T., Abe, M., and Tsuji, M., Mat. Res. Bull, 1989, 24, 1183.

94. Meyn, M., Beneke, K., and Lagaly, G. Inorg Chem, 1990, 29, 5201.

95. Carlino S. and Hudson M. I., ]. Mater. Chem, 1995, 5, 1433.

96. Reichle, W. T., yang, s. Y., and Everhardt, d. s., ]. Catal, 1986, 707, 352.
97. Chibwe K and Iones W., Chem. Of Mater, 1989, 7, 489.

98. Albiston L., Franklin K. R, Lee E., and Smeulders I. B. A. F., ]. Mater:
Chem, 1996, 871.

99.Nakamoto Kazuo, Infrared and Raman Spectra of Inorganic and
Coordination Compouns, third ed.,Iohn and Wiley and Sons, NY, 1978.

100. Ross, G. I. and Kodama, H.,Amer. Min, 1967, 52, 1037.

101. Busetto, C., Del Piero, G., Manara, G., Triﬁro, F., and Vaccari, A., ].
Catal, 1984, 85, 260.

144

 

102. Taylor R. M., Clay Minerals, 1984, 79, 591.

103. Kwon T., thesis, Michigan State University, USA, 1988.

104. Miyata 8., Clays and Clay Miner., 1975, 23, 369.

105. Reichle W. T. Solid States Ionics, 1986, 22, 135.

106. Ghosh, P. K. and Bard, A.I. ]. Am. Chem. S oc., 1983, 705, 5691.

107. Guth, U.; Brosda, S.; and Schomburg, I. Appl. Clay Science, 1996, 77,
229-236.

108. Lopez, T.; Bosch, P.; Ramos, E.; Gomez, R; Novaro, O.; Acosta, D.;
and Figueras, F. Langmuir, 1996, 72, 189-192.

109. Pinnavaia, T. I., Chibwe, M., Constantino, V. R. L., and Yun, 8.,
Applied Clg Science, 1995, 10, 117-129.

110. Sheldon, R. A., I. K. Kochi, Metal Catab'zed Oxidations of Organic
Compounds, Academic Press New York,N. Y., 1981.

111. Haber I., The Role of Mobhdenum in Catalysis: 1981, Climax Molybdenum
Co., Ann Arbor MO.

112. Chibwe, M., and Pinnavaia T. I., ]. Chem Soc. Chem. Commun, 1993, 278.

113. 1. Corma, A., V. Fornis, F. Rey, A. Cervilla, and E. Llopis, ]. of Catal,
1995, 752, 237. b. Cervilla, A., A.Corma, V. Fornes, E. Llopis, R. Palanca,
F. Rey, and A. Ribera, ]. Am. Chem. S oc., 1994, 776, no.4, 1595.

114. Pope M. T. and Muller, A. Angezu. Chem. Int. Ed. Engl, 1991, 30, 34.

115. Hill C. L. and Prosser-McCartha C. M., Coord. Chem. Rea, 1995, 743,
407.

116. Katsoulis D. E., Chem. Rea, 1998, 98, 359.

117. Kozhevnikov I. V., Chem. Rev., 1998, 98, 171.

145

 

118. Recent reviews of polyoxometalate chemistry include: a. Hill, C.L., (ed)
]ournal of Molec. Catal A, 1996 774; b. I. V. Kozhevnikov, Catal Rem-Sci.
Eng, 1995, 37(2) 311; c. Hill, C. L. and Prosser-McCartha C. M., Cord.
Chem. Rem, 1995 743, 407.

119. Neumann R. and Assael 1.,]. Chem. S oc., Chem Commun, 1988, 1286.

120. Sharpless K. B., Townsend]. M., and Williams D. R., ]. Am. Chem. S oc.,
1972, 94, 295.

121. Iorgensen K A., Wheeler R. A., and Hofﬁnan R., ] Am. Chem. Soc.,
1987, 709, 3240.

122. Venturello C., D’Aloisio R., Bart I. GI, and Ricci M., ]. Mol Catal,
1985, 32, 823.

123. Ishii Y. and Ogawa M., in Reviews on Heteroatom Chemisty, Ohno A. and
Furukawa N., Eds., MY: Tokyo, 1990, Vol 3, 121.

124. Aubry C., Chottard G., Plantzer N., Bregault I. M., Thouvenot, R.,
Chauveau f., Huet C., and Ledon H., Inorg Chem. 1991, 30, 4409.

125. Neumann R. and Iuwiler D., Tetrahedron, 1996, 52, 8781.

126. Neumann R., Khenkin A. M., Iuwiler D., Miller H., and Gara M., ].
Mol. Ctal A, 1997, 7 77, 169.

127. Nomiya K., Yanagibayashi H., Nozaki C., Kondoh K., Hiramatsu E.,
and Shirnizu Y., ]. Mol Catal. A, 1996, 774, 181.

128. Kuznetsova L. I., Detusheva L. G., Fedotov M. A., and Likholobov V.
A., ]. Mol. Catal A, 1996, 777, 81.

129. Yamase T., Ichikawa E., Asai Y., and Kanai S., ]. Mol. Catal A, 1996,
774, 237.

130. Nuemann R. and Miller H., ]. Chem. S oc., Chem Commun, 1995, 2277.

146

 

131. Weinstock I. A., Atalla R. H., Reiner R. S., Moen M. A., Hammil K. E.,
Houtrnan C. I., Hill C. L., Harrup M. K., ]. Molec. Catal. A, 1997, 776, 59.

132. Mylonas,A., Hisdia A, and Papaconstantinou E., ]. Molec. Catal A,.
1996, 774, 191. b. Haggin,I., Chem. and Eng News, , 1996, 2530.

133. Harrup M. K. and Hill C. L., j. of Molec. Cato/A: Chemical, 1996, 706, 57.
134. Gal] R. D., Faraj M., and Hill C. L., Inorg Chem. 1994, 33, 5015.

135. Serwicka, E. M., Black]. B., GoodenoughI. B., ]. Catal, 1987, 706, 27.
136. Misono M. and Nojiri N., Appl. Catal, 1990, 64, 1.

137. Ahmed S. and MoffatI. B., Appl Catal, 1988, 40, 101.

138. Berzelius I., Pogg. Ann, 1826, 6, 369,.

139. Marignac C., C. R. Acad. Sci, 1862, 55, 888; b.Ann. Chem, 1862, 25,
362.

140. Rosenheirn A.,Iaenicke H., Z. Anorg Chem, 1917, 700, 304.

141. Gmelin Handhuch der Anorganischen Chemie, Verlag Chemie, Berlin: System-
No. 53 Wohhda'n) 1935; System-No. 54 (Wolfram) 1933.

142. Keggin]. F., Nature, 1933, 73 7, 908.

143. Lipscomb W. N., Inorg Chem, 1965, 4, 132.

144. Izurni Y., Matsuo K., Urabe K., ]. Molec. Catal, 1983, 78, 299.
145. Preyssler, C., Bull Soc. Chim. Fr, 1970, 7, 30.

146. Pope M. T., Alizadeh M. H., Harmalker S. P., Ieanin Y., and Martin-
FrereI., ]. Am. Chem. S oc., 1985, 7 07, 2662.

147. Tytko K.-H. and Glemser O., Inorg Chem, 1990, 27, 71.

148. Baker, L. C. W. and Glick. Chem. Rea, 1998, 98, 3.

147

 

149. Wolterrnann, G. M., US. Pat. 4454244, 1984.

150. Kwon T., Tsigdnios G. A. and Pinnavaia T. I., ]. Amer: Chem. Soc.,
1988, 770, 3653.

151. a. Chibwe K. and Iones W., Chem. Mater, 1998, 7, 489. b. Chibwe K.
and Iones W., ]. Chem. Soc. Chem. Commun, 1989, 926.

152. Narita E., Kaviratna P. and Pinnavaia T. I., Chem. Lett., 1991, 805.
153. Dimotakis E. D. and Pinnavaia T. I., Inorg Chem, 29, 1990, 2393.
154. Sang K. Y. and. Pinnavaia T.I Inorg Chem, 1996, 35, 6853.

155. Han K. S., Guerlou-Demourgues L., and Delmas C., Solid State Ionics,
1996, 84, 227.

156. Serwicka E., Nowak P., Bahranowski K., Iones W., and Kooli F., ].
Mater Chem, 1997, 7, 1937.

157. a. Ulibarri M. A., Labajos F. M., Rives V., Kagunya W., Iones I., and
Turjillano R., Mol Cyst. qu. Cyst, 1994, 244, 167; b. Inog. Chem, 1994, 33,
2592.

158. Kooli F., Iones W., Rives V., and Ulibarri M. A., ] Mater Sci. Lett.,
1997, 76, 27.

159. Kooli F. and Iones W., Inorg Chem, 1995, 34, 6237.

160. Tqu. and Dutta P. K., Chem. Mater, 1992, 4, 398.

161. Hibino T. and Tsumashima A., Chem. Mater, 1997, 9, 2082.

162. Kooli F. and Iones W., Inorg Chem, 1995, 34, 5122.

163. Kwon T. and Pinnavaia T. I., Chem Mater, 1989, 4, 381 and thesis.

164. Kooli F. and Iones W., Inorg Chem, 1995, 34, 5114.

148

 

165. Kooli F., Rives V., and Iones W., Mater Sci. Forum, 1994, 752-753, 375.
166. Chibwe K. and Iones W., Chem. Mater, 1989, 7, 489.

167. Yun, s. K., Constantino V. R L., and Pinnavaia T. I., Microporous Mater,
1995, 4, 21.

168. Battacharyya A., Hall D. B., and Barnes T. I., Appl. Clay Sci, 1995, 70,
57.

169. Depége C., Bigey L., Forano C., de Roy A., and Besse}. P., ]. Solid State
Chemistpt, 1996, 726, 314.

170. Weir M. R., MooreI., and Kydd R. A., Chem. Mater, 1997, 9, 1686.

171. Evans I., Pillinger M., and Zhang I., ]. Chem. S or, Dalton Trans., 1996,
27.

172. Hu C.-W., He Q.-L., Zhang Y.-H., Liu Y-Y., Zhang Y.-F., Tang T.-D.,
ZhangI.-Y., and Wang E.-B., ]. Chem. S or, Chem. Comm, 1996, 727.

173. WangI., Tian Y., Wang R.-C., and Clearﬁeld, A., Chem. Mater, 1992, 4,
1276.

174. Guo I., Sun T., Shen I. P., Iiang d. 2., and Min E. Z., Chem. ]. Chinese
Unir, 1995, 7 6, 512, in Chinese.

175. Liu Y. Y., Hu C. W., Wang Z. P., Zhang I. Y., and Wang E. B., Sci.
China Series B-Chemistgy, 1996, 39, 86, in Chinese.

176. Hu C. W., Zhang X., He Q. L., Wang E. B., Wang S. W., and Guo Q.
L., Transition Metal Chem, 1997, 22., 197.

177. Salinas, L. S., Castillo S. P., Ono Y., Mater Res. Soc. S you). Proc., 1995,
377, 163.

178. Yun S. K. and Pinnavaia T. I., Inorg Chem, 1996, 35, 6853.

179. Pinnavaia T. I., Kwon T., Dirnotakis E. D., and Amarasedera I., US
Patent 5,079,203 1992.

149

 

180. Hu C-W., He Q-L, Zhang Y.-H., Liu Y.-Y., Zhang Y.-F., Tang T.-D.,
ZhangI.-Y., and Wang E.-B., Chem. Commun, 1996, 121.

181. Guo I.,Iiao Q. Z., Shen I. P.,Iiang D. Z., Yang G. H., and Min E. Z.,
CatalLett, 1996, 40, 43.

182. Zheng X., Yue W., Heming H., and Dazhen I., in Hattori, Misono, and
Ono (eds.), Acid-Base Catalysis II, Elsevier, Amsterdam, Stud. Surface Sci.
Catal, 1994, 90, 279.

183. Kayunga W., Hassan Z., and Iones W., Inorg Chem, 1996, 35, 5970.

184. Kwon T. and Pinnavaia, ]. Molec. Catal, 1992, 74, 23.

185. Sels, b. F., De Vos D. E., and Jacobs, P A., 3rd World Conference on
Oxidatiue Catahsis, Grasselli, jR K., Oyama S. T., and Lyons I. E., (Eds.)
1997, 1051.

186. Guo,I.,Iiao, Q. Z., Shen,I. P.,Iiang, D. Z., Yang, G. H. and Min, E.
Z.,. Catal Let, 1996, 40; 43.

187. Payen E., GrimboltI., and Kastelan S., ]. Phys. Chem, 1987, 97, 6642.

188. a. Kwon T. and Pinnaviai T. I., Chem. Mater, 1989, 7, 381. b. Kwon T.,
Thesis, MSU, 1988.

189. Wang I., Tian Y., Wang R., and Clearﬁeld A., Chem. Mater, 1992, 4,
1276.

190. Iannibello A., Marengo S., Tittarelli P., Morelli G., and Zecchina A., ].
Chem. Soc. Farraday Trans. I, 1984, 80, 2209.

191. Fox M. A., Cardona R., and Gaillard P. I., Inorg Chem, 1987, 26, 6347.
192. Harrup M. K. and Hill C. L., ]. Mol Catal, 1994, 5448.

193. Alizadah M. H., Harrnalker S. Pl, Ieanin Y., Martin-Frere I. and Pope
M. T., ]. Amer Chem. S or, 707, 2662.

150