Polymerization of aniline on copper phosphonate materials

Article abstract of DOI:10.1016/S0277-5387(00)00307-7

We report the use of layered copper phosphonate materials for the oxidative polymerization of aniline. The use of copper carboxyethyl-phosphonate results in the formation of the perinigraniline form of polyaniline in the interlayer region of the host structure, giving rise to an increase in the interlayer spacing by 10.2 ?. For copper phenylphosphonate, the emeraldine form of polyaniline is generated, although there is no evidence for intercalation within the interlayer region of the host material. Copper methylphosphonate is destroyed during the process of polymerization of aniline, yielding bulk copper metal as the only solid product. The results demonstrate the diversity of behaviour that may result from the use of structurally related, although chemically different, host materials for the polymerization of aniline. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0277-5387(00)00307-7

www.elsevier.nl/locate/poly
Polyhedron 19 (2000) 765–770
Polymerization of aniline on copper phosphonate materials
Gary B. Hix a,b, Victoria C. Maddocks b, Kenneth D.M. Harris b,
*
a School of Chemistry, University of St Andrews, St Andrews, Fife KY16 9ST, UK
b School of Chemistry, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK
Received 19 April 1999; accepted 13 January 2000
Abstract
We report the use of layered copper phosphonate materials for the oxidative polymerization of aniline. The use of copper carboxyethyl-
phosphonate results in the formation of the perinigraniline form of polyaniline in the interlayer region of the host structure, giving rise to an
˚
increase in the interlayer spacing by 10.2 A. For copper phenylphosphonate, the emeraldine form of polyaniline is generated, although there
is no evidence for intercalation within the interlayer region of the host material. Copper methylphosphonate is destroyed during the process
of polymerization of aniline, yielding bulk copper metal as the only solid product. The results demonstrate the diversity of behaviour that may
result from the use of structurally related, although chemically different, host materials for the polymerization of aniline.
Science Ltd All rights reserved.
q2000 Elsevier
Keywords: Aniline polymerization; Copper phosphonates; Intercalation compounds
1. Introduction
indicate that the polymercan be formedeitherintheinterlayer
region or on the external surfaces of the material [9,10]. In
general, in these preparations, the PANI is produced in the
emeraldine base form (with xs0.5). The presence of Cu(II)
is thought to enhance the oxidation of aniline by ambient O2,
as a consequence of the accessible Cu(I) oxidation state [9].
In a number of these systems, however, complete reduction
to produce bulk copper metal has been observed.
This paper reports the use of layered copper phosphonate
materialsto bringaboutthe polycondensationofaniline.Cop-
per phosphonates, Cu(O3PR)PH2O, are layered materials in
which layers of five-coordinate copper atoms are linked via
–PO3 pyramids. The phosphorus atom is tetrahedrally coor-
dinated, and the fourth vertex is occupied by an organic func-
tional group R [22]. As a variety of groups R can be used,
there is considerable scope for fine-tuning both the structural
and chemical properties of these materials. In particular,
chemical reactions in which the interlayer region is accessed
by the reactant molecules may be influenced significantly by
the nature of the host material. For copper phosphonates, the
presence of a redox active metal leads to the possibility of
electron transfer between the intercalated guest species and
the host material, with potentially important implications
concerning electronic conductivity. The copper phosphonate
materials chosen for this study were copper methylphos-
phonate (RsCH3), copper phenylphosphonate (RsC6H5)
and copper carboxyethylphosphonate (RsCH2CH2CO2H).
Polyaniline (PANI) has been much studied as a conduct-
ing polymer [1–6], and may be described by the general
formula
{[–(C6H4)–NH–(C6H4)–NH–]1yx[–(C6H4)–
N_(C6H4)_N–]x}n, where the value of x (0FxF1) defines
the oxidation state of the polymer [7]. The forms with x/0
have the capacity to be protonated by non-oxidizing acids,
with protonation occurring on the imine (–N_) nitrogen
atoms [8]. The formation of polyaniline chains separated
(insulated) from each other by microporous dielectric host
materials has received considerable attention in recent years
[9–13]. As a result of such confinement of the polymer
chains, the polymer may display quantum-size effects in its
electrical and optical properties [14]. As a consequence,
formation and confinement of a given polymer in different
host materials may allow the physical properties to be con-
trolled and optimized for specific applications of the material.
A number of layered host materials have been reported
[9,10,15–21] to catalyse the polymerization of aniline,
including montmorillonite, vanadium pentoxide, iron oxy-
chloride, copper-exchanged a-zirconium phosphate, zirco-
nium phosphonates and copper-containing layered double
hydroxide materials. Previous reports [9,10,19–21] of the
formation of PANIusing layeredcopper-containingmaterials
* Corresponding author. Fax: q44-121-414-7473; e-mail: k.d.m.harris@
bham.ac.uk
0277-5387/00/$ - see front matter q2000 Elsevier Science Ltd All rights reserved.
PII S0277-5387(00)00307-7
Thursday Apr 13 09:51 AM
StyleTag -- Journal: POLY (Polyhedron) Article: 3384

766
G.B. Hix et al. / Polyhedron 19 (2000) 765–770
2. Experimental
mogravimetric analysis (TGA) was carried out on a TA
Instruments SDT 2960 simultaneous DTA-TGA thermogra-
vimetric analyser. Samples were heated (at a rate of 108C
miny1) in an alumina crucible to a maximum temperature of
12008C in an atmosphere of flowing oxygen (100 ml miny1).
Recalcined alumina was used as the reference material. For
the materials containing PANI (CuPhP–PANI and CuCEP–
PANI, see Section 3.1), the copper content was measured
using a Philips PU9400X atomic absorption spectrometer.
The solutions for analysis were prepared by digesting the
samples in concentrated HNO3 (Analar). The solutions were
gently warmed to aid dissolution; after cooling, the solutions
were made up to the required volume with distilled water.
Standards of 5.0 ppm, 2.5 ppm and 0.5 ppm were prepared
by dilution of a commercial copper solution standard.
2.1. Synthetic aspects
Copper carboxyethylphosphonate (abbreviated CuCEP)
was synthesized following the method of Drumel et al. [23].
Copper nitrate hydrate and 2-carboxyethylphosphonic acid
(molar ratio 1:1) were placed in a solution of 5% deionized
water in acetone, which was then kept at 608C for 5 days. The
solid was recovered by centrifugation, washedwithdeionized
water and dried in air at 608C.
Copper phenylphosphonate (abbreviated CuPhP) was
synthesized using the method of Hayashi and Hudson [24].
Basic copper nitrate was prepared by adding copper oxide
(8.0 g, 0.1 mol) to an aqueous solution of copper nitrate(100
cm3, 2.4 mol dmy3) with vigorous stirring at ambient tem-
perature. After 16 h, the dark green solid was filtered off,
washed with deionized water and ethanol, and then dried in
air at 608C. This solid (1 g) wasthen suspendedinanaqueous
3. Results and discussion
solution of phenylphosphonic acid (160 cm3, 0.1 mol dmy3
)
3.1. General observations
and the pH adjusted to 6 by addition of NaOH solution (1
mol dmy3). The suspension was maintained at 658C and
stirred for 16 h. Copper phenylphosphonate was recovered
by filtration, washed with deionized water, and then dried in
air at 608C.
Copper methylphosphonate (abbreviated CuMeP) was
prepared following asyntheticmethodthatwehavepublished
previously [25]. Copper hydroxide and methylphosphonic
acid were mixed in a 1:1 molar ratio. This mixture was placed
in a sealed glass tube and heated at 1008C for 48 h. After
cooling, the product was washed with deionized water and
dried in air at 608C.
Refluxing the copper phosphonate materials in aniline
resulted in the formation of viscous black solutions in which
solid products were suspended. This colour change is indic-
ative of polymerization of aniline in each of the reactions
carried out. In each case, the solid products were separated
by filtration, and we focus here on characterization of these
materials. The nature of the recovered solids was different
for each of the copper phosphonates considered. For theprep-
aration involving CuMeP, the only crystallinesolidrecovered
(as identified from powder X-ray diffraction) was bulk cop-
per metal. Any other solid residues are presumably either
amorphous (and not directly evident from powder X-ray
diffraction) or have particle sizes sufficiently small that they
are not collected in the filtration procedure. The solid
(denoted CuCEP–PANI) recovered from the preparation
involving CuCEP was violet in colour. The solid (denoted
CuPhP–PANI) recovered from the preparation involving
CuPhP was dark green in colour.
2.2. Polymerization of aniline
A quantity (typically 0.5 g) of the copper phosphonate
materials was placed in a flask with 50 cm3 of pure aniline,
and the mixture was refluxed overnight (at 1848C). The
resulting solutions were black in colour. The solid residues
were recovered from these solutions by filtration, and washed
with acetone prior to drying in air at ambient temperature.
The black colour of the solutions is probably due to colloidal
copper (perhaps with some contribution from any oligomers
of PANI present in the solution), although here we focus on
the solid products recovered by filtration rather than the
remaining liquid (colloidal) phase.
The colour of the solid products gives some indication of
the oxidation state of the PANI, and the reported [26]colours
of different forms of PANI are given together with their
classical names in Table 1. The most frequentlyreportedform
of PANI resulting from reactions of this type [9,10] is the
emeraldine base, which is dark blue in colour and becomes
green on protonation.
The dark green colour of CuPhP–PANI suggests that the
protonated emeraldine form of polyaniline has been formed.
Indeed, this assignment is supported by the fact that, on sub-
sequently suspending the material in a dilute solution of
NaOH, a colour change from green to blue is observed (cor-
responding to deprotonation).
The CuCEP–PANI sample is particularly interesting, as
the violet colour suggests that the PANI is produced in the
fully oxidized perinigraniline form (note that the colour of
this form does not change on protonation). Further reaction
2.3. Characterization
Powder X-ray diffraction patterns were recorded using Cu
Ka1 radiation on a Siemens D5000 diffractometer, operating
in transmission mode with a primary beam Ge monochro-
mator. Fourier transform IR (FTIR) spectra were recorded
on a Perkin Elmer Paragon 1000 FTIR spectrometer at a
spectral resolution of 2 cmy1; the samples were prepared as
KBr pellets (containing approximately 1% sample). Ther-
Thursday Apr 13 09:51 AM
StyleTag -- Journal: POLY (Polyhedron) Article: 3384

G.B. Hix et al. / Polyhedron 19 (2000) 765–770
767
Table 1
Colours and nomenclature for different forms of PANI; in each case the colour of the protonated form is given in brackets
Colour
Name
{–(C6H4)–NH–(C6H4)–NH–}n
{[–(C6H4)–NH–(C6H4)–NH–]0.75[–(C6H4)–N_(C6H4)_N–]0.25
{[–(C6H4)–NH–(C6H4)–NH–]0.5[–(C6H4)–N_(C6H4)_N–]0.5
{[–(C6H4)–NH–(C6H4)–NH–]0.25[–(C6H4)–N_(C6H4)_N–]0.75
yellow
leucoemeraldine base
protoemeraldine base
emeraldine base
nigraniline base
perinigraniline base
}
blue (light green)
dark blue (green)
blue–black (blue)
violet (violet)
n
n
}
n
}
{–(C6H4)–N_(C6H4)_N–}n
of CuCEP–PANI with dilute hydrochloric acid and dilute
NaOH did not result in any colour change. As discussed
below, we believe that the PANI is in the interlayer region in
CuCEP–PANI, which probably renders it inaccessible to
external reagents. To our knowledge, the perinigranilineform
of PANI has not been reported previously to be obtainedfrom
a reaction of this type.
amount of bulk copper metal within the sample is substan-
tially smaller than for the CuCEP–PANI sample discussed
above. The presence of some bulk copper metal in CuPhP–
PANI is again evident from powder X-ray diffraction,
although this amount is apparently not sufficient to give an
observable mass increase in the TGA experiment.
The TGA trace for CuPhP–PANI shows two major mass
losses centred at 185 and 3858C, implying that the organic
material in CuPhP–PANI is lost in two steps. These mass
losses may be attributed to the loss of PANI (see also Section
3.3) and the organic part of CuPhP respectively. On the other
hand, the TGA trace for CuCEP–PANI shows only one major
mass loss centred at 3788C, suggesting that all the organic
material (PANI and the organic part of CuCEP) is lost in
one step. In both cases, there is a further mass loss at ca.
6408C, arising from the final oxidation to yield Cu2P2O7.
3.2. Sample analyses
For CuCEP–PANI, on the initial assumption that the sam-
ple containsonly CuCEP andPANI, theCHNanalysis(Table
2) leads to the composition [Cu(O3PCH2CH2CO2H)]P
[C6H4N]0.95. However, while this accounts for the relative
values of the percentages of C, H and N, it does not account
for the absolute values of these percentages. Furthermore,
this stoichiometry suggests that the residue (Cu2P2O7) in the
TGA experiments should represent 50.0% of the original
mass (whereas the actual residue is 64.7% of the original
mass), and suggests that the copper content is 16.4%
(whereas the actual copper content found from analysis is
40.0%). These discrepancies are due to the presence of bulk
copper metal (as evident from powder X-ray diffraction) in
addition to CuCEP and PANI. During the TGA experiment,
the bulk copper metal is oxidized to CuO, accounting for a
massincreasecentredaround5508C. Onthisbasis,theresidue
from the TGA experiment (after heating to 12008C) com-
prises Cu2P2O7 (from the CuCEP) and CuO (from the bulk
copper metal). In summary, the overall stoichiometry of
the recovered product is [Cu(O3PCH2CH2CO2H)]P
3.3. Powder X-ray diffraction
The powder X-ray diffraction patterns for all the solid
materials contain peaks characteristic of bulk copper metal.
For the experiment involving CuMeP, bulk copper metal is
the only crystalline product, whereas for the experiments
involving CuCEP and CuPhP, bulk copper metal is not the
major product. The powder X-ray diffraction patterns for
CuCEP–PANI and CuPhP–PANI have an intense series of
(00l) peaks, suggesting that the layered nature of the host
materials is retained during the formation of PANI. However,
as now discussed, significant changes in the positions ofthese
peaks indicate changes in the interlayer spacing. ForCuCEP–
PANI, no peaks with h/0 or k/0 are discernible above the
noise level in the powder X-ray diffraction pattern, whereas
for CuPhP–PANI, only very few peaks with h/0 or k/0
are discernible.
[C6H4N]0.95[Cu]1.3
.
For CuPhP–PANI, similar analysis (Table 2) leads to the
following overall stoichiometry of the recovered product:
[Cu(O3PC6H5)]P[C6H4N]1.1[Cu]0.1. In this case, the
Table 2
Results of analyses of CuPhP–PANI and CuCEP–PANI; the experimental errors are ca. 0.3% in the CHN analysis and ca. 0.5% in the Cu analysis; the TGA
residue represents Cu2P2O7qCuO
CuPhP–PANI
CuCEP–PANI
C (%)
46.7
26.9
H (%)
4.2
2.3
N (%)
4.8
3.4
Cu (%)
23.7
40.0
TGA residue (theoretical)
Composition
42.9 (48.5%)
[Cu(O3PC6H5)]P[C6H4N]1.1[Cu]0.1
64.7 (66.5%)
[Cu(O3PC2H4CO2H)]P[C6H4N]0.95[Cu]1.3
Thursday Apr 13 09:51 AM
StyleTag -- Journal: POLY (Polyhedron) Article: 3384

768
G.B. Hix et al. / Polyhedron 19 (2000) 765–770
For CuCEP–PANI, powder X-ray diffraction (Fig. 1)
dicular orientation of the rings with respect to the layers of
the host material.
In contrast, powder X-ray diffraction (Fig. 2) for CuPhP–
shows that the formation of PANI is associated with an
increase in the interlayer spacing of the CuCEP host material,
providing direct evidence that the PANI is located within the
interlayer region. The increase in the interlayer spacing is
PANI shows no evidence for an increase in the interlayer
˚
spacing, but instead a slight decrease from 14.5 to 12.7 A is
˚
10.2 A. It is reported that inclusion of PANI within layered
observed (note that powder X-ray diffraction patterns
recorded at lower diffraction angles (higher d-spacing) than
those shown in Fig. 2 confirm that no new peaks appear at
very low diffraction angle (such peaks would indicate an
increase in the interlayer spacing)). The slight decrease in
the interlayer spacing in CuPhP–PANI suggests that some
host materials leads to an increase in the interlayer spacing
˚
of ca. 4.2 A if the phenyl rings are oriented parallel to the
˚
host layers [9] and ca. 8.6 A if the phenyl rings are oriented
perpendicular to the host layers [10]. On this basis, we
deduce that the PANI included within CuCEP has a perpen-
Fig. 1. Powder X-ray diffraction patterns (Cu Ka1 radiation) for (a) CuCEP and (b) CuCEP–PANI. The d-spacings for selected (00l) peaks due to the layered
metal phosphonate material are indicated. Peaks marked with an asterisk are due to bulk copper metal.
Fig. 2. Powder X-ray diffraction patterns (Cu Ka1 radiation) for (a) CuPhP and (b) CuPhP–PANI. The d-spacings for selected (00l) peaks due to the layered
metal phosphonate material are indicated. Peaks marked with an asterisk are due to bulk copper metal.
Thursday Apr 13 09:51 AM
StyleTag -- Journal: POLY (Polyhedron) Article: 3384

G.B. Hix et al. / Polyhedron 19 (2000) 765–770
769
Fig. 3. FTIR spectra for (a) CuPhP–PANI and (b) CuCEP–PANI.
Table 3
structural rearrangement of the parent host material (perhaps
resulting from loss of water) is associated with the poly-
merization reaction. Our observations suggest that the for-
mation of PANI in this system does not actually involve
intercalation of PANI within the interlayer region, but may
instead occur on the external surface of the CuPhP particles.
Presumably some amount of PANI may be retained on the
external surfaces of the CuPhP particles (giving rise to the
colour of the solid product). However, the relative amount
of PANI in the CuPhP–PANI sample recovered (see Section
3.2) implicates the presence of an additional phase of PANI
(i.e. in additionto PANI ontheexternalsurfacesoftheCuPhP
particles). As there is no evidence from powder X-ray dif-
fraction for any crystalline phase other than CuPhP, we may
conclude that the excess PANI is present as an amorphous
phase.
Diagnostic vibration frequencies observed in FTIR spectra of CuPhP–PANI
and CuCEP–PANI
CuPhP–PANI CuCEP–PANI Assignment
1654
1607
1499
1085
1017
998
751
692
585
506
C_O carboxylic acid
ring mode (quinoid)
ring mode (phenyl)
n(P–O) XPO3 stretching mode
n(P–O) XPO3 stretching mode
n(P–O) XPO3 stretching mode
d(C–H) out-of-plane 5 adjacent H atoms
d(C–H) out-of-plane 5 adjacent H atoms
aromatic ring deformation
1603
1499
1033
1019
996
745
690
592
508
aromatic ring deformation
ing the chain end of the polymer). The fact that these bands
are observed may indicate that the PANI is present as short
oligomers. No bands characteristic of aniline or protonated
aniline are observed for either CuPhP–PANI or CuCEP–
PANI.
3.4. Infrared spectroscopy
Infrared spectroscopy has been used previously [8,10] to
assign the oxidation state of PANI. The assignment relies on
Bands associated with thelayeredcopperphosphonatehost
the relative intensities of peaks at ca. 1500 and 1600 cmy1
,
materials are also observed in the region 1150–990 cmy1
,
which are associated with phenyl and quinoid ring vibrations
respectively. The IR spectrum of the fully reduced form (leu-
coemeraldine) of PANI exhibits only a very weak peak at ca.
1600 cmy1, and this peak becomes apparent as the level of
oxidation of PANI increases (i.e. as the proportion of quinoid
unitsincreases). ForCuPhP–PANIandCuCEP–PANI,peaks
are observed in both regions ca. 1500 and ca. 1600 cmy1
(Fig. 3, Table 3). Other bands characteristic of the formation
of PANI are at ca. 1290 cmy1 (medium) and ca. 820 cmy1
(weak), and are assigned as ring breathing and out-of-plane
CH bending respectively. In addition, two sharp peaks
observed in the ranges 770–730 cmy1 and 710–630 cmy1
are characteristic of five adjacent ring hydrogens (represent-
representing P–O stretching modesofthephosphonategroup.
The peak at 1654 cmy1 in the spectrum of CuCEP–PANI
arises from the C_O group of the carboxylic acid. This fre-
quency is slightly lower than that expected for a free carbox-
ylic acid (ca. 1700 cmy1) and may constitute evidence for
an interaction between the carboxylic acid group and PANI
in the interlayer region.
4. Conclusions
In summary, all three copper phosphonate materials used
in thiswork give rise topolymerizationofaniline. Itisthought
Thursday Apr 13 09:51 AM
StyleTag -- Journal: POLY (Polyhedron) Article: 3384

770
G.B. Hix et al. / Polyhedron 19 (2000) 765–770
that copper aids the oxidation of aniline by atmospheric oxy-
gen, but in doing so, the copper phosphonate material is itself
destroyed in the process. Reduction of Cu(II) to bulk copper
metal is evident in the powder X-ray diffraction patterns of
all the samples recovered. For CuMeP, bulk copper metal is
the only crystalline product isolated following the polymer-
ization reaction. For CuCEP and CuPhP, PANI is formed
with retention (at least partially) of the layered copper phos-
phonate host material. For CuPhP–PANI, the emeraldine
form of PANI is produced, but formation on the surfaces of
the CuPhP particles (rather than within the interlayerregion)
is implicated. As such there is little change in the powder X-
ray diffraction pattern of the layered copper phosphonatehost
material, and, as shown by the analyses, a relatively small
quantity of bulk copper metal is produced. For CuCEP–
PANI, the PANI is present as the perinigraniline form, and
powder X-ray diffraction provides direct evidence that the
PANI is retained within the layered copper phosphonate host
material. The higher oxidation state of the PANI is reflected
by the larger quantity of bulk copper metal produced (with
the implication that the reduction of copper is associatedwith
oxidation of PANI). In conclusion, the results reported here
provide new insights concerning preparative aspectsofPANI
through reactionsin constrainedsolidstateenvironments,and
demonstrate the diversity of behaviour that may result from
the use of structurally related, although chemically different,
host materials for this reaction. Ultimately, the observed dif-
ferences in the course of the reactions in the three different
copper phosphonate materials reflect, inter alia, the differing
relative stabilities of these materials as well as the differing
relative accessibilities of the metal centres to the anilineguest
molecules. Clearly a future programme of experimental
research, including the use of both in situ and ex situ probes
of the structural and chemical evolution of the system as a
function of reaction progress, should shed further light on
mechanistic aspects of the reactions to produce PANI as well
as the concomitant structural degradation of the copper phos-
phonate materials.
References
[1]
A.G. MacDiarmid, J.C. Chiang, A.F. Richter, A.J. Epstein, Synth.
Met. 18 (1987) 285.
[2]
[3]
J.C. Chiang, A.G. MacDiarmid, Synth. Met. 13 (1986) 193.
F. Wudl, R.O. Angus Jr., F.L. Lu, P.M. Allemand, D.F. Vachon, M.
Nowak, Z.X. Lui, A.J. Heeger, J. Am. Chem. Soc. 109 (1987)
3677.
[4]
[5]
[6]
[7]
[8]
D.F. Vachon, R.O. Angus Jr., F.L. Lu, M. Nowak, Z.X. Lui, H.
Shaffer, F. Wudl, A.J. Heeger, Synth. Met. 18 (1987) 297.
E.M. Genies, M. Lapowski, C. Santier, E. Viell, Synth. Met.18(1987)
631.
S. Ghosh, G.A. Bowmaker, R. Cooney, J. Mater. Chem. 7 (1997)
597.
G.E. Asturias, A.G. MacDiarmid, R.P. Mccall, A.J. Epstein, Synth.
Met. 29 (1989) E157.
L.W. Shacklette, J.F. Wolf, S. Gould, R.H. Baughman, J. Chem. Phys.
88 (1988) 3955.
[9]
[10]
[11]
[12]
[13]
[14]
[15]
G.L. Rosenthal, J. Caruso, S.G. Stone, Polyhedron 13 (1994) 1311.
T. Challier, R.C.T. Slade, J. Mater. Chem. 4 (1994) 367.
P. Enzel, T. Bein, J. Phys. Chem. 93 (1989) 6270.
C.-G. Wu, T. Bein, Science 264 (1994) 1757.
C.-G. Wu, T. Bein, J. Mater. Chem. 6 (1994) 1109.
V. Mehrotra, E.P. Giannelis, Solid State Commun. 77 (1991) 155.
P. Cloos, A. Moreale, C. Broers, C. Badot, Clay Miner. 14 (1979)
307.
[16]
[17]
[18]
T.C. Chang, S.Y. Ho, K.J. Chao, J. Chinese Chem. Soc. 39 (1992)
209.
M.G. Kanatzidis, C.G. Wu, H.O. Marcy, D.C. DeGroot, C.R.
Kannewurf, J. Am. Chem. Soc. 111 (1989) 4139.
M.G. Kanatzidis, C.G. Wu, H.O. Marcy, D.C. DeGroot, C.R.
Kannewurf, A. Kostikas, V. Papaefthymiou, Adv. Mater. 2 (1990)
364.
[19]
`
D.J. Jones, R. El Mejjad, J. Roziere, in: T. Bein (Ed.), Supramolecular
Architecture: Synthetic Control in Thin Films and Solids, ACS Symp.
Ser., vol. 499, ACS, Washington, DC, 1992, p. 220.
[20]
[21]
`
B. Bonnet, R. El Mejjad, M.H. Herzog, D.J. Jones, J. Roziere, Mater.
Sci. Forum 91–93 (1992) 177.
´
P. Maireles-Torres, P. Olivera-Pastor, E. Rodrıguez-Castellon, A.
´
´
Jimenez-Lopez, J. Inclusion Phenom. Mol. Recog. Chem. 14 (1992)
327.
[22]
[23]
Y. Zhang, A. Clearfield, Inorg. Chem. 31 (1992) 2821.
S. Drumel, P. Janvier, P. Barboux, M. Bujoli-Doeuff, B. Bujoli, Inorg.
Chem. 34 (1995) 148.
Acknowledgements
`
We thank Professor J. Roziere and Dr D. Jones for useful
[24] H. Hayashi, M.J. Hudson, J. Mater. Chem. 5 (1995) 115.
[25] G.B. Hix, K.D.M. Harris, J. Mater. Chem. 8 (1998) 579.
[26] W.-S. Huang, B.D. Humphrey, A.G. MacDiarmid, J. Chem. Soc.,
Faraday Trans. 82 (1986) 2385.
discussions and Mrs Sylvia Williamson (University of St
Andrews) for sample analyses. We are grateful to EPSRC
for financial support.
Thursday Apr 13 09:51 AM
StyleTag -- Journal: POLY (Polyhedron) Article: 3384