Oriented crystallization of CuSO4·5H2O under a monolayer of a novel amphiphilic ligand, 8-hexadecyloxyquinaldic acid

Article abstract of DOI:10.1039/a701967j

A new amphiphilic ligand, 8-hexadecyloxyquinaldic acid (HQA), has been synthesized and the characteristics of its monolayer on pure water and on CuSO₄ aqueous solution have been examined. The HQA monolayer can interact with copper ions in the subphase, and crystallization of CuSO₄·5H₂O is induced under the monolayer. In order to understand the relationship between the different states of the monolayer and the induced inorganic crystallization, several typical states of the monolayer were chosen in our experiment. The results show that, when the monolayer is in the gas or liquid phase, its ability to control the oriented crystallization is poor. However, rather than the most dense phase, the most favorable state for oriented crystallization is the one with an area per molecule A = 17.0 A², which could be considered as a liquid solid phase. This is attributed to a coordination effect and lattice matching at the inorganic/organic interface. Our research suggests that the state of the monolayer is a very important factor during the process of induced oriented crystallization.

Full text of DOI:10.1039/a701967j

View Article Online / Journal Homepage / Table of Contents for this issue
Oriented crystallization of CuSO · 5H O under a monolayer of a
novel amphiphilic ligand, 8-hexadecyloxyquinaldic acid
4
2
Ruikang Tang, Chaoyang Jiang and Zihou Tai*
State Key L aboratory of Coordination Chemistry, Coordination Chemistry Institute, Nanjing
University, Nanjing 210093, PeopleÏs Republic of China
A new amphiphilic ligand, 8-hexadecyloxyquinaldic acid (HQA), has been synthesized and the characteristics of its monolayer on
pure water and on CuSO aqueous solution have been examined. The HQA monolayer can interact with copper ions in the
4
subphase, and crystallization of CuSO É 5H O is induced under the monolayer. In order to understand the relationship between
4
2
the di†erent states of the monolayer and the induced inorganic crystallization, several typical states of the monolayer were chosen
in our experiment. The results show that, when the monolayer is in the gas or liquid phase, its ability to control the oriented
crystallization is poor. However, rather than the most dense phase, the most favorable state for oriented crystallization is the one
with an area per molecule A \ 17.0 Ó2, which could be considered as a liquid solid phase. This is attributed to a coordination
e†ect and lattice matching at the inorganic/organic interface. Our research suggests that the state of the monolayer is a very
important factor during the process of induced oriented crystallization.
A new Ðeld, usually named biomineralization, has received
considerable attention and has developed very rapidly in
recent years.1h8 The oriented growth of crystals in biological
systems has inspired numerous studies of template-directed
crystallization under organized surfactant ultra-thin Ðlms.
Because such Ðlms can be tailored extensively by speciÐc
chemical modiÐcations, controlled oriented inorganic crys-
tallization is possible.6h15
(Elemental analysis. Found: C, 73.08; H, 8.88; N, 3.70. Calc.
for C
H
NO É H O: C, 73.17; H, 8.98; N, 3.63%.) FTIR
26 39
3
2
(KBr) l/cm~1: 3300 (COOH), 2925, 2861 (CH , CH ), 1642
2
3
(CxO), 721 (CH ). 1H NMR (CDCl ) d: 8.60 (1H, ArwH),
2
3
7.71 (1H, ArwH), 7.65 (1H, ArwH), 7.58 (1H, ArwH), 7.47
(1H, ArwH), 4.36 (3H, OCH ), 2.25 (2H, OwCwCH ), 1.57
3
2
(4H, CH ), 1.40 (6H, CH ), 1.20 [16H, w(CH ) w], 0.84 (3H,
2
2
2 8
CH ).
3
Many papers about the nucleation and growth of amino
Apparatus
acids, proteins, semiconductor particles, organic salts,
a
variety of minerals and soluble inorganic crystals under con-
densed monolayers have been published. All these studies
indicate that concepts such as morphogenesis, replication, self-
organization and metamorphosis could be useful for devising
new synthetic strategies in organic materials chemistry.16h26
However, the inÑuence of di†erent amphiphiles or di†erent
aggregation states of the monolayer on induced crystallization
is not well understood. For example, the monolayers used
in most published works were in a condensed state. Does
this mean the condensed state is the most suitable one for
inducing oriented crystallization? To answer the question, we
C, H and N analyses of HQA and its complex were performed
using a Perkin-Elmer Model 240C elemental analyser. 1H
NMR spectra were obtained using a Bruker Am-500 NMR
spectrometer with tetramethylsilane as an internal reference.
IR spectra were recorded on a Nicolet Model 170SX FTIR
spectrometer. The instrument for XPS study was an Escalab
MK-II electron spectrometer. The Langmuir trough system
was a KSV-5000 and the determinations of the crystal faces
were carried by a D/Max-RA rotating anode X-ray di†rac-
tometer.
have synthesized
a
novel amphiphilic ligand, 8-
Isotherm measurement, preparation of Langmuir–Blodgett
hexadecyloxyquinaldic acid (HQA), and have studied the
inÑuence of various states of its monolayer on its ability to
induce oriented crystallization of CuSO É 5H O. The results
show that the choice of a suitable state in the process is very
important; it is relevant to the compatibility of monolayerÏs
organized structure and lattice structural matching between
the monolayer and the inorganic crystal face.
Ðlms and XPS studies
Measurements of the isotherm curve of a Langmuir Ðlm of
HQA were carried out with a KSV-5000 Langmuir trough
system. The water used in the experiment was from a Milli-Q
4
2
Experimental
Materials
8-Hydroxyquinaldine (General Chemical and Pharmaceutical
Co., Sudbury, Su†olk) and hexadecylamine (Fluka Chemical
Co.) were used without any further puriÐcation. All solvents
were puriÐed by standard procedures. CuSO was of A.R.
4
grade, purchased from Shanghai Chemical Reagents Co.
Synthesis
The synthesis procedure for HQA is illustrated in Scheme 1.
The product is
a
light-brown powder, mp 85È87 ¡C.
Scheme 1 Synthesis of the amphiphilic ligand HQA
J. Chem. Soc., Faraday T rans., 1997, 93(18), 3371È3375
3371

View Article Online
Ultrapur water system. HQA was dissolved in chloroform at a
concentration of 1.0 ] 10~3 mol dm~3 and the solution was
spread on the subphases. Generally 15 min was allowed for
chloroform evaporation, and the monolayer was then com-
pressed by two movable TeÑon barriers at a rate of ca. 20 mm
min~1. The surface pressure (n) vs. area per molecule (A) iso-
therm was recorded automatically. Two subphases were used
The isotherm of HQA on the aqueous Cu2` subphase has
plateau and condensed regions at di†erent surface pressures
and is more expanded than that of HQA on pure water, in
which only the condensed region appears. The plateau in the
curve can be considered as the liquid phase of the monolayer
as described by Fendler.27 By extrapolating the surface pres-
sure to zero for the condensed region of HQA on pure water,
a limiting area of ca. 43 Ó2 per molecule is obtained. This
value is in good agreement with the cross-section of the head-
group plane approximated by a rectangular block of area 46
Ó2 according to a space-Ðlling molecular (CPK) model. This
suggests that the molecular planes of HQA lie almost Ñat on
the surface as shown in Fig. 2(a). On the subphase containing
copper ions, the limiting areas per molecule in the expanded
and condensed regions are ca. 86 and 18 Ó2, respectively.
Apparently, copper ions are able to modify the orientation
and fabrication of an HQA Ðlm, and the orientations of the
molecules in the monolayers are di†erent. Since the “side-onÏ
area of HQA is about 19 Ó2, approximated by the CPK
model, the orientation of CuwHQA in the monolayer on
in the experiment: pure water and aqueous CuSO solution
4
(1.0 ] 10~2 mol dm~3).
After ca. 30 min of self-organization, the monolayer was
transferred onto a clean substrate by liftingÈdipping cycles of
the substrate through the air/water interface at a mean speed
of 1 mm min~1 at a surface pressure of 30 mN m~1 for HQA
on the pure water and 35 mN m~1 for HQA on the aqueous
CuSO subphase. For deposition of LangmuirÈBlodgett (LB)
4
Ðlms on hydrophilic slides, the slides were Ðrst immersed in
the subphases and the monolayer was then formed and com-
pressed to the speciÐc surface pressure.
All the above work was carried out in a dust-free box at
20 ¡C. Both the LangmuirÈBlodgett Ðlms were examined by
XPS to gather information about the interaction between
copper ions and HQA molecules.
aqueous CuSO solution is deduced as shown in Fig. 2(b). The
4
orientation of the ring of HQA in the monolayer is perpen-
dicular to the interface.
Crystal growth of CuSO Æ 5H O under the HQA monolayer
4
2
XPS studies
CuSO (6.5 g) was added to water (20 g) at 40 ¡C. After the
solutes had dissolved, the temperature was reduced to 20 ¡C
4
As a rule, a shift in the binding energy provides a measure-
ment of the electron density of about an atom and the shift
increases with a decrease in electron density. In comparison
with the HQA Ðlm on pure water, the value of CuwHQAÏs
binding energy of the nitrogen atom was increased from 407.6
to 408.2 eV and that of oxygen from 540.2 to 540.4 eV. These
data show that coordination bonds are formed, accompanied
by electron transfer from the N and O atoms of HQA to the
copper ions. The surface composition of the LB Ðlms made on
the aqueous Cu2` subphase was determined by XPS also and
the ratio of the metal ion : nitrogen atom : carbon atom was
about 1 : 2 : 52. This means that two HQA molecules interact
with each copper ion by intermolecular bonds. As a result, a
proposed structure of the HQA monolayer on aqueous
and then the solution was Ðltered. The Ðltrate, a supersatu-
rated solution of CuSO É 5H O, was transferred to a mini-
4
2
trough of the KSV-5000 LB system and the temperature of
the subphase was increased to 25 ¡C and then reduced to
22 ¡C. Next, the monolayer of HQA was carefully and slowly
spread from the 1.0 ] 10~3 mol dm~3 solution in chloroform
and its state was controlled by the LB system through its nÈA
curve; meanwhile, the temperature was decreased to 20 ¡C and
crystals of CuSO É 5H O separated from the supersaturated
4
2
solution.
Results and Discussion
CuSO solution is illustrated [Fig. 2(b)].
4
Surface pressure–molecular area isotherm
Fig. 1 shows the surface pressureÈarea (nÈA) isotherms of an
HQA monolayer on pure water and on the Cu2` subphase.
Evidently, when Cu2` ions are added to the subphase, the
characteristics of the HQA monolayer change remarkably.
Crystal growth of CuSO Æ 5H O under di†erent states of HQA
4
2
monolayer
As mentioned above, the detailed relationship between the dif-
ferent monolayer aggregation states and their ability to induce
oriented nucleation and growth is poorly understood. Fortu-
nately, HQA provides a good chance to study the relationship
thoroughly because its monolayer on CuSO solution has
4
Fig. 1 Surface pressureÈarea isotherm of HQA on pure water
Fig. 2 Stacking modes of HQA molecules at the air/water interface.
The dashed circle represents a copper ion.
(dashed line) and on 1.0 ] 10~2 mol dm~3 CuSO solution (solid line)
4
3372
J. Chem. Soc., Faraday T rans., 1997, V ol. 93

View Article Online
obvious phase transitions. In this experiment, several typical
states of the monolayer were chosen and deÐned as IÈVII (Fig.
1). From I to VI, the values of n and A were accordingly 0.0,
100.0; 20.3, 60.0; 24, 35.0; 26.8; 20.0; 29.0, 17.0; and 41.5 mN
m~1, 10.0 Ó2. VII represents the collapsed monolayer.
is a typical surface active agent that can interact with copper
ions, which results in an increased concentration of the solute
in proximity to the monolayer. Therefore, nucleation of
CuSO É 5H O under the monolayer becomes much easier.28
However, the disordered structure of the monolayer in this
state leads to a poor ability to control oriented crystallization.
According to the statistics of nine independent experiments,
the proportion of ladder-shaped crystals at the interface was
low (6%). The crystallization of the other crystals was still
uncontrolled; they were attached to the monolayer by the
habit faces, (100), (110) and (110), in the proportions 36, 40
and 18%, respectively.
With an area per molecule of the monolayer of 35 Ó2, in
state III, the monolayer was in the liquid phase. The result
was similar to that of state II and the probability of crys-
tallization at the monolayer/water interface increased further.
The proportion of ladder-shaped crystals formed here was
14%, higher than that in state II, which indicated that the
ability of this state to control oriented crystal growth was
greater. This was on account of the monolayer in state III
being more organized. However, as with state II, the mono-
layer in state III also had an undeÐned lattice structure in this
liquid state, so its inÑuence on the oriented crystallization was
limited.
In state IV, the monolayer remained in the liquid phase. All
the crystallization occurred at the monolayer/water interface
and the monolayerÏs ability to induce oriented crystal growth
increased signiÐcantly: ladder-shaped crystals accounted for
ca. 60% of those that formed.
Interestingly, the perfect result was obtained in state V with
a surface pressure of 29.0 mN m~1 and A \ 17.0 Ó2. Under
these conditions, the monolayer could be considered to be in
the transition from the liquid to the solid phase, which is
much more organized than the former, and the monolayer
began to have a deÐned lattice structure. The crystallization
under such a monolayer was di†erent from the cases listed
above. Not only were all the crystals formed under the mono-
layer, but also the proportion of the ladder-shaped crystal
with M010N faces reached about 82%. Meanwhile, the other
15% were the plate-shaped ones whose induced face under the
monolayer was M110N. The crystals with (100) or (110) faces
attached to the monolayer were few.
4
2
When the surface pressure of the monolayer was 0 mN
m~1, it was in the gas phase and therefore not organized at
all. As was expected, the crystallization of CuSO É 5H O in
4
2
the supercooled solution was uncontrolled. Nucleation
occurred both at the air/water interface (minority) and at the
bottom of the containers (majority). These crystals were ran-
domly aggregated and had a heterogeneous size distribution.
The so-formed single crystals of CuSO É 5H O were plate-
4
2
shaped with three main crystal faces: (110), (110) and (001)
[Fig. 3(a)].
In state II, the surface pressure was 20.3 mN m~1 and the
area per molecule A
was 60.0 Ó2. This state could be seen
20.3
as the transition from gas to the liquid phase. Under this con-
dition, crystallization was di†erent from that of state I. The
nucleation of CuSO É 5H O occurred more easily under the
4
2
monolayer and the probability of the crystals forming at
the bottom decreased noticeably. At the monolayer/water
interface, although most of the crystallization was not orient-
ed, a new crystal morphology appeared under the monolayer.
It was ladder-shaped and had a special crystal face (010) [Fig.
3(b)]. This face did not belong to the habit faces of the
CuSO É 5H O crystals which grew from normal aqueous
4
2
solution and it was attached to the monolayer plane during
the whole process of nucleation and growth. Clearly, the
appearance of the ladder-shaped crystal with (010) faces was a
result of the presence of the HQA monolayer on the super-
saturated solution.
The above phenomenon can be explained by the formula,
G \ 16G 3/(3G 2),28 where G is the activation energy for
N
I
B
N
nucleation, G is the energy required to form the new interface
I
as a new phase (crystal) grows from the substable phase
(solution), G is the energy released in the formation of bonds
B
in the bulk of the aggregate and it is a function of the super-
saturation ratio S. It is well known that the surface energy is
decreased when the monolayer forms. The monolayer itself
can also be considered as a di†erent phase formed on the
surface of solution and can provide suitable sites for heter-
ogeneous nucleation, so that the value of G is lowered. HQA
However, further compression weakened the monolayerÏs
ability to control oriented crystallization. When the mono-
I
layer was in the condensed states (VI) with A
\ 10.0 Ó2,
50.0
only 13% of all the induced crystal faces under the monolayer
were (010) while (110) faces increased to 35% and the rest were
(100) and (110) faces. This suggests that although the molecu-
lar organization of the HQA monolayer was more ordered
than that of state V, its ability to control oriented crys-
tallization was reduced.
With increasing n, state VII, the collapsed monolayer, was
reached. Although all the crystallization happened at the air/
water interface, it was disordered and uncontrolled and no
ladder-shaped crystals appeared.
All the above results are summarized in Fig. 4. In the
experiment, four induced crystal faces of CuSO É 5H O, (100),
4
2
(110), (110) and (010), were investigated under an HQA mono-
layer. Except for the (010) faces, all of these are habit faces of
CuSO É 5H O and none of them could have been the relative
4
2
majority of the faces induced by the monolayer; therefore,
they are ascribed to uncontrolled crystallization. Hence, we
can consider the amount of (010) face as a measure of the
oriented crystallization under an HQA monolayer.
As we have stated above and indicated in Fig. 4, state V
shows the most oriented crystallization. Using A, in state V, of
17.0 Ó2 and the (010) faceÏs lattice area, 34.76 Ó2, the result
can be explained. As every two amphiphiles are coordinated
to a copper ion, the lattice area of copper ions under the
monolayer is 17.0 ] 2 \ 34.0 Ó2. This value is in agreement
Fig. 3 Two typical morphologies of CuSO É 5H O single crystals
4
2
obtained in the experiment: (a) is the standard morphology of the
crystals grown in the normal solution without the monolayer and, in
the experiments, some of the crystals formed under monolayer are as
same as (a); (b) is distinct from the former and a new crystal face (010)
appears on its surface, which is attached to the monolayer plane
during the growth. Morphology (b) can only form under the mono-
layer, especially under the organized monolayer whose lattice struc-
ture Ðts the (010) crystals face of CuSO É 5H O well.
4
2
J. Chem. Soc., Faraday T rans., 1997, V ol. 93
3373

View Article Online
under a disorder monolayer, it is impossible for the crys-
tallization to be oriented and controlled. The results of state II
and III also lead to the same conclusion. Conversely, when
the monolayer is in the condensed phase and has the most
organized structure, as in state VI, its area per molecule is too
low and no crystal face of CuSO É 5H O is able to match it
4
2
well. Thus, the induced crystallization is not oriented, as with
unorganized monolayers. This example demonstrates again
that both the organic matrix as a template and the lattice
structural matching between the template and induced crystal
face are very important during inorganic oriented crys-
tallization induced by a monolayer. However, the above two
key factors are not compatible with each other in the crys-
tallization of CuSO É 5H O under an HQA monolayer. This
4
2
is the reason why all of the CuSO É 5H O crystals formed
4
2
under the HQA monolayer were not oriented with the same
kind of induced face.
Fig. 4 Frequency of the induced face (010) of the CuSO É 5H O
4
2
crystal with the HQA monolayer in di†erent states
Fortunately, in another previously reported experiment,30
the monolayer selected for inducing the crystallization of
CuSO É 5H O
was
9-(hexadecylimino)-4,5-diazaÑuorene
4
2
(HIDF). Unlike HQA, the state whose lattice structure
with that of the (010) face of the CuSO É 5H O crystal. This
matches the (010) face of CuSO É 5H O best is in the com-
4
2
4
2
suggests that the monolayer and the crystal face correspond
structurally in this state; two-to-one geometrical matching
between functional groups of the template and the crystallo-
graphic lattice dimensions of a special crystal face occurs and
controlled crystal growth is induced (Fig. 5). However, state V
is not in the compressed phase; many defects are present in
the monolayer, which could possibly result in the formation of
other crystal faces. This is why all the induced crystal faces in
this state are not (010). It is also noted that the lattice area of
the (110) face, next most common to (010) under this mono-
layer, is 35.70 Ó2, almost equal to twice A of the monolayer as
well. We think that this crystal face is induced by the defects
of the monolayer, implying that A of the defects is also ca. 17
Ó2. This agrees well with the experimental results both of the
crystallization and of the nÈA curve.
pressed phase, which means that the two factors can be satis-
Ðed at the same time. As a result, the outcome is ideal as all
the crystals are ladder-shaped with (010) faces attached to the
monolayer.
Conclusion
In this paper, we have reported the synthesis and character-
istics of the novel amphiphilic ligand HQA and focus on the
ability of its monolayer in various states to induce oriented
crystallization of CuSO É 5H O. The results of this experi-
4
2
ment indicate that the selection of a suitable state of the
monolayer is essential for induced oriented crystallization.
Both an organized structure and structural complementarity
play key roles during the process. For a good template, these
two factors should result in compatibility of the monolayer.
These conclusions are helpful for people trying to understand
biomineralization thoroughly and may be relevant to any
process involving transcriptive synthesis.
But how about the other states of the monolayer? For
instance, in state IV where A
\ 20 Ó2, why is its ability to
26.8
induce oriented crystallization much lower than that of V?
First of all, this state is in the liquid phase as its A is larger
than 18 Ó2, the limiting area of the monolayer which is con-
sidered as the critical point of solid phase formation. There-
fore, it retains many more characteristics of the liquid phase
than state V does and is not well organized. In the opinion of
Mann and co-worker,11,29 transcriptive synthesis involves a
reorganized, self-assembled and relatively stable organic archi-
tecture as the chemical and structural template for patterned
materials deposition. Two-dimensional lattice structures of
organized molecular Ðlms are generally considered of the most
fundamental importance in transcriptive synthesis. Obviously,
This research was supported by the National Natural Science
Foundation of China.
References
1
H. A. Lowenstam and S. Weiner, On Biomineralization, Oxford
University Press, Oxford, 1989.
2
Biomineralization: Chemical and Biochemical Perspectives, ed. S.
Mann, J. Webb and R. J. P. Williams, VCH Press, New York,
1989.
3
4
S. Mann, Struct. Bonding, 1983, 54, 125.
A. Berman, L. Addadi and S. Weiner, Nature (L ondon), 1988, 331,
546.
5
6
7
A. Berman, L. Addadi, A. Kuick, L. Leiserowitz and M. Nelson,
Science, 1990, 250, 664.
W. Traub, T. Arad and S. Weiner, Proc. Natl. Acad. Sci. USA,
1989, 86, 9822.
b = 6.0 Å
Cu
HQA
A. H. Heuer, D. J. Frind, V. J. Laraia, J. L. Arias, P. D. Calvert,
K. Kendall, G. L. Messing, J. Blackwell, P. C. Rieke, D. H.
Thompson, A. P. Wheeler, A. Veis and A. I. Caplan, Science,
1992, 255, 1098.
L. Addadi and S. Weiner, Angew. Chem., Int. Ed. Engl., 1992, 31,
153.
= 72°
θ
8
9
a = 6.1 Å
S. Mann, J. Chem. Soc., Dalton T rans., 1993, 1.
10 S. Weiner, Biochemistry, 1983, 22, 4139.
11 S. Mann, Nature (L ondon), 1988, 332, 119.
12 S. Mann, Nature (L ondon), 1993, 365, 499.
13 X. K. Zhao, S. Baral, R. Rolandi and J. H. Fendler, J. Am. Chem.
Soc., 1988, 110, 1012.
Fig. 5 Schematic two-dimensional representation of the proposed
overlap between Cu2` ions and HQA headgroups. The arrangement
of the ions under the monolayer is similar to that on the (010) face of
CuSO É 5H O crystals.
4
2
3374
J. Chem. Soc., Faraday T rans., 1997, V ol. 93

View Article Online
14 N. Herron, Y. Wang, M. M. Eddy, G. D. Stucky, D. E. Cox, K.
Moller and T. Bein, J. Am. Chem. Soc., 1989, 111, 530.
15 R. F. Ziolo, E. P. Giannelies, B. A. Weinstein, M. P. OÏHoro,
B. N. Ganguly, V. Mehrota, M. W. Russell and D. R. Hu†man,
Science, 1992, 257, 219.
23 R. M. Kenn, C. Bohm, A. M. Bibo, I. R. Peterson, H. Mohwald,
J. Als-Nieisan and K. Kjaer, J. Phys. Chem., 1991, 95, 2092.
24 X. K. Zhao, J. Yang, L. D. McCormick and J. H. Fendler, J.
Phys. Chem., 1992, 96, 9933.
25 A. Berman, D. J. Ahn, A. Lio, M. Salmerson, A. Reichert and D.
Charych, Science, 1995, 269, 515.
16 E. M. Landau, M. Levanon, L. Leiserowitz, M. Lahav and J.
Sagiv, Nature (L ondon), 1985, 318, 353.
26 S. Mann, D. D. Archibald, J. M. Didymus, T. Douglas, B. R.
Heywood, F. C. Meldrum and N. J. Reeves, Science, 1993, 261,
1286.
17 E. M. Landau, S. G. Wolf, M. Levanon, L. Leiserowitz, M. Lahav
and J. Sagiv, J. Am. Chem. Soc., 1989, 111, 1436.
18 I. Weissbuch, F. Frolow, L. Addadi, M. Lahav and L. Leisero-
witz, J. Am. Chem. Soc., 1990, 112, 7718.
27 J. H. Fendler, Membrane Mimetic Chemistry, Wiley, New York,
1982.
19 S. Mann, J. M. Didymus, N. P. Sanderson and B. R. Heywood, J.
Chem. Soc., Faraday T rans., 1990, 86, 1873.
28 G. H. Nancollas, Biological Mineralization and Demineralization,
Springer, Berlin/Heidelberg, 1982.
20 S. Rajam, B. R. Heywood, J. B. A. Walker, S. Mann, R. J. Davey
and J. D. Birchall, J. Chem. Soc., Faraday T rans., 1991, 87, 727.
21 B. R. Heywood, S. Rajam and S. Mann, J. Chem. Soc., Faraday
T rans., 1991, 87, 735.
22 S. Mann, B. R. Heywood, S. Rajam and J. D. Birchall, Nature
(L ondon), 1988, 334, 692.
29 S. Mann and G. A. Ozin, Nature (L ondon), 1996, 382, 313.
30 R. Tang, Z. Tai and Y. Chao, J. Chem. Soc., Dalton T rans., 1996,
4439.
Paper 7/01967J; Received 20th March, 1997
J. Chem. Soc., Faraday T rans., 1997, V ol. 93
3375