Detecting a transition-metal ammine at tailored surfaces

Article abstract of DOI:10.1039/a700325k

The fabrication of surfaces by forming Langmuir films, which incorporate amphiphiles containing hydrophilic 18-crown-6 (18C6) derivatives, at a gas/water interface is described. These Langmuir films can be transferred to a hydrophobised quartz crystal microbalance (QCM), using the Langmuir-Blodgett technique. The QCM response has been measured in aqueous solution as a function of the concentration of the transition metal complex [Co(NH₃)₆]Cl₃ which was injected into a vial in which the film-coated QCM had been immersed. By comparing various surfaces covered with hydrophilic polyether and hydroxy functions and hydrophobic methyl groups, and by varying the composition of the films so as to increase the separation between the 18C6 macrocycles, it has been demonstrated that surfaces can be tailored that will enhance the binding of the [Co(NH₃)₆]³⁺ trications.

Full text of DOI:10.1039/a700325k

Detecting a transition-metal ammine at tailored surfaces
Sayeedha Iqbal,a Felix J. B. Kremer,b Jon A. Preece,*a Helmut Ringsdorf,b Martin Steinbeck,b J. Fraser Stoddart,b
Jie Shenc and Nigel D. Tinkerd
aSchool of Chemistry, University of Birmingham, Edgbaston, Birmingham, UK B15 2T T
bInstitut f u¨ r Organische Chemie, Johannes Gutenberg-Universit a¨ t, Becher Weg 18–20, 55099 Mainz, Germany
cDepartment of Applied Physics, De Montford University, T he Gateway, L eicester, UK L E1 9BH
dBNFL , Springfields Works, Salwick, Preston, L ancashire, UK PR4 0XJ
The fabrication of surfaces by forming Langmuir films, which incorporate amphiphiles containing hydrophilic 18-crown-6 (18C6)
derivatives, at a gas/water interface is described. These Langmuir films can be transferred to a hydrophobised quartz crystal
microbalance (QCM), using the Langmuir–Blodgett technique. The QCM response has been measured in aqueous solution as a
function of the concentration of the transition metal complex [Co(NH ) ]Cl which was injected into a vial in which the film-
3
6
3
coated QCM had been immersed. By comparing various surfaces covered with hydrophilic polyether and hydroxy functions and
hydrophobic methyl groups, and by varying the composition of the films so as to increase the separation between the 18C6
macrocycles, it has been demonstrated that surfaces can be tailored that will enhance the binding of the [Co(NH ) ]3+ trications.
3
6
In 1959, Sauerbrey1 showed that, when a quartz crystal comes
into contact with a gas, the change in frequency DF of the
quartz crystal, sandwiched between two excitation electrodes,
ammine complexes depend significantly on the nature of the
solvent molecules. In the times of Werner, these non-covalent
bonding interactions were not well understood at a molecular
level. However today, with the advent of various spectroscopic
techniques and X-ray crystallography, intermolecular inter-
actions are becoming more fully understood. Indeed, new areas
of science are emerging from the study of molecular inter-
actions: they include crystal engineering,27 host–guest28 and
supramolecular chemistry.29 They are all disciplines which rely
upon the natural concepts of self-assembly30 and self-organis-
of natural resonant frequency
F
(s−1), density
r
0
(
2.648 g cm−3), and shear modulus m (2.947×1011 dyn cm−2
for AT-cut quartz) is related to the adsorbed mass Dm by the
relationship:
2
F 2 Dm
A(r m) 1/2
DF=−
o
(1)
where A is the exposed surface area (m2) of the quartz. This
pioneering activity led to the development of the quartz crystal
microbalance (QCM). The QCM has proved to be a highly
versatile instrument for the determination of the amount of
material deposited from the gas phase2 on to a solid surface.
Applications include thickness monitors in metal evaporation
and deposition processes, and the detection of gas-phase
analytes, such as hydrocarbons, water vapour and other vol-
atile compounds. A more demanding, but potentially more
important, area in which the QCM is being employed currently,
is in liquid media,3 where it has been used to measure interfacial
processes at electrode surfaces.4 It has also been employed as
an immunosensor5 at the nanogram level to monitor anti-
bodies,6 bacterial growth,7 cells,8 proteins,9 and microbes,10 as
well as to detect surfactants,11 anaesthetics,12 antibiotics,13
bitter and odorous substances,14 DNA hybridisation,15 pH
changes,16 enzyme reactions,17 liposomes,18 chiral recog-
nition,19 intercalation,20 and even cell growth.21 Many of these
experiments involve the molecular recognition of a substrate
from the subphase to a biological receptor which has been
deposited on the QCM surface. There are, however, very few
examples in the literature where the QCM has been utilised
to detect recognition events involving totally synthetic
systems.22
The research reported in this paper relates to detecting
molecular recognition events within a wholly synthetic system.
It is known from many crystal structures23 that transition-
metal (ammine) complexes (TMCs) hydrogen bond24 via the
hydrogen atoms of their ammine ligands to the oxygen atoms
of crown ethers, e.g. 18-crown-6 (Fig. 1). This type of molecular
recognition is an example of second-sphere coordination,25 a
concept first discussed by Alfred Werner26 in 1912 when he
noted that the optical properties of chiral transition-metal
Fig. 1 Second-sphere coordination in the solid state: (a) illustrating
the first- and second-sphere ligands, and (b) illustrating the supramol-
ecular 151 polymer formed between [Cu(NH ) (H O)]2+ and 18C6
3
4
2
(hydrogen atoms are omitted for clarity)
J. Mater. Chem., 1997, 7(7), 1147–1154
1147

ation31 to construct arrays of molecules held together by non-
covalent bonding interactions.32
Systems I and III form poorly expanded monolayers with
extremely low collapse pressures of approximately 32 and 28
mN m−1, respectively (Fig. 3). Presumably, the poor stability
of these films is a result of (i) the larger area requirement of
the polyether moieties relative to the alkyl chains (especially
in the case of the crown ether lipid), (ii) the electronic repulsion
between electron lone pairs on the oxygen atoms, and (iii) the
high solvation of the polyether moieties by the water molecules.
Conversely, the isotherm of the octadecanol 3 monolayer is
very stable, collapsing at just below 60 mN m−1, forming a
solid analogous phase. Thus, a compromise is required between
the poorly expanded films of 1 and 2 and the well condensed
film of 3. This compromise was achieved by cospreading the
two ether compounds 1 and 2 with octadecanol 3. Fig. 3 shows
the isotherms of the monolayers formed from systems II (1
with 3) and IV (2 with 3) in which one molar equivalent of
octadecanol 3 is cospread with the ether amphipiles 1 and 2,
respectively. The isotherms of systems II and IV are less
expanded, relative to systems I and III, and these two compo-
nent films are stable until around 45 mN m−1. It should be
noted that, at pressures between 30 and 45 mN m−1, the
isotherms have a short phase change from a liquid analogous
phase to a close-packed phase. Isotherms of cospread mixtures
with molar ratios of the ether compounds to octadecanol 3 of
Here, we report the chemical modification of the 18-crown-
6
(18C6) macrocycle in order to facilitate its incorporation33
into Langmuir films which can be deposited on to a QCM by
the vertical dipping method, such that the hydrophilic head
groups are exposed to the aqueous environment. An aqueous
solution of [Co(NH ) ]Cl can then be injected into the
3
6
3
aqueous environment in which the LB film-coated QCM is
immersed, and the frequency response can be measured as a
function of the concentration of [Co(NH ) ]Cl to yield
3
6
3
information about the kinetics and thermodynamics of the
recognition process.34
Results and Discussion
General remarks
The compounds, which were used in the present research, are
listed in Fig. 2. Compound 1 is a chemically modified 18C6
derivative in which one of the methylene hydrogen atoms has
been replaced by an oxymethylene octadecanoate chain to
make it amphiphilic in nature. Compound 2 is a linear
polyether analogue of 1 in which the polyether is bonded
covalently to the aliphatic chain by an ester linkage.
Compounds 3 and 4 are simply the commercially available
octadecanol and thiooctadecanol, respectively. Compound 5 is
the kinetically inert transition-metal ammine complex
1
52 and 154 (and 158 in the case 1) were recorded and showed
the general trend that, as the amount of octadecanol was
increased, the film became less expanded. This point can be
appreciated from inspection of data recorded in Table 1 where
the area per molecule at 25 mN m−1 (the pressure at which
the films were transferred to the QCM for all systems) decreases
as the proportion of octadecanol 3 increases.
[
Co(NH ) ]Cl . Compounds 1–4 were chosen for several
3
6
3
reasons. Firstly, 1 and 2 were utilised to establish if a macro-
cyclic eÂect was operating in addition to purely non-specific
binding. The hydroxy compound 3 was employed to establish
if the transition-metal ammine complex had any aÃnity for a
hydrophilic surface. Conversely, the thiol 4 was utilised to
establish if the [Co(NH ) ]Cl 5 had an aÃnity for a hydro-
QCM studies
The monolayers at the gas/water interface transferred with
good transfer ratios (0.85–0.95) on to hydrophobised quartz
supports by the vertical dipping mode into the aqueous
subphase, thus achieving X-type deposition.
3
6
3
phobic surface. Additionally, the hydroxy amphiphile 3 was
employed to increase the separation of the polyether head-
groups in cospread monolayers incorporating 1 and 2. This
incorporation of 3 into monolayers formed from ether contain-
ing compounds 1 and 2 allows control over the intermolecular
separation between the ether moieties in these monolayers,
something which is crucial for the transition-metal ammine
complex to become inserted into the monolayer. Thus, various
systems were investigated in which the molar ratio of the
octadecanol to the ether amphiphiles was increased systemati-
cally. These systems are illustrated in Fig. 2.
The QCM was covered with a monolayer of the various
systems I–V by the same vertical dipping technique, after the
QCM surface was hydrophobised with a polymer solution,
leading to an X-type deposition which is illustrated in Fig. 4.
The QCM is held on to a Teflon case by a vacuum and
remains immersed in a vial which contains 3 ml of water and
a stirrer bar. Several 50 ml aliquots of a 0.1 m aqueous
[Co(NH ) ]Cl were injected into the vial and the change in
3
6
3
frequency of the QCM was monitored as each injection was
performed.
Monolayer formation
The QCM responses for the 152 ratios for systems II and
IV, together with system V, the one component octadecanol 3
system, are shown as a function of time in Fig. 5. It can be
observed that initially the QCM frequency drops rapidly after
each injection and then reaches a plateau at equilibrium
between the [Co(NH ) ]3+ trications being adsorbed and not
A few points are noteworthy about the Langmuir film forming
ability of the single and mixed component systems. Firstly,
consider the single component systems, systems I, III and V.
3
6
Fig. 2 Lipid compounds and transition-metal ammine complexes used
in this research, showing the orientation of the molecules when
adsorbed on to the QCM. The tabulated information describes the
constitutions of systems I–VI.
Fig. 3 Isotherms of systems I–VI
1
148
J. Mater. Chem., 1997, 7(7), 1147–1154

Table 1 Thermodynamic data for systems II, IV, V, VI adsorbed on to the QCM upon injections of 50 ml of 0.1 m aqueous [Co(NH ) ]Cl
3
3
6
solution into 3 ml of water in which the QCM was immersed
system
relative
binding
indexd
(ratio)
Aa/nm2
n×10−14b
DFc/Hz
K /dm3 mol−1 e
K /dm3 mol−1 f
a
a
II (151)A
II (151)B
II (152)A
II (152)B
II (154)A
II (154)B
II (158)A
II (158)B
0.318
0.318
0.289
0.289
0.253
0.253
0.201
0.201
1.62
1.62
1.78
1.78
2.04
2.04
2.56
2.56
1234
1308
4399
4452
3508
3796
1765
1900
20
22
66
67
46
49
18
20
373
354
121
128
150
131
395
319
297
363
131
141
163
143
448
339
IV (151)A
IV (151)B
IV (152)A
IV (152)B
IV (154)A
IV (154)B
0.299
0.299
0.281
0.281
0.254
0.254
1.71
1.71
1.82
1.82
2.02
2.02
2422
2259
1568
1960
1778
2063
22
35
23
19
26
27
262
253
343
250
361
272
283
278
383
282
328
289
VA
VB
VI
0.189
0.189
—g
2.73
2.73
2.73
1735
2213
102
17
22
1
339
224
87
375
244
97
aArea per molecule when film is transferred to the QCM. bNumber of amphiphiles transferred to the QCM; calculated from the area per molecule
at 25 mN m−1 for the Langmuir films of systems II and VI and 50 mN m−1 for system V multiplied by the area of the gold electrode on to
which the films were transferred (area=0.513 cm2). cChange in frequency at infinite [Co(NH ) ]Cl ; calculated from the Lineweaver Burke plot
3
6
3
of the reciprocal of concentration of TMC against the reciprocal of the change in frequency. dCalculated from normalising the change in
frequency at infinite [Co(NH ) ]Cl concentration. Normalisation achieved by accounting for the fact that diÂerent numbers of amphiphiles
3
6
3
were transferred to the QCM for the various systems. eAssociation constant (k /k ). fAssociation constant calculated from the quotient of the
from solution.
1
−1
gradient and the value of the intercept on the y-axis of the straight line obtained from the Lineweaver Burke plots (LWB). gChemisorbed
Fig. 4 Diagrammatic representation of the QCM experimental set-up with a magnification of a cartoon representation of the recognition event
on the QCM to detect complexation on a receptor-derivatised QCM surface in contact with a solution containing complementary guest molecules
adsorbed on to the surface. The three curves demonstrate that,
after each successive injection of the [Co(NH ) ]Cl aqueous
the self-assembled monolayer of thioocatedecanol 4. The first
point to note is that system V, which presents a purely
hydrophobic surface to the aqueous environment, has a very
small response to the TMC as subsequent injections are made.
Indeed, the same response was recorded when pure water was
injected. Thus, the small changes in frequency are not a result
of the TMC having an aÃnity for this hydrophobic surface,
but merely a result of the physical changes experienced by the
QCM as the water level rises which, in turn, exerts a slightly
greater pressure on the QCM.35 Contrast this result with the
hydrophilic surfaces of systems IV and V. In these cases, the
QCM response is much larger than for system VI, or when
pure water is injected into the vial containing the QCM with
3
6
3
solution, (i) the change in frequency is less, (ii) the crown ether
containing film, system II, has a considerably greater response
than its linear counterpart, namely system IV, and (iii) the
one-component octadecanol 3 system, which presents a purely
hydrophilic surface to the aqueous environment, has a similar
response to system IV.
This data is more clearly presented in a concentration
dependent titration curve for the QCM response. Fig. 6 shows
the QCM response for system IV, the octadecanol 3 and the
amphiphilic polyether 2, in the molar ratios 151, 152 and 154,
together with system V, pure octadecanol 3, and system VI,
J. Mater. Chem., 1997, 7(7), 1147–1154
1149

Fig. 7 QCM concentration-dependent titration curves for the hydro-
philic system II upon injection of 50 ml aliquots of 0.1 m aqueous
[
Co(NH ) ]Cl , illustrating the significantly diÂerent responses for
52 ratio system II, relative to the hydrophilic surfaces of system IV
3
6
3
the various ratios of 153 and the significantly greater response for the
Fig. 5 Comparison of the time-dependent titration curves for system II
are injected into the vial containing the QCM
1
(
152) and system IV (152) as 50 ml aliquots of a 0.1 m [Co(NH ) ]Cl
3
6
3
and V
DF , DF , and DF are the changes in frequency after 10 s
max
t
eq
from the first injection, at equilibrium for the first injection
and at infinite concentration of the TMC, i.e. when all the
surface recognition sites are filled, obtained from the y-intercept
of the Lineweaver Burke plot of the reciprocal of concentration
against the reciprocal of change in frequency.
The results from Table 2 are depicted graphically in Fig. 8.
The ratio of the molar equivalents of octadecanol 3 against
the ether amphiphiles 1 and 2 is plotted on the x-axis and the
rate of complexation (rate on, k ) plotted on the left hand side
1
y-axis and the rate of decomplexation (rate oÂ, k ) plotted
−
1
on the right hand side y-axis.
First of all, consider the rates of complexation and decom-
plexation for the linear polyether lipid in system IV. Here, it
is evident that there is very little variation of these two
parameters, illustrating once again the very non-specific nature
of the complexation event involving the surfaces containing
the linear polyether and octadecanol 3. However, when one
considers the crown ether lipid containing films of system II,
it is evident that, at the 152 to 154 ratio of crown ether lipid
Fig. 6 QCM concentration-dependent titration curves for the hydro-
philic systems IV and V, and the hydrophobic system V, upon
injection of 50 ml aliquots of 0.1 m aqueous [Co(NH ) ]Cl ,
3
6
3
illustrating the non-specificity of all the hydrophilic surfaces
to octadecanol 3, maxima result for both the rate on (k ) and
1
rate o (k ). However, it is slightly surprising that, on closer
−1
these hydrophilic surfaces. However, for all of these hydrophilic
surfaces, the responses are very similar. This result indicates
that the complexation event on the surfaces formed from
system IV and V is not a result of a complementary molecular
recognition event, i.e., it is a result of non-specific non-covalent
bonding interactions.
inspection of Fig. 8, the diÂerence between the complexation
and decomplexation rate is at its smallest at the 152 ratio of
system II. This result means that the binding of the TMC is
apparently weakest for this ratio, even although it has been
established that the QCM response is largest for system II
(152). This apparently anomalous behaviour will be discussed
later in the paper.
The non-discriminatory nature of the molecular recognition
event between the [Co(NH ) ]3+ trications and systems IV
The thermodynamic data37 for the binding events of these
systems are shown in Table 1. This data is summarised graphi-
cally in Fig. 9 and 10. Fig. 9(a) illustrates the maximum38
QCM responses obtained from the Lineweaver Burke plots,
derived from the graphs illustrated in Fig. 6 and 7. It is clearly
evident that, for the linear polyether lipid systems (system IV),
the response is essentially independent of the composition of
the film. This result is in contrast with system II, the crown
ether lipid containing films, where the largest response by the
QCM is with the 152 molar ratio film. Fig. 9(b) represents a
normalisation of the data in Fig. 9(a), to compensate for the
slightly greater number of amphiphilic molecules which are
transferred to the QCM at 25 mN m−1 as the molar proportion
of the octadecanol increases in the film. Thus, one could expect
greater QCM responses for films with more molecules per unit
area. However, upon inspection, this expected increase in
response is clearly not observed. System IV surfaces still have
no distinct feature, and the maximum of system II is still at
the 152 ratio. Thus, the 152 ratio film of system II complexes
to more [Co(NH ) ]3+ trications than any of the other films
3
6
and V, illustrated in Fig. 6, contrasts extremely well with the
amphiphilic crown ether containing films of system II (Fig. 7).
Initially, when only one equivalent of octadecanol 3 is present
in the film with the amphiphilic crown ether, an intermediate
QCM response is obtained between that of the purely hydro-
phobic surface of system VI and the non-specific hydrophilic
surfaces of system IV and V. However, when two molar
equivalents of octadecanol 3 are introduced into the film, the
response of the QCM is much greater than those of the non-
specific hydrophilic surfaces of systems IV and V. When four
equivalents of octadecanol 3 are introduced into the films of
system II, the response of the QCM is then intermediate
between the non-specific hydrophilic surfaces and the film
formed from 1.0 molar equivalent of 1 and 2.0 molar equival-
ents of 3. Finally, when 8.0 molar equivalents of octadecanol
3
are incorporated into the film with the crown ether amphiph-
ile, the QCM response is very similar to the non-specific
hydrophilic surfaces of systems IV and V.
The kinetic data for these systems is shown in Table 2, where
3
6
1
150
J. Mater. Chem., 1997, 7(7), 1147–1154

Table 2 Kinetic dataa for systems II, IV and V adsorbed on to the QCM upon injections of 50 ml of 0.1 m aqueous [Co(NH ) ]Cl solution
3
6
3
into 3 ml of water in which the QCM was immersed
system
DF
b/s−1
DF c/s−1
DFmaxd/s−1
k
k
t=10s
eq
1
−1
II(151)A
II(151)B
II(152)A
II(152)B
II(154)A
II(154)B
II(158)A
II(158)B
222
200
679
682
626
598
319
403
474
487
738
785
702
684
700
659
1234
1308
4399
4452
3508
3796
1765
1900
14.51
13.17
25.50
21.43
26.76
22.36
14.53
19.57
0.039
0.037
0.211
0.167
0.178
0.169
0.037
0.061
IV(151)A
IV(151)B
IV(152)A
IV(152)B
IV(154)A
IV(154)B
315
457
296
287
370
290
672
733
572
577
670
636
2259
2422
1568
1960
1778
2063
11.22
17.58
15.94
12.17
18.09
11.05
0.044
0.067
0.046
0.049
0.050
0.040
VA
277
632
1735
12.58
0.037
VB
284
603
2213
10.44
0.046
aY. Ebara and Y. Okahata, J. Am. Chem. Soc., 1994, 116, 1209. bChange in frequency 10 s after the first injection of the TMC. cEquilibrium
change in frequency after first injection of TMC. dMaximum change in frequency at infinite concentration of TMC extrapolated from the
Lineweaver Burke plot of reciprocal of change in frequency against reciprocal of concentration of TMC.
Fig. 8 Plot of the rate of complexation (rate on, & ) and rate of
decomplexation (rate oÂ, $ ) as the amount of octadecanol increases
in systems II (—) and IV (A)
Fig. 10 Plot of the binding constants obtained both kinetically ($ )
and thermodynamically (& ) for (a) system II and (b) system V as a
function of the amount of octadecanol in the monolayer
studied in this paper, while all system IV (and system V)
surfaces complex only approximately one-third of the number
of trications complexed by system II (152 ratio) and are
independent of surface composition.
Fig. 10 depicts the variation of the binding constants, calcu-
lated both kinetically (k /k
)
and thermodynamically
1
−1
(
Lineweaver Burke plots of data in Fig. 6 and 7) of the films
towards the TMC as a function of the ratio of the ether lipids
1
and 2 and octadecanol 3. These graphs illustrate the extremely
good agreement between the kinetically established binding
constants and the thermodynamically established ones. Again,
it is evident that system IV and V surfaces, containing the
linear polyether lipid, complex to the [Co(NH ) ]3+ trications
3
6
independent of the surface composition, such that they all have
a K of approximately 300 dm3 mol−1. In contrast, it is evident
Fig. 9 Plots of (a) the maximum change in frequency at infinite
a
once again that system II, containing the amphiphilic crown
concentration of [Co(NH ) ]Cl (obtained from Lineweaver Burke
3
6
3
plot), and (b) the normalised change in frequency on taking into
ether 1 and octadecanol 3, has a minimum at the 152 ratio.
account the diÂerent number of amphiphiles transferred to the QCM
However, this minimum is actually illustrating that this surface
J. Mater. Chem., 1997, 7(7), 1147–1154
1151

has the lowest binding constant towards the [Co(NH ) ]3+
has the greatest QCM response, as a result of the 151 (trica-
tion5crown ether) stoichiometry, rather than the 15several
(trication5polyether) stoichiometry for the system IV films.
3
6
with a value for K of approximately 125 dm3 mol−1. This
a
behaviour is apparently anomalous, since it has already been
established that this molar ratio of system II equates with the
largest QCM response, i.e. it complexes to the largest amount
of the [Co(NH ) ]Cl , as illustrated graphically in Fig. 7 and
Furthermore, the low K value for system II (152) is a result
a
of the fine balance between the correct steric fit of the TMC
in the surface cavities and the weak second-sphere hydrogen-
3
6
3
9
. In order to explain this anomalous behaviour, the following
bonding interactions enabling the [Co(NH ) ]3+ trications to
3
6
two models are proposed. Firstly, consider system IV, the
linear polyether 2 and the octadecanol 3 containing films.
These systems have a relatively large binding constant relative
to system II (152) but bind substantially less [Co(NH ) ]Cl .
slip in and out of the surface cavities easily, such that the rate
o is not so diÂerent, relative, to the rate on, leading to a low
K value for system II.
a
3
6
3
This behaviour can be explained by inspection of the model
Conclusions
in Fig. 11 where we consider that each [Co(NH ) ]3+ trication,
3
6
when it reaches the hydrophilic surfaces is ‘captured’ by several
polyether arms. The net eÂect is that, in order to break free
from the surface, several polyether arms have to unravel
simultaneously from the [Co(NH ) ]3+ trications. Thus, the
The observation in the solid state of the so-called second-
sphere coordination between transition-metal ammine com-
plexes and 18C6 ligands, has prompted the chemical modifi-
cation of 18C6 to make it amphiphilic in nature, such that it
could be incorporated into Langmuir–Blodgett films on solid
supports. By utilising a QCM as the solid support, it has
proved possible to detect this second-sphere coordination upon
introduction of the transition-metal ammine complex
[Co(NH ) ]Cl into a solution in which the film-coated QCM
3
6
rate of decomplexation is slow relative to the rate of com-
plexation, resulting in a relatively large binding constant, K .
a
Now consider system II (the 152 ratio variation). Here, the
largest QCM response is observed (Fig. 9) compared with all
the systems studied, yet this system has the lowest binding
constant (Fig. 10). Consider the model in Fig. 12. Here, the
crown ether moiety is spaced out at just the correct distance
to allow the [Co(NH ) ]3+ trication to slip easily into the
3
6
3
was immersed, and has enabled the kinetic and thermodynamic
characterisation of these very weak molecular recognition
events, something which was not possible by other techniques.
It turned out that the complexation was critically dependent
on the composition of the film. The evaluation of the kinetic
and thermodynamic data has enabled a model of the surface
recognition event to be formulated. This model highlights that,
although these very weak NMH, O hydrogen-bonding inter-
actions are competing with the competitive aqueous environ-
ment, the recognition still occurs. By tailoring the film, the
3
6
surface cavity created by two neighbouring crown ethers.
Additionally, the spacing is such that each crown ether binds
to two [Co(NH ) ]3+ trications such that the overall stoichi-
3
6
ometry of the film is 151 with respect to the trication and
1
8C6 head groups. This model then establishes why this film
1
8C6 moieties are preorganised36 in the film such that they
create many surface recognition sites which are stereoelectron-
ically compatible with the [Co(NH ) ]3+. However, the bind-
3
6
ing of these trications is weak, relative to the linear polyether
containing surfaces which bind many fewer trications but in a
significantly stronger manner.
The recognition in natural systems at interfaces by hydrogen-
bonding interactions is of utmost importance, since it is
responsible for immune responses,37 amongst other biologically
important signals. It follows that the study of simpler synthetic
systems38 is of considerable value in shedding light on the
more complex biological recognition events as well as for the
development of new sensors.4
Additionally, we have demonstrated that the QCM oÂers
yet another valuable tool to the research worker who is
studying unnatural supramolecular systems, where the recog-
Fig. 11 Model of the complexation event between the system IV
surfaces and the [Co(NH ) ]3+ trications, illustrating that several
3
6
polyether arms complex with the [Co(NH ) ]3+, leading to a small
in Fig. 12)
3
6
decomplexation rate, relative to the system II (152) model (depicted
nition event is between relatively small molecules (M # 300 u),
r
in contrast with the majority of studies carried out to date
which have been concerned with the detection of naturally
occurring macromolecules5–21 (M >30000 u).
r
Experimental
Solvents were dried using literature methods where necessary
or used directly as obtained from the suppliers. Thin layer
chromatography (TLC) was performed on aluminium sheets
coated with Merck 5554 Kieselgel 60F. Developed plates were
scrutinized in an iodine chamber. Column chromatography
was performed using Kieselgel 60 (0.040–0.063 mm mesh,
Merck 9385). Melting points were determined with an
Electrothermal 9200 melting point apparatus and are uncor-
rected. Microanalyses were performed by the University of
Birmingham and the University of SheÃeld Microanalytical
Services. Low-resolution mass spectra were obtained from a
Kratos Profile mass spectrometer, operating at 4 kV and using
Fig. 12 Model of the complexation event between the system II
surfaces and the [Co(NH ) ]3+ trications, illustrating the 151 molar
3
6
7
0 eV for electron impact mass spectrometry (EIMS). Proton
ratio between the crown ether head groups and the [Co(NH ) ]3+,
3
6
resulting in a high [Co(NH ) ]3+ uptake coupled with the small
nuclear magnetic resonance (NMR) spectra were recorded on
Bruker AC 300 (300 MHz) spectrometer, using the deuteriated
3
6
binding constant, relative to system IV (Fig. 11)
1
152
J. Mater. Chem., 1997, 7(7), 1147–1154

solvents as the lock. 13C NMR spectra were recorded on the
Bruker AC 300 (75 MHz) spectrometer.
were removed in vacuo and the residue was taken up in CH Cl
2
2
(50 ml) and washed with aqueous Na CO (50 ml) and H O
2
3
2
The isotherm measurements were all recorded on a self-
made trough with a Wilhelmy pressure pick up system. The
(2×50 ml). The organic layer was then dried (MgSO ) and
4
the solvents were removed to aÂord a clear liquid which was
spreading solutions consisted of CHCl and the lipids 1–3, and
purified by silica gel column chromatography (eluent:
3
mixtures thereof, in the concentration range 0.5–1.0 mg ml−1,
CH Cl –MeOH) to aÂord 7 as a clear oil. Yield 9.6 g (42%);
2
2
of which between 25 and 50 ml were spread from a syringe on
1H NMR (300 MHz, CDCl )
d
3.70–3.50 (14H, m,
3
to an aqueous subphase or an aqueous subphase with the
OCH CH O), 3.45 (3H, s, OCH ), 2.81–2.78 (2H, t,
2
2
3
[
Co(NH ) ]Cl dissolved in it. Each isotherm was carried out
CH OCH ); 13C NMR (75 MHz, CDCl ) d 72.5, 71.8, 70.5,
3
6
3
2
3
3
over a 20 min period. The [Co(NH ) ]Cl was of analytical
70.5, 70.5, 70.2, 70.2, 61.6, 58.9. EIMS: C H O requires m/z
3
6
3
9 20 5
quality as defined by the commercial supplier and was used
208 [M+]. Found: m/z 209 [M+H]+.
without further purification. [Co(NH ) ]Cl subphases were
3
6
3
prepared freshly each day from Milli Q water (resistivity ca.
Lipid ether 2. A solution of monomethoxytetraethyleneglycol
1
8 V cm−1). The quartz crystal microbalance consisted of a
7 (2.0 g, 9.6 mmol) and NEt (1.2 g, 11.5 mmol) dissolved in
3
quartz crystal (9 MHz, AT-cut, d=8 mm) covered with gold
electrodes (area=0.53 cm2) obtained from Quartzkeramik
GmbH. The quartz crystal was mounted on a Teflon dipstick
dry toluene (25 ml) was added dropwise to a stirred solution
of octadecanoyl chloride (3.5 g, 11.5 mmol) in dry toluene
(10 ml) maintaining the temperature below 10 °C. The reaction
mixture was then allowed to warm to room temperature and
(
Fig. 4). In order to ensure no short circuits between the two
gold electrodes, a silicon sealing ring was placed between the
Teflon holder and the crystal. The QCM was held in place by
a vacuum on the non-covered face. The QCM was hydro-
phobised with a ‘silicon solution’ obtained from Serva. The
quartz crystal was driven by an in-house oscillator (15 V,
H O (20 ml) was added. The solution was then concentrated
2
in vacuo to give a waxy oÂ-white solid which was dissolved
CH Cl (50 ml) and washed with H O (50 ml×2). The organic
2
2
2
layer was dried (MgSO ), filtered, and the filtrate concentrated
4
in vacuo and purified by silica gel column chromatography
1
00 mA), the oscillation shape controlled by a Hameg (HM604)
(eluent: EtOAc–Me CO, 352) to yield the acyclic polyether 2
2
oscilloscope. The frequency change was recorded with an
Iwatsu universal counter (SC7201). Transfer of the Langmuir
films of systems I–IV to the QCM was achieved on a computer-
controlled trough from KSV Instruments (KSV5000) utilising
a vertical dipping method at 20 °C and a film pressure of 25
mN m−1 for systems I–IV and 50 mN m−1 for system V, with
a dipping speed of 2 mm min−1. The films were held at 25 or
as a white waxy solid. Yield 4.4 g (96%); 1H NMR (300 MHz,
CDCl ) d 4.24–4.19 (2H, m, CH CO ), 3.73–3.52 (14H, m,
3
2
2
OCH CH O), 3.46 (3H, s, OCH ), 2.32–2.28 (2H, t,
2
2
3
CH OCH ), 1.65–1.55 (2H, m, CH CH CO ), 1.33–1.20 (28,
2
3
2
2
2
s, CH CH ), 0.89–0.84 (3H, s, CH CH ); 13C NMR (75 MHz,
2
3
2
2
3
CDCl ) d 173.9, 72.6, 71.9, 70.6, 70.3, 69.2, 63.4, 61.7, 59.0, 34.2,
31.9, 29.5, 29.4, 29.1, 24.9, 22.7, 14.1. EIMS: C H O requires
2
7 54 6
5
0 mN m−1 for 20 min before deposition to ensure they were
m/z 474 [M+]. Found: m/z 475 [M+H]+. Anal. Calc.: C,
stable and were compressed with a barrier rate of 5 mm min−1.
68.31; H, 11.46. Found: C, 68.38; H, 11.61%.
System VI was chemisorbed on to the QCM clean gold surface
from a solution (0.1 m) of thiooctadecanol 4 in CHCl . The
Financial support by the Royal Society (J.A.P.) and by BNFL
3
gold surface was cleaned with MeOH and CHCl .
(S.I.) is gratefully acknowledged.
3
Synthesis
References
2
-Oxymethyl-18-crown-6-octadecanoate 1. A solution of 2-
1
2
G. Z. Sauerbrey, Phys., 1959, 155, 206.
hydroxymethyl-18-crown-6 5 (0.50 g, 1.70 mmol) and NEt
Applications of Piezoelectric Quartz Crystal Microbalances, ed.
C. Lu, Elsevier, New York, 1984, vol. 7; J. F. Alder and
J. J. McCallum, Analyst, 1983, 108, 1169; K. Bodenh o¨ fer,
A. Hierlemann, G. Noetzel, U. Weimar and W. G o¨ pel, Anal. Chem.,
3
(
0.260 g, 2.60 mmol) was dissolved in dry toluene (25 ml) and
it was added dropwise to a stirred solution of octadecanoyl
chloride (0.64 g, 2.10 mmol) in dry toluene (10 ml) whilst
maintaining the temperature at 10 °C. The reaction mixture
1
996, 68, 2210.
3
(a) R. Schumacher, Angew. Chem., Int. Ed. Engl., 1990, 29, 329;
was then allowed to warm to room temperature and H O
(
b) D. M. Ward and D. A. Buttry, Science (Washington, DC), 1990,
2
(
20 ml) was added. The solution was then concentrated in
249, 1000; (c) S. Bruckenstein and M. Shay, Electrochim. Acta,
1985, 30, 1295; (d) R. Schumacher, G. Borges and K. K. Kanazawa,
Surf. Sci., 1985, 163, L621; (e) O. Melroy, K. K. Kanazawa,
J. G. Gorgom and D. Buttry, L angmuir, 1986, 2, 697.
vacuo to give a yellow oil, which was dissolved in CH Cl
2
2
(
20 ml) and washed with H O (20 ml×2). The organic layer
2
was dried (MgSO ), concentrated in vacuo and purified by
4
4
5
(a) A. J. T u¨ d o¨ s, P. J. Vandeberg and D. C. Johnson, Anal. Chem.,
1995, 67, 552; (b) Y. Ebara, H. Ebato, K. Ariga and Y. Okhata,
L angmuir, 1994, 10, 2267; (c) A. C. Hillier and D. M. Ward, Anal.
Chem., 1992, 64, 2359; (d) D. A. Buttry and M. D. Ward, Chem.
Rev., 1992, 92, 1355.
silica gel column chromatography (eluent: EtOAc–Me CO,
2
3
52) to yield the amphiphilic crown ether 1 as a colourless oil.
Yield 0.30 g (32%); 1H NMR (300 MHz, CDCl ) d 4.28–4.10
3
(
2H, m, CO CH), 3.90–3.80 [1H, m, OCH(CH O) ], 3.78–3.60
2
2
2
(a) H. Ebato, C. A. Gentry, J. N. Herron, W. Muller, Y. Okahata,
H. Ringsdorf and P. Suci, Anal. Chem., 1994, 114, 8299;
(b) R. C. Ebersole and D. M. Ward, J. Am. Chem. Soc., 1988, 110,
(
4H, m, CH OCH ), 2.40–2.30 (2H, t, CH CO), 1.65–1.58 (2H,
2
2
2
m, CH CH CO), 1.30 [28H, s, M(CH ) CH ], 0.98–0.89 (3H,
2
2
2 14
3
t, CH ). EIMS C H O requires m/z 560 [M]+. Found: m/z
8
623; (c) H. Muramatsu, J. M. Dicks, E. Tamiya and I. Karube,
3
30 60 8
5
61 [M+H]+. Anal. Calc.: C, 66.58; H, 10.72. Found: C,
Anal. Chem., 1987, 59, 2760; (d) M. Thompson, C. L. Arthur and
G. K. Dhaliwal, Anal. Chem., 1986, 58, 1206.
6
6.67; H, 10.81%.
6
7
8
9
N. J. Geddes, E. M. Paschinger, D. N. Furlong, F. Caruso,
C. L. HoÂmann and J. F. Rabolt, Thin Solid Films, 1995, 260, 192.
D. E. Nivens, J. Q. Chambers, T. R. Anderson and D. C. White,
Anal. Chem., 1993, 65, 65.
Monomethoxytetraethyleneglycol 7. A solution of tetraethy-
leneglycol 6 (2 g, 0.11 mol) in dry THF was added dropwise
to a suspension of NaH (60% dispersed in mineral oil) (20 mg,
J. Redepenning, T. K. Schlesinger, E. J. Mechalke, D. A. Puleo and
R. Bizios, Anal. Chem., 1993, 65, 3378.
5
.4 mmol) in dry THF (50 ml) under nitrogen at room tempera-
ture. The reaction mixture was then heated under reflux and
stirred for a further 30 min. A solution of MeI (0.76 g,
(a) M. Masson, K. Yun, T. Haruyama, E. Kobatake and
M. Aizawa, Anal. Chem., 1995, 67, 2212; (b) M. Muratsugu,
F. Ohta, Y. Miya, T. Hosokawa, S. Kurosawa, N. Kamo and
H. Ikeda, Anal. Chem., 1993, 65, 2933; (c) Y. Ebara and Y. Okahata,
L angmuir, 1993, 9, 574; (d) R. C. Ebersole, J. A. Miller, J. R. Moran
and M. D. Ward, J. Am. Chem. Soc., 1990, 112, 3239; (e) E. Tamiya,
M. Suzuki and I. Karube, Anal. Chim. Acta, 1989, 217, 321.
5
2
.4 mmol) in THF (20 ml) was then added over a period of
0 min. The reaction mixture was heated under reflux for a
further 2 h. The reaction mixture was then allowed to cool to
room temperature and MeOH (10 ml) was added. The solvents
J. Mater. Chem., 1997, 7(7), 1147–1154
1153

1
1
0
1
D. E. Nivens, J. Q. Chambers, T. R. Anderson and D. C. White,
Anal. Chem., 1993, 65, 65.
26 A. Werner, Ann. Chem., 1912, 386, 1.
27 (a) M. Shimomura, S. Aiba, N. Tajima, N. Inoue and K. Okuyama,
L angmuir, 1995, 11, 969; (b) J. C. MacDonald and
G. M. Whitesides, Chem. Rev., 1994, 94, 2383.
F. Caruso, T. Serizawa, D. N. Furlong and Y. Okahata, L angmuir,
1
995, 11, 1546.
1
2
Y. Okahata and H. Ebato, J. Chem. Soc., Perkin T rans. 2, 1991, 457.
28 C. J. Pederson (Nobel Lecture), Angew. Chem., Int. Ed. Engl., 1988,
27, 1021.
1
3
Y. Okahata, X. Ye, A. Shimuzu and H. Ebato, Thin Solid Films,
1
989, 180, 51.
29 (a) J-M. Lehn, Supramolecular Chemistry, VCH, Weinheim, 1995;
Science, 1993, 260, 1762; (b) (Nobel Lecture), Angew. Chem., Int.
Ed. Engl., 1988, 27, 89; (c) H. Ringsdorf, Supramol. Sci., 1994, 1, 5;
(d) H. Kuhn, Supramol. Sci., 1994, 1, 75.
1
1
4
5
Y. Okahata, H. Ebato, T rends Anal. Chem., 1992, 11, 344;
Y. Okahata, G. En-na and H. Ebato, Anal. Chem., 1990, 62, 1431.
(a) S. Yamaguchi, T. Shimomura, T. Tatsuma and N. Oyama, Anal.
Chem., 1993, 65, 1925; (b) Y. Okahata, Y. Matsunobo, K. Ijiro,
M. Mukai, A. Murakami and K. Makino, J. Am. Chem. Soc., 1992,
30 (a) D. Philp, J. F. Stoddart, Angew. Chem., Int. Ed. Engl., 1996, 35,
1154; (b) J. A. Preece and J. F. Stoddart, Nanobiology, 1995, 3, 149;
(c) D. B. Amabilino and J. F. Stoddart, Pure Appl. Chem., 1993, 65,
2351; (d) D. Philp and J. F. Stoddart, Synlett, 1991, 445; (e) J. S.
Lindsey, New J. Chem., 1991, 15, 153.
1
14, 8299.
1
1
6
7
J. Wang, M. D. Ward, R. C. Ebersole and R. P. Foss, Anal. Chem.,
1
993, 65, 2553.
(a) D. Liu, K. Ge, K. Chen, L. Nie and S. Yao, Anal. Chim. Acta,
31 (a) T. Kunitake, Angew. Chem., Int. Ed. Engl., 1992, 31, 706; (b) J-
M. Lehn, Angew. Chem., Int. Ed. Engl., 1990, 29, 1304.
32 G. M. Whitesides, E. E. Simanek, J. P. Mathias, C. T. Seto,
D. N. Chin, M. Mammen and D. N. Gordon, Acc. Chem. Res.,
1995, 28, 37.
1
995, 307, 61; (b) Y. Okahata and Y. Ebara, J. Chem. Soc., Chem.
Commun., 1992, 116; G. G. Guilbault, Anal. Chem., 1983, 55, 1682.
S. Tanaka, Y. Iwasaki, K. Ishihara and N. Nakabayashi,
Macromol. Rapid Commun., 1994, 15, 319.
1
1
2
8
9
0
Y. Okahata, K. Yasunaga and K. Ogura, J. Chem. Soc., Chem.
Commun., 1994, 469.
33 (a) A. Ohki, J. P. Lu, J. L. Hallman, X. Huang and R. A. Bartsch,
Anal. Chem., 1995, 67, 2405; (b) S. Zaitsev, M. Belhoradsky,
(a) Y. Okahata, Y. Matsuzaki and K. Ijiro, Sens. Actuators B, 1993,
J. Zavad and D. Mobius, Thin Solid Films, 1994, 248, 78;
¨
1
3/14, 380; (b) Y. Okahata, K. Ijiro and Y. Matsuzaki, L angmuir,
(c) S. M u¨ noz, J. Mallen, A. Nakano, Z. Chen, I. Gay, L. Echegoyen
and G. W. Gokel, J. Am. Chem. Soc., 1993, 115, 1705;
(d) H. Matsumura, T. Watanabe, K. Furusawa, S. Inokuma and
T. Kuwamura, Bull. Chem. Soc. Jpn., 1987, 2747; (e) J. M. Gold,
D. T. Teegarden, K. M. McGrane, D. J. Luca, P. A. Falcigino,
C. C. Chen and T. W. Smith, J. Am. Chem. Soc., 1986, 108, 5827.
R. M. Izatt, J. S. Bradshaw, S. A. Nielson, J. D. Lamb and
J. J. Christiensen, Chem. Rev., 1985, 85, 271.
1
993, 9, 19.
2
2
1
2
R. C. Ebersole, R. P. Foss and M. D. Ward, Biotechnology, 1991,
9
, 450.
A notable example of a totally synthetic system is: K. D.
Schierbaum, T. Weiss, E. U. Thoden van Velzen, J. F. J. Engbersen,
D. N. Reinhoudt and W. G o¨ pel, Science, 1994, 265, 1413.
3
3
4
5
2
3
(a) D. R. Alston, J. F. Stoddart and D. J. Williams, J. Chem. Soc.,
Chem. Commun., 1985, 532; (b) D. R. Alston, A. M. Z. Slawin,
J. F. Stoddart and D. J. Williams, J. Chem. Soc., Chem. Commun.,
(a) G. C. Dunham, N. H. Benson, D. Petelenz and J. Janata, Anal.
Chem., 1995, 67, 267; (b) O-W. Lau, B. Shao and W. Zhang, Anal.
Chim. Acta, 1995, 312, 217; (c) M. I. Ivanchenko, H. Kobayashi,
E. A. Kulik and N. B. Dobrova, Anal. Chim. Acta, 1995, 314, 23;
1
985, 1602; (c) H. M. Colquhoun, D. F. Lewis, J. F. Stoddart and
D. J. Williams, J. Chem. Soc., Dalton Trans., 1983, 607; (d) H. M.
Colquhoun, J. F. Stoddart and D. J. Williams, J. Am. Chem. Soc.,
(
d) Z. Lin and M. D. Ward, Anal. Chem., 1995, 67, 685.
1
(
982, 104, 1426; J. Chem. Soc., Chem. Commun., 1981, (e) 851;
f ) 849; (g) 847.
16, 4497; (b) C. B. Aaker and K. R. Seddon, Chem. Soc. Rev., 1993,
97; (c) G. R. Desiraju, Acc. Chem. Res., 1991, 24, 290; (d) M. C.
3
6
7
D. J. Cram, Angew. Chem., Int. Ed. Engl., 1988, 27, 1009.
G. D. M. Beun, C. J. H. van de Velde and G. J. Fleuren, Immunol.
T oday, 1994, 15, 11.
3
2
4
5
(a) S. Paliwell, S. Geib and C. S. Wilcox, J. Am. Chem. Soc., 1994,
1
3
8
P. Berndt, K. Kurihara and T. Kunitake, L angmuir, 1995, 11, 3083;
T. M. Bohannon, S. Denzinger, R. Fink, H. Ringsdorf and
M. Weck, Angew. Chem., Int. Ed. Engl., 1995, 34, 48; R. C. Ahuja,
P-L. Caruso, D. M o¨ bius, W. Paulus, H. Ringsdorf and
G. Wildburg, Angew. Chem., Int. Ed. Engl., 1993, 32, 1033;
D. Y. Sasaki, K. Kurihara and T. Kunitake, J. Am. Chem. Soc.,
3
Etter, Acc. Chem. Res., 1990, 23, 120; (e) A. D. Hamilton, J. Chem.
Educ., 1990, 67, 821; ( f ) S. K. Burley and G. A. Petsko, Science,
1
985, 229, 23.
2
(a) F. M. Raymo and J. F. Stoddart, Chem. Ber., 1996, 129, 981;
(
b) J. F. Stoddart and R. Zarzycki, Recl. Trav. Chim. Pays-Bas,
1
991, 113, 9685; Y. Honda, K. Kurihara and T. Kunitake, Chem.
1
988, 107, 515; (c) H. M. Colquhoun, J. F. Stoddart and
L ett., 1991, 681.
D. J. Williams, Angew. Chem., Int. Ed. Engl., 1986, 25, 487;
(d) S. J. Loeb, in Comprehensive Supramolecular Chemistry, ed.
G. W. Gokel, Pergamon, 1996, vol. 1, pp. 733–753.
Paper 7/00325K; Received 14th January, 1997
1
154
J. Mater. Chem., 1997, 7(7), 1147–1154