Recognition of Cations by Self-Assembled Monolayers of Crown Ethers

Article abstract of DOI:10.1021/jp990014v

Self-assembled monolayers (SAMs) of crown ether adsorbates on gold reversibly bind metal ions from aqueous solutions. The resulting changes of the electrochemical properties of the monolayers were monitored by impedance spectroscopy. The increased dielect

Full text of DOI:10.1021/jp990014v

J. Phys. Chem. B 1999, 103, 6515-6520
6515
Recognition of Cations by Self-Assembled Monolayers of Crown Ethers
Simon Flink, Frank C. J. M. van Veggel,* and David N. Reinhoudt*
Department of Supramolecular Chemistry and Technology and MESA+ Research Institute,
UniVersity of Twente, P.O. Box 217, 7500 AE Enschede, The Netherlands
ReceiVed: January 4, 1999
Self-assembled monolayers (SAMs) of crown ether adsorbates on gold reversibly bind metal ions from aqueous
solutions. The resulting changes of the electrochemical properties of the monolayers were monitored by
impedance spectroscopy. The increased dielectric constant of the layer due to the complexation of ions results
in an increase of the monolayer capacitance (CML). Analysis of the response curves with a Langmuir isotherm
enables the determination of association constants of the SAMs with various metal ions. The cation binding
2+/3+
also influences the charge-transfer resistance (RCT) of a redox couple Ru(NH3)6
in the electrolyte.
Comparison of both responses allows an accurate interpretation of the origin of the resistive response.
Furthermore, the association constants enable the quantitative determination of interactions between SAMs
and metal ions, using either capacitive or resistive responses.
Introduction
resistance reveals the mechanism of the latter response. This
allowed the interpretation of the titrations in terms of adsorption
The modification of gold surfaces with self-assembled
monolayers has attracted much interest since the first publication
by Nuzzo and Allara.1 It is a simple and versatile method of
attaching molecules to a substrate in a densely packed arrange-
ment and thereby influencing the physical properties (e.g.,
surface free energy) of the exposed surface.2,3 Alternatively, the
introduction of functional groups in the adsorbates allows the
development of rapidly responding chemical sensors. Mono-
layer-based systems that are suitable for sensing devices should
combine at least two properties. First, the existence of a specific
and reversible interaction between guest and SAM which
enables the selective recognition. Second, the binding of a guest
to the monolayer has to be transduced into a detectable signal.
IR spectroscopy,4 mass sensitive devices,5 SPR,6 fluorescence
spectroscopy,7 and electrochemical measurements8 have suc-
cessfully been applied in the detection of host-guest interactions
at SAMs.
Electrochemical impedance spectroscopy (EIS) is a valuable
technique for the characterization of self-assembled monolayers
and the study of recognition processes. It has been used by
Rubinstein et al.9 and Vogel et al.10 to detect capacitance changes
upon the binding of metal ions to SAMs of ionophores.
Recently, we have demonstrated that EIS can also be used to
detect changes of the charge-transfer resistance, caused by the
binding of metal ions.11 It was shown that SAMs of 12-crown-4
and 15-crown-5 adsorbates form sandwich complexes, which
results in sodium selectivity for the 12-crown-4 SAMs and
potassium selectivity for the 15-crown-5 SAMs.
isotherms and gives the values for the association constants of
various metal ions with SAMs of 12-crown-4, 15-crown-5, and
18-crown-6 adsorbates (Chart 1).
Experimental Section
Chemicals. All chemicals for synthesis were obtained from
Aldrich and used as received. The syntheses of 2-(6-mercap-
tohexyloxy)methyl-15-crown-5 (3) and 2-(6-mercaptohexyloxy)-
methyl-12-crown-4 (4) have been described previously.11 All
reactions were conducted under an argon atmosphere. Electrolyte
solutions were freshly prepared from nitrogen-purged, high-
purity water (Millipore). The salts were obtained from Merck
and were of “pro analysis” purity. Ru(NH3)6Cl3 was purchased
from Alfa Products.
2-(p-Toluenesulfonyloxy)methyl-18-crown-6 (5). A solution
of 2-(hydroxymethyl)-18-crown-6 (1.00 g, 3.40 mmol), p-
toluenesulfonyl chloride (1.90 g, 9.97 mmol), and triethylamine
(1.00 g, 9.78 mmol) in dichloromethane (100 mL) was stirred
at room temperature for 16 h. Then, 1 M HCl (100 mL) was
added, and the aqueous layer was extracted with CH2Cl2 (3 ×
100 mL). The combined organic fractions were dried over
MgSO4 and filtered, and the solvent was evaporated to give
the crude product, which was purified by column chromatog-
raphy (Al2O3, eluent gradient CH2Cl2 to EtOAc) to yield 5 as
a colorless oil (1.44 g, 95%): 1H NMR (300 MHz, CDCl3) δ
2.45 (s, 3H), 3.50-3.76 (m, 22H), 3.79-3.88 (m, 1H), 4.04-
4.20 (m, 2H), 7.34 (d, J ) 8.5 Hz, 2H), 7.80 (d, J ) 8.5 Hz
2H); 13C NMR (75 MHz, CDCl3) δ 21.63, 69.78, 70.04, 70.13,
70.63, 70.69, 70.85, 70.98, 128.00, 129.76, 132.99, 144.68; FAB
MS m/z 471.1 ([M + Na]+, calcd for C20H32O9SNa 471.2).
In this paper, we present a detailed study of the binding of
alkali metal ions to self-assembled monolayers of several crown
ether adsorbates. SAMs of two different 18-crown-6 adsorbates
were used to study the capacitive detection of metal ions from
aqueous solutions. We have developed a model that quantita-
tively relates the magnitude of the capacitive response with the
structure of the adsorbates. Comparing the capacitive response
and the simultaneously determined change of the charge-transfer
2-(Mercaptomethyl)-18-crown-6 (1). A solution of 5 (1.24
g, 2.76 mmol) and thiourea (0.42 g, 5.53 mmol) in ethanol (50
mL) was heated under reflux for 16 h. The solvent was
evaporated under reduced pressure. Potassium hydroxide (0.31
g, 5.53 mmol) and nitrogen-purged water (50 mL) were added
to the residue, and the reaction mixture was heated under reflux
for 2 h. After the mixture was acidified with 1 M HClO4 (50
* Corresponding author. Fax: +31 53 4894645. Phone: +31 53
4892980. E-mail: D.N.Reinhoudt@ct.utwente.nl.
10.1021/jp990014v CCC: $18.00 © 1999 American Chemical Society
Published on Web 07/16/1999

6516 J. Phys. Chem. B, Vol. 103, No. 31, 1999
Flink et al.
CHART 1
high purity water. Caution: Piran˜a is a very strong oxidant and
reacts violently with many organic materials. The gold substrates
were cleaned in an oxygen plasma (5 min) and subsequently
immersed in ethanol for 10 min to remove the oxide layer.12
Formation of the self-assembled monolayer was achieved by
immersion of the gold substrate into a 1 mM solution of the
adsorbate in ethanol for 18 h. After the substrate was taken from
the solution, it was rinsed with ethanol (three times) and water
(two times) to remove any physisorbed material.
Instrumentation. 1H NMR and 13C NMR spectra were
recorded with a Brucker AC 250 or a Varian Unity Inova 300
spectrometer in CDCl3 using the traces of nondeuterated solvent
as an internal standard. FAB mass spectra were obtained with
a Finnigan MAT90 mass spectrometer using m-nitrobenzyl
alcohol (NBA) as a matrix. Contact angles were measured on
a Kru¨ss G10 contact angle measuring instrument, equipped with
a CCD camera. Advancing and receding contact angles were
determined automatically during growth and shrinkage of the
droplet by the drop shape analysis routine.
Electrochemical Measurements. Electrochemical measure-
ments were conducted with an Autolab PGSTAT10 using a
three-electrode cell containing a monolayer-covered gold work-
ing electrode (clamped to the bottom of the cell exposing a
geometric area of 0.44 cm2 to the electrolyte), a platinum counter
electrode, and a mercurous sulfate reference electrode (0.61
VSHE). After the cell was filled with the electrolyte solution,
nitrogen was bubbled through the solution for at least 3 min.
During the measurements, a constant flow of nitrogen was
maintained over the solution. Electrochemical measurements
without redox couple were performed in a background electro-
lyte of 0.1 M Et4NCl and titrated with solutions of metal
chlorides ([MCl] ) 0.1 M). The impedance spectra were
collected at a potential of -0.2 VMSE in a frequency range of
10 kHz to 0.1 Hz, with an ac amplitude of 5 mV. Electrochemi-
cal measurements in the presence of a redox couple were
performed in a background electrolyte of 0.1 M Et4NCl + 1
mM Ru(NH3)6Cl3 and titrated with solutions of 0.1 M MCl +
1 mM Ru(NH3)6Cl3 to ensure a constant concentration of the
background electrolyte and the redox couple. The impedance
mL), the aqueous layer was extracted with dichloromethane (3
× 100 mL). The combined fractions were dried over MgSO4
and filtered, and the solvent was evaporated to give the crude
product, which was purified by column chromatography (Al2O3,
eluent gradient EtOAc to EtOAc/EtOH 9/1) to yield 1 as a
colorless oil (0.25 g, 30%): 1H NMR (300 MHz, CDCl3) δ
2.64-2.80 (m, 2H), 3.50-3.57 (m, 1H), 3.58-3.92 (m, 22H);
13C NMR (75 MHz, CDCl3) δ 26.04, 69.81, 70.69, 70.87, 72.10,
80.08; FAB MS m/z 333.1 ([M + Na]+, calcd for C13H26O6-
SNa 333.2).
2-(6-Bromohexyloxy)methyl-18-crown-6 (6). A suspension
of 2-(hydroxymethyl)-18-crown-6 (2.11 g, 7.17 mmol) and
sodium hydride (55% in mineral oil; 0.5 g, 11.5 mmol) in 100
mL of DMF was stirred at room temperature. After 30 min,
1,6-dibromohexane (8.0 mL, 52.0 mmol) was added and the
reaction mixture was stirred overnight. Subsequently, the
reaction was quenched with methanol and the solvent was
evaporated under reduced pressure. The residue was taken up
in CH2Cl2 (100 mL) and washed with water (3 × 100 mL).
After drying over MgSO4, the solvent was evaporated and the
residue was purified by column chromatography (Al2O3, eluent
gradient hexane to EtOAc) to yield 6 as a colorless oil (1.70 g,
52%): 1H NMR (250 MHz, CDCl3) δ 1.28-1.50 (m, 4H),
1.50-1.65 (m, 2H), 1.80-1.92 (m, 2H), 3.35-3.50 (m, 6H),
3.55-3.85 (m, 23H); 13C NMR (62.5 MHz, CDCl3) δ 25.31,
27.97, 29.43, 32.73, 33.94, 69.93, 70.62, 70.64, 70.71, 70.74,
70.79, 70.83, 70.86, 70.88, 71.34, 71.77, 78.40; FAB MS m/z
457.2 ([M + H]+, calcd for C19H38O7Br 457.2), 479.2 ([M +
Na]+, calcd 479.2).
spectra were recorded at the formal redox potential of
2+/3+
Ru(NH3)6
(-0.56 VMSE), in a frequency range of 10 kHz
to 0.1 Hz, with an ac amplitude of 5 mV. The spectra were
analyzed using the software package “Equivalent Circuit”, which
uses a nonlinear least-squares fit to determine the parameters
of the elements in the equivalent circuit.13
Results and Discussion
2-(6-Mercaptohexyloxy)methyl-18-crown-6 (2). 2-(6-Bro-
mohexyloxy)methyl-18-crown-6 (6) was converted into the thiol
2 as described above for 1. The crude product was purified by
column chromatography (Al2O3, eluent gradient hexane to
EtOAc) to yield 2 as a colorless oil (1.25 g, 66%): 1H NMR
(250 MHz, CDCl3) δ 1.25-1.48 (m, 4H), 1.48-1.70 (m, 4H),
2.51 (dt, JA ) JB ) 7.3 Hz, 2H), 3.35-3.55 (m, 4H), 3.56-
3.85 (m, 23H); 13C NMR (62.5 MHz, CDCl3) δ 24.59, 25.59,
28.17, 29.50, 33.96, 69.95, 70.63, 70.65, 70.72, 70.76, 70.81,
70.83, 70.87, 70.89, 71.43, 71.80, 78.42; FAB MS m/z 433.2
([M + Na]+, calcd for C19H38O7SNa 433.2).
Gold Substrates. Gold substrates were prepared by resistive
evaporation of gold (200 nm) on glass slides of 25 mm diameter.
A layer of 2 nm of chromium was evaporated onto the glass
prior to the deposition of the gold layer in order to improve the
adhesion of the gold to the substrate.
Monolayer Characterization. Analysis of the monolayers
by grazing-incidence infrared spectroscopy is a very useful
technique to identify functional groups of the adsorbates. The
SAM of the hexyloxy-containing 18-crown-6 adsorbate 2
showed three absorptions due to methylene stretching vibrations.
These absorptions, centered at 2859, 2904, and 2930 cm-1
,
originate from the hexyl chain and the ethylene groups of the
crown ether ring.11 Due to the absence of the alkoxy spacer,
monolayers of adsorbate 1 lack the absorption at 2930 cm-1
,
which originates from asymmetric CH2 stretching vibrations of
alkyl chains. The remaining peak at 2859 cm-1 and its shoulder
around 2905 cm-1 originate from the symmetric and asymmetric
CH2 stretching vibrations of the crown ether ring, respectively.
Besides this, both monolayers show an intense absorption at
1139 cm-1, which is assigned to the C-O stretching mode of
the ether groups.14 The similar height of the C-O absorptions
(0.0033 and 0.0038 for SAMs of 1 and 2, respectively) suggests
similar densities of adsorbates for both SAMs.
Monolayer Preparation. All glassware used to prepare
monolayers was cleaned in boiling piran˜a (solution of 1:4 30%
H2O2 and concentrated H2SO4) and rinsed several times with

Recognition of Cations by Self-Assembled Monolayers
J. Phys. Chem. B, Vol. 103, No. 31, 1999 6517
Figure 2. Influence of potassium chloride in the electrolyte on the
monolayer capacitance for SAMs of adsorbates 1 (0) and 2 (]).
Figure 1. Capacitance plot of 18-crown-6 modified gold electrodes
(SAM of 1 (b); SAM of 2 (O)) at -0.2 VMSE in 0.1 M Et4NCl, with
C′′ ) 1/jωZ′′ and C′ ) 1/jωZ′. Fits to the spectra are indicated by the
solid lines. The inset shows the equivalent circuit used to fit the spectra.
It contains an electrolyte resistance (REL), a monolayer capacitance
(CML), and a constant-phase element (Q).
Wettability studies of monolayers of 18-crown-6 adsorbates
1 and 2 demonstrated that both surfaces have almost identical
hydrophobicities. SAMs of the longer adsorbate 2 have advanc-
ing and receding contact angles of 60° and 23°, respectively.
The short adsorbate 1 produces SAMs with the same advancing
contact angle, but the receding contact angle of 17° is slightly
smaller. This indicates that in both adsorbates the crown ether
is exposed to the outer interface, which will enable them to
interact with ions from solution.
Cation Binding to SAMs. The electrochemical impedance
spectra of self-assembled monolayers of crown ethers, performed
in a background electrolyte of 0.1 M Et4NCl, are presented in
Figure 1 as the complex capacitance plot.8b In this representation,
the real component of 1/jωZ is plotted versus its imaginary
component. The advantage of the capacitance plot over the more
commonly used Nyquist plot is the simple analysis of the
impedance spectra. The impedance spectra of gold electrodes
covered with a SAM of 1 and 2, respectively, are shown in
Figure 1. Analysis of the spectra showed that the system is best
described by an equivalent circuit consisting of the electrolyte
resistance REL in series with the monolayer capacitance CML
and a constant-phase element Q in parallel (see inset Figure 1).
The constant-phase element, which has values for n between
0.4 and 0.6, is a diffusion-like element that we attribute to the
diffusion of ions in and out of the monolayer under the influence
of the applied ac potential. Deviation from the ideal Warburg,
for which n ) 0.5, is attributed to the roughness of the electrode
surface.15
From the impedance spectra, we determined that monolayers
of the long 18-crown-6 adsorbate 2 have a capacitance of 4.2
( 0.1 µF cm-2, whereas SAMs of the short 18-crown-6
adsorbate 1 have a capacitance of 6.8 ( 0.1 µF cm-2. The
capacitance of a monolayer-covered electrode is inversely
proportional to the layer thickness as given in eq 1 (where ꢀ0 is
the permittivity of vacuum, ꢀr is the relative permittivity of the
monolayer, and d is the average layer thickness). This means
that SAMs of 2 have a larger average thickness than SAMs of
1, in agreement with their molecular structure.
Figure 3. Electrochemical model of modified gold electrodes with
monolayers of 18-crown-6 adsorbates 1 and 2.
The addition of aliquots of a 0.1 M KCl solution to the
background electrolyte of 0.1 M Et4NCl gives rise to an
increased CML in both cases (see Figure 2). We attribute this
effect to the binding of potassium ions to the monolayer. The
binding of ions to the initially neutral monolayer increases the
relative permittivity of the layer and results in the observed
response. Similar findings have previously been reported for
other monolayers that bind metal ions or protons.9,16
To account for the different response amplitudes of both
crown ether SAMs, we assume that the total capacitance of a
gold electrode covered with a monolayer of 2 can be divided
into Ccrown and Cchain, as illustrated in Figure 3. Ccrown represents
the part of the monolayer capacitance that changes due to the
binding of metal ions and the corresponding change of the
relative permittivity. Cchain is the part of the monolayer
capacitance that is not influenced by the binding of metal ions.
Consequently, the total monolayer capacitance CML of 2 can
be described by Ccrown and Cchain which are in series; thus, 1/CML
) 1/Ccrown + 1/Cchain. Binding of metal ions to the crown ether
monolayer, and the subsequent change of Ccrown, results in a
change of CML that is attenuated by the presence of the invariable
C
chain. The absence of the alkyl chain in the short crown ether
adsorbate 1 results in a monolayer capacitance of CML ) Ccrown
.
For this SAM, the change of Ccrown due to the binding of cat-
ions is equal to the detected change of CML and consequently
larger than that for the monolayer of the longer crown ether
adsorbate 2.
ꢀ0ꢀr
A more detailed examination of the data in Figure 2 reveals
that the capacitance of SAM 2 increases from 4.2 ( 0.1 µF
CML
)
(1)
d

6518 J. Phys. Chem. B, Vol. 103, No. 31, 1999
Flink et al.
Figure 4. Adsorption isotherm of potassium cations on crown ether
SAM 1. The surface coverage (Θ) of potassium complexes was
calculated from capacitance changes using eq 2 with Ccomplex ) 8.6 µF
cm-2 and Cligand ) 6.8 µF cm-2. The solid line is the fitted Langmuir
isotherm with K ) 10 400 M-1
.
cm-2 to a maximum value of 4.7 ( 0.1 µF cm-2. The monolayer
capacitance of the short adsorbate 1 increases from 6.8 ( 0.1
µF cm-2 to 8.6 ( 0.1 µF cm-2. From these data, the resulting
Figure 5. Resistive (]) and capacitive (0) response of crown ether
SAM 1 caused by the addition of aliquots 0.1 M KCl + 1 mM Ru-
(NH3)6Cl3 to the background electrolyte of 0.1 M Et4NCl + 1 mM
Ru(NH3)6Cl3 determined simultaneously from impedance measurements
at -0.56 VMSE. The inset shows the Randles equivalent circuit used to
fit the spectra. It contains an electrolyte resistance (REL), a monolayer
capacitance (CML), a charge-transfer resistance (RCT), and a diffusion
element (W).
value for the invariable capacitance was calculated as Cchain
)
10.8 ( 0.2 µF cm-2. The fact that this value is somewhat larger
than the reported value of 8.3 µF cm-2 for a heptanethiol SAM
in 1 M KCl17 is not very surprising, since the large size of the
18-crown-6 headgroup does not allow a close packing of the
alkyl chains such as in a monolayer of alkane thiol.
Changes in the monolayer capacitance (CML) can be used to
characterize the process of ion binding to the SAM. Under the
assumption that total monolayer capacitance is determined by
the fraction of surface covered by complexes and the rest of
the surface that is covered by free ligands, the system can be
described by the corresponding two capacitors (Ccomplex and
cations result in an increase of the charge-transfer resistance
(RCT), which can be measured by impedance spectroscopy.11
Under these conditions, the system is best described by the
Randles equivalent circuit (as shown in the inset of Figure 5).21
Apart from the charge-transfer resistance, also the monolayer
capacitance (CML) is a part of the equivalent circuit used.
Therefore, these two electrochemical properties of the monolayer
(RCT and CML) can be monitored simultaneously during the
titration experiments.
C
ligand) in parallel. Consequently, CML changes according to
eq 2,
CML ) ΘCcomplex + (1 - Θ)Cligand
(2)
The electrochemical properties of crown ether SAM 1 during
the titration of the background electrolyte (0.1 M Et4NCl and 1
mM Ru(NH3)6Cl3) with a solution of 0.1 M KCl and 1 mM
Ru(NH3)6Cl3 are shown in Figure 5. Changes of the monolayer
capacitance are identical to those determined in the absence of
the redox couple. This indicates that the complexation of the
where Θ is the fraction of occupied binding sites, Ccomplex is
the capacitance of the monolayer with metal ions bound to all
binding sites, and Cligand is the capacitance of the monolayer in
the absence of metal ions. Both Cligand and Ccomplex can be
obtained from the titration curves shown in Figure 2. The
resulting adsorption isotherm for the binding of potassium ions
to the monolayer of 1 is shown in Figure 4. We have focused
on the titration curves with SAMs of the short adsorbate 1, since
they have the largest capacitance changes and hence the smallest
relative error.18
To characterize the recognition process in the crown ether
SAM, the adsorption isotherm was fitted with the Langmuir,
Temkin, and Freundlich isotherms.19 Both Temkin and Freun-
dlich isotherms include terms that account for interactions
between bound guests. In contrast, the Langmuir isotherm
assumes equal binding energies for all binding sites. We found
that our titrations with SAMs of 1 are best described by a
Langmuir isotherm, implying that the cation binding sites are
well shielded from each other. From the fitted isotherms, the
association constants for Na+ and K+ were determined to be
1080 ( 90 and 10400 ( 800 M-1, respectively.20
2+/3+
metal ions is not influenced by the presence of Ru(NH3)6
or the more cathodic potential under which these measurements
are performed (-0.56 VMSE vs -0.2 VMSE). A comparison of
the capacitance and the resistance titration curves immediately
shows that both properties respond very differently to the
binding of cations. CML, which changes linearly with the fraction
of occupied binding sites Θ, shows the characteristics of a
Langmuir isotherm. In contrast to this, RCT is increasing linearly
with the KCl concentration of the electrolyte. The fact that both
measurements show different response curves can be qualita-
tively understood by the fact that the capacitance is changing
with the fraction of occupied binding sites (Θ), whereas the
charge-transfer resistance is probed with a redox couple that is
only able to penetrate the monolayer at vacant binding sites (1
- Θ). Since changes of CML and RCT are related to the same
event (the binding of metal ions), it is possible to establish the
exact relation between Θ and the change of RCT. The capacitive
response of the monolayer is linearly proportional to the
Langmuir isotherm,
An alternative method for the detection of metal ion binding
to the SAM uses a positively charged redox couple. Electrostatic
interactions between the redox couple and the monolayer-bound

Recognition of Cations by Self-Assembled Monolayers
J. Phys. Chem. B, Vol. 103, No. 31, 1999 6519
TABLE 1: Association Constants of Metal Ions with
Different Crown Ether SAMs Determined from Changes of
the Charge-Transfer Resistancea
Kc
CML Θ )
1 + Kc
(3)
where K is the association constant and c is the concentration
of metal ions in solution. Linearization of the Langmuir isotherm
gives
18-crown-6
18-crown-6 15-crown-5 12-crown-4
(2) (3) (4)
(1)
i+
KL
100
50 30 90
+
KNa
KK
1000 (1080)
11000 (10400)
4600
230
8500
1500
60
27100
770
15800
500
Θ
1 - Θ
) Kc
(4)
+
+
KCs
50
Consequently, RCT should be multiplied by (1 - Θ) in order to
obtain the desired relation between the charge-transfer resistance
and the fraction of occupied binding sites.
a Association constants determined from capacitance changes are
given in parentheses. The relative error in the association constants is
10%.
RCT(1 - Θ) ) R0
Combination of eqs 4 and 5 gives
(5)
influences the dielectric constant of the layer, resulting in an
increase of the monolayer capacitance. The magnitude of the
capacitive response is related to the chemical structure of the
used adsorbates. Longer adsorbates that have an alkyl spacer
to attach the ionophore to the gold substrate show a smaller
response due to the invariant capacitance of the spacer.
Comparison of capacitive and resistive changes of the crown
ether SAM caused by the binding of metal ions unraveled the
origin of latter response. Both responses can now be used to
determine the association constants of the monolayer with metal
ions. However, the magnitude of the resistance change and its
linear dependency on the metal concentration in the solution
enables its detection over a much wider range compared to the
capacitive detection.
R
Kc ) CT - 1
(6)
R0
where R0 is the charge-transfer resistance of the monolayer in
the absence of metal ions. The linear relation between RCT and
c in eq 6 increases the concentration window in which the metal
ion can be detected compared to the capacitive detection. The
monolayer capacitance is changing for metal ion concentrations
1
between /10Kass and 10/Kass, whereas the charge-transfer
1
resistance is responding at every concentration above /10Kass
.
However, we have found that at very high guest concentrations,
the resistance response is deviating from eq 6, until it eventually
reaches a maximum value.11 We attribute this to the fact that
RCT cannot increase to infinite values.
References and Notes
(1) Nuzzo, R. G.; Allara, D. L. J. Am. Chem. Soc. 1983, 105, 4481.
(2) Ulman, A. An Introduction into Ultrathin Organic Films; Academic
Press: San Diego, CA, 1991. (b) Finklea, H. O. In Electroanalytical
Chemistry; Bard, A. J., Rubinstein, I., Eds.; Marcel Dekker: New York,
1996; Vol. 19. (c) Hickman, J. J.; Ofer, D.; Laibinis, P. E.; Whitesides, G.
M.; Wrighton, M. S. Science 1991, 252, 688. (d) Roscoe, S. B.; Kakkar,
A. K.; Marks, T. J.; Malik, A.; Durbin, M. K.; Lin, W.; Wong, G. K.;
Dutta, P. Langmuir 1996, 12, 4218. (e) Yang, X.; McBranch, D.; Swanson,
B.; Li, D. Angew. Chem., Int. Ed. Engl. 1996, 35, 538. (f) Doron, A.;
Portnoy, M.; Lion-Dagan, M.; Katz, E.; Willner, I. J. Am. Chem. Soc. 1996,
118, 8937.
(3) Thoden van Velzen, E. U.; Engbersen, J. F. J.; de Lange, P. J.;
Mahy, J. W. G.; Reinhoudt, D. N. J. Am. Chem. Soc. 1995, 117, 6853. (b)
Huisman, B.-H.; Thoden van Velzen, E. U.; van Veggel, F. C. J. M.;
Engbersen, J. F. J.; Reinhoudt, D. N. Tetrahedron Lett. 1995, 36, 3273.
(c) Beulen, M. W. J.; Huisman, B.-H.; van der Heijden, P. A.; van Veggel,
F. C. J. M.; Simons, M. G.; Biemond, E. M. E. F.; de Lange, P. J.;
Reinhoudt, D. N. Langmuir 1996, 12, 6170. (d) Scho¨nherr, H.; Vancso, G.
J.; Huisman, B.-H.; van Veggel, F. C. J. M.; Reinhoudt, D. N. Langmuir
1997, 13, 1567.
(4) Sun, L.; Kepley, L. J.; Crooks, R. M. Langmuir 1992, 8, 2101. (b)
Xu, C.; Sun, L.; Kepley, L. J.; Crooks, R. M. Anal. Chem. 1993, 65, 2102.
(5) Sun, L.; Crooks, R. M.; Ricco, A. J. Langmuir 1993, 9, 1775. (b)
Roush, J. A.; Tacker, D. L.; Anderson, M. R. Langmuir 1994, 10, 1642.
(c) Schierbaum, K.-D.; Weiss, T.; Thoden van Velzen, E. U.; Engbersen,
J. F. J.; Reinhoudt, D. N.; Go¨pel, W. Science 1994, 265, 1413. (d) Dermody,
D. L.; Crooks, R. M.; Kim, T. J. Am. Chem. Soc. 1996, 118, 11912. (e)
Sastry, M.; Patil, V.; Mayya, K. S. J. Phys. Chem. B 1997, 101, 1167.
(6) Spinke, J.; Liley, M.; Schmitt, F.-J.; Guder, H.-J.; Angermaier, L.;
Knoll, W. J. Chem. Phys. 1993, 99, 7012. (b) Simpson, T. R. E.; Cook, M.
J.; Petty, M. C.; Thorpe, S. C.; Russel, D. A. Analyst 1996, 121, 1501. (c)
Huisman, B.-H.; Kooyman, R. P. H.; van Veggel, F. C. J. M.; Reinhoudt,
D. N. AdV. Mater. 1996, 8, 561. (d) Friggeri, A.; van Veggel, F. C. J. M.;
Reinhoudt, D. N.; Kooyman, R. P. H. Langmuir 1998, 14, 5457.
(7) Motesharei, K.; Myles, D. C. J. Am. Chem. Soc. 1998, 120, 7328.
(8) Rubinstein, I.; Steinberg, S.; Tor, Y.; Shanzer, A.; Sagiv, J. Nature
1988, 332, 426. (b) Steinberg, S.; Tor, Y.; Sabatani, E.; Rubinstein, I. J.
Am. Chem. Soc. 1991, 113, 5176. (c) Steinberg, S.; Rubinstein, I. Langmuir
1992, 8, 1183. (d) Turyan, I.; Mandler, D. Anal. Chem. 1994, 66, 58. (e)
Rojas, M. T.; Ko¨niger, R.; Stoddart, J. F.; Kaifer, A. E. J. Am. Chem. Soc.
1995, 117, 336. (f) Turyan, I.; Mandler, D. Anal. Chem. 1997, 69, 894.
(9) Gafni, Y.; Weizman, H.; Libman, J.; Shanzer, A.; Rubinstein, I.
Chem.sEur. J. 1996, 2, 759.
The experimentally derived relation between the association
constant and the charge-transfer resistance of the monolayer as
given in eq 6 was used to calculate the association constants of
alkali metal ions with several crown ether monolayers (see Table
1). Since the titration curves deviate from linearity at very high
metal concentrations, only the concentration range where eq 6
is valid was used for the determination of the association
constants. A striking feature of the determined association
constants is that they are higher than those reported for the
complexes in aqueous solutions.22 A much better resemblance
is found to the corresponding association constants of the
complexes in less polar solvents such as methanol. This indicates
that the environment inside the monolayer has a lower polarity
than the contacting aqueous solution.23 In the two cases where
the association constants were determined by capacitive and
resistive measurements, the obtained values are identical within
the experimental error, which confirms the validity of eq 6.
Besides this, both 18-crown-6 adsorbates form monolayers that
+
+
+
+
exhibit similar selectivities: KK > KCs > KNa > KLi . Also,
monolayers of the 15-crown-5 adsorbate have the highest affinity
for potassium ions. The other alkali metal ions have association
constants that are almost 2 orders of magnitude smaller.
Monolayers of adsorbates with the smallest crown ether ring
exhibit sodium selectivity. These remarkable selectivities of the
+
+
+
+
12-crown-4 (KNa /KK ) 30) and 15-crown-5 (KK /KNa ) 450)
SAMs have been attributed to the formation of sandwich
complexes.11 The fact that in these cases the metal ions are
bound in a sandwich complex does not contradict the require-
ments of the Langmuir isotherm (i.e., independence of binding
sites), since the two crown ethers involved in the complexation
can be regarded as being one binding site.
Conclusions
Self-assembled monolayers of crown ether adsorbates are able
to bind cations from aqueous solutions. The binding of ions
(10) Stora, T.; Hovius, R.; Dienes, Z.; Pachoud, M.; Vogel, H. Langmuir
1997, 13, 5211.

6520 J. Phys. Chem. B, Vol. 103, No. 31, 1999
Flink et al.
(11) Flink, S.; Boukamp, B. A.; van den Berg, A.; van Veggel, F. C. J.
M.; Reinhoudt, D. N. J. Am. Chem. Soc. 1998, 120, 4652.
(12) Ron, H.; Rubinstein, I. Langmuir 1994, 10, 4566.
(13) Boukamp, B. A. EquiValent Circuit, version 4.55; University of
Twente, Department Chemical Technology: Enschede, The Netherlands,
1996. (b) Boukamp, B. A. Solid State Ionics 1986, 18-19, 136. (c)
Boukamp, B. A. Solid State Ionics 1986, 20, 31.
(18) The relative error in the capacitance measurements is 0.1 µF cm-2
.
Therefore, the capacitive responses of the SAMs 2-4 that showed maximal
changes of 0.5 µF cm-2 were not fitted to isotherms.
(19) Adamson, A. W. Physical Chemistry of Surfaces; Wiley-Inter-
science: New York, 1990.
(20) Isotherms were fitted with Table Curve, version 1.0, by Jandel
Scientific.
(14) Bruening, M. L.; Zhou, Y.; Aguilar, G.; Agee, R.; Bergbreiter, D.
E.; Crooks, R. M. Langmuir 1997, 13, 770.
(15) Impedance Spectroscopy: Emphasizing Solid Materials and Sys-
tems; Macdonald, J. R., Ed.; John Wiley & Sons: New York, 1987.
(16) Bryant, M. A.; Crooks, R. M. Langmuir 1993, 9, 385.
(17) Porter, M. D.; Bright, T. B.; Allara, D. L.; Chidsey, C. E. D. J.
Am. Chem. Soc. 1987, 109, 3559.
(21) Sluyters-Rehbach, M.; Sluyters, J. H. In Electroanalytical Chem-
istry; Bard, A. J., Ed.; Marcel Dekker: New York, 1970; Vol. 4.
(22) Izatt, R. M.; Pawlak, K.; Bradshaw, J. S.; Bruening, R. L. Chem
ReV. 1991, 91, 1721.
(23) Similar findings have been reported for the binding of ferrocene
to cyclodextrin monolayers; see: Rojas, M. T.; Ko¨niger, R.; Stoddart, J.
F.; Kaifer, A. E. J. Am. Chem. Soc. 1995, 117, 336.