A generic platform for the addressable functionalisation of electrode surfaces through self-induced "electroclick"

Article abstract of DOI:10.1002/chem.201102620

A novel and general strategy for the immobilisation of functional objects onto electrodes is described. The concept is based on the addition of two pendant ethynyl groups onto a bis(pyridyl)amine derivative, which acts as a molecular platform. This platfo

Full text of DOI:10.1002/chem.201102620

DOI: 10.1002/chem.201102620
A Generic Platform for the Addressable Functionalisation of Electrode
Surfaces through Self-Induced “Electroclick”
Christophe Orain,[a] Nicolas Le Poul,[a] Antoine Gomila,[a] Jean-Michel Kerbaol,[a]
Nathalie Cosquer,[a] Olivia Reinaud,[b] FranÅoise Conan,[a] and Yves Le Mest*[a]
Abstract: A novel and general strategy
for the immobilisation of functional ob-
jects onto electrodes is described. The
concept is based on the addition of two
pendant ethynyl groups onto a bis-
thiol self-assembled monolayers on a
gold electrode. Two different functional
copper complexes indicated that a
good surface coverage was achieved
and that a moderately fast electron-
transfer reaction occurs. Remarkably,
in the case of the redox-active ferro-
cenyl-immobilised system, the electro-
chemical response highlighted the in-
volvement of the copper ion of the
platform in the kinetics of the electron
transfer to the ferrocene moiety. This
platform is a promising candidate for
applications in surface addressing in
areas as diverse as biology and materi-
als.
ꢀ
groups, a redox innocent ((CH2)3 Ph)
and an electrochemical probe (ferro-
cene), were immobilised by following
this strategy. The in situ electrochemi-
cal grafting showed, for both systems,
that the kinetics of immobilisation is
fast. The voltammetric characterisation
of the surface-tagged functionalised
ACHTUNGTRENNUNG(pyridyl)amine derivative, which acts
as a molecular platform. This platform
is pre-functionalised with an N3-tagged
object of interest by Huisgen cycload-
dition to one of the ethynyl groups in
biphasic conditions. Hence, when com-
plexed by CuII, this molecular-object
holder can be immobilised, by a “self-
induced electroclick”, through the
second ethynyl group onto N3-alkane-
Keywords: click chemistry · copper ·
electrode functionalisation · N li-
gands · self-assembly
Introduction
ties.[6,11] However, the main drawback remains the lack of
versatility. Indeed, the incorporation of a thiolated arm on
to the targeted object is a prerequisite for direct grafting;
this requires considerable synthetic effort and there is no
guarantee that the molecules will form a structurally well-
defined monolayer. Also, the immobilisation is hampered
when the objects are reactive towards thiol groups, which is
the case for many metallic complexes. These problems are
conveniently overcome by pre-functionalisation of the elec-
trode with reactive groups. The grafting step then operates
by a chemical reaction leading to covalent binding.[6] The
immobilisation of several objects, mainly of biological inter-
est (DNA,[12] peptides,[13] carbohydrates[14]), has been suc-
cessfully achieved by following this strategy.[15] However,
there were some limitations as a result of the experimental
conditions imposed by the chemical reaction (nucleophilic
substitution, esterification, acylation, nucleophilic addition).
A remarkable improvement was achieved by Collman and
Chidsey and their co-workers through the adaptation of the
copper-catalysed azide-alkyne cycloaddition (CuAAC
“click” reaction)[16] to surface immobilisation.[17,18] Indeed,
the mild conditions (room temperature) and high selectivity
(1,3-cycloaddition) of this reaction are very well adapted to
the modification of gold electrodes. This approach led to the
surface immobilisation of several functional systems with
specific properties.[19,20] Recently, Larsen and co-workers de-
veloped a procedure to selectively functionalise conducting
polymers with two different alkyne-substituted fluroro-
phores by electrochemical reduction of a CuII catalyst: the
The chemical functionalisation of conductive surfaces by
specific molecular or biological objects is currently attract-
ing a great deal of attention for the development of efficient
devices in physical,[1] chemical[2] and biological[3] domains.
Among the different strategies for immobilisation (polymer-
ic,[4] layer-by-layer[5]), the self-assembly of monolayers
(SAMs) by thiolate chemisorption on metallic surfaces is
undoubtedly the most popular because it offers the advant-
age of yielding well-organised and controlled modified sur-
faces relatively easily.[6] Beyond the fundamental aspects of
electron-transfer kinetics and surface self-organisation,[7–10]
this approach has led to the direct immobilisation of a vast
number of terminal active groups with specific proper-
[a] C. Orain, Dr. N. Le Poul, Dr. A. Gomila, J.-M. Kerbaol,
Dr. N. Cosquer, Prof. F. Conan, Dr. Y. Le Mest
Laboratoire de Chimie, Electrochimie Molꢀculaire
et Chimie Analytique, UMR CNRS 6521
Universitꢀ de Bretagne Occidentale
6 Avenue Le Gorgeu, 29238 Brest Cedex 03 (France)
Fax : (+33)02-98-01-70-01
E-mail: yves.lemest@univ-brest.fr
[b] Prof. O. Reinaud
Laboratoire de Chimie et de Biochimie Pharmacologique
et Toxicologiques, UMR CNRS 8601
Universitꢀ Paris Descartes, Sorbonne Paris Citꢀ
45 Rue des Saint-Pꢁres, 75006 Paris (France)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201102620.
594
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 594 – 602

FULL PAPER
“electroclick” reaction.[21] This allows spatial control of the
micro-patterning of their objects. In all these examples a CuI
species in solution is necessary for a heterogeneous click re-
action. We recently demonstrated that electroclick onto N3-
terminated SAMs on gold could be achieved in the absence
of any added copper catalyst in solution when applied to a
Results and Discussion
Synthesis of the pre-functionalised platform: The platform 3
(6-TIPSe-BMPPA) was synthesised in two steps from 6-Br-
BMPA (1) (6-Br-BMPA=(6-bromo-2-pyridylmethyl)(2-pyri-
dylmethyl)amine; Scheme 3). In the first step, Sonogashira
specific copper complex with an ethynyl arm: [CuACHTUNTRGNEUNG(6-
eTMPA)] (Scheme 1; 6eTMPA=6-ethynyltris(2-pyridyl-
Scheme 3. Schematic pathway for the synthesis of the platform 3:
1) (iPr)3SiCCH (2 equiv), CuI (10%mol), [PdACHTNUGRTNE(UGN PPh3)2Cl2] (6%mol),
iPr2NH/THF (1:2); 2) propargyl bromide (1.1 equiv), DIPEA (3 equiv),
THF, 808C.
Scheme 1. Schematic representation of CuII complexes of 6-eTMPA (left)
and the pre-functionalised platform (right).
ACHTUNGTRENNUNG
methyl)amine).[22] In our report on that work we emphasised
the concept of a self-induced electroclick reaction, which
allows the direct grafting of the Cu-TMPA complex in a
simple electrochemical step. In this work we show that this
concept can be extended to the surface immobilisation of di-
verse objects (R in Schemes 1 and 2). The proposed strategy
is based on a two-step click modification of a so-called mo-
lecular platform in which one of the pyridine units in the 6-
eTMPA ligand is replaced by an alkyne arm. The latter
allows a click reaction with an N3-tagged object R, the re-
sulting copper complex is designed to retain the self-induced
electroclick property. As shown in Scheme 2, the two-step
procedure involves 1) pre-functionalisation of the platform
by the object R through a chemical click reaction and 2) im-
mobilisation of the pre-functionalised platform on an N3-al-
kanethiol-modified gold electrode by a self-induced electro-
chemically driven click reaction in water.
coupling allowed the insertion of a TIPS-protected (TIPS=
triisopropylsilyl) alkyne group at the 6-position of a pyridyl
group to give 6-TIPSe-BMPA (2; TIPSe=triisopropylsilyl-
AHCTUNGTREGeNNNU thynyl) in a yield of 74%. The second reaction involved a
nucleophilic substitution in basic medium of the bromide
group of propargyl bromide by the activated anionic form of
compound 2, which led to platform 3 (6-TIPSe-BMPPA;
BMPPA=bis(2-pyridylmethyl)propargylamine) in a yield of
65%. The protection of the ethynyl arm on the pyridyl
group by a TIPS group prior to the nucleophilic substitution
was necessary to avoid a double-click reaction on both eth-
ynyl groups.
Platform 3 was then pre-functionalised by a chemical
CuAAC click reaction[16] under biphasic conditions (see
below) to optimise product separation (Scheme 4). Thus, the
organic phase (dichloromethane) containing the organo-
ꢀ
soluble compound 3 and the azido-substituted R N3 reac-
tant was stirred in an aqueous
solution of CuSO4 (10% mol
equiv) and sodium ascorbate
(20% mol equiv) at room tem-
perature for 48 h. The interfa-
cial reaction resulted in the for-
mation of the R-triazolo-substi-
tuted compounds 4 and 5 (R=
ꢀ
ꢀ
AHCTNUGTREN(GUNN CH2)3 Ph and CH2 Fc, respec-
tively) in relatively good yields
(>50%). These compounds
were then treated with NBu4F
to deprotect the ethynyl arm on
the pyridyl group in prepara-
tion for the electroclick reac-
tion, which yielded compounds
6
and
7,
respectively
(Scheme 4).
Complexation of the pre-func-
tionalised platform: The coordi-
nation properties of the N4 core
Scheme 2. Illustration of the two-step approach for the grafting of functional object R onto N3-alkanethiol
SAMs on a gold electrode.
Chem. Eur. J. 2012, 18, 594 – 602
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
595

Y. Le Mest et al.
Scheme 4. Pre-functionalisation of platform 3 before the electroclick re-
ꢀ
action: 1) R N3 (1.1 equiv), CuSO4 (10%mol), Na Ascorbate
(20%mol), CH2Cl2/H2O (1:1); 2) NBu4F (3 equiv), THF. Fc=ferrocene.
Figure 1. Cyclic voltammograms of A) [Cu(6)] and B) [Cu(7)] at a plati-
num electrode in CH3CN/NBu4PF6 (0.1m) under N2 (v=0.02–2 Vsꢀ1).
of the unprotected pre-functionalised platforms 6 and 7
were then explored with CuII and ZnII ions. The ZnII com-
plexation of compound 7 is characterised by a 1H NMR
downfield shift of Ddffi0.5 ppm for most of its signals, which
can be ascribed to the Lewis acidity and positive charge of
the metal ion (Figure S1 in the Supporting Information).
[Zn(7)] and [Cu(7)] in comparison with free ligand 7. This
effect is likely due to electrostatic forces (charge–charge re-
pulsion between the two metal centres in their oxidised
states) and to the Lewis acidity of the zinc or copper ion,
which reduces the electron density on the ferrocenyl group,
as evidenced by NMR studies. In H2O/KNO3 (0.1m), the
CV response is different. With [Cu(6)] (Figure S3 in the
Supporting Information), a broad reduction peak is associat-
ed with a broad reoxidation peak with a significant peak
separation (DEpffi600 mV at v=0.1 Vsꢀ1). Such behaviour is
characteristic of a large reorganisation of the complex as a
result of the electron-transfer reaction. In this medium, a
water molecule coordinated to the CuII centre could rapidly
decoordinate during the reduction to the CuI state at around
Epc =ꢀ0.15 V to produce a stable tetracoordinated cuprous
species that is oxidised at around Epa =0.45 V versus SCE.
Alternatively, a triazole or pyridine unit may undergo de-
coordination in the CuI oxidation state. In both cases, the
CuI would be less easy to reoxidise. With [Cu(7)] (Figure S3
in the Supporting Information), the situation is more com-
plicated because the sluggish oxidation peak of the electro-
generated CuI species lies in the same potential range as the
ferrocene oxidation (Epa =0.45 V for CuI vs. E8=0.37 V for
Fc+/Fc). Hence, this situation results in an intricate elec-
tron-transfer mechanism involving two almost concomitant
oxidations with a possible intramolecular transfer and yields
a composite peak.[24]
The complexation of ligands 6 and 7 by CuII
ACTHNUGTRNE(UNG OTf)2 yielded
mononuclear complexes, which were characterised by elec-
trochemical and spectroscopic means (Table 1).
The EPR spectra of these complexes in CH3CN are both
typical of CuII complexes with a square-pyramidal geometry
(g// >g? ), as has already been observed for [CuII
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
solvent, the [Cu(6)] complex is characterised by cyclic vol-
tammetry (CV) by a single reversible system with E8=
ꢀ0.17 V versus Fc+/Fc at
a moderate scan rate (v=
0.1 Vsꢀ1), which corresponds to the CuII/CuI redox reaction
(Figure 1A). For [Cu(7)], two reversible systems were ob-
served at E8=ꢀ0.18 and 0.14 V (Figure 1B). In rotating disk
electrode voltammetry, these two systems are characterised
by two successive waves of equal intensity, which indicates
two one-electron exchanges. Comparison of the data for
[Cu(6)], [Zn(7)] and 7 (see Table 1) shows that the process
at E8=ꢀ0.17 V can undoubtedly be ascribed to the copper
centre and the one at E8=0.14 V to the ferrocene moiety.
Interestingly, the CuII/I reaction occurs at a quasi-identical
E8 value for the [Cu(6)] and [Cu(7)] complexes. This indi-
cates that the methylferrocenyl and phenylpropyl substitu-
ents have almost the same electronic influence on the metal
ion. Another remarkable point is the positive shift of E8 for
Table 1. EPR, Vis/NIR and electrochemical data for compounds 7, [CuII(6)], [CuII(7)], [CuII(TMPA)] and [CuII
ACTHNUTRGENNUG ACHUTGNTREN(NUGN 6-eTMPA)].
Compound
Solvent
g (A [G])[a]
lmax [nm] (e [mꢀ1 cmꢀ1])
E8 [V] (DEp [mV])
CuII/CuI
Fc+/Fc
7
A
CH3CN
CH3CN
H2O
CH3CN
H2O
CH3CN
H2O
CH3CN
H2O
–
–
437 (167)
402 (126), 718 (70)
–
–
0.10 (90)[b]
0.16 (95)[b]
0.34 (80)[c]
–
[ZnII(7)]
N
g// =2.23 (112), g? =2.06
g// =2.24 (139), g? =2.06
g// =2.01 (65), g? =2.19 (113)
g// =2.24 (162), g? =2.05
652 (112)
ꢀ0.17 (90)[b]
E
E
pc =ꢀ0.15,[c]
E
pa =0.45[c]
N
437 (196), 638 (103)
889 (238), 630 (sh)[d]
809 (152), 640 (104)
ꢀ0.18 (115)[d]
0.14 (90)[b]
0.37 (130)[c]
–
pc =ꢀ0.10[c]
T
A
ꢀ0.40 (65)[b]
ꢀ0.36 (60)[c]
ꢀ0.33 (65)[b]
ꢀ0.16 (120)[c]
T
G
CH3CN
H2O
–
[a] At T=150 K. [b] In CH3CN/NBu4PF6 (0.1m), E vs. Fc+/Fc. [c] In H2O/KNO3 (0.1m), E8 vs. SCE. [d] sh=shoulder. [e] From ref. [22].
596
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 594 – 602

Functionalisation of Electrode Surfaces
FULL PAPER
Surface immobilisation: The [CuII(6)] and [CuII(7)] com-
plexes were grafted onto an azidoalkanethiol-modified gold
electrode by the self-induced electroclick procedure.[22] The
complexes were grafted by cycling between 0.60 and
ꢀ0.20 V with a 3 min hold at ꢀ0.20 V between each cycle in
a 50 mm solution of [Cu(6)] or [Cu(7)] in H2O with KNO3 as
the supporting electrolyte under an atmosphere of N2. As
shown in Figure 2A for the [Cu(7)] complex, the increase in
number of available azido sites have reacted. The shapes of
the voltammetric curves, which deviate from the “ideal” be-
haviour (sharp oxidation peak for ferrocene, more intense
than for copper), indicates that, as observed by CV in H2O,
the grafting occurs by complex mechanisms.[25] Plots of the
cathodic peak intensities versus time for the ferrocenium re-
duction process during the immobilisation of [Cu(7)] al-
lowed the kinetics of triazole formation on the surface to be
estimated.
As a result of the maximum surface coverage calculated
from peak integration after grafting (which corresponds to
the maximum number of available N3 sites for the non-
mixed modified electrode, see below), the ratio of “un-
clicked” available azido sites was determined for each
cycle.[26] As shown in Figure 2B, the resulting decay plots
display a shape that fits reasonably well with a first-order ki-
netic rate law (dashed line). As the reaction is bimolecular,
a second-order rate constant (k) for surface immobilisation
based on the Fc+ reduction peak was calculated on the as-
sumption that the ethynyl complex is in excess over the
azido groups (k=160mꢀ1 sꢀ1). The same procedure was ap-
plied to the decrease in the peak intensity of the CuII reduc-
tion peak at ꢀ0.15 V (Figure 2, insert), taking the maximum
surface coverage from peak integration after grafting (see
Table 2 and Figure 3B). The good match between the
second-order rate constants determined from the Fc+ (k=
160mꢀ1 sꢀ1) and CuII (k=180mꢀ1 sꢀ1) reduction peaks illus-
trates the validity of the process and reflects the concomi-
tant attachment of both redox probes on to the electrode
surface. The same calculation applied to [Cu(6)] and [CuACHTUNGTRENNUNG(6-
eTMPA)] led to similar results, which attests to a fast sur-
face reaction for these complexes with kffi100mꢀ1 sꢀ1. Such
values are, however, 10-fold lower than that found for the
ferrocene-ethynyl compound described previously (k=
1000mꢀ1 sꢀ1),[17c] probably for steric reasons.
Figure 2. A) Cyclic voltammograms (10 cycles) recorded at an N3-
(CH2)11-SH-modified gold electrode (v=0.1 Vsꢀ1) of [Cu(7)] (50 mm) in
H2O/KNO3 (0.1m) under N2 with a 2 min hold at ꢀ0.20 V (Insert: varia-
tion in the cathodic current of the copper system with cycling ). B) Plots
of the number of available azide molecules per cm2 vs. time for the
Surface electrochemical characterisation: After thorough
washing with distilled water, the [Cu(6)]- and [Cu(7)]-modi-
fied gold electrodes were studied in an electroactive-free
sodium-acetate-buffered solution (pH 4.5; Table 2). The
~
copper ( ) and ferrocene (&) systems from the variation of cathodic
peak intensities with time. Dashed line: fitted exponential curve.
the cathodic current intensity at +0.45 V (Fc) and ꢀ0.10 V
(Cu) with cycling indicates that effective immobilisation of
the complex has occurred on the electrode surface. After
around 10 cycles, the intensity of the cathodic peaks reached
[Cu(6)]-modified electrode showed a single reversible
system at E8=0.04 V versus SCE (Figure 3A) with DEp =
40 mV at v=0.05 Vsꢀ1 and peak intensities linearly propor-
tional to the scan rate (not shown), which is indicative of an
immobilised redox centre. The [Cu(7)]-modified electrode
a
maximum value, which suggests that the maximum
Table 2. Electrochemical data for immobilised complexes on the modified gold electrode in H2O/NaBF4 (0.05m)/sodium acetate (0.05m) at pH 4.5.
[a]
Grafted complex
E8 [V] vs. SCE
G [pmolcmꢀ2
]
k8 [sꢀ1
]
DEp,1/2 [mV] at 0.05 Vsꢀ1
(DEp [mV] at 0.05 Vsꢀ1
)
CuII/CuI
Fc+/Fc
CuII/CuI
Fc+/Fc
CuII/CuI
Fc+/Fc
CuII/CuI
Fc+/Fc
A
0.04 (40)
0.04 (70)
–
–
160[b]–110[c]
–
4
4
–
10
–
–
90[d]
7
130
152
–
133
–
–
94
103
–
G
0.42 (10)
0.44 (30)
–
110[b]–75[c]
160[b]–170[c]
160[b]–140[c]
–
G
–
G
G
ꢀ0.05 (20)
160[b]–90[c]
–
–
–-
0.34 (–)[e]
420[f]
500[g]
90–100[g]
[a] Determined from DEp =logv. [b] Calculated from voltammetric peak integration at 0.05 Vsꢀ1. [c] Determined from the slope ip =f(v) for v<0.2 Vsꢀ1
.
[d] Obtained from DEp =logv by using Lavironꢃs approximation for quasi-reversible systems (DEp <200 mV). [e] In 1m HClO4, E vs. Ag/AgCl/KCl (1m)
from ref. [17a]. [f] Determined for mixed monolayers from ref. [17b]. [g] Determined by chronoamperometric methods for mixed SAMs from ref. [17c].
Chem. Eur. J. 2012, 18, 594 – 602
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
597

Y. Le Mest et al.
Figure 4. A) Cyclic voltammograms recorded at a [Cu(7)-N3ACHTNUTRGNE(GNU CH2)11-SH]-
modified gold electrode in H2O/KNO3 (0.1m) under N2 at v=0.01, 0.02,
0.05, 0.1 and 0.2 Vsꢀ1. * See explanation in text. B) Plots of ip vs. v
(0.02<v<1 Vsꢀ1) for the ferrocenium (a and d) and copper (b and c)
processes.
Figure 3. Cyclic voltammograms (v=0.05 Vsꢀ1) recorded in H2O/NaBF4
(0.05m)/sodium acetate (0.05m) at pH 4.5 under N2 at A) a [Cu(6)-N3-
ACHTUNGTRENNUNG(CH2)11-SH]-modified gold electrode and B) a [Cu(7)-N3ACHUTNTGREN(NUNG CH2)11-SH]-
modified gold electrode a) before and b) after dipping in NH4OH
(0.01m) for 10 min.
the same E8 values were found for the CuII/CuI process for
both [Cu(6)]- and [Cu(7)]-modified electrodes indicates that
the propylphenyl and methylferrocenyl groups have similar
electronic effects on the copper centre, in agreement with
previous conclusions (Table 1). To study a putative media-
ting effect of the copper ion on the ferrocenyl moiety, and
for a better characterisation of the processes, the copper
centre was removed from the [Cu(7)]-based immobilised
electrode device by dipping it in a 0.01m NH4OH solution
for 10 min. As shown in Figure 3B (curve b), the resulting
voltammogram shows that the electrochemical signal of the
CuII/CuI system has disappeared, whereas the Fc+/Fc pro-
cess remained almost unaffected. The surface coverage (G)
for each modified electrode was then determined by numer-
ic integration of the voltammetric peaks recorded at low
scan rate on the basis of the geometric area (A=0.071 cm2)
and from the variation of the peak current with scan rate
(Table 2).[27,28] Both methods gave an average value of G=
(120Æ50) pmolcmꢀ2, which is in the same range as that
displayed two redox systems at E8=0.04 and 0.42 V versus
SCE (Figure 3B, curve a), both of which are also typical of
an electron-transfer reaction for an adsorbed species with a
linear variation of the anodic and cathodic peak intensities
with v for both systems (Figure 4). Interestingly, plots of ip
versus v for the copper and ferrocene processes yielded dif-
ferent slopes (smaller for copper, see Figure 4B). Also, the
peak separation (DEp) increased significantly with v for the
CuII/CuI system, whereas it remained almost constant for
ferrocene (Figure 4A and Figure 5) with a low value close to
the ideal case for a very fast electron transfer to an immobi-
lised species (DEp =0 mV). Such differences are ascribed to
slower electron-transfer kinetics related to a reorganisation-
al process in the CuII/I reaction. Indeed, when the kinetics
are no longer limiting for copper (v<0.05 Vsꢀ1), both sys-
tems display similar peak heights (Figure 3B, curve a). The
increase in scan rate also led to the appearance of an extra
anodic peak at around Epa =0.3 V at v=1 Vsꢀ1 (Figure 4A,
peak *); this peak potential shifts positively with increasing
scan rate. As it did not appear in the first cycle when scan-
ning positively, it can be ascribed to the oxidation of a tran-
sient CuI complex that results from the evolution of the
electro-generated CuI species (EC=electrochemical-chemi-
cal mechanism). As observed by CV in solution in an aque-
ous medium (Figure S3 in the Supporting Information), the
chemical reaction (C) following the CuII!CuI reduction
step (E) might involve the release from the CuI centre of
either one one pyridyl or one triazolyl arm, or an H2O
ligand (or a combination of the two processes). The fact that
found for [CuACTHNUGTRNEUNG(6-eTMPA)] when grafted onto non-mixed azi-
doalkanethiol-modified gold surfaces under the same condi-
tions (Table 2). The slight discrepancy found for the G
values obtained for the CuII/CuI process compared with the
Fc+/Fc process for the [Cu(7)]-modified electrode indicates
that the grafted complex does not follow the ideal redox be-
haviour expected for a monolayer with two different reversi-
ble redox systems. The higher G value found for the ferro-
cenyl centre may possibly be due to the presence of “Cu-
free” tagged ferrocenyl complexes (Ø-Fc) on the surface.
Another important parameter is the full-width-at-half-maxi-
598
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 594 – 602

Functionalisation of Electrode Surfaces
FULL PAPER
not too fast (<103 sꢀ1). In the model developed by Laviron,
in which interactions between redox centres are not consid-
ered, the standard rate constant k8 is calculated from the in-
tercept of the linear part of the peak separation DEp versus
logv (DEp >200 mV for a one-electron-transfer reaction).[29]
This allowed k8 values to be evaluated for [Cu(6)] and
[Cu(7)] complexes (Figure 5 and Table 2). The calculations
show that the CuII/I electron-transfer process is moderately
fast (4 sꢀ1) irrespective of the nature of the substituting
group (Fc or Ph). They are also of the same order of magni-
tude as that found for the grafted [CuACTHNUTRGNEUNG(6-eTMPA)] complex
measured under similar conditions (10 sꢀ1). Remarkably, for
the Fc+/Fc process, the peak separation depends on whether
or not the copper centre is present, as clearly depicted in
Figure 5A. Indeed, the calculations give a standard rate con-
stant of 7 sꢀ1 for the tagged metal-free [Cu(7)] complex.
When copper is present, the electron transfer is significantly
faster (k8=90 sꢀ1) and is closer to the value found for the
ethynylferrocene complex grafted on undecanethiol chains
(500 sꢀ1).[17c,30] Two main effects may explain the key role of
the copper ion on the kinetics. The first is the structural
stiffening of the tagged complex in the presence of copper
through the formation of coordinating bonds between the
metal ion and N-donor groups (two pyridyl, two triazoles,
one amine). This induces a decrease in the distance separat-
ing the ferrocenyl moiety from the electrode surface, as de-
picted in Scheme 5. Consequently, the electron transfer be-
tween the electrode and the ferrocenyl moiety is substantial-
ly accelerated.[31] A second possible explanation is that the
copper complex acts as an electron relay in the electron-
hopping process from the electrode to the ferrocene. The
presence of the metal ion would induce a sequential locali-
sation of the charge and a decrease in the activation energy
for the whole electron transfer. However, this second hy-
pothesis is unlikely in the case of [Cu(7)]: the standard po-
tentials of the CuII/CuI and Fc+/Fc processes are not close
enough to each other (DE8=380 mV) to allow a self-ex-
change mechanism by electron hopping. In fact, charge-hop-
ping processes in SAMs have rarely been described in the
literature[32] because most of the studies have been per-
Figure 5. Plots of Epa (a, b and c) and Epc (a’, b’ and c’) vs. logv derived
from the cyclic voltammograms of [Cu(6)]-, [Cu(7)]- and metal-free
[Cu(7)]-modified gold electrodes in H2O/NaBF4 (0.05m)/sodium acetate
(0.05m) at pH 4.5 under N2 at v=5–10 Vsꢀ1 for A) Fc+/Fc and B) CuII/
CuI electron-transfer processes. a,a’) The [Cu(7)]-modified electrode,
b,b’) the metal-free [Cu(7)]-modified electrode prepared by dipping in
NH4OH solution and c,c’) the [Cu(6)]-modified electrode.
mum (fwhm) value (DEp,1/2; Table 2), which gives an indica-
tion of the deviation from the ideal reversible behaviour of
an immobilised redox species (Langmuir isotherm condi-
tions). The DEp,1/2 values obtained for the Fc+/Fc process
are close to the Nernstian case (90.6 mV at 298 K),[27,28]
which suggests that the ferrocenyl groups behave as equiva-
lent redox sites and that the global interactions between
them are null. However, for the
CuII/CuI process, these values
are significantly higher. Such an
effect may result from a combi-
nation of repulsive interactions
between the copper centres and
the sluggish electron-transfer
kinetics, which both broadens
and flattens the peaks.
Kinetics of electron transfer:
The variation in the peak sepa-
ration (DEp) with scan rate (v)
is a well-known method for de-
termining electron-transfer ki-
netics between electrode-tagged
Scheme 5. Hypothesis for the difference in electron-transfer kinetics for the [Cu(7)]-modified electrode before
species when the kinetics are and after copper removal by a NH4OH treatment.
Chem. Eur. J. 2012, 18, 594 – 602
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
599

Y. Le Mest et al.
formed with mono-functional systems with saturated or un-
saturated chains.
case of a gold electrode but should be extendable to the
more robust vitreous and graphitic electrodes.[33,34] Secondly,
thanks to the presence of the N4–Cu redox link, it allows
electron communication between two different probes to be
explored. Indeed, the experimental evidence for electron
relay in molecular devices is of great interest in biology
(e.g., a model for electron-transfer in DNA[35]) and biotech-
nology (design of new biochips]).[36] We are currently work-
ing on these two aspects.
Conclusion
Two main strategies have classically been used for the func-
tionalisation of a surface. The most straightforward ap-
proach is the direct grafting of the object of interest, which,
however, has two problems. It is restricted to compounds
presenting a reactive moiety towards the surface; this raises
the difficulty of functional compatibility with that of the
compound itself. Also, it can hardly insure good covering of
the surface, that is, impermeability to direct electron transfer
when the compound cannot be packed as a dense monolayer
due to steric crowding. The second strategy involves a two-
step procedure: functionalisation of the surface with an
alkyne (generally a thiol) or an aryl (e.g., a diazo) with a
function that will, in the second step, react with the com-
pound of interest. The most popular reaction used for this
second step is the CuAAC procedure, which requires the
use of a copper catalyst.
Experimental Section
Chemicals: Organic solvents were distilled over CaH2 except for THF
and diethyl ether which were distilled over Na/benzophenone. Acetoni-
trile (99.9% BDH, VWR) was used as received and kept under N2 in a
glovebox. All solvents were thoroughly degassed before use. NBu4PF6
was synthesised from NBu4OH (Fluka) and HPF6 (Aldrich). It was then
purified, dried under vacuum for 48 h at 1008C and then stored under N2
[38]
in the glovebox. Compounds 6-Br-BMPA,[37] FcCH2N3
(CH2)2N3
and Ph-
[39]
were prepared according to literature procedures. 11-Azi-
doundecane-1-thiol was synthesised following a previously described pro-
cedure.[17a] All other chemicals were of reagent grade and were used
without purification.
We have proposed in this work a novel and versatile strat-
egy that involves the use of a generic molecular platform
(compound 3). This platform was designed to provide two
sequentially addressable moieties: one for the object to
graft, the other for the pre-functionalised electrode. The key
point of our strategy is the formation, after the first click re-
action, of an N4 ligand bound to the object. The subsequent
stoichiometric N4 complexation of CuII allows the object of
interest to be connected to the pre-functionalised surface di-
rectly by a simple self-induced electroclick reaction. The ad-
vantages of such a strategy based on platform 3 are multi-
ple: 1) the grafting step is electrochemically addressable,
2) it is easily monitored and quantified, 3) the grafting pro-
cedure is “clean” as no exogenous copper complex is re-
quired, 4) each reaction employed in the multistep strategy
is high yielding and has a high chemical tolerance and final-
ly 5) the formation of the intermediate CuII complex in-
creases the hydrophilicity of the object and thus its solubility
in water, a key solvent for avoiding the denaturation of the
alkynethiol monolayer. This last point is particularly inter-
esting for the grafting of organic substrates that are poorly
water soluble. As a proof of concept, two different function-
al groups were immobilised to evaluate this strategy, a
redox-innocent phenyl group and a redox probe, ferrocene.
Monitoring of the in situ electrochemical grafting showed
that the kinetics of immobilisation is fast for both systems
with good surface coverage. Remarkably, also, the study of
the ferrocenyl-based copper complex highlighted the influ-
ence of the copper ion on the kinetics of electron transfer to
ferrocene by combined structural/electronic effects. Hence
the molecular platform 3 appears to be a promising candi-
date for applications in surface addressing in areas ranging
from electronics to biology. Indeed, it addresses two issues:
First, the possibility of selective surface patterning through
electrochemical control; this has been demonstrated in the
Apparatus: NMR spectra were recorded on a Bruker DRX 500 MHz
spectrometer. IR spectra were recorded on a Nicolet Nexus FT-IR spec-
trometer (KBr pellets). EPR spectra were recorded on a Bruker Elexys
spectrometer (X band). UV/Vis/NIR spectroscopy was performed with a
JASCO V-670 spectrophotometer. Single-mass analyses were performed
by the Service Central d’Analyse du CNRS, Solaize, France.
Electrochemical experiments: The electrochemical studies in acetonitrile
were performed in a glovebox (Jacomex, O2 <1 ppm, H2O<1 ppm) with
a home-designed 3-electrode cell. A commercial platinum removable tip
electrode (Metrohm) was used as the working electrode. Before each ex-
periment, it was polished in a slurry of alumina (3 mm) and sonicated in
water (Millipore, 18 MW). The counter electrode was a platinum wire. A
separate frit containing a platinum electrode dipped in an equimolar
electrolytic solution of FcPF6 and ferrocene was used as the reference
electrode. The potential of the cell was controlled by an AUTOLAB
PGSTAT 302 (Ecochemie) potentiostat monitored with a computer pilot-
ed with GPES software. Ferrocene was added at the end of each experi-
ment to determine accurate redox potentials. Voltammetry was per-
formed in an aqueous electrolyte with a standard calomel reference elec-
trode (saturated KCl) as reference.
Electrode modification: A commercial gold removable tip electrode
(Metrohm) was used. Before modification, the surface of the gold elec-
trode (A=0.071 cm2) was prepared following a classical procedure by
polishing in a slurry with alumina (3 mm). It was then sonicated in water
(Millipore, 18 MW) and cycled between +0.5 and 1.4 V versus SCE in
0.1m H2SO4 (40 scans) to remove gold oxide, washed with water, then
ethanol and dried under a slow flow of N2 before being introduced into a
solution containing thiol. This procedure allowed the use of the same
electrode with a renewed appropriate gold surface condition. The elec-
trode was kept in a non-mixed 1 mm 1-azido-11-undecanethiol solution in
EtOH for 12 h under N2. After thorough washing with pure EtOH, the
electrode was tested by voltammetry in an aqueous solution of ferricya-
nide. The absence of a redox signal indicated a blockage of electron
transfer at the electrode surface by the thiol compound.
Synthesis of 2: 6-Br-BMPA (1.80 mmol, 500 mg) was dissolved in a de-
gassed THF/diisopropylamine mixture (2:1 v/v). Under N2, CuI
(0.18 mmol, 35 mg), [PdACHTUNGTRNEG(UN PPh3)2Cl2] (0.11 mmol, 76 mg) and finally triiso-
propylacetyene (3.6 mmol, 0.80 mL) were added. The reaction mixture
was stirred for 16 h at room temperature. The solution was filtered and
the solid washed with THF (3ꢄ25 mL). The solvent was removed with a
rotary evaporator, the black oil obtained was dissolved in CH2Cl2
600
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 594 – 602

Functionalisation of Electrode Surfaces
FULL PAPER
(50 mL) and washed with water (5ꢄ30 mL). The organic phase was dried
over MgSO4, filtered through hydrophilic cotton and the solvent was re-
moved with a rotary evaporator. The product was isolated in a yield of
74% (508 mg, 1.34 mmol) by alumina column chromatography with
CH2Cl2/MeOH (99:1). 1H NMR (300 MHz, CDCl3): d=8.56 (d, J=
4.5 Hz, 1H), 7.69–7.33 (m, 5H), 7.15 (dd, J=4.5, 1 Hz, 1H), 3.98 (s, 4H),
2.71 (brs, 1H), 1.16 ppm (m, 21H).
2H), 2.21 ppm (q, J=7.25 Hz, 2H); 13C NMR (500 MHz, CDCl3): d=
159.4, 158.3, 148.8, 143.5, 141.4, 140.1, 136.7, 128.6–128.3, 126.3, 125.9,
123.4, 122.3, 82.8, 59.2, 49.4, 48.6, 32.5, 31.6 ppm; MS (TOF ES+,
MeOH): m/z: calcd for C26H27N6 [M+H]+: 423.2297; found: 423.2304.
Purification and characterisation of 7: Compound 7 was purified by
column chromatography on silica gel with CH2Cl2/MeOH as eluent
(methanol ratio from 0 to 3% v/v). The dark-red product was obtained
in a yield of 51%. This yield is for the click reaction and deprotection
procedure. 1H NMR (500 MHz, CDCl3): d=8.52 (d, J=4 Hz, 1H), 7.65–
7.61 (m, 5H), 7.34 (m, 1H), 7.14 (dd, J=4, 5.5 Hz, 1H), 5.27 (s, 2H), 4.27
(s, 2H), 4.20 (s, 2H), 4.17 (s, 5H), 3.87 (s, 6H), 3.13 ppm (s, 1H);
13C NMR (500 MHz, CDCl3): d=160, 159, 148.9, 144, 141.4, 136.8, 126.0,
123.7, 123.4, 122.9, 122.3, 28.9, 81.1, 69.0, 59.3, 50.0, 48.8 ppm; MS (TOF
ES+, MeOH): m/z: calcd for C28H27N6Fe [M+H]+: 503.1647; found:
503.1654.
Synthesis of 3: 6-TIPSe-BMPA (1.34 mmol, 508 mg) was dissolved in
THF (20 mL). Under an inert atmosphere, propargyl bromide
(1.47 mmol, 0.15 mL, 80% in toluene) and DIPEA (4 mmol, 0.70 mL)
were added. The reaction mixture was heated at reflux overnight. The so-
lution was filtered and the solid washed three times with THF (25 mL).
The solvent was removed with a rotary evaporator. The product was dis-
solved in CH2Cl2 (90 mL), washed with water (3ꢄ50 mL), dried over
MgSO4, filtered through cotton and the solvent was removed with a
rotary evaporator. The product was purified by column chromatography
on silica gel with CH2Cl2/MeOH as eluent (99:1), to give a yield of 65%.
1H NMR (500 MHz, CDCl3): d=8.49 (d, J=4.8 Hz, 1H), 7.59 (t, J=
7.7 Hz, 1H), 7.55 (t, J=7.5 Hz, 1H), 7.47 (d, J=7.5 Hz, 1H), 7.44 (d, J=
7.8 Hz, 1H), 7.29 (d, J=7.3 Hz, 1H), 7.09 (dd, J=7.3, 4.9 Hz, 1H), 3.88
(s, 2H), 3.85 (s, 2H), 3.37 (d, J=2.3 Hz, 2H), 2.24 (t, J=2.3 Hz, 1H),
1.11 ppm (m, 21H); 13C NMR (500 MHz, CDCl3): d=159.6, 159.1, 149.1,
142.2, 136.3, 126.34, 123.0, 121.9, 107.32, 91.4, 77.8, 73.0, 59.1, 59.0, 43.4,
18.5, 11.2 ppm; IR (KBr): n˜ =3306 (m), 3058 (m), 2943 (s), 2865 (s), 2157
Acknowledgements
This project was supported by the CNRS (Institut de Chimie), the Minis-
tꢁre de l’Enseignement Supꢀrieur et de la Recherche (MESR Ph.D.
grant to C.O.) and Agence Nationale pour la Recherche (Cavity-
zyme(Cu) Project ANR-2010-BLAN-7141).
ꢁ ꢀ
ꢁ ꢀ
(m, n˜C C Si), 2103 (w, n˜C C H) 1683 (m), 1581 (s), 1569 (s), 1447 (s),
1267 (m), 1120 (m), 996 (s), 883 (s), 734 (m), 678 cmꢀ1 (s); MS (TOF
ES+, MeOH): m/z: calcd for C26H36N3Si [M+H]+: 418.2679; found:
418.2683.
General procedure for the synthesis of compounds 4 and 5: 6-TIPSe-
BMPPA (3) and 1-azido-3-phenylpropane (for 4) or azidomethylferro-
cene (5) in a 1:1 ratio were added to a solution of dichloromethane
(15 mL). Aqueous solutions of CuSO4·5H2O (0.1 molequiv, 7.5 mL ) and
sodium ascorbate (0.2 molequiv, 7.5 mL) were prepared. Then the three
solutions were combined (gathered) and stirred vigorously for 48 h under
an inert atmosphere at room temperature. Dichloromethane and water
(50 mL of each) were added to the mixture. The aqueous phase was ex-
tracted with CH2Cl2 (2ꢄ50 mL) and the organic phases were collected,
washed with water (3ꢄ75 mL), dried over MgSO4 and filtered through
cotton. The solvent was removed with a rotary evaporator.
[1] a) A. Pron, P. Gawrys, M. Zagorska, D. Djurado, R. Demadrillea,
Chem. Soc. Rev. 2010, 39, 2577–2632; b) M. Grꢅtzel, Inorg. Chem.
2005, 44, 6841–6851; c) B. Ulgut, H. D. Abruna, Chem. Rev. 2008,
108, 2721–2736.
[2] a) J. Chen, W. Zhang, D. Officer, G. F. Swiegers, G. G. Wallace,
Chem. Commun. 2007, 3353–3355; b) I. Willner, E. Katz, Angew.
Chem. 2000, 112, 1230–1269; c) L. Shen, Z. Chen, Y. Li, S. He, S.
Xie, X. Xu, Z. Liang, X. Meng, Q. Li, Z. Zhu, M. Li, X. C. Le, Y.
Shao, Anal. Chem. 2008, 80, 6323–6328; d) D. Nkosi, J. Pillay, K. I.
Ozoemena, K Nouneh, M. Oyama, Phys. Chem. Chem. Phys. 2010,
12, 604–613; e) E. Bakker, Y. Qin, Anal. Chem. 2006, 78, 3965–
3983.
[3] J. A. Cracknell, K. A. Vincent, F. A. Armstrong, Chem. Rev. 2008,
108, 2439–2461.
[4] a) A. Heller, Curr. Opin. Chem. Biol. 2006, 10, 664–672; b) C.
Bunte, O. Prucker, T. Kçnig, J. Rꢆhe, Langmuir 2010, 26, 6019–
6027; c) B. A. Gregg, A. Heller, J. Phys. Chem. 1991, 95, 5976–5980;
d) S. Cosnier, Biosens. Bioelectron. 1999, 14, 443–456.
Purification and characterisation of 4: Compound 4 was purified by
column chromatography on silica gel with CH2Cl2/MeOH as eluent
(methanol ratio from 0 to 1.5% v/v). The yellow product was obtained in
a yield of 45%. 1H NMR (500 MHz, CDCl3): d=8.55 (d, J=3 Hz, 1H),
7.66–7.56 (m, 4H), 7.36 (dd, J=7.5, 1 Hz, 1H), 7.31 (m, 2H), 7.21–7.15
(m, 4H), 4.34 (t, J=7 Hz, 2H), 3.89 (s, 2H), 3.87 (s, 2H), 3.84 (s, 2H),
2.65 (t, J=7.5 Hz, 2H), 2.25 (q, J=7.5 Hz, 2H), 1.16 ppm (m, 21H).
[5] a) G. Decher, Science 1997, 277, 1232–1237; b) J. Hodak, R. Etche-
nique, E. J. Calvo, K. Singhal, P. N. Bartlett, Langmuir 1997, 13,
2708–2716; c) V. Flexer, E. S. Forzani, E. J. Calvo, S. J. Luduena,
L. I. Pietrasanta, Anal. Chem. 2006, 78, 399–407.
[6] J. C. Love, L. A. Estroff, J. K. Kriebel, R. G. Nuzzo, G. M. White-
sides, Chem. Rev. 2005, 105, 1103–1169.
Characterisation of 5: Compound 5 was used in the next step without ad-
ditional purification. 1H NMR (400 MHz, CDCl3): d=8.50 (d, J=4 Hz,
1H), 7.62–7.53 (m, 5H), 7.32 (d, J=8 Hz, 1H), 7.12 (dd, J=4.6 Hz, 1H),
5.24 (s, 2H), 4.25 (s, 2H), 4.19 (s, 2H), 4.16 (s, 5H), 3.83 (s, 2H), 3.81 (s,
2H), 3.80 (s, 2H), 1.14 ppm (m, 21H).
General procedure for the synthesis of compound 6 and 7: Compound 4
or 5 was dissolved in THF (20 mL) and NBu4F (3 molequiv) was added
to the solution. The reaction mixture was stirred for 15 h and then THF
was removed with a rotary evaporator. The product was dissolved in
CH2Cl2 (50 mL) and extracted with an aqueous solution of 0.5m HCl (3ꢄ
50 mL). The aqueous phase was collected and the pH adjusted to 8.6.
CH2Cl2 (100 mL) was added and reaction mixture was vigorously stirred
for 30 min. Then the aqueous phase was extracted with CH2Cl2 (2ꢄ
75 mL). The organic phases were collected and washed with water (2ꢄ
100 mL), dried over MgSO4, filtered through cotton and the solvent re-
moved with a rotary evaporator.
[7] a) C. E. D. Chidsey, C. R. Bertozzi, . T. M. Putvinski, A. M. Mujsce,
J. Am. Chem. Soc. 1990, 112, 4301–4306; b) C. E. D. Chidsey, Sci-
ence 1991, 251, 919–922.
[8] C. Amatore, S. Gazard, E. Maisonhaute, C. Pebay, B. Schçllhorn, J.-
L. Syssa-Magalꢀ, J. Wadhawan, Eur. J. Inorg. Chem. 2007, 4035–
4042.
[9] a) J. Sumner, S. E. Creager, J. Phys. Chem. B. 2001, 105, 8739–8745;
b) H. O. Finklea, D. D. Hanshaw, J. Am. Chem. Soc. 1992, 114,
3173–3181; c) D. A. Brevnov, H. O. Finklea, H. Van Ryswyk, J.
Electroanal. Chem. 2001, 500, 100–107; d) L. Tender, M. T. Carter,
R. W. Murray, Anal. Chem. 1994, 66, 3173–3181.
Purification and characterisation of 6: Compound 6 was purified by
column chromatography on silica gel with CH2Cl2/MeOH as eluent
(99:1), to give the product in a yield of 65%. 1H NMR (500 MHz,
CDCl3): d=8.53 (d, J=5 Hz, 1H), 7.67–7.55 (m, 4H), 7.33 (dd, J=6,
3 Hz, 1H), 7.28–7.25 (m, 2H), 7.19–7.14 (m, 4H), 4.31 (t, J=7 Hz, 2H),
3.90 (s, 2H), 3.88 (s, 2H), 3.87 (s, 2H), 3.11 (s, 1H), 2.61 (t, J=7.5 Hz,
[10] a) O. AlꢀvÞque, P.-Y. Blanchard, C. Gautier, M. Dias, T. Breton, E.
Levillain, Electrochem. Commun. 2010, 12, 1462–1466; b) S. J.
Green, N. Le Poul, P. P. Edwards, G. Peacock, J. Am. Chem. Soc.
2003, 125, 3686–3687.
[11] a) A. L. Eckermann, D. J. Feld, J. A. Shaw, T. J. Meade, Coord.
Chem. Rev. 2010, 254, 1769–1802; b) E. Tran, M. A. Rampi, G. M.
Chem. Eur. J. 2012, 18, 594 – 602
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
601

Y. Le Mest et al.
Whitesides, Angew. Chem. 2004, 116, 3923–3927; Angew. Chem. Int.
Ed. 2004, 43, 3835–3839.
holding the potential at ꢀ0.20 V for an azido-modified gold elec-
trode. This clearly indicates that the ferrocene moiety of the com-
plex can diffuse into the azidoundecanethiol monolayer, which is no
longer “insulating”. Such mass-transfer processes may thus be re-
sponsible for the appearance of the sharp oxidation peak observed
with higher current intensities for ferrocene during grafting, similar-
ly to what is observed in solution.
[12] a) B. T. Houseman, E. S. Gawalt, M. Mrksich, Langmuir 2003, 19,
1522–1531; b) E. A. Smith, W. D. Thomas, Y. Cheng, S. V. P. Bar-
reira, A. G. Frutos, R. M. Corn, Langmuir 2001, 17, 2502–2507.
[13] G. J. Wegner, H. J. Lee, R. M. Corn, Anal. Chem. 2002, 74, 5161–
5168.
[14] E. A. Smith, W. D. Thomas, L. L. Kiessling, R. M. Corn, J. Am.
Chem. Soc. 2003, 125, 6140–6148.
[26] The theoretical surface coverage of optimally well-packed alkane-
thiols on AuACTHNUTRGNEUNG
(111) is 775 pmolcmꢀ2 (see ref. [17b]). The experimental
[15] The immobilisation of redox proteins onto thioalkane-modified elec-
trodes is not considered here as they specifically bind through non-
covalent modes. See, for example: M. J. Tarlov, E. F. Bowden, J.
Am. Chem. Soc. 1991, 113, 1847–1849.
[16] V. V. Rostovtsev, L. G. Green, V. V. Fokin, K. B. Sharpless, Angew.
Chem. 2002, 114, 2708–2711; Angew. Chem. Int. Ed. 2002, 41, 2596–
2599.
[17] a) J. P. Collman, N. K. Devaraj, C. E. D. Chidsey, Langmuir 2004, 20,
1051–1053; b) J. P. Collman, N. K. Devaraj, T. P. A. Eberspacher,
C. E. D. Chidsey, Langmuir 2006, 22, 2457–2464; c) N. K. Devaraj,
R. A. Decreau, W. Ebina, J. P. Collman, C. E. D. Chidsey, J. Phys.
Chem. B. 2006, 110, 15955–15962.
[18] N. K. Devaraj, P. H. Dinolfo, C. E. D. Chidsey, J. P. Collman, J. Am.
Chem. Soc. 2006, 128, 1794–1795.
[19] J. P. Collman, N. K. Devaraj, R. A. Decreau, Y. Yang, Y.-L. Yan, W.
Ebina, T. A. Eberspacher, C. E. D. Chidsey, Science 2007, 315, 1565–
1568.
surface coverage of the complex from Fc peak integration at
0.05 Vsꢀ1 is 160 pmolcmꢀ2, that is, 20% of the theoretical value.
Such a discrepancy, which has been mentioned by Chidsey and co-
workers,[17b] is due to the steric limitation inherent to the size of the
ethynyl substrate (in the case of ethynylferrocene, a maximum value
of 55% was found[17b]).
[27] A. J. Bard, L. R. Faulkner, Electrochemical Methods: Fundamentals
and Applications, 2nd ed., John Wiley & Sons, New York, 2001,
pp. 580–631.
[28] J.-M. Savꢀant, Elements of Molecular and Biomolecular Electro-
chemistry, John Wiley & Sons, Hoboken, 2006, pp. 1–77.
[29] E. Laviron, J. Electroanal. Chem. 1979, 101, 19–28.
[30] The calculation of the standard rate constant for the Fc+/Fc system
for the [Cu(7)]-modified electrode was performed by using Lavir-
onꢃs procedure for quasi-reversible systems (DEp <200 mV).[29]
[31] The rate of electron transfer falls exponentially with the distance
from the saturated alkane chains, see refs. [7,9].
[20] a) N. K. Devaraj, G. P. Miller, W. Ebina, B. Kakaradov, J. P. Coll-
man, E. T. Kool, C. E. D. Chidsey, J. Am. Chem. Soc. 2005, 127,
8600–8601; b) Y. Zhang, S. Luo, Y. Tang, L. Yu, K.-Y. Hou, J.-P.
Cheng, X. Zeng, P. G. Wang, Anal. Chem. 2006, 78, 2001–2008;
c) J. F. Lutz, Angew. Chem. 2007, 119, 1036–1043; Angew. Chem.
Int. Ed. 2007, 46, 1018–1025.
[32] C. Amatore, E. Maisonhaute, B. Schçllhorn, J. Wadhawan, Chem-
PhysChem 2007, 8, 1321–1329.
[33] D. Evrard, F. Lambert, C. Policar, V. Balland, B. Limoges, Chem.
Eur. J. 2008, 14, 9286–9291.
[34] C. C. L. McCrory, A. Devadoss, X. Ottenwaelder, R. D. Lowe, T. D.
Stack, C. E. D. Chidsey, J. Am. Chem. Soc. 2011, 133, 3696–3699.
[35] a) B. Giese, M. Spichty, ChemPhysChem 2000, 1, 195–198; b) G. B.
Schuster, Acc. Chem. Res. 2000, 33, 253–260.
[21] T. S. Hansen, A. E. Daugaard, S. Hvilsted, N. B. Larsen, Adv. Mater.
2009, 21, 4483–4486.
[22] A. Gomila, N. Le Poul, N. Cosquer, J.-M. Kerbaol, J.-M. Noꢇl, M. T.
Reddy, I. Jabin, O. Reinaud, F. Conan, Y. Le Mest, Dalton Trans.
2010, 39, 11516–11518.
[23] N. Le Poul, B. Douziech, J. Zeitouny, G. Thiabaud, H. Colas, F.
Conan, N. Cosquer, I. Jabin, C. Lagrost, P. Hapiot, O. Reinaud, Y.
Le Mest, J. Am. Chem. Soc. 2009, 131, 17800–17808.
[36] E. M. Boon, D. M. Ceres, T. G. Drummond, M. G. Hill, J. K. Barton,
Nat. Biotechnol. 2000, 18, 1096–1100.
[37] M. Merkel, D. Schnieders, S. M. Baldeau and B. Krebs, Eur. J. Inorg.
Chem. 2004, 783–790.
[38] J. M. Casas-Solvas, A. Vargas-Berenguel, L. F. Capitan-Vallvey, F.
Santo-Gonzalez, Org. Lett. 2004, 6, 3687–3690.
[24] The exact nature of the mechanism has not been scrutinised due to
the instability of the compounds in aqueous media and problems of
insolubility and deposition at the surface of the electrode.
[25] A similar voltammetric shape (but no increase in the peak intensi-
ties) was obtained when cycling with the same solution, but without
[39] G. Colombano, C. Travelli, U. Galli, A. Caldarelli, M. G. Chini, P. L.
Canonico, G. Sorba, G. Bifulco, G. C. Tron, A. A. Genazzani, J.
Med. Chem. 2010, 53, 616–623.
Received: August 23, 2011
Published online: December 9, 2011
602
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 594 – 602