Redox-active Si(100) surfaces covalently functionalised with [60]fullerene conjugates: New hybrid materials for molecular-based devices

Article abstract of DOI:10.1039/b717438a

Herein, we report the covalent immobilisation, through Si-C bonds, of various [60]fullerene derivatives on flat silicon surfaces following three different preparative protocols. Each synthetic strategy comprises a two-step approach that includes a pre-modification step of the Si(100) surface with an organic monolayer bearing a terminal functionality that undergoes a bond-forming reaction with a [60]fullerene synthon as characterized by X-ray photoelectron spectroscopy (XPS) measurements. Water contact angle measurements clearly showed a characteristic change of the surface hydrophobicity upon covalent immobilisation of the carbon functions. The hybrid [4b-Si(100)] surfaces, containing [60]fullerene-ferrocene fragments, were also investigated by means of cyclic voltammetry (CV), and were revealed to be exceptionally robust towards repeated reduction-oxidation cycles. Moreover, several surface-confined redox couples were observed in CH₃CN solution. The surface coverage was measured to be ca. 2.5 × 10^-11 mol cm^-2. The Royal Society of Chemistry.

Full text of DOI:10.1039/b717438a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/materials | Journal of Materials Chemistry
Redox-active Si(100) surfaces covalently functionalised with [60]fullerene
conjugates: new hybrid materials for molecular-based devices†‡
Fabrizio Cattaruzza,a Anna Llanes-Pallas,a Andrea G. Marrani,b Enrique A. Dalchiele,bc Franco Decker,b
b
Robertino Zanoni, Maurizio Prato and Davide Bonifazi
a
ad
*
*
Received 12th November 2007, Accepted 7th January 2008
First published as an Advance Article on the web 1st February 2008
DOI: 10.1039/b717438a
Herein, we report the covalent immobilisation, through Si–C bonds, of various [60]fullerene derivatives
on flat silicon surfaces following three different preparative protocols. Each synthetic strategy
comprises a two-step approach that includes a pre-modification step of the Si(100) surface with an
organic monolayer bearing a terminal functionality that undergoes a bond-forming reaction with
a [60]fullerene synthon as characterized by X-ray photoelectron spectroscopy (XPS) measurements.
Water contact angle measurements clearly showed a characteristic change of the surface
hydrophobicity upon covalent immobilisation of the carbon functions. The hybrid [4b-Si(100)]
surfaces, containing [60]fullerene-ferrocene fragments, were also investigated by means of cyclic
voltammetry (CV), and were revealed to be exceptionally robust towards repeated reduction–oxidation
cycles. Moreover, several surface-confined redox couples were observed in CH3CN solution. The
surface coverage was measured to be ca. 2.5 ꢀ 10ꢁ11 mol cmꢁ2
.
the surface and the bulk have great importance.5,6 Therefore,
controlling the electronic properties of semiconductor surfaces
is indispensable for constructing devices and for fine-tuning their
performance.5,7,8 Among all possible surface treatments, chemical
modification with organic molecules9 provides a promising tool
for the modulation of the superficial electronic properties.10
About 10 years ago, scientists started to study how to chemically
functionalise porous11 and flat12 silicon surfaces by introducing
covalent Si–C bonds.13 Beside the improvement of the chemical
and physical stability of such surfaces,14,15 the covalent modifica-
tion allowed chemists to remotely tune the superficial physical
properties of silicon by immobilising versatile functional organic
molecules the properties of which could be exploited toward the
engineering of advanced hybrid materials for sensor,16 elec-
tronic,8,17–19 magnetic,20 memory storage devices21 and biologi-
cal22,23 applications. Several functionalisation methodologies
have been thus developed to modify free-oxide silicon substrates
viathermalhydrosilylation,15 photochemical(UVorvisible)irradi-
ation,24 electrochemical reduction,25 acid- and metal-catalysed
hydrosilylation.26 The choice of the methodology to form Si–C
bonds strictly depends on the stability and chemical structure of
the attached functional groups. For instance, different electronical
active groups such as ferrocene27 and porphyrins19,28 could be
immobilised on silicon surfaces, showing very promising potential
properties as molecular-based charge-storage devices.
Introduction
The integration of circuits (wires, resistors, capacitors, and
transistors) on silicon substrates was the essential stimulus for
the increase of computational power in the late 1960s. By develop-
ing new techniques to fabricate smaller and smaller components,
scientists have enormously increased the speed and capabilities of
computing.1 According to ‘‘Moore’s Law’’, the technological
advancement should reduce the components to the molecular
scale in 20 years.1–3 Even considering further technological
progress in the manufacturing of integrated circuits, both
materials (such as Si) and processes currently in usewill reach their
fundamental limits some time in the future.1,3 Based on such
considerations, many scientists started to focus their attention
on the development of hybrid electronic devices in which surfaces
are modified with single molecules or molecular assemblies.2,4
Common electronic devices, such as diodes, are essentially
controlled by the electronic properties of their interfaces and
any changes of the chemical and electronic properties between
aCenter of Excellence for Nanostructured Materials, CENMAT,
Dipartimento di Scienze Farmaceutiche, INSTM UdR di Trieste,
`
Universita di Trieste, Piazzale Europa 1, 34127 Trieste, Italy. E-mail:
dbonifazi@units.it
bDipartimento di Chimica, Universita degli Studi di Roma
`
‘‘La Sapienza’’, p.le Aldo Moro 5, 00185 Rome, Italy. E-mail: robertino.
zanoni@uniroma1.it
c
For its part, [60]fullerene is one of the most widely used electron
acceptor components in molecular dyads and many efforts have
been focused on the development of hybrid materials containing
´ ´
Instituto de Fısica, Facultad de Ingenierıa, Herrera y Reissig 565, C.C. 30,
11000 Montevideo, Uruguay
dUniversity of Namur, Chemistry Department, Rue de Bruxelles 61, 5000
Namur, Belgium. E-mail: davide.bonifazi@fundp.ac.be
C
60 and its derivatives with the aim to study their electron transfer
† This paper is part of a Journal of Materials Chemistry theme issue on
carbon nanostructures.
properties.29,30 Surface modification utilising [60]fullerene
modules is currently of high importance, owing to the possibility
of transferring the unique electronic [60]fullerene properties to
bulk materials by surface coating.29,31 For instance, this type of
organic layer containing redox centers at a fixed distance from
‡ Electronic supplementary information (ESI) available: Selected
physical and spectral data of compounds 3a–c, a Zn(II)-centred (Zn 2p)
XPS spectrum, and CV of 3b in CH3CN solution and of [4b-nSi(100)]
under dark and illumination conditions. See DOI: 10.1039/b717438a
1570 | J. Mater. Chem., 2008, 18, 1570–1581
This journal is ª The Royal Society of Chemistry 2008

View Article Online
a surface instead of freely diffusing, represents an important class
of hybrid materials due to their potential technological applica-
tions ranging from medicinal chemistry to advanced nano-
structured devices for electronic applications.31 A large variety
of [60]fullerene-containing self-assembled monolayers formed
upon spontaneous adsorption of molecules on Au,32–34 Ag,35
Hg36 and silicon oxide37–39 surfaces have been described in the
literature. In an earlier work, Cahen and co-workers have also
investigated the immobilisation of [60]fullerenes derivatives on
various semiconducting surfaces.40 Specifically, it has been
observed that a series of [60]fullerene-derived carboxylic acids
could be easily adsorbed on (n- and p-type) GaAs and ZnO
crystals, revealing a strong influence of the electron-accepting
[60]fullerene moieties on the superficial properties (work function
and band bending) of the semiconductor as a consequence of the
interaction between the frontier orbitals of the carbon spheres
with the valence and conduction bands.40 To the best of our
knowledge, if two works on porous silicon substrates41 and one
on Si(111) are excluded,42 only Feng and Miller have described
in 1999 the functionalisation of hydrogenated flat Si(100) with
[60]fullerenes.43 In particular, they reported the self-assembly of
C60 molecules on flat Si(100) surfaces via direct tethering of the
carbon sphere to the surface as shown by electrochemical and
fast atom bombardment mass spectrometry measurements. All
[60]fullerene-containing materials were found to be very stable
in both acidic aqueous and polar non-aqueous solvents.43
General synthetic procedures (Scheme 2). Synthesis of [60]ful-
lerene conjugates 3a–c. To anoven-dried 200 mLround-bottomed
flask charged with a toluene solution of C60 (0.52 mmol in 130 mL)
and amino acid 1 (0.26 mmol), the appropriate aldehyde deriva-
tive (2a–c, 1.04 mmol) was added. After 0.5 h of sonication and
1 h of reflux under vigorous stirring, the solution was concentrated
under reduced pressure, the crude purified by flash chromato-
graphy (SiO2; toluene) and the solvent evaporated in vacuo.
Precipitation of the chromatographic fraction from CHCl3
upon addition of MeOH afforded the desired fulleropyrrolidine
(3a–c) as brown powder. For the spectroscopic characteristics
see ESI.‡
Synthesis of [60]fullerene conjugates 4a–c: general procedure for
the acidic cleavage of the tert-butoxycarbonyl (Boc) protecting
group. To a stirred solution of the appropriate Boc-protected
[60]fullerene conjugate (3a–c) (0.14 mmol) in CH2Cl2 (4 mL,
62.4 mmol), TFA (2 mL, 26.9 mmol) was added dropwise at
0 ꢂC. After stirring for 16 h at 25 ꢂC, hexane (20 mL) was added
to the mixture. The suspension was filtered and the precipitate
washed with methanol, hexane, toluene, and CH2Cl2 until the
supernatant was clear. Subsequent drying under high vacuum
yielded the desired [60]fullerene-amine derivative (4a–c) as
brown powder. In order to prevent extensive decomposition of
the amine [60]fullerene precursor 4a–c, the deprotection reaction
was performed each time prior to the surface functionalisation
step.
In this paper, we thus report our synthetic strategies (Scheme 1)
towards the covalent functionalisation, through Si–C bonds, of
hydrogenated flat Si(100) surfaces with various [60]fullerene
derivatives (e.g., 4a–c). Specifically, three strategies, each
comprising a two-step approach including a pre-modification
step (Step A) of the silicon surface with an organic monolayer
bearing a terminal functionality that undergoes a bond-forming
reaction (Step B) with pristine C60 or with its derivatives, have
been pursued (Scheme 1). Full surface characterization with XPS
and water contact angle measurements showed that all methodo-
logies led to the immobilisation of the [60]fullerene moiety on
the surface. Additionally, comprehensive electrochemical
investigations on materials composed of [4b-Si(100)] showed the
formation of redox-active and very robust hybrid materials.
Functionalisation of Si(100) substrates
40% HF(aq) solution was purchased from Fluka and used as
received. 11-Bromo-1-undecene, 10-undecylenic acid, 10-undecy-
laldehyde, NaN3, pyridine, anhydrous N,N-dimethylformamide
(DMF), CH2Cl2 and MeOH were purchased from Aldrich and
also used as received. All chemical functionalisations of the
hydrogenated silicon samples [H-Si(100)] were carried out in
a N2(g)-purged dry-box (Braun) or using standard preparative
Schlenk-line procedures. The silicon surfaces used in these experi-
ments were single crystals, polished 525 ꢃ 25 mm thick Si(100)
wafers, boron- (p-type) or phosphorus-doped (n-type), with
7–21 U cm resistivity (Si-Mat). Before use, the Si(100) wafers
(1 ꢀ 1 cm) were first cleaned in boiling 1,1,2-trichlorethane for
10 min and then sonicated for 5 min in MeOH at room tempera-
ture. Subsequently the samples were chemically oxidised in H2O2–
HCl–H2O (2 : 1 : 8) at 353 K for 15 min, copiously rinsed with
de-ionised water, chemically etched with 10% aqueous HF for
10 min and again rinsed with de-ionized water. The samples
were then dried under a N2 stream and consequently covered (in
a dry-box) with 11-bromo-1-undecene, 10-undecylenic acid or
10-undecylaldehyde and subjected to a 35 mW cmꢁ2 visible irradi-
ation (quartz–iodine lamp) for 12 h. After functionalisation, the
samples ([Br-Si(100)], [HO2C-Si(100)] and [HOC-Si(100)]) were
subjected to three sonication cycles, 5 min each, with CH2Cl2
and MeOH and then dried under a N2 stream.
Experimental
Synthesis of [60]fullerene derivatives 4a–c
NMR spectra were obtained on a Jeol JNM-EX400 (400 MHz
1H-NMR) and on a Varian Gemini 200 (50 MHz 13C-NMR).
Chemical shifts are reported in ppm using the solvent residual
signal as an internal reference (CDCl3: dH ¼ 7.26 ppm, dC
¼
77.16 ppm). The resonance multiplicity is described as s (singlet),
d (doublet), dt (doublet of triplets), m (multiplet), br (broad
signal). IR spectra (KBr) were recorded on a Perkin Elmer
2000 spectrometer. Electrospray ionisation (ESI) mass spectro-
metry measurements were performed on a Perkin-Elmer API1
at 5600 eV at University of Trieste. 4-Dimethylaminobenzal-
dehyde 2a and ferrocene carboxaldehyde 2b were purchased
from CarloErba and Aldrich respectively, and used as received.
Formyl-substituted porphyrin 2c44 and N-Boc-protected amino-
acid 145 were prepared according to literature methods.
Amidation reaction between [HO2C-Si(100)] and the [60]fuller-
ene conjugates (Scheme 1, path a). The surface derivatisation with
[60]fullerenes was performed immersing the semiconductor
[HO2C-Si(100)] substrate into a 10 ꢀ 10ꢁ3 M solution of EDC
This journal is ª The Royal Society of Chemistry 2008
J. Mater. Chem., 2008, 18, 1570–1581 | 1571

View Article Online
Scheme 1 The three strategies employed to covalently modify the Si(100) surfaces: Step A: Si(100) surfaces functionalisation with a,u-bifunctional
alkyl chain molecules; Step B: [60]fullerene derivatisation of the passivated Si(100) surfaces. hn ¼ Visible irradiation from a quartz–iodine lamp;
TFA: trifluoroacetic acid; DMF: N,N-dimethylformamide; HOBt: 1-hydroxybenzotriazoleꢄH2O; EDC: 1-ethyl-3-(3-dimethylaminopropyl)carbodii-
mide; 5 ¼ N-methyl-3,4-fulleropyrrolidine fragment. 4i ¼ 4a, 4b, or 4c.
and HOBT in dry pyridine for 15 min, followed by the addition
of a 10 ꢀ 10ꢁ3 M solution of the selected [60]fullerene derivative
(4a–c) in dry pyridine. After the reaction was completed
(reaction time: about 48 h), the samples were removed from
the reaction solution, copiously rinsed with dry pyridine,
sonicated three times (5 min each) with a CH2Cl2–TFA (1 : 1)
mixture, washed with MeOH and then dried under a of N2
stream. The samples were stored under vacuum (10ꢁ2 Torr).
solution of C60 (20 ꢀ 10ꢁ3 M) and sarcosine (5 ꢀ 10ꢁ3 M) at
reflux for 16 h. After the reaction was completed, the ([5-
Si(100)]) samples were removed from solution, copiously rinsed
with toluene, sonicated three times (5 min each) with toluene,
CH2Cl2/TFA, MeOH and then dried under a N2 stream.
1,3-Dipolar reaction between [N3-Si(100)] and C60 (Scheme 1,
path c). The nucleophilic substitution of the bromine-terminated
Si(100) surfaces (Scheme 2) was performed using standard
preparative Schlenk-line procedures.46 The [Br-Si(100)] substrates
were immersed in a 10 ꢀ 10ꢁ3 M solution of NaN3 in dry DMF
1,3-Dipolar reaction between [HOC-Si(100)] and C60 (Scheme 1,
path b). The [HOC-Si(100)] surface were immersed in a toluene
1572 | J. Mater. Chem., 2008, 18, 1570–1581
This journal is ª The Royal Society of Chemistry 2008

View Article Online
Scheme 2 Synthesis of [60]fullerene derivatives 4a–c.
for 72 h. After the reaction was completed, the samples
([N3-Si(100)]) were removed from solution, copiously rinsed
with dry DMF, sonicated three times (5 min each) with DMF,
CH2Cl2, MeOH and then dried under a stream of N2. The 1,3-
dipolar reaction with C60 was performed immediately after the
azide treatment. The [N3-Si(100)] surfaces were immersed in
a 10 ꢀ 10ꢁ3 M solution of C60 in dry toluene at reflux for 16 h.
After the reaction was completed, the ([C60N-Si(100)]) samples
were removed from the reacting solution, copiously rinsed with
toluene, sonicated three times (5 min each) with toluene,
CH2Cl2–TFA (1 : 1), MeOH and then dried under a N2 stream.
under extended acquisition times under the X-rays was observed.
The experimental conditions adopted were: excitation by Al Ka
photons (hn ¼ 1486.7 eV), generated operating the anode at 14–
15 kV, 10–20 mA. No charging was experienced by the hybrid
species, as can be inferred from the Si 2p peak position, coinci-
dent with literature reports which assign a value of 99.7 eV to
the Si 2p3/2 bulk component. XPS atomic ratios for the function-
alised hybrids were estimated from experimentally determined
area ratios of the relevant core lines, corrected for the corres-
ponding theoretical atomic cross-sections and for a square root
dependence of the photoelectrons kinetic energies. The effects
on quantitative analysis possibly of photoelectron diffraction
at preferential directions of electron collection were minimised
by mounting the Si(100) wafers always with the same orientation
with respect to the analyzer axis.47
Contact angle measurements
The advancing contact angle of water was determined on sessile
´
drops with a Rame-Hart Model Automated Contact Angle
Goniometer at 25 ꢂC under atmospheric conditions. The advanc-
ing contact angle, qa (H2O), was obtained by RHI 2001 imaging
software at both sides of a forming 5 mL drop of water on the
surface. The reported values are the average of 5–15 measure-
ments (5 measurements in 3 different positions of the same
samples recorded every 15 s).
Electrochemistry
After monolayer formation ohmic contact was made to the back
of the derivatised [4b-Si(100)] samples scratching the Si surface,
rubbing it with Ga–In eutectic and attaching it to a copper
contact. The electrode set-up was obtained by pressing the Si
crystal against an O-ring sealing a small aperture in the PTFE
cell defining an electrode area of 0.3 cm2. The electrochemical
properties of the hybrid [4b-Si(100)] surfaces were explored by
CV measurements in 0.1 M Et4NClO4 in dry CH3CN. The
electrolyte solution deliberately contained no added electroactive
species. All electrochemical measurements were performed inside
a dry box with a three-electrode cell, using an Autolab Electro-
chemical Analyzer (model PGSTAT 12, Eco Chemie BV, The
Netherlands). The counter electrode was a platinum coil wire,
and a silver wire immersed in a 0.01 M AgNO3–0.1 M Et4NClO4
solution in CH3CN, separated from the main solution by a
porous fritted glass + agar plug, served as a reference electrode.
All potentials reported will be henceforth referred to this
reference. The kinetic analysis was performed after correcting
the cyclic voltammograms for IR drop.
X-Ray photoelectron spectroscopy
XPS results were obtained on an experimental apparatus in
UHV consisting of a modified Omicron NanoTechnology
MXPS system, with an XPS chamber equipped with a monochro-
matic X-ray source (Omicron XM-1000) and an Omicron
EA-127 energy analyzer. Samples were transferred between the
various experimental areas by means of linear magnetic transfer
rods or manipulators. All measurements were conducted in the
least possible time after sample preparation with p-type silicon
substrates. Samples were produced and mounted on sample
holders in a dry-box, transferred under N2 from the dry-box to
the XPS facility and introduced into the XPS chamber after
about 1 min air exposure. No sizeable sign of sample degradation
This journal is ª The Royal Society of Chemistry 2008
J. Mater. Chem., 2008, 18, 1570–1581 | 1573

View Article Online
is necessary in order to obtain a complete chemical and struc-
tural description of the prepared organic thin films.30 Therefore,
XPS and water contact angle measurements were conducted with
the aim of studying and confirming the covalent modification of
the silicon surfaces.
Results and discussion
Strategy, preparation, and functionalisation of the Si(100)
surfaces
The criteria for the design and preparation of the C60-based
adsorbates are mainly related to the development of new synthetic
methodologies to easily prepare modified silicon surfaces bearing
redox active molecules that can be in principle used as functional
hybrid materials to engineer molecular-based devices. For these
applications, [60]fullerene layers should possess electronic trans-
ducing properties together with a high chemical and structural
stability. In principle, there are two popular approaches to anchor
[60]fullerenes onto surfaces: covalent (via functionalisation of C60
with suitable groups that undergo a chemisorption on the surface
or via pre-modification of the surface by adsorption of an organic
monolayer bearing a terminal functionality that undergoes
a bond-forming reaction with C60 or its derivatives) or non-cova-
lent (pre-adsorption on the surface of an organic layer which
assembles with pristine C60 or its derivatives through a non-cova-
lent molecular recognition approach).31 As shown in Scheme 1,
our strategies involved the functionalisation of a silicon surface
with [60]fullerene derivatives that exposes their carbon spheres
at the aliphatic u-terminus to the outer surface. Since the linker
attached to the [60]fullerene moiety could have a large influence
on the electrochemical stability of the layer, we have decided to
explore the versatility of these synthetic strategies towards differ-
ent reactions leading to different linkers. Therefore, based on the
above-described requirements, the functionalisation strategies
reported in Scheme 1, in which the [60]fullerene moieties (Step
B) are immobilised to a pre-adsorbed layer (Step A) having
terminal functionalities which undergo bond-forming reactions
(Scheme 1, paths b and c) with C60 (leading to aza[60]fullerene
and [60]fulleropyrrolidine derivatives for [C60N-Si(100)] and
[5-Si(100)], respectively) or its derivatives (4a–c, via an
amidic bond for [4a-Si(100)], [4b-Si(100)] and [4c-Si(100)],
Scheme 1 path a) were chosen for the covalent attachment of
the [60]fullerene moiety.
X-Ray photoelectron spectroscopy
XP spectra were taken in order to both assess the nature of the
substrate–molecule interaction and monitor the progressive
build-up of the final hybrid system along the different function-
alisation steps depicted in Scheme 1. This section is organised as
follows: we first compare the results from the preparation proto-
cols of Si(100) surfaces functionalised with 10-undecylenic acid,
10-undecylaldehyde, and the azide terminus (Step A). The results
of curve-fitting analysis allow identification of the peculiar
features of each terminating functional group. Then we examine
the spectral evolution related to the establishment of a bond
between the C60 or its derivatives and the a,u-bifunctional alkyl
molecules immobilised on the surface (Step B). Each relevant
photoemission region (Si 2p, C 1s, N 1s and Fe 2p) is presented
separately, and its relevance to the overall assignment of the
characteristic nature and electronic structure of the single species
discussed.
Si 2p photoelectron line. Step (a)—Si(100) surfaces passivated
with a,u-bifunctional alkyl chain molecules. The Si 2p regions
are displayed after passivation of the [H-Si(100)] surfaces with
a,u-bifunctional alkyl chain molecules, resulting in azide-,
aldehyde-, and carboxylic-terminated surfaces ([N3-Si(100)],
[HOC-Si(100)], and [HO2C-Si(100)]), respectively, shown in
Fig. 1, spectra (a).
Two resolved spin–orbit split components result from curve-
fitting (not shown here), which are respectively related to the
contribution from bulk Si atoms, located at a binding energy
(BE) of 99.7 eV, and to a surface component, positively shifted
by 0.35 eV, i.e., falling in the region of Si–C and Si–H bonds.
The depth of the valley between the two spin–orbit components
(trough depth) is closely comparable between the different termi-
nated surfaces, as expected from the presence of Si–C bonds to
an equal extent. A flat line results in the energy region typical
for silica (103–104 eV), for both [HOC-Si(100)] and [HO2C-
Si(100)] surfaces, hinting at an oxide-free functionalised surface.
For the [N3-Si(100)] surfaces a modest SiO2 signal could be
detected, probably originating from the Si sites reacting with
residual traces of O2 and H2O still present during the preparation
of the samples in the Schlenk line. This step was necessary for the
substitution of the Br termination with the azide reactive
group.46 A quantification of the SiO2 amount was performed
applying a conventional attenuation model for ultrathin layers
of SiO2 grown on Si substrates based on the relative intensity
ratio between bulk and silica-related Si 2p peak components.
The results are reported in Table 1, where a value of 0.57 nm
is associated to the [N3-Si(100)] surface. Notably, no SiO2 has
been observed for the Br-, HOOC-, and OHC-terminated
surfaces (see Table 1). The low values of SiO2 thickness suggests
a fractional coverage of the surface by the oxide (as further
confirmed by electrochemical characterization, vide infra) and
[60]Fullerene dyads 3a–c were easily prepared through the 1,3-
dipolar cycloaddition of azomethine ylides (formed in situ
starting from the corresponding aldehyde (2a–c) in the presence
of amino acid 1) to C60 (Scheme 2).48 All compounds were fully
characterised by the standard spectroscopic and analytical
techniques, giving satisfactory results. In particular, the struc-
tures of all [60]fullerene-containing dyads 3a–c were assessed
by ESI mass spectrometry, 1H- and 13C-NMR, UV-VIS, and
IR spectroscopy. The broadening of the proton signals in the
NMR spectra can be attributed to the restricted rotation at
298 K suffered by the phenyl (molecules 3a and 3c) and the cyclo-
pentadienyl groups at the a-position of the fulleropyrrolidine
ring.49 Fulleropyrrolidine amine-bearing precursors 4a–c were
prepared by acidic cleavage of the tert-butoxycarbonyl (Boc)
protecting group in the presence of TFA following standard
synthetic protocols.45 Compound 4c was further remetallated
with a saturated solution of Zn(OAc)2 in MeOH.
Given the sensitivity of the monolayer structure towards
various external parameters (reaction conditions and substrate)
and the different types of averaging associated with various spec-
troscopic techniques, the use of several complementary methods
1574 | J. Mater. Chem., 2008, 18, 1570–1581
This journal is ª The Royal Society of Chemistry 2008

View Article Online
Fig. 2 XPS spectra taken for (a) [N3-Si(100)] and (b) [C60N-Si(100)]
surfaces in the C 1s region.
Fig. 1 XPS spectra taken in the Si 2p region. Top: (a) [HO2C-Si(100)]
and (b) [4b-Si(100)]. Middle: (a) [HOC-Si(100)] and (b) [5-Si(100)].
Bottom: (a) [N3-Si(100)] and (b) [C60N-Si(100)].
should be taken as an indication of the relative soft impact of the
preparation procedures on the final surfaces.
Step (b)—[60]Fullerene derivatisation of the modified Si(100)
surfaces. Upon reaction of the [HOC-Si(100)] and [HO2C-
Si(100)] surfaces with C60 and its conjugates 4a–c, respectively,
a slight increase of the signal in the region of silica was observed
for both materials, while for the [N3-Si(100)] surfaces the SiO2
thickness remained comparable (see curves (b) in Fig. 1). Quan-
tification of the Si oxide thicknesses (Table 1) revealed values of
0.74, 0.76, 0.47, and 0.48 nm for the [4b-Si(100)], [4c-Si(100)] (not
shown here), [5-Si(100)] and [C60N-Si(100)] surfaces, respec-
tively. Such values still allow electrical communication between
the Si bulk and the anchored redox species.
C 1s photoelectron line. Step (a)—Si(100) surfaces passivated
with a,u-bifunctional alkyl molecules. The C 1s region for the
[N3-Si(100)], [HOC-Si(100)], and [HO2C-Si(100)] surfaces are
displayed in Fig. 2–4, insets (a). Three different components
are visible for the [N3-Si(100)] sample. The higher peak at
285.4 eV is assigned to carbon atoms of the alkyl chain, while
minor components shifted by ꢁ0.6 and +1.6 eV are related to
C–Si and C–N terminal ends of the aliphatic chains. Both
features are in a 1 : 9 area ratio with the central peak, as expected
from the molecular stoichiometry when the N3– end-group
Fig. 3 XPS spectra taken for (a) [HOC-Si(100)] and (b) [5-Si(100)]
surfaces in the C 1s region.
remains unchanged. For the [HOC-Si(100)] surfaces, four
components were found by curve-fitting procedure, the higher
being related to methylene groups of the aliphatic chains, as in
the previous case. The minor components, shifted by ꢁ0.5,
+0.6, +1.3 eV with respect to the higher peak, are respectively
Table 1 SiO2 thickness as determined by XPS measurements
[Br-Si(100)]
[HO2C-Si(100)]
Not detected
[4b-Si(100)]
[4c-Si(100)]
[HOC-Si(100)]
Not detected
[5-Si(100)]
[N3-Si(100)]
0.57
[C60N-Si(100)]
0.48
dox/nm
Not detected
0.74
0.76
0.47
This journal is ª The Royal Society of Chemistry 2008
J. Mater. Chem., 2008, 18, 1570–1581 | 1575

View Article Online
from C–N bonds are found at +1.7 eV with respect to the
C60-cage atoms.33,52 XPS quantification for the three hybrids
yields peak intensities for the C–N and the chain C atoms in the
right stoichiometric ratio with respect to the carbons of the
fullerene moiety, supporting the proposed elemental composition
of the hybrid materials (Table 2).
N 1s photoelectron line. Fig. 5 (see also Table 3) shows the N 1s
region for the [N3-Si(100)] and [60]fullerene-terminated surfaces
([4b-Si(100)], [5-Si(100)] and [C60N-Si(100)]). The [N3-Si(100)]
substrate shows two peaks [curve (a)] at 401.6 and 405.4 eV,
with a 2 : 1 peak area ratio. The high- and low-BE peaks are
respectively assigned to the central, electron-deficient, N atom
of the N3 group and to the two nearly equivalent azide N
atoms.46,53 After reaction with C60 (curve (b)), the high-BE
peak disappears and a single component is found in the classical
energy range of tertiary aliphatic amines at 400.5 eV.52 This is in
agreement with the presence of aza[60]fullerene derivatives on
the surface as formed from the 1,3-dipolar cycloaddition reac-
tion of the N3 terminal group with C60 followed by the loss of
N2. The same BE value of 400.5 eV is found for the N 1s peak
of [5-Si(100)] surfaces, where a tertiary aminic group is supposed
Fig. 4 XPS spectra taken for (a) [HO2C-Si(100)] and (b) [4b-Si(100)]
surfaces in the C 1s region.
attributed to C–Si, C(–CHO) and C(]O)H carbon atoms. Each
of them is in a stoichiometric 1 : 8 area ratio with the alkyl-
centred peak, hinting at a covalent attachment to the Si(100)
surface through the reaction with the olefinic extremity. Four
different components in the same relative ratios were also found
for the [HO2C-Si(100)] surfaces, the higher, centred at 285.7 eV,
being related to the aliphatic chains of the molecules. The major
difference with the previous case is the presence of a high-BE
component (290.2 eV) that can be assigned to the C(]O)OH
ending group, accompanied by its a-C at 287.4 eV. These results
again support our hypothesis that the terminal alkene group
underwent covalent reaction with the Si surface. Step (b)—
[60]Fullerene functionalisation of the passivated Si(100) surfaces.
The C 1s spectra of the hybrid surfaces ([C60N-Si(100)], [5-
Si(100)] and [4b-Si(100)]) resulting from the reaction with C60
or its derivatives are shown in Fig. 2–4 (insets b). In all cases,
the C atoms of the fullerene cage predominantly contribute to
the peak at 285.3 eV,33 accompanied by a minor peak at 285.6
eV (alkyl chain). The former feature can be reproduced with
a Gaussian–Lorentzian mixed lineshape slightly exponentially
asymmetric towards high BEs.33,50 Two shake-up resonances of
low intensity shifted by +3.5 and +6.0 eV from the parent peak
are associated with the carbon cage atoms.33,51 The contribution
Fig. 5 XPS spectra taken in the N 1s region. Top: [4b-Si(100)]. Middle:
[5-Si(100)]. Bottom: (a) [N3-Si(100)] and (b) [C60N-Si(100)].
Table 2 XPS BE (eV)/FWHM (eV) data characterizing the C 1s photoelectron line
C 1s
[HO2C-Si(100)]
284.6/1.4
[4b-Si(100)]
[HOC-Si(100)]
284.9/1.1
[5-Si(100)]
[N3-Si(100)]
284.8/1.2
[C60N–Si(100)]
C–Si
C]C
C–C
C–N
285.4/1.3
285.6/1.3
287.1/1.5
285.3/1.3
285.6/1.3
287.0/1.2
285.3/1.2
285.6/1.2
287.0/1.3
285.7/1.4
285.4/1.1
285.4/1.2
287.0/1.2
C–COOH
COOH
C–CHO
CHO
287.4/1.4
290.2/1.4
286.0/1.1
286.7/1.1
1576 | J. Mater. Chem., 2008, 18, 1570–1581
This journal is ª The Royal Society of Chemistry 2008

View Article Online
Table 3 XPS BE (eV)/FWHM (eV) data characterizing the N 1s photoelectron line.
N 1s
[4b-Si(100)]
[5-Si(100)]
[4c-Si(100)]
[N3-Si(100)]
[C60N-Si(100)]
400.5/2.8
NR3
400.8/2.5
400.5/2.5
–N¼N+¼Nꢁ
–N¼N+¼Nꢁ
N/Zn
401.6/1.9
405.4/1.9
399.1/2.0
400.9/2.0
[(NH)4]2+
one can note that the corresponding N 1s BE falls at higher
values than that reported for pyrrolic nitrogens in free-base
structures, and its energy separation (N 1s–C 1s) of 115.6 eV is
consistent with the diacid derivative carrying a 2+ charge on
the ring (in Table 3 referred to as [(NH)4]2+) recently reported
in the literature.58 As to the N/Zn atomic ratio, it is worth noting
that only by adding both N 1s peaks is a value of 3.8, close to the
theoretical stoichiometric value of 4, obtained.56 Consistently
with our previous reports with porphyrins, all these results recall
for a demetallation reaction occurring upon anchoring the
molecule on the surface, leaving the porphyrin ring in two related
portions, while the Zn(II) ions remain adsorbed on the surface
(see Fig. S7 in the ESI‡). It is thought that, in the experimental
conditions adopted here, acidity can be locally generated at reac-
tion sites. The presence of protonated N atoms might thus result
from the acidic in-situ demetallation of the Zn(II)-porphyrin
complex. Finally, the low intensity peak located at 403.1 can
be associated to a shake-up feature, resulting from a core-ionized
valence-excited final state.59
Fig. 6 XPS spectrum taken for [4c-Si(100)] in the N 1s region.
to form upon the 1,3-dipolar cycloaddition reaction with C60.52
A small positive N 1s BE shift found in the case of [4b-Si(100)]
is compatible with the additional presence of a N atom employed
in an amidic linkage.52
The N 1s spectrum of [4c-Si(100)] is shown in Fig. 6. A free
base porphyrin has two symmetry-distinct and electronically
non-equivalent N atoms. As widely described in the literature,
these non-equivalent atoms are related to the unprotonated
and protonated porphyrinic N atoms.54 We have inferred from
previous XPS studies on various porphyrins that these two N
atoms present a constant BE separation from the macrocycle C
1s of 113 and 115 eV, respectively, with a small associated uncer-
tainty. Once the free-base porphyrin is complexed, the two
components converge into a single peak, i.e., four equivalent
metal-ligated N atoms, and the energy separation from C 1s
becomes 114 ꢃ 0.3 eV. Such a parameterised analysis allows
for direct comparison among different porphyrins, both before
and after anchoring, largely compensating for the electronic
effects of specific substituents attached to the porphyrin ring,
which can shift sizeably its BE.55 In our previous reports on
metalloporphyrins bound to Si(100) and Si(111) surfaces,56 we
found that the N 1s region can be a complex peak even for the
metallated porphyrins, where at least two components result
from curve fitting, apart from any low-intensity satellite line.
The result is unexpected, and it lacks mention in previous reports
for closely related systems.57 The Zn(II)-porphyrinate moiety
present in the hybrid materials investigated here shows a N 1s
XP spectrum (Fig. 6) where two main components can be located
at 399.1 and 400.9 eV, with an associated shift from the C 1s
peak of 113.7 and 115.6 eV, respectively (see Table 3). These
values allow for a straightforward assignment of the N 1s low-
BE component, which support the presence of N atoms in intact
porphyrinates. As already reported by us for different metallated
porphyrins, the N 1s high-BE component represents a true elec-
tronic state of the tetrapyrrolic system.56 As to its assignment,
Fe 2p photoelectron line. Inspection of the Fe 2p region of the
[4b-Si(100)] sample after curve-fitting (see Fig. 7 and Table 4)
reveals Fe species in both (II) and (III) oxidation states, at 708.7
Fig. 7 XPS spectrum taken for [4b-Si(100)] in the Fe 2p region.
Table 4 XPS BE (eV)/FWHM (eV) data characterizing the Fe 2p
photoelectron line
Fe(II) 2p3/2 Fe(III) 2p3/2 Shake-up Fe(II) 2p1/2 Fe(III) 2p1/2
[4b-Si(100)] 708.7/1.0 712.4/3.6
717.0/4.9 721.3/1.3 725.0/4.9
This journal is ª The Royal Society of Chemistry 2008
J. Mater. Chem., 2008, 18, 1570–1581 | 1577

View Article Online
and 712.4 eV respectively, with Fe(III) being predominant. These
values and the corresponding peak widths are consistent with
our previous findings on ferrocene derivatives anchored on
Si(100). The presence of Fe(III) has been invariably found for
the different ferrocene derivatives explored by us after covalent
attachment to Si(100).17,27,46,60 The high relative Fe(III)/Fe(II)
ratio found for the [4b-Si(100)] samples can be taken as the result
of the strong intramolecular charge-transfer type interaction
between the [60]fullerene and the ferrocenyl moiety. In fact,
such a high Fe(III)/Fe(II) ratio for freshly prepared Si-confined
ferrocene hybrids has never been detected under these conditions
by us, but only invariably observed after extensive and
prolonged redox cycling.27
voltammogram of molecule 3b in solution (see Fig. S8 in ESI‡),
three electrochemically, [60]fullerene-centred, reversible waves
are observed, and the redox potentials for the first, second and
third one-electron electrochemical processes are ꢁ0.93, ꢁ1.34
and ꢁ1.94 V, respectively. In addition, in the anodic scan, a
reversible redox wave centred at 0.22 V corresponding to the
Fc/Fc+ couple is also shown. Fig. 8 shows a typical CV trace
of a p-doped [4b-Si(100)] substrate in the cathodic region under
illumination conditions. Two partially reversible CV waves are
observed, and the redox potentials for the first and second
one-electron reductions are ꢁ0.80 and ꢁ1.18 V, respectively. It
is very likely that these peaks clearly resemble those centred on
the [60]fullerene moiety as has been shown for other [60]fullerene-
containing SAMs.61 The potential for the first [60]fullerene-
centred reduction of [4b-Si(100)] substrate is positively shifted
by 13 mV compared to that measured in solution.
Surface wettability
The [60]fullerene-containing surfaces were also characterised by
wetting contact angle (defined also as water contact angle or
contact angle of wettability) measurements, and the results are
reported in Table 5. The main information that can be extra-
polated from such an experiment is the hydrophobicity (as a
consequence of the presence of a lipophilic organic thin film)
of the surface by measurement of the contact angle (qa) formed
between a drop of water and the coated surface. The experimen-
tal contact angle of wettability on [H-Si(100)] (80 ꢃ 2ꢂ) is higher
than that of non-etched (prior to HF treatment) Si(100) surfaces
(10 ꢃ 2ꢂ), indicating the higher hydrophilicity of the latter
because of the presence of the SiO2 layer. After the linkage of
the [60]fullerene moieties, the water contact angle significantly
change to 66 ꢃ 4ꢂ, 70 ꢃ 3ꢂ, 78 ꢃ 2ꢂ, 81 ꢃ 1ꢂ and 100.9 ꢃ 0.3ꢂ
for [5-Si(100)], [C60N–Si(100)], [4a-Si(100)], [4b-Si(100)] and
[4c-Si(100)], respectively (Table 5), suggesting the presence of
the carbon-based hydrophobic layer that is consistent with our
structural model in which the all-carbon moieties are exposed
at the solid/air interface. The difference in the wettability proper-
ties between the [C60N-Si(100)] and [5-Si(100)] surfaces can be
ascribed to the different organisations of the carbon spheres
with respect to the external surface as a consequence of the
different linking groups (the aziridine-like and the pyrrolidine
ring for [C60N-Si(100)] and [5-Si(100)], respectively), while the
discrepancies between the [4a-Si(100)], [4b-Si(100)] and [4c-
Si(100)] surfaces are mainly attributed to the different hydro-
phobicities (N,N0-dimethylanilinyl < ferrocenyl < porphyrinyl)
of the R groups, which are also externally exposed at the solid/
air interface. These contact angles closely correspond to the values
reported in the literature for thin films of [60]fullerenes confined
on Au,23,32 Hg,36 GaAs,40 ITO,37 Si(111),43 and silicon oxides.38
Fig. 9 shows a typical CV of a [4b-Si(100)] substrate in the
anodic region. An electrochemically reversible one-electron
Fig. 8 Cyclic voltammogram of [4b-Si(100)] in CH3CN (+0.1 M Et4N-
ClO4). Scan rate: 1 V sꢁ1
.
Electrochemical measurements
The representative electrochemical studies of the hybrid surfaces
were essentially performed on the hybrid [4b-Si(100)] samples in
a dry 0.1 M solution of Et4NClO4 in CH3CN. In a typical cyclic
Fig. 9 Cyclic voltammogram of [4b-Si(100)] in CH3CN (+0.1 M Et4N-
ClO4). Scan rate: 1 V sꢁ1
.
Table 5 Advancing contact angles (qa) of water on [60]fullerene-containing Si(100) surfaces.
[C60N-Si(100)]
[4a-Si(100)]
78 ꢃ 2
[4b-Si(100)]
81 ꢃ 1
[4c-Si(100)]
100.9 ꢃ 0.3
[5-Si(100)]
66 ꢃ 4
[HO2C-Si(100)]
58 ꢃ 1
SiO2
[H-Si(100)]
qa
70 ꢃ 4
10 ꢃ 2
80 ꢃ 2
1578 | J. Mater. Chem., 2008, 18, 1570–1581
This journal is ª The Royal Society of Chemistry 2008

View Article Online
process is observed, which can be unambiguously attributed to
the ferrocenyl moiety. The Eꢂ value estimated from the mean
of the anodic and cathodic peak potentials, was 225 mV vs.
Ag/Ag+. In order to study the role of the charge carriers, CV
measurements were performed in the dark and under light illumi-
nation (red laser light). No limiting effects were observed under
illumination conditions, indicating that with such silicon
substrates the electron transfer is not limited by the number of
charge carriers and it can occur via the valence band. The first
CV scan profile shows very symmetric Fc/Fc+ ox/red peaks,
with a full width at half maximum (DEFWHM) of about 100 mV
and 9 mV of potential splitting (DEpp ¼ Epa ꢁ Epc) between the
anodic (Epa) and cathodic (Epc) peak potentials. Such data are
comparable to those obtained with Si(100) surfaces covalently
modified following a similar photochemical reaction as that
used in this work using a vinylferrocene derivative, indicating
reversible behaviour of the SAM attached to the silicon surface.17
Notably, no evidence for substantial repulsive interactions
between neighboring ferrocenyl groups62 was observed due to
the relatively narrow width (< 100 mV) of the CV waves. This find-
ing supports the model for which all molecules constituting the
monolayer are very much accessible for electron transfer from
and to the substrate, behaving independently within the layer.
Representative CVs for the [4b-Si(100)] surfaces in CH3CN
(+0.1 M Et4NClO4) at different scan rates ranging from 0.5 to
10 V sꢁ1 are shown in Fig. 10 (top). As can be inferred from
the inspection of Fig. 10 (bottom), the anodic and cathodic
peak currents were found to linearly scale with the scan rate v
rather than with v1/2, indicating a surface-confined redox process.
This trend is characteristic of a reversible (Nernstian) electro-
chemical process, wherein the relative activity of the Fc/Fc+
couple is uniform throughout the film and at the equilibrium
with each applied electrode potential. The surface coverage
estimated on the scan-rate dependence of the anodic and cathodic
peak currents was found to be approximately 2.5 ꢀ 10ꢁ11 mol
cmꢁ2, which is lower than the theoretically and experimentally
determined value of a densely organized fcc-packed monolayer
of pristine C60 (1.9 ꢀ 10ꢁ10 mol cmꢁ2). This value was found to
be similar to that (1.4 ꢀ 10ꢁ11 mol cmꢁ2) reported by Feng and
Miller for their C60-functionalised Si(100) surfaces.43
Repeated CV experiments showed that after 120 consecutive
potential scans of the [4b-Si(100)] surface in the range between
0.0 and 0.5 V did not significantly change the shape of the CV
traces (Fig. 11). These results clearly indicated that the developed
protocol yields stable modified [4b-Si(100)] surfaces. Addition-
ally, no major shifts of the oxidation peak were observed indica-
ting that the energy and the chemical nature of the oxidation
process remained unaltered. In contrast, the cathodic peak
shifted to more negative potentials as the number of cycles
increased, showing that the reduction process becomes energeti-
cally less favourable (Fig. 12). We tentatively attribute this effect
to the tendency of ionic species such as the Fc+ cations to couple
with anions bound to the surface (for example such as silanol
groups or surface-adsorbed negatively charged species coming
Fig. 11 Cyclic voltammograms for a [4b-Si(100)] electrode in CH3CN
(+0.1 M Et4NClO4): first cycle (––) and after 120 consecutive potential
scans (- - -). Scan rate: 1 V sꢁ1
.
Fig. 10 Top: Cyclic voltammogram of [4b-Si(100)] in CH3CN (+0.1 M
Et4NClO4) recorded at different potential scan rates in the range between
0.5 and 10 V sꢁ1. Bottom: linear dependence of the anodic (-) and
cathodic (d) peak currents on the potential scan rate.
Fig. 12 Dependence of the anodic (B, Epa), cathodic (,, Epc), and
peak–peak separation (D) DEpp potentials of the Fc/Fc+ couple on the
scan rate.
This journal is ª The Royal Society of Chemistry 2008
J. Mater. Chem., 2008, 18, 1570–1581 | 1579

View Article Online
a desired potential range. In principle, such [60]fullerene-confined
surfaces could be used as electronic components for molecular-
based storage devices and for medical-/bio-applications acting
as radical scavengers for the protection of cells against death.
Acknowledgements
This work was supported by the European Union through the
Marie-Curie Research Training Network ‘‘PRAIRIES’’,
contract MRTN-CT-2006-035810, the University of Trieste,
INSTM, MUR (PRIN 2006, prot. 20064372 and FIRB, prot.
RBIN04 HC3S), and the University of Namur. E.A.D. carried
out this work with the support of the ICTP Programme for
Research and Training in Italian Laboratories (Associate
Member), Trieste, Italy, which is kindly acknowledged. We
thank Prof. Zineb Mekhalif (University of Namur) for the access
to the water contact angle instrument and Dr Fabio Hollan
(University of Trieste) for the mass spectrometry measurements.
Fig. 13 Cyclic voltammograms of [4b-p-Si(100)] (––) and [4b-n-Si(100)]
(- - -) in CH3CN (+0.1 M Et4NClO4). Scan rate: 1 V sꢁ1
.
from the electrolyte), resulting in a final stabilisation of the Fc+
species.
References
In the functionalised [4b-nSi(100)] surfaces, in contrast to what
observed on [4b-pSi(100)], the electrochemically active redox
processes were controlled by the photon flux (i.e., the current
of the redox process is proportional to the light intensity shone
on the electrode). The redox peaks were found to be shifted by
ca. 0.4 V towards negative values with respect to the reference
redox potentials as measured for the functionalised pSi(100)
(Fig. 13). The peak current at nSi(100) surfaces (see ESI‡
Fig. S9) is limited by the photon flux, i.e. the irradiation of the
sample with a low-intensity light emphasises the molecular redox
behaviour over the Si(100) oxidation reaction under intense illu-
mination conditions. This explains why such currents are lower
than the peak currents measured for the p-Si surfaces (ESI‡
Fig. S9), due to the thermalised holes already abundant at the
p-Si valence band edge in the dark. A direct comparison of the
CV profiles of the two electrodes is shown in Fig. 13. The peak
potentials, compared to that of the Fc/Fc+ redox couple
dissolved in the same supporting electrolyte and measured at
a Pt wire electrode (0.08 V), indicate the occurrence of a slight
negative and positive photovoltage shift with the n-Si and p-Si
surfaces, respectively. From such data an overlapping of the
valence band of the p-Si with the redox level of the Fc/Fc+ couple
of the grafted fullerene-ferrocene dyads is inferred.
1 P. A. Packan, Science, 1999, 285, 2079.
2 C. B. Gorman and R. L. Carroll, Angew. Chem., Int. Ed., 2002, 41,
4378.
3 N. Mathur, Nature, 2002, 419, 573.
4 J. V. Barth, G. Costantini and K. Kern, Nature, 2005, 437, 671;
M. C. Hersam, N. P. Guisinger and J. W. Lyding, Nanotechnology,
2000, 11, 70; C. Joachim and M. A. Ratner, Proc. Natl. Acad. Sci.
U. S. A., 2005, 102, 8800; C. Joachim and M. A. Ratner, Proc.
Natl. Acad. Sci. U. S. A., 2005, 102, 8801; C. Joachim, Nat. Mater.,
2005, 4, 107; R. M. Metzger, Anal. Chim. Acta, 2006, 568, 146;
G. F. Cerofolini, G. Arena, M. Camalleri, C. Galati, S. Reina,
L. Renna, D. Mascolo and V. Nosik, Microelectron. Eng., 2005, 81,
405; G. F. Cerofolini, G. Arena, C. M. Camalleri, C. Galati,
S. Reina, L. Renna and D. Mascolo, Nanotechnology, 2005, 16, 1040.
5 G. Ashkenasy, D. Cahen, R. Cohen, A. Shanzer and A. Vilan,
Acc. Chem. Res., 2002, 35, 121.
6 O. Seitz, A. Vilan, H. Cohen, C. Chan, J. Hwang, A. Kahn and
D. Cahen, J. Am. Chem. Soc., 2007, 129, 7494.
7 A. Salomon, T. Boecking, O. Seitz, T. Markus, F. Amy, C. Chan,
W. Zhao, D. Cahen and A. Kahn, Adv. Mater., 2007, 19, 445;
N. Armstrong, R. C. Hoft, A. McDonagh, M. B. Cortie and
M. J. Ford, Nano Lett., 2007, 7, 3018.
8 A. Salomon, T. Bocking, J. Gooding and D. Cahen, Nano Lett., 2006,
6, 2873.
9 J. E. Anthony, Chem. Rev., 2006, 106, 5028; A. Salomon, D. Cahen,
S. Lindsay, J. Tomfohr, V. B. Engelkes and C. D. Frisbie,
Adv. Mater., 2003, 15, 1881.
10 A. Vilan and D. Cahen, Trends Biotechnol., 2002, 20, 22; A. Vilan and
D. Cahen, Adv. Funct. Mater., 2002, 12, 795; A. Vilan, J. Ghabboun
and D. Cahen, J. Phys. Chem. B, 2003, 107, 6360; A. Vilan,
A. Shanzer and D. Cahen, Nature, 2000, 404, 166.
11 J. M. Buriak, Chem. Rev., 2002, 102, 1271.
12 D. D. M. Wayner and R. A. Wolkow, J. Chem. Soc., Perkin Trans. 2,
2002, 23.
13 R. J. Hamers and Y. J. Wang, Chem. Rev., 1996, 96, 1261;
H. N. Waltenburg and J. T. Yates, Chem. Rev., 1995, 95, 1589;
A. B. Sieval, A. L. Demirel, J. W. M. Nissink, M. R. Linford,
J. H. van der Maas, W. H. de Jeu, H. Zuilhof and
E. J. R. Sudholter, Langmuir, 1998, 14, 1759.
14 M. R. Linford, P. Fenter, P. M. Eisenberger and C. E. D. Chidsey,
J. Am. Chem. Soc., 1995, 117, 3145.
15 M. R. Linford and C. E. D. Chidsey, J. Am. Chem. Soc., 1993, 115,
12631.
16 G. G. Condorelli, A. Motta, M. Favazza, I. L. Fragala, M. Busi,
E. Menozzi, E. Dalcanale and L. Cristofolini, Langmuir, 2006, 22,
11126; M. Busi, M. Laurenti, G. G. Condorelli, A. Motta,
M. Favazza, I. L. Fragala, M. Montalti, L. Prodi and
E. Dalcanale, Chem.–Eur. J., 2007, 13, 6891; A. Gulino,
S. Bazzano, P. Mineo, E. Scamporrino, D. Vitalini and I. Fragala,
Conclusions
In summary, we have shown that different two-step synthetic
methodologies can be used to covalently immobilise C60 and
its derivatives on flat hydrogenated Si(100) surfaces. All [60]ful-
lerene-modified surfaces were fully investigated by means of XPS
spectroscopy and water contact angle measurements. In particu-
lar, CV measurements performed with the [4b-Si(100)] substrate,
clearly showed the formation of exceptionally stable hybrid
materials exhibiting reversible electrochemical behaviour.
Although more work is needed in order to improve the techno-
logy in terms of surface coverage, the synthetic methodologies
developed here could open new possibilities of preparing func-
tionalized hybrid silicon surfaces bearing redox-active molecules
capable of undergoing multi-electron transfer processes over
1580 | J. Mater. Chem., 2008, 18, 1570–1581
This journal is ª The Royal Society of Chemistry 2008

View Article Online
Chem. Mater., 2005, 17, 521; M. Bollani, R. Piagge and D. Narducci,
Mater. Sci. Eng., C, 2001, 15, 253.
34 M. Matsumoto, J. Inukai, S. Yoshimoto, Y. Takeyama, O. Ito and
K. A. Itaya, J. Phys. Chem. C, 2007, 111, 13297; Y. Shirai,
L. Cheng, B. Chen and J. M. Tour, J. Am. Chem. Soc., 2006, 128,
13479.
35 D. Bonifazi, A. Kieble, M. Stoehr, F. Cheng, T. Jung, F. Diederich
and H. Spillmann, Adv. Funct. Mater., 2007, 17, 1051.
36 F. Song, S. Zhang, D. Bonifazi, O. Enger, F. Diederich and
L. Echegoyen, Langmuir, 2005, 21, 9246.
37 K. Chen, W. B. Caldwell and C. A. Mirkin, J. Am. Chem. Soc., 1993,
115, 1193.
38 V. V. Tsukruk, L. M. Lander and W. J. Brittain, Langmuir, 1994, 10,
996; L. M. Lander, W. J. Brittain and V. Tsukruk, Polym. Prepr.,
1994, 35, 488.
39 A. Gulino, S. Bazzano, G. G. Condorelli, S. Giuffrida, P. Mineo,
C. Satriano, E. Scamporrino, G. Ventimiglia, D. Vitalini and
I. Fragala, Chem. Mater., 2005, 17, 1079.
40 D. Bonifazi, A. Salomon, O. Enger, D. Cahen and F. Diederich,
Adv. Mater., 2002, 14, 802.
41 W. J. Feng and B. Miller, Electrochem. Solid-State Lett., 1998, 1, 172;
D. Dattilo, L. Armelao, M. Maggini, G. Fois and G. Mistura,
Langmuir, 2006, 22, 8764.
`
42 L. Ostrovskaya, A. Podesta, P. Milani and V. Ralchenko, Europhys.
Lett., 2003, 63, 401.
17 F. Decker, F. Cattaruzza, C. Coluzza, A. Flamini, A. G. Marrani,
R. Zanoni and E. A. Dalchiele, J. Phys. Chem. B, 2006, 110, 7374.
18 N. Tajimi, H. Sano, K. Murase, K.-H. Lee and H. Sugimura,
Langmuir, 2007, 23, 3193; B. Fabre and F. Hauquier, J. Phys.
Chem. B, 2006, 110, 6848.
19 L. Wei, D. Syomin, R. S. Loewe, J. S. Lindsey, F. Zaera and
D. F. Bocian, J. Phys. Chem. B, 2005, 109, 6323.
20 F. Cattaruzza, A. Flamini, D. Fiorani, P. Imperatori, G. Scavia,
L. Suber, A. M. Testa, A. Mezzi, G. Ausanio and W. R. Plunkett,
Chem. Mater., 2005, 17, 3311; C. Altavilla, E. Ciliberto,
D. Gatteschi and C. Sangregorio, Adv. Mater., 2005, 1084;
Z. Y. Zhong, B. Gates, Y. N. Xia and D. Qin, Langmuir, 2000, 16,
10369.
21 J. L. He, B. Chen, A. K. Flatt, J. J. Stephenson, C. D. Doyle and
J. M. Tour, Nat. Mater., 2006, 5, 63.
22 N. Shirat, T. Miura, K. Kobayashi and K. Koumoto, Langmuir,
2003, 19, 9107; F. Cattaruzza, A. Cricenti, A. Flamini, M. Girasole,
G. Longo, T. Prosperi, G. Andreano, L. Cellai and E. Chirivino,
Nucleic Acids Res., 2006, 34, 32; B. R. Hart, S. E. Letant,
S. R. Kane, M. Z. Hadi, S. J. Shields and J. G. Reynolds, Chem.
Commun., 2003, 322; A. R. Pike, L. C. Ryder, B. R. Horrocks,
W. Clegg, B. A. Connolly and A. Houlton, Chem.–Eur. J., 2004,
11, 344.
23 N. Higashi, T. Inoue and M. Niwa, Chem. Ber., 1997, 1507.
24 Q. Y. Sun, L. de Smet, B. van Lagen, M. Giesbers, P. C. Thune, J. van
Engelenburg, F. A. de Wolf, H. Zuilhof and E. J. R. Sudholter, J. Am.
Chem. Soc., 2005, 127, 2514; F. Effenberger, G. Gotz, B. Bidlingmaier
and M. Wezstein, Angew. Chem., Int. Ed., 1998, 37, 2462;
R. Boukherroub and D. D. M. Wayner, J. Am. Chem. Soc., 1999,
121, 11513; W. J. Leigh, C. Kerst, R. Boukherroub, T. L. Morkin,
S. I. Jenkins, K. S. Sung and T. T. Tidwell, J. Am. Chem. Soc.,
1999, 121, 4744.
43 W. J. Feng and B. Miller, Langmuir, 1999, 15, 3152.
44 D. Hammel, C. Kautz and K. Mullen, Chem. Ber., 1990, 123, 1353.
45 K. Kordatos, T. Da Ros, S. Bosi, E. Vazquez, M. Bergamin,
C. Cusan, F. Pellarini, V. Tomberli, B. Baiti, D. Pantarotto,
V. Georgakilas, G. Spalluto and M. Prato, J. Org. Chem., 2001, 66,
4915.
46 A. G. Marrani, E. A. Dalchiele, R. Zanoni, F. Decker, F. Cattaruzza,
D. Bonifazi and M. Prato, Electrochim. Acta, 2007, DOI: 10.1016/
j.electacta.2007.10.051.
47 D. Briggs and M. P. Seah, Practical Surface Analysis, J. Wiley &
Sons, Chichester, 2nd edn, 1990, vol. 1.
25 A. Aurora, F. Cattaruzza, C. Coluzza, C. Della Volpe, G. Di Santo,
A. Flamini, C. Mangano, S. Morpurgo, P. Pallavicini and R. Zanoni,
Chem.–Eur. J., 2007, 13, 1240; D. Wang and J. M. Buriak, Surf. Sci.,
2005, 590, 154.
26 M. P. Stewart and J. M. Buriak, Angew. Chem., Int. Ed., 1998, 37,
3257; R. Boukherroub, S. Morin, F. Bensebaa and
D. D. M. Wayner, Langmuir, 1999, 15, 3831.
48 M. Maggini, G. Scorrano and M. Prato, J. Am. Chem. Soc., 1993,
115, 9798; M. Prato and M. Maggini, Acc. Chem. Res., 1998, 31,
519; N. Tagmatarchis and M. Prato, Synlett, 2003, 6, 768.
49 F. Ajamaa, T. M. F. Duarte, C. Bourgogne, M. Holler, P. W. Fowler
and J. F. Nierengarten, Eur. J. Org. Chem., 2005, 3766.
50 A. J. Maxwell, P. A. Bruhwiler, A. Nilsson, N. Martensson and
P. Rudolf, Phys. Rev. B, 1994, 49, 10717.
27 R. Zanoni, F. Cattaruzza, C. Coluzza, E. A. Dalchiele, A. Flamini,
F. Decker, G. Di Santo, L. Funari and A. G. Marrani, Surf. Sci.,
2005, 575, 260; E. A. Dalchiele, A. Aurora, G. Bernardini,
F. Cattaruzza, A. Flamini, P. Pallavicini, R. Zanoni and F. Decker,
J. Electroanal. Chem., 2005, 579, 133.
51 J. A. Leiro, M. H. Heinonen, T. Laiho and I. G. Batirev, J. Electron
Spectrosc. Relat. Phenom., 2003, 128, 205.
52 D. Briggs and G. Beamson, High resolution XPS of organic
polymers—the Scienta ESCA300 database, John Wiley & Sons,
Chichester, 1992.
28 J. Jiao, F. Anariba, H. Tiznado, I. Schmidt, J. S. Lindsey, F. Zaera
and D. F. Bocian, J. Am. Chem. Soc., 2006, 128, 6965; K. M. Roth,
J. S. Lindsey, D. F. Bocian and W. G. Kuhr, Langmuir, 2002, 18,
4030.
29 T. M. Figueira-Duarte, A. Gegout and J. F. Nierengarten, Chem.
Commun., 2007, 109.
30 J. F. Nierengarten, Sol. Energy Mater. Sol. Cells, 2004, 83, 187;
J. L. Segura, N. Martin and D. M. Guldi, Chem. Soc. Rev., 2005,
34, 31; T. Konishi, A. Ikeda and S. Shinkai, Tetrahedron, 2005, 61,
4881; D. M. Guldi, Chem. Soc. Rev., 2002, 31, 22; N. Armaroli,
Photoinduced Energy Transfer Processes in Functionalized
Fullerenes, in Fullerenes: From Synthesis to Optoelectronic
Properties, ed. D. Guldi and N. Martin, Kluwer Academic,
Dordrecht, 2002, p. 137; M. Prato, J. Mater. Chem., 1997, 7, 1097;
R. M. Metzger, Chem. Rev., 2003, 103, 3803; R. M. Metzger,
Synth. Met., 2001, 124, 107.
53 J. P. Collman, N. K. Deveraj, T. P. A. Eberspacher and
C. E. D. Chidsey, Langmuir, 2006, 22, 2457.
54 D. Karweik, N. Winograd, D. G. Davis and K. M. Kadish, J. Am.
Chem. Soc., 1974, 98, 591; Y. Niwa, H. Kobayashi and
T. Tsuchiya, J. Chem. Phys., 1974, 60, 799; M. Seno, S. Tsuchiya
and S. Ogawa, J. Am. Chem. Soc., 1977, 99, 3014; J. P. Macquet,
M. M. Millard and T. Theophanides, J. Am. Chem. Soc., 1978, 100,
4741; P. G. Gassman, A. Ghosh and J. Almlof, J. Am. Chem. Soc.,
1992, 114, 9990.
55 J. G. Goll, K. T. Moore, A. Ghosh and M. J. Therien, J. Am. Chem.
Soc., 1996, 118, 8344.
56 R. Zanoni, A. Aurora, F. Cattaruzza, F. Decker, P. Fastiggi,
V. Menichetti, P. Tagliatesta, A. L. Capodilupo and A. Lembo,
Mater. Sci. Eng., C, 2007, 27, 1351.
57 K. M. Roth, A. A. Yasseri, Z. M. Liu, R. B. Dabke, V. Malinovskii,
K. H. Schweikart, L. H. Yu, H. Tiznado, F. Zaera, J. S. Lindsey,
W. G. Kuhr and D. F. Bocian, J. Am. Chem. Soc., 2003, 125, 505.
58 H. Yamashige, S. Matsuo, T. Kurisaki, R. C. C. Perera and
H. Wakita, Anal. Sci., 2005, 21, 635.
59 A. Ghosh, J. Fitzgerald, P. G. Gassman and J. Almlof, Inorg. Chem.,
1994, 33, 6057.
60 M. Cossi, M. F. Iozzi, A. G. Marrani, T. Lavecchia, P. Galloni,
R. Zanoni and F. Decker, J. Phys. Chem. B, 2006, 110, 22961.
61 V. T. Hoang, L. M. Rogers and F. D’Souza, Electrochem. Commun.,
2002, 4, 50.
31 D. Bonifazi, O. Enger and F. Diederich, Chem. Soc. Rev., 2007, 36,
390.
32 W. B. Caldwell, K. Chen, C. A. Mirkin and S. J. Babinec, Langmuir,
1993, 9, 1945; Y. S. Shon, K. F. Kelly, N. J. Halas and T. R. Lee,
Langmuir, 1999, 15, 5329; O. Enger, F. Nuesch, M. Fibbioli,
L. Echegoyen, E. Pretsch and F. Diederich, J. Mater. Chem., 2000,
10, 2231; M. Fibbioli, K. Bandyopadhyay, S.-G. Liu,
L. Echegoyen, O. Enger, F. Diederich, D. Gingery, P. Buhlmann,
¨
H. Persson, U. W. Suter and E. Pretsch, Chem. Mater., 2002, 14,
1721.
33 D. Benne, E. Maccallini, P. Rudolf, C. Sooambar and M. Prato,
Carbon, 2006, 44, 2896.
62 R. D. Eagling, J. E. Bateman, N. J. Goodwin, W. Henderson,
B. R. Horrocks and A. Houlton, J. Chem. Soc., Dalton Trans.,
1998, 1273.
This journal is ª The Royal Society of Chemistry 2008
J. Mater. Chem., 2008, 18, 1570–1581 | 1581