- Email: [email protected]
Modification of fluorine-doped tin oxide surface: Optimization of the electrochemical grafting of diazonium salt
Accepted Manuscript
Modification of Fluorine-Doped Tin Oxide surface: optimization of the electrochemical grafting of diazonium salt Van Bui-Thi-Tuyet , Caroline Cannizzo , Corinne Legros , Michel Andrieux , Annie Chausse´ PII: DOI: Reference:
S2468-0230(18)30452-8 https://doi.org/10.1016/j.surfin.2019.01.012 SURFIN 289
To appear in:
Surfaces and Interfaces
Received date: Revised date: Accepted date:
17 September 2018 23 January 2019 28 January 2019
Please cite this article as: Van Bui-Thi-Tuyet , Caroline Cannizzo , Corinne Legros , Michel Andrieux , Annie Chausse´ , Modification of Fluorine-Doped Tin Oxide surface: optimization of the electrochemical grafting of diazonium salt, Surfaces and Interfaces (2019), doi: https://doi.org/10.1016/j.surfin.2019.01.012
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Highlights
AC
CE
PT
ED
M
AN US
CR IP T
Covalent functionalization of Fluorine Tin Oxide by carboxyphenyl groups was made by via electroreduction of diazonium salt Parameters such as grafting potential and grafting time were investigated Investigation of FTO surface by XPS and FTIR shows the presence of caboxyphenyl groups at the FTO surface Electrochemical behavior of FTO together with electrochemical characterization in the presence of an electroactive probe allowed us to conclude to the grafting of a poorlypacked organic multi-layer
1
ACCEPTED MANUSCRIPT
Modification of Fluorine-Doped Tin Oxide surface: optimization of the electrochemical grafting of diazonium salt
Van Bui-Thi-Tuyeta, Caroline Cannizzoa, Corinne Legrosb, Michel Andrieuxb and Annie Chausséa LAMBE, Univ Evry, CNRS, CEA, Université Paris-Saclay, 91025, Evry, France
b
ICMMO, Univ Paris-Sud, CNRS, Université Paris-Saclay, 91405, Orsay, France *
CR IP T
a
Corresponding author. Tel.:+33(0)1 69 47 02 22. E-Mail address: [email protected]
AN US
Abstract
Fluorine-doped tin oxide (FTO) electrodes were functionalized by electro reduction of in situ generated 4-carboxyphenyl diazonium salt. Several grafting potentials and times were investigated in order to find the optimal grafting conditions. The electrochemical behavior of bare FTO
M
electrodes in the presence of the diazonium, together with Cyclic Voltammetry (CV) and
ED
Electrochemical Impedance Spectroscopy (EIS) characterization of the grafted electrodes in the presence of redox probes ferricyanide and ferrocene allowed us to conclude that 10 minutes
PT
grafting under a controlled potential of -0.50V lead to the formation of a poorly-packed organic multi-layer on the electrode surface. X-ray Photoelectron Spectroscopy (XPS), together with
CE
Fourier-Transform Infrared Spectroscopy (FTIR) analyses, further evidenced the presence of
AC
carboxylic groups onto modified FTO electrodes.
Keywords
Fluorine-doped tin oxide- FTO- Electrochemical grafting – Diazonium salt- Surface modification
2
ACCEPTED MANUSCRIPT 1. Introduction Among various electrode surfaces, tin oxide (SnO2) has been extensively used in both research and industry for applications such as gas sensors, [1] solar cells [2] and heat mirrors [3]. Commercially, SnO2 is doped by indium oxide (In2O3 :SnO2) or fluorine (SnO2 :F) [4]. Despite the fact that indium tin oxide is the most used, there exists some limitations such as high cost because
CR IP T
of low indium availability [5] and pH sensitivity [6]. For those reasons, Fluorine-doped tin oxide (FTO) may thus be preferred [7, 8] as a prospective optically transparent and electrically conductive oxide [9]. The electro conductivity is achieved by fluorine atoms [10] that give stability to the chemical attack, also at low pH. Its interesting properties make FTO a non-toxic,
various industrial applications [12-14].
AN US
fully biodegradable and cost-affordable material [11], therefore promising for the development of
In literature, a few recent papers concerning FTO modification are available. Bisset et al. reported
M
the grafting of oxidized carbon nanotubes on hydroxylated FTO [15-17]. Hamers et al. focused on the reaction of thiolated or iodated bipyrydiyl compounds with the same substrate [18]. SnO2 was
ED
functionalized by photochemical grafting, followed by click chemistry [19]; electrografting of PEDOT [20] or vinyl-containing poly-pyridyl complexes [21] onto FTO were also reported.
PT
Electrografting of diazonium salts is a convenient way to bring covalently attached functions at
CE
the surface of various materials [22]. Surprisingly, to our knowledge, only a few studies dealing with the grafting of diazonium salts onto FTO are available: the grafting of iodonium [23] or
AC
diazonium salts [23, 24] was considered for the development of dye-sensitized solar cells ; the electrografting of aryldiazonium cations for the development of biological–based sensor devices was also reported [25, 26]. All these reactions were conducted by cyclic voltammetry in either acetonitrile [23, 24] or aqueous HCl medium [25, 26], optimizing the number of cycles. The grafting of diazonium salt at fixed potential was, to our knowledge, only reported once for a parent material : Meyer et al. realized the grafting a Ru(II) diazonium dye on various semi-conductor
3
ACCEPTED MANUSCRIPT materials, including Sn02 [27]. The electrografting time and potential were optimized in acetonitrile, and its feasibility was also demonstrated at one chosen potential in aqueous HCl, appearently with similar characteristics of the grafted layer. The presence of a “turn-off” potential, under which the grafting of diazonium salt is no more possible, was mentioned. It thus appeared important to make a complete study of the optimal grafting conditions of diazonium salts onto
CR IP T
FTO surface in aqueous acidic conditions, and to determine the properties of the as-obtained grafted layer. Herein, starting from the corresponding amine precursor, we studied the immobilization of 4-carboxyphenyldiazonium (Scheme 1) on FTO by means of electrografting, varying the reduction time and potential in order to optimize the grafting conditions. The attached
AN US
organic layer was characterized by means of Cyclic Voltammetry (CV) and Electrochemical Impedance Spectroscopy (EIS) in both aprotic and protic solvent in the presence of two redox probes, one negatively charged (hexacyanoferricyanide (III)) and the other neutral (ferrocene).
M
These easy to handle electrochemical techniques could bring interesting information not only on the presence of the attached organic layer, but also on its thickness and homogeneity. The
ED
presence of carboxylic groups at the FTO surface was further evidenced both by X-ray
AC
CE
PT
Photoelectron Spectroscopy (XPS) and Fourier Transform Infra-Red spectroscopy (FT-IR).
Scheme 1. In situ 4-carboxyphenyl diazonium generation and electrografting on FTO.
4
ACCEPTED MANUSCRIPT 2. Materials and Methods 2.1 Chemicals Chlorhydric acid standard solution (1 M) was purchased from Fluka, dry acetronitrile, tetrabutylammoniumtetrafluoroborate, sodium nitrite, potassium chloride, p-aminobenzoic acid and ferrocene from Aldrich. All aqueous solutions were prepared with purified water (18.2 MΩ
CR IP T
cm-1) from Millipore.
2.2 Electrochemical measurements
A conventional three-electrode cell was used for the electrochemical experiments. Platinum wire
AN US
was used as auxiliary electrode, saturated calomel electrode (SCE) as reference electrode. FTO glass substrates purchased from SOLEMS (France ) were used as working electrodes after isolation of 1cm2 with a solution of polystyrene in mesitylene. Prior to use, the working electrodes
M
were rinsed and sonicated in H2SO4 0.1M during 10 minutes. Experiments were performed with an Autolab (low current PGSTAT302N, Metrohm) at room temperature. All solutions were
ED
deaerated with argon bubbling during 20 minutes prior to the electrochemical experiments. EIS measurements were performed using EG&G (273A, Princeton Applied Research) coupled with a
PT
SI 1255 Frequency Response Analyzer (Solartron). EIS measurements were recorded at room
CE
temperature within the frequency range of 102-106 Hz superimposed on a DC potential of -186 mV for aqueous solution of K3[Fe(CN)6] (1 mM + 0.1 M KCl) and of 420 mV for dry acetonitrile of
AC
Ferrocene (1 mM + 0.1M NBu4BF4) , with AC of 10 mV peak to peak amplitude, and 5 points per decade of frequencies. Three distinct experiments were carried out for each measurement.
2.3 IR characterization Fourier-Transform Infra-Red (FTIR) was carried out with a Vertex 70 Bruker spectrometer equipped with an MCT detector and using the Attenuated Total Reflectance (ATR) technique
5
ACCEPTED MANUSCRIPT which enables samples to be examined directly. Germanium crystal has, by far, the highest refractive index of all the ATR materials available, and the effective depth of penetration is approximately 1 µm. In order to obtain effective signal/noise ratio, each spectrum was obtained by averaging 400 scans in the wavelength range of 400-4000 cm-1 at a spectral resolution of 4 cm-1. Absorption peaks search was automatically realized using the OPUS 7.0 software. As scan spectra
with IR database or data published in the literature.
CR IP T
are of low intensity, all spectra have been normalized to be weighed. The results were compared
Nota : The biggest difference between ATR and Transmission spectra can be the shift of strong absorbing bands moving to lower wavenumbers in ATR mode. For most compounds, the shifts
AN US
observed are small, approximately 1-10 cm-1, but for very strong absorbing compounds, the shifts observed can be as big as 30-50 cm-1. These observed changes are in accordance with the
M
established theory of ATR spectroscopy.
2.4 XPS characterization
ED
XPS measurements were performed on a K Alpha spectrometer (ThermoFisher), equipped with a monochromated X-Ray Source (AlKα, 1486.6 eV). A spot size of 400µm was used. The
PT
hemispherical analyzer was operated in CAE (Constant Analyzer Energy) mode, with a pass
CE
energy of 200 eV and a step of 1 eV for the acquisition of survey spectra, and a pass energy of 50 eV and a step of 0.1 eV for the acquisition of high resolution spectra. A “dual beam” flood gun
AC
was used to neutralize the charge built-up. The spectra were monitored and processed using Advantage software from ThermoFisher. A Shirley type background subtraction and a mixture of Gaussian- Lorentzian shapes (30/70) were used for peak fitting. The peak areas were normalized using the Scofield sensitivity factors.
6
ACCEPTED MANUSCRIPT 3. Results and discussion 3.1 Potential window of FTO and diazotization process The choice of the electrografting potential depends on the potential window of bare electrode in electrolyte solution. Prior to the electrochemical modification of FTO surface, the potential window of FTO was thus checked in 0.5M HCl solution. Fig. 1A shows that the deposition
CR IP T
potential for reductive grafting of diazonium salt should not be conducted below -0.70V, as at lower potential reduction of the electrolyte occurs. Cyclic voltammetry on bare FTO surface was next performed in the presence of 4-carboxyphenyl diazonium salt. The diazonium salt was directly generated in situ in the electrochemical cell, starting from the corresponding aromatic
AN US
amine. In that way, 2 mM 4-carboxyphenylamine and 2.2 mM sodium nitrite (1.1 equivalents) were added in acid electrolytic solution of aqueous HCl 0.5M. Cyclic voltammetry was performed after 5 minutes reaction to promote the chemical diazotization. [28] Fig. 1B presents an Ep = −0.38 V vs SCE on the first scan. This experimental
M
irreversible wave at potential
evidence confirms the formation of diazonium salt. This broad reduction peak is no more observed
ED
on the subsequent cycles, suggesting that covalent grafting takes place from the first scanning cycle, leading to a passivation of the electrode surface. This electrochemical behavior is
PT
characteristic and similar to those observed for various substituted diazonium salts on different
AC
CE
substrates. [22, 29, 30]
7
ACCEPTED MANUSCRIPT Fig. 1. (A) Potential window of bare FTO electrode in HCl 0.5M. (B) Cyclic voltammograms (scans 1 to 3) at bare FTO electrode in HCl 0.5M containing 2 mM 4-carboxyphenylamine and 2.2 mM NaNO2. Red marks: chosen potentials for electrografting experiments. Scan rate of 0.1V s-1.
3.2 Choice of imposition potential
CR IP T
The in-situ diazonium grafting was realized onto FTO surface under potentiostatic conditions at different potentials during 300 seconds. Based on the two above-mentioned experiments, three potentials were chosen: -0.70, -0.50 and -0.35 V (see Fig.1B). As a comparison, similar chronoamperometric experiment was also performed in 0.5 M HCl aqueous solution but without
AN US
diazonium salt in the medium. The total charges (Q) recorded during the grafting process are summarized in Table 1.
Table 1.
|Q| (mC)
Absence of diazonium*
Presence of diazonium**
-0.35
0.08 ± 0.01
1.06 ± 0.08
ED
M
Imposition potential (V)
-0.50
1.28 ± 0.13
16.34 ± 0.49
5.30 ± 2.81
PT
-0.70
0.14 ± 0.05
|Q| value measured during electroreduction of 4-carboxyphenyl diazonium at various potentials.
CE
Mean values based on 3 measurements.
* aqueous solution of HCl 0.5M
AC
** aqueous solution of HCl 0.5M containing 2 mM 4-carboxyphenylamine and 2.2 mM NaNO2
When the applied potential is -0.35 V or -0.50 V, |Q| values in the presence of diazonium are much higher than without diazonium, which is in agreement with the reduction of diazonium on the electrode surface. Differently, when the applied potential is -0.70 V, the |Q| value in the presence of diazonium is much lower than without diazonium. As noted in the first paragraph, 8
ACCEPTED MANUSCRIPT below -0.7 V reduction of the electrolyte occurs; the reductive grafting of diazonium could thus happen together with the reduction of protons in electrolyte solution. It can also be found in the literature that other redox features may appear for indium or tin-based oxides including FTO, in acidic or basic medium, upon cycling at too negative potentials. They were attributed to the reduction of In or Sn to lower valence or metallic states. [31] It has to be pointed out that this
CR IP T
phenomenon could explain the presence of an “on-off” potential, observed by Bangle and al., for the electrochemical grafting of diazonium salts on various semi-conductor materials [25]. When the same experiment was performed in the presence of diazonium, the presence of a passivating aryl film grafted on the electrode surface could inhibit those side reactions with the electrolyte,
AN US
explaining the obtained lower |Q| value when compared to the case without diazonium. Anyway, in order to avoid the possible occurence of any undesirable side reaction during the grafting process, only the grafting at -0.5 V and -0.35 V were considered for the following electrochemical
M
characterization experiments.
The modified electrodes were sonicated in 0.1 M H2SO4 solution during 5 minutes in order to
ED
remove unbound molecules, then their electrochemical behavior was characterized by CV and EIS in aqueous solution of 0.1 M KCl containing an electroactive probe [32]. The very often used
PT
potassium ferricyanide was preferred to another classical electrochemical probe such as ruthenium
CE
hexamine (positively charged), because the use of a negatively-charged probe has been shown to be a very efficient way to characterize electrode surfaces functionalized by 4-carboxyphenyl
AC
groups, due to electrostatic repulsive interactions between the probe and the negatively charged phenylcarboxylates in aqueous medium [33].
9
CR IP T
ACCEPTED MANUSCRIPT
Fig. 2. (A) Cyclic voltammograms and (B) Nyquist plots of electrochemical impedance spectra of bare FTO electrode
AN US
(full line) and modified electrodes grafted at -0.35 V(dotted line) and at -0.50 V (dashed line) recorded in aqueous solution of K3[Fe(CN)6] (1 mM + 0.1 M KCl). For CV: scan rate of 0.1 V s-1. For EIS: fixed potential of 186mV± 10mV within the frequency range 102 -106 Hz, 5 points per decade of frequencies.
M
As can be seen in Fig. 2A, the cyclic voltammogram recorded on bare FTO presents a reversible redox system (ΔE = 0.18 V) centered at E1/2 = 0.19 V vs SCE corresponding to the response of
ED
Fe(CN)63-/4- couple. This signature is also observed for the grafted electrodes but displays a decrease in peak current and an increase in peak-to-peak separation (ΔE) when compared to CV
PT
recorded for the bare FTO. This observation suggests that an organic layer is attached onto the
CE
electrode surface and acts as a barrier for redox processes. Additionally, the blocking effect observed for the electrode grafted at -0.50V (ΔE = 0.47 V) is more evident than for the one grafted
AC
at -0.35 V (ΔE = 0.25 V), suggesting a more important grafting at lower reduction potential. Complementary experiment, performed at different scan rates (see supporting information, Fig. S1), showed that the diffusion coefficient of the electrochemical probe decreases upon grafting of the electrode surface. This is also consistent with a less accessible surface, due to the presence of the organic molecules.
10
ACCEPTED MANUSCRIPT EIS results were in agreement with CV analyses (see Fig. 2B). The Nyquist plots displayed an increase in the semicircle diameter between the bare electrode and the grafted electrodes in order: bare FTO (~210 Ω) < FTO grafted at -0.35V (~ 270 Ω) < FTO grafted at -0.50V (~ 900 Ω). Overall, all these results indicate clearly two features: (i) the diazonium grafting introduces the attachment of an organic layer on electrode surface and (ii) the imposed potential of -0.50V leads
CR IP T
to the strongest blocking effect; this potential was thus chosen for the rest of the study. Electrochemical characterization of bare and grafted FTO electrodes was also performed by CV and EIS in dry acetonitrile containing ferrocene redox probe (see supporting information, Fig. S2). Indeed, the use of a neutral probe in organic solvent was expected to give complementary
AN US
information concerning the grafted layer. Similar behavior (passivation of the surface, especially at -0.50 V) was observed with ferrocene redox probe but remarkably, the electrochemical response for the electrode grafted at -0.50V is not well-defined and there was an increase in the semicircle
M
diameter (~ 5600 Ω) with dry acetonitrile/ferrocene probe couple when compared to the signal recorded with aqueous KCl/ferricyanide probe. This result could indicate that during the
ED
characterization in aqueous solution, the grafted electrode surface is not completely passivated, suggesting the formation of an organic layer with the presence of pinholes. Indeed, in water,
PT
electrostatic repulsions between negatively charged carboxylates (-COO-) could increase the
CE
distance between adjacent grafted molecules, leading to the creation of these pinholes. In dry acetonitrile, neutral carboxylic groups (-COOH) are formed and would maintain a more close-
AC
packed layer, explaining why the barrier effect seems more important. In conclusion, based on CV and EIS characterization, an optimal deposition potential of -0.50 V was chosen for FTO modification, and the grafting of an organic layer on the electrode surface with the presence of pinholes has been considered.
11
ACCEPTED MANUSCRIPT 3.3 Choice of deposition time Another parameter that controls the electrochemical grafting is the deposition time at a given potential. Grafting of FTO electrodes was carried out at -0.50 V for different deposition times and the total charges (Q) were monitored. An increase in the absolute value of total charges |Q| was observed with an increase in the deposition time (data not shown). The modified electrodes with 5
CR IP T
or 10 minutes grafting were then characterized as detailed previously by CV and EIS in the presence of the electroactive probe ferricyanide (Fig. 3A and B, supporting information Fig. S1). Despite of a visible broaden redox peak at E1/2 = 0.19V vs SCE corresponding to the signature of Fe(CN)63-/4- couple, changes in the CV response of modified electrode are clearly observed when
AN US
compared to the signals recorded on the bare electrode, consisting in a substantial decrease in the peak current and a significant increase in the peak-to-peak separation. This effect is more important for 10 minutes (ΔE = 0.72 V) than for 5 minutes grafting (ΔE = 0.47 V). This is further
M
confirmed when the experiment is performed at different scan rates (see supporting information, Fig. S1) ; the diffusion coefficient of the electrochemical probe appears to be lower for 10 minutes
ED
than for 5 minutes grafting. Moreover, in agreement with the CV, the Nyquist plots in Fig. 3B show an increase in the semicircle diameter in the following order: bare FTO (~ 210 Ω) < FTO
PT
grafted 5minutes (~ 900 Ω) < FTO grafted 10 minutes (~2300 Ω). All these observations suggest
CE
that the grafting during 10 minutes leads to the formation of “more closed-pack”/thicker organic layer on electrode surface than a grafting time of 5 minutes. Further increase in the grafting time
AC
to 15 minutes (results not shown) did not seem to improve further the thickness of organic layer. As a consequence, 10 minutes grafting under a fixed potential of -0.50 V seemed to be the optimal grafting conditions.
12
CR IP T
ACCEPTED MANUSCRIPT
Fig. 3. (A) Cyclic voltammograms and (B) Nyquist plots of electrochemical impedance spectra of bare FTO electrode
AN US
(full line) and modified electrodes grafted during 5 minutes (dashed line) and 10 minutes (grey line) recorded in aqueous solution of K3[Fe(CN)6] (1mM + 0.1M KCl). For CV: scan rate of 0.1V s-1. For EIS: fixed potential of 186 mV± 10 mV within the frequency range 102-106 Hz, 5 points per decade of frequencies.
M
Furthermore, in order to evaluate the possibility for a monolayer/or multilayer formation onto the electrode surface, the surface concentration of grafted molecules Γ could be possibly estimated
ED
based on the Faraday’s law, considering that all the reduced diazonium molecules were efficiently grafted at the electrode surface: Γ = Q/nFA, where |Q| corresponds to the charge encountered
PT
during the reduction process of diazonium, n is the number of the electrons, F is the Faraday
CE
constant and A is the area of the electrode surface. In our case, when the grafting was carried out at -0.50V during 10 minutes, the surface concentration was estimated to be around 17.9 10-9 mol cm. A comparison with available data from the literature shows that this quantity of grafted
AC
2
molecules, obtained after optimizing the grafting conditions, is in the same order of magnitude that the one estimated by Meyer and al. by UV-visible absorbance of a Ru (II) diazonium dye grafted under imposed potential in CH3CN or aqueous HCl on a parent material, i.e. SnO2 [27]. This value range is moreover different from the values expected for a grafted monolayer which is from 1.4 to 6.4 10-10 mol cm-2 [34-35] or 5 10-10 mol cm-2 [36]. For 4-carboxyphenyl diazonium
13
ACCEPTED MANUSCRIPT grafted on a different surface (glassy carbon), Γ values reported in the literature are of 5.5x 10-10 mol cm-2 for a monolayer [16] and 3 10-9 mol cm-2 for multilayer [37]. This feature provides an indication that under the above mentioned conditions, the modification of FTO surface with 4carboxyphenyl diazonium would lead to multilayer formation.
CR IP T
3.4 IR characterization To confirm the successful grafting of an organic layer on the electrode surface, the grafted FTO surface was investigated by FT-IR. Fig. 4 shows the FT-IR spectra of the pristine surface and the grafted one which were compared with reported spectra of the literature [38-42] and IR database
AN US
[43]. Several well-defined absorption peaks are clearly observed in both spectra.
The spectrum of the pristine FTO surface reveals absorption peaks at 615 cm-1 and 520 cm-1 that can be assigned respectively to Eu (TO) and Eu (LO) vibration modes of SnO2 [26-29]. This oxide
M
exhibiting a cassierite-type structure has four infrared-active modes belonging to A2u and Eu symmetry but both A2u modes appear below 400 cm-1 and could not be seen here. The presence of
ED
additional peaks at 780 cm-1 and 682 cm-1 is related to the fundamental stretching vibration of tin hydroxyl which is further confirmed in the insert of Fig. 4 by the presence of a broad absorption
PT
peak around 3300 cm-1 and a weak band at 1628 cm-1. This absorption band is caused by the
CE
bending vibration of coordinated H2O as well as Sn-OH. At last, the most intense peak around 996 cm-1 could also be due to the (Sn-OH) or to Sn=O vibrations as well as lattice vibrations [44].
AC
When compared with the spectra of the grafted surface, in addition to the features observed for the pristine FTO, additional absorption bands were detected. The strong wide band between 3700 and 3000 cm-1 is clearly characteristic of carboxyl acid groups. Concomitantly, a new band appeared at about 1648 cm-1 which could be due to C=O stretching bond of an aryl ketone. The presence of the aryl moieties was confirmed by the small peak up at 3021 cm-1 which could be assigned to
14
ACCEPTED MANUSCRIPT aromatic C-H stretching mode and the small band at 420 cm-1 attributed to ring out-of-plane
AN US
CR IP T
deformation of a p-substituted benzene.
M
Fig. 4. Normalised FT-IR spectra of pristine (black) and modified (red) FTO.
ED
3.5 XPS characterization
The chemical surface compositions of bare and modified FTO with optimal grafting conditions
PT
were analyzed by XPS (Fig. 5). The corresponding elements semi-quantitative analysis is
AC
CE
summarized in Table 2.
15
ED
M
AN US
CR IP T
ACCEPTED MANUSCRIPT
Fig. 5. Typical XPS survey spectrum for bare FTO and modified FTO electrode grafted at -0.50V during 10 minutes
PT
with the main peaks indicated (A) XPS peaks of bare FTO (full line) and modified FTO grafted at -0.50V during 10
CE
minutes (dashed line) for C(1s) (B), Sn(3d) (C), O(1s) (D).
AC
Table 2.
%C
% Sn
%O
% N (1s)
(1s)
(3d)
(1s)
Bare FTO
15
33
52
-
Modified
54
13
31
2
FTO Average atomic concentrations extracted from the XPS spectra obtained from bare FTO and modified FTO grafted at -0.5V during 10 minutes.
16
ACCEPTED MANUSCRIPT For bare FTO the main detected elements are C(1s) (15%), Sn(3d) (33%) and O(1s) (52%) (Fig. 5A). The presence of carbon in typical survey of bare FTO reflects mainly adventitious contamination from the ambient air. The peak F(1s) was too weak to be observed in the survey of bare FTO. After modification, typical survey spectrum for modified electrode displayed the contribution of C(1s) (54%), Sn(3d) (13%), O(1s) (31%) and N(1s) (2%) (see Fig. 5A and second
CR IP T
line in Table 2). The comparison of the C(1s), Sn(3d) and O(1s) XPS spectra for bare and grafted electrode is given in Fig. 5B, 5C and 5D. Clearly, an intensity increase of C(1s) together with an intensity attenuation of Sn (3d) and O(1s) were observed, suggesting the coverage of SnO2 surface by an organic layer. Furthermore, C(1s) core level spectrum could be decomposed into three main
AN US
peaks which could correspond to the five C atoms in the aromatic ring (284.7 ± 0.1 eV); the substituted carbon of the aromatic ring (286.1 ± 0.1 eV), the more characteristic change being the apparition of the signal coming from the carbon of the carboxylic function (288.8 ± 0.1 eV). As
M
expected, these results were in agreement with literature [45]. The complete deconvolution of C(1s) core level spectrum is given in supporting information, Fig. S3. Fig. 5C presents the two
ED
defined peaks corresponding to Sn(3d3/2) at 487.3 ± 0.2 eV and Sn(3d5/2) at 495.7 ± 0.2 eV. After electrodeposition, there was no appreciate shift in binding energy and no change in shape of
PT
Sn(3d) peaks related to bare FTO. These observations in Sn(3d) do not clearly show evidence that
CE
the grafting could occur through direct metal-C (Sn-C) bonding. The O(1s) signal of modified FTO showed a decrease in the peak intensity with a broaden form compared to the signal for bare
AC
FTO (see Fig. 5D). These changes are consistent with the presence of O component from carboxylic groups which is expected at high binding energy [30]. The comparison of O(1s) core level spectra before and after modification could thus give an evidence of the presence of carboxylic groups on the electrode surface. However, it is complicated to make a complete deconvolution of O(1s) signal, as it contains several contributions close in energy, coming from the aryl layer as well as from the substrate.
17
ACCEPTED MANUSCRIPT Besides that, a peak of N(1s) core level spectrum (2%) was observed at 400.6 eV for the modified electrode as it was not present on the non-grafted substrate. This additional N(1s) peak could be attributed to the formation of azo (C–N=N-C) bridge at the electrode surface. The presence of azo linkage has also been observed for most electrode modified by electrochemical reduction of diazonium salts. [28, 33, 46]
CR IP T
The attachment of aryl layers on electrode surface through the covalent bonding between C and other atoms of surface has been reported [46, 47]. In our case, there is no clear evidence from the XPS spectra whether the attachment of an aryl layer to the electrode surface occurs via the covalent bonds such as Sn-C or Sn-O-C. Neither C(1s), nor Sn(3d) spectra of the grafted sample
AN US
show significant binding energy shifts or new contributions at lower binding energies, as expected in the case of Sn-C bonds formation. On the other hand, the O(1s) spectrum of the grafted sample is complex, as mentioned in previous paragraph. This makes it difficult to find a clear evidence of
M
the Sn-O-C bonds formation. Taking into account the weakness of the O(1s) spectrum and the difficulties to fit it with several contributions, it would be hazardous to assert the existence of a
ED
bonding rather than another one. This is consistent with other results described in the literature, where the formation of an O-C bond is often only presumed for the grafting of diazonium salt onto
PT
FTO or other Sn02 based materials [24, 27, 48]. The most interesting study is from Harris et al.,
CE
who performed the grafting of diazonium on a parent material, i.e. antimony-doped tin oxide (ATO). In their case it was neither possible to discriminate the presence of Sn-O-C bond from
AC
XPS experiments. They assumed however its presence, based on the good overlapping between the
ATR-IR experimental spectrum of grafted ATO and the theoretical spectrum of phenyl
species bound via Sn-O-C bond to a Sn(OH)3 cluster obtained by DFT calculations [48]. Despite the spectra do not evidence which type of covalent bonds were formed, XPS analysis contributed to more reliable evidences of the organic layer formation on SnO2 electrode surface, i.e. confirms by the way the existence of the grafted organic layer. Further experiments could be
18
ACCEPTED MANUSCRIPT conducted with other salts to better evidence the nature of the covalent bonds responsible for the grafting.
4. Conclusions In summary, we have conducted the functionalization of commercial FTO surface by
CR IP T
electrochemical reduction of in situ generated diazonium cation. Clearly, this work contributes to optimizing the grafting conditions of 4-carboxyphenyl diazonium onto FTO surface. It was demonstrated that the optimal conditions were carried out for 10 minutes under a controlled potential of -0.50 V. The values recorded during chronoamperometric experiments, together with
AN US
the electrochemical behavior of bare and modified electrodes in the presence of redox probes, allowed us to suggest the formation of poorly dense multilayers containing pinholes. IR and XPS surface analysis gave more evidence of the organic film attached to the electrode surface. Towards
M
the application, FTO can be believed as a robust material. Ongoing investigations are focused on the synthesis of tin oxide thin films by metal organic chemical vapor deposition method in order to
ED
control the crystalline structure and the nano(micro) structure of the surface. We believe that the nanostructuration and physicochemistry of SnO2 may induce changes in the grafting reactivity
CE
PT
and/or localization [49]. The as-obtained results will be reported in due course.
Acknowledgements
AC
The authors would like to acknowledge Labex CHARMMMAT for financial support. We gratefully thank Dr. Diana Dragoé for XPS investigations and valuable discussions.
Declaration of interest: None.
19
ACCEPTED MANUSCRIPT
References
[1] B. Yuliarto,G. Gumilar , N. L. W. Septiani, SnO2 nanostructure as pollutant gas sensors:
CR IP T
Synthesis, sensing performances, and mechanism, Adv. Mater. Sci. Engin. (2015) ID 694823. [2] I. Concina , A. Vomiero, Metal Oxide Semiconductors for Dye‐ and Quantum‐ Dot‐ Sensitized Solar Cells, Small 11 (2015) 1744–1774.
[3] R. P. Howson , M. I. Ridge, Proc. SPIE, Heat mirrors on plastic sheet using transparent oxide conducting coatings, Int. Soc. Opt. Eng. 324 (1982) 16-22.
AN US
[4] A. I. Martínez, L. Huerta, J. M. O. R. d. León, D. Acosta, O. Malik , M. Aguilar, Physicochemical characteristics of fluorine doped tin oxide films, J. Phys. D: Appl. Phys. 39 (2006) 5091-5096.
[5] A.C. Tolcin, Mineral Commodity Summaries 2016.
M
[6] T.-H. Tsai , Y.-F. Wu, Wet etching mechanism of ITO films in. oxalic acid, Microelectron. Eng. 83(2006) 536-541.
[7 R. G. Gordon, Criteria for choosing transparent conductors, MRS Bulletin 25 (2000) 52–57.
ED
[8] Q. Qiao, J. Beck, R. Lumpkin, J. Pretko, Jr J.T. McLeskey, A comparison of fluorine tin oxide and indium tin oxide as the transparent electrode for P3OT/ TiO2 solar cells. Solar Energy
PT
Materials and Solar Cells, 90 (2006), 1034–1040. [9] K. L. Chopra, S. Major, D. K. Pandya, Transparent conductors - A status review, Thin solid
CE
films 102 (1983) 1-46.
[10] C. Agashe, S. S. Major, Effect of heavy doping in SnO2:F films, J. Mater. Sci. 31 (1996)
AC
2965-2969.
[11] Q. Gao, H. Jiang, C. Li, Y. Ma, X. Li, Z. Ren, Y. Liu, C. Song, G. Han, Tailoring of textured conductive SnO2:F thin films, J. Alloys Compd 574 (2013) 427–431. [12] J. S. Park, I-K. Park, Performance modulation of contact electrification nanogenerators by controlling the doping concentration of fluorine-doped tin oxide, Ceram. Int. 44 (2018) 1247712482. [13] S. I. Noh, H-J. Han, D-H. Riu, Photovoltaic property dependence of dye-sensitized solar cells on sheet resistance of FTO substrate depositedvia spray pyrolysis, Ceram. Int. 38 (2012) 37353739. 20
ACCEPTED MANUSCRIPT [14] J-W. Bae, B-R. Koo, H-R. An, H-J. Ahn, Surface modification of fluorine-doped tin oxide films using electrochemical etching for dye-sensitized solar cells, ceram. Int. 41 (2015) 1466814673. [15] M. A. Bissett, I. Koper, J. S. Quinton , J. G. Shapter, Dye functionalisation of PAMAM-type dendrons grown from vertically aligned single-walled carbon nanotube arrays for light harvesting antennae, J. Mater. Chem. 21 (2011) 18597-18604. [16] M. A. Bissett, I. Koper, J. S. Quinton , J. G. Shapter, Dendron growth from vertically aligned
CR IP T
single-walled carbon nanotube thin layer arrays for photovoltaic devices, Phys. Chem. Chem. Phys. 13 (2011) 6059-6064.
[17] M. A. Bissett and J. G. Shapter, Photocurrent response from vertically aligned single-walled carbon nanotube arrays, J. Phys. Chem. C 114 (2010) 6778-6783.
[18] B. A. Putans, L. M. Bishop, R. J. Hamers, Versatile Approach to Formation of LightChem. Mater. 26 (2014) 3651−3659.
AN US
Harvesting Complexes on Nanostructured Metal Oxide Surfaces via “On-Surface” Assembly,
[19] M. C. Benson, R. E. Ruther, J. B. Gerken, M. L. Rigsby, L. M. Bishop, Y. Tan, S. S. Stahl, R. J. Hamers, Modular “Click” Chemistry for Electrochemically and Photoelectrochemically Active
M
Molecular Interfaces to Tin Oxide Surfaces, ACS applied. Chem. 3 (2011) 3110-3119. [20] N. Kaur, H. Thakur, N. Prabhakar, Conducting polymer and multi-walled carbon nanotubes nanocomposites based amperometric biosensor for detection of organophosphate, J. Electroanal.
ED
Chem. 775 (2016) 121-128.
[21] D. P. Harrison, L. S. Carpenter, J. T. Hyde, Reductive electropolymerization of a vinyl-
PT
containing poly-pyridyl complex on glassy carbon and Fluorine-doped Tin Oxide electrodes, J. Vis. Exp. 95 (2015) e52035, doi:10.3791/52035.
CE
[22] D. Belanger , J. Pinson, Electrografting, A powerful method of surface modification, Chem. Soc. Rev. 40 (2011) 3995-4048.
AC
[23] C. D. Christiansen, L. A. Sørensen, T. Lund, Modification of fluorine-doped tin oxideelectrodes by electrochemical reduction of di(4-nitrophenyl)iodonium tetrafluoroborate - And its application as a photo-anode in dye-sensitized solar cells, J. Electroanal. Chem. 809 (2018) 44-51. [24] T. Lund, P. T. Nguyen, T. Rulhand, Electrochemical grafting of TiO2-based photo-anodes and its effect in dye-sensitized solar cells, J. Electroanal. Chem. 758 (2015) 85-92. [25] F. Lamberti, S. Agnoli, L. Brigo, G. Granozzi, M. Giomo , N. Elvassore, Surface Functionalization of Fluorine-Doped Tin Oxide Samples through Electrochemical Grafting, ACS Appl. Mater. Interfaces 5 (2013) 12887-12894.
21
ACCEPTED MANUSCRIPT [26] F. Lamberti, L. Brigo, M. Favaro, C. Luni, A. Zoso, M. Cattelan, S. Agnoli, G. Brusatin, G. Granozzi, M. Giomo, N. Elvassore, Optoelectrochemical Biorecognition by Optically Transparent Highly Conductive Graphene-Modified Fluorine-Doped Tin Oxide Substrates, ACS Appl. Mater. Interfaces 6 (2014) 22769−22777. [27] R. Bangle, R. N. Sampaio, L. Troian-Gautier, G. J. Meyer, Surface Grafting of Ru(II) Diazonium-Based Sensitizers on Metal Oxides Enhances Alkaline Stability for Solar Energy Conversion, ACS Appl. Mater. Interfaces 10 (2018) 3121–3132
CR IP T
[28] S. Baranton and D. Bélanger, Electrochemical Derivatization of Carbon Surface by Reduction of in Situ Generated Diazonium Cations, J. Phys. Chem. B 109 (2005) 24401-24410.
[29] C. Combellas, D.-e. Jiang, F. Kanoufi, J. Pinson and F. I. Podvorica, Steric Effects in the Reaction of Aryl Radicals on Surfaces, Langmuir 25 (2009) 286-293.
[30] M. Delamar, R. Hitmi, J. Pinson and J. M. Saveant, Covalent modification of carbon surfaces
AN US
by grafting of functionalized aryl radicals produced from electrochemical reduction of diazonium salts, J. Am. Chem. Soc. 114 (1992) 5883-5884.
[31] J. D. Benck, B. A. Pinaud, Y. Gorlin , T. F. Jaramillo, Substrate Selection for Fundamental Studies of Electrocatalysts and Photoelectrodes: Inert Potential Windows in Acidic, Neutral, and
M
Basic Electrolyte, PLoS ONE 9 (2014) e107942.
[32] A. J. Bard and L. R. Faulkner, Electrochemical Methods-Fundamentals and Applications, Wiley, New York, 2001.
ED
[33] C. Saby, B. Ortiz, G.Y. Champagne, D. Bélanger, Electrochemical Modification of Glassy Carbon Electrode Using Aromatic Diazonium Salts. 1. Blocking Effect of 4-Nitrophenyl and 4-
PT
Carboxyphenyl Groups, Langmuir 13 (1997) 6805-6813. [34] J. Ghilane, G. Trippe-Allard , J. C. Lacroix, Grafting of imidazolium ion-terminated layer
CE
using electrochemical reduction of diazonium: Redox active counter anion exchange within the layer, Electrochem. Commun. 27 (2013) 73-76.
AC
[35] I. Gallardo, J. Pinson , N. Vilà, Spontaneous Attachment of Amines to Carbon and Metallic Surfaces, J. Phys. Chem. B 110 (2006) 19521-19529. [36] S. Q. Lud, M. Steenackers, R. Jordan, P. Bruno, D. M. Gruen, P. Feulner, J. A. Garrido , M. Stutzmann, Chemical Grafting of Biphenyl Self-Assembled Monolayers on Ultrananocrystalline Diamond, J. Am. Chem. Soc. 128 (2006) 16884-16891. [37] T. Menanteau, M. Dias, E. Levillain, A. J. Downard , T. Breton, Electrografting via Diazonium Chemistry: The Key Role of the Aryl Substituent in the Layer Growth Mechanism, J. Phys. Chem. C 120 (2016) 4423-4429.
22
ACCEPTED MANUSCRIPT [38] M. Batzill, U. Diebold, The surface and materials science of tin oxide, Prog. Surf. Sci. 79 (2005) 47–154. [39] R. Summitt, Infrared Absorption in Single-Crystal Stannic Oxide: Optical Lattice-Vibration Modes, J. Appl. Phys. 39 (1968) 3762–3767. [40] R. Drevet, C. Legros, D. Berardan, P. Ribot, A. Chaussé, M. Andrieux, Nanocrystallized SnO2 thin films deposited on Si and LaAlO3 substrates by pulsed-MOCVD technique for electrochemical applications, Surf. Coat.Technol 230 (2013) 180–185.
CR IP T
[41] A. F. Khan, M. Mehmood, M. Aslam, M. Ashraf, Characteristics of electron beam evaporated nanocrystalline SnO2 thin films annealed in air, Appl. Surf. Sci. 256 (2010) 2252–2258.
[42] Wang X, Andrews L., Infrared Spectra of M(OH)1,2,4 (M = Pb, Sn) in Solid Argon, J Phys Chem A. 109 (2005) 9013-20.
[43] Infrared and Raman Characteristic Group Frequencies: Tables and Charts, 3rd Ed., George
AN US
Socrates, Chichester, Wiley, New york, 2001.
[44] D. Almaric-Popescu, F. Bozon-Verduraz, Infrared studies on SnO2 and Pd/SnO2, Catal. Today 70 (2001) 139-154.
[45] D. Quinton, A. Galtayries, F. Prima , S. Griveau, Functionalization of titanium surfaces with
M
a simple electrochemical strategy, Surf. Coat. Technol. 206 (2012) 2302-2307. [46] B. L. Hurley and R. L. McCreery, Covalent Bonding of Organic Molecules to Cu and Al Alloy 2024 T3 Surfaces via Diazonium Ion Reduction, J. Electrochem. Soc. 151 (2004) B252-
ED
B259.
[47] S. Maldonado, T. J. Smith, R. D. Williams, S. Morin, E. Barton , K. J. Stevenson, Surface
PT
Modification of Indium Tin Oxide via Electrochemical Reduction of Aryldiazonium Cations, Langmuir 22 (2006) 2884-2891.
CE
[48] T. G. A. A. Harris, R. Götz, P. Wrzolek, V. Davis, C. E. Knapp, K. Ly, P. Hildebrandt, M. Schwalbe, I. Weidinger, I. Zebger, A. Fisher, Robust electrografted interfaces on metal oxides for
AC
electrocatalysis - an in situ spectroelectrochemical study, J. Mater. Chem. A 6 (2018) 1520015212.
[49] R. Drevet, D. Dragoé, M. G. Barthés-Labrousse, A. Chaussé , M. Andrieux, XPSnanocharacterization of organic layers electrochemically grafted on the surface of SnO2 thin films to produce a new hybrid material coating, Appl. Surf. Sci. 384 (2016) 442-448.
23
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN US
CR IP T
Graphical Abstract
24
Modification of Fluorine-Doped Tin Oxide surface: optimization of the electrochemical grafting of diazonium salt Van Bui-Thi-Tuyet , Caroline Cannizzo , Corinne Legros , Michel Andrieux , Annie Chausse´ PII: DOI: Reference:
S2468-0230(18)30452-8 https://doi.org/10.1016/j.surfin.2019.01.012 SURFIN 289
To appear in:
Surfaces and Interfaces
Received date: Revised date: Accepted date:
17 September 2018 23 January 2019 28 January 2019
Please cite this article as: Van Bui-Thi-Tuyet , Caroline Cannizzo , Corinne Legros , Michel Andrieux , Annie Chausse´ , Modification of Fluorine-Doped Tin Oxide surface: optimization of the electrochemical grafting of diazonium salt, Surfaces and Interfaces (2019), doi: https://doi.org/10.1016/j.surfin.2019.01.012
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Highlights
AC
CE
PT
ED
M
AN US
CR IP T
Covalent functionalization of Fluorine Tin Oxide by carboxyphenyl groups was made by via electroreduction of diazonium salt Parameters such as grafting potential and grafting time were investigated Investigation of FTO surface by XPS and FTIR shows the presence of caboxyphenyl groups at the FTO surface Electrochemical behavior of FTO together with electrochemical characterization in the presence of an electroactive probe allowed us to conclude to the grafting of a poorlypacked organic multi-layer
1
ACCEPTED MANUSCRIPT
Modification of Fluorine-Doped Tin Oxide surface: optimization of the electrochemical grafting of diazonium salt
Van Bui-Thi-Tuyeta, Caroline Cannizzoa, Corinne Legrosb, Michel Andrieuxb and Annie Chausséa LAMBE, Univ Evry, CNRS, CEA, Université Paris-Saclay, 91025, Evry, France
b
ICMMO, Univ Paris-Sud, CNRS, Université Paris-Saclay, 91405, Orsay, France *
CR IP T
a
Corresponding author. Tel.:+33(0)1 69 47 02 22. E-Mail address: [email protected]
AN US
Abstract
Fluorine-doped tin oxide (FTO) electrodes were functionalized by electro reduction of in situ generated 4-carboxyphenyl diazonium salt. Several grafting potentials and times were investigated in order to find the optimal grafting conditions. The electrochemical behavior of bare FTO
M
electrodes in the presence of the diazonium, together with Cyclic Voltammetry (CV) and
ED
Electrochemical Impedance Spectroscopy (EIS) characterization of the grafted electrodes in the presence of redox probes ferricyanide and ferrocene allowed us to conclude that 10 minutes
PT
grafting under a controlled potential of -0.50V lead to the formation of a poorly-packed organic multi-layer on the electrode surface. X-ray Photoelectron Spectroscopy (XPS), together with
CE
Fourier-Transform Infrared Spectroscopy (FTIR) analyses, further evidenced the presence of
AC
carboxylic groups onto modified FTO electrodes.
Keywords
Fluorine-doped tin oxide- FTO- Electrochemical grafting – Diazonium salt- Surface modification
2
ACCEPTED MANUSCRIPT 1. Introduction Among various electrode surfaces, tin oxide (SnO2) has been extensively used in both research and industry for applications such as gas sensors, [1] solar cells [2] and heat mirrors [3]. Commercially, SnO2 is doped by indium oxide (In2O3 :SnO2) or fluorine (SnO2 :F) [4]. Despite the fact that indium tin oxide is the most used, there exists some limitations such as high cost because
CR IP T
of low indium availability [5] and pH sensitivity [6]. For those reasons, Fluorine-doped tin oxide (FTO) may thus be preferred [7, 8] as a prospective optically transparent and electrically conductive oxide [9]. The electro conductivity is achieved by fluorine atoms [10] that give stability to the chemical attack, also at low pH. Its interesting properties make FTO a non-toxic,
various industrial applications [12-14].
AN US
fully biodegradable and cost-affordable material [11], therefore promising for the development of
In literature, a few recent papers concerning FTO modification are available. Bisset et al. reported
M
the grafting of oxidized carbon nanotubes on hydroxylated FTO [15-17]. Hamers et al. focused on the reaction of thiolated or iodated bipyrydiyl compounds with the same substrate [18]. SnO2 was
ED
functionalized by photochemical grafting, followed by click chemistry [19]; electrografting of PEDOT [20] or vinyl-containing poly-pyridyl complexes [21] onto FTO were also reported.
PT
Electrografting of diazonium salts is a convenient way to bring covalently attached functions at
CE
the surface of various materials [22]. Surprisingly, to our knowledge, only a few studies dealing with the grafting of diazonium salts onto FTO are available: the grafting of iodonium [23] or
AC
diazonium salts [23, 24] was considered for the development of dye-sensitized solar cells ; the electrografting of aryldiazonium cations for the development of biological–based sensor devices was also reported [25, 26]. All these reactions were conducted by cyclic voltammetry in either acetonitrile [23, 24] or aqueous HCl medium [25, 26], optimizing the number of cycles. The grafting of diazonium salt at fixed potential was, to our knowledge, only reported once for a parent material : Meyer et al. realized the grafting a Ru(II) diazonium dye on various semi-conductor
3
ACCEPTED MANUSCRIPT materials, including Sn02 [27]. The electrografting time and potential were optimized in acetonitrile, and its feasibility was also demonstrated at one chosen potential in aqueous HCl, appearently with similar characteristics of the grafted layer. The presence of a “turn-off” potential, under which the grafting of diazonium salt is no more possible, was mentioned. It thus appeared important to make a complete study of the optimal grafting conditions of diazonium salts onto
CR IP T
FTO surface in aqueous acidic conditions, and to determine the properties of the as-obtained grafted layer. Herein, starting from the corresponding amine precursor, we studied the immobilization of 4-carboxyphenyldiazonium (Scheme 1) on FTO by means of electrografting, varying the reduction time and potential in order to optimize the grafting conditions. The attached
AN US
organic layer was characterized by means of Cyclic Voltammetry (CV) and Electrochemical Impedance Spectroscopy (EIS) in both aprotic and protic solvent in the presence of two redox probes, one negatively charged (hexacyanoferricyanide (III)) and the other neutral (ferrocene).
M
These easy to handle electrochemical techniques could bring interesting information not only on the presence of the attached organic layer, but also on its thickness and homogeneity. The
ED
presence of carboxylic groups at the FTO surface was further evidenced both by X-ray
AC
CE
PT
Photoelectron Spectroscopy (XPS) and Fourier Transform Infra-Red spectroscopy (FT-IR).
Scheme 1. In situ 4-carboxyphenyl diazonium generation and electrografting on FTO.
4
ACCEPTED MANUSCRIPT 2. Materials and Methods 2.1 Chemicals Chlorhydric acid standard solution (1 M) was purchased from Fluka, dry acetronitrile, tetrabutylammoniumtetrafluoroborate, sodium nitrite, potassium chloride, p-aminobenzoic acid and ferrocene from Aldrich. All aqueous solutions were prepared with purified water (18.2 MΩ
CR IP T
cm-1) from Millipore.
2.2 Electrochemical measurements
A conventional three-electrode cell was used for the electrochemical experiments. Platinum wire
AN US
was used as auxiliary electrode, saturated calomel electrode (SCE) as reference electrode. FTO glass substrates purchased from SOLEMS (France ) were used as working electrodes after isolation of 1cm2 with a solution of polystyrene in mesitylene. Prior to use, the working electrodes
M
were rinsed and sonicated in H2SO4 0.1M during 10 minutes. Experiments were performed with an Autolab (low current PGSTAT302N, Metrohm) at room temperature. All solutions were
ED
deaerated with argon bubbling during 20 minutes prior to the electrochemical experiments. EIS measurements were performed using EG&G (273A, Princeton Applied Research) coupled with a
PT
SI 1255 Frequency Response Analyzer (Solartron). EIS measurements were recorded at room
CE
temperature within the frequency range of 102-106 Hz superimposed on a DC potential of -186 mV for aqueous solution of K3[Fe(CN)6] (1 mM + 0.1 M KCl) and of 420 mV for dry acetonitrile of
AC
Ferrocene (1 mM + 0.1M NBu4BF4) , with AC of 10 mV peak to peak amplitude, and 5 points per decade of frequencies. Three distinct experiments were carried out for each measurement.
2.3 IR characterization Fourier-Transform Infra-Red (FTIR) was carried out with a Vertex 70 Bruker spectrometer equipped with an MCT detector and using the Attenuated Total Reflectance (ATR) technique
5
ACCEPTED MANUSCRIPT which enables samples to be examined directly. Germanium crystal has, by far, the highest refractive index of all the ATR materials available, and the effective depth of penetration is approximately 1 µm. In order to obtain effective signal/noise ratio, each spectrum was obtained by averaging 400 scans in the wavelength range of 400-4000 cm-1 at a spectral resolution of 4 cm-1. Absorption peaks search was automatically realized using the OPUS 7.0 software. As scan spectra
with IR database or data published in the literature.
CR IP T
are of low intensity, all spectra have been normalized to be weighed. The results were compared
Nota : The biggest difference between ATR and Transmission spectra can be the shift of strong absorbing bands moving to lower wavenumbers in ATR mode. For most compounds, the shifts
AN US
observed are small, approximately 1-10 cm-1, but for very strong absorbing compounds, the shifts observed can be as big as 30-50 cm-1. These observed changes are in accordance with the
M
established theory of ATR spectroscopy.
2.4 XPS characterization
ED
XPS measurements were performed on a K Alpha spectrometer (ThermoFisher), equipped with a monochromated X-Ray Source (AlKα, 1486.6 eV). A spot size of 400µm was used. The
PT
hemispherical analyzer was operated in CAE (Constant Analyzer Energy) mode, with a pass
CE
energy of 200 eV and a step of 1 eV for the acquisition of survey spectra, and a pass energy of 50 eV and a step of 0.1 eV for the acquisition of high resolution spectra. A “dual beam” flood gun
AC
was used to neutralize the charge built-up. The spectra were monitored and processed using Advantage software from ThermoFisher. A Shirley type background subtraction and a mixture of Gaussian- Lorentzian shapes (30/70) were used for peak fitting. The peak areas were normalized using the Scofield sensitivity factors.
6
ACCEPTED MANUSCRIPT 3. Results and discussion 3.1 Potential window of FTO and diazotization process The choice of the electrografting potential depends on the potential window of bare electrode in electrolyte solution. Prior to the electrochemical modification of FTO surface, the potential window of FTO was thus checked in 0.5M HCl solution. Fig. 1A shows that the deposition
CR IP T
potential for reductive grafting of diazonium salt should not be conducted below -0.70V, as at lower potential reduction of the electrolyte occurs. Cyclic voltammetry on bare FTO surface was next performed in the presence of 4-carboxyphenyl diazonium salt. The diazonium salt was directly generated in situ in the electrochemical cell, starting from the corresponding aromatic
AN US
amine. In that way, 2 mM 4-carboxyphenylamine and 2.2 mM sodium nitrite (1.1 equivalents) were added in acid electrolytic solution of aqueous HCl 0.5M. Cyclic voltammetry was performed after 5 minutes reaction to promote the chemical diazotization. [28] Fig. 1B presents an Ep = −0.38 V vs SCE on the first scan. This experimental
M
irreversible wave at potential
evidence confirms the formation of diazonium salt. This broad reduction peak is no more observed
ED
on the subsequent cycles, suggesting that covalent grafting takes place from the first scanning cycle, leading to a passivation of the electrode surface. This electrochemical behavior is
PT
characteristic and similar to those observed for various substituted diazonium salts on different
AC
CE
substrates. [22, 29, 30]
7
ACCEPTED MANUSCRIPT Fig. 1. (A) Potential window of bare FTO electrode in HCl 0.5M. (B) Cyclic voltammograms (scans 1 to 3) at bare FTO electrode in HCl 0.5M containing 2 mM 4-carboxyphenylamine and 2.2 mM NaNO2. Red marks: chosen potentials for electrografting experiments. Scan rate of 0.1V s-1.
3.2 Choice of imposition potential
CR IP T
The in-situ diazonium grafting was realized onto FTO surface under potentiostatic conditions at different potentials during 300 seconds. Based on the two above-mentioned experiments, three potentials were chosen: -0.70, -0.50 and -0.35 V (see Fig.1B). As a comparison, similar chronoamperometric experiment was also performed in 0.5 M HCl aqueous solution but without
AN US
diazonium salt in the medium. The total charges (Q) recorded during the grafting process are summarized in Table 1.
Table 1.
|Q| (mC)
Absence of diazonium*
Presence of diazonium**
-0.35
0.08 ± 0.01
1.06 ± 0.08
ED
M
Imposition potential (V)
-0.50
1.28 ± 0.13
16.34 ± 0.49
5.30 ± 2.81
PT
-0.70
0.14 ± 0.05
|Q| value measured during electroreduction of 4-carboxyphenyl diazonium at various potentials.
CE
Mean values based on 3 measurements.
* aqueous solution of HCl 0.5M
AC
** aqueous solution of HCl 0.5M containing 2 mM 4-carboxyphenylamine and 2.2 mM NaNO2
When the applied potential is -0.35 V or -0.50 V, |Q| values in the presence of diazonium are much higher than without diazonium, which is in agreement with the reduction of diazonium on the electrode surface. Differently, when the applied potential is -0.70 V, the |Q| value in the presence of diazonium is much lower than without diazonium. As noted in the first paragraph, 8
ACCEPTED MANUSCRIPT below -0.7 V reduction of the electrolyte occurs; the reductive grafting of diazonium could thus happen together with the reduction of protons in electrolyte solution. It can also be found in the literature that other redox features may appear for indium or tin-based oxides including FTO, in acidic or basic medium, upon cycling at too negative potentials. They were attributed to the reduction of In or Sn to lower valence or metallic states. [31] It has to be pointed out that this
CR IP T
phenomenon could explain the presence of an “on-off” potential, observed by Bangle and al., for the electrochemical grafting of diazonium salts on various semi-conductor materials [25]. When the same experiment was performed in the presence of diazonium, the presence of a passivating aryl film grafted on the electrode surface could inhibit those side reactions with the electrolyte,
AN US
explaining the obtained lower |Q| value when compared to the case without diazonium. Anyway, in order to avoid the possible occurence of any undesirable side reaction during the grafting process, only the grafting at -0.5 V and -0.35 V were considered for the following electrochemical
M
characterization experiments.
The modified electrodes were sonicated in 0.1 M H2SO4 solution during 5 minutes in order to
ED
remove unbound molecules, then their electrochemical behavior was characterized by CV and EIS in aqueous solution of 0.1 M KCl containing an electroactive probe [32]. The very often used
PT
potassium ferricyanide was preferred to another classical electrochemical probe such as ruthenium
CE
hexamine (positively charged), because the use of a negatively-charged probe has been shown to be a very efficient way to characterize electrode surfaces functionalized by 4-carboxyphenyl
AC
groups, due to electrostatic repulsive interactions between the probe and the negatively charged phenylcarboxylates in aqueous medium [33].
9
CR IP T
ACCEPTED MANUSCRIPT
Fig. 2. (A) Cyclic voltammograms and (B) Nyquist plots of electrochemical impedance spectra of bare FTO electrode
AN US
(full line) and modified electrodes grafted at -0.35 V(dotted line) and at -0.50 V (dashed line) recorded in aqueous solution of K3[Fe(CN)6] (1 mM + 0.1 M KCl). For CV: scan rate of 0.1 V s-1. For EIS: fixed potential of 186mV± 10mV within the frequency range 102 -106 Hz, 5 points per decade of frequencies.
M
As can be seen in Fig. 2A, the cyclic voltammogram recorded on bare FTO presents a reversible redox system (ΔE = 0.18 V) centered at E1/2 = 0.19 V vs SCE corresponding to the response of
ED
Fe(CN)63-/4- couple. This signature is also observed for the grafted electrodes but displays a decrease in peak current and an increase in peak-to-peak separation (ΔE) when compared to CV
PT
recorded for the bare FTO. This observation suggests that an organic layer is attached onto the
CE
electrode surface and acts as a barrier for redox processes. Additionally, the blocking effect observed for the electrode grafted at -0.50V (ΔE = 0.47 V) is more evident than for the one grafted
AC
at -0.35 V (ΔE = 0.25 V), suggesting a more important grafting at lower reduction potential. Complementary experiment, performed at different scan rates (see supporting information, Fig. S1), showed that the diffusion coefficient of the electrochemical probe decreases upon grafting of the electrode surface. This is also consistent with a less accessible surface, due to the presence of the organic molecules.
10
ACCEPTED MANUSCRIPT EIS results were in agreement with CV analyses (see Fig. 2B). The Nyquist plots displayed an increase in the semicircle diameter between the bare electrode and the grafted electrodes in order: bare FTO (~210 Ω) < FTO grafted at -0.35V (~ 270 Ω) < FTO grafted at -0.50V (~ 900 Ω). Overall, all these results indicate clearly two features: (i) the diazonium grafting introduces the attachment of an organic layer on electrode surface and (ii) the imposed potential of -0.50V leads
CR IP T
to the strongest blocking effect; this potential was thus chosen for the rest of the study. Electrochemical characterization of bare and grafted FTO electrodes was also performed by CV and EIS in dry acetonitrile containing ferrocene redox probe (see supporting information, Fig. S2). Indeed, the use of a neutral probe in organic solvent was expected to give complementary
AN US
information concerning the grafted layer. Similar behavior (passivation of the surface, especially at -0.50 V) was observed with ferrocene redox probe but remarkably, the electrochemical response for the electrode grafted at -0.50V is not well-defined and there was an increase in the semicircle
M
diameter (~ 5600 Ω) with dry acetonitrile/ferrocene probe couple when compared to the signal recorded with aqueous KCl/ferricyanide probe. This result could indicate that during the
ED
characterization in aqueous solution, the grafted electrode surface is not completely passivated, suggesting the formation of an organic layer with the presence of pinholes. Indeed, in water,
PT
electrostatic repulsions between negatively charged carboxylates (-COO-) could increase the
CE
distance between adjacent grafted molecules, leading to the creation of these pinholes. In dry acetonitrile, neutral carboxylic groups (-COOH) are formed and would maintain a more close-
AC
packed layer, explaining why the barrier effect seems more important. In conclusion, based on CV and EIS characterization, an optimal deposition potential of -0.50 V was chosen for FTO modification, and the grafting of an organic layer on the electrode surface with the presence of pinholes has been considered.
11
ACCEPTED MANUSCRIPT 3.3 Choice of deposition time Another parameter that controls the electrochemical grafting is the deposition time at a given potential. Grafting of FTO electrodes was carried out at -0.50 V for different deposition times and the total charges (Q) were monitored. An increase in the absolute value of total charges |Q| was observed with an increase in the deposition time (data not shown). The modified electrodes with 5
CR IP T
or 10 minutes grafting were then characterized as detailed previously by CV and EIS in the presence of the electroactive probe ferricyanide (Fig. 3A and B, supporting information Fig. S1). Despite of a visible broaden redox peak at E1/2 = 0.19V vs SCE corresponding to the signature of Fe(CN)63-/4- couple, changes in the CV response of modified electrode are clearly observed when
AN US
compared to the signals recorded on the bare electrode, consisting in a substantial decrease in the peak current and a significant increase in the peak-to-peak separation. This effect is more important for 10 minutes (ΔE = 0.72 V) than for 5 minutes grafting (ΔE = 0.47 V). This is further
M
confirmed when the experiment is performed at different scan rates (see supporting information, Fig. S1) ; the diffusion coefficient of the electrochemical probe appears to be lower for 10 minutes
ED
than for 5 minutes grafting. Moreover, in agreement with the CV, the Nyquist plots in Fig. 3B show an increase in the semicircle diameter in the following order: bare FTO (~ 210 Ω) < FTO
PT
grafted 5minutes (~ 900 Ω) < FTO grafted 10 minutes (~2300 Ω). All these observations suggest
CE
that the grafting during 10 minutes leads to the formation of “more closed-pack”/thicker organic layer on electrode surface than a grafting time of 5 minutes. Further increase in the grafting time
AC
to 15 minutes (results not shown) did not seem to improve further the thickness of organic layer. As a consequence, 10 minutes grafting under a fixed potential of -0.50 V seemed to be the optimal grafting conditions.
12
CR IP T
ACCEPTED MANUSCRIPT
Fig. 3. (A) Cyclic voltammograms and (B) Nyquist plots of electrochemical impedance spectra of bare FTO electrode
AN US
(full line) and modified electrodes grafted during 5 minutes (dashed line) and 10 minutes (grey line) recorded in aqueous solution of K3[Fe(CN)6] (1mM + 0.1M KCl). For CV: scan rate of 0.1V s-1. For EIS: fixed potential of 186 mV± 10 mV within the frequency range 102-106 Hz, 5 points per decade of frequencies.
M
Furthermore, in order to evaluate the possibility for a monolayer/or multilayer formation onto the electrode surface, the surface concentration of grafted molecules Γ could be possibly estimated
ED
based on the Faraday’s law, considering that all the reduced diazonium molecules were efficiently grafted at the electrode surface: Γ = Q/nFA, where |Q| corresponds to the charge encountered
PT
during the reduction process of diazonium, n is the number of the electrons, F is the Faraday
CE
constant and A is the area of the electrode surface. In our case, when the grafting was carried out at -0.50V during 10 minutes, the surface concentration was estimated to be around 17.9 10-9 mol cm. A comparison with available data from the literature shows that this quantity of grafted
AC
2
molecules, obtained after optimizing the grafting conditions, is in the same order of magnitude that the one estimated by Meyer and al. by UV-visible absorbance of a Ru (II) diazonium dye grafted under imposed potential in CH3CN or aqueous HCl on a parent material, i.e. SnO2 [27]. This value range is moreover different from the values expected for a grafted monolayer which is from 1.4 to 6.4 10-10 mol cm-2 [34-35] or 5 10-10 mol cm-2 [36]. For 4-carboxyphenyl diazonium
13
ACCEPTED MANUSCRIPT grafted on a different surface (glassy carbon), Γ values reported in the literature are of 5.5x 10-10 mol cm-2 for a monolayer [16] and 3 10-9 mol cm-2 for multilayer [37]. This feature provides an indication that under the above mentioned conditions, the modification of FTO surface with 4carboxyphenyl diazonium would lead to multilayer formation.
CR IP T
3.4 IR characterization To confirm the successful grafting of an organic layer on the electrode surface, the grafted FTO surface was investigated by FT-IR. Fig. 4 shows the FT-IR spectra of the pristine surface and the grafted one which were compared with reported spectra of the literature [38-42] and IR database
AN US
[43]. Several well-defined absorption peaks are clearly observed in both spectra.
The spectrum of the pristine FTO surface reveals absorption peaks at 615 cm-1 and 520 cm-1 that can be assigned respectively to Eu (TO) and Eu (LO) vibration modes of SnO2 [26-29]. This oxide
M
exhibiting a cassierite-type structure has four infrared-active modes belonging to A2u and Eu symmetry but both A2u modes appear below 400 cm-1 and could not be seen here. The presence of
ED
additional peaks at 780 cm-1 and 682 cm-1 is related to the fundamental stretching vibration of tin hydroxyl which is further confirmed in the insert of Fig. 4 by the presence of a broad absorption
PT
peak around 3300 cm-1 and a weak band at 1628 cm-1. This absorption band is caused by the
CE
bending vibration of coordinated H2O as well as Sn-OH. At last, the most intense peak around 996 cm-1 could also be due to the (Sn-OH) or to Sn=O vibrations as well as lattice vibrations [44].
AC
When compared with the spectra of the grafted surface, in addition to the features observed for the pristine FTO, additional absorption bands were detected. The strong wide band between 3700 and 3000 cm-1 is clearly characteristic of carboxyl acid groups. Concomitantly, a new band appeared at about 1648 cm-1 which could be due to C=O stretching bond of an aryl ketone. The presence of the aryl moieties was confirmed by the small peak up at 3021 cm-1 which could be assigned to
14
ACCEPTED MANUSCRIPT aromatic C-H stretching mode and the small band at 420 cm-1 attributed to ring out-of-plane
AN US
CR IP T
deformation of a p-substituted benzene.
M
Fig. 4. Normalised FT-IR spectra of pristine (black) and modified (red) FTO.
ED
3.5 XPS characterization
The chemical surface compositions of bare and modified FTO with optimal grafting conditions
PT
were analyzed by XPS (Fig. 5). The corresponding elements semi-quantitative analysis is
AC
CE
summarized in Table 2.
15
ED
M
AN US
CR IP T
ACCEPTED MANUSCRIPT
Fig. 5. Typical XPS survey spectrum for bare FTO and modified FTO electrode grafted at -0.50V during 10 minutes
PT
with the main peaks indicated (A) XPS peaks of bare FTO (full line) and modified FTO grafted at -0.50V during 10
CE
minutes (dashed line) for C(1s) (B), Sn(3d) (C), O(1s) (D).
AC
Table 2.
%C
% Sn
%O
% N (1s)
(1s)
(3d)
(1s)
Bare FTO
15
33
52
-
Modified
54
13
31
2
FTO Average atomic concentrations extracted from the XPS spectra obtained from bare FTO and modified FTO grafted at -0.5V during 10 minutes.
16
ACCEPTED MANUSCRIPT For bare FTO the main detected elements are C(1s) (15%), Sn(3d) (33%) and O(1s) (52%) (Fig. 5A). The presence of carbon in typical survey of bare FTO reflects mainly adventitious contamination from the ambient air. The peak F(1s) was too weak to be observed in the survey of bare FTO. After modification, typical survey spectrum for modified electrode displayed the contribution of C(1s) (54%), Sn(3d) (13%), O(1s) (31%) and N(1s) (2%) (see Fig. 5A and second
CR IP T
line in Table 2). The comparison of the C(1s), Sn(3d) and O(1s) XPS spectra for bare and grafted electrode is given in Fig. 5B, 5C and 5D. Clearly, an intensity increase of C(1s) together with an intensity attenuation of Sn (3d) and O(1s) were observed, suggesting the coverage of SnO2 surface by an organic layer. Furthermore, C(1s) core level spectrum could be decomposed into three main
AN US
peaks which could correspond to the five C atoms in the aromatic ring (284.7 ± 0.1 eV); the substituted carbon of the aromatic ring (286.1 ± 0.1 eV), the more characteristic change being the apparition of the signal coming from the carbon of the carboxylic function (288.8 ± 0.1 eV). As
M
expected, these results were in agreement with literature [45]. The complete deconvolution of C(1s) core level spectrum is given in supporting information, Fig. S3. Fig. 5C presents the two
ED
defined peaks corresponding to Sn(3d3/2) at 487.3 ± 0.2 eV and Sn(3d5/2) at 495.7 ± 0.2 eV. After electrodeposition, there was no appreciate shift in binding energy and no change in shape of
PT
Sn(3d) peaks related to bare FTO. These observations in Sn(3d) do not clearly show evidence that
CE
the grafting could occur through direct metal-C (Sn-C) bonding. The O(1s) signal of modified FTO showed a decrease in the peak intensity with a broaden form compared to the signal for bare
AC
FTO (see Fig. 5D). These changes are consistent with the presence of O component from carboxylic groups which is expected at high binding energy [30]. The comparison of O(1s) core level spectra before and after modification could thus give an evidence of the presence of carboxylic groups on the electrode surface. However, it is complicated to make a complete deconvolution of O(1s) signal, as it contains several contributions close in energy, coming from the aryl layer as well as from the substrate.
17
ACCEPTED MANUSCRIPT Besides that, a peak of N(1s) core level spectrum (2%) was observed at 400.6 eV for the modified electrode as it was not present on the non-grafted substrate. This additional N(1s) peak could be attributed to the formation of azo (C–N=N-C) bridge at the electrode surface. The presence of azo linkage has also been observed for most electrode modified by electrochemical reduction of diazonium salts. [28, 33, 46]
CR IP T
The attachment of aryl layers on electrode surface through the covalent bonding between C and other atoms of surface has been reported [46, 47]. In our case, there is no clear evidence from the XPS spectra whether the attachment of an aryl layer to the electrode surface occurs via the covalent bonds such as Sn-C or Sn-O-C. Neither C(1s), nor Sn(3d) spectra of the grafted sample
AN US
show significant binding energy shifts or new contributions at lower binding energies, as expected in the case of Sn-C bonds formation. On the other hand, the O(1s) spectrum of the grafted sample is complex, as mentioned in previous paragraph. This makes it difficult to find a clear evidence of
M
the Sn-O-C bonds formation. Taking into account the weakness of the O(1s) spectrum and the difficulties to fit it with several contributions, it would be hazardous to assert the existence of a
ED
bonding rather than another one. This is consistent with other results described in the literature, where the formation of an O-C bond is often only presumed for the grafting of diazonium salt onto
PT
FTO or other Sn02 based materials [24, 27, 48]. The most interesting study is from Harris et al.,
CE
who performed the grafting of diazonium on a parent material, i.e. antimony-doped tin oxide (ATO). In their case it was neither possible to discriminate the presence of Sn-O-C bond from
AC
XPS experiments. They assumed however its presence, based on the good overlapping between the
ATR-IR experimental spectrum of grafted ATO and the theoretical spectrum of phenyl
species bound via Sn-O-C bond to a Sn(OH)3 cluster obtained by DFT calculations [48]. Despite the spectra do not evidence which type of covalent bonds were formed, XPS analysis contributed to more reliable evidences of the organic layer formation on SnO2 electrode surface, i.e. confirms by the way the existence of the grafted organic layer. Further experiments could be
18
ACCEPTED MANUSCRIPT conducted with other salts to better evidence the nature of the covalent bonds responsible for the grafting.
4. Conclusions In summary, we have conducted the functionalization of commercial FTO surface by
CR IP T
electrochemical reduction of in situ generated diazonium cation. Clearly, this work contributes to optimizing the grafting conditions of 4-carboxyphenyl diazonium onto FTO surface. It was demonstrated that the optimal conditions were carried out for 10 minutes under a controlled potential of -0.50 V. The values recorded during chronoamperometric experiments, together with
AN US
the electrochemical behavior of bare and modified electrodes in the presence of redox probes, allowed us to suggest the formation of poorly dense multilayers containing pinholes. IR and XPS surface analysis gave more evidence of the organic film attached to the electrode surface. Towards
M
the application, FTO can be believed as a robust material. Ongoing investigations are focused on the synthesis of tin oxide thin films by metal organic chemical vapor deposition method in order to
ED
control the crystalline structure and the nano(micro) structure of the surface. We believe that the nanostructuration and physicochemistry of SnO2 may induce changes in the grafting reactivity
CE
PT
and/or localization [49]. The as-obtained results will be reported in due course.
Acknowledgements
AC
The authors would like to acknowledge Labex CHARMMMAT for financial support. We gratefully thank Dr. Diana Dragoé for XPS investigations and valuable discussions.
Declaration of interest: None.
19
ACCEPTED MANUSCRIPT
References
[1] B. Yuliarto,G. Gumilar , N. L. W. Septiani, SnO2 nanostructure as pollutant gas sensors:
CR IP T
Synthesis, sensing performances, and mechanism, Adv. Mater. Sci. Engin. (2015) ID 694823. [2] I. Concina , A. Vomiero, Metal Oxide Semiconductors for Dye‐ and Quantum‐ Dot‐ Sensitized Solar Cells, Small 11 (2015) 1744–1774.
[3] R. P. Howson , M. I. Ridge, Proc. SPIE, Heat mirrors on plastic sheet using transparent oxide conducting coatings, Int. Soc. Opt. Eng. 324 (1982) 16-22.
AN US
[4] A. I. Martínez, L. Huerta, J. M. O. R. d. León, D. Acosta, O. Malik , M. Aguilar, Physicochemical characteristics of fluorine doped tin oxide films, J. Phys. D: Appl. Phys. 39 (2006) 5091-5096.
[5] A.C. Tolcin, Mineral Commodity Summaries 2016.
M
[6] T.-H. Tsai , Y.-F. Wu, Wet etching mechanism of ITO films in. oxalic acid, Microelectron. Eng. 83(2006) 536-541.
[7 R. G. Gordon, Criteria for choosing transparent conductors, MRS Bulletin 25 (2000) 52–57.
ED
[8] Q. Qiao, J. Beck, R. Lumpkin, J. Pretko, Jr J.T. McLeskey, A comparison of fluorine tin oxide and indium tin oxide as the transparent electrode for P3OT/ TiO2 solar cells. Solar Energy
PT
Materials and Solar Cells, 90 (2006), 1034–1040. [9] K. L. Chopra, S. Major, D. K. Pandya, Transparent conductors - A status review, Thin solid
CE
films 102 (1983) 1-46.
[10] C. Agashe, S. S. Major, Effect of heavy doping in SnO2:F films, J. Mater. Sci. 31 (1996)
AC
2965-2969.
[11] Q. Gao, H. Jiang, C. Li, Y. Ma, X. Li, Z. Ren, Y. Liu, C. Song, G. Han, Tailoring of textured conductive SnO2:F thin films, J. Alloys Compd 574 (2013) 427–431. [12] J. S. Park, I-K. Park, Performance modulation of contact electrification nanogenerators by controlling the doping concentration of fluorine-doped tin oxide, Ceram. Int. 44 (2018) 1247712482. [13] S. I. Noh, H-J. Han, D-H. Riu, Photovoltaic property dependence of dye-sensitized solar cells on sheet resistance of FTO substrate depositedvia spray pyrolysis, Ceram. Int. 38 (2012) 37353739. 20
ACCEPTED MANUSCRIPT [14] J-W. Bae, B-R. Koo, H-R. An, H-J. Ahn, Surface modification of fluorine-doped tin oxide films using electrochemical etching for dye-sensitized solar cells, ceram. Int. 41 (2015) 1466814673. [15] M. A. Bissett, I. Koper, J. S. Quinton , J. G. Shapter, Dye functionalisation of PAMAM-type dendrons grown from vertically aligned single-walled carbon nanotube arrays for light harvesting antennae, J. Mater. Chem. 21 (2011) 18597-18604. [16] M. A. Bissett, I. Koper, J. S. Quinton , J. G. Shapter, Dendron growth from vertically aligned
CR IP T
single-walled carbon nanotube thin layer arrays for photovoltaic devices, Phys. Chem. Chem. Phys. 13 (2011) 6059-6064.
[17] M. A. Bissett and J. G. Shapter, Photocurrent response from vertically aligned single-walled carbon nanotube arrays, J. Phys. Chem. C 114 (2010) 6778-6783.
[18] B. A. Putans, L. M. Bishop, R. J. Hamers, Versatile Approach to Formation of LightChem. Mater. 26 (2014) 3651−3659.
AN US
Harvesting Complexes on Nanostructured Metal Oxide Surfaces via “On-Surface” Assembly,
[19] M. C. Benson, R. E. Ruther, J. B. Gerken, M. L. Rigsby, L. M. Bishop, Y. Tan, S. S. Stahl, R. J. Hamers, Modular “Click” Chemistry for Electrochemically and Photoelectrochemically Active
M
Molecular Interfaces to Tin Oxide Surfaces, ACS applied. Chem. 3 (2011) 3110-3119. [20] N. Kaur, H. Thakur, N. Prabhakar, Conducting polymer and multi-walled carbon nanotubes nanocomposites based amperometric biosensor for detection of organophosphate, J. Electroanal.
ED
Chem. 775 (2016) 121-128.
[21] D. P. Harrison, L. S. Carpenter, J. T. Hyde, Reductive electropolymerization of a vinyl-
PT
containing poly-pyridyl complex on glassy carbon and Fluorine-doped Tin Oxide electrodes, J. Vis. Exp. 95 (2015) e52035, doi:10.3791/52035.
CE
[22] D. Belanger , J. Pinson, Electrografting, A powerful method of surface modification, Chem. Soc. Rev. 40 (2011) 3995-4048.
AC
[23] C. D. Christiansen, L. A. Sørensen, T. Lund, Modification of fluorine-doped tin oxideelectrodes by electrochemical reduction of di(4-nitrophenyl)iodonium tetrafluoroborate - And its application as a photo-anode in dye-sensitized solar cells, J. Electroanal. Chem. 809 (2018) 44-51. [24] T. Lund, P. T. Nguyen, T. Rulhand, Electrochemical grafting of TiO2-based photo-anodes and its effect in dye-sensitized solar cells, J. Electroanal. Chem. 758 (2015) 85-92. [25] F. Lamberti, S. Agnoli, L. Brigo, G. Granozzi, M. Giomo , N. Elvassore, Surface Functionalization of Fluorine-Doped Tin Oxide Samples through Electrochemical Grafting, ACS Appl. Mater. Interfaces 5 (2013) 12887-12894.
21
ACCEPTED MANUSCRIPT [26] F. Lamberti, L. Brigo, M. Favaro, C. Luni, A. Zoso, M. Cattelan, S. Agnoli, G. Brusatin, G. Granozzi, M. Giomo, N. Elvassore, Optoelectrochemical Biorecognition by Optically Transparent Highly Conductive Graphene-Modified Fluorine-Doped Tin Oxide Substrates, ACS Appl. Mater. Interfaces 6 (2014) 22769−22777. [27] R. Bangle, R. N. Sampaio, L. Troian-Gautier, G. J. Meyer, Surface Grafting of Ru(II) Diazonium-Based Sensitizers on Metal Oxides Enhances Alkaline Stability for Solar Energy Conversion, ACS Appl. Mater. Interfaces 10 (2018) 3121–3132
CR IP T
[28] S. Baranton and D. Bélanger, Electrochemical Derivatization of Carbon Surface by Reduction of in Situ Generated Diazonium Cations, J. Phys. Chem. B 109 (2005) 24401-24410.
[29] C. Combellas, D.-e. Jiang, F. Kanoufi, J. Pinson and F. I. Podvorica, Steric Effects in the Reaction of Aryl Radicals on Surfaces, Langmuir 25 (2009) 286-293.
[30] M. Delamar, R. Hitmi, J. Pinson and J. M. Saveant, Covalent modification of carbon surfaces
AN US
by grafting of functionalized aryl radicals produced from electrochemical reduction of diazonium salts, J. Am. Chem. Soc. 114 (1992) 5883-5884.
[31] J. D. Benck, B. A. Pinaud, Y. Gorlin , T. F. Jaramillo, Substrate Selection for Fundamental Studies of Electrocatalysts and Photoelectrodes: Inert Potential Windows in Acidic, Neutral, and
M
Basic Electrolyte, PLoS ONE 9 (2014) e107942.
[32] A. J. Bard and L. R. Faulkner, Electrochemical Methods-Fundamentals and Applications, Wiley, New York, 2001.
ED
[33] C. Saby, B. Ortiz, G.Y. Champagne, D. Bélanger, Electrochemical Modification of Glassy Carbon Electrode Using Aromatic Diazonium Salts. 1. Blocking Effect of 4-Nitrophenyl and 4-
PT
Carboxyphenyl Groups, Langmuir 13 (1997) 6805-6813. [34] J. Ghilane, G. Trippe-Allard , J. C. Lacroix, Grafting of imidazolium ion-terminated layer
CE
using electrochemical reduction of diazonium: Redox active counter anion exchange within the layer, Electrochem. Commun. 27 (2013) 73-76.
AC
[35] I. Gallardo, J. Pinson , N. Vilà, Spontaneous Attachment of Amines to Carbon and Metallic Surfaces, J. Phys. Chem. B 110 (2006) 19521-19529. [36] S. Q. Lud, M. Steenackers, R. Jordan, P. Bruno, D. M. Gruen, P. Feulner, J. A. Garrido , M. Stutzmann, Chemical Grafting of Biphenyl Self-Assembled Monolayers on Ultrananocrystalline Diamond, J. Am. Chem. Soc. 128 (2006) 16884-16891. [37] T. Menanteau, M. Dias, E. Levillain, A. J. Downard , T. Breton, Electrografting via Diazonium Chemistry: The Key Role of the Aryl Substituent in the Layer Growth Mechanism, J. Phys. Chem. C 120 (2016) 4423-4429.
22
ACCEPTED MANUSCRIPT [38] M. Batzill, U. Diebold, The surface and materials science of tin oxide, Prog. Surf. Sci. 79 (2005) 47–154. [39] R. Summitt, Infrared Absorption in Single-Crystal Stannic Oxide: Optical Lattice-Vibration Modes, J. Appl. Phys. 39 (1968) 3762–3767. [40] R. Drevet, C. Legros, D. Berardan, P. Ribot, A. Chaussé, M. Andrieux, Nanocrystallized SnO2 thin films deposited on Si and LaAlO3 substrates by pulsed-MOCVD technique for electrochemical applications, Surf. Coat.Technol 230 (2013) 180–185.
CR IP T
[41] A. F. Khan, M. Mehmood, M. Aslam, M. Ashraf, Characteristics of electron beam evaporated nanocrystalline SnO2 thin films annealed in air, Appl. Surf. Sci. 256 (2010) 2252–2258.
[42] Wang X, Andrews L., Infrared Spectra of M(OH)1,2,4 (M = Pb, Sn) in Solid Argon, J Phys Chem A. 109 (2005) 9013-20.
[43] Infrared and Raman Characteristic Group Frequencies: Tables and Charts, 3rd Ed., George
AN US
Socrates, Chichester, Wiley, New york, 2001.
[44] D. Almaric-Popescu, F. Bozon-Verduraz, Infrared studies on SnO2 and Pd/SnO2, Catal. Today 70 (2001) 139-154.
[45] D. Quinton, A. Galtayries, F. Prima , S. Griveau, Functionalization of titanium surfaces with
M
a simple electrochemical strategy, Surf. Coat. Technol. 206 (2012) 2302-2307. [46] B. L. Hurley and R. L. McCreery, Covalent Bonding of Organic Molecules to Cu and Al Alloy 2024 T3 Surfaces via Diazonium Ion Reduction, J. Electrochem. Soc. 151 (2004) B252-
ED
B259.
[47] S. Maldonado, T. J. Smith, R. D. Williams, S. Morin, E. Barton , K. J. Stevenson, Surface
PT
Modification of Indium Tin Oxide via Electrochemical Reduction of Aryldiazonium Cations, Langmuir 22 (2006) 2884-2891.
CE
[48] T. G. A. A. Harris, R. Götz, P. Wrzolek, V. Davis, C. E. Knapp, K. Ly, P. Hildebrandt, M. Schwalbe, I. Weidinger, I. Zebger, A. Fisher, Robust electrografted interfaces on metal oxides for
AC
electrocatalysis - an in situ spectroelectrochemical study, J. Mater. Chem. A 6 (2018) 1520015212.
[49] R. Drevet, D. Dragoé, M. G. Barthés-Labrousse, A. Chaussé , M. Andrieux, XPSnanocharacterization of organic layers electrochemically grafted on the surface of SnO2 thin films to produce a new hybrid material coating, Appl. Surf. Sci. 384 (2016) 442-448.
23
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN US
CR IP T
Graphical Abstract
24