Ethers as Oxygen Donor and Carbon Source in Non-hydrolytic Sol–Gel: One-Pot, Atom-Economic Synthesis of Mesoporous TiO2–Carbon Nanocomposites

Article abstract of DOI:10.1002/chem.201800027

Mesoporous TiO₂–carbon nanocomposites were synthesized using an original non-hydrolytic sol–gel (NHSG) route, based on the reaction of simple ethers (diisopropyl ether or tetrahydrofuran) with titanium tetrachloride. In this atom-economic, solvent-free process, the ether acts not only as an oxygen donor but also as the sole carbon source. Increasing the reaction temperature to 180 °C leads to the decomposition of the alkyl chloride by-product and to the formation of hydrocarbon polymers, which are converted to carbon by pyrolysis under argon. The carbon–TiO₂ nanocomposites and their TiO₂ counterparts (obtained by calcination) were characterized by nitrogen physisorption, XRD, solid state ¹³C NMR and Raman spectroscopies, SEM, and TEM. The nanocomposites are mesoporous with surface areas of up to 75 m² g^?1 and pore sizes around 10 nm. They are composed of aggregated anatase nanocrystals coated by an amorphous carbon film. Playing on the nature of the ether and on the reaction temperature allows control over the carbon content in the nanocomposites. The nature of the ether also influences the size of the TiO₂ crystallites and the morphology of the nanocomposite. To further characterize the carbon coating, the behavior of the carbon-TiO₂ nanocomposites and bare TiO₂ samples toward lithium insertion–deinsertion was investigated in half-cells. This simple NHSG approach should provide a general method for the synthesis of a wide range of carbon–metal oxide nanocomposites.

Full text of DOI:10.1002/chem.201800027

A Journal of
Accepted Article
Title: Ethers as oxygen donor and carbon source in non-hydrolytic
sol-gel: one-pot, atom-economic synthesis of mesoporous TiO₂-
carbon nanocomposites
Authors: Angel Manuel Escamilla-Pérez, Nicolas Louvain, Bruno
Boury, Nicolas Brun, and Hubert Mutin
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201800027
Link to VoR: http://dx.doi.org/10.1002/chem.201800027
Supported by

10.1002/chem.201800027
Chemistry - A European Journal
Ethers as oxygen donor and carbon source in non-hydrolytic sol-gel: one-pot, atom-
economic synthesis of mesoporous TiO2-carbon nanocomposites
A. M. Escamilla-Pérez,[a] N. Louvain,*[a,b] B. Boury,[a] N. Brun,[a] and P.H. Mutin*[a]
[a]
Institut Charles Gerhardt Montpellier (ICGM), UMR 5253, Univ.Montpellier-CNRS-
ENSCM, Montpellier, France.
[b]
Réseau sur le Stockage Electrochimique de l’Energie (RS2E), FR CNRS 3459, France.
Mesoporous TiO2/C
nanocomposites were synthesized by
a one-pot, solvent-free, non-
hydrolytic sol-gel route based on the
use of an ether as oxygen donor and
carbon source
Abstract : Mesoporous TiO2-carbon nanocomposites were synthesized using an original non-
hydrolytic sol-gel (NHSG) route, based on the reaction of simple ethers (diisopropyl ether or
tetrahydrofuran) with titanium tetrachloride. In this atom-economic, solvent-free process, the
ether acts not only as an oxygen donor but also as the sole carbon source. Increasing the reaction
temperature to 180 °C leads to the decomposition of the alkyl chloride by-product and to the
formation of hydrocarbon polymers, which are converted to carbon by pyrolysis under argon.
The carbon-TiO2 nanocomposites and their TiO2 counterparts (obtained by calcination) were
characterized by nitrogen physisorption, XRD, solid state 13C NMR and Raman spectroscopies,
SEM, and TEM. The nanocomposites are mesoporous with surface areas of up to 75 m2g-1 and
pore sizes around 10 nm. They are composed of aggregated anatase nanocrystals coated by an
amorphous carbon film. Playing on the nature of the ether and on the reaction temperature
allows control over the carbon content in the nanocomposites. The nature of the ether also
influences the size of the TiO2 crystallites and the morphology of the nanocomposite. To further
characterize the carbon coating, the behavior of the carbon-TiO2 nanocomposites and bare TiO2
samples toward lithium insertion-deinsertion was investigated in half-cells. This simple NHSG
approach should provide a general method for the synthesis of a wide range of carbon-metal
oxide nanocomposites.
This article is protected by copyright. All rights reserved.

10.1002/chem.201800027
Chemistry - A European Journal
Introduction
The sol-gel process is the most common method used to synthesize mesoporous oxides, which
are of paramount importance in the areas of heterogeneous catalysis, separation, adsorption, or
energy.[1] Indeed, this low temperature process offers the possibility to obtain metastable
phases, with control over composition, homogeneity, and texture. The conventional methods
based on hydrolysis and condensation[2] predominate but several alternative non-hydrolytic (or
non-aqueous) sol–gel routes have been developed.[3] Unlike conventional aqueous sol-gel
processes, non-hydrolytic sol–gel (NHSG) is performed in organic medium and involves
organic oxygen-donors such as ethers or alcohols instead of water, and, most importantly,
organic reaction pathways.[4] Over the last 25 years, NHSG has been established as a powerful
methodology for the synthesis of oxide- based materials, notably mesoporous oxide and mixed
oxide xerogels[5] and crystalline metal oxide nanoparticles. NHSG has also been extended to
the synthesis of organic–inorganic hybrids,[6] metal sulfides,[7] and nitrides.[8]
Here we investigate for the first time the extension of the NHSG method to the synthesis of
metal oxide/carbon nanocomposites using the organic oxygen donor as a carbon source. To
demonstrate the feasibility of this NHSG approach we focussed on TiO2/carbon materials.
Metal oxide/carbon nanocomposites marrying the properties of both components are promising
advanced materials, with applications in energy storage, catalysis, or sorption.[9] Owing to its
outstanding electronic and optical properties, high chemical stability and low cost,
nanostructured titanium dioxide has been the subject of intense research.[10] However, pure TiO2
[11]
suffers from a low electronic conductivity (10-12 < Ec < 10-7 s cm-1),
low ion diffusion
coefficient (10-14 < Dc < 10-6 cm2 s-1) [12] and fast electron-hole recombination.[13] Accordingly,
a wide variety of TiO2/carbon nanocomposites has been developed[9b] to improve performances
notably in photocatalysis[14] or lithium batteries applications.[15]
TiO2 is considered a promising alternative to graphite anodes for high-power Li-ion batteries,
but its electronic conductivity is too low.[16] Accordingly, the coating of TiO2 onto highly
conductive carbons such as carbon nanotubes[15c] and graphene-based[15b, 17] materials has been
widely investigated. Conversely, the coating of carbonaceous layers onto TiO2 particles has
also been studied and has proved more advantageous.[18] Such carbon coatings should be
homogenous and thin in order to allow Li ions diffusion through the carbon layer. It has been
demonstrated that the homogeneity of the carbon coating rather than the carbon to TiO2 ratio
plays a key role in tuning electrochemical performances.[15e, 19] Two-step approaches involving
the synthesis of TiO2 nanomaterials by hydrolytic sol-gel routes and their subsequent coating
using various carbon precursors has been widely developed.[20] A large variety of carbon
precursors has been reported including sucrose,[21] glucose,[22] poly(vinyl pyrrolidone),[23]
dopamine,[24] pyrrole,[25] and ethylene.[26] In these multi-step strategies, controlling the
homogeneity of the carbon coating is not straightforward; in addition, besides carbon
precursors, copolymers or surfactants are usually necessary to generate mesoporous or
nanostructured materials, involving extra costs.
Although several NHSG routes have been successfully used to prepare TiO2 porous materials
and nanoparticles,[27] there are very few examples of their application to the preparation of
TiO2–based electrode materials for batteries application. Recently we showed that mesoporous
This article is protected by copyright. All rights reserved.

10.1002/chem.201800027
Chemistry - A European Journal
nanocrystalline TiO2 with good performances in lithium insertion could be prepared from TiCl4
and diisopropylether at 110 °C.[28] Tahir et al.[19a] synthesized carbon-coated TiO2 nanoparticles
via the benzyl alcohol route, using a block copolymer as chelating agent and carbon precursor.
The presence in the copolymer of coordinating hydroxamic acid groups and of graphitizable
polyacrylonitrile blocks afforded after pyrolysis at 650 °C small TiO2 nanoparticles covered
with a uniform carbon coating.
Herein we demonstrate that carbon-coated TiO2 nanomaterials can be obtained by reaction of
TiCl4 with an ether (either diisopropylether, IPE, or tetrahydrofuran, THF), in the absence of
solvent and without adding a carbon precursor. In this atom-economic route, the ether is thus
the oxygen donor and the sole carbon source. The organic by-products, instead of being
eliminated (e.g. by calcination in air) to form the pure oxide, are carbonized by pyrolysis under
argon (Scheme 1). Mesoporous TiO2/C nanocomposites built of anatase nanocrystals coated by
an amorphous carbon film were obtained. The influence of the nature of the ether and of the
reaction temperature on the carbon content and the morphology was investigated. Both the
TiO2/C nanocomposites and the corresponding oxides were thoroughly characterized by X-ray
diffraction, Raman spectroscopy, 13C solid state nuclear magnetic resonance, N2-physisorption
and scanning and transmission electron microscopy. The influence of the carbon coating on
lithium insertion-deinsertion was investigated by galvanostatic cycling in half cells.
Scheme 1. NHSG synthesis of TiO2 and TiO2/C nanocomposite from TiCl4 and an ether (IPE or THF).
Results and Discussion
Choice of the reaction temperature
The NHSG synthesis of TiO2 from TiCl4 and an ether is based on the intermediate formation of
alkoxide groups which then condense with Ti-Cl to form oxo bridges. Both steps lead to the
formation of an alkyl chloride (Scheme 2). The reaction is generally performed under
autogenous pressure at moderate temperature, around 100 °C. Under these conditions, the major
organic by-product is indeed the alkyl chloride, isopropyl chloride in the case of IPE and 1,4-
dichlorobutane in the case of THF.[27a, 29] Both alkyl chlorides are volatile and are easily
removed during the drying step. Accordingly, as shown in Table 1, the TiO2/C nanocomposites
synthesized at 110 °C exhibited low carbon contents, around 3 wt%.
This article is protected by copyright. All rights reserved.

10.1002/chem.201800027
Chemistry - A European Journal
Scheme 2. Reactions involved in the NHSG synthesis of TiO2 from TiCl4 and an ether: alkoxylation (1),
condensation (2), and ideal reaction schemes for the formation of TiO2 at 110 °C from diisopropyl ether (3) and
tetrahydrofuran (4).
The key to increasing the carbon content lies in using a higher reaction temperature, to favor
secondary reactions such as elimination of HCl and polymerization of the resulting alkenes
leading to non-volatile by-products (Scheme 3). These secondary reactions are possibly
catalyzed by the TiO2 particles and by the HCl produced during the synthesis. In the synthesis
of tungsten oxide from tungsten isopropoxide, Niederberger at al. reported that tungsten oxide
catalyzed the polymerization of benzyl alcohol into polyphenylene methylene.[30] In our case it
is noteworthy that the carbon content was higher for the samples prepared with THF than with
IPE, although THF contains only 4 carbon atoms, compared to 6 C atoms for IPE. This can be
explained by the fact that complete dehydrochlorination of 1,4-dichlorobutane would lead to
1,3-butadiene, which can form highly cross-linked polymers leading to higher carbon yields
(Scheme 3). A 180 °C reaction temperature appears as a good compromise: the calculated
pressure in the autoclave remains moderate (<30 bars) and the carbon content is significant: 6.6
wt% and 18.9 wt% for the IPE and THF-derived samples, respectively. Accordingly, in the
following, all the samples were synthesized at 180 °C.
Table 1. Carbon content for TiO2/C composites prepared by reaction at different temperatures, then pyrolyzed at
750 °C under argon. The C content was determined by TGA from the weight loss between 300 and 1000 °C (Figure
SI1).
T reaction C content
Sample
Oxygen donor
IPE
°C
wt%
3.3
110
150
180
260
110
150
180
260
6.6
6.6
TiO2/C-IPE
15.2
3.4
13.7
18.9
21.5
TiO2/C-THF
THF
Scheme 3. Schematic representation of the formation of non-volatile by-products by dehydrochlorination and
polymerization reaction of isopropyl chloride (5) and 1-4,dichlorobutane (6).
This article is protected by copyright. All rights reserved.

10.1002/chem.201800027
Chemistry - A European Journal
Structure of the xerogels
The xerogels (dried gels) prepared at 180 °C from IPE and THF (labeled Xer-IPE and Xer-
THF, respectively), were characterized by elemental analysis, 13C CP-MAS NMR
spectroscopy, and powder X-Ray diffraction (PXRD). The carbon content (determined buy
C/H/N analysis) of Xer-IPE (12.4 wt%) was much lower than that of Xer-THF (34.3 wt%).
13C CP-MAS NMR spectra of the xerogels confirmed the occurence of extensive secondary
reactions (Figure 1). Both spectra showed the presence of sp3 (0-40 ppm) and sp2 (110-140
ppm) carbon atoms. The broad signals corresponding to the sp3 carbons were consistent with
the formation of a cross-linked polyolefin network built of primary, secondary, tertiary, and
possibly quaternary carbons. There was no evidence for aliphatic C-OH or C-O-C groups (55–
75 ppm).[31] This is not surprising, as the gels were prepared using a stoichiometric amount of
oxygen (i.e. 2 O atoms per Ti atom) and that most of these oxygen atoms have been consumed
in the formation of the TiO2 phase. The intense sp2 signals indicated the presence in both
xerogels of significant amounts of vinylic or possibly aromatic carbons. According to EDX
elemental analysis, the xerogels contained significant amounts of chlorine, as shown by the
Cl/Ti ratios of 0.08 and 0.17 found for Xer-IPE and Xer-THF, respectively (Table SI1); carbon
atoms bonded to Cl atoms could thus account for the more intense shoulders found for Xer-
THF at 30 and 135 ppm.
Figure 1. Solid-state 13C CP-MAS NMR spectra of xerogels. The spectra were normalized and the background
signal was removed by subtracting the spectrum obtained for the empty rotor.
X-Ray diffraction (Figure 2) showed the presence in the xerogels of anatase nanocrystals
(JCPDS 00-21-1272). The rutile phase was not detected. The crystallite size estimated from the
Scherrer equation was 7 nm for Xer-IPE and 11 nm for Xer-THF. Accordingly, the xerogels
prepared at 180 °C could be described as built of a nanocrystalline TiO2 anatase phase and a
polymeric, oxygen poor, carbon phase.
This article is protected by copyright. All rights reserved.

10.1002/chem.201800027
Chemistry - A European Journal
Figure 2. Powder XRD patterns of the xerogels, TiO2 and TiO2/C samples. The two weak reflections around 45°
2θ arise from the PTFE support used during analysis.
Choice of pyrolysis temperature
The pyrolysis temperature is an important parameter, which directly impacts the structure and
texture of both the carbon and the TiO2 phases. To find the optimal pyrolysis temperature,
xerogels prepared from IPE and THF at 180 °C were pyrolyzed at three different temperatures
(750, 800, and 900 °C), then analyzed by X-ray diffraction (XRD) and nitrogen physisorption.
XRD patterns (Figure SI2) showed the formation of a nearly pure anatase phase for pyrolysis
temperatures of 750 °C or 800 °C, whereas a significant amount of rutile was detected in the
samples pyrolyzed at 900 °C. Furthermore, the specific surface area was significantly decreased
after pyrolysis at 800 °C (Table SI2). Accordingly, we chose a pyrolysis temperature of 750 °C,
ensuring the formation of virtually pure anatase phase and of an optimal mesoporous texture.
Structural and textural characterization of TiO2/C composites and of the corresponding
TiO2 materials
The nitrogen adsorption-desorption isotherms of TiO2/C and TiO2 samples (Figure 3) and the
associated textural data (Table 2) show that the four samples were mesoporous, with relatively
high specific surface areas and pore diameters considering the high pyrolysis or calcination
temperature used. According to the IUPAC classification,[32] the isotherms were intermediate
between Type II (interparticle porosity) and type IV (mesoporous solids). BJH pore size
distributions showed the presence in all cases of mesopores in the 3 to 11 nm range. The sharp
peaks observed in the pore size distributions at ca. 3-4 nm are due to an artefact related to the
instability of the meniscus at P/P0 lower than 0.42,[32] and simply indicates the presence of small
mesopores (<4 nm). The TiO2/C samples had a significantly higher specific surface areas than
their TiO2 counterparts, suggesting that the carbonaceous phase had a significant surface area,
or that the exothermic calcination led to some sintering of the anatase nanocrystals.
This article is protected by copyright. All rights reserved.

10.1002/chem.201800027
Chemistry - A European Journal
(a)
(b)
Figure 3. (a) N2 physisorption isotherms of TiO2 and TiO2/C samples (filled and open symbols refer to adsorption
and desorption, respectively); (b) BJH pore size distributions (desorption branch).
Table 2. Textural data and TiO2 crystallite size for the composite and oxide samples.
[a]
SBET
Vp[b]
Dp[c] TiO2 crystallite size[d]
Sample
(m2 g-1) (cm3 g-1) (nm)
(nm)
20
16
TiO2/C-IPE
TiO2-IPE
TiO2/C-THF
TiO2-THF
70
50
75
63
0.18
0.17
0.21
0.23
9.3
9.3
10.2
10.3
13
14
[a] Specific surface area, [b] total pore volume; [c] BJH average pore diameter calculated from
the desorption branch; [d] crystallite size determined by the Scherrer equation.
The XRD patterns of TiO2/C composites obtained by pyrolysis at 750 °C and of TiO2 samples
obtained by calcination at 500 °C in air are shown in Figure 2. These patterns revealed the
presence of TiO2 anatase only (JCPDS 00-21-1272), except in the case of TiO2/C-IPE where a
very weak reflection associated with the rutile phase (JCPDS 00-021-1276) was also detected.
The anatase crystallite size in TiO2/C and TiO2 samples ranged from 13 to 20 nm (Table 2), and
was slightly larger in the materials derived from IPE than in those derived from THF. Although
the TiO2/C-IPE and TiO2/C-THF nanocomposites contained 6.6 and 18.9 wt% of carbonaceous
phase, respectively, no reflection related to graphitic carbon (JCPDS 01-075-2078) was
detected, suggesting that the carbon phase was largely amorphous.
The Raman spectra of TiO2 and TiO2/C samples in the 100 to 800 cm-1 range (Figure SI3)
confirm the presence of anatase as shown by the intense peak at ca. 144 cm-1, while no band
This article is protected by copyright. All rights reserved.

10.1002/chem.201800027
Chemistry - A European Journal
related to the rutile phase was detected.[33] It is important to mention that the Raman active
vibrations related to anatase TiO2 were much less intense in the TiO2/C samples than in the
TiO2 ones, as shown by the lower signal to noise ratio. This suggests the presence of carbon
surrounding the TiO2 particles. In order to verify this assumption, the Raman spectrum of
TiO2/C-THF was compared with the spectrum of a physical mixture of TiO2-THF (82 wt%)
and carbon black (18 wt%) prepared by grinding in a mortar until a homogenous powder was
obtained (sample labeled TiO2–THF+CB in Figure SI4). As shown in Figure SI4, the anatase
vibration modes were clearly observed in the spectrum of the physical mixture, while for the
TiO2/C-THF sample these bands were much weaker, which supports the formation of a carbon
coating around the TiO2 nanoparticles. Raman spectroscopy is also a valuable technique to
investigate carbonaceous materials. The Raman spectra in the 900-2000 cm-1 region of TiO2/C
nanocomposites are given in Figure 4. For comparison, the spectrum of the carbon black used
in electrodes preparation is also shown. Both carbon black and TiO2/C samples displayed G
and D bands, characteristic of graphitic and partially graphitic carbons, respectively. While the
G band is usually centered at 1580 cm-1, the disorder-induced band (D band) appears at 1350
cm-1. The former is related to the presence of impurities or defects into the graphitic crystal
framework.[34] Nevertheless, the quantification of the crystallinity of graphitic carbons by
Raman spectroscopy is difficult. The intensity ratio ID/IG can give precious information on the
graphitized crystallite coherence length using the Knight formula.[34c] However, in our case, the
G band is centered at 1597 cm-1 (instead of 1580 cm-1 as expected for graphitic carbons) while
the D band is very broad with a noticeable shoulder at around 1200 cm-1. Thus, the Knight
formula is hardly applicable. Such feature was also observed by Sadezky et al[35] when studying
soot and amorphous carbonaceous materials. Previous reports demonstrated that the D band
could be decomposed into four components[35-36]: D1, D2, D3 and D4 which appear at 1350;
1620; 1500 and 1200 cm-1 respectively. D3 corresponds to amorphous carbon while the three
other bands are related to disordered graphitic layers (A1g and E2g- symmetry). The presence of
these four bands could explain the shift and the shape of both D and G bands in TiO2/C samples
in comparison with theoretical values.[37] These results are in good agreement with XRD
patterns and support the amorphous character of the carbon phase in TiO2/C nanocomposites.
Figure 4. Raman spectra of TiO2/C samples (graphite region).
The nature of the carbon species in the TiO2/C samples was further investigated using 13C solid
state NMR (Figure 5). The spectra of the two TiO2/C samples were similar, showing very broad
resonances indicating the presence of both sp3 (0-90 ppm) and sp2 (90-150 ppm) carbon atoms.
This article is protected by copyright. All rights reserved.

10.1002/chem.201800027
Chemistry - A European Journal
For comparison, carbon black mainly shows resonances at 10 and 170 ppm.[38] The large sp3
contribution in our samples is consistent with the Raman results pointing to the formation of an
amorphous carbon phase.
Figure 5. Solid-state 13C CP-MAS NMR spectra of TiO2/C nanocomposites. Spectra were normalized. The
background signal was removed by subtracting the spectrum obtained without the sample.
Typical SEM images of TiO2 and TiO2/C samples are given in Figure 6. In all cases, the samples
were composed of primary nanoparticles, ca. 15 to 25 nm for IPE-derived samples, 10 to 15
nm for THF-derived samples, in good agreement with the crystallite size derived from XRD
(Table 2), indicating that the particles are monocrystalline. The packing of these dense primary
particles accounted for the mesoporosity of TiO2 and TiO2/C samples. In addition, the samples
derived from IPE showed an interesting hierarchical texture, the primary particles forming
micron-sized spherical secondary particles (Figure 6 a-b), as previously reported for IPE-
derived TiO2 prepared at 110 °C.[28] Such a hierarchical texture was not observed for the THF-
derived samples, in which the primary nanoparticles simply formed large agglomerates. There
is no evidence in the SEM images of TiO2/C samples of a carbonaceous phase ; in addition, the
morphology depends on the oxygen donor (IPE or THF) but not on the heat treatment
(calcination or pyrolysis), again suggesting that the carbonaceous phase forms a thin coating
around the anatase nanocrystals.
Figure 6. Scanning electron micrographs of TiO2-IPE (a,b), TiO2-THF (c,d), TiO2/C-IPE (e,f),
TiO2/C-THF (g,h) at different magnifications.
This article is protected by copyright. All rights reserved.

10.1002/chem.201800027
Chemistry - A European Journal
This was confirmed by TEM and high-angle annular dark-field (HAADF) images of the TiO2/C
nanocomposites (Figure 7) which allowed distinguishing the crystalline TiO2 nanoparticles
from the amorphous carbon phases. Crystal planes of anatase TiO2 particles can be clearly
observed, while the carbon coating appears in light gray on the HAADF images, as shown by
EDX analyses (Figures SI5 and SI6). The particle size (average over 100 particles) obtained by
analysis of TEM micrographs was 16.2 ± 5.7 nm for TiO2/C-IPE and 12.5 ± 3.3 nm for TiO2/C-
THF, confirming the smaller size of THF-derived anatase particles. The thickness of the
coatings (average over 50 different areas) was 1.5 ± 0.9 nm for TiO2/C-IPE and 3 ± 1.5 nm for
TiO2/C-THF. The thicker coating in TiO2/C-THF is consistent with the higher C content in this
sample.
Figure 7. TEM images and HAADF images of TiO2/C-IPE (a,b) and
TiO2/C-THF (c,d) samples.
Electrochemical characterization
The TiO2 samples (formulated with and without carbon black) and the TiO2/C nanocomposites
(formulated without carbon black) were tested in half-cells to evaluate the influence of the
carbon coating on their lithium insertion properties (see Table 3 for electrode formulation
details). The galvanostatic cycling curves are displayed in Figure 8 and the corresponding
coulombic efficiencies in Figure SI8.
When formulated with PVDF and carbon black, electrodes based on TiO2-IPE and TiO2-THF
(labeled TiO2-IPE-18CB and TiO2-THF-18CB, respectively) present the expected voltage
profiles for anatase TiO2 under lithiation conditions (Figure SI7a), with a plateau at 1.8 V vs.
Li+/Li, characteristic of the two-phase reaction between the tetragonal solid-solution LixTiO2
(x < 0.5), formed during the drop of potential from the open circuit, and the orthorhombic
lithium titanate Li0.5TiO2; the pseudo-plateau observed below 1.7 V vs. Li+/Li corresponds to
the formation of LiTiO2 phase.[39] The TiO2-IPE-18CB electrode displayed a very high specific
capacity of 160 mA h g-1 after 30 cycles at a current density of 336 mA g-1 (Figure 8a). In
This article is protected by copyright. All rights reserved.

10.1002/chem.201800027
Chemistry - A European Journal
comparison, the TiO2-THF-18CB electrode showed a lower capacity of 110 mA h g-1 in similar
cycling conditions. As the specific surface area of TiO2-THF is higher than that of TiO2-IPE,
this is possibly related to the hierarchical texture of the TiO2-IPE sample, which could facilitate
the accessibility of lithium ions to the surface of anatase nanocrystals. Both samples showed
good cycling performances with coulombic efficiencies higher than 97 % after 10 cycles at a
current density of 16.8 mA g-1, and above 99% after only three cycles at 336 mA g-1 (Figure
SI8a). These results point out the excellent reversibility of these materials. The capacity loss of
TiO2-IPE-18CB and TiO2-THF-18CB electrodes is approximately 30 mA h g-1 between the 2nd
and the 10th cycle at 16.8 mA g-1, and slightly lower than 10 mA g-1 between the 21st and the
50th cycle at 336 mA g-1 (Figure 8a).
On the other hand, the specific capacity of electrodes based on TiO2-IPE and TiO2-THF
formulated without carbon black (Table 3) was lower than 50 mA h g-1 after 50 cycles (Figures
8b and 8c). The coulombic efficiency remains below 95 % for both materials throughout the
first ten cycles (Figures SI8b and SI8bc). In comparison, the nanocomposite electrodes TiO2/C-
IPE and TiO2/C-THF (also formulated without carbon black) exhibit significantly higher
specific capacities for all current densities. This is a direct evidence of the improvements
stemming from the intrinsic carbon coatings on TiO2 particles. Going further into details,
TiO2/C-THF displays better cycling behavior, especially at 336 mA g-1, than TiO2/C-IPE. Its
specific capacity in charge remains as high as 65 mA h g-1 from cycle 25 to 50 (Figure 8c). This
stability can also be evidenced in the coulombic efficiency (Figure SI8c), where the value is
above 99.7 %. It is then clear that the presence of a carbon coating improves the electrochemical
properties of ether-derived TiO2. Moreover, the higher carbon content for TiO2/C-THF
(18.9 wt.%) than for TiO2/C-IPE (6.6 wt%) leads to better cycling performances at high current
densities.
This article is protected by copyright. All rights reserved.

10.1002/chem.201800027
Chemistry - A European Journal
(a)
(b)
(c)
Figure 8. Comparison of the cycling performance at different current densities: (a) TiO2 samples formulated
with carbon black; (b) TiO2 and TiO2/C materials derived from IPE (without carbon black); (c) TiO2 and TiO2/C
materials derived from THF (without carbon black). Voltage window of 1.2-2.5 V versus Li+/Li; filled and open
symbols refer to reduction (discharge) and oxidation (charge), respectively.
The specific capacities obtained for TiO2/C-IPE and TiO2/C-THF are significantly lower than
that found for the TiO2 electrodes formulated with carbon black, showing that as far as
electrochemical properties are concerned, the intrinsic carbon coating is less effective than a
carbon black additive. Several hypotheses can be proposed to explain this situation. First, the
intrinsic carbon formed at 750 °C is amorphous, whereas the carbon black presents a more
graphitic nature. Accordingly, both electronic conductivity and lithium diffusion should be
more favorable for the carbon black additive than for the intrinsic carbon. Secondly, the intrinsic
carbon forms a homogeneous coating around the anatase nanocrystals, which could hinder
lithium ion diffusion toward the TiO2, thus limiting specific capacity. On the contrary, the
This article is protected by copyright. All rights reserved.

10.1002/chem.201800027
Chemistry - A European Journal
carbon black particles simply bridge the TiO2 nanocrystals, allowing for improved electronic
conductivity while preserving the accessibility of TiO2 to lithium ions.
Conclusions
In conclusion, we have successfully synthesized mesoporous TiO2/C nanocomposites by an
original, solvent-free NHSG route. We proposed for the first time to use simple ethers in a
stoichiometric amount not only as oxygen donors but also as carbon sources. Increasing the
reaction temperature under autogenous conditions favored the occurrence of secondary
reactions leading to increased carbon yields. After pyrolysis at 750 °C, mesoporous
TiO2/carbon nanocomposites built of anatase nanocrystals coated by an amorphous carbon film
were obtained, as shown by XRD, Raman and 13C CPMAS-NMR spectroscopies, and electron
microscopy. The nature of the ether impacted the carbon content and the carbon coating
thickness, as well as the morphology of the nanocomposite. The electrochemical behavior of
TiO2/C nanocomposites and their TiO2 counterparts toward lithium insertion was investigated,
providing additional insight on the carbon coating.
This simple approach opens new opportunities for the NHSG process as it should provide a
general atom-economic method to synthesize mesoporous metal oxide-carbon nanocomposites.
Much remains to be done to explore different oxide/carbon and mixed oxide/carbon systems,
different oxygen and carbon donors, and to investigate the potential applications of these
nanocomposites.
Experimental Section
Materials synthesis
To avoid water, all manipulations were done in a glovebox under argon atmosphere (<10 ppm
of water and O2). Titanium (IV) chloride, (Sigma-Aldrich 99%) was used as received;
diisopropyl ether, IPE, (TCI chemicals 99.9 %) was dried by distillation over sodium;
tetrahydrofuran (THF, Fisher Scientific 99.8 %) was dried over a PureSolve MD5 solvent
purification system from Innovative Technology. The amount of water in the ethers was
controlled with a Karl Fischer coulometer before reaction (H2O < 10 ppm for IPE and H2O < 5
ppm for THF).
TiCl4 (12.52 mmol) then the ether (25.04 mmol) were added into a 23 mL stainless steel
digestion vessel with a PTFE lining from Parr Instrument Company. The sealed digestion vessel
was heated in an oven at 180 °C (unless otherwise mentioned) for 4 days under autogenous
pressure. The gels were dried under vacuum at 120 °C for 5 hours, then the resulting xerogels
were pyrolyzed under argon atmosphere (50 mL min-1) at 750 °C for 4 hours (heating rate 10
°C min-1) to obtain the TiO2/C nanocomposites. For comparison, the corresponding TiO2
samples were obtained by calcination in air (50 mL min-1) of the xerogels at 500 °C for 5 hours
(heating rate 10 °C min-1).
This article is protected by copyright. All rights reserved.

10.1002/chem.201800027
Chemistry - A European Journal
The samples were labeled according to their composition (TiO2 or TiO2/C) and to the ether used
(IPE or THF). For instance, the nanocomposite obtained using THF as an O-donor and C source
is labeled TiO2/C-THF, and the corresponding oxide was labeled TiO2-THF.
Characterization
EDX analyses were acquired with a JEOL CENTURIO detector and electron energy loss
analyses were detected with a GATAN GIF QUANTUM. The carbon content of the
nanocomposites was determined by thermogravimetric analysis (TGA) in air, using a Netzsch
Simultaneous Thermal Analyzer STA 409 PC Luxx with a heating rate of 10 °C min-1 in the
20-1200 °C range, under a dry air atmosphere (50 mL min-1).
X-Ray diffraction (XRD) patterns were obtained using a PANalytical X’Pert Pro MPD
diffractometer, with the Kα radiation of Cu (λ=1.5418 Å) and a step size of 0.033° (2θ scale)
into the 6-90° interval. Crystallite size was estimated using the Scherrer equation from the most
intense reflection (25.28 ° 2θ). Scanning electron microscopy images were acquired with a
Hitachi S-4800 electron microscope. Transmission electron microscopy images were acquired
with a high voltage 200 kV JEOL JEM ARM 200F transmission electron microscope equipped
with a STEM level and a spherical aberration corrector; STEM images were taken with high
angle annular dark field (HAADF). N2-physisorption experiments were carried out at -196 °C
on a Micromeritics TriStar 3000. Samples were out-gassed under vacuum at 110 °C overnight.
Equivalent BET specific surface areas were determined in the relative pressure range P/P0 from
0.08 to 0.25. The total pore volume was measured at P/P0>0.985. Raman spectra were obtained
on a Horiba Jobin Yvon LabRAM ARAMIS microspectrometer with a HeNe laser; the spectra
recorded in the range of 50 to 2000 cm-1 were normalized. Solid-state 13C NMR spectra were
recorded with a VARIAN VNMRS 300 at 75.44 MHz, using a T3 NMR probe with 3.2 mm
ZrO2 rotors; experiments were performed using cross polarization and magic angle spinning
(CP-MAS) with a spinning frequency of 15 kHz, a contact time of 1.5 ms and a 5 s recycle
delay; polymer inserts were removed from the rotor and short caps were used instead to
minimize the 13C background.
Galvanostatic electrochemical characterizations were performed at room temperature on a
BTS3000 (5 V, 1 mA) instrument from Neware Battery using CR2032-type coin cells.
Electrodes were composed of the active material (TiO2 and TiO2/C), polyvinylidene fluoride
(PVDF, Solef® 5130, Solvay) as a binder and eventually carbon black (acetylene black NY50A
from SN2A company). Films composition can be found in Table 3.
After stirring in N-methyl-2-pyrrolidone, the slurry was tape casted uniformly at 150 μm onto
a copper current collector using a 3540 bird film applicator from Elcometer. Electrodes
(diameter 12.7 mm) were cut with a disk cutter then dried under vacuum at 80 °C overnight.
Coin cells were assembled in a glove box under Ar atmosphere (O2<0.5 ppm, H2O<0.5 ppm),
using lithium metal both as reference and as counter-electrode. The electrolyte was 1 M LiPF6
in propylene carbonate, ethylene carbonate, and dimethyl carbonate (PC:EC:3DMC, 1:1:3 vol.)
with 5% fluoroethylene carbonate and 1% vinylene carbonate as additives. Whatman glass fiber
filters were used as the separator. The electrochemical galvanostatic measurements were done
in the voltage range of 2.5-1.2 V vs. Li+/Li at different current densities: 16.8, 168, and
This article is protected by copyright. All rights reserved.

10.1002/chem.201800027
Chemistry - A European Journal
336 mA g-1. The latter densities have been selected as corresponding to specific capacities of
C/20, C/2 and C, respectively (where C=335.6 mA g-1) for pure titanium dioxide. Between each
current density change, the cell was cycled once at 16.8 mA/g. The specific capacities presented
hereafter have been calculated based on the TiO2 content of the electrodes.
Table 3. Nominal active material and carbon black of the electrode films, and TiO2 and intrinsic C content
calculated from the films composition and from the pyrolytic C content of the TiO2/C nanocomposites (6.6 wt%
C for TiO2/C-IPE and 18.9 wt% C TiO2/C-THF). In all cases, the PVDF content was 6 wt%.
Material
content (wt%)
76.0
PVDF content
Carbon black
content (wt%)
TiO2 content
(wt%)
76.0
Intrinsic C
(wt%)
Electrode
(wt%)
6.0
6.0
TiO2-IPE-18CB
TiO2-THF-18CB
TiO2-IPE
18.0
18.0
0
0
0
0
76.0
94.0
76.0
94.0
6.0
TiO2-THF
94.0
6.0
0
94.0
0
TiO2/C-IPE
TiO2/C-THF
94.0
94.0
6.0
6.0
0
0
87.8
76.2
6.2
17.8
Acknowledgements
This work was supported by the “Centre National de la Recherche Scientifique” and the
University of Montpellier, and through a PhD grant of the “Consejo Nacional de Ciencia y
Tecnología” in Mexico.
Keywords: non-hydrolytic sol-gel, TiO2, carbon coating, atom-economic method
[1] a) G. Q. Lu and X. S. Zhao in Nanoporous Materials: Science and Engineering Vol. World Scientific,
Singapore, 2011; b) Y. Ren, Z. Ma and P. G. Bruce, Chem. Soc. Rev. 2012, 41, 4909-4927; c) A. Corma,
Chem. Rev. 1997, 97, 2373-2419.
[2] a) L. L. Hench and J. K. West, Chem. Rev. 1990, 90, 33-72; b) C. J. Brinker and G. W. Scherer, Sol-gel
Science: the physics and chemistry of sol-gel processing, Academic Press, Boston, MA, 1990, p.
[3] a) A. Vioux, Chem. Mater. 1997, 9, 2292-2299; b) N. Pinna and M. Niederberger, Angew. Chem., Int.
Ed. 2008, 47, 5292-5304; c) P. H. Mutin and A. Vioux, Chem. Mater. 2009, 21, 582-596; d) A. Styskalik,
D. Skoda, C. Barnes and J. Pinkas, Catalysts 2017, 7, 168; e) R. Deshmukh and M. Niederberger, Chem.
Eur. J. 2017, 23, 8542-8570.
[4] M. Niederberger and G. Garnweitner, Chem. - Eur. J. 2006, 12, 7282-7302.
[5] a) D. P. Debecker, V. Hulea and P. H. Mutin, Appl. Catal., A 2013, 451, 192-206; b) P. H. Mutin and
A. Vioux, J. Mater. Chem. A 2013, 1, 11504-11512.
[6] a) J. N. Hay and H. M. Raval, Chem. Mater. 2001, 13, 3396-3403; b) N. Pinna, J. Mater. Chem. 2007,
17, 2769-2774.
[7] B. Ludi, I. Olliges-Stadler, M. D. Rossell and M. Niederberger, Chem. Commun. 2011, 47, 5280-5282.
[8] R. Deshmukh, G. Zeng, E. Tervoort, M. Staniuk, D. Wood and M. Niederberger, Chem. Mater. 2015,
27, 8282-8288.
[9] a) M. Zhi, C. Xiang, J. Li, M. Li and N. Wu, Nanoscale 2013, 5, 72-88; b) M. Dahl, Y. Liu and Y. Yin,
Chem. Rev. 2014, 114, 9853-9889; c) H. Chen, X. Huang, L.-J. Zhou, G.-D. Li, M. Fan and X. Zou,
ChemCatChem 2016, 8, 992-1000; d) M. Busch, W. Schmidt, V. Migunov, A. Beckel, C. Notthoff, A.
Kompch, U. Bergmann, M. Winterer and B. Atakan, Appl. Catal. B 2014, 160-161, 641-650.
[10] a) W. Li, Z. Wu, J. Wang, A. A. Elzatahry and D. Zhao, Chem. Mater. 2014, 26, 287-298; b) X. Chen
and S. S. Mao, Chem. Rev. 2007, 107, 2891-2959. ; c) Y. Bai, I. Mora-Seró, F. De Angelis, J. Bisquert and
This article is protected by copyright. All rights reserved.

10.1002/chem.201800027
Chemistry - A European Journal
P. Wang, Chem. Rev. 2014, 114, 10095-10130; d) U. Bach, D. Lupo, P. Comte, J. E. Moser, F. Weissortel,
J. Salbeck, H. Spreitzer and M. Gratzel, Nature 1998, 395, 583-585; e) J. Bai and B. Zhou, Chem. Rev.
2014, 114, 10131-10176; f) T. Song and U. Paik, J. Mater. Chem. A 2016, 4, 14-31; g) J. Jin, S.-Z. Huang,
J. Liu, Y. Li, D.-S. Chen, H.-E. Wang, Y. Yu, L.-H. Chen and B.-L. Su, J. Mater. Chem. A 2014, 2, 9699-9708.
[11] Y. Liu and Y. Yang, J. Nanomater. 2016, 2016, 8123652/8123651-8123652/8123615.
[12] a) L. H. Nguyen, V. Aravindan, S. A. Kulkarni, F. Yanan, R. R. Prabhakar, S. K. Batabyal and S.
Madhavi, ChemElectroChem 2014, 1, 539-543; b) F. Gligor and S. W. de Leeuw, Solid State Ionics 2006,
177, 2741-2746; c) S. Bach, J. P. Pereira-Ramos and P. Willman, Electrochim. Acta 2010, 55, 4952-4959;
d) M. V. Koudriachova, N. M. Harrison and S. W. de Leeuw, Phys. Rev. Lett. 2001, 86, 1275-1278.
[13] A. L. Linsebigler, G. Lu and J. T. Yates, Chem. Rev. 1995, 95, 735-758.
[14] a) W. Wei, C. Yu, Q. Zhao, G. Li and Y. Wan, Chem. Eur. J. 2013, 19, 566-577; b) S. Plumejeau, M.
Rivallin, S. Brosillon, A. Ayral, L. Heux and B. Boury, Chem. Eur. J. 2016, 22, 17262-17268.
[15] a) L. Zhao, X. Chen, X. Wang, Y. Zhang, W. Wei, Y. Sun, M. Antonietti and M.-M. Titirici, Adv. Mater.
2010, 22, 3317-3321; b) G. Qin, X. Wu and H. Zhang, RSC Adv. 2014, 4, 52438-52450; c) F.-F. Cao, Y.-G.
Guo, S.-F. Zheng, X.-L. Wu, L.-Y. Jiang, R.-R. Bi, L.-J. Wan and J. Maier, Chem. Mater. 2010, 22, 1908-
1914; d) K.-T. Kim, C.-Y. Yu, S.-J. Kim, Y.-K. Sun and S.-T. Myung, J. Power Sources 2015, 281, 362-369;
e) B. Oschmann, D. Bresser, M. N. Tahir, K. Fischer, W. Tremel, S. Passerini and R. Zentel, Macromol.
Rapid Commun. 2013, 34, 1693-1700; f) J. M. Lee, I. Y. Kim, S. Y. Han, T. W. Kim and S.-J. Hwang, Chem.
Eur. J. 2012, 18, 13800-13809.
[16] a) T. Song and U. Paik, J. Mater. Chem. A 2016, 4, 14-31; b) M. Agostini, S. Brutti, M. A. Navarra, S.
Panero, P. Reale, A. Matic and B. Scrosati, Sci. Rep. 2017, 7, 1104.
[17] C. Chen, Y. Huang, C. An, H. Zhang, Y. Wang, L. Jiao and H. Yuan, ChemSusChem 2015, 8, 114-122.
[18] B. Wang, H. Xin, X. Li, J. Cheng, G. Yang and F. Nie, Sci. Rep. 2014, 4, 3729.
[19] a) M. N. Tahir, B. Oschmann, D. Buchholz, X. Dou, I. Lieberwirth, M. Panthoefer, W. Tremel, R.
Zentel and S. Passerini, Adv. Energy Mater. 2016, 6, 1501489; b) F. Mueller, D. Bresser, E. Paillard, M.
Winter and S. Passerini, J. Power Sources 2013, 236, 87-94.
[20] a) L. Mai, C. Huang, D. Wang, Z. Zhang and Y. Wang, Appl. Surf. Sci. 2009, 255, 9285-9289; b) Q.
Xiao, J. Zhang, C. Xiao, Z. Si and X. Tan, Sol. Energy 2008, 82, 706-713.
[21] a) J.-J. Zhang, Z. Wei, T. Huang, Z.-L. Liu and A.-S. Yu, J. Mater. Chem. A 2013, 1, 7360-7369; b) S.-
M. Oh, J.-Y. Hwang, C. S. Yoon, J. Lu, K. Amine, I. Belharouak and Y.-K. Sun, ACS Appl. Mater. Interfaces
2014, 6, 11295-11301.
[22] a) D. Pham-Cong, J.-H. Kim, S.-Y. Jeong, J. H. Choi, J. Kim and C.-R. Cho, Electrochem. Commun.
2015, 60, 204-207; b) X. Bai, T. Li, Y.-X. Qi, X.-P. Gao, L.-W. Yin, H. Li, H.-L. Zhu, N. Lun and Y.-J. Bai,
Electrochim. Acta 2015, 169, 241-247; c) L.-Y. Wang, Y. Wu, J.-P. Han, B. Zhang, X. Bai, Y.-X. Qi, N. Lun,
Y.-M. Cao and Y.-J. Bai, Electrochim. Acta 2017, 245, 186-192.
[23] Y. Ge, H. Jiang, J. Zhu, Y. Lu, C. Chen, Y. Hu, Y. Qiu and X. Zhang, Electrochim. Acta 2015, 157, 142-
148.
[24] L. Tan, L. Pan, C. Cao, B. Wang and L. Li, J. Power Sources 2014, 253, 193-200.
[25] S. Guo, J. Liu, S. Qiu, W. Liu, Y. Wang, N. Wu, J. Guo and Z. Guo, J. Mater. Chem. A 2015, 3, 23895-
23904.
[26] N. K. Memon, D. H. Anjum and S. H. Chung, Combust. Flame 2013, 160, 1848-1856.
[27] a) P. Arnal, R. J. P. Corriu, D. Leclercq, P. H. Mutin and A. Vioux, J. Mater. Chem. 1996, 6, 1925-
1932; b) T. J. Trentler, T. E. Denler, J. F. Bertone, A. Agrawal and V. L. Colvin, J. Am. Chem. Soc. 1999,
121, 1613-1614; c) M. Niederberger, M. H. Bartl and G. D. Stucky, Chem. Mater. 2002, 14, 4364-4370;
d) J. Joo, S. G. Kwon, T. Yu, M. Cho, J. Lee, J. Yoon and T. Hyeon, J. Phys. Chem. B 2005, 109, 15297-
15302; e) G. Garnweitner, M. Antonietti and M. Niederberger, Chem. Commun. 2005, 397-399.
[28] A. M. Escamilla-Perez, N. Louvain, M. Kaschowitz, S. Freunberger, O. Fontaine, B. Boury, N. Brun
and P. H. Mutin, J. Sol-Gel Sci. Technol. 2016, 79, 270-278.
[29] P. Arnal, R. J. P. Corriu, D. Leclercq, P. H. Mutin and A. Vioux, Chem. Mater. 1997, 9, 694-698.
[30] I. Olliges-Stadler, M. D. Rossell and M. Niederberger, Small 2010, 6, 960-966.
[31] W. Cai, R. D. Piner, F. J. Stadermann, S. Park, M. A. Shaibat, Y. Ishii, D. Yang, A. Velamakanni, S. J.
An, M. Stoller, J. An, D. Chen and R. S. Ruoff, Science 2008, 321, 1815-1817.
This article is protected by copyright. All rights reserved.

10.1002/chem.201800027
Chemistry - A European Journal
[32] M. Thommes, Chem. Ing. Tech. 2010, 82, 1059-1073.
[33] a) T. Ohsaka, F. Izum and Y. Fujiki, J. Raman Spectrosc. 1978, 7, 321-324; b) R. M. Kadam, B.
Rajeswari, A. Sengupta, S. N. Achary, R. J. Kshirsagar and V. Natarajan, Spectrochim. Acta, Part A 2015,
137, 363-370.
[34] a) A. C. Ferrari and J. Robertson, Phys. Rev. B: Condens. Matter Mater. Phys. 2000, 61, 14095-
14107; b) F. Tuinstra and J. L. Koenig, J. Chem. Phys. 1970, 53, 1126-1130; c) M. J. Matthews, M. A.
Pimenta, G. Dresselhaus, M. S. Dresselhaus and M. Endo, Phys. Rev. B: Condens. Matter Mater. Phys.
1999, 59, R6585-R6588.
[35] A. Sadezky, H. Muckenhuber, H. Grothe, R. Niessner and U. Poschl, Carbon 2005, 43, 1731-1742.
[36] a) N. P. Ivleva, U. McKeon, R. Niessner and U. Poschl, Aerosol Sci. Technol. 2007, 41, 655-671; b)
B. Dippel, H. Jander and J. Heintzenberg, Phys. Chem. Chem. Phys. 1999, 1, 4707-4712.
[37] a) D. R. Tallant, T. A. Friedmann, N. A. Missert, M. P. Siegal and J. P. Sullivan, Mater. Res. Soc. Symp.
Proc. 1998, 498, 37-48; b) H. J. Scheibe, D. Drescher and P. Alers, Fresenius J. Anal. Chem. 1995, 353,
695-697.
[38] H. Darmstadt, C. Roy, S. Kaliaguine, G. Xu, M. Auger, A. Tuel and V. Ramaswamy, Carbon 2000, 38,
1279-1287.
[39] a) M. Wagemaker, W. J. H. Borghols and F. M. Mulder, J. Am. Chem. Soc. 2007, 129, 4323-4327; b)
U. Lafont, D. Carta, G. Mountjoy, A. V. Chadwick and E. M. Kelder, J. Phys. Chem. C 2010, 114, 1372-
1378; c) T. Froeschl, U. Hoermann, P. Kubiak, G. Kucerova, M. Pfanzelt, C. K. Weiss, R. J. Behm, N.
Huesing, U. Kaiser, K. Landfester and M. Wohlfahrt-Mehrens, Chem. Soc. Rev. 2012, 41, 5313-5360.
This article is protected by copyright. All rights reserved.