Chitosan-templated synthesis of few-layers boron nitride and its unforeseen activity as a Fenton catalyst

Article abstract of DOI:10.1002/chem.201405469

A novel procedure for the preparation of fewlayers boron nitride (BN), either as films or as suspended BN platelets, is presented, based on the pyrolysis of chitosan, which serves both as a matrix to embed ammonium borate and as a template for BN synthesis.The resulting BN samples are characterized by XRD, Raman and X-ray photoelectron spectroscopy, and by TEM and AFM imaging.The samples exhibit deep UV emission, which is characteristic of high quality BN. This template synthesis and the easy exfoliation of BN platelets facilitate the use of BN as an extremely highefficiency Fenton catalyst for the generation of highly aggressive hydroxyl radicals in water.

Full text of DOI:10.1002/chem.201405469

DOI: 10.1002/chem.201405469
Full Paper
&
Boron Nitride
Chitosan-Templated Synthesis of Few-Layers Boron Nitride and its
Unforeseen Activity as a Fenton Catalyst
Ana Primo,*[a] Sergio Navalꢀn,[a] Abdullah M. Asiri,[b] and Hermenegildo Garcꢁa*[a, b]
Abstract: A novel procedure for the preparation of few-
layers boron nitride (BN), either as films or as suspended BN
platelets, is presented, based on the pyrolysis of chitosan,
which serves both as a matrix to embed ammonium borate
and as a template for BN synthesis. The resulting BN samples
are characterized by XRD, Raman and X-ray photoelectron
spectroscopy, and by TEM and AFM imaging. The samples
exhibit deep UV emission, which is characteristic of high
quality BN. This template synthesis and the easy exfoliation
of BN platelets facilitate the use of BN as an extremely high-
efficiency Fenton catalyst for the generation of highly ag-
gressive hydroxyl radicals in water.
Introduction
as DMF or N-methylpyrrolidone also results in a negligible dis-
persed mass. Therefore, it is highly desirable to develop alter-
native procedures for the preparation of single-layer or few-
layers BN sheets, either as films or in suspension in water or
low-viscosity solvents.
Hexagonal boron nitride (h-BN) is constituted by layers of alter-
nating boron and nitrogen atoms in hexagonal arrangement
having a graphite-like structure.[1,2] Single-layer BN (sl-BN) is
one atom thick two-dimensional (2D) material isostructural
with graphene.[3] However, in contrast to graphene, BN is an in-
sulator with large band gap and low dielectric constant, shar-
ing with graphene high mechanical resistance.[4–6] Another dis-
tinctive property of BN sheets in comparison to graphene is its
high thermal stability and chemical inertness.[7,8] Whereas gra-
phene undergoes combustion upon heating the material in
the presence of oxygen, sl-BN does not undergo combustion
or chemical oxidation. The main application of sl-BN in elec-
tronics is to act as a nanometric barrier to prevent charge leak-
age.[9, 10] Another unique property of sl-BN is light emission in
the deep UV region (lꢀ200 nm).[11,12] To our knowledge, the
catalytic properties of BN have, to date, not been reported.
Among the different preparation procedures for sl-BN films,
the most important ones are chemical vapor deposition of
aminoborane and other precursors on metal surfaces[8,11,13–17]
or exfoliation of bulk h-BN particles.[18–23] Thin-walled bubbles
of BN or CxBN have been obtained by multistage heating of
aminoborane in the absence or presence of ethanol.[24] Howev-
er, compared to graphene, exfoliation of bulk h-BN particles is
considerably inefficient, due to the reluctance of h-BN to un-
dergo oxidation, which is the key step in the conversion of
graphite into single-layer graphene oxide.[25] The mass of h-BN
dispersed is negligible. Exfoliation of bulk BN in solvents such
Herein, an innovative procedure for the synthesis of few-
layers BN (fl-BN) sheets deposited on arbitrary substrates or
suspended in conventional solvents is presented. Our proce-
dure does not require specific equipment and is reliable,
making sl-BN or fl-BN available for study. We found that the
chemical inertness of fl-BN makes it a suitable and efficient cat-
alyst for the Fenton-like degradation of pollutants.
Results and Discussion
The procedure reported herein can be applied to the prepara-
tion of nanometric fl-BN films or loosely crystallized h-BN pow-
ders that can be easily exfoliated by sonication. The process
consists of the pyrolysis under inert atmosphere at tempera-
tures above 9008C of a homogeneous mixture of (NH4)3BO3
and chitosan (Figure 1). Chitosan is a natural polysaccharide
obtained from the skin of crustaceans and is the largest fisher-
ies industry waste.[26] Pyrolysis under an inert atmosphere of
chitosan, as single source of C and N, affords a turbostratic
graphitic residue that can be easily exfoliated to form N-doped
graphene (7 wt% N).[27–29] Herein, the ability of chitosan to
form uniform films and embed inorganic compounds during
pyrolysis has been used to template the formation of thin fl-
BN films together with the graphitic carbon residue. The tem-
plating effect of chitosan is derived from a combination of fea-
tures, including the ability of chitosan to embed (NH4)3BO3 and
still form films and, in the pyrolysis step, to segregate itself as
graphene separate from the BN phase. Subsequently, combus-
tion of the carbon material in air renders either fl-BN films or
easily exfoliable BN powders. The merit of the present proce-
dure is that, whereas highly crystalline BN powders do not un-
dergo exfoliation upon sonication due to the close packing of
[a] Dr. A. Primo, Dr. S. Navalꢀn, Prof. H. Garcꢁa
Instituto de Tecnologꢁa Quꢁmica CSIC-UPV and Departamento de Quꢁmica
Univ. Politꢂcnica de Valencia, Av. de los Naranjos s/n
46022 Valencia (Spain)
E-mail: hgarcia@qim.upv.es
[b] Dr. A. M. Asiri, Prof. H. Garcꢁa
Center of Excellence for Advanced Materials Research
King Abdulaziz University, Jeddah (Saudi Arabia)
Chem. Eur. J. 2014, 20, 1 – 8
1
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
sidual material is shown to have
BN as chemical composition (see
below).
BN films or powders resulting
after pyrolysis under inert atmos-
phere and air calcination at
10008C were characterized by
XRD, Raman spectroscopy, XPS,
AFM, and electron microscopy.
All data indicate the formation
Figure 1. Preparation procedure for fl-BN films supported on arbitrary substrates. Steps i, iii and iv can be applied
to obtain h-BN powders that can be easily exfoliated into few-layers BN platelets. i) Filtration of aqueous solutions;
ii) spin coating; iii) pyrolysis at 9008C under Ar; iv) calcination under air at 10008C.
the layers and the strong interaction between them, the pres-
ent BN material from chitosan template undergoes a considera-
ble degree of exfoliation upon sonication. This contrasting be-
havior is likely to be attributable to the poor crystallinity and
loose packing of the BN material obtained in pyrolysis, allow-
ing much easier deaggregation of the layers. Up to 50 wt%
loadings of (NH4)3BO3 vs. chitosan were achieved. Most of the
experiments reported below were carried out with a (NH4)3BO3
loading of 30 wt%.
of highly crystalline BN films or powders.
XRD of h-BN is characterized by a (0002) reflection at 278.
Films of the material, resulting from pyrolysis of (NH4)3BO3–chi-
tosan mixtures and subsequent air calcination, supported on
quartz also give rise to this characteristic peak on top of
a broad background arising from the glass substrate. Figure 2
shows the XRD of a thick h-BN film (150 nm) on quartz show-
ing the expected (0002) reflection. This diffraction peak was
also observed for powders prepared according to Figure 1,
except that spin coating (step ii) was replaced by lyophilization
of the solution. Thinner films (50–150 nm) also showed this
narrow peak, albeit with less intensity. However, for films of ini-
tial thickness below 50 nm before pyrolysis, the peak was
barely distinguishable from the background or was completely
obscured. This behavior is expected for very thin films in which
the number of BN layers is small.
Thermogravimetric analysis (TGA) of the samples after pyrol-
ysis under argon containing C, B and N (sample after steps i–iii
in Figure 1) was performed to determine the C vs. BN content.
There was an initial weight loss from ambient temperature to
1508C of about 6.4 wt% that we attribute to co-adsorbed
water (Figure 2). After this weight loss, there was a plateau
from 150 to 5508C, indicating that the material is stable up to
this temperature. From 5508C up to 10008C there was
a second weight loss of 13.8 wt%. This second weight loss was
due to the combustion of residual carbon derived from chito-
san present in the material. The high decomposition tempera-
ture is probably due to the effect of BN retarding the combus-
tion of the carbon residue. According to this TGA, about
79.8 wt% of the material is stable at 10008C under air. This re-
The same relationship between peak intensity and film thick-
ness was shown by Raman spectroscopy. In contrast to gra-
phene-based materials, which have relatively intense Raman
bands, h-BN only gave rise to a sharp medium-intensity peak
at about 1380 cmꢁ1. This expected vibration h-BN peak was de-
tected for powders, but its intensity was too weak to be de-
tectable for h-BN films of nanometric thickness. Figure 2 shows
Figure 2. 1) Raman spectra of powders (a) and films on glass (b) of (NH4)3BO3–chitosan mixtures after pyrolysis under Ar at 9008C and subsequent air calcina-
tion at 10008C. The sharp peak at 1380 cmꢁ1 has been reported as characteristic of BN. The inset shows the emission from BN powders having
a lmax =215 nm upon excitation to 190 nm. This emission at such short wavelengths is due to the collapse by recombination of electrons and holes in high-
quality BN materials; 2) XRD of a glass substrate submitted to pyrolysis at 9008C under argon and subsequent air calcination at 10008C in the absence (a) or
in the presence (b) of a thin h-BN film. The sharp peak appearing at 278 corresponds to h-BN. The inset shows the TGA profile of the residue after pyrolysis at
9008C under inert atmosphere of a mixture of chitosan and ammonium borate (31 wt%).
&
&
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
2
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
the Raman spectra recorded for h-BN powders and films. Since
single-layer graphene could be clearly detected by our Raman
spectrophotometer, the absence of any peak for thin BN films
indicates the total absence of carbon after air calcination.
XPS confirmed that, after calcination, both films and pow-
ders contained N and B with a 1:1 atomic ratio, detected at
398.2 and 190.75 eV, respectively (Figure 3). The B 1s peak
Figure 3. B 1s and N 1s XPS peaks recorded for BN films on glass obtained
by pyrolysis at 9008C under argon of (NH4)3BO3–chitosan films and subse-
quent air calcination. Quantification of the intensity of the peaks indicates
that the atomic B/N ratio is 1:1. The B 1s peak has a single component
whose binding energy agrees with that reported for BN. In contrast, N 1s
peak has two components in 90:10 ratio. The major one has a binding
energy coincident to that reported for BN and the minor one is attributed
to N atoms at the sheet periphery.
Figure 4. a) Optical microscopy image of a BN film on quartz obtained by
pyrolysis of (NH4)3BO3–chitosan at 9008C under argon and subsequent calci-
nation in air at 10008C. b) AFM image of a cut on the surface of a thin film
of BN on quartz. The inset shows that the vertical height along the cut is
about 40 nm. c) AFM image of BN platelets obtained by sonication of h-BN
powders in water, showing dimensions ranging from 10 to 150 nm. The
inset presents the statistical distribution of the height of these platelets
showing that over 80% of the particles are constituted by fewer than four
BN layers.
shows a single component with binding energy coincident
with that reported for B atoms in BN.[13] In the case of the N 1s
peak, we observed two components, at an approximately 9:1
atomic ratio. For the more abundant component (90% of the
total), the binding energy coincided with that reported for N
atoms in BN.[13] The second N component (10% of the total)
appeared at a lower binding energy (394 eV), attributable to N
in the periphery and defects of the material. Accordingly, it is
proposed that in our BN material, N is the element located at
the periphery of the sheets.
layer thickness. Herein, the thickness of the resulting BN layer
was varied from 200 to 20 nm by increasing the spin-coating
velocity and maintaining the (NH4)3BO3–chitosan concentration.
As an example, Figure 4b shows the AFM image of a BN sur-
face with a thickness of about 40 nm (the roughness was mea-
sured for a similar image). It should be noted, however, that
the hardness of the BN film complicates these measurements
since cuts on these surfaces are not easy.
Our synthetic procedure for the preparation of BN films
allows control of the film thickness by modifying the solution
concentration and the spin coating rate. Chitosan forms high-
quality nanometric films without defects and subnanometric
roughness on arbitrary substrates, this being the key point to
form single-layer or few-layers graphene on arbitrary sub-
strates.[27] In the present case, the presence of (NH4)3BO3 em-
bedded in chitosan does not interfere with its ability to form
this type of films. Upon pyrolysis and calcination of (NH4)3BO3–
chitosan films on quartz, optical microscopy showed smooth,
highly cracked, transparent films. Prior studies on the forma-
tion of h-BN films on Ni substrates by chemical vapor deposi-
tion also reported a highly cracked film throughout the area
covered by h-BN.[13] Figure 4 shows the optical microscopy
image of BN films obtained by pyrolysis and calcination of
(NH4)3BO3–chitosan mixtures, which agrees well with reported
images.[13]
High-resolution transmission electron microscopy shows the
crystallinity of the material, and, the single layer morphology
can be observed at the borders. Selected-area electron diffrac-
tion shows intense spots indicating that the material is a single
crystal with hexagonal arrangement at nanometric scale. In
some parts, it has been possible to count the number of
layers. As an example, Figure 5 shows an image of these crys-
tals of 12 and 16 layers in two different particles (Figure 5a). A
frontal view shows the material crystallinity (Figure 5b).
One of the properties that is characteristic of high-quality
BN layers is deep UV emission.[11,12] Typically, organic materials
emit in the UV or in the visible region, whereas materials
having an emission below 220 nm are very scarce. BN is
unique in this property. The BN samples obtained following
the procedure described herein exhibited photoluminescence
The roughness and thickness of thin BN films on quartz
were determined by AFM. Roughness values were measured
by scanning the height variations of a 3ꢃ3 mm2 area and gave
a value about 0.6 nm, agreeing with the expected BN single-
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
3
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
water leading to their oxidative degradation up to the com-
plete mineralization to CO2 and inorganic salts. The Fenton re-
action is typically carried out using stoichiometric amounts of
iron(II) salts that cause OꢁO bond cleavage to give the hydrox-
ide anion and hydroxyl radical [Eq. (1)]. There is an interest in
developing catalytic systems for the Fenton reaction in where
a site capable of transferring electrons to H2O2 initiates the
Fenton reaction.[30] Subsequently, due to the intermediate O
oxidation state in peroxides, H2O2 itself can reduce the oxi-
dized catalytic site evolving O2 in the process.
Fe2þ þ HꢁOꢁOꢁH ! Fe3þ þ ꢂOH þ OHꢁ
ð1Þ
We anticipated that the presence of p and p* orbitals in BN,
as well as N atoms at the periphery of the sheets and possible
dangling bonds and defects, would promote this catalytic ac-
tivity. To check this possibility, we selected the degradation of
phenol as a test reaction and followed its disappearance by
H2O2 oxidation in the presence and absence of fl-BN. This reac-
tion is of interest for waste water treatment of nonbiodegrada-
ble pollutants. Figure 6 shows the temporal profiles of phenol
disappearance and H2O2 consumption, showing that the
Fenton reaction can be catalyzed by fl-BN platelets. Complete
phenol disappearance (Figure 6A) with a H2O2 consumption of
about 8 equivalents (Figure 6B) was achieved. Furthermore,
the fl-BN catalyst was recovered by centrifugation and used in
a second Fenton reaction, giving rise to coincident temporal
profiles (Figure 6A and B, open circles), indicating that fl-BN is
Figure 5. TEM images (a, b and d) of BN flakes after sonication in water.
Panel (c) shows selected-area electron diffraction of the sample.
at 215 nm upon 190 nm excitation, this photoluminescence
being distinctive of BN layers (inset of Figure 2.1).
To date, the unavailability of BN platelets has ren-
dered study of their chemical properties impossible.
In the present case, pyrolysis of (NH4)3BO3–chitosan
powders and subsequent air calcination led to parti-
cles whose XRD and Raman spectra (Figure 2) corre-
spond to h-BN particles. Commercial h-BN powders
cannot be exfoliated by sonication. In contrast, the
loose stacking of h-BN particles obtained in the pres-
ent case makes possible their dispersion upon sonica-
tion in water, ethanol, CH2Cl2, or other solvents, as re-
vealed by dynamic laser scattering. AFM images of
the BN platelets after sonication in water (Figure 4c)
show that they are constituted by fl-BN particles of
about 1.5 nm average height. The high degree of ex-
foliation explains why these fl-BN platelets form in-
definitely persistent suspensions in water. To date,
preparation of fl-BN suspensions in common solvents
had proven difficult.
Having fl-BN suspensions in water, we considered
Figure 6. Phenol degradation (A) and H2O2 consumption (B) using fresh (&) and used
(*) fl-BN or commercial BN (*) as catalyst. C) Product distribution as a function of time:
the possibility that, due to its chemical stability, fl-BN
could be a suitable catalyst to promote reactions in hydroquinone (*), quinone (*), catechol (~). D) Phenol degradation in the absence
(&) or in the presence (*) of DMSO as hydroxyl radical scavenger. Reaction conditions:
which highly aggressive chemical species are formed
Catalyst (200 mgLꢁ1), phenol (100 mgLꢁ1), H2O2 (400 mgLꢁ1), DMSO/H2O2 molar
as reaction intermediates. Based on this hypothesis,
ratio=10:1, pH 3, room temperature. Inset. Experimental EPR spectra (black lines) of
and to show the potential of fl-BN as catalyst, we
a) BN catalyst+H2O2 +PBNO at pH 3; Hyperfine coupling constants of PBNO–OH adduct:
·
tested the activity of fl-BN to promote catalytic AN =15.5 G; AH =2.75 G, coinciding with reported values in the literature for the OH radi-
cal–PBNO adduct (red line).[39] b) BN catalyst+PBNO at pH 3. Hyperfine coupling con-
Fenton decomposition of H2O2 to generate hydroxyl
stants from PBNO degradation AN =14.58 G; AH =13.90 G, coinciding with reported
radicals.[30] Hydroxyl radicals are the most aggressive
values in the literature for tert-butyl aminoxyl (red lines).[39] Conditions of the EPR meas-
chemical species known in aqueous media,[31] and are
urements: BN catalyst (100 mgLꢁ1), H2O2 (200 mgLꢁ1), PBNO/H2O2 molar ratio=1:1, pH 3,
known to react with most organic compounds in room temperature, 30 min reaction time.
&
&
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
4
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
·
not deactivated in the process. In an additional experiment to
assess fl-BN catalytic stability, a 10 times-larger amount of
phenol (1000 mgLꢁ1), close to the saturation limit at the reac-
tion pH, was used alongside a 4 times-lower catalyst amount
(50 mgLꢁ1), which effected complete phenol disappearance
even under these conditions. Furthermore, H2O2 was not com-
pletely consumed (only 80%), maintaining the same 8 equiva-
lent consumption as in the other experiments, indicating that
there is not deactivation even under these conditions.
To support the generation of free OH radicals, we carried
out a test of phenol degradation in the presence of dimethyl
·
sulfoxide (DMSO). DMSO is a known quencher of OH radi-
cals.[36–38] DMSO acting as an OH quencher completely stops
·
·
phenol degradation (Figure 6D). We also performed OH radical
trapping using phenyl tert-butyl nitrone (PBNO) as quencher,
monitoring the process by conventional EPR spectroscopy (Fig-
ure 6D, inset). As anticipated, the characteristic EPR spectra
·
corresponding to the OH/PBNO adduct was recorded, provid-
·
Most of the precedents in the catalytic Fenton degradation
of phenol use a very large H2O2 excess, frequently above
1000 times.[31–34] In contrast, herein only 11 equivalents of H2O2
were employed but only 8 equivalents were consumed. This
efficiency in the utilization of H2O2 is remarkable and close to
the theoretical limit. Thus, oxidation of phenol to benzoqui-
none, via dihydroxybenzene intermediate, requires 3 equiva-
lents of H2O2 [Eq. (2)], plus an additional 3 equivalents for site
regeneration. Therefore, a total of 6 equivalents of H2O2 are
necessary to catalytically convert phenol to benzoquinone,
close to the 8 equivalents of H2O2 consumed in these experi-
ments (Figure 6A and B). This is orders of magnitude more effi-
cient that common Fenton catalysts reported. Therefore, the
negligible H2O2 decomposition by fl-BN is very minor com-
pared to previously reported catalysts.[31]
ing firm evidence of the generation of free OH radicals by BN.
With respect to the nature of the active sites that reduce
·
H2O2 to OH radicals, we propose that they could be N atoms
at the periphery, probably bound to oxygen. To support this
proposal, the catalytic activity of pyridine-N-oxide was
checked. It was observed that this molecule was able to pro-
mote degradation of phenol, providing experimental evidence
for the nature of the fl-BN active sites. Accordingly, the very
low activity as a Fenton catalyst measured for commercial h-
BN (Figure 6A and B, closed circles) compared to fl-BN should
be due to the poor dispersibility in water of h-BN and the
smaller population of active sites, rather than to structural dif-
ferences between the two materials.
Conclusion
Herein, we have presented an innovative procedure based on
the use of biomass waste to template the synthesis of fl-BN on
arbitrary substrates or for fl-BN platelets in suspension. The
success of the film formation derives from the known property
of chitosan to form defect-free nanometric films with subnano-
metric roughness and its capability to embed ammonium
borate, templating its pyrolysis. The pyrolysis of the (NH4)3BO3–
chitosan material led to the spontaneous segregation of BN
and subsequent combustion of the carbon afforded fl-BN. The
process can also be carried out with particles that, after easy
exfoliation, formed fl-BN platelets suspended in water and
other solvents. The quality of the BN materials was determined
by electron microscopy and also by the characteristic deep UV
photoluminescence. The BN platelets could be suspended in
water, which allowed this material’s activity as a Fenton cata-
lyst to be tested, demonstrating an extremely high efficiency
for the generation of free hydroxyl radicals without significant
H2O2 decomposition.
Analysis of the product distribution shows the formation of
hydroquinone, catechol, and benzoquinone. The temporal evo-
lution of these products (Figure 6C) shows that they are pri-
mary degradation products appearing since the initial reaction
time, but that they undergo degradation over the course of
the reaction, reaching concentration maxima and, then, their
concentration diminishes at longer times. The high efficiency
in the use of H2O2 and the low excess required for the com-
plete disappearance of phenol using BN as catalyst can be un-
derstood considering that, once formed in the redox site, hy-
droxyl radicals will not have affinity for the surface of BN due
to its chemical inertness and, therefore, they will become free
·OH radicals in solution. In most of the Fenton catalysts report-
Experimental Section
Synthesis of h-BN films
Prior to coating, the quartz supports were treated to increase their
surface hydrophilicity. For this purpose, quartz plates (1ꢃ1 cm2,
1 mm thick) were immersed overnight in a 1m HCl aqueous solu-
tion and then washed exhaustively with Milli Q water, acetone, and
isopropanol by sonication for 15 min in each solvent. After clean-
ing, the quartz plates were submitted to ozonization during
30 min to decompose any residual organic matter. Afterwards, an
aqueous solution of chitosan (1.12 g, high quality chitosan of
MW 60,000–120,000 from Aldrich, ref. 740063, dissolved in a 25 mL
of 0.3m acetic acid aqueous solution) and (NH4)3BO3 (0.45 g) was
·
ed to date, OH radicals become attached to the metal and do
·
not really form free OH radicals.[31] Only in a few cases, using
·
extremely robust diamond nanoparticles, have OH radicals
generated catalytically been claimed as free radicals in solu-
tion,[35] and the present case would be similar, considering that
the surface of BN is even more inert.
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
5
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
deposited on the freshly clean quartz support by spin coating (ve-
locity varying from 2000 to 6000 rpm). The (NH4)3BO3–chitosan film
was pyrolyzed in an oven under argon flow (1 mL x minꢁ1) increas-
ing the temperature at a rate of 0.98C minꢁ1 up to 9008C. Then,
the pyrolyzed film was calcined in air up to 10008C, heating at
tered through 0.2 mm Nylon filters. In the productivity test, 10-fold
sample dilution was performed prior to analysis.
H2O2 measurements
a rate of 88C minꢁ1
.
H2O2 was determined by 40-fold (or 400-fold in the productivity
test) dilution of the reaction mixture aliquots and using
K2(TiO)(C2O4)2 in H2SO4/HNO3 as colorimetric indicator. The indicator
was prepared by dissolving the mixed potassium and titanyl oxa-
late (2.5 g) in concentrated sulfuric acid (2.5 g) and concentrated
nitric acid (1 mL). After dissolution, the mixture was cautiously di-
luted with milliQ water to reach 100 ml indicator solution. The re-
action was allowed to react for 10 min before monitoring at
420 nm in a UV/Vis spectrophotometer.
Exfoliation of h-BN crystals
h-BN powders were obtained by freeze-drying an aqueous solution
of chitosan (1.12 g dissolved in 25 mL of 0.3m HOAc) and
(NH4)3BO3 (0.45 g). The resulting (NH4)3BO3–chitosan mixture was
pyrolyzed under Ar flow at 9008C and subsequently calcined in air
at 10008C. Exfoliation was performed by suspending the h-BN
powders in water and sonicating with an ultrasound horn (400 W)
for 3 h. The resulting dispersion was left to stand for 24 h to allow
any unstable aggregate to form and precipitate. At this time, the
suspension was centrifuged for 30 min at 1500 rpm to obtain
a homogeneous fl-BN dispersion that was indefinitely persistent.
Quenching and EPR spin-trapping experiments
Hydroxyl radical scavenging was carried out using DMSO in the fol-
lowing conditions: Phenol (100 mgLꢁ1, 1.06 mm), H2O2 (400 mgLꢁ1
,
11.76 mm), pH 3, catalyst (200 mgLꢁ1), DMSO/H2O2 molar ratio=
10:1, and 30 h reaction time. EPR spectra of PBNO and hydroxyl
radical adduct were recorded using the following conditions:
a) PBNO (100 mgLꢁ1), H2O2/PBNO molar ratio=1:1, catalyst
(200 mgLꢁ1), pH 3, 30 min reaction time; b) PBNO (100 mgLꢁ1), cat-
alyst (200 mgLꢁ1), pH 3, 30 min reaction time. A Bruker EMX instru-
ment was employed with the following settings: Frequency=
9.803 GHz, sweep width=3489.9 G, time constant=40.95 ms,
modulation frequency=100 kHz, modulation width 1 G, microwave
power=19.92 mW.
Characterization techniques.
Raman spectra were recorded at ambient temperature with
514 nm excitation on a Renishaw In Via Raman spectrometer with
a CCD detector. TEM images were recorded in a Philips CM300 FEG
system with 100 kV operating voltage. XP spectra were recorded
on a SPECS spectrometer with Phoibos 150 9 MCD detector using
a non-monochromatic X-ray source (Al and Mg) operating at 200
W. The samples were evacuated in the prechamber at 1ꢃ
10ꢁ9 mbar. The intensity ratios of components were obtained from
the area of the corresponding peaks after nonlinear Shirley-type
background subtraction and corrected by the transmission func-
tion of the spectrometer. X-ray diffraction patterns were obtained
Acknowledgements
using a Philips X’Pert diffractometer and copper radiation (CuKa
;
Financial support by the Spanish Ministry of Economy and
Competitiveness (Severo Ochoa and CTQ2012–32315) is grate-
fully acknowledged. Generalidad Valenciana is also thanked for
funding (Prometeo 2012/013 and GV/2013/040).
l=1.541178 ꢄ). AFM measurements were conducted in contact
mode in air at ambient temperature using a Veeco apparatus. AFM
were not measured in a clean room and films on glass substrates
may contain dust.
Catalytic tests
Keywords: boron nitride · heterogeneous catalysis · pyrolysis ·
template synthesis · UV emission
A Milli-Q water solution (100 mL) of phenol (100 mgLꢁ1, 1.06 mm)
was placed in a round-bottom flask. fl-BN platelets or commercial
h-BN catalyst (20 mg) were introduced and the suspension sonicat-
ed (400 W) for 3 h. Then, the initial pH was adjusted to 3 using HCl
(0.1m). Finally, H2O2 (400 mgLꢁ1, 30% v/v, 11.76 mm) was added.
The reaction proceeded under ambient conditions. Once the reac-
tion had finished, the catalyst was recovered by filtration (0.2 mm
Nylon filter), washed with NaOH aqueous solution (pH 10; 20 mL),
and with Milli-Q water (20 mL). Then, the solid was used for
a second run. The productivity test was performed using a phenol
solution (1,000 mgLꢁ1) containing the BN catalyst (50 mgLꢁ1), ini-
tial pH 3, and H2O2 as oxidant (4,000 mgLꢁ1) under ambient condi-
tions.
[1] T. P. Kaloni, Y. C. Cheng, U. Schwingenschlçgl, J. Mater. Chem. 2012, 22,
919.
[2] O. Hod, J. Chem. Theory Comput. 2012, 8, 1360.
[3] D. Golberg, Y. Bando, Y. Huang, T. Terao, M. Mitome, C. Tang, C. Zhi, ACS
Nano 2010, 4, 2979.
[4] J. Furthmꢅller, J. Hafner, G. Kresse, Phys. Rev. B 1994, 50, 15606.
[5] M. Pashangpour, Z. Bagheri, V. Ghaffari, Eur. Phys. J. B 2013, 86, 269.
[6] D. Lahiri, F. Rouzaud, T. Richard, A. K. Keshri, S. R. Bakshi, L. Kos, A. Agar-
wal, Acta Biomater. 2010, 6, 3524.
[7] Y. Lin, J. W. Connell, Nanoscale 2012, 4, 6908.
[8] C. Zhi, Y. Bando, C. Tang, H. Kuwahara, D. Golberg, Adv. Mater. 2009, 21,
2889.
[9] M. P. Levendorf, C. J. Kim, L. Brown, P. Y. Huang, R. W. Havener, D. A.
Muller, J. Park, Nature 2012, 488, 627.
[10] G. Giovannetti, P. Khomyakov, G. Brocks, P. Kelly, J. V. D. Brink, Phys. Rev.
B 2007, 76.
[11] J. Yu, L. Qin, Y. Hao, S. Kuang, X. Bai, Y. M. Chong, W. Zhang, E. Wang,
ACS Nano 2010, 4, 414.
[12] K. Watanabe, T. Taniguchi, H. Kanda, Nat. Mater. 2004, 3, 404.
[13] Y. Shi, C. Hamsen, X. Jia, K. K. Kim, A. Reina, M. Hofmann, A. L. Hsu, K.
Zhang, H. Li, Z. Y. Juang, M. S. Dresselhaus, L. J. Li, J. Kong, Nano Lett.
2010, 10, 4134.
Characterization of phenol degradation products
The organic products present in the aqueous solution were deter-
mined by reverse-phase chromatography using Kromasil-C18
column and
a 69:30:1 (v/v/v) mixture of H2O, MeOH, and
CH3COOH as eluent under isocratic conditions with a UV/Vis photo-
diode array detector. The monitoring wavelengths for phenol, hy-
droquinone, catechol, and quinone were 270, 290, 276, and
245 nm, respectively. Prior to analysis, aliquots of 2 mL were fil-
[14] P. Sutter, J. Lahiri, P. Zahl, B. Wang, E. Sutter, Nano Lett. 2013, 13, 276.
&
&
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
6
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
[15] A. Ismach, H. Chou, D. A. Ferrer, Y. Wu, S. McDonnell, H. C. Floresca, A.
Covacevich, C. Pope, R. Piner, M. J. Kim, R. M. Wallace, L. Colombo, R. S.
Ruoff, ACS Nano 2012, 6, 6378.
[16] K. K. Kim, A. Hsu, X. Jia, S. M. Kim, Y. Shi, M. Dresselhaus, T. Palacios, J.
Kong, ACS Nano 2012, 6, 8583.
[17] L. Song, L. Ci, H. Lu, P. B. Sorokin, C. Jin, J. Ni, A. G. Kvashnin, D. G.
Kvashnin, J. Lou, B. I. Yakobson, P. M. Ajayan, Nano Lett. 2010, 10, 3209.
[18] G. Lian, X. Zhang, M. Tan, S. Zhang, D. Cui, Q. Wang, J. Mater. Chem.
2011, 21, 9201.
[27] A. Primo, P. Atienzar, E. Sanchez, J. M. Delgado, H. Garcꢁa, Chem.
Commun. 2012, 48, 9254.
[28] M. Z. Elsabee, E. S. Abdou, Mater. Sci. Eng. C 2013, 33, 1819.
[29] A. Primo, E. Sꢆnchez, J. M. Delgado, H. Garcꢁa, Carbon 2014, 68, 777.
[30] S. Navalon, M. de Miguel, R. Martin, M. Alvaro, H. Garcꢁa, J. Am. Chem.
Soc. 2011, 133, 2218.
[31] S. Navalon, M. Alvaro, H. Garcꢁa, Appl. Catal. B 2010, 99, 1.
[32] S. Caudo, G. Centi, C. Genovese, S. Perathoner, Appl. Catal. B 2007, 70,
437.
[33] G. Centi, S. Perathoner, T. Torre, M. G. Verduna, Catal. Today 2000, 55,
61.
[34] S. Perathoner, G. Centi, Top. Catal. 2005, 33, 207.
[35] S. Navalon, R. Martin, M. Alvaro, H. Garcꢁa, Angew. Chem. 2010, 122,
8581; Angew. Chem. Int. Ed. 2010, 49, 8403.
[36] M. J. Burkitt, R. P. Mason, Proc. Natl. Acad. Sci. USA 1991, 88, 8440.
[37] S. Navalon, R. Martin, M. Alvaro, H. Garcꢁa, ChemSusChem 2011, 4, 650.
[38] S. Navalon, D. Sempere, M. Alvaro, H. Garcꢁa, ACS Appl. Mater. Interfaces
2013, 5, 7160.
[19] Y. Lin, T. V. Williams, J. W. Connell, J. Phys. Chem. Lett. 2010, 1, 277.
[20] R. J. Smith, P. J. King, M. Lotya, C. Wirtz, U. Khan, S. De, A. O’Neill, G. S.
Duesberg, J. C. Grunlan, G. Moriarty, J. Chen, J. Wang, A. I. Minett, V. Nic-
olosi, J. N. Coleman, Adv. Mater. 2011, 23, 3944.
[21] K. Sohaumburg, Vol. WO 2009047324 A2 20090416, 2009.
[22] Y. Wang, Z. Shi, J. Yin, J. Mater. Chem. 2011, 21, 11371.
[23] J. H. Warner, M. H. Rꢅmmeli, A. Bachmatiuk, B. Bꢅchner, ACS Nano 2010,
4, 1299.
[24] X. Wang, C. Zhi, L. Li, H. Zeng, C. Li, M. Mitome, D. Golberg, Y. Bando,
Adv. Mater. 2011, 23, 4072.
[39] G. R. Buettner, Free Radical Biol. Med. 1987, 3, 259.
[25] S. Stankovich, D. A. Dikin, R. D. Piner, K. A. Kohlhaas, A. Kleinhammes, Y.
Jia, Y. Wu, S. T. Nguyen, R. S. Ruoff, Carbon 2007, 45, 1558.
[26] M. N. V. Ravi Kumar, React. Funct. Polym. 2000, 46, 1.
Received: September 30, 2014
Published online on && &&, 0000
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
7
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
FULL PAPER
C¸a va BN, merci: Few-layers boron ni-
tride (BN) is synthesized, either as films
or as suspended BN platelets, by pyroly-
sis of chitosan, which serves as a matrix
to embed ammonium borate and as
a template. The resulting samples exhib-
it deep UV emission characteristics of
high quality BN. The few-layers BN is
a highly efficient Fenton catalyst for the
generation of hydroxyl radicals in water.
&
Boron Nitride
A. Primo,* S. Navalꢀn, A. M. Asiri,
H. Garcꢁa*
&& – &&
Chitosan-Templated Synthesis of Few-
Layers Boron Nitride and its
Unforeseen Activity as a Fenton
Catalyst
&
&
Chem. Eur. J. 2014, 20, 1 – 8
www.chemeurj.org
8
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!