Boron- and fluorine-containing mesoporous carbon nitride polymers: Metal-free catalysts for cyclohexane oxidation

Article abstract of DOI:10.1002/anie.201000120

(Figure Presented) Beefed up with B,F: A boron- and fluorineenriched carbon nitride polymeric semiconductor was synthesized by a facile onestep process using l-butyl-3-methylimidazolium tetrafluoroborate as a soft template. The resulting materials show an

Full text of DOI:10.1002/anie.201000120

Communications
DOI: 10.1002/anie.201000120
Mesoporous Materials
Boron- and Fluorine-Containing Mesoporous Carbon Nitride Poly-
mers: Metal-Free Catalysts for Cyclohexane Oxidation**
Yong Wang,* Jinshui Zhang, Xinchen Wang, Markus Antonietti, and Haoran Li*
Recently, various lightweight materials with diverse nano-
morphologies that contain heteroatoms such as nitrogen,
boron, or fluorine have been actively pursued because of their
unusual properties, such as in catalytic applications or as
semiconductors.[1] For example, ordered and disordered
modifications of carbon nitride (C3N4) extend the property
profile of carbon nanostructures and have numerous potential
areas of applications ranging from semiconductors to fuel
cells.[1a]
This nanocasting method is good for the proof of
principle, but it is hardly transferable to a practical process,
as the templates have to be removed in an extra step involving
aqueous ammonium bifluoride (NH4HF2) or hydrogen fluo-
ride (HF), which are hazardous and not environmentally
friendly and also prohibit further functionalization. A general
and robust method for the synthesis of porous polymeric
carbon nitrides without the extra step of removing hard
templating silica structures has yet to be developed.
A large number of reports deals with the synthesis of
different modifications of bulk CxNy materials.[2] The synthesis
of these nitrogen-rich species generally includes thermal
condensation of nitrogen-rich precursors, often from mole-
cules containing or generating triazine rings. For example,
through a solid-state reaction of 2,4,6-triamino-1,3,5-triazine
with 2,4,6-trichloro-1,3,5-triazine at high pressure and high
temperature, Wolf and co-workers obtained a well-charac-
terized and highly crystalline graphitic carbon nitride deriv-
ative.[3] However, it was shown that the more ideal bulk
carbon nitride solids perform rather weakly in some catalytic
processes, while more disordered, polymeric versions showed
nice activity, as structural defects or surface terminations
seemed to play a key role for the catalytic activation.[4] To
enhance the performance of carbon nitride both as a support
and as a catalyst, the specific surface had to be enhanced, and
nanocasting with hard templates using porous silicas has been
explored recently for the replication of porous carbon nitride
materials with controlled mesopore structures.[5]
Ionic liquids are an especially promising choice to achieve
this aim. Ionic liquids are generally defined as organic salts
with a low melting point, usually below 1008C. They inherit
many features of inorganic molten salts, such as excellent
chemical and thermal stability (in some cases in excess of
4008C) and negligibly small vapor pressure, with the conven-
ience of being liquid under ambient conditions. They have
found application in numerous fields, for example as solvents
or catalysts in organic synthesis, as electrolytes, or in chemical
separations.[6] Recently, the advantages of ionic liquids in
inorganic synthesis have been gradually realized and have
received more and more attention. For instance, they turned
out to be interesting solvents in the synthesis of nanoparticles
owing to their special solvent structure that derives from the
ion–ion interaction and extended hydrogen-bonding motifs.[7]
From the many different available ionic liquids, we chose a
À
À
class with BF4 counterions, as the anion might enter the C N
condensation scheme, while boron and fluorine doping could
improve the catalytic activity of the discussed systems and
contribute to the performance of such materials.
Herein we demonstrate that the simple, commercially
available room-temperature ionic liquid, 1-butyl-3-methyli-
midazolium tetrafluoroborate (BmimBF4) is a unique soft
template for the easy synthesis of boron- and fluorine-
enriched mesoporous polymeric carbon nitride, in which an
organic precursor, for example dicyandiamide (DCDA), self-
condensed into carbon nitride solids in the presence of
BmimBF4. We show that the resulting materials possess high
nitrogen, boron, and fluorine contents, high surface area, local
graphitic order, and an excellent photoconductivity under
visible light. We also demonstrate the catalytic properties of
these CNBF materials in the heterogeneous, metal-free
oxidation of cyclohexane. The CNBF materials show nice
performance in the oxidation of cyclohexane with good
conversion and excellent cyclohexanone selectivity.
[*] Dr. Y. Wang, Prof. Dr. X. Wang, Prof. Dr. M. Antonietti
Department of Colloid Chemistry
Max-Planck Institute of Colloids and Interfaces
Research Campus Golm, 14424 Potsdam (Germany)
Fax: (+49)331-567-9502
E-mail: yong.wang@mpikg.mpg.de
Dr. Y. Wang, Prof. Dr. H. Li
Department of Chemistry
Zhejiang University, Hangzhou 310027 (China)
E-mail: chemwy@zju.edu.cn
lihr@zju.edu.cn
J. Zhang, Prof. Dr. X. Wang
Research Institute of Photocatalysis
State Key Laboratory Breeding Base of Photocatalysis
Fuzhou University, Fuzhou 350002 (China)
The structures of these hybrid materials were investigated
by powder X-ray diffraction (XRD), nitrogen gas adsorption,
and transmission electron microscopy (TEM) measurements.
The XRD pattern, for example of CNBF-0.3 (r= 0.3, r
denotes the mass ratio of BmimBF4 to DCDA), is dominated
by the (002) interlayer-stacking peak of carbon nitrides,
[**] This work was supported by the ENERCHEM project of the MPI, by
the National Natural Science Foundation of China (No. 20806065),
and by the China Postdoctoral Special Science Foundation funded
project (No. 200902630).
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201000120.
3356
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 3356 –3359

Angewandte
Chemie
corresponding to an interlayer distance of d = 0.338 nm
(Figure 1). In our reaction conditions, the (002) peak is
shifted to slightly lower angles compared with bulk melon
(d = 0.326 nm), thus indicating a small expansion of the
enriched in nitrogen, even compared to ideal C3N4 (the C/N
ratios are around 0.65, the theoretical C/N ratio of bulk C3N4
material is 0.75, see Table S1 in the Supporting Information).
X-ray photoelectron spectroscopy (XPS) results indicate that
both boron and fluorine atoms have been incorporated into
the CN matrix. The boron contents are near 20 mol%, while
the fluorine content is near 4 mol% (Figure S2 in the
Supporting Information). Therefore, the ionic liquid template
not only influences textural features but also enters the
materialsꢀ condensation scheme. Most probably, the B atoms
À
enter C sites in polymeric C N structures (thus balancing
stoichiometry), with F saturating residual bonds. Note that
BN is also an extremely stable covalent material.
The residual boron and fluorine content is not to be seen
as an impurity; instead, it provides potential utility for these
porous CNBF materials in photocatalysis, extending melon as
a polymeric semiconductor.
UV/Vis spectra (Figure 3a) indicate that the optical band
gap and thereby the semiconductor properties of the CNBF
have indeed changed, with a significant extension of the
Figure 1. a) XRD patterns of CNBF-r (r is the mass ratio of BmimBF4
to DCDA, r=0.3, 0.5, or 0.7). b) Typical TEM image of CNBF-0.5.
interlayer. This finding suggests disturbance of the packing
and potential undulations of the single layers by doping units,
which also explains the increased width of the (002) peak. A
typical TEM image of CNBF-0.5 is shown in Figure 1b. The
TEM picture clearly reflects the presence of a well-devel-
oped, sponge-like mesopore structure within the CNBF-0.5
generated by self-assembly of the BmimBF4 template towards
regular, approximately 9.5 nm large micellar units. The whole
structure nicely resembles a nanoscopic morel mushroom.
The surface area of the CN materials could be controlled
by changing the ratio of the template to the DCDA precursor.
Some nitrogen adsorption isotherms of CNBF are shown in
Figure 2. CNBF-0.5 exhibits the largest specific Brunauer–
Figure 3. a) UV/Vis spectra and b) photoluminescence spectra under
420 nm excitation at 298 K for CNBF. c) Current–potential curves and
d) current–time curves under chopped illumination (l>420 nm) for
CNBF-0.5 (dashed line) and bulk melon (solid line) electrodes in 0.2m
aqueous Na2SO4.
Figure 2. N2 adsorption/desorption isotherms and corresponding Bar-
rett–-Joyner–-Halenda (BJH) pore-size distribution curve determined
from the sorption branch of the isotherm of CNBF-0.5. R=pore
radius, v=pore volume.
spectrum towards absorption in the visible range, which we
attribute to dopant states within the band gap. The PL spectra
(Figure 3b) under 420 nm excitation show broad and strong
photoluminescence, which speaks for an effective separation
of charges and weak thermal relaxation of the excited states.
The CNBF material is obviously still an effective semi-
conductor like the undoped parent material melon. The
luminescence is weakened with increasing specific surface
area, thus indicating improved charge localization on the
surface terminal sites, which is believed to promote the
catalytic redox functions.
Emmett–Teller (BET) surface area (444 m2 gÀ1) and a total
pore volume of 0.32 m2 gÀ1, while micropores are essentially
absent in these materials. The relation of the BET surface
area with the content of the template is, however, not at all
linear. First, it strongly increases with r, then it decreases
again, presumably owing to structural pore collapse (Table S1
in the Supporting Information).
The carbon and nitrogen stoichiometry determined by
elemental analysis shows that the products are highly
Figure 3c shows the photocurrent response of the CNBF
electrode in a 0.2m Na2SO4 aqueous solution. Visible light
Angew. Chem. Int. Ed. 2010, 49, 3356 –3359
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3357

Communications
illumination (l > 420 nm) of CNBF layers cast on indium tin
several cycles without losing activity, which is a prerequisite
for practical applications. Furthermore, this catalyst is
expected to give excellent results in other organic reactions,
as it has a large surface area and suitable pore volume with a
large number of boron functional groups on the surface,
which might act as strong Lewis acid sites, thus complement-
ing the basic nitrogen sites.
In conclusion, a new kind of boron- and fluorine-enriched
polymeric carbon nitride solid has been successfully synthe-
sized. Notably, the resulting materials show a technically
advantageous “morel-like” mesopore structure with narrow
pore size distribution and improved photoactivity under
visible-light illumination. Surprisingly, these materials exhibit
good catalytic performance for the selective oxidation of
cycloalkanes. The catalysts afford a remarkably high selec-
tivity towards the formation of cyclohexanone, are highly
stable, and could be recycled several times. Moreover, we
believe that the current ionic-liquid-based strategy to control
structure and composition can be extended to incorporate
other heteroatoms by changing the anion or cation of the
ionic liquids into C,N-based materials. This approach may
widen the materialsꢀ application range.
oxide (ITO) glass plates can generate a photocurrent which
has been remarkably improved as compared to melon. The
material, however, turns out to be electrochemically active
against water, which leads to a decreasing photocurrent with
time. As the reasons for this decrease can be manifold
(including photochemical water splitting), work focusing on
this aspect is ongoing.
To illustrate the usefulness of this organic semiconductor,
selective oxidation of cycloalkanes was tested. The selective
oxidation of cyclohexane is a very important industrial
process, for example in the production of adipic acid and
caprolactam, which in turn are the starting materials for the
synthesis of nylon-6 and nylon-66 polymers, respectively.[8]
The industrial process uses homogeneous metal catalysts and
dioxygen as oxidant at temperatures above 1508C.[9] Apart
from low conversion, this process also produces several by-
products, such as mono- and dicarboxylic acids, esters, and
other oxygenated materials. This low conversion and selec-
tivity, leaching of metal, and over-oxidation of products in the
existing catalytic systems reinforce the need for new systems
using reinforce the need for new catalytic systems, including
heterogeneous catalyst and clean oxidant.[10] Herein, we used
the metal-free CNBF materials as oxidation catalysts and
hydrogen peroxide as oxidant for the highly selective Experimental Section
Synthesis of CNBF: In a typical synthetic procedure, BmimBF4 (0.5 g)
oxidation of cyclohexane.
was dissolved in water (6 mL) and stirred for 5 min. Then dicyandia-
mide (1 g) was added into the aqueous solution and the mixture was
heated in an oil bath at 1008C until the water had completely
evaporated and a white solid had formed. The white solid was then
transferred into a crucible and slowly heated over 2 h to a temper-
ature of 3508C. The material was tempered at this temperature for
4 h, heated over 1 h from 3508C to 5508C, and tempered at this higher
temperature for another 4 h. Then, the sample was allowed to cool in
the oven to room temperature.
Catalytic tests and analytical procedures: The oxidation of
cyclohexane over CNBF samples was carried out in a Teflon-lined
stainless steel batch reactor (40 mL total volume) using acetonitrile as
the reaction solvent. Cyclohexane (0.8 mL), catalyst (0.05 g), 30%
hydrogen peroxide in water solution (0.51 mL), and acetonitrile
(8 mL) were introduced into the reactor. The reactor was heated to
1508C over 30 min with stirring, and the reactor was held at this
temperature for 4 h. The products were analyzed on a GC-MS
instrument (Agilent Technologies, GC6890N, MS5970).
The catalysts are active and afford remarkably high
selectivity towards the formation of cyclohexanone in a
short time. For example, using CNBF-0.3 as catalyst in
acetonitrile, the total conversion of cyclohexane was 7.8%
and the selectivity towards cyclohexanone was as high as 91%
after 4 h (Table 1). Interestingly, the CNBF-1.0 catalyst
Table 1: Conversion and selectivity of cyclohexane oxidation over CNBF
catalysts.[a]
Catalyst
Total
conv. [%]
Selectivity [%]
for one[b]
one/ol[b]
Selectivity [%]
for H2O2
[c]
g-C3N4
1.6
7.8
7.5
5.3
>99
91
89
–
10
43
33
36
CNBF-0.3
CNBF-0.5
CNBF-1.0
10.1
8.4
–
>99
[a] No products were detected without the carbon nitride catalyst.
Reaction conditions: cyclohexane 0.8 mL, H2O2 (30% in water solution)
0.51 mL, catalyst 50 mg, temperature 1508C, time 4 h. [b] cyclohexa-
none=one, cyclohexanol=ol. [c] Graphitic C3N4.
Received: January 8, 2010
Published online: March 25, 2010
Keywords: carbon nitride · ionic liquids · mesoporous materials ·
.
oxidation · polymers
afforded 100% selectivity towards the formation of cyclo-
hexanone, although the conversion of cyclohexane is lower
than observed for CNBF-0.3 and CNBF-0.5 catalysts. The
products of all the reactions were analyzed, but no over-
oxidation by-product was detected in the GC-MS analysis.
This result indicates that, in contrast to the previous metal
catalysts, the CNBF materials are highly selective oxidation
catalysts and do not produce any adipic acid or valeric acid
under these conditions.
[1] a) X. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J. M.
Carlsson, K. Domen, M. Antonietti, Nat. Mater. 2009, 8, 76 – 80;
b) W. L. Wang, X. D. Bai, K. H. Liu, Z. Xu, D. Golberg, Y.
Bando, E. G. Wang, J. Am. Chem. Soc. 2006, 128, 6530 – 6531;
c) Z. Li, G. D. Del Cul, W. Yan, C. Liang, S. Dai, J. Am. Chem.
Soc. 2004, 126, 12782 – 12783; d) F. Goettmann, A. Thomas, M.
Antonietti, Angew. Chem. 2007, 119, 2773 – 2776; Angew. Chem.
Int. Ed. 2007, 46, 2717 – 2720; e) C. Niu, Y. Z. Lu, C. M. Lieber,
Science 1993, 261, 334 – 337.
Furthermore, the catalysts can easily be separated from
the reaction solution by simple filtration. The samples are
highly stable (in the absence of light) and can be reused for
[2] a) D. R. Miller, J. Wang, E. G. Gillan, J. Mater. Chem. 2002, 12,
2463 – 2469; b) Y. Qiu, L. Gao, Chem. Commun. 2003, 2378 –
3358
www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 3356 –3359

Angewandte
Chemie
2379; c) B. Jꢁrgens, E. Irran, J. Senker, P. Kroll, H. Muller, W.
Schnick, J. Am. Chem. Soc. 2003, 125, 10288 – 10300; d) J. R.
Holst, E. G. Gillan, J. Am. Chem. Soc. 2008, 130, 7373 – 7379;
e) M. J. Bojdys, J. Muller, M. Antonietti, A. Thomas, Chem. Eur.
J. 2008, 14, 8177 – 8182.
[7] a) M. Antonietti, D. Kuang, B. Smarsly, Y. Zhou, Angew. Chem.
2004, 116, 5096 – 5100; Angew. Chem. Int. Ed. 2004, 43, 4988 –
4992; b) Z. Ma, J. Yu, S. Dai, Adv. Mater. 2010, 22, 261 – 285;
c) B. Lee, H. Lou, J. Lin, S. Dai, Chem. Commun. 2004, 240 – 241;
d) T. Torimoto, T. Tsuda, K. Okazaki, S. Kuwabata, Adv. Mater.
2009, DOI: 10.1002/anie.200902184; e) J. S. Lee, X. Wang, H.
Luo, S. Dai, Adv. Mater. 2010, 22, 1004 – 1007.
[3] Z. Zhang, K. Leinenweber, M. Bauer, L. A. J. Garvie, P. F.
McMillan, G. H. Wolf, J. Am. Chem. Soc. 2001, 123, 7788 – 7796.
[4] X. Wang, K. Maeda, X. Chen, K. Takanabe, K. Domen, Y. Hou,
X. Fu, M. Antonietti, J. Am. Chem. Soc. 2009, 131, 1680 – 1681.
[5] a) X. Jin, V. V. Balasubramanian, S. T. Selvan, D. P. Sawant,
M. A. Chari, G. Q. Lu, A. Vinu, Angew. Chem. 2009, 121, 8024 –
8027; Angew. Chem. Int. Ed. 2009, 48, 7884 – 7887; b) F.
Goettmann, A. Fischer, M. Antonietti, A. Thomas, Angew.
Chem. 2006, 118, 4579 – 4583; Angew. Chem. Int. Ed. 2006, 45,
4467 – 4471; c) A. Vinu, K. Ariga, T. Mori, T. Nakanishi, S.
Hishita, D. Golberg, Y. Bando, Adv. Mater. 2005, 17, 1648 – 1652;
d) M. Groenewolt, M. Antonietti, Adv. Mater. 2005, 17, 1789 –
1792; e) X. Chen, Y. Jun, K. Takanabe, K. Maeda, K. Domen, X.
Fu, M. Antonietti, X. Wang, Chem. Mater. 2009, 21, 4093 – 4095.
[6] a) J. C. Plaquevent, J. Levillain, F. Guillen, C. Malhiac, A. C.
Gaumont, Chem. Rev. 2008, 108, 5035 – 5060; b) R. D. Rogers,
K. R. Seddon, Science 2003, 302, 792 – 793; c) Y. Wang, H. Li, C.
Wang, Chem. Commun. 2004, 1938 – 1939; d) Y. Wang, C. Wang,
L. Zhang, H. Li, Phys. Chem. Chem. Phys. 2008, 10, 5976 – 5982;
e) J. M. Lu, F. Yan, J. Texter, Prog. Polym. Sci. 2009, 34, 431 –
448.
[8] a) R. Kumar, S. Sithambaram, S. L. Suib, J. Catal. 2009, 262, 304 –
313; b) L. Liu, Y. Li, H. Wei, M. Dong, J. Wang, A. M. Z. Slawin,
J. Li, J. Dong, R. E. Morris, Angew. Chem. 2009, 121, 2240 – 2243;
Angew. Chem. Int. Ed. 2009, 48, 2206 – 2209; c) M. J. L. Kishore,
A. Kumar, J. AICHE 2007, 53, 1550 – 1561.
[9] a) T. F. S. Silva, G. S. Mishra, M. F. G. da Silva, R. Wanke,
L. M. D. R. S. Martins, A. J. L. Pombeiro, Dalton Trans. 2009,
9207 – 9215; b) J. Hao, J. Wang, Q. Wang, Y. Yu, S. Cai, F. Zhao,
Appl. Catal. A 2009, 368, 29 – 34.
[10] a) U. Schuchardt, D. Cardoso, R. Sercheli, R. Pereira, R. S.
da Cruz, M. C. Guerreiro, D. Mandelli, E. V. Spinace, E. L. Pires,
Appl. Catal. A 2001, 211, 1 – 17; b) C. Di Nicola, Y. Y. Karabach,
A. M. Kirillov, M. Monari, L. Pandolfo, C. Pettinari, A. J. L.
Pombeiro, Inorg. Chem. 2007, 46, 221 – 230; c) A. S. Goldstein,
R. H. Beer, R. S. Draggo, J. Am. Chem. Soc. 1994, 116, 2424 –
2429; d) E. Armengol, A. Corma, V. Formes, H. Garcia, J. Primo,
Appl. Catal. A 1999, 181, 305 – 312.
Angew. Chem. Int. Ed. 2010, 49, 3356 –3359
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3359