Metal-free catalysis of sustainable Friedel-Crafts reactions: Direct activation of benzene by carbon nitrides to avoid the use of metal chlorides and halogenated compounds

Article abstract of DOI:10.1039/b608532f

The use of mesoporous graphitic C₃N₄ for the activation of benzene permitted to perform more sustainable Friedel-Crafts reactions by allowing to directly use carboxylic acids, alcohols and even quaternary ammoniums or urea as electrophiles. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b608532f

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Metal-free catalysis of sustainable Friedel–Crafts reactions: direct
activation of benzene by carbon nitrides to avoid the use of metal
chlorides and halogenated compounds{
Frederic Goettmann, Anna Fischer, Markus Antonietti and Arne Thomas*
Received (in Cambridge, UK) 16th June 2006, Accepted 2nd August 2006
First published as an Advance Article on the web 18th September 2006
DOI: 10.1039/b608532f
2
6
The use of mesoporous graphitic C N for the activation of
hydroformylation. This whole trend may lead to a renewed
interest for purely heterogeneous catalysts in fine chemistry. We
recently reported the use of a carbon nitride as a metal-free catalyst
3
4
benzene permitted to perform more sustainable Friedel–Crafts
reactions by allowing to directly use carboxylic acids, alcohols
and even quaternary ammoniums or urea as electrophiles.
20,21
for Friedel–Crafts acylations,
which is in principle strongly
related to the above mentioned approach. Among the various
possible carbonitride molar compositions, C N appeared to be a
Friedel–Crafts reactions are known to be some of the less
3
4
sustainable industrial processes, producing about 88 mass% of
1
promoted version. They usually
highly promising material, especially its graphitic allotrope. Kroke
et al. proposed (and it is now widely admitted) that this material
consists of sheets of highly ordered tri-s-triazine moieties connected
waste, in their standard AlCl
3
require a more than stochiometric amount of co-reactant (like
AlCl ) and rely on halogenated substrates or anhydrides as
22
3
through planarized amino groups as depicted in Fig. 1. The
electrophiles. Despite their low performances these reactions
remain widely used and studied because they are very powerful
synthesis of our catalytic material relied on thermal self-
27
condensation of cyanamide described by Schnick and coworkers.
2
synthetic tools. The search for true catalytic Friedel–Crafts
Mesoporosity was generated by the use of silica nanoparticles as
hard templates. After condensation of the monomers and removal
of the template, the resulting powders exhibited specific surface
reactions has thus attracted a lot of interest. An important step
towards greener Friedel–Crafts reactions was achieved by using
3,4
2
21
2
21
homogeneous Lewis acidic complexes as ‘‘true’’ catalysts. The
5
current focus on catalysts separation and recycling, however, has
areas ranging from 86 m g to 439 m g depending on the
amount of templating agent. Due to their remarkable electronic
properties these solids act as effective Friedel–Crafts catalysts for
the acylation of benzene with acyl chlorides. Although the actual
active centres (which, for symmetry reasons, could be isolated tri-
s-triazine units, the combination of the lone pairs of the nitrogens
pointing towards the structural holes, etc.) are not known yet, the
catalysis most probably relies on the activation of the aromatics
prompted many research groups to look for heterogeneous
6–8
9,10
catalytic systems. Ionic liquids and fluorinated solvents have
proved to be suitable green solvents and/or co-reactants for those
reactions. Nevertheless, the most promising catalytic systems for
Friedel–Crafts reactions are solids, from zeolites to Nafion-
1
1–14
functionalised MCM-41s.
electrophiles are rather atom ineffective, which has encouraged
Besides this issue, the usual
20,21
rather than the activation of the electrophile.
Therefore it was
15,16
the direct use of carboxylic acids for acylation reactions,
17–19
alcohols for alkylations.
or
assumed that this catalyst could also enable the use of less active
and incidentally more sustainable electrophiles.
In this contribution, we propose an
alternative route for using sustainable electrophiles, such as
alcohols, formic acid but also ammonium compounds or urea in
Friedel–Crafts type reactions relying on a Lewis basic solid,
namely mesoporous graphitic C N , mpg-C N , to activate
In this study, the previously described mpg-C N material was
3 4/1
used throughout the catalytic experiments. In a typical synthesis,
1
2.5 g of a 40 wt% water dispersion of 12 nm silica spheres (Ludox
3
4
3
4
HS40, Aldrich) were added dropwise to 1 g of melted cyanamide.
The mixture was subsequently heated to 550 uC in a closed ceramic
crucible for 4 h. The resulting yellow powder was sealed in a quartz
20,21
benzene.
A new trend for enhancing heterogeneous catalysis is to
maximise the interaction between the active site and the surface
2
3
of the support. This approach enables us to take advantage of
the intrinsic properties of the support for achieving multifunctional
24
25
catalysis or enhancing regioselectivity. It can also lead to the
discovery of unpredicted catalytic properties of the support itself,
as exemplified in the case of zirconium oxide catalyzed
Max-Planck-Institute of Colloids and Interfaces, Department of Colloid
Chemistry, Research Campus Golm, D-14424, Potsdam, Germany.
E-mail: arne.thomas@mpikg.mpg.de; Fax: (+49) 331-567-9502;
Tel: (+49) 331-567-9509
{
Electronic supplementary information (ESI) available: Wide angle X-ray
diffraction of mpg-C 4/1, small angle X-ray diffraction of mpg-C
and nitrogen adsorption isotherms of mpg-C 4/1. See DOI: 10.1039/
b608532f
3
N
3 4/1
N
3
N
22
3 4
Fig. 1 Structure proposed by Kroke et al. for g-C N .
4
530 | Chem. Commun., 2006, 4530–4532
This journal is ß The Royal Society of Chemistry 2006

View Article Online
ampoule under vacuum and submitted to an additional heat
treatment at 600 uC for 10 h. Chemical removal of the silica by
washing with a 4 M NH HF solution yielded a mesoporous light
in 5 ml of benzene with 200 mg of the desired electrophile
(methanol, formic acid, tetramethyl ammonium bromide and
urea). The closed autoclave was then heated to 150 uC for 24 h.
The products of the reaction were analysed and quantified by GC
with an internal standard and the products attribution was
4
2
2
21
brown powder (Fig. 2) exhibiting a surface area of 141 m g . The
pore size distribution was narrow with an average pore size of
1
12 nm. Small angle X-ray diffraction patterns featured a peak at
ascertained by H NMR. The results are summarized in Table 1.
0
.071 nm, corresponding to a correlation length of 14 nm, which
The conversion rates indicated in Table 1 were calculated on the
basis of the obtained amount of products compared with the initial
amount of limiting reactant.
was in good agreement with the pore size determined from BET
measurements. The wide angle X-ray diffraction patterns exhibited
the typical graphitic interlayer stacking (002) peak of g-C
3
N
4
,
Concerning alkylation reactions with alcohols as reagent,
isopropanol yielded 13% of cumene under relatively mild
conditions. The current industrial procedure relies on the
alkylation of benzene with propene. In that perspective it is
important to note that no other product was detected and that the
reaction is not entropically disfavoured, which will allow further
optimization and high temperature processes. The reaction with
ethanol yielded 16% (with respect to benzene) of p-diethyl benzene
after 24 h. In the case of methanol another situation arose:
depending on whether methanol is the solvent or not, it is possible
to tune the products towards mesitylene or p-xylene, with good
selectivities (respectively 100 and 80%). A reference test without
catalyst did not provide any detectable amount of alkylation
corresponding to an interlayer distance of d = 0.319 nm. An
intense additional peak is observed at 13.1u, corresponding to an
in-plane structural packing motif, namely the hole-to-hole distance
of the nitride pores (see Fig. 1). The elemental analysis of the
material provided an average atomic C/N ratio value of 0.71
(theoretical value for C N : 0.75). The additional small amounts of
3 4
hydrogen we found (1.5%) were attributed to uncondensed amino
functions, adsorbed water and structural defects on the surface.
No trace of iron or other transition metals could be detected by
Energy Dispersive X-ray analysis (EDX).
The substrates of the catalytic experiments were used without
further purification. Benzene was purchased from Aldrich (99.9%,
HPLC grade). As electrophiles, three alcohols (methanol 99.9%
HPLC grade from Merck, ethanol 99.9% from J. T. Backer, and
isopropanol, 99,5% spectrophotometer grade from Aldrich),
formic acid (88–91% from Fluka), tetramethylammonium bromide
product ascertaining that the observed activity of mpg-C
3
N4/1 is
not an artifact.
The reaction of formic acid with benzene produced benzalde-
hyde in quantitative yield. This result could be the basis of a very
valuable alternative to the usual benzaldehyde synthesis relying on
the oxidation of toluene. Without catalyst, formic acid totally
decomposed.
(98% from Aldrich) and urea (99% from Aldrich) were used. In a
typical experiment, 50 mg of mpg-C 4/1 were placed in a 40 ml
3
N
stainless steel autoclave fitted with a Teflon mantel (to avoid any
contamination by transition metals). As our previous work
showed that the solvent could dramatically affect the efficiency
of the catalysis two series of tests were undertaken. In the first one
the solution placed in the autoclave consisted in 5 ml of the
electrophile with 200 mg of benzene (this is the case for methanol,
ethanol and isopropanol). In the second one the solution consisted
The attempts with molecules bearing nitrogen atoms afforded
unpredicted results. Tetramethylammonium bromide yielded 100%
of toluene, corresponding to the transfer of one methyl group per
ammonium molecule. This reaction is obviously not a sustainable
way of alkylating benzene, but it supports our assumption that the
catalysis proceeds via the activation of the aromatic ring. Indeed,
Table 1 Friedel–Crafts type reactions of benzene with various
a
3 4
electrophiles catalysed by mpg-C N
Conversion
rate (%)
b
Electrophile
Methanol
Products
10
Toluene (20%)
p-Xylene (80%)
Mesitylene (100%)
—
p-Diethylbenzene (100%)
Cumene (100%)
Benzaldehyde (100%)
Decomposition products
of formic acid
c
Methanol
Methanol
Ethanol
20
,0.5
18
13
100
100
c,d
c
c
Isopropanol
Formic acid
Formic acid
d
c
Tetramethylammonium
bromide
Tetramethylammonium
100
Toluene (100%)
e
15
Toluene (100%)
d
bromide
Urea
Urea
20
,0.5
Benzonitrile
—
d
a
Reaction conditions: 50 mg of catalyst, 200 mg of the limiting
b
reactant, 5 ml of solvent, at 150 uC for 24 h. The conversion rate
was calculated as the molar ratio between the products and the
c
initial amount of limiting reactant. In those cases the alcohol was
d
e
used as the solvent. Reference tests made without catalyst. In this
case the conversion rate was calculated on the basis of one methyl
group exchanged per ammonium molecule.
3 4/1
Fig. 2 TEM micropraphs of mpg-C N .
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 4530–4532 | 4531

View Article Online
alkylation agents such as alcohols or acids, it also shows an
unpredicted reactivity towards urea and quaternary ammonium
compounds. Further experimental and theoretical data are needed
to provide a detailed mechanism for the observed activation of
benzene and the collecting work therefore is still in progress.
Notes and references
1
2
J. H. Clark, Green Chem., 1999, 1, 1.
M. Bandini, A. Melloni and A. Umani-Ronchi, Angew. Chem., Int. Ed.,
2
004, 43, 550.
H. Gaspard-Iloughmane and C. Le Roux, Eur. J. Org. Chem., 2004,
517.
G. K. S. Prakash, P. Yan, B. Torok, I. Bucsi, M. Tanaka and
G. A. Olah, Catal. Lett., 2003, 85, 1.
D. J. Cole-Hamilton and R. P. Tooze, Catalyst Separation, Recovery
and Recycling, Springer, Berlin, 2006.
H. Olivier-Bourbigou and L. Magna, J. Mol. Catal. A: Chem., 2002,
3
4
5
6
2
Fig. 3 Possible structure of a ureo- moiety anchored at the surface of
mpg-C 4/1 detected by FT-IR.
3
N
tetramethylammonium is not a Lewis base (and thus cannot be
activated in a standard Friedel–Crafts way) and a moderate
electrophile (the formal positive charge on nitrogen is in fact
distributed on the four methyl groups). A reference test without
catalyst yielded only 15% of toluene.
1
82, 419.
7 R. Sheldon, Chem. Commun., 2001, 2399.
8
9
1
T. Welton, Coord. Chem. Rev., 2004, 248, 2459.
T. Kitazume, J. Fluor. Chem., 2000, 105, 265.
0 A. G. M. Barrett, D. C. Braddock, D. Catterick, D. Chadwick,
Valorization of urea, which is an abundant and cheap starting
molecule, is interesting from an industrial point of view. In our
study, the reaction of benzene and urea yielded benzonitrile. This
reaction formally implies the addition of urea to benzene followed
by the elimination of both a water molecule and an ammonia
molecule. The precise reaction path could not be determined.
Benzamide, which might have been a possible reaction inter-
mediate, failed to react under similar conditions, indicating that
this compound does probably not play any role in the reaction
cycle. Interestingly, after a 20 h treatment with urea at 150 uC,
mpg-C N features three new peaks in FT-IR compared with the
J. P. Henschke and R. M. McKinnell, Synlett, 2000, 847.
11 M. Alvaro, A. Corma, D. Das, V. Fornes and H. Garcia, J. Catal.,
2
005, 231, 48.
2 A. Corma, Chem. Rev., 1997, 97, 2373.
3 G. Sartori and R. Maggi, Chem. Rev., 2006, 106, 1077.
14 A. Taguchi and F. Schuth, Microporous Mesoporous Mater., 2005, 77, 1.
5 M. J. Earle, U. Hakala, B. J. McAuley, M. Nieuwenhuyzen, A. Ramani
and K. R. Seddon, Chem. Commun., 2004, 1368.
1
1
1
16 M. Kawamura, D. M. Cui, T. Hayashi and S. Shimada, Tetrahedron
Lett., 2003, 44, 7715.
17 K. Mantri, K. Komura, Y. Kubota and Y. Sugi, J. Mol. Catal. A:
Chem., 2005, 236, 168.
1
8 G. D. Yadav and G. S. Pathre, Microporous Mesoporous Mater., 2006,
9, 16.
9 G. D. Yadav and G. S. Pathre, Appl. Catal., A, 2006, 297, 237.
3
4/1
8
21
21
21
non-urea treated solid, at 1691 cm , 1588 cm and 1423 cm
.
1
These peaks cannot be attributed to possible remaining urea; on
the contrary, they are consistent with the formation of ureo-
moieties on the surface of our catalyst. Indeed, Lotsch and Schnick
recently showed that primary and secondary amines were present
20 F. Goettmann, A. Fischer, M. Antonietti and A. Thomas, Angew.
Chem., Int. Ed., 2006, 45, 4467.
21 F. Goettmann, A. Fischer, M. Antonietti and A. Thomas, Angew.
Chem., 2006, 118, 4579.
22 E. Kroke, M. Schwarz, E. Horath-Bordon, P. Kroll, B. Noll and
A. D. Norman, New J. Chem., 2002, 26, 508.
28
at the surface of C
urea to form anchored urea species as depicted in Fig. 3. In this
case mpg-C N could be able to activate both the benzene and
3 4
N . The later surface species could react with
23 J. M. Notestein and A. Katz, Chem.–Eur. J., 2006, 12, 3954.
24 F. Goettmann, C. Boissiere, D. Grosso, F. Mercier, P. Le Floch and
3
4/1
C. Sanchez, Chem.–Eur. J., 2005, 11, 7416.
25 F. Goettmann, P. Le Floch and C. Sanchez, Chem. Commun., 2006,
urea.
2
036.
6 F. Goettmann, P. Le Floch and C. Sanchez, Chem. Commun., 2006, 180.
7 B. Jurgens, E. Irran, J. Senker, P. Kroll, H. Muller and W. Schnick,
In this contribution we have shown that mpg-C N4/1 is a
3
2
2
valuable Lewis base catalyst which enables rather unusual
aromatic substitution reactions of a generalized Friedel–Crafts
type. This metal-free catalyst not only allows the use of sustainable
J. Am. Chem. Soc., 2003, 125, 10288.
28 B. V. Lotsch and W. Schnick, Chem. Mater., 2006, 18, 1891.
4
532 | Chem. Commun., 2006, 4530–4532
This journal is ß The Royal Society of Chemistry 2006