CO2 as a C1-building block for the catalytic methylation of amines

Article abstract of DOI:10.1039/c3sc22240c

A novel catalytic reaction has been designed to utilize, for the first time, CO₂ as a C₁ feedstock in the synthesis of N-methylamines. Simple zinc catalysts, based on commercially available zinc salts and ligands, prove highly efficient in promoting both a 6 electron reduction of carbon dioxide and the formation of a C-N bond, using hydrosilanes and amines.

Full text of DOI:10.1039/c3sc22240c

Chemical Science
View Article Online
View Journal | View Issue
EDGE ARTICLE
CO2 as a C1-building block for the catalytic methylation
of amines†
Cite this: Chem. Sci., 2013, 4, 2127
*
Olivier Jacquet, Xavier Frogneux, Christophe Das Neves Gomes and Thibault Cantat
Received 15th December 2012
Accepted 24th February 2013
A novel catalytic reaction has been designed to utilize, for the first time, CO2 as a C1 feedstock in the
synthesis of N-methylamines. Simple zinc catalysts, based on commercially available zinc salts and
ligands, prove highly efficient in promoting both a 6 electron reduction of carbon dioxide and the
formation of a C–N bond, using hydrosilanes and amines.
DOI: 10.1039/c3sc22240c
www.rsc.org/chemicalscience
group is a common functional group in organic compounds,
present in drugs, natural products, nitrogen bases, dyes, etc.
Introduction
Carbon dioxide is an economical, abundant and renewable
carbon source and therefore an advantageous C1 building block
for the production of ne chemicals.1 Yet the scope of organic
molecules directly available from CO2 remains very narrow and
only a handful of industrial applications exists, such as the
production of urea (Bosch–Meiser process) and the syntheses of
cyclic and polymeric carbonates and salicylic acid.1–4 More
recently, pilot plant production of methanol from CO2, H2 and
CO mixtures has been initiated in Japan.5 In fact, multi-electron
chemical reduction of CO2 is undoubtedly the most challenging
difficulty facing CO2 recycling, because highly active and inex-
pensive catalysts are required to activate the stable CO2 mole-
cule and promote its reduction using a mild reductant (e.g. H2,
hydrosilanes, boranes) or an electrochemically generated
hydride. As a consequence, 6-electron and 8-electron reduction
reactions of CO2 are limited to the formation of methanol and
methane, respectively.6 In order to expand the scope of chemical
functions available from CO2, novel catalytic reactions able to
promote the reduction and the functionalization (formation of
C–O, C–N or C–C bonds) of carbon dioxide at the same time are
highly desirable.3,4
Methylamines are basic reagents in nitrogen chemistry.
These include for example methylamine, dimethylamine, tri-
methylamine and dimethylformamide (DMF), which are used
as platform chemicals for the synthesis of fertilizers, fungicides,
synthetic leathers or polymers or directly utilized as solvents,
formulation agents, etc.7 Currently, these important
compounds are produced in industry from ammonia and
methanol and the world production capacity is estimated at
600 000 tons per year worldwide.7 More generally, the N–CH3
Because methylamines feature a C1 group, their synthesis by
CO2 transformation would be an attractive alternative to the
present petrochemical production. Yet, to the best of our
knowledge, the direct synthesis of N-methylamines using CO2
as a building block has never been reported, be it under
heterogeneous or homogeneous catalysis. To circumvent this
limitation, we describe herein an unprecedented catalytic
methylation reaction of amines to N-methylamines, using CO2
and hydrosilanes as reductants. This transformation is the rst
example of a reaction in which a 6-electron reduction of CO2 is
coupled to a C–N bond formation.
Results and discussion
Hydrosilanes are mild reductants with a redox potential well
poised for CO2 reduction and they feature a slightly polar Si–H
bond that can be activated using either organometallic or
organic catalysts.2,3b,4 As such, hydrosilanes are potent candi-
dates to promote the methylation of amines using CO2. Among
secondary amines, N-methylaniline (1a) was selected as a
benchmark substrate for the methylation of N–H bonds,
because it combines a low steric bulk at the nitrogen center with
balanced basic and nucleophilic properties. Recent research
efforts have demonstrated the potential of copper(II), iron(II)
and zinc(II) complexes as efficient catalysts to promote the
hydrosilylation reaction of a large scope of polar multiple
bonds.8 Signicant examples include the hydrosilylation of
esters to alcohols catalyzed by iron(II) complexes or the zinc
catalyzed reduction of amides to amines. In the presence of a
catalytic amount of copper(II) or iron(II) salts (5 mol%), addition
of 2 equiv. PhSiH3 to a THF solution of N-methylaniline (1a),
under an atmosphere of CO2, led to no reaction and the starting
materials were recovered unreacted aer 20 h at 100 ꢀC
(Table S1 in the ESI†). To our delight, simple zinc(II) chloride
was found to promote the methylation of 1a to N,N-dimethyla-
niline (2a) in a low but encouraging 10% yield under the same
reaction conditions, therefore demonstrating the feasibility of
CEA, IRAMIS, SIS2M, CNRS UMR 3299, 91191 Gif-sur-Yvette, France. E-mail: thibault.
cantat@cea.fr; Web: http://iramis.cea.fr/Pisp/thibault.cantat/index.htm; Fax: +33 1
6908 6640
† Electronic supplementary information (ESI) available: Supplementary catalytic
results, kinetic prole of the conversion of 1a to 2a and 3a and experimental
details. See DOI: 10.1039/c3sc22240c
This journal is ª The Royal Society of Chemistry 2013
Chem. Sci., 2013, 4, 2127–2131 | 2127

View Article Online
Chemical Science
Edge Article
the catalytic reaction depicted in eqn (1) (entry 2, Table 1). eqn (1), simple phosphines (PPh3, dppe) proved somewhat
While zinc acetate, zinc acetyl acetonate or zinc sulfate did not benecial in enhancing the catalytic activity of ZnCl2 and
yield any detectable amount of 2a, commercially available conversion yields of up to 33% 2a were obtained from 1a
diethyl zinc afforded 2a in a modest 30% yield aer 20 h at (entries 4–5 in Tables 1 and S1†). More importantly, the most
ꢀ
100 C (entry 3 in Tables 1 and S1†). Results in entries 2 and 3 active catalytic systems were obtained by resorting to the better
(Table 1) were reproduced using 13CO2 and N-methyl,N-[13C]- s-donor N-heterocyclic carbenes (NHCs) (entries 6–11, Table 1).
methylaniline was observed as the only N,N-dimethylaniline Using an equimolar mixture of ZnCl2 (5 mol%) and IPr (5 mol
derivative, conrming that CO2 is the carbon source of the extra %), the methylation of 1a to 2a was successfully achieved in an
methyl group in the synthesis of 2a. These preliminary results excellent 95% yield, within 20 h at 100 ꢀC, using 1 bar CO2 and
establish that bare zinc(II) salts are efficient catalysts for the phenylsilane (2 equiv.) as a reductant. Importantly, a nearly
novel methylation reaction of N–H bonds using CO2 as a carbon quantitative yield (95%) was also maintained in the presence of
source.
a low catalyst charge of 2 mol% of the isolated IPrZnCl2 complex
(entry 11, Table 1). Among the NHCs tested, the more sterically
congested IPr led to a somewhat more reactive catalytic system
(1) than the IMes and ItBu analogues (entries 6–8, Table 1). From
an electronic standpoint, the saturated NHCs s-IPr and s-IMes
exhibit lower activities than their unsaturated counterparts IPr
and IMes, respectively (entries 9–10, Table 1). Using IPr as an
ancillary ligand, Zn(OTf)2, Zn(OAc)2, Zn(acac)2$xH2O, Zn(SO4)$
7H2O or ZnEt2 all showed a signicant catalytic activity in the
methylation of 1a with conversion yields to 2a ranging from
12 to 47% (entries 12–16, Table 1). At this stage, IPrZnCl2 stands
out as the best performing catalyst and was therefore selected to
explore the scope of active substrates in this novel catalytic
reaction.
In order to increase the catalytic activity of zinc(II) salts and
A change of the hydrosilane conrms that PhSiH3 is more
therefore enable the methylation reaction of a wide spectrum of reactive than Ph2SiH2, Ph3SiH, Et3SiH and (EtO)3SiH in the
amines, external neutral donors were screened as supporting methylation of 1a. In fact, conversion of 1a to 2a was obtained in
ligands for ZnII complexes (Table 1). While nitrogen donors 30% yield in the presence of 6 equiv. (EtO)3SiH and 1 bar CO2,
such as pyridine, 1,10-phenanthroline or TMEDA were found to using IPrZnCl2 (5 mol%), whereas the conversion is 20% using 3
deactivate zinc chloride in the benchmark reaction depicted in equiv. Ph2SiH2, under the same conditions (entries 17 and 20,
Table 1 Zinc-catalyzed methylation of N-methylaniline (1a) to N,N-dimethylaniline (2a), as depicted in eqn (1)
Entry
Catalyst (mol%)
Ligand (mol%)
Solvent
R3SiH [n]
Yielda [%]
1
2
3
4
5
6
7
8
—
—
—
—
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
1,4-Dioxane
Toluene
CH2Cl2
PhSiH3 (2)
PhSiH3 (2)
PhSiH3 (2)
PhSiH3 (2)
PhSiH3 (2)
PhSiH3 (2)
PhSiH3 (2)
PhSiH3 (2)
PhSiH3 (2)
PhSiH3 (2)
PhSiH3 (2)
PhSiH3 (2)
PhSiH3 (2)
PhSiH3 (2)
PhSiH3 (2)
PhSiH3 (2)
Ph2SiH2 (3)
Ph3SiH (6)
Et3SiH (6)
(EtO)3SiH (6)
PhSiH3 (2)
PhSiH3 (2)
PhSiH3 (2)
<1
10
30
33
15
95
83
85
45
25
95
47
29
16
12
33
20
<1
<1
30
51
38
<1
ZnCl2 (5)
ZnEt2 (5)
ZnCl2 (5)
ZnCl2 (5)
ZnCl2 (5)
ZnCl2 (5)
ZnCl2 (5)
ZnCl2 (5)
PPh3 (10)
dppe (5)
IPr (5)
IMes (5)
ItBu (5)
s-IPr (5)
s-IMes (5)
—
IPr (5)
IPr (5)
IPr (5)
IPr (5)
IPr (5)
—
—
—
—
—
—
—
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
ZnCl2 (5)
IPrZnCl2 (2)
Zn(OTf)2 (5)
Zn(OAc)2 (5)
Zn(acac)2$xH2O (5)
Zn(SO4)$7H2O (5)
ZnEt2 (5)
IPrZnCl2 (5)
IPrZnCl2 (5)
IPrZnCl2 (5)
IPrZnCl2 (5)
IPrZnCl2 (5)
IPrZnCl2 (5)
IPrZnCl2 (5)
a
Determined by GC/MS.
2128 | Chem. Sci., 2013, 4, 2127–2131
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Edge Article
Chemical Science
Table 1). In contrast, Ph3SiH and Et3SiH were found to be
unreactive in the applied reaction conditions (entries 18–19,
Table 1). It is noteworthy that the solvent has a noticeable
inuence on the conversion and slightly higher yields are
observed for polar solvents such as THF (entries 6 and 21–23 in
Table 1). In contrast, dichloromethane solvent is incompatible
with the reaction conditions (entry 23, Table 1).
The catalytic methylation of various N-methylaniline deriv-
atives (1) was investigated using 2 equiv. PhSiH3 and 1 bar CO2,
in the presence of 5 mol% IPrZnCl2. As outlined in the results
gathered in Table 2, N-methylanilines (1) substituted by either
electron donating or electron withdrawing groups are active
substrates in the methylation reaction and the corresponding
N,N-dimethylanilines (2) were obtained in yields ranging from
10 to 95% (entries 1–10, Table 2). Yet, the substitution scheme
of the aryl ring in 1 has a major electronic inuence on the
reactivity of the amine. The unsubstituted derivate 1a is the
most reactive substrate in the series with a conversion yield to
2a of 95%, aer 20 h at 100 ꢀC. Introduction of electron
donating groups (EDGs) of increasing strength progressively
deactivates the catalytic activity of the amine for which the
conversion to 2 drops to 77%, 47% and 11% for 1d (Xm ¼ Me),
1c (Xp ¼ Me) and 1b (Xp ¼ OMe), respectively; under the same
conditions (entries 1–3, Table 2). Electron withdrawing groups
(EWGs) have a similar impact on the reactivity of 1 and 1i
(Xm ¼ Cl) affords the methylated product 2i in a low 10% yield
Scheme 1 Hammett plot for the catalytic conversion of 1 to 2 at 100 ꢀC, with
10 equiv. PhSiH3 and 1 bar CO2, using IPrZnCl2 (5 mol%) as catalyst.
(entries 5–9, Table 2). These results indicate that substituents
on the aryl ring markedly affect the methylation rates. In fact,
measurements of reaction rates for the early stage of the cata-
lytic reaction (<30% conversion) reveal that their effect does not
follow a direct Hammett correlation, as outlined in the plot of
log(kobs[2]/kobs[2a]) vs. the Hammett parameter s (Scheme 1).
The Hammett plot shows a distinct concave breakdown break
with a maximum observed for 2a. This behavior is diagnostic of
a change in the rate-determining step of an otherwise constant
mechanism.9
From a mechanistic perspective, two main pathways can be
proposed for the methylation of N–H bonds using CO2 and
hydrosilanes (paths A and B in Scheme 2). Reduction of CO2 to
formoxysilane or methoxysilane species by hydrosilylation has
been recently reported using either organic or organometallic
catalysts and can be reasonably proposed as the rst step of
the methylation reaction (path A).2 Both formoxysilane and
methoxysilane are electrophiles known to react with amines to
afford formamides and methylamines, respectively.10 As an
alternative, the catalytic formylation of the N–H bond using
CO2 and the hydrosilane may precede the catalytic hydro-
silylation of the intermediate formamide to the methylamine
product (path B).4,8c Experimentally, CO2 was found to be stable
in the presence of PhSiH3 and 5 mol% IPrZnCl2 and unreacted
(2)
Table 2 Zinc-catalyzed methylation of substituted N-methylanilines (1) to N,N-
dimethylanilines (2), as depicted in eqn (2)
Product
distributionb
ꢀ
Entry
Xp
Xm
Xo
sa
3 [%]
2 [%]
PhSiH3 was recovered as the only product aer 20 h at 100 C,
therefore ruling out the occurrence of path A as a productive
1
2
3
4
5
6
7
8
9
OMe
Me
H
H
F
H
Cl
H
H
H
H
H
Me
H
H
OMe
H
F
Cl
H
H
H
H
H
H
H
H
H
ꢁ0.27
ꢁ0.17
ꢁ0.07
0
+0.06
+0.11
+0.23
+0.34
+0.37
—
3b, <1
3c, 5
3d, 3
3a, <1
3e, 2
3f, 15
3g, <1
3h, 3
3i, 11
3j, 44
2b, 11
2c, 47
2d, 77
2ac, 95
2e, 55
2f, 76
2g, 62
2h, 30
2i, 10
2j, 11
H
OMe
10
a
b
Hammett constants.9
Product distribution of the nitrogen
containing products determined by GC/MS and 1H NMR analyses of
the crude mixture. Unreacted amine substrate (1) is the sole
additional nitrogen compound when the product distributions do not
c
sum up to 100%. Average isolated yield for 2a, over three
experiments: 81%.
Scheme 2 Proposed pathways for the zinc-catalyzed methylation of amines.
Chem. Sci., 2013, 4, 2127–2131 | 2129
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Chemical Science
Edge Article
pathway in the catalytic methylation of amines (eqn (3)).
Nonetheless, formamides 3 were successfully detected in low
yields (up to 15%) in the synthesis of 2 (entries 1–9 in Table 2)
and introducing a methoxy EDG in the ortho position in the
sterically congested 1j reagent favoured the accumulation of
formamide 3j (44%) over 2j (11%) (entry 10 in Table 2). The
intermediate nature of formamide 3 is conrmed by moni-
toring the product distribution 3a/2a as a function of time
upon methylation of 1a (Fig. S1 in the ESI†). In addition,
IPrZnCl2 is an active catalyst in the hydrosilylation of 3a and 2a
is obtained in quantitative yield (>99%) aer 20 h at 100 ꢀC,
using PhSiH3 as a reductant (eqn (4)). All together, these data
suggest that path B in Scheme 2 is a robust mechanistic model
for the methylation reaction, although the elementary steps
involved in the formylation reaction and the reduction of the
formamide intermediate are still to be elucidated. It is also
consistent with the Hammett plot depicted in Scheme 1, as the
two steps likely follow opposite electronic demands: while the
formylation of the amine is kinetically favoured in the pres-
ence of EDGs (negative s values), the reduction of the inter-
mediate formamide is expectedly facilitated in the presence of
EWGs (positive s values) able to enhance the electrophilicity of
the formamide.
(5)
(6)
(7)
(3)
Table 3 Zinc-catalyzed methylation of substituted aniline derivatives (8), as
depicted in eqn (7)
Product distributiona
Entry
RNH2
9b [%]
10b [%]
11b [%]
(4)
1
2
3
C7H15–NH2
PhH2C–NH2
Ph–NH2
9a, <1
9b, 20
9c, 62
10a, 26
10b, 40
10c, 5
11a, <1
11b, <1
11c, 1
The methylation of N–H bonds in a variety of amines was
then undertaken to explore the scope of active substrates. As
exemplied in eqn (5), the catalytic methylation of amines using
CO2 and PhSiH3 is not limited to N-methylaniline derivatives
and secondary aliphatic amines are converted to their methyl-
ated form in good yields (up to 65%). For example, morpholine
is transformed to 6a in 65% yield and the sterically congested
diisopropylamine remains active and is converted to 6d in 46%
yield. In agreement with the mechanistic considerations,
formamides are more difficult to reduce in the presence of
aliphatic substituents on the nitrogen atom and formamides 5a
and 5d are formed in 35% and 17% yield from 4a and 4d,
respectively. Interestingly, N-methyldiphenylamine 6f is
obtained in a low 8% yield, presumably because of the low
reactivity of 4f in the formylation step. Importantly, trimethyl-
amine (TMA) is synthesized in high yield (84%) when dime-
4
5
6
7
9d, 67
9e, 49
9f, 67
9g, 49
10d, 2
10e, 2
10f, 2
10g, 1
11d, <1
11e, 5
11f, <1
11g, 2
8
9
9h, 12
9i, 10
10h, 1
10i, 2
11h, 7
11i, 5
10
Ph2N–NH2
Ph–NH2
9j, 24
10j, <1
10c, 21
11j, 4
11c
9c, <1c
11cd, 79
a
Product distribution of the nitrogen containing products determined
by GC/MS and 1H NMR analyses of the crude mixture. Unreacted amine
substrate (1) is the sole additional nitrogen compound when the
thylammonium dimethylcarbamate (7) is employed as
a
substrate (eqn (6)). In the applied reaction conditions, DMF is
obtained in 16% yield as the side-product. Yet, DMF formation
is not problematic and longer reaction times allow a complete
conversion of 7 to TMA (>99% yield aer 72 h at 100 ꢀC using 3
equiv. PhSiH3).
b
product distributions do not sum up to 100%. Compounds 9d (1b),
10d (3b), 9c (1a), 10c (3a), 11c (2a), 11d (2b), 9f (1e), 10f (3e), 11f (2e),
9g (1g), 10g (3g) and 11g (2g), present in Table 2, have been re-
labelled in Table 3 for clarity. Reaction time: 72 h. Average isolated
yield for 11c, over three experiments: 61%.
c
d
2130 | Chem. Sci., 2013, 4, 2127–2131
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Edge Article
Chemical Science
The reductive functionalization of CO2 is also efficient in the
methylation of N–H bonds in primary amines and reacting
aniline (8c) with 2 equiv. PhSiH3 and 1 bar CO2 affords
N-methylaniline in 62% yield aer 20 h at 100 ꢀC, using
IPrZnCl2 as a catalyst (5 mol%) (entry 3 in Table 3). As outlined
hereinabove, N-methylaniline is a reactive substrate in the
methylation reaction and small amounts of its formyl and
methyl derivatives were observed in the reaction mixture (in 5%
and 1% yield, respectively). In fact, using heptylamine (8a) as an
aliphatic primary amine resulted in the selective production of
the N-formyl,N-methyl species 10a in 26% yield (entry 1, Table
3). As similar behavior is observed with benzylamine (8b), which
is transformed into a mixture of 9b (20%) and 10b (40%), under
identical reaction conditions (entry 2, Table 3). Importantly,
substituted anilines 8d–8i are converted to N-methylanilines
9d–9i in up to 67% yield with a good selectivity, the formyl (10d–
10i) and bis-methyl (11d–11i) derivatives being produced in less
than 8% yield (entries 4–9 in Table 3). The catalytic methylation
of aniline is sensitive to the steric congestion around the
nitrogen atom and 2,6-diisopropylaniline (8i) exhibits a low
conversion yield of 10% to 9i (entry 9 in Table 1). Interestingly,
CO2 can also be utilized as a C1 feedstock to achieve the
methylation of substituted hydrazines and IPrZnCl2 is able to
promote the formation of N,N-diphenyl-N0-methylhydrazine (9j)
in 24% from 8j and PhSiH3. Noticeably, methylation of both
N–H bonds of primary amines can be selectively promoted, as
exemplied by the direct formation of 11c in 79% yield from the
Chem. Soc., 2012, 134, 17462; (c) A. Berkefeld, W. E. Piers
and M. Parvez, J. Am. Chem. Soc., 2010, 132, 10660; (d)
R. Lalrempuia, M. Iglesias, V. Polo, P. J. Sanz Miguel,
´
´
F. J. Fernandez-Alvarez, J. J. Perez-Torrente and
L. A. Oro, Angew. Chem., Int. Ed., 2012, 51, 12824; (e)
K. Motokura, D. Kashiwame, A. Miyaji and T. Baba, Org.
Lett., 2012, 14, 2642; (f) T. Matsuo and H. Kawaguchi, J.
´
Am. Chem. Soc., 2006, 128, 12362; (g) S. Park, D. Bezier
and M. Brookhart, J. Am. Chem. Soc., 2012, 134, 11404;
(h) T. C. Eisenschmid and R. Eisenberg, Organometallics,
1989, 8, 1822.
3 (a) I. I. F. Boogaerts and S. P. Nolan, J. Am. Chem. Soc., 2010,
132, 8858; (b) M. Khandelwal and R. J. Wehmschulte, Angew.
Chem., Int. Ed., 2012, 51, 7323; (c) C. Federsel, A. Boddien,
R. Jackstell, R. Jennerjahn, P. J. Dyson, R. Scopelliti,
G. Laurenczy and M. Beller, Angew. Chem., Int. Ed., 2010,
49, 9777; (d) M. D. Greenhalgh and S. P. Thomas, J. Am.
Chem. Soc., 2012, 134, 11900.
4 (a) C. Das Neves Gomes, O. Jacquet, C. Villiers, P. Thuery,
M. Ephritikhine and T. Cantat, Angew. Chem., Int. Ed.,
2012, 51, 187; (b) O. Jacquet, C. Das Neves Gomes,
M. Ephritikhine and T. Cantat, J. Am. Chem. Soc., 2012,
134, 2934; (c) O. Jacquet, C. Das Neves Gomes,
M. Ephritikhine and T. Cantat, ChemCatChem, 2013,
5, 117.
5 (a) G. A. Olah, A. Goeppert and G. K. S. Prakash, J. Org. Chem.,
2009, 74, 487; (b) M. Saito, Catal. Surv. Jpn., 1998, 2, 175.
6 (a) E. E. Benson, C. P. Kubiak, A. J. Sathrum and J. M. Smieja,
Chem. Soc. Rev., 2009, 38, 89; (b) W. Wang, S. P. Wang,
X. B. Ma and J. L. Gong, Chem. Soc. Rev., 2011, 40, 3703; (c)
P. G. Jessop, F. Joo and C. C. Tai, Coord. Chem. Rev., 2004,
248, 2425; (d) A. A. Peterson and J. K. Norskov, J. Phys.
Chem. Lett., 2012, 3, 251; (e) C. Oloman and H. Li,
ChemSusChem, 2008, 1, 385; (f) S. Wesselbaum, T. vom
Stein, J. Klankermayer and W. Leitner, Angew. Chem., Int.
Ed., 2012, 51, 7499.
7 (a) M. F. Ali, B. M. El Ali and J. G. Speight, Handbook of
Industrial Chemistry – Organic Chemicals, McGraw-Hill,
2005; (b) H.-J. Arpe and S. Hawkins, Industrial Organic
Chemistry, Wiley-VCH, 1997.
8 (a) S. Enthaler, K. Junge and M. Beller, Angew. Chem., Int. Ed.,
2008, 47, 3317; (b) L. C. M. Castro, H. Q. Li, J. B. Sortais and
C. Darcel, Chem. Commun., 2012, 48, 10514; (c) S. Das,
D. Addis, S. L. Zhou, K. Junge and M. Beller, J. Am. Chem.
Soc., 2010, 132, 4971; (d) D. H. Appella, Y. Moritani,
R. Shintani, E. M. Ferreira and S. L. Buchwald, J. Am.
Chem. Soc., 1999, 121, 9473; (e) J. Yun, D. Kim and H. Yun,
Chem. Commun., 2005, 5181.
ꢀ
primary amine 8c, aer 72 h at 100 C (entry 11 in Table 3).
Conclusions
In conclusion, catalytic methylation of N–H bonds using CO2 as
a C1 building block has been achieved for the rst time, using
hydrosilanes as reductants and zinc complexes as catalysts. This
reaction is unique in CO2 transformation as it combines a 6-
electron reduction of CO2 with the formation of a C–N bond to
promote the complete deoxygenation of CO2 to N-methyl-
amines. It therefore opens up the scope of chemical functions
directly available from carbon dioxide.
Acknowledgements
For nancial support of this work, we acknowledge the CEA,
CNRS, ANR (starting grant to T.C.), the FP7 Eurotalents Program
(PD fellowship to O.J.), the University Paris-Sud (fellowship to
X.F.) and the CEA Direction of Materials Science (PD fellowship
to O.J., Basic Research on Low Carbon Energies grant to T.C.).
Notes and references
9 (a) C. Hansch, A. Leo and R. W. Ta, Chem. Rev., 1991, 91,
165; (b) J. E. Leffler and E. Grunwald, Rates and Equilibria
of Organic Reactions, Wiley, New York, 1963.
1 (a) M. Aresta, Carbon Dioxide as Chemical Feedstock, Wiley-
VCH, Verlag GmbH, 2010; (b) K. Huang, C. L. Sun and
Z. J. Shi, Chem. Soc. Rev., 2011, 40, 2435; (c) T. Sakakura, 10 (a) S. Itagaki, K. Yamaguchi and N. Mizuno, J. Mol. Catal. A:
J. C. Choi and H. Yasuda, Chem. Rev., 2007, 107, 2365.
2 (a) S. N. Riduan, Y. G. Zhang and J. Y. Ying, Angew. Chem.,
Int. Ed., 2009, 48, 3322; (b) W. Sattler and G. Parkin, J. Am.
Chem., 2013, 366, 347; (b) G. Sumrell and G. E. Ham, J. Am.
Chem. Soc., 1956, 78, 5573; (c) A. A. Ponaras and
M. Y. Meah, Tetrahedron Lett., 1986, 27, 4953.
This journal is ª The Royal Society of Chemistry 2013
Chem. Sci., 2013, 4, 2127–2131 | 2131