Functionalization of CnH2n+2 Alkanes: Supercritical Carbon Dioxide Enhances the Reactivity towards Primary Carbon-Hydrogen Bonds

Article abstract of DOI:10.1002/cctc.201500610

The functionalization of the primary sites of alkanes is one of the more challenging areas in catalysis. In this context, a novel effect has been discovered that is responsible for an enhancement in the reactivity of the primary C-H bonds of alkanes in a catalytic system. The copper complex Cu(NCMe) (=hydrotris{[3,5-bis(trifluoromethyl)-4-bromo]-pyrazol-1-yl}borate) catalyzes the functionalization of C_nH_2n+2 with ethyl diazoacetate upon inserting the CHCO₂Et unit into C-H bonds. In addition, the selectivity of the reaction toward the primary sites significantly increased relative to that obtained in neat alkane upon using supercritical carbon dioxide as the reaction medium. This was attributed to the effect of the carbon dioxide molecules that withdraw electron density from the fluorine atoms of the ligand, which enhances the electrophilic nature of the metal center. DFT studies validated this proposal.

Full text of DOI:10.1002/cctc.201500610

DOI: 10.1002/cctc.201500610
Communications
Functionalization of C H
Alkanes: Supercritical Carbon
n 2n+2
Dioxide Enhances the Reactivity towards Primary
Carbon–Hydrogen Bonds
[
a]
[a]
[b]
[b]
Riccardo Gava, Andrea Olmos, Bµrbara Noverges, Teresa Varea, Ignacio Funes-
[c]
[a]
[a]
[c, d]
[b]
Ardoiz, Tomµs R. Belderrain, Ana Caballero,* Feliu Maseras,*
Gregorio Asensio,* and
[
a]
Pedro J. PØrez*
The functionalization of the primary sites of alkanes is one of
the more challenging areas in catalysis. In this context, a novel
effect has been discovered that is responsible for an enhance-
ment in the reactivity of the primary CÀH bonds of alkanes in
only in terms of their inertness (mainly because of their poor
[
7]
s nucleophilicity, their high bond dissociation energies, and
their low polarity) but also in terms of the unavailability of fur-
ther subsequent transformations that provide neat functionali-
zation. Moreover, functionalization of the primary sites in
a preferential manner is even more difficult to achieve. In fact,
only the alkane borylation catalytic system developed by Hart-
(CF3)2,Br
a catalytic system. The copper complex Tp
Cu(NCMe)
(CF3)2,Br
(
Tp
=hydrotris{[3,5-bis(trifluoromethyl)-4-bromo]-pyrazol-
1
-yl}borate) catalyzes the functionalization of C H with
n
2n+2
[
6]
ethyl diazoacetate upon inserting the CHCO Et unit into CÀH
wig and co-workers could be considered as selective toward
2
bonds. In addition, the selectivity of the reaction toward the
primary sites significantly increased relative to that obtained in
neat alkane upon using supercritical carbon dioxide as the re-
action medium. This was attributed to the effect of the carbon
dioxide molecules that withdraw electron density from the flu-
orine atoms of the ligand, which enhances the electrophilic
nature of the metal center. DFT studies validated this proposal.
the terminal CÀH bonds of alkanes.
The catalytic selective functionalization of nonactivated alkanes
[
1,2]
CnH2n+2 remains one of the challenges in current chemistry.
In spite of decades of effort toward that end, very few exam-
[
3]
[4]
ples (e.g., electrophilic activation, dehydrogenation, silyla-
[
5]
[6]
tion, and borylation ) of metal-catalyzed transformations of
these compounds into value-added products have been de-
scribed (Scheme 1). This lack of success can be explained not
n
Scheme 1. Catalytic alkane C H2n+2 functionalization reactions.
There is a factor that is common to the examples shown in
Scheme 1: the interaction of the reacting CÀH bond with the
metal center. Actually, this is the origin of the small number of
catalytic systems for alkane functionalization: in many cases,
the formation of very stable metal–carbon and/or metal–hy-
dride bond(s) precludes the intermediate from going further
into the functionalization step. An alternative approach con-
sists of the design of catalytic systems in which the CÀH bond
interacts with an X ligand, very often as part of an in situ gen-
erated unsaturated M=X bond, and not with the metal
[
a] R. Gava, Dr. A. Olmos, Prof. Dr. T. R. Belderrain, Dr. A. Caballero,
Prof. Dr. P. J. PØrez
Laboratorio de Catµlisis HomogØnea, Unidad Asociada al CSIC
CIQSO-Centro de Investigación en Química Sostenible and Departamento
de Química y Ciencia de Materiales
Universidad de Huelva
2
1007-Huelva (Spain)
Fax: (+34)959219942
E-mail: perez@dqcm.uhu.es
[
b] Dr. B. Noverges, Dr. T. Varea, Prof. Dr. G. Asensio
Departamento de Química Orgµnica
Facultad de Farmacia, Universidad de Valencia
Burjassot 46100-Valencia (Spain)
[8]
center. This is the basis of a series of catalytic systems in
which C H
molecules have been functionalized by the
2n+2
n
[
c] I. Funes-Ardoiz, Prof. Dr. F. Maseras
Institute of Chemical Research of Catalonia(ICIQ)
Avgda. Paisos Catalans, 16, 43007-Tarragona (Spain)
E-mail: fmaseras@iciq.es
neat, formal insertion of carbene, nitrene, or oxo units into
their CÀH bonds (Scheme 2). Group 11 metal based catalysts
have been found as efficient catalysts for those transforma-
[
9]
[
d] Prof. Dr. F. Maseras
tions, particularly if they contain pyrazolylborate ligands. In
an example of the potential of this strategy, we first described
silver-based catalysts containing highly fluorinated tris-
indazolylborate ligands that promoted the functionalization of
Departament de Quimica
Universitat Autonoma de Barcelona
0
8193-Bellaterra (Spain)
Supporting Information for this article is available on the WWW under
http://dx.doi.org/10.1002/cctc.201500610.
methane with ethyl diazoacetate as the carbene (CHCO Et)
2
ChemCatChem 2015, 7, 3254 – 3260
3254
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
report the results obtained during the development of the cat-
(
CF3)2,Br
alytic system based on Tp
Cu(NCMe) (1): we found that
this catalyst enhances the functionalization of the primary sites
of alkanes relative to that observed in the neat alkane as the
reaction medium if the reaction was performed in scCO . The
2
discovery of this completely unprecedented behavior opens
up new perspectives to improve the selectivity in alkane
chemistry.
Scheme 2. Functionalization of alkanes by formal insertion of carbene, ni-
trene, or oxo units.
The preliminary study performed with complexes
(
CF3)2,Br
(CF3)2,Br
source by using supercritical carbon dioxide (scCO2)
Tp
Cu(NCMe) (1) and Tp
Ag(thf) (2) showed their cata-
[
10]
(
Scheme 3).
The use of such a medium ensured that no
lytic potential toward the insertion of the carbene CHCO Et
2
other, more reactive CÀH bond than those of methane could
be available in the reaction mixture. Very recently, we extend-
ed this methodology to the use of a copper-based catalyst,
with the ligand hydrotris{[3,5-bis(trifluoromethyl)-4-bromo]-pyr-
groups into the CÀH bonds of C ÀC alkanes, that is, the meth-
1
4
[
11]
ane–butane series. An intriguing observation was made with
propane and butane as the substrates. In these cases, the re-
gioselectivity toward the functionalization of primary and sec-
ondary sites was nearly identical with both the copper and
silver catalysts (Scheme 4). This was in clear contrast with pre-
viously described catalytic systems with these metals by our
(
CF3)2,Br
azol-1-yl}borate (Tp
), introducing this metal for methane
[
11]
activation.
x
group and others in which for the same type of Tp ML cata-
lysts, the Cu-based catalyst was nearly ineffective toward pri-
mary sites, whereas the Ag-containing catalyst provided a cer-
tain degree of functionalization at those sites. However, these
precedents correspond to reactions performed with pentane,
hexane, or other liquid alkanes (Scheme 5) under homogene-
ous conditions and by using the alkane as the reaction sol-
[
13]
vent. The reactions with propane or butane by using cata-
lysts 1 and 2 were performed in scCO as the reaction medium,
2
and therefore, the differences in the alkane nature, concentra-
tion, and reaction phase do not allow appropriate direct com-
[
14]
parison.
We performed a series of experiments with alkanes that
could be functionalized both in homogeneous phase (alkane
as solvent) and in scCO ; pentane, 2-methylbutane, and
2
hexane were chosen as representative examples of nonactivat-
ed CÀH bonds of linear and branched alkanes. The
(
CF3)2,Br
(CF3)2,Br
Tp
Cu(NCMe) (1) and Tp
Ag(thf) (2) complexes were
employed as catalysts in their reactions with ethyl diazoace-
tate. Scheme 6 displays the results obtained with these three
model alkanes in a series of catalytic experiments performed
under identical conditions with the sole difference of the reac-
tion medium. Data available clearly show the different behav-
iors of both catalysts. Functionalization of the primary sites
Scheme 3. Functionalization of methane with copper- or silver-based cata-
lyst.
The above transformation takes place through the
intermediacy of a metallocarbene intermediate, elec-
trophilic in nature, that interacts with the CÀH bond
of methane, which acts as a (weak) nucleophile
(
Scheme 3). This strategy suffers with other alkanes
from the disadvantage of the lack of selectivity:
these carbene-transfer reactions follow the bond dis-
sociation energy trend, and the primary sites are less
prone to activation, in contrast with organometallic
activation that is known to favor primary sites over
[12]
secondary and tertiary sites.
Thus, with alkanes
having available different types of CÀH bonds, mix-
tures of products are obtained, and the primary sites
are usually the less reactive. In this contribution, we
Scheme 4. Previous propane and butane functionalization with ethyl diazoacetate and
(CF3)2,Br
Tp
M-based catalysts (M=Cu, Ag).
ChemCatChem 2015, 7, 3254 – 3260
www.chemcatchem.org
3255
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
ture (therefore out of the supercritical conditions region), the
regioselectivity observed was similar to that in the absence of
carbon dioxide (Scheme 6, bottom). This experiment suggests
that the dramatic change observed in the product distribution
by increasing the amount of carbene insertion into the primary
sites should be attributed to an effect of the supercritical
phase conditions.
A second issue deals with the plausible involvement of free
carbenes in this case. The mechanism of this metal-based car-
bene transfer from diazo compounds (see Scheme 3) is well es-
tablished to occur through the
Scheme 5. Reported functionalization of n-hexane with Cu and Ag catalysts
containing trispyrazolylborate Tp ligands: distinct behavior toward primary
sites depends on the metal.
x
intermediacy of metallocarbene
intermediates that in some
cases have been detected and/
[
15–17]
or isolated.
However, it can
be thought that the observed
reactivity could be related to
the dissociation of the carbene
ligand and that the reaction
takes place through free car-
[
18]
benes. Previous contributions
from our laboratories showed
that the regioselectivity of
hexane functionalization de-
pends on the catalyst structure
and the size of the catalytic
[
19]
pocket, excluding the involve-
ment of free carbenes. However,
to reinforce this proposal, we
performed a series of reactions
in scCO with hexane as the sub-
2
strate and with complex 1 or 2
as the catalyst or catalyst-free
experiments, the latter under ir-
radiation conditions, thus to
generate the free carbene. The
results are shown in Scheme 7,
from which the involvement of
free carbenes can be excluded
on the basis of the distinct re-
gioselectivity observed with or
without the catalysts.
Scheme 6. Influence of the reaction medium in the regioselectivity of the functionalization of alkanes catalyzed
by 1 or 2.
The results presented here in-
with silver catalyst 2 remained nearly identical upon moving
dicate that the supercritical phase induces a change in the re-
gioselectivity of the alkane functionalization reaction with the
copper-based catalyst but not with its silver analogue. Previous
from the neat alkane as the reaction solvent to scCO . On the
2
contrary, with the use of copper catalyst 1 in scCO the inser-
2
tion of the carbene group into the primary sites increased five
times with respect to that observed with silver catalyst 2 in
neat alkane. Given that all the experiments were performed at
4
08C, the observed behavior cannot be related to tempera-
ture.
At this stage, several questions may be raised about this un-
precedented performance. First, it could be thought that the
mere presence of a high pressure of carbon dioxide could be
responsible of the change in regioselectivity. However, upon
performing the reaction in neat hexane under an atmosphere
of carbon dioxide [100 atm (10.1 MPa)] but at room tempera-
Scheme 7. Functionalization of n-hexane under catalytic or photochemical
conditions. Data refer to relative reactivity corrected with the number of H
of each type.
ChemCatChem 2015, 7, 3254 – 3260
www.chemcatchem.org
3256
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
studies showed that increasing the electrophilicity at the metal
center (and subsequently at the carbene ligand) correlates well
with a certain enhancement in the functionalization at the pri-
transition-metal compounds, a computational procedure had
to be devised. From the point of view of dielectric constant,
the medium is very similar to hexane, as dielectric constants
between 1.1 and 1.5 have been reported. The experimental
data, however, point to a potential involvement of carbon di-
oxide molecules in the inner coordination sphere of the car-
bene complex. We represented this by including three explicit
carbon dioxide molecules. The number of molecules was se-
lected from the optimized structure in preliminary calculations
with a single carbon dioxide molecule, which suggests the
[10,20]
mary sites with this methodology.
Diminishing electron
density at the carbene carbon atom must be related to the
lowering of p backdonation from the metal center to the car-
bene ligand, as a result of a decrease in the donation capabili-
ties of the trispyrazolylborate ligand (Figure 1). A plausible ex-
placement of three molecules because of the C symmetry of
3
the ligand. We confirmed through preliminary calculations that
the observed variations do not depend critically on the
number of solvent molecules. For technical reasons, the three
carbon dioxide molecules were introduced in a constrained
optimization for which the structures of the minima and transi-
tion states were kept frozen at the geometries in the continu-
um calculation (without explicit solvent) and only the place-
ment of the solvent molecules was optimized. The frequency
corrections were also taken from the continuum calculation.
The reason for this procedure was purely technical. The poten-
tial energy surfaces are shallow and the transition states are
difficult to locate; the addition of three weakly bound mole-
cules made the geometry optimization extremely troublesome.
We checked that the results from the constrained and full
Figure 1. Decrease in the electron density of the copper catalyst is due to
an electronic flux from the fluorine atoms to carbon dioxide, which pulls
electron density from the ligand and therefore decreases p backbonding
from copper to the carbene ligand.
planation for that picture would be based on the already de-
optimization were very similar in one example. So, we comput-
[
21]
(CF3)2,Br
scribed, so-called “CO -philicity” of fluorocarbons, which, in
ed the reaction of the Tp
Cu(CHCO Et)(CO ) and
2 2 3
2
(
CF3)2,Br
this case, would consist of the interaction of the CF groups
Tp
Ag(CHCO Et)(CO ) complexes derived from catalysts
2 2 3
3
with carbon dioxide molecules (at very high concentrations in
1 and 2, respectively, with the primary and secondary CÀH
bonds of propane. The results are summarized in Figure 2 and
Table 1. The barriers reported in Table 1 are very low, which is
consistent with a very fast reaction between the carbene com-
plex and the alkane and is in agreement with previous compu-
the supercritical phase). In scCO , donation of electron density
2
from the fluorine atoms to the carbon atom of CO would de-
2
x
crease the electron-donating capabilities of the Tp ligand and
would subsequently enhance the electrophilicity of the car-
bene ligand. The lack of this effect in the silver system could
be attributed to the diminished p backdonation of this com-
plex, which eventually reaches the limit of electrophilicity.
This effect could be observed with other electronegative
[
22]
tational studies. The fact that one of the barriers was nega-
tive in terms of free energy is a simple consequence of the
fact that the geometries were optimized in potential energy,
and the chemical meaning is that the barrier, if it exists, is very
low for this process. In the presence of such low barriers, the
diffusion of reactants for the bimolecular process has to be
taken into account. It is in particular relevant that the self-diffu-
sion activation energy of hexane was reported experimentally
x
substituents in the Tp ligands. In one of the first examples of
x
the potential of these Tp Cu complexes to catalyze this trans-
formation, alkanes were functionalized upon treating them
Br3
with ethyl diazoacetate by using the Tp Cu(NCMe) complex
[
13a]
as the catalyst.
Insertion of the carbene group took place
exclusively into secondary and/or tertiary sites and no reactivi-
ty was observed at the primary CÀH bonds. We have now ex-
plored the catalytic capabilities of this complex in scCO in the
2
reaction of ethyl diazoacetate and n-hexane. Although the
complex is not very soluble, a certain degree of hexane func-
tionalization was observed (yield <5%). Interestingly, the prod-
uct derived from functionalization of the primary sites was not
observed. This finding assesses that the enhancement in the
selectivity toward primary sites with complex 1 as the catalyst
is due to the presence of the fluorine atoms.
Accordingly, we should explain the change that occurs at
the copper center if catalyst 1 is dissolved in scCO . For this
2
reason, we decided to perform a computational study with the
B3LYP-D3 functional. As modeling of the supercritical carbon
dioxide environment is not common in DFT calculations of
Figure 2. Optimized structure of the transition state of the primary CÀH acti-
(CF3)2,Br
vation of propane by Tp
2 2 3
Cu(CHCO Et)(CO ) .
ChemCatChem 2015, 7, 3254 – 3260
www.chemcatchem.org
3257
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
and this renders the CÀH bond more reactive. We hypothesize
that the low barriers involved in these processes make them
especially sensitive to these seemingly subtle interactions.
Why is this acceleration phenomenon not observed in reac-
tions below the critical temperature? After all, there will be
some solubility of carbon dioxide in the alkane reaction media.
The explanation is that the low availability of carbon dioxide
makes the concentration of the “activated” catalyst much
lower, which thus results in a negligible effect on the reaction
rate.
Table 1. Computed activation energies and CcarbÀH bond lengths in the
transition state for the Cu and Ag catalysts under both reaction condi-
2
tions (neat hexane and scCO ).
Metal
complex
DGact prim
C
[Š]
carbÀH
DGact sec
C
[Š]
carbÀH
[a]
À1
À1
[kcalmol
]
[kcalmol
]
x
Tp Cu
3.8
2.7
0.8
0.2
0.1
1
1
.657
.878
1.212
x
[b]
Tp Cu(CO
2
)
3
–
–
x
[c]
[c]
Tp Ag
–
x
[c]
[c]
Tp Ag(CO
2
)
3
–
–
x
(CF3)2,Br
[
a] Tp =Tp
. [b] The barrier for this process disappears because of
free energy correction. The computed free energy of the transition state
is 0.4 kcalmol below the free energy of the reactants. [c] Structure
could not be optimized, barrierless process.
Interestingly, the picture emerging from the calculations fits
À1
[21]
well with the already-described “CO -philicity” of fluorocar-
2
bons, which in this case would consist of the interaction of the
CF groups with the carbon dioxide molecules. In scCO , dona-
3
2
tion of electron density from the fluorine atoms to the CO2
carbon atom exists. Such donation would decrease the donat-
ing capabilities of the ligand and would subsequently result in
an enhancement in the electrophilicity of the carbene ligand.
The non-observance of the same effect in the experiments on
the silver system is due to the fact that the barriers are already
below the diffusion limit for these complexes, which are no
longer sensitive to electronic enhancements.
À1 [23]
to be 2.18 kcalmol . Only two of the eight barriers reported
in Table 1 are above the diffusion limit, and both of them cor-
(CF3)2,Br
respond to primary CÀH activation by the Tp
Cu complex
À1
with values of 3.8 and 2.7 kcalmol in alkane medium and
scCO medium, respectively. This fits well with our experimen-
2
tal observation that the secondary/primary ratio for systems
with a silver catalyst is independent from the medium. The re-
gioselectivity for these silver systems does not depend on the
electronic nature of the metal but on the physical phenomena
We found that functionalization of the nonactivated primary
CÀH bonds of alkanes by copper-catalyzed carbene insertion
(
e.g., diffusion, transport) that are more likely independent of
can be enhanced with the aid of supercritical CO as the reac-
2
the complex. If the process is controlled by diffusion, the com-
petition between the different CÀH bonds is no longer ruled
by the very low CÀH activation barriers. Instead, it depends on
the orientation of the approaching molecules. As a first ap-
proach, we associate the probability of competing CÀH activa-
tions to statistical considerations, and the more abundant CÀH
bonds are thus more likely to be activated. This agrees with
the experimental results.
tion medium. This is the result of the increasing electrophilicity
of the carbene complex owing to the decreased flux of elec-
tron density from the fluorinated hydrotris{[3,5-bis(trifluoro-
(
CF3)2,Br
methyl)-4-bromo]-pyrazol-1-yl}borate (Tp
) ligand to the
metal in carbon dioxide. This effect was not observed with the
silver analogue, as this complex has already reached an opti-
mal electronic situation for CÀH activation. Thus, this finding
provides a new perspective in the design of selective catalysts
toward the primary sites of alkanes, a goal we have already tar-
geted in our laboratories. Also, it opens a new window in the
effect of the reaction medium in the catalytic outcome in all
systems based on an electrophilic metal center.
Regardless of what diffusion limits allow us to observe ex-
perimentally, it is clear from Table 1 that the presence of CO2
reduces the activation barrier in all cases. The CcarbÀH distances
in the transition states are also presented in the table. There is
some electronic control in the case of the primary carbon
atoms, and the CÀH distance is longer (1.878 vs. 1.657 Š) in
the case of the silver complex with a lower barrier. For the
case of the secondary carbon atom, the distance is shorter be-
cause the process is controlled by steric effects rather than
electronic effects. The CcarbÀH values are the same in alkane
and scCO media, because we froze them in the scCO calcula-
Experimental Section
General methods
2
2
Preparations and manipulations of metal complexes were per-
formed under an oxygen-free nitrogen atmosphere by using con-
ventional Schlenk techniques. Ethyl diazoacetate (EDA) and the
liquid alkanes were purchased from Aldrich or Alfa Aesar and were
employed without further purification, whereas the gaseous alka-
nes were purchased from Air Liquide. Complexes 1 and 2 were pre-
pared according to literature procedures. The experiments under
supercritical conditions were performed in a commercial THAR
Technologies R100SYS supercritical plant or in a Phenomenex
tions, as indicated above.
We present the optimized structure of one of the transition
states in Figure 2. The carbon dioxide molecules do not favor
the process by getting close to the reaction center but instead
(
CF3)2,Br
by interacting through rather long distances with the Tp
[11]
ligand. There is one CCO2ÀF contact in the 2.75/2.95 Š range for
each carbon dioxide, and the shortest contacts between the
carbon dioxide and the pyrazolyl rings are in the 3.3 Š range.
Although the distances are long, the effect in the energy is
clear and reproduces well the experimental observation. It
seems that electrophilic carbon dioxide is able to abstract
3
3 mL HPLC column (21.2 mm ID”25 mm OD”100 mm) L-con-
nected to a HIP valve through 1.59 mm stainless-steel tubes. NMR
spectroscopy experiments were run with Agilent Technologies 400
1
13
and 500 MHz spectrometers. Chemical shifts for H and C are re-
ported as d values (ppm) relative to the deuterated solvent. GC
studies were performed by using a Varian 4500.
(CF3)2,Br
some density from the Tp
ligand, which ultimately leads
to diminished electron density at the carbene carbon atom,
ChemCatChem 2015, 7, 3254 – 3260
www.chemcatchem.org
3258
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
Reactions performed with neat alkanes as the reaction
media with complexes 1 and 2
cellence Accreditation 2014–2018 SEV-2013-0319), the Junta de
Andalucía (P10-FQM-06292, P12-FQM-01765), the EU COST Action
CM1205, the Intecat Network (CTQ2014-52974-REDC), and the
ICIQ foundation. R.G. and A.O. thank MINECO for FPI and Juan de
la Cierva fellowships. I.F-A. thanks the Severo Ochoa predoctoral
training fellowship (Ref: SVP-2014-068662).
In a Schlenk flask, complex 1 or 2 (0.005 mmol) was dissolved in
the alkane (pentane, hexane, or 2-methylbutane; 5 mL). With the
copper catalyst, a solution of EDA (57 mL) in the alkane (5 mL) was
then added over 8 h with the aid of a syringe pump. In the silver
case, EDA (57 mL) was added in one portion, and the mixture was
stirred for 3 h. The conversions and yields were determined by GC
analysis by using calibration curves. Experiments at room tempera-
ture or 408C (oil bath) gave the same results.
Keywords: CÀH activation · carbene transfer · copper · silver ·
supercritical fluids
[
1] a) K. I. Goldberg, A. S. Goldman, Activation and Functionalization of C-H
Bonds, ACS Symposium Series, Vol. 885; American Chemical Society:
Washington DC, 2004; b) P. J. PØrez, Alkane CÀH Activation by Single-Site
Metal Catalysis, Catalysis by Metal Complexes, Vol. 38, Springer: Dor-
drecht, The Netherlands, 2012; c) T. B. Gunnoe, In Physical Inorganic
Chemistry: Reactions, Processes, and Applications, Vol. 2, (Ed.: A. Bakac),
Wiley, Hoboken, NJ, 2010, pp. 495–549.
Catalytic experiments with liquid alkanes under CO pres-
2
sure
Inside a 100 mL high-pressure reactor, the catalyst (0.005 mmol)
was placed in a polypropylene container with permeable caps at
both ends, and EDA (0.5 mmol) was added to a second open con-
tainer. The alkane was added (10 mL), and the reactor was filled
[
2] a) J. Choi, A. H. R. MacArthur, M. Brookhart, A. S. Goldman, Chem. Rev.
2011, 111, 1761; b) R. H. Crabtree, Chem. Rev. 2012, 112, 1536; c) V. N.
with CO (100 atm). The mixture was stirred for 8 h at either room
2
Cavaliere, B. F. Wicker, D. J. Mindiola, Adv. Organomet. Chem. 2012, 60,
1; d) A. Caballero, P. J. PØrez, Chem. Soc. Rev. 2013, 42, 8809.
[3] A. E. Shilov, G. B. Shul’pin, Chem. Rev. 1997, 97, 2879.
4] a) C. M. Jensen, Chem. Commun. 1999, 2443–2449; b) A. S. Goldman,
A. H. Roy, Z. Huang, R. Ahuja, W. Schinski, M. Brookhart, Science 2006,
temperature or 408C, the latter ensuring supercritical conditions,
and then the system was depressurized before analyzing the mix-
ture by GC.
[
3
12, 257.
[5] a) A. D. Sadow, T. D. Tilley, Angew. Chem. Int. Ed. 2003, 42, 803; Angew.
Chem. 2003, 115, 827; b) A. D. Sadow, T. D. Tilley, J. Am. Chem. Soc.
Experiments under irradiation conditions
2
005, 127, 643.
With hexane as the substrate, the experiments were identical to
[
[
6] I. A. I. Mkhalid, J. H. Barnard, T. B. Marder, J. M. Murphy, J. F. Hartwig,
Chem. Rev. 2010, 110, 890.
7] Y.-R. Luo, Comprehensive Handbook of Chemical Bond Energies, CRC,
Boca Raton, 2007.
[8] a) M. P. Doyle, R. Duffy, M. Ratnikov, L. Zhou, Chem. Rev. 2010, 110, 704;
b) H. M. L. Davies, A. R. Dick, Top. Curr. Chem. 2010, 292, 303; c) D. N. Za-
latan, J. Du Bois, Top. Curr. Chem. 2010, 292, 347.
those above, either in hexane as the solvent or in scCO , but lack-
2
ing the catalyst and by using a UV lamp (340 nm) attached either
to the Schlenk flask or to the high-pressure reactor through the
sapphire window.
[
9] a) M. M. Díaz-Requejo, P. J. PØrez, Chem. Rev. 2008, 108, 3379; b) H. V. R.
Dias, C. J. Lovely, Chem. Rev. 2008, 108, 3223; c) M. M. Díaz-Requejo, T. R.
Belderraín, M. C. Nicasio, P. J. PØrez, Dalton Trans. 2006, 5559; d) A.
Conde, D. Balcells, M. M. Díaz-Requejo, A. Lledós, P. J. PØrez, J. Am.
Chem. Soc. 2013, 135, 3887.
Computational details
All calculations were performed with the Gaussian09 program
[24]
package by using density functional theory. We select the hybrid
[25]
B3LYP functional including the empirical dispersion correction of
[
10] a) A. Caballero, E. Despagnet-Ayoub, M. M. Díaz-Requejo, A. Díaz-Rodrí-
guez, M. E. Gonzµlez-NfflÇez, R. Mello, B. K. MuÇoz, W. Solo Ojo, G. Asen-
sio, M. Etienne, P. J. PØrez, Science 2011, 332, 835; b) M. A. Fuentes, A.
Olmos, B. K. MuÇoz, M. E. Gonzµlez-NfflÇez, R. Mello, G. Asensio, A. Cab-
allero, M. Etienne, P. J. PØrez, Chem. Eur. J. 2014, 20, 11013.
[26]
Grimme (B3LYP-D3)
and including the solvation through the
[27]
SMD model with hexane as solvent (e=1.8819). The nature of all
stationary points as minima (no imaginary frequencies) or transi-
tion states (one imaginary frequency) was confirmed through fre-
quency calculations. Connectivity between minima and transition
states was confirmed by relaxation from the transition state. Two
different basis sets were used. For geometry optimizations and fre-
quency calculations, basis set I was used: 6-31G(d) was used for C,
[11] R. Gava, A. Olmos, B. Noverges, T. Varea, E. lvarez, T. R. Belderrain, A.
Caballero, G. Asensio, P. J. PØrez, ACS Catal. 2015, 5, 3726.
[
[
12] A. Z. Janowicz, R. H. Bergman, J. Am. Chem. Soc. 1982, 104, 352.
13] a) A. Caballero, M. M. Díaz-Requejo, T. R. Belderraín, M. C. Nicasio, S. Tro-
fimenko, P. J. PØrez, J. Am. Chem. Soc. 2003, 125, 1446; b) J. Urbano, T. R.
Belderraín, M. C. Nicasio, S. Trofimenko, M. M. Díaz-Requejo, P. J. PØrez,
Organometallics 2005, 24, 1528; c) E. Despagnet-Ayoub, K. Jacob, L.
Vendier, M. Etienne, E. lvarez, A. Caballero, M. M. Díaz-Requejo, P. J.
PØrez, Organometallics 2008, 27, 4779.
[14] For a silver-based system that activates ethane or propane, see: J. A.
Flores, N. Komine, K. Pal, B. Pinter, M. Pink, C.-H. Chen, K. G. Caulton,
D. J. Mindiola, ACS Catal. 2012, 2, 2066.
15] For representative examples, see: Iron: a) C. G. Hamaker, G. A. Mirafzal,
L. K. Woo, Organometallics 2001, 20, 5171; b) Y. Li, J.-S. Huang, Z.-Y.
Zhou, C.-M. Che, X.-Z. You, J. Am. Chem. Soc. 2002, 124, 13185; c) S. K.
Russell, E. Lobkovsky, P. J. Chirik, J. Am. Chem. Soc. 2009, 131, 36; d) R. L.
Khade, F. Wenchao, Y. Ling, L. Yang, E. Oldfield, Y. Zhang, Angew. Chem.
Int. Ed. 2014, 53, 7574; Angew. Chem. 2014, 126, 7704. Ruthenium:
e) J. P. Collman, P. J. Brothers, L. McElwee-White, E. Rose, L. J. Wright, J.
Am. Chem. Soc. 1985, 107, 4570; f) E. Galardon, P. Le Maux, L. Toupeto,
G. Simmoneaux, Organometallics 1998, 17, 565; g) C.-M. Che, J.-S.
Huang, F.-W. Lee, Y. Li, T.-S. Lai, H.-L. Kwong, P.-F. Teng, W.-S. Lee, W.-C.
Lo, S.-M. Peng, Z.-Y. Zhou, J. Am. Chem. Soc. 2001, 123, 4119; h) S.
[28]
[29]
N, H, B, F, and Br; whereas the LANL2DZ basis set with its as-
sociated pseudopotential was used for Cu and Ag including f po-
larization functions in both cases (exponents of 3.525 and 1.611, re-
spectively). Basis set II was used to refine the potential energies by
[15]
using 6-31G(d,p) for light atoms
and LANL2TZ(f) for Cu and
[
30]
Ag. The full experimental system was used in the calculations
except for the substrate, for which hexane was replaced by pro-
pane, and the reason was that the conformational complexity asso-
ciated to the pending alkane chains would likely have little chemi-
cal relevance.
[
Acknowledgements
Support for this work was provided by the Ministerio de Econo-
mía
y
Competitividad (MINECO) (CTQ2014-52769-C3-1-R,
CTQ2014-52769-C3-2-R, CTQ2014-57761-R and Severo Ochoa Ex-
ChemCatChem 2015, 7, 3254 – 3260
www.chemcatchem.org
3259
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
Iwasa, F. Takezawa, Y. Tuchiya, H. Nishiyama, Chem. Commun. 2001, 59.
Osmium: i) D. A. Smith, D. N. Reynolds, L. K. Woo, J. Am. Chem. Soc.
[23] C. D. Yoo, S. C. Kim, S. I. Lee, Bull. Korean Chem. Soc. 2008, 29, 1059.
[24] Gaussian 09, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A.
Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Pe-
tersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J.
Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R.
Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H.
Nakai, T. Vreven, J. A. Montgomery, Jr. , J. E. Peralta, F. Ogliaro, M. Bear-
park, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi,
J.Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J.
Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V.
Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev,
A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Moro-
kuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dap-
prich, A. D. Daniels, Ö. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski,
D. J. Fox, Gaussian, Inc., Wallingford CT, 2009.
1993, 115, 2511; j) S.-B. Park, N. Sakata, H. Nishiyama, Chem. Eur. J. 1996,
2, 303. Cobalt: k) P. Chattopadhyay, T. Matsuo, T. Tsuji, J. Ohbayashi, T.
Hayashi, Organometallics 2011, 30, 1869; l) H. Lu, W. I. Dzik, X. Xu, L.
Wojtas, B. de Bruin, X. P. Zhang, J. Am. Chem. Soc. 2011, 133, 8518;
m) J. L. Belof, C. R. Cioce, X. Xu, X. P. Zhang, B. Space, H. L. Woodcock,
Organometallics 2011, 30, 2739. Rhodium: n) K. P. Kornecki, J. F. Briones,
V. Boyarskikh, F. Fullilove, J. Autschbach, K. E. Schrote, K. M. Lancaster,
H. M. L. Davies, J. F. Berry, Science 2013, 342, 351; o) J. L. Maxwell, K. C.
Brown, D. W. Bartley, T. Kodadek, Science 1992, 256, 1544; p) D. W. Bart-
ley, T. Kodadek, J. Am. Chem. Soc. 1993, 115, 1656.
[
16] For copper carbenes, see: a) B. F. Straub, P. Hofmann, Angew. Chem. Int.
Ed. 2001, 40, 1288; Angew. Chem. 2001, 113, 1328; b) I. V. Shishkov, F.
Rominger, P. Hofmann, Organometallics 2009, 28, 1049; c) P. Hofmann,
I. V. Shishkov, F. Rominger, Inorg. Chem. 2008, 47, 11755; d) X. L. Dai,
T. H. Warren, J. Am. Chem. Soc. 2004, 126, 10085; e) N. Mankad, J. N.
Peters, Chem. Commun. 2008, 1061.
[25] A. D. Becke, J. Chem. Phys. 1993, 98, 5648.
[26] S. Grimme, J. Antony, S. Ehrlich, H. Krieg, J. Chem. Phys. 2010, 132,
154104.
[
[
17] A. Pereira, Y. Champouret, C. Martín, E. lvarez, M. Etienne, T. R. Belder-
raín, P. J. PØrez, Chem. Eur. J. 2015, 21, 9769.
18] That was the case of a rhodium-based catalyst and diethyl diazomalo-
nate as the carbene source: A. Demonceau, A. F. Noels, J. L. Costa, A.
Hubert, J. Mol. Catal. 1990, 58, 21.
19] a) M. A. Fuentes, B. K. MuÇoz, K. Jacob, L. Vendier, A. Caballero, M. Eti-
enne, P. J. PØrez, Chem. Eur. J. 2013, 19, 1327; b) A. Caballero, M. M.
Díaz-Requejo, T. R. Belderraín, M. C. Nicasio, S. Trofimenko, P. J. PØrez,
Organometallics 2003, 22, 4145.
20] A. Caballero, P. J. PØrez, J. Organomet. Chem. 2015, 793, 108 .
21] a) P. Raveendran, S. L. Wallen, J. Phys. Chem. B 2003, 107, 1473; b) M.
Temtem, T. Casimiro, A. Gil-Santos, A. L. Macedo, E. J. Cabrita, A. Aguiar-
Ricardo, J. Phys. Chem. B 2007, 111, 1318; c) T. Tsukahara, Y. Kachi, Y.
Kayaki, T. Ikariya, Y. Ikeda, J. Phys. Chem. B 2008, 112, 16445.
[27] A. V. Marenich, C. J. Cramer, D. G. Truhlar, J. Phys. Chem. B 2009, 113,
6378.
[28] a) W. J. Hehre, R. Ditchfield, J. A. Pople, J. Chem. Phys. 1972, 56, 2257;
b) P. C. Hariharan, J. A. Pople, Theor. Chem. Acc. 1973, 28, 213; c) M. M.
Francl, W. J. Petro, W. J. Hehre, J. S. Binkley, M. S. Gordon, D. J. DeFrees,
J. A. Pople, J. Chem. Phys. 1982, 77, 3654.
[
[29] P. J. Hay, W. R. Wadt, J. Chem. Phys. 1985, 82, 299.
[30] a) D. Feller, J. Comput. Chem. 1996, 17, 1571; b) K. L. Schuchardt, B. T.
Didier, T. Elsethagen, L. Sun, V. Gurumoorthi, J. Chase, J. Li, T. L. Windus,
J. Chem. Inf. Model. 2007, 47, 1045.
[
[
Received: May 31, 2015
[
22] A. A. C. Braga, F. Maseras, J. Urbano, A. Caballero, M. M. Diaz-Requejo,
P. J. PØrez, Organometallics 2006, 25, 5292.
Revised: July 8, 2015
Published online on September 11, 2015
ChemCatChem 2015, 7, 3254 – 3260
www.chemcatchem.org
3260
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim