Mechanistic investigation of the gold-catalyzed aerobic oxidation of aldehydes: Added insight from Hammett studies and isotopic labelling experiments

Article abstract of DOI:10.1039/b803270j

The gold-catalyzed aerobic oxidation of aldehydes proceeds through development of a partial negative charge and has a significant kinetic isotope effect (k_H/k_D = 2.8-2.9), which illustrates that activation of the C-H bond takes place in the rate-determining step. The Royal Society of Chemistry.

Full text of DOI:10.1039/b803270j

View Article Online / Journal Homepage / Table of Contents for this issue
Chemical Communications
www.rsc.org/chemcomm
Numberꢀ24ꢀꢀ|ꢀꢀ28ꢀJuneꢀ2008ꢀꢀ|ꢀꢀPagesꢀ2701–2816
ISSNꢀ1359-7345
COMMUNICATION
PeterꢀFristrupꢀet al.
FEATURE ARTICLE
SantiagoꢀAlvarezꢀet al.
Mechanisticꢀinvestigationꢀofꢀtheꢀ
gold-catalyzedꢀaerobicꢀoxidationꢀofꢀ Moleculesꢀandꢀcrystalsꢀwithꢀbothꢀ
aldehydes
icosahedralꢀandꢀcubicꢀsymmetry
1359-7345(2008)24;1-2

COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Mechanistic investigation of the gold-catalyzed aerobic oxidation of
aldehydes: added insight from Hammett studies and isotopic labelling
experimentsw
Peter Fristrup,* Louise Bahn Johansen and Claus Hviid Christensen*
Received (in Cambridge, UK) 26th February 2008, Accepted 5th March 2008
First published as an Advance Article on the web 31st March 2008
DOI: 10.1039/b803270j
The gold-catalyzed aerobic oxidation of aldehydes proceeds
through development of a partial negative charge and has a
serve as a key step in the transformation of renewable
1
resources into commodity chemicals.
0
significant kinetic isotope effect (k /kD = 2.8–2.9), which
H
Interestingly, aldehyde oxidation only takes place in the
presence of water (yielding the acid) or an alcohol (yielding the
ester), thus strongly suggesting the intermediacy of a hydrate
or hemiacetal, respectively. However, when the necessary
condition of having either water or alcohol present is satisfied,
the reaction takes place much more readily than alcohol
illustrates that activation of the C–H bond takes place in the
rate-determining step.
Since the pioneering discovery by Haruta and coworkers that
nano-particulate gold catalyzes the oxidation of CO effectively
1
at low temperatures, the scope of gold-catalyzed oxidation
11
oxidation (even significantly below room temperature ), thus
2
reactions has expanded tremendously. In particular, the
making it impossible to investigate when the reaction is started
from the alcohol.
3
oxidation of alcohols has received significant attention, and
rightfully so since the reaction satisfies most of the demands
for a modern, sustainable, and ‘‘green’’ reaction. The high
activity of the supported gold nanoparticles allows the use of
air as the stoichiometric oxidant, thus ensuring optimal atom
economy. Furthermore, the reaction has the potential to play
an important role in the synthesis of both commodity and fine
To study the influence of electronic effects in the substrate
on the rate of the reaction, we invoked the well-known
Hammett methodology, and focused our attention on a range
1
2
of selected para-substituted benzaldehydes (Scheme 1).
A
series of competition experiments were performed in metha-
nol, which allowed a direct conversion into methyl esters
suitable for GC analysis. In all cases, the disappearance of
both benzaldehydes followed first order kinetics, which facili-
tated construction of plots where the relative reactivity could
be easily determined as the slope of the line (Fig. 1).
The inherent approximations (first order reaction in both of
the benzaldehydes, and equal reaction order for other compo-
nents) are justified by the good correlation coefficients, see
Table 1.
4
chemicals, where the use of a heterogeneous catalyst allows
the isolation of product(s) by a simple separation. This
represents a major improvement compared to more ‘‘tradi-
tional’’ oxidation technologies in the speciality chemicals field,
which often rely on stoichiometric amounts of heavy metals
and/or halogens.
2
The mechanism for the oxidation of CO to CO has been
5
studied in great detail, but the mechanism(s) for oxidation of
more complex substrates have only recently received atten-
With the relative reactivities for the six para-substituted
benzaldehydes in hand, the Hammett plot could be con-
6
tion. We have previously studied the oxidation of benzyl
1
3,14
alcohol to benzaldehyde using both Hammett methodology
7
and isotope studies. These investigations clearly identified the
structed using s-values from the literature (Fig. 2).
When
examining the kinetic data it immediately became clear to us
that no single set of s-values was valid over the entire span of
reactivity from the very electron-donating substituent (OMe)
to the very electron-accepting substituent (COOMe). With the
regular s-values the slope is positive (black dots), indicating
that a partial negative charge is built up in the rate-determin-
important step in this reaction to involve the abstraction of a
hydride in the b-position with concomitant build-up of posi-
tive charge in the benzylic position. Depending on the reaction
conditions, the aldehyde can either be isolated or oxidized
8
9
further to the acid, or ester (in the presence of a suitable
alcohol), and in the current work we investigate the mechan-
ism for this oxidation of aldehydes.
ing step, but the CF -substituted aldehyde has a significantly
3
lower reactivity than expected. Instead, we attempted to fit
ꢀ
This reaction is important in itself as the produced methyl
esters are valuable chemicals (e.g. as solvents, extractants,
diluents and ingredients in perfumes), and can in addition
Creary’s s -values (blue dots), and now obtained a good
correlation for the substrates with electron-withdrawing
Center for Sustainable and Green Chemistry, Technical University of
Denmark, Kemitorvet, DK-2800 Lyngby, Denmark.
E-mail: pf@kemi.dtu.dk. E-mail: chc@kemi.dtu.dk; Fax: +45 4588
3136; Tel: +45 4525 2123 (PF); +45 4525 2402 (CHC)
w Electronic supplementary information (ESI) available: Detailed
experimental procedures for the competition experiments and for the
GC analysis (column type, temperature program and retention times).
See DOI: 10.1039/b803270j
Scheme 1 The set of para-substituted benzaldehydes which was
subjected to gold-catalyzed oxidation in the presence of methanol.
2
750 | Chem. Commun., 2008, 2750–2752
This journal is ꢁc The Royal Society of Chemistry 2008

Fig. 2 Hammett plots obtained using four different sets of s-values.
The positive slope indicates build up of a partial negative charge in the
rate-determing step.
Fig. 1 The kinetic plots illustrating the relative reactivity of the six
different para-substituted benzaldehydes relative to benzaldehyde
itself.
substituents (X = Cl, CF
3
, CN, and COOMe) indicating that
gests that proton abstraction is the rate-determining step. For
the more electron-withdrawing substituents the superior fit to
a radical mechanism is operating. For the substrates with an
electron-donating substituent the observed low reactivity is
ꢀ
Creary’s s -values strongly suggests that the radical anion of
the aldehyde is formed. In this case the reaction could be
+
À
expected from several of the possible s-values (s, s , or s )
and it is clear that a radical mechanism is not operating.
Determination of kinetic isotope effects (KIE) is one of the
most effective ways to gain insight into the nature of the rate-
determining step in a complex reaction. In the current study
we decided to employ deuterated benzaldehyde, and investi-
gate its reactivity in comparison to the non-labelled com-
ꢀ
ꢀ
abstraction of H with e.g. an Au–O–O species in a radical
1
recombination. The relative energies for these two pathways
6
+
ꢀ
(H vs. H abstraction) is likely to be heavily influenced by the
inherent electronic properties of the substrate. In the current
study the transferred proton is abstracted from a benzylic
position, allowing for a stabilizing delocalization of the devel-
oped negative charge. In the more general case of alcohol
oxidation (e.g. with ethanol) it is not a priori clear which of the
two paths are favourable, however the build up of negative
charge in the transition state documented in the current study
should still be valid.
À
pound. If addition of a nucleophile (e.g. MeOH or MeO )
took place in the rate-determining step the KIE should be less
than unity (‘‘inverse’’, k
H
/k
D
o 1), due to a loosening of the in-
plane vibration modes in the transition state. On the other
hand, if a weakening of the C–H(D) bond occurred in the
transition state for oxidation, the KIE should be larger than
It is generally accepted that in gold-catalyzed oxidations the
transferred hydrogen is accepted by an oxygen species residing
on the gold surface. Several different oxygen species have been
unity (‘‘normal’’, k
H
/k
D
4 1) with a theoretical maximum
15
value of 7 when the bond is fully broken in TS (298 K). The
KIE was first determined indirectly, by comparing the reactiv-
ities of benzaldehyde and d-benzaldehyde to that of p-CH3-
benzaldehyde in two separate experiments. The resultant KIE
was 2.83, which clearly indicates that the bond to H (or D) is
partially broken in the rate-determining transition state. Using
a GC-MS it was also possible to perform the competition
experiment in a direct fashion using selected ion monitoring
1
7
suggested, including oxo, peroxo and superoxo. In the gas-
phase oxidation of CO to CO the additional oxygen clearly
2
originates from dioxygen, however, it is less clear from which
source the oxygen originates in liquid phase oxidations. The
formal addition of the MeO moiety from methanol seems to be
a reasonable suggestion, which could arise via prior formation
of the hemiacetal, and to investigate this point in detail we
1
8
(
SIM) for the molecular ions of both compounds (M = 106
acquired O -labelled oxygen and methanol. The oxidation
reaction was then conducted as usual, while monitoring the
reaction mixture by GC-MS using SIM. In the experiment
with labelled oxygen we did not observe significant incorpora-
and 107 for the normal and deuterated substrate, respectively).
This experiment resulted in a KIE of 2.88 in excellent agree-
ment with the result from the indirect determination, thus also
adding further support to the validity of the competition
experiments.
1
8
tion of O , only 1%, indicating that oxygen originating from
dioxygen is practically never incorporated into the ester
product. On the other hand, when labelled methanol (95%
isotopic purity) was used as a solvent, the molecular ion of the
The combination of a relatively large primary KIE and
simultaneous development of a partial negative charge sug-
Table 1 Characteristics for the competition experiments
2
para-Substituent X
Relative reactivity k
X
/k
H
Correlation coefficient R
Final conversion PhCHO
Final conversion p-X-PhCHO
CN
19.8
9.82
2.49
1.61
0.80
0.39
0.992
0.991
0.998
0.989
0.988
0.944
12%
15%
35%
38%
46%
68%
95%
95%
66%
56%
38%
35%
COOMe
Cl
CF
3
Me
OMe
This journal is ꢁc The Royal Society of Chemistry 2008
Chem. Commun., 2008, 2750–2752 | 2751

1
8
O -labelled methyl ester was, not surprisingly, significantly
larger than that of the un-labelled ester (92 : 8), indicating that
nearly complete incorporation of the labelled oxygen-atom
had taken place. Thus, it is clear that in the present study
dioxygen functions as the stoichiometric oxidant by providing
reactive oxo functionalities, and that the methoxy moiety
always originates from methanol. This suggests that reversible
formation of the hemiacetal takes place prior to reaction with
Nature, 2005, 437, 1132–1135; (b) A. Abad, P. Conceptio
A. Corma and H. Garcıa, Angew. Chem., Int. Ed., 2005, 44,
066–4069; (c) D. I. Enache, J. K. Edwards, P. Landon,
´
n,
´
4
B. Solsona-Espriu, A. F. Carley, A. A. Herzing, M. Watanabe,
C. J. Kiely, D. W. Knight and G. J. Hutchings, Science, 2006, 311,
362–365.
. P. Fristrup, L. B. Johansen and C. H. Christensen, Catal. Lett.,
2
. (a) A. Corma and M. E. Domine, Chem. Commun., 2005,
4042–4044; (b) C. H. Christensen, B. Jørgensen, J. Rass-Hansen,
K. Egeblad, R. Madsen, S. K. Klitgaard, S. M. Hansen,
M. R. Hansen, H. C. Andersen and A. Riisager, Angew. Chem.,
Int. Ed., 2006, 45, 4648–4651.
. (a) T. Hayashi, T. Inagaki, N. Itayama and H. Baba, Catal.
Today, 2006, 117, 210–213; (b) I. S. Nielsen, E. Taarning,
K. Egeblad, R. Madsen and C. H. Christensen, Catal. Lett.,
7
8
008, 140, 184–190.
1
8
the gold–oxo functionality.
In the present study we have investigated the mechanism for
the gold-catalyzed, aerobic direct oxidation of aldehydes to
methyl esters in methanol, and found that a significant partial
negative charge is developed. The determined kinetic isotope
effect (k /k ) was found to be 2.8–2.9, clearly indicating that
9
2
007, 116(1–2), 35–40.
H
D
1
0. (a) A. Corma, S. Iborra and A. Velty, Chem. Rev.2007, 107,
2411–2502; (b) E. Taarning, I. S. Nielsen, K. Egeblad,
R. Madsen and C. H. Christensen, Chem. Sus. Chem., 2008, 1,
75–78.
breakage of the bond to the neighbouring hydrogen atom
takes place in the rate-determining step. In addition, experi-
1
8
ments which involved either O -labelling of oxygen or metha-
nol clearly indicated that the additional oxygen functionality
present in the product ester originates from methanol rather
than dioxygen, which is compatible with reversible formation
of the hemiacetal prior to the actual proton abstraction.
Center for Sustainable and Green Chemistry is sponsored
by the Danish National Research Council in the period
1
1. C. Marsden, E. Taarning, D. Hansen, L. Johansen,
S. K. Klitgaard, K. Egeblad and C. H. Christensen, Green Chem.,
2008, 10, 168–170.
2. For other Hammett studies conducted using competition experi-
ments, see: (a) L. Keinicke, P. Fristrup, P.-O. Norrby and
R. Madsen, J. Am. Chem. Soc., 2005, 127, 15756–15761;
(b) P. Fristrup, S. Le Quement, D. Tanner and P.-O. Norrby,
Organometallics, 2004, 23, 6160–6166.
1
1
1
1
3. C. Hansch, A. Leo and R. W. Taft, Chem. Rev., 1991, 91,
1
2
005–2010.
65–195.
4. X. Creary, M. E. Mehrsheikh-Mohammadi and S. McDonald,
J. Org. Chem., 1987, 52, 3254–3263.
Notes and references
5. As a simplification tunneling has been neglected, and in addition
only the differences in zero-point energies (ZPE) were considered.
At low temperatures tunneling can be important, and in such cases
there is no theoretical upper limit to the KIE.
1
. M. Haruta, T. Kobayashi, H. Sano and N. Yamada, Chem. Lett.,
987, 16, 405–408.
. (a) A. S. K. Hashmi and G. J. Hutchings, Angew. Chem., Int. Ed.,
006, 45, 7899–7936; (b) P. P. Edwards and J. M. Thomas, Angew.
1
2
2À
2
16. This could result in the formation of H
O
2 2
(or HO ), which have
Chem., Int. Ed., 2007, 46, 5480–5486; (c) G. C. Bond, C. Louis and
D. T. Thompson, in Catalysis by Gold, Imperial College Press,
London, 2006.
been shown also to be active as oxidant, see: M. Comotti,
C. D. Pina, E. Falletta and M. Rossi, Adv. Synth. Catal., 2006,
348, 313–316.
3
4
. T. Mallat and A. Baiker, Chem. Rev., 2004, 104, 3037–3058.
. T. Ishida and M. Haruta, Angew. Chem., Int. Ed., 2007, 46,
17. (a) J. Guzman, S. Carretin, J. C. Fierro-Gonzalez, Y. Hao,
B. C. Gates and A. Corma, Angew. Chem., Int. Ed., 2005, 44,
4778–4781; (b) T. V. W. Janssens, B. S. Clausen, B. Hvolbæk,
H. Falsig, C. H. Christensen, T. Bligaard and J. K. Nørskov, Top.
Catal., 2007, 44, 15–26.
18. For a study on methanol oxidation under UHV conditions, see:
D. A. Outka and R. J. Madix, J. Am. Chem. Soc., 1987
109, 1708–1714.
7
154–7156.
. M. S. Chen and D. W. Goodman, Science, 2004, 316,
52–255.
5
6
2
. (a) M. D. Hughes, Y.-J. Xu, P. Jenkins, P. McMorn, P. Landon,
D. I. Enache, A. F. Carley, G. A. Attard, G. J. Hutchings,
F. King, E. H. Stitt, P. Johnston, K. Griffin and C. J. Kiely,
2
752 | Chem. Commun., 2008, 2750–2752
This journal is ꢁc The Royal Society of Chemistry 2008