Symbiotic Transition-Metal and Organocatalysis for Catalytic Ambient Amine Oxidation and Alkene Reduction Reactions

Article abstract of DOI:10.1002/cctc.201501153

A new oxidation reaction based on two simple catalysts, namely, alloxan and a Cu^I salt, is highly effective for the aerobic oxidation and oxidative cross-coupling of amines. The reaction is operationally simple, reaction atmospheres enriched in dioxygen are obviated, and neither catalyst component requires prior synthesis. Mechanistic investigations have been performed and point towards a complex reaction manifold with evidence that supports a catalytic cycle that does not proceed through a quinone-imine step. Additionally, this dual catalyst system is efficient to effect diimide-mediated hydrogenation reactions of alkenes and alkynes, a transformation that has not been reported previously in the context of quinone catalyst systems.

Full text of DOI:10.1002/cctc.201501153

DOI: 10.1002/cctc.201501153
Communications
Symbiotic Transition-Metal and Organocatalysis for
Catalytic Ambient Amine Oxidation and Alkene Reduction
Reactions
Alexander T. Murray,[a] Rose King,[a] Joseph V. G. Donnelly,[a] Myles J. H. Dowley,[a]
Floriana Tuna,[b] Daniel Sells,[b] Matthew P. John,[c] and David R. Carbery*[a]
A new oxidation reaction based on two simple catalysts,
namely, alloxan and a CuI salt, is highly effective for the aerobic
oxidation and oxidative cross-coupling of amines. The reaction
is operationally simple, reaction atmospheres enriched in di-
oxygen are obviated, and neither catalyst component requires
prior synthesis. Mechanistic investigations have been per-
formed and point towards a complex reaction manifold with
evidence that supports a catalytic cycle that does not proceed
through a quinone-imine step. Additionally, this dual catalyst
system is efficient to effect diimide-mediated hydrogenation
reactions of alkenes and alkynes, a transformation that has not
been reported previously in the context of quinone catalyst
systems.
a structural homology to ortho- and para-quinone structures.
We were struck, however, by the structural similarities between
alloxan and the quinone co-factors tryptophan tryptophylqui-
none (TTQ; 2) and 2,4,5-trihydroxyphenylalanine quinone (TPQ;
3) present in copper-dependent amine oxidases (Scheme 1).[6]
These quinoproteins are an important family of redox-active
enzymes, which function in part through the efficient harness-
ing of ambient oxygen as the terminal oxidant.[7] As an oxi-
dant, ambient oxygen is arguably the ideal oxidant with re-
spect to safety and sustainability. In this regard, synthetic
chemists have often sought to find a balance to reproduce ex-
quisite enzymatic reactivity and selectivity with synthetic gen-
erality and pragmatism. As the synthesis of imines is a valuable
transformation,[8] there is a need for simple air-driven amine
oxidation reactions. In this respect, there has been a recent
growth in interest in the development of quinoprotein-like
amine dehydrogenation reactions, as well as transition-metal
catalysed oxidations under ambient conditions.[9] However, the
development of efficient methodologies based upon readily
available catalysts and effective at room temperature under
ambient air is a pertinent synthetic challenge.
Monoamines such as phenethylamine and serotonin are impor-
tant biochemical substrates that have a wide range of roles.[1]
Nature dehydrogenates monoamines by two key enzymatic
pathways, namely, the flavin-dependent[2] and copper-depen-
dent oxidase pathways.[3] In the case of the flavin-dependent
monoamine oxidase, no consensus has been reached with re-
spect to the mechanism of operation. In contrast, the consen-
sus for the copper-dependent oxidases is that the mechanism
proceeds through a pyridoxal-like transamination mode.[4]
Recently, we presented a model system that mimics the re-
activity of monoamine oxidase B (MAO B) and oper-
ates with a synthetic flavin to convert amine sub-
strates, which are processed readily by MAO B, that
is, benzylamines, to imines under ambient condi-
tions.[5] In this transformation, an additional redox
mediator, alloxan (1; Scheme 1), is used, which was
proposed to act as a single-electron acceptor. Alloxan
The translation of quinoprotein-like amine oxidations into ef-
fective synthetic reactions has led to two key problems. Firstly,
the development of readily available catalyst systems is crucial
to realise the synthetic power of biomimetic chemistry. Sec-
has the capacity to undergo two tautomerisation
Scheme 1. Alloxan tautomerisation and structural parallels to the covalently bound qui-
none co-factors TPQ and TTQ.
events (Scheme 1; 1$1a$1b), which thus offers
[a] A. T. Murray, R. King, J. V. G. Donnelly, M. J. H. Dowley, Dr. D. R. Carbery
Department of Chemistry, University of Bath
Claverton Down, Bath BA2 7AY (UK)
ondly, chemistry that proceeds under ambient conditions is
a key goal. Thus far, there have been examples of the use of
biomimetic quinone molecules in catalytic amine redox trans-
formations as electrochemical oxidants,[9a] as a single catalyst[9c]
and with a Cu co-catalyst,[9b] all of which require multi-step
synthesis of the quinone mediator and, in some cases, with
sensitive synthetic intermediates that require attention en
route.
E-mail: d.carbery@bath.ac.uk
[b] Dr. F. Tuna, D. Sells
EPSRC National EPR Facility
Alan Turing Building, University of Manchester
Oxford Road, Manchester M13 9PL (UK)
[c] Dr. M. P. John
GlaxoSmithKline Research and Development
Gunnels Wood Road, Stevenage SG1 2NY (UK)
The possibility that alloxan may act as a quinone co-factor
mimetic was suggested by two considerations. The readily ac-
Supporting Information for this article is available on the WWW under
http://dx.doi.org/10.1002/cctc.201501153.
ChemCatChem 2016, 8, 510 – 514
510
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
cessible tautomers (1a–b) would offer a structural overlap with
the quinone co-factors 2 and 3, that is, the ortho- and para-
quinone fragments. Additionally, as a commercially available
product, alloxan would offer enhanced practicality, as men-
tioned above. Accordingly, the feasibility of amine dehydro-
genation using alloxan and Cu catalysis was examined, and the
initial results are presented in Table 1.
rings are better substrates.[9c] The only exception to the excel-
lent reactivity of benzylamines was 2-furylamine (4q).[9c,f] This
was partially obviated by the use of copper(3-methylsalicylate),
a less labile Cu ligand (the yield was <20% if CuCl was used).
This effect was also observed in the oxidation of 2-methoxy-
benzylamine, which required 16 h reaction time, whereas the
oxidation of 4-methoxybenzylamine was complete in 3 h.
The more hindered substrate S-a-methylbenzylamine (4r)
was a reasonable substrate for this reaction, although it is
a less common substrate for such oxidations.[11] Additionally,
the coupling of aliphatic amines was attempted, and only
trace amounts of oxidised product were observed.
Table 1. Substrate scope.[a]
Inspired by recent work on the oxidative cross-coupling of
benzylic amines with less oxidisable amines[9b,c,f,12] and the po-
tential applications that relate to dynamic self-sorting chemis-
try,[8,13] the developed CuI/alloxan dual catalyst system was ap-
plied to the reaction of a benzylamine with an aniline (Table 2).
Table 2. Catalytic ambient oxidative cross-coupling reactions.
Entry
R1
R2
Yield [%][a]
Selectivity[b]
1
2
3
4
5
6
7
8
4-Me-C6H4
4-Me-C6H4
Ph
4-Me-C6H4
4-F-C6H4
4-Cl-C6H4
2-thiophenyl
2-OMe-C6H4
4-Me-C6H4
Ph
4-Me-C6H4
4-OMe-C6H4
4-OMe-C6H4
4-F-C6H4
4-OMe-C6H4
4-OMe-C6H4
Ph
Ph
4-pyridyl
n-oct
96
90
64
84
72
55
55
87
68[c]
11 (24)[d]
19 (29)[d]
25:1
>25:1
>25:1
16:1
>25:1
>25:1
1:1.6
11.5:1
<1:25
2.2:1
9
10
11
[a] Changes to standard reaction time shown in parentheses if applicable.
[b] Copper(I) 3-methylsalicylate used. [c] Reaction also completed inde-
pendently using the enantiomeric substrate (R)-a-methylbenzylamine
(50% yield).
4-Me-C6H4
(CH2)2Ph
2:1
[a] Isolated yield if we assumed 100% cross-coupled product. [b] As-
sessed by 1H NMR spectroscopy. [c] Yield of homo-coupled product.
[d] Assessed by comparison to an internal standard of 1,3,5-trimethoxy-
benzene; yield of cross-coupled product only (total imines in parenthe-
ses).
Acetonitrile is the best choice of solvent of those examined.
Ultimately, a loading of 1 mol% CuCl and 2.5 mol% alloxan
monohydrate was reached, which offers a low catalytic load-
ing. Pleasingly, both from an atom economy and price per-
spective, the reaction was possible with such a simple Cu salt,
without recourse to any other ligands. Sonication in a low-
power cleaning bath improves the reproducibility and per-
formance of this amine dehydrogenation reaction. This sonica-
tion may result in both homogenisation of the reaction and
also, potentially, the enhancement of the transfer of atmos-
pheric oxygen to the bulk reaction.[10]
Initial exploration demonstrated promising reactivity and selec-
tivity for the reaction of p-toluidine and 4-methylbenzylamine
(Table 1, entry 1). Good chemoselectivity was generally ob-
served with ꢀ25:1 selectivity in favour of the hetero-coupled
imine 7 observed if anilines with electron-donating groups
were used (entries 1–3, 5 and 6). The oxidative cross-coupling
with the hetero-aromatic aniline 4-aminopyridine was also at-
tempted, and only homo-coupled product 5b was isolated
from the reaction mixture (entry 9). Aliphatic amines were
found to be poor partners in this reaction (entries 10 and 11),
perhaps because of catalyst deactivation effects.
After we had optimised this oxidative dimerisation of ben-
zylamine, a range of arylmethylamine substrates was investi-
gated. Generally, this transformation works well over a relatively
short time with a low loading of the dual commercial catalysts
under an air atmosphere.
An attempt to study the kinetic profile has revealed that this
catalytic reaction is complex (Supporting Information). Over
the range of BnNH2 concentrations studied, the reaction was
best fitted to a linear, zero-order plot but had a fractional and
An examination of the substrates listed in Table 1 suggests
that the reaction is efficient for both electron-rich and elec-
tron-deficient substrates, for example, in the formation of 5c
and 5g. Previous reports on quinoprotein mimicry have shown
a substantial dependence on electronics, and electron-rich aryl
non-linear reaction order by way of a ln[BnNH2] versus ln(kobs
)
plot, which suggests a degree of rate dependence upon amine
ChemCatChem 2016, 8, 510 – 514
www.chemcatchem.org
511
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
concentration. The rate for alloxan displayed first-order kinetics
at relatively low concentrations but deviation at higher alloxan
concentrations.
which CuII would need to be present before the first turnover
to mediate O2 re-oxidation of alloxan from dialuric acid. Addi-
tionally, we saw no evidence of the formation of murexide in
this reaction, which would be expected to form if a dialuric
acid intermediate was involved and would also be expected to
be a catalyst deactivation pathway.[16]
This observation is consistent with that of Eckert and Bruice
who observed amine oxidations mediated by phenanthroline-
quinones to proceed by pseudo-first-order kinetics with re-
spect to quinone at high amine concentrations, which deviat-
ed at low amine concentrations because of the formation of
semiquinone species, that is, the reaction mechanism adopts
a radical character.[14]
We propose, based on the data discussed, that we see
a rapid pre-equilibrium between alloxan and its one-electron-
reduced form, nevertheless far enough towards the starting al-
loxan/CuI pair that we do not observe formation of the CuII by
EPR spectroscopy at this stage. A further reaction of O2 could
Additionally, the reaction also showed a dependence on
[CuCl] that approximates to first order, which suggests that
over most concentrations a possible rate-determining step
might be the CuII-mediated oxidation of a reduced alloxan spe-
cies. Finally, if the reaction was conducted under an atmos-
phere of O2, a faster reaction was observed for the first 15 min,
after which no further amine oxidation was monitored.[15]
X-band EPR spectroscopy of MeCN solutions, at room tem-
perature and 120 K, was used to probe both the oxidation
state of Cu and the presence of any organic radical species
during the reaction turnover. A strong EPR signal that dis-
played a four-line hyperfine structure was observed, commen-
surate with the presence of a Cu2+ (3d9) ion, only after the ad-
dition of benzylamine. No organic radical species were detect-
ed during these experiments. Simulation of the frozen-solution
EPR spectra provides gzz values much larger than the values of
gxx, and gyy, which suggests a symmetry close to axial (square
planar or axially elongated), and with the unpaired electron lo-
add into an alloxan radical to form
(Scheme 2).
a peroxy radical
cated in the dx2
orbital. The spectrum of the reaction mix-
Ày2
ture after 2 h shows three components (at fields Bx(1), By(2) and
Bz(3)), which reveals a distortion toward a rhombic coordination
sphere and corresponds to the three different coordination
axes, x, y and z, of the magnetic tensor. This may be explained
by differences in CuÀligand binding along the x, y and z axes
or differences in the nature of the ligands coordinated along
these axes, which could be caused by the consumption of
benzylamine that may act as a ligand for CuII (Figure 1).
Scheme 2. Plausible reaction mechanism.
Upon amine addition, the small amount of formed peroxyl
abstracts a hydrogen atom to form a hydroperoxide and ex-
trude H2O2 to regenerate alloxan. Alloxan, like similar tricarbon-
yls, would be expected to accept an electron rapidly from an
a-amino radical to form a semi-reduced alloxan, which is re-
oxidised.[17]
The involvement of a Cu-containing species in the rate-de-
termining step but a lack of appreciable build up of CuII before
the addition of amine seemed inconsistent with the quinoidal
mechanism proposed for Cu-containing amine oxidases, in
Figure 1. EPR spectrum that displays CuII-centred paramagnetic species upon amine addition (left) and after 2.5 h (right) at 120 K.
ChemCatChem 2016, 8, 510 – 514
www.chemcatchem.org
512
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
A rate-determining step that involves [Cu] does not fit if the
Cu single-electron transfers are in rapid equilibrium; however,
an alternative rationalisation is that CuCl may be activated by
the presence of amine (given that we would expect Cu amine
complexes to be formed).
purity with this substrate. The reaction tolerated basic amine
functionalities such as that in entries 6 and 9.
Although the oxidative CuII-mediated generation of a diimide
to effect alkene reduction is known,[19] to the best of our
knowledge, a CuI-catalysed aerobic hydrazine oxidation has
not been reported. Notably, this reduction reaction with allox-
an as the sole catalytic species was possible without any Cu
additives. Conversely, CuICl alone did not promote the reaction
(Scheme 3).
This study details the oxidation of amines to imines using
the CuI/alloxan system and parallels our recent flavin/alloxan
oxidation of amines.[5] With this apparent reaction homology,
further synthetic reactions that are catalysed by flavin and con-
cern N-nucleophiles were sought to expand the scope of this
protocol. Accordingly, the generation of diimide from hydra-
zine[18] to enable the pseudo-hydrogenation of alkenes and al-
kynes was assessed.
We worked initially with benzylallyl ether as the substrate,
and a clean olefin pseudo-hydrogenation is attainable using
CuCl (1 mol%) and alloxan (5 mol%) under ambient conditions
with moderate heating (Table 3, entry 1). Reliable alkene reduc-
tion was possible with this protocol, unsaturated sulfides were
also suitable and no sulfoxide was observed (entries 2 and 4)
as has been reported with a flavin catalyst.[18a] This study has
also demonstrated that internal alkenes require more forcing
conditions and regio-selectivity is achievable in cases in which
more than one alkene is present (entries 7 and 8). Finally, ter-
minal alkynes such as the monoamine oxidase inhibitor pargy-
line (8e) were also reduced in good yields to the correspond-
ing alkane 9e (entry 9). No N-oxide was observed from this
ambient reaction. Internal alkynes, such as 8 f, can also be re-
duced fully if extended reaction times are employed (entry 10).
We have also noted that there was no erosion of enantiomeric
Scheme 3. Control reactions that show reduction mediated by 1 but not
CuCl, assayed by H NMR spectroscopy.
1
In conclusion, the combination of alloxan and CuICl has
been developed as an efficient system for the aerobic oxida-
tion of amines to imines. This protocol is also adaptable to an
oxidative hetero-imination reaction if used for electronically
different amines. This catalyst system is able to perform di-
imide-type aerobic alkene and alkyne reductions with hydra-
zine. Initial mechanistic studies suggest that the rate-determin-
ing step involves both Cu species and alloxan but not the sub-
strate amine.
Acknowledgements
We acknowledge GSK, the EPSRC and the University of Bath for
a studentship (A.T.M.); EPSRC (EP/G03768X/1) for studentship
funding (J.V.G.D.) and Syngenta (R.K.) for support. We thank Dr.
Fabienne Pradaux-Caggiano for providing substrate 8j.
Table 3. Catalytic ambient reduction of alkene and alkyne substrates.
Entry
Substrate
Product
t [h]
Yield [%]
Keywords: amines · copper · cross-coupling · homogeneous
catalysis · oxidation
1
18
18
18
93
2[a]
91
93[c]
[1] a) M. D. Berry, J. Neurochem. 2004, 90, 257–271; b) M. Berger, J. A. Gray,
B. L. Roth, Annu. Rev. Med. 2009, 60, 355–366.
3
4[a]
18
18
95
98
[2] C. M. McEwen Jr., J. Biol. Chem. 1965, 240, 2003–2010.
[3] F. Buffoni, H. Blaschko, Proc. R. Soc. London Ser. B 1964, 161, 153–167.
[4] S. Janes, D. Mu, D. Wemmer, A. Smith, S. Kaur, D. Maltby, A. Burlingame,
J. Klinman, Science 1990, 248, 981–987.
5
[5] A. T. Murray, M. J. H. Dowley, F. Pradaux-Caggiano, A. Baldansuren, A. J.
Fielding, F. Tuna, C. H. Hendon, A. Walsh, G. C. Lloyd-Jones, M. P. John,
D. R. Carbery, Angew. Chem. Int. Ed. 2015, 54, 8997–9000; Angew. Chem.
2015, 127, 9125–9128.
6
18
24
90
7[a,b]
77[c]
[6] a) M. Mure, Acc. Chem. Res. 2004, 37, 131–139; b) J. P. Klinman, F.
Bonnot, Chem. Rev. 2014, 114, 4343–4365.
[7] K. Matsushita, H. Toyama, M. Yamada, O. Adachi, Appl. Microbiol. Bio-
technol. 2002, 58, 13–22.
[8] M. E. Belowich, J. F. Stoddart, Chem. Soc. Rev. 2012, 41, 2003–2024.
[9] a) M. Largeron, M.-B. Fleury, J. Org. Chem. 2000, 65, 8874–8881; b) M.
Largeron, M.-B. Fleury, Angew. Chem. Int. Ed. 2012, 51, 5409–5412;
Angew. Chem. 2012, 124, 5505–5508; c) A. E. Wendlandt, S. S. Stahl,
Org. Lett. 2012, 14, 2850–2853; d) M. Largeron, M.-B. Fleury, Chem. Eur.
J. 2015, 21, 3815–3820; e) H. Yuan, W.-J. Yoo, H. Miyamura, S. Kobaya-
shi, J. Am. Chem. Soc. 2012, 134, 13970–13973; f) E. Zhang, H. Tian, S.
8[a]
3
18
72
71[d]
69
9[a]
10[b]
87
[a] N2H4·H2O (5 equiv.). [b] Conducted at 858C. [c] Conversion assessed by
1H NMR spectroscopy. [d] GC yield.
ChemCatChem 2016, 8, 510 – 514
www.chemcatchem.org
513
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
Xu, X. Yu, Q. Xu, Org. Lett. 2013, 15, 2704–2707; g) D. V. Jawale, E.
Gravel, E. Villemin, N. Shah, V. Geertsen, I. N. N. Namboothiri, E. Doris,
Chem. Commun. 2014, 50, 15251–15254; h) A. E. Wendlandt, S. S. Stahl,
J. Am. Chem. Soc. 2014, 136, 506–512; i) A. E. Wendlandt, S. S. Stahl, J.
Am. Chem. Soc. 2014, 136, 11910–11913.
[16] J. Srogl, S. Voltrova, Org. Lett. 2009, 11, 843–845.
[17] J. C. Scaiano, J. Phys. Chem. 1981, 85, 2851–2855.
[18] a) Y. Imada, H. Iida, T. Naota, J. Am. Chem. Soc. 2005, 127, 14544–14545;
b) B. J. Marsh, E. L. Heath, D. R. Carbery, Chem. Commun. 2011, 47, 280–
282; c) Y. Imada, T. Kitagawa, T. Ohno, H. Iida, T. Naota, Org. Lett. 2010,
12, 32–35; d) J. F. Teichert, T. den Hartog, M. Hanstein, C. Smit, B. ter
Horst, V. Hernandez-Olmos, B. L. Feringa, A. J. Minnaard, ACS Catal.
2011, 1, 309–315.
[10] See the Supporting Information.
[11] Y. Qin, L. Zhang, J. Lv, S. Luo, J.-P. Cheng, Org. Lett. 2015, 17, 1469–
1472.
[12] L. Liu, S. Zhang, X. Fu, C.-H. Yan, Chem. Commun. 2011, 47, 10148–
10150.
[19] E. J. Corey, W. L. Mock, D. J. Pasto, Tetrahedron Lett. 1961, 2, 347–352.
ˇ
´
[13] K. Osowska, O. S. Miljanic, J. Am. Chem. Soc. 2011, 133, 724–727.
[14] T. S. Eckert, T. C. Bruice, J. Am. Chem. Soc. 1983, 105, 4431–4441.
[15] See the Supporting Information for full details of the kinetics experi-
ments.
Received: October 22, 2015
Published online on December 22, 2015
ChemCatChem 2016, 8, 510 – 514
www.chemcatchem.org
514
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim