2,2,2-Trifluoroacetophenone as an organocatalyst for the oxidation of tertiary amines and azines to N-oxides

Article abstract of DOI:10.1002/chem.201303360

A cheap, mild and environmentally friendly oxidation of tertiary amines and azines to the corresponding Noxides is reported by using polyfluoroalkyl ketones as efficient organocatalysts. 2,2,2-Trifluoroacetophenone was identified as the optimum catalyst for the oxidation of aliphatic tertiary amines and azines. This oxidation is chemoselective and proceeds in high-to-quantitative yields utilizing 10 mol% of the catalyst and H₂O₂ as the oxidant.

Full text of DOI:10.1002/chem.201303360

DOI: 10.1002/chem.201303360
Full Paper
&
Organocatalysis
2,2,2-Trifluoroacetophenone as an Organocatalyst for the
Oxidation of Tertiary Amines and Azines to N-Oxides
Dimitris Limnios and Christoforos G. Kokotos*[a]
Abstract: A cheap, mild and environmentally friendly oxida-
tion of tertiary amines and azines to the corresponding N-
oxides is reported by using polyfluoroalkyl ketones as effi-
cient organocatalysts. 2,2,2-Trifluoroacetophenone was iden-
tified as the optimum catalyst for the oxidation of aliphatic
tertiary amines and azines. This oxidation is chemoselective
and proceeds in high-to-quantitative yields utilizing
10 mol% of the catalyst and H2O2 as the oxidant.
Introduction
metals can have a great impact, especially in products of phar-
maceutical interest, for which any metal contamination is un-
acceptable. An additional drawback of the abovementioned
procedures is that most of these methods are limited to either
aliphatic tertiary amines or azines, and only a handful of meth-
ods can be applied in both series of substrates. Although
organocatalysis has provided elegant solutions to a number of
reactions,[6] research on oxidation is less documented. A few
examples exist in literature, dealing mainly with the epoxida-
tion reaction.[7–10]
N-Oxides are key components for ubiquitously used materials
such as toilet soaps, toothpastes, detergents, shampoos, cos-
metics, and also found in products for biomedical applica-
tions.[1] They are also employed as oxidants to accomplish im-
portant reactions, such as the osmium-catalyzed dihydroxyl-
ation of olefins,[2a] the ruthenium-catalyzed oxidation of alco-
hols,[2b] the Mn-salen-catalyzed epoxidation of olefins[2c] and
the Pauson–Khand reaction.[2d] The most common approach
for the synthesis of these compounds is the oxidation of
amines (Figure 1) employing stoichiometric amounts of per-
Perhydrates and dioxiranes are among the most promising
organic oxidants,[7] which derive from ketones in conjunction
with an oxygen source. In most cases, a large excess (over
5 equiv) is required for oxidations to reach completion. At-
tempts to reduce the amount of the reaction promoter have
met with limited success. Indeed, the groups of Denmark,[8]
Yang,[9] and Shi[10] have provided elegant solutions to epoxida-
tion reactions. Today, there is an increasing demand to use oxi-
dants such as H2O2, which are environmentally friendly and do
not give rise to any waste products. Our aim was the develop-
ment of a general strategy that enables the use of substoichio-
metric amounts (10 mol%) of an organic compound as the cat-
alyst, to provide a synthetically versatile and operationally trivi-
al mode of activation of H2O2.
Figure 1. Approaches for the synthesis of N-oxides.
acids,[3a,b] activated H2O2,[3c] dioxiranes[3d–f] or oxaziridines.[3g]
These reagents are inexpensive but usually generate large
amounts of waste. Transition-metal-catalyzed processes have
been also developed.[4] Among them Re,[4a] Ti,[4b] W,[4c] Ru[4d]
and V[4e] are the most commonly employed. The flavin family is
the only class of organic molecules utilized as catalysts for the
oxidation of aliphatic tertiary amines.[5] From an environmental
standpoint, the use of precious and potentially hazardous
Results and Discussion
We have been previously engaged in the synthesis of activated
ketones as potent and selective enzyme inhibitors.[11,12] Cou-
pled with our own previous experience in organocatalysis,[13]
we envisaged the use of activated ketones as catalysts. Hydro-
gen peroxide by itself is a poor oxidant for organic oxidations.
Thus, it has to be coupled with a catalyst to create a reactive
intermediate that will efficiently carry out the oxidation
(Scheme 1). Nitriles have been employed for such activa-
tion.[3c,14] Perfluoroketones have been employed in the past for
oxidation reactions, but usually in stoichiometric amounts.[15,16]
For amine oxidation in particular, they have been employed
only for the oxidation of azines,[15] and in some cases in stoi-
[a] D. Limnios, Dr. C. G. Kokotos
Laboratory of Organic Chemistry
Department of Chemistry, University of Athens
Panepistimiopolis 15771, Athens (Greece)
Fax: (+30)2107274761
E-mail: ckokotos@chem.uoa.gr
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201303360.
Chem. Eur. J. 2014, 20, 559 – 563
559
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Table 1. Activated ketones as catalysts for the oxidation of pyridine to
pyridine N-oxide.
Entry
Catalyst
Yield
[%][a]
t
[h]
1
2
3[b]
No catalyst
29
>99(99)
>99(99)
18
18
4
Scheme 1. Design of the synthesis of N-oxides.
4
5
6
92
53
95
18
18
18
chiometric quantities.[15a] It is known that perfluoroalkyl ke-
tones in an aqueous environment exist mainly in their hydrate
form. We believed that this hydrate could react with hydrogen
peroxide to form either a perhydrate or a dihydroperoxide. It
was postulated that this perhydrate will react with MeCN and
H2O2 to generate a reactive oxidant species that will perform
the oxidation and will regenerate the catalyst to be employed
in another catalytic cycle.
7
27
18
8
9
28
44
18
18
To test our hypothesis, activated ketones were tested as cat-
alysts for the oxidation of pyridine to pyridine N-oxide by
using H2O2 as the oxidant (Table 1). When the catalyst was
omitted from the reaction, the product formed in low yields
(entry 1, Table 1). 2,2,2-Trifluoroacetophenone furnished the
best results, because pyridine N-oxide was formed in a quanti-
tative yield (entry 2, Table 1). Increasing the amounts of aceto-
nitrile and H2O2 provided a quantitative yield in just 4 h
(entry 3, Table 1). Decreasing the activation of the carbonyl led
to decreased yields, whereas the use of polyfluoroalkyl sub-
stituents led to similar high yields (entries 4–7, Table 1). Aceto-
phenone led to low yields, highlighting the need of the per-
fluoroalkyl moiety to activate the carbonyl compound to act as
the oxidation catalyst (entry 8, Table 1). Similar results were ob-
tained when acetone was employed (result not shown). Other
activated compounds as diketones, ketoacids, ketoesters, and
ketoamides were also tested but in all cases low-to-moderate
yields were obtained (entries 9–14, Table 1).
10
43
18
11
12
13
39
41
46
18
18
18
14
38
18
[a] Yield was determined by GC-MS analysis; the yield of the isolated
product is given in parenthesis. [b] Two equivalents of MeCN and H2O2
were utilized.
Once the optimum reaction conditions were identified, the
substrate scope of the oxidation was explored (Scheme 2). Ini-
tially substituted pyridines (1a–1i) were utilized leading to 2a–
2i. Substitution at either position had no effect on the reaction
outcome affording the products in almost quantitative yield
(2a–2d). Disusbstituted pyridines can also be oxidized in excel-
lent yields, except in the case of 2,5-disubstituted pyridines in
which the steric bulkiness of the substituents have an impact
in the reaction yield (2 f). Other functional groups can be toler-
ated as substituents since no byproducts were observed and
the N-oxides were obtained in excellent yields (2g–2i). Quino-
line and isoquinoline behave similarly in the reaction, leading
to excellent yields, whereas bipyridine can be doubly oxidized
(2j–2l). One of the main disadvantages of the existing meth-
ods is their lack of generality, since they are usually employed
either for azines or aliphatic tertiary amines. Our protocol pro-
vides an excellent solution to this end, since aliphatic tertiary
amines are also good substrates for oxidation in our protocol
(2m–2w).[17] Cyclic tertiary amines were oxidized to the corre-
sponding N-oxides in high to excellent yields (2m–2p). Mixed
long linear aliphatic- and short non-linear aliphatic alkyl
amines, as well as symmetrical amines are well-tolerated lead-
ing to excellent yields (2q–2v). To further extend this method-
ology, quinine was also employed (2w). The double oxidation
product (N,N’-oxide) was isolated in high yield, whereas the
double bond was left intact, indicating the chemoselectivity of
the method towards the formation of the N-oxides.
To elucidate the reaction mechanism, control experiments
were carried out (Scheme 3). In the absence of H2O2, no reac-
tion took place, confirming that H2O2 is the oxidant of the re-
action. However, H2O2 by itself or in combination with the cat-
alyst was not capable of performing the oxidation, because in
the absence of MeCN, oxidation is negligible. The amount of
the acetonitrile is of critical importance because at least one
Chem. Eur. J. 2014, 20, 559 – 563
www.chemeurj.org
560
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
imidic acid intermediate is the observation of the formation of
acetamide at the end of the reaction both by GC-MS analysis
1
and H NMR spectroscopy. At this stage, the crucial role of the
pH of the solution has to be highlighted for the peroxycarbox-
imidic acid intermediate to be generated (see Supporting In-
formation). To eliminate the possibility of radical intermediates
in this protocol, a number of control experiments were per-
formed (Scheme 3). The reaction outcome was independent
on the addition of the radical traps ((2,2,6,6-tetramethyl-piperi-
din-1-yl)oxyl (TEMPO), azobisisobutylonitrile (AIBN), and 3,5-di-
tert-butyl-4-hydroxytoluene (BHT) indicating that this protocol
does not involve any radical intermediates.
Stemming from previous acquired knowledge,[11,12] it was
postulated that perfluoroalkyl aryl ketones exist mainly in their
hydrate form in the aqueous environment of the reaction.
Indeed, 19F NMR spectroscopic experiments showed that al-
though in organic solvents the equilibrium lies towards the
ketone form, in the aqueous medium of the reaction, the hy-
drate was the main form (Scheme 4, see also the Supporting
Information). Thus, the perfluoroalkyl ketone is hydrated in the
presence of water leading to hydrate form I (Scheme 5). Once
Scheme 2. Substrate scope of the oxidation. Isolated yields. [a] MeCN
(3.0 equiv), H2O2 (2.2 equiv). [b] MeCN (8.0 equiv), H2O2 (8.0 equiv).
Scheme 4. 19F NMR spectroscopic experiments to probe the active oxidant
species.
Scheme 3. Mechanistic investigation of the reaction.
Scheme 5. Proposed reaction mechanism.
equivalent of MeCN is required to result in full oxidation of the
starting material.[18] We assume that an intermediate is formed,
which is a peroxycarboximidic acid, similar to the intermediate
that Payne and co-workers have proposed in their epoxidation
reaction.[3c] This intermediate is sluggish in promoting the reac-
tion by itself, since in the absence of the catalyst only 29%
yield was obtained. Evidence that supports the peroxycarbox-
the optimum pH is utilized, acetonitrile and H2O2 react to form
peroxycarboximidic acid II. The hydrate form of the perfluoro-
alkyl ketone is oxidized by H2O2 and II forming perhydrate IV
and leaving as a byproduct from the peroxycarboximidic acid,
acetamide III. Following the reaction mixture by 19F NMR spec-
Chem. Eur. J. 2014, 20, 559 – 563
www.chemeurj.org
561
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
troscopy and upon addition of tBuOH and MeCN, no change
was observed. However, once H2O2 was added, a new peak
was immediately observed in 19F NMR spectrum (Scheme 4).
The same peak is observed when H2O2 is added to the reaction
mixture in the absence of MeCN (see also the Supporting Infor-
mation). This peak is assigned to perhydrate IV, which can be
formed from the diol with H2O2, but it is not capable to per-
form the oxidation by itself as highlighted in Scheme 3. Perhy-
drate IV then reacts with II forming the active oxidant species
of the oxidation.[19] Upon addition of the tertiary amine, the ox-
idation occurs to the corresponding N-oxides. Finally, N-oxide
is obtained and at the same time recycling of the catalyst
occurs through generation of the hydrate I. If the tertiary
amine is not added in the reaction medium, the perhydrate is
transformed to another species, the dihydroperoxide (based
on the observation of 19F NMR spectroscopic analysis, see the
Supporting Information).
Experimental Section
General procedure for the oxidation of tertiary amines and
azines to N-oxides
Amine (1.00 mmol) was placed in a round-bottom flask followed
by 2,2,2-trifluoro-1-phenylethanone (17.4 mg, 0.10 mmol). tert-Buta-
nol (0.5 mL), aqueous buffer solution (0.5 mL, 0.6m K2CO3/4ꢁ
10À5 m EDTA tetrasodium salt), acetonitrile (0.08 mL, 1.50 mmol)
and 30% aqueous H2O2 (0.13 mL, 1.10 mmol) were added consecu-
tively. The reaction mixture was left stirring for 18 h. The crude
product was purified using flash column chromatography or alumi-
na (10–40% EtOAc in petroleum ether) to afford the desired prod-
uct.
Acknowledgements
The authors gratefully acknowledge the Operational Program
“Education and Lifelong Learning” for financial support
through the NSRF program “ENISCYSH METADIDAK
TORWN EREYNHTWN” (PE 2431) co-financed by ESF and
the Greek State.
To highlight the importance of the products of this protocol,
N-oxide 2w was employed as the catalyst in the enantioselec-
tive allylation of benzaldehyde (Scheme 6). The activation of
Keywords: green chemistry · hydrogen peroxide · N-oxides ·
organocatalysis · oxidation
[1] For selected examples, see: a) K. I. Priyadarshini, M. F. Dennis, M. A.
Naylor, M. R. L. Stratford, P. J. Wardman, J. Am. Chem. Soc. 1996, 118,
5648–5654; b) T. A. Isbell, T. P. Abbott, J. A. Dworak, US Patent
6,051,214, 2000.
Scheme 6. Asymmetric allylation of aldehydes utilizing 2w as the catalyst.
[2] a) M. Schroeder, Chem. Rev. 1980, 80, 187–213; b) S. V. Ley, J. Norman,
W. P. Griffith, S. P. Mardsen, P. Stephen, Synthesis 1994, 639–666; c) S.
Cicchi, F. Cardona, A. Brandi, M. Corsi, A. Goti, Tetrahedron Lett. 1999,
40, 1989–1992; d) S. Shambayani, W. E. Crowe, S. L. Schreiber, Tetrahe-
dron Lett. 1990, 31, 5289–5292.
[3] a) V. Boekelheide, W. J. Linn, J. Am. Chem. Soc. 1954, 76, 1286–1291;
b) L. Kaczmarek, R. Balicki, P. Nantka-Namirski, Chem. Ber. 1992, 125,
1965–1966; c) G. B. Payne, P. H. Deming, P. H. Williams, J. Org. Chem.
1961, 26, 659–663; d) A. R. Gallopo, J. O. Edwards, J. Org. Chem. 1981,
46, 1684–1688; e) R. W. Murray, R. Jeyaraman, J. Org. Chem. 1985, 50,
2847–2853; f) R. W. Murray, M. Singh, R. Jeyaraman, J. Am. Chem. Soc.
1992, 114, 1346–1351; g) W. W. Zajac Jr., T. R. Walters, M. G. Darcy, J.
Org. Chem. 1988, 53, 5856–5860.
[4] a) C. Copꢂret, H. Adolfsson, T.-A. V. Khuong, A. K. Yudin, K. B. Sharpless,
J. Org. Chem. 1998, 63, 1740–1741; b) M. Ramakrishna Prasad, G. Kama-
lakar, G. Madhavi, S. J. Kulkarni, K. V. Raghavan, Chem. Commun. 2000,
1577–1578; c) B. M. Choudary, B. Bharathi, C. Venkat Reddy, M. Lakshmi
Kantam, K. V. Raghavan, Chem. Commun. 2001, 1736–1737; d) S. L. Jain,
B. Sain, Chem. Commun. 2002, 1040–1041; e) L. Rout, T. Punniyamurthy,
Adv. Synth. Catal. 2005, 347, 1958–1960.
allyltrichlorosilane by catalytic amounts of Lewis base, especial-
ly aromatic N-oxides, has already been demonstrated in the
asymmetric allylation of aromatic aldehydes.[20] Although
a number of chiral quinoline oxides have been tested for this
reaction, simple N-oxides deriving from the Cinchona alkaloid
family have not been reported. Herein, N-oxide 2w, which has
been prepared by our protocol, was tested as the catalyst for
the asymmetric allylation of benzaldehyde. In unoptimized re-
action conditions, it was found that 10 mol% of 2w can pro-
mote the reaction at room temperature affording the product
in high yields but low enantioselectivities. A decrease in the
temperature did not lead to any improvement.
Conclusion
[5] K. Bergstad, J.-E. Backvall, J. Org. Chem. 1998, 63, 6650–6655.
[6] For books, see: a) A. Berkessel, H. Groger in Asymmetric Organocatalysis:
From Biomimetic Concepts to Powerful Methods for Asymmetric Synthesis
(Eds.: A. Berkessel, H. Groger), Wiley-VCH, Weinheim, 2005; b) P. I. Dalko
in Enantioselective Organocatalysis Reactions and Experimental Procedure
(Ed.: P. I. Dalko), Wiley-VCH, Weinheim, 2007; For selected reviews, see:
c) S. Mukherjee, J. W. Yang, S. Hoffmann, B. List, Chem. Rev. 2007, 107,
5471–5569; d) A. G. Doyle, E. N. Jacobsen, Chem. Rev. 2007, 107, 5713–
5743; e) D. W. C. MacMillan, Nature 2008, 455, 304–308.
[7] For reviews, see: a) W. Adam, G. A. Ganeshpure, Synthesis 1996, 179–
188; b) W. Adam, C. R. Saha-Moller, P. A. Ganespure, Chem. Rev. 2001,
101, 3499–3548; c) O. A. Wong, Y. Shi, Chem. Rev. 2008, 108, 3958–
3987; d) J. Mlochowski, W. Peczynska-Czoch, M. Pietka-Ottlik, H. Wojto-
wicz-Mlochowska, Open Catal. J. 2011, 4, 54–82.
A highly selective organocatalytic oxidation protocol of tertiary
amines and azines to N-oxides is described. Perfluoroalkyl aryl
ketones can be employed as catalysts (10 mol%) in such a reac-
tion leading to excellent-to-quantitative yields in short reaction
times under mild reaction conditions. The substrate scope of
the reaction is very general and a variety of aliphatic tertiary
amines and azines can be tolerated. The mechanism of the re-
action was studied and active intermediates are proposed.
Chem. Eur. J. 2014, 20, 559 – 563
www.chemeurj.org
562
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[8] a) S. E. Denmark, Z. Wu, C. M. Crudden, H. Matsuhashi, J. Org. Chem.
1997, 62, 8288–8289; b) S. E. Denmark, H. Matsuhashi, J. Org. Chem.
2002, 67, 3479–3486.
e) M. C. A. van Vliet, I. W. C. E. Arends, R. A. Sheldon, Chem. Commun.
1999, 263–264; 2,2,2-Trifluoroacetophenone has been utilized for the
epoxidation of steroids utilizing 40–50 mol% catalyst loading, see: f) K.
Nickisch, H. Arnold, R. Rohde, EP298020A1 19890104, 1989 (Chem.
Abstr. 1989, 111, 39670).
[9] D. Yang, Acc. Chem. Res. 2004, 37, 497–505.
[10] a) Z.-X. Wang, Y. Tu, M. Frohn, J.-R. Zhang, Y. Shi, J. Am. Chem. Soc.
1997, 119, 11224–11235; b) H. Tian, X. She, L. Shu, H. Yu, Y. Shi, J. Am.
Chem. Soc. 2000, 122, 11551–11552.
[11] a) C. G. Kokotos, C. Baskakis, G. Kokotos, J. Org. Chem. 2008, 73, 8623–
8626; b) G. Kokotos, Y.-H. Hsu, J. Burke, C. Baskakis, C. G. Kokotos, V. Ma-
grioti, E. A. Dennis, J. Med. Chem. 2010, 53, 3602–3610.
[17] In the absence of catalyst, but otherwise identical reaction conditions,
N-methyl morpholine and ethyldiisopropylamine gave the correspond-
ing N-oxide in 14 and 16% yield, respectively.
[18] For full experimental procedure, catalyst screening, reaction optimiza-
tion, mechanistic investigations, and full NMR spectroscopic data, see
the Supporting Information.
[12] For a review, see: E. A. Dennis, J. Cao, Y.-H. Hsu, V. Magrioti, G. Kokotos,
Chem. Rev. 2011, 111, 6130–6185.
[19] For a possible structure of the active oxidant species, see the Support-
ing Information.
[13] a) C. G. Kokotos, J. Org. Chem. 2012, 77, 1131–1135; b) M. Tsakos, C. G.
Kokotos, G. Kokotos, Adv. Synth. Catal. 2012, 354, 740–746; c) M.
Tsakos, M. R. J. Elsegood, C. G. Kokotos, Chem. Commun. 2013, 49,
2219–2221; d) C. G. Kokotos, Org. Lett. 2013, 15, 2406–2409.
[14] a) K. Yamaguchi, T. Mizugaki, K. Ebitani, K. Kaneda, New J. Chem. 1999,
23, 799–801; b) P. C. Bulman Page, A. E. Graham, D. Bethell, B. K. Park,
Synth. Commun. 1993, 23, 1507–1514.
[15] For early examples of perfluoroalkylketones utilized in amine oxidation,
see: a) B. Ganem, A. J. Biloski, R. P. Heggs, Tetrahedron Lett. 1980, 21,
689–690; b) K. Neimann, R. Neumann, Chem. Commun. 2001, 487–488.
[16] For early examples of perfluoroalkylketones utilized in oxidations, see:
a) R. D. Chambers, M. Clark, Tetrahedron Lett. 1970, 11, 2741–2742;
b) R. P. Heggs, B. Ganem, J. Am. Chem. Soc. 1979, 101, 2484–2486; c) B.
Ganem, R. P. Heggs, A. J. Biloski, D. R. Schwartz, Tetrahedron Lett. 1980,
21, 685–688; d) A. J. Biloski, R. P. Heggs, G. Bruce, Synthesis 1980, 810;
[20] For selected references, see: a) S. E. Denmark, R. A. Stavenger, Acc.
Chem. Res. 2000, 33, 432–440; b) T. Shimada, A. Kina, T. Hayashi, J. Org.
Chem. 2003, 68, 6329–6337; c) A. V. Malkov, L. Dufkova, L. Farrugia, P.
Kocovsky, Angew. Chem. 2003, 115, 3802–3805; Angew. Chem. Int. Ed.
2003, 42, 3674–3677; d) A. V. Malkov, P. Ramirez-Lopez, L. Biedermanno-
va, L. Rulisek, L. Dufkova, M. Kotora, F. Zhu, P. Kocovsky, J. Am. Chem.
Soc. J. Am. Chem.Soc. 2008, 130, 5341–5348; e) B. Bai, H.-J. Zhu, W. Pan,
Tetrahedron 2012, 68, 6829–6836; f) B. Bai, L. Shen, J. Ren, H. J. Zhu,
Adv. Synth. Catal. 2012, 354, 354–358.
Received: August 27, 2013
Published online on December 2, 2013
Chem. Eur. J. 2014, 20, 559 – 563
www.chemeurj.org
563
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim