Catalytic, PMe3-mediated conversion of secondary nitroalkanes to ketones: a very mild Nef-type process

Article abstract of DOI:10.1016/j.tetlet.2007.11.110

Aliphatic secondary nitro compounds are converted to ketones at room temperature, usually in 90-100% yields, by a one-pot reaction with 220-250 mol % of trimethylphosphine (PMe₃) and 50-100 mol % of ^tBuC₆H₄SSC₆H₄^tBu or PhthN-SePh, or 20 mol % of both additives. Thus, very mild catalytic variants of the reductive Nef-like reactions are disclosed.

Full text of DOI:10.1016/j.tetlet.2007.11.110

Available online at www.sciencedirect.com
Tetrahedron Letters 49 (2008) 441–444
Catalytic, PMe3-mediated conversion of secondary nitroalkanes
to ketones: a very mild Nef-type process
*
´
Jordi Bures, Jaume Vilarrasa
`
´mica,
Departament de Quı´mica Organica, Facultat de Quı
Universitat de Barcelona, Av. Diagonal 647, 08028 Barcelona, Catalonia, Spain
Received 12 October 2007; revised 2 November 2007; accepted 20 November 2007
Available online 24 November 2007
´
The senior author wishes to dedicate this work to Professor Joaquın Plumet (Universidad Complutense) on the occasion of his 60th birthday
Abstract
Aliphatic secondary nitro compounds are converted to ketones at room temperature, usually in 90–100% yields, by a one-pot reaction
t
with 220–250 mol % of trimethylphosphine (PMe3) and 50–100 mol % of tBuC6H4SSC6H4 Bu or PhthN-SePh, or 20 mol % of both addi-
tives. Thus, very mild catalytic variants of the reductive Nef-like reactions are disclosed.
Ó 2007 Elsevier Ltd. All rights reserved.
Keywords: Nitro compounds; Ketones; Catalysed Nef-like reaction; 4,40-Bis-t-butyldiphenyl disulfide; N-(Phenylselenenyl)phthalimide; Trimethyl-
phosphine
Nitronate anions are so easily generated that their reac-
tions with conjugate double bonds (Michael additions),
aldehydes (nitro-aldol, or Henry reaction) and other elec-
trophiles (e.g., imines) have been widely used for the prep-
aration of various starting materials.1 Moreover, the strong
electron-withdrawing character of the NO2 group makes
nitroalkenes very suitable substrates for Michael additions
and cycloadditions.2 The resulting nitro compounds can
then be converted to amines,1c oximes,3 hydrocarbons1b,4
and, in the case of primary nitro groups, into nitrile oxides
(for 3 + 2 cycloadditions)5 and nitriles.3b,6 Many methods,
most of them oxidative, are also known for the conversion
of nitro to carbonyl compounds (Nef-like reactions),7 con-
necting the organic chemistry of the nitrogen compounds
with the carbonyl chemistry. We report here a reductive
method for the quick conversion of secondary nitroalkanes
to ketones that can be performed at room temperature (rt).
The new method relies upon the use of the most reactive
trialkylphosphine (namely PMe3)8 as the reducing agent
and a suitable catalyst, and was inspired by a report of
Zard et al.,9 who treated c-nitroketones and nitro-steroids
with excesses of tributylphosphine (PBu3) and diphenyl
disulfide (PhSSPh) to obtain pyrroles and oxo-steroids,
respectively. As in any reductive method,7 nitro derivatives
or their nitronate ions are expected to give the intermediate
oximes (the main tautomers of the nitroso compounds) or
their anions, which may be then converted to the imines
in situ (Scheme 1); during the workup, the imines may be
hydrolyzed to ketones (or reduced to amines or
transformed to other functional groups).7,9a,b After an
exhaustive search for reagents and catalysts that could
work in an acceptable time and turnover at rt, we have
found that PMe3 and 4,40-bis-tert-butyldiphenyl disulfide
NO
2
NOH
PMe
PMe
3
3
R
B:
R'
R
B:
R'
cat.
cat.
NH
BH+
—
BH+
—
R
R'
NO
NO
2
O=PMe
O=PMe
3
3
R
R
R'
R'
*
Corresponding author. Tel.: +34 934021258; fax: +34 933397878.
Scheme 1. From nitroalkanes to oximes to imines.
E-mail address: jvilarrasa@ub.edu (J. Vilarrasa).
0040-4039/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2007.11.110

442
J. Bure´s, J. Vilarrasa / Tetrahedron Letters 49 (2008) 441–444
t
(tBuC6H4SSC6H4 Bu) or N-(phenylselenenyl)phthalimide
Table 1
(PhthN-SePh) are the reagents of choice.10
Conversion of nitro compounds to ketones
Since the cleavage of oximes to carbonyl compounds is
feasible by several methods,11 our main challenge was to
perform the first step under conditions as mild as possible,
in such a way that the improved procedure(s) could be
applied to polyfunctional substrates (in advanced steps of
total syntheses).
PMe (2.2 equiv)
3
NO
NH
R'
O
2
H O
2
additive A or B
R
R'
R
R
R'
THF, rt, time
rt, 5 min
1
2
O
N
S
SePh
additive B
S
O
These optimized conditions are shown in Table 1. The
reaction is carried out in a commercially available THF
solution of PMe3 (2.20 mL, 2.20 mmol) at rt under N2 or
Ar. The nitro compound (1a–h, 1 mmol) and 0.5 mmol
additive A
Entry
Substrate
Method A:a
time, yield (%)
Method B:b
time (h),
yield (%)
(50 mol %) of catalyst A (tBuC6H4SSC6H4 Bu)12,13 or B
t
NO
2
(PhthN-SePh) are added, without solvent. Although this
is enough to achieve excellent yields of 2a–h within a few
hours, we have also examined the addition of 1 mmol
(100 mol %) of A or B to shorten the reaction times even
more; the times shown in Table 1 are for 100 mol % of
additive. In practice, trimethylphosphine oxide (Me3P@O,
identified by 1H and 31P NMR) begins to precipitate within
a few minutes from the THF solution. When the reaction
via Method A is completed, partition between a nonpolar
organic solvent and water leaves 2 and additive A in the
organic layer and Me3P@O in the aqueous layer; a simple
filtration of the organic layer over silica gel affords pure
ketones 2. In Method B, the products are separated by col-
umn chromatography. Yields are almost quantitative with
both methods. Scaling up to 5 mmol the reaction of entry 2
(1b), quantitative yields of 2b have been maintained. Vari-
ous functional groups (see entries 3–8 of Table 1) are stable
under these reaction conditions.
1
1a
15 min, 97
2, 96
Ph
2
3
1b
1c
15 min, 98
30 min, 95
2, 98
3, 95
NO
2
O
Ph
OMe
NO
2
O
O
4
5
6
1d
1e
1f
30 min, 94
30 min, 98
30 min, 95
3, 96
3, 98
3, 91
Ph
NO
NO
NO
2
OTBS
OAc
Ph
Ph
2
2
OTBS
OTBS
The reactions are slower for the nitro-aldol derivatives
1i and 1j (i.e., with nitro groups sterically more crowded
and/or electronically surrounded by r-EWGs, see entries
9 and 10).14 In these cases, Method A yields a mixture of
ketone and N-(phenylsulphenyl)ketimine; to obtain the
desired ketone, a workup with an aqueous solution of
NaH2PO4 rather than just water alone is required, besides
stirring the final mixture for 1 h. In the light of the results,
Method B is more convenient than A for 1i and 1j.
Furthermore, these last substrates show a drawback,
which may be common to other nitro-aldol reaction prod-
ucts. An inversion of the stereocenter a to the carbonyl
group takes place, as determined by chiral HPLC (Chir-
alpak AD-H). It does not depend on the pH of the buffered
solution added to hydrolyse the imines, as it happens either
at pH 4, 7, or 10. On the other hand, enantiopure ketones
2i and 2j, prepared independently, are configurationally
stable under the reaction conditions. Since a-OBn and a-
OTBS N-(phenylsulfenyl)ketimines (also obtained indepen-
dently)14b do not undergo configuration inversions under
the reaction conditions, the racemization must occur at
the imine stage. In short, quick imine–enamine equilibria
(see Scheme 2) are likely responsible for the epimerization
of the a-stereocenters.
7
8
1g
1h
30 min, 96
30 min, 94
3, 94
3, 96
NO
2
OTBS
O
O
NO
2
OTBS
9
1i
1j
12 h, 84c
12, 90
syn/anti, 1:2
OTBS
NO
2
10
12 h, 70c,d
12, 76d
Ph
NO
2
a
Method A: 1.0 mmol of 1 and 1.0 mmol of additive A are added to a
1.0 M PMe3 solution in THF (2.2 mL) at 0 °C under N2 or Ar; the bath is
removed and stirring is maintained at rt for the time indicated.
b
Method B: 1.0 mmol of additive B is used instead of A.
In these cases, a special workup is required (see the main text).
Disappearance of 1j is complete, but a fragmentation by-product is
c
d
always formed in ca. 20% yields.14a
ArSSAr (Method A) must be the phosphonium salt,
ArS–PMeþ3 ArSꢀ, and/or (ArS)2PMe3, depending on the
medium polarity and reaction conditions. When PhthN-
SePh is employed (Method B), the active species are
expected to be PhSe–PMe3þ PhthNꢀ and/or its P-pentava-
lent species. In practice, we have never detected oximes
As usual for other well-known reactions involving P(III)
and P(V) derivatives, the active species from PMe3 and

J. Bure´s, J. Vilarrasa / Tetrahedron Letters 49 (2008) 441–444
443
18O
ArSH
Me P SAr
3
ArSH
ArS
NH
H218O
—
+ O
N
PMe PMe
O
OPMe
3
3
3
1a
slow
—
H
O
N
+ O
O
N
2a*
Scheme 4. Hydrolysis of the ketimine with O-labeled water.
PMe
3
ArSSAr
OPMe
3
the amounts of A and B can be reduced. For example,
we have examined the combination of 0.2 equiv of A,
0.2 equiv of B, and 2.5 equiv of PMe3 on 1a–d; it is much
more efficient than 0.4 equiv of either A or B with the same
amount of PMe3.15 An explanation for this intriguing fact
is that catalyst A is preferable for the first step (as the nitro-
to-oximate reduction is ‘slower’ with additive B, Scheme 3,
than the first ‘slow’ step with additive A, Scheme 2),
whereas PhthN-SePh show no tendency or a much lower
tendency to give selenenylimines, as mentioned above;
thus, both types of additives co-operate, increasing the
rates of the slowest steps or bypassing them.
As shown in Scheme 4, when a sample of the interme-
diate imine from 1a was quenched with 120 mol % of
18O-labeled water (95% of H218O) in a NMR tube, the
corresponding ketone (18O-labeled 4-tert-butylcyclohexa-
none, 2a*) was obtained. The highfield isotopic shift of
the carbonyl carbon atom (13C NMR) was exactly
50 ppb, as expected.16 A MS of the ketone indicated the
practically complete incorporation of the label. Similarly,
compound 1b gave 2b* (Dd = 0.050 ppm).
ArSH
SAr
ArSH
NH
SAr
slow
PMe
N
3
N
PMe
3
NH
2
Scheme 2. Proposed mechanism for the reaction of nitro compounds with
PMe3/ArSSAr, where Ar = 4-tBuC6H4 (Method A).
as intermediates by NMR (even working with substoichio-
metric amounts of PMe3). It may mean that oximes disap-
pear as soon as they are formed—from independent
experiments we know that oximes react more rapidly than
nitro compounds—and/or that oxime derivatives, more
reactive than oximes themselves, are directly involved in
the next step. A phosphonium oximate is drawn in the
mechanism suggested in Scheme 2 (Method A) as well as
in Scheme 3 (Method B), for the sake of simplification.
With the additive B, we have never detected or trapped
selenenylimines (RR0C@N–SePh) as intermediates under
the reaction conditions of Scheme 3 or any others, by con-
trast to what happens with analogous S derivatives.9,14b
Thus, in Scheme 3 we have depicted that the cleavage of
the oximate takes place by attack of an external molecule
of PMe3 to SePh, loss of O@PMe3 and trapping of
PhSe–PMeþ3 by the N atom. The role of phthalimide anion
as a base (to give phthalimide molecule in the first step) and
that of the resulting phthalimide molecule and/or the start-
ing nitro compound as proton sources have not been indi-
cated in Scheme 4, to simplify the figure.
In conclusion, we have developed a very smooth cata-
lytic procedure for the nitro-to-ketone conversion. Its
scope and limitations have been investigated. The proce-
dure is very useful for the preparation of 18O-labeled
ketones. Owing to the mildness of the protocol, we hope
to show or see soon its application to advanced steps of
total syntheses of complex molecules.
Acknowledgments
This work was funded by the Spanish Ministerio de
It is worth noting that, if catalyst A and B are used
together, the process takes place at a higher rate, so that
´
Educacion y Ciencia (Madrid) through grants SAF2002-
02728 and CTQ-2006-15393, and a studentship to J.B.
The Generalitat de Catalunya (Barcelona) contributed par-
´
tially (Grant 2001SGR065, 2002–2005, GRC de Sıntesi
PhSe
—
`
´
Estereoselectiva d’Antibiotics i Antivırics).
+ O
PMe
O
3
slower
PMe
N
H
3
—
+ O
O
N
O=PMe
Supplementary data
3
SePh
N
Me P
3
Experimental procedures and characterization data
for all new compounds. Supplementary data associated
with this article can be found, in the online version, at
doi:10.1016/j.tetlet.2007.11.110.
O
PMe
3
PhthN-SePh
SePh
PMe
NH
3
PMe
3
N
References and notes
O=PMe
3
1. For very recent reviews and leading references, see: (a) Ballini, R.;
Barboni, L.; Fringuelli, F.; Palmieri, A.; Pizzo, F.; Vaccaro, L. Green
Chem. 2007, 9, 823; (b) Ballini, R.; Palmieri, A. Curr. Org. Chem.
2006, 10, 2145; (c) Ballini, R.; Bosica, G.; Fiorini, D.; Palmieri, A.;
Scheme 3. Proposed mechanism for the reaction of nitro compounds with
PMe3/PhthN-SePh (Method B). Proton exchanges involving PhthNꢀ/
PhthNH are not drawn (see the main text).

444
J. Bure´s, J. Vilarrasa / Tetrahedron Letters 49 (2008) 441–444
Petrini, M. Chem. Rev. 2005, 105, 933; (d) Ono, N. The Nitro Group in
reagent combination had been used in the nucleoside field to
substitute PhS for OH groups: (c) Nakagawa, I.; Hata, T. Tetrahedron
Lett. 1975, 17, 1409; for the conversion of ketoximes to thioimines
and/or imines with the same reagents, see: (d) Lukin, K. A.;
Narayanan, B. A. Tetrahedron 2002, 58, 215.
Organic Synthesis; Wiley-VCH: New York, 2001; (e) Oriyama, T.;
Aoyagi, M.; Iwanami, K. Chem. Lett. 2007, 36, 612; (f) Choudary, B.
M.; Ranganath, K. V. S.; Pal, U.; Kantam, M. L.; Sreedhar, B. J. Am.
Chem. Soc. 2005, 127, 13167; for reviews of aza-Henry reactions, see:
(g) Vilaivan, T.; Bhanthumnavin, W.; Sritana-Anant, Y. Curr. Org.
Chem. 2005, 9, 1315; (h) Westermann, B. Angew. Chem., Int. Ed. 2003,
42, 151.
10. Other commercially available or readily prepared potential activators
were examined: PhSSPh, N-(phenylsulfenyl)phthalimide (PhthN-SPh),
cyclic disulfides naphtho[1,8-cd][1,2]dithiole and dibenzo[1,2]dithiine,
tBuSStBu, 2,20-dipyridyl diselenide (PySeSePy), PhSeCl, I2, CBr4 and
diethyl azodicarboxylate (DEAD). The first two were as efficient as
2. Review of asymmetric additions to nitroalkenes: (a) Berner, O. M.;
Tedeschi, L.; Enders, D. Eur. J. Org. Chem. 2002, 1877; for
cycloadditions, see: (b) Denmark, S. E.; Baiazitov, R. Y. J. Org.
Chem. 2006, 71, 593; and references cited therein.
tBuC6H4SSC6H4 Bu but their use is accompanied by the bad odor of
t
PhSH during the workup. PySeSePy was slower (in the first step, the
nitro-to-oxime conversion). The other additives did not work at all.
´
3. (a) Bartra, M.; Romea, P.; Urpı, F.; Vilarrasa, J. Tetrahedron 1990,
`
46, 587; from optically active nitroalkanes, see: (b) Czekelius, C.;
Carreira, E. Angew. Chem., Int. Ed. 2005, 44, 612.
4. Reviews: (a) Pellissier, H. Tetrahedron 2007, 63, 3235; (b) Harju, K.;
Yli-Kauhaluoma, J. Mol. Div. 2005, 9, 187; (c) Jaeger, V.; Colinas, P.
A. Chem. Heterocycl. Comp. 2002, 59, 361.
11. Reviews: (a) Corsaro, A.; Chiacchio, U.; Pistara, V. Synthesis 2001,
1903; (b) Hajipour, A. R.; Khoee, S.; Ruoho, A. E. Org. Prepn.
Proced. Int. 2003, 35, 527; also see the references cited by the very
recent, illustrative papers that follow: (c) Gupta, P. K.; Manral, L.;
Ganesan, K. Synthesis 2007, 1930; (d) Ali, M. H.; Greene, S.; Wiggin,
C. J.; Khan, S. Synth. Commun. 2006, 36, 1761; (e) Mart´ın, M.;
5. Fessard, T.; Motoyoshi, H.; Carreira, E. Angew. Chem., Int. Ed. 2007,
46, 2078; and references cited therein.
´
´
Martınez, G.; Urpı, F.; Vilarrasa, J. Tetrahedron Lett. 2004, 45, 5559.
12. We obtained it by oxidation of commercially available 4-tert-
butylbenzenethiol with air: (a) Jossi, A. V.; Bhusare, S.; Baidossi,
M.; Qafisheh, N.; Sasson, Y. Tetrahedron Lett. 2005, 46, 3583; also by
oxidation with 200 mol % of TEMPO according to, for example, (b)
Carloni, P.; Damiani, E.; Iacussi, M.; Greci, L.; Stipa, P.; Cauzi, D.;
Rizzoli, C.; Sgarabotto, P. Tetrahedron 1995, 51, 12445; we have also
´
6. (a) Urpı, F.; Vilarrasa, J. Tetrahedron Lett. 1990, 31, 7497; (b)
Mendler, B.; Kazmaier, U. Org. Lett. 2005, 7, 1715.
7. For very recent reviews of the original Nef reaction (nitronates heated
in strong acids), McMurry’s conditions (TiCl3/HCl/NH4OAc/D) and
related reactions, see: (a) Ballini, R.; Petrini, M. Tetrahedron 2004, 60,
1017; (b) Adams, J. P. J. Chem. Soc., Perkin Trans. 1 2002, 2586; (c)
Petrus, L.; Petrusova, M.; Pham-Huu, D.-P.; Lattova, E.; Pribulova,
B.; Turjan, J. Monatsh. Chem. 2002, 133, 383; recent papers: (d) Hwu,
J. R.; Josephrajan, T.; Tsay, S. Synthesis 2006, 3305 (KH, TMSCl/D);
(e) Ballini, R.; Fiorini, D.; Maggi, R.; Oro, C.; Palmieri, A.; Sartori,
G. Synlett 2006, 1849 (bicyclic guanidine, 60 °C); (f) Pradhan, P. K.;
Dey, S.; Jaisankar, P.; Giri, V. S. Synth. Commun. 2005, 35, 913 (Fe/
aq HCl/MeOH/D); (g) Gissot, A.; N’Gouela, S.; Matt, C.; Wagner,
A.; Mioskowski, C. J. Org. Chem. 2004, 69, 8997 (NaNO2, DMSO,
ca. 65 °C); and references cited therein; (h) Ballini, R.; Bosica, G.;
Fiorini, D.; Petrini, M. Tetrahedron Lett. 2002, 43, 5233 (DBU,
60 °C); for a mild oxidative method that we have utilized several times
successfully, see: (i) Ceccherelli, P.; Curini, M.; Marcotullio, M. C.;
Epifano, F.; Rosati, O. Synth. Commun. 1998, 28, 3057 (OxoneÒ,
2KHSO5ꢁKHSO4ꢁK2SO4, buffered alkaline pH, 20 °C).
t
performed experiments by adding directly TEMPO and BuC6H4SH
to the reaction flask with phosphine and the nitro compound, with the
same final yields.
13. We recommend the use of tBuC6H4SH/tBuC6H4SSC6 H4 Bu to avoid
t
the stench of benzenethiol (thiophenol, PhSH) and other relatively
volatile ArSH. The relative odors of thiols have been evaluated: (a)
Nishide, K.; Ohsugi, S.; Miyamoto, T.; Kumar, K.; Node, M.
Monatsh. Chem. 2004, 135, 189; for bis-TMS derivatives, see: (b)
Patra, P. K.; Shanmugasundaram, K.; Matoba, M.; Nishide, K.;
Kajimoto, T.; Node, M. Synthesis 2005, 447.
14. (a) Moreover, the phosphonium oximates of 1j are prone to
fragmentation (the concomitant Beckmann fragmentation that
affords nitriles and lowers the yields of the desired product, 2j). We
also observed such a fragmentation in preparing N-(phenylsulfe-
nyl)ketimines from ketoximes, with the oxime related to 1j. See: (b)
8. Review of trialkylphosphines: Valentine, D. H.; Hillhouse, J. H.
Synthesis 2003, 317.
´
Bures, J.; Isart, C.; Vilarrasa, J. Org. Lett. 2007, 9, 4635.
9. (a) Barton, D. H. R.; Motherwell, W. B.; Simon, E. S.; Zard, S. Z. J.
Chem. Soc., Perkin Trans. 1 1986, 2243; also see: (b) Barton, D. H. R.;
Motherwell, W. B.; Zard, S. Z. Tetrahedron Lett. 1984, 25, 3707; this
15. With the 0.2:0.2:2.5 A/B/PMe3 ratio, the ketone yields were P95%
within 3–5 h.
16. Risley, J. M.; Van Etten, R. L. J. Am. Chem. Soc. 1980, 102, 4609.