Improved procedure for the synthesis of enamine N-oxides

Article abstract of DOI:10.1021/jo8002166

(Chemical Equation Presented) An improved procedure for the preparation of enamine N-oxides involving aminolysis of epoxides, chlorination, N-oxidation, and dehydrochlorination is described. Although isolated β-chloroamine N-oxides are prone to rearrangem

Full text of DOI:10.1021/jo8002166

Improved Procedure for the Synthesis of
Enamine N-Oxides
David Bernier, Alexander J. Blake, and Simon Woodward*
School of Chemistry, The UniVersity of Nottingham,
Nottingham NG7 2RD, United Kingdom
simon.woodward@nottingham.ac.uk
ReceiVed February 4, 2008
FIGURE 1. All known enamine N-oxides reported in the literature
(1900-2008) prior to these studies.9
Direct N-oxidation of enamines is not a viable strategy to
such compounds. Such attempted oxidations usually lead instead
to complex reaction mixtures, comprising of mainly amides,
R-amino ketones, and in some cases, unstable amino epoxides.
All of these products are consistent with a C- rather than
N-oxidation pathway.10 Only two successful approaches to the
synthesis of genuine enamine N-oxides have been reported:
retro-Cope addition of a N,N-dialkylhydroxylamine to activated
alkynes6 and HX elimination of tertiary amine N-oxides
substituted by a suitable ꢀ-leaving group.7,8 However, the
species of the retro-Cope addition have limited stability, often
undergoing subsequent rearrangements.11 The HX elimination
strategy, which we refer to as the “Richmond-O’Neil” procedure,
requires (a) preparation of tertiary amines bearing leaving groups
on the ꢀ-position, (b) oxidation to the N-oxide, and finally (c)
deprotonation at the R-position to the N-oxide to induce the
desired CdC double-bond formation. However, use of this
“Richmond-O’Neil” strategy is complicated by the poor litera-
ture availability and limited stability of many potential ꢀ-leaving
group amine precursors and in some cases the need for aqueous
workup and tedious drying of water-soluble organic salts.7
In connection with another project, we had need of a range
of enamine N-oxides. Here we present a useful form of the
An improved procedure for the preparation of enamine
N-oxides involving aminolysis of epoxides, chlorination,
N-oxidation, and dehydrochlorination is described. Although
isolated ꢀ-chloroamine N-oxides are prone to rearrangements
when isolated, these side reactions can be slowed by the
presence of stabilizing organic acids. The scope and limita-
tions of this strategy are discussed.
Tertiary amine N-oxides [R3NO] are commonly encountered
in organic chemistry,1 being used in transition-metal-catalyzed
oxidations2 as protecting groups for sensitive tertiary amines3
and in some cases as chiral promoters for transition-metal-
catalyzed reactions4 or in organocatalysts.5 However, enamine
N-oxides 1, i.e., tertiary amine N-oxides where the nitrogen bears
vinylic (but not aromatic or heteroaromatic) substituents, have
only been the object of scarce reports (Figure 1).6–8
* To whom correspondence should be addressed. Tel: +44-(0)115-9513541.
Fax: +44-(0)115-9513564.
(1) For reviews on the synthesis and reactivity of tertiary amines N-oxides,
see inter alia: Albini, A. Synthesis 1993, 263–277.
(8) O’Neil, I. A.; Wynn, D.; Lai, J. Y. Q. Tet. Lett. 2000, 41, 271–274.
(9) Searches using Beilstein and SciFinder databases, March 2008.
(2) In Upjohn dihydroxylation, see, for instance: VanRheenen, V.; Kelly,
R. C.; Cha, D. Y. Tetrahedron Lett. 1976, 1973–1976. Choudary, B. M.; Chodari,
N. S.; Jyothi, K.; Kantam, M. L. J. Am. Chem. Soc. 2002, 124, 5341–5349.
Ley, S. V.; Ramarao, C.; Lee, A.-L.; Ostergaard, N.; Smith, S. C.; Shirley, I. M.
Org. Lett. 2003, 5, 185–187. Molander, G. A.; Figueroa, R. Org. Lett. 2006, 8,
75–78. In Sharpless asymmetric dihydroxylation: Ogino, Y.; Chen, H.; Kwong,
H.-L.; Sharpless, K. B. Tetrahedron Lett. 1991, 32, 3965–3968. Branco, L. C.;
Afonso, C. A. M. J. Org. Chem. 2004, 69, 4381–4389. In perruthenate-catalyzed
oxidation of alcohols to carboxylic acids: Xu, Z.; Johannes, C. W.; Houri, A. F.;
La, D. S.; Cogan, D. A.; Hofilena, G. E.; Hoveyda, A. H. J. Am. Chem. Soc.
1997, 119, 10302–10316. In Kornblum-like oxidation of benzylic halides to
aromatic aldehydes: Barbry, D.; Champagne, P. Tetrahedron Lett. 1996, 37,
7725–7726.
(10) Oxidationofenamines:(a)H2O2oxidationyieldinganaminoepoxide:Coffen,
D. L.; Korzan, D. G. J. Org. Chem. 1971, 36, 390–395. (b) m-CPBA oxidation
yielding ꢀ-hydroxy amines (via epoxidation, epoxide opening and nucleophilic
attack of the resulting iminium): Sunose, M.; Anderson, K. M.; Orpen, A. G.;
Gallagher, T.; Macdonald, S. J. F. Tetrahedron Lett. 1998, 39, 8885–8888. (c)
m-CPBA oxidation yielding an amino epoxide: Iwasa, K.; Sugiura, M.; Takao,
N. J. Org. Chem. 1982, 4275–4280. (d) O2 oxidation yielding R-amino ketones
and products of the oxidative cleavage of the CdC double bond: Jerussi, R. A.
J. Org. Chem. 1969, 34, 3648–3650. (e) Blau, K.; Voerckel, V. J.Prakt. Chem.
1989, 331, 285–292. (f) Blau, K.; Kapst, U.; Voerckel, V. J. Prakt. Chem. 1989,
331, 671–676. (g) Dimethyldioxirane oxidation yielding an unstable R-amino
epoxide: Adam, W.; Ahrweiler, M.; Paulini, K.; Reiꢀig, H.-U.; Voerckel, V.
Chem.Ber. 1992, 125, 2719–2721. (h) The related reaction of enamines with
elemental sulphur yields thioamides: Murata, S.; Suzuki, K.; Miura, M.; Nomura,
M. J. Chem. Soc., Perkin Trans. 1 1990, 361–365, and is considered to be a key
step of the Willgerodt-Kindler reaction.
(3) Greene, T. W.; Wuts, P. G. M. ProtectiVe Groups in Organic Synthesis,
3rd ed; John Wiley & Sons; New York, 1999.
(4) (a) Chen, F.-X.; Zhou, H.; Liu, X.; Qin, B.; Feng, X.; Zhang, G.; Jiang,
Y. Chem. Eur. J. 2004, 10, 4790–4797. (b) Kerr, W. J.; Kirk, G. G.; Middlemiss,
D. Synlett 1995, 1085–1086. Derdau, V.; Laschat, S.; Jones, P. G. Heterocycles
1998, 48, 1445–1453.
(11) Addition of Et2N-OH to methyl propiolate (1 equiv) in Et2O at 0°C
provides a white insoluble solid, presumably the expected N-oxide. However,
as described by Winterfelt and Krohn for a similar addition,6a this product
rearranges within minutes in CDCl3 at rt into a less polar one; NMR of the
rearranged product [olefinic signals: 1H δ ) 7.48 (d, J ) 12.5 Hz) and 5.47 (d,
J ) 12.5 Hz); 13C δ ) 164.9, 85.4] closely matches that reported by Hwu et
al.6b as the product of a similar addition. However, we believe these signals
should be assigned to the rearranged product Et2N-O-CHdCH-CO2Me. See
also: Bottle, S.; Busfield, W. K.; Jenkins, I. D.; Skelton, B. W.; White, A. H.;
Rizzardo, E.; Solomon, D. H. J. Chem. Soc., Perkin Trans. 2 1991, 1001–1008.
(5) Chen, F.-X.; Qin, B.; Feng, X.; Zhangb, G.; Jiang, Y. Tetrahedron 2004,
60, 10449–10460. Huang, J.; Liu, X.; Wen, Y.; Qin, B.; Feng, X. J. Org. Chem.
2007, 72, 204–208. Qin, B.; Liu, X.; Shi, J.; Zheng, K.; Zhao, H.; Feng, X. J.
Org. Chem. 2007, 72, 2374–2378.
(6) (a) Winterfelt, E.; Krohn, W. Chem.Ber. 1969, 102, 2336–2345. (b) Hwu,
J. R.; Patel, H. V.; Lin, R. J.; Gray, M. O. J. Org. Chem. 1994, 59, 15771582.
(7) Krouwer, J. S.; Richmond, J. P. J. Org. Chem. 1978, 43, 2464–2466.
10.1021/jo8002166 CCC: $40.75
Published on Web 05/07/2008
2008 American Chemical Society
J. Org. Chem. 2008, 73, 4229–4232 4229

close-contact ion pair.16 Dropwise addition of 7 to a solution
of m-chloroperbenzoic acid17 (m-CPBA) in CHCl3 or CH2Cl2
leads to the easily isolated salts 818 (Supporting Information).
Attempted purification of 8 is fraught with difficulties as it
often unmasks the reactive free N-oxides 9. The free oxides
(pKa ∼ 4.56-4.75)19 have appreciable O-nucleophilicity. In a
remarkable study, Owari and co-workers showed that N-oxides
of nitrogen mustards undergo intramolecular nucleophilic
displacement of a chloride by the oxygen of the N-oxide,
rearranging to form a highly reactive oxazetidinium.20 Trans-
posed to our case, a similar pathway leads to oxazetidinium
10; subsequent opening by either chloride or chlorobenzoate
anions leads to the rearrangement products 11 and 12, respec-
tively, which we could isolate in several instances. Such
rearrangement processes had previously been noted only in
passing as an artifact in attempted N-oxidations of nitrogen
mustards21 and in the synthesis of the anticancer agent “chloram-
bucil”.22 Consistent with this picture, attempted oxidation of
7ea (R1 ) Bui, R2 ) H) fashioned only a mixture of the desired
salt 8 and the corresponding rearranged product 12ea. We
ascribe this behavior to Thorpe-Ingold effect23 promoted
closure to 10. For this reason, the majority of our studies have
concentrated on methyl and pyrrolidino substitutents in the R1
position. Similarly, no attempt was made to prepare 7ef (R2 )
Ph) as it is highly likely that the enhanced benzylic nature of
the leaving group in 8 will favor decomposition via 10.
Deprotonation is also an issue with enamine N-oxide 1de. The
activated allylic/benzylic position in 1de apparently suffers from
competing proton abstraction leading to a complex mixture.
Attempted synthesis of 1ga equally failed, as none of the
expected N-oxide 8ga could be isolated from the complex24
(black) mixture of products obtained after the oxidation of the
corresponding ꢀ-chloro(diphenyl)amine 7ga (this behavior was
attributed to competitive oxidation at the aromatic ring).25 We
could also show that mildly basic conditions (3 equiv of K2CO3,
CH3CN) induced selective “Owari-type” rearrangement of the
SCHEME 1. New Epoxide-Based Route to Enamine
N-Oxides 1
“Richmond-O’Neil” procedure (Scheme 1) and comment on the
pitfalls that can befall the syntheses of these interesting polar
compounds.
(16) See, for instance: D’hooghe, M.; Speybroeck, V. V.; Waroquier, M.;
Kimpe, N. D. Chem. Commun. 2006, 1554–1556.
Several simple (2-chloroethyl)amines 7 (R2 ) H) are com-
mercial as hydrochloride salts, and the corresponding free
amines12 can be used as an entry point into this procedure.13 In
the case of substituted ꢀ-chloroamines 7 (R2 ) n-Bu, c-C6H11,
and PhCH2CH2), our preparation is based on a sequence of
epoxide aminolysis and chlorination of the resulting amino
alcohols. The opening of terminal epoxides 114 with secondary
amines 2 is known to proceed, in general, from the least hindered
side to provide regioisomer 4 (or mixtures with the regioisomer
5).15 Direct treatment of these products with SOCl2 leads to
single ꢀ-chloro substituted amines 7 via regiospecific opening
of the aziridinium ions 6, presumably via the formation of a
(17) This order of addition is essential: we assume that protonation of the
N-oxides prevents the formation of the 1,2-oxazetidinium 10 and therefore limits
the formation of “Owari” rearrangement products.
(18) The salts 8 are very hygroscopic, and their stability varies with their
degree of substitution: when R2 ) H, slow but steady degradation takes place at
rt and can be monitored by 1H NMR. These samples should be stored at 0-4
°C and used within 1 week. When R2 ) alkyl (8db, 8dc, 8dd), no significant
change in the 1H NMR spectra was noted upon storage in a closed flask at room
temperature for 2 weeks.
(19) pKa of queous solutions of Me3NO: Perrin, D. D., Dissociation constants
of organic bases in aqueous solutions; Butterworths: London, 1965; p 473. Lin,
T.-Y.; Timasheff, S. N. Biochemistry 1994, 33, 12695–12701. Qu, Y.; Bolen,
D. W. Biochemistry 2003, 42, 5837–5849. In organic media, see: Chmurzy´nski,
L.; Pawlak, Z. J. Chem. Thermodynam. 1998, 30, 27–35.
(20) Owari, S. Chem. Pharm. Bull. 1953, 1, 353–357.
(21) Sakurai, Y.; Izumi, M. Chem. Pharm. Bull. 1953, 1, 297–301. Kuwada,
Y. Chem. Pharm. Bull. 1960, 8, 807–814.
(12) Hickmott, P. W.; Wood, S.; Murray-Rust, P. J. Chem. Soc., Perkin Trans.
1 1985, 2033–2038, Unhindered ꢀ-chloroamines should be prepared no more
than 24 h before use, as they exhibit a propensity to spontaneous ring closure to
the corresponding aziridinium salts.
(22) Tercel, M.; Wilson, W. R.; Denny, W. A. J. Med. Chem. 1995, 38,
1247–1252.
(23) (a) Thorpe-Ingold effect: Beesley, R. M.; Ingold, C. K.; Thorpe, J. F.
J. Chem. Soc. 1915, 107, 1080–1106. Ingold, C. K. J. Chem. Soc. 1921, 119,
305–329. Ingold, C. K.; Sako, S.; Thorpe, J. F. J. Chem. Soc. 1922, 121, 1117–
1198. (b) Extended concept usually referred to as a gem-dimethyl or gem-dialkyl
effect: Hammond G. In Steric Effects in Organic Chemistry; Newman, M. S.,
Ed.; Wiley: New York, 1956; pp 462-470; Smith, S. W.; Newman, M. S. J. Am.
Chem. Soc. 1968, 88, 1253–1957.
(13) Some 2-chloroethylamines were preparedswhen the corresponding
hydrochlorides were not commercially availablesby reacting R-chloroacetyl
chloride with a dialkylamine in the presence of Et3N to yield the corresponding
R-chloroamide, followed by reduction with BH3 · Me2S to the ꢀ-chloroamine
(Supporting Information). This alternative avoids the use of toxic oxirane.
(14) Non-commercial epoxides were prepared in 74-85% yield by m-CPBA
epoxidation of the corresponding terminal alkenes.
(24) Similar observations were reported by Sakurai and Izumi during the
oxidation of N-aryl nitrogen mustards with peracetic acid (ref. 21).
(25) For examples of ortho-oxidations during oxidative degradation of N,N-
disubstituted anilines (Boyland-Sims oxidation), as well as during side reactions
of the corresponding N-oxides, see: Boyland, E.; Manson, D.; Sims, P. J. J. Chem.
Soc. 1953, 3623–3626. Huisgen, R. Chem. Ber. 1959, 92, 3223–3236. Srinivasan,
C.; Perumal, S.; Arumugam, N. J. Chem. Soc., Perkin Trans. 2 1985, 1855–
1858. Behrman, E. J. J. Chem. Soc., Perkin Trans. 1 1992, 3, 305–306.
(15) Fujiwara, M.; Imada, M.; Babaand, A.; Matsuda, H. Tetrahedron Lett.
1989, 30, 739–742, If R2 ) alkyl, only 4 is obtained. However, if R2 is an aryl
on an electron-withdrawing group such as CO2R, mixtures contaminated with
the minor regioisomer 5 are obtained. The introduction of such groups would
however be detrimental in our synthesis, since in the later steps it would favor
rearrangement via the 1,2-oxazetidinium intermediate 10.
4230 J. Org. Chem. Vol. 73, No. 11, 2008

SCHEME 2. Elimination of trans-Cyclohexyl ꢀ-Chloroamine
N-Oxide vs Its Cis Analogue
FIGURE 2. Enamine N-oxides prepared (1aa, 1ca, 1da-dd, and 1dx)
or unsuccessfully targeted (1de, 1ea-ga, and 1ef) during this work.
salts 8da and 8dd into alkoxyamine 12da (67%) and a mixture
of 11dd (10%) and 12dd (58%), respectively.
The choice of the leaving group appears crucial in this
chemistry: due to their higher propensity to substitution, iodides,
mesylates and tosylates are expected to be poorly suited leaving
groups, favoring formation of intermediate 10. Likewise,
oxidation of ꢀ-(arylthio)amines to generate ꢀ-(arylsulfony-
l)amines N-oxides was shown to be an equally inadequate
strategy: fragmentation via a different pathway generates
vinylsulfones and free amines.26
Spectroscopic differentiation between rearranged products 11/
12 and the actual ꢀ-chloroamine N-oxides 827 is often difficult
but can be achieved by comparing the NMR signals of the alkyl
groups adjacent to the N-oxide functionality: in our experience,
for the N-oxides 8, the 1H NMR signals (typically δ )
3.20-4.80) are 1-1.5 ppm and the 13C NMR signals (typically
δ ) 58.0-70.0) are 10-15 ppm higher than that of the
corresponding amine.
Direct treatment of the crude salts 8 (containing a slight
excess, 1.2-1.7 equiv,of stabilizing m-ClPhCO2H) with excess
KOBut in THF (0 °C to rt) leads to direct clean formation of
the enamine N-oxides 1. Upon scale-up of the deprotonation,
reverse addition (i.e., addition of a THF solution of 8 to a
suspension of excess KOBut in THF) was necessary to prevent
competing “Owari-type” rearrangement. Removal of the THF
under vacuum and CH2Cl2 extraction of the residue leads
directly to 1. This constitutes an attractive simplification over
Richmond’s original procedure (avoiding aqueous workup of
8-like salts, subsequent water removal and drying, and final
sublimation of 1). Figure 2 shows the range of enamine N-oxides
prepared by the new method.
Attempts to apply these conditions of HX elimination for the
preparation of 1fa (R1 ) CH2Ph, R2 ) H) were unsuccessful,
leading only to a complicated mixture. We believe that
deprotonation at the benzylic position induces decomposition.
In most cases, the pure enamine N-oxides obtained during
this work were crystalline, if highly hygroscopic, compounds.
From an NMR-spectroscopic point of view, the olefinic part of
the enamine N-oxides 1 does not clearly indicate the presence
of an electron-withdrawing group (1H: δR ) 6.37-6.49, δꢀ )
5.19-6.11; 13C: δR ) 137.0-148.3, δꢀ ) 107.8-139.4).
However, the NMR criteria indicated above for the salts 8
provides a good indication of the presence of the desired N-oxide
functionality: for the alkyl substituents around the N-oxide, δH
) 3.18-3.48 and δC ) 60.0-69.7. In the case of 1dd, this
connectivity was secured by the first X-Ray crystallographic
structure of an enamine N-oxide (see the Supporting Informa-
tion). Despite their hygroscopicity, and in contrast with earlier
reports, exposure of these compounds to water did not result in
any substantial hydrolytic degradation.
For the preparation of a cyclohexenyl derivative such as
1dx, our assumption of an E2 elimination would require that
only cis-substituted precursors were viable (such as cis-8dx,
Scheme 2).
In the original literature,7 the relative stereochemistry of the
cyclic precursor ꢀ-chloro species used for 1 is not clear. We
have explicitly prepared both relative stereoisomers in a racemic
series starting from the pyrrolidine-opened epoxide product
trans-4dx. Treatment of trans-4dx with thionylchloride afforded
the trans ꢀ-chloro species 7dx via the aziridinium 6dx. After
oxidation to trans-8dx, all attempts to engender elimination to
1dx failed, and only decomposition processes akin to the
formation of 10 were observed. In support of this notion, the
tosylate of cis-4dx N-oxide failed to participate in the desired
elimination (with either KOBut/THF or DBU/CH2Cl2). Cyclo-
hexyl cis related ꢀ-chloroamines are a rare class of compounds.
We prepared cis-7dx by preventing aziridinium formation
through N-benzylation of trans-4dx with BnCl. Subsequent
(26) Griffin, R. J.; Henderson, A.; Curtin, N. J.; Echalier, A.; Endicott, J. A.;
Hardcastle, I. R.; Newell, D. R.; Noble, M. E. M.; Wang, L.-Z.; Golding, B. T.
J. Am. Chem. Soc. 2006, 128, 6012–6013.
(27) Analytical features of the salts 8: (a) EI-MS: loss of the oxide is often
observed, hence [M-O]+• as a main peak; ESI-MS: [M + H]+ is the main peak,
confirming the expected formula. (b) Low chromatographic mobility on SiO2
(elution with CH2Cl2/MeOH up to 70/30), whereas the corresponding products
of the “Owari” rearrangement were eluted with Petrol-Et2O mixtures. (c) TLC
stains: 10% Ce(SO4) in 15% aq H2SO4: blue spots; 2,3,5-triphenyltetrazolium
chloride (TTC), 5 wt % in i-PrOH: pink spots.
(28) Analytical features of enamines N-oxides 1: (a) MS: in ESI, small signals
matching [M + H]+ and [M + Na]+, yet the main signal is [2M + H]+ indicating
propensity to form H-bridged homoconjugate dimers in solution. (b) Low
chromatographic mobility on SiO2 (elution with CH2Cl2/MeOH up to 70/30); c)
TLC stains: Ce(SO4): blue spots, TTC in i-PrOH: pink spots.
J. Org. Chem. Vol. 73, No. 11, 2008 4231

Ar-6), 128.7 and 128.6 (C-2′′/C-6′′ and C-3′′/C-5′′), 126.4 (C-4′′),
73.0 (C-1), 69.5, 67.2 (C-2′ and C-5′), 55.1 (C-2), 38.7 (C-4), 32.1
(C-3), 22.3, 20.6 (C-3′ and C-4′); HRMS calcd for C14H21ClNO+
([M + H]+) m/z 254.1306, found 254.1304.
inversion of trans-13 with SOCl2 and removal of the benzyl
protecting group without further purification afforded the desired
cis-7dx but in low yield. Nevertheless, cis-7dx oxidized
smoothly with m-CPBA to afford cis-8dx, which did eliminate
smoothly with ButOK in THF to fashion 1dx.
In conclusion, we have presented user-friendly conditions for
the synthesis of enamine N-oxides and brought new evidence
of the mechanism of “Owari-type” rearrangements.
Enamine N-Oxides by Dehydrochlorination of ꢀ-Chloroam-
ine N-Oxides. To an ice-cold suspension of KOBut (10.0 g, 3 equiv)
in THF (0.6 M) was added dropwise a solution of the m-
chlorobenzoic acid salt of ꢀ-chloroamine N-oxide 8dd (13.3 g, 28.8
mmol) in THF (0.2 M). The mixture was stirred and allowed to
warm to room temperature. After 7 h, the THF was removed in
vacuo. The resulting solid was triturated in CH2Cl2/MeOH (80/20)
and filtered over a short plug of neutral alumina. Concentration in
vacuo and chromatography (SiO2, CH2Cl2/MeOH from 95/5 to 70/
30) afforded 1dd (3.91 g) as a crystalline solid that became oily if
exposed to atmospheric moisture: yield 67%; IR (CHCl3) ν ) 2975,
2465, 1720, 1680, 1660, 1625, 1605, 1560, 1495, 1455, 1380, 1265,
1240, 1030, 945 cm-1; 1H NMR (CDCl3, 500 MHz) δ ) 7.29-7.24
(m, 2′′-H/6′′-H), 7.19-7.13 (m, 3′′-H/5′′-H, 4′′-H), 6.49 (dt, J )
13.3, 7.5, 2-H), 6.03 (dt, J ) 13.3, 1.4, 1-H), 3.42-3.36 (m, 2H)
and 3.29-3.22 (m, 2H) (2′-/5′-H2), 2.74 (t, J ) 7.5, 4-H2), 2.42
(qd, J ) 7.5, 1.4, 3-H2), 2.55-2.46 (m, 2H) and 2.01-1.92 (m,
3H) (3′-H2/4′-H2); 13C NMR (CDCl3, 101 MHz) δ ) 140.9 (C-
1′′), 139.4 (C-1), 128.5 and 128.4 (C-2′′/C-6′′ and C-3′′/C-5′′), 126.2
(C-4′′), 125.3 (C-2), 69.6 (C-2′/C-5′), 35.0 (C-4), 30.5 (C-3), 21.8
(C-3′/C-4′); HRMS calcd for C14H20NO+ ([M + H]+) m/z 218.1539,
found 218.1556.
Experimental Section
N-Oxidation of Tertiary ꢀ-Chloroamines with m-CPBA:
1-(2-Chloro-4-phenylbutyl)pyrrolidine N-Oxide, 3-Chloroben-
zoic Acid Salt (8dd). To a solution of dried m-CPBA (85-95 wt
%, 1.1-1.5 equiv) in CH2Cl2 (0.2 M), cooled with an ice-brine
bath, was added the amine slowly (1 mL/min) via syringe. The
mixture was then stirred while being allowed to warm to room
temperature. After consumption of the starting amine was complete,
the solvent was removed under reduced pressure to yield quantita-
tive amounts of the m-chlorobenzoic acid salt (8) of the ꢀ-chloro-
amine N-oxide as a brown oil. In all cases, excess m-chlorobenzoic
acid was the main contaminant; the excess m-chlorobenzoic acid
1
was assessed by H NMR to be between 0.2 and 0.7 equiv. The
highly hygroscopic nature of these salts (8) limits their suitability
for combustion analysis: IR (CHCl3) ν ) 3000, 1700, 1570, 1495,
1
1455, 1430, 1365, 1290, 1255, 1145, 1070, 905 cm-1; H NMR
Acknowledgment. We thank the EPSRC for support through
grant EP/025080/1 and EP/E030092/1. SW acknowledges
valuable input from the COST D40 Innovative Catalysis
Programme.
(CDCl3, 400 MHz) δ ) 8.03 (t, J ) 1.6, Ar2-H), 7.92 (dt, J ) 7.8,
1.6, Ar6-H), 7.45 (dt, J ) 7.8, 1.6, Ar4-H), 7.33 (t, J ) 7.8, Ar5-
H), 7.26-7.20 (m, 2′′-H, 6′′-H), 7.20-7.14 (3′′, 4′′, 5′′-H), 4.90
(dddd, J ) 9.1, 8.0, 4.6, 1.3, 2-H), 4.86 (dd, J ) 13.9, 1.3, 1-H),
4.51-4.39 (m, 1H) and 4.11-3.94 (m, 1H) and 3.45-3.34 (m, 2H)
(2′-H2, 5′-H2), 3.52 (dd, J ) 13.9, 8.0, 1-H), 2.91 (ddd, J ) 13.8,
9.9, 5.0, 4-H), 2.81 (ddd, J ) 13.8, 9.5, 6.5, 4-H), 2.55-2.42 (m,
2H) and 2.14-2.04 (m, 2H) (3′-H2, 4′-H2), 2.25 (dddd, J ) 13.9,
9.9, 6.5, 4.6, 3-Ha), 2.15 (dddd, J ) 13.9, 9.5, 9.1, 5.4, 3-H); 13C
NMR (CDCl3, 101 MHz) δ ) 170.0 (ArCO2), 140.2 (C-1′′), 135.8,
134.2 (Ar-1, Ar-3), 131.7, 130.0, 129.4, 127.9 (Ar-2, Ar-4, Ar-5,
Supporting Information Available: Experimental details and
characterization data for all compounds. NMR spectra (1H and
13C NMR) for all new compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
JO8002166
4232 J. Org. Chem. Vol. 73, No. 11, 2008