Hydroxy-group directivity in the nitroso ene reaction: Diastereo- and regioselective amination of chiral allylic alcohols [6]

Full text of DOI:10.1021/ja001752w

9846
J. Am. Chem. Soc. 2000, 122, 9846-9847
Table 1. Product Studies of the Ene Reaction of
p-Nitronitrosobenzene (ArNO) with Allylic Substrates 1
Hydroxy-Group Directivity in the Nitroso Ene
Reaction: Diastereo- and Regioselective Amination of
Chiral Allylic Alcohols
Waldemar Adam and Nils Bottke*
Institut fu¨r Organische Chemie, UniVersita¨t Wu¨rzburg
Am Hubland, D-97074 Wu¨rzburg, Germany
ReceiVed May 24, 2000
ReVised Manuscript ReceiVed July 12, 2000
olefin
(equiv)
convn
mb
diastereoselectivityb
[%]a,b [%]a,b
2:3b
threo:erytho
The ene reactions of singlet oxygen1 and triazolinedione (TAD)2
with 1,3-allylically strained chiral substrates lead to threo-
configured ene products in high diastereoselectivities, a conse-
quence of the hydroxy-group directivity.3 If such stereocontrol
were to operate for the isoelectronic nitroso enophile, an attractive
and convenient synthetic methodology would be available for the
diastereoselective amination of chiral allylic alcohols; however,
the ene reaction of such olefins appears not to have been examined
so far.4 This is presumably due to the undesirable side reactions
known for the nitroso ene reaction, in which the resulting
hydroxylamine ene products are in situ further oxidized.5 To
circumvent these disadvantages, we recently established a new
nitroso enophile, namely p-nitronitrosobenzene (ArNO), which
affords persistent ene products.6 Indeed, presently we demonstrate
that for this nitrosoarene enophile the hydroxy-group directivity
operates effectively when 1,3-allylically strained chiral allylic
alcohols are employed. The desired aminated ene products have
been obtained in high diastereoselectivity and regioselectivity,
as well as in good yield.
1
2
3
4
5
6
7
8
9
1a (1)
1a (2)
1a (1)c
1a (1)d
1b (1)
1b (2)
1b (4)
1c (1)
1c (2)
81
35
56
73
45
18
<5
55
17
<5
51
16
>95
>95
79
67
83
>95
>95
85
>95
>95
61
76:24
92:8
79:21
93:7
79:21
83:17
94:6
51:49
72:28
85:15
> 95:5
>95:5
92:8
>95:5
69:31
66:34
87:13
80:20
77:23
79:21
64:36
60:40
>95:5
>95:5
10 1c (4)
11 Z-1d (1)
12 Z-1d (2)
92
a Conversion and mass balance (mb) relative to the allylic alcohol.
b Determined by 1H NMR spectroscopy, error (5% of the stated value.
c Solvent was CD3OD. d Solvent was d6-DMSO.
Scheme 1. 1,2-Allylic Strain in the Nitrosoarene Ene
Reaction with the Allylic Substrate E-1e
The product studies of the ene reactions were conducted on
the NMR scale, all unknown products were isolated from
preparative runs and fully characterized. The product ratios and
diastereoselectivities were determined directly on the crude
reaction mixture from the peak areas of characteristic signals in
1
the H NMR spectra. The results are summarized in Table 1.
The allylic acohol 1a was converted with an equimolar amount
of the nitrosoarene in the nonpolar d-chloroform with high
(92:8) diastereoselectivity (entry 1) predominantly to the threo-
configured hydroxylamine (2S*,3S*)-2a. In d4-methanol (69:31,
entry 3) and d6-DMSO (66:34, entry 4) significantly lower
diastereoselectivities were observed. Evidently, hydrogen bonding
between the 1,3-allylically strained substrate and the nitrosoarene
enophile is responsible for the pronounced threo selectivity. In
the protic methanol and the polar DMSO, the substrate/enophile
hydrogen bonding is suppressed through competitive inter-
molecular interactions with the solvent.
alcohol 1a in CDCl3. Thus, for the acetate 1b (77:23, entry 7)
and for the methyl ether 1c (60:40, entry 10), the threo selectivity
is substantially lower than that for the allylic alcohol 1a (92:8,
entry 1). Again, these results clearly establish that the hydroxy
group directs the nitrosoarene ene reaction preferentially to the
threo product through hydrogen bonding.
The stereochemically labeled and 1,3-allylically strained alcohol
Z-1d (entry 12) displayed exclusive (>95:5) threo diastereose-
1
lectivity; the erythro ene product was not detected by H NMR
Also chemical masking of the hydroxy group either by
acetylation as in the ester 1b or methylation as in the ether 1c
corraborate that substrate/enophile hydrogen bonding is at work
in expressing the high threo diastereoselectivity for the allylic
spectroscopy. Under identical conditions, the complementary
diastereomer E-1d is unreactive toward the p-nitronitrosobenzene
enophile. That 1,2-allylic strain exercises no appreciable diaste-
reoselective control was demonstrated by the substrate E-1e
(Scheme 1), for which the threo/erythro ratio is only 64:36. In
contrast, the Z-1e isomer did not undergo the nitrosoarene ene
reaction. The reactivities of the various substrates are for
convenience collected in Figure 1 and underline the recently
established high twix regioselectivity of the nitrosoarene ene
reaction.6
The relative configurations of the ene products were determined
by 1H NMR spectroscopy. For this purpose, the hydroxylamines
2 were cyclized with 2,2-dimethoxypropane under acid catalysis
to the corresponding 1,3,4-dioxazines 4 (Scheme 2). The threo
configuration of these conformationally rigid derivatives was
assessed by the J1,3(H,H) coupling constant for the protons of
the vicinal CH groups (threo 8.8 Hz). This clearly speaks for
their 180° arrangement and establishes the threo configuration.
* Address correspondence to this author. Fax: +49931/888 4756. E-mail:
adam@chemie.uni-wuerzburg.de.
(1) Adam, W.; Nestler, B. J. Am. Chem. Soc. 1993, 115, 5041-5049. Adam,
W.; Bru¨nker, H. G. J. Am. Chem. Soc. 1995, 117, 3976-3982. Adam, W.;
Bru¨nker, H.-G.; Kumar, A. S.; Peters, E.-M.; Peters, K.; Schneider, U.; von
Schnering, H. G. J. Am. Chem. Soc. 1996, 118, 1899-1905.
(2) Adam, W.; Nestler, B.; Pastor, A.; Wirth T. Tetrahedron Lett. 1998,
39(17), 2625-2629. Stratakis, M.; Vassilikogiannakis, G.; Orfanopoulos M.
Tetrahedron Lett. 1998, 39(16), 2393-2396. Gau, A.-H.; Lin, G.-L.; Uang,
B.-J.; Liao, F.-L.; Wang, S.-L. J. Org. Chem. 1999, 64, 2194-2201.
(3) Adam, W.; Wirth, T. Acc. Chem. Res. 1999, 32, 703-710.
(4) Seymour, C. A.; Greene, F. D. J. Org. Chem. 1982, 47, 5226-5227.
Greene, F. D. In Stereochemistry and ReactiVity of Systems containing
π-Electrons; Watson, W. H., Ed.; Verlag Chemie International: Deersfield
Beach, FL, 1983; pp 197-240.
(5) Knight, G. T.; Pepper, B. Tetrahedron 1971, 27, 6201-6208.
(6) Adam, W.; Bottke, N.; Krebs, O. J. Am. Chem. Soc. 2000, 122, 6791-
6792.
10.1021/ja001752w CCC: $19.00 © 2000 American Chemical Society
Published on Web 09/26/2000

Communications to the Editor
J. Am. Chem. Soc., Vol. 122, No. 40, 2000 9847
Scheme 4. Hydroxy-Group Directivity (threo Selectivity) in
the Ene Reaction of Nitrosoarenes with 1,3-Allylically (A1,3
Strained Alcohols
)
Figure 1. The ene reactivity with twix regioselectivity between the
nitrosoarene and the chiral allylic alcohols 1.
Scheme 2. Determination of the threo Configuration for the
Nitrosoarene Ene Products
substituent6 along the skew trajectory is directed by the allylic
hydroxy functionality through hydrogen bonding in the chiral
substrate, which is conformationally aligned on account of 1,3-
allylic strain. The threo transition state is favored over the erythro
one because 1,3-allylic strain is minimized and, therefore, the
threo diastereomer is the main product.
Scheme 3. Formation of the Enone 3a and Azoxybenzene 5
from the Primary Ene Product 2a through Subsequent
Oxidation by the Nitrosoarene
This mechanistic model also accounts for the reactivity behavior
of the chiral substrates in Figure 1. The ene-inactive E-1d
possesses no twix substituent, while for Z-1e the twix site is
occupied by the chiral alcohol functionality and hydrogen bonding
presumably prevents the ene reaction.1,8 The poor diastereo-
selectivity of the E-1e substrate (Scheme 1) is due to the lack of
1,3-allylic strain.
In summary, the nitrosoarene ene reactivity has been tamed
sufficiently such that a highly threo-selective allylic amination
of 1,3-allylically strained substrates has been achieved. In fact,
the extent of the diastereoselectivity for the nitrosoarene ene
reaction is superior to that of the well-studied singlet oxygen and
triazolinedione enophiles.1,2 An additional advantage is that a high
twix regioselectivity operates for the nitrosoarene enophile. This
combination of high threo diastereoselectivity and high twix
regioselectivity offers attractive synthetic opportunities for the
controlled amination of chiral 1,3-allylically strained hydroxy-
functionalized olefins.
In all of these nitrosoarene ene reactions, the enones 3 were
formed (Scheme 3). The amount of enone strongly depended on
the substrate/enophile ratio, i.e., the larger the olefin excess, the
less the amount of the enone that formed (Table 1). For example,
in the case of the methyl ether 1c, the product/enone ratio
increased from 51:49 for an equimolar amount of olefin (entry
8) to 85:15 for 4 equiv (entry 10). Mechanistically pertinent, the
diastereomeric ratio increased from 60:40 for 4 equiv (entry 8)
to 79:21 for 1 equiv (entry 10). The same trend was observed for
the ester substrate 1b (entries 5-7). The enone formation is
attributed to dehydrogenation of the hydroxylamine ene product
1 by the nitrosoarene7 to the nitrone and subsequent hydrolysis
by adventitious water. The released hydroxylamine ArNHOH
condenses in situ with the ArNO to generate the azoxyarene 5,
which was observed in amounts nearly equal to the enone 3.
The threo selectivity is rationalized mechanistically in terms
of the established hydroxy-group directivity, which is the
combination of 1,3-allylic strain and hydrogen bonding (Scheme
4).1,2 Thus, the enophilic attack of the nitrosoarene on the twix
Acknowledgment. For financial support we thank the Deutsche
Forschungsgemeinschaft (Schwerpunktprogramm: “Peroxidchemie”) and
the Fonds der Chemischen Industrie (doctoral fellowship for N.B.).
Supporting Information Available: Complete experimental proce-
dures (PDF). This material is available free of charge via the Internet at
http://pubs.acs.org.
JA001752W
(8) Adam, W.; Peters, K.; Peters, E.-M.; Schambony, S. B. J. Am. Chem.
Soc. 2000, 122, 7610-7611.
(7) Knight, G. T.; Loadman, M. J. R J. Chem. Soc. (B) 1971, 2107-2112.