Control of regioselectivity by the lone substituent through steric and electronic effects in the nitrosoarene ene reaction of deuterium-labeled trisubstituted alkenes

Article abstract of DOI:10.1021/jo026198i

For the ene reaction of 4-nitronitrosobenzene (ARNO) with a variety of primary and secondary lone alkyl-substituted substrates, the twix/twin regioselectivity is constant at about 85:15. In contrast, for the lone tert-butyl group and for lone aryl substituents, the twix regioisomer is obtained exclusively. These regioselectivities have been rationalized in terms of steric interactions and coordination between the enophile and the substrates in the transition states of the first reaction step.

Full text of DOI:10.1021/jo026198i

Con tr ol of Regioselectivity by th e lon e Su bstitu en t th r ou gh Ster ic
a n d Electr on ic Effects in th e Nitr osoa r en e En e Rea ction of
Deu ter iu m -La beled Tr isu bstitu ted Alk en es
Waldemar Adam,*,† Oliver Krebs,*,† Michael Orfanopoulos,‡ and Manolis Stratakis‡
Institut fu¨r Organische Chemie, Universita¨t Wu¨rzburg, Am Hubland, D-97074 Wu¨rzburg, Germany, and
Department of Chemistry, University of Crete, P.O. Box 1470, 71409 Heraklion, Crete, Greece
adam@chemie.uni-wuerzburg.de
Received J uly 16, 2002
For the ene reaction of 4-nitronitrosobenzene (ArNO) with a variety of primary and secondary lone
alkyl-substituted substrates, the twix/twin regioselectivity is constant at about 85:15. In contrast,
for the lone tert-butyl group and for lone aryl substituents, the twix regioisomer is obtained
exclusively. These regioselectivities have been rationalized in terms of steric interactions and
coordination between the enophile and the substrates in the transition states of the first reaction
step.
SCHEME 1. Tr a n sition Sta tes TStwix a n d TStwin for
F or m a tion of th e Azir id in e N-Oxid e In ter m ed ia te
in th e Nitr osoa r en e En e Rea ction w ith
In tr od u ction
Experimental work on the deuterium isotope effects
in the ene reaction of nitrosoarene (ArNO) with tetra-
methylethylene disclosed that the same mechanism
operates as in the ene reaction of singlet oxygen (1O2)
and triazolinedione (TAD), namely, through a three-
membered-ring intermediate, for ArNO, the aziridine
N-oxide.1 Nevertheless, the regioselectivity of these three
isoelectronic enophiles shows large differences. Whereas
1O2 favors hydrogen abstraction on the more crowded side
of the olefin (cis effect)2 and triazolinedione (TAD) prefers
to react at the more crowded end (gem effect),3 the
enophile 4-nitronitrosobenzene (ArNO) undergoes the
recently discovered highly regioselective ene reaction at
the twix position (skew effect) of the olefin (Scheme 1).4,5
Deu ter iu m -Ster eola beled Tr isu bstitu ted Alk en es
tion at the twin position is small, and no ene product is
observed from lone abstraction. The skew effect was
rationalized in terms of steric interactions in the transi-
tion states TStwix and TStwin, which lead to the respective
ene products along the enophilic trajectory.4 In the TStwin
structure, the aryl substituent of the ArNO enophile is
located at the more crowded and, thus, the sterically more
encumbered side of the alkene, whereas in TStwix, the aryl
group points to the free corner of the alkene and the steric
repulsions are reduced (skew attack). Consequently,
hydrogen abstraction takes place preferentially from the
twix substituent, as shown in Scheme 1.
To determine the influence of lone substitution on the
reactivity of the double bond, the relative rate constants
(krel) were determined by means of intermolecular com-
petition experiments for a variety of lone-substituted
2-methyl-2-butenes (Figure 1).4c An ethyl group at the
lone site accelerates the reaction rate, but a bulky
substituent such as the tert-butyl group decreases the ene
reactivity significantly.
To distinguish between twix and twin hydrogen ab-
straction in the ene products, one of the methyl groups
must be deuterium-labeled (Scheme 1). No significant
kinetic isotope effects have been observed, and the
regioselectivity is, within the error range (5%), the same
for twix and twin labeled olefins.4a The hydrogen abstrac-
† Universita¨t Wu¨rzburg, http://www-organik.chemie.uni-wuerzburg.de/
ak_adam.
‡ University of Crete, http://www.chemistry.uoc.gr.
(1) Seymour, C. A.; Greene, F. D. J . Org. Chem. 1982, 47, 5226-
5227.
(2) (a) Orfanopoulos, M.; Grdina, M. J .; Stephenson, L. M. J . Am.
Chem. Soc. 1979, 101, 275-276. (b) Schulte-Elte, K. H.; Muller, B. L.;
Rautenstrauch, B. Helv. Chim. Acta 1978, 61, 2777-2783. (c) Stratakis,
M.; Orfanopoulos, M.; Chen, J . S.; Foote, C. S. Tetrahedron Lett. 1996,
37, 4105-4108.
(3) Cheng, C. C.; Seymour, C. A.; Petty, M. A.; Greene, F. D.; Blount,
J . F. J . Org. Chem. 1984, 49, 2910-2916.
(4) (a) Adam, W.; Bottke, N.; Krebs, O. J . Am. Chem. Soc. 2000,
122, 6791-6792. (b) Adam, W.; Bottke, N.; Krebs, O. Org. Lett. 2000,
2, 3293-3296. (c) Adam, W.; Bottke, N.; Krebs, O.; Engels, B. J . Am.
Chem. Soc. 2001, 123, 5542-5548.
(5) For the differentiation of the regioisomeric ene products derived
from twix and twin hydrogen abstraction, deuterium-labeling of one
of the positions is necessary, unless the substrate bears different
substituents.
To probe whether steric interactions of lone substitu-
ents control the twix/twin regioselectivity of the ni-
trosoarene ene reaction, it was decided to use an assort-
10.1021/jo026198i CCC: $22.00 © 2002 American Chemical Society
Published on Web 10/24/2002
J . Org. Chem. 2002, 67, 8395-8399
8395

Adam et al.
SCHEME 3. Syn th esis of twin
Deu ter iu m -Ster eola beled Alk en es 1
F IGURE 1. Relative ene reactivity of lone-substituted 2-meth-
yl-2-butenes with nitrosoarene.
tion of ArNO with styrenes that bear lone aryl substit-
uents. Since the nitrosoarene enophile is more prone to
1
steric and electronic control than O2 and TAD, a highly
selective process might be anticipated for ArNO. The
results of the lone-substituent influence in the ene
reaction of ArNO with the deuterium-labeled alkenes 1
are reported here, and their mechanistic consequences
presented.
F IGURE 2. Expected steric influence of the lone substituent
on twix/twin regioselectivity in the ArNO ene reaction with
trisubstituted alkenes.
SCHEME 2. Coor d in a tion Effects in 1O2 a n d MTAD
En e Rea ction s of Alk en es w ith Ar yl Su bstitu en ts
a t th e Lon e P osition
Resu lts
The twin deuterium-stereolabeled alkenes with various
lone substituents were prepared by stereoselective Wittig
coupling between methyl 2-triphenylphosphoranylidene-
propanoate and the appropriate aldehydes, followed by
reduction of the E-configured esters with LiAlD4 (Scheme
3).9 The resulting alcohols were mesylated7b or, in the
case of the styrenes, transformed into the corresponding
chlorides7b for the subsequent reduction with LiAlD4 to
the twin deuterium-labeled alkenes 1.
The ene reaction of 4-nitronitrosobenzene with the twin
deuterium-labeled lone alkyl- and aryl-substituted al-
kenes 1 gave the expected regioisomeric ene products
from twix and twin hydrogen abstraction, which were
identified and quantified by means of 1H NMR spectros-
copy (Table 1). To confirm the authenticity of the ene
products, the ene reaction with the unlabeled alkenes
1-d 0 was conducted on the preparative scale and the
unlabeled ene products were isolated and fully character-
ized (see Supporting Information).
For the lone alkyl-substituted alkenes 1a -f (entries
1-6), the twix selectivity is quite constant at about 85 (
3%, while for the tert-butyl derivative 1g (entry 7), the
twix regioisomer is heavily favored (> 95%). For the ethyl
substrate 1b (entry 2), about 16% of the ene product
(hydroxylamine) is oxidized in situ by the enophile to the
corresponding nitrone and hydrolyzed to its ketone.10
Fortunately, the regioselectivity is not affected by this
side reaction, since the two regioisomeric ene products
are equally likely oxidized by the nitroso enophile. This
was confirmed by a control experiment for the regio-
isomeric ene products derived from the ene reaction of
E- and Z-3-methyl-2-pentene.11
ment of twin deuterium-labeled alkenes with different
lone substituents (Figure 2). Compared to the parent
2-methyl-2-butene substrate, enhanced twix regioselec-
tivity is expected as a result of steric obstruction by the
lone group along the TStwin approach.
Generally, in contrast to the ArNO enophile, a lower
regioselectivity is observed in the ene reaction of 1O2 and
TAD with simple alkenes.2,6 However, in the case of
styrenes with the aryl group at the lone position, en-
hanced twix regioselectivity was reported.7 This observa-
tion was rationalized in terms of coordination between
the enophile and the incipient delocalized positive charge
of the lone aryl group in TStwix, which promotes twix
hydrogen abstraction (Scheme 2). Moreover, some [4 +
2] cycloaddition occurs with these styrenes,7,8 which is
the consequence of the delocalized positive charge in the
lone aryl group of the zwitterionic intermediates (ZI),
especially when stabilized by electron-rich aryl groups
(Scheme 2, R ) OMe). Such zwitterionic intermediates
have been proposed to be responsible for the loss of
stereoselectivity in the [4 + 2] cycloaddition of MTAD to
the twix-CD3-labeled â,â-dimethyl-(p-methoxy)styrene (R
) OMe).7b
For the styrenes 1h -k with lone aryl substituents, twix
selectivity is also exclusive (>95%). Since the reactivity
of the aryl conjugated double bond in styrenes 1h -k is
lower, their conversion is less than for the alkyl-
substituted cases 1a -g. For all of these alkenes, only the
Of mechanistic interest was the assessment of the
mode selectivity (ene reaction versus [4 + 2] cycloaddi-
tion) and regioselectivity (twix versus twin) in the reac-
(6) (a) Ohashi, S.; Leong, K.; Matyjaszewski, K.; Butler, G. B. J .
Org. Chem. 1980, 45, 3467-3471. (b) Adam, W.; Schwarm, M. J . Org.
Chem. 1988, 53, 3129-3130.
(7) (a) Stratakis, M.; Orfanopoulos, M.; Foote C. S. J . Org. Chem.
1998, 63, 1315-1318. (b) Stratakis, M.; Hatzimarinaki, M.; Froudakis,
G. E.; Orfanopoulos, M. J . Org. Chem. 2001, 66, 3682-3687.
(8) Matsumoto, M.; Kuroda, M. Synth. Commun. 1981, 11, 987-
992.
(9) (a) Stephenson, L. M.; Speth, D. R. J . Org. Chem. 1979, 44,
4683-4689. (b) Ayray, G.; Wong, D. J . J . Labelled Compd. Radio-
pharm. 1978, 14, 935-944.
(10) (a) Adam, W.; Bottke, N. J . Am. Chem. Soc. 2000, 122, 9846-
9847. (b) Knight, G. T.; Pepper, B. Tetrahedron 1971, 27, 6201-6208.
(c) Knight, G. T.; Loadman, M. J . R. J . Chem. Soc. B 1971, 2107-
2112.
(11) Krebs, O. Diplomarbeit, Universita¨t Wu¨rzburg, 2000.
8396 J . Org. Chem., Vol. 67, No. 24, 2002

Control of Regioselectivity by the lone Substituent
TABLE 1. Regioselectivity in th e En e Rea ction of
4-Nitr on itr osoben zen e (Ar NO) w ith Tr isu bstitu ted
Alk en es 1
F IGURE 3. Regioselectivities for the ene reaction of ArNO
and 1O2 with trisubstituted alkenes.
F IGUR E 4. Steric interactions in the cis-configured per-
epoxide-like structures P Ecis of the 1O2 ene reaction.
of comparison, a structural overview of the twin/twix
regioselectivity is given for the ArNO ene reaction with
the simple lone alkyl groups Me (1a ), Et (1b), i-Pr (1e),
and t-Bu (1g), together with the reported data of 1O2
(Figure 3).2a,4,12
It is mechanistically instructive to consider first the
1O2 enophile, for which a regular regiochemical trend is
noted. With increasing steric demand of the lone sub-
stituent, the cis (twix + lone) selectivity decreases. The
very high cis (twix + lone) regioselectivity (cis effect)2a,12a
for 1a and 1b has been attributed to the coordination of
the enophile with the allylic hydrogens in the cis-
configured perepoxide-like structure P Ecis (Figure 4).13a
In substrate 1e, hydrogen abstraction at the lone
isopropyl group is hindered, since unfavorable 1,3-allylic
strain (1,3A) builds up during the hydrogen abstraction
step in P Ecis and, thus, the lone selectivity is diminished.
a
b
Normalized to 1 equiv of alkene. Determined by 1H NMR
spectroscopy; error ca. 5% of the stated value, mass balance > 95%.
c Taken from ref 4a. Values corrected for the stereochemical
d
purity of the alkenes (see Supporting Information). e The E/ Z ratio
of 1g was 85:15 ((5%) and the observed twix/twin product ratio
was 85/15 ((5%).
ene reaction is observed; no [4 + 2] cycloaddition occurs,
as in the reaction of these styrenes with TAD and O2.7,8
1
Discu ssion
Analogous to the previously studied trisubstituted
alkenes,4 the nitrosoarene ene reaction displays a high
twix selectivity for all lone-substituted derivatives 1. For
the alkyl-substituted cases 1a -f (the tert-butyl derivative
1g falls out of line, since the twix regioisomer is exclu-
sively produced, which provides valuable mechanistic
details as we shall see later on), the twix ene products is
the main regioisomer (85 ( 3%), whereas for the lone
aryl-substituted styrenes 1h -k only twix ene product is
obtained. This implies that different steering mechanisms
apply in these two types of substrates: While only steric
effects dictate the twix regioselectivity (85 ( 3%) for the
lone alkyl-substituted substrates 1a -f, for the aryl-
substituted derivatives 1h -k additional coordination of
the enophile by the lone aryl ring enhances twix selectiv-
ity (>95%). This becomes apparent on comparing the
substrates 1f (Rlone ) cyclohexyl) and 1j (Rlone ) phenyl),
which possess similar steric but different electronic
properties and exhibit different efficiencies in the regio-
chemical control (Table 1, entries 6 and 10). The mecha-
nistic features of the two sets of lone-substituted alkenes
shall be discussed separately.
1
For the reaction of O2 with substrate 1g, the sterically
obstructing lone tert-butyl group forces hydrogen abstrac-
tion at the twin side, i.e., the “anti-cis effect”12c operates.
Steric encumbrance by the lone tert-butyl group on the
cis side hinders the formation of P Ecis for the 1g substrate
(Figure 4) and twin reactivity dominates (Figure 4).
In contrast to 1O2 (cis effect),12a for the sterically more
encumbered ArNO enophile (skew effect),4 the aryl group
is sterically more demanding than the terminal oxygen
atom of the nitroso group. Thus, for all of the substrates
1a -g, the steric interactions in the cis-configured transi-
tion structure TStwix are less severe than in the trans-
configured TStwin (Figure 5); consequently, the twix
regioisomers are formed as major ene products.
The fact that about the same (ca. 85:15) twix/twin ratio
for the 1a -f substrates is observed may be rationalized
(12) (a) Orfanopoulos, M.; Stratakis, M.; Elemes, Y. J . Am. Chem.
Soc. 1991, 113, 3180-3181. (b) Orfanopoulos, M.; Stratakis, M.
Tetrahedron Lett. 1995, 36, 4291-4294.
(13) (a) Frimer, A. A.; Bartlett, P. D.; Boschung, A. F.; J ewett, J . G.
J . Am. Chem. Soc. 1977, 99, 7977-7986. (b) Foote, C. S. Acc. Chem.
Res. 1968, 1, 104-110. (c) Hurst, J . R.; McDonald, J . D.; Schuster, G.
B. J . Am. Chem. Soc. 1982, 104, 2065-2067. (d) Hurst, J . R.; Wilson,
S. L.; Schuster, G. B. Tetrahedron 1985, 41, 2191-2197.
Regioselectivity in th e Nitr osoa r en e En e Rea c-
tion w ith Alk yl-Su bstitu ted Alk en es 1a -g. For ease
J . Org. Chem, Vol. 67, No. 24, 2002 8397

Adam et al.
TABLE 2. Mod e Selectivity (En e ver su s [4 + 2]) a n d
Regioselectivity (twix ver su s twin ) in th e En e Rea ction
of Ar NO, 1O2, a n d MTAD w ith Styr en es 1h -k
F IGURE 5. Steric interactions of the lone substituent in the
transition structures TStwix and TStwin for the ArNO ene
reaction
selectivity
mode
[4 + 2]d
regio
enophile
X ) Y
styrene
R
ene
twix
twin
a
1O2
1j
1h
1h -k
H
MeO
all
34
60
>95
66
40
<5
63
>95
>95
37
<5
<5
MTADb
ArNOc
F IGURE 6. Enophile/substrate transition structures TStwix
and TStwin in the ArNO ene reaction with the styrenes 1h -k .
a
b
d
Reference 7a, 8. Reference 7b. c This work. The cycload-
ducts undergo a second addition of the enophile ([4 + 2] for 1O2,
mainly ene for MTAD).
in terms of favored conformations that the lone substitu-
ent may acquire, which minimize steric interaction with
the aryl group of the enophile in TStwin. For the primary
and secondary lone substituents in the substrates 1a -f
(Table 1, entries 1-6), the hydrogen atom may assume
the favorable (minimal 1,3-allylic strain) inside confor-
mation in TStwin, whereas the remaining R1 and R2
groups are located at the periphery, as displayed in
Figure 5. Since the R1 and R2 groups at the lone
substituent are at the periphery in the TStwin structure
for the substrates 1a -f, insignificant differences are
encountered in the steric interaction between the alkene
and the enophile, and consequently, similar twix/twin
ratios are obtained. In contrast, for the substrate 1g with
the tertiary lone substituent, namely, the tert-butyl
group, no conformational arrangement is possible in
which the methyl groups at the lone substituent avoid
1,3-allylic strain (one methyl group is always inside!).
Thus, for substrate 1g, the TStwin encounter is sterically
prohibitive and the twix regioisomer is formed exclusively
(Table 1, entry 7).
Also remarkable for the ArNO ene reaction with these
substrates 1h -k is the absence of [4 + 2] cycloaddition,
especially for the electron-rich (R ) OMe) derivative 1h
(Table 1, entry 8). In contrast, a low mode selectivity (ene
reaction versus [4 + 2] cycloaddition) is observed in the
reaction of 1O2 with the 1j and MTAD with the 1h
substrates, since considerable amounts of [4 + 2] cy-
cloadducts are formed (Table 2).7
Responsible for the [4 + 2] cycloaddition is the zwit-
terionic intermediate ZI (see also Scheme 2), which is
stabilized by aryl conjugation. That no [4 + 2] cyclo-
Regioselectivity in th e Nitr osoa r en e En e Rea c-
tion w ith Styr en es 1h -k . In the ene reaction between
ArNO and the styrenes 1h -k with the lone aryl substi-
tutents, the twix regioisomer is formed exclusively (Table
1). In contrast to the nitrosoarene ene reaction with the
lone alkyl-substituted alkenes 1a -g, an additional co-
ordination of the enophile with the lone aryl group may
be controlling the regioselectivity of the aryl-substituted
addition is observed for ArNO is presumably due to steric
interactions between the aryl group of the enophile and
the twin substituent of the styrene (skew effect), as shown
in the zwitterionic intermediate (ZI) for the ArNO ene
reaction. The twin substituent pushes the aryl group to
the unsubstituted corner of the double bond, such that
twix hydrogen abstraction dominates along the skew
trajectory of the enophile.
1
substrate 1h -k . In analogy to the ene reactions of O2
In summary, for all of the lone alkyl-substituted (1a -
g) and the lone aryl-substituted (1h -k ) substrates, the
ArNO results only in ene reaction. For the styrenes 1h -k
and TAD with these substrates, the aryl group acquires
a partial positive charge during the enophilic attack of
the conjugated double bond. This positively charged ring
coordinates effectively with the negatively charged ter-
minal oxygen atom of the ArNO enophile, and hence, the
cis-configured transition structure TStwix for twix hydro-
gen abstraction is stabilized (Figure 6). Additionally, the
TStwin species is sterically disfavored, as discussed above
for the alkenes 1a -g. In the case of the styrenes 1h -k
this applies all the more, because the aryl group in TStwin
tends to be coplanar with the double bond for optimal
conjugation, which increases the steric obstruction on the
cis side.
1
this exclusive mode selectivity is in contrast to O2 and
TAD, for which both ene reaction and [4 + 2] cycloaddi-
tion are observed. Also the regioselectivity of these lone
aryl-substituted substrates with ArNO contrasts that of
1O2 and TAD, since for ArNO the twix regioisomer is
obtained exclusively. This has been rationalized in terms
of cooperative effects between the steric and electronic
interactions of the ArNO enophile with the lone aryl
group, which favor the skew arrangement for exclusive
twix hydrogen abstraction. Coordination with the lone
substituent is absent in the alkyl-substituted substrates
8398 J . Org. Chem., Vol. 67, No. 24, 2002

Control of Regioselectivity by the lone Substituent
1a -g and as much as about 15% twin hydrogen abstrac-
tion is observed, except in the case of the tert-butyl
derivative 1g, for which again twix abstraction is exclu-
sive. The constancy of the twix/twin ratio (ca. 85:15) for
the diverse primary and secondary lone substituents is
interpreted in terms of the favorable inside conformation
for minimal 1,3-allylic strain that the allylic lone hydro-
gen atom(s) may assume, such that the remaining groups
of the lone substituent are located at the periphery and
do not exert significantly different steric effects on the
attacking enophile. For the substrate 1g with the lone
tert-butyl group, the exclusive twix regioselectivity is due
to the fact that the inside conformational arrangement
is not possible and the massive steric obstruction by the
lone substituent forces the enophile to attack exclusively
along the skew trajectory. Comparison of the regioselec-
tivities for lone-substituted substrates has been instruc-
tive to draw a detailed mechanistic picture for heteroatom
ene reactions.
Ack n ow led gm en t. We thank the “Deutsche Aka-
demische Austauschdienst”, the “Deutsche Forschungs-
gemeinschaft”, and the Greek Ministry of Education for
generous financial support.
Su p p or tin g In for m a tion Ava ila ble: Complete experi-
mental procedures. This material is available free of charge
via the Internet at http://pubs.acs.org.
J O026198I
J . Org. Chem, Vol. 67, No. 24, 2002 8399