On the stereoselectivity of alkenoxyl radical 6-exo-trig cyclizations

Article abstract of DOI:10.1002/ejoc.200801116

The investigated set of 6-exo-trig cyclizations occurs irreversibly and justifies the use of tetrahydropyran-derived transition structures for rationalizing 2,6-cis, 2,5-trans, and 2,4-cis stereoselectivity in ring-closure reactions of 1-, 2-, and 3-pheny

Full text of DOI:10.1002/ejoc.200801116

SHORT COMMUNICATION
DOI: 10.1002/ejoc.200801116
On the Stereoselectivity of Alkenoxyl Radical 6-exo-trig Cyclizations
Nina Schneiders,[a] Thomas Gottwald,[a] and Jens Hartung*[a]
Keywords: Radical reactions / Alkoxyl radical / Carbon radical / Bromination / Cyclization / Stereoselective synthesis /
Stereochemical analysis
The investigated set of 6-exo-trig cyclizations occurs irre- (© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
versibly and justifies the use of tetrahydropyran-derived
transition structures for rationalizing 2,6-cis, 2,5-trans, and
2,4-cis stereoselectivity in ring-closure reactions of 1-, 2-, and
3-phenyl-6-methyl-5-hepten-1-oxyl radicals.
Germany, 2009)
Introduction
Substituents that increase π-nucleophilicity of olefinic
double bonds facilitate 6-exo-trig cyclizations of 5-hexen-1-
oxyl radicals (e.g. 1aǞ2a, Scheme 1) at the expense of an
otherwise significantly interfering 1,5-H-atom transfer.[1–5]
Among the discoveries that followed this perception, dia-
stereoselectivity in the 6-methyl-1-phenylhept-5-en-1-oxyl
radical cyclization 1aǞ2a probably was most important
(Scheme 1).[1] From data available in 2004, the origin of the
2,6-cis selectivity, however, remained unclear. From a book
chapter summarizing state of the art heptenyl and related Scheme 1. Structural formulae and indexing of phenyl-substituted
6-methylhept-5-en-1-oxyl radicals 1a–c, and stereoselective synthe-
radical cyclizations it became obvious that selectivity in
carbon radical 6-exo-trig cyclizations still is explained on a
phenomenological basis using adapted stereochemical con-
ventions.[6] The applied models, however, are neither based
on physical organic investigations nor on computational
analysis. In view of this unsatisfactory theoretical back-
sis of tetrahydropyran 3a via alkenoxyl radical 6-exo-trig cycliza-
tion 1a Ǟ 2a in the presence of BrCCl3 (T = 80 °C).[1]
ground and the necessity to obtain clear cut information on
principles that guide diastereoselection in the synthesis of
Results and Discussion
2,5- and 2,4-disubstituted tetrahydropyrans via 6-exo-trig
Generation of alkenoxyl radicals for the pursuit of
stereoselectivity in alkenoxyl radical 6-exo-trig cyclizations
was achieved under pH neutral non oxidative conditions in
thermally induced transformations starting from N-alkoxy-
4-methyl-5-(p-methoxyphenyl)thiazole-2(3H)-thiones 4b–c
(Scheme 2).[7] BrCCl3 preferentially served as mediator, as
the site of bromination corresponds to the position of car-
bon radical intermediates formed in the alkenoxyl radical
reaction of interest.[8] Required radical precursors 4b–c
were prepared from N-hydroxy-4-methyl-5-(p-methoxy-
phenyl)thiazole-2(3H)-thione tetraethylammonium salt[9,10]
(not shown) and appropriate alkenyl tosylates in 61% (for
4b) and 78% (for 4c) yield, by adapting established pro-
cedures (see the Supporting Information). Compounds 4b–
c were obtained as yellowish crystalline solids that were
stored in amber-colored flasks.
alkenoxyl radical reactions in an ongoing project, an inde-
pendent investigation on this subject was devised. The
major results from this study clarify that the explored type
of 6-exo-trig reaction occurs irreversibly. This finding sug-
gests that 2,6-cis, 2,5-trans, and 2,4-cis selectivity in cycliza-
tions of 1-, 2-, and 3-phenyl-6-methyl-5-hepten-1-oxyl
radicals orginates from kinetically controlled reactions,
justifying the use of transition structures for rationalizing
stereoselectivity in this chemistry.
[a] Fachbereich Chemie, Organische Chemie, Technische Uni-
versität Kaiserslautern,
Erwin-Schrödinger-Straße, 67663 Kaiserslautern, Germany
Fax: +49-631-205-3921
E-mail: hartung@chemie.uni-kl.de
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
Eur. J. Org. Chem. 2009, 797–800
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
797

N. Schneiders, T. Gottwald, J. Hartung
SHORT COMMUNICATION
Scheme 3. Mechanistic interpretation of 6-methyl-2-phenylhept-5-
en-1-oxyl radical reactions other than 6-exo-trig ring closure (for-
mation of formaldehyde was not verified).
leled by a minor loss of products thus providing 39% of
analytically pure 3c (cis/trans = 79:21) and 25% of 5c
(cis/trans Ͻ 5:95, 1H NMR). Stereochemical analysis of
tetrahydropyran 3c was performed in extension to the strat-
egy outlined for isomer 3b.
The origin of diastereoselectivity in alkenoxyl radical 6-
exo-trig reactions was clarified in a stereochemical analysis.
A reversible ring closure would imply that a notable frac-
tion of cyclized radical 2a would undergo β-cleavage to
yield alkenoxyl radical 1a. As a result, the isomeric ratio
of independently generated stereochemically homogeneous
intermediates cis-2a and trans-2a should change (thermo-
dynamic control). If configuration of intermediates cis-2a
and trans-2a were retained over a time span that exceeded
their life-times in comparison to that from synthesis of 3a
by 1–2 orders of magnitude, cyclization 1aǞ2a had to be
classified as irreversible process under such conditions.
The question posed above was experimentally addressed
as follows. Diastereomerically pure tertiary alcohols cis-9
(cis/trans Ͼ 99:1) and trans-9 (cis/trans Ͻ 1:99)[13] (GC)
were prepared and treated with an excess of oxalyl chloride.
This step furnished alkyl chlorosemioxalates (not shown),
which were converted with N-(hydroxy)pyridine-2(1H)-
thione sodium salt (PtO– Na+) into mixed oxalates cis-10
and trans-10, respectively (Scheme 4). Thermally-induced
Scheme 2. Thermally induced transformation of N-(6-methylhept-
5-en-1-oxy)thiazolethiones 4b–c in the presence of BrCCl3 and
AIBN (An = p-MeOC6H4).
Treatment of N-(6-methyl-2-phenylhept-5-en-1-oxy)thi-
azolethione 4b (c0 = 0.08 ) in a solution of BrCCl3 (c0 =
0.8 ) and C6H6 with small portions of AIBN until com-
plete consumption of the starting material had occurred
(ca. 2–3 h), furnished after chromatographic work up of the
reaction
hydropyran (3b) (14%, cis/trans = 21:79), 2-(2-methylprop-
1-enyl)-4-phenyltetrahydrofuran (5b) (20%, cis/trans
mixture
5-phenyl-2-(2-bromoprop-2-yl)tetra-
=
50:50), and 6-bromo-2-methyl-6-phenylhex-2-ene (6) (39%)
(Scheme 2). This material accounted for a mass balance of
73%. Stereochemical analysis of tetrahydropyran 3b was
performed on the basis of one- and two-dimensional NMR-
techniques (1H NMR, 13C NMR, NOESY, HMQC). The
low yield of 6-exo-trig cyclization product 3b was explicable
by including bromoalkene 6 into mechanistic interpretation
(Scheme 3). Formation of the latter compound in notable
yield points to efficient β-C,C-bond cleavage in the 6-
methyl-2-phenylhept-5-en-1-oxyl radical (1b) and liberation
of stabilized secondary benzylic radical 7. Trapping of car-
bon radical 7 with BrCCl3 would explain the formation of
bromoolefin 6 from thiazolethione 4b on the basis of a well
established oxygen radical reaction.[11,12] Formation of
tetrahydrofuran 5b is explicable in a multistep sequence
starting with 1,5-H-atom transfer 1bǞ8, Br atom trapping
of allylic radical 8 and intramolecular cycloetherification
via HBr elimination (Scheme 3).
Conversion of N-(6-methyl-3-phenyl-5-heptenoxy)thi-
azolethione 4c (c0 = 0.08 ) in benzene solution containing
0.8  BrCCl3 at 80 °C furnished 46% of 4-phenyl-2-(2-bro-
moprop-2-yl)tetrahydropyran (3c) (cis/trans = 87:13) and
42% of 2-(2-methylprop-1-enyl)-3-phenyltetrahydrofuran
(5c) (cis/trans = 12:88) (1H NMR, reaction in C6D6, anisole
as internal standard; 88% mass balance). Purification of the
Scheme 4. Generation of (tetrahydropyran-2-yl)prop-2-yl radicals
2a from tertiary alcohols cis/trans-9 [for stereochemical analysis see
text and Table 1; py = pyridin-2-yl; PtO– Na+ = N-hydroxypyr-
reaction mixture via column chromatography was paral- idine-2(1H)-thione sodium salt].
798
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 797–800

Alkenoxyl Radical 6-exo-trig Cyclizations
Table 1. Formation of tetrahydropyrans 3a and 11 from (tetrahydropyran-2-yl)-2-propanols 9.
Entry
(Ϯ)-9/
BrCCl3/
Conversion[a]
(Ϯ)-3a/%
(Ϯ)-11/%
(cis/trans ratio)[b]
(cis/trans ratio)[b]
1
2
3
4
cis-9/8.3ϫ10–3
cis-9/8.3ϫ10–4
trans-9/8.3ϫ10–3
trans-9/8.3ϫ10–4
8.3ϫ10–2
8.3ϫ10–3
8.3ϫ10–2
8.3ϫ10–3
87 %
49 %
77 %
75 %
58 (Ͼ 99:1)
34 (Ͼ 99:1)
27 (Ͻ 1:99)
13 (Ͻ 1:99)
7 (Ͼ 99:1)
8 (Ͼ 99:1)
12 (Ͻ 1:99)
8 (Ͻ 1:99)
[a] Referenced vs. unreacted alcohol 9. [b] Stereochemical analysis was performed via GC of crude reaction mixtures using authentic
1
samples as reference, H NMR, and NOESY investigations of purified products.
decomposition of mixed oxalates 10 (c0 = 8ϫ10–3 to notably stretched (ca. 2.1 Å)[16] C,O-bond, similar to the
8ϫ10–4 ) afforded derived tertiary carbon radicals,[14,15] scenario outlined in the heuristic chair-E model[6] for car-
i.e. cis-2a (from cis-10) and trans-2a (from trans-10). Radi- bon radical 6-exo-trig cyclizations. Sterically demanding
cals cis-2a and trans-2a, respectively, were trapped with an phenyl groups are considered to preferentially adopt sites
at least 10-fold excess of BrCCl3 (c0 = 8ϫ10–2 to that are similar to equatorial positions in tetrahydropyran.
8ϫ10–3 ). The former reaction afforded cis-3a (cis/trans Ͼ In extension to results from a theoretical treatment of the
99:1), olefin cis-11, and minor amounts of unreacted 4-penten-1-oxyl radical ring closure[16] it is furthermore ex-
alcohol cis-9, starting from cis-9 (Table 1, entries 1–2). In a pected that the isopropylidene entity resides in a location,
similar manner trans-3a (cis/trans Ͻ 1:99), olefin trans-11, where it experiences least severe synclinal interactions, i.e.
and unspent substrate trans-9 were obtained, if trans-9 equatorial-like, in order to adhere with the tetrahydropyran
served as starting material (Table 1, entries 3–4). 6-Methyl- notation (Scheme 5). A gradual decrease in stereodirecting
1-phenylhept-5-en-1-ol and/or 6-methyl-1-phenylhept-5-en- effect as the distance between reacting entities increases (1a
1-one were detected (GC) in neither of the runs.
Ͼ 1b Ͼ 1c), and estimated ∆∆G‡ values of 1.8–5.6 kJmol–1
According to results from stereochemical analysis, ring (ca. 80 °C) for stereodifferentiating processes are in accord
opening reactions cis-2aǞ1a and trans-2aǞ1a were not with this interpretation.[16]
relevant under selected conditions. This argumentation is
based on the stereochemical integrity within both tetra-
hydropyran series (cis and trans isomers). Observed dia-
stereoselectivities of product 3a, and presumably also of de-
Concluding Remarks
The use of transition structures as mnemonic device for
interpreting and predicting stereoselecitivity in the investi-
gated type of alkenoxyl radical 6-exo-trig cyclization is ex-
pected to serve as impetus for application of the method in
stereoselective synthesis, possibly by increasing efficiency of
tetrahydropyran synthesis using polar substituents other
than CH3 attached to the terminal position of the vinyl en-
tity. From what is known in 4-penten-1-oxyl radical chemis-
try it is expected that substitution at this site will change
rates but not facial selectivity of intramolecular alkoxyl rad-
ical additions.[17,18] The motivation for pursuing this chem-
istry originates from unique selectivity of 5-hexen-1-oxyl
radicals. Ionic bromocyclizations of ω,ω-dimethyl-substi-
tuted hex-5-en-1-ols (not shown), for example, will not pre-
fer the 6-exo-mode of ring closure.[19,20]
rivatives 3b–c, thus originated from kinetically controlled
reactions. Since 5-hexen-1-oxyl radicals are neutral interme-
diates, their chains are expected to fold upon an approach
of reacting entities in a chair-like manner, for minimizing
strain and steric repulsion. The underlying transition struc-
ture that is formed in this scenario is expected to resemble a
1
distorted C4-conformation of tetrahydropyran having one
Supporting Information (see also the footnote on the first page of
this article): Experimental procedures, characterization data of
compounds 3–6.
Acknowledgments
This work was supported by the Deutsche Forschungsgemeinschaft
(DFG) (Grant Ha1705/5-2).
Scheme 5. Transition structure-based stereochemical model for pre-
dicting major product formation in 6-exo-trig cyclizations of sub-
stituted 5-hexen-1-oxyl radicals 1a–c [1C4 denotes the chair confor-
mation of distorted tetrahydropyran, with atoms 1 (O) and 4 (C)
offset in opposite direction from the tetrahydropyran plane; letters
printed in italics refer to positioning of substituent R1–R3 = H or
Ph (first descriptor) and isopropylidene substituent (second de-
scriptor)].
[1] J. Hartung, T. Gottwald, Tetrahedron Lett. 2004, 45, 5619–
5621.
[2] M. P. Bertrand, J. M. Surzur, M. Boyer, M. L. Mihailovic´, Tet-
rahedron 1979, 35, 1365–1372.
[3] R. D. Rieke, B. J. A. Cooke, J. Org. Chem. 1971, 36, 2674–2677.
Eur. J. Org. Chem. 2009, 797–800
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
799

N. Schneiders, T. Gottwald, J. Hartung
SHORT COMMUNICATION
[4] S. Kim, K. H. Kim, J. R. Cho, Tetrahedron Lett. 1997, 38,
3915–3918.
[5] A. Johns, J. A. Murphy, Tetrahedron Lett. 1988, 29, 837–840.
[6] D. P. Curran, N. A. Porter, B. Giese, Stereochemistry of Radical
Reactions, VCH Verlagsgesellschaft, Weinheim, 1996, pp. 77–
82.
[13]
[14]
J. Hartung, S. Drees, M. Greb, P. Schmidt, I. Svoboda, H.
Fuess, A. Murso, D. Stalke, Eur. J. Org. Chem. 2003, 2388–
2408.
D. H. R. Barton, D. Crich, Tetrahedron Lett. 1985, 26, 757–
760.
B. Giese, J. Hartung, Chem. Ber. 1992, 125, 1777–1779.
J. Hartung, K. Daniel, C. Rummey, G. Bringmann, Org. Bio-
mol. Chem. 2006, 4, 4089–4100.
[15]
[16]
[7] J. Hartung, K. Daniel, T. Gottwald, A. Groß, N. Schneiders,
Org. Biomol. Chem. 2006, 4, 2313–2322.
ˇ
[8] J. Hartung, T. Gottwald, K. Spehar, Synthesis 2002, 1469–
[17] J. Hartung, F. Gallou, J. Org. Chem. 1995, 60, 6706–6716.
[18] J. Hartung, M. Hiller, P. Schmidt, Liebigs Ann. 1996, 1425–
1436.
[19] F. Bravo, F. E. McDonald, W. A. Neiwert, K. I. Hardcastle,
Org. Lett. 2004, 6, 4487–4489.
[20] M. Greb, J. Hartung, F. Köhler, K. Spehar, R. Kluge, R. Csuk,
1498.
ˇ
[9] J. Hartung, K. Spehar, I. Svoboda, H. Fuess, M. Arnone, B.
Engels, Eur. J. Org. Chem. 2005, 869–881.
ˇ
[10] J. Hartung, T. Gottwald, K. Spehar, Synlett 2003, 227–229.
[11] S. Wilsey, P. Dowd, K. N. Houk, J. Org. Chem. 1999, 64, 8801–
ˇ
8811.
Eur. J. Org. Chem. 2004, 3799–3812.
[12] E. Suárez, M. S. Rodriguez, in Radicals in Organic Synthesis
(Eds.: P. Renaud, M. P. Sibi), Wiley-VCH, Weinheim, 2001, vol.
2, pp. 440–454.
Received: November 11, 2008
Published Online: January 13, 2009
800
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 797–800