Diastereocontrol in [4+4]-photocycloadditions of pyran-2-ones: effect of ring substituents and chiral ketal

Article abstract of DOI:10.1016/j.tetlet.2008.12.117

4-Mesyloxypyran-2-ones joined to furan by a three-carbon linker undergo intramolecular-crossed [4+4]-photocycloaddition with high or complete selectivity for the exo cycloadduct. When a C₂-symmetric ketal was present on the tether adjacent to t

Full text of DOI:10.1016/j.tetlet.2008.12.117

Tetrahedron Letters 50 (2009) 1188–1192
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Diastereocontrol in [4+4]-photocycloadditions of pyran-2-ones: effect
of ring substituents and chiral ketal
Lei Li a, John A. Bender b, , F. G. West a,b,
*
a Department of Chemistry, University of Alberta, E3-43 Gunning-Lemieux Chemistry Centre, Edmonton, AB, Canada T6G 2G2
b Department of Chemistry, University of Utah, 315 S. 1400 East, Rm 2020, Salt Lake City, UT 84112-0850, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
4-Mesyloxypyran-2-ones joined to furan by a three-carbon linker undergo intramolecular-crossed [4+4]-
photocycloaddition with high or complete selectivity for the exo cycloadduct. When a C2-symmetric ketal
was present on the tether adjacent to the pyranone ring, moderate levels of asymmetric induction were
obtained.
Received 22 September 2008
Revised 16 December 2008
Accepted 22 December 2008
Available online 10 January 2009
Ó 2009 Elsevier Ltd. All rights reserved.
The prevalence of cyclooctane rings in nature has inspired a
plethora of methods for their construction.1 [4+4]-Cycloaddition
approaches to the cyclooctanoid system offer several appealing
features, including convergent assembly of the eight-membered
ring from two relatively simpler four-carbon fragments, formation
of two new carbon-carbon bonds, and the potential for introduc-
tion of up to four new stereocenters in one step.2 The seminal work
of de Mayo demonstrated that the readily available heterocycle
4,6-dimethylpyran-2-one undergoes photochemical dimerization
to a mixture of [2+2]- and [4+4]-cycloadducts.3 More recently,
we have developed a crossed [4+4]-photocycloaddition process,
using pyran-2-ones linked to furans by a three-atom tether (Eq.
1).4 This method can furnish densely functionalized cyclooctadi-
enes or triquinanes5 in good yields, and in some cases with moder-
ate endo/exo selectivity. On the other hand, control over the
configuration at the newly established bridgehead centers by use
of preexisting stereocenters has been largely unsuccessful, apart
from one rather specialized example.6
auxiliaries have been used with some success in Diels–Alder cyc-
loadditions on the 1,3-diene partner,9 or (in ring-opened form)
on the dienophile.10 In one especially pertinent study, Fukumoto,
Kametani, and co-workers demonstrated that moderate levels of
diastereoselectivity could be achieved in intramolecular Diels–Al-
der cycloadditions of ortho-quinodimethanes bearing chiral ketals
on the tether adjacent to the dienonphile.11 Here, we describe pre-
liminary studies utilizing C2-symmetric ketals as chiral auxiliaries
in the [4+4]-photocycloaddition reactions of pyran-2-ones.
Our initial plan sought to exploit the ready availability of furoyl
substrates, formed via Michael addition of 4-hydroxypyran-2-ones
to acryloylfuran (Scheme 1).12 Mannich base 1 can be prepared
from acetyl furan in the presence of paraformaldehyde and
Me2NHꢀHCl.13 Upon heating with pyran-2-one 2 in acetonitrile, ad-
duct 3 was obtained in moderate yield. With this material in hand,
the C-4 hydroxyl was capped as a triflate, and several C2-symmet-
ric ketals were prepared under standard conditions (catalytic TsOH
in PhH or PhCH3 at reflux), furnishing 4a–e. Due to the extended
heating required to drive these reactions to completion, use of stoi-
chiometric quantities of TsOH was investigated with (R,R)-butane-
diol. However, under these conditions the furan ring was lost,
giving glycol monoester tosylate 5 as the sole product.14 This pro-
cess is believed to proceed through protiodefuranation of the de-
sired ketal 4a, followed by opening of the resulting dioxolanium
ion by tosylate.
O
O
hν
O
O
O
R3
+
R3
O
R3
OR1
O
1 O
1 O
R2 OR
exo
R2 OR
endo
R2
ð1Þ
With 4a–e in hand, we now set out to probe the ability of the
ketal unit to impact the stereochemical course of the photocyclo-
addition reaction (Table 1). Under standard conditions (aq MeOH,
rt), ketals 4a–c were converted to roughly equal quantities of endo
and exo cycloadducts 6a–c and 7a–c. The low endo/exo selectivity
was surprising, in comparison to examples with unfunctionalized
tethers.4b However, poor endo/exo selectivity was also seen in a
number of cases employing a simple ethylene ketal next to the fur-
an.6 In each successful case of the present study, the endo and exo
adducts were obtained as inseparable mixtures of diastereomers,
Use of stereochemical control elements within the chain linking
the two heterocyclic units was of particular interest.7 While place-
ment of methyl-substituted branch points on the tether failed to
achieve substantial diastereocontrol,4a,c we were hopeful that cyc-
lic C2-symmetric ketals might exert a greater influence.8 These
* Corresponding author. Tel.: +1 780 492 8187; fax: +1 780 482 8231.
E-mail address: frederick.west@ualberta.ca (F.G. West).
Present address: Bristol-Myers Squibb,
06492, USA.
5 Research Parkway, Wallingford, CT
0040-4039/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.12.117

L. Li et al. / Tetrahedron Letters 50 (2009) 1188–1192
1189
O
O
Me2N
O
O
Δ
1) Tf2O, Et3N (80%)
1
O
+
O
R
R
O
CH3CN
2)
Me
OH
O
O
3
HO
Δ, cat. TsOH
PhH or PhCH3
OH
(41%)
M
e
OH
2
1) Tf2O, base
2) Δ, TsOH (1 equiv)
R
O
R
PhCH ,
3 Me
Me
O
O
HO
O
OH
O
O
Me
Me
OTf
OTs
4a
4b
O
O
(R = Me; 89%)
(R = Et; 88%)
Me
4c (R = i-Bu; 46%)
4d (R = Ph; 40%)
4e (R = Bn; 79%)
Me
OTf
5 (79%)
Scheme 1.
Table 1
Photocycloaddition reactions of furyl ketals 4a
R
O
R
R
O
+
Me
R
R
R
O
O
O
hν
O
O
O
O
O
Me
O
aq. MeOH
O
O
Me
OTf
O
O
OTf
OTf
7a-c
4a-e
6a-c
Entry
Substrate
% Yield cycloadducts (endo:exo)b
dr for 6c
dr for 7c
1
2
3
4
5
4a
4b
4c
4d
4e
66 (1:1)
62 (1:1)
60 (1.5:1)
—
Trace
1.8:1
1.5:1
1.7:1
—
1.4:1
1.8:1
ndd
—
e
ndd
ndd
a
Standard procedure: ketal substrate 4 was dissolved in aq MeOH (ca. 60% v/v; ca. 6 mM) in a Pyrex tube, and the resulting solution degassed with a slow stream of dry N2
gas. The tube was clamped ca. 10 cm from a Hannovia 450 W medium pressure Hg lamp and irradiated until consumption of 4 (2–5.5 h).
b
All yields given are for homogeneous isolated product following chromatographic purification.
c
Diastereomeric ratios were determined by 1H NMR integration of dihydrofuran alkenyl protons, except in the case of 7b, where integrals of upfield methyl doublets were
measured.
d
Diastereomeric ratios could not be determined in the case of 7c due to spectral overlap, and in the case of 6e/7e due to insufficient quantities.
Only uncharacterizable polar materials were formed.
e
differing in relative configuration between the newly formed
bridgehead stereocenters and the chiral ketals. Ratios were mea-
sured via integration of downfield alkenyl protons of the dihydro-
furan, which were well resolved in all but two cases.
Diphenyl ketal 4d and dibenzyl ketal 4e furnished little or no
[4+4]-cycloadduct upon irradiation; instead, a complex mixture
of polar products was formed, presumably the result of alternative
pyran-2-one photochemical pathways.15 Notably, among the
examples that did undergo the desired cycloaddition, the steric
bulk of the ketal substituents had little effect on the selectivity,
with all cases falling in a range of 1.5–1.8:1 dr, suggesting a rela-
tively small energetic difference between the diastereomeric
photocycloaddition transition states.
pling of the heterocyclic reactants was accomplished through aldol
condensation of dehydroacetic acid and furfural,16 and the result-
ing enone was subjected to catalytic hydrogenation to give 8
(Scheme 2). A variety of protecting groups for the pyranone hydro-
xyl were examined. Pivalate 9a could readily be prepared, but was
found to be labile during the subsequent ketalization. Sulfonates
9b–e were also prepared in good to excellent yield, and proved
to be much more robust to the ketalization conditions.
Ketalization of 9b–e was then examined, using achiral ethylene
glycol to optimize conditions. Standard conditions (HOCH2CH2OH,
cat. acid, benzene or toluene, reflux) caused extensive decomposi-
tion. However, under the Noyori conditions,17 moderate amounts
of ketal 10d were obtained. Reaction rates were greatly increased
when acetonitrile was used as solvent, providing 10d in >90% yield.
These conditions were applied to 9b,c,e and furnished 10b,c,e in
moderate yields.
Given the disappointing levels of diastereoselectivity seen with
furyl ketals 4, we turned our attention to an alternative substitu-
tion pattern, with the ketal adjacent to the pyran-2-one ring. Cou-

1190
L. Li et al. / Tetrahedron Letters 50 (2009) 1188–1192
O
OH
O
O
OH
O
O
O
OHC
O
H2, Pd/C
Me
piperidine
CHCl3
85%
Et3N, CH2Cl2
85%
Me
Me
O
Me
OR
O
OH
O
O
O
conditions
Me
O
O
O
O
8
PivCl, Et3N; 91%
9
9
9
9
9
a (R = Piv)
(R = Tf)
c (R = Ts)
d (R = Ms)
(R = Es)
b
Tf2O, Et3N, -40 °C; 95%
TsCl, Et3N, 0 °C; 92%
MsCl, Et3N, 0 °C; 70%
EsCl, Et3N, 0 °C, 60%
e
Es = EtSO2–
TMSO
OTMS
MsO
O
O
O
9
d
TMSOTf, 0 °C
CH2Cl2 48 h, 30%
,
Me
O
O
or
10d
CH3CN, 2.5h, 92%
RO
O
O
O
as above
9b,c,e
Me
O
O
10b (R = Tf;45%)
10c (R = Ts;39%)
10e (R = Es;68%)
Scheme 2.
R
O
O
O
O
O
O
O
O
+
O
hν, PhH
O
Me
Me
O
O
5–10 °C
O
O
OR
OR
O
Me
O
11b,c
12b,c
10b (R = Tf)
10c (R = Ts)
mixtures
RO
O
O
O
hν
conditions
O
Me
O
O
O
O
Me
O
O
RO
12d
10d (R = Ms)
PhH, 5–10 °C, 8.5 h
PhCH3, –78 °C, 9 h
(72%; single isomer)
10e
12e (58%; single isomer)
(R = Es)
RO
O
O
O
O
hν
+
Me
Me
conditions
O
O
Me
O
OR
OR
O
O
13a (R = Ac)
13b
1:1.6 (R = Ac)
1:10 (R = Ms)
aq. MeOH,
PhCH3, 0 °C (48%)
(68%)
0 °C
(R = Ms)
Scheme 3.
Next, the photochemical behavior of 10b–e was examined
(Scheme 3). Unlike 4a–c, these ketals were not stable under the stan-
dard conditions (aq MeOH, rt), necessitating the use of aprotic sol-
vents. Upon irradiation in benzene, triflate 10b underwent [4+4]-
photocycloaddition to give a mixture of endo 11b and exo 12b, while
tosylate 10c furnished a complex mixture of products that included
some of the expected 11c and 12c. However, under the same condi-
tions, mesylate 10d provided exo adduct 12d in good yield as a single
isomer. The origin of this remarkable diastereoselectivity is not
apparent, but the presence of a small alkanesulfonate at the pyrone

L. Li et al. / Tetrahedron Letters 50 (2009) 1188–1192
1191
9d
Me
Me
TMSO
OTMS
TMSO
OTMS
TMSOTf, rt TMSOTf, 0 °C
CH3CN, 4 h
44%
CH3CN, 3 d
49%
Me
Me
MsO
MsO
O
O
O
O
O
O
Me
O
O
Me
O
O
14a
14b
O
O
hν, conditions
Me
Me
+
O
14a
O
O
O
O
O
O
O
Me
Me
MsO
MsO
Me
Me
15a
PhH
,
5–10 °C, 2h
2.3 : 1
4.2 : 1
1.4 : 1
69%
64%
73%
PhCH3, -78 °C, 3.5h
aq. MeOH, 0 °C, 3h
O
O
hν, PhCH3
Me
Me
+
14b
O
O
O
O
-78 °C, 3.5 h
55%
O
O
O
O
MsO
MsO
15b (3.3:1)
Me Me
1) MsCl, Et3N
hν, aq. MeOH
MsO
3
O
O
O
2)
TMSOTf
CH3CN
0 °C, 4 h
81%
Me
Me
Me
O
O
TMSO
OTMS
4f
(54%, 2 steps)
Me
O
Me
O
Me
+
Me
O
O
O
O
Me
Me
O
O
O
O
OMs
OMs
f (1.9:1)
6
7f (2.4:1)
Scheme 4.
C-4 position seems to be essential. For example, ethanesulfonate 10e
also furnished exo adduct 12e as the only detectable photoproduct.
Further evidence for this substituent effect can be seen with sub-
strates 13a,b. Acetate 13a was previously shown to give a 1:1.6 mix-
ture of endo and exo cycloadducts,4c but the corresponding mesylate
13b furnished a 10:1 exo/endo ratio of adducts.
The complete exo selectivity obtained with 10d was unex-
pected, and it offered a clear advantage relative to the earlier
cases (4a–c) with regard to asymmetric induction studies with
chiral ketals: with the formation of only the exo cycloadduct,
stereochemical analysis would be greatly simplified, as only
two components would be expected in the product mixture.
Thus, ketone 9d was used for all subsequent experiments.18 In
the event, preparation of branched ketals was found to be sub-
stantially slower than for the simple, unsubstituted case (Scheme
4). However, upon extended stirring with TMSOTf and the bis(si-
lyl ether) of racemic butane-2,3-diol or cyclohexane-1,2-diol, ke-
tals 14a and 14b were obtained in moderate yields. Irradiation of
14a in benzene at 5 °C afforded exo [4+4]-cycloadduct 15a in
69% yield and 2.3:1 dr.19 At ꢁ78 °C in toluene, the dr was im-
proved to 4.2:1 with little diminution of chemical yield. These
latter conditions were also applied to 14b, providing 15b in
55% yield and 3.3:1 dr.
The improvement in diastereoselectivity seen for 14a,b was
gratifying, but it was unclear whether it derived from the presence
of mesylate, the position of the ketal on the tether, the conditions

1192
L. Li et al. / Tetrahedron Letters 50 (2009) 1188–1192
RO2S
RO2S
H
O
O
Me
O
O
O
O
S
Me
O
R
O
Me
Me
O
Me
Me
O
R
O
Me
O
O
O
MsO
15a-u
O
O
O
O
A
C
Me
Me
O
Me
Me
O
O
O
O
O
M
e
O
Me
R
O
Me
Me
O
R
O
O
O
O
O
R
O
Me
MsO
Me
O
RO 2S
RO2S
B
15a-l
Me
D
Scheme 5.
of irradiation, or some combination of these factors. In contrast to
10d, 14a tolerated the conditions that had been used for ketals 4a–
c, and irradiation in aq MeOH furnished 15a in 73% and only 1.4:1
dr, indicating the importance of low temperature and aprotic sol-
vent. To further clarify this issue, we also prepared 4f (analogous
to 4a, but with mesylate in place of triflate). Unfortunately, this
substrate was unreactive to extended irradiation in toluene at
ꢁ78 °C, or even at rt. Reaction in aq MeOH furnished a nearly equal
ratio of endo and exo cycloadducts 6f and 7f, each formed in only
moderate diastereoselectivity (1.9–2.4:1 dr). From these results,
we conclude that all of the factors mentioned above affect the
selectivity of the cycloaddition process.
It was not possible to obtain the major diastereomers of 6, 7, or
15 in homogeneous crystalline form, precluding the rigorous
assignment of relative configuration. A working model for the most
selective case (14a) is shown above (Scheme 5). Of the four possi-
ble exo transition states in which the tether adopts a chair-like con-
formation (A–D), A and B are considered less favorable due to
steric interaction between the ketal methyl group adjacent to the
pseudoaxial oxygen and the rest of the tether. Likewise C is ex-
pected to experience unfavorable nonbonded interactions between
the sulfonate group and the pseudoaxial hydrogen, while D seems
to be largely devoid of such issues. This model predicts isomer
15a–l, in which the bridgehead stereocenter adjacent to the ketal
has the same relative configuration as the ketal stereocenters, to
be the major product. However, pending access to a sample suit-
able for single crystal X-ray diffraction analysis, this stereochemi-
cal assignment must remain tentative.
Pyran-2-ones linked to furans and possessing a keto group in
the three-carbon tether are easily prepared. Chiral ketals can be in-
stalled adjacent to either heterocycle, and impose poor to moder-
ate levels of diastereoselectivity in the photochemical [4+4]-
cycloaddition reactions of these substrates. Notably, derivatization
of the 4-hydroxyl on the pyrone ring with a methanesulfonyl group
leads to high or complete selectivity for the exo cycloadduct in
most cases. Improved selectivity and a more detailed understand-
ing of the mechanism of asymmetric induction may be possible
through further optimization of the ketal moiety and the photocy-
cloaddition conditions, and the results of these continuing studies
will be described in due course.
References and notes
1. (a) Evans, P. A.; Baum, E. W.; Fazal, A. N.; Pink, M. Chem. Commun. 2005, 63;
(b) Murakami, M.; Ashida, S.; Matsuda, T. J. Am. Chem. Soc. 2006, 128, 2166;
(c) Clark, J. S.; Walls, S. B.; Wilson, C.; East, S. P.; Drysdale, M. J. Eur. J. Org.
Chem. 2006, 323; (d) Murphy, G. K.; Marmsäter, F. P.; West, F. G. Can. J.
Chem. 2006, 84, 1470; (e) Krishnan, K. S.; Smitha, M.; Suresh, E.;
Radhakrishnan, K. V. Tetrahedron 2006, 62, 12345; (f) DeBoef, B.; Counts,
W. R.; Gilbertson, S. R. J. Org. Chem. 2007, 72, 799; (g) Zhang, Y.-D.; Ren, W.-
W.; Lan, Y.; Xiao, Q.; Wang, K.; Xu, J.; Chen, J.-H.; Yang, Z. Org. Lett. 2008, 10,
665; (h) Jiao, L.; Yuan, C.; Yu, Z.-X. J. Am. Chem. Soc. 2008, 130, 4421;
Reviews: (i) Mehta, G.; Singh, V. Chem. Rev. 1999, 99, 881; (j) Michaut, A.;
Rodriguez, J. Angew. Chem., Int. Ed. 2006, 45, 5740; (k) López, F.; Mazcareñas,
J. L. Chem. Eur. J. 2007, 13, 2172.
2. (a) Zhu, M.; Qiu, Z. L.; Hiel, G. P.; Sieburth, S. McN. J. Org. Chem. 2002, 67, 3487;
(b) Wang, Y.; Schill, B. D.; Arif, A. M.; West, F. G. Org. Lett. 2003, 5, 2747; (c)
Wender, P. A.; Croatt, M. P.; Witulski, B. Tetrahedron 2006, 62, 7505; (d)
Harmata, M.; Rashatasakhon, P.; Barnes, C. L. Can. J. Chem. 2006, 84, 1456.
3. (a) de Mayo, P.; Yip, R. W. Proc. Chem. Soc. London 1964, 84; (b) de Mayo, P.;
McIntosh, C. L.; Yip, R. W.. In Organic Photochemical Synthesis; Srinivasan, R.,
Roberts, T. D., Eds.; Wiley-Interscience: New York, 1971; Vol. 1, pp 99–100.
4. (a) West, F. G.; Chase, C. E.; Arif, A. M. J. Org. Chem. 1993, 58, 3794; (b) Chase, C.
E.; Bender, J. A.; West, F. G. Synlett 1996, 1173; (c) Li, L.; Chase, C. E.; West, F. G.
Chem. Commun. 2008, 4025.
5. Li, L.; McDonald, R.; West, F. G. Org. Lett. 2008, 10, 3733.
6. Song, D.; McDonald, R.; West, F. G. Org. Lett. 2006, 8, 4075.
7. For examples of diastereoselectivity in related metal-catalyzed or
photochemical [4+4]-cycloadditions using tether-based control elements, see:
(a) Wender, P. A.; Nuss, J. M.; Smith, D. B.; Suárez-Sobrino, A.; Vågberg, J.;
Decosta, D.; Bordner, J. J. Org. Chem. 1997, 62, 4908; (b) Sieburth, S. McN.;
McGee, K. F., Jr.; Al-Tel, T. H. J. Am. Chem. Soc. 1998, 120, 587; (c) McGee, K. F.,
Jr.; Al-Tel, T. H.; Sieburth, S. McN. Synthesis 2001, 1185.
8. (a) Mash, E. A.; Hemperly, S. B.; Nelson, K. A.; Heidt, P. C.; Van Deusen, S. J. Org.
Chem. 1990, 55, 2045; Review: (b) Alexakis, A.; Mangeney, P. Tetrahedron:
Asymmetry 1990, 1, 477.
9. Boehler, M. A.; Konopelski, J. P. Tetrahedron 1991, 47, 4519.
10. (a) Sammakia, T.; Berliner, M. A. J. Org. Chem. 1994, 59, 6890; (b) Anderson, J. C.;
Blake, A. J.; Graham, J. P.; Wilson, C. Org. Biomol. Chem. 2003, 1, 2877.
11. Nemoto, H.; Nagai, M.; Kohzuki, K.; Fukumoto, K.; Kametani, T. J. Chem. Soc.,
Perkin Trans. 1 1988, 2835.
12. Bravo, P.; Resnati, G.; Viani, F.; Cavicchio, G. J. Chem. Res. (S) 1986, 374.
13. Joshi, M. V.; Hemler, C.; Cava, M. P.; Cain, J. L.; Bakker, M. G.; McKinley, A. J.;
Metzger, R. M. J. Chem. Soc., Perkin Trans. 2 1993, 1081.
14. For an early report on the scope of this unexpected process, see: Bender, J. A.;
Daves, S.; West, F. G. Tetrahedron Lett. 1998, 39, 2051.
15. (a) Corey, E. J.; Streith, J. J. Am. Chem. Soc. 1964, 86, 950; (b) Chapman, O. L.;
McIntosh, C. L.; Pacanshy, J. J. Am. Chem. Soc. 1973, 95, 244; (c) Huang, B.-S.;
Pong, R. G. S.; Laureni, J.; Krantz, A. J. Am. Chem. Soc. 1977, 99, 4154; (d)
Arnold, B. R.; Brown, C. E.; Lusztyk, J. J. Am. Chem. Soc. 1993, 115, 1576; (e)
Chase, C. E.; Jarstfer, M. B.; Arif, A. M.; West, F. G. Tetrahedron Lett. 1995, 36,
8531.
16. Wiley, R. H.; Jarboe, C. H.; Ellert, H. G. J. Am. Chem. Soc. 1955, 77, 5102.
17. Tsunoda, T.; Suzuki, M.; Noyori, R. Tetrahedron Lett. 1980, 21, 1357.
18. Although ethanesulfonate 10e also demonstrated complete exo selectivity, its
precursor 9e was not used in subsequent studies due to extremely slow rates of
ketalization.
Acknowledgements
19. Diastereomer ratios for 15a and 15b were measured by integration of the enol
mesylate alkene protons, which were well resolved in both cases.
The authors wish to thank NIGMS, NSF, and NSERC (Canada) for
generous support of this work.