Diels-Alder Reactions of Masked o-Benzoquinones: New Experimental Findings and a Theoretical Study of the Inverse Electron Demand Case

Article abstract of DOI:10.1021/jo030307r

Diels-Alder reactions of the masked o-benzoquinone (MOB) 2 with vinylene carbonate (3), the bicyclic derivatives 4, 5, and 6, and the intramolecular version of the 2-hydroxymethylfuran-MOB Diels-Alder reaction are described. In addition, a theoretical study of the Diels-Alder reactions of MOBs with enol and thioenol ethers is presented.

Full text of DOI:10.1021/jo030307r

Diels-Ald er Rea ction s of Ma sk ed o-Ben zoqu in on es: New
Exp er im en ta l F in d in gs a n d a Th eor etica l Stu d y of th e In ver se
Electr on Dem a n d Ca se
Odo´n Arjona,§ Roc´ıo Medel,§ J oaqu´ın Plumet,*,§ Rafael Herrera,‡
Hugo A. J ime´nez-Va´zquez,† and J oaqu´ın Tamariz*,†
Departamento de Qu´ımica Orga´nica, Facultad de Qu´ımica, Universidad Complutense, 28040 Madrid,
Spain, Instituto de Investigaciones Quimicobiolo´gicas, Universidad Michoacana de San Nicola´s de
Hidalgo, Edif. B-1, Ciudad Universitaria, Francisco J . Mujica S/ N, 58066 Morelia, Michoacan, Mexico,
and Departamento de Qu´ımica Orga´nica, Escuela Nacional de Ciencias Biolo´gicas, Instituto Polite´cnico
Nacional, Prol. Carpio y Plan de Ayala, 11340 Me´xico, D.F., Mexico
plumety@quim.ucm.es
Received September 29, 2003
Diels-Alder reactions of the masked o-benzoquinone (MOB) 2 with vinylene carbonate (3), the
bicyclic derivatives 4, 5, and 6, and the intramolecular version of the 2-hydroxymethylfuran-MOB
Diels-Alder reaction are described. In addition, a theoretical study of the Diels-Alder reactions
of MOBs with enol and thioenol ethers is presented.
SCHEME 1
Masked o-benzoquinones (MOBs) 1 are a synthetically
useful class of cyclohexa-2,4-dienones.1 These compounds
can be generated by in situ oxidation of readily available
R-alkoxy phenols, using hypervalent iodine reagents in
the presence of an alcohol (Scheme 1).2
MOBs have been shown to be efficient 4π components
in Diels-Alder reactions undergoing regio- and stereo-
selective cycloaddition processes with electron-deficient3
and electron-rich dienophiles.4 Other 2π components used
in these reactions have been heteroaromatic compounds
such as furan,5 indole,6 pyrrole,7 and thiophene8 deriva-
tives. Hetero-9 and intramolecular3g,10 Diels-Alder reac-
tions have also been considered.
In several cases the resulting bicyclo[2.2.2]octenone
derivatives have been used as starting materials for the
synthesis of different targets, including polysubstituted
cyclohexenes,11 cis-decalins,12 bicyclo[4.2.2]decenones,12g
the triquinane ring system,13 and highly functionalized
tricyclic compounds.14 It should be pointed out that these
Diels-Alder processes were a key step in several complex
synthetic sequences, such as in the syntheses of clerodane
This paper is dedicated to Professor J ose Luis Soto.
§ Universidad Complutense.
‡ Universidad Michoacana de San Nicola´s de Hidalgo.
† Instituto Polite´cnico Nacional.
(1) For selected recent reviews, see: (a) Liao, C. C. Acc. Chem. Res.
2002, 35, 856-866. (b) Quideau, S.; Pouyse´gu, L. Org. Prep. Proced.
Int. 1999, 31, 617-680. (c) Liao, C. C. Modern Methodology in Organic
Synthesis; Kodansha: Tokyo, J apan, 1992; p 409.
(2) (a) Ku¨rti, L.; Herczegh, P.; Visy, J .; Simonyi, M.; Antus, S.; Pelter,
A. J . Chem. Soc., Perkin Trans. 1 1999, 379-380. (b) Varvoglis, A.
Hypervalent Iodine in Organic Synthesis; Academic Press: San Diego,
CA, 1997. (c) Bodajla, M.; J ones, G. R.; Ramsden, C. A. Tetrahedron
Lett. 1997, 38, 2573-2576. (d) Stang, P. J .; Zhdankin, V. V. Chem.
Rev. 1996, 96, 1123-1178.
(3) (a) Lai, C. H.; Shen, Y. L.; Wang, M. N.; Kameswara Rao, N. S.;
Liao, C. C. J . Org. Chem. 2002, 67, 6493-6502. (b) Lai, C. H.; Lin, P.
Y.; Peddinti, R. K.; Liao, C. C. Synlett 2002, 1520-1522. (c) Chittimalla,
S. K.; Liao, C. C. Synlett 2002, 565-568. (d) Liao, C. C.; Chu, C. S.;
Lee, T. H.; Rao, P. D.; Ko S.; Song, L. D.; Shiao, H. C. J . Org. Chem.
1999, 64, 4102-4110. (e) Yang, M. S.; Chang, S. Y.; Lu, S. S.; Rao, P.
D.; Liao, C. C. Synlett 1999, 225-227. (f) Lai, C. H.; Shen, Y. L.; Liao,
C. C. Synlett 1997, 1351-1352. (g) Chu, C. S.; Lee, T. H.; Liao, C. C.
Synlett 1994, 635-636.
(4) (a) Gao, S. Y.; Ko, S.; Lin, Y. L.; Peddinti, R. K.; Liao, C. C.
Tetrahedron 2001, 57, 297-308. (b) Gao, S. Y.; Lin, Y. L.; Rao, P. D.;
Liao, C. C. Synlett 2000, 421-423. (c) Arjona, O.; Medel, R.; Plumet,
J . Tetrahedron Lett. 1999, 40, 8431-8433.
(5) (a) Chou, Y. Y.; Peddinti, R. K.; Liao, C. C. Org. Lett. 2003, 5,
1637-1640. (b) Chen, C. H.; Rao, P. D.; Liao, C. C. J . Am. Chem. Soc.
1998, 120, 13254-13255.
(6) Hsieh, M. F.; Rao, P. D.; Liao, C. C. Chem. Commun. 1999, 1441-
1442.
(7) Hsieh, M. F.; Peddinti, R. K.; Liao, C. C. Tetrahedron Lett. 2001,
42, 5481-5484.
(8) Lai, C. H.; Ko, S.; Rao, P. D.; Liao, C. C. Tetrahedron Lett. 2001,
42, 7851-7854.
(9) Lin, K. C.; Liao, C. C. Chem. Commun. 2001, 1624-1625.
(10) (a) Lin, K. C.; Shen, Y. L.; Kameswara Rao, N. S.; Liao, C. C.
J . Org. Chem. 2002, 67, 8157-8165. (b) Chu, C. S.; Lee, T. H.; Rao, P.
D.; Song, L. D.; Liao, C. C. J . Org. Chem. 1999, 64, 4111-4118.
(11) (a) Arjona, O.; Medel, R.; Plumet, J . Tetrahedron Lett. 2001,
42, 1287-1288. (b) Lee, T. H.; Rao, P. D.; Liao, C. C. Chem. Commun.
1999, 801-802.
(12) (a) Sutherland, H. S.; Higgs, K. C.; Taylor, N. J .; Rodrigo, R.
Tetrahedron 2001, 57, 309-317. (b) Carlini, R.; Higgs, K.; Rodrigo,
R.; Taylor, N. Chem. Commun. 1998, 65-66. (c) Hsu, P. Y.; Lee, Y. C.;
Liao, C. C. Tetrahedron Lett. 1998, 39, 659-662. (d) Rao, P. D.; Chen,
C. H.; Liao, C. C. Chem. Commun. 1998, 155-156. (e) Hsiu, P. Y.; Liao,
C. C. Chem. Commun. 1997, 1085-1086. (f) Carlini, R.; Higgs, K.;
Older, C.; Randhawa, S. J . Org. Chem. 1997, 62, 2330-2331. (g) Lee,
T. H.; Liao, C. C.; Liu, W. C. Tetrahedron Lett. 1996, 37, 5897-5900.
(13) (a) Hsu, D. S.; Rao, P. D.; Liao, C. C. Chem. Commun. 1998,
1795-1796. (b) Hwang, J . T.; Liao, C. C. Tetrahedron Lett. 1991, 32,
6583-6586.
(14) (a) Chen, Y. K.; Peddinti, R. K.; Liao, C. C. Chem. Commun.
2001, 1340-1341. (b) Rao, P. D.; Shen, C. H.; Liao, C. C. Chem.
Commun. 1999, 713-714.
10.1021/jo030307r CCC: $27.50 © 2004 American Chemical Society
Published on Web 02/27/2004
2348
J . Org. Chem. 2004, 69, 2348-2354

Diels-Alder Reactions of Masked o-Benzoquinones
previously described, for the first time, the Diels-Alder
reactions of 2 with enol and thioenol ethers.4c A theoreti-
cal study of this process is also presented.
Resu lts
F IGURE 1. Structures of diene 2 and dienophiles 3-6.
The reaction of 3 with MOB 2 afforded the expected
tricyclic derivative 7 in a totally stereoselective fashion,
albeit in low yield (27% isolated). Dimerization of 2 (35%
of the corresponding dimer was obtained) was the main
competitive process (Scheme 2).
The stereochemical assignment of 7 was confirmed by
correlation of its 1H NMR data with those of adducts
derived from the reaction between 2 and enol and
thioenol ethers.4c
Reactions with the norbornenic derivatives 4-6 were
also considered. In our hands, the orthoquinone monoket-
al 2 reacts with these compounds to give the correspond-
ing tetracyclic cycloadducts 8-10 as a single diastere-
omer (endo) and in acceptable yields (Scheme 3). It should
be pointed out that the two possible regioisomers of 10
(10a and 10b) were obtained as an inseparable 1:1
mixture.
Finally, the intramolecular furan-MOB Diels-Alder
reaction was considered. Thus, the reaction of methyl-
vanillate with (diacetoxy)iodobenzene in CH2Cl2 and in
the presence of 2-hydroxymethylfuran gave cycloadduct
diterpenic acids,15 the iridoid forsythide aglycon dimethyl
ester,16 pallescensin B,17 eremopetasidione,18 the lycopo-
dium alkaloid megallanine,19 reserpine (formal synthe-
sis),20 the calicheamicinone aglycon,21 kadsurenone, den-
udatin B, liliflol B, O-methylliliflodione,22 asatone,23 the
carbocyclic core of CP-263,114,24 xestoquinone,12a,f and
halenoquinone.25
Several theoretical approaches have been considered
to explain the regio- and stereoselective aspects of the
reaction, including simple FMO models3d,4a and quantum
mechanical calculations of transition states and possible
intermediates of these reactions.26 In some cases, second-
ary orbital interactions have been invoked to explain the
total endo stereoselectivity observed.3d
As a new contribution to this chemistry, we wish to
address in this paper the Diels-Alder reactions of the
masked o-benzoquinone 2 with vinylene carbonate (3),
the bicyclic derivatives 4, 5, and 6 (Figure 1), and the
hitherto not considered intramolecular version of the
furan-MOB Diels-Alder reaction. On the other hand, we
SCHEME 2
SCHEME 3
J . Org. Chem, Vol. 69, No. 7, 2004 2349

Arjona et al.
to range between 0.3 and 0.7 eV.3d These quite small gaps
suggest that both interactions (HOMO-diene-LUMO-
dienophile and LUMO-diene-HOMO-dienophile) could
be involved and, consequently, prediction of the regiose-
lectivity becomes difficult.
On the other hand, the high endo stereoselectivity
found in these processes has rarely been studied by a
theoretical model.26 Although secondary orbital interac-
tions havebeen invoked toaccount for such stereoselectivity,3d
additional and probably competitive interactions could
be involved in the case of dienophiles with electron-
releasing substituents, such as steric interactions be-
tween the substituent of the dienophile and the dimethoxy
groups in the ethano bridge of the masked o-benzoquino-
nes.
F IGURE 2. Structures of dienes 13a -c, dienophiles 14 and
15, and the major adducts resulting from the cycloaddition
between them.
Therefore, we hereby describe our ab initio calcula-
tions, using a higher basis set (HF/6-31G*) than that used
in previous studies,3d,4a of the FMOs of the cycloaddends
involved in the cycloadditions under inverse electron
demand (IED),4c in order to account for the observed
regioselectivity. The FMO energies of the dienes (2 and
13a ) and dienophiles (15a ,b ) used in our previous
report,4c along with those of dienes (13b,c) and dieno-
philes (14a ,b) recently described in the literature,3d are
listed in Table 1. Indeed, a preference for IED is found
for the interaction between dienes 2, 13a , and 13c with
the dienophiles substituted with electron-withdrawing
groups 14a and 14b (Table 2, entries 1, 2, 5, 6, 13, and
14). For the reaction of 13b with dienophiles 14a and
14b the normal electron-demand (NED) interaction is
preferred (Table 2, entries 9 and 10), although the
preference is only of 0.1415 and 0.2400 eV, respectively,
with respect to the IED interaction.
SCHEME 4
11 as the only product (47%) (Scheme 4). In this case,
the orthoquinone monoketal intermediate 12 undergoes
intramolecular Diels-Alder reaction to give the exo
product by reaction on the unsubstituted double bond of
the furan ring.
Table 1 also lists the relative magnitudes of the HOMO
and LUMO coefficients of dienes 2and 13a -c and dieno-
philes 14a ,b. The regioselectivity predicted for dienes 2,
13a , and 13c does not agree with the experimental
results, except for the interactions between diene 13b and
olefins 14a and 14b. These results parallel those reported
previously on the basis of 3-21G* calculations.3d The
unreliability of the theoretical prediction may be due to
the small differences in energy gaps between NED and
IED interactions (0.14 to 0.76 eV) (Table 2), leading to
an unclear preference between them.
The structure of 11 was established by 1H and 13C
NMR spectral analyses. The signal at 6.30 ppm, corre-
sponding to a single vinylic furan proton, confirms the
regioselectivity of the reaction that occurs exclusively at
the unsubstituted double bond of the furan ring. On the
other hand, the regiochemistry of the cycloaddition was
clearly established by 1H-1H decoupling experiments.
The coupling observed between H-1 (4.19 ppm) and H-2
(4.65 ppm) (Scheme 4) confirms the position of the oxygen
atom of the furan ring.
The regioselectivity of the Diels-Alder cycloadditions
of dienes 2, 13b, and 13c (Figure 2) with dienophiles 14a
and 14b, which bear electron-withdrawing substituents,
has been analyzed in terms of FMO theory.3d An agree-
ment between ab initio (HF/3-21G*) calculations and
experimental results was found only for diene 13b, which
is substituted with a poor electron-donor group (methyl
group). When the FMO model was applied to the study
of the cycloadditions of dienes 2 and 13a -c, with dieno-
philes substituted with electron-donor groups, such as
15c, the predicted regioselectivity matched the experi-
mental results.4a The main trend predicted for both series
of dienophiles, 14 and 15, was an energetically more
favorable interaction between the LUMO of the diene and
the HOMO of the dienophile, i.e., the cycloadditions
should take place under inverse electron demand.27
However, the energy gaps for both possible HOMO-
LUMO interactions between the diene and dienophiles
substituted with electron-withdrawing groups were found
(15) (a) Liu, W. C.; Liao, C. C. Synlett 1998, 912-913. (b) Lee, T.
H.; Liao, C. C. Tetrahedron Lett. 1996, 37, 6869-6872.
(16) Liao, C. C.; Wei, C. P. Tetrahedron Lett. 1989, 30, 2255-2258.
(17) Liu, W. C.; Liao, C. C. Chem. Commun. 1999, 117-118.
(18) Hsu, D. S.; Hsu, P. Y.; Liao, C. C. Org. Lett. 2001, 3, 263-265.
(19) Yen, C. F.; Liao, C. C. Angew. Chem., Int. Ed. 2002, 41, 4090-
4093.
(20) Chu, C. S.; Liao, C. C.; Rao, P. D. Chem. Commun. 1996, 1537-
1538.
(21) (a) Churcher, I.; Hallett, D.; Magnus P. Tetrahedron 1999, 55,
1597-1606. (b) Churcher, I.; Hallett, D.; Magnus P. J . Am. Chem. Soc.
1998, 120, 3518-3519.
(22) Horne, D. A.; Yakushijin, K.; Bu¨chi, G. Tetrahedron Lett. 1999,
40, 5443-5447.
(23) Ku¨rti, L.; Szila´gyi, L.; Antus, S. Eur. J . Org. Chem. 1999, 2579-
2581.
(24) Njardarson, J . T.; McDonald, I. M.; Spiegel, D. A.; Inoue, M.;
Wood, J . L. Org. Lett. 2001, 3, 2435-2438.
(25) Sutherland, H. S.; Souza, F. E. S.; Rodrigo, R. G. A. J . Org.
Chem. 2001, 66, 3639-3641.
(26) Domingo, L. R.; Aurell, M. J . J . Org. Chem. 2002, 67, 959-
965.
(27) Fleming, I. Frontier Orbitals and Organic Chemical Reactions;
J ohn Wiley & Sons: Chichester, UK, 1976.
2350 J . Org. Chem., Vol. 69, No. 7, 2004

Diels-Alder Reactions of Masked o-Benzoquinones
TABLE 1. Ab In itio HF /6-31G* En er gies (eV) a n d Coefficien ts (Ci) of th e F r on tier Molecu la r Or bita ls for Dien es 2 a n d
13a -c a n d Dien op h iles 14a ,b a n d 15a ,ba
HOMO
LUMO
C3
compdb
E
C1
C2
C3
C4
∆Cic
E
C1
C2
C4
∆Cic
2
-9.5241
-9.5186
0.3173
0.2597
0.2476 -0.2206 -0.2713
0.1853 -0.2941 -0.2474
0.0460 1.4994
0.1879 -0.2029 -0.1880
0.2882 -0.1003
0.1230 -0.1923 -0.0238
0.1537 -0.2167 -0.0158
13a
13b
13c
14a
14b
15a
15b
0.0123 1.4422 -0.1685
0.2228
0.2615
-9.2084 -0.2890 -0.1943
0.2551
0.3166 -0.0274 1.6463 -0.2009
-9.7118
-10.3894
-10.6370
-9.0288
0.3145
0.3391
0.3432
0.3810
0.2293 -0.2301 -0.2718
0.0427 1.2463
0.1697 -0.2216 -0.1679
0.3105 -0.1961 -0.2947
0.3295 -0.2249 -0.2840
0.2733 -0.1036
0.3613 -0.0423 -0.2438 -0.0222 2.6858
0.3715 -0.0239 -0.2172 -0.0283 2.8349
0.2995
0.2582
0.2188
0.1144
0.1046
0.0815 5.5267 -0.3053
0.0902 4.7620 -0.3247
0.3753
0.3509
-0.0700
-0.0262
-8.3812 -0.2875 -0.1973
a
b
These are the values of the pz coefficients; the relative pz′ contributions and their ∆Ci are analogous. For the most stable planar
s-cis conformation for olefins 14a and 14b. c Carbon 1-carbon 2 for the olefins; carbon 1-carbon 4 for the dienes.
TABLE 2. Ab In itio HF /6-31G* En er gy Ga p s (eV)
betw een th e Releva n t F r on tier Or bita ls for Dien es 2 a n d
13a -c a n d Dien op h iles 14a ,b a n d 15a ,b
F IGURE 3. The four possible transition states for the
additions between dienes 2, 13a , and 13b and dienophile 15a .
NEDa
IEDa
HOMO- LUMO-
Despite the agreement between the predicted and
experimental regioselectivity for the strong IED interac-
tion between partners 2, 13b, and 13c and 15a ,b, the
FMO model is unable to account for the high endo
stereoselectivity found for all these cycloadditions, since
it is limited to the evaluation of the ground-state interac-
tions. The interactions controlling the endo or exo selec-
tivity in these processes are the result of a subtle
interplay of electronic and steric effects at the transition
state (TS). Thus, calculation of the four possible transi-
tion states for the reactions between dienes 2, 13a , and
13b and dienophile 15a were carried out, using the HF/
3-21G* level of theory (Figure 3). The energies of the TSs
are shown in Table 3.
entry diene olefin LUMO
HOMO
diffb
interactionc
1
2
3
4
2
2
2
2
14a
14b
15a
15b
12.2099 11.8888
12.3590 12.1364
15.0508 10.5282
0.3211
0.2226
4.5226
4.4155
IED
IED
IED
IED
14.2961
9.8806
5
6
7
8
13a
13a
13a
13a
14a
14b
15a
15b
12.2044 11.8316
12.3535 12.0792
15.0453 10.4688
0.3728
0.2743
4.5765
4.4572
IED
IED
IED
IED
14.2806
9.8234
9
10
11
12
13b
13b
13b
13b
14a
14b
15a
15b
11.8942 12.0357 -0.1415
12.0433 12.2833 -0.2400
14.7351 10.6751
13.9704 10.0275
NED
NED
IED
4.0600
3.9489
IED
13
14
15
16
13c
13c
13c
13c
14a
14b
15a
15b
12.3976 11.6357
12.5467 11.8833
15.2385 10.2751
0.7619
0.6634
4.9634
4.8463
IED
IED
IED
IED
In contrast to previous calculations of the related
cycloaddition of 2 with 2-methylfuran,26 where an inter-
mediate was detected, the potential energy surfaces for
each set of four different TSs, corresponding to additions
between dienes 2, 13a , and 13b to dienophile 15a ,
showed an asynchronous concerted mechanism. In the
case of 13a , the geometries of the two ortho TSs (con-
ducting to the observed regioisomers) indicate that the
σ bond between carbon C-4 of the diene and carbon C-8
(for numbering, see Figure 3) of the dienophile is more
developed than that between carbon C-1 of the diene and
C-7 of the olefin (Figure 4), as suggested by the atom
distances measured between these reaction centers. In
the meta TSs, both distances are very similar, suggesting
a more synchronous process. The internuclear distances
for the addition of 15a to diene 13b, in the four TSs, are
similar to those shown in Figure 4 for diene 13a (for
14.4738
9.6275
a
HOMO-diene/LUMO-dienophile and LUMO-diene/HOMO-di-
enophile. (HOMO-LUMO) - (LUMO-HOMO). c Electron de-
mand: NED ) normal electron demand (HOMO-diene-LUMO-
dienophile); IED ) inverse electron demand (LUMO-diene-
HOMO-dienophile).
b
In contrast to the above, the difference in energy gaps
for the interactions between all of the dienes and the
dienophiles substituted with electron-releasing groups,
15a ,b, range from 3.94 to 4.96 eV, clearly indicating a
preference for the IED interaction (Table 2). The largest
coefficient in the LUMO of the dienes is located at C4,
while the relative magnitude of the largest coefficient in
the HOMO of the dienophiles is located at the double
bond terminus C1 (Table 1). Therefore, the predicted
regioselectivity corresponds to the observed ortho orien-
tation (Figure 3).
example, in the ortho-endo TS: d1-7 ) 2.552 Å; d4-8
2.024 Å).
)
J . Org. Chem, Vol. 69, No. 7, 2004 2351

Arjona et al.
TABLE 3. Ab In itio HF /3-21G Electr on ic (Ee) a n d
Electr on ic + Zer o-P oin t En er gies (ZP E) (E0) (a u ), a n d
Rela tive En er gies (∆E a n d ∆E0)a (k ca l/m ol) of th e
Tr a n sition Sta tes (TS) for th e Cycloa d d ition of Dien es 2,
13a , a n d 13b a n d Dien op h ile 15a
diene
TS
ortho-endo -985.350634 -984.998348 0.000 0.000
ortho-exo -985.346931 -984.995045 2.324 2.073
meta-endo -985.339189 -984.987065 7.182 7.080
meta-exo -985.318261 -984.966944 20.314 19.706
ortho-endo -759.953175 -759.647365 0.000 0.000
ortho-exo -759.951876 -759.645854 0.815 0.948
meta-endo -759.951199 -759.645299 1.240 1.296
meta-exo -759.932177 -759.626673 13.176 12.984
Ee
E0
∆Ee
∆E0
2
2
2
2
13a
13a
13a
13a
13b ortho-endo -798.773788 -798.438778 0.000 0.000
13b ortho-exo -798.772464 -798.437229 0.831 0.972
13b meta-endo -798.772639 -798.437324 0.721 0.912
13b meta-exo
-798.753502 -798.418898 12.730 12.475
a
Relative to the ortho-endo TS.
The presence of a stronger electron-withdrawing (meth-
oxycarbonyl) group at C-3 enhances the asynchrony of
the reaction, particularly for the ortho approaches. This
may be interpreted in terms of the synergistic effects of
the methoxycarbonyl at C-5 and the carbonyl group at
C-2, which reduce the electron density at C-4 (see Table
2), increasing at the same time the positive charge at this
position. Thus, a stronger interaction would be expected
between C-4 in the MOB and C-8 in 15a (Figure 3), which
is the more electron-rich position in the double bond of
the dienophile.
F IGURE 4. Transition states of the four possible approaches
for the cycloaddition between diene 13a and dienophile 15a .
In the case of the cycloaddition between diene 2 and
olefin 15a , the TSs for the unobserved meta adducts were
found to be slightly asynchronous as well, with bond
distances similar to those found for dienes 13a and 13b
(for example, in the meta-endo TS: d1-7 ) 2.321 Å; d4-8
) 2.111 Å). However, the asynchrony was much more
accentuated for the transition states leading to the ortho-
endo and ortho-exo isomers, as can be seen in the bond
distances averaged over the two TSs: d1-7 ) 2.9881 Å
and d4-8 ) 1.9599 Å. The latter was quite a bit shorter
than those calculated for additions of dienes 13a and 13b
while the former, d1-7, was much longer (see Figure 4).
From the electronic energies obtained for the additions
of the three dienes, it turned out that the ortho-endo TS
was more stable than the ortho-exo TS (Table 3). The
electronic energies including the zero-point energy cor-
rection (E0) are also shown in Table 3, and their values
suggest a preference for the ortho-endo TS of ca. 0.96 kcal/
mol for 13a and 13b, while the difference is larger (2.07
kcal/mol) for diene 2 (Table 3). Moreover, the relative
magnitude of the TS energies for the corresponding meta
regioisomers leads to a similar prediction: the endo TS
should be more stable than the exo TS.
Even though the electronic energy differences (∆E)
between the ortho-endo/ ortho-exo TSs for dienes 13a and
13b are relatively small, a preferential trend for the endo
TS is found. The magnitude of the ∆E for diene 2 is,
however, larger, indicating a clear preference for the
ortho-endo approach. The ∆E0 is slightly more differenti-
ated with respect to the stability of the TSs for the ortho/
meta regioisomers for diene 13a . For the endo ap-
proaches, the ortho isomer was ca. 1.30 kcal/mol more
stable than the meta TS (Table 3). This difference is
largely enhanced (ca. 7.08 kcal/mol) for diene 2. This was
not the case for diene 13b, which is slightly smaller.
However, in all cases a much larger difference was
calculated for the meta-exo approaches, with respect to
the energetically more stable ortho-endo TS. Therefore,
it is clear than these TS calculations agree with the FMO
prediction for an ortho preference, as experimentally
observed.
This result suggests that in these cases, the cycload-
ditions proceed through a polar TS, like the one recently
calculated for Domingo et al. for the reaction of diene 2
with 2-methylfuran.26 For the reaction between 2 and
15a , the Mu¨lliken charges determined at the ortho-endo
TS for carbons C-1 (donor, -0.41) and C-7 (acceptor, 0.34)
[C-1 (donor, -0.40) and C-7 (acceptor, 0.33) for the ortho-
exo TS] would support this hypothesis. However, neither
a zwitterionic intermediate nor a second TS were located
in the potential energy surface of this reaction. Instead
of a second step, the meticulous analysis of the energy
surface showed a coherent downward trajectory from the
TSs to the adducts. Therefore, we can propose that the
cycloadditions conducing to the ortho regioisomers pro-
ceed via a nonsychronous concerted pathway, reaching
a dipolaroid-like transition state, which leads directly to
products, without the presence of a charged intermediate.
It appears then that the concertedness of these pro-
cesses depends not only on the dienophile, as is the case
for 2-methylfuran vs the ethyl vinyl ether (15a ), but also
on the electron demand of the substituents at the diene.
Secondary orbital interactions have been invoked to
explain the endo preference, even for dienophiles bearing
2352 J . Org. Chem., Vol. 69, No. 7, 2004

Diels-Alder Reactions of Masked o-Benzoquinones
1
solutions. H and 13C NMR spectra were recorded at 300 and
50 MHz, respectively, in CDCl3 solution with TMS as internal
reference.
The ab initio calculations were carried out with the Gauss-
ian 94 program package30 running on personal computers
under the Linux operating system. Preliminary geometry
optimizations of the starting materials and products were
carried out first with the AM1 semiempirical method, and
these geometries were employed as starting points for opti-
mization with the HF/3-21G and HF/6-31G* models. FMO
energies and coefficients were determined at the HF/6-31G*
level of theory. Characterization of the potential energy surface
of this process was carried out with the HF/3-21G model.
Optimization of reactants, products, and transition states was
followed by frequency analysis to ensure the correct nature of
the stationary points. For the transition states, the normal
modes corresponding to the imaginary frequencies were ana-
lyzed visually to confirm that they led to the desired products.
For the particular case of both ortho approaches in the reaction
between 2 and 15a , IRC calculations were carried out, starting
from the corresponding transition states, as well as an
extensive potential surface scan (HF/3-21G); no additional
stationary points were located.
F IGURE 5. Interaction sites in the ortho-exo Diels-Alder
transition state for the addition between dienes 2, 13a , and
13b, with dienophile 15a .
electron-donating groups.28 In our case, however, the
geometries of the endo TSs do not show any particularly
short distance between the diene and the ether group of
the dienophile that could reveal a possible secondary
interaction.
It seems that, for our additions, steric interactions29
developed at the exo TS between the ethoxy (or thioet-
hoxy) group of the vinyl ether (or vinyl thioether) of the
dienophile and those of the ketal substituent on the
ethano bridge of the diene can destabilize such an
approach (Figures 4 and 5), favoring the endo TS. This
factor could also explain the high energies calculated for
the meta-exo TSs, where the alkoxy groups of diene and
dienophile are in close proximity, thus enhancing the van
der Waals interactions.
Syn th esis of 2,3-exo-Bis(ter t-bu tyld im eth ylsilyloxym -
eth yl)-7-oxa bicyclo[2.2.1]h ep t-5-en e, 5. To a solution of
7-oxabicyclo[2.2.1]hept-5-ene-cis-exo-2,3-dimethanol31 (490 mg,
3.14 mmol) in 31 mL of CH2Cl2 were added 1.09 mL (7.85
mmol) of Et3N and 1.8 mL (7.85 mmol) of tert-butyldimethyl-
silyl trifluoromethanesulfonate. After the mixture was stirred
for 1 h, the reaction was quenched with K2CO3, 5% aqueous
NaHCO3, and water and extracted with CH2Cl2. The combined
organic layers were dried over MgSO4 and concentrated to give
a crude residue, which was purified by column chromatography
(hexane:EtOAc 10:1) to afford 936 mg of 5 as a pale yellow oil
Con clu sion s
The Diels-Alder reactions of the masked o-benzo-
quinone 2 with new electron-rich dienophiles, including
bicyclic systems, have been studied. In addition, we
describe for the first time the intramolecular Diels-Alder
reaction of furan and a MOB. It should be pointed out
that this process occurs with total regio- and stereose-
lectivity.
1
(78%). IR (CHCl3) ν 1637, 1257, 1047 cm-1. H NMR (CDCl3,
300 MHz) δ 0.06 (s, 6 H), 0.07 (s, 6 H), 0.91 (s, 18 H), 1.78 (t,
2 H, J ) 5.4 Hz), 3.54 (t, 2 H, J ) 9.3 Hz), 3.77 (dd, 2 H, J )
5.4, 9.3 Hz), 4.82 (s, 2 H), 6.35 (s, 2 H). 13C NMR (CDCl3, 50
MHz) δ -5.3, -5.3, 25.7, 25.9, 42.5, 62.4, 80.4, 135.5. Anal.
Calcd for C20H40O3Si2: C, 62.50; H, 10.42. Found: C, 62.63;
H, 10.34.
Syn th esis of 7-Oxa bicyclo[2.2.1]h ep t-5-en e-2-en d o-ca r -
boxylic Acid Meth yl Ester , 6. To a solution of 7-oxabicyclo-
[2.2.1]hept-5-ene-2-endo-carboxylic acid32 (1.91 g, 13.6 mmol)
and MeOH (0.6 mL, 15 mmol) in 41 mL of CH2Cl2 cooled to 0
°C was added a solution of 1,3-dicyclohexylcarbodiimide (2.8
g, 13.6 mmol) and (dimethylamino)pyridine (17 mg, 0.14 mmol)
in CH2Cl2 (41 mL) dropwise. The reaction mixture was stirred
for 24 h and then filtered, washed with Et2O, and concentrated.
The residue was purified by chromatography (hexane:Et2O 5:1)
to afford 1.26 g of 6 as a pale yellow oil (60%). IR (CHCl3) ν
The FMO model allows us to predict the correct
regioselectivity for the Diels-Alder cycloadditions of
dienes, 2 and 13a -c, with the dienophiles substituted
by strong electron-releasing groups, 15a ,b, under inverse
electron-demand conditions. However, this theoretical
approach fails to rationalize the orientation observed with
dienophiles substituted with electron-withdrawing groups,
such as 14a and 14b. Ab initio calculations of the four
possible transition states for the cycloaddition of diene
13a with dienophile 15a provide a rationalization of the
regioselectivity and the high endo stereoselectivity ex-
perimentally observed. In addition, these calculations
suggest that an asynchronous concerted transition state
is involved in this type of process.
2849, 1736, 1630, 1132 cm-1 1H NMR (CDCl3, 300 MHz) δ
.
1.52 (dd, 1 H, J ) 3.9, 11.2 Hz), 2.04 (ddd, 1 H, J ) 4.9, 9.3,
11.2 Hz), 3.06 (dt, 1 H, J ) 4.9, 9.1 Hz), 3.58 (s, 3 H), 4.96 (dd,
1 H, J ) 1.2, 4.6 Hz), 5.10 (dd, 1 H, J ) 1.5, 4.9 Hz), 6.16 (dd,
1 H, J ) 1.5, 5.9 Hz), 6.38 (dd, 1 H, J ) 1.5, 5.9 Hz). 13C NMR
(CDCl3, 50 MHz) δ 28.3, 42.4, 51.5, 78.5, 78.7, 132.3, 136.8,
172.3. Anal. Calcd for C8H10O3: C, 62.34; H, 6.49. Found: C,
62.43; H, 6.35.
Exp er im en ta l Section
Syn t h esis of Bicyclo[2.2.2]oct en on es 7-10: Gen er a l
P r oced u r e. A solution of methylvanillate (0.4 mmol) and the
Reagents obtained from commercial sources were used as
received, and solvents were dried prior to use. Silica gel 60
F254 was used for TLC, and the spots were detected with UV
or KMnO4 solution. Flash column chromatography was carried
out on silica gel 60. IR spectra have been recorded as CHCl3
(30) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.;
J ohnson, B. G.; Robb, M. A.; Cheeseman, J . R.; Keith, T.; Petersson,
G. A.; Montgomery, J . A.; Raghavachari, K.; Al-Laham, M. A.;
Zakrzewski, V. G.; Ortiz, J . V.; Foresman, J . B.; Cioslowski, J .;
Stefanov, B. B.; Nanayakkara, A.; Challacombe, M.; Peng, C. Y.; Ayala,
P. Y.; Chen, W.; Wong, M. W.; Andres, J . L.; Reploge, E. S.; Gomperts,
R.; Martin, R. L.; Fox, D. J .; Binkley, J . S.; Defrees, D. J .; Baker, J .;
Stewart, J . P.; Head-Gordon, M.; Gonzalez, C.; Pople, J . A. Gaussian
94; Gaussian, Inc.: Pittsburgh, PA, 1995.
(31) Das, J .; Vu, T.; Harris, D. N.; Ogletree, M. L. J . Med. Chem.
1988, 31, 930-935.
(32) Moore, J . A.; Partain, E. M., III J . Org. Chem. 1983, 48, 1105-
1106.
(28) (a) Apeloig, Y.; Matzner, E. J . Am. Chem. Soc. 1995, 117, 5375-
5376. (b) Arrieta, A.; Cossio, F. P.; Lecea, B. J . Org. Chem. 2001, 66,
6178-6180. (c) Caramella, P.; Quadrelli, P.; Toma, L. J . Am. Chem.
Soc. 2002, 124, 1130-1131.
(29) (a) Cantello, B. C. C.; Mellor, J . M.; Webb, C. F. J . Chem. Soc.,
Perkin Trans. 2 1974, 22-25. (b) Fringuelli, F.; Guo, M.; Minuti, L.;
Pizzo, F.; Taticchi, A.; Wenkert, E. J . Org. Chem. 1989, 54, 710-712.
J . Org. Chem, Vol. 69, No. 7, 2004 2353

Arjona et al.
corresponding dienophile (10 mmol) in 1.2 mL of MeOH was
heated to 50 °C. Then, 1.2 mmol of (diacetoxy)iodobenzene in
3.6 mL of MeOH was added via a syringe pump over 1.5 h.
After the mixture was stirred for 10 min, the solvent was
removed under reduced pressure. The residue was purified by
chromatography (hexane:EtOAc 10:1).
tr id ec-9-en e-9-ca r boxylic Acid Meth yl Ester , 10a , a n d
(1S*,2S*,3R*,4S*,6S*,7R*,8S*)-12,12-Dim et h oxy-4-en d o-
(m eth oxyca r bon yl)-11-oxo-13-oxa tetr a cyclo[6.2.2.13,6.02,7]-
t r id ec-9-en e-9-ca r b oxylic Acid Met h yl E st er , 10b . Pale
1
yellow oil, 70%. IR (CHCl3) ν 2839, 1736, 1161, 1146 cm-1. H
NMR (CDCl3, 300 MHz) δ 2.00-1.82 (m, 4 H), 2.44-2.39 (m,
2 H), 2.62 (dd, 2 H, J ) 2.9, 7.8 Hz), 3.00-2.91 (m, 2 H), 3.25
(s, 6 H), 3.34 (s, 3 H), 3.37 (s, 3 H), 3.42 (dd, 1 H, J ) 2.4, 6.3
Hz), 3.46 (dd, 1 H, J ) 2.4, 6.3 Hz), 3.71 (s, 3 H), 3.74 (s, 3 H),
3.79 (s, 3 H), 3.80 (s, 3 H), 3.87 (t, 1 H, J ) 2.4 Hz), 3.92 (t, 1
H, J ) 2.4 Hz), 4.39 (d, 2 H, J ) 4.4 Hz), 4.51 (d, 2 H, J ) 5.9
Hz), 7.03 (dt, 2 H, J ) 2.4, 6.3 Hz). 13C NMR (CDCl3, 50 MHz)
δ 33.4, 34.0, 38.4, 41.5, 41.9, 42.1, 42.3, 45.3, 47.9, 48.2, 50.0,
50.1, 50.3, 50.3, 52.0, 52.1, 52.2, 52.4, 52.8, 80.5, 81.1, 81.2,
81.6, 93.7, 93.7, 134.4, 135.0, 135.1, 164.8, 171.9, 172.0, 199.3.
Anal. Calcd for C18H22O8: C, 59.02; H, 6.01. Found: C, 58.93;
H, 6.12.
(1R *,2S *,6R *,7S *)-11,11-Dim e t h oxy-4,10-d ioxo-3,5-
d ioxa t r icyclo[5.2.2.02,6]u n d ec-8-en e-8-ca r b oxylic Acid
Meth yl Ester , 7. Pale yellow oil, 27%. IR (CHCl3) ν 2852,
1
1751, 1718, 1630 cm-1. H NMR (CDCl3, 300 MHz) δ 3.36 (s,
3 H), 3.37 (s, 3 H), 3.84 (s, 3 H), 3.92-3.95 (m, 1 H), 4.39 (m,
1 H), 5.08 (dd, 1 H, J ) 2.7, 7.8 Hz), 5.33 (dd, 1 H, J ) 3.7, 7.8
Hz), 7.17 (d, 1 H, J ) 4.4 Hz). 13C NMR (CDCl3, 50 MHz) δ
31.9, 42.0, 50.3, 52.6, 53.4, 73.6, 74.1, 92.5, 133.9, 135.0, 153.5,
163.0, 194.3. Anal. Calcd for C13H14O8: C, 52.35; H, 4.70.
Found: C, 52.27; H, 4.61.
(1R*,2R*,3R*,6S*,7S*,8R*)-13-Meth an e-12,12-dim eth oxy-
11-oxot et r a cyclo[6.2.2.13,6.02,7]t r id ec-9-en e-9-ca r b oxylic
Acid Meth yl Ester , 8. Pale yellow oil, 68%. IR (CHCl3) ν 2837,
Syn th esis of (1R*,2S*,7S*,9S*,12S*)-7-Meth oxy-8-oxo-
3,6-d ioxa tetr a cyclo[5.4.21,9.14,12.01,7]tr id eca -4,10-d ien e-11-
ca r boxylic Acid Meth yl Ester , 11. A solution of (diacetoxy)-
iodobenzene (198 mg, 0.61 mmol) and 2-hydroxymethylfuran
(0.65 mL, 7.5 mmol) in CH2Cl2 (0.8 mL) was heated to reflux.
Then, methylvanillate (75 mg, 0.41 mmol) in 0.6 mL of CH2-
Cl2 was added via a syringe pump over 3 h. After being stirred
12 h at reflux, the reaction mixture was quenched with
saturated aqueous NaHCO3 and saturated aqueous NaCl and
extracted with CH2Cl2. The combined organic layers were dried
over MgSO4 and concentrated to give a crude residue, which
was purified by column chromatography (hexane:EtOAc 5:1)
to afford 53 mg of 11 as a pale yellow oil (47%). IR (CHCl3) ν
1
1740, 1713, 1296 cm-1. H NMR (CDCl3, 300 MHz) δ 0.76 (d,
1 H, J ) 10.1 Hz), 1.11-1.27 (m, 2 H), 1.45 (dd, 2 H, J ) 4.4,
12.7 Hz), 1.72 (d, 1 H, J ) 10.1 Hz), 2.08 (dd, 1 H, J ) 1.5, 8.8
Hz), 2.15 (br s, 1 H), 2.20 (br s, 1 H), 2.23 (dd, 1 H, J ) 2.9,
8.3 Hz), 3.24 (s, 3 H), 3.38 (s, 3 H), 3.36-3.39 (m, 1 H), 3.81
(s, 3 H), 3.81-3.82 (m, 1 H), 7.05 (dd, 1 H, J ) 1.5, 7.5 Hz).
13C NMR (CDCl3, 50 MHz) δ 30.4, 31.6, 35.6, 40.4, 41.1, 42.1,
42.9, 45.3, 49.9, 50.3, 52.0, 54.1, 94.2, 136.3, 137.4, 164.6, 200.5.
Anal. Calcd for C17H22O5: C, 66.67; H, 7.19. Found: C, 66.78;
H, 7.31.
(1S*,2S*,3R*,4S*,5R*,6R*,7R*,8S*)-4,5-Bis(ter t-bu tyldim -
e t h y ls i ly lo x y m e t h y l)-12,12-d i m e t h o x y -11-o x o -13-
oxa tetr a cyclo[6.2.2.13,6.02,7]tr id ec-9-en e-9-ca r boxylic Acid
Meth yl Ester , 9. Pale yellow oil, 70%. IR (CHCl3) ν 2858,
2856, 1751, 1718, 1138 cm-1 1H NMR (CDCl3, 300 MHz) δ
.
3.54 (s, 3 H), 3.71 (dd, 1 H, J ) 3.3, 7.0 Hz), 3.77-3.81 (m, 1
H), 3.83 (s, 3 H), 3.95 (d, 1 H, J ) 7.4 Hz), 4.18 (d, 1 H, J )
7.5 Hz), 4.19 (d, 1 H, J ) 1.9 Hz), 4.65 (t, 1 H, J ) 2.2 Hz),
6.30 (dd, 1 H, J ) 2.2, 2.9 Hz), 7.02 (dd, 1 H, J ) 2.2, 7.0 Hz).
13C NMR (CDCl3, 50 MHz) δ 45.4, 50.9, 51.5, 52.8, 72.6, 77.6,
89.4, 97.5, 99.8, 132.1, 137.8, 149.1, 164.6, 198.9. Anal. Calcd
for C14H14O6: C, 60.43; H, 5.03. Found: C, 60.52; H, 4.91.
1
1736, 1711, 1259 cm-1. H NMR (CDCl3, 300 MHz) δ 0.01 (s,
6 H), 0.02 (s, 6 H), 0.87 (s, 9 H), 0.88 (s, 9 H), 1.94-2.08 (m, 2
H), 2.38 (dd, 1 H, J ) 1.9, 8.3 Hz), 2.55 (dd, 1 H, J ) 2.9, 8.3
Hz), 3.26 (s, 3 H), 3.43-3.34 (m, 1 H), 3.38 (s, 3 H), 3.47 (dd,
1 H, J ) 2.9, 6.8 Hz), 3.45-3.53 (m, 1 H), 3.56-3.58 (m, 2 H),
3.79 (s, 3 H), 3.93 (t, 1 H, J ) 2.4 Hz), 4.16 (s, 1 H), 4.23 (s, 1
H), 7.04 (dd, 1 H, J ) 1.5, 6.8 Hz). 13C NMR (CDCl3, 50 MHz)
δ -5.4, -5.4, 18.1, 18.2, 25.8, 25.9, 41.9, 42.0, 44.8, 49.5, 50.1,
50.2, 52.0, 52.8, 60.4, 60.6, 60.9, 82.1, 82.9, 93.8, 134.0, 135.1,
164.8, 199.8. Anal. Calcd for C30H52O8Si2: C, 60.40; H, 8.72.
Found: C, 60.54; H, 8.61.
Ack n ow led gm en t. The Ministerio de Ciencia y
Tecnolog´ıa of Spain (Project BQU2000-0653) is grate-
fully thanked for financial support. J .T. and H.A.J .-V.
are fellows of the EDD/IPN and COFAA/IPN programs.
(1S*,2S*,3S*,5R*,6R*,7R*,8S*)-12,12-Dim eth oxy-5-en d o-
(m eth oxyca r bon yl)-11-oxo-13-oxa tetr a cyclo[6.2.2.13,6.02,7]-
J O030307R
2354 J . Org. Chem., Vol. 69, No. 7, 2004