Application of D-chiro-inositol as a chiral template for the diels-alder reaction

Article abstract of DOI:10.1002/hlca.201200432

The Diels-Alder reaction can reliably provide the expected endo-product in the presence of secondary orbital overlap. It can be considerably more difficult to access a single enantiomer of the exo-product. In this paper, a D-chiro-inositol derivative is used as a chiral tether to facilitate the regio-, diastereo-, and enatioselective cycloaddition between cinnamic acid and hexa-3,5-dienoic acid. The Diels-Alder reaction between these two substrates, or their respective esters, does not occur under thermal conditions. Because of the ease of removal of the chiral tether from the resulting cyclohexene, this approach could provide a viable technique to access otherwise unavailable systems. Copyright

Full text of DOI:10.1002/hlca.201200432

2026
Helvetica Chimica Acta – Vol. 95 (2012)
Application of d-chiro-Inositol as a Chiral Template for the DielsꢀAlder
Reaction
by John F. Trant, Lee Belding, Travis Dudding, and Tomas Hudlicky*
Department of Chemistry and Centre for Biotechnology, Brock University, 500 Glenridge Avenue,
St Catharines ON, L2S 3A1, Canada
(phone: þ 1905688-5550 (ext. 3406); fax: þ 1905682-9020; e-mail: thudlicky@brocku.ca)
Dedicated to Professor Dieter Seebach on the occasion of his 75th birthday and in recognition of his many
contributions to our discipline
The DielsꢀAlder reaction can reliably provide the expected endo-product in the presence of
secondary orbital overlap. It can be considerably more difficult to access a single enantiomer of the exo-
product. In this paper, a d-chiro-inositol derivative is used as a chiral tether to facilitate the regio-,
diastereo-, and enatioselective cycloaddition between cinnamic acid and hexa-3,5-dienoic acid. The
DielsꢀAlder reaction between these two substrates, or their respective esters, does not occur under
thermal conditions. Because of the ease of removal of the chiral tether from the resulting cyclohexene,
this approach could provide a viable technique to access otherwise unavailable systems.
1. Introduction. – The use of chiral tethers connecting two reactive moieties
provides an excellent solution for asymmetric induction in CꢀC bond-forming reactions
[1]. Highly functionalized chiral cyclic systems, such as inositols, could serve as
templates for such applications. An ideal reaction for the investigation of the use of
inositols as chiral tethers is the DielsꢀAlder cycloaddition, in which the diene and the
dienophile can combine to form up to eight potential products, and four stereogenic
centers in fully substituted systems. In the DielsꢀAlder reaction, regiochemistry is often
controlled through the careful tuning of the electronic nature of the substituents, and
diastereoselectivity can often be controlled through secondary orbital overlap, but
regiochemical control is often less than perfect even for biased systems, and selectivity
is difficult to obtain in cases where one of the reagents lacks a large difference in the
orbital coefficients at the terminal positions of either the diene or dienophile [2][3].
Exclusive selectivity is of great interest, as the regioisomers and diastereoisomers often
have similar chemical and physical properties that complicate their isolation. The
formation of enantiomers, of course, leads to even greater purification difficulties. A
chiral template, such as d-chiro-inositol, may be ideal for conducting otherwise difficult
transformations to possibly provide only a single product. d-chiro-Inositol has been
previously used as a chiral auxiliary, but not, to the best of our knowledge, as a chiral
tether [4]. In this paper, we report the results of a stereoselective DielsꢀAlder
cycloaddition mediated by d-chiro-inositol.
2. Preparation of Cycloadduct 4a and Diol 5. – To investigate the above premise, the
reaction between cinnamic acid and hexa-3,5-dienoic acid was investigated. The
ꢀ 2012 Verlag Helvetica Chimica Acta AG, Zꢁrich

Helvetica Chimica Acta – Vol. 95 (2012)
2027
DielsꢀAlder reaction between these two substrates or between their corresponding
esters has not been reported, although more complex cinnamyl and more highly
substituted b,g,d,e-unsaturated esters have been subjected to DielsꢀAlder reactions [5].
The resulting cycloadducts would be of considerable interest as they contain three
stereogenic centers and an olefin that can be readily functionalized. However, there are
eight possible products that can form (i.e., A – H; Fig. 1). An attempt at an
intermolecular DielsꢀAlder reaction between cinnamic and hexadienoic acid was
unsuccessful; when the substrates were heated to 1008 for 96 h in 1,2,4-trichloroben-
zene, only the starting materials were recovered. At 1408, the hexadienoic acid
completely decomposed, and only the cinnamic acid was quantitatively recovered.
Similarly, when the methyl esters of the two substrates were subjected to reaction at
808, no reaction was observed after 24 h, and after an additional 48 h at 1408, the methyl
3,5-hexadienoate had fully decomposed leading to a quantitative recovery of unreacted
methyl cinnamoate from the reaction mixture. As neither the carboxylic acids nor their
respective methyl esters led to the formation of DielsꢀAlder products under thermal
conditions, d-chiro-inositol-derived templates were investigated to access the cyclo-
adducts.
Fig. 1. Eight possible cyclohexenes from the intramolecular DielsꢀAlder reaction of 3a
Epoxide 1 [6] was found to be chemically inert to many published ring-opening
techniques including Brønsted and Lewis acid catalysis, amine catalysis, hydrolysis with
KOH in DMSO, opening with cesium carboxylates, and reaction with neat carboxylic
acid in a melt [7], but it was cleanly converted to cinnamate 2a (Scheme) in the
presence of 18-crown-6, Bu4NI, and cinnamic acid. The reaction did not proceed
significantly without all the components present. A further analysis of the conditions
and results of epoxide opening under these conditions can be found in the

2028
Helvetica Chimica Acta – Vol. 95 (2012)
Scheme. Preparation of Cyclohexene 5 via Cycloadducts 4a and 4b
a) 3 Equiv. of potassium cinnamate, 3 equiv. of 18-crown-6, 1 equiv. of cinnamic acid, 0.2 equiv. of Bu4NI,
DMF, 1,2-dimethoxyethane, hexamethylphosphoric triamide, 1278, 96 h. b) Hexa-3,5-dienoic acid, N,N’-
dicyclohexylcarbodiimide (DCC), CH2Cl2. c) 1,3,5-Trichlorobenzene, 1418, 72 h. d) 1. LiAlH4, THF,
reflux, 4 h; 2. 60% AcOH, 808, 12 h; 3. MeONa/MeOH.
supplementary information. Standard N,N’-dicyclohaxylcarbodiimide (DCC) coupling
of the resulting alcohol provided fully substituted inositol 3a that was then subjected to
thermal DielsꢀAlder conditions. The reaction was first carried out in deuterated
toluene and DMSO, and monitored by NMR. Reaction mixtures were heated to 608,
and warmed by a further 108 every 4 h after NMR analysis showed no further
conversion. At 1308, trace conversion was noted after 4 h, and conversion increased
slowly after 12 h, providing what appeared to be a single compound. However, this
reaction provided insufficient material for characterization as it contained a significant
amount of decomposed material (presumably because of the instability of DMSO at
higher temperatures); consequently, the solvent was changed to 1,2,4-trichlorobenzene,
and, after 92 h at 140 – 1428, a single tricyclic compound (of the eight possibilities) was
isolated in 62% yield by chromatography as the only product, with the mass balance

Helvetica Chimica Acta – Vol. 95 (2012)
2029
recovered as unreacted starting material. The relative configuration of the cyclohexene
ring was assigned by NMR coupling constant analysis and is consistent with structures
C and G (Fig. 1). However, the conformational flexibility of the nine-membered ring
did not allow for an unambiguous assignment of the configuration using NOESY
studies.
3. Density-Functional-Theory (DFT) Analysis of the Intramolecular DielsꢀAlder
Reaction of 5a. – To better understand the origin of stereoselectivity in the above
DielsꢀAlder reaction, a set of hybrid DFT calculations were performed at the B3LYP/
6 – 31G(d) level of theory at 1008, using the IEFPCM method with the default
parameters of toluene to account for solvent effects. All geometrically viable
DielsꢀAlder transition states were considered; eleven transition states were located,
each possessing only a single imaginary frequency. Of these, at least one transition state
for each of the eight possible diastereoisomers was located. One of the most noticeable
structural features of these first-order saddle points is that all eleven were highly
asynchronous, yet concerted in nature, with C(3)ꢀC(4) bond-formation preceding that
of the C(1)ꢀC(2) bond. Furthermore, in those transition states where the ester groups
had their carbonyl O-atoms directed syn to one another, it was found that an
unfavorable OLP – OLP coulombic interaction forced the [6.9.6] ring system into a
distorted geometry.
The computed Gibbs free-energy values indicated that (R,S,S)-TS-C (Fig. 2) was
energetically favored. The only other transition state within 4 kcal/mol was (S,S,S)-TS-
D, which was higher in energy by 1.4 kcal/mol. To further substantiate these results,
single-point MP2/6-31G(d) calculations were performed on the optimized transition
state structures. Again, (R,S,S)-TS-C was preferred over (S,S,S)-TS-D by 1.9 kcal/mol,
and all other transition states were higher in energy by at least 4 kcal/mol. It is
noteworthy that the predicted stereochemical outcome of this reaction was not only
consistent with experiment, but was also made in advance of the stereochemical
assignment of the observed product. However, it is granted that the computed energetic
difference between (R,S,S)-TS-C and (S,S,S)-TS-D is slightly underestimated. In terms
Fig. 2. Calculated lowest-energy transition-state structures for the formation of diastereoisomer D (TS-D),
and diastereoisomer C (TS-C) at B3LYP/6-31G(d) level of theory with toluene as solvent

2030
Helvetica Chimica Acta – Vol. 95 (2012)
of the key metrics of both (R,S,S)-TS-C and (S,S,S)-TS-D, the ester carbonyl O-atoms
were aligned anti to one another, minimizing the aforementioned distortion in the
[6.9.6] ring system. Following the Intrinsic Reaction Coordinate (IRC), one observes
that the C(1)ꢀC(2) bond actually begins to form after the transition state. It is likely
that this asynchronicity arises primarily from steric constraints imposed by the chiral,
sugar based tether, as opposed to electronic effects. As such, those tethered transition
states which allow C(1) and C(2) to lie closer together, with minimal steric repulsion,
will be favored. The calculated transition-state structures show that geometrical
constraints imposed by the tether affect the allowed C(1) ··· C(2) interatomic distance
in (S,S,S)-TS-D more so than in (R,S,S)-TS-C. In (S,S,S)-TS-D, there exists an
unfavorable steric contact between H(40) of the diene and H(60) of the dienophile
(2.54 ꢂ). The result is an elongated C(1) ··· C(2) interatomic distance (2.70 ꢂ), beyond
that of an optimal bond-forming distance. As such, C(1)ꢀC(2) bond formation is driven
by the preceding C(3)ꢀC(4) bond-forming event, which subsequently forces C(1) and
C(2) into closer proximity. Conversely, due to the orientation of the diene in (R,S,S)-
TS-C, this steric contact does not exist. The closest contact with H(40) becomes O(38)
at 2.62 ꢂ, which still allows C(1) and C(2) to remain at a much closer distance of
2.36 ꢂ.
To confirm the predictions made by the molecular calculation, an X-ray crystal
structure was required. Unfortunately, neither 4a nor any derivative thereof formed a
crystal suitable for analysis under any of the recrystallization conditions tested.
However, when bromine derivative 4b was prepared using the same sequence as used
for 4a with para-bromocinnamic acid instead of cinnamic acid for the epoxide ring
opening, a crystalline solid suitable for x-ray analysis was obtained (Fig. 3)1). The
absolute configuration determined for this derivative demonstrates that the Dielsꢀ
Alder reaction does definitely occur through the predicted exo-transition state to
provide a cyclohexene with configuration of C (Fig. 3). This product would not be
expected from the intermolecular reaction between the corresponding esters, which
should lead to a racemic mixture consisting mostly of the endo-cycloadduct due to the
stabilization provided by the secondary orbital overlap. Consequently, this intra-
molecular methodology using d-chiro-inositol as a chiral tether provides access to a
cycloadduct that would be very difficult to obtain through other means.
This chiral tether is removed in several ways: reductive cleavage of the ester
functionalities of 4a with LiAlH4 provides cyclohexene derivative 5. Similarly,
hydrolysis of the esters can provide the dicarboxylate derivative. Consequently, this
methodology could provide access to substituted cyclohexenes in various oxidation
states, compounds that could serve as important building blocks for organic synthesis.
4. Remarks and Conclusions. – A stereoselective DielsꢀAlder cycloaddition was
performed by utilizing d-chiro-inositol as a reaction tether. The results of this
cycloaddition, which would not have been possible without the tether, matched the
prediction made by computational methods. Furthermore, it appears that not only this,
but no cycloadduct is not obtainable using either the parent carboxylic acids or their
1
)
CCDC-885508 contains the supplementary crystallographic data for this paper. These data can be
obtained free of charge via http://www.ccdc.cam.ac.uk/data_request/cif.

Helvetica Chimica Acta – Vol. 95 (2012)
2031
Fig. 3. X-Ray structure obtained for bromo-aryl derivative 4b corresponding to TS-C
respective methyl esters under thermal conditions. Consequently, d-chiro-inositol-
derived tethers can provide selective access to novel cycloadducts that are difficult to
obtain by other means.
The authors are grateful to Dr. Alan Lough, University of Toronto, for X-ray analysis and the
following agencies for financial support of current research: Natural Sciences and Engineering Research
Council of Canada (Idea to Innovation and Discovery Grants); Canada Research Chair Program;
Canada Foundation for Innovation (CFI); Research Corporation; Noramco, Inc.; TDC Research, Inc.;
TDC Research Foundation; Brock University; and the Ontario Partnership for Innovation and
Commercialization.
Experimental Part
1. General. Reactions were carried out under Ar in flame-dried glassware unless otherwise stated.
Solvents were distilled: CH2Cl2, DMF, Et3N, 1,2,4-trichlorobenzene, ortho-xylene, and pyridine from
CaH2; THF and 1,2-dimethoxyethane from Na/benzophenone. Other reagents were used from
commercial sources (Sigma-Aldrich, Alfa Aesar, Acros Organics, Oakwood Products, Inc.) without
further purification. Qual. TLC was conducted with pre-coated silica gel aluminum sheets (EMD silica
gel 60 F254), detection by UV, or by spraying with Hanessianꢃs stain, vanillin, or aq. KMnO4 soln. followed
by heating. M.p.: uncorrected. Flash chromatography (FC): silica gel SiliaFlash P60 from Silicycle (40 –
66 mm). Optical rotation: 1-dm cell at 20 – 258 and 589 nm with a Perkin-Elmer 341 polarimeter,
concentration (c): in g/100 ml. IR Spectra: in a KBr cell in soln. with a Perkin-Elmer Spectrum One
instrument; ˜n in cmꢀ1. 1H- and 13C-NMR spectra: at 300 or 600 MHz, and 75 or 150 MHz, resp. on Bruker
AVANCE spectrometers; calibrated on the solvent residual peak or TMS signal (CDCl3: 7.26 ppm); the
chemical shifts in ppm, J in Hz. MS: Concept 1S spectrometer (Dr. Tim Jones, Brock University) or on a
Micromass GCT spectrometer (Dr. Kirk Green, McMaster University); m/z (rel. %). Elemental analysis
(C and H): carried out by Atlantic Microlab, Inc., 6180 Atlantic Blvd. Suite M, Norcross, GA 30071.

2032
Helvetica Chimica Acta – Vol. 95 (2012)
2. Syntheses. 2.1. (1R,2R,3S,4S,5S,6R)-5-Cinnamoyloxy-6-hydroxy-1,2;3,4-di(isopropylidenedioxy)-
cyclohexane (¼(3aS,4S,5R,5aR,8aR,8bS)-Hexahydro-5-hydroxy-2,2,7,7-tetramethylbenzo[1,2-d:3,4-
dꢃ]bis[1,3]dioxol-4-yl (2E)-3-Phenylprop-2-enoate; 2a). Epoxide 1 (600 mg, 2.47 mmol) [6], potassium
cinnamate (1.60 g, 8.60 mmol), freshly recrystallized 18-crown-6 (2.40 g, 9.00 mmol) [8], cinnamic acid
(366 mg, 2.47 mmol), Bu4NI (200 mg, 0.54 mmol), DMF (6 ml), 1,2-dimethoxyethane (6 ml), and
HMPA (2 ml) were mixed in a pressure tube, sealed under Ar, and heated to 1278 for 84 h with vigorous
stirring. Reaction progress was monitored by TLC (hexanes/AcOEt 2 :1; ceric ammonium molybdate
(Hannesianꢃs stain)/UV). After final cooling, the reaction mixture was diluted with H2O and CH2Cl2,
and the phases were separated. The aq. phase was twice extracted with additional CH2Cl2, and the
combined org. phases were washed twice with sat. Na2CO3 and with brine. Following drying and
concentration in the usual fashion, the reaction vessel was sealed and stored at 48 for 12 h. FC of the
crude material (hexanes/AcOEt (0.1% Et3N on Et3N-neutralized silica) 8 :1 (5 column volumes) to 4 :1
(5 column volumes) to 3.5 :1 (2 column volumes)) provided 2a (830 mg) in 87% isolated yield. White
solid. Rf (hexanes/AcOEt 2 :1) 0.29. M.p. 73 – 778 (precipitated from Et2O/heptanes). [a]2D0 ¼ 96.5 (c ¼
1.0, CHCl3). IR (CHCl3, c ¼ 26 mm): 3606, 2992, 2938, 1716, 1637, 1450, 1385, 1308, 1163, 1060, 859,
754. 1H-NMR (300 MHz, CDCl3): 7.75 (d, J ¼ 16.0, 1 H); 7.55 – 7.46 (m, 2 H); 7.41 – 7.35 (m, 3 H); 6.50 (d,
J ¼ 16.0, 1 H); 5.09 (dd, J ¼ 11.4, 8.2, 1 H); 4.55 – 4.40 (m, 2 H); 4.35 (dd, J ¼ 8.0, 5.9, 1 H); 4.27 (dd, J ¼
7.9, 6.0, 1 H); 3.68 (dd, J ¼ 10.8, 8.3, 1 H); 2.47 (br. s, OH); 1.53 (s, 3 H); 1.52 (s, 3 H); 1.38 (s, 3 H); 1.35
(s, 3 H). 13C-NMR (75 MHz, CDCl3): 166.9; 146.1; 134.2; 130.5; 128.9; 128.2; 117.3; 109.7; 109.6; 79.0
77.5; 77.0; 76.6; 76.4; 75.6; 75.4; 73.5; 71.8; 27.8; 27.6; 25.6; 25.4. EI-MS: 375 (16.5), 317 (3.5), 227 (7.3),
142 (72.7), 131 (100). HR-EI-MS: 375.1437 ([M ꢀ 15]þ, C20H23O7þ ; calc. 375.1444). Anal. calc. for
C20H23O7: C 64.6, H 6.7; found: C 64.2, H 6.8.
2.2.
5-Cinnamoyloxy-6-(hexa-3,5-dienyloxy)-2,2,7,7-tetramethylhexahydrobenzo[1,2-d;3,4-d’]
bis[1,3]dioxol-4-ol (¼(3aS,4R,5R,5aS,8aS,8bS)-Hexahydro-2,2,7,7-tetramethyl-5-{[(2E)-3-phenylprop-
2-enoyl]oxy}benzo[1,2-d:3,4-d’]bis[1,3]dioxol-4-yl (3E)-Hexa-3,5-dienoate; 3a). Compound 2a (1.50 g,
3.83 mmol) was dissolved in freshly distilled CH2Cl2 (20 ml) along with hexa-3,5-dienoic acid (785 mg,
7.66 mmol) and cat. 4-(dimethylamino)pyridine (DMAP; 10 mg), and the yellow soln. was cooled to 08
under stirring. DCC (1.50 g, 7.27 mmol) was added, and the mixture was stirred for 3 h. The mixture
turned slightly orange and formed insoluble material. TLC showed completion of the reaction, and the
mixture was diluted with Et2O and filtered through a Celite plug. Solvent was removed under reduced
pressure, and FC (hexanes/AcOEt 8 :1, 0.1% Et3N on Et3N-neutralized silica) of the crude product
provided 3a (1.41 g, 78%). White semi-solid. Rf (2 :1 hexanes/AcOEt) 0.63. [a]2D0 ¼ 30.5 (c ¼ 1.0, CHCl3).
IR (CHCl3, c ¼ 21 mm): 3941, 3052, 2987, 1744, 1718, 1265, 1163, 908, 749. 1H-NMR (300 MHz, CDCl3):
7.66 (d, J ¼ 16.0, 1 H); 7.52 – 7.47 (m, 2 H); 7.41 – 7.35 (m, 3 H); 6.38 (d, J ¼ 16.0, 1 H); 6.13 (ddd, J ¼
16.6, 10.1, 10.1, 1 H); 5.99 (dd, J ¼ 14.9, 10.5, 1 H); 5.62 (ddd, J ¼ 14.7, 7.2, 7.2, 1 H); 5.16 (dd, J ¼
19.4, 8.1, 1 H); 5.19 (dd, J ¼ 19.6, 8.0, 1 H); 5.01 – 4.86 (m, 2 H); 4.60 – 4.53 (m, 2 H); 4.44 – 4.34 (m,
2 H); 3.07 (d, J ¼ 7.3, 2 H); 1.55 (s, 3 H); 1.54 (s, 3 H); 1.38 (s, 3 H); 1.37 (s, 3 H). 13C-NMR (75 MHz,
CDCl3): 170.6; 165.8; 146.0; 136.2; 134.7; 130.5; 128.9; 128.3; 124.8; 117.1; 117.0; 109.75; 109.71; 77.2; 76.3;
76.2; 74.9; 71.4; 71.0; 37.9; 27.6; 25.8. EI-MS: 469 (6.5), 426 (12.5), 131 (100). HR-EI-MS: 469.1868
([M ꢀ 15]þ, C27H32O8þ ; 469.1862). Anal. calc. for C27H32O8: C 66.9, H 6.7; found: C 66.2, H 6.6.
2.3. (1S,2S,3S,4S,4aR,6aS,7S,10aR,13aR)-1,2;3,4-Di(isopropylidenedioxy)-7-phenyl-2,3,4,4a,6a,7,
10a,11-octahydro-1H-dibenzo[b,f][1,4]dioxonine-6,12(10H,13aH)-dione (¼(3aS,3bS,6aS,6bR,8aS,
9S,12aR,15aR,15bS)-3a,3b,6a,6b,9,10,12a,13,15a,15b-Decahydro-2,2,5,5-tetramethyl-9-phenyl-8H-ben-
zo[f]bis[1,3]dioxolo[3,4 :5,6]benzo[1,2-b][1,4]dioxonine-8,14(8aH)-dione; 4a). Compound 3a (320 mg,
0.68 mmol) was dissolved in 15 ml of 1,2,4-trichlorobenzene, and the soln. was stirred under vacuum, at
r.t. for 1 h. The vessel was then isolated and sealed under vacuum, and then heated to 1418 for 92 h. The
ratio of product/starting material was monitored by NMR until it reached 73 :27. Following cooling to r.t.
the solvent was removed under high-vacuum distillation, and the residue was purified by FC (50 ml Et3N-
neutralized silica, 21-cm column height, hexanes/AcOEt 7.5 :1, 0.1% Et3N on Et3N-neutralized silica) to
provide 203 mg (62%) of 4a as the only isolable cycloadduct. White solid. Rf (hexanes/AcOEt 6 :1) 0.33.
M.p. 174 – 1768 (CDCl3). [a]2D0 ¼ 73.5 (c ¼ 0.14, CHCl3). IR (CHCl3, c ¼ 3.0 mm): 3051, 2987, 1743, 1387,
1376, 1265, 1217, 1210, 1153, 783. 1H-NMR (300 MHz, CDCl3): 7.36 – 7.27 (m, 2 H); 7.26 – 7.20 (m, 3 H);
5.91 (tdd, J ¼ 9.5, 5.4, 1.9, 1.9, 1 H); 5.53 (m, 1 H); 5.06 (dd, J ¼ 11.6, 8.0, 1 H); 4.86 (dd, J ¼ 11.6, 8.6,

Helvetica Chimica Acta – Vol. 95 (2012)
2033
1 H); 4.47 – 4.36 (m, 3 H); 4.31 (dd, J ¼ 8.5, 6.1, 1 H); 3.22 (dt, J ¼ 11.8, 11.7, 5.0, 1 H); 3.15 – 3.02 (m,
1 H); 2.96 (dd, J ¼ 14.3, 6.4, 1 H); 2.80 (dd, J ¼ 12.0, 8.9, 1 H); 2.33 (dd, J ¼ 14.3, 11.6, 1 H); 2.32 – 2.08
(m, 2 H); 1.54 (s, 3 H); 1.42 (s, 3 H); 1.36 (s, 3 H); 1.29 (s, 3 H). 13C-NMR (75 MHz, CDCl3): 177.4; 171.8;
142.8; 128.6; 128.2; 127.4; 126.9; 126.8; 1110.5; 110.2; 76.7; 74.9; 74.7; 728; 51.0; 42.5; 41.4; 39.4; 32.2; 27.5;
27.4; 25.3; 25.2. HR-EI-MS: 469.1867 ([M ꢀ 15]þ, C27H32O8þ ; calc. 469.1862). Anal. calc. for C27H32O8: C
66.9, H 6.7; found: C 67.1, H 6.9.
2.4. (1R,2R,3S,4S,5S,6R)-5-(4-Bromocinnamoyloxy)-6-hydroxy-1,2;3,4-di(isopropylidenedioxy)cy-
clohexane (¼(3aS,4S,5R,5aR,8aR,8bS)-Hexahydro-5-hydroxy-2,2,7,7-tetramethylbenzo[1,2-d:3,4-d’]
bis[1,3]dioxol-4-yl (2E)-3-(4-Bromophenyl)prop-2-enoate; 2b). Epoxide 1 (600 mg, 2.47 mmol) [6],
potassium para-bromocinnamate (2.40 g, 8.60 mmol), freshly recrystallized 18-crown-6 (2.40 g,
9.00 mmol) [8], para-bromocinnamic acid (576 mg, 2.47 mmol), Bu4NI (200 mg, 0.54 mmol), DMF
(6 ml), 1,2-dimethoxyethane (6 ml), and HMPA (2 ml) were mixed in a pressure tube, sealed under Ar,
and heated to 1278 for 115 h under vigorous stirring. Reaction progress was monitored by TLC (hexanes/
AcOEt 2 :1, CAM/UV). After final cooling, the mixture was diluted with H2O and Et2O, and the phases
were separated. The aq. phase was twice extracted with additional Et2O, and the combined org. phases
were washed twice with sat. Na2CO3 and with brine. Following drying and concentration in the usual
fashion, the residue was purified by FC (hexanes/AcOEt 4 :1, 0.1% Et3N on Et3N-neutralized silica) to
provide 103 mg of the undesired diastereoisomer in the first fraction and 487 mg of the desired isomer,
2b, in the second fraction, in addition to 421 mg of material in a mixed fraction. The mixed fraction was
repurified in the same manner to obtain an additional 303 mg of 2b, for a total of 790 mg in 68% yield.
Colorless amorphous gel. Rf (hexanes/AcOEt 2 :1) 0.29. [a]2D0 ¼ 161.5 (c ¼ 1.0, CHCl3). IR (CHCl3): 3554,
3022, 2995, 2937, 171, 1638, 1488, 1385, 1312, 1167, 1072, 1041, 1010, 900, 860, 803, 734. 1H-NMR
(300 MHz, CDCl3): 7.61 (d, J ¼ 16.0, 1 H); 7.45 (pseudo-d, J ¼ 8.4, 2 H); 7.30 (pseudo-d, J ¼ 8.5, 2 H);
6.43 (d, J ¼ 16.0, 1 H); 5.06 (dd, J ¼ 11.4, 8.3, 1 H); 4.45 (dd, J ¼ 6.2, 3.2, 1 H); 4.42 (dd, J ¼ 6.2, 3.2, 1 H);
4.30 (dd, J ¼ 8.1, 5.9, 1 H); 4.23 (dd, J ¼ 7.9, 6.1, 1 H); 3.64 (dd, J ¼ 11.2, 8.2, 1 H); 2.90 (br. s, 1 H); 1.49
(s, 3 H); 1.48 (s, 3 H); 1.34 (s, 3 H); 1.31 (s, 3 H). 13C-NMR (75 MHz, CDCl3): 166.4; 144.3; 133.0; 132.0;
129.4; 124.6; 118.0; 109.5; 109.4; 78.9; 76.2; 75.5; 75.2; 73.5; 71.4; 27.7; 27.4; 25.4; 25.3. HR-EI-MS:
468.0794 (Mþ, C21H25BrO7þ ; 468.0784). Anal. calc. for C21H25BrO7: C 53.74, H 5.37; found: C 53.73, H
5.47.
2.5. 5-(4-Bromocinnamoyloxy)-6-(hexa-3,5-dienyloxy)-2,2,7,7-tetramethylhexahydrobenzo[1,2-d;
3,4-d’]bis[1,3]dioxol-4-ol (¼(3aS,4R,5R,5aS,8aS,8bS)-5-{[(2E)-3-(4-Bromophenyl)prop-2-enoyl]oxy}-
hexahydro-2,2,7,7-tetramethylbenzo[1,2-d:3,4-d’]bis[1,3]dioxol-4-yl (3E)-Hexa-3,5-dienoate; 3b). Com-
pound 2b (460 g, 1.0 mmol) was dissolved in freshly distilled CH2Cl2 (10 ml) along with hexa-3,5-dienoic
acid (224 mg, 2.0 mmol) and cat. DMAP (10 mg), and the yellow soln. was cooled to 08 under stirring.
DCC (391 mg, 1.9 mmol) was added, and the mixture was stirred for 4 h. The mixture immediately
turned slightly orange and formed insoluble material. TLC showed completion of the reaction, and the
mixture was diluted with Et2O and filtered through a Celite plug. Solvent was removed under reduced
pressure, and FC (hexanes/AcOEt 8 :1, 0.1% Et3N on Et3N-neutralized silica) of the crude mixture
provided 450 mg (80%) of 3b. Colorless gel. Rf (hexanes/AcOEt 4 :1) 0.59. [a]2D0 ¼ 35.3 (c ¼ 0.50,
CHCl3). IR (CHCl3, c ¼ 5 mg/ml): 3027, 2992, 2935, 1740, 1721, 1637, 1488, 1385, 1309, 1209, 1165, 1073,
1
1008, 909, 855, 820, 782, 743, 679. H-NMR (300 MHz, CDCl3): 7.53 (d, J ¼ 16.0, 1 H); 7.45 (pseudo-d,
J ¼ 8.3, 2 H); 7.29 (pseudo-d, J ¼ 8.4, 2 H); 6.31 (d, J ¼ 16.0, 1 H); 6.07 (ddd, J ¼ 16.6, 10.1, 10.1, 1 H);
5.93 (dd, J ¼ 14.8, 10.7, 1 H); 5.56 (ddd, J ¼ 14.7, 7.2, 7.2, 1 H); 5.20 – 5.05 (m, 2 H); 4.90 (d, J ¼ 16.4,
1 H); 4.84 (d, J ¼ 9.8, 1 H); 4.55 – 4.49 (m, 2 H); 4.38 – 4.28 (m, 2 H); 3.02 (d, J ¼ 7.2, 2 H); 1.50 (s, 3 H);
1.48 (s, 3 H); 1.32 (s, 3 H); 1.31 (s, 3 H). 13C-NMR (75 MHz, CDCl3): 1703; 165.3; 144.3; 136.0; 134.4;
132.9; 131.9 (2 ꢁ CH); 129.4 (2 ꢁ CH); 124.7; 124.6; 117.6; 116.8; 109.51; 109.46; 76.1; 75.9; 74.7 (2 ꢁ
CHC2OR); 71.2; 71.0; 37.8; 27.42; 27.39; 25.6; 25.6. HR-EI-MS: 562.1198 (Mþ, C27H31BrO8þ ; calc.
562.1202). Anal. calc. for C27H31BrO8: C 57.56, H 5.55; found: C 57.41, H 5.65.
2.6. (1S,2S,3S,4S,4aR,6aS,7S,10aR,13aR)-7-(4-Bromophenyl)-1,2;3,4-di(isopropylidenedioxy)-
2,3,4,4a,6a,7,10a,11-octahydro-1H-dibenzo[b,f][1,4]dioxonine-6,12(10H,13aH)-dione (¼(3aS,3bS,6aS,
6bR,8aS,9S,12aR,15aR,15bS)-9-(Bromophenyl)-3a,3b,6a,6b,9,10,12a,13,15a,15b-decahydro-2,2,5,5-tetra-
methyl-8H-benzo[f]bis[1,3]dioxolo[3,4 :5,6]benzo[1,2-b][1,4]dioxonine-8,14(8aH)-dione; 4b). Com-
pound 3b (400 mg, 0.71 mmol) was dissolved in 90 ml of 1,2,4-trichlorobenzene, and the soln. was

2034
Helvetica Chimica Acta – Vol. 95 (2012)
stirred under vacuum at r.t. for 1 h. The vessel was then sealed under vacuum and then heated to 1418 for
120 h. Following cooling to r.t., the solvent was removed under high-vacuum destillation, and the residue
was purified by FC (hexanes/AcOEt 8 :1, 0.1% Et3N on Et3N-neutralized silica) to provide 20 mg of
material in the first fraction, which although co-spotting with the starting material, was thoroughly
decomposed, and 320 mg of 4b in 80% yield as the only isolable cycloadduct in the second fraction.
Additionally 10 mg of mixed fractions were obtained but were discarded. A portion of 4b was
recrystallized from a mixture of benzene/abs. EtOH (initially 4 :1) and a drop of H2O through slow
evaporation over three weeks to obtain crystals suitable for X-ray analysis. White crystalline solid. Rf
(hexanes/AcOEt 4 :1, UV/KMnO4) 0.52. M.p. 2128 (dec., CH2Cl2). [a]2D0 ¼ 83.8 (c ¼ 1.0, CHCl3). IR
(CHCl3, c ¼ 10 mg/ml): 3023, 2992, 2963, 1743, 1711, 1490, 1385, 1376, 1262, 1260, 1153, 1126, 1075, 1056,
1009, 909, 856, 803, 712. 1H-NMR (300 MHz, CDCl3): 7.37 (pseudo-d, J ¼ 8.4, 2 H); 7.05 (pseudo-d,
J ¼ 8.4, 2 H); 5.85 (tdd, J ¼ 9.8, 5.8, 1.8, 1.8, 1 H); 5.48 (m, 1 H); 5.03 (dd, J ¼ 11.6, 8.4, 1 H); 4.82 (dd,
J ¼ 11.6, 8.6, 1 H); 4.45 – 4.25 (m, 3 H); 4.29 (dd, J ¼ 8.6, 6.0, 1 H); 3.15 (dt, J ¼ 11.8, 11.7, 4.8, 1 H);
3.10 – 2.98 (m, 1 H); 2.91 (dd, J ¼ 14.3, 6.3, 1 H); 2.72 (dd, J ¼ 12.0, 8.8, 1 H); 2.28 (dd, J ¼ 14.3, 11.7,
1 H); 2.20 (td, J ¼ 17.7, 5.1, 5.1, 1 H); 2.09 (ddd, J ¼ 11.7, 5.0, 2.8, 1 H); 1.50 (s, 3 H); 1.40 (s, 3 H); 1.33 (s,
3 H); 1.27 (s, 3 H). 13C-NMR (75 MHz, CDCl3): 177.0; 171.5; 141.8; 131.6; 129.0; 128.1; 126.3; 120.5;
110.2; 110.1; 76.6; 76.5; 75.1; 74.7; 74.5; 72.6; 50.6; 41.9; 41.2; 39.3; 32.0; 27.35; 27.30; 25.14; 25.10. HR-EI-
MS: 547.0986 ([M ꢀ 15]þ, C27H31BrOþ8 ; calc. 547.0981). Anal. calc. for C27H31BrO8: C 57.56, H 5.55;
found: C 57.30, H 5.55.
2.7. 2-[(1R,5S,6S)-6-(Hydroxymethyl)-5-phenylcyclohex-2-en-1-yl]ethanol (5). Cycloadduct 4a
(34 mg, 0.070 mmol) was dissolved in anh. THF (2 ml) under Ar. LiAlH4 (22 mg, 0.562 mmol) was
added at 08 in one portion, and the mixture was heated to reflux. After 2 h, the reaction was quenched
through successive addition of 100 ml of H2O, 200 ml of 15% NaOH, and 400 ml of H2O. The suspension
was then filtered through Celite with AcOEt, then dried (MgSO4), filtered and concentrated. The crude
mixture was immediately dissolved in 60% AcOH and heated to 808 for 12 h. Solvent was removed under
reduced pressure, and the residue was treated with cat. MeONa in MeOH at r.t. for 2 h, followed by
treatment with Dowex-50X resin. The resin was removed by filtration, the solvent was removed under
reduced pressure, and the residue was purified by FC (AcOEt/hexanes 2 :1) to provide 5 (9 mg, 58%).
Clear oil. Rf (hexanes/AcOEt 1:1) 0.07. [a]D20 ¼ ꢀ47.0 (c ¼ 1.0, MeOH). IR (CDCl3): 3624, 2253, 1816,
1793, 1466, 1381, 1095, 905. 1H-NMR (300 MHz, CDCl3): 7.42 – 7.17 (m, 5 H); 5.87 – 5.77 (m, 1 H); 5.73
(ddd, J ¼ 10.0, 3.6, 2.1, 1 H); 3.87 – 3.75 (m, 2 H); 3.64 (dd, J ¼ 11.5, 3.2, 1 H); 3.38 (dd, J ¼ 11.4, 3.1,
1 H); 2.86 (td, J ¼ 11.1, 8.2, 8.2, 1 H); 2.53 (br. s, 1 H); 2.34 – 2.23 (m, 2 H); 2.02 – 1.86 (m, 1 H); 1.81 –
1.45 (m, 4 H). 13C-NMR (75 MHz, CDCl3): 145.0; 130.9; 128.7; 127.7; 126.4; 126.2; 61.8; 60.6 45.7; 42.9;
36.2; 34.2; 34.1. HR-EI-MS: 214.1362 ([M ꢀ H2O]þ, C15H20O2þ ; calc. 214.1358).
3. X-Ray Analysis of 4b. 3.1. General. All measurements were made by means of a Bruker Kappa
APEX DUO CCD area-detector diffractometer with graphite monochromated CuKa radiation (l
1.54178 ꢂ) using a Bruker ImuS radiation source. Data reduction was performed with SAINT (Bruker,
2007). All calculations were performed with the SHELXL97 program [9], and the crystallographic
diagrams were drawn with PLATON [10].
3.2. Determination of Structure of 4b. The unit cell constants and an orientation matrix for data
collection were obtained from a least-squares refinement of the setting angles of 38907 reflections in the
range 3.618 < 2q < 66.548. A total of 5875 frames were collected by using f and w scans with k offsets.
Equivalent reflections were merged.
Data Collection and Refinement Parameters. Crystallized from benzene/EtOH/H2O; C27H31BrO8; Mr
563.4342 g/mol; crystal dimensions, 0.19 mm ꢁ 0.14 mm ꢁ 0.01 mm; colorless plates; monoclinic; space
group P21: unit cell parameters: a ¼ 12.7006(6), b ¼ 5.7739(3), c ¼ 23.6378(14) ꢂ, a ¼ 90, b ¼
105.522(3)8, g ¼ 90; V¼ 1670.19(15) ꢂ3; Z ¼ 2; F(000), 668.0; T 147 K; Dx ¼ 1.276 g/cm3; wavelength,
1.54178 ꢂ; m(CuKa) ¼ 2.062; scan type, f and w; 2q(max) ¼ 66.54; 2q range for cell determination, 3.61 –
66.548; reflections for cell determination, 38907; transmission factors (min; max), 0.661, 0.7538;
absorption correction, semi-empirical from equivalents; total reflections measured, 5875; symmetry-
independent reflections, 3261; Rint ¼ 0.0373; reflections with I > 2s(I), 5510; reflections used in
refinement, 5657; parameters refined, 383; refinement method, full-matrix least-squares on F2; final

Helvetica Chimica Acta – Vol. 95 (2012)
2035
R(F) (I > 2s(I) reflections), 0.0260 (5510); final wR(F2) (all data), 0.0668 (5657); goodness-of-fit, 1.018;
Dr (max; min) ¼ 0.195; ꢀ 0.229 e ꢂꢀ3
.
Highly disordered solvent molecules were identified as being present in the crystal cell and could not
be modeled. These pockets are assumed to contain disordered H2O/EtOH. The 24-electron contribution
of these molecules was removed during the refinement process as carried out previously [11]. A view of
the molecule is shown in Fig. 3.
REFERENCES
[1] S. Faure, S. Piva-Le-Blanc, C. Bertrand, J.-P. Pete, R. Faure, O. Piva, J. Org. Chem. 2002, 67, 1061; K.
Nakahara, K. Hirano, R. Maehata, Y. Kita, H. Fujioka, Org. Lett. 2011, 13, 2015; K. Takasu, M.
Ueno, M. Ihara, J. Org. Chem. 2001, 66, 4667; T. Sugimura, S. Inoue, A. Tai, Tetrahedron Lett. 1998,
39, 6487; K. J. Shea, D. R. Gauthier Jr., Tetrahedron Lett. 1994, 35, 7311; G. Michaud, M. Bulliard, L.
ˆ
Ricard, J.-P. Genet, A. Marinetti, Chem. – Eur. J. 2002, 8, 3327.
[2] I. Fleming, ꢄFrontier Orbitals and Organic Chemical Reactionsꢃ, John Wiley & Sons, Chichester,
1976.
[3] K. C. Nicolaou, S. A. Snyder, T. Montagnon, G. Vassilikogiannakis, Angew. Chem., Int. Ed. 2002, 41,
1668; K. Ruck-Braun, H. Kunz, ꢄChiral Auxiliaries in Cycloadditionsꢃ, Wiley-VCH, Weinheim, 1999;
M. Juhl, D. Tanner, Chem. Soc. Rev. 2009, 38, 2983; K.-i. Takao, R. Munakata, K.-i. Tadano, Chem.
Rev. 2005, 105, 4779.
[4] T. Akiyama, H. Nishimoto, K. Ishikawa, S. Ozaki, Chem. Lett. 1992, 21, 447; T. Akiyama, H.
Nishimoto, T. Kuwata, S. Ozaki, Bull. Chem. Soc. 1994, 67, 180; J. Singleton, J. O. Hoberg, Mini-Rev.
Org. Chem. 2009, 6, 1.
[5] O. Hoshino, M. Ishizaki, K. Kamei, M. Taguchi, T. Nagao, K. Iwaoka, S. Sawaki, B. Umezawa, Y.
Iitaka, J. Chem. Soc., Perkin Trans. 1 1996, 571 – 580; S.-J. Pang, S.-H. Min, H. Lee, C.-G. Cho, J. Org.
Chem. 2003, 68, 10191; G. A. Kraus, M. E. Krolski, J. Org. Chem. 1986, 51, 3347.
[6] V. P. Bui, T. Hudlicky, Tetrahedron 2004, 60, 641.
[7] G. H. Posner, D. Z. Rogers, J. Am. Chem. Soc. 1977, 99, 8214; G. H. Posner, Angew. Chem., Int. Ed.
1978, 17, 487; G. Stockburger, J. Brandner, J. Am. Oil Chem. Soc. 1966, 43, 6; U. G. Ibatullin, S. A.
Vasilꢃeva, Z. K. Karimova, I. Z. Latypova, M. G. Safarov, Chem. Heterocycl. Compd. (N.Y., NY,
U.S.) 1989, 25, 1335; N. Miyakoshi, D. Aburano, C. Mukai, J. Org. Chem. 2005, 70, 6045; A.
Bukowska, W. Bukowski, J. Chem. Technol. Biotechnol. 1996, 67, 176; E. N. Jacobsen, F. Kakiuchi,
R. G. Konsler, J. F. Larrow, M. Tokunaga, Tetrahedron Lett. 1997, 38, 773.
[8] W. L. F. Armarego, C. L. L. Chai, ꢄPurification of Laboratory Chemicalsꢃ, Butterworth-Heinemann,
2009.
[9] G. M. Sheldrick, Univeristy of Gçttingen, Gçttingen, 1997.
[10] A. Spek, Acta Crystallogr., Sect. D 2009, 65, 148.
[11] R. Stahler, C. Nather, W. Bensch, Acta Crystallogr., Sect. C 2001, 57, 26; S. Athimoolam, J. Kumar, V.
Ramakrishnan, R. K. Rajaram, Acta Crystallogr., Sect. E 2005, 61, m2014; A. A. Mohamed, J. A.
Krause Bauer, A. E. Bruce, M. R. M. Bruce, Acta Crystallogr., Sect. C 2003, 59, m84; P. J. Cox, Y.
Kumarasamy, L. Nahar, S. D. Sarker, M. Shoeb, Acta Crystallogr., Sect. E 2003, 59, o975.
Received August 8, 2012