Internal nucleophilic termination in acid-mediated polyene cyclizations: Part 4. Synthetic access to tetracyclic didehydro and tetradehydro analogues of (±)-Ambrox

Article abstract of DOI:10.1002/hlca.200590247

Treatment of the unsaturated bicyclic homoallylic alcohols (E)- and (Z)-5 and (Z)- and (E)-10 and allenic alcohol 16 with an excess of ClSO₃H in 2-nitropropane or CH₂Cl₂ at -80 afforded, in moderate yields (ca. 30-70%), di

Full text of DOI:10.1002/hlca.200590247

Helvetica Chimica Acta – Vol. 88 (2005)
3055
Internal Nucleophilic Termination in Acid-Mediated Polyene Cyclizations
Part 41)
Synthetic Access to Tetracyclic Didehydro and Tetradehydro Analogues of
(ꢀ)-Ambrox®
by Roger L. Snowden* and Simon Linder
Firmenich SA, Corporate R&D Division, P. O. Box 239, CH-1211 Geneva 8
(phone: +41227803608; fax: +41227803334; e-mail: roger.snowden@firmenich.com)
Dedicated to Dr. Ferdinand Naef on the occasion of his 65th birthday
Treatment of the unsaturated bicyclic homoallylic alcohols (E)- and (Z)-5 and (Z)- and (E)-10 and allenic
alcohol 16 with an excess of ClSO3H in 2-nitropropane or CH2Cl2 at ꢁ808 afforded, in moderate yields (ca. 30–
70%), diastereoisomer mixtures of racemic tetracyclic ethers 12a–c (Table 1) and 17a,b (Table 2), respectively.
These kinetically controlled stereospecific transformations are believed to proceed via concerted or noncon-
certed pathways (see Schemes 4 and 6) and the results are fully consistent with our earlier work. Representing
novel didehydro bridged analogues of known, olfactively active labdane tricyclic ethers, the organoleptic prop-
erties of 12a–c and 17a,b are briefly described, especially those of 12c which, in the context of structure–activity
studies, is a racemic didehydro analogue of the known ambergris odorant Ambrox®
.
1. Introduction. – The relationship between the organoleptic properties of a com-
pound and its chemical structure has been studied extensively [2]. The ambergris-
type odor in particular, exemplified by the benchmark odorant Ambrox®2), has
attracted special attention, and has been the subject of various theories, one of which
is the empirical ‘triaxial rule of odor sensation’ [4]. Thus, a multi-point interaction
between a dimethyl-trans-decalin skeleton and the complementary receptor site is con-
sidered to be associated with the axial orientation of the substituents R’, R’’, and Ra, as
shown in structure A. In this context, we reasoned that the tetracyclic ether 12c, in
which R’ and Ra are replaced by a C2 ethanediyl bridge, would be an interesting test
for this theory. Furthermore, we realized that, in analogy with our recent work on
1
)
)
For Parts 1–3, see [1b], [1c], and [1d], resp.
Ambrox® (tradename of Firmenich SA) is a commercially important naturally occurring odorant; the race-
mate is also commercialized by Firmenich SA under the tradename of Cetalox® [3].
2
© 2005 Verlag Helvetica Chimica Acta AG, Zürich

3056
Helvetica Chimica Acta – Vol. 88 (2005)
acid-mediated polyene cyclizations [1], 12c would be accessible via cyclization of the
cyclohexyl cation I by regioselective protonation of either one of the two homoallylic
alcohols, (E)-5 and (Z)-10, as shown in Scheme 1. We thus describe herein the synthesis
of (E)-5 and (Z)-10, and the outcome of their cyclizations. We also include results con-
cerning the cyclizations of (Z)-5 and (E)-10, as well as that of the dehydro analogue of
(E)- and (Z)-5, allenic alcohol 16.
Scheme 1
2. Results and Discussion. – 2.1. Homoallylic Alcohols (E)- and (Z)-5 (see Scheme
2). Bromination of d-pyronene (=1,1-dimethyl-2,3-bis(methylene)cyclohexane; 1) [5]
afforded dibromide 2 (56% yield), which was treated with dimethyl 3-oxoglutarate
(=dimethyl 3-oxopentanedioate) under basic conditions to furnish, after bis-de(me-
thoxycarbonylation) of the intermediate oxodicarboxylate 3, cycloheptenone 4 in
73% yield (Scheme 2). A subsequent Wittig reaction with the ylid derived from (3-
hydroxypropyl)triphenylphosphonium bromide [6] gave 5 (1:1 (E)/(Z) diaster-
eoisomer mixture; 61% yield), which was chromatographed to afford pure samples
of (E)- and (Z)-5.
2.2. Homoallylic Alcohols (E)- and (Z)-10. Access to (E)- and (Z)-10 was accom-
plished by the route described in Scheme 3. Accordingly, regioselective allylic oxidation
of 6 ((E)/(Z) 1:1), the (tert-butyl)dimethylsilyl ether of 5, was effected by using an
excess of pyridinium dichromate (PDC) and tert-butyl hydroperoxide, to give 7 ((E)/
(Z) 1:1) in low yield (32%); (E)- and (Z)-7 were then separated by chromatography.
Attempted formation of the corresponding tosylhydrazones (E)- and (Z)-8, was accom-
panied by concomitant hydrolysis of the silyl ether moiety, and thus directly afforded
the alcohols (E)- and (Z)-9 in 65% and 42%3) yield, respectively. Finally, hydride
reduction with catecholborane (=1,3,2-benzodioxaborole) furnished (E)- and (Z)-10
in 67 and 71% yield, respectively.
2.3. Acid-Mediated Cyclization of (E)-5, (Z)-5, (E)-10, and (Z)-10 to Tetracyclic
Ethers 12a–c. The acid-mediated cyclizations of (E)- and (Z)-5 and of (E)- and (Z)-
3
)
As shown in Scheme 3, (Z)-9 was accompanied by the non-hydrolyzed, intermediate silyl ether (Z)-8 (29%
yield), which was isolated, characterized, and then separately hydrolyzed under acidic conditions to (Z)-9
(80% yield) (see Exper. Part).

Helvetica Chimica Acta – Vol. 88 (2005)
3057
Scheme 2
a) Br2, CH2Cl2, 08. b) Dimethyl 3-oxoglutarate, NaHCO3, H2O/CH2Cl2, r.t. c) NaCl, H2O, DMSO, 128–1358.
d) [Ph3P(CH2)3OH]+Brꢁ, BuLi (2 mol-equiv.), THF, ꢁ208 ! r.t.
10 were effected by treatment of each substrate with an excess of chlorosulfuric acid
(ClSO3H; 4–6 mol-equiv.) in 2-nitropropane or CH2Cl2 at ꢁ808 during 30 min. Subse-
quent neutralization with aqueous Na2CO3 solution, extractive workup, and distillation
in vacuo afforded mixtures of 11 and 12a–c in 30–71% yield. The product distributions
were determined by anal. GC (Table 1). Pure samples of each product were obtained by
a combination of column chromatography and prep. GC, and were fully characterized
spectroscopically. Structural attributions were determined by in-depth NMR experi-
ments4).
The observed results are explained by kinetically controlled cyclizations analogous
to our previous work [1], and the mechanistic rationale is presented in Scheme 4.
Accordingly, with (E)-5 as substrate, a nonconcerted cyclization, via regioselective
axial protonation of the tetrasubstituted cyclohexene C=C bond to the cyclohexyl cat-
ion l5), and subsequent ring closure involving trans-addition across the trisubstituted
C=C bond with equatorial CꢁC and CꢁO bond formation, leads to 12c. Alternatively,
a concerted cyclization, comprising trans-additions across both of the two C=C bonds,
gives 12a. The anticipated by-product, tetrahydrofuran 11 (see Table 1), is formed via
prior protonation of the trisubstituted C=C bond, followed by ring closure. Unexpect-
edly, our results show an unexplained solvent dependence: changing from 2-nitropro-
pane to CH2Cl2 (Table 1, Entries 1 and 2) results in a higher yield (71% as opposed
to 48%), an almost total suppression of 11, and a reversal of selectivity in favor of
4
)
Structure determinations were greatly aided by comparison with the NMR spectra of analogous tricyclic
ethers from earlier work [1][7].
The observed regioselectivity maximizes the hyperconjugative stabilization of l by the axial or pseudoaxial
H-atoms at the adjacent C-atoms.
5
)

3058
Helvetica Chimica Acta – Vol. 88 (2005)
Scheme 3
a) NaH, THF, then t-BuMe2SiCl, 52–548. b) PDC (4.4 mol-equiv.), t-BuO2H (3.1 mol-equiv.), toluene, r.t. c)
TsNHNH2, EtOH, reflux. d) AcOH, THF/H2O, r.t. e) Catecholborane (2 mol-equiv.), CHCl3, 58, then
NaOAc·3 H2O, reflux.
12c with respect to 12a (3 :1 vs. 0.8 :1)6). In contrast, cyclization of (Z)-10 with 2-nitro-
propane as solvent (Table 1, Entry5 ) led to the highly selective formation of 12c, with
only small amounts of 12a and 117) (12c/12a/11 91:5 :4) as by-products. Despite the low
yield (34%), this result nicely confirms the structure of 12c, as cyclization is presumed
to pass through I via a highly site-selective and regioselective protonation of the trisub-
stituted cyclohexene C=C bond.
6
)
We have no explanation for this reversal in selectivity, though it is interesting to note that the observed
selectivity in CH2Cl2 is reflected in the calculated MM2 energies of 12a and 12c: 48.7 and 46.7 kcal/mol,
respectively.
The formation of 11 from (Z)-10 necessarily involves isomerization of the trisubstituted cyclohexene C=C
bond into the thermodynamically more favored tetrasubstituted position under the strongly acidic condi-
tions.
7
)

Helvetica Chimica Acta – Vol. 88 (2005)
3059
Table 1. Acid-Mediated Cyclizations of (E)- and (Z)-5, and (Z)- and (E)-10
Entrya)
Substrate
Solvent
Product distributionb)c)
Yield [%]d)
11
12a
12b
12c
12d
1
2
3
4
5
6
(E)-5
(E)-5
(Z)-5
(Z)-5
(Z)-10
(E)-10
A
B
A
B
A
A
30
2
50
2
4
9
40
24
3
1
5
–
30
74
–
–
91
–
–
–
–
–
–
–
48
71
42
52
34
31
–
47
97
–
–
91
a) Reaction conditions: substrate (1 mol-equiv.), ClSO3H (4–6 mol-equiv.), solvent (2-nitropropane (A) or
CH2Cl2 (B)), ꢁ808 (see Exper. Part). b) GC Analysis of crude product mixture: relative % of 11 and 12a–d.
c) Another tetracyclic ether, i (structure undetermined), was also detected (ca. 1–5%) in the cyclizations of
(E)- and (Z)-5 (Entries 1–4; see Exper. Part for spectral data). d) Refers to the yield of 11 and 12a–c after chro-
matography and bulb-to-bulb distillation in vacuo.
Analogously, the acid-mediated cyclization of (Z)-5 can theoretically occur via a
nonconcerted pathway, involving the intermediacy of the cyclohexyl cation II, or a con-
certed pathway, to afford either 12b or 12d. In practice however, using once again 2-
nitropropane or CH2Cl2 as solvent (Table 1, Entries 3 and 4), 12d was not detected in
the product mixtures8). In this case, it is logical to assume that the concerted pathway
is strongly disfavored by important nonbonding interactions in the transition state9). As
observed before (vide supra), CH2Cl2 as solvent gave a higher yield (52% vs. 42%), and
once again only trace amounts of 11 (ca. 2%) were formed. As expected, cyclization of
(E)-10 in 2-nitropropane (Table 1, Entry6 ) resulted in the highly selective formation of
12b, whereas 117) (12b/11 91:9) was the only other compound detected in the product
mixture (31% yield).
2.4. Allenic Alcohol 16 (see Scheme 5). By using known methodology [8], base-
mediated 1,2-addition of prop-2-yn-1-ol to 4 afforded diol 13, whose primary OH
group was selectively acetylated to 14 and the tertiary OH group then protected, via
acid-catalyzed addition of ethyl vinyl ether, to furnish 15. Finally, treatment with
LiAlH4 resulted in deprotection of the primary OH group and internal hydride dis-
placement of the O-(1-ethoxyethyl) group, affording 16 in 40% overall yield.
2.5. Acid-Mediated Cyclization of 16 to Tetracyclic Ethers 17a,b and 18. The acid-
mediated cyclization of 16 was effected by treatment with an excess of ClSO3H (9
mol-equiv.) in either 2-nitropropane or CH2Cl2 at ꢁ808 during 1 h. Subsequent neutral-
8
)
However, another tetracyclic ether, i, was detected (ca. 1–5%) in the product mixtures from the cycliza-
tions of (E)- and (Z)-5. Unfortunately, though isolated and characterized (see Exper. Part), structure elu-
cidation was not possible due to lack of material.
9
)
The MM2 energy of 12d is 59.9 kcal/mol as compared to 53.3 kcal/mol for 12b.

3060
Helvetica Chimica Acta – Vol. 88 (2005)
Scheme 4. Acid-Mediated Cyclizations of (E)- and (Z)-5 and of (Z)- and (E)-10: Mechanistic Rationale for
the Formation of 11 and 12a–c
Scheme 5
a) Prop-2-yn-1-ol, KOH, THF, 408. b) Ac2O, Et3N, cat. N,N-dimethylpyridin-4-amine (DMAP), CH2Cl2, r.t.
c) Ethyl vinyl ether, cat. TsOH·H2O, toluene, 08. d) LiAlH4, THF, r.t.
ization with aqueous Na2CO3 solution, extractive workup, chromatography, and evap-
oration in vacuo afforded mixtures containing 17a, 17b, and 18 as major products (ca.
40% yield). The compositions of the mixtures were determined by anal. GC (Table 2).
Pure samples of each component were obtained by chromatography and fully charac-
terized spectroscopically. Structural attributions were established by NMR experi-

Helvetica Chimica Acta – Vol. 88 (2005)
Table 2. Acid-Mediated Cyclization of 16
3061
Entrya)
Solvent
Product distributionb)c)
Yield [%]d)
17a
17b
18e)
1
2
2-nitropropane
CH2Cl2
37
25
54
30
9
45
41
43
a) Reaction conditions: 16 (1 mol-equiv.), ClSO3H (9 mol-equiv.), solvent, ꢁ808 (see Exper. Part). b) GC Anal-
ysis of crude product mixture: relative % of 17a, 17b, and 18. c) Two other minor products, constitutional isomers
of 17a,b and 18, but of unknown structures, were also detected (each ca. 5%). Catalytic hydrogenation (cat. 10%
Pd/C, 1 bar H2, cyclohexane, r.t.) of one of these compounds led stereoselectively to i, the tetracyclic ether of
unknown structure which was formed in trace amounts during the acid-mediated cyclizations of (E)- and
(Z)-5 (see Table 1). d) Refers to the yield of 17a,b and 18 confirmed by catalytic hydrogenation to 19 (see Exper.
Part).
ments, and corroborated by catalytic hydrogenation of the crude product mixture to a
mixture of 12a, 12c, and 19, whose separation was again effected by chromatography.
The purified compounds were then identified by GC and spectral comparison with
authentic samples (for 12a and 12c), or by inspection of the NMR data (for 19).
A mechanistic rationale for the observed results is presented in Scheme 6. A non-
concerted pathway via regioselective protonation of the tetrasubstituted cyclohexene
C=C bond of 16 to cyclohexyl cation III (presumed to be a 1:1 diastereoisomer mix-
ture), followed by concerted ring closure10) via trans-addition across the allenyl
C(3)=C(7’) bond leads to 17a and 17b11). Alternatively, 17a could be the result of a
direct concerted pathway from 16, involving trans-additions across both the two
C=C bonds. The formation of the relatively less strained 1812) is thought to be thermo-
dynamically driven by a successive 1,2-H and 1,2-methyl shift via cyclohexyl cations IV
10
)
It is important to note that, in contrast to previous work in which the ethane-1,2-diyl C2 bridge is replaced by
two axial Me groups [1d], the cyclization of III to 17a and 17b cannot involve the intermediacy of diaster-
eoisomer mixtures of tricyclic allyl cations such as the hypothetical VI and VI’, due to obvious ring strain,
and must proceed in a concerted manner. This means that only one of the two presumed diastereoisomers
of III is able to undergo cyclization, and thus provides an explanation for the low yields.
11
12
)
)
The MM2 energies of 17a and 17b are relatively close: 56.7 and 56.3 kcal/mol, respectively.
MM2 energy of 18: 53.5 kcal/mol.

3062
Helvetica Chimica Acta – Vol. 88 (2005)
Scheme 6. Acid-Mediated Cyclization of 16: Mechanistic Rationale for the Formation of 17a,b and 18
and V (diastereoisomer mixtures), and then stereoselective cyclization to 1813). Cata-
lytic hydrogenation of 17a, 17b, and 18 leads with complete stereoselection to 12a,
12c, and 19, respectively. In the first case, b-face attack is favored14), whereas 17b
and 18 are hydrogenated exclusively from the a-face.
2.6. Organoleptic Properties of 12a–b and 17a–b. Due to the small amounts of com-
pounds available, qualitative rather than quantitative odor evaluations were effected.
Thus, in comparison with racemic Ambrox®, 12a was perceived as amber and woody,
whereas 12b had woody, camphoraceous notes. In contrast, 12c possessed a very power-
ful amber odor which was more tenacious than either of its two diastereoisomers. The
dehydro analogues 17a and 17b also exhibited strong amber and woody notes but, espe-
cially here, the lack of sufficient material at our disposal rendered a precise organolep-
tic evaluation extremely difficult.
3. Conclusion. – Synthetic access to racemic tetracyclic didehydro and tetradehydro
analogues of the labdane tricyclic ether Ambrox® was achieved, and their organoleptic
13
)
)
Once again (cf. Footnote 10), only one of the two diastereoisomers of V is able to cyclize to 18.
In the case of 17b, when the ethane-1,2-diyl bridge is replaced by two axial Me groups, the hydrogenation
stereoselectivity is completely reversed, and a-face attack is favored [1d]. This change in selectivity is
attributed to subtle structural modifications engendered by the presence of the two fused five-membered
rings in 17b.
14

Helvetica Chimica Acta – Vol. 88 (2005)
3063
properties were evaluated. As well as providing further examples of Brønsted acid
mediated polyene cyclizations in which the initiating group is an alkene and the termi-
nating group is an alcohol, the described methodology demonstrates its ability to create
multiple stereogenic centers with a high degree of stereoselectivity. Finally, the targeted
tetracyclic ether 12c was shown to represent another example of the ‘triaxial rule’ in
possessing a strong ambergris-type odor similar to that exhibited by its dihydro tricyclic
analogue, (ꢀ)-Ambrox®.
Experimental Part
General. See [1d].
1,2-Bis(bromomethyl)-3,3-dimethylcyclohex-1-ene (2). A soln. of Br2 (4.4 g, 27.5 mmol) in CH2Cl2 (20 ml)
was added dropwise during 30 min to a stirred soln. of 1,1-dimethyl-2,3-bis(methylene)cyclohexane (=d-pyro-
nene; 1; 3.8 g, 26 mmol) in CH2Cl2 (35 ml) maintained at 08 under N2. After a further 15 min at 08, the pale-red
soln. was allowed to attain 188 during 15 min, and then, after the addition of NaHCO3 (100 mg), was evaporated
at 58/15 mbar. The resulting brown oil (8.4 g) was purified by CC (silica gel+1% NaHCO3, pentane) to afford,
after evaporation at 258/0.05 mbar, 2 (4.3 g, 56%). Pale-yellow oil. 1H-NMR: 1.13 (s, 6 H); 1.49 (2 H); 1.65 (2 H);
2.22 (2 H); 4.08 (4 H). 13C-NMR: 140.9 (s); 136.7 (s); 38.9 (t); 35.3 (s); 32.8 (t); 29.3 (t); 28.1 (q); 26.7 (t); 18.6 (t).
MS: 296 (10, M(79Br81Br)+), 217 (23), 215 (24), 135 (100), 121 (57), 107 (59), 93 (80).
1,2,3,4,5,6,8,9-Octahydro-1,1-dimethyl-7H-benzocyclohepten-7-one (4). A soln. of dimethyl 3-oxoglutarate
(2.8 g, 7.9 mmol) and 2 (3.1 g, 10.6 mmol) in CH2Cl2 (50 ml) was added dropwise to a mechanically stirred mix-
ture of NaHCO3 (5 g, 59.5 mmol), H2O (50 ml), and CH2Cl2 (10 ml) at r.t. under N2. After 48 h, the mixture was
extracted (Et2O), and workup followed by evaporation afforded a pale-yellow viscous oil (3.4 g). CC (cyclohex-
ane/AcOEt 85 :15) furnished crude dimethyl 2,3,4,5,6,7,8,9-octahydro-1,1-dimethyl-7-oxo-1H-benzocyclohep-
tene-6,8-dicarboxylate (3; 2.9 g; pale-yellow viscous oil). Without further purification15), a mixture of 3 (2.8
g), NaCl (0.65 g, 11.1 mmol), H2O (0.6 g, 33.3 mmol), and DMSO (15 ml) was heated at 128–1358 during 4.5
h. The cooled pale-brown mixture was then diluted with Et2O (70 ml) and poured into 10% aq. NaCl soln.
Extraction (Et2O), workup, CC (cyclohexane/AcOEt 9 :1), and bulb-to-bulb distillation i.v. afforded 4 (1.5 g,
73%). Pale-yellow oil. B.p. 130–1408/0.03 mbar. 1H-NMR: 1.01 (s, 6 H); 1.45 (2 H); 1.60 (2 H); 2.02 (2 H);
2.18 (2 H); 2.27 (2 H); 2.41 (2 H); 2.45 (2 H). 13C-NMR: 212.8 (s, C(7)); 140.1 (s, C(9a)); 132.8 (s, C(4a));
44.0 (t, C(8)); 42.4 (t, C(6)); 39.2 (t, C(2)); 35.0 (s, C(1)); 32.5 (t, C(4)); 30.0 (t, C(5)); 27.7 (q, MeꢁC(1));
23.4 (t, C(9)); 19.6 (t, C(3)). MS: 192 (36, M+), 177 (100), 159 (20), 135 (24), 119 (27), 107 (43), 91 (49), 79 (45).
3-(2,3,4,5,6,7,8,9-Octahydro-1,1-dimethyl-1H-benzocyclohepten-7-ylidene)propan-1-ol (5; (E)/(Z) 1:1). At
ꢁ208 1.6
M BuLi in hexane (50 ml, 0.08 mol) was added dropwise during 20 min to a stirred slurry of (3-hydroxy-
propyl)triphenylphosphonium bromide (18 g, 0.05 mol) in THF (130 ml) under N2. The red-brown mixture was
allowed to attain r.t. during 1 h and then stirred for 2 h. The orange mixture was cooled to ꢁ158, (white precip-
itate), and a soln. of 4 (4 g, 0.02 mol) in THF (30 ml) was added dropwise during 15 min. The mixture was
allowed to attain r.t. during 1 h, stirred at this temp. for 16 h, and then poured into sat. aq. NH4Cl soln. Extrac-
tion (Et2O), workup, and evaporation afforded an orange-brown viscous oil (7.7 g) which was triturated with
Et2O/pentane 7:3 (30 ml) at 58. Filtration, and evaporation of the filtrate gave a brown oil (5.3 g). CC (cyclo-
hexane/AcOEt 9 :1) afforded unreacted 4 (0.4 g), and (E)/(Z)-5 1 :1 (2.8 g, 60%). Pale-yellow oil. B.p. 160–
1708/0.01 mbar. MPLC (LiChroprep Si60, 15–25 mm, cyclohexane/AcOEt 19 :1) was used to isolate (Z)-5
(1.1 g, less polar) and (E)-5 (0.8 g, more polar).
Data of (Z)-5: 1H-NMR (after exchange with D2O): 0.98 (s, 6 H); 1.90 (2 H); 1.56 (2 H); 1.95 (dd, J=6, 6, 2
H); 2.03 (2 H); 2.06–2.20 (6 H); 2.29 (dt, J=6, 6, 2 H); 3.61 (br. t, J=6, 2 H); 5.08 (br. t, J=7, 1 H). 13C-NMR:
146.5 (s); 140.4 (s); 133.2 (s); 118.2 (d); 62.5 (t); 39.4 (t); 37.3 (t); 35.6 (t); 35.0 (s); 33.0 (t); 30.6 (t); 30.3 (t); 27.8
(q); 27.2 (t); 19.9 (t). MS: 234 (52, M+), 219 (44), 175 (41), 145 (39), 119 (58), 105 (72), 91 (100), 79 (73).
15
)
NMR Analysis indicated 3 to be a partially tautomerized (keto/enol ca. 5:1) cis/trans-diastereoisomer mix-
ture.

3064
Helvetica Chimica Acta – Vol. 88 (2005)
Data of (E)-5: 1H-NMR (after exchange with D2O): 0.97 (s, 6 H); 1.90 (2 H); 1.57 (2 H); 1.94–2.02 (4 H);
2.10 (s, 4 H); 2.18 (2 H); 2.29 (dt, J=6, 6, 2 H); 3.59 (br. t, J=6, 2 H); 5.08 (br. t, J=7, 1 H). 13C-NMR: 146.4 (s);
140.6 (s); 133.2 (s); 118.4 (d); 62.5 (t); 39.4 (t); 38.8 (t); 34.8 (s); 34.1 (t); 33.0 (t); 30.6 (t); 28.6 (t); 28.3 (t); 27.8
(q); 19.9 (t). MS: 234 (49, M+), 219 (46), 175 (38), 145 (39), 119 (55), 105 (71), 91 (100), 79 (72).
7-{3-{[(tert-Butyl)dimethylsilyl]oxy}propylidene}-2,3,4,5,6,7,8,9-octahydro-1,1-dimethyl-1H-benzocyclo-
heptene (6;(E)/(Z) 1:1). A soln. of (E)/(Z)-5 1:1 (1 g, 4.3 mmol) in THF (3 ml) was added dropwise during 10
min to a stirred slurry of NaH (80% dispersion in oil; 0.2 g, 6.6 mmol) in THF (10 ml) at r.t. under N2. The mix-
ture was then heated at 52–548 during 1 h and cooled to 108 prior to the dropwise addition of a soln. of (tert-
butyl)chlorodimethylsilane (1.1 g, 7.1 mmol) in THF (5 ml). After attaining r.t. during 15 min, the mixture
was heated at 52–548 for 1 h, cooled to 108, and poured into sat. aq. NaHCO3 soln. Extraction (Et2O), workup,
CC (cyclohexane/AcOEt 92 :8), and bulb-to bulb distillation i.v. afforded (E)/(Z)-6 1 :1 (1.2 g, 81%). Colorless
1
oil. B.p. 190–2008/0.1 mbar. H-NMR: 0.06 (s, 6 H); 0.90 (s, 9 H); 0.97, 0.99 (2s, 6 H); 1.40 (2 H); 1.56 (2 H);
1.90–2.17 (10 H); 2.23 (q, J=6, 2 H); 3.57, 3.58 (2t, J=7, 2 H); 5.07 (t, J=7, 1 H). 13C-NMR: 144.8 (s); 140.7,
140.6 (2s); 133.4 (s); 118.8 (d); 63.6, 63.5 (2t); 39.6, 39.5 (2t); 38.9 (t); 37.4 (t); 35.7 (t); 35.1, 34.9 (2s); 34.4
(t); 33.2 (t); 31.1 (t); 30.3 (t); 28.8 (t); 28.5 (t); 27.9 (q); 27.4 (t); 26.1 (q); 20.0 (t); 18.5 (s); ꢁ5.1 (q). MS: 348
(<0.5, M+), 291 (77), 215 (24), 189 (62), 129 (75), 75 (100).
7-{3-{[(tert-Butyl)dimethylsilyl]oxy}propylidene}-2,3,4,5,6,7,8,9-octahydro-4,4-dimethyl-1H-benzocyclo-
hepten-1-one (7; (E)/(Z) 1 :1). To a stirred soln. of (E)/(Z) 6 1 :1 (1.1 g, 3.15 mmol) in toluene (25 ml) containing
Celite® (2 g) at 158 was added portionwise pyridinium dichromate (5.2 g, 13.8 mmol), and then dropwise 80% aq.
tert-butyl hydroperoxide (1.25 ml, 9.9 mmol). After a further 22 h at r.t., the mixture was filtered through Celite®
and the solid thoroughly washed with Et2O. Evaporation of the filtrate afforded an orange oil (850 mg). Sus-
pension of the filter cake in warm H2O and further extraction (Et2O), followed by workup and evaporation
gave more orange oil (300 mg). CC (CH2Cl2) of the combined oils furnished unreacted (E)/(Z)-6 1:1 (200
mg; Rf (CH2Cl2) 0.70), (E)-7 (180 mg, 16%), and (Z)-7 (120 mg, 10%).
Data of (E)-7: Pale-yellow oil. Rf (CH2Cl2) 0.38. 1H-NMR: 0.02 (s, 6 H); 0.86 (s, 9 H); 1.12 (s, 6 H); 1.79 (dd,
J=6, 6, 2 H); 2.02 (2 H); 2.15–2.50 (10 H); 3.54 (t, J=7, 2 H); 5.09 (t, J=7, 1 H). 13C-NMR: 198.0 (s); 167.9 (s);
142.8 (s); 136.7 (s); 120.3 (d); 63.4 (t); 37.2 (t); 37.1 (t); 36.5 (s); 34.3 (t); 31.1 (t); 29.7 (t); 28.7 (t); 26.3 (q); 26.1
(q); 24.5 (t); 18.5 (s); ꢁ5.1 (q). MS: 362 (<0.5, M+), 347 (2), 305 (100), 181 (14), 73 (43).
Data of (Z)-7: Beige, semi-crystalline solid. Rf (CH2Cl2) 0.30. 1H-NMR: 0.05 (s, 6 H); 0.89 (s, 9 H); 1.13 (s, 6
H); 1.80 (m, 2 H); 2.05–2.50 (12 H); 3.57 (t, J=7, 2 H); 5.13 (t, J=7, 1 H). 13C-NMR: 197.9 (s); 168.0 (s); 142.7
(s); 136.7 (s); 120.7 (d); 63.4 (t); 37.2 (t); 37.1 (t); 36.5 (s); 34.3 (t); 31.0 (t); 30.7 (t); 28.6 (t); 26.3 (q); 26.1 (q); 23.4
(t); 18.5 (s); ꢁ5.1 (q). MS: 362 (<0.5, M+), 347 (2), 305 (100), 181 (9), 73 (38).
(7E)-2,3,4,5,6,7,8,9-Octahydro-7-(3-hydroxypropylidene)-4,4-dimethyl-1H-benzocyclohepten-1-one Tosyl-
hydrazone ((E)-9). A stirred mixture of (E)-7 (200 mg, 0.55 mmol) and tosylhydrazide (=4-methylbenzenesul-
fonic acid hydrazide; 120 mg, 0.65 mmol) in EtOH (2 ml) was heated at reflux during 1 h, cooled to r.t., and then
poured into 10% aq. NaCl soln. Extraction (Et2O), workup, and recrystallization (Et2O/pentane 1.5 :1) afforded
(E)-9 (150 mg, 65%). White crystals. M.p. 137–1418. 1H-NMR (after exchange with D2O): 1.02 (s, 6 H); 1.59 (br.
t, J=6, 2 H); 2.05 (2 H); 2.13–2.36 (7 H); 2.42 (s, 4 H); 2.54 (2 H); 3.60 (t, J=6, 2 H); 5.09 (br. t, J=7, 1 H); 7.29
(d, J=7, 2 H); 7.86 (d, J=7, 2 H). 13C-NMR: 156.0 (s); 144.6 (s); 144.0 (s); 135.5 (s); 132.6 (s); 129.4 (d); 128.3
(d); 119.5 (d); 62.5 (t); 37.0 (t); 35.8 (t); 35.3 (s); 30.7 (t); 29.0 (t); 28.7 (t); 26.4 (2q); 25.7 (t); 21.7 (q); 21.3 (t).
MS: 416 (<0.5, M+), 261 (4), 217 (5), 91(100).
(7Z)-2,3,4,5,6,7,8,9-Octahydro-7-(3-hydroxypropylidene)-4,4-dimethyl-1H-benzocyclohepten-1-one Tosyl-
hydrazone ((Z)-9). As described for (E)-9, with (Z)-7 (280 mg, 0.77 mmol), tosylhydrazide (170 mg, 0.91
mmol), and EtOH (3 ml) (75 min). After workup, CC (cyclohexane/AcOEt 4 :1) furnished (7Z)-7-{3-{[(tert-
butyl)dimethylsilyl]oxy}propylidene}-2,3,4,5,6,7,8,9-octahydro-4,4-dimethyl-1H-benzocyclohepten-1-one Tosyl-
hydrazone ((Z)-8; 120 mg, 29%) and (Z)-9 (135 mg, 42%).
Data of (Z)-8: White crystals (from pentane/Et2O 4 :1). Rf (cyclohexane/AcOEt 4 :1) 0.50. M.p. 57–608.
1H-NMR: 0.04 (s, 6 H); 0.88 (s, 9 H); 0.97 (s, 6 H); 1.54 (br. t, J=6, 2 H); 1.97–2.08 (4 H); 2.13–2.28 (5 H);
2.39 (s, 4 H); 2.47 (2 H); 3.52 (t, J=6, 2 H); 5.06 (br. t, J=7, 1 H); 7.28 (d, J=7, 1 H); 7.85 (d, J=7, 2 H).
MS: 530 (<0.5, M+), 213 (18), 199 (29), 149 (39), 73 (100).
Data of (Z)-9: White crystals (from pentane/Et2O 1.5 :1). Rf (cyclohexane/AcOEt 4 :1) 0.10. M.p. 145–
1498. 1H-NMR (after exchange with D2O): 1.00 (s, 6 H); 1.58 (br. t, J=6, 2 H); 2.05–2.35 (10 H); 2.43 (s, 4
H); 2.52 (2 H); 3.53 (t, J=6, 2 H); 5.07 (br. t, J=7, 1 H); 7.32 (d, J=7, 2 H); 7.88 (d, J=7, 2 H). 13C-NMR:
144.0 (s); 135.3 (s); 129.4 (d); 128.3 (d); 120.2 (d); 62.5 (t); 37.1 (t); 35.6 (t); 35.1 (s); 30.6 (t); 29.2 (t); 27.9
(t); 26.3 (q); 24.2 (t); 21.6 (q); 21.2 (t). MS: 416 (1, M+), 261 (9), 217 (10), 91 (100).

Helvetica Chimica Acta – Vol. 88 (2005)
3065
Conversion of (Z)-8 to (Z)-9. A mixture of (Z)-8 (100 mg, 0.19 mmol), THF (2 ml), and 10% aq. AcOH
soln. (1 ml) was stirred at r.t. during 2 h. Extraction (Et2O), workup, and recrystallization (pentane/Et2O
1.5 :1) afforded (Z)-9 (62 mg, 80%), identical to an authentic sample (vide supra).
(3E)-3-(1,2,3,5,6,8,9,9a-Octahydro-1,1-dimethyl-7H-benzocyclohepten-7-ylidene)propan-1-ol ((E)-10). Cat-
echolborane (60 mg, 0.5 mmol) was added to a stirred soln. of (E)-9 (100 mg, 0.24 mmol) in CHCl3 (2 ml) at 28
under N2. After 2 h at 58, NaOAc·3H2O (200 mg, 1.47 mmol) was added portionwise, and the mixture was
allowed to attain r.t. prior to refluxing during 1 h. Then 1N HCl (100 mg) was added to the cooled mixture,
and stirring was continued at r.t. during 15 min. Extraction (Et2O), workup, CC (cyclohexane/AcOEt 4 :1),
and evaporation afforded (E)-10 (37 mg, 67%). Colorless oil. 1H-NMR (after exchange with D2O): 0.84 (s, 3
H); 0.88 (s, 3 H); 1.10–1.22 (2 H); 1.39 (t, J=6, 1 H); 1.43 (t, J=6, 1 H); 1.65 (1 H); 1.84 (m, 1 H); 1.94–2.42
(9 H); 3.62 (2 H); 5.09 (br. t, J=7, 1 H); 5.32 (1 H). 13C-NMR: 144.4 (s); 140.4 (s); 120.9 (d); 119.9 (d); 62.3
(t); 49.3 (d); 38.2 (t); 38.0 (t); 32.3 (s); 31.6 (t); 31.1 (t); 29.6 (t); 29.0 (t); 27.4 (q); 26.2 (q); 23.2 (t). MS: 234
(100, M+), 219 (57), 175 (80), 119 (78), 105 (72), 91 (87).
(3Z)-3-(1,2,3,5,6,8,9,9a-Octahydro-1,1-dimethyl-7H-benzocyclohepten-7-ylidene)propan-1-ol ((Z)-10). As
described for (E)-10, with (Z)-9 (95 mg, 0.23 mmol): (Z)-10 (38 mg, 71%). Colorless oil. 1H-NMR (after
exchange with D2O): 0.85 (s, 3 H); 0.88 (s, 3 H); 1.04–1.18 (2 H); 1.39 (2 H); 1.65 (1 H); 1.92 (m, 1 H); 1.92–
2.38 (9 H); 3.60 (2 H); 5.15 (br. t, J=7, 1 H); 5.32 (1 H). 13C-NMR: 144.8 (s); 141.1 (s); 120.6 (d); 120.2 (d);
62.3 (t); 49.4 (d); 37.4 (t); 35.8 (t); 32.2 (s); 31.7 (t); 31.5 (t); 31.1 (t); 30.0 (t); 27.3 (q); 26.7 (q); 23.2 (t). MS:
234 (100, M+), 219 (60), 175 (80), 119 (79), 105 (74), 91 (92).
Acid-Mediated Cyclization (ClSO3H/2-nitropropane) of (E)-5. 1,2,3,4,4’,5,5’,6,8,9-Decahydro-1,1-dimethyl-
spiro[7H-benzocycloheptene-7,2’(3’H)-furan] (11), and (1RS,6RS,9SR,13RS)- and (1RS,6SR,9SR,13RS)-5,5-
Dimethyl-10-oxatetracyclo[7.4.2.01,6.09,13]pentadecane16) (=(1RS,4SR,9SR,10SR)- and (1RS,4RS,9SR,10SR)-5,
5-Dimethyl-13-oxatetracyclo[7.4.2.01,10.04,9]pentadecane, resp., or (3aRS,5aSR,9aSR,9bSR)- and (3aRS,5aRS,
9aSR,9bSR)-Octahydro-6,6-dimethyl-2H,4H-3a,9a-ethanonaphtho[2,1-b]furan, resp.; 12a and 12c, resp.). A
soln. of (E)-5 (0.77 g, 3.29 mmol) in 2-nitropropane (12 ml) was added dropwise during 10 min to a mechanically
stirred soln. of ClSO3H (0.9 ml, 13.4 mmol) in 2-nitropropane (12 ml) at ꢁ808 under N2. After 20 min at ꢁ808,
sat. aq. Na2CO3 soln. (22 ml) was added dropwise to the orange-brown mixture (temp. rise to ca. ꢁ158), which
was then allowed to attain 108 during 30 min prior to extraction (Et2O). Workup and GC analysis of the crude
product showed the presence of five components, in ascending order of retention times: 4 (1%), 11 (27%), 12a
(37%), 20c (28%), and i (7%, undetermined structure). MPLC (LiChroprep Si60, 15–25 mm, cyclohexane/
AcOEt 19 :1) and bulb-to-bulb distillation at 130–1408/0.01 mbar enabled the isolation of 11 (110 mg, 14%),
12a (160 mg, 20%), 12c (110 mg, 14%), and i (4 mg, 5%), all as colorless oils.
Data of 11: 1H-NMR: 1.96 (s, 6 H); 1.35–2.00 (16 H); 2.23 (2 H); 3.82 (t, J=6.5, 2 H). 13C-NMR: 140.8 (s);
133.5 (s); 85.5 (s); 66.5 (t); 39.4 (t); 38.9 (t), 37.5 (t); 34.8 (s); 32.9 (s); 32.9 (t); 30.1 (t); 27.8 (q); 27.6 (q); 25.7 (t);
23.3 (t); 20.0 (t). MS: 234 (81, M+), 219 (44), 175 (42), 105 (70), 97 (100).
Data of 12a: 1H-NMR: 0.91 (s, 3 H); 0.98 (s, 3 H); 1.08–1.90 (17 H); 2.01 (dd, J=11, 8, 1 H); 3.92 (m, 1 H);
4.07 (m, 1 H). 13C-NMR: 90.9 (s, C(9)); 67.3 (t, C(11)); 52.6 (d, C(6)); 50.2 (d, C(13)); 46.9 (s, C(1)); 42.0 (t,
C(4)); 37.5 (t, C(14)); 36.3 (t, C(8)); 36.0 (t, C(2)); 34.8 (s, C(5)); 34.4 (q, Mea-C(5)); 31.0 (t, C(15)); 26.2 (t,
C(12)); 24.0 (q, Meb-C(5)); 20.9 (t, C(7)); 20.8 (t, C(3)). MS: 234 (3, M+), 205 (100), 163 (10), 96 (26).
Data of 12c: 1H-NMR: 0.88 (s, 3 H); 0.92 (s, 3 H); 1.00–2.00 (18 H); 3.92 (dd, J=18, 6.5, 1 H); 4.09 (m, 1 H).
13C-NMR: 89.6 (s, C(9)); 67.9 (t, C(11)); 61.9 (d, C(13)); 53.2 (d, C(6)); 45.7 (s, C(1)); 42.5 (t, C(4)); 36.4 (t,
C(8)); 36.3 (t, C(2)); 33.4 (q, Mea-C(5)); 33.4 (t, C(15)); 33.4 (s, C(5)); 29.3 (t, C(14)); 25.9 (t, C(12)); 21.5 (q,
Meb-C(5)); 21.5 (t, C(7)); 19.6 (t, C(3)). MS: 234 (31, M+), 219 (11), 135 (25), 97 (100).
Data of i: 1H-NMR: 0.95 (s, 3H); 0.96 (s, 3 H); 1.00–2.00 (17 H); 2.07 (m, 1 H); 3.66 (m, 1 H); 3.86 (m, 1 H).
13C-NMR: 83.2 (s); 66.5 (t); 54.7 (d); 48.8 (s); 43.0 (t); 41.7 (d); 37.0 (t); 36.3 (s); 35.6 (t); 30.1 (t); 29.4 (t); 29.0 (t);
26.4 (q); 26.1 (t); 25.6 (t); 23.0 (q). MS: 234 (5, M+), 219 (11), 124 (100), 109 (6).
Acid-Mediated Cyclization (ClSO3H/CH2Cl2) of (E)-5. As described in the foregoing experiment (vide
supra), except that 2-nitropropane was replaced by CH2Cl2. Workup and GC analysis of the crude product
showed the presence of 11 (2%), 12a (23%), 12c (73%), and i (2%), whose bulb-to-bulb distillation at 130–
1408/0.01 mbar afforded a colorless oil (560 mg, 73%).
Acid-Mediated Cyclization (ClSO3H/2-nitropropane) of (Z)-5. (1RS,6SR,9SR,13SR)-5,5-Dimethyl-10-oxa-
tetracyclo[7.4.2.01,6.09,13]pentadecane16) (=(1RS,4RS,3SR,10RS)-5,5-Dimethyl-13-oxatetracyclo[7.4.2.01,10.04,9]-
pentadecane or (3aRS,5aRS,9aSR,9bRS)-Octahydro-6,6-dimethyl(-2H,4H-3a,9a-ethanonaphtho[2,1-b]furan;
16
)
Systematic names of 12a–c and 17a,b are given in parentheses.

3066
Helvetica Chimica Acta – Vol. 88 (2005)
12b). A soln. of (Z)-5 (1 g, 4.27 mmol) in 2-nitropropane (15 ml) was added dropwise during 10 min to a
mechanically stirred soln. of ClSO3H (1.2 ml, 18 mmol) in 2-nitropropane (15 ml) at ꢁ808 under N2. Following
the same procedure as for the cyclization of (E)-5 (vide supra), GC analysis of the crude product showed the
presence of 4 (8%), 11 (45%), 12a (3%), 12b (42%), and i (2%). MPLC (cyclohexane/AcOEt 19 :1) and
bulb-to-bulb distillation at 130–1408/0.01 mbar enabled the isolation of 4 (50 mg, 6%) and 11 (220 mg, 22%),
both identical to authentic samples (vide supra), and 12b (190 mg, 19%).
Colorless oil. 1H-NMR: 0.88 (s, 3 H); 0.90 (s, 3 H); 1.00–2.00 (17 H); 2.32 (m, 1 H); 4.35 (2 H). 13C-NMR:
89.6 (s, C(9)); 74.0 (t, C(11)); 60.9 (d, C(13)); 43.5 (d, C(6)); 42.5 (t, C(4)); 39.5 (t, C(2)); 39.2 (s, C(1)); 38.7 (t,
C(15)); 33.0 (s, C(5)); 32.7 (q, Mea-C(5)); 31.0 (t, C(14)); 30.6 (t, C(8)); 21.7 (q, Meb-C(5)); 21.5 (t, C(7)); 20.4 (t,
C(12)); 19.9 (t, C(3)). MS: 234 (35, M+), 219 (13), 205 (79), 149 (51), 96 (100).
Acid-Mediated Cyclization (ClSO3H/CH2Cl2) of (Z)-5. As described in the foregoing experiment (vide
supra), except that 2-nitropropane was replaced by CH2Cl2. Workup and GC analysis of the crude product
showed the presence of 11 (2%), 12a (1%), 12b (95%) and i (2%), whose bulb-to-bulb distillation at 130–
1408/0.01 mbar afforded a colorless oil (530 mg, 53%).
Acid-Mediated Cyclization (ClSO3H/2-nitropropane) of (Z)-10. A soln. of (Z)-10 (35 mg, 0.15 mmol) in 2-
nitropropane (1 ml) was added dropwise during 5 min to a mechanically stirred soln. of ClSO3H (0.06 ml, 0.9
mmol) in 2-nitropropane (2 ml) at ꢁ808 under N2. After 25 min at ꢁ808, sat. aq. Na2CO3 soln. (2 ml) was
added dropwise to the orange mixture (temp. rise to ca. ꢁ158). Extraction (Et2O), workup, and evaporation
afforded a viscous oil (25 mg), shown by GC analysis to contain three major products: 11 (4%), 12a (5%),
and 12b (91%). Subsequent CC (cyclohexane/AcOEt 19 :1) and bulb-to-bulb distillation at 180–1908/0.01
mbar afforded 12b (11 mg, 31%), identical to an authentic sample (vide supra).
Acid-Mediated Cyclization (ClSO3H/2-nitropropane) of (E)-10. As described in the foregoing experiment
(vide supra), (E)-10 (35 mg, 0.15 mmol) was converted to a crude product mixture, shown by GC analysis to con-
tain two major products: 11 (9%) and 12b (91%). The latter compound was isolated (10 mg, 28%) and shown to
be identical to authentic material.
2,3,4,5,6,7,8,9-Octahydro-7-(3-hydroxyprop-1-ynyl)-1,1-dimethyl-1H-benzocyclohepten-7-ol (13). A soln. of
4 (3 g, 15.1 mmol) in prop-2-yn-1-ol (5.1 g, 91 mmol) was added dropwise during 10 min to a stirred slurry of
powdered KOH (6 g, 90 mmol) in THF (30 ml) at 408 under N2. The mixture was then refluxed during 2.5 h,
cooled, and poured into cold sat. aq. NH4Cl soln. Extraction (Et2O), workup, CC (cyclohexane/AcOEt 7 :3),
and recrystallization (pentane/Et2O 95 :5) afforded 13 (2.5 g, 68%). White crystals. M.p. 80–818. 1H-NMR
(after exchange with D2O): 0.95 (s, 3 H); 0.97 (s, 3 H); 1.39 (2 H); 1.42–1.60 (4 H); 1.80–2.30 (8 H); 4.30 (s,
2 H). 13C-NMR: 140.8 (s); 133.6 (s); 89.9 (s); 82.1 (s); 72.2 (s); 50.7 (t); 40.7 (t); 39.3 (2t); 34.8 (s); 33.0 (t);
29.5 (t); 27.7 (q); 27.6 (q); 22.8 (t); 19.9 (t). MS: 248 (2, M+), 215 (31), 187 (57), 105 (60), 91 (100).
3-(2,3,4,5,6,7,8,9-Octahydro-7-hydroxy-1,1-dimethyl-1H-benzocyclohepten-7-yl)propy-2-nyl Acetate (=7-
[3-(Acetyloxy)prop-1-ynyl)-2,3,4,5,6,7,8,9-octahydro-1,1-dimethyl-1H-benzocyclohepten-7-ol; 14). Et3N (2.7
ml, 19.1 mmol) was added dropwise during 5 min to a stirred soln. of 13 (3.6 g, 14.5 mmol) and Ac2O (1.7 ml,
17.9 mmol) in CH2Cl2 (10 ml) at 28. After a further 2 h at 58, GC analysis showed 94% conversion. Addition
of DMAP (2 mg) and stirring during a further 30 min completed the reaction. The mixture was poured into
cold 5% aq. HCl soln. and extracted (Et2O). Workup, CC (cyclohexane/AcOEt 4 :1), and evaporation afforded
14 (4.05 g, 96%). Pale-yellow oil. 1H-NMR (after exchange with D2O): 0.96 (2s, 6 H); 1.39 (2 H); 1.54 (4 H);
1.80–2.30 (8 H); 2.10 (s, 3 H); 4.72 (s, 2 H). 13C-NMR: 170.4 (s); 140.8 (s); 133.6 (s); 91.0 (s); 77.7 (s); 72.1
(s); 52.4 (t); 40.7 (t); 39.3 (t); 39.2 (t); 34.8 (s); 33.0 (t); 29.4 (t); 27.7 (q); 27.6 (q); 22.7 (t); 20.8 (q); 19.9 (t).
MS: 290 (< 0.5, M+), 220 (31), 205 (33), 187 (35), 91 (41), 43 (100).
3-[7-(1-Ethoxyethoxy)-2,3,4,5,6,7,8,9-octahydro-1,1-dimethyl-1H-benzocyclohepten-7-yl]prop-2-ynyl Ace-
tate (15; 1:1 diastereoisomer mixture). A soln. of ethyl vinyl ether (2.4 g, 34.6 mmol) in toluene (5 ml) was
added dropwise during 10 min to a stirred soln. of 14 (3.9 g, 13.4 mmol) in toluene (15 ml) containing TsOH·
H2O (20 mg) at ꢁ158 under N2. After a further 2 h at ꢁ108 and 2 h at 08, the mixture was poured into sat.
aq. NaHCO3 soln. Extraction (Et2O), workup, CC (cyclohexane/AcOEt 93 :7) and evaporation afforded 15
(3.9 g, 80%). Pale-yellow oil. 1H-NMR: 0.96 (s, 6 H); 1.19 (2t, J=6, 3 H); 1.33 (2d, J=7, 3 H); 1.39 (2 H);
1.54 (2 H); 1.50–2.30 (10 H); 2.09 (s, 3 H); 3.50 (m, 1 H); 3.69 (m, 1 H); 4.74 (s, 2 H); 5.17 (q, J=6, 1 H).
13C-NMR: 170.2 (2s); 140.9, 140.6 (2s); 133.6, 133.3 (2s); 96.1 (2d); 60.8, 60.7 (2t); 52.3 (2t); 39.6, 39.4 (2t);
39.3 (2t); 38.2, 38.0 (2t); 34.8 (2s); 32.9 (2t); 29.3, 29.1 (2t); 27.6 (4q); 22.5 (2q); 22.3 (2t); 20.7 (2q); 19.9 (2t);
15.3 (2q). MS: 362 (<0.5, M+), 272 (2), 187 (10), 73 (100).
3-(1,2,3,4,5,6,8,9-Octahydro-1,1-dimethyl-7H-benzocyclohepten-7-ylidene)prop-2-en-1-ol (16). A soln. of 15
(3.4 g, 9.1 mmol) in THF (20 ml) was added dropwise during 20 min to a stirred slurry of LiAlH4 (0.42 g, 10.7
mmol) in THF (15 ml) at 18–208 under N2. After a further 30 min at r.t., the mixture was cooled to 58 prior

Helvetica Chimica Acta – Vol. 88 (2005)
3067
to the successive addition of H2O (0.4 ml), 15% aq. NaOH soln. (0.4 ml), Et2O (10 ml), and H2O (1.2 ml). Stir-
ring was continued for 30 min and then followed by filtration through Hyflo. Concentration of the filtrate
afforded a yellow oil (2.4 g) which was purified by CC (cyclohexane/AcOEt 85 :15) and bulb-to-bulb distillation
under high vacuum: 16 (1.6 g, 76%). Colorless oil. B.p. 180–1908/0.01 mbar. 1H-NMR (after exchange with
D2O): 0.97 (s, 6 H); 1.40 (2 H); 1.57 (2 H); 1.98 (t, J=6, 2 H); 2.05–2.25 (8 H); 4.07 (d, J=6, 2 H); 5.18 (br.
t, J=6, 1 H). 13C-NMR: 199.3 (s); 140.7 (s); 133.6 (s); 109.6 (s); 89.0 (d); 61.1 (t); 39.4 (t); 34.9 (t); 34.8 (t);
34.8 (s); 33.0 (t); 32.9 (t); 31.5 (t); 28.0 (t); 27.8 (q); 19.9 (t). MS: 232 (12, M+); 204 (72), 189 (81), 145 (66),
105 (63), 91 (100).
Acid-Mediated Cyclization (ClSO3H/2-nitropropane or CH2Cl2) of 16. (1RS,6RS,9SR)- and (1RS,6SR,
9SR)-5,5-Dimethyl-10-oxatetracyclo[7.4.2.01,6.09,13]pentadec-12-ene16) (=(1RS,4SR,9SR)- and (1RS,4RS,9SR)-
5,5-Dimethyl-13-oxatetracyclo[7.4.2.01,10.04,9]pentadec-10-ene or (3aRS,5aSR,9aSR)- and (3aRS,5aRS,9aSR)-
5,5a,6,7,8,9-Hexahydro-6,6-dimethyl-2H,4H-3a,9a-ethanonaphtho[2,1-b]furan; 17a and 17b, resp.), and (1RS,
6RS,10RS,11SR)-6,11-Dimethyl-2-oxatetracyclo[8.3.2.01,5.06,11]pentadec-4-ene (=(3aRS,5aSR,6RS,9aRS)-4,5,
5a,6,7,8,9,9a-Octahydro-5a,9a-dimethyl-2H-3a,6-ethanonaphtho[2,1-b]furan; 18). A soln. of 16 (150 mg, 0.65
mmol) in 2-nitropropane (5 ml) was added dropwise during 10 min to a mechanically stirred soln. of ClSO3H
(0.4 ml, 6 mmol) in 2-nitropropane (5 ml) at ꢁ808 under N2. After 1 h at ꢁ808, sat. aq. Na2CO3 soln. (10 ml)
was added dropwise to the yellow mixture (temp. rise to ca. ꢁ108), followed by Et2O (10 ml), and the mixture
was allowed to attain 108 during 30 min prior to extraction (Et2O). Workup, filtration (silica gel, cyclohexane/
AcOEt 19 :1), and evaporation afforded a yellow oil (70 mg), shown by GC analysis to contain three major
products: 17a (37%; 15% yield), 17b (54%; 22% yield), and 18 (9%; 4% yield), subsequently separated by
prep. GC. Colorless oils. B.p. 130–1408/0.1 mbar (bulb-to-bulb distillation).
Data of 17a: 1H-NMR: 0.88 (s, 3 H); 0.91 (s, 3 H); 1.10–2.00 (15 H); 4.78 (dd, J=12, 2, 1 H); 4.88 (dd, J=12,
1.5, 1 H); 5.22 (m, 1 H). 13C-NMR: 153.1 (s, C(13)); 107.5 (d, C(12)); 96.1 (s, C(9)); 79.0 (t, C(11)); 52.8 (d, C(6));
42.8 (t, C(4)); 42.5 (s, C(1)); 40.5 (t); 36.6 (t); 35.5 (t); 35.2 (s, C(5)); 32.8 (q, Mea-C(5)); 31.6 (t); 22.6 (q,
Meb-C(5)); 20.9 (t, C(3)); 20.4 (t, C(7)). MS: 232 (80, M+), 204 (100), 189 (25), 136 (62), 91 (61).
Data of 17b: 1H-NMR: 0.88 (s, 3 H); 0.95 (s, 3 H); 1.10–1.85 (11 H); 1.96 (br. dd, J=12, 5, 2 H); 2.23 (m, 2
H); 4.81 (dd, J=12, 2, 1 H); 4.89 (d, J=12, 1 H); 4.99 (br. s, 1 H). 13C-NMR: 157.9 (s, C(13)); 104.3 (d, C(12));
95.3 (s, C(9)); 79.6 (t, C(11)); 51.4 (d, C(6)); 42.2 (t, C(4)); 41.6 (s, C(1)); 37.7 (t, C(8)); 34.8 (t, C(2)); 33.7 (t,
C(14)); 33.5 (s, C(5)); 32.9 (q, Mea-C(5)); 32.8 (t, C(15)); 21.6 (q, Meb-C(5)); 21.4 (t, C(7)); 19.4 (t, C(3)).
MS: 232 (99, M+), 203 (71), 135 (75), 91 (80), 41 (100).
Data of 18: 1H-NMR: 0.95 (s, 3 H); 1.10 (s, 3 H); 1.38–1.85 (12 H); 2.00–2.10 (3 H); 4.55 (dd, J=12, 2, 1
H); 4.68 (br. d, J=12, 1 H); 5.36 (br. s, 1 H). 13C-NMR: 155.4 (s, C(5)); 115.5 (d, C(4)); 88.6 (s, C(1)); 73.6 (t,
C(3)); 44.0 (d, C(10)); 39.5 (t, C(13)); 38.5 (s, C(6)); 38.0 (s, C(11)); 37.6 (t, C(7)); 32.2 (t, C(14)); 31.2 (t,
C(12)); 30.7 (t, C(9)); 29.7 (t, C(15)); 29.1 (q, Me-C(6)); 25.7 (q, Me-C(11)); 22.4 (t, C(8)). MS: 232 (100,
M+), 217 (22), 203 (33), 189 (34), 175 (22), 161 (39), 149 (35), 133 (38), 109 (68).
Repetition of the above procedure, replacing 2-nitropropane by CH2Cl2, afforded a yellow oil (65 mg),
shown by GC and NMR analysis to contain 17a (25%; 11% yield), 17b (30%; 13% yield), and 18 (45%, 19%
yield).
Catalytic Hydrogenation of 17a, 17b, and 18: (1RS,5SR,6SR,10RS,11SR)-6,11-Dimethyl-2-oxatetracy-
clo[8.3.2.01,5.06,11]pentadecane
(=(3aRS,5aSR,6RS,9aSR,9bSR)-Decahydro-5a,9a-dimethyl-2H-3a,6-ethano-
naphtho[2,1-b]furan; 19). A stirred soln. of 17a (26%), 17b (24%), and 18 (50%) (95 mg, 0.41 mmol) in cyclo-
hexane (25 ml) containing 10% Pd/C (30 mg) was hydrogenated (1 bar H2) at r.t. during 5 h. A further quantity
of 10% Pd/C (30 mg) was then added, and hydrogenation was continued for a total of 27 h. Filtration (Celite®),
evaporation, and bulb-to-bulb distillation at 130–1408/0.1 mbar afforded a colorless oil (90 mg), shown by GC
analysis to consist of unreacted 17a (3%), 12a (23%), 12c (23%), and 19 (49%). The identities of 17a, 12a, and
12c were confirmed by GC and NMR comparison with authentic samples (vide supra). A sample of 19 (13 mg)
was isolated by extensive CC purification (cyclohexane/AcOEt 19 :1 and CH2Cl2): Colorless oil.
1H-NMR: 0.92 (s, 3 H); 0.93 (s, 3 H); 1.23 (m, 1 H); 1.36–2.00 (16 H); 2.08 (m, 1 H); 3.71 (m, 1 H); 3.86 (dd,
J=10, 9, 1 H). 13C-NMR: 82.3 (s, C(1)); 65.4 (t, C(3)); 51.2 (d, C(5)); 44.5 (d, C(10)); 40.4 (t, C(13)); 38.2 (t,
C(7)); 37.0 (s, C(11)); 34.3 (s, C(6)); 31.9 (t, C(12)); 30.3 (t, C(4)); 29.7 (t, C(9)); 29.1 (t, C(15)); 28.0 (t,
C(14)); 26.6 (q, Me-C(11)); 24.4 (q, Me-C(6)); 20.1 (t, C(8)). MS: 234 (16, M+), 219 (100), 150 (65), 135 (17),
109 (25).
We thank Mr. W. Thommen and Mr. R. Brauchli for the measurements and interpretations of the NMR
spectral data, and Dr. B. Winter and Dr. J.-Y. de Saint Laumer for the MM2 calculations.

3068
Helvetica Chimica Acta – Vol. 88 (2005)
REFERENCES
[1] a) K. H. Schulte-Elte , R. L. Snowden, C. Tarchini, B. Baer, C. Vial, to Firmenich SA, Eur. Pat. EP 403945,
1990 (Chem. Abstr. 1991, 114, 24717); b) R. L. Snowden, J.-C. Eichenberger, S. M. Linder, P. Sonnay, C. Vial;
K. H. Schulte-Elte, J. Org. Chem. 1992, 57, 955; c) R. L. Snowden, J.-C. Eichenberger, W. Giersch, W. Thom-
men, K. H. Schulte-Elte, Helv. Chim. Acta 1993, 76, 1608; d) R. L. Snowden, J.-C. Eichenberger, S. Linder, P.
Sonnay, Helv. Chim. Acta 2004 87, 1711.
[2] G. Ohloff, B. Winter, C. Fehr, in ‘Perfumes: Art, Science, and Technology’, Eds. P. M. Muller and D. Lamp-
arsky, Elsevier, Amsterdam, 1991, p. 287; K. J. Rossiter, Chem. Rev. 1996, 96, 3201; G. Fráter, J. A. Bajgro-
wicz, P. Kraft, Tetrahedron 1998, 54, 7633; P. Kraft, J. A. Bajgrowicz, C. Denis, G. Fráter, Angew. Chem., Int.
Ed. 2000, 39, 2980.
[3] C. Chauffat, A. Morris, Perfum. Flavor. 2004, 29, 34.
[4] S. Escher, W. Giersch, Y. Niclass, G. Bernardinelli, G. Ohloff, Helv. Chim. Acta 1990, 73, 1935.
[5] M. J. Quirin, M. Taran, B. Delmond, Can. J. Chem. 1996, 74, 1852.
[6] M. E. Garst, E. A. Tallman, J. N. Bonfiglio, D. Harcourt, E. B. Ljungwe, A. Tran, Tetrahedron Lett. 1986, 27,
4533.
[7] G. Ohloff, W. Giersch, W. Pickenhagen, A. Furrer, B. Frei, Helv. Chim. Acta 1985, 68, 2022.
[8] J. S. Cowie, P. D. Landor, S. R. Landor, J. Chem. Soc., Perkin Trans. 1 1973, 720.
Received July12, 2005