Recent results in the search for new molecules with ambergris odor

Article abstract of DOI:10.1002/cbdv.201400153

The synthesis of new odorant molecules is still a challenging task for the fragrance chemist, because now as ever it is difficult to predict the odor properties of small organic molecules. Therefore, certain tools, such as, e.g., lead-structure optimizati

Full text of DOI:10.1002/cbdv.201400153

CHEMISTRY & BIODIVERSITY – Vol. 11 (2014)
1639
Recent Results in the Search for New Molecules with Ambergris Odor1)
by Johannes Panten*, Horst Surburg, and Bernd Hçlscher
Symrise AG, Scent & Care Innovations, Mꢀhlenfeldstrasse 1, D-37603 Holzminden
(
e-mail: johannes.panten@symrise.com)
The synthesis of new odorant molecules is still a challenging task for the fragrance chemist, because
now as ever it is difficult to predict the odor properties of small organic molecules. Therefore, certain
tools, such as, e.g., lead-structure optimization of existing odorants, are helpful techniques. In this article,
we describe the synthesis and the odor properties of a new molecule derived by the so-called ꢁsecoꢂ lead-
ꢃ
structure optimization of the ambergris compound Ambroxide . Based on these results, more
representatives with similar structures have been synthesized and evaluated for their olfactory properties.
Introduction. – The odor of ambergris is one of the most desired in perfumery. The
discovery of the ambergris odor goes back to the very early history. In ancient times,
grey lumps had been found at the beaches of the oceans. Solutions of these grey lumps
in alcohol tincture exhibit a sweet, earthy odor. Because of this nice odor, the lumps
(also called ambra) had been described as the ꢁgold of the seaꢂ. In antiquity, this
material became a desirable commodity and a commercial product. In the 10th century,
the Arabs in Spain used this raw material for fine perfumery. In the 18th century, whale
fishermen were successful to clarify the mystery of the origin of the grey lumps: it is a
metabolic product of the gut of diseased sperm whales. In more recent times, the ambra
lumps were as valuable as gold. Scarcity of the resource and the animal origin has
motivated the fragrance chemist to look for alternative approaches towards this
odorant.
In 1820, Caventou and Pelletier were the first to identify the triterpene ambrein (1)
as an important part of ambra lumps [1]. The structure of 1 was first described by
Lederer in 1946. He also postulated that 1 acts as a precursor, which is cleaved by a
photolytic singlet O reaction, leading to the tricyclic molecule 2 and the bicyclic
2
alcohol a-ambrinol (3; Scheme 1) [2].
Scheme 1. Photolytic Cleavage of Ambrein (1) by Singlet O Reaction to Ambrinol (3) and Ambroxideꢀ
2
(2)
1)
Registered names, trademarks, etc., used in this article, even when not specifically marked as such,
are not to be considered unprotected by law.
ꢄ 2014 Verlag Helvetica Chimica Acta AG, Zꢀrich

1
640
CHEMISTRY & BIODIVERSITY – Vol. 11 (2014)
Scheme 2. Industrial Synthesis of Ambroxideꢀ (2) Starting from Sclareol (4)
In 1977, Ohloff determined the absolute configuration of 2 [3]. As it became
obvious that 2 is the sensorially valuable compound behind the ambergris odor, an
industrial synthesis has been developed and established in the mid 20th century by
Stoll, making this molecule available from a non-animal source and an indispensable
ingredient in modern perfumery [4]. Several companies are producing this material
ꢀ
ꢀ
ꢀ
under trade names like such as Ambroxan , Ambrox , or Ambroxide . Starting
material of the industrial process is sclareolide (5), which is readily available from
sclareol (4), a natural product obtained by steam distillation of clary sage oil
(
Scheme 2).
Due to its outstanding olfactory character, 2 has not only become an important part
of the perfumerꢂs shelf but has also been the object of numerous investigations on
understanding and improvement of the molecular structure and the olfactophore
model of ambergris odor. More particularly, one of the first serious structureꢀodor
relationship (SOR) study has been carried out with odor materials preferred in this
type of odor and naturally with these kind of structures: the ꢁtriaxial rule of amber
compoundsꢂ postulated by Ohloff [5]. The main principle of this rule describes the
position and the distances of the substituents at C(3a), C(5a), and C(9a) within the
ꢀ
trans-decalin ring system of the Ambroxide skeleton; for an ambergris odor sensation,
it is essential that they all should have axial positions. Nowadays, there are numerous
publications covering the synthesis of new ambergris molecules and the evaluation of
their odors with respect to the ꢁtriaxial ruleꢂ and, despite some minor exceptions (see
below), it has been recognized as the main rule of the ambergris olfactophore model. It
is impossible to mention all investigations on structural modifications of 2 and related
reports on SORs; a recent and good overview including references is given in [6]. Three
examples should be exemplarily mentioned here: the contributions of Ohloff et al. [7],
and of Klein and Brunke [8] on the influence of the geminal Me groups at C(6), both
highlighting the 6-nor-derivative 6 with strong ambergris odor but slightly higher odor
threshold than 2; the synthesis of 7 by Boden and Jones [9] by varying the chain length
of the axial substituent at C(3a), and the insertion of a C¼C bond between C(5) and
C(5a) in the tricyclic ring leading to 8 by Snowden and Escher [10] (Fig. 1).
ꢃ
In contrast to the ꢁtriaxial ruleꢂ, compound 8 (tradename Superambrox ) lacks a
substituent at C(5a) but exhibits an outstanding powerful and strong ꢁambery headnote,
with much volume and fullnessꢂ [11].
In our search for new structures with ambergris-like odor, we have been interested
to synthesize molecules with variations of the tricyclic skeleton by using the ꢁseco
principleꢂ.

CHEMISTRY & BIODIVERSITY – Vol. 11 (2014)
1641
Fig. 1. Selected molecules with Ambroxideꢃ skeleton: Norambroxide (6), Grisalvaꢃ (7), and Super-
ambroxꢃ (8)
Results and Discussion. – The ꢁseco principleꢂ for targeted rational drug design has
first been introduced by Sestanj in 1962 [12]. ꢁCutting outꢂ of two C-atoms from the b-
ionone skeleton (9) leads to an acyclic fragment 10 (Fig. 2). The directed synthesis of
such a fragment should lead to a molecule with a similar or preferably stronger effect.
In [12], Sestanj reported that the target molecule had the same violet odor as b-ionon.
Another example for a ꢁseco approachꢂ to new fragrance molecules was described in a
Takasago patent [13]. Cutting out two atoms from damascone (11) led to the acyclic
molecule 12. In this case, it was realized that the synthesis of 12 was impractical and did
not necessarily lead to a molecule with a better performance. But at least, it has inspired
ꢃ
Kraft et al. to design a similar molecule (tradename, Pomarose ) [14]. As 2 has a
tricyclic structure, it offers theoretically several possibilities to create mono- or bicyclic
ring systems by ꢁcutting outꢂ some C-atoms. The already mentioned a-ambrinol (3)
could be seen as an unsaturated ꢁseco-ambroxideꢂ derived by cutting out atoms C(1),
C(2), and the Me group at C(9a). In our project for new molecules with ambergris
odor, we were interested to synthesize and to evaluate the odor of the seco derivative
ꢃ
1
3a, derived from the Ambroxide skeleton by cutting out three C-atoms (C(8), C(9),
and the Me group at C(9a)). As 13a does not contain a decalin skeleton, and only two
Fig. 2. Seco derivatives of b-ionone, b-damascone, and ambroxide

1
642
CHEMISTRY & BIODIVERSITY – Vol. 11 (2014)
out of three axial substituents are present, we were interested to compare it with
molecules relevant for the triaxial rule.
After analyzing several retrosynthetic pathways for seco-Ambroxide (13a), we
decided to carry out the synthesis by starting from the readily available 4-(tert-
butyl)cyclohexanone (14) and using a reaction sequence which is similar to that
published by Bꢁchi and Wuest in 1989 [15] as alternative pathway for the synthesis of
racemic 2 (Scheme 3).
In the first step, 14 was to be converted into the a-allyl keton 15. Several methods
have been published in the literature for this type of transformation, e.g., alkenylation
of the corresponding pyrrolidine enamine with allyl bromide [16] or with allyl acetate
by transition-metal catalysis [Pd(Ph P)4] [17], alkenylation using strong bases, e.g.,
3
i
LiN Pr (LDA) and allyl bromide [18], or radical addition of (tributyl)(prop-2-enyl)tin
2
to 4-(tert-butyl)-2-(phenylseleno)cyclohexanone [19]. In all investigated reactions, the
cis/trans ratio of the alkenylated product has been one major issue (see also [20]): cis-
selectivities have been favored by using the metal enolate pathway, whereas trans-
selectivities have been favored by using the enamine or a radical pathway. However, an
isomerization of the trans-isomer to the thermodynamically more stable cis-isomer
could be easily afforded by treatment with EtONa [19]. Even the synthesis of the
enantiomerically pure form of 15 has been published using the enamine pathway
supported by l-proline esters as chiral auxiliary [21].
As our main interest has been to obtain an odor description for compound 13a, we
decided, for practical reasons, to use the Claisen rearrangement of the diallyl ketals of
14. The detour via the dimethyl ketal was not necessary, but the yield towards the diallyl
ketal was slightly higher in case of transketalization of dimethyl ketal with allylic
alcohol than via a direct ketalization without purification. Thermal elimination of one
allyl moiety led to an enol ether, which was readily converted via Claisen rearrange-
ment to the a-allyl keton 15. Using these conditions, 15 was obtained as a mixture of cis/
trans isomers in a ratio of 3 :1. The determination of the configuration was carried out
1
by GC and H-NMR analysis of the mixture according to the published data [19]. The
cis-isomer content has been increased by stirring with MeONa according [19], and was
used without further purification in the next step where an introduction of the Me
Scheme 3. Retrosynthesis of Seco-Ambroxide (13)

CHEMISTRY & BIODIVERSITY – Vol. 11 (2014)
1643
group via Grignard reaction with MeMgCl was achieved. The attack of the Me
nucleophile at the C¼O function proceeded predominantly to provide a trans-
configuration of the Me group relative to the allyl moiety (more precisely, a 7:1 trans/
cis ratio was observed). The yield of this transformation was 85%. The Grignard
product 16 was then ozonized, followed by reductive workup, to provide diol 17 in 73%
yield. No change in the isomeric ratios was observed. In the last step, diol 17 was
cyclized. To achieve the desired configuration at C(7a), the orientation of the OH
group had to be inverted. To achieve such an inversion, we used the same conditions as
described in [15] with MeNO as solvent and catalytic amounts of TsOH. Unfortu-
2
nately, we were not able to achieve an inversion by this method, and we isolated 13 as a
mixture of two isomers 13a/13b in the same ratio, 1:7, as by using the ꢁclassicalꢂ way
with TsCl as cyclization reagent. Isomers 13a and 13b were separated by GC, and their
1
structures were determined by H-NMR spectroscopy according to the published data
for 2 and its isomer (ꢀ)-3a-epiambroxide [22].
The odors of the two isomers differ significantly. The odor of seco-ambroxide 13a
was described as strong ambergris, balsamic and erogenic, whereas stereoisomer 13b
was found much weaker, more woody, and greasy-like (Fig. 3).
As 13a exhibits a strong ambergris odor, the odor strength of this molecule was
compared with that of 2 on a qualitative and time-dependent basis. Panelists were
asked to compare the strength of both molecules on blotters on a scale from 1 to 9. At
the beginning, panelists rated 13a stronger than 2. This impression disappeared after
3
h, at which time 2 became stronger. After 24 h on the blotter, there was no dry-down
odor impression at all for 13a, whereas 2 gave nearly the same strong odor impression
as at the beginning (Fig. 4). This result implies that 13a could be preferably more useful
in a perfume for the creation of an ambergris top note rather than for the substantive
part of the fond as is the case for 2.
For a discussion on SORs, it is difficult to correlate these findings with the ꢁtriaxial
ruleꢂ, as this rule was clearly confined to a trans-decalin system. However, as 13a
exhibits a strong ambergris odor, we carried out modeling studies [23] to determine the
superposition of 13a on 2 (Fig. 5).
Not surprisingly, apart of the three omitted C-atoms, 13a and 2 showed a very good
superposition. To the best of our knowledge, the odor of a ꢁ9a-norambroxideꢂ has not
yet been described. It would be of interest to compare this molecule with 13a to
examine if an alkyl substituent at C(9a) in 2 is necessary to generate an ambergris odor
sensation or not. Our results, taken together with the odor impression of a-ambrinol
Fig. 3. Odor descriptions of 13a and 13b

1
644
CHEMISTRY & BIODIVERSITY – Vol. 11 (2014)
Fig. 4. Intensity profile 13a vs. 2
Fig. 5. Superposition of 13a (lilac) and 2 (yellow)
(also a seco-ambroxide), indicate that the Me group at C(9a) should not be of major
importance for the olfactory properties.
As 13a exhibits a powerful strong ambergris odor, we decided to further investigate
the influence of the substitution pattern of the bicyclic ring system, as well as by
t
variation of the Bu group. The syntheses of these molecules could be achieved almost
t
by the same reaction sequence as described above (Scheme 4). Variations of the Bu
group could easily be accomplished by using 4-isopropylcyclohexanone (14b) or 4-
isoamylcyclohexanone (14c) as starting material. In addition, a higher alkyl group
could be incorporated at C(7a) by using EtMgBr during the Grignard reaction
sequences. A reduction of 15 with NaBH led to a trisubstituted molecule at ꢁC(7a)ꢂ
4

CHEMISTRY & BIODIVERSITY – Vol. 11 (2014)
1645
Scheme 4. Synthesis of Derivatives of 13
(16c; see also [22]). A Me substitution at C(2) could be attained by cyclization of
precursor 16 with TsOH. At least by carrying out a hydroboration of 16, alcohol 18 was
successfully obtained. Cyclization of 18 resulted in the heptahydrobenzopyrane
derivative 20. All reactions were well-known and afforded good yields (generally
70–80%).
In the Table, all newly synthesized compounds are compiled, together with their
odor descriptions.
Apparently, a substituent at C(7a) in 13a is necessary, for ambergris odor. In
analogy to a comparison of 2 and 7, replacement of the Me group at C(7a) with an Et

1
646
CHEMISTRY & BIODIVERSITY – Vol. 11 (2014)
Table. Odor Descriptions of Diverse Derivatives of 13a, 19, and 20
Compound
Odor description
Tabac, tonka, green, phenolic
Strong ambergris, balsamic, erogenic
Animalic, ambergris, weaker than 13a
Woody, ambergris, weaker than 13a
Woody, phenolic, leather, weak, cedarwood
Woody, ambergris, creamy, powdery
group (!13d) decreases the strength of the ambergris odor. This is also the case for
t
variations of the Bu group at C(5), and for replacement of the tetrahydrofuran ring by
a hexahydropyran ring. In summary, it was observed that none of the variations have
resulted in an improved olfactory performance of 13a.
Conclusions. – The motivation behind the design of the ꢁseco-Ambroxideꢂ 13a was
described. This molecule has been synthesized in four-steps starting from readily
ꢀ
available educts. The olfactory properties are very similar to those of Ambroxide (2),
but 13a has not the same substantivity as 2. Several molecules with structural variations
at different positions of 13a have been synthesized, to clearly demonstrate the scope of
these structures as ambergris odorants.

CHEMISTRY & BIODIVERSITY – Vol. 11 (2014)
1647
Experimental Part
General. Unless stated otherwise, all reactions were performed under N . Reagents and solvents
2
were purchased from SigmaꢀAldrich (DE-Deisenhofen) or Acros Organics (DE-Schwerte), and used
without further purification. All solvents were purified by distillation before use. Mixtures were washed
with 5% soln. of Na CO , NaOH, and brine, or 10% H SO and dried (Na SO ), unless noted otherwise.
2
3
2
4
2
4
Column chromatography (CC): Merck silica gel 60 (63–200 mm). Isomer (e.g., (E)/(Z)) ratios were
determined via GC unless otherwise stated. GC: Hewlett Packard 6890 with FID and GC-sniffing-port
(
0
He; column, DB-Wax, 60 mꢁ0.25 mmꢁ0.25 mm; 50–2408 at 48/min; or HP5, 60 mꢁ0.25 mmꢁ
1 13
.25 mm; 60–2508 at 48/min). NMR Spectra: Varian Mercury ( H: 400, C: 100 MHz) in CDCl ; d in ppm
3
rel. to Me Si as internal standard, J in Hz. EI-MS: Hewlett Packard 5973N (He; column, DB-Wax, 60 mꢁ
4
0
.25 mmꢁ0.25 mm; 60–2408 at 48/min) in EI mode (70 eV); in m/z (%). HR-MS: MAT-8200 Finnigan in
EI mode (70 eV).
General Procedure for a-Alkenylation of 14 to 4-Alkylcyclohexanone (15). 4-(tert-Butyl)cyclohex-
anone (14; 107.8 g, 0.7 mol), acetone dimethyl acetal (79.8 g, 0.76 mol), allyl alcohol (98 g), TsOH
(
9
0.14 g), and cyclohexane (230 ml) were heated under normal pressure until the bottom temp. raised to
18. The org. phase was treated with a soln. of MeONa (0.15 g) in MeOH (14 ml), and hexane (200 ml)
was added to the org. phase. The org. layer was washed with H O to neutrality and concentrated on a
2
rotary evaporator (crude yield, 174.3 g). Crude product (174.3 g), TsOH (0.15 g), and toluene (200 ml)
were heated to a bottom temp. of 1358. Allylic alcohol (35 g) passed over. For workup, the org. phase was
washed with aq. Na CO soln. to neutrality, followed by H O, and concentrated on a rotary evaporator
2
3
2
(
crude yield, 143.9 g). The crude oil was stirred with EtONa as described in [19]. After washing with H O
2
and phase separation, the residue was purified by distillation (768/0.2 mbar) at a 20-cm Vigreux column to
afford pure 15a (98.7 g, 73%).
(
2R,4R)-4-(tert-Butyl)-2-(prop-2-en-1-yl)cyclohexanone (15a) and (2R,4R)-4-(1-Methylethyl)-2-
1
13
(
(
prop-2-en-1-yl)cyclohexanone (15b): H-NMR: cis and trans, see [19]. C-NMR: see [19]. MS: 137
50), 109 (36), 95 (31), 93 (17), 83 (19), 67 (24), 57 (100), 55 (38), 41 (37), 29 (17).
General Procedure for Grignard Reaction of a-Alkenyl-4-alkylcyclohexanones 15 to 2-Allyl-
cyclohexanols 16. In a 500-ml three-necked flask, MeMgCl soln. (100 g, 0.294 mol) was placed under
N as a 22% soln. in THF. Then, without cooling, 15a (44 g, 0.226 mol), dissolved in dry THF (226 ml),
2
was added dropwise during 1–1.5 h, upon which the temp. rose to a maximum of 478. The mixture was
refluxed (668) for a further 5 h, prior to cooling to 0–58, and poured into 600 g of ice/H O 1:1. THF was
2
t
removed on the rotary evaporator at 608/150–350 mbar. BuOMe (220 ml) and solid NH Cl (106 g)
4
t
dissolved in H O (300 ml) were added. After phase separation, the aq. phase was extracted with BuOMe
2
(
220 ml). The combined org. phases were washed twice with 5% NaCl soln. (110 ml), and the solvent was
evaporated at 608 on the rotary evaporation. The crude product was purified by bulb-to-bulb distillation
120–1458/0.4 mbar) to afford pure 16a and 16b (40 g, 85%; GC: 93% sum of two isomers) as a colorless
oil.
(
(
1S,2R,4R)- and (1R,2R,4R)-4-(tert-Butyl)-1-methyl-2-(prop-2-en-1-yl)cyclohexanol (16a and 16b,
1
resp.). Main isomer: H-NMR: 5.83 (dddd, J¼17.1, 10.1, 8.4, 5.8, 1 H); 5.09–4.97 (m, 2 H); 2.40 (dddt,
J¼14.1, 5.8, 3.4, 1.7, 1 H); 1.93 (dddt, J¼14.1, 9.4, 8.4, 1.0, 1 H); 1.76–1.60 (m, 2 H); 1.60–1.50 (m, 1 H);
13
1
7
.38 (td, J¼13.3, 3.6, 1 H); 1.33–1.24 (m, 2 H); 1.22 (s, 3 H); 1.03–0.94 (m, 2 H). C-NMR: 138.3; 115.7;
0.9; 47.9; 45.7; 41.1; 34.7; 32.5; 28.6; 28.1; 27.6; 22.7. Peak 1 (main isomer, GC: 81%): MS: 152 (62), 111
(
51), 95 (65), 83 (57), 71 (100), 69 (43), 57 (90), 55 (43), 43 (68), 41 (48); Peak 2 (GC: 12%): 135 (44),
1
11 (50), 95 (59), 83 (52), 71 (100), 69 (44), 57 (98), 55 (46), 43 (74), 41 (53).
1
(
1S,2R,4R)-4-(tert-Butyl)-1-ethyl-2-(prop-2-en-1-yl)cyclohexanol (16d). H-NMR: 5.84 (dddd, J¼
1
7.1, 10.1, 8.5, 5.9, 1 H); 5.07–4.98 (m, 2 H); 2.44–2.19 (m, 1 H); 1.89 (dt, J¼14.1, 9.0, 1 H); 1.69–1.57
(
m, 3 H); 1.57 (q, J¼7.6, 2 H); 1.45–1.21 (m, 3 H); 1.10–0.92 (m, 2 H); 0.87 (t, J¼7.6, 3 H); 0.86 (s, 9 H).
C-NMR: 138.3; 115.8; 73.2; 47.8; 42.7; 36.4; 34.2; 33.2; 32.5; 27.9; 27.6; 27.5; 22.4; 8.3. MS: 195 (60), 95
20), 85 (66), 83 (26), 69 (27), 57 (100), 55 (32), 43 (26), 41 (31), 29 (23).
1
3
(
(
1S,2R,4R)-4-(tert-Butyl)-2-(prop-2-en-1-yl)cyclohexanol (16c). Compounds 15a and 15b (75 g,
0
.387 mol) were dissolved in MeOH (300 ml) and a soln. of NaBH (4.9 g, 0.129 mol) in NaOH (1%)
4
(
49 g) was added during 1 h at 0–108. Stirring was continued for an additional 3 h, prior to removal of the

1
648
CHEMISTRY & BIODIVERSITY – Vol. 11 (2014)
t
MeOH on the rotary evaporator. The resulting residue was dissolved in BuOMe (200 ml), washed twice
with H O, and concentrated in vacuo. The crude product was distilled on a 5-cm Vigreux column (1088/
2
1
1
5
1
mbar) to afford 16c (65 g, 85%, GC: 90%, two isomers). 16c: Peak 1: H-NMR: 5.94–5.72 (m, 2 H);
.11–4.97 (m, 4 H); 3.86 (q, J¼2.7, 1 H); 3.22 (td, J¼10.4, 4.5, 1 H); 2.50–2.34 (m, 1 H); 2.16 (dtt, J¼
5.1, 7.5, 1.2, 1 H); 2.10–1.95 (m, 3 H); 1.95–1.86 (m, 1 H); 1.84–1.68 (m, 2 H); 1.58–1.40 (m, 4 H); 1.40–
1
.24 (m, 3 H); 1.14–0.96 (m, 4 H); 0.86 (s, 9 H); 0.84 (s, 9 H); 0.79–0.68 (m, 1 H). 13C-NMR: 137.5; 137.4;
1
2
16.1; 115.8; 74.9; 67.8; 47.9; 47.3; 44.9; 42.0; 37.9; 37.8; 35.8; 33.8; 32.5; 32.3; 31.5; 27.6; 27.5; 27.2; 25.7;
0.6. MS: Peak 1 (GC: 52%): 154 (42), 139 (62), 122 (33), 121 (58), 93 (44), 81 (95), 80 (38), 79 (51), 57
(
(
100), 41 (43); Peak 2 (GC: 38%): 139 (38), 136 (44), 123 (38), 122 (36), 121 (64), 95 (33), 93 (33), 81
56), 57 (100), 41 (38).
General Procedure for the Ozonolysis of 2-Alkenyl-1-alkyl-4-alkylcyclohexanols (16) to 17. In a 500-
ml three-necked flask equipped with magnetic stirrer, gas inlet, O generator (Ozonia), and external
3
i
cooling (dry ice/ PrOH), 16a (21.4 g, 0.1 mol) was dissolved in MeOH (214 mol). Then, in ca. 1 h at ꢀ10
to 08 O -containing O (O content, ca. 40 g O /ccm) was supplied to the soln. until no more O was
3
2
3
3
3
absorbed. For the reduction, NaBH (5.7 g, 0.15 mol) was dissolved in 1% NaOH (57 g), and added in 1 h
4
dropwise at ꢀ108 to 08, upon which a strong exothermic reaction was observed. To complete the reaction,
it was stirred for 3 h at r.t., H O (200 ml) and toluene (200 ml) were added to the mixture. After addition
2
of solid NaCl (72 g), the mixture was stirred for additional 20 min. After phase separation, the org. layer
was washed with 5% brine (50 ml), and the solvent was removed on the rotary evaporator. The crude
product (20.5 g, 96%; GC: 94% sum of two isomers) was directly used in the next step.
(
1S,2S,4R)- and (1R,2S,4R)-4-(tert-Butyl)-2-(2-hydroxyethyl)-1-methylcyclohexanol (17a and 17b,
1
resp.). Peak 1: H-NMR: 3.76 (ddd, J¼11.0, 7.7, 5.2, 1 H); 3.63 (q, J¼9.8, 7.7, 1 H); 1.77–1.67 (m, 2 H);
1
.61–1.53 (m, 1 H); 1.52–1.37 (m, 3 H); 1.24 (s, 3 H); 1.04 (m, 1 H); 0.87 (s, 9 H). 13C-NMR: 70.6; 59.9;
4
7
8.2; 43.5; 41.2; 33.2; 32.5; 28.6; 27.8; 27.6; 22.7. Main isomer (GC: 80%): MS: 129 (40), 111 (29), 83 (45),
2 (23), 71 (100), 69 (35), 57 (52), 55 (29), 43 (44), 41 (27).
(
1S,2S,4R)-4-(tert-Butyl)-2-(2-hydroxyethyl)cyclohexanol (17c). MS: 83 (26), 81 (28), 80 (23), 79
37), 69 (25), 67 (24), 57 (100), 55 (45), 41 (39), 29 (23).
1S,2R,4R)-4-(tert-Butyl)-2-(3-hydroxypropyl)-1-methylcyclohexanol (18). Under N , 16a/16b
(
(
(
2
10.5 g, 0.05 mol) was dissolved in THF (17 ml). Then, BH ꢀTHF complex (1m in THF; 20 ml,
3
0
.02 mol) was added with slight cooling. The mixture was stirred for 1 h at r.t. After addition of H O
2
(
3
5 ml), first 3m NaOH soln. (6.7 ml) and then 30% H O soln. (7.4 g, 0.065 mol) were added dropwise at
0–358. The mixture was stirred for 1 h at 508, prior to dilution with BuOMe (50 ml), washed with 20%
2 2
t
Na SO soln. (40 ml), and then evaporated in vacuo. The crude product (12.3 g, GC: 91%) was directly
2
3
1
used in the next step. H-NMR: 3.64 (t, J¼6.1, 2 H); 1.74–1.70 (m, 5 H); 1.60–1.55 (m, 1 H); 1.50–1.44
(
0
m, 1 H); 1.38–1.34 (m, 1 H); 1.27–1.24 (m, 1 H); 1.20 (s, 3 H); 1.17–1.05 (m, 2 H); 1.01–0.97 (m, 1 H);
.86 (s, 9 H). C-NMR: 71.0; 63.3; 47.9; 45.8; 41.1; 32.5; 31.1; 28.5; 28.3; 27.6; 25.8; 22.6.
13
General Procedure for Cyclization of Diols (17) to 13. To 17a/17b (23.3 g, 0.1 mol) as a 8.6% soln. in
toluene, 50% NaOH soln. (20 g, 0.26 mol), tris(2,6-dioxaheptyl)amine (TDA-1; 1 g), and TsCl (26 g,
.14 mol) were added at 608 in 1.5 h. The reaction was slightly exothermic and stirred for 3 h at 508 and
washed two times at 50–608 with 50 ml of H O. Toluene was removed by rotary evaporation at 608/200–
0
2
2
0 mbar. The residue (24 g) was dissolved in EtOH (100 ml), and the resulting mixture was stirred for 4 h
at 608 with 10% aq. NaOH soln. (44 g) to remove traces of sulfonates. EtOH was evaporated at 608/200–
t
5
0 mbar, and the resulting residue was treated with H O (25 ml), and then extracted twice with BuOMe
2
(
75 ml). The org. phases were combined and washed twice with 5% NaCl soln. (50 ml). After
evaporation (608/500–20 mbar), a crude product (18 g) was obtained, which was further purified by bulb-
to-bulb distillation (64–738/2 mbar) to afford 13a/13b (13 g, 66%; GC: 94%, sum of two isomers). Two g
of the mixture were separated by prep. GC to afford the pure isomers.
(
3aS,5R,7aR)-5-(tert-Butyl)octahydro-7a-methyl-1-benzofuran (13a). 1H-NMR: 3.96–3.78 (m,
2
2
9
2
H); 2.11–1.93 (m, 2 H); 1.93–1.84 (m, 1 H); 1.84–1.76 (m, 1 H); 1.71–1.61 (m, 1 H); 1.59–1.42 (m,
H); 1.40–1.28 (m, 1 H); 1.27 (s, 3 H); 1.26–1.17 (m, 1 H); 0.97 (tdd, J¼13.0, 11.7, 3.9, 1 H); 0.86 (s,
13
H). C-NMR: 80.32; 77.33; 77.02; 76.70; 64.70; 44.11; 41.48; 33.79; 32.11; 30.19; 27.59; 26.20; 24.81;
4.50. MS: 182 (16), 181 (100), 97 (98), 83 (10), 81 (12), 57 (29), 55 (22), 43 (30), 41 (21), 29 (11). Odor:
strong ambergris, balsamic, erogenic.

CHEMISTRY & BIODIVERSITY – Vol. 11 (2014)
1649
(
3aS,5R,7aS)-5-(tert-Butyl)octahydro-7a-methyl-1-benzofuran (13b). 1H-NMR: 3.99–3.81 (m,
2
H); 2.36–2.20 (m, 1 H); 2.05–1.93 (m, 1 H); 1.77 (dt, J¼12.1, 6.3, 1 H); 1.63–1.46 (m, 3 H); 1.50–
1
.36 (m, 1 H); 1.25–1.11 (m, 1 H); 1.08 (s, 3 H); 1.02–0.89 (m, 1 H); 0.84 (s, 9 H); 0.89–0.77 (m, 1 H).
13
C-NMR: 80.28; 77.34; 77.02; 76.70; 64.37; 46.81; 42.90; 35.95; 32.45; 32.32; 30.79; 27.47; 26.80; 22.74. MS:
96 (9), 181 (14), 98 (11), 97 (100), 84 (9), 57 (10), 55 (9), 43 (32), 41 (12), 29 (6). Odor: weak, woody,
1
greasy.
5
-(tert-Butyl)octahydro-1-benzofuran (13c). Two isomers (1:1). 1H-NMR: 3.99–3.88 (m, 3 H);
3
.84–3.77 (m, 2 H); 2.95 (ddd, J¼11.1, 9.9, 3.9, 1 H); 2.15–2.05 (m, 3 H); 2.05–1.93 (m, 3 H); 1.93–1.80
(
m, 2 H); 1.72–1.45 (m, 4 H); 1.45–1.25 (m, 3 H); 1.24–1.02 (m, 3 H); 1.01–0.89 (m, 2 H); 0.88 (s, 9 H);
13
0
.85 (s, 9 H). C-NMR: 83.5; 77.4; 67.2; 65.8; 48.0; 46.8; 45.6; 38.7; 33.1; 32.4; 32.4; 31.1; 30.8; 29.7; 29.0;
2
8.8; 27.9; 27.5; 25.3; 21.3. MS: 126 (33), 93 (16), 83 (98), 81 (18), 79 (25), 67 (18), 57 (100), 55 (40), 41
(
36), 29 (21). Odor: tabac, tonka, green, phenolic.
-(tert-Butyl)-7a-ethyloctahydro-1-benzofuran (13d). Two isomers (GC: 10 :1). Main isomer:
5
1
H-NMR: 3.95–3.87 (m, 1 H); 3.86–3.75 (m, 1 H); 2.21 (dddd, J¼12.1, 9.8, 8.4, 6.7, 1 H); 1.95 (dt, J¼
1
4.2, 3.7, 1 H); 1.89–1.77 (m, 1 H); 1.64–1.57 (m, 1 H); 1.57–1.44 (m, 3 H); 1.38 (ddd, J¼14.2, 12.4, 4.6,
H); 1.32–1.21 (m, 1 H); 1.21–1.09 (m, 1 H); 0.96–0.83 (m, 2 H); 0.85 (t, J¼7.5, 3 H); 0.84 (s, 9 H).
C-NMR: 82.6; 64.1; 46.5; 41.1; 32.6; 32.3; 31.8; 31.3; 31.0; 27.4; 22.5; 8.2. MS: 182 (13), 181 (100), 111
16), 97 (20), 81 (7), 69 (9), 57 (25), 55 (16), 41 (11), 29 (8). Odor: animalic, ambergris, weaker than 13a.
1
13
(
5-(1,1-Dimethylpropyl)octahydro-7a-methyl-1-benzofuran (13e). GC: Two isomers. MS: Peak 1: 196
(
(
13), 195 (100), 97 (95), 95 (10), 81 (10), 71 (15), 69 (15), 55 (19), 43 (30), 41 (11). Peak 2: 210 (4), 195
6), 98 (7), 97 (100), 84 (5), 71 (5), 69 (4), 55 (7), 43 (18), 41 (5). Odor: woody, ambra.
5
-(1,1-Dimethylpropyl)octahydro-2-methyl-benzofuran (19). Starting material, 16e (22.5 g); as
1
described for 13a/13b. Yield: 13 g (70%, GC: 98.7%, sum of two isomers). Main isomer: H-NMR:
4
.06 (ddt, J¼10.0, 6.1, 6.0, 1 H); 3.89 (dt, J¼10.0, 6.7, 1 H); 2.51 (dddd, J¼14.8, 12.8, 6.3, 2.4, 1 H); 1.93–
1
.68 (m, 3 H); 1.65–1.53 (m, 2 H); 1.41–1.15 (m, 6 H); 1.29 (d, J¼6.1, 3 H); 0.83–0.75 (m, 9 H).
13
C-NMR: 77.5; 75.3; 39.1; 38.8; 37.3; 34.6; 32.7; 31.7; 26.4; 24.3; 24.0; 23.7; 22.8; 8.1. MS: 163 (29), 150
31), 139 (36), 121 (47), 97 (100), 95 (74), 81 (45), 79 (37), 71 (69), 43 (30). Odor: woody, phenolic,
(
leather, weak, cedarwood.
6
-(tert-Butyl)octahydro-8a-methyl-2H-1-benzopyran (20). Starting material, 18 (12.3 g, 0.05 mol); as
1
described for 13a/13b. Yield: 5.4 g (51%; GC: 98.7%, sum of two isomers). Main isomer: H-NMR: 3.78
(
1
(
2
ddd, J¼12.7, 11.7, 3.0, 1 H); 3.73–3.65 (m, 1 H); 2.03 (ddt, J¼13.1, 8.8, 4.4, 1 H); 1.92–1.80 (m, 1 H);
.79–1.73 (m, 1 H); 1.66–1.55 (m, 1 H); 1.50–1.44 (m, 1 H); 1.40–1.25 (m, 5 H); 1.24 (s, 3 H); 1.25–1.14
13
m, 1 H); 1.04–0.95 (m, 1 H); 0.86 (s, 9 H). C-NMR: 71.4; 61.6; 48.3; 41.2; 38.6; 32.4; 28.0; 27.6; 25.1;
3.3; 22.60; 20.6. MS: 196 (11), 195 (81), 111 (100), 95 (11), 71 (15), 69 (11), 57 (21), 55 (14), 43 (25), 41
(
22). Odor: woody, ambra.
We are grateful to Marc Mansfeld and Rolf Borchers for experimental work, Dr. Wiebke Zander,
Stephan Seilwind, Carsten Strempel, Beate Hartmann, Andreas Kretzer, and Ferdinand Schrçder for
NMR, GC/MS, and EI-MS recordings, Matthias Werner for the olfactory evaluations, Dr. Marco Singer
for the sensory tests, and Bodo Schwarze for molecular modeling.
REFERENCES
[
[
[
[
[
1] K. Mori, H. Tamura, Liebigs Ann. Chem. 1990, 361.
2] E. Lederer, F. Marx, D. Mercier, G. P e´ rot, Helv. Chim. Acta 1946, 29, 1354.
3] G. Ohloff, K.-H. Schulte-Elte, B. L. Mꢀller, Helv. Chim. Acta 1977, 60, 2763.
4] M. Stoll, M. Hinder, Helv. Chim. Acta 1950, 33, 1251.
5] G. Ohloff, ꢁRelationship in between odor sensation and configuration of decalin ring compoundsꢂ,
Gustation and Olfaction, International Symposium, Geneva, June 1970, Academic Press, London,
New York, 1971, p. 178.
[
6] G. Ohloff, W. Pickenhagen, P. Kraft, ꢁScent and Chemistry – The Molecular World of Odorsꢂ, Verlag
Helvetica Chimica Acta, Zꢀrich, WileyꢀVCH, Weinheim, 2012, p. 98–100.

1
650
CHEMISTRY & BIODIVERSITY – Vol. 11 (2014)
[
[
[
7] G. Ohloff, W. Giersch, W. Pickenhagen, A. Furrer, B. Frei, Helv. Chim. Acta 1985, 68, 2022.
8] E. Klein, E.-J. Brunke, Dragoco Report 1979, 278.
9] R. Boden, P. D. Jones, to International Flavors & Fragrances, US 20070942020, 2007.
[
10] R. L. Snowden, S. D. Escher, to Firmenich, Eur. Pat. Appl. 558, 928, 1993 (Chem. Abstr. 1994, 120,
07408).
1
[
[
[
11] C. Fehr, I. Farris, Angew. Chem. 2006, 118, 7058; Angew. Chem., Int. Ed. 2006, 45, 6904.
12] K. Sestanj, Croat. Chem. Acta 1962, 43, 211.
13] T. Kobayashi, H. Sugiyama, H. Tsuruta, T. Yoshida, to Takasago Perfumery, Jpn. Kokai Tokkyo
Koho 1, 323, 532 (1978) (Chem. Abstr. 1980, 93, 149797).
[
[
[
[
[
[
[
[
[
[
14] P. Kraft, C. Denis, W. Eichenberger, Eur. J. Org. Chem. 2001, 2363.
15] G. Bꢀchi, H. Wꢀest, Helv. Chim. Acta 1989, 72, 996.
16] S. Karady, M. Lenfant, R. E. Wolff, Bull. Chim. Fra. 1965, 9, 2472.
17] I. Ibrahem, A. Cordova, Angew. Chem., Int. Ed. 2006, 45, 1952.
18] G. A. Molander, K. O. Kimberly, J. Org. Chem. 1993, 58, 5931.
19] T. Toru, T. Okumura, Y. Ueno, J. Org. Chem. 1990, 55, 1277.
20] J. M. Conia, P. Briet, Tetrahedron Lett. 1964, 39, 2797.
21] K. Hiroi, S. Yamada, Chem. Pharm. Bull. 1973, 21, 47.
22] J. Weißmꢀller, D. Berg, S. Dutzmann, to Bayer AG, DE 3740127 (1987).
23] Software Package Maestro v. 9.5, Fa. Schrçdinger.
Received April 14, 2014