Synthesis and properties of conformationally constrained analogues of floral-type odorants

Helvetica Chimica Acta Vol. 87 (2004)

1767

Synthesis and Properties of Conformationally Constrained Analogues of

Floral-Type Odorants¹)

by Serge Lamboley^a), Ce¬dric Morel^a), Jean-Yves de Saint Laumer^a), Andre¬ F. Boschung^a),

Nigel G. J. Richards^b), and Be¬at M. Winter*^a)

^a) Firmenich SA, Corporate R&D Division, P.O. Box 239, CH-1211 Geneva 8

^b) Department of Chemistry, University of Florida, Gainesville, FL 32611, USA

Dedicated to Dr. G¸nther Ohloff on the occasion of his 80th birthday

The twelve bridged analogues 8 19 of floral-type odorants related to cyclamenaldehyde (1) were

synthesized (Schemes 1 5) to investigate the relationship between the structural and conformational features

of these compounds and their odor properties. Comparison of the data from sensory evaluation and molecular

modeling suggests that the side chain of both the unconstrained and the constrained active analogues is not

extended (anti) but rather folded (gauche) in the −bioactive× conformation. However, it is mainly the nature of

the substituents at the a position of the aldehyde function that critically influences the odor quality and strength.

These studies provide new information that should aid ongoing efforts to develop models of odorantÀreceptor

interactions.

1. Introduction.

A commercially significant class of floral-type odorants is

represented by 3-(4-isopropylphenyl)-2-methylpropanal (1; cyclamenaldehyde) [2], 3-

¾

[4-(tert-butyl)phenyl]-2-methylpropanal ( Lilial ²); 2) [3a] and 3-[4-(tert-butyl)phen-

¾

yl]propanal ( Bourgeonal ³); 3) [3b], which all exhibit a desirable lily-of-the-valley

(−muguet×) tonality. Though many other compounds with different functional groups

(alcohols, hydroxyaldehydes, acetals) belong to this class of odorants, in the present

work, we will focus on aldehydes related to compounds 1 3. StructureÀodor

relationships in this class of odorants, among others, have been reviewed [4]. A

computational model for lily-of-the-valley-scented compounds, based on AM1

calculations, was proposed by Pelzer et al. [5], which distinguished between two

structural fragments bearing either a carbonyl or a hydroxy functional group. More

recently, conformational analysis of a series of −muguet×-type odorants was used to

design and synthesize candidates anticipated to possess a lily-of-the-valley odor [6].

Interest in this group of compounds has also been stimulated by the demonstration that

¾

a mixture of Lilial (2) and Lyral ⁴) (4) [7], a compound related to hydroxycitronellal

(5) and that also exhibits a characteristic −muguet× odor, was able to stimulate a

heterologously expressed rat olfactory receptor protein, OR5, and to generate second-

messenger responses at submicromolar concentrations [8]. Shortly after, computer-

1

)

Parts of this work have been presented as a poster at the 12th Congress of the European Chemoreception

Research Organization (ECRO XII), Zurich, Switzerland, August 25 31, 1996, see [1].

Trade name registered by Givaudan.

Trade name registered by Quest International.

Trade name registered by International Flavors & Fragrances Inc.

2

3

4

)

1768

Helvetica Chimica Acta Vol. 87 (2004)

¾

aided molecular-modeling studies of the interactions of OR5 with Lyral (4) [9] and

¾

with (À)-(R)-Lilial (2) [10] were published. Subsequently, specific binding of Lilial

(2) with OR5 was assessed by photoaffinity labeling experiments and by tryptophan

¾

fluorescence measurements [11]. And, more recently, Lilial (2) and Bourgeonal (3)

were shown to be powerful agonists of an odorant receptor that may be involved in

mediating human sperm chemotaxis [12]. In previous work aimed at defining the

−bioactive× conformation of floral odorants represented by compounds 1 3, we had

prepared the conformationally restricted ether-type analogues 6 and 7 as mimics of a

folded (gauche) side-chain conformation of 2 and 3, respectively, and neither 6 nor 7

exhibited the typical −muguet× activity [13].

In continuation of that work, we prepared a series of aldehyde-type analogues (see

8 19) of 2 and 3, with an identical lipophilic region, and in which the side chain is

constrained into a more-limited number of conformations intended to mimic extended

side-chain conformations, except for analogues 18 and 19. We now report the synthesis

and sensory characterization of these novel odorants, together with the results of

molecular-modeling experiments aimed at a better understanding of structureÀodor

relationships for this type of odorants.

2. Results. 2.1. Synthesis. As shown in Scheme 1, we initially synthesized the indane

¾

derivative 8, the proximate conformationally constrained analogue of Lilial (2), and

the corresponding methylated derivative 9. Compound 9 can be seen as a conforma-

tionally restricted analogue of aldehyde 20 [14]. We note that compound 20 has been

patented as a stabilizer for perfume formulations [14a] and as an odorous compound

[14b] with a fresher note than 1 and 2. The related compound 21 [15], having the same

¾

carbonyl-bearing side chain as 20, is known as Floralozone ⁵).

Thus, the enolate of 5-(tert-butyl)indan-1-one 22 [16] was treated with diethyl

carbonate to afford b-keto ester 23, which was methylated by reaction with MeI in the

presence of K₂CO₃to give keto ester 24. Hydrogenation of 23 in AcOEt gave the ester

25, while hydrogenation of 24 needed AcOH as solvent to go to completion and give

ester 26. Reduction of esters 25 and 26 to the alcohols 27 and 28, respectively, was

5

)

Trade name registered by International Flavors & Fragrances Inc.

Helvetica Chimica Acta Vol. 87 (2004)

1769

accomplished with LiAlH₄, and subsequent pyridinium chlorochromate (PCC)

oxidation gave the desired indane-aldehydes 8 and 9, respectively (Scheme 1).

Analogue 8 turned out to be a valuable odorant [17a], having an odor quality very

similar to that of 2 and 3, with superior intensity. In contrast, 9 [17a] had none of the

odor characteristics of 2, 3, or 8 (see Table 1).

Scheme 1

a) NaH, (EtO)₂CO, toluene, 608, 8 h. b) MeI, K₂CO₃, THF, 658, 3 h. c) H₂(1 atm), 5% Pd/C, AcOEt, r.t., 3 h.

d) H₂(1 atm), 5% Pd/C, AcOH, r.t., 17 d. e) LiAlH₄, Et₂O, 258, 2 h. f) PCC, CH₂Cl₂, 258, 5 h.

1770

Helvetica Chimica Acta Vol. 87 (2004)

Table 1. Odor Description of Compounds 2, 3, and 8 19. With the exception of 3, all compounds are racemates.

Odor description

¾

2 (Lilial )

−Sweet, yet refreshing and intensely floral; green odor of considerable radiance×^a)

−Floral, lily-of-the-valley, aldehydic, green×

−Floral, green, −muguet×, Bourgeonal , powerful, tenacious×

¾

3 (Bourgeonal )

¾

8

9

10

−Watery, metallic, aldehydic, green, somewhat fatty, vaguely phenolic×

¾

−Floral, Lilial , white flower, metallic, slightly insecticide, weak×

¾

11

12

13

14

15

16

17

18

19

−Bourgeonal , aldehydic, powerful, pleasant×

−Aldehydic, phenolic, leathery, waxy, green, weak×

−Aldehydic, phenolic, weak×

¾

−Lilial , Bourgeonal , hydroxycitronellal, too weak×

−Floral, vaguely green, very weak×

−Woody, dry, very weak×

−Floral, chemical, plastic, very weak×

−Woody, dry, cedar, old wood, very weak×

−Odorless×

^a) Taken from [17b].

¾

Given the attractive properties of the conformationally constrained Lilial

analogue 8, we next synthesized the slightly less-constrained indane analogues 10

and 11 [17a], as shown in Scheme 2. Compounds 10 and 11 were prepared by identical

routes starting again from the indanone 22 [16] and from the indanone 32 [18],

respectively. Treatment of 22 with methyl (dimethoxyphosphinyl)acetate gave a

mixture of esters 29 31, which was hydrogenated to afford ester 36; LiAlH₄reduction

of 36 ( ! 38) followed by PCC oxidation gave aldehyde 10. The same sequence of

reactions on indanone 32 gave aldehyde 11 in good overall yield via 33 35 and 37 and

39. Analogue 10 had only a weak floral odor with metallic side notes; analogue 11

possessed a powerful and pleasant odor close to 3 and 8 (see Table 1).

Scheme 2

a) (MeO)₂POCH₂COOMe, NaOMe, petroleum ether (30 508), r.t., 24 h. b) H₂, 5% Pd/C, AcOEt, r.t., 1 h.

c) LiAlH₄, Et₂O, r.t. to reflux, 2 h. d) PCC, CH₂Cl₂, r.t., 3 5 h.

Helvetica Chimica Acta Vol. 87 (2004)

1771

We then synthesized the homologues of indane 8, namely the tetralin derivatives 12

and 13, again by identical routes starting from the known tetralones 40 [19] and 42 [19]

[20], respectively (Scheme 3). Treatment of 40 with the Vilsmeier reagent [21] gave

chloro-aldehyde 41 in good yield. Reduction and dechlorination to give a 2 :1 mixture

of alcohol 44 and the desired aldehyde 12 was accomplished by medium-pressure

hydrogenation over 5% Pd/C as catalyst under basic conditions. PCC Oxidation of

alcohol 44 in the mixture then gave the substituted tetralinaldehyde 12. The same

sequence of procedures, starting from 42, afforded the target aldehyde 13 via 43 and 45.

Both tetralinaldehydes 12 and 13 had only weak aldehydic odors (see Table 1).

Scheme 3

a) DMF, POCl₃, 8 08, 4 h. b) H₂(50 psi), 5% Pd/C, K₂CO₃, MeOH/H₂O, r.t., 56 h. c) PCC, NaOAc, CH₂Cl₂, r.t.,

3 h.

Continuing on this path, we envisaged the preparation of the homologues of 12 and

13, namely the 6,7,8,9-tetrahydro-5H-benzocycloheptene derivatives 14 and 15, by a

route similar to that described above, using the ketones 47 and 48, respectively, as

starting materials (see Scheme 4). Although the synthesis of 48 had been reported [22],

its regioisomer 47 seemed unknown. Detailed examination of the literature describing

the preparation of 48, however, revealed a number of inconsistencies. For example,

initial work on the cyclization of 5-[4-(tert-butyl)phenyl]pentanoyl chloride (49) with

AlCl₃in cold nitrobenzene was reported to give a solid product (m.p. 408; semi-

carbazone, m.p. 192 1938; oxime, m.p. 122 1238), which was assigned structure 48

[22a]. Interestingly, it was also noted that the product ketone produced a powerful

odor, reminiscent of that obtained when burning sandalwood, when placed on a hot

plate⁶). In a subsequent paper, the same authors effected cyclization of 49 using AlCl₃

in CS₂and obtained a liquid product that could not be crystallized [22b]. More

importantly, no sandalwood odor was produced on heating the product on a hot plate,

6

)

Although compound 48 was meant in the odor description [23], the structure corresponding to 3-(tert-

butyl)-6,7,8,9-tetrahydro-5H-benzocyclohepten-6-one was printed in error.

1772

Helvetica Chimica Acta Vol. 87 (2004)

Scheme 4

a) CF₃SO₃H, ClCH₂CH₂Cl, 848, 12 h. b) SOCl₂, 608, 15 h. c) AlCl₃, CH₂Cl₂, 08 (1 h) ! r.t. (1 h). d) DMF,

POCl₃, 70 8 08, 1 h. e) H₂(50 psi), 5% Pd/C, K₂CO₃, MeOH/H₂O, r.t., 15 h. f) PCC, AcONa, CH₂Cl₂, r.t., 1.5 h.

and the semicarbazone of this compound melted at 2208. Despite these discrepancies,

the product of the reaction under these conditions was also assigned as 48. In a third

study on methods for the cyclization of the acid chlorides derived from variously

substituted d-phenylvaleric acid, it was stated that whatever the solvent and the catalyst

used, only a low yield of the desired ketones was obtained, except when working under

high dilution [22c]. Therefore, an alternative method was investigated, based on work

of Gilmore and Horton [24], in which the carboxylic acids were treated with a large

excess of P₂O₅in 85% phosphoric acid. Under these conditions it was claimed that the

cyclized products could be obtained in 60 90% yield, including 48 from 46 [22c]. The

Helvetica Chimica Acta Vol. 87 (2004)

1773

oxime derivative of the cyclized material 48 was, however, reported to melt at 1718

rather than 122 1238 as given in the initial paper on the cyclization reaction [22a]. We

suspected that the structural assignments in these studies were being complicated by an

unrecognized migration of the tert-butyl group and that, in fact, ketone 47 was being

formed in these experiments. This phenomenon has been carefully investigated for the

intermolecular acetylation of p-(tert-butyl)toluene [25], and was later clarified in the

case of a related intramolecular cyclization [26].

To substantiate this hypothesis, we reinvestigated the acid-catalyzed cyclization of

acid 46 and the cognate acid chloride 49 (Scheme 4). We observed indeed that ketone

48 was the major product (ratio 47/48 24 :70) when acid chloride 49 was treated with

AlCl₃in CH₂Cl₂at 08; moreover, as reported [22a], ketone 48 was a solid that could be

purified from the reaction mixture by crystallization. By contrast, reaction of acid 46

with trifluoromethanesulfonic acid [27a]⁷) in refluxing 1,2-dichloroethane (848)

predominantly afforded ketone 47 (ratio 47/48 90 :4), which proved to be an oil upon

purification. The regioisomerism of the products 47 and 48 was assessed by using the

chemical shifts of the aromatic-ring protons. Having clarified the details of the

cyclization reaction, we proceeded to prepare the target aldehydes 14 and 15 by

subjecting the ketones 47 and 48 to a Vilsmeier reaction ( ! 50 and 51, resp.), followed

by hydrogenation ( ! 52 and 53, resp.) and oxidation, as used for the preparation of

tetralin derivatives 12 and 13. Aldehyde 14 had the typical −muguet× odor present in 2, 3,

and hydroxycitronellal 5 (used as a descriptor) but was too weak, whereas 15 was found

to exhibit only a weak floral, vaguely green odor (Table 1).

Finally, the synthesis of the tricyclic analogues 16 19 was accomplished by starting

again from the indanones 22 and 32 (Scheme 5). Thus, 22 was reduced with LiAlH₄to

alcohol 54, which was dehydrated by distillation in the presence of KHSO₄to indene 56.

Treatment of 56 with ethyl diazoacetate in the presence of CuSO₄(cf. [28]) afforded a

ca. 2 :1 mixture of −exo×- and −endo×-esters 58 and 59. After separation by column

chromatography, each of the esters 58 and 59 was reduced to the alcohols 62 and 64

respectively, which in turn were oxidized to the aldehydes 16 and 18, respectively. A

similar sequence of reactions starting from indanone 32 gave aldehydes 17 and 19 via

55, 57, 60 61, 63, and 65, respectively. The odor descriptions of compounds 16 19,

compared to 2 and 3, are given in Table 1.

The odor descriptions for the compounds 8 19 synthesized in this study showed

that the extremely constrained analogues 16 19 not only were very weak but had lost

all lily-of-the-valley odor activity (Table 1). Among the less-constrained analogues 8

15, compounds 8 and 11 were the most powerful and the most similar to 2 and 3. In an

attempt to rationalize these findings and to define more precisely the structural/

conformational features associated with the odorants of this family, we carried out two

series of computer-aided molecular-modeling experiments.

2.2 Molecular Modeling. 2.2.1. Preamble. Preliminary molecular-modeling calcu-

lations (MM2, Monte Carlo procedure) indicated that compounds 8 15 could adopt

both extended and folded conformations, although at various energy costs. In contrast,

for compounds 16 and 17, the side chain is restricted to extended conformations,

7

)

For a rare example of CF₃SO₃H-mediated intramolecular acylation of a 4-arylbutanoic acid derivative, see

[27b].

1774

Helvetica Chimica Acta Vol. 87 (2004)

Scheme 5

^a) Eluted first on CC, but second on GC (Sil. and Carb.). ^b) Eluted second on CC, but first on GC (Sil. and

Carb.).

a) LiAlH₄, Et₂O, r.t., 1.5 h. b) KHSO₄, 160 1758/100 25 mbar, 0.5 h. c) KHSO₄, 90 1008/0.2 mbar, 0.5 h.

d) N₂CHCOOEt, CuSO₄, cyclohexane, 75 808, 1 h. e) LiAlH₄, Et₂O, r.t., 0.5 h. f) PCC, CH₂Cl₂, r.t., 3 h.

whereas for compounds 18 and 19 the side chain is constrained into folded

conformations.

2.2.2. Investigation of the Conformational Features. For small molecules, modern,

automated conformational search methods have been developed to the extent where a

complete determination of the populated conformations is feasible [29][30]. We

therefore sought to determine the conformational preferences of 2 and 3, and to

Helvetica Chimica Acta Vol. 87 (2004)

1775

compare them to those of the constrained analogues 8 19. Our approach was to use

the −jump between wells× (JBW) method, which has been successfully used to examine

the conformational properties of both linear and cyclic hydrocarbons [31][32]. This

method combines a −smart× Monte Carlo algorithm for identifying energy minima

populated by the molecule at a given temperature with a stochastic molecular-dynamics

algorithm [33] that allows exploration of the conformational space local to these low-

energy wells. Since the acceptance or rejection of a conformation is based on energetic

changes, the distributions obtained reflect the Boltzmann-weighted populations of

individual molecular conformations [31]. Assuming that all of the compounds

exhibiting −muguet×-type odors were acting at the same receptor, or set of receptors,

the position of the aldehyde function with respect to the hydrophobic region of the

molecule should be crucial. Therefore, we took the distance between the quaternary C-

atom (C_quat) of the tert-butyl substituent and the O-atom of the aldehyde group (O_carb

)

as a simple conformational measure. The C_quatÀO_carbdistances distribution was then

evaluated from the conformer populations calculated for 2, 3, and 8 19, and some

representative examples are shown in Fig. 1. The computational results indicate, as

expected, that the free-chain compounds 2 and 3 have the broadest conformer

population, as reflected by the C_quatÀO_carbdistance domain of 4 ä (5.5 9.5 ä). A

similarly broad conformer population is shared by the more flexible of the constrained

analogues, i.e., by 10 (5.5 9.5 ä) and 11 (4.5 8.5 ä). In contrast, the C_quatÀO_carb

distance domain of compounds 8 and 9 (5.7 8.7 ä), 18 (5.0 8.0 ä), and 19 (4.7

7.7 ä) covers only 3 ä, that of compound 17 (6.7 8.7 ä) only 2 ä, and that of

compound 16 (7.8 9.3 ä) only 1.5 ä.

Intriguingly, the two inactive analogues 18 and 19, constrained in a folded

conformation, have a C_quatÀO_carbdistance domain equal in breadth to 8, the most-active

constrained analogue, although for 18 and 19 the maximum C_quatÀO_carbdistance is

hardly above 7.5 ä, while for 8 it is above 8.5 ä. Nevertheless, all inactive or weakly

active analogues possess C_quatÀO_carbdistances in common with those of the most-active

analogues in the present series (2, 3, 8, and 11). Moreover, in the case of compounds 2

and 3, which have almost identical conformational populations as reflected by the

C_quatÀO_carbdistances, it has been observed that the odor activity of 3 is approximately

two- to four-times stronger than that of 2 [3b][34]. Taken together, these observations

indicated that the conformational features of each individual compound, reflected by

the distance parameter, were not discriminatory enough. Our hypothesis was that all

the compounds eliciting a strong −muguet×-type odor could adopt a defined

conformation which was able to interact similarly with the same olfactory receptor

or set of receptors. In other words, we were looking for a set of structural features

responsible for a defined odor-type sensation, which has been termed −olfactophore×

[35]. Computational olfactophore models have been proposed for several families of

odorants and in particular for −muguet× alcohols and hydroxyaldehydes [4b]. Thus, we

envisaged to construct a computational model ligand integrating the conformational

features of the most-active compounds in this study, which we named −consensus

ligand×.

2.2.3. Construction of a −Consensus Ligand× by Induced Superimpositions. The most

active compounds (R)-2, 3, and 8 (no chirality constraint was imposed upon compound

8, since all evaluations employed racemic mixtures) were minimized in parallel (see

1776

Helvetica Chimica Acta Vol. 87 (2004)

Fig. 1. Calculated C_quatÀO_carbdistances [ä] in the conformational populations (MM2* calculation) of selected

compounds

Exper. Part), while a set of atoms (C_quat, C(1) and C(2) of the benzene ring, C_carband

O_carb) of one molecule were forced to coincide with the corresponding atoms of the

other molecules. As in a classical Monte Carlo multiple conformational search, the

minimizations were repeated starting from randomly chosen conformations of each

individual compound. The result of this procedure was a list of superimpositions of

increasing energy, which could be classified into four types (see Fig. 2). The lowest-

energy superimposition A had both the carbonyl functions and the a-methyl group of

(R)-2 pointing towards the aromatic ring. The next, slightly higher-in-energy super-

imposition B had both the carbonyl functions and the a-methyl group pointing away

from the aromatic ring, and, notably, the chiral C-atom of compound 8 adopted an

Helvetica Chimica Acta Vol. 87 (2004)

1777

Fig. 1 (cont.)

inverse absolute configuration (we noticed that superimposition B reflected a

conformation of (R)-2 similar to the one seen in the published pictures [10] [35b] of

(R)-2 docked at a possible binding site in a model of the putative OR5 odorant

receptor). The next higher-energy superimposition C had again the carbonyl functions

turned towards the aromatic ring, but the a-methyl group of (R)-2 was pointing away

from the aromatic ring, and the chiral C-atom of compound 8 had again adopted an

inverse absolute configuration, as in superimposition B (we noticed that super-

imposition C comprised a conformation of (R)-2 similar to the one obtained in the

superimposition of (R)-2 with the inactive 1,3,4,5-tetrahydro-2-benzoxepin derivative

6, tested earlier as a conformationally restricted analogue [13]). Finally, the highest-

energy superimposition D had the carbonyl function turned towards the −outside×, while

the a-methyl group of (R)-2 was oriented towards the aromatic ring.

The conformations revealed by the superimpositions B and D, although not the

lowest ones in energy, were thought to be more probable as −bioactive× conformations

because the carbonyl function would be more accessible for H-bonding. In previous

studies [36], we observed a correlation between the solvent-accessible surface area

(SASA) of a H-bond-acceptor atom and activity, in two series of odorants. Thus, we

carried out a SASA calculation for the carbonyl O-atom of each compound in the four

superimpositions A D; the results are shown in Table 2. All the compounds in

superimpositions A and C were in the group of small SASAvalues (<42.7 ä²), whereas

all the compounds in superimpositions B and D were in the group of large SASAvalues

(>47.9 ä²), except for compound (R)-2 in superimposition B. In consequence, we

1778

Helvetica Chimica Acta Vol. 87 (2004)

Fig. 2. Types and energies of superimpositions of compounds (R)-2 (white), 3 (red) and 8 (green) obtained by

parallel minimization (the atoms used for the superimpositions are marked in pink)

Table 2. Solvent-Accessible Surface Areas (SASA) of the Carbonyl O-Atom for Compounds (R)-2, 3, and 8 in

the Four Superimpositions A D (probe radius 1.4 ä)

SASA [ä²] of carbonyl O-atom

Superimp. A

Superimp. B

Superimp. C

Superimp. D

(R)-2

3

8

36.7

42.7

40.0

41.2

49.3

48.3

41.3

42.0

39.4

47.9

49.3

48.4

selected the set of positions obtained from superimposition D, in which all the carbonyl

O-atoms have a large SASA value, to define our −consensus-ligand×.

2.2.4. Evaluation of the Energy Penalty for Each Compound of the Set to Fit to the

−Consensus Ligand×. Compounds (R)- and (S)-2, 3, and 8 19 were superimposed onto

the −consensus ligand× at defined positions (C_quat, C(1) and C(2) of the benzene ring,

C_carb, and O_carb). The energy difference (DE) between the −consensus-ligand×-super-

imposed and the ground-state conformers of a given compound, which represents how

well the compound fits the model, was calculated, and the results are shown in Table 3.

The data show that compounds 2, 3, 8 13, 18, and 19 can be superimposed on the

−consensus ligand× with relatively low energy penalties (DE < 11.1 kJ/mol), whereas

Helvetica Chimica Acta Vol. 87 (2004)

1779

analogues 14 17 have considerably higher DE values (DE > 22.5 kJ/mol). However,

among the first group of compounds, the DE values are not able to discriminate

¾

between active and inactive compounds. For example, (S)-2 ( (S)-Lilial ) has a lower

¾

penalty than (R)-2 ((R)-Lilial ), which is in contradiction with the fact that (R)-Lilial

¾

is the active compound, while (S)-Lilial is significantly weaker [37]. Compound 9,

with no lily-of-the-valley odor activity, has a very low penalty compared to the very

active compound 8. Similarly, compound 11, having a much more powerful and typical

odor than 10, has a higher DE value. These results prompted us to investigate in more

detail the crucial substrateÀreceptor(s)-interaction event.

Table 3. Energy Differences (DE) between the Ground-State Conformers of Compounds 2, 3, and 8 19 and

Their Corresponding Conformers Superimposed onto the −Consensus Ligand× and the −Interaction Model×

Odor activity^a)

DE [kJ/mol]

−Consensus ligand×

−Interaction model×

(R)-2

(S)-2

3

8

9

10

11

12

13

14

15

16

17

18

19

1

0

1

0

1

0

7.1

2.1

5.7

4.5

0.2

8.0

9.83.7

8.9

2.6

9.2

1.2

À 1.5

11.0

8.6

25.8

21.9

52.7

34.4

32.0

28.0

35.0

3.3

9.1

22.5

25.7

39.0

32.2

10.3

11.0

^a) The compounds with a strong lily-of-the-valley odor were coded −1×, those with weak or inexistent lily-of-the-

valley odor were coded −0×.

2.2.5. Generation of an −Interaction Model×. We added to the −consensus ligand× the

simplest possible external steric constraint, representing a steric demand of the

receptor, by placing a methane molecule in proximity to the carbonyl group. The

position was chosen to create unfavorable interactions with the inactive compounds

(S)-2, 9, and 10 (see Exper. Part), and a new model, the −interaction model×, including a

methane molecule as a part of the presumed receptor protein(s) was created (see Fig. 3).

Again, the energy difference (DE) between the −interaction-model×-superimposed

and the ground-state conformers of each compound of the set was calculated, and the

results are given in Table 3. As expected, all inactive compounds, except 19⁸), suffered

an unfavorable steric interaction with the added methane molecule, which resulted in

increased DE values. In contrast, we found that the active compounds ((R)-2, 3, 8, and

11) gained favorable Van der Waals interactions with the added methane molecule in

8

)

The very weak odor activity of compounds 16 19 puts into question the significance of this particular

outcome.

1780

Helvetica Chimica Acta Vol. 87 (2004)

Fig. 3. −Interaction model×

the −interaction model×, leading to a decrease in the DE values. The superimpositions of

some selected compounds onto the −interaction-model× are shown in Fig. 4.

3. Discussion. The cloning of the genes encoding olfactory receptors (ORs) in

humans [38] and other mammals (rat, [39a]; mouse, [39b]) has permitted significant

progress in elucidating the molecular mechanisms by which odorants interact with their

target proteins [40]. Specifically, it is likely that receptor activation by odorant ligands

proceeds by a molecular mechanism similar to that already determined for cellular

receptors that interact with hormones, neurotransmitters, and light [41]. In this model,

active (R*) and inactive (R) conformational forms of the OR are in an equilibrium that

is perturbed by ligand binding. Hence an odorant initially binds to the R form of the

OR to yield a complex that undergoes a conformational transition to give R*, which

can then activate G-protein-mediated cell signaling pathways. As a result, some of the

initial binding energy is employed to drive conformational changes of the OR within

the membrane. Elucidation of binding-site structure is, therefore, complicated when

sets of conformationally flexible ligands are employed to −map× receptor recognition, a

problem that is often overcome by preparing and characterizing the activity of

conformationally constrained analogues.

Moreover, in contrast to studies on the interaction of drugs with G-protein-coupled

receptors (GPCRs), the development of structureÀactivity relationships (SARs) for

odorants is complicated by the observation that odorants can (in principle) interact

with several different receptors, so that the olfactory response then arises from a

combination of signals within the brain. This is a particular problem in modern

structure-based methods for constructing structureÀodor relationships (SOR),

because all current methods require that a family of ligands all bind within an identical

target site [42].

Helvetica Chimica Acta Vol. 87 (2004)

1781

Fig. 4. Superimposition of selected compounds onto the −interaction model×

Given that there is strong evidence that the same receptor does accommodate both

¾

Bourgeonal and Lilial [12], however, we believed that access to conformationally

constrained analogues of these odorants would permit us to develop an appropriate

SOR model for this class of compounds with computational approaches. More

specifically, we undertook to explore the ability of two computational approaches to

rationalize the odorant qualities of a series of conformationally well-defined ligands,

both of which have not been applied previously to this problem.

1782

Helvetica Chimica Acta Vol. 87 (2004)

In our first approach, we used the JBW method to determine the conformational

¾

preferences of Bourgeonal , Lilial , and our series of conformationally constrained

analogues. This algorithm provides an estimate of populated conformers based upon

free energy rather than simple molecular-mechanics strain enthalpies. As might have

been expected, however, while these calculations permitted us to evaluate the effects of

molecular constraints upon the ability of a given compound to adopt multiple

conformations, and therefore molecular flexibility, the SOR model developed by

modeling the structural preferences of the free ligands failed to explain the olfactory-

ligand set in a completely consistent manner. Efforts to identify common conforma-

tional features of our set of odorants by constrained molecular superimposition led to a

−consensus ligand× and revealed a C_quatÀO_carbdistance of ca. 7.7 ä, which is in-between

the fully extended (anti) and the fully folded (gauche) conformations. A comparison of

the dimensions given by Pelzer et al. [5] for the odor-active side-chain fragment of

carbonyl-type −muguet× odorants (AM1 calculations) with the ones measured on the

side chain of our −consensus ligand× indicated slightly shorter distances in our model

(Table 4). However, this model was also problematic, given that steric constraints

imposed by receptor-based functional groups are not included in such an approach.

Table 4. Comparison of the Dimensions for the Putative Odor-Active Side-Chain Fragment Obtained by Pelzer×s

Model^a) and by the −Consensus-Ligand×

Pelzer×s model

−Consensus ligand× model (present work)

C(1) ! C(4)

C(1) ! C(5d)

C(1) ! C(5e)

3.90.3 ä

4.4 Æ 0.3 ä

5.0 Æ 0.3 ä

3.14 ä

3.33 ä

4.22 ä

a

)

Structural fragment identified by Pelzer et al. [5] in lily-of-the-valley odorants bearing a carbonyl function.

As a result, it was necessary to introduce additional features to model the existence

of such interactions, without the computational expense of modeling the three-

dimensional structure of the OR explicitly, and thereby to obtain an −interaction model×

to which novel structures could be compared. Importantly, the inclusion of a methane

¾

molecule to introduce steric effects allows our model to reflect the fact that (S)-Lilial

¾

((S)-2) is far less intense than (R)-Lilial ((R)-2) [37]. As far as we are aware, this

represents the first ligand-based model that includes the effects of receptor environ-

ment; at this point, however, there is still room for refinement and improvement of the

model with respect to this particular feature.

Work to elaborate and extend this model is continuing, and results will be reported

in due course.

We gratefully acknowledge the collaboration of Mrs. S. Gallo-Fl¸ckiger for her experimental skill and of

Mr. W. Thommen and Mr. R. Brauchli for the NMR analysis. We wish to thank Dr. P.-A. Blanc for the evaluation

of olfactory properties and Dr. R. L. Snowden for stimulating discussions.

Helvetica Chimica Acta Vol. 87 (2004)

1783

Experimental Part

General. All reactions were performed under N₂. GLC: Hewlett-Packard-5890 instrument equipped with a

flame-ionization detector coupled to a Hewlett-Packard-3395 or -3396A integrator; capillary columns

Chrompack CP-Wax-52 CB (10 m, 0.25 mm i.d.) and CP-Sil-5 CB (10 m, 0.25 mm i.d.). TLC: silica gel 60

(Merck F 254, layer thickness 0.25 mm). Column chromatography (CC): silica gel 60 (Merck, 0.063 0.2 mm,

70 270 mesh, ASTM). Bulb-to-bulb distillation: B¸chi-GKR-50 or -GRK-51 oven; b.p. correspond to the air

temp. IR Spectra (liquid film): Perkin-Elmer-297 or -1600-FTIR spectrometers; nÄ in cm^À^{1 1}H- and ¹³C-NMR

.

Spectra (CDCl₃): Bruker-AMX-360, -DPX-400, or -AV-500 spectrometers; d in ppm downfield from SiMe₄, J in

Hz. MS: HP 5972 or 5973 MSD (70 eV); in m/z (intensity in % rel. to the base peak (100%)).

1. Computer-Aided Molecular Modeling. Conformational Properties. All calculations were carried out with

the MM2*-force-field parameters [43], and the JBW algorithm [29], as implemented in the MacroModel/

BATCHMIN 6.0 software package [44]. Each simulation was run for 10×000 search steps, and all conformations

that possessed a strain energy that was within 50 kJ/mol of the lowest-energy structure were sampled. Duplicate

conformations were identified by using standard superimposition methods [45].

Parallel Minimization. This type of calculation, already used in a previous study [46], was performed by

using the classical Monte Carlo conformational search procedure, as implemented in MacroModel. A single file

containing the structures of the molecules to be superimposed was used as input. The constraints between the

atoms were introduced by fixing interatomic distances equal to 0 (the force constant was 100 kJ/mol ¥ ä²). The

force-field interactions between the atoms of different molecules were removed by using the ASNT key in the

command (.com) file. The result of such calculations was a list of superimpositions classified by energy, the latter

representing the summation of the force-field energy of each molecule.

Testing Molecules on the −Consensus Ligand×. The model is constituted by the atoms fixed at positions

determined in the parallel-minimization procedure. The superimposition of a molecule on the model was

performed by the same procedure as above for the parallel-minimization calculations, with one exception: for

the C-atom of the carbonyl function, a flat-bottom constraint was used, which started the energy penalty only

when the inter-atomic distance was greater than 0.5 ä. This allowed a certain flexibility of the carbonyl-function

orientation. All the atoms of the model were frozen at their initial positions. The force-field interactions

between all the atoms of the molecules and the atoms of the model were removed.

Positioning of the Methane Molecule in the −Interaction Model×. The methane molecule was placed initially

to create unfavorable interactions with the inactive compounds ((S)-2, 9, and 10). The methane position was

then optimized to best interact with the three active compounds ((R)-2, 8, and 11) fixed on the −consensus

ligand×. In this procedure, the energies of the three active molecules and the methane molecule was minimized;

as in the parallel minimization (see above), the force-field interactions between the three active molecules were

removed, but not the ones with the methane molecule. After this, the methane molecule was moved towards the

C-atom in a position of the carbonyl function in compound (R)-2 by 1/10 of the original distance, to maximize

the unfavorable interactions with the inactive compounds, relative to the favorable interactions with the active

compounds. This position of the methane molecule defined the −interaction model×. The coordinates of this

model are available; please contact J.-Y. de Saint Laumer (e-mail: jean-yves.de.saint.laumer@firmenich.com).

Testing Molecules on the −Interaction Model×. The same procedure as above was used, except that the C-

atom of methane was also frozen, while the H-atoms of methane were not constrained. The force-field

interactions between all the atoms of the molecules and the atoms of the model, except the methane, were

removed.

2. Reduction of Esters or Ketones to Alcohols with LiAlH₄: General Procedure A (G.P. A). To a stirred

suspension of LiAlH₄(1.5 equiv.) in Et₂O at r.t. was added dropwise a soln. of the ester or ketone (1.0 equiv.) in

Et₂O, and the mixture was heated to reflux during 2 h. The mixture was cooled to 48, acetone (4.0 equiv.) was

added to consume the excess reagent and then a stoichiometric amount of 1n aq. NaOH. The mixture was stirred

during 0.5 h ( ! r.t.), then Na₂SO₄was added, the solid filtered off, and the filtrate evaporated.

3. Oxidation of Alcohols to Aldehydes with Pyridinium Chlorochromate (PCC): General Procedure B (G.P.

B). To a stirred suspension of PCC (1.5 equiv.) in CH₂Cl₂at r.t. was added dropwise a soln. of the alcohol

(1.0 equiv.) in CH₂Cl₂, and the mixture was stirred at r.t. during 3 5 h. The mixture was diluted with an excess of

¾

Et₂O, filtered through a short column of Florisil (Acros Organics), and evaporated.

4. Compounds 8 and 9. (Æ)-Ethyl 5-(tert-Butyl)-2,3-dihydro-1-oxo-1H-indene-2-carboxylate (23). An 80%

NaH dispersion in oil (4.32 g; corresponding to 3.45 g, 144 mmol) was washed with pentane. Then toluene

(140 ml) and diethyl carbonate (34 g, 288 mmol) were added, and the mixture was heated to 608. A soln. of 5-

(tert-butyl)-2,3-dihydro-1H-inden-1-one (22; 7.52 g, 40 mmol) in toluene (20 ml) was added during 2 h, and

1784

Helvetica Chimica Acta Vol. 87 (2004)

stirring at 608 was continued for 6 h. The mixture was poured on an excess of H₂O/AcOH 1 :1 and extracted with

petroleum ether (30 508) and the extract washed with sat. aq. NaHCO₃soln. and brine, dried (Na₂SO₄) and

evaporated (10.8g). Bulb-to-bulb distillation (oven temp. ! 2508/0.1 mbar) gave 23 (7.12 g, 62%, purity (NMR)

ca. 90%). Oil. IR: 2960, 2860, 1730 1700, 1598, 1360, 1320, 1250, 1205, 1150, 1080, 1010. ¹H-NMR: 7.70 (d, J 8,

1 H); 7.50 (s, 1 H); 7.45 (d, J 8, 1 H); 4.25 (q, J 7, 2 H); 3.71 (dd, J 8, 4, 1 H); 3.54 (dd, J 17, 4, 1 H); 3.35

(dd, J 17, 8, 1 H); 1.36 (s, 9 H); 1.32 (t, J 7, 3 H ) . ¹³C-NMR: Table 5. CI-MS: 260 (61, M ), 245 (25), 215 (24),

199 (60), 186 (77), 171 (100), 157 (16), 143 (11), 131 (34), 115 (31), 91 (18), 57 (76), 41 (31).

Table 5. ¹³C-NMR Chemical Shifts (d [ppm] relative to SiMe₄) of Compounds in Schemes 1 and 2 (CDCl₃

solutions)^a)

R

C(1) C(2) C(3) C(4) C(5) C(6) C(7) C(8) C(9) C(10) Me₃C Me₃C C(R)

22 H

155.5 123.2 158.8 125.0 123.3 134.7 206.5 36.5 25.9

35.4 31.2

23 COOEt 154.0 123.1 159.8125.6 124.3 132.9 199.2 53.6 30.4

35.6 31.2 169.4, 61.4, 14.2

24 COOEt 152.9 123.0 159.6 125.6 124.5 132.3 203.0 56.2 40.2 21.1 35.6 31.2 172.2, 61.4, 14.0

25 COOEt 141.5 121.2 149.8123.7 123.813.87 35.843.8 36.3 34.5 31.6 175.4, 60.6, 14.3

26 COOEt 141.2 121.5 149.7 123.6 124.1 138.3 43.6 49.6 44.1 25.2 34.5 31.6 177.7, 60.6, 14.2

27 CH₂OH 142.5 121.5 149.5 123.4 124.1 139.7 35.3 41.7

28 CH₂OH 142.3 121.7 149.4 123.3 124.2 139.4 42.9 45.0 42.4 24.1 34.5 31.6

35.9

34.5 31.6

66.1

70.7

8 CHO

9 CHO

141.0 121.5 150.1 124.0 124.1 138.1 32.5 50.9 33.1

140.8121.6 150.1 124.0 124.2 137.9 40.6 54.4 41.1 21.1 34.6 31.5 204.1

30.7 106.4 35.0 31.3 16.81, 50.9

137.0 119.9 150.8132.5 126.3 152.6 25.3 36.6 207.4 34.8 31.3

34.6 31.5 203.0

31 COOMe 149.7 122.3 154.8124.4 121.3 137.4 163.4 31.5

32 H

35 COOMe 146.9 118.0 150.1 125.2 128.8 139.7 30.1 31.6 163.8 106.7 34.7 31.4 168.0, 51.0

36 COOMe 143.7 121.5 149.9 123.4 122.9 142.7 41.0 32.7 31.3 39.7 34.6 31.6 173.3, 51.5

37 COOMe 145.5 120.2 149.5 123.9 124.1 140.9 30.7 32.6 41.5 39.9 34.6 31.6 173.3, 51.5

38 CH₂OH 143.7 121.4 149.6 123.2 123.0 144.0 41.1 32.5 31.6 38.0 34.5 31.6

39 CH₂OH 146.8120.4 149.3 123.5 124.0 140.9 30.9 32.5 41.6 3.80 34.6 31.6

61.7

61.6

10 CHO

11 CHO

143.6 121.5 150.1 123.5 122.9 142.5 38.6 35.7 31.5 49.5 34.6 31.6 202.0

145.3 120.2 149.7 124.0 124.2 140.830.9 32.7 39.1 49.5 34.6 31.6 202.0

^a) Arbitrary numbering.

(Æ)-Ethyl 5-(tert-Butyl)-2,3-dihydro-(2-methyl-1-oxo-1H-indene-2-carboxylate (24). To a stirred soln. of 23

(5.36 g, 20.6 mmol) in THF (75 ml) at r.t. were added K₂CO₃(5.7 g, 41.2 mmol) and MeI (1.92 ml, 4.37 g,

31 mmol), and the mixture was heated to reflux (bath temp. ! 658) during 3 h. The cooled mixture was diluted

with Et₂O and H₂O and the org. phase washed with brine (2Â), dried (Na₂SO₄), and evaporated: yellow liquid

(5.5 g). Bulb-to-bulb distillation (oven temp. ! 1108/0.16 mbar) gave 24 (5.19 g, 94%; purity 94%). Colorless

oil. IR: 2957, 1747, 1699, 1601, 1428, 1374, 1322, 1267, 1189, 1172, 1096, 1080, 965, 921, 849. ¹H-NMR: 7.72 (d, J

8, 1 H); 7.47 (s, 1 H); 7.46 (d, J 8, 1 H); 4.15 (q, J 7, 2 H); 3.69 (d, J 18, 1 H); 2.97 (d, J 18, 1 H); 1.51

(s, 3 H); 1.37 (s, 9 H); 1.20 (t, J 7, 3 H ) . ¹³C-NMR: Table 5. MS: 274 (58, M ), 259 (24), 246 (10), 229 (10), 217

(17), 201 (100), 185 (75), 171 (13), 157 (10), 143 (18), 128 (29), 115 (36), 91 (11), 77 (5), 57 (12), 41 (6).

(Æ)-Ethyl 5-(tert-Butyl)-2,3-dihydro-1H-indene-2-carboxylate (25). A soln. of 23 (4.65 g, 16.2 mmol; purity

90%) in AcOEt (50 ml) was shaken at r.t. in presence of 5% Pd/C (0.48g) under H ₂(1 atm) during 12 h. The

¾

catalyst was filtered off through Celite and the soln. evaporated (4.25 g). CC (SiO₂(106 g), pentane/Et₂O 9 :1),

followed by bulb-to-bulb distillation (oven temp. ! 1608/0.5 mbar), gave pure 25 (3.47 g, 86%). Oil. IR: 2950,

1729, 1445, 1363, 1256, 1213, 1200, 1159, 1030, 817. ¹H-NMR: 7.24 (s, 1 H); 7.20 (d, J 8, 1 H); 7.13 (d, J 8,

Helvetica Chimica Acta Vol. 87 (2004)

1785

1 H); 4.18( q, J 7, 2 H); 3.40 3.10 (m, 5 H); 1.31 (s, 9 H); 1.28( t, J 7, 3 H ) . ¹³C-NMR: Table 5. MS: 246 (25,

M ), 231 (100), 201 (3), 172 (25), 157 (60), 129 (20), 115 (20), 91 (4), 79 (4), 57 (20), 41 (7).

(Æ)-Ethyl 5-(tert-Butyl)-2,3-dihydro-2-methyl-1H-indene-2-carboxylate (26). To a soln. of 24 (5.19 g,

17.9 mmol, purity 94%) in AcOH (50 ml) was added 10% Pd/C (2.1 g), and the mixture was shaken under H₂

(1 atm) at r.t. during 17 days. The catalyst was filtered off and the soln. was evaporated. The residue was

evaporated from toluene (3Â), then treated with active charcoal in toluene soln., filtered, and evaporated:

colorless oil (5.33 g). Bulb-to-bulb distillation (oven temp. ! 1058/0.1 mbar) afforded 26 (4.44 g, 74%; purity

77%). Colorless oil. An anal. sample was obtained by CC (SiO₂, toluene/AcOEt 9 :1). IR: 2950, 2890, 2860,

1720, 1455, 1355, 1300, 1210, 1195, 1110, 1025, 820. ¹H-NMR: 7.21 (s, 1 H); 7.19 (d, J 8, 1 H); 7.10 (d, J 8,

1 H); 4.16 (q, J 7, 2 H); 3.49 (d, J 15, 1 H); 3.44 (d, J 15, 1 H); 2.79 (d, J 15, 1 H); 2.77 (d, J 15, 1 H);

1.35 (s, 3 H); 1.30 (s, 9 H); 1.26 (t, J 7, 3 H ) . ¹³C-NMR: Table 5. MS: 260 (18, M ), 245 (51), 217 (3), 186 (65),

171 (100), 157 (10), 143 (13), 129 (23), 115 (16), 91 (8), 77 (4), 57 (27), 41 (11).

(Æ)-5-(tert-Butyl)-2,3-dihydro-1H-indene-2-methanol (27). From 25 (2.2 g, 8.9 mmol), according to the

G.P. A. Bulb-to-bulb distillation (oven temp. ! 2008/0.4 mbar) gave 27 (1.86 g, 100%). Colorless oil. IR: 3306,

2949, 2864, 1493, 1361, 1264, 1032, 816, 716. ¹H-NMR: 7.24 (s, 1 H); 7.18( d, J 8, 1 H); 7.13 (d, J 8, 1 H); 3.66

(d, J 6, 2 H); 3.04 (m, 2 H); 2.71 (m, 3 H); 1.62 (br. s, OH); 1.31 (s, 9 H). ¹³C-NMR: Table 5. MS: 204 (21,

M ), 189 (100), 171 (15), 143 (14), 129 (16), 115 (13), 91 (6), 77 (3), 57 (7), 41 (5).

(Æ)-5-(tert-Butyl)-2,3-dihydro-2-methyl-1H-indene-2-methanol (28). From 26 (2.28g, 8.2 mmol; purity

94%), according to the G.P. A. Bulb-to-bulb distillation (oven temp. ! 1808/0.1 mbar) afforded 28 (1.85 g, 98%;

1

purity 94%). Colorless oil. IR: 3330, 2960, 2920, 2880, 1605, 1490, 1455, 1360, 1255, 1040, 820. H-NMR: 7.20

(s, 1 H); 7.17 (d, J 8, 1 H); 7.09 (d, J 8, 1 H); 3.51 (s, 2 H); 2.90 (d, J 16, 1 H); 2.86 (d, J 16, 1 H); 2.65

(d, J 16, 1 H); 2.62 (d, J 16, 1 H); 1.68( s, OH); 1.30 (s, 9 H); 1.18( s, 3 H). ¹³C-NMR: Table 5. MS: 218(25,

M ), 203 (100), 185 (25), 171 (4), 157 (9), 143 (15), 129 (12), 115 (8), 91 (5), 77 (3), 57 (17), 41 (8).

(Æ)-5-(tert-Butyl)-2,3-dihydro-1H-indene-2-carboxaldehyde (8). From 27 (2.04 g, 10 mmol), according to

the G.P. B. Bulb-to-bulb distillation (oven temp. ! 1608/0.4 mbar) gave 8 (1.47 g, 72%). Colorless oil. IR: 2950,

1

2713, 1720, 1494, 1436, 1391, 1361, 1265, 1200, 1056, 900, 818, 716. H-NMR: 9.77 (d, J 2, 1 H); 7.27 (s, 1 H);

7.22 (d, J 8, 1 H); 7.16 (d, J 8, 1 H); 3.35 3.10 (m, 5 H); 1.31 (s, 9 H). ¹³C-NMR: Table 5. MS: 202 (36, M ),

187 (100), 169 (10), 157 (10), 141 (11), 129 (18), 115 (19), 91 (6), 77 (3), 57 (6), 41 (6).

(Æ)-5-(tert-Butyl)-2,3-dihydro-2-methyl-1H-indene-2-carboxaldehyde (9). From 28 (1.85 g, 8 mmol; purity

94%) according to the G.P. B. Bulb-to-bulb distillation (oven temp. ! 1108/0.1 mbar) afforded 9 (1.56 g, 81%;

purity 90%). Colorless oil. IR: 2960, 2890, 2860, 2690, 1720, 1600, 1490, 1455, 1430, 1360, 1265, 885, 820.

¹H-NMR: 9.65 (s, 1 H); 7.23 (s, 1 H); 7.21 (d, J 8, 1 H); 7.13 (d, J 8, 1 H); 3.36 (d, J 16, 1 H); 3.32 (d, J 16,

1 H); 2.76 (d, J 16, 1 H); 2.73 (d, J 16, 1 H); 1.31 (s, 9 H); 1.30 (s, 3 H). ¹³C-NMR: Table 5. MS: 216 (36,

M ), 201 (100), 183 (3), 171 (4), 157 (16), 141 (9), 129 (18), 115 (2), 105 (2), 91 (8), 71 (6), 57 (24), 41 (8).

5. Compounds 10 and 11. Methyl (E)-[5-(tert-Butyl)-2,3-dihydro-1H-inden-1-ylidene]acetate (31) and

Isomers. To a stirred soln. of 22 (10.2 g, 53 mmol) in petroleum ether (30 508) (125 ml) at r.t. were added

methyl (dimethoxyphosphinyl)acetate (11.6 ml, 80 mmol) and dropwise 5.4m NaOMe in MeOH (13.4 ml,

72 mmol) during 5 min. After 24 h at r.t., the mixture was heated under reflux (358) during 17 h. The cooled

mixture was diluted with Et₂O and H₂O and the org. layer washed with brine (2Â), dried (Na₂SO₄), and

evaporated: solidifying oil (12.6 g). GC: 2% of 22, 14% of deconjugated isomer 29, 5% of (Z)-isomer 30, and

78% of (E)-isomer 31 (yield: 93%). Crystallization from Et₂O at À 308 gave enriched 31 (5.56 g, 42%; purity

94%, the balance consisting of 5% of 29 and 1% of 30). M.p. 116.5 117.58. IR (CHCl₃): 2960, 1690, 1630, 1605,

1435, 1355, 1315, 1290, 1175, 1090, 830. ¹H-NMR: 7.53 (d, J 8, 1 H); 7.37 (d, J 2, 1 H); 7.30 (dd, J 8 , 2, 1 H);

6.27 (t, J 2.5, 1 H); 3.76 (s, 3 H); 3.29 (m, 2 H); 3.06 (m, 2 H); 1.33 (s, 9 H). ¹³C-NMR: Table 5. MS: 244 (43,

M ), 229 (100), 213 (15), 197 (14), 188(9), 169 (15), 155 (27), 141 (25), 128(34), 115 (23), 85 (10), 57 (13), 41

(7).

Methyl (E)-[6-(tert-Butyl)-2,3-dihydro-1H-inden-1-ylidene]acetate (35) and Isomers. As described for 31,

with 32 (1.4 g, 7.3 mmol; purity 98%): crude material (1.5 g). GC: 19% of deconjugated isomer 33, 4% of (Z)-

isomer 34, and 53% of (E)-isomer 35. Bulb-to-bulb distillation (oven temp. 110 1228/0.25 mbar) afforded a

product mixture (1.20 g, 64%; purity 96%). GC: 26% of 33, 5% of 34, and 65% of 35. IR: 2980, 1740, 1710,

1680, 1450, 1355, 1200, 1170, 860, 835. ¹H-NMR (major isomer 35): 7.61 (d, J 2, 1 H); 7.42 (dd, J 8 , 2, 1 H);

7.28( d, J 8, 1 H); 6.33 (t, J 2, 1 H); 3.76 (s, 3 H); 3.31 (m, 2 H); 3.03 (m, 2 H); 1.34 (s, 9 H). ¹³C-NMR:

Table 5. MS (major isomer 35): 244 (47, M ), 229 (100), 213 (19), 197 (41), 188 (43), 169 (17), 155 (30), 141 (21),

129 (38), 115 (16), 85 (15), 57 (11).

(Æ)-Methyl 5-(tert-Butyl)-2,3-dihydro-1H-indene-1-acetate (36). A soln. of 31 (7.4 g, 30 mmol; purity 94%,

the balance consisting of 5% of 29 and 1% of 30) in AcOEt (150 ml) at r.t. was shaken under H₂(1 atm) in the

1786

Helvetica Chimica Acta Vol. 87 (2004)

¾

presence of 5% Pd/C (0.68g, 0.3 mmol) during 1 h. The catalyst was filtered off through Celite and the soln.

evaporated (7.39 g). Bulb-to-bulb distillation (oven temp. ! 1108/0.35 mbar) gave 36 (7.1 g, 93%; purity 99%).

Colorless oil. IR: 2970, 1735, 1435, 1360, 1270, 1190, 1170, 830. ¹H-NMR: 7.26 (d, J 2, 1 H); 7.20 (dd, J 8 , 2,

1 H); 7.00 (d, J 8, 1 H); 3.72 (s, 3 H); 3.55 (m, 1 H); 2.89 (m, 2 H); 2.79 (m, 1 H); 2.40 (m, 2 H); 1.73 (m, 1 H);

1.32 (s, 9 H). ¹³C-NMR: Table 5. MS: 246 (28, M ), 231 (58), 189 (52), 173 (74), 157 (65), 143 (32), 129 (100),

115 (54), 91 (15), 77 (7), 57 (31), 41 (7).

(Æ)-Methyl 6-(tert-Butyl)-2,3-dihydro-1H-indene-1-acetate (37). As described for 36, with 35 (1.13 g,

4.4 mmol; purity 65%, the balance consisting of 26% of 33 and 5% of 34): crude material (1.11 g) which was

bulb-to-bulb distilled (oven temp. 105 1108/0.12 mbar): 37 (1.10 g, 99%; purity 98%). Colorless oil. IR: 2990,

2940, 1740, 1500, 1440, 1370, 1270, 1180, 830. ¹H-NMR: 7.22 (dd, J 8, 2, 1 H); 7.20 (d, J 2, 1 H); 7.15 (d, J 8,

1 H); 3.72 (s, 3 H); 3.58( m, 1 H); 2.85 (m, 2 H); 2.80 (m, 1 H); 2.40 (m, 2 H); 1.75 (m, 1 H); 1.31 (s, 9 H).

¹³C-NMR: Table 5. MS: 246 (27, M ), 231 (59), 189 (41), 173 (31), 157 (100), 129 (70), 115 (29), 91 (8), 57 (15).

(Æ)-5-(tert-Butyl)-2,3-dihydro-1H-indene-1-ethanol (38). Treatment of 36 (6.3 g, 25 mmol) according to

the G.P. A gave crude 38. Bulb-to-bulb distillation (oven temp. ! 1158/0.3 mbar) afforded 38 (5.5 g, 99%).

Colorless oil. IR: 3320, 2960, 2860, 1490, 1360, 1265, 1060, 1025, 825. ¹H-NMR: 7.27 (d, J 2, 1 H); 7.21 (dd, J

8, 2, 1 H); 7.14 (d, J 8, 1 H); 3.81 (m, 2 H); 3.20 (m, 1 H); 2.87 (m, 2 H); 2.32 (m, 1 H); 2.15 (m, 1 H); 1.70

(m, 2 H); 1.58(br., OH); 1.32 ( s, 9 H). ¹³C-NMR: Table 5. MS: 218(56, M ), 203 (99), 200 (3), 185 (25), 173

(100), 161 (34), 143 (30), 128(22), 115 (20), 91 (8), 77 (4), 57 (25), 41 (6), 31 (9).

(Æ)-6-(tert-Butyl)-2,3-dihydro-1H-indene-1-ethanol (39). Treatment of 37 (0.99 g, 3.9 mmol; purity 98%)

according to the G.P. A gave crude 39 (0.81 g), which was bulb-to-bulb distilled (oven temp. 100 1158/

0.07 mbar): 39 (0.78 g, 87%; purity 97%). Colorless oil. IR: 3320, 2960, 2860, 1480, 1355, 1255, 1150, 820.

¹H-NMR: 7.24 (d, J 2, 1 H); 7.21 (dd, J 8, 2, 1 H); 7.15 (d, J 8, 1 H); 3.81 (m, 2 H); 3.22 (m, 1 H); 2.85

(m, 2 H); 2.31 (m, 1 H); 2.16 (m, 1 H); 1.70 (m, 2 H); 1.42 (br., OH); 1.32 (s, 9 H). ¹³C-NMR: Table 5. MS: 218

(60, M ), 203 (100), 185 (29), 173 (84), 161 (84), 157 (52), 143 (84), 129 (60), 117 (70), 91 (22), 77 (10), 57 (50).

(Æ)-5-(tert-Butyl)-2,3-dihydro-1H-indene-1-acetaldehyde (10). Treatment of 38 (3.8 5 g, 17.6 mmol)

according to the G.P. B gave crude 10 (2.89 g). Bulb-to-bulb distillation (oven temp. ! 1508/0.4 mbar) gave 4

(2.34 g, 62%; purity 98%). Colorless oil. IR: 2960, 2900, 2860, 2710, 1490, 1360, 1260, 830. ¹H-NMR: 9.88 (t, J

2, 1 H); 7.28( d, J 2, 1 H); 7.22 (dd, J 8, 2, 1 H); 7.08 (d, J 8, 1 H); 3.62 (m, 1 H); 2.90 (m, 3 H); 2.62

(m, 1 H); 2.43 (m, 1 H); 1.70 (m, 1 H); 1.32 (s, 9 H). ¹³C-NMR: Table 5 . MS: 216 (44, M ), 201 (100), 173 (79),

157 (98), 143 (37), 129 (72), 115 (83), 102 (5), 91 (23), 77 (10), 57 (33), 41 (8), 29 (8).

(Æ)-6-(tert-Butyl)-2,3-dihydro-1H-indene-1-acetaldehyde (11). Treatment of 39 (8.53 g, 32.1 mmol; purity

82%) according to the G.P. B gave a crude product, which was bulb-to-bulb distilled (oven temp. ! 1208/

0.15 mbar): 11 (6.01 g, 72%; purity 84%). Colorless oil. CC (silica gel, cyclohexane/Et₂O 4 :1) gave pure 11. IR:

2980, 2880, 2620, 1730, 1495, 1370, 1270, 1125, 830. ¹H-NMR: 9.88 (br. s, 1 H); 7.22 (dd, J 8 , 2, 1 H); 7.19

(d, J 2, 1 H); 7.17 (d, J 8, 1 H); 3.64 (m, 1 H); 2.89 (m, 3 H); 2.62 (m, 1 H); 2.42 (m, 1 H); 1.71 (m, 1 H); 1.29

(s, 9 H). ¹³C-NMR: Table 5. MS: 216 (16, M ), 201 (42), 172 (45), 159 (27), 157 (100), 141 (21), 131 (25), 129

(61), 117 (33), 115 (51), 91 (16), 77 (6), 57 (11).

6. Compounds 12 and 13. 6-(tert-Butyl)-1-chloro-3,4-dihydronaphthalene-2-carboxaldehyde (41). To stirred

dimethylformamide (DMF; 2 ml, 1.89 g, 26 mmol) was added POCl₃(2.31 g, 15 mmol), while maintaining the

temp. below 308 with an ice-water bath. After 30 min at r.t., ketone 40 (1.01 g, 5 mmol) was added, and the

mixture was heated to 808 during 4 h. The mixture was cooled to r.t., 30% aq. NaOH soln. (8.1 g) was added

slowly (exothermic), and the mixture was heated to 708 during 0.5 h. H₂O (2.5 ml) was added, the mixture

extracted with CH₂Cl₂, and the extract dried (Na₂SO₄) and evaporated. Bulb-to-bulb distillation (oven temp. !

1908/0.2 mbar) gave 41 (1.11 g, 77%; purity 86%). Colorless oil. IR: 2954, 2849, 1653, 1584, 1547, 1362, 1347,

1267, 1161, 969, 888, 870, 827, 820, 795, 713, 682. ¹H-NMR: 10.37 (s, 1 H); 7.78( d, J 8, 1 H); 7.35 (dd, J 8 , 2,

1 H); 7.23 (d, J 2, 1 H); 2.83 (m, 2 H); 2.64 (m, 2 H); 1.33 (s, 9 H). ¹³C-NMR: Table 6. MS: 248(44, M ), 235

(32), 233 (100), 205 (14), 193 (22), 191 (50), 152 (16), 141 (19), 128(26), 115 (10), 57 (15), 41 (16), 29 (16).

7-(tert-Butyl)-1-chloro-3,4-dihydronaphthalene-2-carboxaldehyde (43). As described for 41, with 7-(tert-

butyl)-3,4-dihydronaphthalen-1(2H)-one (42; 1.01 g, 5 mmol): crude 43 (1.57 g). Bulb-to-bulb distillation (oven

temp. ! 1908/0.2 mbar) afforded 43 (1.08g, 87%; purity 97%). Yellow oil, which solidified on standing.

Crystallization from heptane at À 308 gave yellowish crystals. M.p. 67 688. IR: 2957, 2855, 1659, 1582, 1555,

1

1351, 1250, 1130, 1228, 967, 895, 835, 818, 778, 739, 715, 655. H-NMR: 10.40 (s, 1 H); 7.90 (d, J 2, 1 H); 7.41

(dd, J 8, 2, 1 H); 7.15 (d, J 8, 1 H); 2.80 (m, 2 H); 2.63 (m, 2 H); 1.35 (s, 9 H). ¹³C-NMR: Table 6. MS: 248

(36, M ), 235 (33), 233 (100), 205 (16), 193 (20), 191 (35), 165 (8), 152 (14), 141 (14), 128 (22), 115 (9), 102 (8),

57 (10), 41 (13), 29 (10).

Helvetica Chimica Acta Vol. 87 (2004)

1787

Table 6. ¹³C-NMR Chemical Shifts (d [ppm] relative to SiMe₄) of Compounds in Schemes 3 and 4 (CDCl₃

solutions)^a)

R

C(1) C(2) C(3) C(4) C(5) C(6) C(7) C(8) C(9) C(10) C(11) Me₃C Me₃C C(R)

40

42

H

144.3 125.4 157.1 124.0 127.1 130.3 198.2 39.2 23.4 30.1

132.2 123.6 149.7 130.8 128.0 141.7 29.3 23.4 39.3 198.7

35.1

34.6

35.0

34.7

34.3

31.1

31.2

31.1 190.6

31.3 190.7

31.4 204.0

31.4 203.9

41 CHO 138.7 124.9 155.3 124.1 126.1 131.3 146.1 129.4 21.7 27.4

43 CHO 132.0 123.3 150.2 128.5 127.5 136.2 26.5 21.8 131.5 146.5

12 CHO 135.4 125.7 149.0 123.2 129.0 131.3 28.1 47.1 23.1 28.4

13 CHO 133.7 126.0 148.9 123.3 128.6 133.0 27.7 23.1 47.0 28.8

47

48

H

141.4 126.7 155.8123.7 12.78 136.1 205.5 40.9 21.0 25.3

138.4 125.4 149.6 129.3 129.6 138.6 32.1 25.3 21.1 41.0 206.3 34.5

33.0 34.9

31.1

31.3

50 CHO 140.7 126.0 154.2 123.812.38 135.0 147.8136.0 22.6 33.9

32.4 34.9

31.2 190.3

51 CHO 136.3 125.3 149.7 127.8 128.8 138.0 31.4 33.7 22.6 137.3 148.0 34.6

31.3 190.3

31.4 203.8

14 CHO 142.4 126.3 149.6 123.0 129.2 136.4 34.7 50.9 31.7 26.6

15 CHO 139.8126.6 149.2 123.4 12.98 139.1 35.3 26.6 31.7 51.0

36.3 34.3

35.6 34.3

31.4 203.7

^a) Arbitrary numbering.

(Æ)-6-(tert-Butyl)-1,2,3,4-tetrahydronaphthalene-2-carboxaldehyde (12). In a Parr hydrogenation flask

were placed 41 (1.12 g, 4.5 mmol), 5% Pd/C (0.1 g), K₂CO₃(0.66 g, 4.8mmol), H ₂O (1.55 ml), and MeOH

(3 ml). This mixture was shaken under H₂(50 psi) for 6 h at r.t. The mixture was diluted with CH₂Cl₂and

¾

filtered through Celite . The aq. layer was separated, and the org. layer evaporated: crude product (1.03 g). GC:

45% of alcohol 44, 22% of aldehyde 12, and 26% of 6-(tert-butyl)-1,2,3,4-tetrahydro-2-methylnaphthalene.

This crude mixture was added at r.t. to a suspension of PCC (1.51 g, 7 mmol) and NaOAc (0.4 g, 5.5 mmol)

¾

in CH₂Cl₂(15 ml). The mixture was stirred at r.t. for 3 h, diluted with Et₂O, filtered through Celite and then

¾

through a short column of Florisil , and evaporated (0.76 g). CC (SiO₂(50 g), cyclohexane/Et₂O 9 :1) gave a

fraction, which was bulb-to-bulb distilled (oven temp. ! 1408/0.2 mbar) 12 (0.41 g, 41% from 41; purity 97%).

Colorless oil. IR: 2954, 2707, 1720, 1610, 1503, 1435, 1361, 1271, 813, 717. ¹H-NMR: 9.78( d, J 1, 1 H); 7.18

(dd, J 8, 2, 1 H); 7.11 (d, J 2, 1 H); 7.08( d, J 8, 1 H); 2.97 (m, 2 H); 2.87 (m, 2 H); 2.68( m, 1 H); 2.21

(m, 1 H); 1.78( m, 1 H); 1.30 (s, 9 H). ¹³C-NMR: Table 6. MS: 216 (36, M ), 201 (100), 159 (14), 141 (10), 129

(20), 128(18), 115 (14), 91 (10), 57 (14), 41 (8), 29 (8).

(Æ)-7-(tert-Butyl)-1,2,3,4-tetrahydronaphthalene-2-carboxaldehyde (13). To a soln. of 43 (0.86 g, 4 mmol)

in MeOH (0.42 g) and H₂O (0.55 g) were added K₂CO₃(0.55 g, 4 mmol) and 5% Pd/C (50 mg), and the mixture

was shaken at r.t. in a Parr apparatus under H₂O (50 psi) during 22 h. The mixture was diluted with CH₂Cl₂and

¾

filtered through Celite ; the aq. layer was separated and the org. layer evaporated: mixture (0.8g). GC: 78% of

aldehyde 13 and 10% of alcohol 45. CC (SiO₂(30 g), cyclohexane/Et₂O 4 :1) gave a first fraction of 13 (0.49 g,

57%; purity > 99%), which solidified on standing. Crystallization from pentane at À 308 afforded colorless

crystals. M.p. 36 378. IR: 2919, 2864, 2811, 2713, 1721, 1608, 1569, 1503, 1433, 1360, 1268, 924, 912, 819, 788, 723.

¹H-NMR: 9.78( d, J 1, 1 H); 7.17 (dd, J 8, 2, 1 H); 7.15 (br. s, 1 H); 7.03 (d, J 8, 1 H); 2.97 (m, 2 H); 2.83

(m, 2 H); 2.67 (m, 1 H); 2.20 (m, 1 H); 1.77 (m, 1 H); 1.30 (s, 9 H). ¹³C-NMR: Table 6. MS: 216 (36, M ), 201

(100), 159 (10), 155 (12), 141 (12), 129 (26), 128(20), 115 (14), 57 (20), 41 (11), 29 (14).

7. Compounds 14 and 15. 2-(tert-Butyl)-6,7,8,9-tetrahydro-5H-benzocyclohepten-5-one (47). To a stirred

soln. of acid 46 (1.13 g, 4.8mmol) in 1,2-dichloroethane (20 ml) at r.t. was added trifluoromethanesulfonic acid

(Fluka purum; 1.5 ml, 2.58g, 16.5 mmol), and the mixture was heated to reflux (84 8, bath temp. ! 908) during

12 h. The cooled mixture was diluted with Et₂O, washed with 4% aq. NaOH soln. and brine, dried (Na₂SO₄), and

evaporated: brown oil (0.77 g). Bulb-to-bulb distillation (oven temp. ! 1808/0.3 mbar) gave a colorless oil

(0.68g, 61%). GC: 90% of 47 (eluted second on GC) and 4% of 48 (eluted first on GC). 47: IR: 2960, 2860,

1788

Helvetica Chimica Acta Vol. 87 (2004)

1665, 1595, 1450, 1400, 1360, 1290, 1260, 1100, 965, 830. ¹H-NMR: 7.70 (d, J 8, 1 H); 7.32 (dd, J 8 , 2, 1 H); 7.19

(d, J 2, 1 H); 2.94 (m, 2 H); 2.72 (m, 2 H); 1.92 1.78( m, 4 H); 1.32 (s, 9 H). ¹³C-NMR: Table 6. MS: 216 (26,

M ), 201 (100), 187 (9), 173 (18), 145 (8), 131 (22), 129 (10), 91 (12), 77 (5), 41 (7).

3-(tert-Butyl)-6,7,8,9-tetrahydro-5H-benzocyclohepten-5-one (48). A stirred mixture of acid 46 (13.7 g,

57 mmol) and thionyl chloride (Fluka; 9.87 g, 6.02 ml, 83 mmol) was heated to 608 during 15 h. Vacuum (from

the water-jet) was applied during 1 h, then the residue was dissolved in CH₂Cl₂(100 ml) and cooled to 0 48 (ice-

water bath). AlCl₃(9.1 g, 68mmol, 1.2 equiv.) was added portionwise during 10 min. ( ! dark mixture). After

1 h, the mixture was allowed to reach r.t. during 1 h and then diluted with Et₂O and 5% aq. HCl soln. The org.

layer was washed with brine (2Â), sat. aq. NaHCO₃soln., and brine, dried (Na₂SO₄), and evaporated: brownish

oil (10.6 g). Bulb-to-bulb distillation (oven temp. ! 1508/0.6 mbar) gave a mixture of 47 (eluted second on GC)

and 48 (eluted first on GC) as a colorless oil (9.72 g, 75%; purity 94%; 47/48 24 :70). After three crystallizations

from pentane at À 308, the major product 48 was obtained (1.68g). Colorless crystals. M.p. 40 8 ([22a]: m.p. 408

(benzene)). IR: 2940, 2860, 1655, 1590, 1450, 1360, 1240, 1120, 1100, 965, 835. ¹H-NMR: 7.76 (d, J 2, 1 H); 7.44

(dd, J 8, 2, 1 H); 7.13 (d, J 8, 1 H); 2.90 (m, 2 H); 2.73 (m, 2 H); 1.90 1.77 (m, 4 H); 1.32 (s, 9 H).

¹³C-NMR: Table 6. MS: 216 (18, M ), 201 (100), 187 (2), 173 (7), 157 (3), 145 (3), 131 (18), 129 (10), 115 (11),

91 (10), 77 (5), 41 (7).

3-(tert-Butyl)-9-chloro-6,7-dihydro-5H-benzocycloheptene-8-carboxaldehyde (50). To DMF (1.5 ml) at

0

48 (ice-water bath) was added POCl₃(1 ml, 1.64 g, 11 mmol), and the mixture was stirred at 0 48 during

30 min. A soln. of 47 (0.36 g, 1.5 mmol; purity 94%) in DMF (2 ml) was added dropwise, and the mixture was

heated to 70 808 (bath temp. ! 838) during 1 h ( ! red). The mixture was diluted with Et₂O, and 2.5n (10%)

aq. NaOH (10 ml) was added carefully (exothermicity). The org. layer was washed with H₂O, sat. aq. NaHCO₃

soln. and brine, dried (Na₂SO₄), and evaporated: yellow oil (0.42 g), which solidified on standing. Bulb-to-bulb

distillation (oven temp. ! 2008/0.2 mbar) gave solid 50 (0.40 g). Crystallization from heptane (2 ml) at À 308

afforded colorless crystals (0.30 g, 76%). M.p. 104 1058. IR (CHCl₃): 3000, 2950, 2850, 1650, 1590, 1570, 1280,

1200, 1150, 1035, 925, 720, 660. ¹H-NMR: 10.38( s, 1 H); 7.60 (d, J 8, 1 H); 7.40 (dd, J 8, 2, 1 H); 7.25 (d, J 2,

1 H); 2.61 (t, J 7, 2 H); 2.24 (m, 2 H); 2.15 (m, 2 H); 1.35 (s, 9 H). ¹³C-NMR: Table 6. MS: 262 (53, M ), 247

(100), 234 (17), 227 (40), 219 (25), 211 (14), 203 (13), 197 (14), 191 (13), 183 (16), 171 (11), 165 (25), 153 (30),

141 (40), 128(33), 115 (35), 102 (9), 91 (8), 77 (13), 57 (70), 41 (26), 29 (31).

2-(tert-Butyl)-9-chloro-6,7-dihydro-5H-benzocycloheptene-8-carboxaldehyde (51). To DMF (3 ml) at 0 48

(ice-water bath) was added dropwise POCl₃(3.44 g, 2.09 ml, 22.5 mmol), and the mixture was stirred at r.t.

during 30 min. Ketone 48 (1.63 g, 7.5 mmol) was added and the mixture heated to 70 808 (bath temp.). After ca.

15 min, an exothermic reaction took place (temp. ! 958) and the mixture became deep red. After 15 min, the

temp. had decreased to 70 808 and the mixture solidified. It was cooled to 408 and covered with Et₂O (20 ml).

Then 2.5n (10%) aq. NaOH (10 ml) was added dropwise and cautiously (exothermicity). The org. layer was

washed with sat. aq. NaHCO₃soln. and brine, dried (Na₂SO₄) and evaporated: red syrup (1.84 g). Bulb-to-bulb

distillation (oven temp. ! 1808/0.3 mbar) gave solid 51 (1.78 g, 88%; purity 97%). Crystallization from heptane

(5 ml) at À 308 gave colorless crystals (1.04 g). M.p. 67 698. IR (CHCl₃): 2950, 2850, 1655, 1570, 1280, 1240,

1155, 1020, 950, 890, 835, 640. ¹H-NMR: 10.39 (s, 1 H); 7.67 (d, J 2, 1 H); 7.40 (dd, J 8, 2, 1 H); 7.18 (d, J 8,

1 H); 2.58( t, J 7, 2 H); 2.23 (m, 2 H); 2.12 (m, 2 H); 1.35 (s, 9 H). ¹³C-NMR: Table 6. MS: 262 (33, M ), 247

(100), 234 (12), 227 (13), 219 (10), 211 (6), 205 (5), 197 (8), 191 (5), 183 (9), 171 (5), 165 (16), 153 (19), 141

(28), 128 (24), 115 (24), 102 (7), 91 (5), 77 (11), 57 (40), 41 (20), 29 (18).

(Æ)-2-(tert-Butyl)-6,7,8,9-tetrahydro-5H-benzocycloheptene-6-carboxaldehyde (14). To a soln. of 50 (0.25 g,

0.9 mmol) in boiling MeOH (10 ml) were added H₂O (2 ml) (some product precipitated on cooling), K₂CO₃

(0.55 g, 4 mmol), and 5% Pd/C (200 mg, 0.1 mmol). The mixture was hydrogenated at r.t./60 psi (ca. 4 bar) in

¾

the Parr device during 4 h. The mixture was filtered through a short column of Celite , the latter rinsed with

MeOH and AcOEt, and the soln. diluted with Et₂O, washed with sat. aq. NaHCO₃soln. and brine, dried

(Na₂SO₄), and evaporated: colorless syrup (0.25 g) containing 20% of 14 and 74% of alcohol 52 (by GC). A

soln. of this crude in CH₂Cl₂(5 ml) was added to a stirred suspension of PCC (430 mg, 2 mmol) in CH₂Cl₂

¾

(20 ml). The mixture was stirred at r.t. during 2 h, diluted with Et₂O, filtered through a short column of Florisil ,

the latter rinsed with Et₂O, and the soln. evaporated: colorless syrup (0.20 g). Bulb-to-bulb distillation (oven

temp. ! 1908/0.2 mbar) gave 14 (0.18g, 80%; purity 92%). Syrup, which solidified on standing. Crystallization

from pentane (1 ml) at À 308 afforded colorless crystals. M.p. 63 648. IR (CHCl₃): 2950, 2920, 2840, 2700, 1715,

1600, 1495, 1435, 1355, 1265, 1215, 1115, 885, 810. ¹H-NMR: 9.68( s, 1 H); 7.14 (dd, J 8, 2, 1 H); 7.12 (d, J 2,

1 H); 7.09 (d, J 8, 1 H); 3.11 (m, 1 H); 2.80 (m, 3 H); 2.30 (m, 2 H); 2.05 (m, 1 H); 1.68( m, 1 H); 1.50

(m, 1 H); 1.30 (s, 9 H). ¹³C-NMR: Table 6. MS: 230 (20, M ), 215 (100), 197 (6), 184 (12), 173 (6), 169 (9), 155

(7), 145 (13), 141 (9), 128(18), 115 (15), 91 (11), 77 (5), 57 (17), 41 (12), 29 (12).

Helvetica Chimica Acta Vol. 87 (2004)

1789

(Æ)-3-(tert-Butyl)-6,7,8,9-tetrahydro-5H-benzocycloheptene-6-carboxaldehyde (15). To a soln. of 51 (0.98g,

3.6 mmol) in MeOH (10 ml) were added H₂O (2 ml), K₂CO₃(0.55 g, 4 mmol), and 10% Pd/C (100 mg,

0.1 mmol), and the mixture was hydrogenated in the Parr device at r.t./50 psi during 15 h. The mixture was

¾

filtered through paper and Celite and the latter rinsed with Et₂O. The org. phase was washed with sat. aq.

NaHCO₃soln., dried (Na₂SO₄), and evaporated: pink oil (0.9 g) containing 95% of alcohol 53 (by GC). A soln.

of this crude in CH₂Cl₂(10 ml) was added dropwise to a stirred suspension of PCC (1.07 g, 5 mmol) in CH₂Cl₂

(20 ml) at r.t., and the mixture was stirred at r.t. during 2 h. The mixture was diluted with Et₂O (150 ml) and

¾

filtered through a short column of Florisil , the latter rinsed with Et₂O, and the soln. evaporated: yellowish oil

(0.75 g). Bulb-to-bulb distillation (oven temp. ! 1808/0.3 mbar) gave 15 (0.70 g, 82%; purity 97%). Colorless

oil. IR: 2960, 2920, 2840, 2700, 1720, 1600, 1495, 1435, 1355, 1265, 1200, 1120, 820. ¹H-NMR: 9.70 (s, 1 H); 7.18

(d, J 2, 1 H); 7.15 (dd, J 8, 2, 1 H); 7.03 (d, J 8, 1 H); 3.13 (m, 1 H); 2.83 (m, 1 H); 2.77 (m, 2 H); 2.31

(m, 2 H); 2.04 (m, 1 H); 1.67 (m, 1 H); 1.48( m, 1 H); 1.30 (s, 9 H). ¹³C-NMR: Table 6. MS: 230 (21, M ), 215

(100), 197 (6), 184 (15), 169 (10), 155 (8), 145 (14), 128 (17), 115 (14), 105 (5), 91 (10), 77 (5), 57 (15), 41 (14),

29 (15).

Table 7. ¹³C-NMR Chemical Shifts (d [ppm] relative to SiMe₄) of Compounds in Scheme 5 (CDCl₃solutions)^a)

R

C(1) C(2) C(3) C(4) C(5) C(6) C(7) C(8) C(9) C(10) Me₃C Me₃C C(R) MeCH₂O MeCH₂O

54

55

56

57

142.2 121.7 151.7 124.0 123.7 143.3

144.8120.9 150.0 125.6 124.4 140.3

143.7 120.8147.7 123.3 120.3 142.2 131.7 133.6 39.1

144.9 117.9 149.4 121.7 123.2 140.6831834.2 132.3

76.2 36.1 29.9

29.3 36.1 76.4

34.7 31.5

34.7 31.6

34.6 31.7

58 COOEt 141.8122.2 149.6 123.4 123.3 140.7

59 COOEt 144.3 121.0 149.5 123.1 124.1 137.2

60 COOEt 143.6 120.9 149.6 123.5 124.7 138.8

61 COOEt 139.9 121.6 149.1 123.6 123.5 141.3

62 CH₂OH 142.3^b) 122.2 148.8123.0 122.8142.4

63 CH₂OH 145.1 120.4 149.2 122.7 124.7 139.4

64 CH₂OH 144.1 123.4 149.1 120.9 123.7 138.4

65 CH₂OH 141.4 121.2 149.6 123.0 123.4 141.0

16 CHO

17 CHO

18 CHO

19 CHO

34.0 26.6 35.4 30.834.5 31.5 172.680.4

14.3

31.1 23.5 32.6 25.2 34.5 31.5 169.859.8

34.9 26.7 34.6 30.8 34.6 31.6 172.8 60.5

13.8

14.3

32.0 23.4 31.5 25.0 34.5 31.6 169.6 59.813.9

b

)

29.0 21.5 35.2 32.1 34.5 31.6

34.7 21.6 29.6 32.1 34.5 31.6

27.9 20.7 31.8 24.2 34.5 31.6

65.2

58.0

31.2 20.82.58 24.1 34.5 31.6

57.9

141.6^b) 122.3 149.9 123.6 123.3 140.0^b) 35.1 27.5 35.3 40.5 34.6 31.5 199.3

138.6 120.9 149.9 123.8 124.8 142.9 34.9 27.6 35.6 40.6 34.6 31.5 199.3

142.5 121.4 150.4 124.2 124.3 136.834.0 27.4 33.3 32.5 34.6 31.5 203.1

139.6 121.6 150.5 124.2 123.9 139.4 32.7 27.4 34.6 32.4 34.6 31.5 203.1

^a) Arbitrary numbering. ^b) Interchangeable.

8. Compounds 16 19. (Æ)-5-(tert-Butyl)-2,3-dihydro-1H-inden-1-ol (54). According to the G.P. A, 5-(tert-

butyl)-2,3-dihydro-1H-inden-1-one (22; 30.08g, 160 mmol) was converted to crude 54 (24.3 g, solid).

Crystallization from petroleum ether (50 708) at À 308 gave colorless crystals (22.5 g, 72%; purity 98%).

1

M.p. 88 908. IR (solid): 3185 (br.), 2952, 2840, 1440, 1363, 1215, 1083, 1060, 887, 827. H-NMR: 7.35 (d, J 8,

1 H); 7.29 (d, J 2, 1 H); 7.28( dd, J 8, 2, 1 H); 5.21 (m, 1 H); 3.05 (m, 1 H); 2.80 (m, 1 H); 2.48( m, 1 H); 1.94

(m, 1 H); 1.80 (br., OH); 1.32 (s, 9 H). ¹³C-NMR: Table 7. MS: 190 (23, M ), 175 (100), 157 (11), 133 (25), 131

(28), 115 (12), 105 (5), 91 (9), 77 (5), 57 (3), 41 (4).

(Æ)-6-(tert-Butyl)-2,3-dihydro-1H-inden-1-ol (55). According to the G.P. A, 6-(tert-butyl)-2,3-dihydro-1H-

inden-1-one (32; 8.83 g, 45.6 mmol; purity 97%) was converted to crude 55 (8.83 g, solid). Two crystallizations

1790

Helvetica Chimica Acta Vol. 87 (2004)

from Et₂O/pentane at À 308 afforded colorless crystals (7.81 g, 90%). M.p. 77.5 788. IR: 3560, 3400 (br.), 2950,

2850, 1480, 1355, 1255, 1075, 1035, 950, 820. ¹H-NMR: 7.45 (d, J 2, 1 H); 7.30 (dd, J 8, 2, 1 H); 7.17 (d, J 8,

1 H); 5.18( m, 1 H); 2.99 (m, 1 H); 2.75 (m, 1 H); 2.46 (m, 1 H); 2.08(br., OH); 1.91 ( m, 1 H); 1.32 (s, 9 H).

¹³C-NMR: Table 7. MS: 190 (20, M ), 175 (100), 173 (2), 172 (2), 157 (14), 131 (30), 129 (13), 115 (14), 91 (10),

77 (6), 57 (4), 41 (5).

6-(tert-Butyl)-1H-indene (56). To 54 (1.9 g, 10 mmol) was added KHSO₄(68mg, 0.5 mmol), and the

mixture was heated in the bulb-to-bulb oven to 1608 (oven temp.) at 100 mbar. After 0.5 h, no more bubbling

was observed; the vacuum was lowered to 25 mbar and the oven temp. raised to 1758. After 0.5 h, nearly all

product had distilled into the bulb just outside of the oven: 56 (1.65 g, 96%). IR: 3060, 2957, 2902, 1477, 1361,

1261, 868, 823, 749, 690. ¹H-NMR: 7.52 (d, J 2, 1 H); 7.33 (d, J 8, 1 H); 7.30 (dd, J 8 , 2, 1 H); 6.8 4 m( , 1 H);

6.48( m, 1 H); 3.37 (br. s, 2 H); 1.35 (s, 9 H). ¹³C-NMR: Table 7. MS: 172 (32, M ), 157 (100), 142 (12), 129 (26),

115 (19), 91 (2), 77 (2), 57 (3), 41 (2).

5-(tert-Butyl)-1H-indene (57). To 55 (0.97 g, 5 mmol) was added KHSO₄(35 mg, 0.25 mmol), and the

mixture was heated in the bulb-to-bulb oven to 90 1008 (oven temp.) at 0.2 mbar during 0.5 h. The product was

collected in the bulb just outside of the oven: 57 (0.7 g, 78%; purity 96% (4% of 56)). Colorless liquid. IR:

3062, 2957, 2902, 1476, 1391, 1361, 1262, 944, 879, 813, 701. ¹H-NMR: 7.45 (d, J 2, 1 H); 7.39 (d, J 8, 1 H); 7.23

(dd, J 8 , 2, 1 H); 6.8 6 m( , 1 H); 6.52 (m, 1 H); 3.33 (br. s, 2 H); 1.35 (s, 9 H). ¹³C-NMR: Table 7. MS: 172 (32,

M ), 157 (100), 142 (13), 129 (28), 115 (22), 91 (2), 77 (2), 57 (4), 41 (2).

Ethyl (1RS,1aRS,6aSR)-4-(tert-Butyl)-1,1a,6,6a-tetrahydrocyclopropa[a]indene-1-carboxylate (58; −exo×)

and Ethyl (1RS,1aSR,6aRS)-4-(tert-Butyl)-1,1a,6,6a-tetrahydrocyclopropa[a]indene-1-carboxylate (59; −endo×).

To a stirred soln. of 56 (1.65 g, 9.6 mmol) in cyclohexane (20 ml) at r.t. was added CuSO₄(25 mg, 0.15 mmol)

and 20 drops of a soln. of ethyl diazoacetate (Fluka purum; 6.0 g, 52 mmol) in cyclohexane (15 ml). The mixture

was then heated to 75 808 (bath temp. ! 908), and the rest of the reagent soln. was added dropwise during 1 h

( ! N₂evolution). At the end of the addition, no more 56 was detectable by GC. The mixture was diluted with

Et₂O, washed with brine (2Â), dried (Na₂SO₄), and evaporated: brown oil (6 g) containing 6% of diethyl

maleate, 13% of diethyl fumarate, 20% of −endo×-ester 59, and 55% of −exo×-ester 58 (by GC). Bulb-to-bulb

distillation (oven temp. ! 958/0.1 mbar) gave a first fraction (1.73 g) of volatiles; raising the oven temp. to 1808

gave a second fraction containing the desired products as a colorless oil (2.51 g, 97%; purity > 99%; 59/58

26 :74). CC (SiO₂(200 g), cyclohexane/Et₂O 19 :1) gave as a first fraction 58 (1.26 g, 48%). IR: 2958, 1718, 1382,

1303, 1264, 1154, 1038, 822. ¹H-NMR: 7.26 (d, J 8, 1 H); 7.19 (d, J 2, 1 H); 7.17 (dd, J 8, 2, 1 H); 4.13 (q, J

7, 2 H); 3.26 (dd, J 17, 7, 1 H); 3.02 (dd, J 17, 2, 1 H); 2.91 (m, 1 H); 2.42 (m, 1 H); 1.28( s, 9 H); 1.25 (t, J 7,

3 H); 1.22 (m, 1 H). ¹³C-NMR: Table 7. MS: 258(81, M ), 243 (99), 229 (61), 213 (24), 185 (25), 169 (45), 155

(35), 129 (44), 115 (18), 57 (100), 41 (8).

After a second (mixed) fraction, a third fraction was collected, which contained 59 (0.1 g, 4%; purity 98%).

Similar rechromatography of the second (mixed) fraction afforded an additional crop of 59 (0.31 g, 11%; purity

95%). IR: 2960, 1725, 1460, 1380, 1360, 1265, 1180, 1140, 1035, 825. ¹H-NMR: 7.22 (d, J 8, 1 H); 7.16 (d, J 2,

1 H); 7.14 (dd, J 8, 2, 1 H); 3.81 (m, 2 H); 3.32 (dd, J 17, 2, 1 H); 3.19 (dd, J 17, 7, 1 H); 2.89 (m, 1 H); 2.23

(m, 1 H); 1.99 (t, J 8, 1 H); 1.28 (s, 9 H); 0.86 (t, J 7, 3 H ) . ¹³C-NMR: Table 7. MS: 258(35, M ), 243 (43),

229 (26), 213 (12), 185 (12), 169 (22), 155 (17), 141 (16), 129 (25), 57 (100), 41 (9).

Ethyl (1RS,1aRS,6aSR)-3-(tert-Butyl)-1,1a,6,6a-tetrahydrocyclopropa[a]indene-1-carboxylate (60; −exo×)

and Ethyl (1RS,1aSR,6aRS)-3-(tert-Butyl)-1,1a,6,6a-tetrahydrocyclopropa[a]indene-1-carboxylate (61; −endo×).

As described for 58/59, with 57 (0.5 g, 2.9 mmol), cyclohexane (10 ml), CuSO₄(16 mg, 0.1 mmol), and 20 drops

of ethyl diazoacetate (Fluka purum; 2 g, 17 mmol) in cyclohexane (10 ml); addition of the rest of the reagent

soln. within 15 min. Workup (washing with H₂O and brine (2Â)) gave a brown oil (0.87 g) containing 8% of

diethyl maleate, 8% of diethyl fumarate, 20% of −endo×-ester 61, and 54% of −exo×-ester 60 (by GC). Bulb-to-

bulb distillation (oven temp. ! 1008/0.3 mbar) gave a first fraction (0.74 g) containing the volatiles. The oven

temp. was raised to 1808 to afford the desired products as a colorless oil (0.7 g, 90%; purity 95%; 61/60 24 :71).

CC (SiO₂(100 g), cyclohexane/Et₂O 9 :1) gave as a 1st fraction 60 (0.34 g, 41%; purity 94%). IR: 2958, 2905,

1

2867, 1718, 1397, 1380, 1291, 1224, 1172, 1155, 1038, 818. H-NMR: 7.38( d, J 2, 1 H); 7.16 (dd, J 8 , 2, 1 H);

7.08( d, J 8, 1 H); 4.14 (q, J 7, 2 H); 3.22 (dd, J 17, 6, 1 H); 3.00 (dd, J 17, 2, 1 H); 2.93 (m, 1 H); 2.42

(m, 1 H); 1.30 (s, 9 H); 1.26 (t, J 7, 3 H); 1.22 (m, 1 H). ¹³C-NMR: Table 7. MS: 258(37, M ), 243 (43), 229

(31), 213 (12), 185 (13), 169 (24), 155 (20), 129 (31), 115 (11), 85 (8), 57 (100), 41 (10).

After a second (mixed) fraction (0.23 g), a third fraction was collected, which contained 61 (0.03 g, 3%;

1

purity 94%). IR: 2955, 1727, 1381, 1362, 1264, 1134, 1035, 814. H-NMR: 7.33 (d, J 2, 1 H); 7.15 (dd, J 8 , 2,

1 H); 7 05 (d, J 8, 1 H); 3.80 (m, 2 H); 3.32 (dd, J 17, 2, 1 H); 3.15 (dd, J 17, 7, 1 H); 2.91 (m, 1 H); 2.23

Helvetica Chimica Acta Vol. 87 (2004)

1791

(m, 1 H); 2.00 (t, J 8, 1 H); 1.29 (s, 9 H); 0.86 (t, J 7, 3 H ) . ¹³C-NMR: Table 7. IR: 258(34, M ), 243 (38), 229

(24), 213 (12), 185 (11), 169 (22), 155 (20), 141 (16), 129 (30), 57 (100), 41 (10).

(1RS,1aRS,6aSR)-4-(tert-Butyl)-1,1a,6,6a-tetrahydrocyclopropa[a]indene-1-methanol (62). Following the

G.P. A, 58 (2.21 g, 8.6 mmol) was converted to crude 62 (1.94 g). Bulb-to-bulb distillation (oven temp. ! 1608/

0.2 mbar) afforded 62 (1.8g, 97%; purity > 99%). Colorless viscous syrup. IR: 3310 (br.), 2950, 2880, 2840, 1460,

1355, 1260, 1100, 1025, 825. ¹H-NMR: 7.21 (d, J 8, 1 H); 7.16 (d, J 2, 1 H); 7.14 (dd, J 8, 2, 1 H); 3.55 (d, J

7, 2 H); 3.18( dd, J 17, 7, 1 H); 2.95 (d, J 17, 1 H); 2.29 (m, 1 H); 1.76 (m, 1 H); 1.54 (br., OH); 1.28( s, 9 H);

0.79 (m, 1 H). ¹³C-NMR: Table 7. MS: 216 (45, M ), 201 (100), 185 (45), 170 (16), 155 (31), 141 (27), 129 (44),

115 (22), 91 (6), 57 (54), 41 (7).

(1RS,1aRS,6aSR)-3-(tert-Butyl)-1,1a,6,6a-tetrahydrocyclopropa[a]indene-1-methanol (63). Following the

G.P. A, 60 (3.04 g, 11.8mmol) was converted to crude 63 (3.2 g). Bulb-to-bulb distillation (oven temp. ! 1758/

0.3 mbar) afforded 63 (2.56 g, 98%; purity 98%). Colorless oil. IR: 3300 (br.), 2940, 2890, 2840, 1480, 1350, 1255,

1

1100, 1025, 815. H-NMR: 7.34 (d, J 2, 1 H); 7.12 (dd, J 8, 2, 1 H); 7.06 (d, J 8, 1 H); 3.57 (d, J 7, 2 H ) ;

3.14 (dd, J 17, 7, 1 H); 2.93 (d, J 17, 1 H); 2.32 (m, 1 H); 1.77 (m, 1 H); 1.53 (br., OH); 1.31 (s, 9 H); 0.80

(m, 1 H). ¹³C-NMR: Table 7. MS: 216 (52, M ), 201 (100), 185 (62), 170 (12), 155 (32), 141 (33), 129 (56), 115

(26), 91 (7), 57 (66), 41 (9).

(1RS,1aSR,6aRS)-4-(tert-Butyl)-1,1a,6,6a-tetrahydrocyclopropa[a]indene-1-methanol (64). Following the

G.P. A, 59 (0.44 g, 1.7 mmol) was converted to crude 64 (0.36 g). Bulb-to-bulb distillation (oven temp. ! 1758/

0.25 mbar) gave 64 (0.35 g, 99%). Colorless oil. IR: 3316 (br.), 2958, 2904, 1489, 1360, 1266, 1014, 825. ¹H-NMR:

7.20 (d, J 8, 1 H); 7.14 (dd, J 8, 2, 1 H); 7.12 (d, J 2, 1 H); 3.30 (dd, J 11, 6.5, 1 H); 3.14 (m, 2 H); 2.82

(d, J 17, 1 H); 2.60 (m, 1 H); 1.98( m, 1 H); 1.45 (m, 1 H); 1.28( s, 9 H); 1.20 (br., OH). ¹³C-NMR: Table 7.

MS: 216 (30, M ), 201 (90), 185 (35), 170 (11), 155 (24), 141 (22), 129 (37), 115 (17), 91 (5), 57 (100), 41 (11).

(1RS,1aSR,6aRS)-3-(tert-Butyl)-1,1a,6,6a-tetrahydrocyclopropa[a]indene-1-methanol (65). Following the

G.P. A, 61 (0.8g, 2.9 mmol; purity 94%) was converted to crude 65 (0.72 g). Bulb-to-bulb distillation (oven

temp. ! 1758/0.25 mbar) afforded 65 (0.62 g, 90%; purity 91%). Colorless oil. IR: 3311 (br.), 2957, 2902, 1490,

1

1361, 1262, 1014, 816. H-NMR: 7.32 (d, J 2, 1 H); 7.13 (dd, J 8, 2, 1 H); 7.02 (d, J 8, 1 H); 3.29 (m, 1 H);

3.17 (m, 1 H); 3.11 (dd, J 17, 6, 1 H); 2.80 (d, J 17, 1 H); 2.62 (m, 1 H); 1.99 (m, 1 H); 1.47 (m, 1 H); 1.30

(s, 9 H); 1.10 (br., OH). ¹³C-NMR: Table 7. MS: 216 (18, M ), 201 (47), 185 (27), 170 (4), 155 (17), 141 (19), 129

(33), 115 (16), 91 (4), 57 (100), 41 (11).

(1RS,1aRS,6aSR)-4-(tert-Butyl)-1,1a,6,6a-tetrahydrocyclopropa[a]indene-1-carboxaldehyde (16). Follow-

ing the G.P. B, 62 (1.8g, 8.3 mmol) was converted to crude 16 (1.57 g). Bulb-to-bulb distillation (oven temp. !

1408/0.25 mbar) afforded 16 (1.5 g, 84%). Colorless solid. Crystallization from pentane (2 ml) at À 308 gave

colorless crystals, M.p. 68 69 8. IR: 3020, 2960, 2900, 2860, 2820, 2710, 1695, 1485, 1360, 1265, 1110, 1010, 890.

¹H-NMR: 9.42 (d, J 4, 1 H); 7.26 (d, J 8, 1 H); 7.22 (d, J 2, 1 H); 7.19 (dd, J 8, 2, 1 H); 3.31 (dd, J 17, 6,

1 H); 3.07 (m, 1 H); 3.04 (d, J 17, 1 H); 2.58( m, 1 H); 1.52 (m, 1 H); 1.29 (s, 9 H). ¹³C-NMR: Table 7. MS: 214

(36, M ), 199 (97), 185 (100), 170 (25), 155 (42), 141 (33), 129 (65), 115 (27), 91 (7), 57 (96), 41 (12).

(1RS,1aRS,6aSR)-3-(tert-Butyl)-1,1a,6,6a-tetrahydrocyclopropa[a]indene-1-carboxaldehyde (17). Follow-

ing the G.P. B, 63 (2.53 g, 11.7 mmol) was converted to 17 (2.02 g), which was bulb-to-bulb distilled (oven

temp. ! 1408/0.3 mbar): 17 (1.98g; purity 95%). CC (SiO ₂(200 g), cyclohexane/Et₂O 4 :1), followed by bulb-

to-bulb distillation (oven temp. ! 1408/0.2 mbar) afforded purified 17 (1.8g, 71%). This material was dissolved

in pentane (3 ml) and left at À 308 during 7 days, whereby a solid had deposited. The cold solvent was removed

and the solid allowed to reach r.t. under vacuum, whereby it melted to give pure 17 (1.01 g). IR: 3020, 3000,

2960, 2900, 2880, 2820, 2710, 1695, 1485, 1365, 1260, 1110, 1015, 920, 820. ¹H-NMR: 9.44 (d, J 4, 1 H); 7.38

(d, J 2, 1 H); 7.20 (dd, J 8, 2, 1 H); 7.11 (d, J 8, 1 H); 3.27 (dd, J 17, 6, 1 H); 3.09 (m, 1 H); 3.03 (d, J 17,

1 H); 2.59 (m, 1 H); 1.53 (m, 1 H); 1.31 (s, 9 H). ¹³C-NMR: Table 7. MS: 214 (55, M ), 199 (100), 185 (72), 171

(16), 155 (27), 141 (30), 129 (62), 115 (26), 91 (7), 57 (85), 41 (11).

(1RS,1aSR,6aRS)-4-(tert-Butyl)-1,1a,6,6a-tetrahydrocyclopropa[a]indene-1-carboxaldehyde (18). Follow-

ing the G.P. B, 64 (0.36 g, 1.7 mmol) was converted to crude 18 (0.26 g). Bulb-to-bulb distillation (oven temp. !

1608/0.25 mbar) afforded 18 (0.25 g, 53%; purity 81%). Colorless oil. CC (SiO₂(30 g), cyclohexane/Et₂O 4 :1)

gave purified product (0.24 g), which solidified on standing. Crystallization from pentane at À 308 afforded

colorless crystals. M.p. 67 688. IR: 2958, 2913, 2866, 1693, 1490, 1466, 1438, 1362, 1267, 1190, 1065, 971, 924, 834.

¹H-NMR: 8.56 (d, J 7, 1 H); 7.27 (d, J 8, 1 H); 7.23 (dd, J 8, 2, 1 H); 7.22 (d, J 8, 1 H); 3.39 (dd, J 17, 6,

1 H); 3.22 (d, J 17, 1 H); 3.11 (m, 1 H); 2.52 (m, 1 H); 1.88 (m, 1 H); 1.30 (s, 9 H). ¹³C-NMR: Table 7. MS: 214

(18, M ), 199 (37), 185 (41), 170 (9), 155 (15), 141 (12), 129 (27), 115 (11), 57 (100), 41 (8).

(1RS,1aSR,6aRS)-3-(tert-Butyl)-1,1a,6,6a-tetrahydrocyclopropa[a]indene-1-carboxaldehyde (19). Follow-

ing the G.P. B, 65 (0.55 g, 2.3 mmol; purity 91%) gave crude 19 (0.5 g, 95%; purity 93%). Yellowish solid. Two

1792

Helvetica Chimica Acta Vol. 87 (2004)

crystallizations from Et₂O at À 308 afforded colorless crystals. M.p. 89 908. IR: 2948, 2902, 1689, 1487, 1359,

1194, 1068, 971, 927, 889, 829, 817, 774. ¹H-NMR: 8.55 (d, J 7, 1 H); 7.38( d, J 2, 1 H); 7.22 (dd, J 8 , 2, 1 H);

7.11 (d, J 8, 1 H); 3.35 (dd, J 17, 6, 1 H); 3.20 (d, J 17, 1 H); 3.13 (m, 1 H); 2.53 (m, 1 H); 1.90 (m, 1 H);

1.30 (s, 9 H). ¹³C-NMR: Table 7. MS: 214 (60, M ), 199 (100), 185 (75), 171 (14), 155 (27), 141 (28), 129 (60),

115 (25), 57 (75), 41 (11).

REFERENCES

[1] S. Lamboley, C. Morel, A. F. Boschung, N. G. J. Richards, B. M. Winter, Chem. Senses 1997, 22, 223.

[2] W. Berends, L. M. van der Linde, Perfum. Essent. Oil Rec. 1967, 58, 372.

[3] a) M. S. Carpenter, W. M. Easter, to Givaudan Corp., U.S. 2.875.131, priority 11 June 1956 (Chem. Abstr.

1959, 53, 10130c); b) J. Dorsky, W. M. Easter, to Givaudan Corp., U.S. 2.976.321, priority 18June, 1959

(Chem. Abstr. 1961, 55, 16485h).

¬

[4] a) K. J. Rossiter, Chem. Rev. 1996, 96, 3201; b) P. Kraft, J. A. Bajgrowicz, C. Denis, G. Frater, Angew.

Chem., Int. Ed. 2000, 39, 2980.

[5] R. Pelzer, U. Harder, A. Krempel, H. Sommer, H. Surburg, P. Hoever, in −Recent Developments of Flavor

and Fragrance Chemistry×, Proceedings of the 3rd International Haarmann & Reimer Symposium, Eds. R.

Hopp, K. Mori, VCH Publishers, Weinheim, Germany, 1993, p. 29.

[6] F. Yoshii, T. Nakamura, S. Hirono, Y. Shimizu, T. Hoshi, M. Ando, H. Hagiwara, Helv. Chim. Acta 2001, 84,

2051.

[7] R. W. Teegarden, L. Steinbach, to International Flavors & Fragrances Inc., U.S. 2.947.780, priority 2 August,

1960 (Chem. Abstr. 1961, 55, 2522c).

[8] K. Raming, J. Krieger, J. Strotmann, I. Boekhoff, S. Kubick, C. Baumstark, H. Breer, Nature (London)

1993, 361, 353.

[9] M. S. Singer, G. M. Shepherd, NeuroReport 1994, 5, 1297; M. Afshar, R. E. Hubbard, J. Demaille,

Biochimie 1998, 80, 129.

[10] J. A. Bajgrowicz, C. Broger, −Proceedings of the 13th Int. Congress of Flavours, Fragrances and Essential

Oils×, Istanbul, Turkey, October 1995, in −Flavours, Fragrances and Essential Oils×, Ed. K. H. Baser, AREP

Publ., Istanbul, 1995, p. 1.

[11] H. Kiefer, J. Krieger, J. D. Olszewski, G. von Heijne, G. D. Prestwich, H. Breer, Biochemistry 1996, 35,

16077.

[12] M. Spehr, G. Gisselmann, A. Poplawski, J. A. Riffell, C. H. Wetzel, R. K. Zimmer, H. Hatt, Science

(Washington, D.C.) 2003, 299, 2054.

[13] G. Skouroumounis, B. Winter, Helv. Chim. Acta 1996, 79, 1095.

[14] a) Naarden en Shell Aroma Chemicals B. V., Neth. Appl. 7905,175, priority 03 July, 1979 (Chem. Abstr.

1981, 95, 24555x); b) S. A. Voitkevich, A. V. Gurevich, L. A. Kheifits, A. G. Bel×fer, T. E. Greshneva, N. V.

Lepikhina, V. A. Kovalenko, N. A. Strykov, I. S. Uspenskaya, G. V. Fedotov, to All-Union Scientific

Research Institute of Synthetic and Natural Perfumes, USSR, SU 1.047.122, priority 7 February, 1992 (Chem.

Abstr. 1993, 118, 27299w).

[15] S. Arctander, −Perfume and Flavor Chemicals×, Published by the Author, Montclair, N.J., U.S.A, 1969,

No. 1211.

[16] E. Hannig, Pharmazie 1965, 20, 762; D. M. Brouwer, J. A. van Doorn, A. A. Kiffers, P. A. Kramer, Recl.

Trav. Chim. Pays-Bas 1974, 93, 189; F. Boberg, K. Deters, J. Schutz, K.-F. Torges, Phosphorus, Sulfur, Silicon

1994, 91, 69.

[17] a) B. Winter, P.-A. Blanc, S. Lamboley, to Firmenich SA, Eur. Pat. Appl. EP 685444, priority 31 May, 1994

(Chem. Abstr. 1996, 124, 155725t); b) S. Arctander, −Perfume and Flavour Chemicals×, Published by the

Author, Montclair, N.J., U.S.A., 1969, No. 496.

[18] P. Cagniant, Buu-Hoi, Bull. Soc. Chim. Fr. 1942, 9, 119; D. M. Brouwer, J. A. van Doorn, A. A. Kiffers,

P. A. Kramer, Recl. Trav. Chim. Pays-Bas 1974, 93, 189.

[19] J. W. Burnham, W. P. Duncan, E. J. Eisenbraun, G. W. Keen, M. C. Hamming, J. Org. Chem. 1974, 39, 1416.

[20] P. Cagniant, Buu-Hoi, Bull. Soc. Chim. Fr. 1942, 9, 111; N. G. Bromby, A. T. Peters, F. M. Rowe, J. Chem.

Soc. 1943, 144; R. A. Jackson, C. J. Rhodes, J. Chem. Soc., Perkin Trans. 2 1993, 53.

[21] J. A. Virgilio, E. Heilweil, Org. Prep. Proced. Int. 1982, 14, 9; T. Thiemann, B. Gehrcke, A. de Meijere,

Synlett 1993, 483.

Helvetica Chimica Acta Vol. 87 (2004)

1793

¬

[22] a) Buu-Hoi, P. Cagniant, C.R. Hebd. Seances. Acad. Sci. 1942, 214, 115; b) P. Cagniant, D. Cagniant, Bull.

¬

Soc. Chim. Fr., 1956, 1152; c) R. Legros, P. Cagniant, C.R. Hebd. Seances. Acad. Sci. 1959, 248, 3179.

[23] S. Arctander, −Perfume and Flavor Chemicals×, Published by the Author, Montclair, N.J., U.S.A, 1969,

No. 406.

[24] R. C. Gilmore Jr., W. J. Horton, J. Am. Chem. Soc. 1951, 73, 1411.

[25] W. T. Heintzelman, B. B. Corson, J. Org. Chem. 1957, 22, 25.

[26] P. Cagniant, A. Reisse, D. Cagniant, Bull. Soc. Chim. Fr. 1969, 985.

[27] a) R. D. Howells, J. D. McCown, Chem. Rev. 1977, 77, 69; b) G. J. Quallich, T. M. Woodall, Tetrahedron

1992, 48, 10329.

[28] S. S. Hixson, J. Borovsky, J. Am. Chem. Soc. 1976, 98, 2840.

[29] M. Saunders, K. N. Houk, Y.-D. Wu, W. C. Still, M. Lipton, G. Chang, W. C. Guida, J. Am. Chem. Soc. 1990,

112, 1419.

[30] A. R. Leach, in −Reviews in Computational Chemistry×, Eds. K. B. Lipkowitz, D. B. Boyd, VCH Publishers,

New York, 1991, Vol. 2, p. 1.

[31] F. Guarnieri, W. C. Still, J. Comput. Chem. 1994, 15, 1302.

[32] H. Senderowitz, F. Guarnieri, W. C. Still, J. Am. Chem. Soc. 1995, 117, 8211.

[33] D. Berkowitz, J. A. McCammon, Chem. Phys. Lett. 1982, 90, 215.

[34] M. Boelens, H. J. Wobben, J. Heydel, Perfum. Flavor. 1980, 5(6), 1.

¬

[35] a) C. L. Ham, P. C. Jurs, Chemical Senses 1985, 10(4), 491; b) G. Frater, J. A. Bajgrowicz, P. Kraft,

Tetrahedron 1998, 54, 7633.

[36] B. Winter, Helv. Chim. Acta 1989, 72, 1278; B. Winter, Pure Appl. Chem. 1990, 62, 1377.

[37] D. Enders, H. Dyker, Liebigs Ann. Chem. 1990, 1107; D. Bartschat, S. Bˆrner, A. Mosandl, J. W. Bats, Z.

Lebensm.-Unters. Forsch. A 1997, 205, 76.

[38] P. Mombaerts, Annu. Rev. Genomics Hum. Genet. 2001, 2, 493.

[39] a) L. Buck, R. A. Axel, Cell 1991, 65, 175; b) X. Zhang, S. Firestein, Nat. Neurosci. 2002, 2, 124.

[40] L. Buck, Annu. Rev. Neurosci. 1996, 19, 517; L. Dryer, A. Berghard, Trends Pharmacol. Sci. 1999, 20, 413;

W. B. Floriano, N. Vaihedi, W. A. Goddard, III, M. S. Singer, G. M. Shepherd, Proc. Natl. Acad. Sci. U.S.A.

2000, 97, 10712; N. Vaidehi, W. B. Floriano, R. Trabanino, S. E. Hall, P. Freddolino, E. J. Choi, G.

Zamanakos, W. A. Goddard, III, Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 12622.

[41] P. Leff, Trends Pharmacol. Sci. 1995, 16, 89; P. Mombaerts, Science (Washington, D.C.) 1999, 286, 707.

[42] R. D. Cramer, D. E. Patterson, J. D. Bunce, J. Am. Chem. Soc. 1988, 110, 5959; A. M. Davis, S. J. Teague,

G. J. Kleywegt, Angew. Chem., Int. Ed. 2003, 42, 2718.

[43] N. L. Allinger, J. Am. Chem. Soc. 1977, 99, 8127.

[44] F. Mohamadi, N. G. J. Richards, W. C. Guida, R. Liskamp, M. Lipton, C. Canfield, G. Chang, T.

Hendrickson, W. C. Still, J. Comput. Chem. 1990, 11, 440.

[45] M. Lipton, W. C. Still, J. Comput. Chem. 1988, 9, 343.

[46] C. Chapuis, J.-Y. de Saint Laumer, Poster at the −XI^thEuropean Symposium on Quantitative Structure-

Activity Relationships×, Lausanne, Switzerland, September 1 6, 1996.

Received April 14, 2004

Article Doi

Synthesis and properties of conformationally constrained analogues of floral-type odorants

DOI: 10.1002/hlca.200490159

Source and publish data:

Authors:

Article abstract of DOI:10.1002/hlca.200490159

Full text of DOI:10.1002/hlca.200490159

Products guided by the article

R&D Labs maybe for 102003-61-2

Relevant to this article

Hot Product