Regioselectivity of acid-catalyzed cyclization of 1-(3,4-dialkylaryl)-3- chloropropan-1-ones to indanones. Comparison of experimental data and results of computer simulation

Article abstract of DOI:10.1007/s11172-009-0117-0

The acid-catalyzed cyclization of 1-(3,4-dialkylaryl)-3-chloropropan-1-ones to dialkyl-indanones via the intermediate formation of (3,4-dialkylaryl) propenones was studied. This reaction affords isomeric products: 5,6-dialkylindan-1-ones and 4,5-dialkylin

Full text of DOI:10.1007/s11172-009-0117-0

Russian Chemical Bulletin, International Edition, Vol. 58, No. 5, pp. 929—935, May, 2009
929
Regioselectivity of acidꢀcatalyzed cyclization
of 1ꢀ(3,4ꢀdialkylaryl)ꢀ3ꢀchloropropanꢀ1ꢀones to indanones.
Comparison of experimental data and results of computer simulation
P. V. Ivchenko, I. E. Nifant´ev, L. Yu. Ustynyuk, and V. A. Ezerskii
Department of Chemistry, M. V. Lomonosov Moscow State University,
1 Leninskie Gory, 119992 Moscow, Russian Federation.
Fax: +7 (495) 939 4523. Eꢀmail: inpv@org.chem.msu.ru
The acidꢀcatalyzed cyclization of 1ꢀ(3,4ꢀdialkylaryl)ꢀ3ꢀchloropropanꢀ1ꢀones to dialkylꢀ
indanones via the intermediate formation of (3,4ꢀdialkylaryl)propenones was studied. This
reaction affords isomeric products: 5,6ꢀdialkylindanꢀ1ꢀones and 4,5ꢀdialkylindanꢀ1ꢀones. The
DFT quantum chemical calculation results correlate with the experimental data and suggest
that the structural factors affect the ratio of products.
Key words: indanones, Friedel—Crafts acylation, Nazarov reaction, quantum chemical
calculations.
Substituted indanones1 are widely used in organic
synthesis and organometallic chemistry. Among many
methods for their synthesis, one of the most attractive is
the acylation of aromatic hydrocarbons with unsaturated
acid chlorides in the presence of electrophilic catalysts
followed by the cyclization of intermediate aryl vinyl ketones
to indanones (Scheme 1).2,3 This method is experimenꢀ
tally simple and the starting reactants are accessible. Along
with unsaturated acid derivatives, 2ꢀ or 3ꢀalkanoyl chloꢀ
rides can be used.4,5 In the case of 3ꢀhalogencarboxylic
acids, the reaction is sometimes carried out in two steps,
and in the second step 2ꢀhaloalkyl aryl ketone is subjected
to the acidꢀcatalyzed cyclization (Scheme 1). The reaction
also proceeds through the intermediate formation of the
protonated form of aryl vinyl ketone.6
tuted to the benzene ring, the reaction affords mixtures of
isomeric indanones. In this case, hydrocarbons formally
similar in structure and electronic properties sometimes
give different products. For example, tetralin is predomiꢀ
nantly transformed into the αꢀcyclization product, whereas
indane mainly yields the βꢀproduct3 (Scheme 2). Reasons for
the observed difference in reactivity are unclear and canꢀ
not be explained by general concepts with allowance for
only the ortho/paraꢀorienting effects of alkyl substituents.
We attempted to find reasons for these differences in
reactivity. For this purpose we studied the acidꢀcatalyzed
cyclization of the series of 1ꢀ(3,4ꢀdialkylaryl)ꢀ3ꢀchloroꢀ
propanꢀ1ꢀones 1—4. To interpret the data obtained, we
studied this process by the DFT method and theoretically
estimated the ratio of αꢀ and βꢀisomers. Since the results
of calculations qualitatively coincided with the experiꢀ
mental data, we proposed the factors determining the
regioselectivity of this reaction.
The key step of the processes discussed is the Nazarov
reaction, viz., the acidꢀcatalyzed cyclization of aryl vinyl
ketones to indanones.6 In the case of substrates substiꢀ
Scheme 1
Published in Russian in Izvestiya Akademii Nauk. Seriya Khimicheskaya, No. 5, pp. 909—915, May, 2009.
1066ꢀ5285/09/5805ꢀ0929 © 2009 Springer Science+Business Media, Inc.

930
Russ.Chem.Bull., Int.Ed., Vol. 58, No. 5, May, 2009
Ivchenko et al.
Scheme 2
Table 1. Yields and ratios of cyclization products of
ketones 1—4 to indanones 5—12
Comꢀ
pound
RR
Yield
(%)
Ratio
1
2
3
4
Me, Me
(СН2)4
(СН2)3
(СН2)2
98
98
94
63
5 : 6
7 : 8
9 : 10
1 : 2
1 : 0.7
1 : 4.5
1 : 13
11 : 12
the NMR spectra of the crude materials. The reaction
rate decreasing in the series of compounds 4 > 3 > 1 > 2
was approximately evaluated. Compound 2 is least reacꢀ
tive, although in the series of fused aliphatic rings the
sixꢀmembered fused fragment is most electronꢀdonating.9
The obtained experimental data were further verified
using computer simulation.
The reaction affording isomeric dialkylindanꢀ1ꢀones
occurs via the mechanism shown in Scheme 4.6 In the
first step, aryl vinyl ketones 13—16 and the corresponding
protonated forms 17—20 are formed upon the action of
H2SO4 on 1ꢀarylꢀ3ꢀchloropropanꢀ1ꢀones. Compounds
17—20 undergo cyclization via the Nazarov mechanism
to form cationic intermediates 21≠—28≠, which then elimiꢀ
nate a proton to produce enolic forms 29—36. Their
isomerization affords indanones 5—12.
Results and Discussion
1ꢀ(3,4ꢀDialkylaryl)ꢀ3ꢀchloropropanꢀ1ꢀones 1—4 have
The key step of the reaction is the cyclization of aryl
vinyl ketones 17—20 protonated at oxygen (Scheme 4),
which can exist as different conformers. The conformers
involved in the Nazarov reaction are shown in Scheme 4.
The cyclization of αꢀ17—αꢀ20 affords indanones 5, 7, 9,
and 11, respectively; compounds βꢀ17—βꢀ20 are cyclized
to form isomeric indanones 6, 8, 10, and 12, respectively.
The ratio of the reaction products is determined by the
step of cyclization of protonated vinyl ketones 17—20.
The structures and energies of intermediates 17—20 and
transitions states 21≠—28≠ were calculated. The unique
PRIRODA program was used for the calculations.10 The
calculations were performed for the gas phase. This simꢀ
plification of the calculation procedure is acceptable,
because the task is the qualitative estimation of the relaꢀ
tive energies and the search for structural factors deterꢀ
earlier been synthesized by the reaction of the correspondꢀ
ing arenes with 3ꢀchloropropionyl chloride in the presꢀ
ence of AlCl3 (Scheme 3).7,8 For compounds 1—3 we
used CH2Cl2 as a solvent. An attempt to obtain comꢀ
pound 4 in this solvent resulted in resinification; the prodꢀ
uct was synthesized in 63% yield when the reaction was
carried out in nitromethane. The yields of the products
(except for compound 4) were high. The moderate yield
of compound 4 is due, most likely, to side electrophilic
reactions accompanied by strained fourꢀmembered ring
opening. It should be mentioned that the acylation proꢀ
ceeds selectivity to position 4 in all cases.
The cyclization of ketones 1—4 to indanones was
carried out on heating in sulfuric acid. The ratios of the
reaction products listed in Table 1 were calculated from
Scheme 3
R = Me (1, 5, 6)
R, R = (CH2)4 (2, 7, 8), (CH2)3 (3, 9, 10), (CH2)2 (4, 11, 12)

Regioselectivitity of indanone formation
Russ.Chem.Bull., Int.Ed., Vol. 58, No. 5, May, 2009
931
Scheme 4
mining the regioselectivity in the series of the sameꢀtype
compounds.
The free energies that allow one to determine the actiꢀ
vation energies were calculated for ketones 17—20 and
transition states 21≠—28≠. The results obtained are given
in Table 2 and as a diagram in Fig. 1. The data presented
show that βꢀcyclization is more favorable for intermediꢀ
ates 17, 19, and 20, whereas for compound 18 αꢀ and βꢀ
cyclizations are almost equiprobable.
The activation energy values for the competitive proꢀ
cesses make it possible to calculate the ratios of αꢀ and
βꢀcyclization products and compare the data obtained
with the experimental results (Table 3). It is seen that the
theoretical and experimental data are rather close. In
addition, the calculated energy values correlate with
the experimentally observed cyclization rates of aryl vinyl
ketones 13—16.
Table 2. Activation energies of the competitive processes for
indanones 5—12 at 360 K
Compound
Isomer
ΔGact/kcal mol–1
5
7
9
11
6
8
10
12
α
α
α
α
β
β
β
β
20.18
19.55
20.19
20.04
19.53
19.55
19.09
18.48
20.18
20.04
20.0
19.5
19.0
19.55
19.53
19.55
19.09
18.48
18.5
0
R = Me (5, 6, 13α,β, 17α,β, 21≠, 22≠, 21, 22, 29, 30)
R, R = (CH2)4 (7, 8, 14α,β, 18α,β, 23≠, 24≠, 23, 24, 31, 32)
R, R = (CH2)3 (9, 10, 15α,β, 19α,β, 25≠, 26≠, 25, 26, 33, 34)
R, R = (CH2)2 (11, 12, 16α,β, 20α,β, 27≠, 28≠, 27, 28, 35, 36)
Fig. 1. Calculated activation energies of the competitive reacꢀ
tions of cyclization of vinyl ketones 17—20 at 360 K.

932
Russ.Chem.Bull., Int.Ed., Vol. 58, No. 5, May, 2009
Ivchenko et al.
Table 3. Products and their ratio
Scheme 6
Products
Ratio
α : β
Experiment
Calculation
5 : 6
7 : 8
9 : 10
11 : 12
1 : 2
1 : 2.5
1 : 1
1 : 4.6
1 : 8.8
1 : 0.7
1 : 4.5
1 : 13
Still in the step of intermediates 17—20, it is seen that
the С(6)—С(7) (α) and С(7)—С(8) (β) bonds are shortꢀ
ened with a decrease in the fused cycle size, which elonꢀ
gates the С(4)—С(5) bond and facilitates the cyclization
to occur.
Both the experiment and calculation show that the
fraction of the βꢀcyclization product increases with a deꢀ
crease in the size of the fused cycle. In order to establish
the factors determining the regioselectivity of formation
of isomeric indanones, it is reasonable to analyze the
geometry of the initial cationic intermediates and transiꢀ
tion states leading to the formation of isomeric indanones
(Schemes 5 and 6). Selected structural parameters of
inermediates αꢀ17—αꢀ20 → 21≠, 23≠, 25≠, 27≠ are given in
Tables 4 and 5, and those for intermediates βꢀ17—βꢀ20 →
22≠, 24≠, 26≠, 28≠ are presented in Tables 6 and 7.
Figure 2 shows the change in the bond lengths in the
aromatic rings during ketone transformation 17—20 →
→ 21≠—28≠. The key structural fragment determining the
regioselectivity of cyclization is the distortion of the aroꢀ
matic ring on going to the transition state: the regular
elongation of the С(4)—С(5) and С(5)—С(6) bonds at
the C(5) atom at which the electrophilic attack occurs is
accompanied by the elongation of the С(7)—С(8) and
С(8)—С(9) symmetric bonds, and the lengths of other
bonds decrease. These distortions are of the same type:
the С(4)—С(5) bond elongates by 0.024—0.029 Å, and the
С(7)—С(8) symmetric bond elongates by 0.009—0.011 Å
in the case of αꢀcyclization and by 0.012—0.014 Å for
βꢀcyclization. The С(6)—С(7) bond becomes shorter by
0.016—0.018 Å (αꢀcyclization) and by 0.017—0.019 Å
(βꢀcyclization).
Scheme 5
Distortions of the benzene ring structure necessarily
involve the alicyclic fragment fused with this ring, beꢀ
Table 4. Selected bond lengths in intermediates α-17—α-20 and in transition states 21≠, 23≠, 25≠,
and 27≠
Bond
d/Å
αꢀ17
6a = 6b, 7a = 7b*
21≠
αꢀ18
23≠
αꢀ19
25≠
αꢀ20
6a = 6b, 7a = 7b*
27≠
6b = 7b*
4—5
5—6
6—7
7—8
8—9
4—9
6—6a
7—7a
6a—6b
7a—7b
6b—7b
1.423
1.389
1.425
1.413
1.380
1.421
1.504
1.495
—
1.449
1.429
1.407
1.422
1.393
1.401
1.502
1.504
—
1.421
1.391
1.420
1.416
1.377
1.424
1.515
1.503
1.532
1.531
1.530
1.448
1.430
1.402
1.425
1.390
1.404
1.511
1.513
1.535
1.534
1.533
1.427
1.382
1.416
1.405
1.384
1.426
1.511
1.499
1.548
1.548
—
1.456
1.422
1.399
1.414
1.396
1.405
1.507
1.510
1.553
1.552
—
1.434
1.377
1.410
1.398
1.390
1.432
1.522
1.512
—
1.463
1.416
1.394
1.409
1.400
1.411
1.519
1.520
—
—
—
—
—
—
1.580
—
1.583
* The positions are shown in Scheme 5. The coinciding positions are marked with the sign “=”.

Regioselectivitity of indanone formation
Russ.Chem.Bull., Int.Ed., Vol. 58, No. 5, May, 2009
933
Table 5. Selected bond angles in intermediates αꢀ17—αꢀ20 and in transition states 21≠, 23≠, 25≠,
and 27≠
Angle
ω/deg
αꢀ19
6b = 7b
αꢀ17
6a = 6b, 7a = 7b
21≠
αꢀ18
23≠
25≠
αꢀ20
6a = 6b, 7a = 7b
27≠
4—5—6
5—6—7
6—7—8
7—8—9
8—9—4
9—4—5
7—6—6a
6—6a—6b
6a—6b—7b
6b—7b—7a
7b—7a—7
7a—7—6
121.54
118.75
119.56
121.68
119.32
119.12
120.73
—
—
—
—
120.51
120.59
121.63
118.71
119.61
121.73
119.24
119.06
121.18
113.10
110.08
110.47
114.55
121.37
120.63
118.15
119.68
123.09
118.20
119.73
122.01
113.15
110.15
110.56
113.76
121.54
119.25
120.15
121.08
119.33
120.28
119.90
110.13
103.28
104.86
—
118.28
119.58
121.11
120.75
119.20
120.68
111.03
103.23
105.15
—
115.91
122.20
122.98
116.03
121.70
121.16
93.15
86.49
—
114.91
121.40
100.40
117.37
120.56
122.08
93.69
86.38
—
118.30
119.49
123.15
118.29
119.77
121.67
—
—
—
—
120.78
—
87.0
93.28
—
86.49
93.43
103.72
110.20
103.31
110.38
Table 6. Selected bond lengths in intermediates βꢀ17—βꢀ20 and in transition states
22≠, 24≠, 26≠, and 28≠
Bond
d/Å
βꢀ17
7a = 7b, 8a = 8b
22≠
βꢀ18
24≠
7b = 8b
1.448
1.420
1.392
1.438
1.401
1.400
1.515
1.511
1.534
1.535
1.532
βꢀ19
26≠
βꢀ20
7a = 7b, 8a = 8b
28≠
4—5
5—6
6—7
7—8
8—9
4—9
7—7a
8—8a
7a—7b
8a—8b
7b—8b
1.421
1.383
1.410
1.429
1.387
1.422
1.494
1.503
—
1.466
1.422
1.391
1.443
1.399
1.402
1.502
1.499
—
1.424
1.380
1.413
1.424
1.389
1.421
1.503
1.514
1.532
1.532
1.530
1.426
1.386
1.402
1.420
1.381
1.427
1.499
1.510
1.548
1.548
—
1.451
1.425
1.384
1.434
1.393
1.406
1.509
1.505
1.551
1.549
—
1.431
1.393
1.395
1.413
1.375
1.433
1.513
1.521
—
1.459
1.431
1.379
1.425
1.386
1.413
1.522
1.518
—
—
—
—
—
—
—
1.580
1.581
Table 7. Selected bond angles (ω) in intermediates βꢀ17—βꢀ20 and in transition states 22≠, 24≠, 26≠,
and 28≠
Angle
ω/deg
24≠
βꢀ17
22≠
βꢀ18
βꢀ19
26≠
βꢀ20
28≠
4—5—6
5—6—7
6—7—8
7—8—9
8—9—4
9—4—5
7—8—8a
8—8a—8b
8a—8b—7b
8b—7b—7a
7b—7a—7
7a—7—8
119.39
121.59
119.56
118.81
121.48
119.15
120.65
—
—
—
—
120.44
118.44
121.18
119.51
119.98
120.51
119.96
120.07
—
—
—
—
120.04
119.32
121.64
119.59
118.79
121.55
119.07
121.09
113.21
110.18
110.43
114.39
121.24
118.39
121.26
119.51
120.00
120.57
119.91
120.88
114.22
110.29
110.02
113.66
113.66
120.35
119.22
121.04
120.26
119.14
119.95
110.02
103.39
104.81
—
119.35
118.83
120.96
121.50
118.16
120.76
109.77
103.78
104.50
—
121.77
115.87
122.98
122.36
115.74
121.24
93.10
86.58
—
120.71
115.39
123.05
123.57
114.80
122.02
93.01
87.13
—
—
87.11
93.20
—
86.98
92.87
103.76
110.14
103.36
109.68

934
Russ.Chem.Bull., Int.Ed., Vol. 58, No. 5, May, 2009
Ivchenko et al.
1
2
,
Fig. 2. Direction of changing the bond lengths in the aromatic rings during ketone transformation 17—20 → 21≠—28≠: 1, bond
elongation; 2, bond shortening.
cause on going to the transition state the С(6)—С(7) bond
lengths decreases in the case of αꢀcyclization and the
С(7)—С(8) bond length increases in the case of βꢀcyꢀ
clization. For compounds 17 and 18 the factor discussed
is not principally significant: in the case of compound 18,
the conformational mobility of the sixꢀmembered fused
fragment compensates the changes in the geometry of the
aromatic ring at any direction of cyclization. In the case
of intermediates 19 and particularly intermediates 20, the
changes in the geometry of the fused cycles with the
retention and even enhancement of the angular strain in
them correspond to the transition state of αꢀcyclization
accompanied by a decrease in the С(6)—С(7) bond
lengths. On the contrary, in the case of βꢀcyclization the
transition to the transition state is facilitated by the partial
elimination of angular strain.
The discussed structural changes can be illustrated for
the transitions 20α → 27≠ and 20β → 28≠, using for the
qualitative estimation of the angular strain the change in
the total deviation of the bond angles in the fourꢀmemꢀ
bered ring from 90°. In the case of αꢀcyclization this
value is +1.386°, whereas for βꢀcyclization it is –0.831°.
Thus, the experimentally determined ratio of the
Nazarov reaction products corresponds to that found by
the quantum chemical calculations. For the compounds
with the fiveꢀ and fourꢀmembered alicycles, the experiꢀ
mental data coincides well with the computer simulation
results. The influence of the saturated fused fragment on
the regioselectivity of cyclization cannot be explained only
by the electronic effects of alkyl substituents. The reacꢀ
tion is facilitated by distortions in the aromatic ring caused
by the fused cycle. The predominant occurrence of the
cyclization in the βꢀposition is determined by the partial
elimination of the angular strain in the fused cyclic fragꢀ
ments in the corresponding transition state. The presence
of the small fused cyclic fragment can be considered
as the general structural factor affecting both the reactivꢀ
ity of aromatic compounds and regioselectivity of the
process.
Experimental
Calculation procedure. Calculations were performed using
the PRIRODA program10 and PBE functional11 including
the electron density gradient and the extended basis sets of the
Gaussian functions. In the present study, the orbital basis sets
have the following compression samples: (5s1p)/[3s1p] for the
H atom and (5s5p2d)/[3s3p2d] for the C, O, and Cl atoms. The
auxiliary basis sets are the uncompressed sets of the Gaussian
functions of the sizes (5s1p) for the H atom and (6s3p3d1f) for
the C, O, and Cl atoms. Ten electrons of internal shells for the
Cl atom and two electrons for the C and O atoms were described
using the pseudoꢀpotentials (ECP).12—14
The full geometry optimization of all structures described in
the work (both the complexes and transition state) was performed
without restrictions to the symmetry of the molecule using
analytical gradients. The character of the determined stationary
points was found on the basis of the analytical calculation of the
secondary energy derivatives with respect to coordinates.
The thermodynamic characteristics presented in this work
and the vibrational frequencies were calculated in the approxiꢀ
mations of harmonic oscillator, rigid rotator, and ideal gas. The
vibrational spectrum of all calculated reactants, intermediates,
and reaction products contain no imaginary frequency, and the
vibrational spectrum of all optimized transition states has one
imaginary mode corresponding to the reaction coordinate.
Synthetic procedures. All experiments were carried out in an
argon atmosphere. Methylene chloride was purified by distillaꢀ
tion over CaH2. Nitromethane, 3ꢀchloropropionyl chloride,
oꢀxylene, tetralin, and indane (ACROS) were used without
additional purification. 1H and 13C NMR spectra were recorded
on a Varian VXRꢀ400 instrument at 20 °С in CDCl3.
All compounds synthesized in the present work are described,
but no NMR data are presented in the literature. In addition,
some procedures for the synthesis and isolation were modified
and, hence, their description is given in this work along with the
NMR spectral data for compounds 1—8.
Synthesis of compounds 1—4 (general procedure). Aluminum
trichloride AlCl3 (24 g, 0.18 mol) was added with vigorous stirring
to a solution (cooled to 0 °C) of 3ꢀchloropropionyl chloride
(19.7 g, 0.155 mol) in CH2Cl2 (200 mL). The corresponding
arene (0.15 mol) was added dropwise to the resulting suspension
for 30 min, maintaining the temperature of the reaction mixture

Regioselectivitity of indanone formation
Russ.Chem.Bull., Int.Ed., Vol. 58, No. 5, May, 2009
935
about 5 °C. After 1 h the cooling was removed, and the mixture
was left to stay at ambient temperature. Then the reaction
mixture was carefully quenched with a mixture of a saturated
solution of HCl (30 mL), cold water (150 mL), and ice (150 g).
The organic layer was separated, and the aqueous solution was
extracted with chloroform (3×50 mL). The combined organic
phases were washed with water, a 5% solution of Na2CO3, and
again with a minimum water, dried over Na2SO4, and conꢀ
centrated in vacuo.
3ꢀChloroꢀ1ꢀ(3,4ꢀdimethylphenyl)propanꢀ1ꢀone (1). The yield
was 98% (light gray powder). 1H NMR, δ: 2.33 (br.s, 6 H); 3.42
(t, 2 H, J = 6.8 Hz); 3.92 (t, 2 H, J = 6.8 Hz); 7.23 (d, 1 H,
J = 7.7 Hz); 7.69 (d, 1 H, J = 7.7 Hz); 7.74 (s, 1 H).
3ꢀChloroꢀ1ꢀ(5,6,7,8ꢀtetrahydroꢀ2ꢀnaphthyl)propanꢀ1ꢀone
(2). The yield was 98% (beige powder). 1H NMR, δ: 1.83 (m,
4 H); 2.82 (m, 4 H); 3.42 (t, 2 H, J = 6.8 Hz); 3.91 (t, 2 H,
J = 6.8 Hz); 7.16 (d, 1 H, J = 8.4 Hz); 7.69—7.65 (m, 3 H).
3ꢀChloroꢀ1ꢀ(2,3ꢀdihydroꢀ1Hꢀindenꢀ5ꢀyl)propanꢀ1ꢀone (3).
The yield was 94%. 1H NMR, δ: 2.13 (q, 2 H, J = 7.5 Hz);
2.97 (t, 4 H, J = 7.5 Hz); 3.44 (t, 2 H, J = 6.9 Hz); 3.92 (t, 2 H,
J = 6.9 Hz); 7.31 (d, 1 H, J = 7.9 Hz); 7.76 (d, 1 H, J = 7.9 Hz);
7.82 (s, 1 H).
1ꢀ(Bicyclo[4.2.0]octaꢀ1,3,5ꢀtrienꢀ3ꢀyl)ꢀ3ꢀchloropropanꢀ
1ꢀone (4) was obtained according to the general procedure using
nitromethane as a solvent. The residue after evaporation was
recrystallized from hexane. The yield was 63% (pale pink
powder). 1H NMR, δ: 3.23 (br.s, 4 H); 3.43 (t, 2 H, J = 6.8 Hz);
3.92 (t, 2 H, J = 6.8 Hz); 7.15 (d, 1 H, J = 8 Hz); 7.65 (s, 1 H);
7.85 (d, 1 H, J = 8 Hz).
compound 8, δ: 1.80—1.90 (m, 4 H); 2.62—2.71 (m, 4 H); 2.83
(t, 2 H, J = 5.7 Hz); 3.04 (t, 2 H, J = 5.7 Hz); 7.16 (s, 1 H); 7.45
(s, 1 H).
1,6,7,8ꢀTetrahydroꢀasymmꢀ(2H)ꢀindacenꢀ3ꢀone (9) and
3,5,6,7ꢀtetrahydroꢀsymmꢀ(2H)ꢀindacenꢀ1ꢀone (10). The overall
yield was 75%, white powder. The ratio 9 : 10 was 1 : 4.5.
1H NMR spectrum of compound 9, δ: 2.20 (q, 2 H, J = 7.3 Hz);
2.68—2.71 (2 H); 2.90—3.02 (6 H); 7.24 (d, 1 H, J = 7.7 Hz);
7.58 (d, 1 H). 1Н NMR spectrum of compound 10, δ: 2.13
(q, 2 H, J = 7.3 Hz); 2.68 (t, 2 H, J = 7.3 Hz); 2.92 (t, 2 H,
J = 7.3 Hz); 2.95 (t, 2 H, J = 5.8 Hz); 3.07 (t, 2 H, J = 5.8 Hz);
7.29 (s, 1 H); 7.57 (s, 1 H).
1,2,5,6ꢀTetrahydroꢀ4Hꢀcyclobuta[f]indenꢀ4ꢀone (12). The
yield was 45%, pale pink powder. 1H NMR, δ: 2.65 (t, 2 H,
J = 5.8 Hz); 3.10 (t, 2 H, J = 5.8 Hz); 3.19 (br.s, 4 H); 7.14 (s,
1 H); 7.41 (s, 1 H).
References
1. N. B. Ivchenko, P. V. Ivchenko, I. E. Nifant´ev, Zh. Org.
Khim., 1999, 35, 641 [Russ. J. Org. Chem. (Engl. Transl.),
1999, 35].
2. W. Kaminsky, O. Rabe, A.ꢀM. Schauwienold, G. U. Schupfner,
J. Hanss, J. Kopf, J. Organomet. Chem., 1995, 497, 181.
3. U. Dietrich, M. Hackmann, B. Rieger, M. Klinga, M. Leskela,
J. Am. Chem. Soc., 1999, 121, 4348.
4. L. Resconi, S. Guidotti, I. Camurati, R. Frabetti, F. Focante,
I. E. Nifant’ev, I. P. Laishevtsev, Macromol. Chem., 2005,
206, 1405.
5. R. T. Hart, R. F. Hart, J. Am. Chem. Soc., 1950, 72, 3286.
6. K. R. Koltunov, S. Walspurger, J. Sommer, Tetrahedron Lett.,
2005, 46, 8391.
7. W. Bartmann, E. Konz, W. Rueger, J. Heterocycl. Chem.,
1987, 24, 677.
8. L. Horner, K. Muth, H.ꢀG. Schmelzer, Chem. Ber., 1959, 92,
2953.
9. D. V. Avila, A. G. Davies, E. R. Li, K. M. Ng, J. Chem. Soc.,
Perkin Trans. 2, 1993, 355.
10. D. N. Laikov, Chem. Phys. Lett., 1997, 281, 151.
11. J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett.,
1996, 77, 3865.
Synthesis of compounds 5—8 (general procedure). The
corresponding 1ꢀarylꢀ3ꢀchloropropanꢀ1ꢀone (0.1 mol) was
added in small portions with vigorous stirring to concentrated
H2SO4 (150 mL) heated to 80 °C. The stirring was continued
at 85—90 °C until the end of gas evolution (1 h), monitoring
by TLC in chloroform. Then the reaction mixture was cooled to
20 °С, carefully poured into water (150 mL) with ice (150 g),
and extracted with chloroform (3×50 mL). The combined
extracts were dried over Na2SO4 and concentrated by evaporaꢀ
tion. The ratio of the cyclization products was determined by
1
the analysis of the H NMR spectra.
4,5ꢀDimethylindanꢀ1ꢀone (5) and 5,6ꢀdimethylindanꢀ1ꢀone
(6). The overall yield was 80%, white powder. 1H NMR spectrum
of compound 5, δ: 2.25 (s, 3 H); 2.37 (s, 3 H); 2.67 (t, 2 H,
J = 5.8 Hz); 3.00 (t, 2 H, J = 5.8 Hz); 7.17 (d, 1 H, J = 7.7 Hz);
12. W. J. Stevens, H. Basch, M. Krauss, J. Chem. Phys., 1984,
81, 6026.
13. W. J. Stevens, M. Krauss, H. Basch, P.G.Jasien, Can. J.
Chem., 1992, 70, 612.
1
7.51 (1 H). H NMR of compound 6, δ: 2.30 (s, 3 H); 2.34 (s,
3 H); 2.64 (t, 2 H, J = 5.9 Hz); 3.04 (t, 2 H, J = 5.9 Hz); 7.24
(s, 1 H); 7.50 (s, 1 H).
14. T. R. Cundari, W. J. Stevens, J. Chem. Phys., 1993, 98, 5555.
1,2,6,7,8,9ꢀHexahydroꢀ3Hꢀcyclopenta[a]naphthalenꢀ3ꢀone
(7) and 2,3,5,6,7,8ꢀhexahydroꢀ1Hꢀcyclopenta[b]naphthalenꢀ
1ꢀone (8). The overall yield was 53%, white powder. 1H NMR
spectrum of compound 7, δ: 1.80—1.90 (m, 4 H); 2.62—2.71
(m, 4 H); 2.84 (t, 2 H, J = 5.6 Hz); 2.93 (t, 2 H, J = 5.6 Hz);
7.08 (d, 1 H, J = 7.9 Hz); 7.48 (d, 1 H, J = 7.9 Hz). 1H NMR of
Received July 15, 2008;
in revised form January 12, 2009