The chemical behavior of 3,4-benzocycloocten-1,5-diyne

Article abstract of DOI:10.1002/ejoc.200901371

The title compound 3,4-benzocycloocten-1,5-diyne (1) is a highly reactive hydrocarbon that has been shown to undergo addition reactions with tetraphenyl-cyclopentadienone (tetracyclone) to the 2:1 adduct 6, with octacarbonyldicobalt to the bis metal complex 7, with lithium aluminium hydride to the bis diene 9, and with various electrophilic reagents. In these latter cases cationic intermediates are generated from 1 and a transannular cyclization takes place leading to novel derivatives of 1,2-dihydropentalene (31, 32, 35, 36, 43, 44).

Full text of DOI:10.1002/ejoc.200901371

FULL PAPER
DOI: 10.1002/ejoc.200901371
The Chemical Behavior of 3,4-Benzocycloocten-1,5-diyne[‡]
Christian Werner,[a] Henning Hopf,*[a] Jörg Grunenberg,[b] and Peter G. Jones[c]
Keywords: Alkynes / Strained molecules / Ring contraction / Transannular interactions
The title compound 3,4-benzocycloocten-1,5-diyne (1) is a
highly reactive hydrocarbon that has been shown to undergo
addition reactions with tetraphenyl-cyclopentadienone
(tetracyclone) to the 2:1 adduct 6, with octacarbonyldicobalt
to the bis metal complex 7, with lithium aluminium hydride
to the bis diene 9, and with various electrophilic reagents. In
these latter cases cationic intermediates are generated from
1 and a transannular cyclization takes place leading to novel
derivatives of 1,2-dihydropentalene (31, 32, 35, 36, 43, 44).
Theoretical Structure and Properties of 1
Introduction
Since no X-ray crystallographic structure could be ob-
tained, we carried out gas phase calculations at the Möller–
Plesset level of theory[3] in combination with a correlation
consistent basis set[4] to get an impression of the geometric
structure and the strain of 1. All optimizations and fre-
quency calculations were performed using the Gaussian03
set of programs.[5] According to our MP2/cc-pvtz calcula-
tions, the cycloocten-1,5-diyne substructure minimum in C2
symmetry displays pronounced deviations from the triple
bond linearity of 25° and 20.7°, respectively (see Scheme 1),
introducing an angle strain of 6.4 kcal/mol per “yne” moi-
ety relative to an unperturbed acetylene. This cis distortion
of the carbon–carbon triple bonds in 1 is not only attended
by a weakening in terms of relaxed force constants (inverse
CϵC compliance constants;[6] 13.88 mdynÅ–1 in 1 vs.
15.62 mdynÅ–1 in acetylene, respectively), but also gives
rise to a pronounced reactivity.
Recently we described a simple synthesis of 3,4-benzocy-
cloocten-1,5-diyne (1, Scheme 1), which provides sufficient
material (0.3 g lots can readily be obtained in one prepara-
tive experiment)[2] to study the chemical properties of this
strained (see below) hydrocarbon. Although the compound
crystallizes as colorless needles, an X-ray crystallographic
structure determination failed because of pseudo-symmetry
effects. When the neat compound is exposed to air, its mass
spectrum shows – besides the molecular ion peak – signals
at M+16 and M+32, which might be caused by epoxides
and/or peroxides (endo-peroxides?) formed from 1. Clearly,
we are dealing with a reactive compound making other ad-
dition reactions very likely. As this publication shows, this
is indeed the case.
Cycloadditions to 1
Assuming that the distorted triple bonds of 1 would in-
crease its propensity towards cycloadditions, we first carried
out a Diels–Alder reaction with 1,2,4,5-hexatetraene (2), a
hydrocarbon that has frequently proved its value as a diene
component.[7] If two molecules of it added to 1, the bis(p-
quinodimethane) intermediate 3 could be formed, which
under the reaction conditions could dimerize to the layered
compound 5 (Scheme 2).
However, heating the two components in benzene at
80 °C provided polymeric material only. In a second [2+4]
cycloaddition experiment 1 was heated with excess tet-
raphenylcyclo-pentadienone (4, tetracyclone) in toluene un-
der reflux. After 2 d the bis adduct 6 had been formed in
excellent yield (91%). The compound was characterized by
its spectroscopic data (see Exp. Section) and also a crystal
structure determination (Figure 1).
Scheme 1. 3,4-Benzocycloocten-1,5-diyne (1) and its calculated
structure.
[‡] Alkynes and Cumulenes, XXVI. Part XXV: Ref.[1]
[a] Institut für Organische Chemie, Technische Universität
Braunschweig,
Hagenring 30, 38106 Braunschweig, Germany
Fax: +49-531-391-5388
E-mail: H.Hopf@tu-bs.de
[b] Institut für Organische Chemie, Technische Universität
Braunschweig, Abtlg. Computerchemie,
Hagenring 30, 38106 Braunschweig, Germany
Fax: +49-531-391
E-mail: Joerg.Grunenberg@tu-bs.de
[c] Institut für Anorganische und Analytische Chemie, Technische
Universität Braunschweig,
Postfach 3329, 38023 Braunschweig, Germany
Fax: +49-531-391-5387
E-mail: P.Jones@tu-bs.de
Eur. J. Org. Chem. 2010, 4027–4034
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4027

C. Werner, H. Hopf, J. Grunenberg, P. G. Jones
FULL PAPER
In a second cycloaddition experiment an ethereal solu-
tion of 1 was treated with excess octacarbonyldicobalt (7)
at room temperature (Nicholas reaction).[8] After 20 min
the bis adduct 8 had been produced as the only product; it
was isolated after chromatography and recrystallization as
deep black needles in 65% yield (Scheme 3).
Scheme 3. Reaction of 1 with excess octacarbonyldicobalt.
On standing, the compound is air-oxidized, but it is far
more stable than the starting hydrocarbon. The structure
assignment of the adduct rests on its spectroscopic proper-
ties (see experimental section), with the mass spectrum be-
ing particularly revealing since it displays the successive loss
of all 12 carbonyl ligands. Again, the X-ray structure deter-
mination failed because of pseudo-symmetry effects.
Scheme 2. Cycloadditions with 1.
Ionic Additions to 1
In a first ionic reaction the title compound was treated
with LiAlH4 in ether at room temperature. To our surprise
the two triple bond were reduced to double bonds within
1 h at room temperature and 1,2-benzo-1,3,7-cyclooctatri-
ene (9) was obtained in 91% yield (Scheme 4). The com-
pound was identified by comparison of its spectroscopic
data with those described in the literature.[9]
Scheme 4. The reduction of 1 with lithium aluminum hydride.
Figure 1. The molecule of compound 6 in the crystal. Ellipsoids
represent 30% probability levels. Phenyl hydrogen atoms are omit-
ted for clarity.
We assume that hydride is added to the highly strained
triple bond and that a vinyl anion such as 8 is produced
first. The process is repeated and on work-up the ionic in-
The structure analysis determines unambiguously the termediate is protonated to 9. Whereas functionalized al-
chemical nature of compound 6. The molecule has no im- kynes can be reduced readily by lithium aluminum hydride
posed symmetry. The bond length C4–C5 of 1.551(3) Å (see the often used reduction of propargylic to allylic
clearly establishes this as a single bond. The eight-mem- alcohols[10]) the hydrogenation of acetylenic hydrocarbons
bered ring displays an approximate tub conformation, by this route is rare. Interestingly, we could add neither
whereby the atoms C1,4,5,8 and C2,3,6,7 each form a co- phenyl nor methyllithium to the triple bonds of 1.
planar set (mean deviations 0.13, 0.11 Å) with an in-
Whereas electron-rich intermediates such as 8 do not
terplanar angle of 4°. Strain in the ring system is expressed show any tendency for intraannular constriction by neigh-
e.g. in the torsion angles C1–C2–C3–C4 –17.5(3), C5–C6– boring group participation, a completely different situation
C7–C8 –11.9(3) and C2–C1–C8–C7 9.3(4)° (ideal values 0°) is encountered when electron-deficient intermediates are
and in the formally sp3 bond angle C3–C4–C5 119.6(2)°; generated from 1. As shown in general form in Scheme 5,
surprisingly, the angle C4–C5–C6 is relatively unaffected at an electrophile E+ has two options to add to 1, producing
111.6(2)°.
the regioisomeric cations 10 or 11.
4028
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 4027–4034

The Chemical Behavior of 3,4-Benzocycloocten-1,5-diyne
to two categories, those possessing a [3.3.0]skeleton (12 and
14) and those having a [4.2.0]structure (13 and 15). From
these, by external attack of Y–, the four isomeric adducts
16–19 can be derived. Intuitively one would expect the
routes via 13 and 15 to be disfavored since these contain
strain-producing four-membered rings.
To gain a better understanding of the possibly preferred
addition/cyclization modes we first carried out gas phase
calculations again at the MP2/cc-pvtz level of theory. Ac-
cording to our calculations, regioselective protonation of 1
leads to two different non-classical cations, favoring 10 by
12.9 kcalmol–1 with respect to cation 11 (intrinsic thermo-
dynamic stability, no solvent effects included). Both cations
can be viewed as precursors of the ring closure: A nucleo-
philic attack – the chloride anion, in our theoretical case
study – on 10 leads to the products 16 and 17 depending
on the entry trajectory of the nucleophile. Whereas, in terms
of thermodynamic stability, the products 16 and 17 are
quite similar (16 lying 1.4 kcalmol–1 lower than 17), they
are separated by a substantial kinetic barrier. The transition
structure (one negative Hessian eigenvalue, MP2/cc-pvtz
level of theory) connecting 16 and 17 is only passed through
an energy maximum of 94.3 kcalmol–1. On the other hand,
regioselective nucleophilic attack on cation 11 leads to
product 18 and 19. While the thermodynamic stability of
product 18 is again not far from the product global mini-
mum 16 (+1.4 kcalmol–1), isomer 19 is destabilized by more
than 35 kcalmol–1.
We conclude that in case of an electrophilic attack on 1
both the thermodynamic and kinetic stability should lead
to product 16 as the only observable product (Scheme 6).
To test the above hypotheses we first treated 1 with BF3–
OEt2 in benzene at room temperature (Scheme 7), a process
Scheme 5. The addition of electrophiles to 1 followed by a transan-
nular ring closure: possible reaction pathways.
Excluding the direct trapping of 10 and 11 by a nucleo- yielding a single product in excellent yield (91%). To this
phile Y– to a 1,2-acetylene addition product, but allowing we assign the benzo dihydropentalene structure 24 largely
ring closure by the opposing triple bonds, these cations on the basis of its NMR spectra given in the experimental
could cyclize to the four vinyl cations 12–15, which belong section. In principle the cyclization product could be de-
Scheme 6. Calculated reaction pathways for the addition of HCl to 1.
Eur. J. Org. Chem. 2010, 4027–4034
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4029

C. Werner, H. Hopf, J. Grunenberg, P. G. Jones
FULL PAPER
Scheme 7. Electrophilic additions to 1: experimental results.
rived from all four alternatives 16–19 with a phenyl group has been generated, which then undergoes an electrophilic
as the substituent (Cl in Scheme 6). Evidence against the substitution reaction with the solvent. The resulting borate
two alternatives 18 and 19 is furnished by the observation complex 22 may either be protonated in situ or during hy-
that the proton H-8 of our product is registered as a sharp drolytic work-up. Whereas various dihydropentalenes have
singlet at δ = 7.51 ppm. For structures such as 18 and 19 been described in the literature,[12–15] we could find no refer-
we would expect a triplet or a pseudo-triplet for the corre- ence to benzannelated derivatives such as 24. In an attempt
sponding proton because of coupling with the neighboring to oxidize this hydrocarbon to the corresponding pentalene
methylene group. Besides, both adducts would be formed we heated it in boiling toluene in the presence of DDQ:
via energetically disfavored pathways. To differentiate be- even after several weeks of refluxing no dehydrogenation
tween 16 and 17 is more difficult, and at the present time had taken place.
we favor alternative 16, e.g. structure 24, largely because of
When the same process was repeated in dichloromethane
its very well-resolved 13C NMR spectrum, in which 16 of as the solvent, the chloro dihydro-pentalene 23 was pro-
its 18 carbon atoms can be assigned. For a structure such duced in practically quantitative yield (95%) and presum-
as 17, with its inherently more symmetrical structure, we ably by a very similar mechanism (Scheme 7). Again we
would expect partial overlap of signals. It should be pointed cannot assign its structure unambiguously, since we cannot
1
out that on the basis of the H chemical shifts a distinction rigorously exclude a structure of type 17. That we prefer
between 16 and 17 is not possible. The methylene protons the structure given in Scheme 7 results from the similarity
of 24 absorb as a pseudo-singlet at δ = 3.32–3.39, whereas of its spectroscopic data with those of 24. We were unable
its carbon atoms are registered at two triplets at δ = 28.48 to determine an X-ray structure of 23 or 24.
and 29.04 ppm. The NMR spectra of the corresponding
In the next series of experiments the species E+ and Y–
phenyl derivative of naphtha[b]cyclobutene have not been were varied. In a first experiment we attempted to hydrate
reported, but for the parent system, the 1H NMR spectrum either one or both of the triple bonds of 1, hoping to obtain
displays a singlet at δ = 3.28 ppm and the 13C spectrum a either the diketone 25 or the monoketone 26, which again
signal at δ = 29.4, i.e. with practically identical chemical could be produced by the above transannular cyclization
shifts, as observed for 24.[11] As an additional argument for mechanism. However, applying
a
well-established pro-
favoring a structure of type 16 (e.g. 24) we cite the energeti- cedure[16] for triple bond hydration (Scheme 8) only gave the
cally more favorable pathway leading to this adduct. benzo-dihydropentalene derivatives 27 and 28, presumably by
To rationalize the transannular cyclization we suggest a mechanism analogous to that discussed above (Scheme 7).
that it is initiated by complexation of one of the triple The structure assignment of both products follows from the
bonds of 1 to the Lewis acid, and that subsequently the spectroscopic data given in the experimental section.
facing triple bond will begin to provide electron density to
Recently Sankararaman and co-workers[17] reported that
this now positively polarized section of the substrate (see various enediynes can be cyclized via radical cation inter-
20). When the ring closure is complete, the vinyl cation 21 mediates; a case in point is the cyclization of 29 to the ind-
4030
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 4027–4034

The Chemical Behavior of 3,4-Benzocycloocten-1,5-diyne
Scheme 8. Oxymercuration of 1.
enone derivative 30. However, when the same conditions
From the complex product mixture (containing at least
[tris(p-bromophenyl)ammonium
hexachloroantimonate five products) generated when iodine is added to 1, the two
(TBPA·SbCl6), oxygen gas, dichloromethane] were applied dihydropentalenes 35 and 36 were isolated and charac-
to 1, only 23 was produced (48%), besides 20% of unre- terized. Whether the expected diiodide (ethoxy in 35 re-
acted substrate.
placed by iodine) is also produced, is presently unknown.
As shown in Scheme 9 iodine has been added to various All products are extremely air-sensitive, and even the
cyclic oligoynes; for example, iodine addition to cyclodeca- eventually characterized adducts 35 and 36 are so sensitive
1,6-diyne (31) leads to the constricted diiodide 32,[18] towards oxidation that they can only be kept at low tem-
whereas with the tetrayne 33 the s-indacene derivative 34 is perature (–24 °C) under inert gas for a limited amount of
produced.[19]
time. The iodides 35 and 36 could be formed by a similar
ionic mechanism as discussed above or involve radical inter-
mediates, considering the low heat of dissociation of the I–
I bond.
Conclusions
3,4-Benzocycloocten-1,5-diyne (1) is a strained acetylenic
hydrocarbon that undergoes various types of addition reac-
tions (Diels–Alder addition, hydride addition, electrophilic
additions) readily. Whenever cation intermediates are gener-
ated from 1 a transannular cyclization takes place leading
to novel derivatives of 1,2-dihydropentalene.
Experimental Section
1. General: Melting points up to 200 °C: Büchi 510, above 200 °C:
Reichert melting point stage. Column chromatography: silica gel 60
(70–230 mesh, Merck). NMR: in deuteriochloroform with TMS as
internal standard; Bruker AM-400 (1H NMR, 400 MHz. 13C
NMR, 100 MHz); Bruker AC-200 (1H NMR, 200 MHz. 13C
NMR, 50 MHz). IR: Nicolet 320 FT-IR under the condition given
in the respective experiments. UV/Vis: HP 8542A, in acetonitrile.
MS: MAT 8430 (EI, 70 eV). GC/MS: Carlo–Erba HRGC 5160/
Finnigan MAT 4515 (EI, 40 eV).
Scheme 9. Iodine addition to 1.
Eur. J. Org. Chem. 2010, 4027–4034
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4031

C. Werner, H. Hopf, J. Grunenberg, P. G. Jones
2. 5,6,7,8,11,12,13,14-Octaphenyl-9,10-dihydrotribenzo[a,c,e]cyclo- methane, and the combined organic phases were passed through a
FULL PAPER
octene (6): A solution of 1 (50 mg, 0.32 mmol) and 4 (384 mg,
1 mmol) in toluene (50 mL) was heated under reflux for 2 d. The
solvent was removed in vacuo and the remaining product mixture
purified by column chromatography (silica gel; cyclohexane/dichlo-
romethane = 1:1): 259 mg of 6 (91%), colorless prisms (dichloro-
methane), m.p. 160–170 °C. 1H NMR (400 MHz, CDCl3): δ =
pad of silica gel. After solvent removal and silica gel chromatog-
raphy (cyclohexane) 138 mg (91%) of 24 was obtained as a color-
less, glassy solid; m.p. 81 °C. 1H NMR (400 MHz, CDCl3): δ =
3.32–3.39 (m, 4 H, CH2CH2), 7.37–7.47 (m, 3 H, 6-, 5-, 16-H), 7.51
(s, 1 H, 8-H), 7.52–7.55 (m, 4 H, Ar-H), 7.88 (d, Jo = 6.1 Hz, 1 H,
7-H), 7.99 (d, Jo = 7.4 Hz, 1 H, 4-H) ppm. 13C NMR (100 MHz,
CDCl3): δ = 28.48 (t, C-1), 29.04 (t, C-2), 120.12 (d, C-8), 124.43
(d, C-4), 124.70 (d, C-7), 125.45 (d, C-6), 127.14 (d, C-16), 128.33
3
2.50–2.55 (m, 2 H, CH2), 2.72–2.80 (m, 2 H, CH2), 6.04 (d, Jo =
7.7 Hz, 2 H, Ar-H), 6.56–7.14 (m, 44 H, Ar-H) ppm. 13C NMR
(100 MHz, CDCl3): δ = 26.90 (t, CH2), 32.12 (t, CH2), 124.87 (d), (d, C-5), 128.37 (d, C-15), 129.96 (d, C-14), 131.59 (s, C-12), 132.42
124.91 (d), 125.19 (d), 125.69 (d), 125.78 (d), 126.16 (d), 126.37 (d),
126.41 (d), 126.63 (d), 126.97 (d), 127.14 (d), 130.91 (d), 131.06 (d),
131.16 (d), 131.34 (d), 131.60 (d), 131.70 (d), 132.10 (d), 132.36 (d),
136.87 (s), 139.18 (s), 139.27 (s), 140.10 (s), 140.13 (s), 140. 35 (s),
140.45 (s), 140.77 (s), 141.00 (s), 141.30 (s, Ar-C) ppm. IR (KBr):
(s, C-9), 134.69 (s, C-10), 136.94 (s, C-13), 142.75 (s, C-3), 143.92
(s, C-11) ppm. IR (KBr): ν = 3052 (w), 2971 (m), 2932 (s), 1509
˜
(m), 1403 (m), 871 (s), 757 (s), 747 (s), 697 (s) cm–1. UV/Vis (aceto-
nitrile): λmax (logε) = 192 (4.66), 228 (4.75), 294 nm (4.01). MS
(70 eV): m/z (%) = 230 (97) [M+], 215 (100), 202 (28). C18H14
(230.31): calcd. C 93.87, H 6.13; found C 93.73, H 6.07.
ν = 3055 (s), 2925 (s), 1442 (s), 1028 (s), 738 (s), 699 (s) cm–1.
˜
UV/Vis (acetonitrile): λmax (logε) = 194 (5.23), 240 nm (4.86). MS
(70 eV): m/z (%) = 866 (27), 865 (75), 864 (100) [M+], 787 (4), 695
(4), 432 (10), 382 (12). HRMS: calcd. 864.3756; found 864.375. X-
ray: see main section and below.
6. 3-Chloro-1,2-dihydrocyclopenta[a]indene (23): As described under
5 a solution of 1 (100 mg, 0.66 mmol) was prepared, but in dichlo-
romethane (100 mL). Work-up as described above furnished 14
(118 mg, 95%) as a colorless, glassy solid; m.p. 52 °C. 1H NMR
(400 MHz, CDCl3): δ = 3.31–3.33 (m, 2 H, CH2), 3.36–3.38 (m, 2
H, CH2), 7.41 (s, 1 H, 8-H), 7.47 (m, Jo = 7.9, Jm = 1.5 Hz, 1 H,
3. 1,2;5,6-Bis(dicobalthexacarbonyl)-3,4-benzocyclooctene (7): To a
solution of 1 (100 mg, 0.658 mmol) in anhydrous diethyl ether
(50 mL) was added under nitrogen Co2(CO)8 (674 mg, 1.97 mmol).
The mixture was stirred for 20 min at room temp. (gas formation),
and the solvent was removed by rotary evaporation. After 4 h un-
der vacuum the adduct was dissolved under nitrogen in dichloro-
methane, the solution was filtered through a pad of silica gel, and
the solvent was removed in vacuo. The remaining black residue was
recrystallized from dichloromethane/methanol: 310 mg (65%) of
dark brown/black needles, decomp. temp. Ͼ300 °C. 1H NMR
(400 MHz, CDCl3): δ = 3.48 (s, 4 H, CH2CH2), 7.31–7.33 (m, 2 H,
Ar-H), 7.59–7.61 pm (m, 2 H, Ar-H) ppm. 13C NMR (100 MHz,
CDCl3): δ = 38.44 (t, CH2CH2), 89.90 (s, C-2), 96.24 (s, C-1),
128.88 (d, Ar-C), 134.57 (d, Ar-C), 136.15 (s, Ar-C), 199.37 (s, CO)
6-H), 7.54 (m, Jo = 8.3, Jm = 1.5 Hz, 1 H, 5-H), 7.82 (d, Jo
=
7.9 Hz, 1 H, 7-H), 8.24 (d, Jo = 8.3 Hz, 1 H, 4-H) ppm. 13C NMR
(100 MHz, CDCl3): δ = 28.23 (t, C-1), 28.36 (t, C-2), 119.72 (d, C-
4), 123.14 (s, C-9), 123.56 (d, C-7), 125.25 (d, C-8), 125.51 (d, C-
6), 128.21 (d, C-5), 130.62 (s, C-12), 135.28 (s, C-10), 141.56 (s, C-
3), 144.09 (s, C-11) ppm. IR (KBr): ν = 2970 (m), 2929 (s), 1417
˜
(m), 1249 (s), 938 (s), 871 (s), 849 (s), 756 (s) cm–1. UV/Vis (acetoni-
trile): λmax (logε) = 228 (4.89), 274 (3.79), 284 (3.87), 294 nm (3.75).
MS (70 eV): m/z (%) = 190 (33) [M+, 37Cl], 188 (100) [M+, 35Cl],
154 (10), 153 (83), 152 (68), 151 (22), 150 (12), 76 (25). HRMS:
C12H9Cl: calcd. 188.0392; found 188.039.
7. Oxmercuration/Hydration of 1. 3-Methoxy-1,2-dihydrocyclo-
penta[a]indene (28) and (3-Methoxy-1,2-dihydrocyclopenta[a]inden-
8-yl)mercury Chloride (27): A mixture of HgO (40 mg), tri-
chloroacetic acid (10 mg) and BF3·Et2O (0.1 mL) was heated at
60 °C for several min and then added to a solution of 1 (100 mg,
0.66 mmol) in methanol (50 mL) under nitrogen. The mixture was
stirred for 1 h at 20 °C, and water (200 mL) was added. After thor-
ough extraction with diethyl ether the combined organic phases
were passed through a pad of silica gel, dried (sodium sulfate), and
the solvent was removed by rotary evaporation. The residue thus
obtained was separated by silica gel column chromatography (cy-
clohexane/dichloromethane = 1:1). First fraction: 30.3 mg (25%)
ppm. IR (KBr): ν = 3043 (w), 2934 (w), 2082 (vs), 2007 (vs), 1431
˜
(w), 514 (s), 494 (s) cm–1. UV/Vis (acetonitrile): λmax (logε) = 200
(4.51), 222 (4.47), 232 nm (4.53). MS (70 eV): m/z (%) = 724 (12)
[M+], 696 (3), 668 (96), 640 (31), 612 (17), 584 (79), 556 (100),
528 (72), 500 (55), 472 (51), 444 (42), 416 (19), 388 (79), 329 (33).
C24H8Co4O12 (724.06): calcd. C 39.81, H 1.11; found C 38.81, H
0.99.
4. 1,2-Benzo-1,3,7-cyclooctatriene (9): To a solution of 1 (30 mg,
0.2 mmol) in anhydrous diethyl ether (50 mL) was added under ni-
trogen LiAlH4 (38 mg, 1 mmol), and the mixture was stirred for
1 h at room temperature. For work-up water (50 mL) was added,
and the product mixture was extracted carefully with diethyl ether
(4ϫ50 mL). After drying (sodium sulfate) the solvent was removed
by rotary evaporation and the residue purified by column
chromatography (silica gel, cyclohexane) to provide 29 mg (93%)
1
of 28, colorless needles, m.p. 56 °C. H NMR (400 MHz, CDCl3):
δ = 3.32–3.35 (m, 2 H, CH2), 3.58–3.61 (m, 2 H, CH2), 4.13 (s, 3
H, CH3O), 7.10 (s, 1 H, 8-H), 7.35–7.42 (m, 2 H, 5-, 6-H), 7.73 (d,
Jo = 7.4 Hz, 1 H, 7-H), 8.16 (d, Jo = 7.5 Hz, 1 H, 4-H) ppm. 13C
NMR (100 MHz, CDCl3): δ = 29.69 (t, CH2), 29.93 (t, CH2), 57.11
(q, CH3O), 113.32 (d, C-4), 121.56 (s, C-9), 122.03 (d, C-7), 123.74
(d, C-8), 125.06 (s, C-12), 125.18 (d, C-5), 127.32 (d, C-6), 135.22
1
of 9 as a colorless oil. H NMR (400 MHz, CDCl3): δ = 2.30–2.34
(m, 4 H, CH2CH2), 5.90–5.98 (m, 2 H, 4-H), 6.55 (d, 3J = 11.8 Hz,
2 H, 3-H), 7.10–7.14 (m, 2 H, Ar-H), 7.18–7.25 (m, 2 H, Ar-H)
ppm. 13C NMR (100 MHz, CDCl3): δ = 27.03 (t, CH2CH2), 126.49
(d, C-9, –12), 129.57 (d, C-10, –11), 129.87 (d, C-2, –5), 132.57 (d,
C-1, –6), 135.92 (s, C-3, –4) ppm. MS (70 eV): m/z (%) = 156 (22)
[M+], 141 (21), 128 (100), 115 (38). The spectroscopic and analyti-
cal data agree with those reported in the literature.[9].
(s, C-10), 145.30 (s, C-11), 147.80 (s, C-3) ppm. IR (KBr): ν = 3005
˜
(w), 2923 (s), 2854 (m), 1591 (s), 1364 (s), 1330 (s), 1279 (s), 1110
(s), 835 (s), 739 (s) cm–1. UV/Vis (acetonitrile): λmax (logε) = 232
(4.57), 294 nm (3.61). MS (70 eV): m/z (%) = 184 (100) [M+], 169
(99), 153 (22), 141 (86), 115 (53). HRMS: C13H12O: calcd:
184.0888; found 184.088. Second fraction: 30.3 mg (11%) of 27,
colorless needles, m.p. 235 °C. 1H NMR (400 MHz, CDCl3): δ =
3.30–3.33 (m, 2 H, CH2), 3.51–3.89 (m, 2 H, CH2), 4.12 (s, 3 H,
CH3O), 7.40 (m, 1 H, 6-H), 7.46 (m, 1 H, 5-H), 7.71 (d, Jo = 7.8 Hz,
1 H, 7-H), 8.21 (m, 1 H, 4-H) ppm. 13C NMR (100 MHz, CDCl3):
δ = 29.02 (t, CH2), 31.08 (t, CH2), 57.24 (q, CH3O), 121.41 (s, C-
5. 3-Phenyl-1,2-dihydrocyclopenta[a]indene (24): To a solution of 1
(100 mg, 0.66 mmol) in anhydrous benzene (100 mL) was added
under nitrogen BF3·Et2O (0.2 mL) and the mixture kept at room
temperature for 30 min. The mixture was hydrolyzed with hydrogen
carbonate solution (100 mL), extracted carefully with dichloro-
4032
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 4027–4034

The Chemical Behavior of 3,4-Benzocycloocten-1,5-diyne
9), 123.04 (d, C-7), 124.50 (d, C-4), 126.40 (s, C-5), 126.41 (d, C-
10), 129.31 (d, C-6), 138.59 (s, C-11), 151.47 (s, C-3) ppm; the sig-
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
nals for C-8 and C-12 could not be detected. IR (KBr): ν = 2936
˜
(w), 1602 (m), 1333 (s), 1270 (s), 1110 (s), 754 (s) cm–1. UV/Vis
Acknowledgments
(acetonitrile): λmax (logε) = 192 (4.47), 2.36 (4.76), 300 nm (4.01).
MS (70 eV): m/z (%) = 422 (25) [M+
– –
37Cl], 420 (67) [M+ 35Cl],
We thank Prof. Dr. Ludger Ernst (Braunschweig) for helpful dis-
cussions concerning the NMR spectra of the new compounds and
Prof. Dr. H. Meier (Mainz) for the spectroscopic data of 1,2-dihy-
dropentalene.
203 (12), 183 (44), 168 (82), 139 (100). HRMS: = C13H11ClHgO:
calcd. 420.01964; found 420.0196. C13H11ClHgO (419.27): calcd. C
37.24, H 2.64; found C 37.05, H 2.55.
8. Iodine Addition to 1. 8-Iodo-1,2-dihydrocyclopenta[a]indene (36)
and 3-Ethoxy-8-iodo-1,2-dihydrocyclopenta[a]indene (35): To a solu-
tion of 1 (30 mg, 0.197 mmol) in anhydrous ether (50 mL) was
added whilst stirring under N2 at 0 °C a solution of iodine (50 mg,
0.197 mmol) in diethyl ether (10 mL). After 1 h the solvent was
removed by rotary evaporation and the product mixture separated
by silica gel column chromatography (cyclohexane). Fraction 1:
20 mg (36%) of 36, brown-violet solid, m.p 50–60 °C (decomp.).
1H NMR (400 MHz, CDCl3): δ = 2.68–2.70 (m, 2 H, CH2), 3.10–
[1] Part 25: H. Hopf, H. Maurer, Eur. J. Org. Chem. 2005, 2702–
2707; for part 24 see: A. Krüger, H. Hopf, Chem. Eur. J. 2001,
7, 4378–4385.
[2] C. Werner, I. Dix, H. Hopf, P. Bubenitschek, P. G. Jones, Chem.
Eur. J. 2007, 13, 9462–9477.
[3] C. Moeller, M. S. Plesset, Phys. Rev. 1934, 46, 618–622.
[4] R. A. Kendall, T. H. Dunning Jr., R. J. Harrison, J. Chem.
Phys. 1992, 96, 6796–6806.
3.13 (m, 2 H, CH2), 6.82 (t, 1 H, 3-H), 7.15–7.19 (m, 2 H, 5-, 6- [5] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, J. A. Montgomery, Jr., T.
Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar,
J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N.
Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y.
Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P.
Hratchian, J. B. Cross, C. Adamo, J. Jaramillo, R. Gomperts,
R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pom-
elli, J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P.
Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich,
A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D.
Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui,
A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu,
A. Liashenko, P. Piskorz, I. Komaromi, R. L. Martin, D. J.
Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara,
M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W.
Wong, C. Gonzalez, J. A. Pople, Gaussian 03, Revision B.02,
Gaussian, Inc., Pittsburgh PA, 2003.
H), 7.34 (m, Jo = 7.6 Hz, 1 H, 7-H), 7.52 (m, Jo = 7.3 Hz, 1 H, 4-
H) ppm. 13C NMR (100 MHz, CDCl3): δ = 24.77 (t, CH2), 38.01
(t, CH2), 77.87 (s, C-8), 120.65 (d, C-3), 121.49 (d, C-4), 124.35 (d,
C-7), 128.14 (d, C-5), 129.28 (s, C-10), 134.89 (d, C-6), 149.01 (s,
C-11), 150.26 (s, C-12), 160.23 (s, C-9) ppm. IR (KBr): ν = 3059
˜
(w), 2914 (w), 1484 (s), 1444 (s), 1339 (s), 1216 (s), 754 (s) cm–1.
UV/Vis (acetonitrile, qualitative): λmax = 198, 236, 242, 266, 290,
362 nm. MS (70 eV): m/z (%) = 280 (8) [M+], 153 (18), 152 (15),
126 (18). HRMS: C12H9I: calcd. 279.9749; found 279.974. Fraction
2: 12 mg (19%) of 35, brown-violet crystals; m.p. 50–60 °C (de-
comp. point). 1H NMR (400 MHz, CDCl3): δ = 1.48 (t, 3J =
7.2 Hz, 3 H, CH3), 3.15–3.18 (m, 2 H, CH2), 3.39 (m, 2 H, CH2),
4.36 (q, 3J = 7.2 Hz, 2 H, CH2CH3), 7.41 (m, Jo = 8.3, Jm = 1.0 Hz,
1 H, 7-H), 7.50 (m, Jo = 7.4, Jm = 1.0 Hz, 4-H), 7.96 (m, Jo
=
8.3 Hz, 1 H, 6-H), 8.18 (m, Jo = 7.4, Jm = 1.0 Hz, 1 H, 5-H) ppm.
13C NMR (100 MHz, CDCl3): δ = 15.35 (q, CH3), 27.55 (t, CH2),
32.63 (t, CH2), 65.14 (t, CH2CH3), 80.48 (s, C-8), 121.83 (s, C-9),
122.75 (d, C-4), 124.45 (d, C-7), 126.51 (s, C-10), 126.75 (d, C-5),
129.62 (d, C-6), 134.53 (s, C-12), 147.81 (s, C-11), 151.54 (s, C-3)
[6] K. Brandhorst, J. Grunenberg, Chem. Soc. Rev. 2008, 37, 1558–
1567.
[7] H. Hopf, Angew. Chem. 1972, 84, 471; Angew. Chem. Int. Ed.
Engl. 1972, 11, 419; cf.H. Hopf, G. Schön, Liebigs Ann. Chem.
1981, 165–180.
[8] D. H. Bradley, M. A. Khan, K. M. Nicholas, Organometallics
1992, 11, 2598–2607; cf.P. Magnus, T. Pitterna, J. Chem. Soc.,
Chem. Commun. 1991, 541–543.
[9] M. E. Günther, R. Aydin, W. Buchmeier, B. Engelen, H.
Günther, Chem. Ber. 1984, 117, 1069–1076.
ppm. IR (KBr): ν = 3062 (w), 2923 (m), 1579 (s), 1418 (s), 1413
˜
(s), 1337 (s), 1270 (s), 1111 (s), 757 (s) cm–1. UV/Vis (acetonitrile):
λmax (logε) = 220 (4.56), 236 (4.60), 304 nm (4.02). MS (70 eV): m/z
(%) = 324 (100) [M+], 295 (35), 254 (10), 168 (32), 139 (32). HRMS:
C14H13IO: calcd. 324.0011; found 324.001.
9. Crystal Structure Determination: The data were recorded on a
Siemens P4 diffractometer at –100 °C using Mo-Kα radiation. The
structure was refined anisotropically on F2.[20] Hydrogen atoms
were included using a riding model. Two significant areas of resid-
ual electron density were tentatively interpreted as disordered sol-
vent, probably dichloromethane, but could not be refined satisfac-
torily. The program SQUEEZE[21] was therefore used to remove
mathematically the effects of the solvent. Because of the approxi-
mations involved, molecular dimensions should be interpreted with
caution. Numerical values such as density are based on an idealised
solvent content of two dichloromethane per asymmetric unit.
[10] R. A. Raphael, Acetylenic Compounds in Organic Synthesis,
Butterworths, London, 1955, pp. 29.
[11] We thank Prof. Dr. R. Thummel (University of Houston) for
making these spectra available to us; cf.: R. P. Thummel, W. E.
Cravey, W. Nutakul, J. Org. Chem. 1978, 43, 2473–2477.
[12] R. Kaiser, K. Hafner, Angew. Chem. 1970, 82, 877–878; Angew.
Chem. Int. Ed. Engl. 1970, 9, 892–893.
[13] L. Eisenhuth, H. Hopf, J. Am. Chem. Soc. 1974, 96, 5667–
5668.
[14] N. Hanold, Th. Molz, H. Meier, Angew. Chem. 1982, 94, 921;
Angew. Chem. Int. Ed. Engl. 1982, 21, 917.
[15] H. Meier, A. Pauli, P. Kochhan, Synthesis 1987, 573–574.
[16] Autorenkollektiv, Organikum, 20th ed., Wiley-VCH, Weinheim,
1999, p. 286.
Crystal Data for 6·2CH2Cl2: Monoclinic, space group P21/n, a =
12.361(2), b = 21.422(4), c = 21.788(3) Å, β = 97.168(14)°, Z =
4, crystal size 0.8ϫ 0.6ϫ 0.4 mm, 13719 reflections to 2θmax 50°;
refinement to wR2 = 0.129, R1 = 0.054 for 613 parameters and
10044 unique reflections; max. ∆ρ = 0.25 eÅ–3, S = 0.98.
[17] D. Ramkumar, M. Kalpana, B. Varghese, S. Sankararaman, J.
Org. Chem. 1996, 61, 2247–2250.
[18] X. Gu, M. B. Sponsler, Tetrahedron Lett. 1996, 37, 1571–1574.
For transannular cyclizations in medium-sized alicyclic diynes
and tetraynes the reader is referred to the extensive work of
Gleiter et al., for example: a) J. Ritter, R. Gleiter, Liebigs Ann.
Chem./Recueil 1997, 1179–1188; b) E. M. Schmidt, R. Gleiter,
CCDC-752109 (for 6) contains the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
Eur. J. Org. Chem. 2010, 4027–4034
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4033

C. Werner, H. Hopf, J. Grunenberg, P. G. Jones
FULL PAPER
F. Rominger, Chem. Eur. J. 2003, 9, 1814–1822; c) B. Esser,
R. Gleiter, J. Org. Chem. 2006, 71, 5126–5135, and ref. cited
therein.
1986, 98, 646–648; Angew. Chem. Int. Ed. Engl. 1986, 25, 630–
632.
[20]
G. M. Sheldrick, Acta Crystallogr., Sect. A 2008, 64, 112–122.
[21] A. L. Spek, University of Utrecht, The Netherlands.
Received: January 8, 2010
[19] Q. Zhou, P. J. Carroll, T. M. Swager, J. Org. Chem. 1994, 59,
1294–1301 and refs. reported therein. For the preparation of
other stable s-indacenes see K. Hafner, B. Stowasser, H.-P.
Krimmer, S. Fischer, M. C. Böhm, H. J. Lindner, Angew. Chem.
Published Online: June 11, 2010
4034
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 4027–4034