Efforts toward distorted spiropentanes

Article abstract of DOI:10.1021/jo1017298

Tetravinylbenzene 4 was prepared in nearly quantitative yield from commercially available tetrabromobenzene; the improved, one-step procedure now employs Suzuki-Miyaura cross-coupling conditions. Intermolecular cyclopropanation of 4 with dibromocarbene gave a series of gem-dibromide adducts. Intramolecular cyclopropanation of monoadduct 5, putatively by its methyllithium-generated cyclopropylidene(oid), produced compound 11, which features a highly distorted spiropentane having a C-C-C bond angle of 163.5°. The stability of the reported spiropentanes was investigated using DFT calculations.

Full text of DOI:10.1021/jo1017298

pubs.acs.org/joc
Efforts toward Distorted Spiropentanes†,‡
Kuan-Jen Su,§ Jean-Luc Mieusset,§ Vladimir B. Arion,
Wolfgang Knoll,§ Lothar Brecker,§ and Udo H. Brinker*,§
§Chair of Physical Organic and Structural Chemistry,
Institute of Organic Chemistry, University of Vienna,
FIGURE 1. Calculated bond lengths in picometers (B3LYP/6-
31G(d)) of 1, 2, and 3 (aX-ray data12).
€
Wahringer Strasse 38, A-1090 Vienna, Austria, and Institute
€
of Inorganic Chemistry, University of Vienna, Wahringer
Strasse 42, A-1090 Vienna, Austria
addition of a cyclopropylidene(oid)4 to a double bond.5
Thus, we were encouraged by this strategic milestone, which
generates the corresponding carbenoid intermediate6 by
reaction of its gem-dibromocyclopropane precursor with
methyllithium. Besides its elegance, another incentive for
studying olefin 1 resides in its nearly collinear arrangement
of three carbon atoms,7 which includes the spiro atom of
the distorted spiropentane segment.8 Furthermore, various
rearrangements of unstable 1 to lower-energy isomers were
observed.1 However, as noted, all experimental attempts to
isolate 1 have failed,1 as corroborated by computational
evidence.9 Nevertheless, Brinker and Streu did successfully
prepare 4,5-benzotricyclo[4.1.0.01,3]hept-4-ene (2),10 which
retains the two sp2 carbon centers of the double bond of
1 (Figure 1). Delocalization of the juxtaposing alkene func-
tion within the aromatic ring stabilizes the tricycloheptene
moiety. Hence, 2 is isolable at room temperature, although
it quantitatively isomerizes at 30 °C with a measurable rate
constant.11
udo.brinker@univie.ac.at
Received September 2, 2010
Tetravinylbenzene 4 was prepared in nearly quantitative
yield from commercially available tetrabromobenzene;
the improved, one-step procedure now employs Suzuki-
Miyaura cross-coupling conditions. Intermolecular cyclo-
propanation of 4 with dibromocarbene gave a series of
gem-dibromide adducts. Intramolecular cyclopropanation
of monoadduct 5, putatively by its methyllithium-gener-
ated cyclopropylidene(oid), produced compound 11, which
features a highly distorted spiropentane having a C-C-C
bond angle of 163.5°. The stability of the reported spiro-
pentanes was investigated using DFT calculations.
In order to study the influence of the spiropentane moiety
on the bond lengths of the attached benzene ring, a single
crystal X-ray structure determination of 2 was performed.12
The annulated bond of the benzene ring was found to be
˚
elongated to 1.414 A, and the distances of the other C-C
ꢀ
bonds tended toward one Kekule structure, i.e., 2 (Figure 1).
We wanted to determine the consequences of placing
a second spiropentane moiety on the opposite side of the
benzene ring of 2. In addition, we set out to investigate the
influence of the geometry on the stability and bonding
characteristics of this molecule and possibly its rearrange-
ment behavior. Therefore, we first embarked on the synthesis
of the aesthetically appealing bisspiropentane 3, which can
exhibit meso, d, and l forms (Figure 1).13 We also explored
theoretical aspects of 3, using computations correlated with
experimental results.
Tricyclo[4.1.0.01,3]hept-4-ene (1),1 like other small bridged
alicyclic molecules,2 has aesthetic appeal (Figure 1), but its
alluring beauty has eluded synthesis due to its high strain
energy.1 Regardless, tricyclo[4.1.0.01,3]heptane has been pre-
pared by Skattebøl,3 who reported the first intramolecular
(5) Backes, J.; Brinker, U. H. In Houben-Weyl, Methoden der Orga-
nischen Chemie; Regitz, M., Ed.; Thieme: Stuttgart, 1989; Vol. 19b, 391-510.
(6) Jones, M., Jr.; Moss, R. A. In Reactive Intermediate Chemistry; Moss,
R. A., Platz, M. S., Jones, M., Jr., Eds.; Wiley: Hoboken, 2004; pp 273-328.
† Dedicated to my distinguished colleague, Professor John J. Eisch, Department of
Chemistry, SUNY-Binghamton, USA, on the occasion of his 80th birthday (UHB).
‡ Carbene Rearrangements 81. For part 80, see: Mieusset, J.-L.; Brinker, U. H.
In Molecular Encapsulation: Organic Reactions in Constrained Systems; Brinker,
U. H., Mieusset, J.-L., Eds.; Wiley: Chichester, 2010; pp 269-308.
(1) (a) Brinker, U. H.; Gomann, K.; Zorn, R. Angew. Chem., Int. Ed.
Engl. 1983, 22, 869. (b) Brinker, U. H.; Gomann, K.; Zorn, R. Angew. Chem.
Suppl. 1983, 1241 and references therein.
(2) (a) Wiberg, K. B. J. Am. Chem. Soc. 1983, 105, 1227. (b) de Meijere, A.
In Houben-Weyl, Methods of Organic Chemistry; Thieme: Stuttgart, 1996;
Vol. E17a-d, pp 1-3635. (c) Takeuchi, K.; Horiguchi, A.; Inagaki, S.
Tetrahedron 2005, 61, 2601 and references therein. (d) Minyaev, R. M.;
Minkin, V. I. Russ. J. Gen. Chem. 2008, 78, 732. (e) Keese, R. Chem. Rev.
2006, 106, 4787.
ꢀ
ꢀ
(7) Dodziuk, H.; Leszczynski, J.; Nowinski, K. S. J. Org. Chem. 1995, 60,
6860.
(8) Kuznetsova, T. S.; Eremenko, O. V.; Kokoreva, O. V.; Zatonsky,
G. V.; Zefirov, N. S. Russ. Chem. Bull. 1996, 45, 1662.
(9) Wiberg, K. B.; Snoonian, J. R. J. Org. Chem. 1998, 63, 1390 and
references therein.
(10) Brinker, U. H.; Streu, J. Angew. Chem., Int. Ed. Engl. 1980, 19, 631.
(11) (a) Brinker, U. H.; Wilk, G.; Gomann, K. Angew. Chem., Int. Ed.
Engl. 1983, 22, 868. (b) Brinker, U. H.; Wilk, G.; Gomann, K. Angew. Chem.
Suppl. 1983, 1228.
€
(12) Boese, R.; Blaser, D.; Gomann, K.; Brinker, U. H. J. Am. Chem. Soc.
(3) Skattebøl, L. Chem. Ind. (London) 1962, 2146.
(4) Azizoglu, A.; Balci, M.; Mieusset, J.-L.; Brinker, U. H. J. Org. Chem.
2008, 73, 8182.
1989, 111, 1501.
(13) For reasons of simplicity, only one of the possible stereoisomers is
shown in the Abstract and Figure 1.
7494 J. Org. Chem. 2010, 75, 7494–7497
Published on Web 10/12/2010
DOI: 10.1021/jo1017298
r
2010 American Chemical Society

Su et al.
JOCNote
SCHEME 1. Procedures for the Synthesis of 1,2,4,5-Tetravi-
nylbenzene (4)
SCHEME 3
aTrivinylboroxane complex, Pd(OAc)2, PPh3, K2CO3, DME/H2O,
100 °C, 24 h.
SCHEME 2. Dibromocarbene Addition to 4
aNot observed. bWhen using fresh silica gel.
TABLE 1. Product Distributionsa from Dibromocarbene Reactions
with 1-6 equiv of Bromoform
compound 1 equiv 2 equiv 3 equiv 4 equiv 5 equiv 6 equiv
4
5
6
84
11
1
39
18
2
22
17
3
9
18
5
17
15
3
2
8
4
7
1
1
1
3
1
4
8
2
3
4
6
4
5
9
10
1
0
2
35
5
48
9
50
4
56
8
69
aDetermined by 1H NMR peak integration of crude mixtures.
aFrom t-BuOk and HCBr3
The conversion requires heating of the reactants at 100 °C
in DME/H2O for 24 h in an autoclave apparatus.
We surmised that 1,2,4,5-tetravinylbenzene (4)14 would
be an appropriate starting material for the construction of
two spiropentane subunits (Scheme 1), because the ana-
logous o-benzeno-bridged spiropentane 2 was synthesized
from o-divinylbenzene.10 We envisioned a 2-fold intramo-
lecular cyclopropylidene(oid) addition to the proximal double
bonds of the intermediary bisadduct(s) 6 and/or 7 (Scheme 2)
to furnish bisspiropentane 3 (Scheme 3).
The established procedure for 4 begins with 1,2,4,5-tetra-
methylbenzene and affords the compound in four steps
with an overall yield of 12-15%14a (Scheme 1). However,
we developed a more efficient synthetic route, exploiting
Suzuki-Miyaura cross-coupling,15 that yields a larger sup-
ply of 4 in just one step. Substitution of the four bromine
atoms in 1,2,4,5-tetrabromobenzene by vinyl groups was
readily achieved, and optimization of the reaction conditions
led to tetravinylbenzene 4 with an almost quantitative yield.
Regrettably, the subsequent dibromocarbene addition16
lacks selectivity, and for a series of dienes,17 indiscriminate
additions were reported. Depending on the molar ratio of the
generated dibromocarbene, the reaction with 4 (Scheme 2)
gave a complex mixture of starting material and mono-, bis-,
tris-, and tetrakisadducts (Table 1).
Chromatographic separation of 5-10 allowed the hitherto
unknown dibromocarbene adducts to be characterized. The
separation of isomers 6 and 7,18 which have nearly identical
retention times, was achieved by repeated chromatography.
Single crystals of bisadducts 7 and 8 were obtained and
analyzed by X-ray diffraction. Crystals of two stereoisomers
of tetrakisadduct 10 were suitable for X-ray analysis as well.
In contrast, no crystals suitable for X-ray analysis could be
obtained for monoadduct 5 and trisadduct 9.
To investigate the dependence of product distribution on
the molar ratio of tetravinylbenzene 4 and the dibromocar-
bene precursor, e.g., bromoform, a test series was run. Six
reaction compositions starting with 1-6 equiv of bromo-
form were analyzed. The relative product distributions of the
(14) (a) Schrievers, T.; Brinker, U. H. Synthesis 1988, 330. (b) Nakamura,
Y.; Haashida, Y.; Wada, Y.; Nishimura, J. Tetrahedron 1997, 53, 4593.
(15) (a) Miyaura, N.; Yamada, K.; Suzuki, A. Tetrahedron Lett. 1979,
3437. (b) Miyaura, N.; Suzuki, A. Chem. Commun. 1979, 866. (c) Suzuki, A.
Pure Appl. Chem. 1991, 63, 419. (d) Miyaura, N.; Suzuki, A. Chem. Rev. 1995,
95, 2457. (e) Suzuki, A. J. Organomet. Chem. 1999, 576, 147. (f) Mantel,
M. L. H.; Søbjerg, L. S.; Huynh, T. H. V.; Ebran, J. P.; Lindhardt, A. T.;
Nielsen, N. C.; Skrydstrup, T. J. Org. Chem. 2008, 73, 3570.
1
samples were determined by peak area integrations of H
NMR spectra of each crude reaction mixture directly after
workup. As expected, six dibromocarbene adducts (and their
(16) (a) Dehmlow, E. V. In Houben-Weyl, Methoden der Organischen
Chemie; Regitz, M., Ed.; Thieme: Stuttgart, 1989; Vol. 19b, p 1461. (b)
Doering, W. v. E.; Hoffmann, A. K. J. Am. Chem. Soc. 1954, 76, 6162. (c) Su,
K.-J.; Mieusset, J.-L.; Arion, V. B.; Brecker, L.; Brinker, U. H. Org. Lett.
2007, 9, 113.
(17) (a) Skattebøl, L. J. Org. Chem. 1964, 29, 2951. (b) Brinker, U. H.;
Fleischhauer, I. Angew. Chem., Int. Ed. Engl. 1979, 18, 396.
(18) The methine protons of the cyclopropane rings of both isomers are
positioned either on the same or on different sides.
J. Org. Chem. Vol. 75, No. 21, 2010 7495

Su et al.
JOCNote
stereoisomers), including the 4-fold addition product 10,
were obtained plus some unreacted 4 (Scheme 2). Table 1
shows the dependence of the product ratios in relation to the
equivalents of bromoform used. However, it should be taken
into consideration that the effective in situ generated dibro-
mocarbene concentration may be lower than the actual
bromoform quantity used.
SCHEME 4. Experimental Ratio10 and Results of DFT Com-
putations for the Synthesis of Spiropentane 2 and Its Rearrange-
ment to 20
To accumulate monoadduct 5 and bisadducts 6 and 7 in
one experiment, the use of 4 equiv of HCBr3 was optimal.
The decisive step in constructing the spiropentane subunit
is an intramolecular cyclopropanation of the vinyl group.
This was achieved by adding methyllithium to monoadduct
5 at -78 °C and allowing the temperature to rise overnight to
9 °C (Scheme 3).19 This procedure did indeed yield the expected
spiropentane 11 and allene 12 in a ratio of 70:30 after chroma-
tography. Using NMR, the half-lives were determined to be
nearly the same (t1/2 = ca. 24 h at room temperature). The low
isolated yield, i.e., 19%, of the isomeric mixture of 11 and 12
may be due to the known instability of styrenes, which tend to
polymerize.
and the cyclopropylidene addition to give spiropentane for the
methyllithium reaction of bisadducts 6 and 7. For two con-
secutive reactions of the aforementioned type, the statistic
predicts a product distribution of allenes 14 þ 15 to mono-
spiropentane 13 to bisspiropentanes 3 in a ratio of 72:26:2.
Combining these assumptions with the experimentally ob-
tained total isolated yield of 15% for the product mixture
from the reaction of 6 and 7, compound 3 would be produced
in a low yield of 0.3% (15% ꢀ 2% = 0.3%).
Accordingly, treatment of either bisadduct 6 or 7 with
methyllithium, under the same conditions applied for 5, should
lead to a 2-fold intramolecular cyclopropylidene addition
giving bisspiropentane 3. Thus, a mixture of 6 and 7 could be
used for further reaction (Scheme 3). The crude product result-
ing from the cyclopropanation reaction of mixed 6 and 7
However, the synthesis of 3 might be viable using transi-
tion-metal catalysis, thereby minimizing the occurrence of
competing cyclopropylidene-allene rearrangements. This en-
tails that extrusion of the second bromine nucleofuge should
be induced by nucleophilic attack of the nearby vinyl group.
Next, a closer inspection of the effect of benzene ring
bifunctionalization by two spiropentylene groups was under-
taken.12 Calculations for bisspiro compound 3 gave essentially
the same results as those for monospiro compound 2. Com-
pared with 3, only one barrier is slightly lower in energy than
in 2. It involves the decomposition by a rDA that is energeti-
cally less favorable for 2 than for 3 [TS(2/20) 124.4 kJ/mol vs
120.7 kJ/mol]. Consequently, 3 is expected to be kinetically less
stable than 2. An explanation can be found in the bond order21
of the C-C bonds of the aromatic ring. In contrast to 2, the
alternation of shortened and lengthened C-C bonds in 3 is
rather suitable for the generation of the central cyclohexadiene
subunit of the rDA product (Figure 1). exo-Methylene 20 is
thermodynamically unstable and partially rearranges to meth-
ylnaphthalene.11 Finally, in all of the synthesized spiropen-
tanes, the augmentation of the C-C-C bond angle remains
similar [2, 163.6° (164.0° experimentally12); 3, 163.6°; 11, 163.5°;
16, 163.2°].
1
showed, however, mostly broad H NMR signals that are
likely due to polymers. Chromatography on silica gel afforded
a 15% isolated yield of an isomeric mixture of spiroallene 13
and bisallenes 14 and 15. The 13:14:15 product distribution
was determined by 1H peak integration to be 68:23:9, and their
structures were determined by 2D NMR techniques. Curi-
ously, no evidence for the desired bisspiropentane 3 was found
under the conditions applied. We also noted, when using fresh
silica gel, that methylnaphthalene 16 was found in the mix-
ture. It can arise from an electrocyclization of 13 followed by
rearomatization via a 1,3-H shift. Alternatively, a retro-Diels-
Alder reaction (rDA) of one tricyclo[4.1.0.01,3]hept-4-ene sub-
unit would also lead to 16 (vide infra).
The feasibility of the aforementioned spirocyclizations
was assessed by ab initio molecular orbital calculations.
Electron-correlated single-point energies were computed
using the hybrid DFT method B3LYP and a split-valence
6-31G(d) basis set.20 We first performed computations using
2-(2-ethenylphenyl)cyclopropylidene (18) as an unsubsti-
tuted model compound and then compared the results with
those from experimental data (Scheme 4).10 Carbene 18 is
predicted to form spiropentane 2 by energetically overcom-
ing a low activation barrier of 16.6 kJ/mol, but rearrange-
ment of 18 to allene 19 is expected to be even faster, i.e., Ea =
14.1 kJ/mol. Indeed, these values are compatible with the
observed ratio of 2:19 at 25 °C, i.e., 15:85.10
In conclusion, the highly distorted spiropentane 11 was
prepared in only three steps from commercially available
1,2,4,5-tetrabromobenzene. Compound 11 possesses an al-
most linear arrangement of three C atoms with a C-C-C
bond angle of 163.5° about the spiro atom. Thus far, bisspiro
compound 3 has not been obtained. Statistical considera-
tions and theoretical calculations provided insight into the
difficulties of the formation and instability of 3. The use of
carbenoid reactions induced by transition-metal catalysis
may help to prevent the competitive formation of allenes.
And finally, an efficient procedure giving 4 in nearly quanti-
tative yield, i.e., 99.7%, is introduced.
Statistical considerations reveal that this adverse ratio of
spirocycle to allene offers only a small opportunity for the
construction of a second spiropentane to form bisspiropentane
3. Therefore, we presumed the same product distributions
from the competitive cyclopropylidene-allene rearrangement
Experimental Section
1,2,4,5-Tetravinylbenzene (4). 1,2,4,5-Tetrabromobenzene
(776 mg, 2 mmol), Pd(OAc)2 (89 mg, 0.4 mmol), PPh3 (270 mg,
(19) The intramolecular addition reaction of the generated carbene(oid)
intermediate is temperature-dependent and competes with the cyclopropy-
lidene-allene rearrangement; a lower temperature range favors production
of the spirocyclic compound.
(20) The energy values given represent E þ ZPVE.
(21) Stanger, A. J. Am. Chem. Soc. 1998, 120, 12034.
7496 J. Org. Chem. Vol. 75, No. 21, 2010

Su et al.
JOCNote
TABLE 2. Compositions for Dibromocarbene Reactions
was further stirred overnight (from -78 to 9 °C) and then
quenched with water. The reaction mixture was extracted with
dichloromethane (3 ꢀ 20 mL), and the combined organic phases
were washed with brine. After drying over MgSO4, the solvent
was removed in vacuo. By using column chromatography with
hexane as eluant, a mixture of spiro compound 11 and allene 12
was obtained (30 mg, 19% of 11 and 12). Ratio as determined by
1H NMR peak integration: 11:12 = 70%:30%. NMR signal
allocation for 11 and 12 was verified by 2D NMR techniques.
Data for 11 and 12: see the Supporting Information.
Reaction of Bisadducts 6 and 7 with Methyllithium. Bisadducts
6 and 7 (258 mg, 0.49 mmol) were dissolved with 100 mL of dry
ether and cooled in a dry ice-acetone bath at -78 °C. A fresh
ethereal solution of methyllithium (∼1.6 M, 1.55 mL, 5 equiv)
was added by syringe with stirring. The reaction mixture was
further stirred overnight (from -78 to 9 °C) and then quenched
with water. The reaction mixture was extracted with dichloro-
methane (3 ꢀ 20 mL), and the combined organic phases were
washed with brine. After drying over MgSO4, the solvent was
removed in vacuo. Column chromatographic separation with
hexane as eluant gave a mixture of the spiro compound 13 and
the corresponding allene isomers 14 and 15 (15 mg, 15% of 13,
14, and 15). Ratio determined by 1H NMR peak integration:
13:14:15 = 68:23:9. When using fresh, untreated silica gel for
chromatography, the rearranged product 16 was also obtained.
NMR signal allocation for 13-16 was verified by 2D NMR
techniques. Data of 13-16: see the Supporting Information.
equiv of HCBr3
A
B
C
1
2
3
4
5
6
195 mg
319 mg
338 mg
331 mg
203 mg
328 mg
90 mg (0.75 equiv)
294 mg (1.5 equiv)
468 mg (2.25 equiv)
611 mg (3 equiv)
468 mg (3.75 equiv)
908 mg (4.5 equiv)
0.93 mL (1 equiv)
3.03 mL (2 equiv)
4.82 mL (3 equiv)
6.29 mL (4 equiv)
4.82 mL (5 equiv)
9.35 mL (6 equiv)
1 mmol), K2CO3 (1100 mg, 8 mmol), and 2,4,6-trivinylcyclotri-
boroxane pyridine complex (1.465 g, 6 mmol) were added to an
autoclave apparatus with 1,2-dimethoxyethane (ethylene glycol
dimethyl ether) (35 mL) and H2O (10 mL) as solvents. The closed
autoclave apparatus was heated to 95-100 °C for 24 h and then
cooled to room temperature. The crude reaction mixture was
extracted with dichloromethane (3 ꢀ 20 mL), and the combined
organic phases were washed with brine. After drying over
MgSO4, the solvent was removed in vacuo. Bulb-to-bulb distilla-
tion of the crude product in vacuo (0.01 mmHg) at 75 °C afforded
the title compound as a white solid (during distillation cooled by
dry ice), which melts to a clear colorless liquid (361 mg, 99.7%).
Data for 4: see the Supporting Information.
Dibromocarbene Addition to 1,2,4,5-Tetravinylbenzene (4). To
a mixture of A mg of 1,2,4,5-tetravinylbenzene and B mg of
potassium tert-butoxide in ca. 150 mL of dichloromethane,
C mL of a dichloromethane/bromoform (1:9) solution was
added dropwise while cooling with an ice-salt bath (see Table 2).
Stirring was continued overnight, while the reaction mixture was
allowed to warm to room temperature. Next, ca. 200 mL of water
was added, the organic phase was separated, and the aqueous
phase was extracted with dichloromethane. The combined
organic phases were washed with water and dried over MgSO4.
The solvent was evaporated, and the compounds con-
tained in the crude mixture of 4-10 were separated by column
chromatography with hexane as an eluant. Data for 5-10: see the
Supporting Information.
€
Acknowledgment. We are indebted to Dr. Verena Tellstrom
(Bruker Daltonik, Bremen, Germany) and Dr. Carsten
Krantz (Thermo Fischer Scientific, Reinach, Switzerland)
for the HRMS characterization of compound 9. We thank
Alexander Roller (Institute of Inorganic Chemistry, University
of Vienna, Austria) for single crystal X-ray diffraction measure-
ꢁ
ments and Mirjana Pacar for experimental assistance.
Reaction of 1-(2,2-Dibromocyclopropyl)-2,4,5-trivinylbenzene
(5) with Methyllithium. 1-(2,2-Dibromocyclopropyl)-2,4,5-trivi-
nylbenzene (5, 282 mg, 0.8 mmol) was dissolved with 50 mL of
dry ether and cooled in a dry ice-acetone bath at -78 °C. A
fresh ethereal solution of methyllithium (∼1.6 M, 1.13 mL, 2.25
equiv) was added by syringe with stirring. The reaction mixture
Supporting Information Available: Spectral data of all new
compounds and crystallographic data in CIF format for 7, 8 and
two stereoisomers of 10 and computational data. This material
is available free of charge via the Internet at http://pubs.acs.org.
J. Org. Chem. Vol. 75, No. 21, 2010 7497