Synthesis of 1,2,3,4-tetramethyl- and 1,2,3,4-tetraethylfluorene through a Dewar benzene pathway

Article abstract of DOI:10.1002/ejoc.201201343

A simple and selective synthesis of 1,2,3,4-tetra(m)ethylfluorenones and -fluorenes was achieved by using Dewar benzenes as important intermediates. The corresponding Dewar benzenes were prepared by reactions of tetraalkylcyclobutadiene-AlCl₃ c

Full text of DOI:10.1002/ejoc.201201343

SHORT COMMUNICATION
DOI: 10.1002/ejoc.201201343
Synthesis of 1,2,3,4-Tetramethyl- and 1,2,3,4-Tetraethylfluorene through a
Dewar Benzene Pathway
[a]
[b]
[c]
[b]
ˇ
ˇ
ˇ
ˇ
Stepánka Janková, Jirí Schulz, Simona Hybelbauerová, Ivana Císarová,
[b]
[a]
ˇ
ˇ
ˇ
Petr Stepnicka, and Martin Kotora*
Keywords: Cycloaddition / Arenes / Rearrangement / Reduction / Fused-ring systems
A simple and selective synthesis of 1,2,3,4-tetra(m)ethyl-
fluorenones and -fluorenes was achieved by using Dewar
arrangement provided substituted biphenylcarboxylic acids.
Treatment of the prepared acids with thionyl chloride gave
benzenes as important intermediates. The corresponding De- rise to the fluorenones, which were reduced to the respective
war benzenes were prepared by reactions of tetraalkylcyclo-
butadiene–AlCl3 complexes with methyl phenylpropynoate.
Their subsequent hydrolysis followed by photochemical re-
fluorenes. A Ru complex with 1,2,3,4-tetraethylfluorenone
was prepared.
and other alkylated derivatives, only a procedure based on
the rearrangement of bis(indenyl)zirconacyclopentadienes
has been devised.[5] Interestingly, but perhaps not surpris-
ingly, the presence of 1,2,3,4-tetraalkylated fluorenes has
been confirmed in NatureϪthey were detected in aromatic
fractions of Athabasca oil sand bitumens.[6]
Recently, it was shown that highly substituted Dewar
benzenes could be used as convenient synthetic intermedi-
ates for the preparation of selectively substituted benz-
enes[7–9] and ter- and quaterphenyls alkylated at the ter-
minal benzene rings.[10] Given that one of the pathways for
their preparation utilizes the reaction of a tetralkylcyclobut-
adiene–aluminum chloride complex with activated alkynes
bearing at least one ester group, we envisioned that 1,2,3,4-
tetraalkylated fluorenes could be also prepared by a Dewar
benzene pathway by taking advantage of the ester function-
ality present in the molecule. Herein, we would like to re-
port a general and selective synthetic approach toward
1,2,3,4-tetramethyl- and 1,2,3,4-tetraethylfluorenones and
the corresponding fluorenes by this strategy.
Introduction
Fluorenes have been shown to coordinate to various
transition metal fragments to form either η5-cyclopen-
tadienyl or η6-arene complexes. In this respect, the synthesis
of such compounds bearing an unsymmetrically substituted
fluorene scaffold is of general interest, as it may allow elec-
tronic effects in transition-metal complexes and coordina-
tion preferences of various metal fragments to be studied.
Although several complexes with octamethyl- and nona-
methylfluorene ligands have been described,[1,2] to the best
of our knowledge, reports concerning the complexation of
1,2,3,4-tetraalkylfluorenes are not available in the literature.
Moreover, so far only a handful of methods have been re-
ported for the preparation of such ligands. For instance,
two methods for the synthesis of 1,2,3,4-tetramethylfluor-
ene (1a) have been described: one is based on the Lewis acid
catalyzed alkylation of pristine fluorene, but this reaction
is unselective and gives rise to numerous regioisomers (the
highest yield of 1a was 21%),[3] whereas the other method
makes use of a palladium-catalyzed cross-coupling/C–H ac-
tivation processes (93% yield of the isolated product from
the reaction of pentamethylphenylmagnesium chloride with
1,2-dibromobenzene).[4] In the case of tetraethylfluorene
Results and Discussion
[a] Department of Organic Chemistry, Faculty of Science, Charles
University in Prague,
Hlavova 8, 128 43 Praha 2, Czech Republic
Fax: +420-221951324
At the outset, the starting Dewar benzenes were synthe-
sized according to
a previously reported procedure
E-mail: kotora@natur.cuni.cz
(Scheme 1).[11] Tetramethylcyclobutadiene- and tetraethyl-
cyclobutadiene–aluminum chloride adducts 2a and 2b,[12]
respectively, generated in situ from the respective internal
alkyne and anhydrous AlCl3, were treated with methyl
phenylpropynoate (3) in dichloromethane at –15 °C to give
corresponding products 4a and 4b in 54 and 38% isolated
yield, respectively.
Homepage: http://www.natur.cuni.cz/chemie/orgchem/kotora
[b] Department of Inorganic Chemistry, Faculty of Science,
Charles University in Prague,
Hlavova 8, 128 43 Praha 2, Czech Republic
[c] Department of Teaching and Didactics of Chemistry, Faculty
of Science, Charles University in Prague,
Albertov 3, 128 43 Praha 2, Czech Republic
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201201343.
44
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 44–47

Synthesis of 1,2,3,4-Tetramethyl- and 1,2,3,4-Tetraethylfluorene
ferent strategy was applied. Dewar benzenes 4a and 4b were
used again as the starting materials, but they were first hy-
drolyzed to Dewar benzene carboxylic acids 8a and 8b by
using NaOH in methanol (65 °C, 1 week). The long reac-
tion times that were typically required[13] probably reflect
the steric hindrance of the ester substrates; the use of higher
reaction temperatures to increase the reaction rate may lead
to undesirable rearrangement of the Dewar benzene scaf-
fold to the benzene one. Acids 8a and 8b were isolated in
45 and 34% yield, respectively. Given that conventional
thermal rearrangement of carboxylic acid 8a in DMSO at
120 °C[14] took 1 week to reach full conversion and given
that the corresponding benzoic acid was isolated in only
27% yield, the photochemical rearrangement of both acids
was attempted next (254 nm, in CDCl3).[15] Gratifyingly, the
photochemical conversion of acids 8 into 9 proceeded
quantitatively at 20 °C within 24 h. Simple evaporation of
the solvent followed by column chromatography on silica
gel yielded 9a and 9b in 56 and 90% yield, respectively (in
the case of the rearrangement of 8a, the unreacted starting
material was recovered). Although it was initially envi-
sioned that these benzoic acids would be converted stepwise
into the corresponding acid chlorides and then treated with
a Lewis acid to bring about intramolecular acylation, our
early experiments showed that the cyclization could be done
in one step. Eventually, after trying several reaction condi-
tions, it was found that heating benzoic acids 9a and 9b in
pure thionyl chloride at 75 °C resulted in smooth, spontane-
ous intramolecular acylation to give 1,2,3,4-tetralkylfluor-
enones 10a and 10b in high isolated yields (Scheme 3).
Thus, 1,2,3,4-tetramethyl-9H-fluoren-9-one (10a) was ob-
tained in 93% yield, whereas 1,2,3,4-tetraethyl-9H-fluoren-
9-one (10b) was obtained in 92% yield. Recrystallization of
the latter compound from hexane furnished crystals suitable
for X-ray analysis, which unequivocally corroborated its
structure (Figure 1). It showed alternating up–down ethyl
groups analogically to hexaethylbenzene.[16]
Scheme 1. Preparation of Dewar benzenes 4.
Compound 4a was expected to be converted into 1,2,3,4-
tetramethylfluorene (1a) in a straightforward manner
(Scheme 2). The prepared Dewar benzene was quantita-
tively thermally rearranged in DMSO (150 °C, 20 h) to bi-
phenyl 5, which was isolated in 67% yield. Subsequent re-
duction with LiAlH4 provided alcohol 6 in 49% isolated
yield. We assumed that alcohol 6 could be converted into
expected fluorene 1 upon intramolecular alkylation induced
by a Lewis acid. We tried to bring about the intramolecular
cyclization by addition of several Lewis acids, including
AlCl3, Bi(OTf)3, and BF3·OEt2. However, no reaction took
place in the presence of AlCl3 (c = 0.04 mmolL–1) and the
starting material was recovered unchanged. The use Bi-
(OTf)3 (c = 0.04 mmolL–1) resulted in the formation of an
inseparable mixture of products. Carrying out the reaction
under dilute conditions [c(6) = 0.004 molL–1] did not have
any positive effect on the course of the reaction. Finally, the
reaction of 6 with BF3·OEt2 also provided an inseparable
mixture of various products. In neither case was the pres-
ence of desired fluorene 1a detected in the crude mixture.
Instead, compound 7 was isolated, and its structure was
tentatively assigned on the basis of 1H NMR and 13C NMR
spectroscopy. It is the product of sequential intra- and inter-
molecular alkylations, and it was present in various
amounts along with higher alkylated products.
Scheme 2. Lewis acid mediated reaction of 6.
Scheme 3. Synthesis of fluorenones 10a and 10b.
An attempt to synthesize tetramethylfluorene by using
the previously reported procedure based on the rearrange-
In the last step, the carbonyl moiety in fluorenones 10a
ment of bis(indenyl)tetramethylzirconacyclopentadiene in and 10b was reduced to a methylene group to give 1,2,3,4-
the presence of TiCl4[5] was also unsuccessful; and this reac- tetramethylfluorene (1a) and 1,2,3,4-tetraethylfluorene (1b;
tion led to a complex mixture in which 1,2,3,4-tetrameth- Scheme 4). Attempts to use the classical Wolff–Kishner re-
ylfluorene (1a) was not detected. Because the approach out- duction or its later modification[18] for the reduction of 10a
lined above failed to provide the desired products, a dif- were unsuccessful. The reaction did not take place and the
Eur. J. Org. Chem. 2013, 44–47
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
45

M. Kotora et al.
SHORT COMMUNICATION
lower steric hindrance of the π ligands surrounding the RuII
ion, which may outweigh the anticipated preference for the
coordination to the more electron-rich alkylated benzene
ring.
Figure 1. PLATON[17] drawing of 1,2,3,4-tetraethylfluorenone
(10b) showing atom labeling and displacement ellipsoids at the 30%
probability level. Note the alternating up–down arrangement of the
ethyl substituents.
Scheme 5. Synthesis of Cp*Ru–fluorenone complex 11.
starting material was recovered. In contrast, catalytic hy-
drogenation with Pd/C in acetic acid[19] proved to be the
method of choice for this transformation, as it provided de-
sired fluorene 1a in 66% isolated yield. Interestingly, this
methodology was not successful for the reduction of 10b.
In this instance, reduction of the carbonyl group was
achieved by using the Me2NH·BH3 complex in the presence
of titanium(IV) chloride,[20] and after workup, 1b was iso-
lated in 38% yield. The same reaction conditions were also
used for the reduction of 10a to afford fluorene 1a in 59%
yield.
Figure 2. PLATON plot of the cation in the structure of 11·MeCN.
Displacement ellipsoids are scaled to the 30% probability level.
According to a search in the Cambridge Structural Data-
base,[21] structurally characterized transition-metal com-
plexes with π-coordinated fluorenone are restricted to only
several representatives, namely, to two cobalt complexes in
which unsubstituted fluorenone forms a η5:η6-bridge be-
tween two (η5-C5R5)CoI fragments[22] and the CpRu com-
plex [Ru(η5-C5H5)(η6-fluorenone)][PF6].[23] The π-coordi-
Scheme 4. Reduction of fluorenones 10 to fluorenes 1.
Finally, complexation of fluorenone 10a with [Ru(η5- nated aromatic rings in 11 are practically parallel [dihedral
C5Me5)(MeCN)3][PF6] was attempted. Heating a mixture angle 1.2(1)°], which in turn corresponds to marginal varia-
of these two compounds in warm dioxane resulted in the tion in the individual Ru–C distances [2.190(2)–2.233(2)
formation of η6-fluorenone complex 11 in 25% isolated and 2.177(2)–2.192(2) Å for the coordinated cyclopen-
yield (Scheme 5). Its structure was unequivocally confirmed tadienyl and benzene ring, respectively]. The 14 atoms con-
by single-crystal X-ray analysis for the solvate 11·MeCN stituting the fluorenone skeleton [i.e., C(1–13) and O] are
(Figure 2). Analysis of the crude reaction mixture (after coplanar within ca. 0.09 Å, and the C–C distances within
evaporation) by NMR spectroscopy revealed only the pres- the coordinated benzene ring [1.411(3)–1.423(3) Å] are
ence of complex 11 and unreacted 10a. Other products (in- slightly longer than those found in the uncoordinated benz-
cluding possible isomers of 11) were not detected. The pref- ene ring in 11 [1.388(3)–1.415(3) Å] and even for the unsub-
erential coordination of the bulky (η5-pentamethylcyclo- stituted ring in the molecule of 1b [1.382(2)–1.405(2) Å].
pentadienyl)ruthenium cation to the unsubstituted benzene The C=O bond remains unaffected by coordination [cf.,
ring of the fluorenone moiety could be attributed to the C=O 1.218(2) Å for 1b and 1.215(3) Å for 11].
46
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 44–47

Synthesis of 1,2,3,4-Tetramethyl- and 1,2,3,4-Tetraethylfluorene
Hu, Org. Lett. 2006, 8, 5057–5060; c) C.-G. Dong, Q. S. Hu,
Tetrahedron 2008, 64, 2537–2552.
Conclusions
[5] S. Ren, E. Igarashi, K. Nakajima, K. Kanno, T. Takahashi, J.
Am. Chem. Soc. 2009, 131, 7492–7493.
In conclusion, we have presented a new and concise
method for the synthesis of 1,2,3,4-tetra(m)ethylfluor-
enones and -fluorenes by making use of Dewar benzenes as
the key synthetic intermediates. Moreover, we demonstrated
that borane-based reductants are suitable for the reduction
of fluorenones to fluorenes. Last, but not least, the prefer-
ential coordination of the ruthenium atom in complex 11
to the less sterically hindered and the less electron-rich
benzene ring suggests that the unsymmetrically substituted
fluorenones and fluorenes accessible by this newly discov-
ered route might exert interesting coordination chemistry
with transition metals.
[6] a) T. W. Mojelsky, O. P. Strausz, Org. Geochem. 1986, 9, 31–37;
b) T. W. Mojelsky, O. P. Strausz, Org. Geochem. 1986, 9, 39–45.
[7] a) M. Ohkita, K. Ando, K. Yamamoto, T. Suzuki, T. Tsuji,
Chem. Commun. 2000, 83–84; b) M. Ohkita, K. Ando, T. Su-
zuki, T. Tsuji, J. Org. Chem. 2000, 65, 4385–4390; c) M. Ohkita,
K. Ando, T. Tsuji, Chem. Commun. 2001, 2570–2571.
[8] For other applications, see: a) M. J. Marsella, M. M. Meyer,
F. S. Tham, Org. Lett. 2001, 3, 3847–3849; b) M. J. Marsella,
S. Estassi, L. S. Wang, K. Yoon, Synlett 2004, 192–194.
[9] For studies regarding the mechanism of the Dewar benzene
rearrangement to benzene, see: a) R. P. Johnson, K. J. Daoust,
J. Am. Chem. Soc. 1996, 118, 7381–7385; b) R. W. A. Havenith,
L. W. Jenneskens, J. H. Lenthe, THEOCHEM 1999, 492, 217–
224; c) J. E. Norton, L. P. Olson, K. N. Houk, J. Am. Chem.
Soc. 2006, 128, 7835–7845; d) M. Dracˇínský, O. Castaño, M.
Kotora, P. Bourˇ, J. Org. Chem. 2010, 75, 576–581.
CCDC-873451 (for 10b) and -873452 (for 11) contain the supple-
mentary crystallographic data for this paper. These data can be
obtained free of charge from The Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
ˇ
[10] a) S. Janková, M. Dracˇínský, I. Císarˇová, M. Kotora, Eur. J.
ˇ
Org. Chem. 2008, 47–51; b) S. Janková, I. Císarˇová, F. Uhlík,
ˇ
ˇ
P. S teˇpnicˇka, M. Kotora, Dalton Trans. 2009, 3137–3139; c) S.
Janková, S. Hybelbauerová, M. Kotora, Synlett 2011, 396–398.
[11] J. B. Koster, G. J. Timmermans, H. van Bekkum, Synthesis
1971, 139–140.
Supporting Information (see footnote on the first page of this arti-
cle): Experimental procedures, copies of the 1H NMR and 13C
NMR spectra for all new compounds, and crystallographic data
for 10b and 11.
[12] P. B. J. Driessen, H. Hogeveen, J. Am. Chem. Soc. 1978, 100,
1193–1200.
[13] J. H. Dopper, B. Greijdanus, H. Wynberg, J. Am. Chem. Soc.
1975, 97, 216–218.
Acknowledgments
[14] E. E. van Tamelen, S. P. Pappas, J. Am. Chem. Soc. 1963, 85,
Financial support from the Grant Agency of the Academy of Sci-
ences of the Czech Republic (project no. IAA400550702) and the
Ministry of Education, Youth and Sports of the Czech Republic
(project nos. LC06070 and MSM0021620857) is gratefully ac-
knowledged.
3297–3298.
[15] Hanamoto, Y. Koga, T. Kawanami, H. Furuno, J. Inanaga, Tet-
rahedron Lett. 2006, 47, 493–495.
[16] D. J. Iverson, G. Hunter, J. F. Blount, J. R. Danewood Jr., K.
Mislow, J. Am. Chem. Soc. 1981, 103, 6073–6083.
[17] A. L. Spek, J. Appl. Crystallogr. 2003, 36, 7–13.
[18] L. Caglioti, Org. Syntheses, Coll. Vol. 1988, 6, 62–63.
[19] K. Ofosu-Asante, L. M. Stock, J. Org. Chem. 1987, 52, 2938–
2939.
[1] For examples of transition-metal coordination to the benzene
ring of fluorenes, see: a) J. Moss, J. Thomas, A. Ashley, A. R.
Cowley, D. O’Hare, Organometallics 2006, 25, 4279–4285; b)
J. Moss, J. Thomas, A. Ashley, A. R. Cowley, D. O’Hare, J.
Organomet. Chem. 2007, 692, 2071–2075.
[2] For examples of transition-metal coordination to the central
cyclopentadiene ring in fluorenes, see: a) P. Bazinet, T. D. Til-
ley, Organometallics 2006, 25, 4286–4291; b) P. Bazinet, T. D.
Tilley, Organometallics 2008, 27, 1267–1274; c) P. Bazinet, T. D.
Tilley, Organometallics 2009, 28, 2285–2293.
[20] E. V. Dehmlov, T. Niemann, A. Kraft, Synth. Commun. 1996,
26, 1467–1472.
[21] Cambridge Structural Database, version 5.33 of November
2011 with updates from November 2011 and February 2012.
[22] S. Guo, R. Hauptmann, J. J. Schneider, Z. Anorg. Allg. Chem.
2007, 633, 2332–2337.
[23] M. H. Garcia, A. Valente, P. Florindo, T. S. Morais, F. M. Pie-
dale, M. T. Duarte, V. Moreno, F. X. Avilés, J. Loreno, Inorg.
Chim. Acta 2010, 363, 3768–3775.
[3] L. H. Klemm, D. R. Taylor, Prep. Proc. Int. 1976, 8, 163–168.
[4] a) C.-G. Dong, Q. S. Hu, Angew. Chem. 2006, 118, 2347; An-
gew. Chem. Int. Ed. 2006, 45, 2289–2292; b) C.-G. Dong, Q. S.
Received: October 11, 2012
Published Online: November 21, 2012
Eur. J. Org. Chem. 2013, 44–47
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
47