Synthesis of benzonorbornadienes: Regioselective benzyne formation

Article abstract of DOI:10.1021/jo001277k

This report details the synthesis of several benzonorbornadienes by Diels-Alder cycloaddition of cyclopentadiene derivatives with substituted benzyne intermediates, which were generated by low-temperature metal-halogen exchange of halobenzenes. General conditions were developed, allowing synthesis of most benzonorbornadienes described herein at the multigram scale with isolated yields approaching 90% in some cases. Cycloaddition of the benzyne produced by substitution of a chlorodifluorobenzene for a bromodifluorobenzene in the metal-halogen exchange reaction unexpectedly gave a different benzonorbornadiene. The benzyne, which resulted by a deprotonation pathway rather than by metal-halogen exchange, formed in a highly regioselective elimination step.

Full text of DOI:10.1021/jo001277k

2932
J . Org. Chem. 2001, 66, 2932-2936
Syn th esis of Ben zon or bor n a d ien es: Regioselective Ben zyn e
F or m a tion
Kenneth C. Caster,* Christopher G. Keck, and Russell D. Walls
Lord Corporation, Materials Division, 110 Lord Drive, Cary, North Carolina 27511
ken•caster@lord.com
Received August 23, 2000
This report details the synthesis of several benzonorbornadienes by Diels-Alder cycloaddition of
cyclopentadiene derivatives with substituted benzyne intermediates, which were generated by low-
temperature metal-halogen exchange of halobenzenes. General conditions were developed, allowing
synthesis of most benzonorbornadienes described herein at the multigram scale with isolated yields
approaching 90% in some cases. Cycloaddition of the benzyne produced by substitution of a
chlorodifluorobenzene for a bromodifluorobenzene in the metal-halogen exchange reaction
unexpectedly gave a different benzonorbornadiene. The benzyne, which resulted by a deprotonation
pathway rather than by metal-halogen exchange, formed in a highly regioselective elimination
step.
Sch em e 1
In tr od u ction
Several years ago, we began a program to prepare
polymers with designed, controlled properties using ring-
opening metathesis polymerization (ROMP). Structur-
ally, ROMP monomers require a disubstituted olefin to
be part of a strained ring system such as a norbornene,
norbornadiene, benzonorbornadiene, or cyclooctene.1 Many
synthetic routes to the benzonorbornadiene ring system
may be imagined with a primary method involving
Diels-Alder [4 + 2] cycloaddition of cyclopentadiene
derivatives with benzyne intermediates. Although the
rich chemistry of these intermediates has been investi-
gated in great detail, they are short-lived, fleeting species
that must be generated and trapped in situ; many
different approaches to their generation have been
reported.2 Only recently have they been isolated and
spectroscopically characterized in an argon matrix3 and
in solution.4 Through these intermediates, we have
prepared the substrates for our study (Figure 1). It was
during the preparation of halogenated benzonorborna-
dienes that we encountered an unusual regioselective
elimination during benzyne generation.
Meta l-Ha logen Exch a n ge Rea ction s. Metalation
of 1,2-dihalogenated benzenes either through Grignard
or aryllithium formation is an extremely useful method
to generate benzynes: metal-halogen exchange is fol-
lowed by elimination of a stable MX compound (M )
metal, X ) halogen) with simultaneous formation of the
benzyne. Grignard chemistry usually requires elevated
temperatures for both metalation and elimination steps,
and solvent choice depends on the halogen being ex-
changed5 (Scheme 1).
In contrast to arylmagnesium formation, lithium-
halogen exchange occurs at low temperature. Monoha-
logenated benzenes undergo metalation, but regiocontrol
may be lost during the elimination step. 1,2-Disubstituted
dihalogens 1 provide the best regiocontrol in the elimina-
tion step (Scheme 1). When different halogens are present
on the ring, two opportunities for lithium-halogen
exchange exist. In general, the more electropositive
halogen exchanges first.2,5 Once formed, several pathways
are available for the aryllithium to react. Side-reactions
are known, but these can be controlled by proper reagent
choice, reaction stoichiometry, and solvent.5,6 At low
temperature, the aryllithium 2 is stable to elimination.
This temperature is highly dependent on what group is
being eliminated (e.g., triflates, tosylates, mesylates)7 and
on the overall electronic effects in the ring induced by
other substitution. For example (G ) H), the relative
stability of halogenated aryllithium 2 toward elimination
of LiX follows the order LiBr (-100 °C) < LiCl (-90 °C)
< LiF (-60 °C).8 It is expected that substitution will
greatly affect these elimination temperatures. Many
reactions can occur once the benzyne has been generated.
For example, biphenylenes readily form by dimerization
while nucleophilic alkyllithium reagents can add to give
new aryllithium reagents.9 In the presence of a suitable
diene, cycloaddition can give a benzonorbornadiene de-
rivative.2
(1) Ivin, K. J .; Mol, J . C. Olefin Metathesis and Metathesis Polym-
erization; Academic: New York, 1997.
(2) Hoffmann, R. W. Dehydrobenzene and Cycloalkynes; Academic:
New York, 1967.
(3) Moriyama, M.; Ohana, T.; Yabe, A. J . Am. Chem. Soc. 1997, 119,
10229.
(6) Negishi, E.-I. Organometallics in Organic Synthesis; J ohn
Wiley: New York, 1980; p 116.
(7) Matsumoto, T.; Hosoya, T.; Katsuki, M.; Suzuki, K. Tetrahedron
Lett. 1991, 32, 6735.
(4) Warmuth, R. Angew. Chem., Int. Ed. Engl. 1997, 36, 1347.
(5) Stowell, J . C. Carbanions in Organic Synthesis; J ohn Wiley: New
York, 1979; Chapter 1.
(8) Reference 5, pp 43-44.
(9) Cava, M. P.; Mitchell, M. J . Cyclobutadiene and Related Com-
pounds; Academic: New York, 1967; Chapter 10.
10.1021/jo001277k CCC: $20.00 © 2001 American Chemical Society
Published on Web 04/03/2001

Synthesis of Benzonorbornadienes
J . Org. Chem., Vol. 66, No. 9, 2001 2933
F igu r e 1. Stereochemistry between norbornene monomer and ROMP polymer. Substituent control in the Diels-Alder cycloaddition
reaction.
Resu lts a n d Discu ssion
ation of 1-bromo-2,4-difluorobenzene 3 with n-BuLi (1.2
equiv) in THF at low temperature and trapping of the
generated benzyne 18 on warming with furan (1.6 equiv)
yielded a product mixture containing benzonorbornadiene
10 and many impurities as determined by analysis of the
Of the two dienes used in this work, furan actively
participated in cycloaddition reactions, whereas N-meth-
ylpyrrole was more sluggish as a result of its increased
aromaticity.10,11 Six different halogenated benzenes (3-
8) were used as starting materials. Compounds 3, 4 and
1
crude H NMR spectrum and by thin-layer chromatog-
raphy (nine spots). Identifiable from the NMR spectrum
was n-butylbenzene 17. This side-product arises by
alkylation of the aryllithium 16 with n-butyl bromide,
which is formed as a byproduct of the initial metal-
halogen exchange.6 As primary alkyl bromides are good
alkylating agents, this process is competitive with cy-
cloaddition to give norbornadiene 10 or dimerization to
give biphenylene 19 (Scheme 2). While no direct evidence
was obtained for 19, it is a likely side-product, as
biphenylenes are known to be produced by benzyne
dimerization.9 Butylbenzene 17 was minimized by using
an excess of n-BuLi (2 equiv). However, while benzonor-
bornadiene 10 formed, so were other impurities as
5, 6 should give mono- and difluoro-benzonorbornadienes,
respectively, and were chosen to understand reactivity
differences between Cl-F, Br-F, and Br-Br substitution
patterns. It is with compounds 3 and 4 that regioselective
benzyne formation was observed. The parent 7-oxaben-
zonorbornadiene 9 (1,4-diepoxy-1,4-dihydronaphthyl-
ene12) was commercially available and was used to aid
in NMR analysis of products 10-15.
1
revealed by H NMR.
By changing the reaction solvent to diethyl ether and
trapping fluorobenzyne 18 with 10 equiv of furan, reac-
tion yield and product selectivity to 10 were dramatically
increased giving isolated yields ranging from 85 to 95%.
Even on the 25 g scale, the isolated yield was 94%, which
compared favorably with lower scale runs. This improve-
ment is clearly the result of a strong solvent effect. We
wondered if this same effect was operable when chlo-
robenzene 4 was substituted for bromobenzene 3. This
substitution, which will be described below, gave very
different results.
Again, with chlorobenzene 4 we saw mixed results
when performing the metal-halogen exchange with
n-BuLi in THF. At least five compounds, with a primary
1
one, were observed by TLC and H NMR spectroscopy.
As described above, changing the reaction solvent to
diethyl ether and increasing the number of equivalents
of diene to 10 yielded a pure compound as an oily, clear-
yellow liquid in 90% crude yield. Kugelrohr distillation
yielded it as a waxy, white, low-melting solid which
displayed very clean 1H and 13C NMR spectra and was
determined to be another benzonorbornadiene 11 (Scheme
2). This reaction was easily scaled-up to yield larger
quantities of 11 in 86% (5.3 g) and 97% (30.4 g) yields.
Significant differences in metalation between bro-
mobenzene 3 and chlorobenzene 4 resulted in different
benzonorbornadienes. Specifically, norbornadiene 10 was
the primary product isolated from the bromobenzene
whereas norbornadiene 11 was derived from the chlo-
robenzene. Potential deprotonation pathways leading to
this other benzonorbornadiene were envisioned as shown
Ben zon or bor n a d ien e Syn th esis by Meta l-Ha lo-
gen Exch a n ge. A variety of reactions were examined
while identifying the best conditions for metal-halogen
exchange and trapping of the intermediate benzyne. The
synthesis of benzonorbornadienes via lithium-induced
benzyne formation has been examined by Coe. They
reported on the strong solvent effects associated with the
metal-halogen exchange reaction and concluded that
THF was a poor solvent for this chemistry and that more
highly selective reactions resulted when performed in
diethyl ether.13 Our results support their findings. Lithi-
(10) Acheson, R. M. An Introduction to the Chemistry of Heterocyclic
Compounds; 3rd ed.; J ohn Wiley: New York, 1976; p 174.
(11) Katritzky, A. R. Handbook of Heterocyclic Chemistry; Perga-
mon: New York, 1985; p 76.
(12) Stiles, M.; Miller, R. G.; Burckhardt, U. J . Am. Chem. Soc. 1963,
85, 1792.
(13) Coe, P. L.; Waring, A. J .; Yarwood, T. D. J . Chem. Soc., Perkin
Trans. 1 1995, 2729.

2934 J . Org. Chem., Vol. 66, No. 9, 2001
Caster et al.
Sch em e 2
in Scheme 2. Formation of benzonorbornadiene 10 in-
volves metal-halogen exchange of bromobenzene 3 to
give aryllithium 16 followed by elimination of LiF to give
benzyne 18 which then undergoes cycloaddition with
furan. Benzonorbornadienes 11 and 23 are formed by
deprotonation of chlorobenzene 4 to give the common
aryllithium intermediate 20 which on elimination of LiF
gives either benzyne 21 or 22 followed by cycloaddition
with furan. Benzonorbornadienes 24 and 25 would be
produced by metalation at C-5 and C-6, respectively,
subsequent loss of LiF or LiCl, and cycloaddition to furan.
That elemental analysis of the crude product isolated
from reaction of 4 showed it to be analytically pure
(>99.7%), and that the NMR showed other benzonorbor-
nadienes in addition to the main product 11, suggests
that isomers 23 and 24 were generated in the reaction.
C-13 NMR data obtained on a component isolated by
column chromatography during one of the THF solvated
reactions showed characteristic doublet-of-doublet carbon-
fluorine coupling patterns associated with an 1,3-difluoro
substitution on the aromatic ring, as shown in norbor-
nadiene 25. This differential reactivity was unantici-
pated.
zene derivatives. For example, 1,3-difluorobenzene is
deprotonated at the relatively acidic 2-position.14 Many
similar examples have been reported in the litera-
ture.2,13,15,16
The 1H NMR data allow assignment of the correct
fluoro- and chloro-regiochemistry in the primary ben-
zonorbornadiene reaction product. Bridgehead proton
resonances were key to determining the structure of 11
over its isomer 23, and substituent effects from the
aromatic ring on these resonances are clearly observed
by comparison of differently substituted halogenated
benzonorbornadienes. Halogen substitution in the aro-
matic ring at C-5,8 deshields C-4,1 protons, respectively,
whereas substitution at C-6,7 has no affect on them.
Benzonorbornadienes 24 and 25, while present in small
quantities in the reaction mixture, are easily eliminated
as the primary reaction product based on their expected
NMR spectral data and on the low probability of meta-
lation at C-5 and C-6 positions of the chlorobenzene.
Most puzzling was the high regioselectivity observed
in the elimination step to generate benzyne 21 over 22.
This result may be explained by considering that aryne
reactivity and stability are determined by substituent
inductive rather than resonance effects.17 As fluoride is
eliminated from the aromatic ring, chlorine proximity to
the developing positive charge on the eliminated C-F
In an attempt to understand and explain these unusual
results, literature on metalation of halogen-containing
benzene derivatives was critically evaluated. Aryl bro-
mides are known to undergo metal-halogen exchange
at much lower temperatures than the corresponding aryl
chloride. Besides metal-halogen exchange, alkyllithium
reagents can deprotonate appropriately substituted ben-
(15) Bu¨ker, H. H.; Nibbering, N. M. M.; Espinosa, D.; Mongin, F.;
Schlosser, M. Tetrahedron Lett. 1997, 38, 8519.
(16) Schlosser, M. Angew. Chem., Int. Ed. 1998, 110, 1496.
(17) Modena, G.; Scorrano, G. Directing, Activating and Deactivating
Effects. In The Chemistry of the Carbon-Halogen Bond; Patai, S. Ed.;
J ohn Wiley: New York, 1973; Part 1, Chapter 6, pp 355-361, 366-
374.
(14) Tamborski, C.; Soloski, E. J . J . Org. Chem. 1966, 31, 746.

Synthesis of Benzonorbornadienes
J . Org. Chem., Vol. 66, No. 9, 2001 2935
Sch em e 3
efficient pathway to occur for other appropriately sub-
stituted aromatic and heteroaromatic compounds.
Exp er im en ta l Section
Gen er a l Meth od s. 1H and 13C NMR spectra were recorded
on
a Bruker 250 MHz NMR at 250.13 and 62.9 MHz,
respectively. All chemical shifts (δ) are positive and referenced
downfield from tetramethylsilane (TMS); coupling constants
(J ) are recorded in Hz unless otherwise noted.
All reagents and solvents were ordinary commercial grade
and obtained from either Aldrich, Acros, Fluka, or PCR, Inc.
Furan, N-methylpyrrole, 1-bromo-2,4-difluorobenzene 3 (P2O5),
1-chloro-2,4-difluorobenzene 4 (P2O5), 1-bromo-4-chloro-2-fluo-
robenzene 7 (P2O5), and 1-chloro-2,3,4,5,6-pentafluorobenzene
8 (CaSO4) were purified by distillation at reduced pressure
from the drying agent shown. 1-Bromo-2,4,5-trifluorobenzene
5, 1,2-dibromo-4,5-difluorobenzene 6, and 9-oxabenzonorbor-
nadiene 9 were used as received. THF and diethyl ether were
freshly distilled from benzophenone ketyl before use. n-BuLi
(2.5 M solution in hexanes) was purchased from Aldrich and
used without analysis. All handling of air-sensitive materials
was done under an argon or nitrogen atmosphere using
standard syringe and cannula techniques.
Thin-layer chromatography (TLC) was performed on glass-
backed silica gel plates (0.25 mm thickness, E. Merck). Spots
were visualized under UV light (254 nm) and then stained with
an ethanolic solution (5%) of phosphomolybdic acid and
charred with a heat gun. Flash column chromatography was
performed using Merck grade 9385, 230-400 mesh, 60 Å silica
gel (available from Aldrich, 22,719-6). Compounds were
diluted to 25% solution with hexanes and applied to the top
of the adsorbent bed. Elution began with 100% hexanes
followed by hexanes/EtOAc (75%:25%) and finally with 100%
EtOAc. Melting points and boiling points are uncorrected.
Kugelrohr distillation temperatures are oven temperatures.
Combustion elemental analyses were performed by Galbraith
Laboratories (Knoxville, TN).
bond determines relative transition state stability. Hence,
chlorine substitution destabilizes positive charge forma-
tion at C-2 more than at C-4 and leads to selective
formation of 21 (Scheme 3).
Oth er Ben zon or bor n adien es. The knowledge gained
as a result of the syntheses of benzonorbornadienes 10
and 11 allowed the straightforward synthesis of several
heterocyclic benzonorbornadiene derivatives. On exami-
nation of its 1H and 13C NMR spectra, benzonorborna-
diene 12 appeared cleaner when prepared from dibro-
modifluorobenzene 6 (65-85%) than from bromotrifluoro-
benzene 5 (60%). Benzonorbornadiene 13 (52-65%) was
prepared from bromochlorobenzene 7. NMR analysis
showed one primary product and a small amount of an
isomeric benzonorbornadiene side-product. Heavily flu-
orinated benzonorbornadienes 1418 (28-70%) and 1519
(36-64%) were also prepared. The benzonorbornadienes
prepared in this study are stable oils or crystalline solids
and are readily isolated by column chromatography.
While undistillable using short-path distillation (i.e., they
tend to decompose), high mass balances were obtained
by Kugelrohr distillation, which left only small amounts
of resinous products in distillation pot. Full physical
1
property and combustion data and H and 13C NMR and
Typ ica l P r oced u r e: 9-Oxa -6-flu or ob en zon or b or n a -
d ien e (10).13 To a 1000 mL, 4-neck, round-bottom flask, which
was equipped with an overhead stirrer, condenser, thermom-
eter, rubber septum, gas adapter, and stirring bar and flame-
dried under Ar, were charged 400 mL of diethyl ether and 17.6
mL of bromobenzene 3 (0.156 mol). To the solution, which was
cooled to -78 °C in an acetone/dry ice bath, was added via
gastight syringe 72 mL of a 2.5 M solution of n-BuLi in
hexanes (0.180 mol) over a 1 h period at such a rate to keep
the reaction mixture below -70 °C. The reaction mixture was
then stirred for 35 min at -78 °C before 120 mL of furan (1.650
mol) was added over 20 min at such a rate to keep the reaction
temperature below -73 °C. It was then allowed to warm slowly
to room-temperature overnight. The cloudy-orange reaction
mixture was poured into 600 mL of stirring deionized water,
vacuum filtered, separated, and the aqueous phase was
washed (2 × 75 mL) with ether. The combined organic phases
were dried over MgSO4, vacuum filtered, and concentrated by
rotoevaporation and high vacuum to give 23.76 gm of clear-
gold 10 (94% yield) which was purified by Kugelrohr distilla-
tion to give a clear-yellow, liquid: bp 100 °C/0.25 Torr; Rf 0.36
(95:5 hexane/EtOAc); 1H NMR (250 MHz, CDCl3) δ 5.648 (pt,
J ) 1.6, 1.6 Hz, 2H), 6.597 (ddd, J ) 9.8, 7.5, 2.0 Hz, 1H),
6.968 (d, J ) 1.5 Hz), 6.985 (d, J ) 1.7 Hz), 6.94-7.01 (m),
7.100 (dd, J ) 7.9, 4.7 Hz, 1H); 13C NMR (62.5 MHz, CDCl3)
δ 81.77, 82.09, 109.37 (d, J ) 24.9 Hz), 110.10 (d, J ) 22.5
Hz), 120.47 (d, J ) 8.8 Hz), 142.34, 143.33, 144.14 (d, J ) 2.4),
151.85 (d, J ) 8.8 Hz), 160.27 (d, J ) 244.6 Hz); IR (neat, NaCl)
ν 3016, 1608 (s), 1459 (s), 1350, 1281 (s), 1217 (s), 1185, 1083,
993 (s), 966, 928, 852 (s), 834 (s), 816 (s), 740 (s), 698 (s), 645
(s), 603 (s), 576 (s) cm-1. Anal. Calcd for C10H7FO: C 74.06; H
4.36; F 11.72. Found: C 74.25; H 4.75; F 11.36; Br 123 ppm;
Li <0.03%.
IR spectra including analysis and assignment are avail-
able in Supporting Information.
Su m m a r y
A series of benzonorbornadienes were prepared by
trapping in situ generated benzynes with furan and
N-methylpyrrole. Benzynes were generated by metal-
halogen exchange of substituted halobenzenes followed
by elimination of a lithium salt. High product yields and
selectivities were obtained when (a) bromoaromatics were
used as starting materials, (b) diethyl ether was used as
the reaction solvent, (c) low temperatures (r70 °C) were
used in the metal-halogen exchange step, and (d) excess
diene (10 equiv) was used to trap the intermediate
benzyne. In some cases, reactions were performed on up
to 50 g of starting halobenzene and could easily be done
on a larger scale. We expect this procedure to be general
and should be applicable to other halogen-containing
benzene derivatives.
Cycloaddition of the benzyne produced by substitution
of 1-chloro-2,4-difluorobenzene for its 1-bromo analogue
in the metal-halogen exchange reaction unexpectedly
gave benzonorbornadiene 11 instead of 10. The benzyne,
which resulted by a deprotonation pathway rather than
by metal-halogen exchange, formed in a highly regiose-
lective elimination step. We expect this surprisingly
(18) Gribble, G. W.; LeHoullier, C. S.; Sibi, M. P.; Allen, R. W. J .
Org. Chem. 1985, 50, 1611.
(19) Gribble, G. W.; Allen, R. W.; LeHoullier, C. S.; Eaton, J . T.;
Easton, N. R.; Slayton, R. I.; Sibi, M. P. J . Org. Chem. 1981, 46, 1025.
9-Oxa -5-flu or o-6-ch lor oben zon or bor n a d ien e (11). The
reaction was carried out according to the typical procedure
using 50 mL of ether and 2.2 mL of chlorobenzene 4 (0.022

2936 J . Org. Chem., Vol. 66, No. 9, 2001
Caster et al.
mol), 9 mL of a 2.5 M solution of n-BuLi in hexanes (0.022
mol), and 14.5 mL of furan (0.200 mol). This gave 2.915 g of
11 as a clear-orange, oily liquid (90% yield), which on Kugel-
rohr distillation yielded a waxy, white solid.
procedure A using 5.396 g of freshly distilled 8 (26.65 mmol)
in 30 mL of dry diethyl ether, 10.8 mL of 2.6 M n-BuLi in
hexanes (27 mmol), and 8.70 g of freshly distilled N-meth-
ylpyrrole (107.22 mmol). The combined ether layers were
concentrated by rotoevaporation to give 5.435 g of crude
product. 1H NMR showed this to be contaminated with
unreacted N-methylpyrrole (42%). The crude product was
purified by staged Kugelrohr distillation (rt/5 Torr) to remove
the pyrrole. The receiving flask was changed, and the product
was sublimed (125 °C/5 Torr) in the Kugelrohr apparatus. This
gave 2.56 g of yellow-white crystals which on sublimation a
second time (75 °C/10 Torr) gave 2.073 g of 15 as white crystals
(32% yield).
9-Oxa -6,7-d iflu or oben zon or bor n a d ien e (12).13 The reac-
tion was carried out according to the typical procedure using
15.318 g of dibromobenzene 6 (0.056 mol), 250 mL of ether,
29.0 mL of 2.5 M n-BuLi (0.072 mol) in hexanes, and 41.0 mL
of furan (0.564 mol). Workup gave 8.63 g of crude product (85%
yield) which on purification by Kugelrohr distillation gave
5.282 g of 12 (52% yield) as a clear, light-yellow liquid.
Compound 12 was also prepared according to the typical
procedure using 4.128 g of bromobenzene 5 (0.020 mol), 50 mL
of ether, 9.0 mL of 2.5 M n-BuLi in hexanes (0.022 mol), and
14.5 mL of furan (0.199 mol). Workup gave 2.12 g of crude
compound 12 (60% yield).
Ack n ow led gm en t. We thank the Management of
Lord Corporation for supporting this work and allowing
us to publish our results. We would like to thank
Professor Louis D. Quin for his helpful comments during
the preparation of this manuscript. Discussions with Dr.
J ohn D. Anderson (Milliken, Spartanburg, SC) regard-
ing elemental analysis of fluorine compounds are most
appreciated.
9-Oxa -6-ch lor oben zon or bor n a d ien e (13). The reaction
was carried out according to the typical procedure using 4.327
g of freshly distilled chlorobenzene 7 (20.66 mmol) in 30 mL
of dry diethyl ether, 8.4 mL of 2.5M n-BuLi in hexanes (21.00
mmol), and 1.5 g of freshly distilled furan (22.00 mmol).
Workup gave 2.25 g of crude 13 (61%) as a cloudy orange
liquid. Purification by Kugelrohr distillation gave 1.93 g (52%)
of 13 as a clear liquid.
9-Oxa-5,6,7,8-tetr aflu or oben zon or bor n adien e (14).18 The
reaction was carried out according to the typical procedure
using 2.04 g of freshly distilled 8 (10.07 mmol) in 30 mL of
dry diethyl ether, 3.9 mL of 2.6 M n-BuLi in hexanes (10.14
mmol), and 0.84 g of freshly distilled furan (12.37 mmol).
Workup gave 1.635 gms of crude 14 as an orange crystalline
solid which was sublimed (80 °C/1 Torr) to give 0.613 g (28%)
as near white crystals.
Su p p or tin g In for m a tion Ava ila ble: Physical property
1
data, combustion analysis data, and H and 13C NMR assign-
ments for most compounds are available. Copies of spectra for
3, 4, 5, 7, 9, 10, 11, 12, 13, 14, 15 (1H NMR); 3, 4, 7, 9, 10, 11,
12, 13, 14, 15 (13C NMR); and 9, 10, 11, 13, 15 (IR). This
material is available free of charge via the Internet at
http://pubs.acs.org.
9-Me t h yl-9-a za -5,6,7,8-t e t r a flu or ob e n zon or b or n a -
d ien e (15).19 The reaction was carried out according to typical
J O001277K