Nickel-mediated cyclobutabenzene syntheses. trans-7,8-dibromo-cyclobutabenzenes: 1 Their one-pot preparation, x-ray structure, and Diels-Alder reactions

Full text of DOI:10.1021/jo9516852

J . Org. Chem. 1996, 61, 2549-2552
2549
of the respective bis(iodoaryls). We intended to use the
Ni-mediated coupling reaction intramolecularly to form
a four-membered ring. In order to enable better control
of the reaction, we decided to use defined Ni(0) complexes
instead of the usual in-situ prepared cyclization media-
tors, formed from Ni(II), phosphines, and a reducing
agent. We did this because in several of the reported
cases the product may have resulted from a reaction
mediated by the reducing agent (e.g., Zn) rather than
from the Ni-mediated reaction. Thus, when a solution
of R,R,R′,R′-tetrabromo-o-xylene was stirred with Zn in
DMF, with or without the presence of ca. 10% (Bu3P)2-
Ni(COD) (COD ) 1,5-cyclooctadiene), identical product
mixtures were obtained. We have therefore used definite
Ni(0) complexes: (Bu3P)2NiL (L ) anthracene, COD, 2a
and 2b,13 respectively).14
Nick el-Med ia ted Cyclobu ta ben zen e
Syn th eses. tr a n s-7,8-Dibr om o-
cyclobu ta ben zen es:1 Th eir On e-P ot
P r ep a r a tion , X-r a y Str u ctu r e, a n d
Diels-Ald er Rea ction s
Amnon Stanger,*,2a Nissan Ashkenazi,2a
Alona Shachter,2a Dieter Bla¨ser,2b Peter Stellberg,2b and
Roland Boese*,2b
Department of Chemistry, TechnionsIsrael Institute of
Technology, Haifa 32000, Israel, and Institut fu¨r
Anorganische Chemie der Universita¨t-GH Essen,
Universita¨tsstrasse 5-7, D-45117, Essen, Germany
Received J anuary 24, 1996
The initial attempts to close a four-membered ring
were focused on R,R′-dibromo-o-xylene, but the only
reaction products were dimers, mainly 1,5-dibenzocy-
clooctadiene. Under no conditions (including addition of
reagents by a syringe pump) could we obtain the in-
tramolecular ring-closure product.15 Thus, as the inter-
molecular coupling is extremely facile, it was necessary
to decrease the rate of the bimolecular coupling and/or
increase the rate of the intramolecular coupling in order
to force the reaction to proceed via the intramolecular
pathway. Steric congestion near the reaction center(s)
was thought to retard the bimolecular coupling, since it
should prevent close contact between the reacting mol-
ecules. The intramolecular reaction should be enhanced
by stabilization of the intermediate in order to prolong
its lifetime and thus let the entropy (which favors the
intramolecular ring closure) play a more important role
in the determination of the reaction rate. In case of the
nickel-mediated coupling, one of the key intermediates
is probably an insertion product of Ni into a C-X bond;16
thus, an electronegative substituent on the reacting
carbon atom should stabilize it. A substituent that
fulfills both demands is Br; therefore, we have reacted
R,R,R′,R′-tetrabromo-o-xylene (1) with a Ni(0) complex (2a
or 2b) under cyclization conditions. The reaction is
described in eq 1. The products were assigned by 1H
Cyclobutabenzenes are of considerable interest, both
from a synthetic point of view, being synthons for
o-xylylenes,3,4 and from that of theory, as they are key
compounds for the study of the Mills Nixon effect.5
However, the number of methods for their preparation
is limited. Cyclobutabenzene is usually prepared in
moderate yields from NaI/KI-mediated cyclization of
R,R,R′,R′-tetrabromo-o-xylene (1).6 This reaction yields
dibromo- and diiodocyclobutabenzenes, each as a mixture
of cis and trans isomers.7 Thus, the method’s only
practical use is for the preparation of unsubstituted
cyclobutabenzene (after a medium yield reduction of the
halogen atoms)8 or as precursors for the preparation of
systems in which the stereochemistry is unimportant.9
Another synthetic method for the preparation of cyclo-
butabenzenes is based on the CpCoL2 (L ) CO, ethene)-
mediated [2 + 2 + 2] cyclization between 1,5-hexadiyne
and an alkyne.10 This method, however, yields cyclo-
butabenzenes that are disubstituted in the aromatic ring,
whereas the four-membered ring is not functionalized.11
We report here a simple, single-step, and high-yield
preparation of the title compounds, the X-ray structure
of the parent system, and preliminary results concerning
their Diels-Alder reactivity.
Nickel-mediated coupling of RX (where R is aryl,
benzyl, ethenyl, allyl and X ) Br, I) is a well-known and
widely used synthetic method.12 In all cases of which we
are aware, this method have been used for bimolecular
coupling, except for Semmelhack’s work12a in which he
reports intramolecular ring formation (where the ring
sizes are equal or larger than six) by the Ni(0) coupling
(1) Alternative name: trans-1,2-dibromobenzocyclobutene.
(2) (a) Technion. (b) Essen.
(3) Alternative name: o-quinodimethanes.
(4) Oppolzer, W. Synthesis 1978, 793.
NMR,17 and the X-ray structure of the major product
(shown in Figure 1, see below) shows that the tentative
assignments of the NMR signals of 3 and 4 were correct.
When the reaction is carried out in an NMR tube the
convergence of 1 to 3 and 4 is quantitative and no side
products are observed. The yield of the mixture (after
(5) (a) Boese, R.; Bla¨ser, D. Angew. Chem., Int. Ed. Engl. 1988, 27,
304. (b) Boese, R.; Bla¨ser, D.; Billups, W. E.; Haley, M. M.; Maulitz,
A. H.; Mohler, D. L.; Vollhardt, K. P. C. Angew. Chem., Int. Ed. Engl.
1994, 33, 313. (c) Stanger, A. J . Am. Chem. Soc. 1991, 113, 8277 and
references therein. (d) Siegel, J . S. Angew. Chem., Int. Ed. Engl. 1994,
33, 1721.
(6) Cava, M. P.; Napier, D. R. J . Am. Chem. Soc. 1956, 78, 500
(7) In a control experiment using Cava’s procedure we obtained
trans- and cis-dibromo and diiodocyclobutabenzene in a ratio of 63:
19:12:6, respectively. The yield of pure 3 obtained from this procedure
(mp ) 52.4-52.8 °C) after repeated recrystallization is 4.8%. See:
Cava, M. P.; Napier, D. R. J . Am. Chem. Soc. 1957, 79, 1701.
(8) (a) Sanders, A.; Giering, W. P. J . Org. Chem. 1973, 38, 3055. (b)
Cava, M. P.; Napier, D. R. J . Am. Chem. Soc. 1958, 80, 2255.
(9) Cava, M. P.; Deana, A. A.; Muth, K. J . Am. Chem. Soc. 1959,
81, 6458.
(12) (a) Semmelhack, M. F.; Helquist, P. M.; J ones, D. J . Am. Chem.
Soc. 1971, 93, 5908. (b) Semmelhack, M. F.; Ryono, L. S. J . Am. Chem.
Soc. 1975, 97, 3873. (c) Kende, A. S.; Liebeoking, L. S.; Braitsch, D.
M. Tetrahedron Lett. 1975, 3375. (d) Zembayashi, M.; Tameo, K.;
Yoshida, J .; Kumada, M. Tetrahedron Lett. 1977, 4089. (e) Tsou, T.
T.; Kochi, J . K. J . Am. Chem. Soc. 1979, 101, 7549. (f) Semmelhack,
M. K.; Helquist, P.; J ones, L. D.; Keller, L.; Mendelson, L.; Ryono, L.
S.; Smith, J . G.; Stauffer, R. D. J . Am. Chem. Soc. 1981, 103, 6460.
(13) Stanger, A.; Shazar, A. J . Organomet. Chem. 1993, 458, 233.
(14) It is unimportant which L is used as the (Bu3P)2Ni ligand,
because the reaction solvent (DMF) replaces L after a few minutes
and the mediator is actually (Bu3P)2Ni(DMF)n.
(10) Funk, R. L.; Vollhardt, K. P. C. J . Am. Chem. Soc. 1976, 98,
6755.
(11) Another drawback is the necessity to prepare the diyne that is
not commercially available.
0022-3263/96/1961-2549$12.00/0 © 1996 American Chemical Society

2550 J . Org. Chem., Vol. 61, No. 7, 1996
Notes
chromatography) is 83%, and the isolated yield of 3 (after
crystallization) is 70%.18
Attempts to carry out the reaction catalytically, using
a catalytic amount of Ni(0) complex and a reducing agent
(Zn, Mg, Na-ascorbate) were unsuccessful. The maxi-
mum yield of 3 and 4 obtained was the same as the molar
percentage of the Ni complex.19 It occurred to us that
nickel powder should serve as a good reducing agent.
Being a solid heterophase, its chemical potential is 1,
whereas the concentration of the (Bu3P)2Ni(II) species in
solution is small. Therefore, the reduction of the dis-
solved Ni(II) species by the solid Ni powder should be
thermodynamically favorable. This reasoning also holds
for the other reducing agents that we have tried to utilize,
but Ni powder does not introduce any other metal (and
therefore different chemistry) to the reaction mixture.
This, indeed, seemed to be a correct approach because
relatively high yields of dibromocyclobutabenzene were
obtained from the reaction containing 1, 5 mol % of
(Bu3P)2Ni(COD), and Ni powder. However, in a control
experiment, it was found that nickel powder can mediate
ring closure in the absence of 2. The drawbacks, relative
to the (Bu3P)2Ni(COD)-mediated reaction, are the lower
total yield (81%) and a lower 3:4 ratio (ca. 90:10). The
advantages, however, are that the nickel powder is much
easier to handle, the reaction can be carried out under
air, and the process is much less expensive than the Ni(0)
complex-mediated reaction. This process can also be
utilized for substituted systems (see below).
F igu r e 1. X-ray structure of 3. See ref 20 for details. The
ellipsoids correspond to 50% probability of electron distribu-
tion. Distances (Å) and angles (deg) (standard deviations in
parentheses): C1-C2 1.375(7), C2-C3 1.391(7), C3-C4
1.407(8), C4-C5 1.393(7), C5-C6 1.398(6), C6-C7 1.507(6),
C7-C8 1.571(8), C7-Br1 1.964(6), C8-Br2 1.951(5), C1-C2-
C3 114.9(5), C2-C3-C4 122.4(5), C3-C4-C5 122.3(4), C4-
C5-C6 114.8(5), C1-C6-C5 122.1(4), C2-C1-C6 123.4(4),
C6-C1-C8 93.2(4), C1-C8-C7 86.2(3), C6-C7-C8 87.1(4),
C1-C6-C7 93.3(4) Br1-C7-C8 114.0(3), Br2-C8-C7 115.2(3).
Ta ble 1. Som e Deta ils of th e Str u ctu r e Deter m in a tion
of 3
chem formula
C8H6Br2
fw 261.94
space group P21/c
(No. 14)
Z ) 4
λ ) 0.710 69 Å µ ) 9.59 mm-1
T ) 125 K
a ) 7.488(2) Å b ) 14.980(4) Å
c ) 8.316(2) Å
γ ) 90°
R ) 90°
â ) 116.07(2)°
V ) 837.9(3) Å3 Fcalcd ) 2.076 g cm-3
2
The X-ray structure of 320 (Figure 1 and Table 1) shows
that the substitution of C(1) and C(2) by Br does not alter
the structure of the carbon skeleton substantially, and
the major geometrical parameters are the same for 3 as
for the nonsubstituted cyclobutabenzene.5a There is no
evidence for bond alternation in the benzene ring, and
the C(1)-C(2)-C(3) angle (114.9°) is significantly de-
creased (relative to benzene) due to the strain implied
onto the benzene ring.
R ) 0.0451
Rw ) 0.0473
w-1 ) (σ2(Fo)+0.0015Fo
)
The fact that 3 (the trans isomer) can be prepared
easily in high yield makes it a potential starting material
for the syntheses of organic systems on the basis of its
Diels-Alder reactivity. There are two stereoselective
requirements for making this reaction useful: the open-
ing of 3 to the o-xylylene and its addition to a dienophile.
In order to check this 3 was reacted with several
dienophiles (Scheme 1). Two of the products clearly
indicate that the opening of 3 is conrotatory, leading only
to 5 as an intermediate, and that the Diels-Alder
addition is exclusively endo. Thus, 621 is the only product
obtained from the reaction of 3 with maleic anhydride
(Scheme 1, i). The reaction of 3 with styrene (Scheme 1,
iv) leads directly to the elimination product 9. The
reaction conditions (benzene as a solvent with no ad-
ditional base except the reactants) suggest that the
elimination of two HBr molecules is very facile, indicating
that the hydrogen and bromine atoms are antiperiplanar.
This arrangement is possible only if the styrene adds
endo to 5. The reaction of 3 with tetracyanoethylene and
phenylacetylene (Scheme 1, ii and iii, respectively) yields
only one product in each case, tentatively assigned as 7
and 8, respectively. Attempts to measure NOE enhance-
ment between H(1) and H(4) of the newly formed six-
membered ring in 8 failed. However, molecular modeling
at the PM3 level suggests that the distance between these
protons is 4.857 Å (4.738 Å at a force field level); thus,
the absence of NOE enhancement is not proof for the
trans arrangement of these protons. We are currently
trying to crystallize the product and obtain its stereo-
chemistry from X-ray analysis.
(15) If (Bipy)Ni(COD) is used instead of (Bu3P)2Ni(COD) the ben-
zonickelacyclopentane can be isolated instead of the dimer. However,
this nickelacycle does not yield the cyclobutabenzene upon reductive
elimination. See: Matsunaga, P. T.; Mavropoulos, J . C.; Hillhouse, G.
L. Polyhedron 1995, 14, 175. See also: Khomik, L. I. Deposited Doc.
1981, SPSTL 836, khp-D81; Izv. Vyssh. Uchebn. Zaved. Kim. Kim.
Tekhnol. 1983, 26, 664. We have tried to mediate the reaction by using
(Depe)Ni(anthracene) instead of (Bu3P)2NiL (L ) COD, anthracene),
but only starting materials were isolated. Thus, no coupling is observed
when a bidentate ligand is used, and a bisphosphine leads only to
dimerization.
(16) Yamamoto, T.; Wakabayashi, S.; Osakada, K. J . J . Organomet.
Chem. 1992, 428, 223.
(17) 1H NMR: 3, 7.39, 7.19 (AA′BB′, 4H), 5.41 (s, 2H, CHBr); 4,
7.43, 7.26 (AA′BB′, 4H), 5.81 (s, 2H, CHBr). See Fraenkel, G.; Asahi,
Y.; Mitchell, M. J .; Cava, M. P. Tetrahedron 1964, 20, 1179.
(18) Mp 52.7-52.8 °C (uncorrected) and GC analysis of the crystals
show that they contain 99.7% 3 and 0.3% 4.
(19) No starting material, however, was recovered, and most
products were Wurtz type.
(20) A crystal of approximately 0.33 × 0.27 × 0.17 mm3 was
measured on a Nicolet R3m/V four circle diffractometer with Mo KR-
radiation (graphite monochromator). The crystal system is monoclinic
and the cell dimensions are refined from the diffractometer angles of
50 centered reflections in the 2Θ-range of 20-25°. Extinction correction
(F* ) Fc/[(1 + 0.002xF2c)/sin 2Θ]0.25; x ) 0.000 59) and empirical
absorption correction in the 2Θ-range of 3-35° has been performed
(max/min transmission: 0.29/0.86, Rmerg before/after correction: 0.199/
0.036). The maximum scan angle in 2Θ was 80°, which led to 2444
unique intensities (Rmerg 0.0329) and 1733 observed intensities (Fo
g
4σ(Fo)). The structure solution with direct methods and the refinement
with full-matrix least-squares was performed with the SHELXTL-Plus
program package (Vers.4.11/V). One hundred sixteen parameters were
We have investigated the scope of the nickel-mediated
cyclization method in two directions. One is the pos-
refined with anisotropic displacement parameters for
C and Br,
hydrogen atoms, located from a difference map and refined without
any constraints, with isotropic displacement parameters. The maxi-
mum residual electron density is 1.32 eÅ-3. Further details are given
in Table 1.
(21) (a) Mueller, P.; Rey, M. Helv. Chim. Acta 1981, 64, 354. (b) Ito,
Y.; Nakatsuka, M.; Saegusa, T. J . Am. Chem. Soc. 1982, 104, 7609.

Notes
J . Org. Chem., Vol. 61, No. 7, 1996 2551
Sch em e 1
sibility to close more than one four-membered ring in a
“one pot” fashion. Indeed, hexabromotricyclobutabenzene
can be prepared in high yield in a single reaction.22 The
second is a study of the generality of the ring closure in
substituted 1. We have attempted few of these reactions;
those reported here are our preliminary results. Equa-
tion 2 describes two examples of substituted systems (10)
reacting with a Ni(0) complex or Ni metal to form the
respective dibromocyclobutabenzenes 11 and 12. In all
cases, the nickel-mediated cyclizations yielded the ex-
pected products. For example, when 10a or 10b was
subjected to cyclization conditions, the total isolated yield
was only ca. 25%, with an 11:12 ratio of 8:1. However,
when 10b was reacted with (Bu3P)2Ni(anthracene) the
yield was 80%, and 11a was shown by NMR to be the
sole product. Metallic nickel cyclizes 10a to 11a and 12a
with a total yield of 38%. The solvent employed is of
prime importance: When the cyclization of 10a is carried
out in THF with metallic nickel and stoichiometric
amount of DMF the yield is only 38% but the only product
is 11a .
3,4-Dim eth ylp h en yl Aceta te. Acetyl chloride (8.5 mL, 0.12
mol) was added dropwise to a cold (0 °C) solution of 3,4-
dimethylphenol (10 g, 0.082 mol) and Et3N (20 mL) in CH2Cl2
(150 mL). The reaction was stirred at rt for 4 h. Water (100
mL) was added, the organic phase separated, washed with HCl
2 M, NaOH 2 M, and water, and dried over MgSO4, and the
solvent removed under reduced pressure. The product was
purified over a Florisil column (hexane) to give 3,4-dimethyl-
phenyl acetate as a light yellow oil (11.0 g, 82%). NMR: 1H δ
Preliminary results suggest that 11 undergoes re-
giospecific Diels-Alder cyclization (eq 3), where only one
of the two possible products (13 or 14) is obtained. Under
3
7.14 (d, 3J ) 8.4 Hz, 1H), 6.93 (s, 1H), 6.89 (d, J ) 8.4 Hz, 1H),
2.30 (s, 3H), 2.28 (s, 6H); 13C (50.4 MHz) δ 169.57, 148.64, 137.74,
133.97, 130.24, 122.40, 118.53, 20.94, 19.70, 19.01.
Meth yl 3,4-Dim eth ylben zoa te. Five equiv of diazomethane
in dry ether was added dropwise to a cold (0 °C) solution of 3,4-
dimethylbenzoic acid in dry CH2Cl2 (40 mL). The reaction was
stirred overnight at rt. The crude product was washed with
NaOH 2 M and water and dried over MgSO4, and the solvent
was removed under reduced pressure to yield the product (a light
yellow oil) which was used as is in the subsequent bromination.
3
3
NMR: 1H δ 7.74 (s, 1H), 7.70 (d, J ) 8.0 Hz, 1H), 7.08 (d, J )
8.0 Hz, 1H), 3.61 (s, 3H), 2.20 (s, 6H); 13C (50.4 MHz) δ 167.17,
142.02, 136.52, 130.53, 129.52, 127.71, 127.05, 51.66, 19.76,
19.45.
Diels-Alder conditions, 12a is completely unreactive and
is fully recovered. We are currently optimizing the
cyclization conditions and investigating the Diels-Alder
reactivity of 11 and 12. Note that 11 and 12 are chiral,
so enatioselective closure of the four-membered ring
using chiral phosphines as ligands to nickel is also under
investigation.
3,4-Bis(d ibr om om eth yl)p h en yl Aceta te (10a ). A solution
of Br2 (6.6 mL, 0.13 mol) in CCl4 (30 mL) was added to an
irradiated refluxing solution (using a 375W sun-light lamp as
the light and heat sources) of (3,4-Dimethyl)phenyl acetate (5 g
30 mmol) in CCl4 (70 mL) during 24h. The reaction mixture
was cooled, washed with 10% Na2S2O3, water, dried over MgSO4
and the solvent was removed under reduced pressure. The
product was filtered on a florisil column (hexane:CH2Cl2 3:1) to
Exp er im en ta l Section
Gen er a l. All solid starting materials were recrystallized
prior to use. THF and DMF were freshly distilled from potas-
sium-benzophenone ketyl and CaH2, respectively, prior to use.
C6D6 and CD2Cl2 were dried and kept on molecular sieves 4A
and vacuum transferred to the NMR tubes which were fire-
sealed. Metallic nickel was Nickel sponge gd.II b2152 (J ohnson,
Matthey and Co. Ltd.) used without any treatment. Alumina
and florisil used for chromatography were Merck neutral
alumina activity I and Riedel-de Hae¨n Florisil, respectively.
Elemental analyses were performed in the Hebrew university
microanalysis laboratory. NMR spectra were recorded on a 400
MHz spectrometer (100.8 MHz for 13C) in CDCl3 unless other-
wise noted. Reported melting points are uncorrected.
give 10a (8 g, 62%). Mp: 109 °C. NMR: 1H δ 7.64 (d, J ) 8.4
3
Hz, 1H), 7.43 (s, 1H), 7.11 (dd, 3J ) 8.7, 2.2 Hz, 1H), 7.06 (s,
1H), 7.04 (s, 1H), 2.31 (s, 3H); 13C δ 168.40, 151.45, 138.96,
134.56, 130.58, 123.53, 122.48, 35.38, 35.31, 21.02. Anal. Calcd
for C10H8Br4O2: C, 25.03; H, 1.68. Found: C, 25.35; H, 1.72.
Meth yl 3,4-Bis(d ibr om om eth yl)ben zoa te (10b). The com-
pound was prepared using the same procedure that was used
for the preparation of 10a with similar yields. Mp: 96.2 °C.
3
NMR: 1H δ 8.26 (s, 1H), 7.99 (d, J ) 8.1 Hz, 1H), 7.78 (s, 1H),
7.15 (s, 1H), 7.06 (s, 1H), 3.93 (s, 3H); 13C δ 165.07, 141.85,
137.54, 131.82, 131.16, 130.72, 130.18, 52.54, 35.45, 35.29. Anal.
Calcd for C10H8Br4O2: C, 25.03; H, 1.68. Found: C, 25.31; H,
1.65.
tr a n s-7,8-Dibr om ocyclobu ta ben zen e (3). (A) Un d er 2b/
DMF Cycliza tion Con d ition s. A solution of 1 (153 mg, 0.36
(22) Stanger, A; Ashkenazi, N.; Bla¨ser, D.; Stellberg, P.; Mauliz, A.;
Boese, R. manuscript in preparation.

2552 J . Org. Chem., Vol. 61, No. 7, 1996
Notes
mmol) in dry and degassed DMF (25 mL) was cannulated under
argon into a Schlenk flask containing 2b (225 mg, 0.39 mmol).
The reaction was stirred at 70 °C for 3 h. The volatiles were
removed under vacuum, and the residue was extracted with
CHCl3, washed with water, dried over MgSO4, and the solvent
was removed under reduced pressure. Filtration on alumina
(hexane:CHCl3 1:1) gave a mixture of 3 and 4 (78.4 mg, 83%
yield) in a ratio of 98.5:1.5. Recrystallization from hexane at
-85 °C gave pure 3 (66 mg, 70% yield).18 (B) Un d er Meta llic
Nick el/DMF Cycliza tion Con d ition s. Nickel (0.3 g, 5.11
mmol) and 1 (490.4 mg, 1.16 mmol) in DMF (25 mL) were stirred
at 85-90 °C for 24 h. The workup was identical to that used
for the 2b-mediated reaction. Obtained were 3 and 4 (246 mg,
81% yield) in a 9:1 ratio. Recrystallization from hexane at -85
°C yielded pure 3 (197 mg, 65% yield). 1H NMR for 3 and 4 are
identical to that reported in ref 17. 3. 13C NMR: δ 142.4, 131.5,
123.1, 49.8. MS: 262 (20), 181 (68), 102 (100). 4. 13C NMR: δ
133.9, 128.2, 51.3 (the quaternary carbon was not detected).
MS: 262 (10), 181 (40), 102 (100).
Reaction of tr a n s-7,8-Dibr om ocyclobu taben zen e (3) with
Ma leic An h yd r id e. A solution of 3 (14 mg, 0.05 mmol) and
maleic anhydride (5.3 mg, 0.05 mmol) in benzene-d6 (0.4 mL)
was fire-sealed in an NMR tube under vacuum (10-2 mbar). The
tube was allowed to stand at 155 °C for 72 h and the reaction
monitored by 1H-NMR spectroscopy. The sole product was 6
when 3 had completely reacted.21 NMR: 1H (200 MHz, C6D6) δ
3
6.82 (AA′BB′, 2H), 6.70 (AA′BB′, 2H), 5.22 (d, J ) 6.0 Hz, 2H),
3.85 (d, 3J ) 6.0 Hz, 2H); 13C (50.4 MHz, C6D6) δ 170.21, 130.16,
128.19, 123.13, 49.94, 29.82. Upon exposure to air 6 is decom-
posed to 2,3-naphthalic anhydride.23
Reaction of tr a n s-7,8-Dibr om ocyclobu taben zen e (3) with
Tetr a cya n oeth ylen e. A fire-sealed NMR tube containing 3
(36.5 mg, 0.14 mmol), tetracyanoethylene (19 mg, 0.14 mmol),
and benzene-d6 (0.4 mL) was prepared as described above. The
mixture was allowed to react at 185 °C for 1 h until complete
disappearance of 3 to give 7 as the only observed product. The
NMR tube was opened and the reaction mixture stirred over
active charcoal. Solvent evaporation and crystallization from
CH2Cl2/hexane at -85 °C yielded 7 (35 mg, 64% yield). Mp 133-
135 °C sub. NMR: 1H (200 MHz, C6D6) δ 6.75 (AA′BB′, 2H),
6.62 (AA′BB′, 2H), 4.64 (s, 2H). 1H (200 MHz, CDCl3) δ 7.67
(AA′BB′, 2H), 7.55 (AA′BB′, 2H), 5.86 (s, 2H); 13C (C6D6) δ 131.08,
130.85, 127.24, 110.04, 109.01, 45.13, 43.01. FTIR (neat): 2196,
tr a n s-7,8-d ib r om o-cyclob u t a [3,4]p h en yl Acet a t e (11a )
an d m eth yl tr a n s-7,8-dibr om ocyclobu ta[3,4])ben zoate (11b)
were prepared using the same procedures that were used for
the preparation of 3.
tr a n s-7,8-Dib r om ocyclobu t a [3,4]p h en yl Acet a t e (11a ).
(A) Un d er 2b/DMF Cycliza tion Con d ition s. 10a (170 mg,
0.354 mmol), 2b (200 mg, 0.35 mmol), DMF (25 mL). 11a (30
mg, 26% yield) was obtained as a light yellow oil. (B) Un d er
Meta llic Nick el/DMF Cycliza tion Con d ition s. 10a (0.6 g,
1.25 mmol), nickel (0.3 g, 5.11 mmol), DMF (25 mL). 11a and
2240 cm-1
. HRMS: calcd for C14H6N4Br2 387.8959/389.8939/
391.8918, found 387.8952/389.8927/391.8896.
Reaction of tr a n s-7,8-Dibr om ocyclobu taben zen e (3) with
P h en yla cetylen e. A fire-sealed NMR tube that had been
prepared as described above, containing 3 (17 mg, 0.06 mmol),
phenylacetylene (7 mg, 0.06 mmol), and benzene-d6 (0.4 mL),
was allowed to react at 155 °C for 27 days. The only observed
product was 8. NMR: 1H (200 MHz, C6D6) δ 7.38 (m, 2H), 6.97
(m, 1H), 6.94 (m, 2H), 6.84 (AA′BB′, 2H), 6.68 (AA′BB′, 2H), 5.68
3
12a (120 mg, 30%) were obtained. NMR: 1H δ 7.22 (d, J ) 8.3
Hz, 1H), 7.11 (d, 3J ) Hz, 1H), 6.98 (s, 1H), 5.75 (cis-dibromo, s,
2H), 5.37 (trans-dibromo, s, 2H), 2.29 (s, 3H); 13C δ 168.99,
153.04, 143.00, 139.33, 125.65, 124.68, 116.89, 48.91, 48.69,
21.06. FTIR (neat): 1772 cm-1. HRMS: calcd for C10H8O2Br2
317.8891/319.8870/321.8850, found 317.8898/319.8874/321.8847.
Anal. Calcd for C10H8Br2O2: C, 37.54; H, 2.52. Found: C, 37.84;
H, 2.60.
3
3
1
(d, J ) 2.0 Hz, 1H), 5.51 (d, J ) 2.0 Hz, 1H), 5.12 (s, 1H); H
(200 MHz, CD2Cl2) δ 7.46 (m, 5H), 7.33 (AA′BB′, 2H), 7.24
3
3
(AA′BB′, 2H), 6.14 (d, J ) 2.0 Hz, 1H), 5.86 (s, 1H), 5.79 (d, J
) 2.0 Hz, 1H); 13C (50.4 MHz, C6D6) δ 132.35, 131.32, 129.12,
128.80,24 128.73, 128.29, 127.85,24 127.80, 127.27,24 126.76,
123.12, 117.64, 77.69, 44.94. Under air 8 slowly decomposes to
9.25
Meth yl tr a n s-7,8-Dibr om ocyclobu ta [3,4]ben zoa te (11b).
(A) Un d er 2b/DMF Cycliza tion Con d ition s. 10b (182 mg,
0.379 mmol), 2b (330 mg, 0.577 mmol), DMF (10 mL). 11b was
obtained as a light yellow oil (24 mg, 20%). (B) Un d er Meta llic
Ni/THF Cycliza tion Con d ition s. A solution of 10b (5 g, 10
mmol) and DMF (3 mL) in THF (150 mL) was refluxed for 5
days with metallic nickel (2.5 g, 40 mmol) under Ar. Water (70
mL) was added, and the THF was removed under reduced
pressure. The crude product was extracted with CHCl3, washed
with water, and dried over MgSO4, and the solvent was removed
under reduced pressure. The product was purified using an
alumina column (hexane:CHCl3 1:1) to give pure 11b (900 mg,
30% yield). (C) Un d er 2a /DMF Cycliza tion Con d ition s. A
solution of 10b (175 mg, 0.365 mmol) and 2a (295 mg, 0.460
mmol) in DMF (10 mL) was stirred overnight at 65 °C. The
volatiles were removed in vacuo, and the residue was extracted
with CHCl3. NMR of the crude shows 80% yield (relative to
anthracene). The reaction was not treated further. NMR: 1H
δ 8.11 (d, 3J ) 8.0 Hz, 1H), 7.86 (s, 1H), 7.26 (d, 3J )8.0 Hz, 1H),
5.39 (s, 2H), 3.89 (s, 3H); 13C δ 166.07, 147.00, 142.48, 133.32,
132.95, 124.68, 123.22, 52.43, 48.76, 48.64. FTIR (neat): 1730
cm-1. HRMS: calcd for C10H8O2Br2 317.8891/319.8870/321.8850,
found 317.8862/319.8883/321.8862.
Reaction of tr a n s-7,8-Dibr om ocyclobu taben zen e (3) with
Styr en e. A fire-sealed NMR tube that had been prepared as
described above containing 3 (22 mg, 0.08 mmol), styrene (9 mg,
0.08 mmol), and benzene-d6 (0.4 mL) was reacted for 31 days at
155 °C. 9 was the only observed product after 3 had completely
reacted.25
Ack n ow led gm en t. We thank Volkswagen Stiftung,
the basic research fund administrated by the Israeli
Academy for Sciences and Humanities, the fund for
promotion of research at the Technion and the Fonds
der Chemischen Industrie, for financial support.
Su p p or tin g In for m a tion Ava ila ble: 1H NMR spectra of
3,4-dimethylphenyl acetate, methyl 3,4-dimethylbenzoate, 11b,
12b, and 7 and 13C spectrum of 7 are available (6 pages). This
material is contained in libraries on microfiche, immediately
follows this article in the microfilm version of the journal, and
can be ordered from the ACS; see any current masthead page
for ordering information.
Met h yl cis-7,8-Dib r om ocyclob u t a [3,4]b en zoa t e (12b ).
Crystallization of a crude reaction product described above
(containing 11b and 12b) from ethanol gave pure 12b. Mp:
131-2 °C. NMR: 1H δ 8.10 (d, 3J ) 8.0 Hz, 1H), 7.87 (s, 1H),
7.27 (d, 3J ) 9.0 Hz, 1H), 5.79 (s, 2H), 3.90 (s, 3H); 13C δ 166.42,
148.02, 143.42, 133.24, 132.97, 124.85, 123.45, 52.70, 50.47,
50.41. HRMS: calcd for C10H8O2Br2 317.8891/319.8870/321.8850,
found 317.8957/319.8903/321.8781.
J O9516852
(23) Cava, M. P.; Van Meter, J . P. J . Org. Chem. 1969, 34, 538.
(24) These signals were obtained by subtracting the C6D6 spectrum
from the spectrum of 8 in C6D6.
(25) (a) Guyot, M.; Mentzer, C. Bull. Soc. Chem. Fr. 1967, 1843. (b)
Wenkert, E.; Michelotti, E. L.; Swindel, C. S.; Tingoli, M. J . Org. Chem.
1984, 49, 4894.