Naphtho[1,2-c:5,6-c]difuran: A Reactive Linker and Cyclophane Precursor

Article abstract of DOI:10.1021/jo034777x

Naphtho[1,2-c:5,6-c]difuran has been isolated in a 9-step synthesis from 2,6-dimethylnaphthalene. It is a highly reactive diene similar in nature to the related isobenzofuran. In Diels-Alder reactions, its intermediate monoadducts are actually less reactive that the parent difuran making possible sequential Diels-Alder reactions with different dienophiles. Reaction with a tethered bis(dienophile) leads to the production of a naphthalenic cyclophane.

Full text of DOI:10.1021/jo034777x

Na p h th o[1,2-c:5,6-c]d ifu r a n : A Rea ctive Lin k er a n d Cyclop h a n e
P r ecu r sor
Michelle E. Thibault, Taunia L. L. Closson, Sidney C. Manning, and Peter W. Dibble*
Department of Chemistry and Biochemistry, University of Lethbridge, Lethbridge AB, Canada T1K 3M4
dibble@uleth.ca
Received J une 6, 2003
Naphtho[1,2-c:5,6-c]difuran has been isolated in a 9-step synthesis from 2,6-dimethylnaphthalene.
It is a highly reactive diene similar in nature to the related isobenzofuran. In Diels-Alder reactions,
its intermediate monoadducts are actually less reactive that the parent difuran making possible
sequential Diels-Alder reactions with different dienophiles. Reaction with a tethered bis(dienophile)
leads to the production of a naphthalenic cyclophane.
The chemistry of isobenzofuran (IBF, 1, Figure 1) and
related compounds is of great interest and has been
reviewed on a regular basis.1-5 The commercially avail-
able diphenylisobenzofuran is used as a trap for reactive
olefins,6,7 as a singlet oxygen scavenger,8 and for a host
of other applications. Isobenzofurans are also of consider-
able theoretical interest and have found application as
intermediates in the synthesis of various natural prod-
ucts.2 Much of the utility of isobenzofurans lies in their
substantial reactivity in Diels-Alder reactions. Recently,
we have been interested in the preparation of molecules
which contain two IBF moieties linked by a phenyl or
biphenyl ring. In an earlier communication9 we reported
the synthesis of naphtho[1,2-c:5,6-c]difuran, 2, and we
now wish to give the full experimental details of its
synthesis and some new results associated with its
reactivity.
F IGURE 1. Isobenzofuran (1), naphtho[1,2-c:5,6-c]difuran (2),
and naphtho[1,2-c]furan (3).
protocol.11 Applying the same approach to the synthesis
of 2 required the preparation of dibromodiol 4 (Scheme
1). This synthesis was achieved by electrophilic bromi-
nation of 2,6-dimethylnaphthalene followed by radical
bromination of the methyl groups. Hydrolysis gave the
dibromodiol 4. Attempts to dimetalate 4 with excess
n-butyllithium failed. The starting material is not very
soluble even before deprotonation by the first two equiva-
lents of butyllithium and the subsequent metal-halogen
exchange would give a species with a significant amount
of negative charge that may not be well-accommodated.
In any event, decomposition appeared to be the result of
attempts to metalate 4 since starting material was not
recovered.
In our earlier work involving the synthesis of related
naphtho[1,2-c]furan, 3, we generated the furan ring via
metal-halogen exchange of an o-bromobenzyl alcohol
followed by reaction with DMF.10 The resulting hemiace-
tal served as a precursor to the furan under conditions
of acid catalysis or could be converted into an acetal that
gave the furan with use of Rickborn’s base-induced
Protection of the alcohols as tert-butyldimethylsilyl
ethers followed by metalation fared no better, decomposi-
tion being the problem again. We chose to oxidize the
alcohols to aldehydes, protect as acetals, and subse-
quently metalate to eliminate the difficulty of charge
buildup. This procedure worked well despite the poor
solubility of dibromobis(acetal) 8 in the metalation reac-
tion. Reduction of the aldehydes followed by cyclization
gave bis(acetal) 11 as a mixture of syn and anti diaster-
eomers.
* Corresponding author.
(1) Rickborn, B. In Advances in Theoretically Interesting Molecules;
Thummel, R. P., Ed.; J .A.I. Press: Greenwich, CT, 1989; Vol. 1.
(2) Rodrigo, R. Tetrahedron 1988, 44, 2093.
(3) Friedrichsen, W. In Houben-Weyl, Methoden der Organtschen
Chemie; Kreher, R., Ed.; Thieme Verlag: Stuttgart, Germany, 1994;
Vol. E6b, pp 163-216.
(4) Wege, D. In Advances in Theoretically Interesting Molecules;
Thummel, R. P., Ed.; J .A.I. Press: Greenwich, CT, 1998; Vol. 4, pp
1-52.
(5) Friedrichsen, W. In Advances in Heterocyclic Chemistry; Katritz-
ky, A. R., Ed.; Academic Press: London, UK, 1999; Vol. 73, pp 1-96.
(6) Pitt, I. G.; Russell, R. A.; Warrener, R. N. J . Am. Chem. Soc.
1985, 107, 7176-7178.
Treatment of bis(acetal) 11 with excess LDA in ether
afforded difuran 2. Unlike isobenzofuran, it is a fairly
stable solid at room temperature and could be chromato-
graphed on alumina. Its NMR spectrum consists of two
doublets for the furan protons (J ) 1.6 Hz) and an AB
quartet for the protons of the carbocyclic ring. The
(7) Torun, L.; Mohammad, T.; Morrison, H. Tetrahedron Lett. 1999,
40, 5279-5282.
(8) Rabek, J . F.; Rånby, B. J . Polym. Sci. 1974, 12, 273-294.
(9) Yu, D. W.; Preuss, K. E.; Cassis, P. R.; Dejikhangsar, T. D.;
Dibble, P. W. Tetrahedron Lett. 1996, 49, 8845-8848.
(10) Smith, J . G.; Dibble, P. W.; Sandborn, R. E. J . Org. Chem. 1986,
51, 3762-3768.
(11) Naito, K.; Rickborn, B. J . Org. Chem. 1980, 45, 4061.
10.1021/jo034777x CCC: $25.00 © 2003 American Chemical Society
Published on Web 10/07/2003
J . Org. Chem. 2003, 68, 8373-8378
8373

Thibault et al.
SCHEME 1
SCHEME 2
coupling constant in the AB system of 9.3 Hz indicates a
high degree of double bond character in this bond, and
is very similar to the 9.2-Hz coupling observed in 1.12 This
observation is consistent with the picture of 1 and 2 as
being mainly polyene in character.
experiment. While performing a competition experiment
between phenanthro[2,3-c]furan and isobenzofuran,13
both species were formed simultaneously by treating a
mixture of their acetal precursors with LDA. NMR
spectra obtained before elimination is complete show that
phenanthro[2,3-c]furan forms more readily than IBF
despite the fact that the former product sacrifices slightly
more resonance energy. This outcome suggests that some
factor, such as the acidity of the acetal CH2, plays a
significant role in the kinetics of elimination rather than
thermodynamic considerations concerning loss of aro-
matic character.
Wege has examined the reactivity of isobenzofuran and
its benzologues in great detail and has determined that
their second-order rate constants give a linear Herndon
relationship against a parameter known as the structure
count ratio.14 The application of this Herndon relation-
ship to difurans has some interesting ramifications.
Scheme 3 shows the stepwise reaction of 2 with 2 equiv
of N-methylmaleimide, 13, and gives the structure of the
intermediate monoaddition adduct, 14. Difuran 2 would
Acetal 11 was treated with a single equivalent of LDA
(Scheme 2) in an effort to selectively eliminate only one
of the two methoxy groups to explore the possibility of
stepwise formation and trapping of the two furan moie-
ties. Both eliminations involve a significant loss of
resonance energy associated with the naphthalenic ring.
The first elimination involves the loss of only four
electrons worth of resonance energy, the second, six.
There was some hope, therefore, that the first elimination
might take place significantly faster than the second. The
result of this experiment was a mixture of starting
material, difuran 2, and the monoelimination product 12
in a ratio of about 2:2:1. The proportions of 2 and 12
suggest that, despite the fact that in the former case the
reactant sacrifices slightly more resonance energy, it is
the second elimination that is faster. We have seen a
similar example of this phenomenon since doing this
(13) Thibault, M. E.; Pacarynuk, L. A.; Closson, T. L. L.; Dibble, P.
W. Tetrahedron Lett. 2001, 42, 789-791.
(14) Moursounidis, J .; Wege, D. Aust. J . Chem. 1988, 41, 235-249.
(12) Chacko, E.; Bornstein, J .; Sardella, D. J . J . Am. Chem. Soc.
1977, 99, 8248.
8374 J . Org. Chem., Vol. 68, No. 22, 2003

Naphtho[1,2-c:5,6-c]difuran
SCHEME 3
SCHEME 4
be expected to be comparable in reactivity to 1. In the
second reaction, however, intermediate 14 contains the
substructure of naphtho[1,2-c]furan (3). The parent
hydrocarbon is a known compound15,16 and Wege has
shown it to be 40 times less reactive than IBF itself. On
this basis, one might predict that reaction of 2 with a
single equivalent of 13 should give intermediate 14 fairly
cleanly and that it would be sufficiently stable to be
isolated. This experiment is somewhat akin to a competi-
tion experiment in that during the course of reaction,
reactant 2 and product 14 (once its concentration builds)
will compete for 13 as it is added. Upon performing this
experiment, intermediate 14 was indeed the only observ-
able product formed. Given the results of this experiment,
it seems very likely that it will be possible to trap difuran
2 with two different dienophiles, something that would
be quite useful.
for all benzodifurans. We are currently working on a
synthesis of phenanthrodifuran 16. In this example, the
difuran should be as reactive as 1, but the intermediate
17 has the substructure of the more reactive benzologue
phenanthro[2,3-c]furan.13 Therefore, in the case of 16, the
second Diels-Alder reaction should be faster than the
first, opposite to the case of 2. This should also be true
for a known isomer of 2:16 naphtho[1,2-c:3,4-c]difuran 18
should have reactivity comparable to that of furan. Its
monoaddition product, however, has 3 as a substructure
and should be much more reactive than 18. Details of
its reactivity have not been published. Sequential diene
reactivity similar to ours has been reported for diene 19.17
Diels-Alder adducts of 2 could be prepared in two
ways (Scheme 4). A mixture of acetal 11 and excess
dimethyl fumarate in the presence of TsOH gave the bis-
(fumarate) adduct as a complex mixture of syn and anti
diastereomers. Presumably, this adduct is formed via
stepwise Diels-Alder reactions via a monofuran inter-
mediate rather than by the intermediacy of 2 itself. The
adduct was smoothly aromatized to chrysene 20 upon
treatment with methanesulfonic acid, demonstrating that
In the case described above, Wege’s data accurately
predict the reactant to be more reactive than the inter-
mediate. It is worth pointing out that this is not the case
(15) Cornejo, J . J .; Ghodsi, S.; J ohnson, R. D.; Woodling, R.;
Rickborn, B. J . Org. Chem. 1983, 48, 3870.
(16) Stringer, M. B.; Wege, D. Tetrahedron Lett. 1980, 21, 3831-
3834.
(17) Pilet, O.; Vogel, P. Helv. Chim. Acta 1981, 64, 2563-2570.
J . Org. Chem, Vol. 68, No. 22, 2003 8375

Thibault et al.
SCHEME 5
2 is a useful precursor to substituted chrysene deriva-
tives. Alternatively, 2 can be generated with LDA and
subsequently treated with a dienophile. This was done
by using 13 as the dienophile giving 15, again as a
mixture of syn and anti diastereomers, exclusively endo.
The endo, endo, syn diastereomer of adduct 15 bears
a topological similarity to molecular rack 21 reported by
Warrener.18-20 Warrener has shown that it is possible to
tether various carbon chains, with and without functional
groups, between the two nitrogen atoms (6 Å apart) to
form alicyclophanes. In a recent paper, an isobenzofuran
was tethered on a rack and was found to be markedly
more stable.20 To make a comparable tethered naphtha-
lenic cyclophane from 2 requires the use of tethered bis-
(dienophile)s to obtain exclusively the desired endo, endo,
syn product. Based on Warrener’s racks, the N-N bond
distance of 5.7 Å in 15 (determined by AM1 modeling)
suggested that two maleimides tethered by three meth-
ylene units (22 Scheme 5) would be suitable for the
formation of a cyclophane, 23. The other likely outcome
of this reaction is the formation of polymer. A similar
reaction using a bis(isobenzofuran) to prepare a cyclo-
phane has been reported.21
Our first attempt at the preparation of 23 was per-
formed under acid-catalyzed conditions. A mixture of 11
and tethered dienophile 22 was refluxed in diethyl ether
in the presence of TsOH. An NMR spectrum of the crude
reaction mixture indicated a mixture of polymer and 23
in a ratio of 3:2. The presence of 23 was indicated by
sharp doublets corresponding to the bridgehead protons
as opposed to broad signals for polymeric or oligomeric
products. Unmistakable, however, was the appearance
of the middle CH2 of the tether as a multiplet at -1.4
ppm. AM1 models of 23 show that the tether is locked
into a staggered conformation such that the protons of
the middle CH2 are pointing directly into the shielding
region of the naphthalenic ring. This upfield multiplet
is unmistakable evidence for the formation of 23. AM1
models of 23 have an N-N bond distance of 4.9 Å, slightly
shorter than the distance of 5.7 Å in the untethered 15.
There is a slight warp to the naphthalene ring and small
distortions in the bridged rings to accommodate this
strain.
F IGURE 2. Intermediate in the formation of 23.
We have found that the best way to maximize the
production of cyclophane over polymer is to add a dilute
solution of difuran 2 to a dilute solution of 22 at -40 °C.
When the reaction is performed in this way, one can
obtain the cyclophane in better yield (29%) with very
little evidence of polymer in the crude NMR spectrum of
the reaction mixture.
In the desired reaction, endo intermediate 24 (Figure
2) must undergo intramolecular Diels-Alder reaction to
form the cyclophane. Intermolecular reaction of 24 with
2, 22, or itself leads to polymer. In the two sequential
Diels-Alder reactions, the first (between 2 and 22)
involves the more reactive diene as discussed above. The
second reaction, while involving a less reactive diene, is
intramolecular. In an effort to determine whether the
first Diels-Alder reaction is faster than the second, we
tried to obtain a proton NMR spectrum of intermediate
24 by mixing solutions of 2 and 22 at -80 °C in
methylene chloride-d2 followed by data acquisition in a
cold probe. Only signals of 23 were observed, indicating
that both reactions are fast, even at -80 °C.
At this point, we have established some interesting
features in the reactivity of naphtho[1,2-c:5,6-c]difuran
and shown that it can be used to prepare chrysene
derivatives and cyclophanes that are comparable to
tethered molecular racks. We are in the process of
preparing several more cyclophanes to see what tether
lengths can be accommodated and what functional groups
might be incorporated. While there is clear evidence of
polymer formation in the reactions of 2 with bis(dieno-
phile)s, we have not attempted to isolate or characterize
any of these materials. Similar polymers have been
reported and the potential utility of 2 has been mentioned
in this context.22
Exp er im en ta l Section
(18) Butler, D. N.; Shang, M.; Warrener, R. N. Chem. Commun.
2001, 2, 159-160.
Solvents were used without additional purification with the
exception of diethyl ether and THF, both of which were
distilled from sodium metal prior to use. Melting points were
(19) Warrener, R. N.; Schultz, A. C.; Houghton, M. A.; Butler, D.
N. Tetrahedron 1997, 53, 3991-4012.
(20) Warrener, R. N.; Shang, M.; Butler, D. N. Chem. Commun.
2001, 1550-1551.
(21) Chiba, T.; Kenny, P. W.; Miller, L. L. J . Org. Chem. 1987, 52,
4327-4331.
(22) Watson, K. A.; Bass, R. G. High Perform. Polym. 2000, 12, 299-
313.
8376 J . Org. Chem., Vol. 68, No. 22, 2003

Naphtho[1,2-c:5,6-c]difuran
performed in open capillaries and are uncorrected. The absence
of 13C data or missing 13C signals is due to poor solubility.
Molecular modeling calculations (MM+ and MNDO) were
performed with HyperChem release 4.5, Hypercube, Inc.,
Waterloo, Ontario, N2L 3X2, Canada. Complex 1H NMR
coupling patterns for 23 were simulated by using MestRe-C23
to determine the coupling constants.
8.41 (d, J ) 9 Hz, 2H); 13C NMR (62.5 MHz, CDCl3) δ 54.4,
104.0, 123.7, 126.5, 127.5, 133.5, 136.8; IR (KBr) 1307, 1141,
1112, 1064, 1054, 961, 827, 709 cm-1; MS (EI) m/e 436 (M +
4, 11), 434 (M + 2, 22), 432 (M+, 11), 405 (46), 403 (97), 401
(45), 359 (21), 357 (48), 355 (21), 75 (54).
Anal. Calcd for C16H18Br2O4: C, 44.27; H, 4.18. Found: C,
44.40; H, 4.29.
2,6-Bis(d im eth oxym eth yl)-1,5-n a p h th a len ed ica r boxa l-
d eh yd e, 9. Bis(acetal) 8 (1.26 g, 2,9 mmol) was added to 200
mL of dry diethyl ether and the slurry cooled to -80 °C under
nitrogen. n-Butyllithium (2.5 mL of a 2.5 M solution, 6.3 mmol)
was added, the solution swas tirred for 5 min and then warmed
to 0 °C for 40 min. A small aliquot quenched with water had
the following proton NMR spectrum: 3.38 (s, 12H), 5.55 (s,
2H), 7.57 (d, J ) 8 Hz, 2H), 7.87 (d, J ) 8 Hz, 2H), 7.92 (s,
2H), idicating complete metalation. DMF (0.7 mL, 10 mmol)
was added and the mixture was stirred for 10 h. Water (50
mL) was added, the ether phase was separated, dried over
MgSO4, and filtered, and the solvent was removed. The residue
was recrystallized from diethyl ether to give 0.83 g (86%) of 8:
1,5-Dibr om o-2,6-bis(br om om eth yl)n a p h th a len e, 6. Di-
bromide 5 (7.3 g, 23 mmol) and N-bromosuccinimide (12.6 g,
71 mmol) were brought to reflux in 200 mL of carbon
tetrachloride. A small amount of benzoyl peroxide was added
and, after 1.5 h, another portion of benzoyl peroxide was added.
Refluxing was continued for a total of 4 h. The mixture was
cooled, saturated NaHCO3 solution (200 mL) was added, and
the mixture was filtered giving a white solid that was then
air-dried to give 7.0 g (64%) of 6, which was used without
further purification. An analytical sample obtained by recrys-
tallization from toluene/hexanes gave the following: mp 255-
1
259 °C; H NMR (250 MHz, CDCl3/DMSO-d6) δ 4.95 (s, 4H),
7.83 (d, J ) 10 Hz, 2H), 8.33 (d, J ) 10 Hz, 2H); IR (KBr)
1312, 1222, 1205, 1149, 819, 728, 699 cm-1; MS (EI) m/e calcd
for C12H8Br4 467.7362, found 467.7358; 472 (M + 4, 25), 470
(M + 2, 15), 468 (M+, 5), 395 (33), 393 (100), 391 (100), 389
(35), 314 (34), 312 (65), 310 (35), 233 (12), 152 (43).
1
mp 132-133 °C; H NMR (250 MHz, CDCl3) δ 3.40 (s, 12H),
5.91 (s, 2H), 7.89 (d, J ) 9 Hz, 2H), 8.98 (d, J ) 9 Hz, 2H),
10.92 (s, 2H); 13C NMR (62.5 MHz, CDCl3) δ 53.9, 102.1, 126.9,
130.5, 130.8, 140.5, 194.0; IR (KBr) 1698, 1351, 1104, 1071,
912, 843 cm-1; MS (EI) m/e calcd for C18H20O6 332.1260, found
332.1263; 332 (M+, 10), 302 (10), 301 (55), 300 (100), 286 (17),
285 (88), 255 (17), 225 (27).
Anal. Calcd for C12H8Br4: C, 30.81; H, 1.72. Found: C,
30.52; H, 1.90.
1,5-Dibr om o-2,6-n a p h th a len ed im eth a n ol, 4. Tetrabro-
mide 6 (6.7 g, 14 mmol) was stirred in 200 mL of dioxane. A
slurry of 31 g (31 mmol) of CaCO3 in 100 mL of water was
added and the mixture refluxed for 48 h. The dioxane was
removed by rotary evaporation and the residue treated with
6 M HCl until acidic. A white solid was filtered from the
mixture, washed well with water, and air-dried to give 4.72 g
(96%) of crude 4, which was used without purification. An
analytical sample was obtained from THF, which had mp 260-
264 °C; 1H NMR (250 MHz, CDCl3/DMSO-d6) δ 4.79 (d, J ) 6
Hz, 4H), 5.58 (t, J ) 6 Hz, 2H), 7.85 (d, J ) 9 Hz, 2H), 8.29 (d,
J ) 9 Hz, 2H); 13C NMR (62.5 MHz, CDCl3/DMSO-d6) δ 61.8,
118.7, 124.2, 125.1, 130.1, 138.4; IR (KBr) 3254, 1482, 1208,
1064, 809, 729 cm-1; MS (EI) m/e calcd for C12H10Br2O2
343.9049, found 343.9039; 348 (20), 346 (M+, 39), 344 (23), 237
(12), 235 (12), 139 (17), 129 (29), 128 (100), 127 (28), 126 (23).
Anal. Calcd for C12H10Br2O2: C, 41.91; H, 2.93. Found: C,
41.86; H, 2.78.
Anal. Calcd for C18H20O6: C, 65.10; H, 6.07. Found: C, 64.89;
H, 6.26.
2,6-Bis(d im eth oxym eth yl)-1,5-n a p h th a len ed im eth a n -
ol, 10. Dialdehyde 9 (1.25 g, 3.8 mmol) was stirred in 100 mL
of methanol. Sodium borohydride (0.6 g, 16 mmol) was added
and the mixture was stirred for 4 h. The reaction was then
heated until homogeneous, cooled, and filtered to give 0.83 g
of 10. The filtrate was stripped of solvent and the residue
recrystallized from methanol to give an additional 0.36 g of
product (total yield 95%): mp 195-197 °C; 1H NMR (250 MHz,
CDCl3) δ 2.52 (t, J ) 7 Hz, 2H), 3.42 (s, 12H), 5.21 (d, J ) 7
Hz, 4H), 5.76 (s, 2H), 7.77 (d, J ) 9 Hz, 2H), 8.34 (d, J ) 9 Hz,
2H); 13C NMR (62.5 MHz, CDCl3/DMSO-d6) δ 52.6, 55.3, 100.9,
123.5, 123.8, 131.9, 133.1, 134.3; IR (KBr) 3412, 1356, 1113,
1065, 1018, 956 cm-1; MS (EI) m/e calcd for C18H24O6 336.1573,
found 336.1577; 336 (M+, 23), 305 (29), 304 (55), 273 (100),
272 (47), 241 (84), 182 (20), 105 (24).
Anal. Calcd for C18H24O6: C, 64.31; H, 7.20. Found: C, 64.12;
H, 7.34.
1,5-Dibr om o-2,6-n a p h th a len ed ia ld eh yd e, 7. To a slurry
of 9 g of crushed molecular sieves and 7.5 g (35 mmol) of PCC
in 250 mL of methylene chloride was added 3.0 g (8.7 mmol)
of diol 4. The solution was brought to reflux for 4 h. After
cooling, an equal volume of diethyl ether was added and the
mixture was filtered through 8 cm of silica gel. The silica gel
was washed with 8 200-mL portions of hot chloroform. The
solution was stripped of solvent, and the residue was recrys-
tallized from chloroform to give 2.0 g (67%) of dialdehyde 7:
3,8-Dim eth oxy-1,3,6,7-tetr a h yd r on a p h th o[1,2-c:5,6-c]-
d ifu r a n , 11. In 30 mL of methanol were refluxed 0.54 g of 10
and a catalytic amount of PPTS for 10 h. Methanol was added
at reflux until the solution was homogeneous. After cooling, a
white solid was collected by filtration that proved to be 11,
0.37 g (84%): 1H NMR (250 MHz, CDCl3) δ 3.47 (S, 6H), 5.46
and 5.61 (AB of ABX, J AB ) 13 Hz, J AX ) 0 Hz, J BX ) 2.5 Hz,
4H), 6.40 (d, J ) 2.5 Hz, 2H), 7.56 and 7.67 (AB, J ) 9 Hz,
4H); 13C NMR (62.5 MHz, CDCl3) δ 54.5, 72.6, 108.9, 121.6,
124.8, 128.0, 135.1, 137.8; IR (KBr) 1383, 1087, 1042, 1001,
958, 805 cm-1; MS (EI) m/e calcd for C16H16O4 272.1049, found
272.1046; 272 (M+, 20), 271 (11), 242 (15), 241 (100), 210 (9),
182 (24), 153 (15), 152 (15).
1
mp 245-249 °C dec; H NMR (250 MHz, CDCl3) δ 8.12 (d, J
) 8 Hz, 2H), 8.61 (d, J ) 8 Hz, 2H), 10.69 (s, 2H); 13C NMR
(62.5 MHz, CDCl3/DMSO-d6) δ 125.0, 127.0, 132.5, 134.4,
190.6; IR (KBr) 1683, 1297, 1213, 935, 820, 765 cm-1; MS (EI)
m/e 344 (M + 4, 50), 343 (36), 342 (M + 2, 100), 341 (51), 340
(51, M+), 313 (11), 204 (27), 126 (32), 125 (27), 124 (28), 74
(28).
Anal. Calcd for C16H16O4: C, 70.63; H, 5.93. Found: C, 70.38;
H, 6.00.
Anal. Calcd for C12H6Br2O2: C, 42.15; H, 1.77. Found: C,
42.30; H, 1.75.
2,3,8,9-Ch r ysen etetr a ca r boxylic Acid , Tetr a m eth yl Es-
ter , 20. Bis(acetal) 11 (100 mg, 0.37 mmol) and dimethyl
fumarate (112 mg, 0.78 mmol) were dissolved in 40 mL of 1:1
diethyl ether/methylene chloride. A catalytic amount of TsOH
was added and the solution refluxed for 4 h. The solvent was
removed and the residue dissolved in 10 mL of methane-
sulfonic acid. After the mixture was stirred for 1 h, water was
added and a solid product was collected by filtration. Recrys-
tallization from methanol gave 109 mg of chrysene 20 (64%):
mp 265-266 °C; Rf 0.38 (5% MeOH in CH2Cl2); 1H NMR (250
MHz, CDCl3) δ 4.02 (s, 6H), 4.04 (s, 6H), 8.14 (d, J ) 9 Hz,
2H), 8.41 (s, 2H), 8.88 (d, J ) 9 Hz, 2H), 9.18 (s, 2H); IR (KBr)
1,5-Dibr om o-2,6-bis(d im eth oxym eth yl)n a p h th a len e, 8.
Dialdehyde 7 (2.0 g, 5.8 mmol) was refluxed for 10 h in 125
mL of methanol and 15 mL of trimethylorthoformate with a
catalytic amount of PPTS. A white solid was obtained from
1
methanol, 2.4 g (95%): mp 183-185 °C; H NMR (250 MHz,
CDCl3) δ 3.44 (s, 12H), 5.86 (s, 2H), 7.83 (d, J ) 9 Hz, 2H),
(23) Cobas, C.; Cruces, J .; Sardina, F. J . MestRe-C Magnetic
Resonance Companion, version 2.3a; Universidad de Santiago de
Compostela, 2000.
J . Org. Chem, Vol. 68, No. 22, 2003 8377

Thibault et al.
1723 (CdO), 1298, 1273, 1222, 1133 cm-1; MS (EI) m/e calcd
for C26H20O8 460.1158, found 460.1162; 460 (M+, 100), 430 (30),
429 (75), 383 (8), 266 (7), 200 (10), 199 (39).
Anal. Calcd for C26H20O8: C, 67.87; H, 4.38. Found: C, 68.07;
H, 4.06.
added 41 mg of N-methyl maleimide (0.36 mmol) in 3 mL of
THF in a dropwise manner. After the mixture was stirred for
1 h, the reaction was stripped, taken up in diethyl ether, dried
over MgSO4, and filtered. After removal of the solvent, the
residue was recrystallized from diethyl ether, giving 52 mg
(44%) of monoadduct 14: mp 195 °C dec; 1H NMR (250 MHz,
CDCl3) δ 2.06 (s, 3H), 3.76 (m, 2H), 5.74 (d, 1H), 6.03 (d, 1H),
7.06 (d, J ) 10 Hz, 1H), 7.27 (d, J ) 8 Hz, 1H), 7.28 (dd, J )
1, 9 Hz, 1H), 7.77 (dd, J ) 1, 8 Hz, 1H), 7.86 (d, J ) 1.5 Hz,
1H), 8.17 (dd, J ) 1, 1.5 Hz, 1H); IR (KBr) 1771 and 1701,
Na p h th o[1,2-c:5,6-c]d ifu r a n , 2. A suspension of 0.205 g
(0.75 mmol) of 11 in 100 mL of dry ether was cooled to 0 °C
under N2. Diisopropylamine (20 µL) was added followed by 3.0
mL of 1.0 M MeLi (3 mmol). After 0.5 h, the reaction was
quenched with water and sufficient ether added to dissolve
all material. The ether layer was separated, dried over MgSO4,
and filtered. Approximately 5 g of neutral alumina was added
and the solvent removed by rotary evaporation. The alumina/
difuran was added to the top of a column of neutral alumina
and eluted with 3% EtOAc/pet ether graded to 8%. The
progress of elution of the difuran was monitored by shining
long-range UV light on the columnsthe difuran fluoresces
blue. Removal of the eluant gave 0.10 g of the difuran, 64%:
1H NMR (250 MHz, CDCl3) δ 7.37 and 7.41 (ABq, J ) 9.2 Hz,
4H), 8.02 (d, J ) 1.5 Hz, 2H), 8.25 (d, J ) 1.5 Hz, 2H); 13C
NMR (62.5 MHz, acetone-d6) δ 119.4 (d, J ) 65 Hz), 122.7 (s),
123.1 (d, J ) 61 Hz), 123.8 (s), 123.9 (s), 136.3 (dd, J ) 5.8,
206 Hz), 137.7 (dd, J ) 5.8, 206 Hz); IR (KBr) 1044, 851, 805,
753 cm-1; UV-vis (c 3.8 × 10-5, cyclohexane) λmax 228 (4.4),
235 (4.5), 245 (4.4), 253 (4.3), 335 (3.3), 352 (3.6), 371 (3.9),
391 (4.0) nm; MS (EI) m/e calcd for C14H8O2 208.05243, found
208.05245; 208.05245 (M+, 100), 180.05701 (15), 179.04961 (9),
152.01679 (22), 151.05412 (26), 150.04666 (18), 104.02605 (17),
75.02340 (8).
Syn a n d An ti Ad d u cts of 2 w ith N-Meth yl Ma leim id e:
N,N′-Dim eth yl 1,4,7,10-Diepoxy-1,2,3,4,7,8,9,10-octah ydr o-
2,3,8,9-ch r ysen etetr a ca r boxylic Acid , Diim id e, 15. Acetal
11 (50 mg, 0.18 mmol) was dissolved in 30 mL of dry ether
under nitrogen and cooled to 0 °C. LDA (1 mL of 2.0 M, 2
mmol) was added and the reaction stirred for 30 min. Water
was added and the ether phase was isolated. The ether phase
was added to a solution of N-methyl maleimide (55 mg, 0.50
mmol) in 25 mL of THF. The reaction was stirred overnight,
taken up in methylene chloride, and washed with brine. The
solution was dried with MgSO4 and filtered and the residue
was recrystallized from methanol to give 55 mg (70%) of 15
as a 1:1 mixture of diastereomers: mp >360 °C; 1H NMR (250
MHz, CDCl3) δ 1.95 (s, 3 H), 2.06 (s, 3 H), 3.76-3.82 (m, 4H),
5.76-5.81 (m, 2H), 6.05-6.11 (m, 2H), 7.17-7.76 (m, 4H); IR
(KBr) 1773 and 1697, 1434, 1296, 1131, 865 cm-1; MS (EI)
m/e calcd for C24H18N2O6 430.1165, found 430.1168; 430 (M+,
1), 319 (12), 208 (100), 209 (17), 111 (5), 69 (10), 55 (11).
Mon oa d d u ct of 2 a n d N-Meth yl Ma leim id e, 14. Acetal
11 (100 mg, 36 mmol) in 50 mL of THF was cooled to -70 °C
under nitrogen. LDA (1.2 mL of a 2.0 M solution, 1.2 mmol)
was added by syringe and the reaction was stirred for 30 min.
An equal volume of water was added and the THF was
removed by rotary evaporation. The residue was extracted with
diethyl ether to a total volume of 75 mL. To this solution was
1435, 1283, 1130, 969, 869 cm-1; MS (EI) m/e calcd for C19H13
-
NO4 319.0845, found 319.08361; 319 (M+, 19), 209 (16), 208
(100), 152 (9), 151 (10), 104 (15), 87 (6).
3-Ca r bon Cyclop h a n e 23. Acetal precursor 11 (263 mg,
0.966 mmol) was dissolved in 40 mL of dry diethyl ether and
cooled to 0 °C under nitrogen. LDA (5.0 mL of a 2.0 M solution
in heptane/THF/ethylbenzene, 10.0 mmol) was added via
syringe and the mixture was stirred for 30 min. The reaction
was quenched with water and extracted several times with
ether. The combined ether extracts were washed once with
brine, dried with MgSO4, and filtered. Meanwhile, a solution
of 228 mg (0.973 mmol) of the 3C tethered bismaleimide 22 in
∼250 mL of THF was cooled to -40 °C. The difuran solution
was added dropwise to the bis(dienophile) solution at -40 °C
and the reaction mixture was stirred overnight, warming to
room temperature. The solvent was removed by using a rotary
evaporator, and the product passed through an alumina
column with CHCl3. The CHCl3 was removed and the product
recrystallized from CHCl3/hexanes to give 124 mg of 23
(29%): mp 225 °C dec; 1H NMR (250 MHz, CDCl3) δ -1.41
2
3
(AA′ of AA′MM′NN′, symmetrical m, J AA′ ) -12 Hz, J AM
)
)
3J AN′ ) J A′M′ ) J A′N ) 4.0 Hz, J AM′ ) J AN ) J A′N′ ) J A′M
3
3
3
3
3
3
13.0 Hz, 2H), 2.59 and 2.80 (MM′NN′ of AA′MM′NN′, sym-
2
2
3
3
3
metrical m, J MM′ ) J NN′ ) -12 Hz, J AM ) J AN′ ) J A′M′
)
3J A′N ) 4.0 Hz, J AM′ ) J AN ) J A′N′ ) J A′M ) 13.0 Hz, 4H),
3
3
3
3
3
3
3.82 and 3.90 (AB of ABMN, dABq, J AB ) 8.0 Hz, J AN ) 5.5
3
3
Hz, J BM ) 5.5 Hz, 4H), 5.90 (M of ABMN, d, J BM ) 5.5 Hz,
3
2H), 6.21 (N of ABMN, d, J AN ) 5.5 Hz, 2H), 7.55 and 7.74
(ABq, J ) 8.0 Hz, 4H); 13C NMR (62.5 MHz, CDCl3) δ 25.7,
35.0, 46.7, 49.6, 80.5, 81.6, 120.5, 125.4, 128.3, 138.8, 141.3,
173.27, 173.34; IR (KBr) 3005, 2949, 1776 and 1701 (CdO),
1458, 1396, 1344, 1220, 1128, 1040, 866; MS (EI) m/e calcd
for C25H18N2O6 442.1165, found 442.1165; 442 (M+, 17), 209
(18), 208 (100), 152 (5), 151 (5), 110 (5), 57 (7), 55 (5).
Ack n ow led gm en t. The authors wish to thank the
Natural Sciences and Engineering Research Council for
funding and the Centre for Molecular Architecture,
Central Queensland University, Queensland, Australia
for analytical support. We are grateful to the members
of the Centre and its directors Ron Warrener and Doug
Butler for many constructive discussions.
J O034777X
8378 J . Org. Chem., Vol. 68, No. 22, 2003