Synthesis of 9-alkylidene-9H-fluorenes by a novel, palladium-catalyzed cascade reaction of aryl halides and 1-aryl-1-alkynes

Article abstract of DOI:10.1021/jo010561o

In the presence of a palladium catalyst and NaOAc, aryl iodides react with 1-aryl-1-alkynes to afford 9-alkylidene-9H-fluorenes in good yields. The products from this reaction are highly dependent on the base employed. This process appears to involve (1)

Full text of DOI:10.1021/jo010561o

7
372
J . Org. Chem. 2001, 66, 7372-7379
Syn th esis of 9-Alk ylid en e-9H-flu or en es by a Novel,
P a lla d iu m -Ca ta lyzed Ca sca d e Rea ction of Ar yl Ha lid es a n d
1
-Ar yl-1-a lk yn es
Richard C. Larock* and Qingping Tian
Department of Chemistry, Iowa State University, Ames, Iowa 50011
larock@iastate.edu
Received J une 4, 2001
In the presence of a palladium catalyst and NaOAc, aryl iodides react with 1-aryl-1-alkynes to
afford 9-alkylidene-9H-fluorenes in good yields. The products from this reaction are highly dependent
on the base employed. This process appears to involve (1) oxidative addition of the aryl iodide to
Pd(0), (2) alkyne insertion, (3) rearrangement of the resulting vinylic palladium intermediate to
an arylpalladium species, and (4) aryl-aryl coupling with simultaneous regeneration of the Pd(0)
catalyst. Consistent with this mechanism is the fact that 9-alkylidene-9H-fluorenes can also be
prepared by the Pd-catalyzed rearrangement of 1,1-diaryl-2-iodo-1-alkenes.
In tr od u ction
Annulation processes are extremely important in or-
ganic synthesis for the construction of heterocycles and
carbocycles. Palladium-mediated annulations are par-
1
ticularly important. Among these processes, the annu-
lation of alkynes, which often involves cascade insertion
processes, has proven a useful route for the synthesis of
a wide variety of heterocycles and carbocycles from
2
simple synthetic building blocks. One unique feature of
organopalladium chemistry is the extraordinary effect
seemingly minor variations in the reaction conditions can
have on the course of the reaction and the yield of the
product. For example, Heck has reported the formation
of a substituted naphthalene 1 from the palladium-cat-
alyzed reaction of iodobenzene and 2 equiv of dipheny-
3
lacetylene when PPh
3
and Et
3
N are used (eq 1). Dyker
reversed the ratio of the reactants but added n-Bu
4
NBr
and K CO as base instead and observed a totally new
2
3
4
product, the substituted phenanthrene 2 (eq 2). Cacchi
has employed the same organic substrates in the pres-
ence of a formate salt and obtained triphenylethylene (3)
(
eq 3).5
(
1) (a) Schore, N. E. Chem. Rev. 1988, 88, 1081. (b) Tsuji, J .
Palladium Reagents and Catalysts; J ohn Wiley & Sons: New York,
996. (c) Pfeffer, M. Recl. Trav. Chim. Pays-Bas 1990, 109, 567. (d)
Ojima, I.; Tzamarioudaki, M.; Li, Z.; Donovan, R. J . Chem. Rev. 1996,
6, 635. (e) Larock, R. C. Palladium-Catalyzed Annulation. J . Orga-
1
9
nomet. Chem. 1999, 576, 111. (f) Larock, R. C. Palladium-Catalyzed
Annulation. In Perspectives in Organopalladium Chemistry for the XXI
Century; Tsuji, J ., Ed.; Elsevier Press: Lausanne, Switzerland, 1999.
(
1
g) Larock, R. C. Palladium-Catalyzed Annulation. Pure Appl. Chem.
999, 71, 1435.
2) (a) Trost, B. M.; Dumas, J . Tetrahedron Lett. 1993, 34, 19. (b)
Zhang, Y.; Wu, G.; Angel, G.; Negishi, E. J . Am. Chem. Soc. 1990, 112,
(
8
1
1
590. (c) Grigg, R.; Loganathan, V.; Sridharan, V. Tetrahedron Lett.
996, 37, 3399. (d) Zhang, Y.; Negishi, E. J . Am. Chem. Soc. 1989,
11, 3454. (e) Lee, G. C. M.; Tobias, B.; Holmes, J . M.; Harcourt, D.
With our own ongoing interests in palladium annula-
tion chemistry,1
e-g
we had reason to reinvestigate the
reaction of iodobenzene and diphenylacetylene. Employ-
ing our more-or-less standard palladium reaction condi-
A.; Garst, M. E. J . Am. Chem. Soc. 1990, 112, 9330. (f) Silverberg, L.
J .; Wu, G.; Rheingold, A. L.; Heck, R. F. J . Organomet. Chem. 1991,
4
09, 411.
(3) Wu, G.; Rheingold, A. L.; Geib, S. J .; Heck, R. F. Organometallics
tions with NaOAc as the base, an unusual 1:1 adduct,
1
987, 6, 1941.
6
9
-benzylidene-9H-fluorene (4), has been observed (eq 4).
(
4) Dyker, G.; Kellner, A. Tetrahedron Lett. 1994, 35, 7633.
This result encouraged us to further investigate this
unusual reaction. Herein, we wish to report our “optimal”
(5) Cacchi, S.; Felici, M.; Pietroni, B. Tetrahedron Lett. 1984, 25,
3
137.
1
0.1021/jo010561o CCC: $20.00 © 2001 American Chemical Society
Published on Web 10/06/2001

Synthesis of 9-Alkylidene-9H-fluorenes
J . Org. Chem., Vol. 66, No. 22, 2001 7373
Ta ble 1. P a lla d iu m -Ca ta lyzed Rea ction of Iod oben zen e
a n d Dip h en yla cetylen e (Eq 4)a
%
isolated
yield
chloride
source
(equiv)
alkyne
entry (equiv)
base
(equiv)
PPh
(%)
3
time
(h)
2
4
1
2
3
4
5
6
7
8
9
0
1
2
2
1
1
1
1
1
1
1
1
1
1
1
NaOAc (2) LiCl (1)
NaOAc (2) LiCl (1)
NaOAc (2) n-Bu NCl (1)
4
NaOAc (2) LiCl (1)
NaOAc (2)
17
30
24
8
24
24
24
24
24
24
48
48
23
37
44
41
37
10
10
10
10
10
10
10
10
10
n-Bu
NaOAc (2) n-Bu
NaOAc (2) n-Bu
NaOAc (2) n-Bu
4
4
4
4
4
4
4
NCl (1)
NCl (1)
NCl (2)
NCl (3)
NCl (1)
NCl (1)
NCl (1)
NRb
62
61
57
42
20
8
reaction conditions for this reaction and our efforts to
extend this reaction to a variety of aryl iodides and 1-aryl-
1
1
1
KOAc (2)
Na CO (2) n-Bu
CO (2) n-Bu
n-Bu
12
30
71
1
-alkynes.
2
3
K
2
3
Resu lts a n d Discu ssion
Our original reaction conditions employing NaOAc as
a
All reactions were run in the presence of 5 mol % Pd(OAc)2 in
DMF at 100 °C. b No reaction.
the base for the reaction shown in eq 4 afforded a 23%
yield of 9-benzylidene-9H-fluorene (4) (Table 1, entry 1).
Obviously, the reaction conditions needed to be optimized
before further testing the scope and limitations of the
reaction. The effect of various bases, the chloride source,
NaOAc was employed as the base in the reaction (Table
1
, entry 7).
This investigation led to the following standard reac-
tion procedure. One equivalent of aryl halide, 1 equiv of
alkyne, 5 mol % of Pd(OAc) , 10 mol % of PPh , 2 equiv
NCl were heated in DMF
2
3
3
and the ligand (PPh ) were therefore examined. The
results are summarized in Table 1.
of NaOAc, and 1 equiv of n-Bu
4
at 100 °C.
As indicated in eq 4, the formation of product 4 is the
result of a 1:1 adduct of iodobenzene and diphenylacety-
lene; therefore, only 1 equiv of diphenylacetylene is
actually required in the reaction. Indeed, we have found
that the yield of the reaction was actually improved when
only 1 equiv of diphenylacetylene was employed in the
reaction (compare entries 1 and 2 of Table 1).
With this standard procedure in hand, we next ex-
plored the scope and limitations of the reaction by first
examining other aryl alkynes. As shown in Table 2,
entries 1-3, the alkynes that have been successful in this
reaction all have an aryl group on one end of the carbon-
carbon triple bond and another sterically hindered group,
such as an aryl, tert-butyl, or similar group, on the other
end (Table 2, entries 1-3). Thus, 1-phenyl-3,3-dimethyl-
In an attempt to further improve the yield of the
reaction, we examined the use of PPh
3
in the reaction.
1
-butyne (Table 2, entry 2) and 4-phenyl-2-methyl-3-
The addition of 10 mol % PPh , which presumably
3
butyn-1-ol (Table 2, entry 3) gave 9-alkylidene-9H-
fluorenes in yields similar to the parent system (Table
behaves as a ligand for palladium, showed a significant
effect on the reaction. Good yields were obtained when
2
, entry 1). 1-Phenylpropyne gave a messy reaction with
the reactions employed only catalytic amounts of PPh
Table 1, entries 4, 5, and 7).
An appropriate chloride source is also important for
3
little or no fluorene evident. The structural features
required of the alkyne can be rationalized by the mech-
anism proposed for the reaction, which will be discussed
later.
(
the reaction. Without chloride present, the reaction
afforded only a 37% yield of 4 (Table 1, entry 5). The
addition of LiCl improved the yield to 41% (Table 1, entry
The product of the reaction of iodobenzene and phenyl-
(trimethylsilyl)acetylene was quite unexpected (Table 2,
4
). The use of n-Bu
improved the yield to 62% (Table 1, entry 7). Different
amounts of n-Bu NCl were employed in the reaction, but
no significant difference in the yield has been observed
Table 1, entries 7-9). One equivalent of n-Bu NCl
appears to be enough.
4
NCl as the chloride source further
entry 4). Fluorene 4 was obtained in 66% yield. There
appear to be two different paths by which the product 4
may be formed. As shown in Scheme 1, the reaction may
involve formation of the anticipated product 17, followed
by desilylation and a subsequent Heck reaction with
iodobenzene to form the final product 4. An alternative
route to compound 4 starts with the desilylation of
phenyl(trimethylsilyl)acetylene, followed by cross-cou-
pling with iodobenzene (Scheme 2). The resultant diphe-
nylacetylene can then react in the usual fashion with
iodobenzene to give the final product 4.
4
(
4
Like much of our previous palladium annulation
chemistry,1
e-g
the choice of base is critical to the reaction.
Without any base, no reaction is observed (Table 1, entry
). Thus, a number of bases have been examined in the
reaction. When NaOAc was used as the base, compound
6
4
was the only product obtained from the reaction (Table
To further examine the scope of this reaction, a variety
of aryl iodides have been employed in the reaction. The
results are summarized in Table 2, entries 5-14. Various
functionally substituted aryl iodides generally work as
well as iodobenzene. The functional group can be either
an electron-donating or an electron-withdrawing group.
Aryl iodides bearing an electron-withdrawing substituent
in the ortho position, such as 2-iodobenzotrifluoride
1
, entries 1-9). However, a mixture of products 2 and 4
was obtained when other bases were employed (Table 1,
entries 10-12). The base KOAc favored formation of the
2 3
fluorene (Table 1, entry 10), while Na CO gave a 3:2
mixture favoring the phenanthrene 2 (Table 1, entry 11).
Using conditions similar to Dyker’s4 (only replacing
4 4
n-Bu NBr with n-Bu NCl), the reaction afforded 2 as the
major product in 71% yield (Table 1, entry 12). This result
is dramatically different from that observed when
(Table 2, entry 5) and 2-iodobenzonitrile (Table 2, entry
6
), afforded the expected E isomers in better yields than
iodobenzene, indicating that steric hindrance is not a
(6) For a preliminary communication, see: Tian, Q.; Larock, R. C.
Org. Lett. 2000, 2, 3329.
major factor in this reaction. The substrate 1-tert-butyl-

7
374 J . Org. Chem., Vol. 66, No. 22, 2001
Ta ble 2. P a lla d iu m -Ca ta lyzed Rea ction of Ar yl Iod id es a n d 1-Ar yl-1-a lk yn esa
Larock and Tian

Synthesis of 9-Alkylidene-9H-fluorenes
J . Org. Chem., Vol. 66, No. 22, 2001 7375
Ta ble 2 (Con tin u ed )
a
Reaction conditions: 5 mol % Pd(OAc)2, 10 mol % PPh3, 1 equiv of alkyne, 2 equiv of NaOAc, and 1 equiv of n-Bu4NCl in DMF at 100
°
C. a The yield is based upon the consumption of iodobenzene.

7
376 J . Org. Chem., Vol. 66, No. 22, 2001
Larock and Tian
Sch em e 1
information to assign the stereochemistry. For example,
the H NMR spectra of compounds 14a and 14b (Table
1
2
8
, entry 12) exhibit doublets for proton H-4 at 8.37 and
.39 ppm. In the Z isomer 14a , the phenyl ring present
on the vinylic carbon C-10 can exhibit an anisotropic
effect8 on the ring that bears the ester group. This
interaction may shield the protons on that ring, and as
a result, the chemical shift of proton H-4 on that ring
should appear at higher field. No such anisotropic
interaction can exist in the E isomer 14b, and the signal
for H-4 should appear at lower field. Therefore, the
configurations of 14a and 14b are tentatively assigned
as Z and E, respectively.
The reactions of 1-tert-butyl-2-iodobenzene (Table 2,
entry 7) and 2-iodoanisole (Table 2, entry 9) deserve
special mention here. Dyker has reported that 1-tert-
butyl-2-iodobenzene alone reacts with a palladium cata-
lyst to give a strained 1,2-dihydrocyclobutabenzene
Sch em e 2
9
derivative, while 2-iodoanisole produces a substituted
1
0
dibenzopyran under the same reaction conditions.
Under our reaction conditions, we only observe fluorene
products (Table 2, entries 7 and 9). Clearly, addition of
an alkyne completely changes the nature of the reaction.
As shown in entry 9 (Table 2), the reaction of 2-iodoani-
sole did not afford a good yield even after a long reaction
time (158 h). GC-MS analysis of the reaction indicated
that a significant amount of the 2-iodoanisole still had
not participated in the reaction. On the other hand,
4-iodoanisole reacted with diphenylacetylene much faster
and also gave a much better yield (Table 2, entry 11).
Ta ble 3. Effects of Tem p er a tu r e a n d Rea ction Tim e on
th e Rea ction of Eth yl 4-Iod oben zoa te a n d
a
Dip h en yla cetylen e
reaction
time (h)
% isolated
yield
ratio
14a /14b
b
entry
T (°C)
1
2
3
4
5
6
7
8
9
80
80
80
4
8
16
48
4
5
10
20
33
10
30
44
44
44
21:79
30:70
37:63
42:58
32:68
40:60
42:58
39:61
40:60
Heteroaromatic iodides have also been examined in the
reaction. The reaction of 3-iodopyridine afforded a good
yield of a mixture of regioisomers and Z/ E isomers (Table
80
100
100
100
100
100
2
, entry 13). All of these isomers are known compounds,
8
16
32
72
and the structural assignments are thus based on the
11
literature. On the other hand, only one isomer 16 was
obtained from the reaction of 2-iodothiophene, although
the yield was low (Table 2, entry 14).
a
All reactions were run in the presence of 5 mol % of Pd(OAc)2,
1
0 mol % of PPh3, 2 equiv of NaOAc, and 1 equiv of n-Bu4NCl in
As one can see from Table 2, in many cases, the E
isomers are the sole or predominant products in the
reaction. However, mixtures are often obtained. Previous
literature has shown that these types of fluorenes
undergo interconversions when heated to 140 °C in
Decalin.1 We, therefore, suspected that the formation
of isomers is due to thermal isomerization of the initially
formed E isomer, which is expected by our mechanism
to be produced in the reaction. This has been proven to
be true by the following experiments.
b
1
DMF at 100 °C. The yield was determined by H NMR spectros-
copy using undecane as an internal standard.
2
-iodobenzene with a highly hindered electron-donating
group also afforded a single stereoisomer, but in a
considerably lower yield presumably due to the sub-
stantial steric hindrance present in the alkyne and the
product (Table 2, entry 7). However, other substrates
with electron-donating groups, such as 2-iodotoluene
1a
(
Table 2, entry 8) and 2-iodoanisole (Table 2, entry 9),
The Z and E isomers 14a and 14b from the reaction of
ethyl 4-iodobenzoate and diphenylacetylene have been
separated by preparative TLC (Table 2, entry 12). When
submitted to the standard palladium reaction conditions,
each pure isomer 14a and 14b gave the same 40:60
mixture of isomers 14a and 14b as that obtained from
the reaction of ethyl 4-iodobenzoate and diphenyl-
produced mixtures of Z and E isomers. Furthermore, aryl
iodides bearing either electron-donating or electron-
withdrawing groups in the para position also afforded
mixtures of Z and E isomers (Table 2, entries 10-12). In
all systems where Z and E stereoisomers have been
produced, the ratio has been approximately 40:60.
The structural assignment of the Z and E isomers is
2
acetylene. However, without Pd(OAc) present, simple
heating of the E isomer 14b for the same period of time
based on 1D and 2D NMR spectroscopy. 2D NOESY
7
spectroscopy is a powerful tool to identify the structure
of these isomers. For example, the 2D NOESY spectrum
of 9 (Table 2, entry 7) clearly shows a cross-peak between
the protons of the tert-butyl group and the vinylic proton
H-10. This confirms that 9 exists in the E configuration.
(8) G u¨ nther, H. NMR Spectroscopy, 2nd ed.; J ohn Wiley & Sons:
New York, 1995; pp 85-93.
(
9) Dyker, G. Angew. Chem., Int. Ed. Engl. 1994, 33, 103.
10) Dyker, G. Chem. Ber. 1994, 127, 739.
(
1
In some cases, the 1D H NMR spectra provide sufficient
(11) (a) Prostakov, N. S.; Moiz, S. S.; Soldatenkov, A. T.; Zvolinskii,
V. P.; Cherenkova, G. I. Khim. Geterotsikl. Soedin. 1971, 1398. (b)
Prostakov, N. S.; Varlamov, A. V.; Anismov, B. N.; Mikhailova, N. M.;
Vasil’ev, G. A.; Zakharov, P. I.; Galiullin, M. A. Khim. Geterotsikl.
Soedin. 1978, 1234.
(7) J eener, J .; Meier, B. H.; Bachmann, P.; Ernst, R. R. J . Chem.
Phys. 1979, 71, 4546.

Synthesis of 9-Alkylidene-9H-fluorenes
J . Org. Chem., Vol. 66, No. 22, 2001 7377
F igu r e 1. Resonance structures for the E isomer 14b.
Sch em e 3
in DMF generated a 12:88 mixture of isomers 14a and
4b . This indicates that Pd(OAc) or perhaps reduced
Pd(0) may play an important role in the isomerization
process and that it is not a simple thermal isomerization.
We have also examined the effects of temperature and
reaction time on the isomerization process. From Table
The mechanism of the isomerization process about the
exocyclic double bond in the fluorene system has not been
fully elucidated. However, examination of the resonance
structures for the E isomer 14b reveals particularly
favorable resonance structures 18 and 19 (Figure 1). In
resonance structure 18, the two π-electrons of the exo-
cyclic double bond are shifted to the five-membered ring
of the fluorene ring system to give a stable 6π-electronic
1
2
3
, one can see that different ratios were observed at
different reaction times. At the beginning of the reaction,
the percent of the E isomer 14b in the mixture was
relatively high, presumably because there has been
insufficient time to effect significant isomerization. After
a period of time (16 h for the reaction at 80 °C or 8 h for
the reaction at 100 °C), the ratio of 14a to 14b levels off
at approximately 40:60 and little further isomerization
is observed (Table 3, entries 7-9). We have also found
that the actual temperature is very important for the
isomerization process. Higher temperatures (100 °C)
accelerate the process. After the same period of time (4
h), the reaction at the higher temperature (100 °C)
afforded a higher percentage of isomer 14a in the mixture
than that at the lower temperature (80 °C) (compare
entries 1 and 5 of Table 3). It appears to take somewhat
more than 16 h at 80 °C or approximately 8 h at 100 °C
for this isomerization to reach equilibrium.
1
2
arrangement, which is similar to that of fulvene.
Resonance structure 19 is also particularly favorable due
to the electron-withdrawing effect of the ester group. We
suggest that, due to the resonance contributions of 18
and 19, the exocyclic double bond may be of low enough
energy to allow thermal isomerization to the Z isomer
14b. We suspect that the Pd(0) catalyst may also form
an olefin complex with this exocyclic double bond further
weakening the double bond and thus accelerating the
isomerization process.
Based on the structure of the products from this
reaction (Table 2) and our present understanding of
organopalladium chemistry, especially the active role of
(12) Lo, D. H.; Whitehead, M. A. Tetrahedron 1969, 25, 2615. (b)
Griffiths, J .; Lockwood, M. J . Chem. Soc., Perkin Trans. 1 1976, 48.

7
378 J . Org. Chem., Vol. 66, No. 22, 2001
Larock and Tian
Sch em e 4
Sch em e 5
Pd(IV) as an intermediate in organopalladium chemis-
try,13 we propose the following mechanism for this
reaction (Scheme 3).
The mechanism appears to involve the formation,
transformation, and reductive elimination of Pd(IV)
intermediates. The oxidative addition of Pd(0) to iodo-
benzene produces an arylpalladium intermediate 20,
which rapidly inserts the alkyne to produce a vinylic
palladium(II) species 21. It is obvious why the alkyne
must have an aryl group at one end of the carbon-carbon
triple bond and an even more hindered group on the other
end, since previous work on the carbopalladation of
alkynes has indicated that this process is controlled by
producing a palladium(IV) intermediate 28 (Scheme 4).
This intermediate can undergo reductive elimination to
either vinylic palladium(II) species 29 or arylpalladium-
(II) intermediate 30, either of which can further cyclize
by oxidative addition or electrophilic aromatic substitu-
tion pathways similar to those shown for the formation
of the fluorene in Scheme 3. Apparently, the presence of
K CO favors formation of palladacycle 27 and the use
2 3
of NaOAc favors the complete migration of palladium
steric effects with the carbon moiety adding to the less
hindered end of the triple bond.1
c,e-g
Intermediate 21 in
from the vinylic position to the aromatic species 24.
turn apparently undergoes oxidative addition to the
neighboring aryl C-H bond to generate a Pd(IV) inter-
mediate 22, which isomerizes to afford a new Pd(IV)
intermediate 23. Reductive elimination of 23 leads to
arylpalladium(II) intermediate 24. It is not clear if either
the former or the latter process is reversible. Intermedi-
ate 24 can further cyclize to the fluorene product 4 by
either of two possible mechanisms. It can undergo single
bond rotation and oxidative addition to the neighboring
phenyl ring to afford new organopalladium(IV) interme-
diate 25. Two consecutive reductive eliminations finally
afford the product 4 and HI and regenerate the Pd(0)
catalyst. Alternatively, arylpalladium intermediate 24
may undergo electrophilic aromatic substitution on the
neighboring arene to directly produce 26, which affords
the final product and Pd(0) by simple reductive elimina-
tion.
14
Indeed, Dyker has proposed the formation of an inter-
mediate like 27 by oxidative addition of vinylic halides
and removal of HI, but he has not previously observed
the apparent migration of palladium from a vinylic
position to an aryl position as required for formation of
the fluorene.
As shown in the proposed mechanism in Scheme 3,
vinylic palladium intermediate 21 is a proposed inter-
mediate in the formation of the fluorene 4. Since this
intermediate should be easily generated by the oxidative
addition of Pd(0) to the corresponding vinylic iodide,
1
,2,2-triphenyl-1-iodoethylene, we might expect to ob-
serve the formation of product 4 from this vinylic iodide
under our usual reaction conditions. Indeed, this turned
out to be true. The fluorene 4 was obtained from the
reaction in 70% yield (eq 5).
The profound effect that the choice of base has on the
products formed in this process (Table 1, entries 9-12)
raises the question as to how formation of the phenan-
threne 2 and flourene 4 are related mechanistically. It
appears that the phenanthrene 2 arises by either cy-
clization of 21 to palladacycle 27 or reductive elimination
of HI from intermediates 22 or 23 to produce the five-
membered ring palladacycle 27, which in turn undergoes
oxidation addition by another molecule of iodobenzene,
Encouraged by this result, we have also examined the
analogous reaction of 1-iodo-2,2-diphenylethylene (eq 6).
(
13) (a) Canty, A. J . Acc. Chem. Res. 1992, 25, 83 and references
therein. (b) Catellani, M.; Frignani, F.; Rangoni, A. Angew. Chem., Int.
Ed. Engl. 1997, 36, 119. (c) Catellani, M.; Chiusoli, G. P.; Costa, M.
Pure Appl. Chem. 1990, 62, 623.
(14) (a) Dyker, G.; Nerenz, F.; Siemsen, P.; Bubenitschek, P.; J ones,
P. G. Chem. Ber. 1996, 129, 1265. (b) Dyker, G.; Siemsen, P.; Sostman,
S.; Wiegand, A.; Dix, I.; J ones, P. G. Chem. Ber. 1997, 130, 261.

Synthesis of 9-Alkylidene-9H-fluorenes
J . Org. Chem., Vol. 66, No. 22, 2001 7379
1
5
Farchan Scientific Co. 3,3-Dimethyl-1-phenyl-1-butyne, 3-io-
dopyridine,16 1-tert.-butyl-2-iodobenzene, 1-iodo-1,2,2-triph-
enylethylene,18 and 1-iodo-2,2-diphenylethylene were pre-
pared according to previous literature procedures.
17
19
Gen er a l P r oced u r e for th e P a lla d iu m -Ca ta lyzed Re-
a ction of Ar yl Iod id es a n d In ter n a l Alk yn es. Palladium
acetate (2.8 mg, 0.0125 mmol), PPh
3
(6.7 mg, 0.025 mmol),
NaOAc (42 mg, 0.50 mmol), n-Bu NCl (70 mg, 0.25 mmol), the
4
aryl iodide (0.25 mmol), the alkyne (0.25 mmol), and 5 mL of
DMF (or appropriate modifications) were placed in a 4 dram
vial, which was heated in an oil bath at 100 °C for the period
of time indicated in Tables 1-3. The reaction mixture was
To our surprise, this reaction provided the annulated
pentafulvene 31 in 69% yield. This is the same product
that Dyker obtained from the same starting material
under similar reaction conditions. This reaction pre-
sumably proceeds as shown in Scheme 5 via key inter-
mediates 32 and 33, instead of forming the anticipated
fluorene product. Intermediate 32 apparently undergoes
reductive elimination of HI to form 33 faster than
rearrangement to the arylpalladium species required to
form the fluorene. Subsequent oxidative addition of the
cooled, diluted with ether, washed with saturated NH
4
Cl, dried
1
4a
over anhydrous MgSO , and filtered. The solvent was evapo-
4
rated under reduced pressure, and the product was isolated
by chromatography or preparative TLC. The following com-
pound is representative of those prepared by this procedure.
All other fluorenes are reported in the Supporting Information.
9-Ben zylid en e-9H-flu or en e (4) (Ta ble 2, En tr y 1). Ob-
tained as a white solid in 62% yield after purification by
2
0
column chromatography (hexanes): mp 73-74 °C (lit. mp
1
7
7
7
1
3-74 °C); H NMR (CDCl
3
) δ 7.06 (dt, J ) 1.2, 7.5 Hz, 1 H),
2
vinylic halide Ph CdCHI to 33 eventually leads to the
fulvene product 31. The intervening steps may involve
reductive elimination to an arylpalladium intermediate
.29-7.50 (m, 6 H), 7.54-7.61 (m, 3 H), 7.70-7.74 (m, 3 H),
.78-7.82 (m, 1 H); 13C NMR (CDCl
) δ 119.6, 119.7, 120.2,
3
24.4, 126.7, 127.0, 127.3, 128.1, 128.2, 128.6, 129.3, 136.5,
3
5 as shown or reductive elimination to a vinylic pal-
2
1
36.6, 136.9, 139.2, 139.5, 141.3 (one sp C missing due to
overlap); IR (CDCl ) 3053, 1490 cm ; HRMS m/z 254.1088
3
ladium intermediate (not shown). Subsequent intramo-
lecular Heck coupling of either species would provide the
fulvene 31.
-1
(calcd for C20H14 254.1096).
The following compounds were prepared using the general
procedure reported earlier, except that no alkynes were
employed.
Con clu sion
9
-Ben zylid en e-9H-flu or en e (4) (Eq 5). Obtained as a
white solid in 70% yield from the reaction of 1-iodo-1,2,2-
The Pd-catalyzed coupling of aryl or heterocyclic
iodides and 1-aryl-1-alkynes in the presence of NaOAc
provides an efficient, general synthetic route to 9-alkyl-
idene-9H-fluorenes. The process proceeds by an unusual
cascade migration/coupling process involving the novel
rearrangement of a vinylic palladium intermediate to an
arylpalladium species. On the basis of the proposed
mechanism, the synthesis of a 9-alkylidene-9H-fluorene
has also been achieved starting with a vinylic iodide.
triphenylethylene18 after purification by column chromatog-
1
13
raphy (hexanes). The melting point and H and C NMR
spectra were identical to those in the literature.2
0
Com p ou n d 31 (Eq 6). Obtained as an orange solid in 69%
19
yield from the reaction of 1-iodo-2,2-diphenylethylene after
purification by column chromatography (hexanes). The melting
1
13
point and H and C NMR spectra match those in the
literature.1
4a
Ack n ow led gm en t. We thank the donors of the
Petroleum Research Fund, administered by the America
Chemical Society, for partial support of this research.
Thanks also go to J ohnson Matthey, Inc., and Kawaken
Fine Chemicals Co., Ltd., for donating Pd(OAc)2.
Exp er im en ta l Section
Gen er a l Meth od s. All 1H and 13C NMR spectra were
recorded at 300 and 75.5 MHz, respectively. Thin-layer chro-
matography (TLC) was performed using commercially pre-
pared 60 mesh silica gel plates (Whatman K6F), and visual-
ization was effected with short wavelength UV light (254 nm)
Su p p or tin g In for m a tion Ava ila ble: Full characteriza-
tion of compounds 5-16 and 1H and 13C NMR spectra for
or a basic KMnO
mL of NaOH (5%) + 300 mL of H
Rea gen ts. All reagents were used directly as obtained
commercially unless otherwise noted. Anhydrous forms of
NaOAc, LiCl, DMF, CH Cl , hexanes, and ethyl acetate were
purchased from Fisher Scientific. Pd(OAc) was donated by
4
solution (3 g of KMnO
4
+ 20 g of K
2
CO
3
+
1
13
compounds 8, 11a ,b, 12a ,b, 16, and 31 (the H and C NMR
5
2
O).
spectra for the other major products can be found in the
6
Supporting Information for the communication ). This material
is available free of charge via the Internet at http://pubs.acs.org.
2
2
J O010561O
2
J ohnson Matthey, Inc., and Kawaken Fine Chemicals Co., Ltd.
Iodobenzene, 2-iodobenzotrifluoride, 2-iodotoluene, 2-iodoani-
sole, 4-iodoanisole, 2-iodothiophene, triphenylphosphine, phe-
nyl(trimethylsilyl)acetylene, and diphenylacetylene were ob-
tained from Aldrich Chemical Co., Inc. Ethyl 4-iodobenzoate
(15) Larock, R. C.; Doty, M. J .; Cacchi, S. J . Org. Chem. 1993, 58,
4579.
(16) Wang, Y. Ph.D. Dissertation, Iowa State University, 1995.
(17) Lesslie, M. S.; Mayer, U. J . H. J . Chem. Soc. 1961, 611.
(18) Miller, L. L.; Kaufman, D. A. J . Am. Chem. Soc. 1968, 90, 7282.
(19) Curtin, D. Y.; Flynn, E. W. J . Am. Chem. Soc. 1959, 81, 4714.
and n-Bu
-Iodobenzonitrile was purchased from Trans World Chemi-
cals, Inc. 4-Phenyl-2-methyl-3-butyn-2-ol was obtained from
4
NCl were purchased from Lancaster Synthesis, Inc.
2
(20) Sprinzak, Y. J . Am. Chem. Soc. 1956, 78, 466.
(21) Miller, R. B.; McGarvey, G. Synth. Commun. 1978, 8, 291.