Microwave-assisted synthesis of cyclopentadienyl-cobalt sandwich complexes from diaryl acetylenes

Article abstract of DOI:10.1021/om7011634

Sealed tube microwave dielectric heating of diaryl acetylenes with cyclopentadienyl cobalt dicarbonyl at elevated temperature in p-xylene provides access to metallocenes in both the cyclobutadiene (Ar₄C ₄CoCp) and cyclopentadienone (Ar₄C₄(C=O)CoCp) families. When compared with the traditional thermal approach, the current method offers dramatically reduced reaction times and, especially with respect to cyclopentadienone complexes, increased yields. In the case of an especially bulky diarylacetylene the microwave approach allows access to a complex that cannot be readily obtained under traditional thermal conditions. An initial microwave-promoted Sonogashira coupling may be employed for in situ generation of the diarylacetylene, although lower yields of the metallocene complexes are ultimately obtained.

Full text of DOI:10.1021/om7011634

Organometallics 2008, 27, 1653–1656
1653
Microwave-Assisted Synthesis of Cyclopentadienyl–Cobalt Sandwich
Complexes from Diaryl Acetylenes
Emily M. Harcourt,† Shifra R. Yonis,† Daniel E. Lynch,‡ and Darren G. Hamilton*,†
Department of Chemistry, Mount Holyoke College, 50 College Street, South Hadley, Massachusetts 01075,
and Faculty of Health and Life Sciences, CoVentry UniVersity, Priory Street, CoVentry CV1 5FB, U.K.
ReceiVed NoVember 19, 2007
Scheme 1. Traditional Thermal Preparation of 18-Electron
Cobalt Sandwich Complexes from Acetylenes
Summary: Sealed tube microwaVe dielectric heating of diaryl
acetylenes with cyclopentadienyl cobalt dicarbonyl at eleVated
temperature in p-xylene proVides access to metallocenes in both
the cyclobutadiene (Ar4C4CoCp) and cyclopentadienone
(Ar4C4(CdO)CoCp) families. When compared with the tradi-
tional thermal approach, the current method offers dramatically
reduced reaction times and, especially with respect to cyclo-
pentadienone complexes, increased yields. In the case of an
especially bulky diarylacetylene the microwaVe approach allows
access to a complex that cannot be readily obtained under
traditional thermal conditions. An initial microwaVe-promoted
Sonogashira coupling may be employed for in situ generation
of the diarylacetylene, although lower yields of the metallocene
complexes are ultimately obtained.
proceeds rapidly and efficiently under microwave conditions
to yield both cyclobutadienyl–CoCp and cyclopentadienon-
e–CoCp complexes. Total conversion efficiencies, based on
consumption of the starting diarylacetylene, exceed 90% in the
optimal case. The approach is illustrated with the preparation
of a number of known and novel metallocenes and with a
representative X-ray structure of a new mononuclear complex.
Rausch and co-workers reported the preparation of 18-electron
cobalt metallocene complexes from the reaction of diaryl
acetylenes with CpCo(CO)2 in 1970.4 With the exception of
the example noted above an essentially unchanged preparative
approach, when CpCo(CO)2 has been the source of cyclopen-
tadienyl–cobalt, has been employed ever since: lengthy reflux
of a p-xylene solution of CpCo(CO)2 with 2 molar equiv of the
appropriate diaryl acetylene 1 (Scheme 1). In this manner a
number of tetraarylcyclobutadienyl–CoCp complexes 2 have
been prepared, alongside typically smaller production of the
corresponding tetraarylcyclopentadieneone–CoCp complexes 3
that result from CO insertion during complex formation.5
Rausch’s initial report quotes (for R ) phenyl) approximately
50% and 10% yields of 2 and 3, respectively. A more recent
application of this methodology has realized yields of 2 bearing
a variety of aryl substituents of up to 75%.6 It was against this
background that we approached the use of microwaves to
promote complex assembly.
The application of microwave irradiation to chemical trans-
formations has received a great deal of attention in recent years.1
Numerous important classes of organic reactions have been
examined that benefit from this approach, with issues of time
and energy economy and waste minimization adding further to
the technique’s appeal. Among this wealth of investigative effort
the preparation of organometallic compounds has received
comparatively little attention,2 while the use of microwaves to
promote many important metal-catalyzed or metal-mediated
organic transformations has met with great success.1
A highly pertinent, yet rare, example of the preparation of
an organometallic compound using a microwave protocol was
published during prosecution of the work described here.
Richards and co-workers have reported a number of intramo-
lecular acetylene cycloadditions, mediated by cyclopentadienyl
cobalt dicarbonyl (CpCo(CO)2), leading to exclusiVe cyclopen-
tadienone–CoCp complex formation.3 It is explicitly noted in
this work that none of the corresponding cyclobutadienyl
complexes were obtained. Contrasting with this outcome we
report here that intermolecular cobalt-mediated cycloaddition
of diarylacetylenes and capture by cyclopentadienyl cobalt
A 2:1.3 ratio of diphenylacetylene 1a (3 mmol scale) and
CpCo(CO)2 was added to 3 mL of p-xylene in a capped
microwave reactor tube. The tube was maintained at 175 °C
via microwave irradiation for 10 min, after which thin-layer
* To whom correspondence should be addressed. E-mail: hamilton@
mtholyoke.edu.
† Mount Holyoke College.
(4) (a) Rausch, M. D.; Genetti, R. A. J. Org. Chem. 1970, 35, 3888–
3897. This work dramatically improved access to CpCoC4Ph4, thus allowing
full investigation of its chemistry, especially its aromatic reactivity. (b)
Selected cyclobutadienyl–CoCp complexes may also be obtained via a more
complex route, but one that avoids acetylene cyclodimerization: Yamazaki,
H.; Wakatsuki, Y. J. Organomet. Chem. 1978, 149, 377–384.
(5) For diverse examples of the preparation and utility of these classes
of complex see: (a) Nguyen, H. V.; Butler, D. C. D.; Richards, C. J. Org.
Lett. 2006, 8, 769–772(asymmetric organic synthesis). (b) Waybright, S. M.;
Singleton, C. P.; Wachter, K.; Murphy, C. J.; Bunz, U. H. F. J. Am. Chem.
Soc. 2001, 123, 1828–1833(oligonucleotide chemistry). (c) Johannessen,
S. C.; Brisbois, R. G.; Fischer, J. P.; Grieco, P. A.; Counterman, A. E.;
Clemmer, D. E. J. Am. Chem. Soc. 2001, 123, 3818–3819(supramolecular
chemistry).
‡ Coventry University.
(1) (a) For a comprehensive overview see: MicrowaVes in Organic
Synthesis; Loupy, A., Ed.; Wiley-VCH: Weinheim, 2002. (b) For a review
of recent advances in the field see: Kappe, C. O. Angew. Chem., Int. Ed.
2004, 43, 6250–6284.
(2) (a) Ardon, M.; Hogarth, G.; Oscroft, G. T. W. J. Organomet. Chem.
2004, 689, 2429–2435. (b) Whittaker, A. G.; Mingos, D. M. P. J. Chem.
Soc., Dalton Trans. 2002, 21, 3967–3970. (c) VanAtta, S. L.; Duclos, B. A.;
Green, D. B. Organometallics 2000, 19, 2397–2399. (d) Dabirmanesh, Q.;
Roberts, R. M. G. J. Organomet. Chem. 1997, 542, 99–103. (e) Dabir-
manesh, Q.; Fernando, S. I. S.; Roberts, R. M. G. J. Chem. Soc., Perkin
Trans. 1 1995, 743–749. (f) Mingos, D. M. P.; Baghurst, D. R. Chem. Soc.
ReV. 1991, 20, 1–47.
(3) Taylor, C. J.; Motevalli, M.; Richards, C. J. Organometallics 2006,
25, 2899–2902.
(6) Harrison, R. M.; Brotin, T.; Noll, B. C.; Michl, J. Organometallics
1997, 16, 3401–3412.
10.1021/om7011634 CCC: $40.75
2008 American Chemical Society
Publication on Web 03/05/2008

1654 Organometallics, Vol. 27, No. 7, 2008
Notes
Table 2. Results of Microwave Reactions of a Variety of
Scheme 2. Microwave-Assisted Assembly of 18-Electron Cobalt
Sandwich Complexes from Diarylacetylenes
a
Diarylacetylenes with CpCo(CO)2
diarylacetylene
R group
phenyl
4-pyridyl
4-methylphenyl
4-fluorophenyl
2-thienyl
yield 2 (%)b yield 3 (%)b
1a
1b
1c
1d
1e
1f
52
3
40
85
10
20
44
4c
14
28
33
1-napthyl
26c
1g
2,4,6-trimethylphenyl no reaction
no reaction
Table 1. Exploration of the Sealed Tube Reaction of
Diphenylacetylene with CpCo(CO)2
a 3 mmol of diarylacetylene/p-xylene/175 °C/10 min. b Isolated
yields, averaged over a minimum of three reactions. c Isolated yields
after 1 h.
temp yield 2a yield 3a
entry solvent methoda scaleb
time
(°C)
(%)c
(%)c
1
2
3
4
5
6
7
8
9
p-xylene MW
3 mmol 10 min 175
3 mmol 10 min 175
3 mmol 30 min 120f
3 mmol 60 ming 138
3 mmol 10 min 200
52d
37
21
45
48
40d
25
17
35
38
constraints of the reaction vessel) with only a small decrease in
product yield, affording approximately 1 g amounts of each
complex (entry 8).
Decaline MW
DME
MW
p-xylene MW
p-xylene MW
p-xylene MW
p-xylene MW
p-xylene MW
The final two entries in Table 1 (9 and 10) detail attempts to
draw as direct as possible a comparison between the microwave
reactions and a traditional thermal approach. These reactions
were prepared in an identical manner to those destined for
microwave irradiation, using microwave-dedicated reaction tubes
and caps, and were then maintained at 175 °C in an oil bath.
The 10 min reaction afforded around 30% yields of each
complex. Periodic TLC analysis of the lengthier reaction
revealed diphenylacetylene consumption to be complete after
24 h, while isolated yields of each of the product complexes
rose by roughly an additional 10%. While the yields after 10
min for the traditionally heated reaction are certainly lower than
those for the corresponding microwave reaction, the difference
is not such to suggest that any microwaVe-specific effect is in
operation. Rather, microwave irradiation appears to provide a
moderate enhancement of yields of the desired complexes via
simultaneous heating of the whole reaction volume, in contrast
to conventional heating methods, where heat is transferred to
the reacting species from the surface of the vessel.10
3 mmol
2 min 175
20
14
3 mmol 120 min 175
<10h
44
<10h
35
12 mmol 20 mini 175
p-xylene oil bathj 3 mmol 10 minj 175
30
42
35
44
10 p-xylene oil bathj 3 mmol 24 h
175
a MW ) microwave reaction. b With respect to diphenylacetylene.
c Based on consumption of diphenylacetylene. d Averaged over three
reactions. e cis/trans isomeric mix. f A higher temperature was not
explored for safety reasons. g Monitored at 10 min intervals by TLC for
consumption of diphenylacetylene. h Decomposition materials made clean
product isolation problematic. i Complete consumption of diphenylacetylene
required a doubling of the reaction time. j 2 min were allowed for
equilibration of the reaction to the bath temperature.
chromatographic analysis revealed complete consumption of the
starting acetylene and confirmed the presence of complexes 2a
and 3a (Scheme 2). After flash chromatographic isolation 2a
and 3a were obtained in analytically pure form in yields of
53% and 45%, respectively (see Experimental Section). The
initial choices of solvent, reaction temperature, and reaction
duration subsequently proved optimal. The results of multiple
iterations of this benchmark reaction (entry 1) are summarized
in Table 1 and reveal the following: (1) employment of the high
boiling point hydrocarbon solvent Decalin led to small reduc-
tions in the yields of both complexes (entry 2),7 while use of
1,2-dimethoxyethane (DME)8 led to further yield decreases, even
after extending the reaction time to 30 min (entry 3);9 (2)
reducing the reaction temperature to the boiling point of p-xylene
(138 °C) considerably extended the time required for consump-
tion of the starting acetylene but did not otherwise substantially
affect the final yield of complexes, nor their relative ratio (entry
4); (3) further elevation of the reaction temperature (to 200 °C)
did not noticeably alter the reaction outcome (entry 5); (4) both
far shorter (2 min; entry 6) and far longer (2 h; entry 7) reaction
times led to substantial reductions in the yields of both
complexes for reasons of incomplete reaction and product
degradation, respectively; (5) the reaction scale could be
increased by a factor of 4 (the maximum possible due to volume
Notably, these sealed tube syntheses, whether microwave
irradiated or thermally heated, realize far higher yields of
cyclopentadienone complex 3a than was observed in the original
studies of Rausch (10%).4 This outcome may be rationalized
as a result of the relatively high local concentrations of CO
generated during these sealed tube transformations, in contrast
to reactions where the CO is allowed to escape under a nitrogen
flow. While approaches have been developed to selectively
prepare cyclobutadienyl–CoCp complexes from diarylacety-
lenes,11 there exists no corresponding preparative approach to
cyclopentadienone–CoCp metallocenes. The production of
comparable amounts of both complexes (for certain diarylacety-
lenes) under the current method may therefore be viewed as a
reasonable synthetic compromise, especially given the ready
separation of the products.
Under the optimal microwave reaction conditions (i.e., Table
1, entry 1) diarylacetylenes 1b–f are each converted into a
mixture of the two product complexes, although none of these
reactions approach the overall efficiency of conversion for
diphenylacetylene. The results of these experiments are sum-
marized in Scheme 2 and Table 2 (the results for dipheny-
(7) Decalin was investigated, as this was the solvent employed for the
previously reported microwave-assisted preparations of cyclopentadienone
cobalt metallocenes (see ref 3).
(8) DME was investigated, as this solvent has proven particularly
effective in reactions involving cobalt-mediated acetylene cycloadditions:
Boñaga, L. V. R.; Zhang, H.-C.; Gauthier, D. A.; Reddy, I.; Maryanoff, B.
E Org. Lett. 2003, 24, 4537–4540.
(10) For a critical overview of the origin of differences in reaction
outcome between microwave and traditional heating approaches, see: De
la Hoz, A.; Díaz-Ortiz, Á.; Moreno, A. Chem. Soc. ReV. 2005, 34, 164–
178.
(9) Reactions in DME were maintained at temperatures of no more than
120 °C for reasons of safety. Inevitably rates of production of the desired
metallocenes would be reduced at this temperature when compared with
reactions in either p-xylene or Decalin. However, lengthier reaction times
did not produce comparable yields to reactions conducted in either of these
hydrocarbon solvents.
(11) (a) Stevens, A. M.; Richards, C. J. Organometallics 1999, 18, 1346–
1348. (b) Uno, M.; Ando, K.; Komatsuzaki, N.; Tsuda, T.; Tanaka, T.;
Sawada, M.; Takahashi, S. J. Organomet. Chem. 1994, 473, 303–311.

Notes
Organometallics, Vol. 27, No. 7, 2008 1655
aromatic substituents of the cyclopentadienone ligand occupy
constrained environments in this complex. They cannot readily
rotate into downward (i.e., Cp facing) positions because of
unfavorable interactions with the Cp ring, nor can they adopt
orientations that would bring them into coplanarity with the
cyclopentadienone ligand. The result is that all four naphthyl
units are directed away from the vertical axis of the complex,
and the strikingly well-resolved proton NMR spectrum presented
by this material argues against the presence of significant barriers
to rotation about the 1-naphthylcyclopentadienone bonds. In
stark contrast, the corresponding cyclobutadiene complex offers
evidence of severe structural constraints. In fact, the very low
yield of 2f reported in the table appears to represent a mixture
of atropisomers resulting from hindered rotation of the naphthyl
substituents about their 1-naphthylcyclobutadiene bonds. Proton
NMR analysis reveals the cyclobutadienyl complex containing
fractions obtained from an initial chromatographic purification
to be composed of a minimum of five materials of very similar
polarity. However, these individual components could not be
satisfactorily separated by additional chromatography (see
Supporting Information).16
Figure 1. Ball-and-stick representation of the X-ray structure of
metallocene 3f.
lacetylene are included for comparative purposes).12 No attempt
was made to optimize yields in each individual case in the
manner performed for diphenylacetylene, although periodic TLC
of reactions involving di(1-naphthyl)acetylene (1f) revealed
comparatively slow consumption of starting material and
prompted an increase in reaction time to 1 h. All materials,
whether previously prepared (2a, 3a, 2b, 3b, and 2d) or novel,
gave spectroscopic data in accordance with their proposed
structures (see Supporting Information).13
For R ) 4-fluorophenyl, 2-thienyl, or 1-naphthyl complexes
in both series are obtained in reasonable yield (with the
exception of 2f, vide infra). For R ) 4-pyridyl the outcome of
the microwave experiment (2b, 3%; 3b, 85%) is even more
extreme than that obtained under a standard thermal protocol
(2b, 15%; 3b, 45%),6 indicative of a favorable pathway for CO
inclusion in this specific case. The total failure of dimesity-
lacetylene (mesityl ) 2,4,6-trimethylphenyl) to produce either
complex 2g or 3g indicates that this degree of steric crowding
of the carbon–carbon triple bond entirely shuts down the
metallocene formation pathways. Complex formation from this
particular acetylene is also known to fail under a standard
thermal protocol.14
The reaction of di(1-naphthyl)acetylene with CpCo(CO)2
appeared a sensible choice to further compare the microwave
and traditional thermal approaches: a preparative advan-
tage conferred by the unique heating mode offered by micro-
wave irradiation would likely be highlighted in a case where
the product complexes are sterically constrained (implying
significant energetic barriers to formation).17 From a sealed tube
reaction maintained at 175 °C for 10 min in an oil bath only
traces of complexes 2f and 3f could be detected by TLC.
Production of the desired complexes did not increase with
prolonged reaction time.18 The whole volume heating conferred
by microwave irradiation therefore offers, in this specific case,
a viable preparative route to a metallocene product that cannot
be conveniently prepared by a traditional thermal route.
As a possible extension of the utility of microwave-assisted
chemistry in the preparation of these complexes, it was
intriguing to explore whether the diarylacetylene component
could be prepared as an immediate prelude to metallocene
formation. The Sonogashira reaction, coupling of an aryl halide
to a terminal acetylene, has itself previously been adapted for
the application of microwaves: an approach using catalytic
copper iodide with cesium carbonate in a high-boiling solvent
appeared potentially compatible with the conditions described
The yield of cyclopentadienone complex 3f, derived from
di(1-napthyl)acetylene, appears impressive for such a crowded
metallocene. This material crystallized readily, and the structure
reveals the expected cobalt sandwich complex of cyclopenta-
dienyl and cyclopentadienone ligands buried at the center of a
propellor-like array of 1-naphthyl substituents (Figure 1).15 The
(16) Hindered rotation about the 1-naphthylcyclobutadiene bonds would
result in up (away from the Cp ring) or down (toward the Cp ring)
orientations. Accordingly, a total of six atropisomers are possible for 2f:
all up, three up/one down, two up/two down (two possibilities), one up/
three down, and all down. Proton NMR analysis of the cyclobutadienyl
complex containing fractions revealed five resonances of varying intensity
attributable to Cp ring protons in the range 5.28–4.91 ppm, and an extremely
complex aromatic region. Integration of the entire Cp proton versus aromatic
region gave an excellent fit to the required ratio (5:28). Potential simplifica-
tion of this putative mixture of atropisomers was not observed on heating
at high temperature (d6-DMSO, 140 °C). Preparative TLC did not allow
for separation of any individual isomer, but did allow two components to
be isolated from the remainder (see Supporting Information).
(17) An additional bulky diarylacetylene, di(9-phenanthrenyl)acetylene,
was also prepared. This material failed to provide any evidence of
metallocene complex production from reactions with CpCo(CO)2 under
microwave conditions, lending credence to the view that 1-naphthyl
substituents are at the upper limit of those aryl groups that may be
successfully incorporated.
(12) All quoted yields are the average of a minimum of three independent
runs and are based on conversion of the respective starting acetylene to
complexes 2 and 3. These reactions also typically produced very small
amounts of hexasubstituted benzene derivatives, the products of metal-
catalyzed cyclotrimerization of the parent acetylene, and non-cobalt-
complexed tetraarylcyclopentadieneones. The presence of these materials
was confirmed by TLC comparison with, where available, authentic samples,
but the amounts were generally too small for practical isolation.
(13) Complexes 2a and 3a have long been known (see ref 4a), while
2b and 3b have also been previously prepared (see ref 6). All other materials
reported are new, with the exception of cyclobutadiene complex 2c,
which has previously been prepared by an alternative route: Wang, H.;
Tsai, F.–Y.; Nakajima, K.; Takahashi, T. Chem. Lett. 2002, 6, 578–579.
(14) Rausch, M. D.; Westover, G. F.; Mintz, E.; Reisner, G. M.; Bernal,
I.; Clearfield, A.; Troup, J. M. Inorg. Chem. 1979, 18, 2605–2615.
(15) Crystal data for 3f: a ) 12.4176(4) Å, b ) 21.1018(6) Å, c )
14.7313(4) Å, ꢀ ) 114.4950(10)°, V ) 3512.68(18) Å3, M ) 708.69,
R1 ) 0.0614. CCDC entry 643536.
(18) The small amount of complex 2f produced in this reaction could
be isolated by column chromatography and corresponded to a yield of 1–
2%. While TLC analysis also indicated the presence of complex 3f, this
material could not be separated from a number of additional products of
similar polarity.

1656 Organometallics, Vol. 27, No. 7, 2008
Notes
Scheme 3. Microwave-Promoted Assembly of Complexes 2a
and 3a via in Situ Generation of Diphenylacetylene
higher than the corresponding conventional reactions, while
simultaneously both temperature and pressure are significantly
enhanced. The result of the combination of these effects is to
provide a very convenient route to a variety of cobalt metal-
locenes, at least one of which cannot be effectively prepared
by traditional means. The reaction outcomes do not point to
the operation of a specific microwave effect, but product yields
are enhanced over traditional thermal reactions as a result of
the whole volume heating conferred by microwave irradiation.
here for cobalt complex formation.19 Accordingly, a single tube
sequential reaction was performed where diphenylacetylene,
formed from an initial microwave-promoted Sonogashira cou-
pling, was used without isolation or purification as the acetylene
source for metallocene formation. Scheme 3 summarizes this
sequential approach. While yields of the eventual product
complexes 2a and 3a are considerably lower than those obtained
from direct inclusion of diphenylacetylene, this one-tube ap-
proach is clearly effective and augers well for development of
other metallocene syntheses from simple aromatic starting
materials.20
Experimental Section
Example Preparative Procedure. To a mixture of anhydrous
p-xylene (3 mL) and diphenylacetylene (0.48 g, 2.8mmol) in
a capped, thick-walled, microwave reactor tube was added
CpCo(CO)2 (0.34 g, 0.25 mL, 1.9 mmol) via syringe. The tube
was then heated at 175 °C for 10 min with stirring in a microwave
reactor (ramp time to the target temperature varied from 1 to 2
min at full power, with 0–50 W power required to maintain the
target temperature throughout the reaction period). After cooling
to room temperature the reaction mixture was dissolved in
chloroform (10 mL) and purified by flash chromatography on silica
gel to yield 2a (0.36 g, 53%) and 3a (0.32 g, 45%). See Supporting
Information for full spectroscopic data. Cautionary note: the high
temperatures and pressures involved in these reactions demand that
only dedicated microwave reactors with accurate temperature
feedback controls, and their matching thick-walled reaction tubes,
be used.
Detailed examinations of parallel microwave and conventional
heating approaches are complicated by the fact that direct
comparisons are frequently difficult to establish.10 For the
reactions described here reactant concentrations are typically
(19) He, H.; Wu, Y.-J. Tetrahedron Lett. 2004, 45, 3237–3239.
(20) Procedure: Cesium carbonate (650 mg, 2 mmol), copper(I) iodide
(20 mg, 0.1 mmol), NMP (1.35 mL), iodobenzene (204 mg, 110 µL, 1
mmol), and phenylacetylene (106 mg, 120 µL, 1.06 mmol) were added, in
that order, to a microwave reactor tube. The mixture was held, with stirring,
at 195 °C for 30 min (ramp time to target temperature ≈ 10 min). The tube
was cooled to room temperature, and through the tube cap septum were
then added anhydrous p-xylene (1.6 mL) and CpCo(CO)2 (162 mg, 120
µL, 0.9 mmol); the mixture was heated, with stirring, at 175 °C for an
additional 10 min. After final cooling the reaction mixture was worked up
and purified according to the example preparative procedure to afford
complex 2a (50 mg, 21%) and complex 3a (30 mg, 12%). Yields quoted in
Scheme 3 are the average of three independent runs at this scale. No attempt
was made to fully optimize solvent or temperature conditions, nor reaction
times.
Acknowledgment. We thank the Donors of the American
Chemical Society Petroleum Research Fund for support of
this research (Awards 38308-B3 and 45312-B3).
Supporting Information Available: General experimental
details, preparative and isolation procedures, and selected spectro-
scopic data for 1b–g, 2a–f, and 3a–f. CIF file for 3f. This material
is available free of charge via the Internet at http://pubs.acs.org.
OM7011634