A soluble fluorous palladium complex that promotes heck reactions and can be recovered and reused

Article abstract of DOI:10.1055/s-2004-829094

A new fluorous SCS pincer palladium complex is synthesized and shown to efficiently promote typical Heck reactions under microwave or thermal heating. The complex is air stable and can be recovered after reactions for reuse by fluorous solid phase extraction. By analogy to related complexes, it may function as a precursor of soluble ligandless palladium metal.

Full text of DOI:10.1055/s-2004-829094

LETTER
1379
A Soluble Fluorous Palladium Complex that Promotes Heck Reactions and
Can Be Recovered and Reused
A
S
oluble Fluorou
e
s
Palladium
n
Complexth
n
at Promotes
i
H
eck
R
s
eactions P. Curran,* Karin Fischer, Gustavo Moura-Letts
Department of Chemistry, University of Pittsburgh, Pittsburgh, PA 15260, USA
Fax +1(412)6249861; E-mail: curran@pitt.edu
Received 8 March 2004
Dedicated to Professor Clayton Heathcock
tions and the solid phase extraction conditions. Tridentate
Abstract: A new fluorous SCS pincer palladium complex is syn-
thesized and shown to efficiently promote typical Heck reactions
under microwave or thermal heating. The complex is air stable and
can be recovered after reactions for reuse by fluorous solid phase
‘pincer’ palladacycles like 1 are robust complexes that can
be used for Heck and Suzuki reactions,9 so easily separa-
ble variants that do not sacrifice reactivity could be very
extraction. By analogy to related complexes, it may function as a useful. Bergbreiter and coworkers have reported that sul-
precursor of soluble ligandless palladium metal.
fur-carbon-sulfur (SCS) palladacycle 2a efficiently pro-
motes Heck reactions, and they have also introduced
PEG-tagged variant 2b.10a Dupont described Heck reac-
tions with the related complex 2c.10b Herein we report the
synthesis and crystal structure of fluorous SCS palladacy-
cle 3. This complex promotes Heck reactions, can be re-
covered intact after the reactions, and can be reused
(Figure 1).
Key words: fluorous chemistry, Heck reaction, palladium pincer
complex, solid phase extraction, catalysis
Fluorous reagents, reactants, and protecting groups are in-
creasingly used in organic synthesis.1,2 Light fluorous
molecules are often soluble in common organic reaction
solvents yet can still be separated from organic products
by a simple solid phase extraction over fluorous silica
gel.3 A diverse collection of light fluorous reagents for
stoichiometric use are now known,4 and a significant sub-
set is commercially available. In contrast, catalyst re-
search has focused on the heavy fluorous approach where
many more fluorines are appended and a liquid-liquid or
liquid-solid biphasic separation is used after the reaction.5
Optimized fluorous biphasic reactions may have great
value in large scale process chemistry, but they often lack
the generic applicability that is so important in small scale
discovery chemistry. This is because appending large
numbers of fluorines to ligands or catalysts imparts un-
usual physical and chemical properties.
Y
Rn
Pd X
Rn
S
R2
R1
Pd Cl
R2
Y
S
1
R1
R2
generic pincer complex
Y = P, N, S, etc.
X = counterion
2a NHAc Ph
2b OPEG Ph
2c
H
tBu
PEG is polyethylene glycol
S
C6H4-p-C6F13
Pd Cl
We described the first examples of light fluorous catalysis
in 2000.6 Fluorous phosphines were used as ligands for
platinum6a and later palladium-catalyzed6b reactions, and
the catalyst residues were removed from the target prod-
ucts by solid phase extraction (SPE). However, the cata-
lysts studied did not survive the reaction conditions and
could not be recovered for reuse. Recently, Hope, Stuart
and coworkers described a fluorous-tagged nickel catalyst
that promoted additions of 1,3-diketones to ethyl cyano-
formate and could be recovered and reused.7 Herein we
report the first example of a light fluorous palladium
S
C6H4-p-C6F13
3
Figure 1 Palladium pincer complexes
Complex 3 is readily prepared on gram scale as outlined
in Scheme 1. Commercially available11 bromide 4 was
lithiated and thionated prior to treatment with acetyl chlo-
ride. Rapid chromatography of the crude product provided
crystalline 5 in 52% yield.
complex8 that can be recovered largely intact by solid Deacylation with methyllithium followed by direct treat-
phase extraction after Heck reactions.
ment of the thiolate with a,a¢-dibromo-m-xylene provided
6 in 58% yield. A mixture of ligand 6 and bis-acetonitrile
palladium dichloride [Cl2Pd(CH3CN)2] was heated for 48
hours in acetonitrile–benzotrifluoride (BTF). Upon cool-
ing, complex 3 precipitated as a yellow solid in 71% yield
(mp ca. 224 °C with decomposition). The complex was
stable in air and was characterized by 1H NMR, 13C NMR
and 19F NMR spectroscopy and mass spectrometry.12
Prerequisites for recovering any light fluorous catalyst or
catalyst precursor are stability to both the reaction condi-
SYNLETT 2004, No. 8, pp 1379–1382
0
1.
0
7.
2
0
0
4
Advanced online publication: 22.06.2004
DOI: 10.1055/s-2004-829094; Art ID: Y01804ST
© Georg Thieme Verlag Stuttgart · New York

1380
D. P. Curran et al.
LETTER
1) tBuLi
recovery of complex 3 was not quantitated. Several reac-
tions in Table 1 also succeeded under standard thermal
conditions, although longer reaction times were needed.
In contrast to these successes, haloaromatics did not add
to 2,3-dihydrofuran in good yields under these conditions.
2) S8
Br
C6F13
AcS
C6F13
3) AcCl
5, 52%
4
Table 1 Heck Reactions with Fluorous Complex 3
S
S
C6F13
Cl2Pd(CH3CN)2
CH3CN/BTF
R2
1) MeLi
2)
X
3 mol% 3
3
R2
+
Br
Br
R1
R1
Bu3N, DMA
140 °C
71%
C6F13
R1
X
I
R2
Time (min) Yield (%)
6, 58%
Scheme 1
1
2
MeO
CN
CO2Me
45
45
30
30
30
30
30
45
45
45
45
45
89
84
90
94
87
79
82
76
78
89
92
89
Br
I
CO2Me
CO2Me
CO2Me
CO2Me
CO2Me
Ph
Crystals of 3 suitable for X-ray analysis were grown from
hexanes, and the structure one of the three closely related
molecules in the asymmtric unit is shown in Figure 2.13
The features of the fluorous complex are quite similar to
those of a previously reported non-fluorous complex.10a In
this molecule, one of the fluoroalkyl groups has a single
gauche C–C bond, while all the others are anti.
3
COMe
COMe
COMe
NO2
4
Br
OTf
I
5
6
7
H
I
8
MeO
CN
I
Ph
9
Br
I
Ph
10
11
12
MeCO
MeCO
MeO
Ph
Br
I
Ph
Figure 2 ORTEP diagram of the X-ray crystal structure of one of
three related molecules of complex 3
CO2Me
To quantitate the recovery of complex 3, we conducted a
1.25 mmol scale reaction of bromoacetophenone and
methyl acrylate promoted by 41.5 mg (3 mol%) of catalyst
3 (Equation 2). After heating and solid phase extraction as
above, the product 9 was isolated in 94% yield. The crude
recovered catalyst 3 (31.4 mg, 76%) was then reused in a
second 1.25 mmol scale experiment. The yield of 9 was
again good (97%) and now 26.0 mg (83%) of 3 was recov-
ered. A third cycle with this sample (now down to about
1.9 mol% complex 3) again provided 98% yield of 9 and
rendered back 19.6 mg (75%) of 3.
Rapid, high temperature Heck reactions were promoted
by complex 3 under microwave conditions.14 For exam-
ple, a mixture of p-bromoacetophenone 7 (0.30 mmol),
methyl acrylate 8 (0.38 mmol), tributylamine (0.30 mmol)
and complex 3 (10 mg, 3.00 mol%) in dimethylacetamide
(DMA, 1 mL) was heated for 30 minutes at 140 °C
(Equation 1). The mixture was cooled and then charged
directly to a FluoroFlash™ fluorous silica gel cartridge.
Elution with 90% MeOH–H2O gave an organic fraction of
the product, DMA and ammonium salt. Filtration of this
through silica gel gave pure product 9 in 94% yield. Sec-
ond pass elution of the fluorous silica gel with diethyl
ether gave recovered complex 3.
Finally, we also conducted two larger scale reactions to
test scalability and to analyze the Heck product for traces
of palladium. In the first reaction, bromoacetophone 7 (1
g, 5.05 mmol) and methyl acrylate 8 were coupled with
complex 3 (167 mg, 3 mol%) under the usual microwave
conditions. The mixture was subjected to the standard
fluorous spe followed by rapid chromatography. Product
9 was isolated in 95% yield and 160 mg (96%) of complex
The results of a series of similar Heck reactions with
methyl acrylate and styrene are collected in Table 1. Aryl
iodides, bromides and a triflate all added successfully
over 30–45 minutes at 140 °C and products were isolated
in 76–94% yields. In these small-scale reactions, the
O
O
3 mol% 3
CO2Me
+
Br
CO2Me
Bu3N, DMA
140 °C, 30 min
7, 1 equiv
8, 1.25 equiv
9, 94%
Equation 1
Synlett 2004, No. 8, 1379–1382 © Thieme Stuttgart · New York

LETTER
A Soluble Fluorous Palladium Complex that Promotes Heck Reactions
1381
O
O
2-3 mol% 3
CO2Me
+
Br
CO2Me
Bu3N, DMA
140 °C, 30 min
7 1.25 mmol/cycle
8
9
Amt 3 used
Recovered 3
Yield 9
Cycle
41.5 mg
31.4 mg
26.0 mg
31.4 mg
26.0 mg
19.6 mg
94%
97%
98%
1
2
3
Equation 2
3 was recovered from the ether fraction of the spe. This resents an attractive new breed of organometallic complex
was recrystallized to provide 116 mg (69%) of pure yel- that is soluble under typical reaction conditions, can ready
low complex 3.15 The reaction product 9 (951 mg) was an- be recovered from the reaction mixture fluorous spe, and
alyzed for trace palladium and contained 74 ppm. This can be purified prior to reuse on either small or large scale.
corresponds to 0.45% of the original amount of palladium It is clearly of interest to develop other catalysts or cata-
that was added.
lyst precursors with these features.
The second reaction was treated the same way, except that
the fluorous spe was omitted and the crude product was
directly subjected to chromatography to give product 9 in
91% yield. However, we were not able to recover either
complex 3 or ligand 6 from this experiment. The fluorous
procedure is crucial for complex recovery.
General Procedure for the Microwave-Mediated Heck Reac-
tion: The haloarene (0.3 mmol) was mixed with 3 (10 mg, 3%) in a
microwave tube with a stirring bar. To this was added DMA (1 mL),
tributylamine (0.3 mmol) and ethyl acrylate or styrene (0.38 mmol).
The reaction tube was sealed and reacted in a CEM Discover Micro-
wave reactor for 30 to 45 min at 140 °C with continuous stirring. Af-
ter 5 min of cooling, the reaction mixture was loaded on a 10 g
FluoroFlashTM cartridge that had been preconditioned with 90:10
MeOH–H2O. The cartridge was eluted with 60 mL of the solvent
mixture to give the non-fluorous compounds, followed by elution
with 30 mL of Et2O to give the fluorous complex 3. The non-fluo-
rous compounds were purified by a quick column chromatography
to afford the Heck product in high yields. The products were ana-
lyzed by 1H NMR spectroscopy and GC.
The function of complex 3 in these Heck reactions is not
known with certainty. Early workers suggested that com-
plexes like this were catalysts or pre-catalysts in a cycle
with an intact ligand.9 But more recent work has suggest-
ed that such palladacycles might be precursors for trace
quantities of soluble or nanoparticulate palladium metal.16
Gladysz and Rocaboy have provided experimental evi-
dence for a nanoparticle8l pathway in a different class of
heavy fluorous palladacycle. If the small amount of palla-
dium remaining in the product from the large scale reac-
tion represents the actual catalyst that was released from
complex 3, then the Heck reaction was promoted by about
0.01 mol% palladium.
Gram-Scale Reaction and Complex Recovery Experiment: Bro-
moacetophenone (1.0 g, 5.05 mmol) was mixed with palladacycle 3
(167 mg, 3 mol%). To this was added DMA (8 mL), tributylamine
(5.05 mmol) and methyl acrylate (6.3 mmol). The reaction tube was
sealed and reacted in a CEM Discover Microwave reactor for 30
min at 140 °C with continuous stirring. After 5 min of cooling, the
reaction mixture was loaded on a 50 g FluoroFlashTM cartridge that
Fluorous complex 3 is an attractive promoter for Heck re- had been preconditioned with 90:10 MeOH–H2O. The cartridge
was eluted with 100 mL of the mentioned solvent mixture to give
actions because it is a readily available, stable solid. It is
the non-fluorous compounds, followed by elution with 80 mL of
soluble in DMA and related solvents and promotes rapid
Et2O to give fluorous complex 3 (160 mg). The ethereal solution
reactions when added in small amounts. The complex can
was concentrated and recrystallized from hexanes, giving dark yel-
be recovered by fluorous solid phase extraction and re-
low crystals (116 mg, 60%).
used either crude or after recrystallization. For use with
different Heck partners, we recommend the use of recrys-
Acknowledgment
tallized complex. The ability to return light fluorous cata-
lysts to a state ‘as good as new’ by recrystallization
differentiates them from heavy fluorous catalysts used in
liquid biphasic processes as well as from PEG-bound and
resin bound catalysts. Such recovered catalysis are gener-
ally reused on an ‘as is’ basis along with any catalyst or
ligand degradation products. When such catalysts are used
for a series of different reactions, there is a risk of cross
contamination of a product of a later reaction with some-
thing formed in an earlier one.
We thank Dr. Steven Geib for solving the crystal structure. We
thank the National Institutes of Health for funding this work. We
also thank the CEM Corporation for the Discover Microwave
reactor.
References
(1) (a) Curran, D. P. Pharm. News 2002, 9, 179. (b) Zhang, W.
Tetrahedron 2003, 59, 4475.
(2) Handbook of Fluorous Chemistry; Gladysz, J. A.; Curran, D.
P.; Horvath, I., Eds.; Wiley-VCH: Weinheim, 2004, ; in
press.
Whether or not fluorous palladacycle 3 is a catalyst or a
precursor for small amounts of soluble palladium, it rep-
Synlett 2004, No. 8, 1379–1382 © Thieme Stuttgart · New York

1382
D. P. Curran et al.
LETTER
(3) (a) Curran, D. P. Synlett 2001, 1488. (b) Curran, D. P. In
Handbook of Fluorous Chemistry; Gladysz, J. A.; Curran, D.
P.; Horvath, I., Eds.; Wiley-VCH: Weinheim, 2004, Chap. 7,
in press. (c) Curran, D. P. In Handbook of Fluorous
Chemistry; Gladysz, J. A.; Curran, D. P.; Horvath, I., Eds.;
Wiley-VCH: Weinheim, 2004, Chap. 8, in press.
(4) Representative recent references: (a) Dienophiles: Werner,
S.; Curran, D. P. Org. Lett. 2003, 5, 3293. (b)FCBzgroups:
Curran, D. P.; Amatore, M.; Guthrie, D.; Campbell, M.; Go,
E.; Luo, Z. Y. J. Org. Chem. 2003, 68, 4643. (c) Tin
reagents: Bucher, B.; Curran, D. P. Tetrahedron Lett. 2000,
41, 9617. (d) FDEAD: Dandapani, S.; Curran, D. P.
Tetrahedron 2002, 58, 3855. (e) Ester: Chen, C. H. T.;
Zhang, W. Org. Lett. 2003, 5, 1015. (f) Scavengers: Zhang,
W.; Chen, C. H. T.; Nagashima, T. Tetrahedron Lett. 2003,
44, 2065. (g) Acetals: Read, R. W.; Zhang, C. T.
(10) (a) Bergbreiter, D. E.; Osburn, P. L.; Liu, Y. S. J. Am. Chem.
Soc. 1999, 121, 9531. (b) Gruber, A. S.; Zim, D.; Ebeling,
G.; Monteiro, A. L.; Dupont, J. Org. Lett. 2000, 2, 1287.
(11) (a) This bromide, FluoroFlash silica gel and complex 3 are
commercially available from Fluorous Technologies, Inc.,
www.fluorous.com. (b) DPC holds an equity interest in this
company.
(12) 1H NMR (300 MHz, CDCl3): d = 4.68 (s, 4 H, CH2), 7.08 (m,
4 H), 7.59 (d, 4 H, 3J = 8.3 Hz), 7.93 (d, 4 H, 3J = 8.3 Hz).
19F NMR (282.4 MHz, CDCl3): d = –79.57 (m, 3 F, 3J = 8.5
Hz), –109.86 (m, 2 F, CaF2), –120.32 (m, 4 F, CbF2, CdF2),
–121.59 (m, 2 F, CeF2), –124.96 (m, 2 F, CgF2). 13C NMR (75
MHz, CDCl3): d = 50.32 (CH2), 122.79, 125.59, 128.33,
130.23, 130.68, 136.75, 149.04, 161.84. EIMS: m/z = 1100
(M+), 1063 (M – Cl), 958 (M – PdCl), 513 (M – C8F17
PdCl), 428 (HSPhC6F13). LCMS (APCl, neg. mode):
–
Tetrahedron Lett. 2003, 44, 7045.
m/z = 1037, 957 (M – HPdCl), 427 (M – SPhC6F13). MS
(APCl, pos. mode): m/z = 1063 (M – Cl).
(5) (a) Gladysz, J. A.In Handbook of Fluorous Chemistry;
Gladysz, J. A.; Curran, D. P.; Horvath, I., Eds.; Wiley-VCH:
Weinheim, 2004, Chap. 5, in press. (b) Dobbs, A. P.;
Kimberley, M. R. J. Fluorine Chem. 2002, 118, 3.
(c) Barthel-Rosa, L. P.; Gladysz, J. A. Coord. Chem. Rev.
1999, 192, 587. (d) Horváth, I. T. Acc. Chem. Res. 1998, 31,
641. (e) deWolf, E.; van Koten, G.; Deelman, B. J. Chem.
Soc. Rev. 1999, 28, 37. (f) Hope, E. G.; Stuart, A. M. J.
Fluorine Chem. 1999, 100, 75.
(6) (a) Zhang, Q.; Luo, Z.; Curran, D. P. J. Org. Chem. 2000, 65,
8866. (b) Vallin, K. S. A.; Zhang, Q. S.; Larhed, M.; Curran,
D. P.; Hallberg, A. J. Org. Chem. 2003, 68, 6639.
(7) Croxtall, B.; Hope, E. G.; Stuart, A. M. Chem. Commun.
2003, 2430.
(8) For work on heavy fluorous palladium complexes, see:
(a) Betzemeier, B.; Knochel, P. Angew. Chem., Int. Ed. Engl.
1997, 36, 2623. (b) Betzemeier, B.; Lhermitte, F.; Knochel,
P. Tetrahedron Lett. 1998, 39, 6667. (c) Shezad, N.; Oakes,
R. S.; Clifford, A. A.; Rayner, C. M. Tetrahedron Lett. 1999,
40, 2221. (d) Kling, R.; Sinou, D.; Pozzi, G.; Choplin, A.;
Quignard, F.; Busch, S.; Kainz, S.; Koch, D.; Leitner, W.
Tetrahedron Lett. 1998, 39, 9439. (e) Moineau, J.; Pozzi,
G.; Quici, S.; Sinou, D. Tetrahedron Lett. 1999, 40, 7683.
(f) Chechik, V.; Crooks, R. M. J. Am. Chem. Soc. 2000, 122,
1243. (g) Schneider, S.; Bannwarth, W. Angew. Chem. Int.
Ed. 2000, 39, 4142. (h) Nishimura, T.; Maeda, Y.; Kakiuchi,
N.; Uemura, S. J. Chem. Soc., Perkin Trans. 1 2000, 4301.
(i) Schneider, S.; Bannwarth, W. Helv. Chim. Acta 2001, 84,
735. (j) Moreno-Mañas, M.; Pleixats, R.; Villarroya, S.
Organometallics 2001, 20, 4524. (k) Rocaboy, C.; Gladysz,
J. A. Tetrahedron 2002, 58, 4007. (l) Rocaboy, C.; Gladysz,
J. A. Org. Lett. 2002, 4, 1993. (m) Markert, C.; Bannwarth,
W. Helv. Chim. Acta 2002, 85, 1877. (n) Rocaboy, C.;
Gladysz, J. A. New J. Chem. 2003, 27, 39. (o) Grigg, R.;
York, M. Tetrahedron Lett. 2000, 41, 7255.
(13) Complex 3 crystallizes in space group P21/n with three
crystallographically independent but chemically similar
molecules per asymmetric unit. The metric symmetry is
suggestive of orthorhombic, with b equal to 90.02 (2)° but
the crystal packing does not show any symmetry higher than
monoclinic. Each molecule consists of a Pd atom
coordinated to a tridentate 1,3-bis(thiomethyl)benzene SCS
ligand with a Pd–Cl bond trans to the Pd–C bond. The
environment about Pd is slightly distorted square planar.
Distortions are caused by the rigid SCS ligand, which results
in non-linear S–Pd–S angles of 171.07 (7)°, 170.11 (7)° and
167.40 (7)° in the three independent molecules of 3. Pd–Cl
bonds average 2.391 (2) Å, Pd–S bonds average 2.302 (2) Å,
and Pd-C bonds average 1.977 (8) Å. All three molecules
have similar, well resolved Pd(SCS)Cl cores, but differ in
the orientation of the fluoroalkylaryl groups attached to S.
For all molecules, the two fluoroalkylaryl groups are on
opposite sides of the Pd(SCS)Cl plane. Differences lie in the
twisting of the aryl rings. In two of the three molecules, one
of the fluoroalkyl chains is highly disordered and some
fluorines and carbons at the ends of these chains cannot be
resolved.
(14) (a) Larhed, M.; Moberg, C.; Hallberg, A. Acc. Chem. Res.
2002, 35, 717. (b) Olofsson, K.; Kim, S. Y.; Larhed, M.;
Curran, D. P.; Hallberg, A. J. Org. Chem. 1999, 64, 4539.
(c) Larhed, M.; Hoshino, M.; Hadida, S.; Curran, D. P.;
Hallberg, A. J. Org. Chem. 1997, 62, 5583.
(15) The recovered complex exhibited the same spectra as that
starting complex and had a bromine content of <0.15% by
elemental analysis.
(16) (a) de Vries, A. H. M.; Mulders, J. M. C. A.; Mommers, J. H.
M.; Hendricks, H. J. W.; de Vries, J. G. Org. Lett. 2003, 5,
3285. (b) Consorti, C.; Zanini, M. L.; Leal, S.; Eberling, G.;
Dupont, J. Org. Lett. 2003, 5, 983. (c) Nowotny, M.;
Hanefeld, U.; van Konigsveld, H.; Maschmeyer, T. Chem.
Commun. 2000, 1877.
(9) van der Boom, M. E.; Milstein, D. Chem. Rev. 2003, 103,
1759.
Synlett 2004, No. 8, 1379–1382 © Thieme Stuttgart · New York