Novel η3-allylpalladium-pyridinylpyrazole complex: Synthesis, reactivity, and catalytic activity for cyclopropanation of ketene silyl acetal with allylic acetates

Article abstract of DOI:10.1021/ja982269c

Novel cationic η³-allylpalladium-pyridinylpyrazole complexes 1a and 1b were synthesized from 3-alkyl-5-(2-pyridinyl)pyrazole and η³-allylpalladium chloride dimer in the presence of AgBF₄. Cationic complexes 1a and 1b were converted into neutral complexes 2a and 2b under basic conditions. These complexes were characterized by ¹H, ¹³C, and ¹⁵N NMR studies. Neutral complexes 2a and 2b have high catalytic activity for cyclopropanation of ketene silyl acetals with allylic acetates. Comparison of the cationic and neutral complexes and the reaction mechanism of cyclopropanation were discussed.

Full text of DOI:10.1021/ja982269c

J. Am. Chem. Soc. 1998, 120, 10391-10396
10391
Novel η3-Allylpalladium-Pyridinylpyrazole Complex: Synthesis,
Reactivity, and Catalytic Activity for Cyclopropanation of Ketene
Silyl Acetal with Allylic Acetates
Akiharu Satake*,† and Tadashi Nakata
Contribution from The Institute of Physical & Chemical Research (RIKEN), 2-1 Hirosawa,
Wako, Saitama 351-0198, Japan
ReceiVed June 29, 1998
Abstract: Novel cationic η3-allylpalladium-pyridinylpyrazole complexes 1a and 1b were synthesized from
3-alkyl-5-(2-pyridinyl)pyrazole and η3-allylpalladium chloride dimer in the presence of AgBF4. Cationic
complexes 1a and 1b were converted into neutral complexes 2a and 2b under basic conditions. These complexes
were characterized by 1H, 13C, and 15N NMR studies. Neutral complexes 2a and 2b have high catalytic activity
for cyclopropanation of ketene silyl acetals with allylic acetates. Comparison of the cationic and neutral
complexes and the reaction mechanism of cyclopropanation were discussed.
Introduction
Scheme 1
Palladium complexes are well-known to exhibit good per-
formance in various catalytic organic reactions.1 In these
reactions, ligands attached to palladium play an important role.
Ligands affect metals electronically and sterically by complex-
ation. The resulting complexes can control the reactivity of
substrates and selectivity of products. In nucleophilic substitu-
tion reactions of allylic compounds via η3-allylpalladium
complexes, especially, differences in the nature and reactivity
between neutral monophosphine η3-allylpalladium complex A
and cationic bisphosphine complexes B have been fully
discussed.2 In contrast to the phosphine palladium complexes,
there are few detailed studies of η3-allylpalladium complexes
coordinated with nitrogen ligands (C and D in Scheme 1)
because neutral mononitrogen η3-allylpalladium complexes may
be unstable and the isolation is difficult.3 Therefore, it is a
challenging task to synthesize stable cationic and neutral η3-
allylpalladium complexes having nitrogen ligands, and to
compare the properties of these complexes. Pyrazoles are
known as both monodentate and exobidentate ligands, and their
nitrogen atoms coordinate with the metal center as both anionic
and donor groups.4 Pyrazole having a 2-pyridinyl group5 that
we focus on here could be a good candidate for production of
both cationic and neutral η3-allylpalladium complexes (1 and 2
Scheme 2
in Scheme 2). Further, we observed that the neutral complex
2 prepared from 1 worked as an effective catalyst for cyclo-
propanation of ketene silyl acetals with allylic acetates. The
palladium-catalyzed cyclopropanation by nucleophilic attack to
the central carbon on the η3-allylpalladium intermediate followed
by reductive elimination of palladacyclobutane is very rare6
* To whom correspondence should be addressed.
† Special Postdoctoral Researcher, Special Postdoctoral Researchers
Program in RIKEN.
(5) (a) Sugiyarto, K. H.; Goodwin, H. A. Aust. J. Chem. 1988, 41, 1645.
(b) Jones, P. J.; Jeffery, J. C.; McCleverty, J. A.; Ward, M. D. Polyhedron
1997, 16, 1567. (c) Amoroso, A. J.; Jeffery, J. C.; Jones, P. L.; McCleverty,
J. A.; Thornton, P.; Ward, M. D. Angew. Chem., Int. Ed. Engl. 1995, 34,
1443.
(1) (a) Tsuji, J. Palladium Reagents and Catalysts; John Wiley & Sons:
Chichester, 1995. (b) ComprehensiVe Organic Synthesis; Trost, B. M.,
Fleming, I., Eds.; Pergamon Press: Oxford, 1991; Vol. 3, p 435; Vol. 4, p
585; Vol. 4, p 833.
(2) (a) Trost, B. M. Tetrahedron 1977, 33, 2615. (b) Trost, B. M.; Strege,
P. E. J. Am. Chem. Soc. 1975, 97, 2534. (c) Oshima, M.; Shimizu, I.;
Yamamoto, A.; Ozawa, F. Organometallics 1991, 10, 1221. (d) Åkermark,
B.; Krakenberger, B.; Hansson, S.; Vitagliano, A. Organometallics 1987,
6, 620.
(6) (a) Hegedus, L. S.; Darlington, W. H.; Russell, C. E. J. Org. Chem.
1980, 45, 5193. (b) Carfagna, C.; Mariani, L.; Musco, A.; Sallese, G. J.
Org. Chem. 1991, 56, 3924. (c) Hoffmann, H. M. R.; Otte, A. R.; Wilde,
A. Angew. Chem., Int. Ed. Engl. 1992, 31, 234. (d) Carfagna, C.; Galarini,
R.; Musco, A.; Santi, R. J. Mol. Catal. 1992, 72, 19. (e) Wilde, A.; Otte,
A. R.; Hoffmann, H. M. R. J. Chem. Soc., Chem. Commun. 1993, 615. (f)
Formica, M.; Musco, A.; Pontellini, R.; Linn, K.; Mealli, C. J. Organomet.
Chem. 1993, 448, C6. (g) Hoffmann, H. M. R.; Otte, A. R.; Wilde, A.;
Menzer, S.; Williams, D. J. Angew. Chem., Int. Ed. Engl. 1994, 33, 1280.
(h) Otte, A. R.; Wilde, A.; Hoffmann, H. M. R. Angew. Chem., Int. Ed.
Engl. 1995, 34, 100.
(3) Although syntheses of many cationic bisnitrogen η3-allylpalladium
complexes were reported, there are few reports on neutral η3-allylpalladium
complexes. Huttel, R.; Rau, B. J. Organomet. Chem. 1977, 139, 89.
(4) (a) Minghetti, G.; Banditelli, G.; Bonati, F. J. Chem. Soc., Dalton
Trans. 1979, 1851. (b) Bushnell, G. W.; Dixon, K. R.; Eadie, D. T.; Stobart,
S. R. Inorg. Chem. 1981, 20, 1545.
S0002-7863(98)02269-0 CCC: $15.00 © 1998 American Chemical Society
Published on Web 09/25/1998

10392 J. Am. Chem. Soc., Vol. 120, No. 40, 1998
Satake and Nakata
Table 1. NMR Data of 1a and 2a in CD2Cl2:CD3OD (1:1)a
chem shifts in ppm (coupling constant in Hz)
1a 2a
Scheme 3
N1
N2
N11
C3
C4
C5
C6
C7
C8
219.7
189.9
225.3
145.3
103.9
153.4
151.8
122.7
141.4
126.5
154.4
11.0
236.3
288.3
218.7
150.7
101.9
152.4
155.4
120.4
140.2
122.8
153.6
13.4
Scheme 4
Scheme 5
C9
C10
C12
C14
C13, C15
H4
118.5
58.7, 63.7
6.79 (s)
116.7
56.2, 60.8
6.42 (s)
H7
H8
7.98 (ddd, 7.8, 1.5, 1.0) 7.64 (br.dd, 7.8, 1.5)
8.13 (dd, 7.8, 7.8, 1.5) 7.88 (ddd, 7.8, 7.8, 1.5)
H9
7.54 (ddd, 7.8, 5.4, 1.5) 7.18 (ddd, 7.8, 5.4, 1.5)
8.70 (ddd, 5.4, 1.5, 1.0) 8.52 (br.dd, 5.4, 1.5)
H10
H12
H14
H13, H15 syn
2.46 (s)
2.31 (s)
5.86 (tt, 12.7, 7.3)
4.52 (br.), 4.37 (br.)
5.68 (tt, 12.7, 6.8)
4.31, 3.96 (dd, 6.8, 1.5)
3.23, 3.11 (d, 12.7)
H13, H15 anti 3.52 (br.), 3.30 (br.)
a
1H NMR (600 MHz):13C NMR (150 MHz):15N NMR spectra were
obtained by the 1H-15N PEG-HMBC method (ref NH4NO3 in DMSO-
d6 at 0 ppm).
(path B in Scheme 3) although palladium-catalyzed allylation
is well-known1 (path A in Scheme 3). In this paper, we report
the synthesis and reactivity of novel cationic and neutral
palladium-pyridinylpyrazole complexes 1a, 1b, 2a, and 2b.
Catalytic cyclopropanation of ketene silyl acetals with 2 and
its reaction mechanism are also reported.
Results
A. Synthesis and Properties of Palladium Complexes.
Pyridinylpyrazole ligands 3a (MePPH)7 and 3b (tBuPPH) were
easily obtained by Claisen condensation of ethyl pyridinecar-
boxylate with methyl ketones and subsequent formation of the
pyrazole ring with hydrazine (Scheme 4). Reaction of 3a and
η3-allylpalladium chloride dimer in the presence of AgBF4 in
dichloromethane gave air-stable complex 1a in 89% yield as a
white powder (Scheme 5). Rotation of the allyl moiety in 1a
was observed by decoupling study in 1H NMR at room
temperature.8 A hydrogen on nitrogen at the 2-position was
observed at 14.25 ppm as a broad signal in DMSO-d6. The
acidic hydrogen was easily abstracted by a base to give a
deprotonated complex (X); treatment of 1a with sodium
methoxide produced neutral η3-allylpalladium complex 2a in a
1:1 mixture of CD2Cl2 and CD3OD. Data of 1a and 2a in 1H,
13C, and 15N NMR are shown in Table 1. Major signals moved
to upfield by conversion from cationic 1a into neutral 2a except
for N(1), N(2), C(3), C(6), and C(12). On the contrary, there
was pronounced downfield shift of the N(2) nitrogen in 15N
NMR because of deprotonation.
Figure 1. 1H NMR spectra of palladium complexes: (a) 1b in DMSO-
d6 (300 MHz), (b) 2b in DMSO-d6 (300 MHz), and (c) 2b in CDCl3
(300 MHz).
of CD2Cl2 and CD3OD, complicated signals were observed in
DMSO-d6 or DMF-d7. From the NMR spectra, a mixture of
the monomer form and other structures seemed to exist in
DMSO or DMF. FAB mass spectrum of X indicated the
existence of a dimer complex.10
A bulky group on the 3-position of pyrazole prevented
production of a dimer. We synthesized η3-allylpalladium
complex 1b from 3-tert-butyl-5-(2-pyridinyl)pyrazole (3b) and
η3-allylpalladium chloride dimer using a similar procedure to
that of 1a. The proton NMR spectrum of the deprotonated
complex prepared from 1b showed monomer complex 2b even
in DMSO-d6 [(b) in Figure 1]. The FAB mass spectrum of 2b
also indicated peaks derived mainly from the monomer complex.
Deprotonated complex X9 indicated different 1H NMR spectra
in various solvents. Although X showed mainly a simple
spectrum derived from a monomer form in CDCl3 or a mixture
(7) Ferles, M.; Kafka, S.; Sˇilha´nkova´, A.; Sˇputova´, M. Collect. Czech.
Chem. Commun. 1981, 46, 1167.
(8) Because each anti proton was exchanged by rotation of the allyl
moiety at room temperature, the signal of one proton disappeared by a
decoupling study on another proton. A similar phenomenon was observed
in the case of a decoupling study on syn protons.
(10) Pyrazoles can usually coordinate as exobidentate ligands to give a
(9) See Experimental Section.
bimetallic complex. Trofimenko, S. Chem. ReV. 1972, 72, 497.

NoVel η3-Allylpalladium-Pyridinylpyrazole Complex
J. Am. Chem. Soc., Vol. 120, No. 40, 1998 10393
Table 3. Reaction of Allylic Acetates with Ketene Silyl Acetal 10
Scheme 6
Scheme 7
Scheme 8
Table 4. Catalytic Cyclopropanation of Ketene Silyl Acetal
Table 2. Catalytic Cyclopropanation of Ketene Silyl Acetal
ylpropionate 10. When ketene silyl acetals of tert-butyl acetate
and ethyl propionate were used instead of 10, no cyclopropane
was obtained. Reaction of substituted allyl acetate with 10
proceeded stereoselectively to give 13.
We changed the amounts of 1a and 1b to examine the
turnover number (TON) of the palladium catalysts in the reaction
of allyl acetate and 4. Although the reaction was almost
completed with 5 mol % of 1a in 0.5 h, in the case of using 1
mol % of 1a the reaction terminated, giving only 14% yield of
5 (Runs 1 and 2 in Table 4). Although use of 1b lowered
selectivity of 5, the turnover number was heightened to 88 times
(Runs 3 and 4 in Table 4).
B. Reactivity. Complex 1a reacted with nucleophiles to give
allylated products and cyclopropane compounds. Reaction of
1a with the sodium enolate of dimethyl malonate in THF at
room temperature gave dimethyl allylmalonate in 46% yield
(Scheme 7). On the other hand, reaction of 1a with ketene silyl
acetal 4 in the presence of NaOAc in DMSO at room
temperature gave cyclopropane 5 and allylated compound 6 in
85% and 8% yields, respectively (Scheme 8).
Further, complex 1a catalyzed cyclopropanation of ketene
silyl acetal 4 with allyl acetate. Reaction of allyl acetate (1
mmol) and ketene silyl acetal 4 (2 mmol) with 1a (0.05 mmol)
and sodium acetate (0.2 mmol) in DMSO (4 mL) gave
cyclopropane 5 and allylated ester 6 in 87% and 7% yields,
respectively (Run 1 in Table 2). As described in the Discussion
section, the present cyclopropanations proceeded via neutral
complex 2a. To compare other cationic palladium-bisnitrogen
complexes which could not transform into neutral complexes,
we examined cationic complexes 7 and 8 (Runs 2 and 3 in Table
2). Although allyl acetate disappeared after 24 h in the case of
7 and 8, the yields of 5 and 6 were very low. When 7 or 8 was
used as a catalyst, R,â-unsaturated ester 9 was obtained in about
60% yield based on ketene silyl acetal 4.
Discussion
A. Comparison of Cationic and Neutral Complexes. 13C
NMR shifts of the allyl moiety are a good indication to
understand the properties and reactivity of η3-allyl complexes.2d
In the case of neutral η3-allylpalladium complex having a
phosphine and an anionic ligand (A in Scheme 1), two terminal
13C NMR shifts appear in different positions, and the difference
is usually large (>16 ppm). However, the difference in two
terminal 13C NMR shifts of neutral complex 2a is only 4.6 ppm
and is close in the case of cationic complex 1a (5.0 ppm). This
result means that the two terminal carbons do not lean to one
side electronically, and the pyridinylpyrazole ligand behaves
like a bisnitrogen ligand, not an anionic group such as halogen
or cyanide.
Cationic complex 1a is soluble in DMSO and DMF, but
almost insoluble in CHCl3, CH2Cl2, MeOH, acetone, and H2O.
Interestingly, 1a is soluble in a mixture of CH2Cl2 and MeOH.11
On the other hand, neutral complex 2a is very soluble in various
organic solvents, such as CHCl3, CH2Cl2, MeOH, acetone,
DMSO, and DMF. The solubility is advantageous in terms of
the broad choice of solvents.
Results of reactions of another ketene silyl acetal with allylic
acetates are shown in Table 3. Cyclopropane compounds were
obtained only in the case of the ketene silyl acetal of 2-meth-
(11) At the present time, the reason the combination of CH2Cl2 and
MeOH is effective to solve 1a is not clear.

10394 J. Am. Chem. Soc., Vol. 120, No. 40, 1998
Satake and Nakata
It is apparent in Table 2 that there is a great difference
between 1a, which is a precursor for neutral complex 2a, and
cationic complexes 7 and 8, which cannot be converted into
neutral form. When 7 and 8 were used as catalysts, a
considerable amount of 9 was obtained. Tsuji et al. reported12
that 9 was obtained in the reaction of 4 with allyl carbonate
and palladium acetate without phosphine in benzonitrile. They
explained formation of 9 by attack of enolate on palladium and
subsequent â-elimination. Thus, formation of 9 in the case of
7 and 8 means that ligand dissociation and attack of enolate on
palladium occurred. On the contrary, because 2a generated from
1a prevents ligand dissociation even in DMSO, cyclopropanation
proceeded smoothly.
B. Cyclopropanation. Formation of cyclopropane from
allylic acetate and ketene silyl acetal is considered to take place
by regioselective nucleophilic attack to the central carbon of
the allyl moiety on palladium and subsequent reductive elimina-
tion of palladacyclobutane13 (path B in Scheme 3). Completion
of cyclopropanation is dependent on the kind of nucleophile
and ligand. Generally, when phosphine is used as a ligand,
allylated products are obtained mainly. On the contrary, in the
case of nonphosphine ligands cyclopropanes are obtained.6
Nucleophiles for the cyclopropanation are limited to less
stabilized carbon nucleophiles (pKa ) 20-30). When the allyl
moiety has a good leaving group on the center carbon, soft
nucleophiles also react with the η3-allylpalladium complex.
However, only disubstituted propene, not cyclopropane, is
obtained.13c,d,e
Figure 2.
Scheme 9. Pd(0)/Pd(II) Cycle (eq 1) and Pd(II)/Pd(IV)
Cycle (eq 2)
In the catalytic cyclopropanation, regeneration of an active
palladium species that can react with allylic compounds to
produce η3-allylpalladium is also important. Musco and co-
workers succeeded in selective cyclopropanation of ketene silyl
acetals with allyl bromide using a palladium-TMEDA catalyst.6f
In their catalytic system, however, use of reactive allyl bromide
and a stoichiometric amount of thulium acetate14 was essential.
Therefore, it is desirable that allylic acetate is used instead of
allyl bromide and thulium acetate. Complex 1a can react with
allyl acetate to give cyclopropanes. This is the first example
of selective cyclopropanation with use of allyl acetate directly.
Two major mechanisms, the Pd(0)/Pd(II) cycle [(1) in Scheme
9] and the Pd(II)/Pd(IV) cycle [(2) in Scheme 9], are considered
in the present catalytic cyclopropanation. The Pd(II)/Pd(IV)
cycle was proposed in Musco’s reports.6f Actually, a dimeth-
ylpalladium complex having a bisnitrogen ligand reacts with
additional allyl bromide to give a four-valent palladium species,
whose reductive elimination proceeds to give η3-allylpalladium
species and ethane.15 In our case, however, no additional allyl
acetate was necessary to produce cyclopropane (Scheme 8). This
result means that reductive elimination of palladacyclobutane
directly occurred. Therefore, the catalytic cycle of 2a must be
the Pd(0)/Pd(II) cycle. A plausible mechanism is shown in
Figure 2. Cationic complex 1a is converted into neutral
complex 2a under basic conditions, and nucleophilic attack
occurs on the central carbon of the η3-allyl moiety to produce
palladacyclobutane. The palladacyclobutane gives cyclopropane
5 and an active Pd(0) species, which react with allyl acetate to
generate 2a and sodium acetate again. We consider that DMSO
plays an important role because no reaction proceeded in other
solvents, even in DMF. DMSO may accelerate the reaction in
the oxidative addition and reductive elimination steps. There-
fore, the difference in catalytic ability among complexes 2a, 7,
and 8 may arise from their stability and that of their derivatives16
in DMSO.
(12) Tsuji, J.; Takahashi, K.; Minami, I.; Shimizu, I. Tetrahedron Lett.
1984, 42, 4783.
(13) Center carbon attack of a nucleophile against an allyl ligand to
produce metalacyclobutanes was observed in various metals; however, there
are few examples of the reductive elimination of metalacyclobutanes to
produce cyclopropanes. Center carbon attack: Pd and Pt: (a) Carfagna,
C.; Galarini, R.; Musco, A.; Santi, R. Organometallics 1991, 10, 3956. (b)
Carfagna, C.; Galarini, R.; Linn, K.; Lo´pez, J. A.; Mealli, C.; Musco, A.
Organometallics 1993, 12, 3019. (c) Ohe, K.; Matsuda, H.; Morimoto, T.;
Ogoshi, S.; Chatani, N.; Murai, S. J. Am. Chem. Soc. 1994, 116, 4125. (d)
Aranyos, A.; Szabo´, K. J.; Castan˜o, A. M.; Ba¨ckvall, J.-E. Organometallics
1997, 16, 1058. (e) Organ, M. G.; Miller, M. Tetrahedron Lett. 1997, 47,
8181. Other metals: (f) Periana, R. A.; Bergman, R. G. J. Am. Chem. Soc.
1984, 106, 7272. (g) McGhee, W. D.; Bergman, R. G. J. Am. Chem. Soc.
1985, 107, 3388. (h) Periana, R. A.; Bergman, R. G. J. Am. Chem. Soc.
1986, 108, 7346. (i) Ephritikhine, M.; Green, M. L. H.; MacKenzie, R. E.
J. Chem. Soc., Chem. Commun. 1976, 619. (j) Ephritikhine, M.; Francis,
B. R.; Green, M. L. H.; MacKenzie, R. E.; Smith, M. J. J. Chem. Soc.,
Dalton Trans. 1977, 1131. Reductive elimination of metalacyclobutane: (k)
Benyunes, S. A.; Brandt, L.; Green, M.; Parkins, A. W. Organometallics
1991, 10, 57. (l) Wakefield, J. B.; Stryker, J. M. J. Am. Chem. Soc. 1991,
113, 7057. (m) Tjaden, E. B.; Stryker, J. M. Organometallics 1992, 11, 16.
(n) Tjaden, E. B.; Stryker, J. M. J. Am. Chem. Soc. 1993, 115, 2083. (o)
Schwiebert, K. E.; Stryker, J. M. Organometallics 1993, 12, 600.
(14) Thulium acetate is very toxic. Musco reported that no reaction took
place with use of allyl acetate instead of allyl bromide and thulium acetate.
The turnover number of 1a was lower than that of complex
1b, and complex 2a can easily form dimeric structures in DMSO
whereas 2b mainly exists as a monomer form. These two facts
seem to suggest dimers of 2a were unreactive. Formation of
(15) (a) Byers, P. K.; Canty, A. J.; Traill, P. R.; Watson, A. A. J.
Organomet. Chem. 1990, 390, 399. (b) de Graaf, W.; Boersma, J.; van
Koten, G. Organometallics 1990, 9, 1479.
(16) The derivatives include the palladacyclobutane and palladium(0)
species in Figure 2.

NoVel η3-Allylpalladium-Pyridinylpyrazole Complex
J. Am. Chem. Soc., Vol. 120, No. 40, 1998 10395
149.3, 136.7, 122.5, 119.9, 99.7, 31.6, 30.4; mp 106 °C. Anal. Calcd
for C12H15N3: C, 71.61; H, 7.51; N, 20.88. Found: C, 71.48; H, 7.56;
N, 20.64.
dimers depends on a substituent group on the pyrazole ring,
the solvent, and concentration. The exact structures of the
dimers are not clear at the present time. At least, more than
three kinds of methyl groups on the pyrazole ring were observed
in 13C NMR in DMF-d7. Structures and ratios of dimers varied
in the time-course, and this is the reason for the difficulty in
determining the structures.
[(η3-C3H5)Pd(MePPH)]+BF4- (1a). Dichloromethane (40 mL) was
added at 0 °C to a brown two-necked round-bottom flask containing
AgBF4 (216 mg, 1.11 mmol) and (η3-C3H5)PdCl dimer (203 mg, 0.555
mmol). Ligand 3a (177 mg, 1.11 mmol) was added to the mixture.
After the mixture was stirred at room temperature for 1.5 h, methanol
(40 mL) was added to the mixture. A white precipitate was filtered
off with Celite and membrane filter (Millipore, LCR25-LH), and the
filtrate was concentrated to give white solid. The white solid was
washed with CH2Cl2 on a filter paper and dried under reduced pressure
to give 1a (387 mg, 89%). 1H NMR (500 MHz, DMSO-d6) δ 14.25
(H2, br), 8.81 (H10, d, J ) 5.4 Hz), 8.24 (H8, dd, J ) 7.8, 7.8 Hz),
8.19 (H7, d, J ) 7.8 Hz), 7.62 (H9, dd, J ) 7.8, 5.4 Hz), 7.06 (H4, s),
5.93 (H14, tt, J ) 11.7, 6.4 Hz), 4.47 (H13syn and H15syn, 2H, d, J
) 6.4 Hz), 3.43 (H13anti and H15anti, 2H, d, J ) 11.7 Hz), 2.40 (H12,
Conclusion
We have developed stable η3-allylpalladium cationic and
neutral bisnitrogen ligand complexes 1a, 1b, 2a, and 2b. The
neutral complexes 2a and 2b generated from 1a and 1b have
high catalytic activity for cyclopropanation of ketene silyl acetals
with allylic acetates. Other cationic complexes which cannot
be converted into neutral form are ineffective for the cyclopro-
panation. It is considered that the catalytic ability of the neutral
complexes 2a and 2b depends on the stability of their derivatives
in DMSO.
1
3H, s); H NMR, 13C NMR, and 15N NMR in CD2Cl2-CD3OD (1:1),
see Table 1; mp >285 °C dec. Anal. Calcd for C12H14N3BF4Pd: C,
36.63; H, 3.59; N, 10.68. Found: C, 36.73; H, 3.56; N, 10.64.
[(η3-C3H5)Pd(tBuPPH)]+BF4 (1b) (480 mg, 97%) was prepared
-
from 3b (229 mg, 1.14 mmol) by using a similar procedure to that of
1a. 1b: 1H NMR (500 MHz, DMSO-d6) δ 13.95 (br, 1H), 8.83 (d,
1H, J ) 5.1 Hz), 8.28-8.22 (m, 2H), 7.62 (ddd, 1H, J ) 2.2, 5.1, 6.6
Hz), 7.19 (s, 1H), 5.93 (dt, 1H, J ) 6.5, 12.4 Hz), 4.54 (d, 1H, J ) 6.5
Hz), 3.46 (d, 1H, J ) 12.4 Hz), 1.38 (s, 9H); 13C NMR (125 MHz,
DMSO-d6) δ 157.8, 154.0, 151.4, 150.1, 141.0, 126.0, 122.1, 117.9,
101.1, 62.1 (2C), 31.3, 29.6; mp >140 °C dec. Anal. Calcd for
C15H20N3BF4Pd: C, 41.37; H, 4.63; N, 9.65. Found: C, 41.17; H,
4.59; N, 9.54.
(η3-C3H5)Pd(MePP) (2a). To a mixture of 1a (4.1 mg) and sodium
methoxide (5.2 mg, excess) were added dichloromethane-d2 (0.25 mL)
and methanol-d4 (0.25 mL) at room temperature, and the mixture was
stirred under sonication for 5 min. After the precipitate was filtered
off, the solution containing 2a was obtained: 1H NMR, 13C NMR, and
15N NMR in CD2Cl2-CD3OD (1:1), see Table 1; HRMS (FAB) calcd
for C12H13N3Pd 306.0228, found 306.0229.
Experimental Section
General Procedure. Methylene chloride, ethanol, and methanol
were dried over activated molecular sieves 4A or 3A prior to use.
Dehydrated ether and THF were purchased from Kanto Chemical Co.,
Inc. Commercially available DMSO (special grade) was used without
purification or dryness. NMR spectra were obtained from JEOL R-600,
JEOL GSX-500, and JEOL AL-300. 1H NMR chemical shift are
reported in ppm from tetramethylsilane (0 ppm) in CDCl3 and a mixture
of CD2Cl2 and CD3OD, and residual DMSO (3.35 ppm) in DMSO-d6.
13C NMR chemical shifts are reported in ppm from tetramethylsilane
(0 ppm) in a mixture of CD2Cl2 and CD3OD, and residual CDCl3 (77.0
ppm) or DMSO (39.5 ppm). 15N NMR spectra were obtained by the
1H-15N PEG-HMBC method, and chemical shifts are reported in ppm
from NH4NO3 (0 ppm) in DMSO-d6. Melting points were determined
on a Yanaco MP-500 melting point apparatus and were not corrected.
Analytical gas chromatography was performed on a Hewlett-Packard
5890 with J&W Scientific DB-5 (15 m × 0.25 mm). Mass spectral
analyses were performed on a JEOL JMS-HX100. Microanalyses were
performed by the Division of Chemical Analysis in RIKEN. Flash
chromatographies were performed by using Merck silica gel 60 (230-
400 mesh).
2-H-3-Methyl-5-(2-pyridynyl)pyrazole (MePPH, 3a).7 To a sus-
pension of NaH (60 wt % in mineral oil, 5.2 g, 130 mmol) in THF (50
mL) at 0 °C was added acetone (9.54 mL, 130 mmol). The mixture
was stirred at room temperature for 20 min, and then allowed to heat
to 60 °C. Ethyl 2-pyridinecarboxylate (16.5 g, 100 mmol) in THF (50
mL) was slowly added to the mixture. After the mixture was stirred
at 70 °C for 20 min, dilute HCl solution was added to the mixture
until pH 8-9 at 0 °C. The mixture was extracted with diethyl ether
(10 mL X 4), and the combined organic layer was washed with brine,
dried over MgSO4, and concentrated in vacuo to give yellow oil (17.9
g). To a refluxing solution of the above oil (17.9 g) in EtOH (180
mL) was added dropwise for 10 min, hydrazine monohydrate (9.7 mL,
200 mmol) in EtOH (20 mL). After the mixture was refluxed for 1.5
h, the solvent was removed under reduced pressure. The residue was
dissolved in CH2Cl2, and the solution was washed with water, dried
over MgSO4, and concentrated in vacuo. The residual solid was washed
with ether on a filter paper and dried under reduced pressure to give
white solid 3a (11.3 g, 71%): 1H NMR (500 MHz, CD3OD) δ 8.55 (br
d, 1H, J ) 5.0 Hz), 7.84 (br d, 1H, J ) 8.2 Hz), 7.78 (dt, 1H, J ) 1.4,
7.8 Hz), 7.28-7.24 (m, 1H), 6.62 (s, 1H), 2.36 (s, 3H).
Preparation of X from 1a. Suspension of 1a (18.5 mg) in CH2Cl2
(5 mL) was washed with saturated aqueous NaHCO3 in a separatory
funnel. An organic layer was dried over MgSO4 and evaporated under
1
reduced pressure to give X (16.0 mg) as pale yellow oil. Data for H
and 13C NMR in DMF-d7 and FAB mass spectra of X are available in
the Supporting Information.
(η3-C3H5)Pd(tBuPP) (2b). White solid 2b (24 mg) was prepared
1
from 1b (49.3 mg) by using a similar procedure to that of X. 2b: H
NMR (300 MHz, CDCl3) δ 8.41 (br d, 1H, J ) 5.3 Hz), 7.75 (dt, 1H,
J ) 1.3, 7.9 Hz), 7.56 (br d, 1H, J ) 7.9 Hz), 7.03 (br t, 1H, J ) 6.0
Hz), 6.46 (s, 1H), 5.60 (tt, 1H, J ) 7.0, 12.5 Hz), 4.39 (br d, 1H, J )
7.0 Hz), 3.79 (br d, 1H, J ) 7.0 Hz), 3.17 (br t, 2H, J ) 12.5 Hz),
1.39 (s, 9H); 13C NMR (75 MHz, CDCl3) δ 164.7, 155.2, 152.4, 150.9,
138.9, 121.1, 119.5, 115.6, 98.4, 59.4, 56.5, 32.2, 31.1; 1H NMR (300
MHz, DMSO-d6) δ 8.59 (br d, 1H, J ) 5.3 Hz), 7.94 (br t, 1H, J ) 7.7
Hz), 7.72 (br d, 1H, J ) 8.3 Hz), 7.20 (br t, 1H, J ) 6.2 Hz), 6.49 (s,
1H), 5.72 (tt, 1H, J ) 9.2, 9.9 Hz), 4.00 (br, 2H), 3.13 (br, 2H), 1.25
(s, 9H); mp >150 °C dec; HRMS (FAB) calcd for C15H19N3Pd
348.0698, found 348.0698.
2, 3-Dimethyl-5-(2-pyridynyl)pyrazole (Me2PP). A mixture of 1,3-
dimethyl-5-(2-pyridynyl)pyrazole and 2,3-dimethyl-5-(2-pyridynyl)-
pyrazole (Me2PP) was obtained from ethyl 2-pyridinecarboxylate,
acetone, and methylhydrazine according to the procedure for 3a. These
products were isolated by flash chromatography (25% EtOAc in
hexane). 2,3-Dimethyl-5-(2-pyridynyl)pyrazole (Me2PP, 29%): 1H
NMR (500 MHz, CDCl3) δ 8.62-8.58 (m, 1H), 7.85 (br d, 1H, J )
8.2 Hz), 7.69-7.63 (m, 1H), 7.17-7.11 (m, 1H), 6.62 (s, 1H), 3.83 (s,
3H), 2.30 (s, 3H); 13C NMR (67.9 MHz, CDCl3) δ 152.1, 149.7, 149.0,
139.6, 136.1, 121.8, 119.4, 103.6, 36.2, 11.2; mp 97-99 °C dec. Anal.
Calcd for C10H11N3: C, 69.34; H, 6.40; N, 24.26. Found: C, 69.24;
H, 6.45; N, 23.83. 1,3-Dimethyl-5-(2-pyridynyl)pyrazole) (27%,
colorless oil, less polar product than Me2PP): 1H NMR (500 MHz,
CDCl3) δ 8.62 (br d, 1H, J ) 4.6 Hz), 7.69-7.65 (m, 1H), 7.49 (br d,
1H, J ) 8.2 Hz), 7.19-7.15 (m, 1H), 6.34 (s, 1H), 4.14 (s, 3H), 2.29
2-H-3-tert-Butyl-5-(2-pyridynyl)pyrazole (tBuPPH, 3b) was pre-
pared in 65% yield (1.18 g) from ethyl 2-pyridinecarboxylate (1.65 g,
10 mmol), 3,3-dimethyl-2-butanone (1.1 g, 11 mmol), and hydrazine
monohydrate (9.7 mL, 200 mmol) by using a similar procedure to that
1
of 3a. 3b: H NMR (300 MHz, CDCl3) δ 8.61 (d, 1H, J ) 4.6 Hz),
7.81 (br d, 1H), 7.72 (dt, 1H, J ) 1.3, 7.3 Hz), 7.20 (dd, 1H, J ) 4.6,
7.2 Hz), 6.67 (s, 1H), 1.39 (s, 9H); 13C NMR (75 MHz, CDCl3) δ

10396 J. Am. Chem. Soc., Vol. 120, No. 40, 1998
Satake and Nakata
(s, 3H); 13C NMR (67.9 MHz, CDCl3) δ 149.6,148.6, 146.6, 141.5,
as a internal standard. Pure cyclopropane derivatives were obtained
after dihydroxylation with OsO4 and purification by flash chromatog-
raphy.
136.2, 122.2, 121.8, 105.6, 38.7, 13.2.
[(η3-C3H5)Pd(Me2PP)]+BF4 (8) was prepared in 53% yield from
-
1
Me2PP by using a similar procedure to that of 1a. 8: H NMR (300
5:6a 1H NMR (270 MHz, CDCl3) δ 3.71 (s, 3H), 2.62-1.94 (m,
2H), 1.68-1.53 (m. 3H), 1.37-1.03 (m, 5H), 0.94-0.81 (m, 1H), 0.31
(d, 4H); 13C NMR (67.9 MHz, CDCl3) δ 176.3, 51.3, 46.4, 32.6, 25.8,
23.6, 20.8, 0.71.
MHz, DMSO-d6) δ 8.79 (br d, 1H, J ) 5.1 Hz), 8.23 (br t, 1H, J ) 7.6
Hz), 8.17 (br d, 1H, J ) 7.8 Hz), 7.60 (br t, 1H, J ) 6.2 Hz), 7.10 (s,
1H), 5.94 (tt, 1H, J ) 7.0, 12.6 Hz), 4.59 (d, 2H, J ) 7.0 Hz), 3.86 (s,
3H), 3.54 (d, 2H, J ) 12.6 Hz), 2.42 (s, 3H); 13C NMR (75 MHz,
DMSO-d6) δ 153.9, 150.1, 149.8, 144.4, 141.1, 125.9, 121.9, 117.9,
104.8, 62.5, 37.6, 11.6; mp >170 °C dec. Anal. Calcd for C13H16N3-
BF4Pd: C, 38.32; H, 3.96; N, 10.31. Found: C, 38.25; H, 3.91; N,
10.25.
11:6d (volatile) 1H NMR (500 MHz, CDCl3) δ 4.08 (q, 2H), 1.21 (t,
3H), 1.01 (s, 3H), 1.00 (m, 1H), 0.35-0.22 (m, 4H); 13C NMR (125.8
MHz, CDCl3) δ 177.9, 60.2, 41.1, 22.9, 19.5, 14.5, 0.70; MS, m/z (rel
intensity) 156 (M+, 3.3), 141 (17.5), 128 (6.6), 113 (9.9), 110 (10.4),
100 (24.8), 83 (100), 67 (11.6), 55 (99.3).
1
Stoichiometric Reaction. (a) Allylation of Dimethyl Malonate.
To a suspension of 1a (20 mg, 0.0508 mmol) in THF (2 mL) was
added sodium enolate of dimethyl malonate [generated from dimethyl
malonate (15 mg, 0.112 mmol) and NaH (60 wt % in mineral oil, 4.9
mg, 0.122 mmol) in THF (1 mL) at 0 °C]. After the mixture was
stirred for 1 h, precipitate was removed through Celite, and the filtrate
was concentrated in vacuo. The residue was purified by flash
chromatography to give dimethyl allylmalonate (4.0 mg, 46%).
(b) Cyclopropanation of Ketene Silyl Acetal. To a mixture of 1a
(31.5 mg, 0.080 mmol) and NaOAc (26.2 mg, 0.32 mmol) in DMSO
(5 mL) was added ketene silyl acetal 4 (35 µL, ca. 0.16 mmol) at room
temperature. After the mixture was stirred for 0.5 h, ether and dilute
HCl solution were added to the mixture, and the mixture was stirred
for 10 min. The mixture was extracted with ether, and the combined
organic layer was dried over MgSO4 and concentrated in vacuo. Decane
(9.4 mg) was added as an internal standard to the residue, and yields
and the ratio of a mixture of 5 and 6 were obtained (see in Scheme 8).
Catalytic Reaction. All the catalytic reactions were carried out
under argon. The appropriate allyl acetate (1 mmol) and ketene silyl
acetal (2 mmol) were added to a solution containing cationic palladium
catalyst (0.05 or 0.01 mmol) and NaOAc (0.2 or 0.04 mmol, 4 equiv
of palladium) in DMSO. The reaction mixture was stirred at room
temperature, and progress of the reaction was monitored by GLC or
TLC analysis. Diethyl ether and diluted HCl solution were added to
the reaction mixture, and the mixture was extracted with ether. The
organic layer was washed with saturated NaHCO3 solution and brine,
dried over MgSO4, and concentrated in vacuo. From the residue, a
mixture of cyclopropane derivatives and allylated compounds was
obtained by distillation (Kugelrohr apparatus) and flash chromatography.
Yields and ratios were determined by GLC and NMR with n-decane
13: H NMR (500 MHz, CDCl3) δ 7.29 (t, 2H), 7.18 (t, 1H), 7.13
(d, 2H), 4.17 (m, 2H), 1.99 (s, 3H), 1.93 (ddd, 1H, J ) 5.0, 5.7, 9.2
Hz), 1.39 (ddd, 1H, J ) 5.0, 5.7, 8.7 Hz), 1.26 (t, 3H, J ) 7.1 Hz),
1.20 (s, 3H), 1.01 (ddd, 1H, J ) 5.5, 5.7, 8.7 Hz), 0.90 (ddd, 1H, J )
5.5, 5.7, 9.2 Hz); 13C NMR (125 MHz, CDCl3) δ 177.3, 143.0, 128.2,
126.1, 125.4, 60.4, 41.5, 31.1, 23.2 (2C), 19.1, 14.2, 11.3. Anal. Calcd
for C15H20O2: C, 77.55; H, 8.69. Found: C, 77.27; H, 8.85.
14 could not be isolated from a mixture of 13 and 14. 14: 1H NMR
(500 MHz, CDCl3) δ 7.30-7.05 (m, 5H), 6.35 (d, 1H, J ) 15.3 Hz),
6.11 (dt, 1H, J ) 6.8, 15.3 Hz), 4.10-4.04 (m, 2H), 2.39 (d, 2H, J )
6.8 Hz), 1.19 (t, 3H, J ) 7.1 Hz), 1.18 (s, 6H); MS, m/z (rel intensity)
232 (M+, 31.6), 159 (18.4), 117 (100), 91 (12.3).
Reaction with Complexes 7 and 8. The reactions with complexes
7 and 8 were carried out according to the procedure described above.
The unsaturated ester 912 was obtained in 60% (in the case of 7) and
58% (in the case of 8) yields, respectively, based on ketene silyl acetal
4.
Acknowledgment. A.S. gratefully acknowledges the finan-
cial support by the Special Postdoctoral Researchers Program.
We thank Dr. Hiroyuki Koshino for measurement of NMR
spectra and Ms. Kumiko Harata for operation of FAB mass
spectra.
Supporting Information Available: NMR data of X in
DMF-d7, mass spectra of X and 2b (3 pages, print/PDF). See
any current masthead page for ordering information and Web
access instructions.
JA982269C