Palladium-catalyzed allylic alkylations via titanated nucleophiles: A new early-late heterobimetallic system

Full text of DOI:10.1021/jo982266i

2
962
J . Org. Chem. 1999, 64, 2962-2965
P a lla d iu m -Ca ta lyzed Allylic Alk yla tion s via
Tita n a ted Nu cleop h iles: A New Ea r ly-La te
Heter obim eta llic System
Sch em e 1
Giovanni Poli,* Giuliano Giambastiani,* and
Alessandro Mordini
Dipartimento di Chimica Organica “Ugo Schiff” and Centro
CNR di Studio per la Chimica e la Struttura dei Composti
Eterocicli e loro Applicazioni, Via G. Capponi 9,
I-50121 Firenze, Italy
7
enolate, which cyclizes onto the concomitantly generated
iodonium ion (eq 1).
Received November 13, 1998
The palladium(0)-catalyzed allylic alkylation of soft
carbon nucleophiles is a very useful organometallic
1
transformation that allows the clean allylation of a great
variety of active methylene compounds with high and
predictable regio-, diastereo-, and enantioselection. In the
presence of suitable ligands, the allylic system is known
to interact with the palladium(0) complex to generate a
cationic π-allyl palladium species, which is in turn
intercepted by the anion of the carbon nucleophile
Since the chemistry of the palladium-catalyzed allylic
alkylations has been mainly focused on alkali metal
enolates of active methylenes, we decided to investigate
such postulated titanium enolates in the contest of allylic
alkylations. Our results are reported in Table 1.
In a preliminary experiment, we heated diethyl mal-
8
(Scheme 1).
onate (1a ) with cinnamyl acetate (2a ) in CH
in the presence of Ti(OPr-i) (1.2 equiv), Pd (dba)
equiv), and PPh (0.5 equiv) (entry 1A). Much to our
satisfaction, we isolated the expected allylated product
a a in a very encouraging 86% yield. Two comments are
worthy of note: (a) No apparent Lewis acid-base conflict
between Ti(OPr-i) and PPh ensued. (b) The same allylic
alkylation under neutral conditions failed, since the pK
2
Cl
2
for 9 h
The way the nucleophile is deprotonated is dependent
on a few parameters. In particular, when the pK value
a
4
2
3
(0.05
3
of the conjugated acid of the displaced group is higher
than that of the active methylene, deprotonation may
take place in situ by the displaced anion itself. Such an
endogenous deprotonation mode can be at work on a wide
3
2
4
3
variety of carbon acids when allylic carbonates, phenox-
3
4
a
ides, and oxiranes are used. In the case of the more
common allylic acetates, the carbanionic species is gener-
ally, either preformed or generated in situ by the explicit
addition of a stoichiometric amount of base. We recently
found that endogenous deprotonation is viable with allylic
value of diethyl malonate is not in the suitable base-free
5
window. When, for comparison, the same reaction was
run in the presence of BSA/AcOK9 as the enolizing
system, 3a a was obtained in a slightly decreased 83%
yield (entry 1B). In order to evaluate the scope of this
new method, we extended such a comparative study
window.5
acetates too, though on a more restricted pK
a
As an extension of this investigation, we next decided to
look for alternative enolizing conditions, so as to enable
reaction of a broader range of carbanionic nucleophiles.
Taguchi and co-workers reported that the iodocarbocy-
4
(method A, Ti(OPr-i) ; method B, BSA/AcOK cat.) to other
palladium-catalyzed allylic alkylations. Methyl nitroac-
etate (1b) reacted with cinnamyl acetate (2a ) only under
method A, with method B affording just recovered start-
ing material (entries 2A and 2B). Method A was also
successful in the reaction between diphenyl disulfone (1c)
and either geranyl acetate (2b) (entry 3) or cinnamyl
acetate (2a ) (entry 4A). The reactions between (Z)-1,4-
diacetoxy-but-2-ene (2c) and diethyl malonate (1a ) (en-
clization of 4-pentenyl malonates could be achieved in
6
the presence of I
2
and Ti(OBu-t)
4
.
According to the
authors, the mechanism of this cyclization involves the
titanium alkoxide-mediated generation of a titanium
*
To whom correspondence should be addressed. E-mail: poli@
risc1.chimorg.unifi.it.
1) (a) Hegedus L. S. In Organometallics in Synthesis; Schlosser,
(
(7) For other recent reports on titanium enolates, see: (a) Evans,
D. A.; Urpi, F.; Somers, T. C.; Clark, J . S.; Bilodeau, M. T. J . Am. Chem.
Soc. 1990, 112, 8215. (b) Williams, R. M.; Esslinger, C. S. Tetrahedron
Lett. 1991, 32, 3635. (c) Mahrwald, R. Chem. Ber. 1995, 128, 919. (d)
Bernardi, A.; Cavicchioli, M.; Potenza, D.; Scolastico, C. J . Org. Chem.
1994, 59, 3690. (e) Mahrwald, R. Tetrahedron 1995, 51, 9015. (f)
Mahrwald, R.; Costisella, B. Synthesis 1996, 1087. (g) Matsumura, Y.;
Nishimura, M.; Hiu, H.; Watanabe, M.; Kise, N. J . Org. Chem. 1996,
61, 2809.
M., Ed.; J ohn Wiley: Chichester, 1994; Chapter 5, pp 385-459. (b)
Tsuji, J . Palladium Reagents and Catalysts; J ohn Wiley: New York,
1
995. (c) Heck, R. F. Palladium Reagents in Organic Synthesis;
Academic Press: Orlando, 1985. (d) Frost, C. G.; Howarth, J .; Williams,
J . M. J . Tetrahedron: Asymmetry 1992, 3, 1089. (e) Harrington P. J .
In Comprehensive Organometallic Chemistry II; Abel, E. W., Stone, F.
G. A., Wilkinson, G., Eds.; Pergamon: Oxford, 1995; Vol. 12, Chapter
8
9
.2, pp 798-903. (f) Heumann, A.; R e´ glier, M. Tetrahedron 1995, 51,
75.
(8) To the best of our knowledge, the only precedents combining
(
2) (a) Tsuji, J .; Minami, I. Acc. Chem. Res. 1987, 20, 140. (b) Tsuji,
J . Tetrahedron 1986, 42, 4361. (c) Safi, M.; Sinou, D. Tetrahedron Lett.
991, 32, 2025.
3) Tsuji, J .; Okumoto, H.; Kobayashi, Y.; Takahashi, T. Tetrahedron
Lett. 1981, 22, 1357.
4) Tsuji, J .; Kataoka, H.; Kobayashi, Y. Tetrahedron Lett. 1981, 22,
575.
palladium catalysis and Ti(OPr-i)
4
, deal with the addition of CO, Zn-
(acac) , and phenols to allylic alcohols: (a) Itoh, K.; Hamaguchi, N.;
2
1
Miura, M.; Nomura, M. J . Mol. Catal. 1992, 75, 117. (b) Itoh, K.;
Hamaguchi, N.; Miura, M.; Nomura, M. J . Chem. Soc., Perkin Trans.
1 1992, 2833. (c) Satoh, T.; Ikeda, M.; Miura, M.; Nomura, M. J . Org.
Chem. 1997, 62, 4877. For the structure of a Ti(IV)/Pd(II)-based
complex, see: (d) Kless, A.; Lefeber, C.; Spannenberg, A.; Kempe, R.;
Baumann, W.; Holz, J .; B o¨ rner, A. Tetrahedron 1996, 52, 14599.
(9) BSA/AcOK cat. is a standard enolizing system in palladium-
catalyzed allylic alkylations. Trost, B. M.; Murphy, D. J . Organome-
tallics 1985, 4, 1143.
(
(
2
(
(
5) Poli, G.; Giambastiani, G. J . Org. Chem. 1998, 63, 9608.
6) (a) Kitagawa, O.; Inoue, T.; Hirano, K.; Taguchi, T. J . Org. Chem.
1
993, 58, 3106. (b) Inoue, T.; Kitagawa, O.; Oda, Y.; Taguchi, T. J .
Org. Chem. 1996, 61, 8256.
1
0.1021/jo982266i CCC: $18.00 © 1999 American Chemical Society
Published on Web 04/01/1999

Notes
J . Org. Chem., Vol. 64, No. 8, 1999 2963
Ta ble 1. P a lla d iu m -Ca ta lyzed Allylic Alk yla tion s: A Com p a r ison betw een Ti(OP r -i)4 a n d BSA/AcOK a s P r om oter sa
a
All the reactions were run at reflux of the solvent. b Method A: Pd2(dba)3 0.05 equiv, PPh3 0.5 equiv, Ti(OPr-i)4 1.2-1.5 equiv. Method
B: Pd2(dba)3 0.05 equiv, PPh3 0.5 equiv, BSA (1.2-1.5 equiv)/AcOK (0.1 equiv). Yields of isolated product. d Since the calculation is
based on the reacted acetate, a theoretical yield of 50% has to be considered.
c
tries 5A and 5B) and those between cinnamyl acetate (2a )
and dimethyl methylmalonate (1d ) (entries 6A and 6B)
all worked straightforwardly. The reaction between cin-
namyl acetate (2a ) and malononitrile (1e) gave the
diallylated product 4ea exclusively, with both methods
2 2
the IR spectrum of the same mixture in CH Cl showed
-
1
the presence of two bands at 1729 and 1625 cm , which
can be assigned to the free and Ti-coordinated carbonyl
stretching, respectively.1 Since no apparent enolization
2,13
1
4,15
was detectable in the presence of only Ti(OPr-i)4,
it
(
entries 7A and 7B). As a further experiment, we tested
was concluded that either the enolization equilibrium is
dramatically shifted to the left side or another anionic
species, present in the medium, had to be responsible for
the deprotonation step. Indeed, when a catalytic amount
the intramolecular allylic alkylation of the amide 5, a
precursor of the 3,4-disubstituted pyrrolidin-2-one 6,
recently studied by us.10 Here again, method A proved
to be superior with respect to the reference method B
+
-
of n-Bu N AcO was added to the solution of spectrum
4
(eq 2).
c, as a soluble acetate source, the immediate and total
disappearance of the methylene singlet of the malonate
indicated that a rapidly equilibrating enolate species was
generated (spectrum d). This fact suggested that the
acetate counterion of the transiently formed π-allyl
complex is the previously postulated external deproto-
nating agent.16 The results obtained are consistent with
the following sequence of events: (a) coordination of the
titanium complex on the basic site(s) of the nucleophile
and concomitant oxidative addition of the Pd(0) catalyst
to form a cationic π-allyl-complex, (b) acetate anion-
mediated deprotonation of the Ti-coordinated nucleophile
4
Worthy of note, although Ti(OPr-i) is a known trans-
esterification promoter,11 no trace, or only tiny amounts,
of isopropyl esters have been detected in the experiments
involving esters as nucleophiles. Some preliminary NMR
experiments helped to clarify the mechanism of this new
process (Figure 1).
(12) Finn, M. G. Sharpless, K. B. In Asymmetric Synthesis; Morrison,
J . D., Ed.; Academic Press: New York, 1985; Vol. 5, Chapter. 8.
(13) The Lewis base/acid interaction between carbonyl esters and
Ti(OPr-i) is a rapid process on the NMR time scale at room temper-
4
ature, involving the fluxional exchange between bound and unbound
carbonyl esters. See ref 12.
1
In particular, the H NMR spectrum of an equimolar
4
mixture of diethyl malonate and Ti(OPr-i) (spectrum c)
in CDCl showed no significant deviation with respect
3
to the simple superimposition of the spectra of the two
single components (spectra a and b). On the other hand,
(14) A slow Ti(OPr-i)
only detectable reaction.
4
15) For a hint on the weak basic character of Ti(OR) , see: Sasai,
4
-mediated monotransesterification was the
(
H.; Suzuki, T.; Arai, S.; Arai, T.; Shibasaki, M. J . Am. Chem. Soc. 1992,
114, 4418.
(16) This result suggests that the mechanism leading to the titanium
enolates postulated by Taguchi in the iodocarbocyclization process
might need a slight revision, i.e., the requirement of an external
deprotonating agent, such as the iodide anion released in the iodionium
ion formation, or an added base.
(
10) Giambastiani, G.; Pacini, B.; Porcelloni, M.; Poli, G. J . Org.
Chem. 1998, 63, 804.
11) Seebach, D.; Weidmann, B.; Widler, L. In Modern Synthetic
Methods; Scheffold, R., Ed. Salle + Sauerlander: Aarau, 1983; pp 217-
51. As expected, copious or total transesterification products can be
detected with highly prolonged reaction times.
(
3

2
964 J . Org. Chem., Vol. 64, No. 8, 1999
Notes
Sch em e 2
window, and it often compares favorably with the most
9
popular enolizing system based on BSA/AcOK. Exten-
F igu r e 1. Generation of the titanium enolate of diethyl
sion of the present method to asymmetric variants by the
use of chiral titanium alkoxides, possibly in substoichio-
metric amounts, is currently under investigation.
1
malonate. H NMR 200 MHz spectra in CDCl
3
at rt: (a) diethyl
1.0:
malonate; (b) Ti(OPr-i) ; (c) diethyl malonate/Ti(OPr-i)
4
4
-
+
1
1
4 4
.0 equiv; (d) diethyl malonate/Ti(OPr-i) /n-Bu N AcO 1.0:
.0:0.1 equiv.
Exp er im en ta l Section
to generate a titanium enolate, or titanated species,17 and
c) reductive coupling between the two organometallic
species to give the allylated product (Scheme 2).
Gen er a l P r oced u r e for th e P a lla d iu m -Ca ta lyzed Allylic
Alk yla tion via Tita n a ted Sp ecies (Table 1, Entries 1A, 2A,
, 4A, 5A, 6A, 7A, and 8). To a solution of the allylic acetate (0.5
(
3
It should be noted that the acetate anion-promoted
deprotonation is a mechanistic feature in common with
the previously reported acetate/base-free mechanism.
However, in the present case the greater Lewis acidity
mmol) and the carbon nucleophile (0.6 mmol) in the appropriate
solvent (10 mL) at room temperature and under nitrogen
atmosphere was added Ti(OPr-i)
resulting solution stirred for 10 min. Pd
PPh (0.25 mmol) were then premixed in a separated vessel and
4
(0.7 equiv) dropwise and the
2
(dba) (0.025 mmol) and
3
3
of Ti(IV) with respect to that of Pd(II) lowers the pK
a
added. After the appropriate reflux time (see Table 1), the
value of the coordinated active methylene18 to the point
that deprotonation of a wider range of nucleophilic
species becomes now possible.
reaction was then treated with H O (10 mL), and the resulting
2
white precipitate (TiO
with CH Cl . The organic phase was separated and the aqueous
phase extracted with CH Cl
2
) was filtered off on a Celite pad, washing
2
2
2
2
(3 × 5 mL). The collected organic
4
In summary, we have shown that Ti(OPr-i) promotes
the palladium-catalyzed addition of a wide range of
carbon acids on allylic acetates. This new protocol is
layers were dried, and the solvent was removed in vacuo to afford
the crude product.
Gen er a l P r oced u r e for th e P a lla d iu m -Ca ta lyzed Allylic
Alk yla tion via BSA/AcOK (Table 1, Entries 1B, 2B, 4B, 5B,
operationally very simple, it spans a wide nucleophile pK
a
6
B, and 7B). To a solution of the allylic acetate (0.5 mmol) and
the carbon nucleophile (0.6 mmol) in THF (10 mL) were added
bis(trimethylsilyl)acetamide (0.6 mmol) and AcOK (0.05 mmol)
(
17) At present we do not have structural information on these
transiently generated titanated species. In particular, study of the
titanated derivative of the disulfone (1c), for which a true enolate form
is forbidden, might prove of structural interest. For studies on the
structure of R-lithiated sulfones, see: (a) Gais, H.-J .; Hellmann, G.;
G u¨ nther, H.; Lopez, F.; Lindner, H. J .; Braun, S. Angew. Chem., Int.
Ed. Engl. 1989, 28, 1025. (b) Hollstein, W.; Harms, K.; Marsch, M.;
Boche, G. Angew. Chem., Int. Ed. Engl. 1988, 27, 846. (c) Boche, G.
Angew. Chem., Int. Ed. Engl. 1989, 28, 277.
in this order with stirring under nitrogen atmosphere. Pd
0.025 mmol) and PPh (0.25 mmol), were then premixed in a
separated vessel and added. After the appropriate reflux time
(see table) a saturated aqueous solution of NH Cl (10 mL) was
added and the organic phase was extracted with Et
2 3
(dba)
(
3
4
2
O (5 × 3
mL). The collected organic layers were dried and the solvent was
removed in vacuo to afford the crude product.
(18) (a) Constable, E. C. Metals and Ligand Reactivity; VCH:
(
E)-5-P h en yl-2-n itr open t-4-en oic Acid Meth yl Ester (3ba).
Flash chromatography (hexanes/AcOEt 80:20) gave pure 3ba as
an oil. 1H NMR (CDCl
): δ 7.34-7.24 (5H), 6.56 (d, 1H, J )
Weinheim, 1995. (b) Murahashi, S.-I.; Naota, T.; Taki, H.; Mizuno, M.;
Takaya, H.; Komiya, S.; Mizuho, I.; Oyasato, N.; Hiraoka, M.; Hirano,
M.; Fukuoka, A. J . Am. Chem. Soc. 1995, 117, 12436.
3

Notes
J . Org. Chem., Vol. 64, No. 8, 1999 2965
1
5
5
1
3
6.1 Hz), 6.09 (dt, 1H, J ) 16.1, 7.3 Hz), 5.24 (dd, 1H, J ) 8.8,
.9 Hz), 3.84 (s, 3H), 3.17-3.06 (2H). 13C NMR (CDCl
): δ 33.74,
3.66, 87.49, 121.06, 126.43, 128.05, 128.65, 135.48, 136.22,
117 (100); 91 (24). Anal. Calcd for C21
N, 9.39. Found: C, 84.29; H, 5.97; N, 9.31.
E)-2-Met h yl-2-(3-p h en yla llyl)m a lon ic Acid Dim et h yl
Ester (3d a ). Flash chromatography (hexanes/Et O 90:10) gave
): δ ) 7.35-7.20 (5H),
.45 (d, 1H, J ) 15.7 Hz), 6.08 (dt, 1H, J ) 15.8, 7.3 Hz), 3.73
18 2
H N : C, 84.53; H, 6.08;
3
(
-
1
+
64.52. IR (CDCl
); 188 (14); 129 (100); 117 (13); 91 (58). Anal. Calcd for C12
: C, 61.27; H, 5.57; N, 5.95. Found: C, 61.32; H, 5.70; N,
.03.
E)-4,4-Bis(p h en ylsu lfon yl)-1-p h en ylbu ten e (3ca ). Flash
chromatography (hexanes/AcOEt 80:20) gave the pure 3ca as
3
): 3033, 2960, 1754 cm . MS m/z: 235 (M ,
2
H
13
-
1
pure 3d a as a colorless oil. H NMR (CDCl
3
NO
6
4
6
(
13
s, 6H), 2.77 (dd, 2H, J ) 7.5, 1.4 Hz), 1.45 (s, 3H). C NMR
CDCl ): δ 19.76, 39.37, 52.31, 54.01, 124.03, 125.96, 127.20,
28.22, 134.04, 137.29, 172.33. IR (CDCl ): 3032, 3002, 2956,
728 cm . MS m/z: 262 (M , 16); 202 (39); 143 (74); 117 (100),
(
(
3
1
1
9
6
3
1
white crystals. Mp 151-152 °C. H NMR (CDCl
3
): δ 8.00-7.95
-1
+
(4H), 7.73-7.51 (6H), 7.34-7.19 (5H), 6.31 (d, 1H, J ) 15.8 Hz),
1 (27). Anal. Calcd for C15
8.50; H, 7.01.
18 4
H O : C, 68.69; H, 6.92. Found: C,
6
3
1
.06 (dt, 1H, J ) 15.8, 6.6 Hz), 4.56 (t, 1H, J ) 6.2 Hz), 3.13-
13
.06 (2H). C NMR (CDCl
27.89, 128.58, 129.18, 129.71, 134.15, 134.70, 136.34, 138.01.
3
): δ 29.39, 82.72, 123.43, 126.36,
2
-(4-Acetoxybu t-2-en yl)m alon ic Acid Dieth yl Ester (3ac).
-
1
Flash chromatography (hexanes/AcOEt 75:25) gave pure 3a c as
IR (CDCl
3
): 3067, 3030, 2929, 1329, 1147 cm . MS m/z: 413
1
+
a colorless oil. H NMR (CDCl
.80 Hz), 4.20 (q, 4H, J ) 7.3 Hz), 3.40 (t, 1H, J ) 7.7 Hz), 2.65
2H), 2.04 (s, 3H), 1.26 (t, 6H, J ) 7.3 Hz). 13C NMR (CDCl
): δ
14.04, 20.91, 31.36, 51.55, 61.49, 64.55, 127.12, 130.95, 168.78,
3
): δ 5.67 (2H), 4.49 (d, 2H, J )
(M + 1 , 2); 271 (14); 129 (57); 128 (100); 91 (27); 77 (78). Anal.
4
(
)
Calcd for C22
.07.
6,7-E)-9,9-Bis(p h en ylsu lfon yl)-2,6-d im eth yln on a -2,6-d i-
en e (3cb). Flash chromatography (hexanes/AcOEt 80:20) gave
20 4 2
H O S : C, 64.06; H, 4.89. Found: C, 64.31; H,
3
5
(
-
1
+
1
3
70.75. IR (CDCl ): 2985, 1729 cm . MS m/z: 272 (M , 22);
1
212 (20); 199 (29); 113 (100).
pure 3cb as white crystals. Mp: 61-63 °C). H NMR (CDCl
δ 7.95 (4H), 7.72-7.51 (6H), 5.03 (2H), 4.27 (t, 1H, J ) 6.3 Hz),
.90-2.84 (2H), 2.05-1.84 (4H), 1.67 (s, 3H), 1.58 (s, 3H), 1.45
3
):
2
Ack n ow led gm en t. This work was supported in part
by grants from Consiglio Nazionale delle Ricerche (CNR
Roma) and Ministero dell’Universit a` e della Ricerca
Scientifica e Tecnologica (MURST, Cofin 98-99).
1
3
(s, 3H). C NMR (CDCl
3
): δ 16.10, 17.70, 24.60, 25.67, 26.24,
9.42, 84.26, 118.02, 123.81, 129.09, 129.63, 134.55, 138.19,
39.61. IR (CDCl
): 3073, 2970, 2918, 2856, 1329, 1150 cm-1
3
1
3
+
.
MS m/z: 432 (M , 2); 291 (18).
E)-Bis(3-p h en ylp r op -2-en yl)m a lon on itr ile (4ea ). Flash
chromatography (hexanes/AcOEt 90:10) gave the pure 4ea as
(
Su p p or tin g In for m a tion Ava ila ble: 1H NMR spectra of
compounds 3a a , 3ba , 3cb, 3ca , 3a c, 3d a , and 4ea . This
material is available free of charge via the Internet at
http://pubs.acs.org.
1
white crystals. Mp 116-118 °C. H NMR (CDCl
3
): δ 7.48-7.25
(
2
1
10H), 6.71 (d, 2H, J ) 15.7 Hz), 6.26 (dt, 2H, J )15.7, 7.7 Hz),
.90 (dd, 4H, J )7.7, 1.1 Hz). 13C NMR (CDCl
15.04, 119.02, 126.74, 128.58, 128.76, 135.70, 138.01. IR
3
): δ 37.91, 40.37,
-
1
+
(CHCl
3
): 3686, 3019, 2930, 2230 cm . MS m/z: 298 (M , 2);
J O982266I