Ruthenium-catalysed linear-selective allylic alkylation of allyl acetates

Article abstract of DOI:10.1039/b614015g

The regioselectivity in the ruthenium-catalysed allylic alkylation of mono substituted allyl acetates with the malonate anion was highly controlled by Ru₃(CO)₁₂ with 2-(diphenylphosphino)benzoic acid, and the linear-type alkylated product was obtained. The Royal Society of Chemistry.

Full text of DOI:10.1039/b614015g

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Ruthenium-catalysed linear-selective allylic alkylation of allyl
acetates{
Motoi Kawatsura,*a Fumio Ata,a Shohei Wada,a Shuichi Hayase,a Hidemitsu Unob and Toshiyuki Itoh*a
Received (in Cambridge, UK) 26th September 2006, Accepted 17th October 2006
First published as an Advance Article on the web 27th October 2006
DOI: 10.1039/b614015g
We discovered that Ru3(CO)12 with an appropriate ligand
system accomplished the highly regioselective allylic alkylation of
unsymmetrical allyl substrates with the malonate anion. Typically,
the reaction was carried out as follows: for the ruthenium catalyst,
which was prepared in situ by mixing 3.3 mol% Ru3(CO)12 with a
ligand, c-substituted allyl acetates 1 and a-substituted allyl acetates
2 were allowed to react with the dimethyl methylmalonate anion in
THF at 0 to 60 uC for 12 h (Scheme 1). When PPh3 or dppe was
used as a ligand for the reaction of the c-substituted allyl acetate 1a
with the malonate anion, the reaction proceeded smoothly, but the
regioselectivity was low, and the branched isomer 3a was obtained
as the major product. Fortunately, we discovered that the reaction
using 2-(diphenylphosphino)benzoic acid (L1)10 as a ligand
exhibited an unusual regioselectivity. The alkylation of 1a with
the malonate anion in the presence of the ruthenium catalyst
coordinated with L1 took place with a 99% linear selectivity in a
97% isolated yield (Table 1, entry 3). The reaction with 1 mol% of
Ru3(CO)12 also proceeded with a perfect linear type of regio-
selectivity, though it required a longer reaction time to complete
(entry 4). The c-substituted acetates 1b–f exhibited a similar
reactivity with an excellent high linear selectivity (entries 5–9). The
linear selectivity which we observed here was higher than that of a
palladium-catalysed reaction.2e,11 We next investigated the reaction
with a-substituted allyl acetates 2a–f, which are regioisomers of
1a–f. Most of the a-substituted allyl acetates 2 also exhibited the
same linear selectivity as obtained for the reaction of the
c-substituted acetates 1, while a slight reduction in reactivity was
recorded (entries 10–16). The fact that the same regioselectivities
are observed for the reactions of 1 and 2 suggests that the reaction
proceeds through a shared p-allylruthenium intermediate as is
usual in transition metal-catalysed allylic substitutions.
The regioselectivity in the ruthenium-catalysed allylic alkylation
of mono substituted allyl acetates with the malonate anion was
highly controlled by Ru3(CO)12 with 2-(diphenylphosphino)-
benzoic acid, and the linear-type alkylated product was
obtained.
Transition metal-catalysed allylic substitution is a useful process in
organic synthesis, and many excellent results have been reported
for the reaction of symmetrically disubstituted substrates.1 On the
other hand, the regioselective allylic substitution of unsymmetrical
allyl substrates is a difficult problem and very challenging topic in
this field of chemistry. In particular, it is known that the control of
mono-substituted allyl substrates that forms both linear and
branched isomers is very difficult and the regioselectivity depends
on the type of transition metal–catalyst which was employed in the
reaction. Palladium-catalysed reactions tend to cause nucleophilic
substitution at the sterically less hindered allylic terminus and
produces linear-type products,2,3 while other metals such as
tungsten,4 molybdenum,5 rhodium,6 iridium7 and ruthenium8,9
preferentially yield branched-type products, by attack at the more
hindered allylic terminus. On the other hand, common ruthenium
catalysts exhibited a slightly low reactivity for the allylic alkylation
of unsymmetrical monosubstituted allyl substrates when compared
to palladium. The first regioselective ruthenium-catalysed reaction
with a carbon nucleophile was reported by Mitsudo; the
[Ru(cod)(cot)] catalysed highly regioselective reactions with several
carbon nucleophiles. However, no regioselectivity was obtained for
the reaction of the malonate anion.8e Recently, Trost reported that
a branched type of alkylation with the malonate anion proceeded
with a high selectivity using [Cp*Ru(NCCH3)3]PF6.8b Bruneau
also reported some ruthenium catalysed reactions that exhibited a
branched-type selectivity,9a–e and Pregosin also reported examples
of a number of p-allylruthenium intermediates that generated
branched products.9f–i However, there are still no examples of a
ruthenium-catalysed linear-type allylic alkylation of monosubsti-
tuted allyl substrates by the malonate anion. We here report the
first example of a ruthenium-catalysed perfect linear-type of allylic
alkylation with malonate anion.
aDepartment of Materials Science, Faculty of Engineering, Tottori
University, Koyama, Tottori, 680-8552, Japan.
E-mail: kawatsur@chem.tottori-u.ac.jp; titoh@chem.tottori-uac.jp;
Fax: +81 857 31 5179; Tel: +81 857 31 5179
bIntegrated Center for Science, Ehime University, Matsuyama Ehime,
790-8577, Japan
{ Electronic supplementary information (ESI) available: Experimental
details, NMR spectra of all new compounds and X-ray data for 5. See
DOI: 10.1039/b614015g
Scheme 1 Ruthenium-catalysed allylic alkylation of 1a–f and 2a–f.
This journal is ß The Royal Society of Chemistry 2007
298 | Chem. Commun., 2007, 298–300

View Article Online
Table 1 Ru3(CO)12 catalysed allylic alkylation of 1a–f and 2a–fa
Entry
Acetate
L
Yieldb (%) of 3 and 4
Ratioc 3 : 4
1
2
3
1a
1a
1a
1a
1b
1c
1d
1e
1f
2a
2a
2b
2c
2d
2e
2f
PPh3
dppe
L1
L1
L1
L1
L1
L1
L1
L1
L1
L1
L1
L1
L1
L1
70
90
97
90
98
98
96
98
87
96
86
86
98
96
81
79
77 : 23
79 : 21
1 : 99
1 : 99
3 : 97
1 : 99
1 : 99
1 : 99
1 : 99
1 : 99
1 : 99
3 : 97
1 : 99
1 : 99
1 : 99
1 : 99
4d,e
5
6
7
8
9
10
11d,e
12
13
14
15
16
a
All reactions were carried out in THF at 0 to 60 uC for 12 h under
nitrogen unless otherwise noted: THF (1.0 mL), allylic acetate
(1.0 mmol), MeCH(CO2Me)2 (1.5 mmol), base (1.4 mmol),
Fig. 1 Crystal structure of 5 with thermal ellipsoids at the 30%
b
Ru3(CO)12 (0.03 mmol), and ligand (0.1 mmol). Isolated yield by
silica gel column chromatography. The ratio was determined by
probability level. All hydrogen atoms and ethyl acetate were omitted for
˚
clarity. Selected bond lengths (A): Ru(1)–C(1), 2.224(5); Ru(1)–C(2),
c
500 MHz 1H NMR analysis of crude materials.
Ru3(CO)12 and 3 mol% of L1 were used. 24 h.
1 mol% of
d
e
2.239(5); Ru(1)–C(3), 2.298(5); Ru(1)–C(29), 1.853(6), Ru(1)–C(30),
1.914(6); Ru(1)–P(1), 2.3345(14); Ru(1)–O(1), 2.121(4).
p-allylruthenium(II) complex 5 with the lithium enolate of dimethyl
methylmalonate in THF gave a linear-type product 4a in 100%
conversion with a 99% regioselectivity, which is in good agreement
with our catalytic reactions. Furthermore, we confirmed that the
same p-allyl complex 5 was also formed from the allyl acetate 2a.
The 1H NMR spectra showed that the major ruthenium complex 5
slowly converts to another new p-allyl complex 59. We isolated the
minor p-allylruthenium species 59 as a yellow powder, which was
then subjected to a stoichiometric reaction with the malonate
anion; the reactivity of 59 was, however, very low and the reaction
gave a mixture of branched- and linear-type isomers in a 79 : 21
ratio with an 80% conversion. From these results, we concluded
that the p-allyl complex 5 might be the key intermediate in the
linear selective allylic alkylation reaction, and nucleophilic attack
selectively occurred at the sterically less hindered p-allyl terminus.
The regioselectivity might be due to the difference in oxidation
state [Ru(0)/Ru(II) or Ru(II)/Ru(IV)] and the absence of the Cp*
ligand.8e These mechanistic studies will be performed in the
future.{
Scheme 2 Reaction pathway in the ruthenium-catalysed allylic alkyla-
tion of 1a–f and 2a–f.
Following these reaction results, we attempted to prepare the
p-allylruthenium complex, which might be a key intermediate of
our highly linear selective allylic alkylation reaction (Scheme 2).
Mixing Ru3(CO)12, the ligand L1 and allyl acetate 1a at 60 uC gave
no new specific signals in the 31P NMR spectrum. However, when
the lithium salt of dimethyl methylmalonate (1 equivalent to Ru
metal) was added to the reaction mixture, two new signals (major
isomer 5: d 28.2 ppm. minor isomer 59: d 34.6 ppm) appeared in
1
the 31P NMR spectrum. The H NMR spectra of this complex
clearly indicated the formation of p-allylruthenium complexes. The
subsequent workup and recrystallization afforded orange crystals
of the major p-allylruthenium complex [Ru(1-phenyl-
allyl)(L1)(CO)2] (5) in a 71% isolated yield, based on the Ru
metal. The neutral ruthenium(II) complex 5 was stable under
atmospheric conditions for the solid state and we succeeded in
solving the structure by an X-ray crystallographic analysis (Fig. 1).
Notes and references
{ Preparation of p-allylruthenium complex 5 [Ru(1-phenyl-allyl)(L1)(CO)2]
and 59. Ru3(CO)12 (1.0 g, 1.56 mmol), 2-(diphenylphosphino)benzoic acid
(L1) (1.4 g, 4.69 mmol), 1-phenyl-2-propenyl acetate (1a) (3.3 g, 18.7 mmol)
and dimethyl methylmalonate (2.1 g, 14.0 mmol) were suspended in 15 mL
of THF in a screw-capped vial. The mixture was stirred at 0 uC for 5 min,
then LiHMDS (14.0 mmol. 14 mL of 1.0 M in THF) was slowly added at
same temperature. The reaction mixture was then stirred at 60 uC for 2 h.
The mixture was absorbed onto silica gel and chromatographed with ethyl
acetate/hexane (50 : 50 to 100 : 0), then evaporated. The gummy residue
was dissolved in a minimum amount of ethyl acetate and recrystallized by
slow diffusion of pentane into the concentrated ethyl acetate solution at
room temperature, yielding orange prismatic crystals of 5; yield 660 mg
(71% as ethyl acetate co-crystals). A suitable crystal was selected for the
X-ray study. On the other hand, the mother liquor was concentrated and
gave ruthenium complex 59 as a yellow powder; ca. 100 mg. The 1H NMR
and 31P NMR revealed that the ruthenium complex 5 (28.2 ppm in 31P
NMR) slowly converts to ruthenium complex 59 (34.6 ppm in 31P NMR).
˚
The bond length of Ru–C3 (2.298(5) A) is slightly longer than that
˚
of Ru–C1 (2.224(5) A), and this relative difference in the bond
lengths between Ru and the two p-allyl termini, C1 and C3, is
similar to the reported results9c,d,f–i for the key intermediates of the
ruthenium-catalysed branch selective allylic alkylations by
Bruneau and Pregosin. Although branched-type alkylations were
reported for Bruneau’s and Pregosin’s cationic ruthenium (IV)
complexes, the stoichiometric reactions of the neutral
This journal is ß The Royal Society of Chemistry 2007
Chem. Commun., 2007, 298–300 | 299

View Article Online
Ruthenium complex 5: 1H NMR (500 MHz, CDCl3) d 2.58 (dd, J = 4.55,
12.90 Hz, 1H), 3.12 (d, J = 12.90 Hz, 1H), 4.42 (dd, J = 4.55, 7.80 Hz, 1H),
5.94 (dt, J = 7.80, 12.90 Hz, 1H), 6.32–6.37 (m, 3H), 6.50–6.54 (m, 2H),
7.07–7.13 (m, 5H), 7.32–7.40 (m, 4H), 7.44–7.48 (m, 3H), 7.65–7.68 (m,
1H), 8.38 (ddd, J = 1.35, 4.6, 7.8 Hz, 1H). selected 13C NMR (125 MHz,
CDCl3) d 14.16, 21.00, 61.64 (d, J = 17.30), 80.53 (d, J = 4.80), 102.96,
139.51, 140.37 (d, J = 11.54), 169.28 (d, J = 7.69), 171.09, 195.36 (d, J =
4.81), 197.76 (d, J = 11.54). 31P NMR (202 MHz, CDCl3) d 28.2. mp 144–
147 uC (decomp).
S. Hildbrand and K. Dogra, J. Am. Chem. Soc., 1999, 121,
10416–10417; (c) F. Glorious, M. Neuburger and A. Pfatltz,
Org. Lett., 1999, 1, 141–144.
6 (a) B. L. Ashfeld, K. A. Miller and S. F. Martin, Org. Lett., 2004, 6,
1321–1324; (b) T. Hayashi, A. Okada, T. Suzuka and M. Kawatsura,
Org. Lett., 2003, 5, 1713–1715; (c) P. A. Evans and D. K. Leahy, J. Am.
Chem. Soc., 2002, 124, 7882–2883; (d) J. Tsuji, I. Minami and I. Shimizu,
Tetrahedron Lett., 1984, 25, 5157–5160; (e) R. Takeuchi and
N. Kitamura, New J. Chem., 1998, 22, 659–660.
7 (a) T. Ohmura and J. F. Hartwig, J. Am. Chem. Soc., 2002, 124,
15164–15165; (b) R. Takeuchi, N. Ue, K. Tanabe, K. Yamashita and
N. Shiga, J. Am. Chem. Soc., 2001, 123, 9525–9534; (c) B. Bartels and
G. Helmchen, Chem. Commun., 1999, 741–742.
8 (a) E. C. Burger and J. A. Tunge, Chem. Commun., 2005, 2835–2837; (b)
B. M. Trost, P. L. Fraisse and Z. T. Ball, Angew. Chem., Int. Ed., 2002,
41, 1059–1061; (c) T. Kondo, H. Ono, N. Satake, T. Mitsudo and
Y. Watanabe, Organometallics, 1995, 14, 1945–1953; (d) S.-W. Zhang,
T. Mitsudo, T. Kondo and Y. Watanabe, J. Organomet. Chem., 1995,
485, 55–62; (e) S.-W. Zhang, T. Mitsudo, T. Kondo and Y. Watanabe,
J. Organomet. Chem., 1993, 450, 197–207; (f) I. Minami, I. Shimizu and
J. Tsuji, J. Organomet. Chem., 1985, 296, 269–280; (g) C. Bruneau,
J.-L. Renaud and B. Demerseman, Chem.–Eur. J., 2006, 12, 5178–5187.
Ruthenium complex 59: selected 1H NMR (500 MHz, CDCl3) d 1.74 (d,
J = 12.58 Hz, 1H), 3.60 (d, J = 7.80 Hz, 1H), 4.48 (dd, J = 12.58 Hz, JHP
=
5.50 Hz, 1H), 5.88 (dt, J = 7.80, 12.58 Hz, 1H). selected 13C NMR
(125 MHz, CDCl3) d 14.20, 21.06, 57.07, 60.41, 89.12, 89.32, 104.76, 139.65,
139.68, 140.36. 31P NMR (202 MHz, CDCl3) d 34.6. mp 174–176 uC
(decomp).
Crystal data for 5?EtOAc: C34H31O6RuP, MW = 667.66, T = 173 K,
Monoclinic, space group P21/c, a = 12.846(2), b = 16.061(3), c =
14.566(4) A, U = 3005.1(11) A3, Z = 4, m(Mo-Ka) = 0.711 mm21, 25 776
reflections measured, 6696 unique [Rint = 0.125], Final R1 [I . 2sI] = 0.075,
wR2 = 0.173 (all data). CCDC 622374. For crystallographic data in CIF or
other electronic format see DOI: 10.1039/b614015g
¨
9 (a) N. Gu¨rbu¨z, I. Ozdemir, B. Cetinkaya, J.-L. Renaud, B. Demerseman
1 (a) J. Tsuji, Palladium Reagents and Catalysts: New Perspectives for the
21st Century, John Wiley & Sons, Chichester, 2004; (b) B. M. Trost and
M. L. Crawley, Chem. Rev., 2003, 103, 2921–2943; (c) J. Tsuji,
Transition Metal Reagents and Catalysts: Innovations in Organic
Synthesis,Wiley & Sons, Chichester, 2000.
2 (a) B. M. Trost, Tetrahedron, 1977, 33, 2615–2649; (b) G. Consiglio and
M. Waymouth, Chem. Rev., 1989, 89, 257–276; (c) G. C. Frost,
J. Holwarth and J. M. J. Williams, Tetrahedron: Asymmetry, 1992, 3,
1089–1122; (d) M. P. T. Sjo¨gren, S. Hansson, B. Akermark and
A. Vitagliano, Organometallics, 1994, 13, 1963–1971; (e) M. Kawatsura,
Y. Uozumi and T. Hayashi, Chem. Commun., 1998, 217–218.
3 (a) J. W. Faller and N. Sarantopoulos, Organometallics, 2004, 23,
2179–2185; (b) R. Pre´toˆt and A. Pfaltz, Angew. Chem., Int. Ed., 1998, 37,
323–325; (c) T. Hayashi, M. Kawatsura and Y. Uozumi, Chem.
Commun., 1997, 561–562.
and C. Bruneau, Tetrahedron Lett., 2006, 47, 535–538; (b) M. D. Mbaye,
B. Demerseman, J.-L. Renaud and C. Bruneau, J. Organomet. Chem.,
2005, 690, 2149–2158; (c) M. D. Mbaye, B. Demerseman, J.-L. Renaud,
L. Toupet and C. Bruneau, Adv. Synth. Catal., 2004, 346, 835–841; (d)
M. D. Mbaye, B. Demerseman, J.-L. Renaud, L. Toupet and
C. Bruneau, Angew. Chem., Int. Ed., 2003, 42, 5066–5068; (e) J.-L.
Renaud, C. Bruneau and B. Demerseman, Synlett, 2003, 408–410; (f)
R. Hermatschweiler, I. Ferna´ndez, P. S. Pregosin and F. Breher,
Organometallics, 2006, 25, 1440–1441; (g) I. Ferna´ndez,
R. Hermatschweiler, P. S. Pregosin, A. Albinati and S. Rizzato,
Organometallics, 2006, 25, 323–330; (h) R. Hermatschweiler,
I. Ferna´ndez, F. Breher, P. S. Pregosin, L. F. Veiros and
M. Calhorda, Angew. Chem., Int. Ed., 2005, 44, 4397–4400; (i)
R. Hermatschweiler, I. Ferna´ndez, P. S. Pregosin, E. J. Watson,
A. Albinati, S. Rizzato, L. F. Veiros and M. J. Calhorda,
Organometallics, 2005, 24, 1809–1812.
4 (a) B. M. Trost and M.-H. Hung, J. Am. Chem. Soc., 1983, 105,
7757–7759; (b) J. Lehmann and G. C. Lloyd-Jones, Tetrahedron, 1995,
˚
51, 8863–8874; (c) F. Ha˚kan and B. Akermark, Organometallics, 1995,
14, 561–563; (d) G. C. Lloyd-Jones and A. Pfaltz, Angew. Chem., Int.
Ed. Engl., 1995, 34, 462–464.
10 (a) D. Ikeda, M. Kawatsura and J. Uenishi, Tetrahedron Lett., 2005, 46,
6663–6666; (b) J. Uenishi, M. Kawatsura, D. Ikeda and N. Muraoka,
Eur. J. Org. Chem., 2003, 3909–3912.
5 (a) B. M. Trost, K. Dogra, I. Hachiya, T. Emura, D. L. Hughes,
S. W. Krska, R. A. Reamer, M. Palucki, N. Yasuda and P. J. Reider,
Angew. Chem., Int. Ed., 2002, 41, 1929–1932; (b) B. M. Trost,
11 We confirmed the regioselectivities in the reaction of 1a by
[PdCl(p-allyl)]2 with PPh3, dppe and L1 instead of Ru3(CO)12/L1 were
80%, 92% and 86%, respectively.
300 | Chem. Commun., 2007, 298–300
This journal is ß The Royal Society of Chemistry 2007