Catalytic tail-to-tail selective dimerization of methyl methacrylate promoted by a ruthenium(0) complex

Article abstract of DOI:10.1021/om100646q

Ru(η⁶-naphthalene)(η⁴-1,5-COD) (1) (10 mol %) catalyzes the tail-to-tail dimerization of methyl methacrylate (MMA) in MeCN in 74% yield at 70 °C for 4 h. Without use of solvent complex 1 (1 mol %) rapidly catalyzes the tail-to-tail dimerization and trimerization of MMA in 59% and 23% yields, respectively, at 70 °C for 5 min.

Full text of DOI:10.1021/om100646q

3690 Organometallics 2010, 29, 3690–3693
DOI: 10.1021/om100646q
Catalytic Tail-to-Tail Selective Dimerization of Methyl Methacrylate
Promoted by a Ruthenium(0) Complex
Masafumi Hirano,* Yuki Hiroi, Nobuyuki Komine, and Sanshiro Komiya
Department of Applied Chemistry, Graduate School of Engineering, Tokyo University of Agriculture and
Technology, 2-24-26 Nakacho, Koganei, Tokyo 184-8588, Japan
Received July 5, 2010
Summary: Ru(η6-naphthalene)(η4-1,5-COD) (1) (10 mol %)
catalyzes the tail-to-tail dimerization of methyl methacrylate
(MMA) in MeCN in 74% yield at 70 °C for 4 h. Without use of
solvent complex 1 (1 mol %) rapidly catalyzes the tail-to-tail
dimerization and trimerization of MMA in 59% and 23%
yields, respectively, at 70 °C for 5 min.
conversion of MMA at 80 °C,7 by in situ formed Ni(II) to
give E/Z-2b (26%) at rt for 20 h,8 and by PdCl2(p-nonyl-
C6H4CN)2/LiBF4 to give a 6:94 mixture of 2a and Z/E-2b
with 50% conversion of MMA at 20-50 °C (eq 1).9
Quite recently, catalytic tail-to-tail dimerization of MMA
by N-heterocyclic carbene as an organic catalyst has been
reported.10
The tail-to-tail dimers of substituted olefins are potential
precursors to monomers for condensation polymerization,
and a variety of transition metal complexes have been docu-
mented to catalyze such tail-to-tail dimerization.1 Although
such examples involve dimerization of methyl acrylate,2
acrolein,3 and acrylonitrile,4 the catalytic tail-to-tail dimeri-
zation of methyl methacrylate (MMA) is generally very
difficult even though the product is potentially important.5
To the best of our knowledge, only a few examples have been
reported of the dimerization of MMA by transition-metal-
based catalysts, and the catalytic activities are generally
poor. For examples, these pioneering works report the dim-
erization of MMA promoted by [Pd(NCMe)4]2þ[PF6]- in
2
the presence of LiBF4 to give of E/Z-2b (59%) and 2c (4%) at
25 °C for 3 days,6 by [Pd(η3-allyl)(1,5-COD)]þBF4-/PR3
(COD = cyclooctadiene (C8H12)) to give E/Z-2b with 12%
We have documented the first solid evidence for an
oxidative coupling mechanism in the selective tail-to-tail
coupling reaction of methyl acrylate promoted by Ru(η6-
naphthalene)(η4-1,5-COD) (1).11 As an extension of this
study, we have found that complex 1 also catalyzes the tail-
to-tail dimerization of MMA, and interestingly, this process
occurs much faster than the corresponding reaction with
methyl acrylate. We disclose details of this work here and
discuss possible mechanisms.
The naphthalene complex of ruthenium(0), 1 (10 mol %),
catalyzed the tail-to-tail dimerization of MMA in the pre-
sence of MeCN (10 equiv to 1) at 70 °C for 4 h in 74% yield
(2a/Z-2b/E-2b = 3/5/92) with concomitant formation of tri-
mers in 13% yield (3x/Z-3a/E-3a = 54/38/8) (eq 2, entry 1 in
Table 1).12 The tail-to-tail dimers 2a, Z-2b, and E-2b and the
trimer 3a were characterized by GC-MS, 1H NMR, 1H-1H
COSY, 13C NMR, DEPT135, and 1H-13C HETCOR NMR
spectra, and E-2b and 3a were separated by silica gel column
chromatography and preparative GLC. A small amount of
*To whom correspondence should be addressed. Tel and fax: þ81 423
88 7044. E-mail: hrc@cc.tuat.ac.jp.
(1) Parshall, G. W.; Ittel, S. D. In Homogeneous Catalysis: The
Applications and Chemistry of Catalysis by Soluble Transition Metal
Complexes, 2nd ed.; Wiley: New York, 1992; pp 82-85.
(2) (a) DiRenzo, G. M.; White, P. S.; Brookhart, M. J. Am. Chem.
Soc. 1996, 118, 6225. (b) Brookhart, M.; Sabo-Etienne, S. J. Am. Chem.
Soc. 1991, 113, 2777. (c) Broookhart, M.; Hauptman, E. J. Am. Chem. Soc.
1992, 114, 4437. (d) Hauptman, E.; Sabo-Etienne, S.; White, P. S.; Brookhart,
M.; Garner, J. M.; Fagan, P. J.; Calabrese, J. C. J. Am. Chem. Soc. 1994, 116,
8038. (e) McKinney, R. J.; Colton, M. C. Organometallics 1986, 5, 1080.
(f) Strehblow, C.; Schupp, T.; Sustmann, R. J. Organomet. Chem. 1998, 561,
181. (g) Guibert, I.; Neibecker, D.; Tkatchenko, I. J. Chem. Soc., Chem.
Commun. 1989, 1850. (h) Picquet, M.; Stutzmann, S.; Tkachenko, I.;
Tommasi, I.; Zimmermann, J.; Wasserscheid, P. Green Chem. 2003, 5,
153. (i) Pertici, P.; Ballantini, V.; Salvadori, P.; Bennett, M. A. Organome-
tallics 1995, 14, 2565.
(3) Ohgomori, Y.; Ichikawa, S.; Sumitani, N. Organometallics 1994,
13, 3758.
(4) (a) Fukuoka, A.; Nagano, T.; Furuta, S.; Yoshizawa, M.; Hirano,
M.; Komiya, S. Bull. Chem. Soc. Jpn. 1998, 71, 1409. (b) Kashiwagi, K.;
Sugise, R.; Shimakawa, T.; Matuura, T.; Shirai, M.; Kakiuchi, F.; Murai, S.
Organometallics 1997, 16, 2233. (c) Tembe, G. L.; Ganeshpure, P. A.; Satish,
S. React. Kinet. Catal. Lett. 1998, 63, 151.
(5) Thermal dimerizations of MMA (ref 5a) and calcium methacry-
late followed by treatment of diazomethane were reported (ref 5b).
(a) Albisetti, C. J.; England, D. C.; Hogsed, M. J.; Joyce, J. J. Am. Chem.
Soc. 1956, 78, 472. (b) Naruchi, K.; Yamamoto, O.; Miura, M.; Nagakubo, K.
Nippon Kagakukaishi (J. Chem. Soc. Jpn, Chem. Ind. Chem.) 1976, 11,
1794.
(8) Wilke, G.; Sperling, K. Ger. Patent 3336691, 1985.
(9) (a) Oehme, G. J. Prakt. Chem. 1984, 5, 779. (b) Oehme, G.; Grassert,
I.; Mennenga, H.; Baudisch, H. J. Mol. Catal. 1986, 37, 53.
(10) Matsuoka, S.; Washio, A.; Ohta, Z.; Katada, A.; Ichioka, K.;
Takagi, K.; Suzuki, M. Abstracts of Papers, the 90th Annual Meeting of
the Chemical Society of Japan, Osaka, 2010; 3F703.
(11) Hirano, M.; Sakate, Y.; Komine, N.; Komiya, S.; Bennett, M. A.
Organometallcs 2009, 28, 4902.
(12) The structure of trimer 3x is not clear at present, but 3x is
estimated to be two regioisomers of trimer 3a on the basis of the GC-MS
analysis. The yield of 3x was calculated on the basis of the GLC analysis
with assumption of the same relative intensity as 3a.
(6) Nugent, W. A. U.S. Patent, 4 451 665, 1984.
(7) Grenouillet, P.; Neibecker, D.; Tkatchenko, I. Organometallics
1984, 3, 1130.
r
pubs.acs.org/Organometallics
Published on Web 08/18/2010
2010 American Chemical Society

Communication
Organometallics, Vol. 29, No. 17, 2010 3691
Table 1. Catalytic Tail-to-Tail Dimerization of MMA Promoted by 1a
conditions
dimers
2a/Z-2b/E-2b, %
trimers
3x/Z-3a/E-3a, %
entry
cat., %
temp, °C
time, h
solvent
yield, %
yield, %
1
2
3
4
5
6
7
8
10
1
1
1
1
1
1
1
1
1
1
1
70
140
0
4
4
97
24
2
2
2
8
2
MeCN
MeCN
neat
neat
neat
neat
neat
C6H6
neat
neat
74
33
40
47
55
66
67
61
3/5/92
3/9/88
13
1
60
54
40
24
20
15
20
trace
trace
23
54/38/8
0/64/36
2/58/40
7/33/60
9/39/56
13/25/62
20/10/70
13/13/74
10/45/55
15/35/50
2/13/85
3/22/75
2/11/87
3/7/90
30
50
70
90
70
70
70
70
70
2/7/91
2/23/75
9b
10c
11d
12e
65
2
2
6
trace
trace
62
neat
neat
2/10/88
12/23/65
a Yields are estimated on the basis of MMA. b 1,5-COD (5 mol %) added. c PPh3 (5 mol %) added. d P(OMe)3 (5 mol %) added. e NEt3 (100 mol %)
added. Typical conditions: 1 (0.100 mmol), MMA (10.0 mmol).
isobutyric acid (2%) was observed and compound 2c was not
observed in this catalysis.
Without use of MeCN, complex 1 (1 mol %) at 70 °C cata-
lyzed the reaction to give a mixture of the tail-to-tail dimer in
59% yield (tail-to-tail selectivity: 100%, 2a/Z-2b/E-2b=14/30/
56) in 5 min, with concomitant formation of the trimers in 23%
yield (3x/Z-3a/E-3a=4/63/33). After 1 h, the product ratios
changed to 2a/Z-2b/E-2b=2/15/83 (61% yield) and 3x/Z-3a/
E-3a = 11/37/62 (27% yield), and after 24 h to 2a/Z-2b/E-2b=
2/5/93 (56% yield) and 3x/Z-3a/E-3a = 3/7/90 (27% yield).
The transient formations of 2a, Z-2b, and Z-3a suggest that 2a
is converted into 2b and/or 3a and that Z-2b and Z-3a are
similarly converted into E-2b and E-3a, respectively.
This catalysis proceeded even at 0 °C, but the trimers
became the dominant product (entry 3 in Table 1). The
Figure 1. Molecular structure of Ru(κ4-cisoid-MMA)(η2-ci-
dimer/trimer ratio increased at higher temperature (entries
3-7), but the conversion decreased due to catalyst deactiva-
tion (entry 2). The change of the dimer/trimer ratio depend-
ing on the reaction temperature may be due to rapid
isomerization of the initial product 2a to 2b at high tempera-
ture, while 2a is converted into the trimers by the coupling
with MMA (vide infra). Addition of 1,5-COD did not cause
significant change in this catalysis (entry 9), but trivalent
phosphorus compounds, such as PPh3 or P(OMe)3, were
found to inhibit the reaction, probably by preventing co-
ordination of MMA to the ruthenium center or by displace-
ment of the 1,5-COD ligand (entries 10, 11).
In order to clarify the mechanism of the tail-to-tail dimeri-
zation of MMA promoted by 1, we performed a stoichio-
metric reaction of 1 with MMA. Treatment of 1 with 3 equiv
of MMA in C6D6 resulted in the tail-to-tail dimer in 74%
yield (2a/Z-2b/E-2b = 2/5/93) along with Ru(η6-C6D6)(η4-
1,5-COD) (4)13 (23%) and [Ru(η4-1,5-COD)]2(η6:η4-nap-
hthalene) (5)14 (3%) in 57% conversion of 1 without any
soid-MMA)(PMe3)2 (6). Hydrogen atoms in the PMe3 ligands
were omitted for clarity. Ellipsoids represent 50% probability.
Although two crystallographically independent molecules are
observed in a unit cell, one of them is depicted because they are
isostructural to each other.
indication of intermediates. Although the catalyst precursor
1 does not give a single product during this process, the
formation of 4 and 5 suggests formation of a zerovalent
ruthenium species such as “Ru(η4-1,5-COD)”.15 When 1 was
treated with 2 equiv of MMA in the presence of 2 equiv of
PMe3 at 50 °C for 1 day in benzene, the zerovalent complex
Ru(κ4-cisoid-MMA)(η2-cisoid-MMA)(PMe3)2 (6) was iso-
lated as pale yellow plates in 32% yield. It was characterized
by IR and NMR spectroscopy and, most unambiguously, by
X-ray structure analysis (Figure 1).
In complex 6, two MMA fragments are coordinated to the
ruthenium center, but the bonding modes are different—
κ4-cisoid and η2-cisoid—and the original 1,5-COD fragment
has been displaced by two PMe3 ligands. The bond distances
to the side-bonded CdO group of the κ4-MMA ligand
(13) Vitulli, G.; Pertici, P.; Salvadori, P. J. Chem. Soc., Dalton Trans.
1984, 2255.
(14) Bennett, M. A.; Neumann, H.; Thomas, M.; Wang, X.-Q.;
(15) Bennett, M. A.; Lu, Z.; Wang, X. Q.; Bown, M.; Hockless,
D. C. R. J. Am. Chem. Soc. 1998, 120, 10409.
Pertici, P.; Salvadori, P.; Vitulli, G. Organometallics 1991, 10, 3237.

3692 Organometallics, Vol. 29, No. 17, 2010
Hirano et al.
˚
˚
[Ru(1)-C(3) = 2.448(4) A and Ru(1)-O(1) = 2.333(2) A]
are significantly longer than those to the CdC group of this
Scheme 1
˚
ligand [Ru(1)-C(1) = 2.125(3) A and Ru(1)-C(2) = 2.169(3)
A]. This feature suggests that the κ -MMA unit in 6 is close
4
˚
to the η2-CdC extreme. On the other hand, in the η2-MMA
˚
unit in 6, the distances Ru(1)-C(6) [2.141(3) A] and Ru(1)-
C(7) [2.221(3) A] clearly show the η -coordination through
the CdC bond, while the Ru(1)-C(8) [3.186(3) A] and
Ru(1)-O(3) [3.699(3) A] distances clearly rule out CdO
2
˚
˚
˚
coordination. In the NMR spectrum of 6 in benzene-d6,
these two MMA fragments were observed separately at
room temperature, showing that they do not exchange
rapidly on the NMR time scale. Complex 6 does not bring
about dimerization (or trimerization) of MMA, when it is
heated either alone or in the presence of excess MMA at
70 °C. We believe that complex 6 is obtained as an isolable
species by displacement of the 1,5-COD ligand of an active
species with PMe3. It is worth noting that complexes of
general formula [Ru(L)(η2-1,5-COD)(κ4-MMA)] (L = PMe3,
PEt3, P(OMe)3) have also been isolated as yellow, crystalline
solids from the reaction of 1 with MMA in the presence of the
P-donor ligands.16
The isolation of complex 6 in the case of MMA and of the
ruthenacyclopentane in the case of methyl acrylate11 are
consistent with the catalytic cycle shown in Scheme 1 for
dimerization of MMA on Ru(0). In the first step, MMA
initiates displacement of the naphthalene ligand in 1 to give
Ru(η2-MMA)2(η4-1,5-COD) (A). Interestingly, the two
MMA molecules in complex 6 are coordinated through
different prochiral faces (si and re), whereas the observed
trans-configuration of the ruthenacyclopentane indicates
that the two methyl acrylate molecules coordinate through
the same prochiral faces.11 Although the face selectivity of
MMA in the active species is unknown, the oxidative cou-
pling reaction should proceed with retention of configura-
tion to give a ruthenacyclopentane B. In this cycle, we believe
the 1,5-COD ligand remains attached because (i) addition of
1,5-COD does not affect the catalysis (entry 9 in Table 1), (ii)
formation of the “Ru(1,5-COD)” fragment is observed in the
stoichiometric reaction, (iii) in situ generated naked Ru(0)
that the formation of 3a also supports the previously
suggested11 oxidative coupling mechanism for this catalysis,
for the following reasons: (i) the yield of 3a increased with
decrease of the exomethylene compound 2a as described
above; (ii) 3a has the trimer structure that formally formed
two C-C bonds at the methylene and the methyl carbons of a
MMA molecule, and (iii) 3a has a CdC bond at the 4-position.
These features can best be explained only by co-dimerization
between 2a and MMA, as illustrated in path 1 in Scheme 2,
where the tail-to-tail type oxidative coupling reaction be-
tween 2a and MMA gives ruthenacyclopentane E, from
which subsequent β-hydride elimination and reductive elim-
ination between the hydride and alkyl groups release the trimer
3a. The predominant formation of E over F may be caused by
the coordination of the methoxycarbonyl groups at the 2- and
5-positions, or the difficulty in achieving the C-C bond form-
ation in the oxidative coupling reaction prevents the formation
of ruthenacyclopentane F.
An alternative insertion mechanism, in which a MMA
molecule inserts into the M-C bond in the ruthenacyclopen-
tane B, giving the putative ruthenacycloheptane G or H, can
also be considered. Subsequent β-hydride elimination and
reductive elimination between the hydride and alkyl would
then give the trimers (path 2). However, it is not poss-
ible to account for the formation of compound 3a from either
G or H. Although we cannot rule out the formation of trimers
from intermediates G and H, the dominant process for the
formation of the trimer seems to be the oxidative coupling
reaction between 2a and MMA, which is also consistent with
the time course for the formation of 3a with decrease of 2a.18
Another potential mechanism may be the hydride-inser-
tion mechanism and the C-H bond activation mechanism.
species by reduction of RuCl3 3H2O with Zn2e does not give
3
dimers and trimers of MMA under these conditions at all,
and (iv) an analogue of B has been isolated in the case of
methyl acrylate. Then, successive β-hydride elimination and
the reductive elimination release the tail-to-tail dimer 2a or
2b. It is noteworthy that the dimerization of MMA occurs
under much milder conditions (even at 0 °C) than dimeriza-
tion of methyl acrylate (at 140 °C) in this system. We believe
this is because the R-methyl groups induce facile β-hydride
elimination, thus producing the dimer rapidly. Distortion of
the ruthenacyclopentane caused by the methyl and methox-
ycarbonyl groups at the 2- and 5-positions may also promote
β-hydride elimination from the methylene protons in the
ruthenacyclopentane.17
The dominant trimer in the mixture of trimers also ob-
served in this catalysis was characterized as a E/Z-mixture of
2,5,8-trimethoxycarbonylnon-4-ene (3a) (eq 2). We believe
(16) Lu, Z. Ph.D. Thesis, Australian National University, Canberra,
1998.
(17) One of the reviewers mentioned a possibility for the formation of
dimers by degradation of the trimer at high temperature because the
selectivity of the dimer was increased with increase of the temperature.
Because 3a is thermally stable under these conditions, we can rule out
this possibility (see SI Figure S8).

Communication
Scheme 2. Possible Pathways for the Formation of 3aa
Organometallics, Vol. 29, No. 17, 2010 3693
with solid evidence, in the present case unexpected pro-
tonation of the catalyst precursor, leading to the hydride-
insertion mechanism, is less likely because the catalysis in the
presence of a large excess of NEt3 does not give a significant
change (Table 1, entry 12). Moreover, Mitsudo and his co-
workers proposed the hydride-insertion mechanism for the
stoichiometric dimerization of dimethyl fumarate in Ru(η6-
1,3,5-COT)(η2-dimethyl fumarate)2 (COT = cyclooctatriene
(C8H10)) via intramolecular C-H bond oxidative addition.19
We believe this is also less likely because it is not possible to
account for the formation of 3a by this mechanism.
In summary, the present work provides a rare example of
the catalytic tail-to-tail dimerization of MMA promoted by a
Ru(0) precursor that is considered to be carried out by the
oxidative coupling mechanism, although the other potential
mechanisms cannot be completely excluded. While the di-
merization of methyl acrylate promoted by 1 requires high
temperature (ca. 140 °C), this process proceeds even at 0 °C.
The key step is expected to be the facile β-hydrogen elimina-
tion from the R-methyl group and/or the methylene groups
in the ruthenacyclopentane ring formed by the oxidative
coupling reaction. Further detailed mechanistic studies are
now in progress.
a The 1,5-COD ligand is omitted.
Acknowledgment. We are grateful to Prof. M. A. Bennett
(Australian National University) for discussions. We
thank Ms. Y. Sakate for preliminary work.
Brookhart and his co-workers reported a detailed mechan-
istic study for dimerization of methyl acrylate promoted by
the combination of [RhCp*(C2H4)2] with a proton.2d Alth-
ough they have proposed the hydride-insertion mechanism
Supporting Information Available: Text, tables, and figures
giving full experimental details involving the characterizations
of 2a, 2b, 3a, and 6, crystallographic analyses of 6, a CIF file
giving X-ray crystal data, and time-yield curves for this cata-
lysis at 0 and 70 °C. This material is available free of charge via
the Internet at http://pubs.acs.org.
(18) Characteristic time-yield curves, initial formation of 2a fol-
lowed by increase of 3a and 2b with decrease of 2a, was observed at 0 °C
(see SI Figure S7).
(19) Shiotsuki, M.; Suzuki, T.; Kondo, T.; Wada, K.; Mitsudo, T.
Organometallics 2000, 19, 5733.