Ruthenium/C5Me5/bisphosphine- or bisphosphite-based catalysts for normal-selective hydroformylation

Article abstract of DOI:10.1002/anie.201108396

A way into normality: By using [Cp*Ru] complexes with bisphosphite or bisphosphine ligands, selective hydroformylation of propene and 1-decene to homologated normal aldehydes was accomplished with the highest levels of activity and selectivity ever reported for ruthenium catalyst systems (see scheme). The reaction mechanism was investigated using stoichiometric amounts of [Cp*Ru(Xantphos)H] and [D₂]-1-decene. Copyright

Full text of DOI:10.1002/anie.201108396

Angewandte
Chemie
DOI: 10.1002/anie.201108396
Synthetic Methods
Ruthenium/C5Me5/Bisphosphine- or Bisphosphite-Based Catalysts for
normal-Selective Hydroformylation**
Kohei Takahashi, Makoto Yamashita,* Yoshiyuki Tanaka, and Kyoko Nozaki*
Hydroformylation of alkenes is the most widely industrially
applied homogeneous transition-metal-catalyzed reaction.[1]
Since the discovery by Roelen that [HCo(CO)4] can promote
this transformation,[2] cobalt complexes have been used
extensively as hydroformylation catalysts. However, more
recently developed rhodium/phosphine complexes generally
afford higher selectivity compared to cobalt catalysts for the
production of normal aldehydes (n-aldehydes), which are
utilized as intermediates for the synthesis of plasticizers,
solvents, and detergents. Additionally, rhodium complexes
typically produce negligible quantities of alkanes from
competing alkene hydrogenation unlike cobalt-based hydro-
formylation catalysts. Thus, rhodium catalysts have come to
play a major role in industry.
Intensive studies have been devoted both in industry and
in academia to the development of bisphosphine or bisphos-
phite ligands that promote highly normal-selective (n-selec-
tive) rhodium-catalyzed hydroformylation of 1-alkenes.[3]
However, recent global increases in demand for rhodium,
particularly in the automotive industry, is increasing the price
of this already expensive precious metal.[4] In contrast, more
cost-effective cobalt-based systems suffer from concomitant
formation of alkane by-products. Considering these factors,
the development of hydroformylation catalysts using other
transition metals is desirable. In contrast to the extensive
studies on cobalt and rhodium systems, other transition
metals such as palladium,[5] iridium[6], and ruthenium[7] have
received less attention as potential hydroformylation cata-
lysts. We have targeted ruthenium, which is less expensive
compared to rhodium. Only two examples of ruthenium-
catalyzed hydroformylation with high n-selectivity (normal/
iso (n/i) > 30) have been reported.[7d,h]
Our design for ruthenium hydroformylation catalysts is
illustrated in Figure 1. In conventional rhodium catalysts for
hydroformylation, the hydridorhodium(I) A is a key species
Figure 1. Comparison of conventional rhodium (A) and ruthenium (B)
hydroformylation catalysts with the ruthenium catalyst C used in this
work.
to which alkenes insert to generate the corresponding
alkylrhodium species. In contrast, the dihydrido rutheniu-
m(II) B is known as a catalytically active species for hydro-
genation of alkenes, which is a problematic side reaction in
hydroformylation. By comparing the species A and B, we
envisioned replacing one hydride in B with a cyclopentadienyl
anion. Through this strategic change, the [CpRuII] complex C
was expected to be inactive as an alkene hydrogenation
catalyst because the ruthenium center lacks one of the
requisite hydride ligands present in the active hydrogenation
intermediate B. Herein, we report a new catalyst system that
consists of a [Cp*RuII] (Cp* = C5Me5) complex ligated by
chelating bisphosphines. These complexes effect the hydro-
formylation of 1-alkenes with the highest level of activity and
n to i selectivity to date for any ruthenium-based catalyst,[7d,h]
even though the activity remains well below that of the best
rhodium-based catalysts. Additionally, these ruthenium cata-
lysts successfully suppress the rate of competing alkene
hydrogenation.
First, hydroformylation of propene was examined using
a combination of [{Cp*Ru(acac)}2] and the bidentate phos-
phorus ligands A4N3,[3i] Xantphos,[3a] and bisbi[3e] (Figure 2)
which were previously developed to promote highly active
and n-selective rhodium hydroformylation catalysts. The
results are summarized in Table 1. The reaction in toluene
at 1608C using a combination of [{Cp*Ru(acac)}2] and A4N3
as the catalyst proceeded with a turnover frequency (TOF) of
[*] K. Takahashi, Dr. M. Yamashita,[+] Prof. Dr. K. Nozaki
Department of Chemistry and Biotechnology
Graduate School of Engineering, The University of Tokyo
7-3-1 Hongo, Bunkyo-ku 113-8656, Tokyo (Japan)
E-mail: nozaki@chembio.t.u-tokyo.ac.jp
Homepage: http://park.itc.u-tokyo.ac.jp/nozakilab/
Dr. Y. Tanaka
Mitsubishi Chemical Corporation, Mizushima R&D Center
3-10 Ushiodori, Kurashiki, 712-8054, Okayama (Japan)
[+] Current address: Department of Applied Chemistry, Faculty of
Science and Engineering, Chuo University, Tokyo (Japan)
[**] K.N. is grateful to Prof. Charles P. Casey (Univ. of Wisconsin) for
helpful discussions. K.T. acknowledges a JSPS research fellowship.
This work was supported by Funding Program for Next Generation
World-Leading Researchers, Green Innovation, JSPS.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201108396.
Figure 2. Bidentate bisphosphine and bisphosphite ligands used in
this study.
Angew. Chem. Int. Ed. 2012, 51, 4383 –4387
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4383

.
Angewandte
Communications
Table 1: Hydroformylation of propene by ruthenium complexes.[a]
product compared to the reaction using [Ru3(CO)12]
and no additional ligand (Table 1, entry 4). Low
n/i selectivity is also observed, which suggests that
the active species in entry 1 is different from that in
entry 4. The use of Xantphos and bisbi (Table 1,
entries 5 and 6) resulted in TOFs of 1.7 and 0.6 hꢀ1
,
Entry Catalyst
T
t
Aldehydes Alcohols
[8C] [h] TOF [hꢀ1
(n/i)
]
TOF [hꢀ1
]
(n/i)
respectively, and with n/i selectivity comparable to
that observed for A4N3. The normal selectivity can
be improved (n/i = 43) by lowering the reaction
temperature to 1208C (Table 1, entry 7). The n/i
value of 43 in entry 7 is comparable to that obtained
at 708C using [NEt4][HRu3(CO)11],[7b] which is the
ruthenium catalyst reported to have the best n-
selectivity (Table 1, entry 8).
1[b]
2[c]
3[d]
4[e]
5[f]
6[f]
7
[{Cp*Ru(acac)}2]/A4N3
[{Cp*Ru(acac)}2]
[Ru3(CO)12]
[Ru3(CO)12]/A4N3
[{Cp*Ru(acac)}2]/Xantphos
[{Cp*Ru(acac)}2]/bisbi
[{Cp*Ru(acac)}2]/A4N3
[NEt4][HRu3(CO)11]
[{(indenyl)Ru(CO)2}2]/A4N3
[{(1,2,3-trimethylindenyl)Ru(CO)2}2]/
A4N3
160 24 1.3 (17)
0.29 (14)
160 24 0.92 (1.8) 0.07 (2.5)
160 24 8.3 (1.7) 8.5 (1.9)
160 24 0.03 (8.0) 0.85 (8)
160 24 1.7 (13)
160 24 0.6 (14)
120 24 0.48 (43) 0.06 (> 100)
70 66 0.11 (45) n.d.
120 24 2.2 (32)
120 24 4.3 (41)
0.05 (> 100)
0.04 (> 100)
8[g]
9
Substitution of Cp* with an indenyl ligand leads
to improved activity and selectivity. In entries 9 and
10 in Table 1, Ru–Ru bonded dimers [{(indenyl)-
0.14 (12)
0.14 (>100)
10
Ru(CO)2}2][8]
and
[{(1,2,3-trimethyindenyl)-
Ru(CO)2}2],[9] respectively, were employed. The
TOF for formation of the n-aldehyde reached 4.3
and an aldehyde n/i of 41 by using a combination of
[{(1,2,3-trimethylindenyl)Ru(CO)2}2] and A4N3.
Next, we investigated the hydroformylation of 1-
decene. The results are summarized in Table 2. High
n-selectivity (n/i = 79) was observed using [{Cp*Ru-
(acac)}2]/A4N3 as the catalyst system (Table 2,
[a] The molar quantity of the ruthenium complexes is based on the total mol of Ru
atoms. TOF=(mol of product)/[(mol of Ru)ꢀ(reaction time)]. The amounts of
charged H2 and CO were so high that the changes in their partial pressure during the
reaction time were negligible. No significant amount of by-product was detected by
GC unless otherwise mentioned. [b] Unidentified high-boiling products were
observed. The TOF of their formation was roughly estimated to be 0.08 hꢀ1
[c] Unidentified high-boiling products similar to those for entry 1 were observed by
GC. The TOF of their formation was was roughly estimated to be 0.07 hꢀ1
[d] Unidentified high-boiling products different from those of entry 1 were observed
by GC. The TOF of their formation was roughly estimated to be 3 hꢀ1
.
.
.
entry 1).
[{Cp*Ru(acac)}2]/Xantphos
(Table 2,
[e] Unidentified high-boiling products different from those of entry 1 were observed
by GC. The TOF of their formation was roughly estimated to be 0.9 hꢀ1. [f] 1,4-
dioxane was used as a solvent. [g] The reaction conditions were the same as the best
conditions reported in literature (Ref. [7d]). Ru complex (102 mmol), propene
(0.5 MPa), H2 (0.17 MPa), CO (0.34 MPa) in dimethoxyethane (2 mL), 708C, 66 h.
acac=acetyl acetonate, n.d.=not determined.
entry 2) also exhibited a high level of selectivity (n/
i = 27). In these experiments, however, reproduci-
bility was problematic, thus resulting in fluctuating
amounts of internal alkene products. In fact, the use
of [{(1,2,3-trimethylindenyl)Ru(CO)2}2]/A4N3 as the
catalyst resulted in significant isomerization to
1.3 hꢀ1 and n/i of 17 (Table 1,
Table 2: Hydroformylation of 1-decene by ruthenium complexes.[a]
entry 1). Reduction of some of the
desired aldehyde to the alcohol was
observed (18%). Without the addi-
tion of the A4N3 ligand, the reac-
tion proceeded with poor n/i selec-
tivity and a slightly lower TOF
(Table 1, entry 2). When the ruthe-
nium source was changed to
[Ru3(CO)12], which is one of the
most active ruthenium catalysts
reported, a higher TOF of 8.3 hꢀ1
but lower n/i of 1.7 were observed
(Table 1, entry 3). Alcohol forma-
tion was also a problem when
[Ru3(CO)12] was used without the
added ligand; equimolar amounts
of alcohol were formed together
with the desired aldehyde. Signifi-
cant amounts of unidentified high-
boiling by-products also formed
under these reaction conditions. A
combination of [Ru3(CO)12] and
A4N3 as the catalyst formed
Entry Catalyst
T
t
Recovery of
[8C] [h] starting
material [%]
Aldehydes Alkane Isomerized
[%] (n/i)
[%]
alkenes
[%]
1
[{Cp*Ru(acac)}2]/A4N3
100 18 13
66 (79)
56 (27)
13 (5.5)
29 (31)
60 (28)
20 (14)
<0.1 (ꢀ)
58[e] (28)
1.5
2.5
10
1.2
3.2
19
19
81
8.5
8.4
56
2
[{Cp*Ru(acac)}2]/Xantphos 160 21 15
3
4
5
[{Cp*Ru(acac)}2]
160 12
160 24 60
160 48 23
0
1/Xantphos[c]
1/Xantphos[c]
6
[Ru3(CO)12]/Xantphos
160 18
9
10
7[c]
8[d]
1/Xantphos[b]
160 24 86
1.4
4.6
1/Xantphos[b]
160 48 23[e]
5.6[e]
15[e]
[a] The molar quantity of the ruthenium complexes is based on the total mol of Ru atoms. The amounts
of charged H2 and CO were such that the changes in their partial pressure during the reaction time were
negligible. No significant amount of by-product was detected by GC unless otherwise mentioned. In the
blank experiment, loss of 1-decene because of its volatility was confirmed. The recovery of 1-decene was
94% in the absence of catalyst under the same reaction conditions.[10] [b] Xantphos 2.5 mol%. [c] (Z)-2-
decene (purity 95%, containing decane 1.6%, (E)-2-decene 2.5%, and other C10 alkenes 0.9%) was
1
used as substrate. [d] 1-eicosene was used as a substrate. [e] Yield determined by H NMR analysis
a lower fraction of reduced alcohol using trimethoxybenzene as an internal standard.
4384
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 4383 –4387

Angewandte
Chemie
internal alkenes.[10] The rapid alkene isomerization that
occurred in the absence of phosphorus ligands (Table 2,
entry 3) suggests that a trace amount of a non-phosphine-
coordinated ruthenium species is present and acts as an
isomerization catalyst. To avoid this problem, we examined an
ꢀ
isolated phosphine-coordinated Ru H complex. We success-
fully isolated the [Cp*Ru(Xantphos)H] complex (1) as shown
in Scheme 1. The reaction of dimeric of [{Cp*Ru(Cl)(m-Cl)}2]
Scheme 2. A possible reaction mechanism for the [Cp*Ru]-catalyzed
hydroformylation of alkenes.
Scheme 1. Synthesis and isolation of [Cp*Ru(Xantphos)H] (1).
entry 7 of Table 2 provides further evidence for the slow
insertion reaction of internal alkenes with 1.
with Xantphos first formed [Cp*Ru(Xantphos)Cl], which was
then converted into [Cp*Ru(Xantphos)H] (1) upon treat-
ment with sodium methoxide in methanol at 508C. Complex
The stoichiometric reaction of the bisdeuterated (C1) 1-
decene ([D2]-1-decene, D content 96%) and 1 was conducted
(Scheme 3) to determine the relative rate of 1,2- and 2,1-
alkene insertion and b-hydride elimination. Following 2,1-
1
1 was fully characterized by H, 13C, 31P NMR spectroscopy
and HRMS/ESI.[10] The n-selective hydroformylation with
suppressed alkene isomerization and good reproducibility
was achieved by using complex 1 and one additional
equivalent of Xantphos (aldehyde 29%, n/i = 31; Table 2,
entry 4)). When the reaction time was prolonged to 48 hours,
the aldehyde yield increased to 60% with an n/i of 28
(Table 2, entry 5), thus suggesting the absence of an induction
period and slow catalyst deactivation. Longer reaction times
did not change the amount of isomerized alkenes (8.5% for
24 h, 8.4% for 48 h), thereby suggesting that their formation
occurs predominantly during the initial stage of the reaction.
In contrast, the use of an in situ generated catalyst from the
combination of [Ru3(CO)12]/Xantphos resulted in significant
alkene isomerization (Table 2, entry 6). When (Z)-2-decene
was used as a substrate, no hydroformylation was observed
over 24 hours at 1608C (Table 2, entry 7). Using 1-eicosene,
a less volatile substrate, a similar result to that with 1-octene
was accomplished with complete mass recovery (Table 2,
compare entry 8 to entry 5). The use of other phosphorus
ligands such as PtBu3, P(o-tol)3, and DPPP resulted in low
selectivity and activity.[10]
We also conducted a series of mechanistic studies related
to hydroformylation catalyzed by [Cp*Ru(Xantphos)H]/
Xantphos complexes (Scheme 2). First, reversible 1-alkene
insertion into 1 was studied. When 1 was treated with
a stoichiometric amount of 1-decene at 1008C, the Ru/alkyl
complex 2 was not detected, but the isomerization of 1-decene
to (E)- and (Z)-2-decene (E/Z = 85:15) did occur over the
same period of time. No further isomerization to 3- or 4-
decenes was detected.[10] The production of 2-decenes is
Scheme 3. Control study for the isomerization of 1-decene.
insertion of [D2]-1-decene into 1, there are two potential b-
hydride elimination pathways (Scheme 4). One is b-hydride
elimination from C1 to reform 1-decene. The other is b-
hydride elimination from C3 to give 2-decene. From the
ꢀ
increase of C1 H and the increase of 2-decenes, the ratio of
kꢀ2D/k3 was estimated to be 0.8:1. Taking into account the
reported kinetic isotope effect for b-hydride elimination as
1.0–3.3, kꢀ2H/k3 could be estimated to be 0.8:1–2.6:1, thus
showing that the two pathways are comparable to each
other.[11] In contrast, the ratio of [D3]-1-decene to [D]-
1 remained constant at 1:1 during the reaction. These data
imply that 1,2-insertion and subsequent b-hydride elimination
is much faster than the 2,1-insertion and subsequent b-
hydride elimination (k+1H @ k+2H) such that there exists
a rapid equilibrium between [D2]-1-decene plus [D]-1 and
[D3]-1-decene plus 1; otherwise the ratio of [D3]-1-decene/
[D]-1 would have increased as the reaction proceeded.[12]
Reversible 1,2-insertion and nearly reversible 2,1-inser-
tion of the 1-alkene were suggested under the hydroformy-
lation conditions. The hydroformylation of [D2]-1-decene
catalyzed by 1/Xantphos afforded the product n-aldehyde
with the deuterium content, as well as recovered 1-decene
(Scheme 5). In the recovered 1-decene, the deuterium content
was 88% at C1 and 3% at C2. In the obtained n-aldehyde, it
was less than 1% at C1, 79% at C2, and 5% at C3. Thus, the
decrease of the terminal D content in [D2]-1-decene indicates
that 2,1-insertion/C1-b-deuteride elimination took place to
ꢀ
explained by 2,1-insertion of 1-decene into the Ru H and
subsequent b-hydride elimination from the C3-position.
ꢀ
These data suggest the insertion of 1-decene into the Ru H
bond in 1 is reversible while the insertion reaction of internal
alkenes to 1 is slow under these reaction conditions. The
almost quantitative recovery of (Z)-2-decene as shown in
Angew. Chem. Int. Ed. 2012, 51, 4383 –4387
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4385

.
Angewandte
Communications
Scheme 4. Possible reaction in the initial stage of the stoichiometric reaction of [D2]-1-decene and 1.
The catalytic cycle based on our mechanistic studies is as
follows (Scheme 2): First, 1-alkene coordination to the
ruthenium complex 1 and successive insertion takes place
initially through generation of an open coordination site by
either slippage of Cp* from an h5 coordination to
h3 coordination or by dissociation of one phosphorus atom
of the diphosphorus ligand. This step (1!2) is nearly
reversible. The CO coordination to 2 again requires the ring
slippage of Cp* or P dissociation. Migratory insertion of CO
to form 3 and subsequent hydrogenolysis will release product
aldehyde to complete the cycle.
Scheme 5. Deuterium distributions in the reaction mixture of hydro-
formylation using [D2]-1-decene.[13]
some extent. Since 1,2-insertion was estimated to be much
faster than 2,1-insertion as mentioned in the previous para-
graph, the 1,2-insertion step should occur reversibly at a rate
much faster than that of 2,1-insertion.
The mechanism of coordination of one of the substrates
(alkene, CO or H2) to the 18e, coordinatively saturated
ruthenium center of 1 was also considered. One may expect
either slippage of the h5-cyclopentadienyl ligand to give an
h3 coordination, or dissociation of one phosphorus atoms of
the bidentate ligand to k1 coordination (Figure 3). An h3-Cp*
Coordination and insertion of CO (2!3) or hydrogenol-
ysis (2!1) seems to be the rate-determining step for the
production of n-aldehyde since 1,2-insertion of the alkene was
determined to be reversible. Furthermore, the reaction rate in
the 1/Xantphos system exhibited a first-order dependence on
the 1-alkene concentration.[10] A higher concentration of 1-
alkene accelerates the reaction by changing the ratio of the
pre-equilibrium complexes (1 + 1-alkene versus 2 or 3) to
favor either the intermediate 2 or 3. Finally, the higher n-
selectivity obtained when using bulky bidentate phosphorus
ligands may be attributed to the destabilization of any
ruthenium/branched alkyl or acyl intermediates by creating
a sterically hindered environment around the ruthenium
center.
In conclusion, we have developed a new class of ruthe-
nium catalyst that is effective for the n-selective hydro-
formylation of terminal alkenes. The active species [Cp*Ru-
(Xantphos)H] (1) was isolated and the reaction mechanism
was investigated. The reaction rate is still far below that of the
industrially applied rhodium systems (about 3–4 orders of
magnitude), and future improvement of catalyst activity
(TOF) is desirable. Nevertheless, the obtained n/i ratios are
similar to what is typically obtained in commercial processes,
and [Cp*Ru] complexes may offer a new direction to further
catalyst development for selective hydroformylation reac-
tions and could obviate the use of expensive rhodium
complexes in the future.
Figure 3. Possible 16e intermediates.
intermediate is consistent with the accelerating effect
observed by substitution of an indenyl ligand for Cp*
(Table 1, entries 7, 9, and 10), although the effect is small.[14]
Dissociation of one phosphorus atom is still possible, but this
mechanism seems unlikely considering the lower yields of the
aldehyde product obtained when using monophosphines
instead of bisphosphines.[10]
4386 www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 4383 –4387

Angewandte
Chemie
mueller, B. R. Proft, B. A. Matter, D. R. Powell, J. Am. Chem.
Experimental Section
Soc. 1999, 121, 63; i) M. Takai, I. Nakajima, T. Tsukahara, Y.
Tanaka, H. Urata, A. Nakanishi, US2002049355, 2002.
[4] “Other Platinum Group Metals”: J. Butler in PLATINUM 2011,
Johnson Mattey Public Limited Company, Hertfordshire, 2011.
[5] D. Konya, K. Q. Almeida LeÇero, E. Drent, Organometallics
2006, 25, 3166 – 3174.
[6] I. R. J. Piras, R. J. Jackstell, A. Spannenberg, R. Franke, M.
Beller, Angew. Chem. 2011, 123, 294 – 298; Angew. Chem. Int.
Ed. 2011, 50, 280 – 284.
[7] a) D. Evans, J. A. Osborn, F. H. Jardine, G. Willkinson, Nature
1965, 208, 1203 – 1204; b) E. Cesarotti, A. Fusi, R. Ugo, G. M.
Zanderighi, J. Mol. Catal. 1978, 4, 205 – 216; c) M. Bianchi, G.
Menchi, P. Frediani, U. Matteoli, F. Piacenti, J. Organomet.
Chem. 1983, 247, 89 – 94; d) G. Sussfink, G. F. Schmidt, J. Mol.
Catal. 1987, 42, 361 – 366; e) J. F. Knifton, J. Mol. Catal. 1987, 43,
65 – 77; f) T. Hayashi, Z. H. Gu, T. Sakakura, M. Tanaka, J.
Organomet. Chem. 1988, 352, 373 – 378; g) M. M. T. Khan, S. B.
Halligudi, S. H. R. Abdi, J. Mol. Catal. 1988, 48, 313 – 317; h) T.
Mitsudo, N. Suzuki, T. Kondo, Y. Watanabe, J. Mol. Catal. A
1996, 109, 219 – 225; i) T. Mitsudo, N. Suzuki, T. Kobayashi, T.
Kondo, J. Mol. Catal. A 1999, 137, 253 – 262; j) K. I. Tominaga,
Catal. Today 2006, 115, 70 – 72; k) P. J. Baricelli, K. Segovia, E.
Lujano, M. ModroÇo-Alonso, F. Lꢀpez-Linares, R. A. Sꢁnchez-
Delgado, J. Mol. Catal. A 2006, 252, 70 – 75.
General procedure for the hydroformylation of propene: The Ru
complex (25 mmol) and bisphosphine (50 mmol) were charged with
2 mL of solvent in a 50 mL stainless steel autoclave with a magnetic
stirring bar. The autoclave was pressurized with 0.8 MPa of propene
(ca. 16 mmol). Soon after that, it was additionally pressurized with
2 MPa of H2/CO. After completion of the reaction under the
conditions indicated in the tables, the autoclave was cooled to 08C
with a water/ice bath and the temperature maintained for 30 min. The
gas pressure was released and then, to the the resulting solution,
dodecane (75 mg, 0.44 mmol) was added as an internal standard for
GC analysis. Judging from the propene conversion into aldehydes, the
initial charges of H2 and CO were high enough that the drops in
partial pressures were negligible.
General procedure for the hydroformylation of 1-decene: The Ru
complex and bisphosphine were charged with 2 mL of solvent in
a 50 mL stainless steel autoclave with a magnetic stirring bar. A
mixture of 1-decene and dodecane (2:1 molar ratio, 300 mL, ca. 1-
decene 1 mmol) was then added and the autoclave was pressurized
with appropriate pressure of H2/CO. After completion of the reaction
under the conditions indicated in the tables, the autoclave was cooled
to 08C with a water/ice bath. The gas pressure was released and the
resulting solution was analyzed by GC. The initial charges of H2 and
CO was high enough that the drops of their partial pressures were
negligible.
[8] V. S. Sridevi, W. K. Leong, J. Organomet. Chem. 2007, 692,
4909 – 4916.
[9] P. Ghosh, P. J. Fagan, W. J. Marshall, E. Hauptman, R. M.
Bullock, Inorg. Chem. 2009, 48, 6490 – 6500.
Detailed procedures and results of all the control experiments
can be found in the Supporting Information.
Received: November 29, 2011
Revised: January 18, 2012
[10] See the Supporting Information.
[11] Among the reported values, kH/kD = 2.28 ꢁ 0.20 for the thermol-
ysis of [Ir(CO)(n-octyl)(PPh3)2] better represents the KIE of b-
hydride elimination. In this case, kꢀ2H/k3 = 1.8:1. See: J. Evans, J.
Schwartz, P. W. Urquhart, J. Organomet. Chem. 1974, 81, C37 –
C39. Other reported systems are Co: T. Ikariya, A. Yamamoto, J.
Organomet. Chem. 1976, 120, 257 – 284. Pt: H. E. Bryndza, J.
Chem. Soc. Chem. Commun. 1985, 1696 – 1698; and review: M.
Sierra, M. Gomez-Gallego, Chem. Rev. 2011, 111, 4857 – 4963.
[12] For detailed discussions, see the Supporting Information.
[13] The values were determined by using 2,4,6-trimethoxybenzene
as an internal standard. Each signal of the 1H and 2H NMR
spectrum was assigned according to literature values. P. C. Casey,
L. M. Petrovich, J. Am. Chem. Soc. 1995, 117, 6007 – 6014. The
yield of i-aldehyde was too low for reliable analysis.
[14] “Ligand Substitution Ractions”: J. F. Hartwig in Organotransi-
tion Metal Chemistry: From Bonding to Catalysis, University
Science Books, Sausalito, CA, 2010, chap. 5, pp. 250 – 253.
Related articles discussing indenyl effect are Ru: M. P.
Gamasa, J. Gimeno, C. Gonzalez-Bernardo, B. M. Martin-
Vaca, Organometallics 1996, 15, 302 – 308; M. Bassetti, P. Case-
llato, Organometallics 1997, 16, 5470 – 5477; Rh: M. E. Rerek, F.
Basolo, J. Am. Chem. Soc. 1984, 106, 5908 – 5912; Mo: A. J. Hart-
Davis, C. White, R. J. Mawby, Inorg. Chim. Acta 1970, 4, 441; and
review Chem. Rev. 1987, 87, 307 – 318. Since the acceleration
effect is rather small, however, there might be other mechanisms
involved.
Published online: March 21, 2012
Keywords: aldehydes · cyclopentadienyl ligands ·
.
hydroformylation · ruthenium · synthetic methods
[1] a) P. W. N. M. van Leeuwen, C. Claver in Rhodium Catalyzed
Hydroformylation, Springer, Dordrecht, 2000; b) P. W. N. M.
van Leeuwen Homogeneous Catalysis: Understanding the Art,
Springer, Dordrecht, 2004.
[2] O. Roelen, US2327066, 1943.
[3] a) M. Kranenburg, Y. E. van der Burgt, P. C. J. Kamer,
P. W. N. M. van Leeuwen, K. Goubitz, J. Fraanje, Organometal-
lics 1995, 14, 3081 – 3089; b) L. A. van der Veen, M. D. K. Boele,
F. R. Bregman, P. C. J. Kamer, P. W. N. M. van Leeuwen, K.
Goubitz, J. Fraanje, H. Schenk, C. Bo, J. Am. Chem. Soc. 1998,
120, 11616 – 11626; c) L. A. van der Veen, P. C. J. Kamer,
P. W. N. M. van Leeuwen, Angew. Chem. 1999, 111, 349 – 351;
Angew. Chem. Int. Ed. 1999, 38, 336 – 338; d) L. A. van der Veen,
P. C. J. Kamer, P. W. N. M. van Leeuwen, Organometallics 1999,
18, 4765 – 4777; e) T. J. Devon, G. W. Phillips, T. A. Puckette,
J. L. Stavinoha, J. J. Vanderbilt, US4694109, 1987; f) C. P. Casey,
G. T. Whiteket, M. G. Melville, L. M. Petrovich, J. A. Gaveney,
D. R. Powell, J. Am. Chem. Soc. 1992, 114, 5535 – 5543; g) C. P.
Casey, E. L. Paulsen, E. W. Beuttenmuller, B. R. Proft, L. M.
Petrovich, B. A. Matter, D. R. Powell, J. Am. Chem. Soc. 1997,
119, 11817 – 11825; h) C. P. Casey, E. L. Paulsen, E. W. Beutten-
Angew. Chem. Int. Ed. 2012, 51, 4383 –4387
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4387