Ruthenium Carboxylate Complexes as Efficient Catalysts for the Addition of Carboxylic Acids to Propargylic Alcohols

Article abstract of DOI:10.1002/ejic.201500203

Ruthenium complexes [Ru(CO)₂(PPh₃)₂(O₂CR)₂] - 3a (R = CH₂OCH₃), 3b (R = iPr), 3c (R = tBu), 3d (R = 2-^cC₄H₃O), and 3e (R = Ph) - were synthesized

Full text of DOI:10.1002/ejic.201500203

FULL PAPER
DOI:10.1002/ejic.201500203
Ruthenium Carboxylate Complexes as Efficient Catalysts
for the Addition of Carboxylic Acids to Propargylic
Alcohols
Janine Jeschke,[a] Christian Gäbler,[a] Marcus Korb,[a]
Tobias Rüffer,[a] and Heinrich Lang*[a]
COVER PICTURE
Keywords: Synthetic methods / Homogeneous catalysis / Ruthenium / Propargylic alcohols
Ruthenium complexes [Ru(CO)2(PPh3)2(O2CR)2] – 3a (R =
CH2OCH3), 3b (R = iPr), 3c (R = tBu), 3d (R = 2-cC4H3O), and
3e (R = Ph) – were synthesized by treatment of Ru(CO)3-
(PPh3)2 with the corresponding carboxylic acids. The mo-
lecular structures of the newly synthesized complexes in the
solid state are discussed. Compounds 3a–e were successfully
applied as catalysts in the addition of carboxylic acids to
reaction conditions, as was confirmed by 31P{1H} NMR spec-
troscopic studies. The addition of catalytic amounts of
Na2CO3 resulted in an increase in β-oxo ester formation. The
reaction is tolerant to the use of versatile functional groups
on the propargylic alcohols and carboxylic acids, revealing a
broad substrate generality. In contrast to most other known
catalytic systems, even sterically hindered substrates, includ-
propargylic alcohols to give the corresponding β-oxo esters in ing 2,4,6-trimethylbenzoic acid, 1,1-diphenylprop-2-yn-1-ol,
good to excellent yields even in air. The different carboxylate
ligands do not have an influence on the productivities, be-
cause the carboxylates exchange rapidly under the applied
or the biologically active steroid ethisterone, were success-
fully converted.
C6Me6; PR3 = PPh3, PMe3, P(c-C4H3O)2(CϵCFc), phos-
phoramidite], the mononuclear bis(allyl)ruthenium(IV)
complex trans-[Ru(η3:η3-C10H16)(PPh3)Cl2], and the di-
nuclear species [Ru(CO)2(PPh3)(μ2-O2CH)]2 have shown the
best catalytic productivities for this reaction.[1,17–21]
Introduction
β-Oxopropyl esters are important substrates in, for ex-
ample, pharmaceutical chemistry, because the acetonyl ester
motif can be found in many natural products and bio-
logically active steroids.[1–4] Furthermore, β-oxopropyl
esters are basic intermediates in the synthesis of α-hydroxy
ketones,[5] furanones,[6,7] imidazoles,[8] and amides and
peptides.[9,10] Established methodologies for the preparation
of β-oxo esters generally include the hydration and esterifi-
cation of propargylic alcohols,[11] the oxidation of ketones
with metal acetates,[12] the carboxylation of α-halo
ketones,[13] and the Wacker oxidation of allyl carboxyl-
ates.[14] However, the ruthenium-catalyzed addition of carb-
oxylic acids to propargylic alcohols, which was first de-
scribed by Mitsudo and Watanabe in 1987,[15] is the most
straightforward and atom-economical approach for the syn-
thesis of β-oxopropyl esters. The proposed mechanism for
this reaction involves an initial Markovnikov addition of
the carboxylic acid to the ruthenium-coordinated alkyne,
followed by an intramolecular transesterification step
(Scheme 1).[16] To date, the mononuclear arene-RuII deriva-
tives Ru(η6-arene)(PR3)Cl2 [arene = p-cymene, C6H6,
Scheme 1. Possible mechanism for the Ru-catalyzed synthesis of β-
oxo esters.[16,19]
Here we describe the synthesis of the mononuclear RuII
carboxylate complexes Ru(CO)2(PPh3)2(O2CR)2 (R
=
[a] Technische Universität Chemnitz, Faculty of Natural Sciences,
Institute of Chemistry, Inorganic Chemistry,
CH2OCH3, iPr, tBu, 2-cC4H3O, Ph) and their use in the
catalytic formation of diverse β-oxopropyl esters. The im-
pact of the carboxylate ligands and the addition of catalytic
amounts of the base Na2CO3 on the reactivity of the cata-
lytic system are discussed as well.
09107 Chemnitz, Germany
E-mail: heinrich.lang@chemie.tu-chemnitz.de
https://www.tu-chemnitz.de/chemie/anorg/
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201500203.
Eur. J. Inorg. Chem. 2015, 2939–2947
2939
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
arrangement of the CO ligands, because for a trans-3a–e
assembly only one strong carbonyl stretching would be ex-
Results and Discussion
Ruthenium carboxylates [Ru(CO)2(PPh3)2(O2CR)2] – 3a pected.[28] In contrast, the IR spectrum of dinuclear 4a
(R = CH2OCH3), 3b (R = iPr), 3c (R = tBu), 3d (R = shows three strong ν(CO) absorptions between 1945–
2-cC4H3O), and 3e (R = Ph) – were synthesized through 2023 cm–1, typical for a sawhorse structure of Ru2(CO)4
straightforward reactions between Ru(CO)3(PPh3)2 (1) and moieties.[29–31] In addition, characteristic asymmetric and
the corresponding carboxylic acids 2a–e (Scheme 2). It symmetric carboxylate stretching vibrations were observed
could be shown that 4-methylpentan-2-one is best suited as for 3a–e and 4a. Carboxylates can coordinate to transition
solvent for this conversion because it prevents the decompo- metals in a monodentate,[32] chelating,[33] or μ-bridging[34]
sition of 3a–e and the formation of dinuclear species of the fashion. From the difference between the asymmetric
type [Ru(CO)2(PPh3)(O2CR)]2 (4a, R = CH2OCH3). After (νasym) and symmetric (νsym) CO vibrations (Δν, Δν =
appropriate workup, organometallic compounds 3a–e, as νasym – νsym) it can be concluded which structural bonding
well as 4a, could be isolated as air- and moisture-stable motif is present.[35] Monodentate-bonded R(O)CO– carb-
colorless solids soluble in most common organic solvents, oxylates show increased separation of the corresponding
including dichloromethane and toluene.
νCO2 frequencies, whereas chelating or μ-bridging carboxyl-
ates show reduced Δν values.[35] For mononuclear ruth-
enium complexes 3a–e, for example, Δν values ranging from
262 to 312 cm–1 were found (Experimental Section),
indicating a monodentate coordination of the carboxylate
groups to the RuII ion. This finding is consistent with
similar Ru carboxylate complexes – that is, Ru(CO)2(PPh3)2-
(O2CCH3)2[32a] – and was confirmed by single-crystal X-ray
structure determination (see below). In contrast, the Δν
value of 160 cm–1 for 4a indicates a μ-bridging coordina-
tion, which is representative of complexes of structural type
4.[29]
Scheme 2. Synthesis of compounds 3a–e and 4a. (i) 4-Meth-
ylpentan-2-one, 100 °C, 1 h (Experimental Section).
The structures of 3a–d and 4a in the solid state were
determined by single-crystal X-ray diffraction analysis.
The identities of 3a–e and 4a were confirmed by elemen- Suitable crystals were obtained by slow diffusion of hexane
tal analysis, IR, NMR (1H, 13C{1H}, 31P{1H}) spec- into a dichloromethane solution containing the appropriate
troscopy, and ESI mass spectrometry (Experimental Sec- ruthenium complex at ambient temperature. ORTEP
tion). In addition, the structures of 3a–d and 4a in the solid diagrams, along with key structural data for 3a–d and 4a,
state were determined by single-crystal X-ray structure are presented in Figures 1, 2, and 3. Crystal and structure
analysis.
refinement data are presented in the Experimental Section.
The 31P{1H} NMR spectra of ruthenium complexes 3a–
Compounds 3a and 3b crystallize in the monoclinic space
e and 4a each exhibit one characteristic singlet for the PPh3 group P21/n, whereas compounds 3c, 3d, and 4a crystallize
¯
groups, in a range from 29.7 to 32.4 ppm for 3a–e and at in the triclinic space group P1 with one molecule in the
13.7 ppm for 4a (see Experimental Section); this is typical asymmetric unit. Complexes 3a–d each consist of a ruth-
for these types of ruthenium complexes.[22,23] These reso- enium atom hexacoordinated with two trans-positioned tri-
nance signals are shifted to higher field than in the case phenylphosphine groups (P1, P2) in the apical positions,
of 1 (δ =55.4 ppm), and so 31P{1H} NMR spectroscopy is two cis-carbonyl groups, and two cis-monodentate O-
suitable to monitor the progress of the reactions.
bonded carboxylate groups in the equatorial plane of a dis-
The H and 13C{1H} NMR spectra of 3a–e and 4a are torted octahedron (Figures 1 and 2). The Ru–P bond
consistent with the proposed structures. Eye-catching is the lengths in complexes 3a–d are not affected by the different
splitting of the PPh3 ipso-, ortho-, and meta-carbon atoms carboxylate ligands and are comparable to those in pre-
into triplets in the 13C{1H} spectra, a typical phenomenon viously published species.[36–38] However, the angles be-
for symmetry-related nuclei present in transition metal tween the carbonyl groups are influenced by sterically
phosphine compounds.[24] The spin system can be described demanding groups in the carboxylate ligands, reducing
as an AAЈXXЈ pattern (A = P, X = C), in which the ruth- from 86.38(9)° in 3a to 83.5(3)° in 3d. For complexes 3a–d
enium serves as a bridge between the phosphorus atoms. it is striking that both non-binding carboxylate oxygen
Calculations by Metzinger and Harris confirmed that such atoms are directed away from each other but towards the
1
splitting occurs when the coupling constant JP,P is signifi- carbonyl ligands, revealing an interaction with the central
[25,26]
cantly larger than that of JC,P
.
Ru1 atom, due to Ru–O distances significantly below the
The IR spectra of mononuclear complexes 3a–e each ex- sum of the van der Waals radii [Σ = 3.54 Å,[39] but
hibit, as expected, two strong stretching vibrations for the 3.1621(18) in 3c and 3.234(3) Å in 3b]. The carboxylate
terminal carbonyl groups, between 1974 and 2047 cm–1 (Ex- moiety itself is not in plane with the square-planar coordi-
perimental Section).[27] The appearance of two absorptions nation environment, with a twist angle ranging from 19.8(3)
in octahedrally structured 3a–e confirms the cis (3b) to 5.86(15)° (3a).[40]
Eur. J. Inorg. Chem. 2015, 2939–2947
2940
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
Figure 1. ORTEP diagram (50% probability level) of the molecular structures of 3a (left) and 3b (right), together with the atom numbering
scheme. All hydrogen and disordered atoms have been omitted for clarity. Selected interatomic distances [Å] and angles [°]. For 3a: Ru1–
P1 2.4070(6), Ru1–C1 1.880(2), Ru1–O3 2.1063(14), C1–O1 1.139(3), C3–O3 1.290(3), C3–O4 1.224(3), P1–Ru1–P2 176.52(2), Ru1–C1–
O1 173.6(2), P1–Ru1–C1 90.57(7), O3–C3–O4 125.9(2). For 3b: Ru1–P1 2.4105(12), Ru1–C1 1.884(5), Ru1–O3 2.090(3), C1–O1 1.148(6),
C3–O3 1.291(6), C3–O4 1.227(7), C3–C4 1.528(8), C4–C5 1.453(9), C4–C6 1.498(8), P1–Ru1–P2 169.99(5), Ru1–C1–O1 173.5(5), P1–
Ru1–C1 94.74(16), O3–C3–O4 125.2(5).
Figure 2. ORTEP diagram (50% probability level) of the molecular structures of 3c (left) and 3d (right), together with the atom numbering
scheme. All hydrogen atoms and solvent molecules (3c; CH2Cl2) have been omitted for clarity. Selected interatomic distances [Å] and
angles [°]. For 3c: Ru1–P1 2.4089(7), Ru1–C1 1.883(3), Ru1–O3 2.0947(17), C1–O1 1.137(3), C3–O3 1.291(3), C3–O4 1.228(3), C3–C4
1.534(4), C4–C5 1.521(4), C4–C6 1.526(4), C4–C7 1.527(4), P1–Ru1–P2 171.58(2), Ru1–C1–O1 171.1(2), P1–Ru1–C1 93.70(8), O3–C3–
O4 124.6(2). For 3d: Ru1–P1 2.4098(17), Ru1–C11 1.848(8), C11–O7 1.171(9), Ru1–O1 2.075(5), C1–O1 1.281(8), C1–O2 1.229(8), C1–
C2 1.475(10), C2–O3 1.383(8), C2–C3 1.308(11), C3–C4 1.440(13), C4–C5 1.280(12), O3–C5 1.378(10), P1–Ru1–P2 174.31(7), Ru1–O1–
C1 120.0(5), Ru1–C11–O7 171.7(6), O1–C1–O2 126.8(7), P1–Ru1–C11 91.1(2).
Dinuclear compound 4a (Figure 3) belongs to a family of ligands. A number of such tetracarbonyl-1κ2C,2κ2C-bis(tris-
μ-carboxylato-bridged complexes in which different family phenylphosphine-1κP,2κP)-bis(μ-carboxylato-κO:κOЈ)di-
members possess varying organic groups in the carboxylate ruthenium(Ru-Ru) complexes have already been crystallo-
Eur. J. Inorg. Chem. 2015, 2939–2947
2941
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
graphically characterized.[23] Characteristic for dinuclear ture, solvent and by addition of catalytic amounts of
compounds of this kind is the Ru–Ru backbone, in which Na2CO3 (Table 1). The efficiencies of all reactions were
1
each of the two Ru atoms shows a pseudo-octahedral coor- determined by H NMR spectroscopy, with acenaphthene
dination geometry with two CO ligands in cis-equatorial as internal standard. In a typical experiment, benzoic acid
positions (Figure 3). The Ru2(CO)4 core in 4a resembles a (2.0 mmol), propargylic alcohol (3.0 mmol), acenaphthene
sawhorse with two μ-η2-O,O-O2CCH2OMe carboxylate (1.0 mmol), and Na2CO3 (0.08 mmol) were dissolved in
bridges. The two triphenylphosphine ligands are located in toluene (20 mL) and treated with 1.0 mol-% of 3a for 24 h
the axial positions around the Ru atoms. The Ru–Ru dis- at elevated temperature. Special precautions against air or
tance is, at 2.7433(10) Å, in accordance with a Ru–Ru single moisture were not necessary.
bond and compares well with those in other family
members.[41] The alignment of P1–Ru1–Ru2–P2 is not
Table 1. Optimization of the reaction conditions for the ruthenium-
catalyzed addition of benzoic acid to prop-2-yn-1-ol.
linear [Є P1–Ru1–Ru2, 166.567(17) and 165.639(18)°], with
both phosphorus atoms directed towards the carboxylate
ligands with decreased bond angles between both apical
moieties.[42] The phosphine ligands are rotated by
–17.49(11)° (C23–P1–P2–C35) with respect to each other,
revealing a conformation between fully eclipsed and
staggered, contrary to mononuclear 3a–d, each of which
shows an almost ideally staggered conformation [41.8(4)° in
3d to 58.67(12)° in 3a]. Furthermore, the Ru–P bond
lengths in 4a [2.4450(11) and 2.4314(11) Å] are significantly
increased relative to those in 3a–d [2.3993(7) Å in 3c to
2.4105(12) Å in 3b], but are still within the range for Ru–P
bond lengths in already reported family members.[43] The
carbonyl groups in 4a are linear, with Ru–C–O angles be-
tween 177.5(2) and 178.5(2)° and C–O bond lengths be-
tween 1.157(3) and 1.159(3) Å. The angles between the cis-
bonded carbonyl groups vary from 85.94(12)° for C7–Ru1–
C8 to 88.08(12)° for C9–Ru2–C10. The equatorial planes of
the two square-planar RuC2O2 sub-coordination units are
tilted towards each other by 14.456(10)°, due to the carb-
oxylate substituents; this is average (11.15 to 18.82°) for
reported examples of cis-bridged complexes.[44]
Entry Catalyst Solvent
Na2CO3 [mol-%] Yield[a] [%]
1
2
3
4
3a
3a
3a
3a
3a
3a
3a
3a
3a
3a
3a
3a
3b
3c
3d
3e
4a
toluene
toluene
toluene
toluene
toluene
toluene
toluene
chlorobenzene
toluene/acetonitrile (9:1)
dioxane
ethanol
cyclohexane
toluene
toluene
toluene
toluene
toluene
0
1
4
10
40
4
4
4
4
4
4
4
4
4
4
4
4
61
67
90
78
31
97
34
83
69
34
6
50
88
89
86
89
76
5
6[b]
7[c]
8
9
10
11
12
13
14
15
16
17
[a] Reaction conditions: benzoic acid (2.0 mmol), propargylic
alcohol (3.0 mmol), catalyst (0.02 mmol), acenaphthene
(1.0 mmol), 24 h at 80 °C. The yields were determined by 1H NMR
spectroscopy with acenaphthene as internal standard. [b] Reaction
performed at 90 °C. [c] Reaction performed at 70 °C.
For the ruthenium-catalyzed addition of carboxylic acids
to terminal alkynes, Gooßen et al. could show that the ad-
dition of catalytic amounts of Na2CO3 not only improves
the selectivity of the enol ester synthesis, but also enhances
the reaction rate.[45,46] These results prompted us to explore
the impact of Na2CO3 on the related ruthenium-catalyzed
β-oxopropyl ester synthesis systematically. From Table 1 it
can be seen that the obtained yields strongly depend on the
amount of added Na2CO3. Without additional base a yield
of 61% could be achieved (Entry 1). The best catalytic re-
sults, with a 90% yield, were obtained by applying four
equivalents (based on ruthenium) of Na2CO3 (Entry 3,
Table 1). On further increasing the amount of the Na2CO3
base the yields were reduced, due to the poor solubility of
Na2CO3 and the formed sodium benzoate in toluene, with
Figure 3. ORTEP diagram (50% probability level) of the molecular
structure of 4a, together with the atom numbering scheme. All
hydrogen atoms are omitted for clarity. Selected interatomic dis-
tances [Å] and angles [°]: Ru1–Ru2 2.7433(10), Ru1–P1 2.4450(10),
Ru1–C7 1.836(3), C7–O7 1.157(3), Ru1–O1 2.1461(19), C1–O1
1.258(3), C1–O2 1.247(3), Ru1–C7–O7 178.4(2), P1–Ru1–Ru2
65.639(19), O1–Ru1–C7 92.50(10), O1–C1–O2 126.6(2), O1–Ru1–
Ru2 82.65(5).
For initial catalytic studies and optimization reactions increasing viscosity of the reaction mixture hence occurring
the conversion of benzoic acid (2e) and propargylic alcohol (Entries 4–5, Table 1).
(5a) into 2-oxopropyl benzoate (6a) in the presence of cata-
We also screened the dependency of the productivities on
lytic amounts of 3a was chosen as the model reaction. The the reaction temperature. Complete conversion was
reaction conditions were optimized by varying the tempera- achieved at a working temperature of 90 °C (Entry 6,
Eur. J. Inorg. Chem. 2015, 2939–2947
2942
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
Table 1). However, a decrease in the yield from 97% to 90%
was observed when the temperature was reduced to 80 °C
(Entry 3, Table 1). A further decrease in the reaction tem-
perature resulted in a significant drop of productivity down
to 34% (Entry 7, Table 1).
In addition, a survey of the influence of the solvent on
the productivity of the conversion was made. The most
commonly utilized solvent in β-oxopropyl ester synthesis is
toluene.[17,18,37] With this solvent the best results could be
obtained (Entry 3, Table 1). Only slightly lower yields were
achieved when chlorobenzene was used (Entry 8, Table 1).
The application of coordinating solvents, however, reduced
the productivity drastically (Entries 9–11, Table 1). Never-
theless, when nonpolar cyclohexane was used instead, a de-
crease in the yield of the β-oxopropyl ester was observed,
which can be explained by poor solubility of the corre-
sponding ruthenium complex (Entry 12, Table 1).
Figure 4. 31P{1H} NMR spectra of 1:1 mixtures of (a) Ru(CO)2-
(PPh3)2(O2CtBu)2 (3c) and PhCO2H, and (b) Ru(CO)2(PPh3)2-
(O2CPh)2 (3e) and tBuCO2H in CDCl3 after 1 h at 80 °C.
Subsequently, ruthenium complexes 3a–e and 4a were
screened under optimized reaction conditions (Entry 3 and
13–17, Table 1). The obtained results reveal that all com-
Diverse propargylic alcohols were treated with benzoic
plexes are catalytically active in the addition of benzoic acid acid under the optimized reaction conditions (vide supra)
to propargylic alcohol to give β-oxopropyl benzoate. How- in order to assess the substrate scope of the reaction. From
ever, use of the dinuclear compound 4a leads to a somewhat Table 2 it can be seen that besides the primary propargylic
reduced yield (76%, Entry 17, Table 1), relative to the re- alcohol prop-2-yn-1-ol (Entry 1), secondary (Entries 2–3)
sults obtained with mononuclear 3a–e, with equal ruth- and tertiary propargylic alcohols (Entries 4–9) can also be
enium loadings. It is interesting to note that although the successfully converted into the corresponding β-oxopropyl
mononuclear ruthenium complexes screened were elec- esters at catalyst loadings of 1.0 mol-% in good to excellent
tronically and sterically modified by varying the carboxyl- yields (Table 2, Experimental Section). The yields show
ate ligands, no significant differences in productivities were only moderate substrate dependency. Even the addition of
observed. This prompted us to analyze the reaction mixture benzoic acid to the tertiary diaromatic propargylic alcohol
by 31P{1H} NMR spectroscopy. For example, a solution of 1,1-diphenylprop-2-yn-1-ol was successful, with a yield of
3c was treated with benzoic acid in a ratio of 1:1. The re- 79% (Entry 6, Table 2). Most other known catalytic systems
sulting 31P{1H} NMR spectrum exhibited three signals failed in this conversion, with detection of only trace
(Figure 4, a), which could be assigned to the initial catalyst amounts of the expected β-oxopropyl ester 6f.[16,37]
3c (δ = 29.7 pm), the corresponding benzoate complex 3e (δ
Complex 3a also promotes the addition of benzoic acid
= 31.2 ppm), and the mixed carboxylate complex [Ru(CO)2- to cyclic propargylic alcohols. The best result (94% yield)
(PPh3)2(O2CtBu)(O2CPh)] (δ = 30.1 ppm). When benzoate was obtained with 1-ethynylcyclopentanol, due to least
complex 3e was treated with pivalic acid, the same three steric hindrance (Entry 7, Table 2). A somewhat reduced
signals in the 31P{1H} NMR spectrum could be observed yield (87%) was achieved with 1-ethynylcyclohexanol (En-
(Figure 4, b). This implies that an equilibrium between the try 8, Table 2). Even the bulky steroid ethisterone could be
carboxylate ligands exists and is also detectable at ambient successfully converted into the corresponding β-oxopropyl
temperature. The location of the equilibrium is affected by ester 6i (Entry 9, Table 2). Subsequent purification by col-
the acidity of the appropriate carboxylic acid and the steric umn chromatography on silica gel provided analytically
parameters. As a result of the high temperature and the pure samples of all β-oxopropyl esters 6a–i, characterized
large excess of carboxylic acid applied during the conver- by standard analytical methods (Supporting Information),
sion a rapid exchange of the carboxylate ligands occurs. In in yields of up to 93%.
all 31P{1H} NMR spectra of the reaction mixtures after the
After evaluation of the scope with regard to the prop-
catalytic conversions (Table 1) only one main signal of the argylic alcohols, the influence of the carboxylic acid was
benzoate complex 3e was found, due to a large excess of examined. Therefore, several aromatic and aliphatic carb-
benzoic acid. This also explains why the observed produc- oxylic acids were treated with propargyl alcohol. From Fig-
tivities for complexes 3a–e are identical under the same re- ure 5 it can be seen that the catalytic addition is not limited
action conditions. Related investigations concerning the ex- to benzoic acid, with the reaction also proceeding with
change of carboxylate ligands in [Ru(PPh3)2(κ2-O2CMe)2] heteroaromatic substrates, affording the corresponding β-
were performed by Lynam’s group.[37] In contrast to our oxopropyl esters 7a–c in excellent isolated yields (86–92%).
findings, they determined differences in productivities be- The reaction is also compatible with olefinic substrates such
tween the complexes [Ru(PPh3)2(κ2-O2CMe)2] (81%) and as 3-(2-furyl)acrylic acid (7c). A survey of various electroni-
[Ru(PPh3)2(κ2-O2CPh)2] (68%), in the addition of benzoic cally modified carboxylic acids showed that acids with elec-
acid to 1-phenylprop-2-yn-1-ol.[37]
tron-withdrawing functionalities have higher reactivity than
Eur. J. Inorg. Chem. 2015, 2939–2947
2943
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
Table 2. Variation of the propargylic alcohol in the formation of β-
oxopropyl esters.
Figure 5. Variation of the carboxylic acid in the synthesis of β-
oxopropyl esters. Reaction conditions: carboxylic acid (2.0 mmol),
propargylic alcohol (3.0 mmol), 3a (0.02 mmol), Na2CO3
(0.08 mmol), acenaphthene (1.0 mmol), 24 h in toluene (20 mL) at
90 °C. Reaction times are not optimized. The yields were deter-
mined by 1H NMR spectroscopy, with acenaphthene as internal
standard. Isolated yields are given in brackets.
Conclusions
In summary, in these studies we could show for the first
time that mononuclear ruthenium compounds of type
Ru(CO)2(PPh3)2(O2CR)2 – 3a (R = CH2OCH3), 3b (R =
iPr), 3c (R = tBu), 3d (R = 2-cC4H3O), and 3e (R = Ph) –
are efficient catalysts for the synthesis of β-oxopropyl esters
from carboxylic acids and propargylic alcohols. The ruth-
enium complexes 3a–e described here are easily accessible
in good yields by a “one-pot” synthetic methodology: the
reaction between Ru(CO)3(PPh3)2 (1) and the appropriate
carboxylic acids HO2CR (2a–e). Electronic or steric modifi-
cation of the catalysts by variation of the carboxylate li-
gands does not lead to different productivities because the
carboxylates are able to exchange rapidly under the applied
reaction conditions. In contrast, the addition of catalytic
amounts of the base Na2CO3 resulted in an
obvious increase in the obtained yields. The screening of
variable propargylic alcohols and carboxylic acids revealed
a remarkable substrate generality. Unlike in cases of other
catalysts known to promote this reaction,[16,37] even bulky
tertiary propargylic alcohols such as 1,1-diphenylprop-2-
yn-1-ol or the biologically active steroid ethisterone could
be successfully employed. Further synthetic and catalytic
investigations relating to the electronic impact of the phos-
phine ligands, as well as experiments to understand the
mechanism of the reaction better, are currently in progress.
[a] Reaction conditions: benzoic acid (2.0 mmol), propargylic
alcohol (3.0 mmol), 3a (0.02 mmol), Na2CO3 (0.08 mmol), ace-
naphthene (1.0 mmol), 24 h in toluene (20 mL) at 90 °C. The yields
were determined by H NMR spectroscopy, with acenaphthene as
internal standard. Isolated yields are given in brackets.
1
substrates featuring electron-donating groups (7g–j, Fig-
ure 5). It is also interesting to note that sterically hindered
substrates such as 2,4,6-trimethylbenzoic acid, 1-naphthoic
acid, or 9-anthracenecarboxylic acid can be successfully
converted without increasing the catalyst loading or tem-
perature, generating new β-oxopropyl esters 7j–l (Figure 5)
in good to excellent isolated yields (61–87%, see Experi-
mental Section).
Experimental Section
General: Compound 1 was prepared by published procedures.[47,48]
Toluene was dried with a solvent purification system (MB SPS-
800, MBraun). All other reagents were obtained from commercial
suppliers and used without further purification. For column
Eur. J. Inorg. Chem. 2015, 2939–2947
2944
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
chromatography, silica with a particle size of 40–60 μm [230–
400 mesh (ASTM), Fa. Macherey–Nagel] was used.
(w), 1162 (w), 1095 (m), 745 (m), 708 (m), 696 (s), 602 (m), 520
(s) cm–1. HRMS (ESI-TOF): calcd. for C42H37O4P2Ru [M – O2C-
iPr]+ 769.1217; found 769.1182. C46H44O6P2Ru (855.87): calcd. C
64.55, H 5.18; found C 64.24, H 5.18.
Instruments: NMR spectra were recorded with a Bruker Advance
III 500 spectrometer operating at 500.3 MHz for 1H, 125.7 MHz
for 13C{1H}, and 202.5 MHz for 31P{1H} in the Fourier transform Crystal Data for 3b: C46H44O6P2Ru, Mr = 855.82, monoclinic, P21/
mode at 298 K. Chemical shifts are reported in δ (ppm) downfield
n, λ = 0.71073 Å, a = 11.5362(8) Å, b = 18.5522(16) Å, c =
from tetramethylsilane with the solvent as reference signal [1H
19.1499(17) Å, β = 99.681(7)°, V = 4040.1(6) Å3, Z = 4, ρcalcd.
=
NMR, CHCl3 δ = 7.26; 13C(1H) NMR, CDCl3 δ = 77.16; 31P(1H) 1.407 gcm–3, μ = 0.516 mm–1, T = 110 K, θ range 2.923–24.998°,
NMR, standard external relative to 85% H3PO4 δ = 0.0 ppm].
FTIR spectra were recorded with a Thermo Nicolet IR 200 instru-
ment. The melting points were determined with a Gallenkamp
MFB 595 010 M melting point apparatus. Elemental analyses were
measured with a Thermo FlashAE 1112 instrument. High-resolu-
tion mass spectra were recorded with a Bruker Daltonite micrO-
TOF-QII spectrometer and use of electron spray ionization (ESI).
18044 reflections collected, 7106 independent reflections (Rint
0.0855), R1 = 0.0686, wR2 = 0.1738 [IϾ2σ(I)].
=
Synthesis of Ru(CO)2(PPh3)2(O2CtBu)2 (3c): Ru(CO)3(PPh3)2 (1,
400 mg, 564 μmol) was treated with pivalic acid (230 mg,
2.25 mmol) by the synthesis procedure described above to afford
3c, yield 234 mg (265 μmol, 47% based on 1), m.p. 230 °C. 1H
NMR (CDCl3): δ = 0.55 (s, 18 H, CH3), 7.36–7.42 (m, 18 H, m-CH,
p-CH), 7.67–7.74 (m, 12 H, o-CH) ppm. 13C{1H} NMR (CDCl3): δ
= 28.1 (s, CH3), 39.4 [s, C(CH3)3], 128.6 (t, JC,P = 5.0 Hz, m-CH),
General Preparation Procedure for Ru(CO)2(PPh3)2(O2CR)2 (3a–e):
The complexes Ru(CO)2(PPh3)2(O2CR)2 (3a–e) were prepared by
modified literature procedures.[28,36] A suspension of Ru(CO)3-
(PPh3)2 (1, 400 mg, 564 μmol) and the appropriate carboxylic acid
(4 equiv., 2.25 mmol) in 4-methylpentan-2-one (10 mL) was heated
at 100 °C for 1 h. After solvent removal the oily residue was sus-
pended in diethyl ether (10 mL). The colorless precipitate that
formed was separated by filtration, washed with diethyl ether (2ϫ
10 mL), and dried in vacuo.
130.63 (s, p-CH), 130.67 (t, JC,P = 23.4 Hz, i-C), 134.5 (t, JC,P
=
5.7 Hz, o-CH), 183.4 (s, O2CtBu), 196.9 (t, JC,P = 11.7 Hz,
CO) ppm. 31P{1H} NMR (CDCl3): δ = 29.7 (s) ppm. IR data
(KBr): ν = 3056 (w), 2952 (w), 2924 (w), 2898 (w), 2863 (w), 2042
˜
(vs), 1980 (vs), 1601 (s), 1562 (w), 1479 (m), 1434 (m), 1390 (m),
1332 (s), 1212 (m), 1159 (w), 1095 (m), 1029 (w), 998 (w), 889 (w),
783 (w), 747 (m), 706 (m), 694 (m), 617 (w), 603 (m), 520 (s).
HRMS (ESI-TOF): calcd. for C43H39O4P2Ru [M – O2CtBu]+
783.1373; found 783.1363. C48H48O6P2Ru (883.92): calcd. C 65.22,
H 5.47; found C 64.99, H 5.58.
Synthesis of Ru(CO)2(PPh3)2(O2CCH2OCH3)2 (3a): Ru(CO)3-
(PPh3)2 (1, 400 mg, 564 μmol) was treated with methoxyacetic acid
(203 mg, 2.25 mmol) by the synthesis procedure described above to
give 3a, yield 310 mg (361 μmol, 64% based on 1), m.p. 242 °C. 1H Crystal Data for 3c: C48H48O6P2Ru·0.8CH2Cl2, Mr = 951.81, tri-
¯
NMR (CDCl3): δ = 2.78 (s, 6 H, OCH3), 2.97 (s, 4 H, CH2), 7.36– clinic, P1, λ = 0.71073 Å, a = 12.0218(4) Å, b = 12.2938(4) Å, c =
7.46 (m, 18 H, m-CH, p-CH), 7.64–7.71 (m, 12 H, o-CH) ppm. 16.4757(6) Å, α = 102.909(3)°, β = 99.681(7)°, γ = 90.151(3)°, V =
13C{1H} NMR (CDCl3): δ = 57.6 (s, OCH3), 70.5 (s, CH2), 127.5
2288.77(14) Å3, Z = 2, ρcalcd. = 1.381 gcm–3, μ = 0.553 mm–1, T
(t, JC,P = 5.0 Hz, m-CH), 128.9 (t, JC,P = 23.7 Hz, i-C), 129.8 (s, p- = 110 K, θ range 2.961–25.500°, 16149 reflections collected, 8499
CH), 133.3 (t, JC,P = 5.6 Hz, o-CH), 174.2 (s, O2C), 195.7 (t, JC,P independent reflections (Rint = 0.0276), R1 = 0.0383, wR2 = 0.0902
= 10.8 Hz, CO) ppm. 31P{1H} NMR (CDCl3): δ = 32.4 (s) ppm. [IϾ2σ(I)].
IR (KBr): ν = 3061 (w), 3025 (w), 2988 (w), 2960 (w), 2925 (w),
˜
Synthesis of Ru(CO)2(PPh3)2(O2CcC4H3O)2 (3d): Ru(CO)3-
(PPh3)2 (1, 400 mg, 564 μmol) was treated with furan-2-carboxylic
acid (253 mg, 2.25 mmol) by the synthesis procedure described
above to afford 3d, yield 243 mg (269 μmol, 48% based on 1), m.p.
2896 (w), 2869 (w), 2824 (w), 2046 (vs), 1987 (vs), 1641 (s), 1607
(s), 1484 (m), 1436 (s), 1385 (m), 1367 (m), 1339 (w), 1295 (s), 1195
(m), 1161 (w), 1120 (s), 1095 (s), 1030 (w), 1020 (w), 998 (w), 943
(m), 916 (w), 752 (s), 709 (s), 696 (s), 625 (w), 590 (m) cm–1. HRMS
(ESI-TOF): calcd. for C44H40O8P2Ru + Na [M + Na]+ 883.1146;
found 883.1099. C44H40O8P2Ru (859.81): calcd. C 61.46, H 4.69;
found C 61.16, H 4.84.
1
3
232 °C. H NMR (CDCl3): δ = 6.16 (dd, JH,H = 3.3, 1.7 Hz, 2 H,
3
4
4-H/C4H3O), 6.42 (dd, JH,H = 3.3, JH,H = 0.8 Hz, 2 H, 3-H/
3
4
C4H3O), 7.19 (dd, JH,H = 1.6, JH,H = 0.8 Hz, 2 H, 5-H/C4H3O),
7.25–7.29 (m, 18 H, m-CH, p-CH), 7.71–7.77 (m, 12 H, o-
Crystal Data for 3a: C44H40O8P2Ru, Mr = 859.77, monoclinic, P21/ CH) ppm. 13C{1H} NMR (CDCl3): δ = 110.3 (s, C-4/C4H3O),
n, λ = 0.71073 Å, a = 10.1053(2) Å, b = 20.5452(3) Å, c =
19.1152(3) Å, β = 91.093(2)°, V = 3967.89(12) Å3, Z = 4, ρcalcd.
1.439 gcm–3, μ = 0.529 mm–1, T = 106.9(5) K, θ range 3.01–26.00°,
112.9 (s, C-3/C4H3O), 128.2 (t, JC,P = 5.1 Hz, m-CH), 129.6 (t, JC,P
= 23.9 Hz, i-C/Ph), 130.5 (s, p-CH), 134.4 (t, JC,P = 5.7 Hz, o-CH),
142.7 (s, C-5/C4H3O), 150.0 (s, C-2/C4H3O), 163.7 (s, O2C), 196.7
(t, JC,P = 11.0 Hz, CO) ppm. 31P{1H} NMR (CDCl3): δ = 31.0
=
31206 reflections collected, 7777 independent reflections (Rint
0.0381), R1 = 0.0318, wR2 = 0.0727 [IϾ2σ(I)].
=
(s) ppm. IR data (KBr): ν = 3112 (w), 3051 (w), 2042 (vs), 1979
˜
(vs), 1606 (s), 1578 (m), 1485 (m), 1434 (m), 1385 (m), 1347 (vs),
1186 (m), 1094 (m), 1077 (w), 1012 (w), 929 (w), 886 (w), 808 (m),
781 (m), 743 (s), 708 (m), 691 (m), 604 (m) cm–1. HRMS (ESI-
TOF): calcd. for C43H33O5P2Ru [M – O2CcC4H3O]+ 793.0853;
found 793.0851; calcd. for C48H36O8P2Ru + Na [M + Na]+
927.0835; found 927.0757. C48H36O8P2Ru (903.83): calcd. C 63.79,
H 4.01; found C 63.75, H 3.99.
Synthesis of Ru(CO)2(PPh3)2(O2CiPr)2 (3b): Ru(CO)3(PPh3)2 (1,
400 mg, 564 μmol) was treated with isobutyric acid (199 mg,
2.25 mmol) by the synthesis procedure described above to give 3b,
1
yield 276 mg (322 μmol, 57% based on 1), m.p. 226 °C. H NMR
3
3
(CDCl3): δ = 0.45 (d, JH,H = 7.0 Hz, 12 H, CH3), 1.58 (sep, JH,H
= 7.0 Hz, 2 H, O2CCH), 7.34–7.44 (m, 18 H, m-CH, p-CH), 7.64–
7.74 (m, 12 H, o-CH) ppm. 13C{1H} NMR (CDCl3): δ = 19.7 (s,
¯
CH3), 35.7 (s, O2CCH), 128.4 (t, JC,P = 5.0 Hz, m-CH), 130.4 (t, Crystal Data for 3d: C48H36O8P2Ru, Mr = 903.78, triclinic, P1, λ
JC,P = 23.6 Hz, i-C), 130.6 (s, p-CH), 134.4 (t, JC,P = 5.7 Hz, o-
= 0.71073 Å, a = 12.1102(9) Å, b = 12.5567(10) Å, c =
CH), 181.9 (s, O2CiPr), 197.1 (t, JC,P = 11.4 Hz, CO) ppm. 31P{1H}
14.6692(11) Å, α = 78.959(7)°, β = 68.009(7)°, γ = 76.312(7)°, V =
1996.5(3) Å3, Z = 2, ρcalcd. = 1.503 gcm–3, μ = 0.530 mm–1, T =
110 K, θ range 2.878–26.000°, 18514 reflections collected, 7812 in-
dependent reflections (Rint = 0.0771), R1 = 0.0900, wR2 = 0.2201
NMR (CDCl ): δ = 32.1 (s) ppm. IR data (KBr): ν = 3060 (w),
˜
3
3026 (w), 2983 (w), 2962 (w), 2931 (w), 2917 (w), 2868 (w), 2036
(vs), 1974 (vs), 1615 (m), 1601 (s), 1571 (w), 1483 (m), 1465 (w),
1436 (m), 1384 (m), 1339 (m), 1302 (w), 1280 (w), 1261 (m), 1191 [IϾ2σ(I)].
Eur. J. Inorg. Chem. 2015, 2939–2947
2945
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
Ru(CO)2(PPh3)2(O2CPh)2 (3e): Ru(CO)3(PPh3)2 (1, 400 mg, and 13C{1H} NMR spectra of all catalysts and β-oxopropyl esters
564 μmol) was treated with benzoic acid (275 mg, 2.25 mmol) by
the synthesis procedure described above to afford 3e, yield 273 mg
(295 μmol, 52% based on 1), m.p. 253 °C. H NMR (CDCl3): δ =
synthesized in this work.
CCDC-1049384 (for 3a), -1049385 (for 3b), 1049387 (for 3c),
1049386 (for 3d) and 1048846 (for 4a) contain the supplementary
crystallographic data for this paper. These data can be obtained
free of charge from The Cambridge Crystallographic Data Centre
via www.ccdc.cam.ac.uk/data_request/cif.
1
7.09–7.14 (m, 4 H, m-CH/O2CPh), 7.19–7.26 (m, 20 H, m-CH/
PPh3, p-CH/PPh3, p-CH/O2CPh), 7.46–7.49 (m, 4 H, o-CH/
O2CPh), 7.70–7.76 (m, 12 H, o-CH/PPh3) ppm. 13C{1H} NMR
(CDCl3): δ = 126.8 (s, m-CH/O2CPh), 128.4 (t, JC,P = 5.1 Hz, m-
CH/PPh3), 129.3 (s, o-CH/O2CPh), 129.6 (s, p-CH/O2CPh), 129.9
(t, JC,P = 23.8 Hz, i-C/PPh3), 130.5 (s, p-CH/PPh3), 134.2 (t, JC,P
= 5.7 Hz, o-CH/PPh3), 136.0 (s, i-C/O2CPh), 172.1 (s, O2CPh),
197.0 (t, JC,P = 11.3 Hz, CO) ppm. 31P{1H} NMR (CDCl3): δ =
Acknowledgments
The Fonds der Chemischen Industrie is thanked for generous finan-
cial support and Chemiefond fellowships to J. J. and M. K.
31.2 (s) ppm. IR data (KBr): ν = 3080 (w), 3060 (w), 3051 (w),
˜
3025 (w), 2047 (vs), 1986 (vs), 1622 (w), 1610 (m), 1572 (w), 1486
(w), 1435 (m), 1351 (m), 1340 (s), 1191 (w), 1172 (w), 1140 (w),
1098 (m), 1025 (w), 744 (m), 729 (w), 720 (m), 710 (m), 693 (m),
[1] C. Darcel, C. Bruneau, P. H. Dixneuf, G. Neef, J. Chem. Soc.,
Chem. Commun. 1994, 333–334.
596 (m) cm–1. HRMS (ESI-TOF): calcd. for C45H35O4P2Ru [2] D. L. Blithe, L. K. Nieman, R. P. Blye, P. Stratton, M. Passaro,
803.1061 [M – O2CPh]+; found 803.1031. C52H40O6P2Ru (923.90):
calcd. C 67.60, H 4.36; found C 67.82, H 4.48.
Steroids 2003, 68, 1013–1017.
[3] G. Scheid, W. Kuit, E. Ruijter, R. V. A. Orru, E. Henke, U.
Bornscheurer, L. A. Wessjohann, Eur. J. Org. Chem. 2004,
1063–1074.
Ru(CO)4(PPh3)2(O2CCH2OCH3)2 (4a): The yellow title complex
4a was obtained as a side product during the synthesis of com-
pound 3a in variable amounts, depending on the solvent used (espe-
cially in toluene), m.p. 225 °C. 1H NMR (CDCl3): δ = 2.91 (s, 6 H,
OCH3), 3.60 (s, 4 H, CH2), 7.35–7.44 (m, 18 H, m-CH, p-CH),
7.52–7.58 (m, 12 H, o-CH) ppm. 13C{1H} NMR (CDCl3): δ = 58.7
(s, OCH3), 71.6 (s, CH2), 128.4 (t, JC,P = 4.6 Hz, m-CH), 129.9 (s,
p-CH), 133.3 (t, JC,P = 16.1 Hz, i-C), 134.0 (t, JC,P = 5.9 Hz, o-
CH), 185.1 (t, JC,P = 7.2 Hz, O2C), 205.1 (t, JC,P = 4.1 Hz,
CO) ppm. 31P{1H} NMR (CDCl3): δ = 13.7 (s) ppm. IR data
[4] H.-L. Lu, Z.-W. Wu, S.-Y. Song, X.-D. Liao, Y. Zhu, Y.-S. Hu-
ang, Org. Process Res. Dev. 2014, 18, 431–436.
[5] B. M. Trost, H. Urabe, J. Org. Chem. 1990, 55, 3982–3982.
[6] D. Erdmann, K. Schuehrer, W. Koch, G. Koch, German
patent, DE 2116416, 1976.
[7] B. Schickmous, J. Christoffers, Eur. J. Org. Chem. 2014, 4410–
4416.
[8] V. Stoeck, W. Schunack, Arch. Pharm. 1976, 309, 421–425.
[9] R. Schwyzer, B. Iselin, M. Feurer, Helv. Chim. Acta 1955, 38,
69–79.
[10] C. Bruneau, M. Neveux, Z. Kabouche, C. Ruppin, P. H.
Dixneuf, Synlett 1991, 11, 755–763.
[11] P. Yates, R. S. Grewal, P. C. Hayes, J. F. Sawyer, Can. J. Chem.
1988, 66, 2805–2815.
[12] A. S. Demir, N. Camkerten, H. Akgun, C. Tanyeli, A. S. Mah-
asneh, D. S. Watt, Synth. Commun. 1990, 20, 2279–2289.
[13] M. A. Ashraf, M. A. Jones, N. E. Kelly, A. Mullaney, J. S.
Snaith, I. Williams, Tetrahedron Lett. 2003, 44, 3151–3154.
[14] M. G. Kulkarni, S. M. Bagale, M. P. Shinde, D. D. Gaikwad,
A. S. Borhade, A. P. Dhondge, S. W. Chavhan, Y. B. Shaikh,
V. B. Ningdale, M. P. Desai, D. R. Birhade, Tetrahedron Lett.
2009, 50, 2893–2894.
(KBr): ν = 3058 (w), 2990 (w), 2923 (w), 2819 (w), 2819 (w), 2023
˜
(vs), 1976 (s), 1945 (vs), 1593 (s), 1480 (m), 1433 (s), 1414 (m), 1333
(m), 1184 (m), 1134 (m), 1107 (w), 1093 (m), 1028 (w), 998 (w),
950 (w), 921 (w), 751 (m), 706 (m), 695 (m) cm–1. HRMS (ESI-
TOF): calcd. for C46H40O10P2Ru2 1018.0182 [M]+; found
1018.0146. C46H40O10P2Ru2 (1016.90): calcd. C 54.33, H 3.96;
found C 54.34, H 3.97.
¯
Crystal Data for 4a: C46H40O10P2Ru2, Mr = 1016.86, triclinic, P1,
λ = 0.71073 Å, a = 9.893(3) Å, b = 10.434(2) Å, c = 21.187(6) Å, α
= 83.90(2)°, β = 80.57(3)°, γ = 80.35(2)°, V = 2119.9(11) Å3, Z =
2, ρcalcd. = 1.593 gcm–3, μ = 0.847 mm–1, T = 110 K, θ range 2.93–
25.00°, 15138 reflections collected, 7449 independent reflections
(Rint = 0.0298), R1 = 0.0330, wR2 = 0.0655 [IϾ2σ(I)].
[15] T. Mitsudo, Y. Hori, Y. Yamakawa, Y. Watanabe, J. Org. Chem.
1987, 52, 2230–2239.
[16] V. Cadierno, J. Francos, J. Gimeno, Green Chem. 2010, 12, 135–
143.
Typical Procedure for the Catalytic Reactions: In a screw-capped
vial, carboxylic acid (2.0 mmol), propargylic alcohol (3.0 mmol),
acenaphthene (1.0 mmol), Na2CO3 (0.08 mmol), and the catalyst
(0.02 mmol, 1.0 mol-%) were dissolved in toluene (20 mL). The
sealed vial was immersed in a heating mantle preheated to 90 °C.
After 24 h of stirring, all volatiles were evaporated under reduced
pressure. The yields were determined by 1H NMR spectroscopy
with use of acenaphthene as internal standard. Analytically pure
products were isolated by column chromatography on silica gel.
[17] D. Devanne, C. Ruppin, P. H. Dixneuf, J. Org. Chem. 1988, 53,
925–926.
[18] C. Bruneau, Z. Kabouche, M. Neveux, B. Seiller, P. H.
Dixneuf, Inorg. Chim. Acta 1994, 222, 155–163.
[19] S. Costin, N. P. Rath, E. B. Bauer, Adv. Synth. Catal. 2008, 350,
2414–2424.
[20] V. Cadierno, J. Francos, J. Gimeno, Organometallics 2011, 30,
852–862.
[21] B. Milde, T. Rüffer, H. Lang, Inorg. Chim. Acta 2012, 387, 338–
345.
[22] a) M. Czaun, A. Goeppert, J. Kothandaraman, R. B. May, R.
Haiges, G. K. S. Prakash, G. A. Olah, ACS Catal. 2014, 4, 311–
320; b) F. Micoli, L. Salvi, A. Salvini, P. Frediani, C. Giannelli,
J. Organomet. Chem. 2005, 690, 4867–4877.
[23] a) J. P. Johnpeter, L. Plasseraud, F. Schmitt, L. Juillerat-
Jeanneret, B. Therrien, J. Coord. Chem. 2013, 66, 1753–1762;
b) S. P. Oh, B. Y. Chor, W. Y. Fan, Y. Li, W. K. Leong, Organo-
metallics 2011, 30, 6774–6777; c) J. P. Johnpeter, B. Therrien,
Zh. Strukt. Khim. 2011, 52, 155; d) M. Auzias, J. Mattsson, B.
Therrien, G. Süss-Fink, Z. Anorg. Allg. Chem. 2009, 635, 115–
119; e) D. J. Morris, G. J. Clarkson, M. Wills, Organometallics
2009, 28, 4133–4140; f) M. Auzias, B. Therrien, G. Labat, H.
Single-Crystal X-ray Diffraction Analysis: The data were collected
with an Oxford Gemini S diffractometer and use of graphite-mono-
chromated Mo Kα radiation (λ = 0.71073 Å) at 110 K. The struc-
tures were solved by direct methods and refined by full-matrix
least-squares procedures on F2.[49–51] All non-hydrogen atoms were
refined anisotropically, and a riding model was employed in the
refinement of the hydrogen atom positions. Graphics of the molec-
ular structures have been created by using ORTEP.[52]
Supporting Information (see footnote on the first page of this arti-
cle): Characterization data for the β-oxopropyl esters as well as 1H
Eur. J. Inorg. Chem. 2015, 2939–2947
2946
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
Stoeckli-Evans, G. Süss-Fink, Inorg. Chim. Acta 2006, 359,
1012–1017; g) M. Auzias, B. Therrien, G. Süss-Fink, Inorg.
Chim. Acta 2006, 359, 3412–3416; h) R. Gavino, S. Pellegrini,
Y. Castanet, A. Mortreux, Acta Crystallogr., Sect. E 2005, 61,
m1762–m1764; i) K.-B. Shiu, S.-M. Peng, M.-C. Cheng, J. Or-
ganomet. Chem. 1993, 452, 143–149.
[40]
[41]
The central plane just includes Ru1 and the two bonded carb-
oxylate group oxygen atoms, due to a slight deviation of one
carbonyl group resulting in too high rms deviations.
The median/mean of Ru–Ru bond lengths in sawhorse-type di-
ruthenium tetracarbonyl complexes with triphenylphosphine li-
gands is 2.725/2.728 Å with a minimum of 2.711 Å (QOP-
GEI[23d]) and a maximum of 2.748 Å (DEXCEQ;[23a] 14 obser-
vations) according to entries in the CSD database.[53]
[24] J. H. Nelson, Concepts Magn. Reson. 2002, 14, 19–78.
[25] H. G. Metzinger, Org. Magn. Reson. 1971, 3, 485–494.
[26] R. K. Harris, Can. J. Chem. 1964, 42, 2275–2281.
[27] E. W. Abel, Q. Rev. Chem. Soc. 1963, 17, 133–159.
[28] B. F. G. Johnson, R. D. Johnston, J. Lewis, I. G. Williams, J.
Chem. Soc. A 1971, 689–691.
[29] G. R. Crooks, B. F. G. Johnson, J. Lewis, I. G. Williams, G.
Gamlen, J. Chem. Soc. A 1969, 2761–2766.
[30] H. Schumann, J. Opitz, J. Pickardt, J. Organomet. Chem. 1977,
128, 253–264.
[42]
[43]
The median/mean of Ru–P bond angles in sawhorse-type di-
ruthenium tetracarbonyl complexes with triphenylphosphine li-
gands is 166.413/166.410° with
a minimum of 161.37°
(KEHZIH[23g]) and a maximum of 171.828° (QOPGEI;[23d] 28
observations) according to entries in the CSD database.[53]
The median/mean of Ru–P bond lengths in sawhorse-type di-
ruthenium tetracarbonyl complexes with triphenylphosphine li-
gands is 2.446/2.443 Å with a minimum of 2.425 Å (QOP-
[31] B. Therrien, G. Süss-Fink, Coord. Chem. Rev. 2009, 253, 2639–
FOR[23d]
) and a
maximum of 2.457 Å (QUATMII,[23h]
2664.
RAGVAY;[23c] 28 observations) according to entries in the CSD
database.[53]
[32] See, for example: a) S. D. Robinson, M. F. Uttley, J. Chem.
Soc., Dalton Trans. 1973, 1912–1920; b) N. W. Alcock, V. M.
Tracy, T. C. Waddington, J. Chem. Soc., Dalton Trans. 1976,
2243–2246.
[33] See, for example: a) D. Rose, J. D. Gilbert, R. P. Richardson,
G. Wilkinson, J. Chem. Soc. A 1969, 2610–2615; b) W. K.
Dean, G. L. Simon, P. M. Treichel, L. F. Dahl, J. Organomet.
Chem. 1973, 50, 193–207.
[44]
A CSD database search reports 91 examples of dinuclear cis-
bridged compounds. The minimum and maximum values are
given in: J. S. Field, R. J. Haines, C. J. Parry, J. Chem. Soc.,
Dalton Trans. 1997, 2843–2848; K. Panneerselvam, T.-H. Lu,
C.-H. Huang, S.-F. Tung, K.-B. Shiu, Acta Crystallogr., Sect.
C 1997, 53, 1782–1784.
[34] See, for example: T. A. Stephenson, S. M. Morehouse, A. R.
Powell, J. P. Heffer, G. Wilkinson, J. Chem. Soc. 1965, 3632–
3640.
[45]
[46]
L. J. Gooßen, J. Paetzold, D. Koley, Chem. Commun. 2003,
706–707.
L. J. Gooßen, N. Rodríguez, K. Gooßen, Angew. Chem. Int.
Ed. 2008, 47, 3100–3120; Angew. Chem. 2008, 120, 3144–3164.
[35] G. B. Deacon, R. J. Phillips, Coord. Chem. Rev. 1980, 33, 227–
250.
[47] N. Ahmad, S. D. Robinson, M. F. Uttley, J. Chem. Soc., Dalton
Trans. 1972, 843–847.
[48] N. Ahmad, J. J. Levison, S. D. Robinson, M. F. Uttley, Inorg.
Synth. 1974, 15, 50.
[49] G. M. Sheldrick, Acta Crystallogr., Sect. A 1990, 46, 467–473.
[50] G. M. Sheldrick, Acta Crystallogr., Sect. A 2008, 64, 112–122.
[51] G. M. Sheldrick, SHELXL-2013, Program for Crystal Struc-
ture Refinement, University of Göttingen, Germany, 2013.
[52] L. J. Farrugia, J. Appl. Crystallogr. 2012, 45, 849–854.
[53] ConQuest, version 1.16, Cambridge Crystallographic Database
(CSD), 2013.
[36] M. Rotem, Z. Stein, Y. Shvo, J. Organomet. Chem. 1990, 387,
95–101.
[37] N. P. Hiett, J. M. Lynam, C. E. Welby, A. C. Whitwood, J. Or-
ganomet. Chem. 2011, 696, 378–387.
[38] D. L. Hughes, J. N. Wingfield, J. Chem. Soc., Dalton Trans.
1984, 739–742.
[39] The sum consists of 1.52 Å for oxygen and 2.02 Å for ruth-
enium, which is the smallest value reported in the literature.
For a comparison of different van der Waals radii reported for
ruthenium, see: S. Alvarez, Dalton Trans. 2013, 42, 8617–8636;
S.-Z. Hu, Z.-H. Zhou, B. E. Robertson, Z. Kristallogr. 2009,
224, 375–383; S. S. Batsanov, Inorg. Mater. 2009, 37, 871–885;
A. Bondi, J. Phys. Chem. 1964, 68, 441–451.
Received: February 25, 2015
Published Online: April 27, 2015
Eur. J. Inorg. Chem. 2015, 2939–2947
2947
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim