(Ethynylferrocenyl)phosphine ruthenium complexes in catalytic β-oxopropyl benzoate formation

Article abstract of DOI:10.1016/j.ica.2012.02.025

The synthesis of a series of (ethynylferrocenyl)phosphino ruthenium compounds of type (FcCC)R₂P(RuCl₂(η⁶-p- cymene)) (3a, R = C₆H₅; 3b, R = 2-CH₃C ₆H₄; 3c, R = ^c<

Full text of DOI:10.1016/j.ica.2012.02.025

Inorganica Chimica Acta 387 (2012) 338–345
Contents lists available at SciVerse ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
(Ethynylferrocenyl)phosphine ruthenium complexes in catalytic b-oxopropyl
benzoate formation
⇑
Bianca Milde, Tobias Rüffer, Heinrich Lang
Chemnitz University of Technology, Faculty of Science, Institute of Chemistry, Department of Inorganic Chemistry, Straße der Nationen 62, 09111 Chemnitz, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
The synthesis of a series of (ethynylferrocenyl)phosphino ruthenium compounds of type (FcC„C)R2-
P(RuCl2(
p-cymene = 1-iC3H7-4-CH3-C6H4; Fc = Fe(
Ph2(RuCl2(
6-p-cymene)))2 (10) resulting from the addition of ferrocenylphosphines P(C„CFc)R2 (1a–
1e) or P(C„CFc)(C„CPPh2)2 (9) to [RuCl2(
6-p-cymene)]2 (2) is described. The structures of 3b, 3c and
g ;
6-p-cymene)) (3a, R = C6H5; 3b, R = 2-CH3C6H4; 3c, R = cC4H3O; 3d, R = t-Bu; 3e, R = cC6H11
Received 28 November 2011
Received in revised form 8 February 2012
Accepted 15 February 2012
Available online 23 February 2012
g
5-C5H4)( 5-C5H5)) and (RuCl2( 6-p-cymene))(FcC„C)P(C„CP-
g g
g
g
10 in the solid state are reported confirming the expected tetrahedral coordination sphere about the
phosphorus atom as well as the ‘‘piano-stool’’ arrangement about the ruthenium atom(s). Ruthenium
complexes 3 and 10 are catalytically active under mild conditions in the alkyne-to-carboxylic acid cou-
pling as it was shown for the reaction of propargyl alcohol with benzoic acid. A comparison with litera-
Keywords:
(Ethynylferrocenyl)phosphine
b-Oxopropyl esters
Ruthenium
ture known [RuCl2(PR3)(g
6-p-cymene)] catalysts is presented.
Ferrocene
Ó 2012 Elsevier B.V. All rights reserved.
Catalysis
1. Introduction
even better yields are obtained using phosphines with strong p-
acceptor ability, e.g. P(cC4H3O)3 [9]. In addition, also water-soluble
[3,35] or on MCM-41-immobilized ruthenium(II) [19] catalysts
have been developed.
Since the pioneering work of Dixneuf and co-workers [1a], the
ruthenium-promoted synthesis of b-oxo esters developed rapidly
and is now-a-days an integral part in homogeneous catalysis,
applicable on bulky or functionalized carboxylic acids [2]. This
reaction is atom-economic, which provides an elegant alternative
to classical synthesis methodologies of b-oxo esters. Established
standard methods for the preparation of b-oxo esters include, for
example, the preparation from propargylic alcohols by hydration-
We here report on the synthesis of diverse (ethynylferroce-
nyl)phosphino ruthenium(II) complexes and their application in
the catalytic formation of b-oxopropyl benzoate to clarify the ques-
tion whether strong or weak electron donating groups at the phos-
phorus atom are responsible for the activity. Given the fact that PR3
groups are known to be highly sensitive toward oxygen we intro-
duced a ferrocenyl group for more stability and an additional
alkynyl functionality due to its electron-withdrawing nature. Also,
the preparation of a molecule featuring three ruthenium dichloro
p-cymene units is discussed to evaluate possible synergistic and
cooperative effects in the catalytic performance.
esterification steps [3–6] or the carboxylation of
[6–13]. In general, b-oxo esters are of versatile interest in organic
synthesis and industry, because they easily form -hydroxy ke-
a-halo ketones
a
tones which are structural building blocks in, for example, the syn-
thesis of natural products [3–6], antibacterial compounds [14] and
intermediates for furanones and imidazoles, respectively [1a,14].
In addition, b-oxo esters can be used as photolabile protecting
groups for carboxylic acids [6,15] or even as activated esters for
peptide synthesis [14]. Early works on the catalytic addition of ter-
minal alkynes to carboxylic acids include the application of
[Ru3(CO)12] [1,16] or [RuCl3 Â 3H2O] [1a] as catalyst precursors.
In recent years, several phosphine-carrying catalysts have been
2. Experimental
2.1. General procedure and materials
All reactions were performed under an atmosphere of nitrogen
using standard Schlenk techniques. Diethyl ether and dichloro-
methane were dried over sodium/benzophenone and calcium
hydride, respectively, and purified by distillation. For filtrations
Celite (purified and annealed, Erg. B.6, Riedel de Haen) was used.
Column chromatographies were performed using silica with a
developed, for example, [RuCl2(
OPh) [1,17] or [Ru( O2CH)(CO)2(PPh3)]2 [1b,2], which show high
conversions and regioselectivity under mild reaction conditions
g
6-p-cymene)(PR3)] (R = Ph, Me,
l
using basic phosphines. In contrast, Goossen et al. reported that
particle size of 40–60
pounds Fc–C„C–PR2 (1a–1e) [20,21], HC„C–PPh2 (4) [22],
P(NEt2)Cl2 (5) [23] and [RuCl2(
6-p-cymene)]2 (7) [24] were
lm (230–400 mesh (ASTM), Becker). Com-
⇑
Corresponding author. Tel.: +49 (0)371 531 21210; fax: +49 (0)371 531 21219.
E-mail address: heinrich.lang@chemie.tu-chemnitz.de (H. Lang).
g
0020-1693/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2012.02.025

B. Milde et al. / Inorganica Chimica Acta 387 (2012) 338–345
339
3
synthesized according to published procedures. All other chemicals
were obtained from commercial suppliers and used without fur-
ther purification.
CDCl3, d): 17.5 (s, CH3), 21.8 (s, CH(CH3)2), 22.5 (d, JCP = 4.9 Hz,
2-CH3C6H4), 30.0 (s, CH(CH3)2), 62.7 (d, JCP = 3.4 Hz, Ci/C5H4),
3
69.9 (s, Cb/C5H4), 70.3 (s, C5H5), 72.1 (d, JCP = 0.6 Hz, C /C5H4),
4
a
2
2
The 1H NMR spectra were recorded with a Bruker Avance III 500
spectrometer working at 500.3 MHz. The 13C{1H} and 31P{1H} NMR
spectra were recorded at 125.7 MHz and 202.5 MHz, respectively.
Chemical shifts are reported in d units (parts per million) down-
field from tetramethylsilane with the solvent as reference signal
(1H NMR: standard internal CDCl3, d 7.26; 13C{1H} NMR: standard
internal CDCl3, d 77.16; 31P{1H} NMR: standard external rel. 85%
H3PO4, d 0.0 and P(OMe)3, d 139.0). High resolution mass spectra
were recorded with a Bruker Daltonik micrOTOF-QII spectrometer
(ESI-TOF). Elemental analyses were carried out with a Thermo
FlashAE 1112 series instrument. Melting points of analytical pure
samples were determined by a Gallenkamp MFB 595 010 M melt-
ing point apparatus. FT IR spectra were recorded with a Thermo
Nicolet IR 200 spectrometer using either KBr pellets or NaCl plates.
79.6 (C„C–P⁄)), 86.6 (d, JCP = 5.3 Hz, C6H4), 91.5 (d, JCP = 5.1 Hz,
C6H4), 95.2 (s, Ci/C6H4), 109.5 (d, JCP = 13.5 Hz, C„C–P), 109.6 (s,
2
Ci/C6H4), 125.7 (d, JCP = 12.5 Hz, 2-CH3C6H4), 130.4 (d, JCP = 16.7 Hz,
2-CH3C6H4), 130.9 (d, JCP = 2.4 Hz, 2-CH3C6H4), 131.8 (d,
JCP = 7.9 Hz, 2-CH3C6H4), 135.4 (m, 2-CH3C6H4), 141.9 (d,
JCP = 5.9 Hz, 2-CH3C6H4). 31P{1H} NMR (202.53 MHz, CDCl3, d):
À9.1. HRMS (ESI-TOF) C36H37Cl2FePRu [M]+ m/z: calcd.: 728.0403,
found: 728.0413; [MÀCl]+ m/z: calcd.: 693.0717, found:
⁄)
693.0709. Signal concealed by CDCl3.
2.2.3. Synthesis of (FcC„C)(cC4H3O)2P(RuCl2( 6-p-cymene)) (3c)
g
0.5 g (1.34 mmol) of 1c were reacted with 0.41 g (0.67 mmol) of
2. After appropriate work-up, 3c was isolated as an orange solid.
Yield: 0.88 g (1.26 mmol, 94% based on 2). Anal. Calc. for
C
30H29Cl2FeO2PRu  1/5 CH2Cl2 (697.33 g/mol): C, 52.02; H, 4.25.
2.2. General procedure for the synthesis of ruthenium complexes 3a–
3e and 10
Found: C, 52.03; H, 4.49%. Mp.: 175 °C. IR (KBr, m
/cmÀ1): 1007 (s,
C–O), 1458 (w, P–C), 2159 (s, C„C). 1H NMR (500.30 MHz, CDCl3,
d): 1.18 (d, JHH = 6.9 Hz, 6H, CH(CH3)2), 2.06 (s, 3H, CH3), 2.91
3
3
0.5 g of 1 or 9 and 0.5 or 1.5 equiv. of [RuCl2(
g
6-p-cymene)]2 (2)
(sept, JHH = 6.9 Hz, 1H, CH(CH3)2), 4.27 (s, 5H, C5H5), 4.31 (pt,
3
were dissolved in 40 mL of dry dichloromethane. The solution was
stirred for 2 h at ambient temperature. Afterwards, the solvent was
removed in vacuum and the residue was washed 5–6 times with
5 mL portions of diethyl ether. After drying in vacuum the appro-
priate complexes were obtained as orange solids.
3JHH = 1.9 Hz, 2H, C5H4), 4.58 (pt, JHH = 1.9 Hz, 2H, C5H4), 5.30 (s,
CH2Cl2), 5.49–5.51 (m, 2H, C6H4), 5.54–5.57 (m, 2H, C6H4), 6.48
4
3
3
(dt, JHP = 1.6 Hz, JHH = 3.4 Hz, JHH = 1.6 Hz, 2H, H4/C4H3O), 7.21
(m, 2H, H3/C4H3O), 7.68 (m, 2H, H5/C4H3O). 13C{1H} NMR
(125.81 MHz, CDCl3, d): 17.8 (s, CH3), 22.0 (s, CH(CH3)2), 30.4 (s,
3
CH(CH3)2), 53.53 (s, CH2Cl2), 61.7 (d, JCP = 3.7 Hz, Ci/C5H4), 70.1
4
2.2.1. Synthesis of (FcC„C)(C6H5)2P(RuCl2(
g
6-p-cymene)) (3a)
(s, Cb/C5H4), 70.7 (s, C5H5), 72.5 (d, JCP = 0.8 Hz, C /C5H4), 74.0 (d,
a
2
Following the synthesis methodology described above, 0.5 g
(1.27 mmol) of 1a were reacted with 0.39 g (0.63 mmol) of 2. After
appropriate work-up, 3a was isolated as an air stable orange solid.
Yield: 0.88 g (1.22 mmol, 97% based on 2). Anal. Calc. for
1JCP = 112.6 Hz, C„C–P), 86.9 (d, JCP = 6.7 Hz, C6H4), 90.9 (d,
2JCP = 5.4 Hz, C6H4), 96.7 (s, Ci/C6H4), 109.0 (d, JCP = 18.1 Hz,
2
C„C–P), 109.5 (s, Ci/C6H4), 111.6 (d, JCP = 7.6 Hz, C4/C4H3O),
3
123.0 (d, JCP = 17.7 Hz, C3/C4H3O), 144.5 (d, JCP = 81.0 Hz, C2/
C4H3O), 147.4 (d, 4JCP = 5.6 Hz, C5/C4H3O). 31P{1H} NMR
(202.53 MHz, CDCl3, d): À26.0. HRMS (ESI-TOF) C30H29Cl2FeO2PRu
[M]+ m/z: calcd.: 679.9674, found: 679.9673.
2
1
C
34H33Cl2FePRu  1/4 CH2Cl2 (721.65 g/mol): C, 57.00; H, 4.68.
Found: C, 56.96; H, 4.66%. Mp.: 200 °C (dec.). IR (KBr,
m
/cmÀ1):
1436 (m, P–C), 2153 (m, C„C). 1H NMR (500.30 MHz, CDCl3, d):
3
1.23 (d, JHH = 6.9 Hz, 6H, CH(CH3)2), 2.00 (s, 3H, CH3), 2.95 (sept,
3JHH = 7.0 Hz, 1H, CH(CH3)2), 4.30 (s, 5H, C5H5), 4.37 (pt,
2.2.4. Synthesis of (FcC„C)(tBu)2P(RuCl2( 6-p-cymene)) (3d)
g
3JHH = 1.9 Hz, 2H, C5H4), 4.62 (pt, JHH = 1.9 Hz, 2H, C5H4), 5.23–
Reaction of 0.5 g (1.41 mmol) of 1d with 0.42 g (0.69 mmol) of 2
gave, after appropriate work-up, complex 3d which was isolated as
an air stable orange solid. Yield: 0.87 g (1.28 mmol, 93% based on
2). Anal. Calc. for C30H41Cl2FePRu  1/4 CH2Cl2 (681.68 g/mol): C,
53.30; H, 6.14. Found: C, 53.26; H, 6.22%. Mp.: 151 °C (dec.). IR
3
5.26 (m, 2H, C6H4), 5.30 (s, CH2Cl2), 5.31–5.33 (m, 2H, C6H4),
7.33–7.39 (m, 6H, Hm,p/C6H5), 8.01–8.09 (m, 4H, Ho/C6H5). 13C{1H}
NMR (125.81 MHz, CDCl3, d): 17.7 (s, CH3), 22.2 (s, CH(CH3)2),
3
30.5 (s, CH(CH3)2), 53.57 (s, CH2Cl2), 62.1 (d, JCP = 3.0 Hz, Ci/
4
a
C5H4), 70.1 (s, Cb/C5H4), 70.4 (s, C5H5), 72.4 (d, JCP = 1.0 Hz, C /
(KBr,
m
/cmÀ1): 1467 (w, P–C), 2158 (s, C„C), 2959 (s, C–H). 1H
1
2
3
C5H4), 78.3 (d, JCP = 53.3 Hz, C„C–P), 86.8 (d, JCP = 6.2 Hz, C6H4),
NMR (500.30 MHz, CDCl3, d): 1.30 (d, JHH = 6.9 Hz, 6H, CH(CH3)2),
90.6 (d, JCP = 4.3 Hz, C6H4), 96.0 (s, Ci/C6H4), 109.6 (s, Ci/C6H4),
1.51 (d, JHP = 14.8 Hz, 18H, C(CH3)3), 2.15 (s, 3H, CH3), 3.12 (sept,
2
3
110.4 (d, JCP = 13.4 Hz, C„C–P), 128.1 (d, JCP = 10.8 Hz, Cm/
3JHH = 6.5 Hz, 1H, CH(CH3)2), 4.31 (s, 5H, C5H5), 4.37 (m, 2H,
C5H4), 4.57 (m, 2H, C5H4), 5.30 (s, CH2Cl2), 5.40–5.48 (m, 4H,
C6H4). 13C{1H} NMR (125.81 MHz, CDCl3, d): 17.7 (s, CH3), 22.2 (s,
2
3
4
1
C6H5), 130.4 (d, JCP = 2.7 Hz, Cp/C6H5), 132.7 (d, JCP = 54.2 Hz, Ci/
C6H5), 133.3 (d, 2JCP = 10.3 Hz, Co/C6H5). 31P{1H} NMR
(202.53 MHz, CDCl3, d): À3.3. HRMS (ESI-TOF) C34H33Cl2FePRu
[M+nK]+ m/z: calcd.: 740.9717, found: 740.9639; [MÀCl]+ m/z:
calcd.: 665.0404, found: 665.0448.
2
CH(CH3)2), 29.5 (s, CH(CH3)2), 30.6 (d, JCP = 3.5 Hz, C(CH3)3), 39.3
1
3
(d, JCP = 14.8 Hz, C(CH3)3), 53.52 (s, CH2Cl2), 62.8 (d, JCP = 2.3 Hz,
Ci/C5H4), 69.9 (s, C5H4), 70.0 (s, C5H5), 72.0 (s, C5H4), 80.8 (d,
1JCP = 33.0 Hz, C„C–P), 89.2 (d, JCP = 5.0 Hz, C6H4), 89.3 (d,
2
2.2.2. Synthesis of (FcC„C)(2-CH3C6H4)2P(RuCl2(
g
6-p-cymene)) (3b)
2JCP = 4.6 Hz, C6H4), 97.2 (s, C6H4), 106.5 (s, C6H4), 108.1 (d,
2JCP = 2.2 Hz, C„C–P). 31P{1H} NMR (202.53 MHz, CDCl3, d): 26.7.
HRMS (ESI-TOF) C30H41Cl2FePRu [MÀCl]+ m/z: calcd.: 625.1029,
0.5 g (1.18 mmol) of 1b were reacted with 0.36 g (0.59 mmol) of
2. After appropriate work-up, complex 3b was isolated as orange
solid. Yield: 0.71 g (0.97 mmol, 82% based on 2). Anal. Calc. for
found: 625.0936; [M–(
found: 354.1188.
g
6-p-cymen)RuCl2]+ m/z: calcd.: 354.1194,
C
36H37Cl2FePRu (728.47 g/mol): C, 59.35; H, 5.12. Found: C,
59.42; H, 5.13%. Mp.: 195 °C. IR (KBr,
m
/cmÀ1): 1468 (m, P–C),
2150 (s, C„C). 1H NMR (500.30 MHz, CDCl3, d): 1.08 (d,
3JHH = 7.0 Hz, 6H, CH(CH3)2), 2.10 (s, 3H, CH3), 2.16 (s, 6H, 2-
2.2.5. Synthesis of (FcC„C)(cC6H11)2P(RuCl2( 6-p-cymene)) (3e)
g
Reaction of 0.5 g (1.23 mmol) of 1e with 0.38 g (0.62 mmol) of 2
gave, after appropriate work-up, 3e which was isolated as an or-
ange solid. Yield: 0.86 g (1.17 mmol, 94% based on 2). Anal. Calc.
3
CH3C6H4), 2.86 (sept, JHH = 7.0 Hz, 1H, CH(CH3)2), 4.18 (s, 5H,
3
3
C5H5), 4.30 (pt, JHH = 1.9 Hz, 2H, C5H4), 4.52 (pt, JHH = 1.9 Hz, 2H,
C5H4), 5.04–5.07 (m, 2H, C6H4), 5.31–5.34 (m, 2H, C6H4), 7.09–
7.14 (m, 2H, Hp/2-CH3C6H4), 7.29–7.35 (m, 4H, Hm/2-CH3C6H4),
8.43–8.52 (m, 2H, Ho/2-CH3C6H4). 13C{1H} NMR (125.81 MHz,
for
6.25. Found: C, 56.37; H, 6.49%. Mp.: 201 °C. IR (KBr,
1447 (m, P–C), 2154 (m, C„C), 2924 (vs C–H). 1H NMR
C
34H45Cl2FePRu  1/4 CH2Cl2 (733.75 g/mol): C, 56.06; H,
m
/cmÀ1):

340
B. Milde et al. / Inorganica Chimica Acta 387 (2012) 338–345
(500.30 MHz, CDCl3, d): 1.25 (d, 3JHH = 6.9 Hz, 6H, CH(CH3)2), 1.22–
1.35 (m, 6H, C6H11), 1.54–1.69 (m, 6H, C6H11), 1.76–1.81 (m, 4H,
C6H11), 1.90–1.94 (m, 2H, C6H11), 2.01–2.03 (m, 2H, C6H11), 2.11
2.5. Synthesis of (RuCl2(g -
6-p-cymene))(FcC„C)P(C„CPPh2(RuCl2(g6
p-cymene)))2 (10)
3
(s, 3H, CH3), 2.47–2.53 (m, 2H, H1/C6H11), 3.02 (sept, JHH = 6.9 Hz,
0.5 g (0.76 mmol) of 9 were reacted with 0.70 g (1.14 mmol) of
2. After appropriate work-up (Section 2.2), 10 was isolated as an
orange solid. Yield: 1.18 g (0.75 mmol, 99% based on 9). Anal. Calc.
for C70H71Cl6FeP3Ru3 (1577.01 g/mol): C, 53.31; H, 4.54. Found: C,
CH(CH3)2), 4.28 (s, 5H, C5H5), 4.36 (pt, 3JHH = 1.8 Hz, 2H, C5H4), 4.57
3
(pt, JHH = 1.8 Hz, 2H, C5H4), 5.29 (s, CH2Cl2), 5.39–5.42 (m, 4H,
C6H4). 13C{1H} NMR (125.81 MHz, CDCl3, d): 17.7 (s, CH3), 22.2 (s,
CH(CH3)2), 26.2 (s, C6H11), 26.9 (d, JCP = 11.0 Hz, C6H11), 27.3 (d,
JCP = 13.3 Hz, C6H11), 27.9 (d, JCP = 4.0 Hz, C6H11), 28.7 (m, C6H11),
53.05; H, 4.45%. Mp.: 150 °C. IR (NaCl,
2155 (m, C„C), 2179 (m, C„C). 1H NMR (250.130 MHz, CDCl3,
m
/cmÀ1): 1435 (m, P–C),
1
3
3
29.9 (s, CH(CH3)2), 34.3 (d, JCP = 26.4 Hz, C1/C6H11), 53.55 (s,
d): 0.98 (d, JHH = 4.5 Hz, 6H, CH(CH3)2), 1.00 (d, JHH = 4.5 Hz, 6H,
3
3
CH2Cl2), 62.8 (d, JCP = 2.7 Hz, Ci/C5H4), 69.9 (s, C5H4), 70.3 (s,
CH(CH3)2), 1.23 (d, JHH = 6.0 Hz, 6H, CH(CH3)2), 1.81 (s, 6H, CH3),
1
C5H5), 72.3 (s, C5H4), 79.0 (d, JCP = 65.1 Hz, C„C–P), 88.5 (d,
2.00 (s, 3H, CH3), 2.55 (sept, 3JHH = 6.9 Hz, 2H, CH(CH3)2), 2.88 (sept,
3JHH = 6.9 Hz, 1H, CH(CH3)2), 4.23 (s, 5H, C5H5), 4.36 (pt,
2
2JCP = 4.8 Hz, C6H4), 90.0 (d, JCP = 4.5 Hz, C6H4), 97.3 (s, C6H4),
2
3
105.1 (s, C6H4), 108.1 (d, JCP = 4.5 Hz, C„C–P). 31P{1H} NMR
3JHH = 1.9 Hz, C5H4), 4.58 (pt, JHH = 1.9 Hz, C5H4), 5.22–5.25 (m,
(202.53 MHz, CDCl3, d): 14.4. HRMS (ESI-TOF) C34H45Cl2FePRu [M
2H, C6H4), 5.39–5.49 (m, 6H, C6H4), 5.56–5.58 (m, 2H, C6H4),
5.64–5.67 (m, 2H, C6H4), 7.27–7.31 (m, 6H, Hm,p/C6H5), 7.35–7.38
(m, 6H, Hm,p/C6H5), 8.01–8.09 (m, 8H, Ho/C6H5). 13C{1H} NMR
(125.81 MHz, CDCl3, d): 17.6 (s, CH3), 18.1 (s, CH3), 21.7 (s,
CH(CH3)2), 22.2 (s, CH(CH3)2), 22.3 (s, CH(CH3)2), 30.4 (s, CH(CH3)2),
- Cl]+m/z: calcd.: 677.1343, found: 677.1249.
3
2.3. Synthesis of (Et2N)P(C„C–P(C6H5)2)2 (6)
30.8 (s, CH(CH3)2), 60.1 (d, JCP = 3.6 Hz, Ci/C5H4), 70.8 (s, Cb/C5H4),
a
1
70.9 (s, C5H5), 72.7 (m, C /C5H4), 73.5 (d, JCP = 127.4 Hz, FcC„CP),
2
2
Phosphine 6 was synthesized by a modified literature proce-
dure. [22] To 2.0 g (9.52 mmol) of 4 dissolved in 50 mL of dry
diethyl ether, 3.8 mL (9.5 mmol) of nBuLi were added dropwise at
À50 °C. After stirring the solution for 30 min at ambient tempera-
ture it was again cooled to À30 °C and 0.69 mL (826 mg,
4.75 mmol) of Cl2PNEt2 (5) were added dropwise. The reaction
mixture was stirred at ambient temperature for 1 h and then fil-
tered through a pad of Celite. The resulting solution was evapo-
rated to dryness and the product was obtained as brown viscous
oil in high purity. Phosphine 6 was used without further purifica-
tion steps. Yield: 2.4 g (4.68 mmol, 97% based on 5). C32H30NP3
(521.51 g/mol). 31P{1H} NMR (101.249 MHz, CDCl3, d): À34.0 (d,
86.4 (d, JCP = 5.4 Hz, C6H4), 87.3 (d, JCP = 6.0 Hz, C6H4), 88.0 (d,
2JCP = 6.2 Hz, C6H4), 89.9 (d, JCP = 7.0 Hz, C6H4), 90.0 (d,
2
2JCP = 4.3 Hz, C6H4), 91.4 (d, JCP = 5.0 Hz, C6H4), 97.1 (s, Ci/C6H4),
2
97.6 (s, Ci/C6H4), 97.9 (s, Ci/C6H4), 102.0 (d, 1JCP = 9.1 Hz, Ph2PC„C),
102.5 (d, JCP = 8.7 Hz, Ph2PC„C), 109.0 (s, Ci/C6H4), 110.5 (d,
1
2JCP = 23.4 Hz, FcC„CP), 110.9 (s, Ci/C6H4), 128.3 (d, JCP = 10.9 Hz,
3
Cm/C6H5), 128.8 (d, 3JCP = 10.9 Hz, Cm/C6H5), 130.1 (d, 1JCP = 53.4 Hz,
Ci/C6H5), 130.8 (d, 4JCP = 2.3 Hz, Cp/C6H5), 130.9 (d, 4JCP = 2.2 Hz, Cp/
C6H5), 132.3 (d, JCP = 52.5 Hz, Ci/C6H5), 133.0 (d, JCP = 10.8 Hz, Co/
C6H5), 134.4 (d, 2JCP = 10.5 Hz, Co/C6H5). 31P{1H} NMR
(101.249 MHz, CDCl3, d): À44.8 (s, FcC„CP), À0.8 (s, Ph2PC„C).
HRMS (ESI-TOF) C70H71FeP3Ru3Cl6 [M]+ m/z: calcd.: 1578.9370,
1
2
3
3JPP = 2.7 Hz, Ph2P), À0.9 (t, JPP = 3.2 Hz, Et2NP).
found: 1578.9253; [M–RuCl2(
1270.9866, found: 1270.9751.
g
6-p-cymene)]+ m/z: calcd.:
2.6. General procedure for the catalytic reactions
2.4. Synthesis of P(C„CFc)(C„CPPh2)2 (9)
122 mg (1.0 mmol) of benzoic acid, 77 mg (0.5 mmol) of
acenaphthene (internal standard) and 1.0 mol% (based in Ru)
of the respective catalyst (3a–3e or 10) were dissolved in 15 mL
of chloro benzene. After addition of 0.87 mL (84 mg, 1.5 mmol) of
propargyl alcohol the reaction mixture was stirred at 80 °C and
samples (0.5 mL) were taken in periods of 1 h. The samples were
dried in vacuum and the conversions were determined by 1H
NMR spectroscopy.
To a solution of 2.4 g (4.68 mmol) of 6 in 50 mL of dry diethyl
ether, 10 mL of a 1.0 M solution of HCl (2 equiv.) in diethyl ether
were added slowly at ambient temperature. The resulting mixture
was stirred for 1 h and then added dropwise to a cooled solution
(À50 °C) of 8 in dry diethyl ether. Compound 8 was prepared by
dropwise addition of 1.7 mL (4.25 mmol) of nBuLi to a solution of
0.89 g (4.26 mmol) of ethynyl ferrocene in 30 mL of dry diethyl
ether. The resulting mixture was stirred for 1 h at ambient temper-
ature and was then concentrated in vacuum. The residue was puri-
fied by column chromatography on silica gel (column size:
4 Â 20 cm) using n-hexane as eluent. Phosphine 9 was obtained
as a red solid. Yield: 1.63 g (2.48 mmol, 58% based on nBuLi). Anal.
Calc. for C40H29FeP3 (658.42 g/mol): C, 72.97; H, 4.44. Found: C,
2.7. Crystal structure determination
The crystal and intensity collection data for 3b, 3c, and 10 are
summarized in Table 1. The data were collected with an Oxford
Gemini S diffractometer with graphite monochromatized Mo K
a
73.33; H, 4.62%. IR (NaCl,
m
/cmÀ1): 1434 (m, P–C), 2151 (s, C„C),
radiation (k = 0.71073 Å) at 100 K. The structures were solved by
direct methods using SHELXS-97 [25] and refined by full-matrix
least-square procedures on F2 using SHELXL-97 [26]. All non-hydro-
gen atoms were refined anisotropically and a riding model was
employed in the refinement of the hydrogen atom positions.
2175 (m, C„C). 1H NMR (250.130 MHz, CDCl3, d): 4.26 (s, 5H,
3
3
C5H5), 4.30 (pt, JHH = 1.9 Hz, 2H, C5H4), 4.57 (pt, JHH = 1.9 Hz,
C5H4), 7.32–7.40 (m, 12H, Hm,p/C6H5), 7.63–7.71 (m, 8 H, Ho/
C6H5). 13C{1H} NMR (62.895 MHz, CDCl3, d): 62.8 (d, JCP = 0.5 Hz,
3
Ci/C5H4), 69.8 (s, Cb/C5H4), 70.4 (s, C5H5), 72.4 (d, JCP = 1.8 Hz, C /
4
a
1
4
C5H4), 74.1 (dpt, JCP = 13.4 Hz, JCP = 2.3 Hz, FcC„CP), 99.7 (dpt,
1JCP = 3.8 Hz, JCP = 2.3 Hz, Ph2PC„C), 105.6 (dd, JCP = 19.1 Hz,
3. Results and discussion
2
1
2JCP = 4.2 Hz, Ph2PC„C), 107.5 (d, JCP = 13.7 Hz, FcC„CP), 128.9
2
(d, JCP = 7.8 Hz, Cm/C6H5), 129.4 (s, Cp/C6H5), 132.9 (d, JCP
=
The (ethynylferrocenyl)phosphino ruthenium(II) complexes
3
2
20.6 Hz, Co/C6H5), 135.1 (d, JCP = 6.2 Hz, JCP = 1.0 Hz, Ci/C6H5).
(FcC„C)R2P(RuCl2(
H4; 3c, R = cC4H3O; 3d, R = t-Bu; 3e, R = cC6H11
1-iC3H7-4-CH3-C6H4; Fc = Fe( 5-C5H4)( 5-C5H5)) were synthesized
by treatment of P(C„CFc)R2 (1a–1e) [20] with 0.5 equiv. of dimeric
g
6-p-cymene)) (3a, R = C6H5; 3b, R = 2-CH3C6-
1
4
31P{1H} NMR (101.249 MHz, CDCl3, d): À88.2 (t, JPP = 4.8 Hz,
; p-cymene =
3
3
FcC„CP), À32.7 (d, JPP = 4.8 Hz, Ph2PC„C). HRMS (ESI-TOF)
g
g
C
40H29FeP3 [M]+ m/z: calcd.: 658.0827, found: 658.0778.

B. Milde et al. / Inorganica Chimica Acta 387 (2012) 338–345
341
Table 1
chlorophosphine 7 failed due to its high reactivity and hence it
was used in the synthesis of 9 without additional purification
(Section 2). Addition of 7 to LiC„CFc (8) in diethyl ether at low
temperature gave by concomitant precipitation of LiCl red
P(C„CFc)(C„CPPh2)2 (9) in moderate yield. Treatment of 9 with
Crystal and intensity collection data for 3b, 3c and 10.
3b
3c
10
Formula weight
Chemical formula
Crystal system
Space group
a (Å)
847.81
680.32
3681.84
C
37H38Cl5FePRu C30H29Cl2FeO2PRu C145H149Cl25Fe2P6Ru6
2 produced (RuCl2(
cymene))2 (10). In heterometallic Ru3Fe 10 the building blocks
FcC„C, C„CPPh2(RuCl2(
6-p-cymene) and RuCl2( 6-p-cymene)
give rise to coordination number 4 at phosphorus.
g g
6-p-cymene))(FcC„C)P(C„CPPh2(RuCl2( 6-p-
triclinic
triclinic
triclinic
ꢀ
ꢀ
ꢀ
P1
P1
P1
10.1745(3)
12.3193(4)
15.2195(4)
81.358
87.010
69.003
1760.76(9)
1.599
860
9.7028(12)
10.8658(6)
13.8545(7)
74.085
78.058
85.551
1373.9(2)
1.644
688
12.1915(5)
14.3215(5)
25.1595(10)
96.926
103.679
96.136
4195.2(3)
1.457
1850
0.2 Â 0.2 Â 0.1
g
g
b (Å)
c (Å)
Newly synthesized organometallic compounds 3, 9 and 10 have
been identified by elemental analysis, IR and NMR (1H, 13C{1H},
31P{1H}) spectroscopy. ESI TOF mass-spectrometry and single crys-
tal X-ray structure analysis of 3b, 3c and 10 were additionally car-
ried out. However, all complexes show the tendency to enclose
solvent molecules which, even after 2–3 days in vacuum, could
not be completely removed. These solvents include chloroform,
dichloromethane and diethyl ether, whereas dichloromethane is
able to displace the other solvents.
a
(°)
b (°)
c
(°)
V (Å3)
qcalc (g cmÀ3
F(000)
)
Crystal dimensions 0.3 Â 0.3 Â 0.3 0.2 Â 0.2 Â 0.1
(mm)
Z
2
2
1
Maximum and
minimum
transmission
Absorption
coefficient
1.00000,
0.95682
1.00000, 0.94896 1.00000, 0.93550
The IR spectra of 3, 9 and 10 show very characteristic absorp-
tions for the FcC„C and Ph2PC„C alkynyl units in the expected re-
1.293
1.357
1.192
gion, i.e. at 2151 cmÀ1
2155 cmÀ1
of the phosphorus atom in 1 and 9 to the 16-valence electron com-
plex fragment RuCl2(
6-p-cymene) (formation of 3 and 10) a small
(mC„CPPh2) and 2175 (
mC„CFc) for 9 as well as
(k, mmÀ1
Scan range (°)
)
(m
C„CPPh2) and 2179 ( C„CFc) for 10. Upon coordination
m
3.03–25.00
À8 6 h 6 12
À14 6 k 6 14
À18 6 l 6 18
14838
2.78–25.00
À11 6 h 6 11
À12 6 k 6 12
À16 6 l 6 16
12978
4798
0.0234
4798/0/334
3.27–25.00
À14 6 h 6 14
À17 6 k 6 16
À29 6 l 6 29
33943
14712
0.0234
14712/66/868
Index ranges
g
shift of the ferrocenyl acetylide and the phosphine alkynyl stretch-
ing frequencies to higher wavenumbers is induced, whereby the
ferrocenyl-bonded C„C triple bond is more affected (i.e. 1c,
Total reflections
Unique reflections 6158
Rint
0.0234
6158/0/406
m
C„C = 2153 [20]; 3c, m
C„C = 2159 cmÀ1; Section 2). As conse-
Data/restraints/
para-meters
Goodness-of-fit
quence thereof, IR spectroscopy is suited to monitor the progress
1.043
1.107
1.059
of the reactions.
(GOF) on F2
In the 13C{1H} NMR spectra the alkynyl groups create two (3) or
four (9 and 10) exceptional resonance signals of which the phos-
phorus- (73–80 ppm, 1JPC = 13–127 Hz) and the ferrocenyl-bonded
R1a, wR2 [I 2
r
(I)] 0.0508, 0.1249 0.0200, 0.0508
0.0527, 0.1430
0.0616, 0.1494
2.737, À1.691
a
R1a, wR2 (all data) 0.0622, 0.1298 0.0252, 0.0526
a
Largest differences 1.821, À1.155 0.450, À0.391
2
in peak and
hole peak in
final Fourier
acetylide carbon atoms (107–110 ppm, JPC = 4–23 Hz) in 3a–3e
and 10 appear as doublets (Section 2). Complexation of the phos-
phorus atom in 1 and 9 to a RuCl2(g
6-p-cymene)-fragment induces
map (e ÅÀ3
)
a shift of the phosphorus atom signal to lower field, which is char-
acteristic in phosphorus transition metal chemistry [18]. While the
coupling patterns in 1, 3 and 10 are as expected (Section 2), com-
pound 9 possesses a more complex signal splitting which is attrib-
uted to the increased number of phosphorus atoms present.
Through the formation of a dative phosphorus–ruthenium bond
P
P
P
P
P
a
2
2
R1 = [ (||Fo| À |Fc|)/ |Fo|]; wR2 = [ (w(Fo À Fc2)2)/ (wFo4)]1/2; S = [ w(Fo
À
Fc2)2]/(n À p)1/2. n = number of reflections, p = parameters used.
[RuCl2( 6-p-cymene)]2 (2) in dichloromethane at ambient temper-
g
ature (Reaction 1). After appropriate work-up, compounds 3a–3e
could be isolated as orange solid materials which are stable
towards air and moisture for months. They dissolve in common or-
ganic solvents including dichloromethane, chloroform and tetrahy-
drofuran, while in diethyl ether, n-hexane and toluene they are not
soluble.
2
in 10 the JPC coupling constant diminishes and hence is not any-
more detectable in the spectrum. Also a shift of the p-cymene
carbon atoms is observed when going from non-complexed to
the coordinated species (Section 2). In 10 a set of two cymene units
in the ratio of 2:1 is visible due to their different chemical environ-
Cl Ru
Cl
C C PR2
C C
P
1/2 [RuCl2(η6-p-cymene)]2 (2)
R2
Fe
Fe
(1)
1
3a - e
Tetrametallic 10 was prepared applying the consecutive synthe-
sis sequence shown in Scheme 1 of which the first three steps
could be carried out in a one-pot procedure. Attempts to isolate
ments of which the signal of the inner phosphorus-bonded ruthe-
nium dichloro p-cymene moiety is found at lower magnetic field.
Notable in the spectrum of 10 is the observation of three signal sets

342
B. Milde et al. / Inorganica Chimica Acta 387 (2012) 338–345
PPh2
C
PPh2
PPh2
C
C
1) 2 nBuLi
C
C
C
C
HCl / Et2O
2
HC C PPh2
Et2N P
Cl
P
2) P(NEt2)Cl2
C
5
( )
PPh2
4
7
6
C CLi
(8)
PPh2
PPh2
Ru
C
C
Cl
Cl
Fe
C
C
C C P
C
C P
3/2 [RuCl2 (η6-p-cymene)]2 (2)
C
C
Ph
Ru
2 Cl
Fe
Fe
C
P
Cl
C
Ph2P
Ru
Cl
Cl
9
10
Scheme 1. Consecutive synthesis of tetrametallic 10.
for the iso-propyl groups which can be explained by hindered
rotation. The same behavior is found for the diphenylphosphino
building blocks (Section 2).
p-cymene moiety at phosphorus (Fig. 1). For 3b and 3c, respec-
tively, the bond angles around phosphorus P1 range from 108–
123 ° and those around Ru1 from 84–95° (Table 2) indicating a
characteristic ‘‘piano-stool’’ geometry (Fig. 1). The carbon–carbon
triple bond distances C11–C12 are 1.198(8) (3b) and 1.204(3)
(3c) Å, which is typical for this type of bonding [20,21,27]. The
P1–C11–C12 and C11–C12–C6 units are with 178.8(5)° and
178.1(6)° (3b) as well as 170.91(18)° and 177.5(2)° (3c) essen-
tially linear. The two ferrocenyl cyclopentadienyl rings in mole-
cule 3c are in a nearly eclipsed conformation (3.1°), whereas in
complex 3b both cyclopentadienyl rings are with 22.5° about in
the middle between the fully eclipsed (0°) and the fully stag-
gered (36°) conformation. The D1–Fe1 and D2–Fe1 separations
are between 1.634–1.675 Å (D1 = centroid of C5H5, D2 = centroid
of C5H4) and are similar to those of related compounds
[20,21,27].
As might have been expected, the 1H NMR spectra of 3, 9 and 10
consist of distinctive signal patterns as typical for the ferrocenyl
and p-cymene units, respectively (Section 2).
A more expressive method than IR spectroscopy to verify the
progress of the reaction is 31P{1H} NMR spectroscopy. A significant
shift to lower field is observed upon coordination of the phos-
phines 3 and 9 to ruthenium (i.e. 1c, À83.4 [20]; 3c, À26.0 ppm,
Section 2). Peculiar for 9 is the detection of a triplet at À88.2
3
(FcC„CP) and a doublet at -32.7 (C„CPPh2) with JPP = 4.8 Hz,
while in 10 this coupling diminishes.
The structures of 3b, 3c, and 10 in the solid state were deter-
mined by single X-ray diffraction studies confirming the half-sand-
wich configuration about the ruthenium(II) center and the
tetrahedral environment at phosphorus. Single crystals of 3b and
3c could be grown from a saturated chloroform solution at ambient
temperature, while crystals of 10 were accessible by slow diffusion
of n-hexane into a saturated dichloromethane-chloroform solution
The key structural data of 10 (Fig. 2) confirm the half-sandwich
structure about Ru1, Ru2 and Ru3. The coordination number
around P1–P3 is four and along with the appropriate bond lengths
and angles a ‘‘piano-stool’’ geometry is setup (Fig. 2). Mentionable
are the distances Ru1–P1 (2.2772(12) Å, Ru2–P2 (2.3309(13) Å)
and Ru3–P3 (2.3185(12) Å) proving, as expected, the stronger
(1:1, v:v) containing 10 at 25 °C. It was found that all molecules crys-
ꢀ
tallize in the triclinic space group P1. The molecular structures of 3b
and 3c are shown in Fig. 1, Fig. 2 displays tetrametallic 10. Geometric
details of 3b and 3c are listed in Table 2, while the ones of 10 are
summarized in the caption of Fig. 2. The crystallographic and refine-
ment data of all compounds can be found in Table 1 (Section 2). Bond
distances (Å), angles (°) and torsion angles (°) of the ethynylferroce-
nyl and p-cymene units are as expected and similar to those re-
ported for closely related organometallic compounds [19–21,27].
Compounds 3b and 3c are set-up by the ferrocenylethynyl
unit, the two organic groups R and the ruthenium dichloro
binding of the RuCl2(
lower -donor capability compared with the respective alkynyl
g
6-cymene) unit by P1 explainable by the
r
phosphine moieties [20,27]. All other bond distances and angles
agree well with those building blocks reported for similar
compounds [19–21,27].
The application of 3a–3e and 10 in the homogeneous ruthe-
nium-catalyzed addition of benzoic acid to propargyl alcohol for
the synthesis of b-oxopropyl benzoate was studied as model
system (Reaction 2).
O
CO2H
[Ru]
chloro benzene,
80 °C
OH
O
(2)
+
O
[Ru] = 3a - 3c, 10

B. Milde et al. / Inorganica Chimica Acta 387 (2012) 338–345
343
Fig. 1. ORTEP diagram (50% probability level) of the molecular structures of 3b (left) and 3c (right) with the atom numbering scheme. (Hydrogen atoms and chloroform as
packing solvent of 3b are omitted for clarity.)
Table 2
Selected bond lengths (Å), bond angles and torsion angles (°) for 3b and 3c.a
3b
3c
Ru1–P1
Ru1–Cl1
Ru1–Cl2
P1–C12
P1–C23
P1–C27
P1–C30
C11–C12
O1–C23
O1–C26
O2–C27
O2–C30
D1–Fe1b
D2–Fe1b
D3–Ru1b
2.3799(13)
2.4118(13)
2.4069(13)
1.764(6)
2.3035(5)
2.4137(5)
2.4121(6)
1.754(2)
1.790(2)
1.804(2)
1.831(5)
1.839(5)
1.198(8)
1.204(3)
1.383(2)
1.362(2)
1.368(2)
1.371(3)
1.640
1.634
1.675
1.706
1.646
1.696
Cl1–Ru1–Cl2
Cl1–Ru1–P1
Cl2–Ru1–P1
Ru1–P1–C12
Ru1–P1–C23
Ru1–P1–C27
Ru1–P1–C30
P1–C12–C11
C12–C11–C6
C26–O1–C23
C27–O2–C30
D1–Fe1–D2b
87.93(5)
84.06(5)
95.62(4)
108.98(17)
109.27(16)
88.29(2)
85.891(18)
91.442(19)
111.60(6)
119.87(6)
116.38(6)
123.96(18)
178.8(5)
178.1(6)
170.91(18)
177.5(2)
105.78(15)
106.50(16)
179.6
Fig. 2. ORTEP diagram (50% probability level) of the molecular structure of 10 with
the atom numbering scheme. (Hydrogen atoms, packing solvent molecules and the
phenyl groups (except the ipso-carbon atoms) are omitted for clarity.) Selected bond
distances (Å), angles (°) and torsion angles (°): Fe1–D1 = 1.661, Fe1–D2 = 1.630,
Ru1–D3 = 1.699, Ru2–D4 = 1.703, Ru3–D5 = 1.702, C11–C12 = 1.205(7), C23–
C24 = 1.196(7), C47–C48 = 1.205(7), Ru1–Cl1 = 2.3995(12), Ru1–Cl2 = 2.4095(13),
Ru1–P1 = 2.2772(12), Ru2–Cl3 = 2.4072(13), Ru2–Cl4 = 2.4171(13), Ru2–P2 =
2.3309(13), Ru3–Cl5 = 2.4077(12), Ru3–Cl6 = 2.4068(12), Ru3–P3 = 2.3185(12),
P1–C12 = 1.742(5), P1–C23 = 1.764(5), P1–C47 = 1.758(5), P2–C24 = 1.770(5), P2–
C25 = 1.831(5), P2–C31 = 1.822(5), P3–C48 = 1.763(5), P3–C49 = 1.829(5), P3–C55 =
1.826(5); D1–Fe1–D2 = 179.6, P1–C12–C11 = 170.1(5), P1–C23–C24 = 177.7(5),
P1–C47–C48 = 170.3(4), C6–C11–C12 = 177.8(6), P2–C24–C23 = 175.9(4), P3–C48–
C47 = 173.8(4); P1–C12–C11–C6 = -71(16), P1–C23–C24–P2 = -96(13), P1–C47–
C48–P3 = 73(5). Standard uncertainties of the last significant digit(s) are shown in
parenthesis. D1 = denotes the centroid of C5H5; D2 = denotes the centroid of C5H4,
D3–D5 = denotes the centroids of C6H4.
178.1
P1–C12–C11–C6
P1–C23–C24–C29
P1–C30–C35–C36
Ru1–P1–C12–C11
Ru1–P1–C23–O1
Ru1–P1–C27–O2
À115(24)
12.4(7)
À1.1(8)
21(6)
69.7(11)
À167.51(11)
68.11(16)
a
Standard uncertainties of the last significant digit(s) are shown in parenthesis.
D1 denotes the centroid of C5H5 at Fe1; D2 denotes the centroid of C5H4 at Fe1;
D3 denotes the centroid of C6H4 at Ru1.
b
phosphines can be quantified by measuring the 1J(31P–77Se) cou-
pling constant of the appropriate seleno phosphines. Results there-
of were previously published [20] indicating that phosphine 3c
Various ruthenium complexes featuring different electron-rich
or electron-poor (ferrocenylethynyl)phosphino entities were
screened in order to identify factors that may influence the
catalytic activity and productivity. The donor capacity of the
possesses the weakest
phines 3d, e show the best
After screening different solvents, temperatures and catalyst
concentrations we chose as best suited reaction conditions a
r
donor ability whereat the aliphatic phos-
donor ability [20].
r

344
B. Milde et al. / Inorganica Chimica Acta 387 (2012) 338–345
us (Fig. 3). We believe that it is a combination of both issues,
60
50
40
30
20
10
0
whereas electron-poor ligands at the phosphorus atom are best
suited, which is achieved by the introduction of an ethynylferroce-
nyl functionality. Moderate electron-rich species are only effective
catalysts, when they possess at the same time bulky ligands, e.g.
ortho-tolyl groups (Fig. 3). Also, some of our systems need longer
reaction times and therefore, suffer from deactivation due to
decomposition of the active species. Finally, it must be noted that,
however, our catalysts only work at 80 °C, which differs from liter-
ature known species described by, for example, Dixneuf et al. [1] or
Goossen et al. [9].
3c/10
3b
3e
3a
3d
Cversion
4. Conclusions
0
5
10
15
20
25
In this work, we presented the synthesis of novel heterobime-
tallic dinuclear and tetranuclear complexes of type (FcC„C)R2P-
t
/ h
(RuCl2(
p-cymene = 1-iC3H7-4-CH3-C6H4; Fc = Fe(
(RuCl2(
6-p-cymene))(FcC„C)-P(C„CPPh2(RuCl2(
g ;
6-p-cymene)) (R = C6H5, 2-CH3C6H4, cC4H3O, t-Bu, cC6H11
5-C5H4)(
g
5-C5H5)) and
Fig. 3. Reaction profiles for catalysts 3a–3e and 10 for the reaction of benzoic acid
with propargyl alcohol to give b-oxopropyl benzoate (Reaction 2, catalyst loading
1.0 mol% based on Ru, 80 °C, chlorobenzene). Conversions equal 1H NMR spectro-
scopic yields and are based on benzoic acid.
g
g
g
6-p-cymene)))2,
respectively. All molecules were used as catalysts in the formation
of b-oxopropyl benzoate by treatment of propargyl alcohol with
benzoic acid. All complexes show a catalytic activity with moder-
ate to good conversions (23–58%). However, neither electronic
properties nor steric factors alone are responsible for the catalytic
performance, which differs from statements recently made [1,9].
We believe that it is more or less a combination of both criteria,
whereas electron-poor ligands R are best suited. Moderate elec-
tron-rich species are only effective catalysts, when they possess
bulky ligands.
temperature of 80 °C, a concentration of 1.0 mol% and chloro ben-
zene as solvent. However, in contrast to well-established systems
[1a,9], below 80 °C no catalytic activity was observed. As best sol-
vent chloro benzene was found which benefits from the high polar-
ity and hence better solubility when compared to the commonly
used solvent toluene [28]. The results of the catalytic investigations
are summarized in Fig. 3.
From Fig. 3 it can be seen that under the reaction conditions
mentioned above all complexes are catalytically active in the for-
mation of b-oxo propyl benzoate in moderate to good yields. The
most active catalyst is 3b which shows a conversion of 45% within
10 h. Somewhat less active is furyl-substituted 3c but shows the
highest productivity (conversion 58%) after 25 h of all complexes
3a–3e. Nevertheless, the lowest conversion (42%) within the series
of aromatic/heteroaromatic phosphines is observed for the phenyl
derivative 3a. In addition, this phosphine needs a longer induction
period of ca. 5 h to form the catalytically active species which
might be responsible for the poor performance and low yield. Com-
paring the aromatic with the aliphatic phosphine substituents it is
obvious that more electron-rich systems 3d and 3e are less active
and only show productivities of 46% (3e) or 23% (3d) after 25 h
(Fig. 3). However, due to the long reaction times necessary, all cat-
alysts suffer from a loss of activity, which can be attributed to grad-
ually decomposition of the catalyst during the course of the
reaction. Nevertheless, no formation of ‘‘ruthenium black’’ which
indicates the formation of ruthenium particles could be observed.
Also from Fig. 3 it can be seen that tetrametallic 10 with its
three ruthenium(II) centers shows a significantly higher productiv-
ity than 3a featuring only one ruthenium dichloro p-cymene build-
ing block. Furthermore, no induction period is observable. This
phenomenon can most probably be ascribed to synergistic and
cooperative effects between the appropriate transition metals
which improves the catalytic activity with increasing number of
active centers present in one molecule, i.e. dendritic carbene–
palladium complexes [29], phosphino palladium-functionalized
PAMAM dendrimers [30,31] and carbosilane dendrimers with
end-grafted NCN pincer-nickel(II) groups [32]. Further examples
include metallo-enzymes [33] and metal oxide supported catalysts
[34].
Acknowledgements
We are grateful to the Deutsche Forschungsgemeinschaft and
the Fonds der Chemischen Industrie for financial support.
References
[1] (a) D. Devanne, C. Ruppin, P.H. Dixneuf, J. Org. Chem. 53 (1988) 925;
(b) C. Bruneau, P.H. Dixneuf, Angew. Chem., Int. Ed. 45 (2006) 2176.
[2] (a) C. Bruneau, Z. Kabouche, M. Neveux, B. Seiller, P.H. Dixneuf, Inorg. Chim.
Acta 222 (1994) 155;
(b) M. Neveux, B. Seiller, F. Hagedorn, C. Bruneau, P.H. Dixneuf, J. Organomet.
Chem. 451 (1993) 133.
[3] V. Cadierno, J. Francos, J. Gimeno, Green Chem. 12 (2010) 135.
[4] D.W. Hansen Jr., R. Pappo, R.B. Garland, J. Org. Chem. 53 (1988) 4244.
[5] P. Yates, R.S. Grewal, P.C. Hayes, J.F. Sawyer, Can. J. Chem. 66 (1988) 2805.
[6] S. Costin, N.P. Rath, E.B. Bauer, Adv. Synth. Catal. 350 (2008) 2414.
[7] A.J. Fry, D. Herr, Tetrahedron Lett. 19 (1978) 1721.
[8] M.A. Ashraf, M.A. Jones, N.E. Kelly, A. Mullaney, J.S. Snaith, I. Williams,
Tetrahedron Lett. 44 (2003) 3151.
[9] L.J. Goossen, J. Paetzold, D. Koley, Chem. Comm. (2003) 706.
[10] A.D. Cort, J. Org. Chem. 56 (1991) 6708.
[11] R.C. Cambie, R.C. Hayward, J.L. Jurlina, P.S. Rutledge, P.D. Woodgate, J. Chem.
Soc., Perkin Trans. 1 (1978) 126.
[12] S. Torii, T. Inokuchi, S. Misima, T. Kobayashi, J. Org. Chem. 45 (1980) 2731.
[13] S. Stavber, B. Sket, B. Zajc, M. Zupan, Tetrahedron 45 (1989) 6003.
[14] M.M. Salunkhe, A.R. Sande, A.S. Kanade, P.P. Wadgaonkarb, Synth. Commun. 27
(1997) 2885.
[15] M. Arfan Ashraf, A.G. Russell, C.W. Wharton, J.S. Snaith, Tetrahedron 63 (2007)
586.
[16] M. Rotem, Y. Shvo, Organometallics 2 (1983) 1689.
[17] (a) C. Ruppin, P.H. Dixneuf, Tetrahedron Lett. 27 (1986) 6323;
(b) C. Ruppin, P.H. Dixneuf, S. Lecolier, Tetrahedron Lett. 29 (1988) 5365;
(c) K. Philippot, D. Devanne, P.H. Dixneuf, J. Chem. Soc., Chem. Commun.
(1990) 1199.
[18] R. Packheiser, H. Lang, Eur. J. Inorg. Chem. (2007) 3786.
[19] P. Stepnicka, J. Demel, J. Cejka, J. Molec. Catal. A: Chem. 224 (2004) 161.
[20] B. Milde, D. Schaarschmidt, P. Ecorchard, H. Lang, J. Organomet. Chem., in
press, doi:10.1016/j.jorganchem.2012.01.017.
Compared to [RuCl2(
basic phosphines and [RuCl2(
g
6-p-cymene)(PR3)] (R = Ph, Me) [1a] with
6-p-cymene)]2/P(cC4H3O)3 [9] with
[21] T. Baumgartner, M. Fiege, F. Pontzen, R. Arteaga-Müller, Organometallics 25
(2006) 5657.
[22] Vi. Huc, A. Balueva, R.-M. Sebastian, A.-M. Caminade, J.-P. Majoral, Synthesis 5
(2000) 726.
g
its electron-poor phosphine we could not find a general trend con-
cerning basicity or steric factors under reaction conditions used by

B. Milde et al. / Inorganica Chimica Acta 387 (2012) 338–345
345
[23] S.-Z. Luo, Y.-M. Li, Z.-Z. Chen, H. Abe, L.-P. Cui, H. Nakanishi, X.-R. Qin, Y.-F.
Zhao, Lett. Pept. Sci. 10 (2003) 57.
[30] M.T. Reetz, G. Lohmer, R. Schwickardi, Angew. Chem., Int. Ed. Engl. 36 (1997)
1526.
[24] S. Suravajjala, J.R. Polam, L.C. Porter, J. Organomet. Chem. 461 (1993) 201.
[25] G.M. Sheldrick, Acta Crystallogr., Sect. A 46 (1990) 467.
[26] G.M. Sheldrick, SHELXL-97, Program for Crystal Structure Refinement, University
of Göttingen, 1997.
[27] A. Jakob, B. Milde, P. Ecorchard, C. Schreiner, H. Lang, J. Organomet. Chem. 693
(2008) 3821.
[31] S. Dietrich, A. Nicolai, H. Lang, J. Organomet. Chem. 696 (2011) 739.
[32] J.W.J. Kapen, A.W. van der Made, J.C. de Wilde, P.W.N.M. van Leeuwen, P.
Wijkens, D.M. Grove, G. van Koten, Nature 372 (1994) 659.
[33] E.K. van den Beuken, B.L. Feringa, Tetrahedron 54 (1998) 12985.
[34] B. Coq, F. Figueras, Coord. Chem. Rev. 178–180 (1998) 1753.
[35] V. Cadierno, J. Francos, J. Gimeno, Organometallics 30 (2011) 852.
[28] C. Schreiner, Ph.D. Thesis, TU Chemnitz, Germany, 2010.
[29] V. Hornillos, J. Guerra, A. de Cozar, P. Prieto, S. Merino, M.A. Maestro, E. Diez-
Barra, J. Tejeda, Dalton Trans. 40 (2011) 4095.