Synthesis and structural characterization of new chiral mixed phosphine phosphoramidite complexes of ruthenium

Article abstract of DOI:10.1016/j.ica.2008.09.003

New phosphoramidite complexes of ruthenium chiral at the metal were synthesized, structurally characterized and their electrochemical and catalytic properties were studied. Reaction of the known chiral phosphoramidites (RO)₂ PNR₂

Full text of DOI:10.1016/j.ica.2008.09.003

Inorganica Chimica Acta 362 (2009) 1935–1942
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Synthesis and structural characterization of new chiral mixed phosphine
phosphoramidite complexes of ruthenium
*
Stephen Costin, Nigam P. Rath, Eike B. Bauer
University of Missouri – St. Louis, Department of Chemistry and Biochemistry, One University Boulevard, St. Louis, MO 63121, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 16 July 2008
Received in revised form 2 September 2008
Accepted 4 September 2008
Available online 11 September 2008
New phosphoramidite complexes of ruthenium chiral at the metal were synthesized, structurally charac-
terized and their electrochemical and catalytic properties were studied. Reaction of the known chiral
phosphoramidites ðROÞ2PNR02 (R = naphthyl, R0 = CH3, 1a; R = naphthyl, R0 = benzyl, 1b; R = octahydro-
naphthyl, R0 = benzyl, 1c) with CpRu(PPh3)2Cl afforded the title compounds CpRu(PPh3)(1a–c)(Cl)
(2a–c) in 46–74% isolated yields. Fractional crystallization of 2b and 2c afforded the corresponding dia-
stereopure complexes which are chiral both at the metal and at the ligand. The molecular structures of 2b
and 2c were determined, revealing a pseudo octahedral coordination geometry about the ruthenium cen-
ter. Electrochemical studies by cyclic voltammetry showed reversible electrochemical behavior of the
metal complexes 2a–c. The new metal complexes are catalytically active in the Mukaiyama aldol reaction
(24 h, room temperature, 31–53% yield), but almost no enantiomeric excesses for the products were
obtained.
Keywords:
Ruthenium
Phosphoramidites
Cyclic voltammetry
X-ray
Chiral at metal
Mukaiyama aldol reaction
Ó 2008 Elsevier B.V. All rights reserved.
1. Introduction
dite complexes of ruthenium and they are chiral at the metal. We
investigated the electrochemistry of the new complexes and applied
Phosphoramidites are a monodentate, P-coordinating ligand
class, featuring two oxygen atoms and one nitrogen atom bonded
to the phosphorus center (Fig. 1) [1]. Originally described by Ferin-
ga [1b], they are increasingly applied as ligands in transition metal
catalyzed organic transformations such as enantioselective conju-
gate enone addition reactions [1f,2], hydrogenations [1a,1c,3],
allylic alkylations [4], hydrosilylations [5], vinylations [6], cycload-
ditions [7], Diels–Alder [8] and Heck reactions [9]. Due to second-
ary interactions of aromatic ring systems on the ligand with the
metal center, phosphoramidites can serve as two-, four-, six- or
eight-electron donors [10].
It is known that phosphoramidite ligands (L) form ruthenium
complexes with the general formula (p-cymene)Ru(L)(Cl)2 [11].
The p-cymene ligand detaches from these metal complexes at ele-
vated temperatures, potentially opening decomposition pathways.
We were looking for a more robust platform for piano stool type
phosphoramidite complexes of ruthenium and assumed that replac-
ing the p-cymene ligand with a cyclopentadienyl (Cp) ligand would
give rise to more stable architectures. We also were interested in
synthesizing phosphoramidite complexes that are chiral at the me-
tal. Herein we describe the synthesis and structural characterization
of ruthenium phosphoramidite complexes with the general formula
CpRu(PPh3)(L)(Cl). These complexes are the first Cp phosphorami-
them as catalysts in the Mukaiyama aldol reaction.
2. Results and discussion
To achieve our objectives, commercial CpRu(PPh3)2Cl seemed to
be a suitable precursor. This complex is known to give half-sand-
wich complexes of the type CpRu(PPh3)(L)(Cl) by PPh3 exchange
when treated with monodentate phosphines [12]. Accordingly,
the known [13] ligand 1a was reacted with an equimolar amount
of CpRu(PPh3)2Cl in CHCl3 under reflux for 8 h, as shown in Scheme
1 (top reaction). Chromatographic workup of the reaction mixture
provided the new chiral target complex CpRu(PPh3)(1a)(Cl) (2a) as
a 5:3 mixture of diastereomers (assessed by 1H NMR). Efforts to
separate the stereoisomers and to obtain diastereomerically pure
material have failed so far. We speculated that the relatively small
methyl groups on nitrogen create only slight steric congestion at
the metal, rendering the two diastereomers close in energy. Thus,
the known ligand 1b [13] featuring two bulkier benzyl groups on
nitrogen was next employed in complex synthesis. Under the stan-
dard conditions described above (but with 16 h reaction time), li-
gand 1b gave an 8:1 mixture of diastereomers (1H NMR) of
complex 2b after flash chromatography. It was possible to isolate
the major diastereomer by fractional crystallization from CH2Cl2/
MeOH to obtain diastereopure 2b in 74% yield.
To investigate if changes in the naphthyl backbone have an
influence on the properties of the corresponding metal complex,
* Corresponding author.
E-mail address: bauere@umsl.edu (E.B. Bauer).
0020-1693/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2008.09.003

1936
S. Costin et al. / Inorganica Chimica Acta 362 (2009) 1935–1942
the aromatic carbon atoms couple with the phosphorus atoms
(JCP = 12–60 Hz). As a consequence, a complex 13C NMR spectrum
appears in the aromatic region. However, the number of signals
did not reach the number of diastereotopic aromatic carbons, and
some of the peaks were broad and some of them had shoulders (or
other non-Gaussian features). Similarly, the four protons of the two
NCH2 groups in 2b and 2c are diastereotopic and gave four distinct
signals in the 1H NMR. The Cp ligands gave sharp singlets between
4.48 and 4.71 ppm.
O
O
O
O
P
NR2
P
NR2
1a
1c
R=CH2Ph,
R=CH3,
1b
R=CH2Ph,
The FAB mass spectra showed a strong molecular ion peak for
the metal complexes, and a somewhat weaker peak arising from
loss of Cl.
Fig. 1. Phosphoramidite ligands.
To unequivocally establish the structures of the new ruthenium
phosphoramidite complexes, the crystal structures of the com-
plexes 2b and 2c described above were determined (Table 1 and
Section 4). The molecular structures are depicted in Fig. 2 and
Table 2 shows key structural data.
In general, the two ruthenium complexes do not show large
structural differences. Key bond lengths and angles are similar, as
well as the distance between the ruthenium center and the Cp cen-
troid (Table 2).
CpRuCl(PPh3)2
+
R2N
Ru
P
O
PPh3
CHCl3
8 -16 h, reflux
P NR2
Cl
O
O
O
R = Me, 2a, 64%
The bond angles around ruthenium range from 87.37(3)° for the
Cl(1)–Ru–P(1) angle to 99.11(3)° for the P(1)–Ru–P(2) angle. Thus,
the coordination geometry of the complexes is best described as
octahedral with slight distortions and with the largest angles ob-
served between PPh3 and the bulky phosphoramidite ligands
(99.11(3)° and 97.88(6)°, respectively).
5:3 mixture of diastereomers
1a-b
R = CH2Ph, 2b, 74%
diastereopure after recrystallization
The PPh3 Ru–P(1) bond lengths for both complexes (2.3294(8)
and 2.3231(16) Å) are slightly longer than the corresponding phos-
phoramidite Ru–P(2) bond lengths (2.2426(8) and 2.2404(14) Å).
The stronger bond between the phosphorus atom of the phospho-
ramidite and the ruthenium center might be explained by stronger
backbonding for this ligand.
CpRuCl(PPh3)2
+
R2N
Ru
P
PPh3
O
CHCl3
Cl
O
8 h, reflux
O
O
For both ruthenium complexes 2b and 2c, the two phenyl rings
P NR2
connected to the phosphorus form an angle
a (Fig. 3). It is larger for
complex 2c (59.74°) than for complex 2b (49.51°) and is the big-
gest structural difference between the two complexes. In complex
2c, half of each naphthyl ring is hydrogenated, and thus non-aro-
matic. Repulsive forces between the two non-planar rings might
R = CH2Ph, 2c, 46%
diastereopure after recrystallization
1c
contribute to a larger angle a. In turn, the angle a has no influence
Scheme 1. Phosphoramidite complex syntheses.
on the O(1)–P(2)–O(2) angle, which is for both complexes almost
identical (Table 2, entry 8).
the ligand 1c was synthesized following a literature procedure
[14]. Phosphoramidite ligand 1c was converted to its ruthenium
complex 2c as described above for 2b (Scheme 1, bottom).
After flash chromatography, an 8:1 mixture of diastereomers was
obtained (1H NMR), and the major diastereomer was isolated by
fractional crystallization from CH2Cl2/MeOH to obtain 2c diaste-
reopure in 46% yield.
All new mixed phosphoramidite PPh3 ruthenium complexes
2a–c were obtained as orange powders and characterized by
NMR (1H, 13C, 31P), mass spectrometry, IR, and microanalysis.
In the 31P NMR, two doublets were observed for the phospho-
ramidite and the PPh3 ligands. The coordination of the phospho-
ramidite ligand was best observed by a downfield shift of the
phosphorus signal. The free ligands have chemical shifts between
140 and 150 ppm, whereas the ruthenium complexes 2a–c exhib-
ited signals for the coordinated phosphoramidite ligands between
163.8 and 177.8 ppm. The PPh3 ligands showed signals between
45.2 and 48.0 ppm. The JPP coupling constants for the two different
phosphorus atoms in the metal complexes ranged from 148.5 to
168.0 Hz.
Complexes 2b and 2c are chiral at the metal. According to the
priority order C5H5 > Cl > P(2) > P(1) [15,16], the absolute configu-
ration of these complexes is (R), as determined from the X-ray
structures.
To obtain additional information for the new metal complexes,
additional experiments were performed. We were interested in
whether the new diastereopure ruthenium complexes are configu-
rationally stable at elevated temperatures, which would have
importance in their application in catalysis. Thus, an NMR sample
of diastereopure 2b in CDCl3 was heated at 90 °C for 3 h and mon-
itored by 31P NMR (Scheme 2). The sample slowly decomposed
over time, but no signals for the other diastereomer were observed
in the 31P NMR spectrum suggesting that complex 2b is configura-
tionally stable even at elevated temperatures.
We also were interested in whether both PPh3 ligands could be
substituted for phosphoramidite ligands. Thus, 2.5 equiv. of the
phosphoramidite ligand 1b were combined with one equivalent
of CpRu(PPh3)2Cl in CDCl3 and heated at 90 °C for 24 h. The 31P
NMR spectra revealed that besides monosubstitution, some disub-
stitution took place. In addition to the two doublets for the mono-
substituted product 2b, two new doublets at 171.5 and 179.9 ppm
were observed. These peaks are broad and of much lower intensity
than the peaks for 2b. The 31P NMR also showed that unreacted 1b
still was present in the reaction mixture. A FAB mass spectrum
Due to the coordination of the phosphoramidite ligand to the
ruthenium center, all aromatic ring atoms become diastereotopic,
and give in principle different 13C NMR signals for each carbon
atom (but the peaks are in practice not always resolved). Some of

S. Costin et al. / Inorganica Chimica Acta 362 (2009) 1935–1942
1937
Table 1
Crystallographic parameters
2b ꢀ (THF)2
2c ꢀ (CHCl3)2
Empirical formula
Formula weight
Temperature (K)/wavelength (Å)
Crystal system
Space group
C65H62ClNO4P2Ru
1119.62
100(2)/0.71073
monoclinic
P21
C59H56Cl7NO2P2Ru
1222.21
100(2)/0.71073
orthorhombic
P212121
Unit cell dimensions
a (Å)
b (Å)
11.036(2)
13.859(2)
17.620(3)
90
12.4499(9)
15.5701(12)
28.516(2)
90
c (Å)
a
(°)
b (°)
99.349(8)
90
c
(°)
90
90
V (Å3)/Z
2659.0(8)/2
1.398
0.457
1164
5527.8(7)/4
1.621
0.724
Dcalc (Mg/m3)
Absorption coefficient (mmꢁ1
F(000)
)
2504
Crystal size (mm3)
0.28 ꢂ 0.23 ꢂ 0.21
0.24 ꢂ 0.22 ꢂ 0.20
Theta range for data collection (°)
Index ranges
Reflections collected
2.94–27.00
2.62–26.47
ꢁ14 6 h 6 14, ꢁ17 6 k 6 17, ꢁ22 6 l 6 22
72549
ꢁ15 6 h 6 14, ꢁ19 6 k 6 19, 35 6 l 6 35
147736
Independent reflections
Absorption correction
Maximum and minimum transmission
Data/restraints/parameters
Goodness-of-fit on F2
11479 [R(int) = 0.059]
semi-empirical from equivalents
0.9102 and 0.8812
11479/1/697
11368 [R(int) = 0.071]
semi-empirical from equivalents
0.8699 and 0.8006
11368/0/658
1.032
1.114
Final R indices [I > 2sigma(I)]
R indices (all data)
R1 = 0.0326, wR2 = 0.0708
R1 = 0.0373, wR2 = 0.0731
0.663 and ꢁ0.484
ꢁ0.009(16)
R1 = 0.0639, wR2 = 0.1562
R1 = 0.0703, wR2 = 0.1599
1.793 and ꢁ1.622
0.03(4)
Largest difference in peak and hole (e Åꢁ3
Absolute structure parameter
)
exhibited, in addition to a peak for 2b, also a peak for the doubly
substituted product. The NMR and MS data suggest that the second
substitution occurs but it is a sluggish and incomplete process,
potentially leading to a less stable product showing some dynamic
behavior in solution. The bulky phosphoramidite ligands presum-
ably create too much steric constraint about the ruthenium center
resulting in an unstable material.
In order to obtain preliminary insight into the electronic prop-
erties of the new metal complexes, cyclic voltammograms of the
phosphoramidite ligands 1a–c and the metal complexes 2a–c were
recorded as described in Section 4. For comparison, a cyclic voltam-
mogram of the ruthenium precursor CpRu(PPh3)2Cl was recorded
as well. Key data are listed in Table 3 and two representative traces
for ligand 1a and its ruthenium complex 2a are depicted in Fig. 4.
The traces for the other ligands and complexes are given in the
Supplementary material.
The cyclic voltammograms of the phosphoramidite ligands sug-
gest complex and irreversible chemical reactions taking place after
oxidation [17]. Scans towards increasing positive potential show
an irreversible oxidation wave without a reduction wave on the re-
verse scan. The potentials of the free phosphoramidite ligands for
the first oxidation differed between 1.37 and 1.70 V (vs. ferro-
cene/ferrocenium). Thus, the oxidation potentials of the phospho-
ramidite ligands show a slight dependency on their substituents.
The cyclic voltammograms for the corresponding metal com-
plexes 2a–c show a fair degree of reversibility in that a reduction
wave is seen for each complex that appears similar in size and is
of complementary shape to the first oxidation wave. The E1/2 val-
ues for the oxidation range from 0.56 to 0.63 V (vs. ferrocene/ferro-
cenium) and the peak current ratios ic/ia are P0.95 (Table 3). While
the ideal separation between the oxidation and reduction peaks for
a diffusing species undergoing a one electron oxidation–reduction
Whereas the ligands alone show irreversible electrochemical
behavior, the formation of the complex yields much greater signs
of reversibility. This suggests that a one-electron oxidation and
reduction is occurring that involves a highest occupied molecular
orbital found only for the complex and not for the ligand alone.
The precursor complex CpRu(PPh3)2Cl has an E1/2 value of 0.49 V,
which is somewhat lower than those of the ruthenium phospho-
ramidite complexes 2a–c. Strictly speaking, the Ea values for the
free ligands and the E1/2 values for the metal complexes cannot
be compared. However, the potentials of the ligands seem to be
generally higher than those of their respective metal complexes,
suggesting a metal-centered HOMO.
Finally, we were interested to determine if the new complexes
are suitable as catalysts in organic transformations. To maintain
the stereochemical properties of the metal complexes, we targeted
chloride abstraction as a suitable method for activation. It is known
that reagents such as TlPF6 and (Et3O)(PF6) selectively remove
chloride ligands from metal complexes [18]. If performed with
our metal complexes, a chiral Lewis acid would be the result of
the activation. Many organic transformations are catalyzed by Le-
wis acids, and we found the Mukaiyama aldol reaction especially
appealing, as it allows for asymmetric carbon–carbon bond forma-
tion [19].
First, we tested if (Et3O)(PF6) removes selectively the chloride
from complex 2b. An NMR tube experiment in CDCl3 showed that
combination of 1 equiv. of complex 2b and (Et3O)(PF6) resulted in
spectra, which show no remaining starting material (1H, 31P). Three
new broad signals showed up in the 31P NMR spectrum (one singlet
at 14.4 ppm and two doublets at 47.2 and 185.6 ppm, JPP = 62 Hz),
suggesting dynamic processes as a result of chloride abstraction.
After removal of the solvent, a FAB mass spectrum of the residue
no longer exhibited a molecular ion peak for 2b, but a molecular
ion peak for a compound of the general formula [CpRu(PPh3)(1b)]+,
missing the chloride ligand was present (Scheme 2). Identical
results were obtained when AgPF6 in CH2Cl2 was employed as
chloride scavenger. The chloride abstraction was sluggish and
cycle is 59 mV, the peak-to-peak separations
DE seen here are be-
tween 0.23 and 0.41 V suggesting sluggish electron transfer. At
higher potentials up to 2 V, no additional oxidation waves were
observed.

1938
S. Costin et al. / Inorganica Chimica Acta 362 (2009) 1935–1942
Fig. 2. Molecular structures of the ruthenium phosphoramidite complexes 2b (top) and 2c (bottom). Solvent molecules are omitted for clarity.
Table 2
Selected bond lengths, angles and calculated values for ruthenium phosphoramidite
complexes
α
α
Entry
Complex
2b ꢀ (THF)2
2c ꢀ (CHCl3)2
O
O
O
O
Bond lengths (Å)
P
P
NR2
NR2
1
2
3
4
Ru–P(1)
Ru–P(2)
Ru–Cl(1)
P(2)–N(1)
2.3294(8)
2.2426(8)
2.4302(7)
1.653(3)
2.3231(16)
2.2404(14)
2.4415(16)
1.646(5)
Fig. 3. Schematic representation of angle a.
incomplete when TlPF6 was used, and does not work with
(Et3O)(PF6) in diethylether or THF, leaving the starting material
2b unreacted. Due to its high reactivity, all efforts to isolate
[CpRu(PPh3)(1b)]PF6 have failed to date.
Next, we tested if the activated metal complex is catalytically
active in the Mukaiyama aldol reaction. A solution of the catalyst
activated by (Et3O)(PF6) was added to tert-butyl(1-methoxyvinyl-
oxy)dimethylsilane (3) and the corresponding aromatic carbonyl
Bond angles (°)
5
6
7
8
Cl(1)–Ru–P(1)
Cl(1)–Ru–P(2)
P(1)–Ru–P(2)
O(1)–P(2)–O(2)
87.37(3)
92.28(3)
99.11(3)
99.21(9)
88.05(6)
93.85(6)
97.88(6)
99.5(2)
Calculated values (Mercury 1.4.2)
9
10
Angle
a
(Fig. 3) (°)
49.51
1.850
59.74
1.850
Ru–Cp distance (Å)

S. Costin et al. / Inorganica Chimica Acta 362 (2009) 1935–1942
1939
-
+ PF6
R2N
Ru
R2N
Ru
P
some
O
PPh3
P
O
1 equiv. (Et3O)(PF6)
90 ºC for 3h
Cl
PPh3
decomposition
but no change
in configuration
O
O
2b
R=CH2Ph,
detected by MS
Scheme 2. Experiments to understand the reactivity of complex 2b.
Table 3
compound 4, and the sample was analyzed and worked up after
24 h at room temperature. The results are compiled in Table 4. A
control experiment showed that (Et3O)(PF6) also catalyzes the
reaction at least of 4a and 4b with tert-butyl(1-methoxyvinyl-
oxy)dimethylsilane (3). Thus, special care was taken when com-
plex 2b was activated; it was employed in slight excess to make
sure that the activated metal complex did not contain residual
(Et3O)(PF6).
Cyclic voltammetry data of ruthenium phosphoramidite complexes
E1/2/Va
D
E/Va
ic/ia
Epa/Va
Epa/Vb free ligand 1
a
Complex
CpRu(PPh3)2Cl
0.49
0.56
0.63
0.56
0.23
0.23
0.27
0.41
0.95
1.0
0.96
1.0
0.61
0.76
0.80
0.76
2a
2b
2c
1.37
1.63
1.70
a
Referenced against ferrocene as internal standard (see Section 4).
First oxidation potential.
b
After 24 h at room temperature, the silane 3 was completely con-
sumed, and some carbonyl starting material 4 was left over in the
Ligand 1a
Ruthenium Complex 2a
4.00E-06
2.00E-06
0.00E+00
1.00E-05
5.00E-06
0.00E+00
-2.00E-06
-4.00E-06
-6.00E-06
-8.00E-06
-1.00E-05
-1.20E-05
-1.40E-05
-1.60E-05
0
0.5
1
1.5
2
0
0.2
0.4
0.6
0.8
1
1.2
-5.00E-06
-1.00E-05
-1.50E-05
Potential (V)
Potential (V)
Fig. 4. Representative cyclic voltammograms for the free ligand 1a and its ruthenium complex 2a (before referencing to ferrocene/ferrocenium).
Table 4
Catalytic results of the Mukaiyama aldol reaction
Entrya
1
Substrate
1 = R2 = R3 = H
Product
Yieldb (%)
ee (%)
3c
R
R1 = R2 = R3 = H,
53
4a
R4 = SiMe2tBu, 5a
(42)d
2
3
4
a
R1 = CH3
R1 = CH3, R2 = R3 = H,
48
50
31
5e
R2 = R3 = H, 4b
R4 = SiMe2tBu, 5b
R1 = R2 = H
R1 = R2 = H, R3 = OMe,
f
R3 = OMe, 4c
R4 = SiMe2tBu, 5c
R1 = R3 = H
R1 = R3 = H, R2 = Cl,
2c
R2 = Cl, 4d
R4 = H, 5d
Conditions: catalyst 2b (0.007 mmol) was preactivated with (Et3O)(PF6) in CH2Cl2 and after 2 h added to the substrates 3 and 4 (0.472 mmol each). The resulting solution
was maintained at room temperature for 24 h.
b
Isolated yield after chromatography.
Determined by chiral GC.
c
d
Isolated yield when the reaction was performed in the presence of 20 mol% 2,6-di-tert-butylpyridine.
e
Determined by 1H NMR with a chiral shift reagent.
f
Determination of the enantiomeric excess was not possible.

1940
S. Costin et al. / Inorganica Chimica Acta 362 (2009) 1935–1942
reaction mixture, as determined by GC/MS. A strong product peak
was observed. Chromatographic workup afforded the correspond-
ing silyloxy esters (5) in 31–53% isolated yields (Table 4). However,
as assessed by chiral GC or 1H NMR with a chiral shift reagent, only
small enantiomeric excesses for the products 5 were obtained.
The lack of enantioselectivity can be rationalized as follows. It
might indeed be that the architecture of the metal complex after
chloride abstraction does not allow for stereodifferentiation under
the reaction conditions. It also might be that the fragment
[CpRu(PPh3)(1b)]+ (Scheme 2) is configurationally not stable. On
the other hand, it is known that besides Lewis acids, strong
Brønsted acids catalyze the Mukaiyama aldol reaction [20]. Thus,
another possibility is that the strong Lewis acid resulting from
chloride abstraction of 2b creates in the solvent a strong (uniden-
tified) Brønsted acid which is the catalytically active species. How-
ever, when the catalytic reaction was performed in the presence of
20 mol% 2,6-di-tert-butylpyridine, the product 5a was isolated in
42% yield. Thus, it appears to be unlikely that a strong Brønsted
acid is the only catalytically active species in solution. Further
investigations of these processes are currently underway.
(15 mL) was added and the solids dissolved. The orange solution
was then heated to reflux for 8 h. Upon cooling, the solvent was re-
moved under vacuum, giving an orange solid. The solid was purified
by flash chromatography (2.5 ꢂ 17 cm FlorisilÒ, CH2Cl2/Et2O 49:1 v/
v) to obtain 2a as a yellow solid as mixture of diastereomers (5:3, 1H
NMR) (0.535 g, 0.650 mmol, 64%), m.p. 202–203 °C dec. (capillary).
Anal. Calc. for C45H38ClNO2P2Ru: C, 65.65; H, 4.65. Found: C, 65.27;
H, 4.77%.
NMR (d, CDCl3) 1H 7.86 (d, JHH = 8.6 Hz, 2H, binaphthyl), 7.82 (d,
JHH = 4.5 Hz, 0.6H, binaphthyl0), 7.80 (d, JHH = 3.5 Hz, 1H, binaph-
thyl), 7.77 (d, JHH = 2.9 Hz, 1H, binaphthyl), 7.06 (d, JHH = 8.2 Hz,
0.6H, binaphthyl0) 7.63–7.53 (m, 3H, aromatic), 7.40–7.35 (m, 4H,
aromatic), 7.34–7.21 (m, 10H, aromatic), 7.19–7.08 (m, 10H, aro-
matic), 7.06 (d, JHH = 1.7 Hz, 1.6H, aromatic), 7.03 (d, JHH = 1.6 Hz,
1.6H, aromatic), 7.02–6.95 (m, 7H, aromatic), 4.48 (s, 5H, Cp), 4.39
(s, 3H, Cp0), 2.42 (s, 3H, NCH3), 2.39 (s, 3H, NCH3), 2.30 (s, 1.8H,
NCH3), 2.27 (s, 1.8H, NCH3); 13C{1H} (partial) 82.9 (s, Cp), 81.9 (s,
Cp’), 39.04 (s, NCH3), 38.96 (s, NCH3), 38.4 (s, NCH0 ), 38.3 (s,
3
NCH03); 31P{1H} 177.8 (d, JPP = 168.0 Hz, phosphoramidite) 176.0
2
2
2
(d, JPP = 160.2 Hz, phosphoramidite0), 48.7 (d, JPP = 168.0 Hz,
PPh3), 46.9 (d, 2JPP = 160.2 Hz, PPh0 ).
3
HRMS calcd for C45H3835ClNO2P2102Ru 823.1109, found
823.1094. MS (FAB): 823 (2a+, 95%), 788 ([2a–Cl]+, 60%), 526
([2a–PPh3–Cl]+, 30%), 429 ([CpRuPPh3]+, 100%). IR (cmꢁ1, neat so-
lid) 3050 (w), 2840 (w), 2796 (w), 1617 (w), 1589 (m), 1463 (m),
1432 (m), 1229 (s).
3. Conclusion
In conclusion, we have synthesized and structurally character-
ized the first chiral at metal Cp phosphoramidite PPh3 complexes
of the general formula CpRu(PPh3)(L)(Cl) (2a–c). The X-ray structure
analyses showed a slightly distorted octahedral coordination geom-
etry about the ruthenium center. Electrochemical analyses by cyclic
voltammetry of the metal complexes CpRu(PPh3)(L)(Cl) revealed
that they undergo reversible oxidation reactions, whereas their cor-
responding phosphoramidite ligands show irreversible electro-
chemical behavior. The chloride ligand of complex 2b can be
removed by (Et3O)(PF6). The resulting fragment is catalytically ac-
tive in the Mukaiyama aldol reaction but produced almost no enan-
tiomeric excess.
4.2. Synthesis of ‘‘((R)-BINOL-N,N-dibenzyl-
phosphoramidite)CpRuCl(PPh3)” (2b)
To a Schlenk flask containing phosphoramidite 1b (0.299 g,
0.568 mmol) and CpRu(PPh3)2Cl (0.387 g, 0.533 mmol), CHCl3
(8 mL) was added and the solids dissolved. The orange solution
was then heated to reflux for 16 h. Upon cooling, the solvent was re-
moved under vacuum, giving an orange solid. The solid was purified
by flash chromatography (2 ꢂ 17 cm silica, CH2Cl2/diethyl ether
49:1 v/v) to obtain 2b as an orange solid as a mixture of diastereo-
mers (>8:1, 1H NMR) (0.444 g, 0.455 mmol). The compound was
recrystallized from CH2Cl2/MeOH to obtain 2b as a single diastereo-
mer (0.386 g, 0.395 mmol, 74%), m.p. 189–190 °C dec. (capillary).
Anal. Calc. for C57H46ClNO2P2Ru: C, 70.18; H, 4.75. Found: C, 69.89;
H, 4.72%.
4. Experimental
4.1. General
Chemicals were treated as follows: diethyl ether, distilled from
Na/benzophenone; CH2Cl2, distilled from CaH2, MeOH and CHCl3
used as received. CpRu(PPh3)2Cl (Strem), (Et3O)(PF6), silica (Al-
drich), FlorisilÒ (Fisher) and other materials, used as received.
‘‘(R)-BINOL-N,N-dimethyl-phosphoramidite” 1a and ‘‘(R)-BINOL-
N,N-dibenzyl-phosphoramidite” 1b were synthesized according
to literature procedures [13] as well as ‘‘(R)-BINOL(8H)-N,N-diben-
zyl-phosphoramidite” (1c) [14]. All reactions were carried out un-
der nitrogen employing standard Schlenk techniques, and workups
were carried out in the air.
NMR (d, CDCl3) 1H 8.06 (d, JHH = 4.9 Hz, 1H, binaphthyl), 8.01 (d,
JHH = 5.3 Hz, 1H, binaphthyl), 7.74 (d, JHH = 4.9 Hz, 1H, binaphthyl),
7.55 (t, JHH = 4.3 Hz, 1H, binaphthyl), 7.47 (d, JHH = 5.3 Hz, 1H,
binaphthyl), 7.42–7.32 (m, 11H, aromatic), 7.31–7.27 (m, 6H, aro-
matic), 7.15–7.09 (m, 7H, aromatic), 7.08–6.92 (m, 7H, br, aromatic),
2
6.73 (d, JHH = 5.3 Hz, 1H, binaphthyl), 4.92 (d, JHH = 6.8 Hz, 1H,
NCHH0), 4.89 (d, 2JHH = 6.8 Hz, 1H, NCHH0), 4.71 (s, 5H, Cp), 3.83 (d,
2
2JHH = 6.1 Hz, 1H, NCHH0), 3.79 (d, JHH = 6.8 Hz, 1H, NCHH0);
13C{1H} (500 MHz) 151.1 (t, JCP = 24.1 Hz, aromatic), 149.3 (s, br, aro-
matic), 139.6 (s, br, aromatic), 137.9 (s, br, aromatic), 137.6 (s, br,
aromatic), 135.0 (s, br, aromatic), 133.8 (d, JCP = 16.5 Hz, aromatic),
132.8 (d, JCP = 17.4 Hz, aromatic), 131.5 (s, br, aromatic), 131.1 (s,
br, aromatic), 130.6 (d, JCP = 12.0 Hz, aromatic), 130.4 (d,
JCP = 12.0 Hz, aromatic), 129.7 (s, br, aromatic), 129.3 (d,
JCP = 12.0 Hz, aromatic), 129.1 (d, JCP = 15.3 Hz, aromatic), 128.4 (s,
br, aromatic), 128.2 (s, br, aromatic), 127.8 (d, JCP = 16.2 Hz, aro-
matic), 127.1 (d, JCP = 18.0 Hz, aromatic), 126.9 (s, br, aromatic),
126.7 (s, br, aromatic), 126.5 (t, JCP = 14.5 Hz, aromatic), 125.8 (d,
JCP = 14.3 Hz, aromatic), 125.6 (d, JCP = 14.7 Hz, aromatic), 125.5 (d,
JCP = 14.7 Hz, aromatic), 125.0 (d, JCP = 14.8 Hz, aromatic), 123.6 (s,
aromatic), 123.4 (s, aromatic), 122.8 (s, aromatic), 122.3 (s, aro-
matic), 121.5 (s, aromatic), 81.6 (d, J = 176 Hz, br, Cp), 51.2 (d,
J = 102.3 Hz, br, NCH2), 50.1 (d, J = 328.5 Hz, br, NCH2); 31P{1H}
NMR spectra were obtained at room temperature on a Bruker
Avance 300 MHz or a Varian Unity Plus 300 MHz instrument and
referenced to a residual solvent signal; all assignments are tentative.
GC/MS spectra were recorded on a Hewlett Packard GC/MS System
Model 5988A. Exact masses were obtained on a JEOL MStation [JMS-
700] Mass Spectrometer. Melting points are uncorrected and were ta-
ken on an Electrothermal 9100 instrument. IR spectra were recorded
on a Thermo Nicolet 360 FT-IR spectrometer. Elementalanalyses were
performed by Atlantic Microlab Inc., Norcross, GA, USA.
4.1. Synthesis of ‘‘((R)-BINOL-N,N-dimethyl-
phosphoramidite)CpRuCl(PPh3)” (2a)
To a Schlenk flask containing phosphoramidite 1a (0.400 g,
1.11 mmol) and CpRu(PPh3)2Cl (0.735 g, 1.01 mmol), CHCl3

S. Costin et al. / Inorganica Chimica Acta 362 (2009) 1935–1942
1941
171.8 (d, 2JPP = 152.2 Hz, phosphoramidite), 45.2 (d, 2JPP = 152.2 Hz,
PPh3).
4.5. Cyclic voltammetry
HRMS calcd for C57H4635ClNO2P2102Ru 975.1735, found
The voltamogramms were recorded with a Princeton Applied
Research Parstat 2273 Electrochemical System and analyzed with
POWERSUITE 2.58 software. Cells were fitted with a glassy carbon
working electrode, and an Ag wire pseudoreference electrode and
a platinum wire as auxiliary electrode. All CH2Cl2 solutions were
975.1702. MS (FAB) 975 (2b+, 90%), 940 ([2b–Cl]+, 45%), 678
([2b–PPh3–Cl]+, 92%), 429 ([CpRuPPh3]+, 100%). IR (cmꢁ1, neat so-
lid) 3052 (m), 1617 (w), 1591 (w), 1460 (w), 1432 (m), 1229 (s).
1 ꢂ 10ꢁ3 M in substrate and 0.05 M in n-Bu4N+ ꢀ PF6 and degassed
ꢁ
4.3. Synthesis of ‘‘((R)-BINOL(8H)-N,N-dibenzyl-
phosphoramidite)CpRuCl(PPh3)” (2c)
with argon. All cyclic voltammograms were recorded with a scan
rate of 100 mV sꢁ1. Ferrocene was subsequently added, and cali-
bration voltammograms recorded. All potentials are references to
the observed E1/2 value for the ferrocene/ferrocenium couple. The
ambient laboratory temperature was 22 1 °C.
To a Schlenk flask containing phosphoramidite 1c (0.200 g,
0.385 mmol) and CpRu(PPh3)2Cl (0.225 g, 0.310 mmol), CHCl3
(8 mL) was added and the solids dissolved. The orange solution
was then heated to reflux for 16 h. Upon cooling, the solvent was
removed under vacuum, giving an orange solid. The solid was puri-
fied by flash chromatography (2 ꢂ 10 cm silica, CH2Cl2/diethyl
ether 49:1 v/v) to obtain 2c as an orange solid as a mixture of dia-
stereomers (8:1, 1H NMR) (0.181 g, 0.184 mmol). The compound
was recrystallized from CH2Cl2/MeOH to obtain 2c as a single dia-
stereomer (0.141 g, 0.143 mmol, 46%), m.p. 218–219 °C dec. (capil-
lary). Anal. Calc. for C57H54ClNO2P2Ru: C, 69.61; H, 5.53. Found: C,
69.54; H, 5.68%.
4.6. Catalytic experiments
A batch solution of the activated catalyst was prepared first.
Complex 2b (0.029 g, 0.0297 mmol) and (Et3O)(PF6) (0.006 g,
0.0241 mmol) were combined in CH2Cl2 (6 mL) under nitrogen
and maintained at room temperature for 2 h. A deep orange solu-
tion resulted, and 1.5 mL of the solution (equivalent to 0.007 g,
0.007 mmol catalyst 2b) was added to the silane 3 (0.472 mmol)
and the carbonyl compound 4 (0.472 mmol). After 24 h at room
temperature, the solvent was removed and the residue purified
by column chromatography (1 ꢂ 8 cm silica gel, CH2Cl2/cyclohex-
ane 2:1 v:v) to obtain 5 as slightly yellow oils.
NMR (d, CDCl3) 1H 7.37–7.25 (m, 10H, aromatic), 7.24–7.09 (m,
3
16H, aromatic), 6.80 (d, JHH = 8.2 Hz, 1H, binaphthyl), 6.57 (d,
3
3JHH = 8.2 Hz, 1H, binaphthyl), 6.21 (d, JHH = 8.0 Hz, 1H, binaph-
2
2
thyl), 4.91 (d, JHH = 10.0 Hz, 1H, NCHH0), 4.85 (d, JHH = 10.0 Hz,
2
1H, NCHH0), 4.60 (s, 5H, Cp), 3.41 (d, JHH = 10.8 Hz, 1H, NCHH0),
2
3
3.35 (d, JHH = 10.8 Hz, 1H, NCHH0), 2.96 (t, JHH = 6.0 Hz, 2H,
CH2), 2.72–2.38 (m, 4H, 2CH2), 2.11–1.95 (m, 2H, CH2), 1.92–1.71
(m, 3H), 1.70–1.55 (m, 4H, 2CH2), 1.40–1.25 (m, 1H); 13C{1H}
149.3 (d, JCP = 60.0 Hz, aromatic), 148.6 (d, JCP = 24.0 Hz, aromatic),
139.8 (d, JCP = 7.8 Hz, aromatic), 139.1 (s, aromatic), 138.2 (s, aro-
matic), 137.8 (s, aromatic), 137.6 (s, aromatic), 134.4 (s, br, aro-
matic), 133.5 (s, aromatic), 133.4 (s, aromatic), 129.3 (s,
aromatic), 129.2 (s, aromatic), 129.0 (s, aromatic), 128.4 (s, aro-
matic), 128.1 (s, aromatic), 127.7 (s, aromatic), 127.6 (s, aromatic),
127.4 (d, JCP = 36.0 Hz, aromatic), 127.1 (d, JCP = 9.1 Hz, aromatic),
126.9 (s, aromatic), 81.0 (s, Cp), 50.1 (s, NCH2), 50.0 (s, NCH2),
29.7 (s, CH2), 29.2 (s, CH2), 28.2 (s, CH2), 27.6 (s, CH2), 23.1 (s,
CH2), 23.0 (s, CH2), 22.9 (s, CH2), 22.8 (s, CH2); 31P{1H} 163.8 (d,
Acknowledgments
We thank Dr. Keith Stine (University of Missouri – St. Louis) for
assistance with the cyclic voltammograms. We thank the Univer-
sity of Missouri – St. Louis (Research Board) for support. Funding
from the National Science Foundation for the purchase of the Apex-
II diffractometer (MRI, CHE-0420497), the purchase of the NMR
spectrometer (CHE-9974801) and the purchase of the mass spec-
trometer (CHE-9708640) is acknowledged.
Appendix A. Supplementary material
CCDC 694648 and 694647 contain the supplementary crystallo-
graphic data for 2b and 2c. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif. NMR spectra (1H, 13C) for
compounds 5a–d (Table 4) as well as spectra for the determination
of enantiomeric excesses and the cyclic voltamogramms for com-
pounds 1b–c and 2b–c. Supplementary data associated with this
article can be found, in the online version, at doi:10.1016/
j.ica.2008.09.003.
2
2JPP = 148.5 Hz, phosphoramidite), 46.0 (d, JPP = 148.5 Hz, PPh3).
HRMS calcd for C57H5435ClNO2P2102Ru 983.2361, found
983.2329. MS (FAB): 983 (2c+, 45%), 948 ([2c–Cl]+, 15%), 686
([2c–PPh3–Cl]+, 100%), 429 ([CpRuPPh3]+, 95%). IR (cmꢁ1, neat so-
lid) 3052 (m), 3022 (m), 2929 (s), 2858 (m), 1582 (w), 1469 (s),
1433 (s).
4.4. X-ray structure analyses
References
To obtain crystals of X-ray quality, 2b was dissolved in THF and
layered with hexanes and 2c was dissolved in CHCl3 and layered
with MeOH and both samples stored at ꢁ18 °C. X-ray structure
data were collected and analyzed as described before [21].
[1] (a) N. Mršic´, L. Lefort, J.A.F. Boogers, A.J. Minnaard, B.L. Feringa, J.G. de Vries,
Adv. Synth. Catal. 350 (2008) 1081;
(b) A.J. Minnaard, B.L. Feringa, L. Lefort, J.G. de Vries, Acc. Chem. Res. 40 (2007)
1267;
Crystal data and intensity data collection parameters are listed in
Table1. Structuresolutionandrefinementwerecarriedoutusingthe
SHELXTL-PLUS software package [22]. The structures were solved by di-
rect methods and refined successfully in the space groups P212121
and P21, respectively. Full matrix least-squares refinement was car-
(c) W. Zhang, X. Zhang, J. Org. Chem. 72 (2007) 1020;
(d) B. Zhao, Z. Wang, K. Ding, Adv. Synth. Catal. 348 (2006) 1049;
(e) M. van den Berg, A.J. Minnaard, R.M. Haak, M. Leeman, E.P. Schudde, A.
Meetsma, B.L. Feringa, A.H.M. de Vries, C.E.P. Maljaars, C.E. Willans, D. Hyett,
J.A.F. Boogers, H.J.W. Henderickx, J.G. de Vries, Adv. Synth. Catal. 345 (2003) 308;
(f) A.H.M. de Vries, A. Meetsma, B.L. Feringa, Angew. Chem., Int. Ed. Engl. 35
(1996) 2375.
P
2
ried out by minimizing
wðF2o ꢁ F2c Þ . The non-hydrogen atoms
´
[2] (a) B. Maciá, A. Fernández-Ibáñez, N. Mršic, A.M. Minnaard, B.L. Feringa,
were refined anisotropically to convergence. All hydrogen atoms
were calculated in their idealized geometry and were treated using
appropriate riding model (AFIX m3). Disorder in the solvent mole-
cule (CHCl3) in case of 2c and the 2 THF molecules in case of 2b were
resolved with partial occupancy atoms.
Tetrahedron Lett. 49 (2008) 1877;
(b) T. Pfretzschner, L. Kleemann, B. Janza, K. Harms, T. Schrader, Chem. Eur. J.
10 (2004) 6048.
[3] (a) L. Eberhardt, D. Armspach, J. Harrowfield, D. Matt, Chem. Soc. Rev. 37
(2008) 839;
(b) I.D. Kostas, K.A. Vallianatou, J. Holz, A. Börner, Tetrahedron Lett. 49 (2008)
331;
(c) W. Zhang, X. Zhang, Angew. Chem., Int. Ed. 45 (2006) 5515.
Table of calculated and observed structure factors are available
in electronic format.

1942
S. Costin et al. / Inorganica Chimica Acta 362 (2009) 1935–1942
[4] (a) S. Shekar, B. Trantow, A. Leitner, J.F. Hartwig, J. Am. Chem. Soc. 128 (2007)
11770;
[13] (a) R. Hulst, N.K. de Vries, B.L. Feringa, Tetrahedron: Asymmetry 5 (1994) 699;
(b) L.A. Arnold, R. Imbos, A. Mandoli, A.H.M. de Vries, R. Naasz, B.L. Feringa,
Tetrahedron 56 (2000) 2865;
(c) A. Rimkus, N. Sewald, Org. Lett. 5 (2003) 79.
[14] S. Costin, N.P. Rath, E.B. Bauer, Adv. Synth. Catal. (2008), doi: 10.1002/
adsc.200800355.
(b) E. Rauly, C. Claver, O. Pàmies, M. Diéguez, Org. Lett. 9 (2007) 49;
(c) M.D.K. Boele, P.C.J. Kamer, M. Lutz, A.L. Spek, J.G. de Vries, P.W.N.M. van
Leeuwen, G.P.F. Strijdonck, Chem. Eur. J. 10 (2004) 6232;
(d) M. Kimura, Y. Uozumi, J. Org. Chem. 72 (2007) 707.
[5] J.F. Jensen, B.Y. Svendsen, T.V. la Cour, H.L. Pedersen, M. Johannsen, J. Am.
Chem. Soc. 124 (2002) 4558.
[15] (a) D. Carmona, C. Vega, N. García, F.J. Lahoz, S. Elipe, L.A. Oro, M.P. Lamata, F.
Viguri, R. Borao, Organometallics 25 (2006) 1592;
[6] A. Duursma, J.-G. Boiteau, L. Lefort, J.A.F. Boogers, A.H.M. de Vries, J.G. de Vries,
A.J. Minnaard, B.L. Feringa, J. Org. Chem. 69 (2004) 8045.
[7] (a) N. Toselli, D. Martin, M. Achard, A. Tenaglia, T. Bürgi, G. Buono, Adv. Synth.
Catal. 350 (2008) 280;
(b) D. Carmona, F.J. Lahoz, S. Elipe, L.A. Oro, M.P. Lamata, F. Viguri, F. Sánchez, S.
Martínez, C. Cativiela, M.P. López-Ram de Víu, Organometallics 21 (2002)
5100;
(c) K. Hiroi, K. Watanabe, Tetrahedron: Asymmetry 13 (2002) 1841.
[16] (a) R.S. Cahn, C. Ingold, V. Prelog, Angew. Chem., Int. Ed. Engl. 5 (1966) 385;
(b) K. Stanley, M.C. Baird, J. Am. Chem. Soc. 97 (1975) 6598.
[17] W.E. Geiger, in: P.T. Kissinger, W.R. Heinemann (Eds.), Laboratory Techniques
in Electroanalytical Chemistry, Marcel Dekker, New York, 1996, pp. 684–717.
[18] D. Huber, P.G.A. Kumar, P.S. Pregosin, A. Mezzetti, Organometallics 24 (2005)
5221.
[19] (a) T. Mukaiyama, K. Banno, K. Narasaka, J. Am. Chem. Soc. 96 (1974) 7503;
(b) C. Palomo, M. Oiarbide, J.M. García, Chem. Eur. J. 8 (2002) 37.
[20] T.K. Hollis, B. Bosnich, J. Am. Chem. Soc. 117 (1995) 4570.
[21] S.L. Sedinkin, N.P. Rath, E.B. Bauer, J. Organomet. Chem. 693 (2008) 3081.
[22] G.M. Sheldrick, Acta Crystallogr., Sect. A 64 (2008) 112.
(b) B.M. Trost, S.M. Silverman, J.P. Stambuli, J. Am. Chem. Soc. 129 (2007) 12398;
(c) B.M. Trost, N. Cramer, S.M. Silverman, J. Am. Chem. Soc. 129 (2007) 12396;
(d) B.M. Trost, J.P. Stambuli, S.M. Silverman, U. Schwörer, J. Am. Chem. Soc. 128
(2006) 13328.
[8] J.W. Faller, P.P. Fontaine, Organometallics 24 (2005) 4132.
[9] R. Imbos, A.J. Minnaard, B.L. Feringa, Dalton Trans. (2003) 2017.
[10] I.S. Mikhel, H. Rüegger, P. Butti, F. Camponovo, D. Huber, A. Mezzetti,
Organometallics 27 (2008) 2937.
[11] D. Huber, P.G. Anil Kumar, P.S. Pregosin, I.S. Mikhel, A. Mezzetti, Helv. Chim.
Acta 89 (2006) 1696.
[12] C.A. Mebi, R.P. Nair, B.J. Frost, Organometallics 26 (2007) 429. and literature
cited therein.