Transfer hydrogenation of ketones catalyzed by complexes of ruthenium(II) with the heterotridentate [P,N,O] ligands (S)-2-[{2-(diphenylphosphanyl)benzylidene}amino]propan-1-ol, (S)-2-[{2-(diphenylphosphanyl)benzyl}amino]propan-1-ol or the [P,N,S] ligand (S)-2-(dimethylamino)-1-(diphenylphosphino)-3-(methylthio)propane

Article abstract of DOI:10.1016/j.jorganchem.2016.12.008

Ruthenium complexes [mer-κ³-{(S)-Ph₂PC₆H₄CH?=?NCH(CH₃)CH₂OH}PPh₃RuCl₂[sbnd]P,N,O] 1, [mer-κ³-{(S)-Ph₂PC₆H₄CH₂NHCH(CH₃)CH₂OH}PPh₃RuCl₂[sbnd]P,N,O] 2 and [{(S)-Ph₂PCH₂C(H)N(CH₃)₂CH₂SCH₃}PPh₃RuCl₂] 3 have been prepared and their activity in the catalytic transfer hydrogenation of ketones investigated. Using isopropanol-isopropoxide as the hydrogen donor, 1.OPPh₃.toluene and 2.OPPh₃.CH₂Cl₂both showed low activity with average turnover frequency up to 1.1 per hour and enantiomeric excess up to 39.9% where 3 showed no activity. With dimethylamine borane as the hydrogen donor, 1.OPPh₃.toluene and 3 proved active catalysts showing average activity up to 3.0 turnovers per hour and selectivity up to 46% ee. New structural information on [(HNMe₂)₂BH₂]Cl, a side product in the reaction, is presented.

Full text of DOI:10.1016/j.jorganchem.2016.12.008

Accepted Manuscript
Transfer hydrogenation of ketones catalyzed by complexes of
ruthenium(II) with the heterotridentate [P,N,O] ligands (S)-2-[{2-
(diphenylphosphanyl)benzylidene}amino]propan-1-ol, (S)-2-[{2-
(diphenylphosphanyl)benzyl}amino]propan-1-ol or the [P,N,S] ligand (S)-2-
(dimethylamino)-1-(diphenylphosphino)-3-(methylthio)propane
Abeer H. Obaid, Nuru G. Stracey, Christopher D. Incarvito, Brandon Q. Mercado,
Jonathan Parr
PII:
S0022-328X(16)30559-9
10.1016/j.jorganchem.2016.12.008
JOM 19731
DOI:
Reference:
To appear in: Journal of Organometallic Chemistry
Received Date: 17 August 2016
Revised Date: 6 December 2016
Accepted Date: 7 December 2016
Please cite this article as: A.H. Obaid, N.G. Stracey, C.D. Incarvito, B.Q. Mercado, J. Parr, Transfer
hydrogenation of ketones catalyzed by complexes of ruthenium(II) with the heterotridentate
[P,N,O] ligands (S)-2-[{2-(diphenylphosphanyl)benzylidene}amino]propan-1-ol, (S)-2-[{2-
(diphenylphosphanyl)benzyl}amino]propan-1-ol or the [P,N,S] ligand (S)-2-(dimethylamino)-1-
(diphenylphosphino)-3-(methylthio)propane, Journal of Organometallic Chemistry (2017), doi: 10.1016/
j.jorganchem.2016.12.008.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Graphical Abstract
Cl
Ru
Cl
Ph2
P
PPh3
OH
N
OH
O
Me
R1 R2
R1 R2
PriOH/PriO- or Me2NHBH3
1

ACCEPTED MANUSCRIPT
Transfer hydrogenation of ketones catalyzed by complexes of ruthenium(II) with the
heterotridentate [P,N,O] ligands (S)-2-[{2-(diphenylphosphanyl)benzylidene}amino]propan-1-ol,
(S)-2-[{2-(diphenylphosphanyl)benzyl}amino]propan-1-ol or the [P,N,S] ligand (S)-2-
(dimethylamino)-1-(diphenylphosphino)-3-(methylthio)propane.
Abeer H. Obaid,† Nuru G. Stracey‡, Christopher D. Incarvito,ǁ Brandon Q. Mercado†§ and
Jonathan Parr*†
Department of Chemistry, Sterling Chemistry Laboratories, Yale University, New Haven, CT 06520
† Department of Chemistry, Sterling Chemistry Laboratory, Yale University, New Haven, CT 06520-
8107, USA.
‡ Current address: Department of Chemistry, University of Wisconsin-Madison, 1101 University
Avenue, Madison, WI 53706.
ǁ Current address: Director of Research Operations and Technology, West Campus, Yale University, 100
West Campus Drive, Orange, CT 06477: Crystallography (Ruthenium).
§ Chemical and Biological Instrument Centre, Department of Chemistry, Yale University, New Haven,
CT 06520-8107, USA: Crystallography (Ruthenium, Boron).
Keywords: Transfer hydrogenation; ruthenium; heterotridentate ligands; dimethylamine borane; x-ray
crystallography
Abstract: Ruthenium complexes [mer-κ3-{(S)-Ph2PC6H4CH=NCH(CH3)CH2OH}PPh3RuCl2-P,N,O] 1,
[mer-κ3-{(S)-Ph2PC6H4CH2NHCH(CH3)CH2OH}PPh3RuCl2-P,N,O] 2 and [{(S)-
Ph2PCH2C(H)N(CH3)2CH2SCH3}PPh3RuCl2] 3 have been prepared and their activity in the catalytic
transfer hydrogenation of ketones investigated. Using isopropanol-isopropoxide as the hydrogen donor,
1.OPPh3.toluene and 2.OPPh3.CH2Cl2 both showed low activity with average turnover frequency up to
1.1 per hour and enantiomeric excess up to 39.9% where 3 showed no activity. With dimethylamine
1

ACCEPTED MANUSCRIPT
borane as the hydrogen donor, 1.OPPh3.toluene and 3 proved active catalysts showing average activity
up to 3.0 turnovers per hour and selectivity up to 46% ee. New structural information on
[(HNMe2)2BH2]Cl , a side product in the reaction, is presented.
1.0 Introduction.
Catalytic transfer hydrogenation is an attractive protocol that has found a number of important
applications[1-3] particularly in the reduction of prochiral carbonyl and imine compounds, where
relatively small molar amounts of high-value chiral alcohols or amines are sought.[4-8] The two most
widely used hydrogen donors in catalytic transfer hydrogenation are isopropanol and formic acid. Both
typically require the use of a strong base, isopropoxide or triethylamine respectively, and the role of base
in such transfer hydrogenations has been examined in terms of the reaction mechanism.[9] Recently,
some examples of catalytic transfer hydrogenation using isopropanol as the hydrogen donor that work
without added base have been reported.[10,11] In cases where isopropanol is the hydrogen donor, the
reverse reaction is more favourable than in those reactions that use formic acid as the hydrogen donor
due in large part to the fugitive nature of carbon dioxide, the dehydrogenated product of formic acid.
Such a reversible reaction has been exploited in kinetic resolutions using transfer hydrogenation
catalysts.[12,13]
Amine borane complexes have been shown to be useful hydrogen donors for both heterogeneous
palladium-catalyzed[14,15] and homogeneous ruthenium-[16,17] and rhenium-catalyzed reactions.[18]
Such amine borane complexes, as with formic acid-triethylamine, are less likely to show reversible
reaction behaviour due in this case to the inert nature of the dehydrogenated boron-nitrogen compounds
produced[19,20] and as a result may prove of particular interest in reactions where enantiomeric excess
in the product is desired.
2

ACCEPTED MANUSCRIPT
A number of groups have reported complexes of transition metals that comprise tridentate
[P,N,O] ligands derived from the condensation of 2-(diphenylphosphino)benzaldehyde and any of a
range of amino alcohols, amino phenols, amino acids and hydrazones.[21-32] While many of the
complexes discussed in these papers comprise metals associated with homogeneous catalysis,
Scheme 1. The synthesis of 1-3.
few feature the application of these complexes in catalysis.[21,23,24,28,29,31] Here we report the
transfer hydrogenation of aryl ketones catalyzed by 1.OPPh3.tol, 2.OPPh3.CH2Cl2 and the related
complex 3, containing the [P,N,S] ligand (S)-2-(dimethylamino)-1-(diphenylphosphino)-3-
(methylthio)propane reported by Kellogg.[33] This ligand probably binds to the ruthenium in a facial
3

ACCEPTED MANUSCRIPT
configuration; ruthenium complexes of [P,N,O] ligands that bind in a facial fashion have been reported
and their application in transfer hydrogenation investigated.[34,35]
The [P,N,O] ligand in 1 is prepared from the Schiff-base condensation of 2-
(diphenylphosphino)benzaldehyde with (S)-2-amino propanol and has an imine nitrogen donor whereas
2 has the reduced form of the same ligand, a [P,N,O] donor with an amine nitrogen. This difference in
nitrogen donors, an imine and a secondary amine respectively, speaks to the observations made by
Noyori et al. on the requirement for a nitrogen-containing ligand that can act as a hydrogen shuttle in
effective hydrogen transfer catalysis.[12] The chiral ligands used to prepare the most successful chiral
catalysts rely on steric effects[36], on differential electronic donor-acceptor interactions[37] or both to
control selectivity. The chiral ligands used here do not have large groups at or near the chiral centre but
polydentate ligands that comprise different donor groups have the opportunity to influence the geometry
and conformation of reaction intermediates through differences in bonding interactions that depend upon
the identity and relative position of the donors.
Although the heterotridentate [P,N,S] ligand (S)-2-(dimethylamino)-1-(diphenylphosphino)-3-
(methylthio)propane in complex 3 is expected to show a fac rather than a mer configuration, it does
comprise a nitrogen donor, in this case a tertiary amine, coupled in this case with a diphenylphosphino
group, as do the ligands in 1 and 2, and a thioether. It was anticipated that the nitrogen in this complex
would be less competent to participate in any hydrogen-shuttling pathways than the nitrogens in 1 and 2.
An examination of the efficiency of these three complexes in transfer hydrogenation reactions was
intended to throw some light on the role of the different nitrogen donors. In addition, each ligand was
also prepared using chiral starting materials that are readily available and inexpensive in order to
examine the possibility of preparing effective chiral ligands of low cost.
4

ACCEPTED MANUSCRIPT
2. Experimental
2.1 General considerations: The syntheses of catalyst precursors were carried out under nitrogen using
Schlenk teachniques. Reactions were carried out in reagent grade solvents used without additional
drying, degassing or distillation. NMR Spectra were recorded on Ge ꢀ300MHz, Bruker Avance
400MHz or 500MHz or Agilent 400MHz spectrometers. GC-MS Analyses were carried out using a
Thermo Focus GC – DSQII with a BetaDex 120 column. Enantiomeric excess was also measured by
NMR using the shift reagent (trifluoromethyl)(anthryl)ethanol. Optical rotations were measured using a
Perkin Elmer 341 polarimeter. The ligands were prepared according to published procedures.[33,38]
2.2.1 Synthesis of κ3-(S)-2-[{2-(diphenylphosphanyl)benzylidene}amino]propan-1-ol-
P,N,O](PPh3)RuCl2, 1.OPPh3.tol: A solution of (S)-2-[{2-
(diphenylphosphinyl)benzylidene}amino]propan-1-ol (0.3011g, 0.867mmol) and RuCl2(PPh3)3 (0.863g,
0.90mmol) in toluene (20mL) was refluxed under nitrogen for 4h. Cooling the resulting deep red-purple
solution to room temperature gave 1 as a purple microcrystalline solid. Recrystallization of this crude
product in air from toluene gave crystals of 1.OPPh3.tol suitable for x-ray analysis. Yield: 0.552g, 79%.
Anal. Calc. for 1.OPPh3.tol: C 67.76 H 5.21 N 1.21%. Found: C 67.42 H 5.11 N 1.33%. 31P{1H},
CD2Cl2, 121MHz , δ: 67.59 (1P, d, 35Hz, RuPPh2), 35.41 (1P, d, 35Hz, Ru-PPh3,), 28.0 (1P, s, OPPh3).
1H NMR, CD2Cl2, 400MHz, δ: 8.76 (1H, HC=N, d, 8.8Hz), 7.72-6.84 (50H, m. arom.), 4.95 (1H, br,
CH2OH), 3.90 (1H, CHHOH, dd, 8.0, 1.6Hz), 3.68 (1H, C(CH3)H, dq, 8.0, 4.1Hz), 3.37 (1H, CHHOH,
dd, 8.0, 1.6Hz), 2.34 (3H, C6H5CH3), 1.53 (3H, HCCH3, d, 4.1Hz).
2.2.2 Synthesis of κ
3-(S)-2-[{2-(diphenylphosphanyl)benzyl}amino]propan-1-ol-P,N,O](PPh3)RuCl2 2:
A solution of (S)-2-[{(2-diphenylphosphinyl)benzyl}amino]propan-1-ol (0.138g, 0.39mmol) and
RuCl2(PPh3)3 (0.757g, 0.78mmol) in toluene (15mL) was refluxed under nitrogen for 4h. Cooling the
red-orange solution under nitrogen gave compound 2 as an orange-brown precipitate. The crude product
5

ACCEPTED MANUSCRIPT
was purified using column chromatography (silica gel) using methylene chloride as the eluting solvent.
Recrystallization of this crude product in air from a methylene chloride solution gave crystals of
2.OPPh3.CH2Cl2 suitable for x-ray analysis. Yield: 0.125g, 40.0%. Anal. Calc for 2.OPPh3.CH2Cl2: C
61.70 H 4.92 N 1.22%. Found: C 61.65 H 4.46 N 1.19%. 31P{1H} NMR, CD2Cl2, 162MHz, δ: 57.0
1
(1P, RuPPh2, d, 34Hz), 40.5 (1P, RuPPh3, d, 34Hz ), 28.0 (1P, s, OPPh3). H NMR, CD2Cl2, 400MHz,
δ: 7.5-6.5 (44H, m, arom.) 5.95 (1H, br, OH or NH), 5.32 (2H, s, CH2Cl2) 4.42 (1H, CH2N, d, 14Hz)
4.36 (1H, CH2N, d, 14Hz) 4.00 (1H, CH2O, dd, 8.9, 3.6Hz), 3.81 (1H, CH2O, dd, 8.9, 3.6Hz), 3.52 (1H,
CHCH3, m), 3.15 (1H, HCN, m), 1.40 (3H, HCCH3, d, 4.8Hz).
2.2.3 Synthesis of {(S)-2-(dimethylamino)-1-(diphenylphosphino)-3-
(methylthio)propane}(PPh3)RuCl2 3: A solution of (S)-2-(dimethylamino)-1-(diphenylphosphino)-3-
(methylthio)propane (0.160g, 0.50mmol) and RuCl2(PPh3)3 (0.475g, 0.48mmol) in toluene (50mL) was
refluxed under nitrogen for 24h. The resulting yellow solution was cooled to room temperature and the
volume reduced to 5mL under vacuum. Addition of diethyl ether (20mL) gave a yellow precipitate of 3.
Yield: 0.328g, 85.7%. Anal. Calc. for 3: C 57.50 H 5.22 N 1.86%. Found: C 57.37 H 5.28 N 1.62%.
31P NMR, CD2Cl2, 162MHz, δ: 57.0 (1P, RuPPh2, d, 34Hz), 40.5 (1P, RuPPh3, d, 34Hz ): 1H NMR,
CD2Cl2, 400MHz, δ: 8.38-6.21 (25H, m, arom.) 2.95 (1H, m), 2.79 (3H, s, N(CH3)2) 2.72 (3H, s,
N(CH3)2), 2.48 (4H, m, CH2) 1.06 (3H, s, SCH3).
2.3.1 Transfer hydrogenation using isopropanol-isopropoxide: Catalytic reactions were carried out
under an atmosphere of dry nitrogen according to the following general method. To a solution of ketone
(1.945mmol) and catalyst (0.0255mmol) in 18.0 mL isopropanol was added 0.076 mL of a 0.040M
solution of potassium isopropoxide. The reaction was allowed to proceed at the required temperature for
the allotted time, at the end of which the solvent was removed under vacuum and the characterized using
1H NMR and GC-MS analysis.
6

ACCEPTED MANUSCRIPT
2.3.2 Transfer hydrogenation using dimethylamine borane: Catalytic reactions were carried out in air
according to the following general method. In a 2-dram screw-top vial, 1 mL of THF was added to a
mixture of catalyst (0.0065 mmol) and the substrate ketone (1.00 mmol). Separately, dimethylamine
borane (120 mg, 2.03 mmol) was dissolved in a second 1 mL aliquot of THF. The two solutions were
mixed in one vial and the screw cap placed on the vial. The reaction mixture was allowed to stand at the
appropriate temperature for the chosen duration whereupon the reaction was quenched by the sequential
addition of dichloromethane (1 mL) and water (1 mL). The organic layer was collected and dried over
solid magnesium sulfate, filtered and the volatile components removed by rotary evaporation. The
residue was analyzed by 1H NMR and GC-MS.
2.3.3 NMR Tube Reactions: Catalytic reactions were monitored by NMR using solutions of catalyst
(2.0 mg), 4-methoxy acetophenone (18mg) and dimethylamine borane (20mg) in THF-d8 (0.50 mL)
prepared in air.
2.4 X-Ray Crystallography: Low-temperature diffraction data (ω-scans) were collected on a Rigaku R-
AXIS RAPID diffractometer coupled to an R-AXIS RAPID imaging plate detector with Mo Kα
radiation (λ = 0.71073 Å) for the structure of 1.OPPh3.tol. The diffraction images were processed and
scaled using the Rigaku CrystalClear software. The structure was solved with SHELXT and was refined
against F2 on all data by full-matrix least squares with SHELXL.[39] All non-hydrogen atoms were
refined anisotropically. The hydrogen atoms were included in the model at geometrically calculated
positions and refined using a riding model. The isotropic displacement parameters of all hydrogen
atoms were fixed to 1.2 times the U value of the atoms to which they are linked, 1.5 times for methyl
groups. The crystal was not of high quality, and the reflections were subsequently affected. This fact is
probably the cause of the globally poor thermal parameters. The model reported here was restrained to
have internally similar thermal parameters; this treatment was accomplished with rigid bond restraints.
7

ACCEPTED MANUSCRIPT
The thermal parameters of carbon atoms C24 and C16 required additional constraints; their thermal
parameters were defining as equal to their nearest neighbor. CCDC number 1485849 contains the
supplementary crystallographic data for this structure.
For compound 2.OPPh3.CH2Cl2, the diffraction data were collected in the same way. The
hydrogen atoms were also included in the model at geometrically calculated positions and refined using
a riding model. The isotropic displacement parameters of all hydrogen atoms were fixed to 1.2 times the
U value of the atoms to which they are linked, 1.5 times for methyl groups. The only exceptions are H1
and H1a, which were semi-freely refined with the aid of distance restraints. Specifically, N1-H1a was
restrained to 0.93(2); O1-H1 to 0.96(2). The full numbering scheme of compound 2.OPPh3.CH2Cl2 can
be found in the full details of the X-ray structure determination (CIF), CCDC number 1485850.
For compound 4, Low-temperature diffraction data (ω-scans) were collected on a Rigaku SCX
Mini diffractometer coupled to a Rigaku Mercury275R CCD with Mo Kα radiation. The diffraction
images were processed and scaled using Rigaku Oxford Diffraction software. The structure was also
solved with SHELXT and was refined against F2 on all data by full-matrix least squares with
SHELXL.[39] All non-hydrogen atoms were refined anisotropically. Hydrogen atoms were included in
the model at geometrically calculated positions and refined using a riding model. The isotropic
displacement parameters of all hydrogen atoms were fixed to 1.2 times the U value of the atoms to
which they are linked (1.5 times for methyl groups). The only exceptions are the hydrogen atoms on the
nitrogen and bromine atoms. Two hydrogen bonds were refined between atoms N1 ... Cl1 and N2 ... Cl2
which converged to 3.1574(17) Å and 3.1557(17) Å, respectively. The full numbering scheme of
compound 4 can be found in the full details of the X-ray structure determination (CIF), CCDC number
1485851. These data can be obtained free of charge from The Cambridge Crystallographic Data Center
at www.ccdc.cam.ac.uk/data_request/cif.
8

ACCEPTED MANUSCRIPT
3.0 Results and Discussion
Complexes 1.OPPh3.tol and 2.OPPh3.CH2Cl2 were characterized by x-ray crystallography and the
structures are presented in Figures 1 and 2. The two structures are similar in that the tridentate ligand is
bound to the ruthenium(II) centre in a meridional fashion in each case with the two chloro ligands
disposed trans. The bonds lengths Ru-Nimine and Ru-Namine are quite similar to one another at 2.104(7)
and 2.173(5) Å respectively, as are the angles Ph2P-Ru-N, 88.8(2) and 91.01(13)o, and N-Ru-O, 79.1(3)
and 77.74(16)o, data that shows that there is little macroscopic structural difference on switching
between the imine and amine version of this ligand framework (Tables 1 and 2). The angle Cl-Ru-Cl is
noticeably compressed in 2 compared to 1 (167.13(5)o vs. 172.17(9)o) which is probably due to the
Figure 1. ORTEP diagram of the structure of 1.OPPh3.toluene with ellipsoids drawn at the 50%
probability level. Distance O(1)-O(2) 2.603 Å. See Table 1 for details of the structure determination
and Table 2 for selected bond lengths and angles
.
increased ruffling of the tridentate ligand. Neither of these compounds could be made to crystallize in
the absence of triphenylphoshine oxide, which acts in the solids as a hydrogen-bonding partner with the
9

ACCEPTED MANUSCRIPT
alcohol moiety of the ligand and as a result, compounds 1 and 2 were prepared and used as their
triphenylphosphine oxide adducts throughout this study.
Figure 2. ORTEP diagram of the structure of 2.OPPh3.CH2Cl2 with ellipsoids drawn at the 50%
probability level. Distance O(1)-O(2) 2.697 Å. See Table 1 for details of the structure determination
and Table 2 for selected bond lengths and angles.
3.1 Isopropanol as the hydrogen donor.
Both 1.OPPh3.tol and 2.OPPh3.CH2Cl2 proved competent catalysts for the transfer hydrogenation of
acetophenone using isopropanol-isopropoxide reaction conditions (Table 3) but compound 3 did not.
Using identical reaction conditions and 3 resulted in the immediate decomposition of the catalyst as
evidenced by the formation of a black precipitate of ruthenium metal and a very small conversion,
probably due entirely to the isopropoxide-driven reaction. Seemingly, 3 is not stable in the reaction
conditions and, in a related vein, a number of reports have appeared detailing 'base free' transfer
10

ACCEPTED MANUSCRIPT
Table 1. Crystallographic data and details of data collection and structure refinement for
1.OPPh3.toluene and 2.OPPh3.CH2Cl2.
______________________________________________________________________________
1.OPPh3.tol
2.OPPh3.CH2Cl2
______________________________________________________________________________
Empirical Formula
RuCl P O NC H
RuCl P O NC H
4 3 2 59 56
2 3 2
65 59
Formula Weight
Crystal System
Space Group
1151.01
Orthorhombic
1146.82
Monoclinic
P2 2 2 (#19)
P2 (#4)
1 1 1
1
a (Å)
b (Å)
c (Å)
10.6992(8)
12.2926(9)
41.627(3)
14.8086(10)
10.2131(7)
17.2537(12)
3
V (Å )
5474.8(7)
2600.72(8)
Z
4
2
90
94.698
90
α (o)
β (o)
γ (o)
90
90
90
3)
(g/cm
D
1.396
1.464
5.554
1180
0200 x 0.200 x 0.200
2.569 to 66.592
calc
Absorption coefficient
4.404 mm-1
2380
F
000
Crystal size (mm)
0200 x 0.200 x 0.200
Theta range for collection (o) 3.749 to 66.590
Index ranges
-7≤h≤12, -14≤k≤13, -35≤l≤49
26581
9102[R(int)0.0829]
Semi empirical from equivalents
-15≤h≤17, -12≤k≤12, -19≤l≤20
19436
Reflections collected
Independent reflections
Absorption correction
Max. and min. transmission 1.000, 0.785
Refinement method
Data/restraints/parameters
Goodness-of-fit on F2
Final R indices {I>2σI]
R indices (all data)
8492 [R(int) 0.0412]
Semi-empirical from equivalents
1.000, 0.667
Full-matrix least-squares on F2
9102/621/657
0.978
R1 = 0.0659, wR2 = 0.0821
R1 = 13.03, wR2 = 0.0959
Full-matrix least-squares on F2
8492/423/634
0.870
R1 = 0.0356, wR2 = 0.0521
R1 = 0.0495, wR2 = 0.0537
-0.012(5)
Absolute structure parameter 0.000(8)
Largest diff. peak and hole 0.670 and -0.604 eÅ3
0.914 and -0.626 eÅ3
-0.016
Flack parameter
0.000
11

ACCEPTED MANUSCRIPT
Table 2. Selected bond lengths (Å) and angles (o) for 1 and 2.
______________________________________________________________________________
1
2
______________________________________________________________________________
Ru-P(1)
2.2364(15)
2.239(3)
2.356(3)
2.104(7)
2.206(6)
2.403(2)
2.432(3)
Ru-P(2)
Ru-N
Ru-O
2.2364(15)
2.173(5)
2.207(4)
2.4278(14)
2.4079(14)
Ru-Cl
P(1)-Ru-N
P(2)-Ru-N
P(1)-Ru-O
P(2)-Ru-O
N-Ru-O
88.8(2)
91.01(13)
168.58(13)
168.72(11)
91.00(11)
77.74(16)
167.13(5)
170.1(3)
167.71(17)
92.51(17)
79.1(3)
Cl-Ru-Cl
172.17(9)
hydrogenations which may suggest that other catalysts have proven equally intolerant of reaction
conditions that require a base.[10,11] Repeating this reaction with the omission of isopropoxide resulted
in no product formation.
The highest activity is seen for reactions catalyzed by 2.OPPh3.CH2Cl2, the complex with the
reduced form of the ligand, (Table 3, entry 5) and the greatest selectivity was observed for the reaction
catalyzed by 1.OPPh3.tol at low temperature with low conversion (Table 3, entry 2). The pronounced
difference in both the efficiency and selectivity between 1.OPPh3.tol and 2.OPPh3.CH2Cl2 suggests that
there is no large concentration of a common reactive species formed by the two catalysts independently,
for example by the reduction of the imine in 1 forming 2 in situ. Additionally, it may well reflect the
differential ability of the two nitrogen donors to participate in hydrogen-shuttling as 2 contains a
secondary amine which has a proton available whereas 1 does not. Another factor to consider is the
increased flexibility of the tridentate ligand associated with the amine moiety in 2 compared to the imine
12

ACCEPTED MANUSCRIPT
in 1, which may lead to a lower barrier to rearrangement within the coordination sphere during the
reaction. This in turn may lead to a different geometry in the transition state and therefore a different
Table 3: Transfer hydrogenation of acetophenone using 2 mmol acetophenone, 1.3 mol% catalyst,
3 µmol potassium in 18 mL 2-propanol. Enantionmeric excess and conversion determined using GC-
MS and 1H NMR.
Entry
catalyst
1.OPPh3.tol
1.OPPh3.tol
1.OPPh3.tol
2.OPPh3.CH2Cl2
2.OPPh3.CH2Cl2
3
T (oC)
25
t (h) %Conversion
%ee
(rot. at 589nm)
nd
1
2
3
4
5
6
45
112
45
1.1
12
33
21
63
3
25
39.9 (+)
17 (+)
8.3 (-)
1.7 (-)
nd
82
25
45
82
45
82
45
product distribution. Repeating the reaction described in Table 3, entry three omitting 1.OPPh3.tol and
using instead 0.18 mmol of triphenylphosphine oxide resulted no formation of the product alcohol over
that seen in the blank reaction indicating that triphenylphosphine oxide alone does not promote this
transfer hydrogenation.
Both the selectivity and turnover frequency for 1.OPPh3.tol and 2.OPPh3.CH2Cl2 were found to
be very much reduced compared to the best catalysts reported.[12] A closely related complex of
ruthenium(II) prepared in situ by combining a ligand prepared from the condensation of 2-
(diphenylphosphino)benzaldehyde and (S)-phenylglycinol with tetrakis(dimethylsulfoxide)ruthenium(II)
chloride in a 1:4 ruthenium(II):ligand ratio is reported to show a turnover frequency (TOF) of 35/hour
13

ACCEPTED MANUSCRIPT
and an enantiomeric excess (ee) of 66%[29] in a reaction analogous to that reported here (Table 3, entry
3) that has a TOF of 0.56/hour and an ee of 17%. Though no direct comparison can be made since there
is an excess of the chiral ligand in the reaction reported by Kwong et al. and they isolate no discrete
complex, what is evident from these two examples is that changing the size of the group that invests the
ligand with chirality from a phenyl to a methyl alters the activity far more than it does the selectivity.
In the same paper, Kwong's group also reports a crystallographically-characterized 2:1 bis-κ3-
[P,N,O] ligand:ruthenium(II) complex of the same (S)-phenylglycinol-derived ligand that also catalyzes
the transfer hydrogenation of acetophenone under similar temperature and concentration conditions, in
this case with a TOF of 4/hour at 23h and an ee of 23%. This result is much closer to that seen for our
complex 1.OPPh3.tol and indeed there is evidence of extensive ligand scrambling in some of our
reactions (vide infra). They attribute the reduction in activity to the coordination of the alcohol group of
the ligand to the metal, which is also present in 1.OPPh3.tol and 2.OPPh3.CH2Cl2, but which they do not
anticipate in their in situ prepared catalyst. Similarly, the same reaction catalyzed by the complex
formed in situ between the [P,N,O] ligand prepared from the condensation of (R)-1-[(S)-2-
(diphenylphosphino)-ferrocenyl]ethylamine with 3,5-dinitro salicylaldehyde and
tris(triphenylphosphine)ruthenium(II) chloride in a 2:1 ligand:ruthenium ratio gave the corresponding
alcohol in 45% ee[24], again comparable to the selectivity of our catalysts (Table 3, entry 2). Equally,
the selectivity of the complex [fac-κ3-(R)-1-(diphenylphosphino)-2-((1R,2S,5R)-menthoxy)-1-(2-
pyridyl)- ethane-P,N,O)RuCl2(PPh3)] in the analogous reaction reported by Mathieu is also superior to
that reported here (60%) and their TOF is far greater (100 /h).[34]
3.2 Dimethylamine borane as the hydrogen donor.
Following on from the failure of 3 to survive the isopropanol-isopropoxide reaction conditions,
reactions using dimethylamine borane as the hydrogen donor modelled after those reported by
14

ACCEPTED MANUSCRIPT
Table 4: Transfer hydrogenation of acetophenone using 1mmol acetophenone, 0.67 mol% catalyst,
2mmol H3BNHMe2, 2.0mL THF, 48h. Enantionmeric excess and conversion determined using GC-MS
and 1H NMR.
Entry
catalyst
T (oC)
25
t (h) %Conversion
%ee
(rot. at 589nm)
30.6 (+)
1
2
3
4
1.OPPh3.tol
48
48
48
48
98
97
56
46
25
43.0 (-)
4.4 (-)
3.7 (-)
3
3
3
0
-25
Williams were examined (Table 4).[17,40] These reactions proved more successful, both in terms of the
activity and the selectivity of the catalysts 1.OPPh3.tol and 3. Catalyst 2.OPPh3.CH2Cl2 was not pursued
further in this study because of the low selectivity seen in the products of the isopropanol reactions.
The TOF of 1.OPPh3.tol was higher in these reactions (Table 4, entry 1) and the ee comparable
with the isopropanol/isopropoxide system (Table 3, entries 2 and 1) despite being carried out in air.
Compound 3 performed better than did 1.OPPh3.tol (Table 4, entry 2) proving in this case tolerant of the
reaction conditions, although the selectivity remained modest.
Lowering the temperature of the reaction did not lead to an increase in the selectivity, but rather
the reverse, as seen in Table 4, entries three and four and in Table 5, entries one through seven.
Precisely why this happens in unclear but it may reflect differences in how the rates of competing
reactions change with changing temperature, for example the uncatalyzed reduction of the ketone by the
dimethylamine borane or its dehydrogenated products. In blank reactions that did not contain catalyst
the hydrogenations proceeded to between five and eight percent completion. Repeating the
15

ACCEPTED MANUSCRIPT
Table 5: Transfer hydrogenation of ketones (a-e) using using 1mmol ketone, 0.67 mol% catalyst,
2mmol H3BNHMe2, 2mL THF, 48h. Enantionmeric excess and conversion determined using GC-MS
and 1H NMR. a 4'-(fluoro)acetophenone; b 4'-(methyl) acetophenone; c 4'-(methoxy)acetophenone; d
benzylacetone; e 2-chloro acetophenone.
Entry
1
catalyst
substrate T (oC) %Conversion
%ee
(rot. at 589nm)
11.4 (+)
1.OPPh3.tol
a
a
a
a
b
b
b
c
c
d
d
e
e
25
25
0
84.5
98.5
62
2
33.9 (-)
6.2 (-)
3
3
3
4
-25
25
25
0
28
0.4 (-)
3
5
1.OPPh3.tol
77
33.6 (+)
39.0 (-)
11.0 (-)
46.4 (+)
26.0 (-)
17.4 (+)
18.3 (-)
25.5 (+)
2.9 (-)
6
92
3
7
29
3
8
1.OPPh3.tol
25
25
25
25
25
25
39.5
54
9
3
10
11
12
13
1.OPPh3.tol
98
96.5
89
3
1.OPPh3.tol
>99
3
reaction in Table 4, entry one omitting 1.OPPh3.tol and using instead 0.18 mmol of triphenylphosphine
oxide leads to 54% conversion to the phenethyl alcohol. This is an approximately twenty-seven fold
16

ACCEPTED MANUSCRIPT
increase in the concentration of triphenylphosphine oxide compared to the reactions in Tables 4 and 5
resulting in a five-fold increase in the formation of the product with respect to the blank reaction. From
this it seems that the triphenylphosphine oxide present in 1.OPPh3.tol has a small direct influence on the
outcome of the reaction.
Williams reports[17] an enantiomeric excess of 76% for the transfer hydrogenation of
acetophenone using dimethylamine borane and [(R)-xylylBINAP][(R,R)-DPEN]RuCl2 as catalyst at
room temperature which, given the proven efficacy of the catalyst in other hydrogenations, might fairly
be viewed as an upper limit to the expected selectivity. It is perhaps of note that none of the transfer
hydrogenation reactions catalyzed by complexes of ruthenium with tridentate [P,N,O] ligands achieve
the selectivity seen for a range of comparable complexes with bidentate ligands.
The scope of the reaction was further explored using analogues of acetophenone and the results
are presented in Table 5. In all cases, the conversion achieved by 3 is greater than that of 1.OPPh3.tol
and, in most cases, the selectivity is better for 1.OPPh3.tol, although they are similar. The results of the
reductions of benzylacetone, entries ten and eleven, are the most consistent between the two catalysts
and this may be because in this substrate, of all the ketones examined, the two groups on the carbonyl
carbon are the least sterically and electronically differentiated. The substrates 4'-
(methoxy)acetophenone (entries eight and nine) and α-chloroacetophenone (entries twelve and thirteen)
show the greatest differences in selectivity. The former substrate is expected to be relatively
deactivated, which is seen in the low conversion with both catalysts, but with pronounced electronic and
steric asymmetry due to the differences between the two groups on the carbonyl carbon, whereas the
latter is far less obviously sterically differentiated. The variation in rates seen for all of these reactions is
quite marked, with the general trend of increasing TOF associated with increasing electron-donating
17

ACCEPTED MANUSCRIPT
character in the ketone. Previously, this trend in similar transfer hydrogenation reactions was taken as
evidence that ketone complexation by the catalyst is a rate-limiting step.[41]
The most effective catalysts for stereoselective transfer hydrogenation achieve far greater
selectivity and turnover frequency than those reported here, which prompts questions as to how the
reactions reported here proceed. In order to probe the mechanism of our reactions that use the
dimethylamine borane complex adduct, a number of simple experiments were carried out.
There is an extensive literature detailing the metal-catalyzed evolution of hydrogen from amine
borane complexes[42,43] and so one possible mode of action for 1.OPPh3.tol and 3 might involve a two-
step process in which the catalyst first liberates hydrogen from the dimethylamine borane before the
hydrogen is added to the ketone in a subsequent step. None of the catalysts were found to be active
catalysts for the direct hydrogenation of acetophenone in THF solution using pressures of hydrogen gas
as high as twenty-five atmospheres and reaction times of twenty-four hours, which would suggest that
such a coupled reaction is not contributing to the observed reaction.
The reactions reported in entries one and two of Table 4 were repeated using two separate
reaction chambers within a larger vessel, one containing a solution of dimethylamine borane complex
and the other a solution of the catalyst and the ketone, under which conditions no product formation was
observed. Similarly, running the same reaction again using two separate reaction chambers within a
larger vessel, one containing a solution of dimethylamine borane complex with the catalyst and the other
a solution of the catalyst and the ketone also resulted in no product formation.[44] These results indicate
that all of the reaction components need to be in contact in order for the reaction to proceed.
Following the reaction of 4'-(methoxy)acetophenone, dimethylamine borane complex and 3 by
1H NMR in THF-d8 solution revealed some interesting features. Upon mixing dimethylamine borane
and 3, a resonance in the upfield region typical of a metal hydride of intensity close to one proton very
18

ACCEPTED MANUSCRIPT
quickly appears at -6.25 ppm as a doublet of doublets (J = 133, 29Hz), probably indicative of coupling
to two inequivalent phosphorus nuclei, Figure 4. From the coupling it seems likely that one phosphorus
is close to a cis and the other to a trans position with respect to the hydride. This could imply transfer of
a single hydride to the metal or the κ1 coordination of a
Figure 4. 1H NMR spectrum (THF-d8, 400MHz) of compound 3 with Me2NHBH3 at 24h.
borohydride group. Examination of the spectrum in the region 0.00 - -2.00ppm does not reveal any
quadrupolar broadened resonances such as those observed by Williams et al. in the spectra of complexes
containing the [Ru…H…BH2-N] group[40] and so the former explanation of the observation seems the
19

ACCEPTED MANUSCRIPT
more likely. The chemical shift is somewhat downfield with respect to typical ruthenium(II) hydrides
and may indicate that the hydride occupies a bridging position.
Upon addition of 4'-(methoxy)acetophenone, the reaction begins and, after 24h, the presence of
both the ketone and the product alcohol is confirmed by the 1H NMR spectrum of the reaction mixture
while the hydride resonance remains at the same chemical shift, Figure 5. At the same time point, the
1
Figure 5 H NMR spectrum (THF-d8, 400MHz) of compound 3 (pale blue) and compound 3 with
Me2NHBH3 and 4'(methoxy)acetophenone (maroon).
31P NMR spectrum shows predominantly an AB pattern (δ = 67.9, 23.0, J = 12Hz) which is significantly
different from that of 3. At forty-eight hours, the hydride resonance in the 1H NMR has shifted to -
9.95ppm and is a doublet of doublets with couplings of 31 and 22Hz, indicative of two inequivalent cis
20

ACCEPTED MANUSCRIPT
phosphorus nuclei, Figure 6. This suggests a change in the conformation of the catalyst as the reaction
proceeds and indeed the 31P{1H}NMR confirms this, showing at forty-eight hours two AB sets, one at
78.6 and 72.8ppm with a coupling of 34Hz and another more abundant set at 45.3 and 43.5ppm with a
coupling of 29Hz of relative intensity 1:2.1 respectively. This data indicates a significant scrambling of
1
Figure 6. H NMR spectrum (THF-d8, 400MHz) of compound 3 with Me2NHBH3 and 4-MeO-
acetophenone at 48h.
the ligands to form two distinct complexes, and based upon the chemical shift information, it seems that
the the [PPh2] moiety, and therefore the chiral ligand, is at forty-eight hours in a bis-ligand complex with
some of the ruthenium centres, with the balance of the ruthenium complexes containing two
21

ACCEPTED MANUSCRIPT
triphenylphosphine ligands. If both species are catalytically active, it certainly would go some way to
explaining the limited enantiomeric excess seen in the reactions catalyzed by 3. By analogy to the bis-
complex reported by Kwong et al.(vide supra), a reduced selectivity might be the expected outcome of
any reaction catalyzed by such a complex whereas a catalytically competent species that no longer
comprised the chiral ligand would surely make racemic product. Superposition of the spectrum of 3 and
this mixture, Figure 5, reveals that the resonance at 1.06 ppm due to the methyl protons of the
coordinated thioether is no longer present in the spectrum of the reaction mixture suggesting
displacement of the thioether from the ruthenium in the reaction.
It has previously been observed that a colourless or white precipitate is thrown down in the
course of transfer hydrogenations of ketones that use amine boranes as the hydrogen donor. In the
course of several of our reactions a solid was seen to precipitate, in some cases as well-formed
colourless crystals. Crystallographic analysis of these showed the material to be
[Me2NHBH2NHMe2]Cl, 4, Figure 7. The structure of 4 has been reported[45] but the original report
gives no positional parameters, a deficiency we have sought to correct here.
The finding that 4 could be isolated from some of these reactions, possibly formed in a reaction
similar to that proposed by Berke et al. for a rhenium-catalyzed reaction[18] and Conojero et al. for a
platinum-driven dehydrocoupling of dimethylamine borane,[46] prompted an examination of the 11B
NMR spectra recorded on the reactions. The only significant peaks seen were assigned to either the
starting dimethylamine borane or the compound usually identified as (Me2NBH2)2, the expected product
of the dehydrogenation. Attempts to redissolve 4 in THF-d8 were not very successful however and so it
is proposed that even though this compound was formed it was not evident from the 11B NMR spectra
due to its low solubility. Repeating the transfer hydrogenation of acetophenone using dimethylamine
borane catalyzed by (para-cymene)RuCl2[17] we were also able to collect well-formed crystals of 4
22

ACCEPTED MANUSCRIPT
indicating that it may be a product in many such reactions. Indeed, the reaction of dimethylamine
borane with a metal catalyst to give a metal hydride and a compound similar to 4 has been proposed in a
recent publication that examined the mechanism of iron-catalyzed dehydrocoupling and
dehydrogenation of amine borane complexes.[47]
Figure 7. ORTEP diagram of the structure of 4 with ellipsoids drawn at the 50% level. Monoclinic;
P21/c; a = 5.904, b = 10.852, c = 13.899 Å; β = 99.76o; M = 138.448 g/mole; V = 877.62 Å3; Z = 4;
Bond Lengths N1-B1 1= .588; N2-B1 = 1.576; N2-Cl1 3.156 Å; Angles N2-B1-N1 110.81o; N-Cl1-N =
96.451o.
Interestingly, following the corresponding reaction catalyzed by 1.OPPh3.tol by 1H NMR reveals
no sharp hydride resonances but instead broad features at -2.2, -6.0 and -10ppm which may indicate the
coordination of a borohydride or of a hydrogen molecule, Figure 8.[48] The NMR tube reaction
catalyzed by 1.OPPh3.tol stopped at 39.5% conversion of the starting ketone because, at that point, the
dimethylamine borane complex was completely consumed. Since the reaction begins with an excess of
dimethylamine borane with respect to the ketone it seems that dehydrogenation of the amine borane
23

ACCEPTED MANUSCRIPT
1
Figure 8. H NMR spectrum (THF-d8, 400MHz) of compound 1.OPPh3.tol and Me2NHBH3 at 24h.
without concommitant transfer of the hydrogen to the ketone is occurring. The rate of hydrogen
evolution observed with all of the complexes studied here is low with respect to many reported
examples.[49] This observation, taken with the absence of conversion in the direct catalytic
hydrogenation reactions, further argues against a two-stage dehydrogenation-ketone hydrogenation
mechanism. The 31P{1H}NMR of the reaction mixture at the end of the reaction shows singlets at
75.0ppm and 63.0ppm in addition to the two doublets associated with the starting complex, again
suggestive of an extensive redistribution of the ligands during the reaction.
24

ACCEPTED MANUSCRIPT
Both catalysts begin as coordinatively saturated complexes and it seems likely from the
formation of 4 that at least one chloro ligand is exchanged for a hydride during the course of the
catalysis. It is also clear from the 31P{1H}NMR spectra that there is extensive ligand scrambling in the
course of the reaction and the 1H NMR indicates that decomplexation of one or more of the donor atoms
takes place under the reaction conditions. Since there is a correlation between the TOF and the electron-
donating character of the carbonyl oxygen of the ketone, it may be that such partial dissociation of the
ligand is required to allow coordination of the substrate ketone during the reduction. Recently,
examples of bimetallic rhodium complexes have been shown to form in the activation of amine
boranes.[50,51] The dimer [Rh2(Pcy3)2(H)2(µ-H2BNMe3)2(µ-H3B·NMe3)][BArF4]2 has been isolated
from the reaction between trimethylamine borane and [Rh(Pcy3)(binor-R)][BArF4][50] which is
suggestive since the formation of a dimeric complex in our system could also promote the observed
ligand scrambling.
If it is the case that the selectivity is affected by ligand scrambling then it follows that the ee
should change during the course of the reaction and further that a higher ee should be observed at an
earlier time in the reaction. Intercepting the reaction in Table 5, entry nine at 180 minutes shows a
conversion of 26.6% and an ee of 76.3% which supports this interpretation, at least qualitatively.
The ruthenium catalyzed transfer hydrogenations reported here seem to follow complex
mechanisms with both TOF and ee controlled by the interplay of several competing processes. The
tridentate ligands are labile under the reaction conditions leading to mixtures of species in the reactions,
each of which may be influencing the conversion and selectivity differently. The ability of any ligand to
influence stereoselection in a catalyst depends upon conserving conformation in the complex and that
stability is absent in these complexes. It is likely that the catalytically competent species require partial
dissociation of the tridentate ligand in order to coordinate the ketone, as evidenced by the different TOF
25

ACCEPTED MANUSCRIPT
observed on changing the ketone, or hydride or borohydride, as evidenced by the 1H NMR studies of the
reactions. This partial deconplexation opens pathways to complete dissociation and ligand scrambling,
which is in contrast to complexes of bidentate ligands which allow for the coordination of substrates or
hydride with no requirement for dissociation of the chiral ligand.
In the case of the reactions using isopropanol as the hydrogen donor the complex of the ligand
containing the imine nitrogen donor showed greater selectivity and lower activity than the corresponding
complex with the secondary amine nitrogen, which is entirely consistent with a mechanism such as
Noyori proposed, whereas the complex of the ligand with the tertiary amine nitrogen donor was
ineffective. With the reactions in Tables 4 and 5, however, both the imine- and tertiary amine-
containing complexes were active catalysts, suggesting that a different mechanism is in effect when the
hydrogen donor is changed.
4.0 Conclusion
The tridentate ligands used in this study are relatively inexpensive to prepare, using some of the
least costly homochiral compounds available in their synthesis, but the low cost of these ligands may not
be sufficient compensation for the comparatively low selectivity and activity shown by ruthenium(II)
complexes of these ligands when used as catalysts for transfer hydrogenation. There is evidence of
ligand dissociation and ligand scrambling under the reaction conditions which must play a significant
part in reducing the efficacy of these catalysts.
Although the ee observed in the reduction of α-chloroacetophenone is not large, the relatively
high TOF is of note (Table 5, entries 12 and 13). The alcohol formed has, according to the CDC NIOSH
database, very low toxicity and no particular hazardous properties associated with it, whereas the
starting ketone is widely used as a chemical weapon, being a major component of CN gas, and as such
26

ACCEPTED MANUSCRIPT
has been manufactured in large quantities and stockpiled worldwide. An efficient method for the
conversion of this powerful lachrymator to the alcohol such as that reported here might point the way to
a useful method for the destruction of CN gas stockpiles.
5.0 Acknowledgement
The authors thank the Department of Chemistry, Yale University and the STARS program for financial
support.
27

ACCEPTED MANUSCRIPT
6.0 References
[1]
[2]
[3]
D. Wang, D. Astruc, Chemical Reviews 115 (2015) 6621.
J. Ito, H. Nishiyama, Tetrahedron Letters 55 (2014) 3133.
Y.M.Y. Haddad, J. Husbands, H.B. Henbest, T.R. Mitchell, Proceedings of the Chemical Society
of London (1964) 361.
[4]
[5]
Y.D. Ju, L.W. Xu, L. Li, G.Q. Lai, H.Y. Qiu, J.X. Jiang, Y.X. Lu, Tetrahedron Letters 49 (2008)
6773.
A. Fujii, S. Hashiguchi, N. Uematsu, T. Ikariya, R. Noyori, Journal of the American Chemical
Society 118 (1996) 2521.
[6]
[7]
[8]
X.F. Wu, J.L. Xiao, Chemical Communications (2007) 2449.
R. Noyori, T. Ohkuma, Angewandte Chemie-International Edition 40 (2001) 40.
K. Mikami, T. Korenaga, M. Terada, T. Ohkuma, T. Pham, R. Noyori, Angewandte Chemie-
International Edition 38 (1999) 495.
[9]
J.S.M. Samec, J.E. Backvall, P.G. Andersson, P. Brandt, Chemical Society Reviews 35 (2006)
237.
[10] L.T. Ghoochany, C. Kerner, S. Farsadpour, F. Menges, Y. Sun, G. Niedner-Schatteburg, W.R.
Thiel, European Journal of Inorganic Chemistry 2013 (2013) 4305.
[11] M.C. Carrion, F. Sepulveda, F.A. Jalon, B.R. Manzano, A.M. Rodriguez, Organometallics 28
(2009) 3822.
[12] R. Noyori, M. Yamakawa, S. Hashiguchi, Journal of Organic Chemistry 66 (2001) 7931.
[13] R. Noyori, S. Hashiguchi, Accounts of Chemical Research 30 (1997) 97.
[14] M. Couturier, B.M. Andresen, J.L. Tucker, P. Dube, S.J. Brenek, J.T. Negri, Tetrahedron Letters
42 (2001) 2763.
[15] M. Couturier, J.L. Tucker, B.M. Andresen, E. Dube, S.J. Brenek, J.T. Negri, Tetrahedron Letters
42 (2001) 2285.
[16] N. Blaquiere, S. Diallo-Garcia, S.I. Gorelsky, D.A. Black, K. Fagnou, Journal of the American
Chemical Society 130 (2008) 14034.
[17] T.D. Nixon, M.K. Whittlesey, J.M.J. Williams, Tetrahedron Letters 52 (2011) 6652.
[18] Y.F. Jiang, O. Blacque, T. Fox, C.M. Frech, H. Berke, Organometallics 28 (2009) 5493.
[19] E.M. Leitao, T. Jurca, I. Manners, Nature Chemistry 5 (2013) 817.
[20] A. Rossin, M. Peruzzini, Chemical Reviews 116 (2016) 8848.
[21] B. Crociani, S. Antonaroli, M. Burattini, P. Paoli, P. Rossi, Dalton Transactions 39 (2010) 3665.
[22] J. Wencel, I. Laurent, L. Toupet, C. Crevisy, M. Mauduit, Organometallics 29 (2010) 1530.
[23] P. Pelagatti, A. Bacchi, M. Balordi, S. Bolano, F. Calbiani, L. Elviri, L. Gonsalvi, C. Pelizzi, M.
Peruzzini, D. Rogolino, European Journal of Inorganic Chemistry (2006) 2422.
[24] H.C. Dai, X.P. Hu, H.L. Chen, C.M. Bai, Z. Zheng, Tetrahedron-Asymmetry 14 (2003) 1467.
[25] J.W. Faller, G. Mason, J. Parr, Journal of Organometallic Chemistry 626 (2001) 181.
[26] J. Parr, A.M.Z. Slawin, Inorganica Chimica Acta 303 (2000) 116.
[27] P. Bhattacharyya, M.L. Loza, J. Parr, A.M.Z. Slawin, Journal of the Chemical Society-Dalton
Transactions (1999) 2917.
[28] H.L. Kwong, L.S. Cheng, W.S. Lee, Journal of Molecular Catalysis a-Chemical 150 (1999) 23.
[29] H.L. Kwong, W.S. Lee, T.S. Lai, W.T. Wong, Inorganic Chemistry Communications 2 (1999)
66.
[30] P. Bhattacharyya, J. Parr, A.M.Z. Slawin, Journal of the Chemical Society-Dalton Transactions
(1998) 3609.
28