Synthesis and catalytic activity of new, water-soluble mono- and dinuclear ruthenium(II) complexes containing 1,3,5-triaza-7-phosphaadamantane: Study of the effect of the visible light

Article abstract of DOI:10.1016/j.ica.2017.04.054

The newly synthesized mer-trans-[RuCl₂(OH₂)(PTA)₃] (1) is the first compound isolated in solid state containing the Ru(PTA)₃-fragment (PTA = 1,3,5-triaza-7-phosphatricyclo[3.3.1.1^3,7]decane). Dissolution of 1 in aqueous HCl leads to mer-[RuCl₃(HPTA)₃]Cl₂ (2) which is stable in the dark but is transformed into fac-[RuCl₃(HPTA)₃]Cl₂ (3) under visible light. Irradiation with visible light of an aqueous solution of 1 at room temperature or refluxing of the same solution in the dark leads to the formation of [{Ru(PTA)₃}₂(μ-Cl)₃]Cl (4). The dinuclear complex 4 was also formed upon irradiation of solutions of PTA and various Ru(II)-complexes ([RuCl₂(DMSO)₄], [{RuCl₂(η⁶-p-cymene)}₂]) or cis-cis-trans-[RuCl₂(DMSO)₂(PTA)₂]). All complexes were characterized by elemental analysis and NMR spectroscopy, furthermore solid state structures of 2·1.25H₂O, 3·HCl·2H₂O and 4·9H₂O were also determined by single crystal X-ray diffraction. We have investigated the influence of the above photochemical processes on reduction of benzaldehyde and cinnamaldehyde with trans-[RuCl₂(PTA)₄] and cis-cis-trans-[RuCl₂(DMSO)₂L₂] (L = PTA, (PTA-Me)CF₃SO₃, (PTA-Bn)Cl; Me = methyl, Bn = benzyl) complexes as catalysts. The effect of visible light on benzonitrile hydration with various Ru(II)-PTA catalysts is also reported.

Full text of DOI:10.1016/j.ica.2017.04.054

Accepted Manuscript
Research paper
Synthesis and catalytic activity of new, water-soluble mono- and dinuclear
ruthenium(II) complexes containing 1,3,5-triaza-7-phosphaadamantane: study
of the effect of the visible light
Antal Udvardy, Manuel Serrano-Ruiz, Vincenzo Passarelli, Evelin Bolyog-
Nagy, Ferenc Joó, Ágnes Kathó, Antonio Romerosa
PII:
S0020-1693(17)30143-3
http://dx.doi.org/10.1016/j.ica.2017.04.054
ICA 17559
DOI:
Reference:
Inorganica Chimica Acta
To appear in:
Received Date:
Revised Date:
Accepted Date:
30 January 2017
26 April 2017
27 April 2017
Please cite this article as: A. Udvardy, M. Serrano-Ruiz, V. Passarelli, E. Bolyog-Nagy, F. Joó, A. Kathó, A.
Romerosa, Synthesis and catalytic activity of new, water-soluble mono- and dinuclear ruthenium(II) complexes
containing 1,3,5-triaza-7-phosphaadamantane: study of the effect of the visible light, Inorganica Chimica Acta
(2017), doi: http://dx.doi.org/10.1016/j.ica.2017.04.054
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.

1
Synthesis and catalytic activity of new, water-soluble mono- and dinuclear
ruthenium(II) complexes containing 1,3,5-triaza-7-phosphaadamantane:
study of the effect of the visible light.
Antal Udvardy,a Manuel Serrano-Ruiz,b Vincenzo Passarelli,b Evelin Bolyog-Nagy,a Ferenc
Joó,a,c Ágnes Kathóa* and Antonio Romerosab*
a Department of Physical Chemistry, University of Debrecen, P.O. Box 400, H-4002, Hungary
b Área de Química Inorgánica-CIESOL, Facultad de Ciencias, Universidad de Almería, 04120
Almería, Spain
c
MTA-DE Homogeneous Catalysis and Reaction Mechanisms Research Group, P.O. Box
400, H-4002, Hungary
Abstract
The newly synthesized mer-trans-[RuCl2(OH2)(PTA)3] (1) is the first compound isolated in
solid
state
containing
the
Ru(PTA)3-fragment
(PTA
=
1,3,5-triaza-7-
phosphatricyclo[3.3.1.13,7]decane). Dissolution of 1 in aqueous HCl leads to mer-
[RuCl3(HPTA)3]Cl2 (2) which is stable in the dark but is transformed into fac-
[RuCl3(HPTA)3]Cl2 (3) under visible light. Irradiation with visible light of an aqueous
solution of 1 at room temperature or refluxing of the same solution in the dark leads to the
formation of [{Ru(PTA)3}2(-Cl)3]Cl (4). The dinuclear complex 4 was also formed upon
irradiation of solutions of PTA and various Ru(II)-complexes ([RuCl2(DMSO)4], [{RuCl2(η6-
p-cymene)}2]) or cis-cis-trans-[RuCl2(DMSO)2(PTA)2]). All complexes were characterized
by elemental analysis and NMR spectroscopy, furthermore solid state structures of
2·1.25H2O, 3·HCl·2H2O and 4·9H2O were also determined by single crystal X-ray
diffraction. We have investigated the influence of the above photochemical processes on
reduction of benzaldehyde and cinnamaldehyde with trans-[RuCl2(PTA)4] and cis-cis-trans-
[RuCl2(DMSO)2L2] (L=PTA, (PTA-Me)CF3SO3, (PTA-Bn)Cl; Me=methyl, Bn=benzyl)
complexes as catalysts. The effect of visible light on benzonitrile hydration with various
Ru(II)-PTA catalysts is also reported.

2
Keywords
ruthenium, visible light, water, hydrogenation, hydration, 1,3,5-triaza-7-phosphaadamantane
(PTA)
Introduction
Water is an environment-friendly solvent and has gained increasing application recently in
organic synthesis and in homogeneous catalysis.[1-5] Water-soluble transition metal complex
catalysts frequently contain tertiary phosphine ligands,[6] although hydrosoluble N-
heterocyclic carbenes,[7-10] salens and salans[11] also receive attention. Among tertiary
phosphines, monosulfonated and trisulfonated triphenylphosphines, usually as their sodium
salts (mtppms-Na[12,13] and mtppts-Na3,[14,15] respectively) are the most commonly used. In
addition,
the
small
aliphatic
phosphine,
PTA[16-18]
(1,3,5-triaza-7-
phosphatricyclo[3.3.1.13,7]decane) and its various derivatives have often been used for
synthetic and catalytic purposes. Due to the importance of aqueous organometallic catalysis in
green chemical processes, a wide range of water-soluble catalysts with ruthenium, rhodium,
iridium, palladium, and other metal ions has been investigated.[1-21] The present study is
focused on new mono- and dinuclear chloro-Ru(PTA)-complexes prepared from various
Ru(II)-complexes such as [RuCl2(PPh3)3], cis-[RuCl2(DMSO)4] and [{RuCl2(η6-p-
cymene)}2].
The first PTA-containing Ru(II)-complex, trans-[RuCl2(PTA)4] was synthesized by
refluxing ethanolic solutions of PTA and hydrated RuCl3 or reacting aqueous solutions of
PTA (nPTA:nRu≥4) with toluene solutions of [RuCl2(PPh3)3].[16]
Scheme 1. Water-soluble phosphines used in this study
In the presence of Na-formate, trans-[RuCl2(PTA)4] proved catalytically active both in
transfer hydrogenation of aldehydes[16,22] and in redox isomerization of allylic alcohols.[23]
Conversely, reduction of aldehydes by complexes with the general formula cis-cis-trans-
[RuCl2(DMSO)2L2] (L=PTA, (PTA-Me)CF3SO3, (PTA-Bn)Cl; Me=methyl, Bn= benzyl) is

3
unexplored. The latter complexes showed at least twice as high catalytic activities than trans-
[RuCl2(PTA)4] in transformation of oct-1-en-3-ol to octan-3-one. These catalysts were
obtained in reactions of the water-soluble cis-[RuCl2(DMSO)4] with PTA and its N-alkylated
derivatives and were applied either synthesized in situ or as isolated solid compounds.[23]
Light-sensitivity of cis-[RuCl2(DMSO)4] is well known[24] and therefore these catalytic redox
isomerization reactions were run with strict exclusion of light unlike several earlier
studies[16,22,25] in which no attention was paid to possible photochemical effects.
It has been reported that upon irradiation trans-[RuCl2(PTA)4] can be isomerized to cis-
[RuCl2(PTA)4] and the latter can be aquated to cis-[RuCl(OH2)(PTA)4]+.[26,27] The
photoactivity of trans-[RuCl2(PTA)4] brings up the question of the influence of visible light
irradiation on the rate or selectivity of the reactions catalyzed by this complex.
Trans-[RuCl2(PTA)4] efficiently catalyzed the hydration of various nitriles, as well.[28]
We have also reported on the catalytic application of cis-cis-trans-[RuCl2(DMSO)2L2]
complexes and that of trans-[RuCl2(PTA)4] in hydration of various nitriles,[29,30] however,
these studies did not include investigation of photochemical effects.
In several cases[16,22,25,29,30] the catalytically active species was presumed to be a
Ru(II)-trisphosphine complex, however to date such species have not been isolated in pure
form. Earlier, [Ru(OH2)3(PTA)3]2+ (formed in the reaction of [Ru(OH2)6](tos)2 and PTA) was
characterized in aqueous solution[21] and very recently cis-mer-[Ru(S)Cl2(PTA)3] complexes
were identified in refluxing solutions of trans-[RuCl2(PTA)4] using S = DMSO or CH3CN as
solvents.[27]
We deemed important the synthesis and characterization of Ru(II)-PTA complexes
with a PTA:Ru=3:1 ratio including also the study of their photoactivity. In addition, we also
investigated the hitherto unexplored photochemistry of cis-cis-trans-[RuCl2(DMSO)2L2]
complexes. This paper describes detailed synthetic investigations on the interconversions of
various Ru(II)-PTA complexes under visible light irradiation together with structural
characterization of the products both in solution and in the solid state. It is also essential to
learn, how these photochemical processes may influence the catalytic reactions. With this aim
in mind we have investigated the reduction of benzaldehyde and cinnamaldehyde both in the
dark and under light both with trans-[RuCl2(PTA)4] and cis-cis-trans-[RuCl2(DMSO)2L2]
complexes as catalysts. The effect of visible light on benzonitrile hydration with various
Ru(II)-PTA catalysts is also reported.

4
Experimental Section
All reactions were carried out under an N2 atmosphere using standard Schlenk techniques.
Solvents were dried and deoxygenated under nitrogen/vacuum before use. Doubly distilled
water was used throughout. PTA,[31] (PTA-Me)CF3SO3,[32]
(PTA-Bn)Cl,[33] trans-
[(DMSO)3Ru(μ-Cl)3RuCl(DMSO)2],[35]
p-isopropyltoluene),[36]
cis-cis-trans-
[RuCl2(PTA)4],[16]
[{RuCl2(η6-p-cymene)}2]
cis-[RuCl2(DMSO)4],[34]
(p-cymene
=
[RuCl2(DMSO)2L2] [23] were prepared according to the literature. UV-Vis spectra of 10-3-10-5
M solutions were recorded on a Jasco V-650 spectrometer in quartz cells (10 mm optical
path). Specific absorption coefficients were obtained by linear regression over 5-7 points (R2
≥ 0.999). When necessary, the bands were analysed by the Spectra Manager software of
JASCO. Elemental analyses (C, H, N) were performed on a Fisons Instrument EA 1108
elemental analyzer or on an Elementar Vario Micro (CHNS) equipment. NMR spectra were
recorded on a Bruker AVANCE DRX300 spectrometer operating at ca. 300 MHz (1H), ca.
75.47 MHz (13C) and ca. 121 MHz (31P), respectively and on a BRUKER AVANCE 360
spectrometer operating at 360 MHz (1H), ca. 95 MHz (13C) and ca. 145 MHz (31P). Peak
positions were referenced to TMS or 3-(trimethylsilyl)-1-propanesulfonic acid Na-salt, DSS
(1H and 13C), H3PO4 (0.1 M in D2O) with downfield values taken as positive and were
measured at 293 K (if not stated otherwise). Gas chromatographic measurements were made
on an Agilent 7890A gas chromatograph with Chrompack WCOT Fused Silica
30 m×32 mm×0.25 m CP WAX52CB, column, injector temperature 250 °C, FID (300 °C).
Photolysis
Irradiation of the NMR tubes with broad-band UV-visible light was carried out using a home-
made photo-reactor with a built-in standard 150 W halogen lamp[37] and a LOT-Oriel 150 W
Xenon lamp. Irradiation at selected wavelengths was carried out with the Xenon system
equipped with a water filter (5 cm) and appropriate band-pass or interference filters (Schott).
Solutions for photolysis and reactions in the dark were prepared under Ar and transferred to
10 mm path-length quartz cells. During photolysis the solutions were stirred using a small
magnetic stirring bar inside the cell.
Synthesis of mer-trans-[RuCl2(OH2)(PTA)3]·3H2O (1·3H2O)
A solution of PTA (183.2 mg, 1.166 mmol) in 50 mL of water was mixed with a solution of
[RuCl2(PPh3)3] (372.2 mg, 0.388 mmol) in 50 mL of CH2Cl2 and stirred at room temperature
in the dark overnight. The water phase was separated, filtered and evaporated to dryness to

5
give a yellow powder which was washed with acetone (2 mL) and Et2O (6x6 mL) and finally
dried under vacuum. Yield: 168.1 mg (65%). S25°C(H2O) = 190 mg/mL.
Elemental analysis: Found C, 30.03; H, 5.85; N, 17.16 %; C18H38N9Cl2OP3Ru·3H2O
(715.53) calculated C, 30.22; H, 6.20; N, 17.62 %.
UV-vis λmax(H2O)/nm (ε/dm3 mol-1 cm-1): 457 (199), 336 (914).
IR (KBr): max/cm-1 3434b (OH), 2930 and 2873m (CH), /cm-1 1706w (OH).
1H NMR (300 MHz, D2O):  (ppm) 3.94 (bs, 6H, PCH2N), 4.24 (bs, 12H, PCH2N),
4.49 (bs, 6H, NCH2N), 4.55 (bs, 12H, NCH2N).
1
13C{1H} NMR (75.47 MHz, D2O):  (ppm) 48.27 (t, JCP = 6.4 Hz, PCH2N, Ptrans-O),
53.74 (d, 1JCP = 19.1 Hz, PCH2N, Ptrans-P), 70.62 (d, 3JCP = 6.3 Hz, NCH2N, Ptrans-P), 70.78, (t,
3JCP = 2.4 Hz, NCH2N, Ptrans-O
)
2
31P{1H} NMR (121 MHz, D2O):  (ppm) -5.08 (t, JPP = 34.6 Hz, Ptrans-O, 1 P),
-46.78 (d, 2JPP = 34.6 Hz, Ptrans-P, 2 P).
ESI-MS(+) (H2O): m/z=644.072, [M-H2O]H+ ([RuCl2P3N9H37C18]+) calculated
644.081.
Synthesis of mer-[RuCl3(HPTA)3]Cl2 ·2H2O (2·2H2O)
This complex was obtained by reaction of 1·3H2O (200 mg, 0.280 mmol) dissolved in 5 mL
aqueous 1 M HCl solution. After 15 min at room temperature the resulting bright yellow
solution was evaporated to dryness and the powder obtained was washed with EtOH (2x3
mL), Et2O (2x3 mL) and vacuum dried. Yield 185 mg (83 %).
Elemental analysis: Found C, 27.26; H, 6.12; N, 15.89 %; C18H39Cl5N9P3Ru·2H2O
(788.85) calculated C, 27.41; H, 5.49; N, 15.98 %.
IR (KBr): max/cm-1 3480 and 3377b (OH), 2979 and 2815m (CH), 2726-2071m
(NHPTA), /cm-1 1706w (OH).
NMR data of aqueous solutions of mer-[RuCl3(HPTA)3]Cl2 (2):
Two sets of NMR signals are detected in aqueous solutions of 2: mer-[RuCl3(HPTA)3]2+ and
mer-[RuCl2(OH2)(HPTA)3]3+ are in equilibrium
NMR data of mer-[RuCl3(HPTA)3]2+
1H NMR (300 MHz, D2O):  (ppm) 4.16 (bs, NCH2Ptrans-Cl, 6 H), 4.46 (bs, NCH2Ptrans-
P, 12 H), 4.85 (bs, NCH2Ntrans-P, 12 H), 4.84 (bs, NCH2Ntrans-Cl, 6 H).
1
13C{1H} NMR (75.47 MHz, D2O):  (ppm) 47.85 (t, JCP = 6.9 Hz, NCH2Ptrans-Cl),
52.14 (d, 1JCP = 17.2 Hz, NCH2Ptrans-P).

6
31P{1H} NMR (121 MHz, D2O):  (ppm) -4.62 (t, 2JPP = 33.3 Hz, Ptrans-Cl, 1 P), -40.62
(d, 2JPP = 33.3 Hz, Ptrans-P, 2P).
NMR data of mer-[RuCl2(OH2)(HPTA)3]3+:
1H NMR (300 MHz, D2O):  (ppm) 4.12 (bs, NCH2PtransO, 6 H), 4.40 (bs, NCH2Ptrans-
P, 12 H), 4.83 (bs, NCH2Ntrans-O, 6 H), 4.89 (bs, NCH2Ntrans-O, 6 H).
1
13C{1H} NMR (75.47 MHz, D2O):  (ppm) 46.83 (t, JCP = 6.5 Hz, NCH2Ptrans-O) ,
1
3
52.02 (d, JCP = 19.3 Hz, NCH2Ptrans-P), 70.84 (d, JCP = 5.60 Hz, NCH2Ntrans-P), 70.90 (bs,
NCH2Ntrans-O).
31P{1H} NMR (121 MHz, D2O):  (ppm) 2.31 (t, 2JPP = 35.6 Hz, Ptrans-O, 1 P), -36.28
(d, 2JPP = 35.6 Hz, Ptrans-P, 2 P).
UV-vis λmax(H2O)/nm (ε/dm3 mol-1 cm-1): 453 (218), 330 (939).
Single crystals of 2·1.25H2O were obtained by slow diffusion of 5 mL of isopropanol into a
solution of 100 mg 1·3H2O in 5 mL aqueous 0.1 M HCl. The resulting orange crystals were
filtered and air dried. Yield: 80 mg (74 %). S25°C(H2O) = 250 mg/mL.
Synthesis of fac-[RuCl3(HPTA)3]Cl2·HCl·2H2O (3·HCl·2H2O)
Single crystals of 3·HCl·2H2O were obtained by irradiation with visible light (150 W halogen
lamp) of a solution of 2·1.25H2O (30 mg, 0,039 mmol) in 1 mL of H2O at 80 ºC. The
crystals were filtered and air dried. Yield: 25 mg (78 %). S25°C(H2O) = 20 mg/mL.
Elemental analysis: Found C, 26.01; H, 5.49; N, 15.48 %; C18H44Cl6N9O2P3Ru
(825.30) calculated C, 26.20; H, 5.37; N, 15.27 %.
IR (KBr): max/cm-1 3392b (OH), 2981 and 2921m (CH), 2773-2535bm (NHPTA),
/cm-1 1625m (OH).
NMR data of aqueous solutions of fac-[RuCl3(HPTA)3]Cl2 (3):
Two sets of NMR signals are detected in aqueous solutions of 3: fac-
[RuCl2(OH2)(HPTA)3]2+ and fac-[RuCl3(HPTA)3]2+ are in equilibrium, however, only signals
of the latter are observed when 10 eq. of NaCl are dissolved into the solution.
NMR data of fac-[RuCl3(HPTA)3]2+
1H NMR (300 MHz, D2O):  (ppm) 4.27 (bs, NCH2Ptrans-Cl, 18 H), 4.75-4.85 (bm,
NCH2Ntrans-Cl, 18 H).

7
13C{1H} NMR (75.47 MHz, D2O):  (ppm) 51.71-52.07 (m, NCH2Ptrans-Cl), 70.67 (bs,
NCH2Ntrans-P and NCH2Ntrans-O).
31P{1H} NMR (121 MHz, D2O):  (ppm) -12.83 (s).
NMR data of fac-[RuCl2(OH2)(HPTA)3]3+.
1H NMR (300 MHz, D2O):  (ppm) 4.23 (bs, NCH2Ptrans-O, 6 H), 4.15-4.40 (bm,
NCH2Ptrans-Cl, 12 H), 4.78 (bs, NCH2Ntrans-O + NCH2Ntrans-Cl, 18 H).
1
13C{1H} NMR (75.47 MHz, D2O):  (ppm) 50.86 (t, JCP = 8.7 Hz, NCH2Ptrans-O),
51.71-52.07 (m, NCH2Ptrans-Cl), 70.67 (bs, NCH2NtransP and NCH2NtransO).
31P{1H} NMR (121 MHz, D2O):  (ppm) -4.57 (t, 2JPP = 36.5 Hz, Ptrans-O, 1 P), -12.49
(d, 2JPP = 36.5 Hz, Ptrans-Cl, 2 P).
UV-vis λmax(H2O)/nm (ε/dm3 mol-1 cm-1): 350 (1715), 312 (1061).
Synthesis of [{Ru(PTA)3}2(-Cl)3]Cl·9H2O (4·9H2O)
The complex can be prepared by various procedures:
a) PTA (105.9 mg, 0.674 mmol) was added to an aqueous solution (2 mL) of cis-
[RuCl2(DMSO)4] (109 mg, 0.225 mmol) and the solution was irradiated overnight with
visible light. The reaction mixture was evaporated to dryness and the residue dissolved in
acetone (2 mL). The product was precipitated by adding Et2O (10 mL). The solid was filtered
off and dried under vacuum. Yield: 160 mg (98 %).
b) A solution of mer-trans-[RuCl2(OH2)(PTA)3]·3H2O (1·3H2O) (100 mg, 0.140
mmol) in 5 mL of water was irradiated overnight with a 150 W halogen lamp. The initial
orange solution changed to pale yellow. The resulting solution was evaporated and the
obtained powder washed with EtOH (3x5 mL), Et2O (5x10 mL) and vacuum dried. Yield: 87
mg (86%).
c) A solution of mer-trans-[RuCl2(OH2)(PTA)3]·3H2O (1·3H2O) (100 mg, 0.140
mmol) in 10 mL of water was refluxed for 1 h. The resulting solution was treated as described
in b). Yield: 78 mg (77%).
Slow diffusion of 1:1 mixture of isopropanol/Et2O (5 mL) into 5 mL aqueous solution
of 200 mg of the product yielded light yellow single crystals of 4·9H2O. Crystals: 90 mg,
S25°C(H2O) = 179 mg/mL.
Elemental analysis: Found C, 30.06; H, 6.09; N, 16.96 %; C36H90N18O9Cl4P6Ru2
(1449.02) calculated C, 29.84; H, 6.26; N, 17.40 %.

8
UV-vis λmax(H2O)/nm (ε/dm3 mol-1 cm-1): 362 (3013).
1H NMR (300 MHz, D2O):  (ppm) 4.09 (bs, 36H, NCH2N), 4.48 (bs, 36H, NCH2P).
13C{1H} NMR (90 MHz, D2O):  (ppm) 51.50 (s, NCH2P), 70.68 (s, NCH2N).
31P{1H} NMR (121 MHz, D2O): -14.60 (s) ppm.
ESI-MS(+) (H2O): m/z
calculated: 1251.177.
=
1251.163 [Ru2Cl3(PTA)6]+ (C36H72N18P6Cl3Ru2),
Synthesis of trans-[RuCl2(PTA)4] (5) from cis-[RuCl2(DMSO)4] and PTA
A solution of cis-[RuCl2(DMSO)4] (100 mg, 0.21 mmol) in 5 mL of water was irradiated
overnight with visible light from a 150 W halogen lamp. The irradiation was stopped and
solid PTA (129.8 mg, 0.83 mmol) was added to the resulting deep green solution. After 1
night in the dark a yellow solution was obtained, which was evaporated to dryness and the
residue was dissolved in 2 mL of acetone. Addition of 10 mL of Et2O gave rise to a yellow
precipitate, which was filtered and dried in vacuum. The product was identified as trans-
[RuCl2(PTA)4]. Yield: 96 mg (58%).
Reaction of cis-cis-trans-[RuCl2(DMSO)2(PTA)2] with PTA under visible light
A deep yellow solution was obtained during an overnight irradiation with a 150 W halogen
lamp of a 2 mL aqueous solution of cis-cis-trans-[RuCl2(DMSO)2(PTA)2] (162 mg, 0.25
mmol) and PTA (37 mg, 0.24 mmol) at room temperature. The solvent was evaporated and
the residue was crystallized from acetone/Et2O. The obtained yellow powder was identified as
the dinuclear complex, 4·9H2O.
Reaction of cis-[RuCl2(DMSO)4] and trans-[RuCl2(PTA)4] under visible light
A solution of trans-[RuCl2(PTA)4] (10 mg, 0.01 mmol) and cis-[RuCl2(DMSO)4] (2.1 mg,
0.004 mmol) in 1 mL water was introduced into a 5 mm NMR tube. After irradiation with a
150 W halogen lamp overnight the starting compounds were fully transformed into 4. When
the same reaction was made without irradiation no significant transformation of the starting
complexes was observed.
Reaction of [(DMSO)3Ru(μ-Cl)3RuCl(DMSO)2] and PTA under visible light
A solution of [(DMSO)3Ru(μ-Cl)3RuCl(DMSO)2] (10 mg, 0.014 mmol) and PTA (12.8 mg,
0.081 mmol) in 1 mL of water was placed into a 5 mm NMR tube and was irradiated

9
overnight with a 150 W halogen lamp. The unique compound observed by NMR was 4. The
same reaction in the dark gave rise to a mixture of [RuCl2(DMSO)2(PTA)2] and trans-
[RuCl2(PTA)4].
Reaction of [{RuCl2(η6-p-cymene)}2] and PTA under visible light
A solution of [{RuCl2(η6-p-cymene)}2] (5 mg, 0.008 mmol) and PTA (7.7 mg, 0.049 mmol)
in 1 mL of water was placed into a 5 mm NMR tube and was irradiated with a 150 W halogen
lamp 6 hours. 31P NMR analysis of the resulting solution showed that 4 was the unique
compound formed.
Reaction of [{RuCl2(η6-p-cymene)}2] and trans-[RuCl2(PTA)4] under visible light
A solution of [{RuCl2(η6-p-cymene)}2] (2.4 mg, 0.004 mmol) and trans-[RuCl2(PTA)4]
(17.8 mg, 0.02 mmol) in 1 mL of water was placed into a 5 mm NMR tube and was irradiated
with a 150 W halogen lamp overnight. 31P NMR analysis of the resulting solution showed that
4 was the unique compound formed. With exclusion of light only the starting complexes were
recovered from the same reaction mixture.
Reactivity of 1 in D2O with NaCl
Compound 1·3H2O (15 mg, 0.021 mmol) was dissolved in 0.5 mL of D2O in a 5 mm NMR
31
tube and the P{1H} NMR spectrum was recorded. Increasing amounts of NaCl were added
31
into the solution and P{1H} NMR spectra were recorded in each step. Upon addition of 10
eq of NaCl signals of a new compound, mer-[RuCl3(PTA)3]- (20 %) were observed (a triplet
around -11.5 ppm and a doublet around -51.3 ppm). With increasing NaCl concentration, the
intensities of these resonances increased with a gradual shift of the signals. When the solution
arrived close to saturation in NaCl (ca. 80 eq) 31P{1H} NMR showed the following data (121
MHz, D2O):  (ppm) -10.60 (bt, Ptrans-Cl, 1 P), -51.30 (bd, Ptrans-P, 2 P). Nevertheless, signals
of mer-[RuCl2(H2O)(PTA)3]2- were still observed (34 %) and all efforts on isolation of mer-
[RuCl3(PTA)3]- in solid form failed.
Reaction of 1 in D2O with HBF4
Compound 1·3H2O (15 mg, 0.021 mmol) was dissolved in 0.5 mL of D2O in a 5 mm NMR
tube. After recording its 31P{1H} NMR spectrum 0.46 mmol HBF4 (60 L 48 w/w % solution)
31
was introduced into the solution. The P{1H} NMR spectrum observed after 15 min showed

10
the exclusive signals of mer-[RuCl2(OH2)(HPTA)3]3+: 31P{1H} NMR (121 MHz, D2O): 
(ppm) 2.31 ppm (t, 2JPP = 35.6 Hz, Ptrans-O, 1 P), -36.28 ppm (d, 2JPP = 35.6 Hz, Ptrans-P, 2 P).
Study of the behaviour of 1, 2 and 3 in D2O at 25 ºC, 50 ºC and 80 ºC in the dark
The complexes 1·3H2O (10 mg, 0.014 mmol), 2·1.25H2O (15 mg, 0.021 mmol) and
3·HCl·2H2O (15 mg, 0.018 mmol) were dissolved in 0.5 mL of D2O each in 5 mm NMR
31
tubes. The P{1H} NMR spectra obtained at 25 ºC, 50 ºC and 80 ºC showed that complex 1
was stable at 50 ºC for more than 30 min, however, under the same time at 80 ºC the dinuclear
complex 4 could also be detected in addition to 2. In contrast, complex 2 was stable even at
80
ºC.
The
proportion
between
mer-[RuCl3(HPTA)3]2+
and
mer-trans-
[RuCl2(OH2)(HPTA)3]3+ at room temperature (67/33) was reversibly modified at higher
temperatures in favour of mer-[RuCl3(HPTA)3]2+, being 73/27 at 80 ºC. At this temperature
no significant modifications in the solution composition were observed after 12 h in the dark,
however, under visible light irradiation the fac-isomer of complex 2 was slowly formed.
Complex 3 is stable at 80 ºC for 12 h both in the dark and under visible light. At this
temperature the only signal observed in the 31P{1H} NMR spectrum is the singlet
corresponding to fac-[RuCl3(HPTA)3]2+.
Catalysis
Aqueous-organic biphasic transfer hydrogenation of aldehydes
Appropriate amounts of the Ru(II)-PTA catalysts containing 0.0625 mmol Ru were placed
into a Schlenk tube which was carefully deoxygenated and filled with N2. The catalyst was
dissolved in 5 mL of 5 M aqueous Na-formate solution, followed by addition of 4.92 mmol
benzaldehyde or 3.96 mmol cinnamaldehyde in 5 mL chlorobenzene using hypodermic
syringes. The reaction mixture was rapidly stirred at T=80 °C for t=3 h using a magnetic
stirrer. After this time the mixture was cooled, the phases separated, a sample of the organic
phase was passed through a short MgSO4 plug and analyzed by gas chromatography.
In certain cases the catalyst was obtained by 20 min irradiation of a solution of trans-
[RuCl2(PTA)4].
Hydration of nitriles
0.05 mmol of the appropriate complex and the required amount of substrate and additives
were placed into a Schlenk tube equipped with a reflux condenser and a bubble counter and

11
were dissolved in 3 mL water. Under argon atmosphere 1 mmol benzonitrile was added and
the tube was immersed into an oil bath of 100-108 °C temperature and refluxed for 3 h.
Samples (50 μL) were taken from the hot reaction mixture at 1, 2 and 3 h, and were extracted
with 3×2 mL dichloromethane. Part of the combined organic phases was passed through a
short plug of anhydrous MgSO4 and analyzed by gas chromatography. The same yields were
obtained in reactions carried out under air. In certain cases the catalyst was obtained by 20
min irradiation of a solution of trans-[RuCl2(PTA)4].
X-ray structure determinations
Crystals of complexes 2·1.25H2O, 3·HCl·2H2O and 4·9H2O suitable for single crystal X-ray
structure determination were obtained from water. Crystal data and data collection details are
given in Table 1. Data collection for all compounds were performed on a Bruker APEX CCD
diffractometer (XDIFRACT service of the University of Almería) in the range 2.82  2 
50.12, 3.06  2  50.10 and 3.66  2  50.08 at 150 ºK, 100 ºK and 100 ºK, respectively.
No empirical absorption correction was applied for 4·9H2O but for 2·1.25H2O (Tmin=0.732;
Tmax=0.949) and for 3·HCl·2H2O (Tmin=0.863; Tmax=0.906). Structures were determined by
direct methods (SIR97[38] or SHELXS-XTL[39]) and refined by least-squares procedures on F2
(SHELX-XTL). A disordered PTA molecule was found in 2·1.25H2O and an average model
was refined. All non-hydrogen atoms were anisotropically refined. C-H and hydrogen atoms
for all the complexes were placed in riding positions and refined isotropically but NPTA-H in
2·1.25H2O were localized by synthesis difference and those for the PTA disordered molecule
were not included. The positions of the H atoms of the water molecules coordinated to Ru in
4·9H2O were determined and refined altogether isotropically but those for 2·1.25H2O could
not be localized although there was residual electron density close to the O atom. Final
geometrical calculations, the graphical manipulations and the analysis of H-bond network and
other crystallographic calculations were carried out with SHELXS-XTL package.

12
Table 1. Crystal data and structure refinement information for complexes 2·1.25H2O,
3·HCl·2H2O and 4·9H2O.
2·1.25H2O
C18H41.5Cl5N9O1.25P3Ru
775.33
Orthorhombic
Pccn
3·HCl·2H2O
C18H44Cl6N9O2P3Ru
825.30
Monoclinic
P21/n
4·9H2O
C36H90Cl4N18O9P6Ru2
1449.02
Empirical formula
Formula weight
Crystal system
Space group
Triclinic
a (Å)
b (Å)
c (Å)
 (°)
26.6270(10)
13.2360(10)
16.8950(10)
90
90
90
5954.4(6)
8
1.695
11.1793(8)
12.2900(9)
22.3109(16)
90
94.599(2)
90
3055.5(4)
4
1.794
13.0395(8)
15.4809(10)
17.2746(11)
113.0690(10)
111.6010(10)
93.1250(10)
2901.3(3)
2
 (°)
 (°)
V (Å3)
Z
Calculated density (g cm-3)
1.659
0.71073
0.71073
1.232
0.71073
0.935
 / (Å)
Absorption coefficient (mm-1) 1.169
F(000)
3048
1688
1500
Crystal size / mm
Index ranges
0.290x0.230x0.045
-31≤ h ≤30
-13≤ k ≤15
-20≤ l ≤16
31517
4402 (Rint = 0.0443)
5279/0/ 328
R1=0.0641, wR2=0.1675
R1=0.0757, wR2=0.1762
1.052
0.165x0.100x0.080
-13≤ h ≤13
-14≤ k ≤14
-13≤ l ≤26
16207
4769 (Rint = 0.0320)
5370/0/ 357
R1=0.0445, wR2=0.1136 R1=0.0332; wR2=0.0667
R1=0.0505, wR2=0.1176 R1=0.0442; wR2=0.0692
0.148x0.110x0.090
-15≤ h ≤14
-18≤ k ≤16
-10≤ l ≤20
16165
Reflections collected
Reflections unique
7918 (Rint = 0.0286)
10154/27/ 733
Data/restraints/parameters
Final R indices [I2(I)]a,b
R indices (all data)a,b
Goodness-of-fit on F2
1.050
1.292 and -1.431
1035048
0.918
1.436 and -0.919
1035049
Largest diff. peak; hole (e·Å-3) 2.049 and -1.264
CCDC
1035047

13
Results and Discussion
Synthesis and properties of mer-trans-[RuCl2(OH2)(PTA)3] (1)
Reaction of [RuCl2(PPh3)3] dissolved in CH2Cl2 with 3 equivalents of PTA in aqueous
solution immediately gave a dark yellow aqueous phase and a greenish organic phase. The
mixture was stirred overnight at room temperature in the dark during which the aqueous phase
turned orange yellow. (The organic phase remained pale green probably due to the presence
of unreacted [RuCl2(PPh)3] and/or its oxidized derivatives – this was not investigated in
detail.) According to 31P{1H} NMR spectra no free PTA was present in the aqueous phase
which contained [RuCl2(OH2)(PTA)3] (1) exclusively. A similar reaction but using of 4
equivalents of PTA over [RuCl2(PPh3)3] led to a colorless organic phase and a yellow aqueous
phase (Scheme 2); the latter contained trans-[RuCl2(PTA)4] (5) in almost quantitative
yield.[16]
Scheme 2. Formation of mer-trans-[RuCl2(OH2)(PTA)3] (1)
[RuCl2(OH2)(PTA)3] (1) mer-trans-[RuCl2(OH2)(PTA)3]·3H2O (1·3H2O) is the first isolated
compound containing a Ru(PTA)3-fragment. Its elemental analysis is in agreement with this
formulation. with that of a compound containing one ruthenium, three PTA, two Cl atoms and
four water molecules. The IR spectra shows a broad absorption band at 3434 cm-1 and a clear
sharp band at 1706 cm-1. These absorptions were also observed when the compound was dried
at 70 ºC for 1 day, that is in agreement with an OH2 coordinated to a Ru atom.[26] The 31P{1H}
NMR spectrum of 1 corresponds to an AM2 system constituted by a triplet at -5.08 ppm (1P)
and a doublet at -46.78 ppm (2P). This indicates two Ru-coordinated PTA ligands trans to
each other (-57.64 ppm for cis-[RuCl2(PTA)4] and -61.0 ppm for cis,mer-[RuCl2(DMSO-
S)(PTA)3] [27]).
In similar compounds with trans Cl-Ru-PTA structures the chemical shift of the PTA
trans to a Cl arises at -23.40 ppm.[40] The known complexes containing a water molecule
trans to a PTA, such as cis-[RuCl(OH2)(PTA)4]+ [26] and cis-[Ru(OH2)2(PTA)4]2+ [21] both
display the corresponding signal at -16.5 ppm that is far from the chemical shift observed for
complex 1 but far also from the normal chemical shift for Ru-coordinated PTA trans to a Cl.

14
In contrast, mer-[Ru(OH2)3(PTA)3]2+ gives rise to a triplet Ptrans-OH resonance at -7.4 ppm,
2
and a doublet Ptrans-PTA resonance at -48.3 ppm;[21] both are close to the relevant resonances of
1. Therefore the most plausible assignation for the triplet signal observed in the 31P{1H} NMR
of 1 is a PTA trans to a H2O coordinated to the metal by the O atom. The presence of a water
molecule coordinated to the ruthenium was confirmed by TG analysis that evidenced that in
the complexes there are three lattice water molecules that are eliminated below 90 ºC and one
1
above 100 ºC. The H NMR spectrum of 1 displays only broad signals in the region
determined for the P-CH2-N and N-CH2-N protons of known PTA-Ru complexes and its
31C{1H} NMR data are also in agreement with it’s the proposed composition of 1 (ref.[17-18]
;
and references therein).
Finally, reaction of 1 in water with NaCl showed that the compound was gradually
transformed into mer-[RuCl3(PTA)3]- upon the increase of salt concentration in the solution.
The transformation was not complete within the solubility range of NaCl and it was not
possible to isolate the new compound. Addition of H2O into the solution returned the starting
complex, which is in agreement with the initial substitution of the coordinated water molecule
by a Cl- what was reverted as the chloride concentration was reduced (Scheme 3).
Scheme 3. Reactivity of mer-trans-[RuCl2(OH2)(PTA)3] (1) in water towards PTA, HBF4 and
Cl- at room temperature
1 was reacted with PTA and its N-alkylated derivatives [(PTA-Bn)Cl, (PTA-Me)CF3SO3] in
the dark. Only the neutral PTA afforded a uniform product, the formation of which was
followed by UV-vis spectroscopy. Only one new species could be detected in the solution as
shown by a single isosbestic point at λ = 322 nm. Accordingly, the 31P{1H} NMR spectra
display a singlet signal (-49.7 ppm) which corresponds to trans-[RuCl2(PTA)4] (5).
Dissolution of 1 in aqueous HCl results in protonation of one nitrogen atom of each of
the coordinated PTA ligands yielding mer-[RuCl3(HPTA)3]Cl2 (2). Crystals of 2·1.25H2O
suitable for single crystal X-ray diffraction were obtained by diffusion of isopropanol to

15
solutions of 2 in aqueous HCl. In mer-[RuCl3(HPTA)3]2+, ruthenium is bonded to the P atoms
of three mono-protonated PTA and three Cl in the meridional position. The 31P{1H} NMR in
D2O showed two sets of signals with relative intensities of 1 to 2, a triplet at 2.31 ppm
connected to a doublet at -36.28 ppm and a triplet at -4.62 ppm coupled to a doublet at -40.62
ppm. Ratio of the intensity of these sets of signals varies with the temperature as well as with
the amount of added NaCl at room temperature (Scheme 4).
Ratio of mer-[RuCl3(HPTA)3]2+
mer-trans-[RuCl2(OH2)(HPTA)3]3+
66.6% : 33.3%
85% : 15%
75% : 25%
:
T(°C) nNaCl/nRu
25
25
80
0
40
0
Scheme 4. Aquation of mer-[RuCl3(HPTA)3]2+ as a function of temperature and chloride
concentration of aqueous solutions
This behaviour suggests that the main signals at -4.62 ppm and -40.62 ppm correspond
to mer-[RuCl3(HPTA)3]2+ and that the other set belongs to mer-trans-[RuCl2(OH2)(HPTA)3]3+
which is produced by the substitution of a Cl- in mer-[RuCl3(HPTA)3]2+ by a water molecule.
This assumption was confirmed when 1 was reacted with 5 equivalents of HBF4 in water: the
31P{1H} NMR spectrum of the solution showed only the doublet and triplet assigned to mer-
trans-[RuCl2(OH2)(HPTA)3]3+ (Scheme 3).
When a solution of 2 was irradiated by a 150 W halogen lamp[37] single crystals
suitable for X-ray determination separated in good yield (see Experimental). The crystal
structure showed that the new complex is the fac-isomer of 2 isolated as fac-
[RuCl3(HPTA)3]Cl2·HCl·2H2O (3·HCl·2H2O). Therefore the visible light promotes the
isomerization of 2 into 3 in water (Scheme 5).
31
At room temperature, the P{1H} NMR spectrum of irradiated solutions of 2 showed
a triplet at -4.57 ppm and a doublet at -12.49 ppm connected to each other and a singlet at -
31
12.83 ppm. At 80 ºC, P{1H} NMR of the solution showed only the singlet at -12.83 ppm

16
which is also the only signal observed when NaCl is added to the solution in high excess at
room temperature (Scheme 5).
2+
2+
Cl
Ru
Cl
Cl
Ru
P
h
436 nm
P
Cl
P
P
Cl
Cl
P
P
mer-[RuCl3(HPTA)3]2+
fac-[RuCl3(HPTA)3]2+
Cl-
Cl-
P = HPTA+
H2O
H2O
Cl
OH2
Ru
P
3+
3+
P
OH2
P
P
Cl
Cl
h
436 nm
Ru
P
P
Cl
fac-[RuCl2(OH2)(HPTA)3]3+
mer-[RuCl2(OH2)(HPTA)3]3+
Ratio of fac-[RuCl3(HPTA)3]2+ :
fac-[RuCl2(OH2)(HPTA)3]3+
25% : 75%
99% : 1%
99% : 1%
T(°C) nNaCl/nRu
25
25
80
0
40
0
Scheme 5. Synthesis of fac-[RuCl3(HPTA)3]2+ and its aquation in water as a function of
temperature and chloride concentration.
The expected signal for 3 should be a singlet as the ligands are in a fac-disposition. The
chemical shift of the singlet (-12.83 ppm) is very close to that of the doublet (-12.49 ppm)
therefore it is reasonable to assign both signals to a PTAtrans-Cl, while the observed triplet (-
4.57 ppm) at room temperature can be assigned to a PTAtrans-OH2. This is further supported by
the triplet’s chemical shift being close to the chemical shift of PTAtrans-OH in mer-
2
[Ru(OH2)3(PTA)3]2+ (-7.4 ppm) and in mer-trans-[RuCl2(OH2)(PTA)3] (-5.08 ppm).
Altogether these observations show that irradiated solutions of 1 in aqueous HCl contain both
fac-[RuCl3(HPTA)3]2+ and fac-[RuCl2(OH2)(HPTA)3]3+.
Synthesis and characterization of [{Ru(PTA)3}2(-Cl)3]Cl (4)

17
[RuCl2(PR3)3-4] complexes prepared in the reactions of RuCl3 and aliphatic tertiary
phosphines are known to rearrange spontaneously to [{Ru(PR3)3}2(-Cl)3]Cl in alcohols at
reflux temperature.[41] With P(Bu)3 and P(CH2OH3)3 the dimerization took place even at room
temperature, albeit in slow processes.[42-44]
We have found that [{Ru(PTA)3}2(μ-Cl)3]Cl (4) can be obtained in several ways from various
Ru(II)-complexes (Scheme 6).
mer-trans-[RuCl2(OH2)(PTA)3]
1
6
[{RuCl2( -p-cymene)}2]
[RuCl2(DMSO)2(PTA)2]
i)
h
v)
ii)
h
PTA
h
6 PTA
[{Ru(PTA)3}2(µ-Cl3)]Cl
iv)
iii)
h
4
6 PTA
h
3 PTA
[RuCl2(DMSO)4]
[Ru(DMSO)3(µ-Cl3)Ru(DMSO)2(H2O)]Cl
Scheme 6. Formation of [{Ru(PTA)3}2(-Cl)3]Cl (4) from various precursors at room
temperature in water under illumination
[{Ru(PTA)3}2(μ-Cl)3]Cl (4) can be prepared by heating of an aqueous solution of
[RuCl2(OH2)(PTA)3] (1) for 1 hour at elevated temperatures (80 °C) even with exclusion of
light. Replacement of the 31P NMR signals of the starting complex (1) by a new one at -14.60
ppm in D2O shows the formation of the dinuclear complex. 4 was also obtained in 1 hour
already at room temperature when an aqueous solution of 1 was irradiated with white light
(150 W halogen lamp) or light of >460 nm (process i) on Scheme 6). In comparison, without
irradiation, aqueous solutions of trans-[RuCl2(PTA)4] are stable in the dark even at 80 °C,
while under conditions of process i) trans-[RuCl2(PTA)4] isomerizes to the
thermodynamically stable cis-[RuCl2(PTA)4] and small amount of cis-[RuCl(OH2)(PTA)4]+
can be also detected[26,27]
.
Our earlier studies have shown that an aqueous solution of cis-cis-trans-
[RuCl2(DMSO)2(PTA)2] obtained in the reaction of cis-[RuCl2(DMSO)4] and 2 eq. PTA is
stable in the dark.[23] However, upon irradiation (process ii) on Scheme 6), the solution turns
31
green, and after addition of 1 eq PTA it becomes lemon yellow; the P{1H} NMR spectrum

18
of this yellow solution displays only the signal of 4 at -14.60 ppm. Formation of 4 is
facilitated if the solution of cis-cis-trans-[RuCl2(DMSO)2(PTA)2] contains PTA in the ratio
(PTA/Ru=3) right from the beginning of irradiation. Increasing the nPTA/nRu ratio from 3 to 4,
the exclusive product was trans-[RuCl2(PTA)4] in aqueous solution in dark at room
temperature, however, upon white light irradiation cis-[RuCl2(PTA)4] was detected as a main
product accompanied with [Ru(OH2)Cl(PTA)4]+.
Other water-soluble Ru(II)-complexes such as cis-[RuCl2(DMSO)4], [(DMSO)3Ru(μ-
Cl)3RuCl(DMSO)2] and [{RuCl2(η6-p-cymene)}2] were also found suitable for preparation of
4. In fact, reaction of cis-[RuCl2(DMSO)4] with 3 equivalents of PTA under visible light
irradiation (process iii), Scheme 6) provides the simplest synthesis of 4 with almost
quantitative isolated yield (98%; see Experimental).
Irradiation with visible light of a solution of the known dinuclear complex
[(DMSO)3Ru(μ-Cl)3RuCl(DMSO)2] in the presence of 6 equivalents of PTA (process iv),
Scheme 6) also led to the formation of 4. However, with the exclusion of light a 1:1 mixture
of cis-cis-trans-[RuCl2(DMSO)2(PTA)2] and trans-[RuCl2(PTA)4] (with a 1:2 ratio of
31P{1H} NMR integrals) was obtained in the same solution, and no traces of 4 were detected.
In comparison, although cis-cis-trans-[RuCl2(DMSO)2(PTA-Me)2](CF3SO3)2 is also light-
sensitive, irradiation by visible light in the presence of 1 eq (PTA-Me)(CF3SO3) yields the
known[40] trans-[RuCl2(OH2)(PTA-Me)3]3+ and cis-[RuCl2(PTA-Me)4]4+ complexes instead of
a dinuclear complex similar to 4. Irradiation of aqueous solutions of cis-cis-trans-
[RuCl2(DMSO)2(PTA-Bn)2]Cl2 did not yield uniform products independent of presence or
absence of 1 eq (PTA-Bn)Cl.
Finally, the dinuclear compound 4 was formed exclusively upon irradiation or
refluxing of an aqueous solution containing [{RuCl2(η6-p-cymene)}2] and 6 eq of PTA
(PTA/Ru=3) for one hour (the colour of the solution turned from orange to yellow with strong
smell of free p-cymene) (process v), Scheme 6).
In contrast to the analogous water-soluble dinuclear complex containing
P(CH2OH3)3,[44] compound 4 is air-stable both in solution and in solid form. Single crystals of
4·9H2O were obtained by layering 2-propanol on an aqueous solution of the complex.
Crystal structures of 2·1.25H2O, 3·HCl·2H2O and 4·9H2O

19
The crystal structures of the metal complex units of 2·1.25H2O, 3·HCl·2H2O and 4·9H2O
determined by X-ray crystallography are shown in Figures 1, 2 and 3. Crystal data are given
in Table 1 (selected bond lengths and angles in Suppl. Table S1).
Complex 2·1.25H2O is constituted by one cationic mer-[RuCl3(HPTA)3]2+, three chlorides
and 1.25 lattice water molecules while the asymmetric unit of 3·HCl·2H2O consists of a fac-
+
[RuCl3(HPTA)3]2+ cation, three chlorides, and a protonated water dimer, H5O2 . Single crystal
X-ray crystallography showed that the asymmetric unit of 4·9H2O consisted of one dinuclear
ion [{Ru(PTA)3}2(-Cl)3]+, one chloride and nine water molecules. In 2·1.25H2O,
3·HCl·2H2O and 4·9H2O, the C-N distances of the PTA ligands are found to be in the range
consistent with those for other PTA complexes. The angles between ligands for a distorted
octahedral geometry deviate significantly from 90º in 2 and 4: the smallest and largest angles
in 4 are Cl2-Ru1-Cl1 (79.63(3)°) and P2-Ru1-P1 (97.60(3)º), those in 2 are P1-Ru1-Cl2
(80.60(5)º) and P2-Ru1-P3 (97.31(7)º) while in 3 deviations from 90° are smaller: Cl1-Ru1-
Cl2 (85.66(4)º) and P3-Ru1-Cl1 (91.04(4)º).
Figure 1. Ball and stick model of the mer-[RuCl3(HPTA)3]2+ cation in 2·1.25H2O with partial
atom numbering scheme. Hydrogen atoms are omitted for clarity.
Figure 2. Ball and stick model of the fac-[RuCl3(HPTA)3]2+ cation in 3·HCl·2H2O with
partial atom numbering scheme. Hydrogen atoms are omitted for clarity.

20
Figure 3. Ball and stick model of the [{Ru(PTA)3}2(-Cl)3]+ cation in 4·9H2O with partial
atom numbering scheme. Hydrogen atoms are omitted for clarity.
The metal ion of 2·1.25H2O is in a distorted octahedral environment bound to three chlorides
in a mer-disposition, completing its coordination geometry by three protonated PTA ligands
bonded through the P atoms. The Ru-P bond distances in 2 are in the range from 2.2218(16) Å
(Ru-P2; Ptrans-Cl) to 2.3387(17) Å (Ru-P1; Ptrans-P) what is similar to those found in mer-
[RuCl3(PTA-Me)3]2+ (Ptrans-Cl = 2.220(2) Å - Ptrans-P = 2.333(2) Å),[40] showing a similar short
distance for the P atom trans to the Cl. Nevertheless the Ru-Ptrans-Cl distance is longer than
those for Ru-Ptrans-P and they are quite different from those found in cis-[RuCl2(PTA-Me)4]4+
(Ru-Ptrans-Cl = 2.267(2) Å, 2.277(2) Å; Ru-Ptrans-P = 2.375(2) Å, 2.377(2) Å).[40] The three Ru-
Cl bond lengths in 2 (Ru1-Cl3 = 2.4033(19) Å, Ru1-Cl1 = 2.4198(18) Å, Ru1-Cl2 =
2.5025(16) Å) are somewhat different but similar to those observed for mer-[RuCl3(PTA-
Me)3]2+ (Ru1-Cl1 = 2.427(1) Å, Ru1-Cl2 = 2.496(1) Å, Ru1-Cl3 = 2.422(1) Å);[40] however, a
new feature is that the Ru-Cl distance trans to P is the longest.
The ruthenium ion in 3·HCl·2H2O is in an octahedrally distorted coordination
geometry bonded to three PTA ligands by the P atom and to three Cl- in a fac disposition. The
distances Ru-P are more different among them than the Ru-Cl bond lengths and similar to
those found in 4 (Ru1-P1 = 2.2494(11) Å; Ru1-P2 = 2.2567(11) Å; Ru1-P3 = 2.2381(11) Å;
Ru1-Cl1 = 2.4803 (10) Å; Ru1-Cl2 = 2.4734(11) Å; Ru1-Cl3 = 2.4708(11) Å), what shows
the distortion of the molecule.
Finally, 4 is constituted by two {Ru(PTA)3} units linked through the metals by three
bridging Cl-; the metal atoms are coordinated in a distorted octahedral geometry to three PTA
ligands through the P atoms. According to the Cambridge Structural Database[45] to date only

21
8 chloro-bridged dinuclear Ru(II)-complexes with monodentate phosphine ligands (similar to
4) have been characterized in solid state [42-44,46-49]. The distances Ru-P and Ru-Cl for the two
metals are not the same (Ru1-P1 = 2.2672(9) Å; Ru1-P2 = 2.2601(10) Å; Ru1-P3 = 2.2518(9)
Å; Ru2-P4 = 2.2499(9) Å; Ru2-P5 = 2.2438(10) Å; Ru2-P6 = 2.2525(9) Å; Ru1-Cl1 =
2.5335(8) Å; Ru1-Cl2 = 2.4988(9) Å; Ru1-Cl3 = 2.4797(8) Å; Ru2-Cl1 = 2.4857(8) Å; Ru2-
Cl2 = 2.5202(9) Å; Ru2-Cl3 = 2.4949(9) Å) showing the distortion of the entire molecule. In
the similar dinuclear ruthenium complexes containing trimethylphosphine, [Ru{P(CH3)3}3(-
Cl)3Ru{P(CH3)3}3]X (X=BF4[46], Cl[47]) both ruthenium atoms have equal distances to
phosphines as well as to chlorides; in the chloride salt the average Ru-P distance (2.266(2) Å)
was found larger while the average Ru-Cl distance (2.503(6) Å) almost the same as in 4. It is
known that PTA and P(CH3)3 have similar cone angles and basicity, therefore the observed
differences in the complex structure must be due to the crystal packing of the complex. In
fact, complex 4 is involved in an extensive strong hydrogen-bond network made by
connections of N atoms of the PTA molecules with the neighbouring molecules (Suppl.
Figure S3).
Catalysis
Biphasic reduction of aldehydes catalyzed by water-soluble Ru(II)-PTA complexes
Since upon visible light irradiation trans-[RuCl2(PTA)4] can be isomerized to cis-
[RuCl2(PTA)4] and the latter can be aquated to cis-[RuCl(OH2)(PTA)4]+,[26,27] it is important
to learn how these photochemical processes may influence the reactions catalyzed by the
various Ru(II)-PTA complexes. With this aim in mind we have investigated the reduction of
cinnamaldehyde (Scheme 7) and benzaldehyde with aqueous Na-formate as hydrogen donor
and with trans-[RuCl2(PTA)4] as catalyst; the results are shown in Table 2. It was established,
that the results of the experiments performed in dark were not different from those obtained
under light: benzaldehyde was converted to benzyl alcohol in a fast reaction and
cinnamaldehyde was reduced to cinnamalcohol with 100% selectivity (Table 2, entry 1).

22
Scheme 7. Possible products of hydrogenation of cinnamaldehyde
Reduction of aldehydes was also performed in such a way that an intensively stirred solution
of trans-[RuCl2(PTA)4] was irradiated with visible light for 20 min and this solution was then
used as catalyst. At this point there was already no trans-[RuCl2(PTA)4] in the solution which,
however, contained cis-[RuCl2(PTA)4] and cis-[RuCl(OH2)(PTA)4]+ in a 1:3 ratio. This in situ
prepared mixture was somewhat less active in the reduction of benzaldehyde, leading to 85%
conversion instead of 92%. Furthermore, reduction of cinnamaldehyde was less selective than
with trans-[RuCl2(PTA)4], and small amounts of 3-phenyl-1-propanal and 3-phenyl-1-
propanol could also be detected (entry 3).
Table 2. Biphasic reduction of benzaldehyde and cinnamaldehyde catalyzed by various
Ru(II)-PTA complexes using aqueous HCOONa as hydrogen donor.*
Benz-
Cinnamaldehyde
aldehyde
Product yield
Catalyst
Benzyl
alcohol
(%)
Total
conversion
(%)
Cinnam-
alcohol
(%)
3-phenyl-
3-phenyl-
1-propanal 1-propanol
(%)
(%)
-
1
2
trans-[RuCl2(PTA)4] (5)
92
90
30
29
30
29
-
trans-[RuCl2(PTA)4] + 5 eq. NaCl
-
-
"in situ" cis-[RuCl(OH2)(PTA)4]
and cis-[RuCl2(PTA)4]**
3
4
85
91
32
29
24
28
5
3
cis-[RuCl2(PTA)4]
**
1
-
(as in entry 3 but + 5 eq. NaCl)
5
6
7
8
mer-trans-[RuCl2(OH2)(PTA)3] (1) 99
55
5
55
5
-
-
-
-
-
-
-
-
-
[{Ru (PTA)3}2(µ-Cl)3]Cl (4)
cis-[RuCl2(DMSO)4]
25
0
0
[(DMSO)3Ru(μ-Cl)3RuCl(DMSO)2]
0
0
-

23
cis-cis-trans-
-[RuCl2(DMSO)2(PTA)2]
9
71
68
78
34
16
28
36
7
2
2
11
3
cis-cis-trans-
-[RuCl2(DMSO)2(PTA-Bn)2]2+
10
11
33
40
cis-cis-trans-
2
-[RuCl2(DMSO)2(PTA-Me)2]2+
Conditions: 0.0625 mmol Ru, 4.92 mmol benzaldehyde or 3.96 mmol cinnamaldehyde, 5 mL 5 M aqueous
solution of HCOONa, 5 mL chlorobenzene, t = 3 h, T = 80 °C.
*Yields were determined by gas chromatography. ** This catalyst solution was obtained by 20 min irradiation of
a solution of trans-[RuCl2(PTA)4].
Addition of 5 eq NaCl reverted this change almost completely while it had negligible effect
on reactions catalyzed by trans-[RuCl2(PTA)4] with no irradiation (entries 4 and 2). Since
already such a small excess of chloride is sufficient to prevent aquation[26,27] this result shows
that cis- and trans-[RuCl2(PTA)4] have about the same catalytic activity in this hydrogen
transfer reaction. The highest activity in benzaldehyde reduction was shown by the
trisphosphine complex, mer-trans-[RuCl2(OH2)(PTA)3] (entry 5). The use of the dinuclear
[{Ru(PTA)3}2(-Cl)3]Cl (4) resulted in 25% benzaldehyde conversion (entry 6). This lower
efficiency relative to mer-trans-[RuCl2(OH2)(PTA)3] (1) may be the consequence of the
difficulty of splitting three chloride bridges in [{Ru(PTA)3}2(-Cl)3]+ in order to form
catalytically active mononuclear complexes.
The Ru(II)-complexes containing only chloride and DMSO ligands were completely
inactive in the biphasic hydrogen transfer from aqueous formate (entries 7,8). However, the
mixed [RuCl2(DMSO)2L2] complexes with L=phosphaurotropine ligands showed catalytic
activities in reduction of benzaldehyde with 68-78% conversions, and the activity increased in
the order L = (PTA-Bn)Cl < PTA < (PTA-Me)(CF3SO3).
Transfer hydrogenation of cinnamaldehyde was also investigated with the same Ru(II)
complexes. The highest catalytic activity was again shown by mer-[RuCl2(OH2)(PTA)3] (55%
conversion, almost double than that obtained with trans-[RuCl2(PTA)4]; 30%), with 100%
selectivity to cinnamalcohol. [{Ru(PTA)3}2(-Cl)3]Cl was a poor catalyst (5% conversion)
while [RuCl2(DMSO)4] and [(DMSO)3Ru(-Cl)3RuCl(DMSO)2] did not catalyze the reaction.
[RuCl2(DMSO)2L2] complexes showed medium catalytic activity, however, in their cases the
reduction was not selective, and the saturated aldehyde and saturated alcohol products were
also detected in the reaction mixtures.

24
From these measurements it can be concluded that visible light has only a small effect
on the transfer hydrogenation of aldehydes by [RuClx(OH2)y(PTA)6-x-y] catalysts (x= 1, y =1;
x = 2, y = 0; x=2, y=1) with 5 M aqueous sodium formate as hydrogen source. This
observation may be explained by the fast formation of the same catalytic species from the
various Ru(II)-catalyst precursors in the concentrated HCOONa solution (probably Ru(II)-
hydrido-complexes, such as those identified in ref.[44]). These results lend further credit to
earlier investigations on such catalytic systems with no exclusion of light and suggest that
photoactivity of [RuClx(OH2)y(PTA)6-x-y] catalysts is more likely to influence reactions (such
as hydrogenation or redox isomerization) in the absence of a high concentrations of other
coordinating ligands such as, for example, HCOO-.
Hydration of nitriles
For hydration of nitriles earlier we applied catalysts formed in situ from –among others– cis-
[RuCl2(DMSO)4] and PTA and (PTA-Bn)Cl.[29,30] The isolated complex, cis-cis-trans-
[RuCl2(DMSO)2(PTA)2] showed only modest catalytic activity, however, under the same
conditions an almost doubled activity was observed with cis-cis-trans-[RuCl2(DMSO)2(PTA-
Bn)2]Cl2.[29] Here we report on the catalytic activity of various new, isolated Ru(II)-
complexes with PTA-derived ligands as catalysts for benzonitrile hydration (Scheme 8) as
well as the effect of light on the catalytic reactions. The results are shown in Table 3 (entries
1-3 and 10 are included for comparison from ref.[29]).
Scheme 8. Hydration of benzonitrile to benzamide
It has been found that addition of 1 eq PTA to cis-cis-trans-[RuCl2(DMSO)2(PTA)2]
considerably increased the catalytic activity; however an even larger rate increase could be
achieved by addition of 1 eq (PTA-Bn)Cl (entries 1-3): with (PTA-Bn)Cl the reaction was
practically complete in 1 hour. Similar effects were observed with PTA and (PTA-Bn)Cl as
additives applying [{Ru(PTA)3}2(µ-Cl)3]Cl (entries 4-6), mer-trans-[RuCl2(OH2)(PTA)3]
(entries 7-9) and trans-[RuCl2(PTA)4] (entries 10,11). In all cases the use of (PTA-Bn)Cl led
to ≥90% conversions in 1 hour. We have already observed such a striking accelerating effect
of (PTA-Bn)Cl (relative to PTA) with catalysts prepared in situ from cis-[RuCl2(DMSO)4]

25
and 3 eq (PTA-Bn)Cl.[29] Similarly, an earlier study on catalysis of nitrile hydration with half-
sandwich [RuCl2(arene)L] (L=PTA, (PTA-Bn)Cl, mtppms-Na) complexes also showed the
PTA-Bn containing catalysts the most effective (independent of the arene ligand).[50] The
catalysts shown in Table 3 entries 3, 6, 9, 11 are approximately twice as active {TOFs = 18-
20 h-1; TOF=turnover frequency=(mol product)×(mol catalyst)-1×h-1} in hydration of
benzonitrile than the best half-sandwich complex, [RuCl2{η6-C6(CH3)6)}(PTA-Bn)]Cl
(TOF=10 h-1) with no microwave irradiation.[51] At present we do not know the origin of this
phenomenon. In the cases of entries 3, 6 and 9 one may assume the fast coordination of a
(PTA-Bn)+ ligand to Ru(II), however, this is less likely with the six-coordinated and stable
trans-[RuCl2(PTA)4] (entries 10-11). In separate experiments it was established, that (PTA-
Bn)Cl itself had no catalytic effect in hydration of benzonitrile. Furthermore, under the same
conditions but in the absence of benzonitrile, attempted reactions of (PTA-Bn)Cl with cis-cis-
trans-[RuCl2(DMSO)2(PTA)2], mer-trans-[RuCl2(OH2)(PTA)3] (1), [{Ru(PTA)3}2(µ-Cl)3]Cl
31
1
(4) or trans-[RuCl2(PTA)4] (5) did not yield uniform products as shown by P and H NMR
measurements. Our efforts to obtain a well defined Ru(II)-PTA-Bn complex from hydrated
RuCl3 and PTA-Bn in refluxing ethanol (analogous to the synthesis of trans-[RuCl2(PTA)4])
also failed. Clearly, further experiments are needed to clarify the mechanism of this catalytic
nitrile hydration and that of the “magic” effect of (PTA-Bn)Cl.
Table 3. Hydration of benzonitrile catalyzed by various Ru(II)-PTA/PTA-Bn complexes*
Conversion (%)
[Ru]:[PTA]:
[(PTA-Bn)Cl]
1:2:0
Catalyst
1 h
0
2 h
13
56
99
84
98
99
65
91
94
65
95
13
3 h
46
79
99
89
99
99
80
99
98
80
98
29
1
2
3
4
5
6
7
8
9
cis-cis-trans-[RuCl2(DMSO)2(PTA)2]
cis-cis-trans-[RuCl2(DMSO)2(PTA)2] + PTA
1:3:0
2
cis-cis-trans-[RuCl2(DMSO)2(PTA)2] + (PTA-Bn)Cl 1:2:1
99
81
96
99
29
43
90
32
94
3
[{Ru(PTA)3}2(µ-Cl)3]Cl
1:3:0
1:4:0
1:3:1
1:3:0
1:4:0
1:3:1
1:4:0
1:4:1
[{Ru(PTA)3}2(µ-Cl)3]Cl + 2 PTA
[{Ru (PTA)3}2(µ-Cl)3]Cl + 2 (PTA-Bn)Cl
mer-trans-[RuCl2(OH2)(PTA)3]
mer-trans-[RuCl2(OH2)(PTA)3] + PTA
mer-trans-[RuCl2(OH2)(PTA)3] + (PTA-Bn)Cl
10 trans-[RuCl2(PTA)4]
11 trans-[RuCl2(PTA)4] + [(PTA-Bn)Cl]
12 „in
and cis-[RuCl2(PTA)4]**
situ”cis-[RuCl(OH2)(PTA)4] 1:4:0
13 cis-[RuCl2(PTA)4]** (+5 eq. NaCl)
14 trans-[RuCl2(PTA)4] (+5 eq. NaCl)
1:4:0
1:4:0
14
17
47
41
64
87
Reaction conditions: 0.05 mmol (5 mol%) Ru, 1 mmol benzonitrile, 3 mL H2O, reflux. *Conversions were
determined by gas chromatography. **This catalyst solution was obtained by 20 min irradiation of a solution of
trans-[RuCl2(PTA)4].

26
The catalytic activity of [{Ru(PTA)3}2(µ-Cl)3]Cl formed by refluxing or irradiation of
aqueous solutions of mer-trans-[RuCl2(OH2)(PTA)3] is higher than that of the mother
mononuclear complex (entries 4 vs 7). The same conclusion can be drawn by comparison of
catalytic activities of 4 and cis-cis-trans-[RuCl2(DMSO)2(PTA)2] (or cis-cis-trans-
[RuCl2(DMSO)2(PTA)2] + 1 eq PTA; entries 4 vs 1, 2).
We have no evidence on the nuclearity of the active catalytic species in case of
[{Ru(PTA)3}2(µ-Cl)3]Cl as a precatalyst. However, it is striking that in the presence of 2 eq
PTA the activity of the formed catalytic species was much higher than that of trans-
[RuCl2(PTA)4] despite the same overall Ru:Cl:PTA=1:2:4 molar ratio (entries 5 vs. 10),
which may refer to the presence of an active dinuclear catalyst.
As discussed above, 20 min irradiation by visible light of aqueous solutions of trans-
[RuCl2(PTA)4] yields a solution containing cis-[RuCl2(PTA)4] and cis-[RuCl(OH2)(PTA)4]+
in 1:3 ratio. This solution showed a catalytic activity much inferior to that of trans-
[RuCl2(PTA)4] (entries 12 vs 10). NaCl (5 eq) disfavoured the photochemically assisted
formation of cis-[RuCl(OH2)(PTA)4]+, however, the catalytic activity of cis-[RuCl2(PTA)4]
still did not reach that of trans-[RuCl2(PTA)4] also measured in the presence of 5 eq NaCl
(entries 13 vs 14). In general, NaCl seems to decrease the rate of benzonitrile hydration
catalyzed by Ru(II)-PTA phosphaurotropine species (entries 13, 14 vs 10).
Conclusion
The results described above show that in many cases the composition of in situ prepared
Ru(II)-PTA phosphaurotropine catalysts may critically depend on the ligand/metal
concentration ratio and the variety of the obtained complexes is further increased by thermal
and photochemical effects. For example, in solutions of [PTA]:[RuCl2(DMSO)4]=4
concentration ratio, upon irradiation with visible light, in addition to trans- and cis-
[RuCl2(PTA)4], cis-[RuCl(OH2)(PTA)4]+ is also formed via photoaquation. When the
[PTA]:[RuCl2(DMSO)4] ratio is lowered to 3, visible light irradiation leads to formation of
the dinuclear [{Ru(PTA)3}2(μ-Cl)3]Cl. The same dinuclear complex is obtained in aqueous
solution from the thermal or photochemical reaction of mer-trans-[RuCl2(OH2)(PTA)3].
However, in aqueous hydrochloric acid solutions mer-trans-[RuCl2(OH2)(PTA)3] yields mer-
[RuCl3(HPTA)3]2+ which is also photoactive and gives fac-[RuCl3(HPTA)3]2+ upon
irradiation.

27
In the catalytic transfer hydrogenation of aldehydes in concentrated aqueous solutions
of Na-formate only slight photochemical effects were observed. Conversely, the
photochemical activity of the Ru(II)-PTA-type complexes had more pronounced influence on
the catalytic hydration of benzonitrile. From the results it can be concluded that irradiation
with visible light of trans-[RuCl2(PTA)4] leads to formation of catalytically less active Ru(II)-
complexes for hydration of nitriles. One must also consider the photochemical or thermal
formation of [{Ru(PTA)3}2(µ-Cl)3]Cl which can take place from various Ru(II)-PTA
derivatives and which itself is active in benzonitrile hydration especially in the presence of N-
benzylated PTA.
Acknowledgement
Thanks are given to the European Commission FEDER program for co-financing the projects
CTQ2015-67384-R (MINECO) and P09-FQM-5402 (Junta de Andalucía). Thanks are also
given to Junta de Andalucía PAI-research group FQM-317 and COST Action CM1302 (WG1,
WG2). The research was supported by the EU and co-financed by the European Social Fund
under the project ENVIKUT (
). A.U. acknowledges
funding from the Hungarian Academy of Sciences of a postdoctoral fellowship. Thanks are
also due to the Hungarian Research, Development and Innovation Office - Hungarian
Research Fund (NKFIH-OTKA K101372) for financial support.
Contributions of Dr. Attila Kiss-Szikszai (elemental analyses) and of Dr. Lajos Nagy (ESI-
MS measurements) both of the University of Debrecen are thankfully acknowledged.
Appendix A. Supplementary data
CCDC 1035047, 1035048, 1035049 contain the supplementary crystallographic data for
compounds 2, 3 and 4 respectively. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif .
Supplementary data associated with this article can be found, in the online version, at
http://dx.doi.org/10.1016/xxxxxxxxxxxxxxxxx

28
6. References
[1] F. Joó, Aqueous Organometallic Catalysis, Kluwer Academic Publishers, Dordrecht,
2001.
[2] B. Cornils, W.A. Herrmann, Aqueous-Phase Organometallic Catalysis, 2nd ed., Wiley-
VCH, Weinheim, 2004.
[3] D.J. Adams, P.J. Dyson, S.J. Tavener, Chemistry in Alternative Reaction Media, Wiley,
Chichester, 2004.
[4] C.J. Li, Chem. Rev. 105 (2005) 3095-3166.
[5] P.H. Dixneuf, V. Cadierno, Metal-Catalyzed Reactions in Water, Wiley-VCH, Weinheim,
Germany, 2013.
[6] K.H. Shaughnessy, Chem. Rev. 109 (2009) 643-710.
[7] C.E. Czégéni, G. Papp, Á. Kathó, F. Joó, J. Mol. Catal. A: Chemical 340 (2011) 1-8.
[8] H.D. Velázquez, F. Verpoort, Chem. Soc. Rev. 41 (2012) 7032-7060.
[9] L.A. Schaper, S. J. Hock, W.A. Herrmann, F.E. Kühn, Angew. Chem. Int. Ed. 52 (2013)
270-289.
[10] E. Levin, E. Ivry, C.E. Diesendruck, N. G. Lemcoff, Chem. Rev. 115 (2015) 4607-4692.
[11] K. Voronova, L. Homolya, A. Udvardy, A.C. Bényei, F. Joó, ChemSusChem 7 (2014)
2230-2239.
[12] S. Ahrland, J. Chatt, N.R. Davies, A. A. Williams, J. Chem. Soc. (1958) 276-288.
[13] F. Joó, J. Kovács, Á. Kathó, A.C. Bényei, T. Decuir, D.J. Darensbourg, Inorg. Synth. 32
(1998) 1-8.
[14] E.G. Kuntz, CHEMTECH 17 (1987) 570-575.
[15] W.A. Herrmann, C.W. Kohlpaintner, Inorg. Synth. 32 (1998) 8-25.

29
[16] D.J. Darensbourg, F. Joó, M. Kannisto, A. Katho, J.H. Reibenspies, Organometallics 11
(1992) 1990-1993.
[17] A.D. Phillips, L. Gonsalvi, A. Romerosa, F. Vizza, M. Peruzzini, Coord. Chem. Rev. 248
(2004) 955-993.
[18] J. Bravo, S. Bolaño, L. Gonsalvi, M. Peruzzini, Coord. Chem. Rev. 254 (2010) 555-607.
[19] B. Cornils, W.A. Herrmann, I.T. Horváth, W. Leitner, S. Mecking, H. Olivier-
Bourbigou, D. Vogt, Multiphase Homogeneous Catalysis, Wiley-VCH, Weinheim, 2005.
[20] F. Joó, Á. Kathó, in Handbook of Homogeneous Hydrogenation (Ed.: J. G. de Vries, C.
J. Elsevier), Wiley-VCH, Weinheim, (2007) 1327-1360.
[21] J. Kovács, F. Joó, A.C. Bényei, G. Laurenczy, Dalton Trans. (2004) 2336-2340.
[22] D.J. Darensbourg, F. Joó, M. Kannisto, Á. Kathó, J.H. Reibenspies, D.J. Daigle, Inorg.
Chem. 33 (1994) 200-208.
[23] A. Udvardy, A.C. Bényei, Á. Kathó J. Organomet. Chem. 717 (2012) 116-122.
[24] M. Brindell, G. Stochel, V. Bertolasi, R. Boaretto, S. Sostero, Eur. J. Inorg. Chem.
(2007) 2353-2359.
[25] A.C. Bényei, S. Lehel, F. Joó, J. Mol. Catal. A: Chemical 116 (1997) 349-354.
[26] R. Girotti, A. Romerosa, S. Mañas, M. Serrano-Ruiz, R.N. Perutz, Inorg. Chem. 48
(2009) 3692-3698.
[27] F. Battistin, G. Balducci, E. Iengo, N. Demitri, E. Alessio, Eur. J. Inorg. Chem. (2016)
2850-2860
[28] W.-C. Lee, B. J. Frost, Green Chem. 14 (2012) 62-66.
[29] E. Bolyog-Nagy, A. Udvardy, F. Joó, Á. Kathó, Tetrahedron Lett. 55 (2014) 3615-3617.
[30] A. K. Misra, É. Bokor, S. Kun, E. Bolyog-Nagy, Á. Kathó, F. Joó, L. Somsák,
Tetrahedron Lett. 56 (2014) 5995-5998.
[31] D.J. Daigle, Inorg. Synth. 32 (1998) 40-41.