Base-free transfer hydrogenation of ketones using arene ruthenium(II) complexes

Article abstract of DOI:10.1021/om9001268

Ruthenium derivatives of the type [RuCl(arene)(NN)][BPh₄] (arene = benzene, p-cymene) have been synthesized with NN = 2- hydroxyphenylbis(pyrazol-l-yl)methane (bpzmArOH) or 2-hydroxyphenylbis(3,5- dimethylpyrazol-l-yl)methane (bpz*mArOH). In th

Full text of DOI:10.1021/om9001268

3822 Organometallics 2009, 28, 3822–3833
DOI: 10.1021/om9001268
Base-Free Transfer Hydrogenation of Ketones Using Arene
Ruthenium(II) Complexes
ꢀ
M. Carmen Carrion, Francisco Sepulveda, Felix A. Jalon, and Blanca R. Manzano*
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
Departamento de Quꢀımica Inorganica, Organica y Bioquꢀımica, IRICA, Universidad de Castilla-La Mancha,
Facultad de Quꢀımicas, Avenida C. J. Cela, 10, 13071 Ciudad Real, Spain
Ana M. Rodrꢀıguez
ꢀ
ꢀ
Departamento de Quꢀımica Inorganica, Organica y Bioquꢀımica, IRICA, Universidad de Castilla-La Mancha,
ꢀ
Escuela Tecnica Superior de Ingenieros Industriales, Avenida C. J. Cela, 3, 13071-Ciudad Real, Spain
Received February 17, 2009
Ruthenium derivatives of the type [RuCl(arene)(NN)][BPh4] (arene = benzene, p-cymene) have
been synthesized with NN = 2-hydroxyphenylbis(pyrazol-1-yl)methane (bpzmArOH) or 2-hydro-
xyphenylbis(3,5-dimethylpyrazol-1-yl)methane (bpz*mArOH). In the p-cymene derivative contain-
ing bpzmArOH, activation of the hydroxy group is observed and a scorpionate complex is obtained
with the ligand behaving in a tridentate manner, [Ru(bpzmArO-κ3-N,N,O)(p-cymene)][BPh4].
The structure of this derivative and that of [RuCl(p-cymene)(bpz*mArOH)][BPh4] were determined
by X-ray diffraction. The new derivatives, along with other compounds previously described by us
that contain similar ligands, were tested in the transfer hydrogenation of benzophenone and other
carbonyl compounds under base-free conditions without any other additive to promote the reaction.
The precursors were active in this process, and the p-cymene derivatives exhibited a higher activity
than the benzene complexes. For the p-cymene complexes the activity was even higher than that
found in the presence of base (KOH). The effect of the substituents on the methylene carbon was also
studied. Mechanistic and kinetic studies were carried out, and hydride species were observed after a
pretreatment of the precatalyst in 2-propanol. Taking into account all of the results, a proposal for
the mechanism taking place during the hydrogenation of carbonyl groups under base-free conditions
has been made.
Introduction
that generate active species for the homogeneous and
asymmetric hydrogenation of ketones1-4 and imines4,5 with
hydrogen gas in the presence of a base. 2-Propanol has also
been used as a conventional hydrogen source, and this has
favorable properties: it is stable, easy to handle, nontoxic,
and inexpensive and has a moderate boiling point (82 °C).2,6
Numerous different Ru-arene systems bearing PP, NN, NO,
and NPN ligands have been described for this process.7
Catalytic hydrogenation of ketones is a fundamental and
key process for the production of a wide range of alcohols
including chiral compounds, which are valuable final pro-
ducts and precursors for the pharmaceutical, agrochemical,
flavor, fragrance, materials, and fine chemical industries.1,2
The catalytic hydrogenation processes developed by
Noyori and co-workers are very attractive, since the pre-
catalysts consist of well-defined air-stable [RuCl2(PR3)2-
(diamine)] or [RuCl2(diphosphine)(diamine)] complexes
(4) (a) Abdur-Rashid, K.; Lough, A. J.; Morris, R. H. Organometal-
lics 2000, 19, 2655. (b) Abdur-Rashid, K.; Lough, A. J.; Morris, R. H.
Organometallics 2001, 20, 1047. (c) Leyssens, T.; Peeters, D.; Harvey,
J. N. Organometallics 2008, 27, 1514. (d) Doherty, S.; Knight, J. G.; Bell,
A. L.; Harrington, R. W.; Clegg, W. Organometallics 2007, 26, 2465.
*Corresponding author. E-mail: blanca.manzano@uclm.es.
(1) (a) Ohkuma, T.; Ooka, H.; Ikariya, T.; Noyori, R. J. Am. Chem.
Soc. 1995, 117, 10417. (b) Doucet, H.; Ohkuma, T.; Murata, K.;
Yokozawa, T.; Kozawa, M.; Katayama, E.; England, A. F.; Ikariya,
T.; Noyori, R. Angew. Chem., Int. Ed. 1998, 37, 1703. (c) Mikami, K.;
Korenaga, T.; Terada, M.; Ohkuma, T.; Pham, T.; Noyori, R. Angew.
Chem., Int. Ed. 1999, 38, 495.
(e) Clapham, S. E.; Hadzovic, A.; Morris, R. H. Coord. Chem. Rev. 2004,
248, 2201. (f) Samec, J. S. M.; Backvall, J.-E.; Andersson, P. G.; Brandt,
P. Chem. Soc. Rev. 2006, 35, 237.
::
(5) (a) Cobley, C. J.; Henschke, J. P. Adv. Synth. Catal. 2003, 345,
195. (b) Sahoo, S.; Kumar, P.; Lefebvre, F.; Halligudi, S. B. J. Mol.
Catal. A: Chem. 2007, 203, 102.
(2) (a) Wu, X.; Xiao, J. Chem. Commun. 2007, 2449. (b) Blaser, H.-U.;
Malan, C.; Pugin, B.; Spindler, F.; Steiner, H.; Studer, M. Adv. Synth.
Catal. 2003, 345, 103. (c) Noyori, R.; Ohkuma, T. Angew. Chem., Int. Ed.
Engl. 2001, 40, 40.
(3) (a) Ohkuma, T.; Ooka, H.; Yamakawa, M.; Ikariya, T.; Noyori,
R. J. Org. Chem. 1996, 61, 4872. (b) Ohkuma, T.; Koizumi, M.; Doucet,
H.; Pham, T.; Kozawa, M.; Murata, K.; Katayama, E.; Yokozawa, T.;
Ikariya, T.; Noyori, R. J. Am. Chem. Soc. 1998, 120, 13529. (c) Ohkuma,
T.; Doucet, H.; Pham, T.; Mikami, K.; Korenaga, T.; Terada, M.;
Noyori, R. J. Am. Chem. Soc. 1998, 120, 1086.
(6) Rautenstrauch, V.; Hoang-Cong, X.; Churlaud, R.; Abdur-Rashid,
K.; Morris, R. H. Chem.Eur. J. 2003, 9, 4954.
(7) (a) Ghebreyessus, K. Y.; Nelson, J. H. J. Organomet. Chem. 2003,
669, 48. (b) Rath, R. K.; Nethaji, M.; Chakravarty, A. R. Polyhedron
2001, 20, 2735. (c) Braunstein, P.; Fryzuk, M. D.; Naud, F.; Rettig, S. J.
J. Chem. Soc., Dalton Trans. 1999, 589. (d) Braunstein, P.; Naud, F.;
Rettig, S. J. New J. Chem. 2001, 25, 32. (e) Ogo, S.; Abura, T.; Watanabe,
Y. Organometallics 2002, 21, 2964. (f ) Reetz, M. T.; Li, X. J. Am. Chem.
Soc. 2006, 128, 1044. (g) Ikariya, T.; Blacker, A. J. Acc. Chem. Res. 2007,
40, 1300.
r
pubs.acs.org/Organometallics
Published on Web 05/28/2009
2009 American Chemical Society

Article
Organometallics, Vol. 28, No. 13, 2009 3823
In this context, an important development was Noyori’s
discovery of the use of “Ru(II)(arene)” precursors with chiral
amino alcohols or diamines8 in processes that lead to
excellent activities and enantioselectivities. As a result of
the participation of the ligand in the catalytic reaction,
Noyori proposed the term bifunctional metal-ligand cata-
lysis.9 Other authors have also developed systems based on
Ru-arene fragments with amino alcohol,10 aminoamide,11
and aminocarboxylate12 ligands. Besides the metal-ligand
bifunctional mechanism, other mechanisms involving mono-
hydride or dihydride species have also been proposed.4 It is
worth noting that Baratta et al.13,14 recently designed a new
type of Ru(II) nonarene catalyst containing the ligand
2-(aminomethyl)pyridine (or its derivatives), and these were
very active in the transfer hydrogenation of acetophenone
using NaOH as a cocatalyst. As in the approaches described
by Noyori, in Baratta’s systems the existence of a cis-RuH/
-NH2 motif that assists the hydrogen transfer process plays a
fundamental role in the high activity.14a The main difference
between the Baratta and Noyori systems is that in the former
the alcohol and the ketone coordinate to the metallic center
during the catalytic process in an inner-sphere mechanism,
whereas in Noyori’s system an outer-sphere mechanism has
been established.
been proposed.16 However, the presence of a base is not
desirable from an industrial point of view due to corrosion,
and, more importantly, it can adversely affect stereoselectiv-
ity and cannot be used for base-sensitive ketones17 or
aldehydes, a fact that is very important in the pharmaceutical
industry. Recent studies have focused on the transfer hydro-
genation of ketones in 2-propanol in the absence of base.17
However, in these cases the precursors that led to success are
usually metal-hydride complexes or an additive is present to
generate the hydride derivative. These hydride complexes are
usually much more difficult to handle than the correspond-
ing halides due to their greater sensitivity to air and higher
reactivity. A hydroxy-osmium derivative obtained by treat-
ment of the chloride precursor with NaOH has been found to
show good activity in the transfer hydrogenation of alde-
hydes without the addition of base during the catalytic cycle,
although the activity toward ketones was quite low.18 In
addition, Peris reported an iridium system that is active in the
transfer hydrogenation of ketones, aldehydes, and imines at
room temperature under base-free conditions. The catalyst,
[Ir(NHC)(Cp*)Cl2], is activated in this case by the addition
of AgTfO as a chloride abstractor.19
As far as hydrogenation with H2 is concerned, Noyori8b,8e,20
and Ohkuma21 recently reported excellent results in the
asymmetric hydrogenation of ketones without the parti-
cipation of a base.
The presence of a strong base such as KOH, NaOH, or
KOtBu is usually necessary for the reduction to occur.2,4,6,15
In general, it is proposed that the base is needed to generate a
hydride, which is the catalytically active species, although
cases where the base participates in the catalytic cycle have
We have previously reported the synthesis of chloro-arene
Ru(II) complexes with bis(pyrazol-1-yl)methane ligands
containing different phenyl groups on the central carbon.
The ruthenium derivatives were used as precursors for the
hydrogenation of benzophenone in the presence of base and
using 2-propanol as the hydrogen source.22 It was observed
that there was an interconversion of the two possible isomers
of the arene complexes (see below for the two isomers), a
process that could take place through rupture of the Ru-
N (pyrazolyl) bond. Besides, we have reported that [RuCl2(p-
cymene)(dpim)] [dpim = 2-(diphenylphosphino)-1-methyli-
midazole] reinforces the nucleophilic character of MeOH
by interchange of one of the chlorides with the alcohol
and deprotonation induced by a decoordinated methylimi-
dazole moiety of the phosphane.23 Taking these results into
account, we decided to explore the possibility that our bis
(pyrazol-1-yl)methane derivatives could activate 2-propanol
and possibly generate hydride species that would be active in
the base-free transfer hydrogenation process. Thus, the
partial decoordination of the NN ligand will give rise to
the formation of unsaturated species that could allow the
coordination of the alcohol. This decoordinated pyrazolyl
ring could behave as an internal base and activate the
(8) (a) Ohkuma, T.; Ooka, H.; Hashiguchi, S.; Ikariya, T.; Noyori, R.
J. Am. Chem. Soc. 1995, 117, 7562. (b) Fujii, A.; Hashiguchi, S.;
Uematsu, N.; Ikariya, T.; Noyori, R. J. Am. Chem. Soc. 1996, 118,
2521. (c) Noyori, R.; Hashiguchi, S. Acc. Chem. Res. 1997, 30, 97. (d)
Haack, K.-J.; Hashiguchi, S.; Fujii, A.; Ikariya, T.; Noyori, R. Angew.
Chem., Int. Ed. Engl. 1997, 36, 285. (e) Ohkuma, T.; Utsumi, N.;
Tsutsumi, K.; Murata, K.; Sandoval, C.; Noyori, R. J. Am. Chem.
Soc. 2006, 128, 8724. (f) Sandoval, Ch. A.; Ohkuma, T.; Utsumi, N.;
Tsutsumi, K.; Murata, K.; Noyori, R. Chem. Asian J. 2006, 1-2, 102.
(9) Ikariya, T.; Murata, K.; Noyori, R. Org. Biomol. Chem. 2006, 4,
393.
(10) (a) Nordin, S. J. M.; Roth, P.; Tarnai, T.; Alonso, D. A.; Brandt,
P.; Andersson, P. G. Chem.Eur. J. 2001, 7, 1431. (b) Faller, J. W.;
ꢀ
Lavoie, A. R. Organometallics 2002, 21, 3493. (c) Pasto, M.; Riera, A.;
Pericas, M. A. Eur. J. Org. Chem. 2002, 2337. (d) Everaere, K.;
ꢁ
Mortreux, A.; Carpentier, J. F. Adv. Synth. Catal. 2003, 345, 67.
(e) Petra, D. G. I.; Kamer, P. C. J.; Van Leeuwen, P. W. N. M.; Goubitz,
K.; Van Loon, A. M.; De Vries, J. G.; Schoemaker, H. E. Eur. J. Inorg.
Chem. 1999, 2335. (f ) Brandt, P.; Roth, P.; Andersson, P. G. J. Org.
Chem. 2004, 69, 4885.
(11) Pelagatti, P.; Carcelli, M.; Calbiani, F.; Cassi, C.; Elviri, L.;
Pelizzi, C.; Rizzotti, U.; Rogolino, D. Organometallics 2005, 24, 5836.
(12) (a) Carmona, D.; Lamata, M. P.; Oro, L. A. Eur. J. Inorg. Chem.
2002, 2239. (b) Carmona, D.; Lamata, M. P.; Viguri, F.; Dobrinovich, I.;
Lahoz, F. J.; Oro, L. A. Adv. Synth. Catal. 2002, 344, 499. (c) Bogevig,
A.; Pastor, I. M.; Adolfsson, H. Chem.Eur. J. 2004, 10, 294.
(13) (a) Baratta, W.; Schutz, J.; Herdtweck, E.; Herrmann, W. A.;
Rigo, P. J. Organomet. Chem. 2005, 690, 5570. (b) Baratta, W.;
Herdtweck, E.; Siega, K.; Toniutti, M.; Rigo, P. Organometallics 2005,
24, 1660. (c) Baratta, W.; Chelucci, G.; Gladiali, S.; Siega, K.; Toniutti,
M.; Zanette, M.; Zangrando, E.; Rigo, P. Angew. Chem., Int. Ed. 2005,
44, 6214.
(14) (a) Baratta, W.; Ballico, M.; Esposito, G.; Rigo, P. Chem.Eur. J.
2008, 14, 5588. (b) Baratta, W.; Chelucci, G.; Herdtweck, E.; Magnolia,
S.; Siega, K.; Rigo, P. Angew. Chem., Int. Ed. 2007, 46, 7651. (c) Baratta,
W.; Siega, K.; Rigo, P. Chem.Eur. J. 2007, 13, 7479. (d) Baratta, W.;
Siega, K.; Rigo, P. Adv. Synth. Catal. 2007, 349, 1633. (e) Baratta, W.;
Bosco, M.; Chelucci, G.; Del Zotto, A.; Siega, K.; Toniutti, M.;
e
(17) (a) Ohkuma, T.; Koizumi, M.; Muniz, K.; Hilt, G.; Kabuto, C.;
Noyori, R. J. Am. Chem. Soc. 2002, 124, 6508. (b) Abdur-Rashid, K.;
Clapham, S. E.; Hadzovic, A.; Harvey, J. N.; Lough, A. J.; Morris, R. H.
J. Am. Chem. Soc. 2002, 124, 15104. (c) Dong, Z. R.; Li, Y. Y.; Chen, J.
S.; Li, B. Z.; Xing, Y.; Gao, J. X. Org. Lett. 2005, 7, 1043. (d) Esteruelas,
::
~
M. A.; Garcꢀıa-Yebra, C.; Onate, E. Organometallics 2008, 27, 3029.
~
(18) Castarlenas, R.; Esteruelas, M. A.; Onate, E. Organometallics
2008, 27, 3240.
(19) Corberan, R.; Peris, E. Organometallics 2008, 27, 1954.
(20) Ohkuma, T.; Tsutsumi, K.; Utsumi, N.; Arai, N.; Noyori, R.;
Murata, K. Org. Lett. 2007, 9, 255.
(21) Ohkuma, T.; Utsumi, N.; Watanabe, M.; Tsutsumi, K.; Arai, N.;
Murata, K. Org. Lett. 2007, 9, 2565.
ꢀ ꢀ
(22) Carrion, M. C.; Jalon, F. A.; Manzano, B. R.; Rodrꢀıguez, A. M.;
ꢀ
Sepulveda, F.; Maestro, M. Eur. J. Inorg. Chem. 2007, 3961.
Zangrando, E.; Rigo, P. Organometallics 2006, 25, 4611.
(15) (a) Backvall, J. E. J. Organomet. Chem. 2002, 652, 105.
::
ꢀ
ꢀ
(b) Palmer, M. J.; Wills, M. Tetrahedron: Asymmetry 1999, 10, 2045.
(16) Sandoval, C. A.; Ohkuma, T.; Noyori, R. J. Am. Chem. Soc.
2003, 125, 13490.
(23) Caballero, A.; Jalon, F. A.; Manzano, B. R.; Espino, G.; Perez-
Manrique, M.; Mucientes, A.; Poblete, F. J.; Maestro, M. Organome-
tallics 2004, 23, 5694.

ꢀ
Carrion et al.
3824 Organometallics, Vol. 28, No. 13, 2009
2-propanol, yielding hydrides that could be the active
species. In these hydrides, the partial decoordination of the
NN ligand will allow coordination of the substrate to be
hydrogenated. In this way, the bis(pyrazol-1-yl)methane
derivatives could act as hemilabile ligands and assist the
hydrogen transfer reaction in a similar way to the amido
ligand in the acetamido complex [RuH(PCy3)2(CO)-
(CH3CONH)(iPrOH)] reported by Yi.24 It was therefore
decided to explore the behavior of our derivatives in base-
free hydrogenation processes. We were also interested in
evaluating the influence on the base-free catalytic process of
the different functional groups (in some cases of basic
character) present in the phenyl ring. The effect of groups
such as 2-(NH2)Ph and 2-(NO2)Ph in the previously
described derivatives was analyzed along with 2-(OH)Ph,
which was introduced into the derivatives described in
this work for the first time.
changing the color to yellow. After half an hour, 3/4 of
the solution volume was evaporated and the solid obtained
was filtered off. The yellowish solid was washed once
with MeOH and crystallized from 1,2-dichloroethane/hexane.
Yield: 180.6 mg, 87%. Anal. Calcd for RuC47H46N4OBCl 0.5
3
C2H4Cl2: C, 65.54; H, 5.50; N, 6.37. Found: C, 65.39; H, 5.32; N,
6.29. 1H NMR (acetone-d6, 500 MHz, 298 K): 2.64 (s, 6H, Me3-
Pz); 2.66 (s, 6H, Me5-Pz); 5.61 (s, 6H, C6H6); 6.32 (d, 1H, H3-
ArOH, J = 7.5 Hz); 6.36 (s, 2H, H4-Pz); 6.79 (t, 4H, Hp-Ph
(BPh4), J = 7.1 Hz); 6.93 (t, 8H, Hm-Ph(BPh4), J = 7.3 Hz);
7.01 (t, 1H, H4-ArOH, J = 7.5 Hz); 7.07 (d, 1H, H6-ArOH, J =
7.5 Hz); 7.35 (bs, 8H, Ho-Ph(BPh4)); 7.40 (t, 1H, H5-ArOH);
7.83 (s, 1H, HR); 9.16 (s, 1H, OH). 13C{1H} NMR (acetone-d6,
125 MHz, 298 K): 11.37 (Me5-Pz); 16.15 (Me3-Pz); 67.13 (CR);
86.29 (C6H6); 109.17 (C4-Pz); 117.35 (C6-ArOH); 120.43
(C4-ArOH); 121.60 (Cp-Ph(BPh4)); 121.85 (C2-ArOH); 125.36
(Cm-Ph(BPh4)); 128.47 (C3-ArOH); 131.75 (C5-ArOH); 136.39
(Co-Ph(BPh4)); 146.47 (C5-Pz); 155.58 (C1-ArOH); 157.11
(C3-Pz); 164.28 (m, Cipso-B, JC-11B = 49.4 Hz). IR (cm-1):
3473, ν(O-H); 1564, ν(C-N).
Experimental Section
Preparation of [RuCl(p-cymene)(bpz*mArOH)][BPh4] (2).
[RuCl2(p-cymene)]2 (79.6 mg, 0.13 mmol) and bpz*mArOH
(77.1 mg, 0.26 mmol) were mixed in 10 mL of MeOH. After
stirring for 12 h, a solution of 10 mL of MeOH with 174 mg
(0.5 mmol) of NaBPh4 was added to the initial reddish solution.
After 1 h, the solution was completely evaporated, obtaining an
oil, which was triturated under ultrasound after adding 2 mL of
MeOH. The reddish solid obtained was then filtered off. The
product was crystallized from 1,2-dichloroethane/hexane.
General Considerations. All manipulations were carried out
under an atmosphere of dry oxygen-free nitrogen using standard
Schlenk techniques. Solvents were distilled from the appropriate
drying agents and degassed before use. Elemental analyses were
performed with a Thermo Quest FlashEA 1112 microanalyzer;
IR spectra, on a Shimadzu IRPrestige-21 IR spectrometer
equipped with a Pike Technologies ATR. 1H and 13C{1H}
NMR spectra were recorded on a Varian Unity 500 spectro-
meter. Chemical shifts (ppm) are relative to TMS (1H, 13C
Yield: 177.4 mg, 77%. Anal. Calcd for RuC51H54N4OBCl
3
C2H4Cl2: C, 64.61; H, 5.93; N, 5.69. Found: C, 64.54; H, 5.72;
N, 4.96. 1H NMR (acetone-d6, 500 MHz, 298 K): 0.96 (d, 6H,
Me(iPr), J = 6.8 Hz); 1.07 (m, 1H, CH(iPr)); 2.17 (s, 3H,
Me(Tol)); 2.61 (s, 6H, Me5-Pz); 2.62 (s, 6H, Me3-Pz); 5.64 (d,
1
NMR). H-1H COSY spectra: standard pulse sequence with
an acquisition time of 0.214 s, pulse width 10 ms, relaxation
delay 1 s, number of scans 16, number of increments 512. For
1H-13C g-HMBC and g-HMQC spectra the standard VAR-
IAN pulse sequences were used (VNMR 6.1 C software). The
spectra were acquired using 7996 Hz (1H) and 25133.5 Hz (13C)
widths; 16 transients of 2048 data points were collected for
each of the 256 increments. NOESY spectra were acquired using
8000 Hz width, and 16 transients of 2048 data points were
collected for each of the 256 increments, pulse time = 1 s and
mixing time of 1 s. For variable-temperature spectra, the probe
temperature ((1 K) was controlled by a standard unit calibrated
with a methanol reference. o, m, and p stand for ortho, meta,
and para. s, d, t, and p stand for singlet, doublet, triplet, and
apparent for the NMR resonances. Unless otherwise stated, the
13C{1H} NMR signals are singlets. The ligands bpz*mArNO2,
bpzmArNO2, and bpz*mArNH2,22 bpz*m,25 bpz*mPh,26,27
bpzmArOH and bpzm*ArOH28 were prepared according to
the methods described in the literature. Starting materials:
[RuCl2( p-cymene)]2,29 [RuCl2(C6H6)]2,29 and [RuCl2(C6H6)-
(CH3CN)]30 were prepared according to literature procedures.
NaBPh4 was used as purchased from Aldrich.
0
0
2H, H3/3 (p-cym), J = 6.5 Hz); 5.72 (d, 2H, H2/2 (p-cym), J =
6.8 Hz); 6.34 (s, 2H, H4-Pz); 6.42 (d, 1H, H3-ArOH, J =
7.3 Hz); 6.79 (t, 4H, Hp-Ph(BPh4), J = 7.3 Hz); 6.94 (t, 8H,
Hm-Ph(BPh4), J = 7.3 Hz); 7.01 (t, 1H, H4-ArOH, J = 7.3 Hz);
7.08 (d, 1H, H6-ArOH, J = 7.3 Hz); 7.35 (bs, 8H, Ho-Ph(BPh4));
7.41 (t, 1H, H5-ArOH); 7.75 (s, 1H, HR); 9.41 (s, 1H, OH). 13C-
{1H} NMR (acetone-d6, 125 MHz, 298 K): 11.66 (Me5-Pz);
16.54 (Me3-Pz); 17.40 (Me(Tol)); 22.34 (Me(iPr)); 66.89 (CR);
0
82.63 (CH2/2’(p-cym)); 85.17 (CH3/3 (p-cym)); 109.49 (C4-Pz);
117.03 (C6-ArOH); 120.28 (C4-ArOH); 121.40 (C2-ArOH);
121.60 (Cp-Ph(BPh4)); 125.36 (Cm-Ph(BPh4)); 129.11 (C3-
ArOH); 132.07 (C5-ArOH); 136.39 (Co-Ph(BPh4)); 146.54
(C5-Pz); 156.53 (C1-ArOH); 157.26 (C3-Pz); 164.28 (m, Cipso-
B, JC-11B = 49.4 Hz). IR (cm-1): 3448, ν(O-H); 1565, ν(C-N).
Preparation of [RuCl(benzene)(bpzmArOH)][BPh4] (3). The
method was similar to that used for complex 1. Amounts were as
follows: 72.8 mg (0.25 mmol) of [RuCl2(C6H6)(CH3CN)] and
60.1 mg (0.25 mmol) of bpzmArOH. 3 was obtained as a
yellowish solid and crystallized from 1,2-dichloroethane/hex-
Preparation of [RuCl(benzene)(bpz*mArOH)][BPh4] (1).
[RuCl2(C6H6)(CH3CN)] (72.8 mg, 0.25 mmol) and bpz*mAr-
OH (74.1 mg, 0.25 mmol) were mixed in 10 mL of MeOH. After
stirring for 12 h, a solution of NaBPh4 (200 mg, 0.6 mmol) in
10 mL of MeOH was added to the initial orange solution,
ane. Yield: 150.4 mg, 75%. Anal. Calcd for RuC43H38N4OBCl
3
0.5C2H4Cl2: C, 64.17; H, 4.95; N, 6.80. Found: C, 64.39; H,
4.95; N, 6.63. Isomer ratio A/B = 2.5:1, determined by
1H NMR. Isomer A: 1H NMR (acetone-d6, 500 MHz, 298 K):
6.27 (s, 6H, C6H6); 6.32 (t, 1H, H4-ArOH, J = 7.2 Hz); 6.65
(d, 1H, H3-ArOH, J = 7.2 Hz); 6.69 (t, 2H, H4-Pz); 6.79 (t, 4H,
Hp-Ph(BPh4), J = 7.1 Hz); 6.93 (t, 8H, Hm-Ph(BPh4), J =
7.3 Hz); 7.04 (td, 1H, H5-ArOH, J = 7.2 Hz/J = 1.3 Hz); 7.09
(bs, 1H, H6-ArOH); 7.35 (bs, 8H, Ho-Ph(BPh4)); 7.71 (s, 1H,
HR); 8.23 (d, 2H, H5-Pz, J = 2.7 Hz); 8.92 (d, 2H, H3-Pz, J =
2.3 Hz). 13C{1H} NMR (acetone-d6, 125 MHz, 298 K): 86.82
(C6H6); 108.05 (C4-Pz); 116.84 (C-ArOH); 119.96 (C-ArOH);
121.60 (Cp-Ph(BPh4)); 125.36 (Cm-Ph(BPh4)); 128.04 (C-
ArOH); 133.47 (C5-Pz); 136.39 (Co-Ph(BPh4)); 148.04 (C3-Pz);
(24) Yi, Ch. S.; He, Z.; Guzei, I. A. Organometallics 2001, 20, 3641.
(25) Sebastian, J.; Sala, P.; Del Mazo, J.; Sancho, M.; Ochoa, C.;
Elguero, J.; Fayet, T. A.; Vertut, M. C. J. Heterocycl. Chem. 1982, 19,
1141.
(26) The, K. I.; Peterson, L. K. Can. J. Chem. 1973, 51, 422.
(27) (a) The, K. I.; Peterson, L. K.; Kiehlmann, E. Can. J. Chem.
1974, 52, 2448. (b) Peterson, L. K.; Kiehlmann, E.; Sanger, A. R.; The,
K. I. Can. J. Chem. 1974, 52, 2367.
(28) Higgs, T. C.; Carrano, J. C. Inorg. Chem. 1997, 36, 291.
(29) Bennett, M. A.; Huang, T.-N.; Matheson, T. W.; Smith, A. K.;
Ittel, S.; Nickerson, W.; Fackler, J. F. Inorg. Synth., 1974, 21, 75.
(30) Takahashi, H.; Kobayashi, K.; Osawa, M. Anal. Sci. 2000, 16,
n777.
1
164.28 (m, Cipso-B, JC-11B = 49.4 Hz). Isomer B: H NMR
(acetone-d6, 500 MHz, 298 K): 6.30 (s, 6H, C6H6); 6.55 (t, 2H,
H4-Pz, J = 2.9 Hz); 6.79 (t, 4H, Hp-Ph(BPh4), J = 7.1 Hz); 6.93

Article
Organometallics, Vol. 28, No. 13, 2009 3825
(CH(iPr)); 59.13 (CR); 82.33 (CH2/2 (p-cym)); 84.98 (CH3/3 (p-
cym)); 100.34 (C4(p-cym)); 107.87 (C1(p-cym)); 109.32 (C4-Pz);
143.78 (C5-Pz); 156.06 (C3-Pz). IR (cm-1): 1556, ν(C-N); 1049,
ν(BF4).
0
0
(t, 8H, Hm-Ph(BPh4), J = 7.3 Hz); 7.20 (d, 1H, H3-ArOH, J =
7.2 Hz); 7.30 (td, 1H, H4-ArOH, J = 7.2 Hz/J = 0.8 Hz); 7.35
(bs, 8H, Ho-Ph(BPh4)); 7.67 (td, 1H, H5-ArOH, J = 7.2 Hz/J =
1.6 Hz); 7.76 (d, 2H, H5-Pz, J = 2.9 Hz); 7.86 (bs, 1H, H6-
ArOH); 8.17 (d, 2H, H3-Pz, J = 1.4 Hz). 13C{1H} NMR
(acetone-d6, 125 MHz, 298 K): 76.19 (CR); 87.00 (C6H6);
107.88 (C4-Pz); 118.00 (C-ArOH); 121.60 (Cp-Ph(BPh4));
125.36 (Cm-Ph(BPh4)); 134.48 (C-ArOH); 135.13 (C5-Pz);
136.39 (Co-Ph(BPh4)); 148.86 (C3-Pz); 164.28 (m, Cipso-B,
Preparation of [RuCl(p-cymene)(bpz*m)]Cl (6). [RuCl2-
(p-cymene)]2 (79.6 mg, 0.13 mmol) and bpz*m (53.1 mg,
0.26 mmol) were mixed in 10 mL of MeOH. After stirring
for 24 h, the solvent was evaporated and the solid obtained
was washed with pentane and hexane. The yellowish solid
obtained was dried under vacuum. Complex 6 was obtained
as an orange solid. Yield: 111.5 mg, 84%. Anal. Calcd for
J
C-11B = 49.4 Hz). IR (cm-1): 3203, ν(O-H); 1577, ν(C-N).
Preparation of [RuCl(p-cymene)(bpzmArOH)][BPh4] (4) and
RuC21H30N4Cl2F4 0.5C2H4Cl2: C, 46.70; H, 5.65; N, 10.13.
[Ru(bpzmArO)(p-cymene)][BPh4] (4C). The method was similar
to that used for complex 2. Amounts were as follows: 79.6 mg
(0.13 mmol) of [RuCl2(p-cymene)]2 and 62.5 mg (0.26 mmol)
of bpzmArOH. A mixture of 4 + 4C was obtained as a yellow
solid (163.9 mg). 4C was crystallized from 1,2-dichloroethane/
hexane. Anal. Calcd for C47H45BN4ORu C2H4Cl2 (4C
3
Found: C, 46.22; H, 5.98; N, 10.30. 1H NMR (CDCl3, 500 MHz,
298 K): 1.17 (d, 6H, Me(iPr), J = 6.6 Hz); 2.33 (s, 3H, Me(Tol));
2.51 (s, 3H, Me3-Pz); 2.52 (s, 3H, Me5-Pz); 3.19 (m, 1H,
CH(iPr)); 5.86 (d, 2H, H3/3 (p-cym), J = 6.2 Hz); 6.04 (s, 2H,
0
0
H4-Pz); 6.06 (d, 2H, H2/2 (p-cym), J = 6.3 Hz); 6.49 (d, 1H, HR,
0
3
3
J = 15.4 Hz); 6.80 (d, 1H, HR , J = 15.7 Hz). 13C{1H} NMR
(CDCl3, 125 MHz, 298 K): 11.59 (Me5-Pz); 16.03 (Me3-Pz);
C2H4Cl2): C, 65.93; H, 5.53; N, 6.28 . Found: C, 65.45; H,
5.32; N, 6.69. Isomer ratio from the reaction, A/B/C =
1:1.8:3.2, determined by 1H NMR. Complex 4: Isomer A:
1H NMR (acetone-d6, 500 MHz, 298 K): 1.24 (d, 6H,
Me(iPr), J = 7.7 Hz); 1.85 (s, 3H, Me(Tol)); 2.63 (bs, 1H,
18.76 (Me(Tol)); 22.92 (Me(iPr)); 31.62 (CH(iPr)); 57.77 (CR);
0
0
82.01 (CH2/2 (p-cym)); 84.82 (CH3/3 (p-cym)); 100.36 (C4(p-
cym)); 108.06 (C1(p-cym)); 109.30 (C4-Pz); 143.75 (C5-Pz);
156.01 (C3-Pz). IR (cm-1): 1556, ν(C-N).
CH(iPr)); 5.82 (bs, 2H, H3/3 (p-cym)); 5.89 (bs, 2H, H2/2’(p-
0
Hydrogen-Transfer Catalysis. A typical procedure for the
catalytic hydrogen-transfer reaction under base-free conditions
is as follows. A mixture of benzophenone (368.1 mg, 2 mmol)
and the catalyst (0.002 mmol) was dissolved in 2-propanol
(10 mL), and the mixture was heated under reflux under nitro-
gen. At the desired reaction times, aliquots were extracted from
the reaction vessel, and the yields were determined by 1H NMR
spectroscopy.
For the reaction under basic conditions the procedure is
similar: a mixture of benzophenone (368.1 mg, 2 mmol), KOH
(10 mL, 0.008 M in 2-propanol), and the catalyst (0.004 mmol)
was heated under reflux under nitrogen.
cym)); 6.79 (t, 4H, Hp-Ph(BPh4), J = 7.1 Hz); 6.94 (t, 8H, Hm-
Ph(BPh4), J = 7.3 Hz); 7.35 (bs, 8H, Ho-Ph(BPh4)); 7.42 (bs,
1H, H5-ArOH); 7.54 (bs, 1H, H6-ArOH); 8.39 (bs, 2H, H5-Pz);
8.52 (bs, 2H, H3-Pz). Isomer B: 1H NMR (acetone-d6,
500 MHz, 298 K): 1.30 (d, 6H, Me(iPr), J = 6.8 Hz); 2.20
(s, 3H, Me(Tol)); 3.06 (m, 1H, CH(iPr)); 5.99 (bs, 2H, H3/3 (p-
0
cym)); 6.08 (d, 2H, H2/2’(p-cym), J = 5.4 Hz); 6.58 (bs, 2H, H4-
Pz); 6.79 (t, 4H, Hp-Ph(BPh4), J = 7.1 Hz); 6.94 (t, 8H, Hm-Ph
(BPh4), J = 7.3 Hz); 7.23 (bs, 1H, H4-ArOH); 7.35 (bs, 8H,
Ho-Ph(BPh4)); 7.68 (t, 1H, H5-ArOH, J = 7.8 Hz); 7.79
(d, 2H, H5-Pz, J = 2.4 Hz); 7.91 (bs, 1H, H6-ArOH); 8.11
(bs, 2H, H3-Pz). 13C{1H} NMR (acetone-d6, 125 MHz, 298 K):
17.94 (Me(Tol)); 21.79 (Me(iPr)); 31.25 (CH(iPr)); 121.60 (Cp-
Ph(BPh4)); 125.36 (Cm-Ph(BPh4)); 136.39 (Co-Ph(BPh4));
164.28 (m, Cipso-B, JC-11B = 49.4 Hz). Complex 4C:
1H NMR (acetone-d6, 500 MHz, 298 K): 1.19 (d, 6H,
Me(iPr), J = 6.8 Hz); 2.30 (s, 3H, Me(Tol)); 2.97 (m, 1H,
All of the catalytic reactions were run at least three times to
ensure the reproducibility of the results.
Special Catalytic Tests. Apart from the standard catalytic
tests, several special tests were carried out in order to obtain
information about the mechanism of the reaction. Following the
general procedure described above, in the pretreatment tests the
benzophenone was added after the desired incubation period of
the precatalyst at 82 °C in 10 mL of 2-propanol, and the catalytic
reaction started at that moment. Other special tests were carried
out with 300 molar equiv of Hg or 20 molar equiv of free
p-cymene added to the catalytic media from the beginning of
the reaction. Finally, some tests were carried out in which the
precatalysts of the reaction were generated “in situ” by adding
the corresponding amounts of ligand, [RuCl2(p-cymene)]2, and
NaBPh4 to the reaction media.
CH(iPr)); 5.96 (d, 2H, H3/3 (p-cym), J = 6.5 Hz); 6.29 (d, 2H,
0
H
2/2’(p-cym), J = 6.5 Hz); 6.33 (t, 1H, H4-ArOH, J = 8.1 Hz);
6.68 (d, 1H, H6-ArOH, J = 8.1 Hz); 6.70 (bs, 2H, H4-Pz); 6.79
(t, 4H, Hp-Ph(BPh4), J = 7.1 Hz); 6.94 (t, 8H, Hm-Ph(BPh4),
J = 7.3 Hz); 6.99 (td, 1H, H5-ArOH, J = 8.1 Hz/J = 1.4 Hz);
7.09 (dd, 1H, H3-ArOH, J = 8.1 Hz/J = 1.4 Hz); 7.35 (bs, 8H,
Ho-Ph(BPh4)); 7.68 (s, 1H, HR); 8.23 (d, 2H, H5-Pz, J =
2.9 Hz); 8.89 (d, 2H, H3-Pz, J = 1.9 Hz). 13C{1H} NMR
(acetone-d6, 125 MHz, 298 K): 16.91 (Me(Tol)); 21.77 (Me-
(iPr)); 30.47 (CH(iPr)); 76.26 (CR); 83.19 (CH2/2’(p-cym));
0
86.18 (CH3/3 (p-cym)); 108.70 (C4-Pz); 113.77 (C4-ArOH);
Kinetic Studies. The precatalyst 2 was placed (acetophenone/
catalyst ratio = 1000:1, 1000:4.4, 1000:9.6, and 1000:18,
where 1 equiv means 0.0001 mmol) in a Young Teflon-capped
NMR tube followed by the addition of 0.5 mL of 2-propanol-
d8. Acetophenone (11.8 μL, 0.1 mmol) was then added.
Immediately after the addition, the 1H NMR spectra to
determine the yield of the process were registered at 80 °C
every 10 min for the first hour, every 15 min for the second
hour, and every 30 min for the third and fourth hours. The
same experiment was performed at 70 and 75 °C in order to
determine the activation parameters, which were evaluated
from the Eyring equation.
121.60 (Cp-Ph(BPh4)); 123.37 (C6-ArOH); 125.36 (Cm-Ph-
(BPh4)); 129.99 (C3-ArOH); 131.95 (C5-ArOH); 133.62
(C5-Pz); 136.39 (Co-Ph(BPh4)); 147.77 (C3-Pz); 164.28
(m, Cipso-B, JC-11B = 49.4 Hz). IR (cm-1): 3344, ν(O-H);
1598, ν(C-N).
Preparation of [RuCl(p-cymene)(bpz*m)][BF4] (5). The meth-
od was similar to that used for complex 2. Amounts were as
follows: 79.6 mg (0.13 mmol) of [RuCl2(p-cymene)]2, 53.1 mg
(0.26 mmol) of bpz*m, and 122.5 mg (0.60 mmol) of NaBF4. 5
was obtained as a dark yellow solid. Yield: 119.8 mg, 82%. Anal.
Calcd for RuC21H30N4BClF4 2C2H4Cl2: C, 37.76; H, 4.68; N,
3
7.74. Found: C, 37.68; H, 4.05; N, 7.74. 1H NMR (CDCl3,
500 MHz, 298 K): 1.18 (d, 6H, Me(iPr), J = 6.9 Hz); 2.31 (s, 3H,
Me(Tol)); 2.48 (s, 3H, Me3-Pz); 2.52 (s, 3H, Me5-Pz); 2.98 (sept,
X-ray Crystallography. A summary of crystal data collection
and refinement parameters for 2 and 4C is given in Tables S1 and
S10, respectively.
1H, CH(iPr), J = 6.6 Hz); 5.69 (d, 2H, H3/3 (p-cym), J =
0
Single orange crystals of the compounds [RuCl(p-cymene)
(bpz*mArOH)][BPh4] (2) and [Ru(p-cymene)(bpzmArO)]-
[BPh4] C2H4Cl2 (4C C2H4Cl2) were placed in a Bruker-Non-
6.2 Hz); 5.81 (d, 2H, H2/2’(p-cym), J = 6.2 Hz); 6.06 (s, 2H,
0
H4-Pz); 6.15 (d, 1H, HR, J = 15.7 Hz); 6.42 (d, 1H, HR , J =
3
3
15.7 Hz). 13C{1H} NMR (CDCl3, 125 MHz, 298 K): 11.97 (Me5-
ius X8 APEXII CCD area-detector diffractometer, equipped
with a graphite-monochromated Mo KR radiation source
Pz); 15.95 (Me3-Pz); 18.79 (Me(Tol)); 22.97 (Me(iPr)); 31.32

ꢀ
Carrion et al.
3826 Organometallics, Vol. 28, No. 13, 2009
˚
(λ = 0.71073 A). The data were collected by using the full
Chart 1. Ligands Used in This Work and Their Abbreviations
sphere data collection routine with 0.3° wide ω scans.
The substantial redundancy for the three compounds allows
empirical absorption corrections (SADABS31) to be applied
using multiple measurements of symmetry-equivalent reflec-
tions (ratio of minimum to maximum apparent transmission:
0.8401 for 2 and 0.8573 for 4C C2H4Cl2). The raw intensity data
3
frames were integrated with the SAINT32 program, which also
applied corrections for Lorentz and polarization effects.
The software package SHELXTL33 was used for space group
determination, structure solution, and refinement. The space
group determination was based on a check of the Laue symme-
try and systematic absences and was confirmed using the
structure solution. The structures were solved by direct meth-
ods, completed with difference Fourier syntheses, and refined
with full-matrix least-squares minimization w(F2o - Fc2)2.
Weighed R factors (Rw) and all goodness-on-fit S values are
based on F2; conventional R factors (R) are based on F. All non-
hydrogen atoms were refined with anisotropic displacement
parameters. All hydrogen atoms were located in difference
Fourier maps, but their positions were calculated geometrically
and were allowed to ride on their parent carbon atoms with fixed
isotropic U.
(p-cymene)]BPh4. This complex could be formed from the
equatorial isomer B according to the reaction outlined in
Scheme 2.
Isomers A and B would be in equilibrium in solution, and
from the equatorial isomer B, a boat-to-boat inversion
process could lead to a transient species in which the OH
group would be near the chloride and the metallic center. In
this situation the activation of the OH group should take
place and the loss of one HCl molecule would lead to
complex 4C. Other possible mechanisms for the formation
of 4C are possible. For example, decoordination of one
pyrazolyl ring in isomer A could allow the interaction of
the OH group with the metallic center and the subsequent
formation of the tridentate ligand after HCl elimination.
However, considering the fact that this latter mechanism
does not involve the existence of the equatorial isomer the
formation of a complex similar to 4C from complex 2 that
differs only in the presence of methylated pyrazolyl groups
should be possible, but this has not been observed. It is
noteworthy that the formation of a similar complex to 4C
was not observed with the benzene derivative 3, which also
has the equatorial isomer. Probably, the higher basicity of
the p-cymene ring as compared to benzene favors the O-H
activation.
Results and Discussion
Synthesis and Characterization. The ligands used in this
work and their abbreviations are summarized in Chart 1.
The synthesis and characterization of all of these ligands
have been described previously.22,25-28 The synthesis of the
intermediate bispyrazolyl ketone was carried out with a safer
methodology22 involving triphosgene instead of phosgene.
This compound was then reacted with the corresponding
aldehyde to obtain the ligand.
The Ru(II) arene (arene = benzene, p-cymene) complexes
were obtained by reacting the corresponding Ru(II) starting
material, the nature of which depends on the arene, with the
desired ligand in MeOH with the addition of NaBPh4 or
NaBF4 at the end of the reaction (Scheme 1). Salt was not
added in the case of complex 6.
Endobidentate coordination with the formation of metal-
lacycles in a boat conformation was found in all cases
(Chart 2). In principle, the aryl substituent can adopt an
axial (isomer A) or equatorial orientation (isomer B) in the
metallacycle. We have previously found that when non-
methylated pyrazolyl rings are present, both isomers are
formed,22 while with 3,5-dimethylated pyrazolyl rings isomer
B is not found, possibly due to steric hindrance between this
substituent and the Me5 of the pyrazolyl rings. This situation
is in agreement with the results for complexes 1-3, 5, and 6
(see Experimental Section for the isomer ratio obtained)
since only in complex 3 is isomer B present (29%). The
isomer interconversion has been observed, and partial
decoordination of the nitrogenated ligand has been proposed
as the most plausible route for interconversion.22 In the case
of complex 4, besides the two isomers, a new type of
derivative (4C) is also obtained as the major component. In
4C the bpzmArOH ligand has been deprotonated and is
coordinated in a tridentate fashion: [Ru(bpzmArO-κ3-N,N,O)-
The new complexes were characterized by elemental ana-
lysis as well as by IR, 1H NMR, and 13C{1H} NMR spectro-
scopy. In certain cases X-ray diffraction studies were carried
out. The assignment of the resonances of the new complexes
1
was possible with the aid of H,1H-COSY and g-HMQC
spectra (in some cases also with g-HMBC) and by consider-
1
ing the values of coupling constants. The H NMR and
13C{1H} NMR data are gathered in the Experimental
Section. The structures of both isomers (where applicable)
were elucidated with the aid of NOESY spectra by following
the procedure described in our previous publication22 on this
kind of derivative. The most informative NOE (see Chart 2)
was that observed between H5 (or Me5) of the pyrazolyl
groups and HR (isomers A) or ortho protons of the R group
(isomers B). As in the previously described derivatives,22 the
H5 pyrazolyl protons of isomer B appear more shielded than
those of isomer A.
Crystal Structures. The molecular structures of complexes 2
and 4C were determined by X-ray diffraction. The crystal-
lographic data are gathered in Tables S1 and S10. The corres-
ponding ORTEP diagrams are reflected in Figures 1 and 2.
Complex 2 has a three-legged half-sandwich structure
consisting of the coordinated arene, a chloride, and the
bidentate nitrogenated ligand. The phenyl group has an axial
disposition on the boat metallacycle. The bond distances are
in the expected range.22 The metallacycle is quite flat in the
region of the ruthenium atom: the dihedral angle NRuN/N4
is 165.62°. The angle formed by the phenyl ring and the plane
formed by the atoms HR-CR-C(ipso) is 69.5°, not the 90°
(31) Sheldrick, G. M. SADABS version 2004/1, A Program for
::
::
Empirical Absorption Correction; University of Gottingen: Gottingen,
Germany, 2004.
(32) SAINT+ v7.12a, Area-Detector Integration Program; Bruker-
Nonius AXS: Madison, WI, 2004.
(33) SHELXTL-NT version 6.12, Structure Determination Package;
Bruker-Nonius AXS: Madison, WI, 2001.

Article
Organometallics, Vol. 28, No. 13, 2009 3827
Scheme 1. New Derivatives Synthesized in This Work
Chart 2. Isomers A and B for the Arene-Ru(II) Derivatives with
the Substituent on the Aryl Group in the Axial (A) or the
Equatorial (B) Positiona
Scheme 2. Possible Pathway for the Formation of Complex 4Ca
a The arrows show the most representative NOEs observed. Both
isomers are monopositive cations.
that one would expect to reduce the steric hindrance with
the arene group. In addition, the isopropyl group of the
p-cymene ring is oriented toward the inside of the molecule, a
situation not favored from steric requirements. The explana-
tion for these facts could be the existence of weak hydrogen
bonds34 between the oxygen or chloride atoms and different
groups of the molecule (see Table S6 for the distances and
a All of the species are monopositive cations.
vacuum, and analyzed by NMR spectroscopy in order to
study the evolution of the products with the reaction time. In
Tables 2-4 the yield at 24 h is reflected and also the TOF
value. The TOF was calculated when the yield of hydro-
genated product was 50% in order to minimize the influence
of the concentration of substrate. In the Supporting Infor-
mation plots reflecting the evolution with time for all the
experiments and more complete tables with the yield and
TOF values at 1.5, 3, 6, 9, and 24 h are also included.
The catalytic behavior was studied for the complexes
described for the first time in this work (1-6) and also with
other compounds previously reported by us (see Table 1
for numbering).22 For the sake of comparison, the results
obtained in the presence of base (results obtained in this work
or taken from ref 22) were also introduced (see Tables 2 and 3,
runs b). The substrate/catalyst/KOH ratio used in these cases
was500:1:20, and it was previouslyconfirmedthatthe amount
of base used was not sufficient to catalyze the reaction without
the addition of a metal complex.22 The product was not
observed at all without the addition of the precatalyst or when
the reaction was performed at room temperature (verified for
complex 2). The results were divided into two sets depending
on the arene group present in the precatalyst. It was observed
that the p-cymene complexes generally exhibit much higher
activities than the benzene derivatives.
angles) and also a C-H π interaction35 between the
3 3 3
isopropylic proton and the phenol ring (H25 centroid
˚
distance of 2.443 A). All of these interactions are shown in
3 3 3
Figure 1b. In addition, in both compounds the chloride
atoms are involved in hydrogen bonds with protons of the
Me3 groups of both pyrazolyl rings (Table S6).
Complex 4C also has a half-sandwich structure, but this is
clearly different because it is a scorpionate derivative due to
the tridentate N,N,O coordination of the anionic nitroge-
nated ligand. The distances between the Ru center and the
coordinated atoms are in the expected range for this kind of
derivative.22 Concerning the orientation of the p-cymene
ring, the isopropyl group is situated between the two pyr-
azolyl rings, and a weak hydrogen bond between the toluenic
methyl group and the oxygen atom is observed (Table S6).
Catalytic Transfer Hydrogenation of Benzophenone. As
previously stated, our goal was to perform base-free transfer
hydrogenation processes of ketones using our stable arene
derivatives as precursors. The results were positive for all
derivatives, although there were differences between them,
which will be explained below (Tables 2 and 3, runs a).
The hydrogenation of benzophenone was studied using
the following general conditions: substrate/catalyst ratio =
1000:1, no base, 2-propanol as solvent, and a temperature of
82 °C. Samples were collected at different times, dried under
In the different tables where catalytic results are gathered,
the column corresponding to the complex also contains
(in brackets) the corresponding nitrogenated ligand in order
to facilitate the comparisons between different ligands.
The activity obtained with the benzene derivatives
(Table 2) is only moderate. The starting compound [RuCl2-
(benzene)(CH3CN)] gives a yield of 12% The complexes that
(34) Jeffrey, G. A. An Introduction to Hydrogen Bonding; Oxford
University Press: New York, 1997.
(35) (a) Takahashi, H.; Tsuboyama, S.; Honda, K.; Nishio, M.
Tetrahedron 2000, 56, 6185. (b) Re, S.; Nagase, S. Chem. Commun.
2004, 658.

ꢀ
Carrion et al.
3828 Organometallics, Vol. 28, No. 13, 2009
Figure 1. (a) ORTEP diagram of the cation of complex 2, [RuCl(p-cymene)(bpz*mArOH)]+. Ellipsoids are at the 50% level.
˚
Hydrogen atoms have beenomitted for clarity. Distances (A):Ru(1)-N(3) = 2.113(4);Ru(1)-N(1) = 2.133(4);Ru(1)-Cl(1) =2.389(2);
Ru(1)-Caverage = 2.203. Angles (deg): N(3)-Ru(1)-N(1) = 87.0(2); N(3)-Ru(1)-Cl(1) = 83.4(1); N(1)-Ru(1)-Cl(1) = 82.6(1).
(b) Some of the weak interactions observed in 2 (dashed lines).
detected. The presence of a functional group (NO2, NH2,
or OH) in the aryl ring of the precatalysts increases the
yield slightly, but the effect on the TOF values is more
marked. The best results are obtained when the ligands
bpz*mArNO2, bpzmArNO2, and bpz*mArOH are present,
and in these cases the reaction is almost complete at 9 h
(see Figure 3). The poor behavior of 4 (containing ligand
bpzmArOH, run 17a), with a yield much lower than that of
the dimethylated derivative (run 15a), must be due to the
formation of the scorpionate derivative 4C, with the ligand
acting in a tridentate fashion. The efect of the methyl groups
in the pyrazole rings is not marked.
When the precatalyst was formed “in situ” by mixing
[RuCl2(p-cymene)]2, the corresponding ligand, and NaBPh4
(runs 18a-20a in Table 3), a lower activity was found in
comparison with the preformed precursors (biggest differ-
ence for the phenol derivative), which indicates that it is more
convenient to use a preformed complex.
The results we have described here are particularly rele-
vant because a base has not been added and the chloride
Figure 2. ORTEP diagram of the cation of complex 4C, [Ru-
(p-cymene)(bpzmArO)]+. Ellipsoids are at the 50% level.
˚
Hydrogen atoms have been omitted for clarity. Distances (A):
derivatives, which are robust and air-stable, have been used
as precursors. Furthermore, for the p-cymene derivatives it is
also noteworthy that the use of the same precursors in
processes with the addition of base (KOH) always gave a
lower final yield at 24 h22 even with a lower substrate:catalyst
ratio. In any case, some experiments were performed with the
same substrate/catalyst ratio, and the yields were clearly
higher without base in the media (compare runs 9c/9b and
15c/15b, see Figure 4). The marked reduction in yield in the
case of the bpz*mArOH derivative may be due to deproto-
nation of the phenol group and coordination of the ligand as
a tridentate system. However, the significant decrease in
activity observed with base for other complexes suggests
the existence of other deactivation processes.
For the benzene derivatives a clear trend is not observed
on comparing the experiments with the addition of KOH and
those performed under base-free conditions.
When the results of the derivatives 5, 6, and 8 (see Table 3,
runs 21a, 22a, and 3a) are compared, it is concluded that the
influence of the counteranion is not marked. The TOF value
Ru(1)-N(4) = 2.093(3); Ru(1)-N(1) = 2.107(2); Ru(1)-O(1) =
2.053(2); Ru(1)-Caverage = 2.185. Angles (deg): N(4)-Ru(1)-
N(1) = 86.61(10); O(1)-Ru(1)-N(4) = 82.78(10); O(1)-
Ru(1)-N(1) = 88.17(9).
gave the best results are those that contain the ligands
bpzmArNO2 and bpzmPh (runs 10a and 6a, respectively).
The complexes with methylated pyrazolyl ligands generally
show worse behavior than the corresponding nonmethylated
ones (Table 2, cf. runs 4a/6a, 8a/10a, and 14a/16a).
The results obtained for the p-cymene derivatives are
clearly better than the others (Table 3). A selection of data
is included in Figure 3. The [RuCl2(p-cymene)]2 derivative is
active, although the results are poor (run 1a).
When the ligand bpz*m, which does not have any sub-
stituent on the methylene group, is present in the complex
(8, run 3a), a period of induction of about 2 h is observed, but
the final yield is high. When a phenyl group is included in the
ligand (precatalysts 10 and 12), this period of induction is not

Article
Organometallics, Vol. 28, No. 13, 2009 3829
Table 1. [RuCl(arene)(NN)][BPh4] Complexes Previously Described22 and Compound Numberinga
[RuCl(benzene)(NN)][BPh4]
[RuCl(p-cymene)(NN)][BPh4]
[RuCl(benzene)(bpz*m)][BPh4] (7)
[RuCl(p-cymene)(bpz*m)][BPh4] (8)
[RuCl(benzene)(bpz*mPh)][BPh4] (9)
[RuCl(benzene)(bpzmPh)][BPh4] (11)
[RuCl(benzene)(bpz*mArNO2)][BPh4] (13)
[RuCl(benzene)(bpzmArNO2)][BPh4] (15)
[RuCl(benzene)(bpz*mArNH2)][BPh4] (17)
[RuCl(p-cymene)(bpz*mPh)][BPh4] (10)
[RuCl(p-cymene)(bpzmPh)][BPh4] (12)
[RuCl(p-cymene)(bpz*mArNO2)][BPh4] (14)
[RuCl(p-cymene)(bpzmArNO2)][BPh4] (16)
[RuCl(p-cymene)(bpz*mArNH2)][BPh4] (18)
a See Scheme 1 for the new complexes.
Table 2. Transfer Hydrogenation of Benzophenone Using [RuCl-
(benzene)(NN)][BPh4] Complexes As Precatalyst
Table 3. Transfer Hydrogenation of Benzophenone Using [RuCl-
(p-cymene)(NN)][BPh4] Complexes As Precatalyst
with KOH22
with KOH22
TOFb/
TOFb/
TOFb/
TOFb/
complex(NN-ligand) runa h-1 yieldc/% rund h-1 yieldc/%
complex(NN-ligand) runa h-1 yieldc/% rund h-1 yieldc/%
no ligande
7 (bpz*m)
9 (bpz*mPh)
11 (bpzmPh)
0
2a
4a
6a
12
45
20
54
41
65
25
9
no ligande
8 (bpz*m)
10 (bpz*mPh)
12 (bpzmPh)
14 (bpz*mArNO2)
14 (bpz*mArNO2)f
16 (bpzmArNO2)
18 (bpz*mArNH2)
2 (bpz*mArOH)
2 (bpz*mArOH)f
4 (bpzmArOH)
bpz*mh
1a
3a
5a
7a
9a
9c
11a
13a
15a
15c
17a
18a
19a
20a
21a
22a
34
96
96
87
1b
3b
5b
7b
9b
5
61
82
10
60
2b
4b
6b
8b
10b
12b
14b
16b
18
54
62
97
37
60
18
74
26
16
142
82
83
259
178
258
105
240
455
15
28
23
80
13 (bpz*mArNO2)
15 (bpzmArNO2)
17 (bpz*mArNH2)
1 (bpz*mArOH)
3 (bpzmArOH)
a Reaction conditions: benzophenone/catalyst = 1000:1, 0.002 mmol
of precatalyst, solvent: 2-propanol, 10 mL, no base. b Determined at
yield = 50%. c Determined by 1H NMR at t = 24 h. d Reaction
conditions: benzophenone/catalyst/KOH = 500:1:2, solvent: 2-propa-
nol. e The precatalyst used was [RuCl2(benzene)(CH3CN)].
8a
13
28
96
11
18
10a
12a
14a
16a
>99
95
91
11b
13b
15b
8
66
23
31
96
>99
14
75
69
47
17b
10
71
176
h
bpz*ArNO2
bpz*ArOHh
5 (bpz*m)
6 (bpz*m)
118
100
94
90
is higher for the derivative with BPh4, but the yield at 24 h is
similar for all three compounds.
a Reaction conditions: benzophenone/catalyst = 1000:1, 0.002 mmol
of precatalyst, solvent: 2-propanol, 10 mL, no base. b Determined at
yield = 50%. c Determined by 1H NMR at t = 24 h. d Reaction
conditions: benzophenone/catalyst/KOH = 500:1:2, solvent: 2-propa-
nol. e The precatalyst is [RuCl2(p-cymene)]2. f Reaction conditions:
benzophenone/catalyst = 500:1, using 0.004 mmol of precatalyst.
Solvent: 2-propanol, no base. h Precatalyst prepared “in situ”.
The fact that our p-cymene precatalysts show a good
activity under base-free conditions prompted us to study
the transfer hydrogenation of aldehydes. The absence of base
prevents side reactions such as, the deprotonation of the
R-CH group, which would lead to aldol condensation pro-
ducts. The hydrogenation of other ketones besides benzo-
phenone was also tested in order to widen the scope of our
derivatives and to explore the hydrogenation of base-sensi-
tive ketones such as benzylphenylketone. The behavior of the
best catalysts, [RuCl(p-cymene)(bpz*mArOH)][BPh4] (2)
and [RuCl(p-cymene)(bpz*mArNO2)][BPh4] (14), was ana-
lyzed using the unsaturated substrates shown in Table 4. The
hydrogenation of acetophenone (Table 4, runs 23a-24a) was
slightly easier than that of benzophenone, probably due to
the lower level of steric hindrance. In any case, both sub-
strates were almost completely reduced to the alcohol under
the reaction conditions employed. When benzylphenyl
ketone was used (Table 4, runs 25a-28a), the hydrogenation
proved to be more difficult. Furthermore, low yields were
obtained with an aliphatic ketone such as cyclohexanone
(runs 29a-32a). This finding is in contrast to that described
by Peris,19 who found that his iridium catalyst was more
active for aliphatic than for aromatic ketones. In this con-
text, it is interesting to note that for arene-Ru complexes the
possible existence of CH-π interactions between the arene
group of the precatalyst and the aromatic ring of the ketone
has been described.36 It is possible that such interactions are
operating in our derivatives, and this would explain the
difference in activity. For these two substrates, a lower
substrate to precatalyst ratio was also used in order to obtain
good yields. The reduction of benzaldehyde is easier than the
reduction of ketones, and indeed, benzaldehyde was com-
pletely reduced after 9 h (Table 4, runs 33a-34a). If we
compare these results with the reported data on benzalde-
hyde transfer hydrogenation in the absence of base, only the
system described by Esteruelas18 is more active, although its
activity in ketone hydrogenation is very poor. The use of
2-phenylpropanal led to a dramatic decrease in the value of
conversion, and a lower substrate to precatalyst ratio is
necessary for total conversion to the alcohol (Table 4, runs
35a-38a). Once again a plausible explanation for this low
conversion could be the absence of productive CH-π inter-
actions with the arene group of the catalyst.
Mechanistic and Kinetic Studies. Some other special experi-
ments were carried out in order to obtain information on the
mechanism that is operating under base-free conditions.
First, we studied the homogeneous or heterogeneous
character of the reaction. When Hg (300 equiv) (Table 5,
run 39a) was added to the reaction medium (using complex
2), a reduction in the yield of around 20% was observed
(compare to run 15a, Table 3). The fact that the hydrogena-
tion was not inhibited is consistent with a homogeneous
catalytic process.37 The same conclusion had been obtained
for the reaction in the presence of base.22
(37) (a) Anton, D. R.; Crabtree, R. H. Organometallics 1983, 2, 855.
(b) Foley, P.; DiCosimo, R.; Whitesides, G. M. J. Am. Chem. Soc. 1980,
102, 6713. (c) Eberhard, M. R. Org. Lett. 2004, 6, 2125.
(36) Yamakawa, M.; Yamada, I.; Noyori, R. Angew. Chem., Int. Ed.
Engl. 2001, 40, 2818.

ꢀ
Carrion et al.
3830 Organometallics, Vol. 28, No. 13, 2009
Table 4. Transfer Hydrogenation of Different Substrates, Using [RuCl(p-cymene)(NN)]-
[BPh4] Complexes As the Precatalysts
a Reaction conditions: solvent: 10 mL of 2-propanol, 0.002 mmol of precatalyst, no base.
b Precatalyst: 0.006 mmol used. c Precatalyst: 0.004 mmol used. d TOF determined at yield =
50%. e Determined by 1H NMR at t = 24 h. f Yield determined at t = 9 h.
Figure 4. Comparison of the catalytic results for the transfer
hydrogenation of benzophenone for [RuCl(p-cymene)(NN)]
[BPh4] with base and under base-free conditions. (9) 14
(bpz*mArNO2); (O) 2 (bpz*mArOH); (-) no base added; (- -)
40 equiv of KOH added.
Figure 3. Evolution with time of the yield of product for the
transfer hydrogenation of benzophenone with selected [RuCl-
(p-cymene)(NN)][BPh4] complexes. The lines for the precursors
containing bpz*mArNO2 and bpzmArNO2 are very similar to
that of 2. (1) 8 (bpz*m); ([) 10 (bpz*mPh); (9) 2 (bpz*mAr-
OH); (b) 18 (bpz*mArNH2).
induction period was not observed for this reaction or,
alternatively, it was so short that it could not be detected,
except for the complexes [RuCl(p-cymene)(bpz*m)][X] (X =
BPh4, BF4), where it was clearly observed (Figure S6). In
order to minimize this induction period and also to observe
the effect of a pretreatment, several catalytic experiments
were carried out using derivatives that had been pretreated
Different experiments were run with short reaction times
in order to ascertain whether there is an induction period for
the catalytic transfer hydrogenation of benzophenone under
base-free conditions. As stated above, as a general rule an

Article
Organometallics, Vol. 28, No. 13, 2009 3831
Table 5. Transfer Hydrogenation of Benzophenone Using [RuCl(p-cymene)(NN)][BPh4] As the Precatalyst and Different Periods of
Pretreatment or with Different Changes in the Reaction Media
runa
complex(NN- ligand)
pretreatment (h)
yieldb (%)
runa
complex(NN- ligand)
product added(product/cat. ratio)
yieldc (%)
40a
41a
42a
43a
45a
46a
47a
2 (bpz*mArOH)
2
16
24
1.5
16
1
57
19
4
22
6
39a
48a
2 (bpz*mArOH)
2 (bpz*mArOH)d
Hg (300:1)
benzophenone (2000:1)
74
66
8 (bpz*m)
49a
50a
2 (bpz*mArOH)
14 (bpz*mArNO2)
p-cymene (20:1)
p-cymene (20:1)
54
32
10 (bpz*mPh)
44
15
24
a Reaction conditions: benzophenone/catalyst = 1000:1, 0.002 mmol of precatalyst, solvent: 2-propanol, 10 mL, no base. b Determined by 1H NMR
at t = 6 h. c Determined by 1H NMR at t = 24 h. d After 24 h of normal conditions, 2 mmol (1000 equiv) of ketone was added. Yield considering 4 mmol
of ketone calculated at t = 48 h.
with refluxing 2-propanol for a certain number of hours
(see Table 5, runs 40a-47a, and Figures S7-S9). It was
found that the induction period for the bpz*m derivative
disappeared (see Figure S9) after the pretreatment, but in
general the TOF and the final yield were inversely propor-
tional to the number of hours of pretreatment for all of the
precatalysts tested. This would indicate that although the
precatalyst is being activated, and therefore the induction
period;in cases where it existed;disappeared, the active
species are probably decomposing to some extent in the
refluxing alcohol.
It was verified that the catalysts are still active after 24 h of
reaction. In the case of derivative 2, a further 2 mmol of
benzophenone was added after 24 h of reaction (see Table 5,
run 48a and Figure S10). The activity was only slightly lower
than for the first cycle, especially in the initial moments. The
observed reduction in activity is reasonable considering that
nearly 2 mmol of hydrogenated alcohol are already present
in the reaction medium. However, the activity is higher than
when a pretreatment period of 24 h was used, a situation that
makes the decomposition reactions less likely to occur when
the substrate is in the medium.
One of the possible activation pathways could be decoor-
dination of the arene. In order to determine whether this
process occurs, other catalytic experiments were run in which
free p-cymene was added to the reaction medium (p-cymene:
complex = 20) (see Table 5, runs 49a-50a). Although a
reduction in the activity was observed, total inhibition of the
catalytic activity did not occur, a fact that led us to propose
that decoordination of the arene does not occur during the
catalytic process. In order to confirm that the p-cymene is not
lost in the process, treatment of complex 2 in an NMR tube in
refluxing 2-propanol-d8 during 2 h was carried out, and free
p-cymene was not detected. The same result was obtained
when the experiment was carried out in a Schlenk tube using
2-propanol, with the samples examined by GC. Neither free
p-cymene nor coordinated benzene were observed when 20
equiv of benzene was added in refluxing 2-propanol at 82 °C
during 6 h.
Considering that our systems do not have N-H groups,
and therefore the bifunctional metal-ligand mechanism8,38
is not applicable, the process probably involves the forma-
tion of a monohydride that initiates the catalytic cycle.15,39
Consequently the formation of hydrides in the absence of
added base in 2-propanol was explored.
The precatalysts were first treated under the same condi-
tions as in the catalytic reactions (dissolved in 2-propanol
and heated under reflux for different periods of time) without
the presence of substrate. After removing the solvent, the
products were analyzed by 1H NMR spectroscopy. Interest-
ingly, it was observed that the p-cymene systems are capable
of generating hydride species without the addition of either a
base or other additives. We consider that this is an important
finding that explains the activity of our derivatives. It is also
noteworthy that a broad resonance at low field, which can be
ascribed to an NH+ group of the ligand, is also observed
(free pyrazole is not present in the reaction medium). The
precursor [RuCl2(p-cymene)]2 also gave rise to the formation
of a hydride species that was identified as [Ru2(p-cym-
ene)2(μ-H)(μ-Cl)Cl2] (complex I, < 5% after 4 h).40 The
p-cymene precatalysts used in these studies were those con-
taining the ligands bpz*mArNO2 (14), bpz*mArOH (2), and
bpz*mPh (10). One to four resonances were observed with
ratios that changed with time (see Table S11 and Figures S12
and S13). In the case of complex 14 (ligand bpz*mArNO2),
the hydride resonances corresponding to I and to another
hydride species were observed. Considering the better results
obtained with the precatalyst containing this ligand, it is
envisaged that this second species that could be [RuH-
(p-cymene)(NN)]+ is mainly responsible for the catalytic
activity. For the derivatives with the other two ligands the
hydride resonance of I was not observed, but other hydride
signals were present. It was found that the more active
precatalysts gave rise to higher ratios of hydride species,
and these were formed at a faster rate. The precursor of the
benzene derivatives, [RuCl2(benzene)(CH3CN)], or the more
active benzene complexes did not give rise to hydride reso-
nances detectable by NMR spectroscopy. However, it can-
not be ruled out that such species are formed in a very low
ratio. In any case, the different behavior explains the lower
activity obtained with the benzene derivatives under base-
free conditions.
(38) (a) Takehara, J.; Hashiguchi, S.; Fujii, A.; Shin-ichi, I.; Ikariya,
T.; Noyori, R. Chem. Commun. 1996, 233. (b) Murata, K.; Okano, K.;
Miyagi, M.; Iwane, H.; Noyori, R.; Ikariya, T. Org. Lett. 1999, 1, 1119.
(c) Yamakawa, M.; Ito, H.; Noyori, R. J. Am. Chem. Soc. 2000, 122,
1466. (d) Hashiguchi, S.; Fujii, A.; Haack, K. J.; Matsumura, K.;
Ikariya, T.; Noyori, R. Angew. Chem., Int. Ed. Engl. 1997, 36, 288.
(39) (a) Gladiali, S.; Mestroni, G. Transit. Met. Org. Synth. 1998, 2,
97. (b) Matharu, D. S.; Morris, D. J.; Kawamoto, A. M.; Clarkson, G. J.;
In a similar way, some experiments were carried out in
order to ascertain whether the evolution of the hydride
species is influenced by the presence of substrate, with
acetophenone or benzophenone used in a large excess (sub-
strate/precatalyst = 100:1) and sufficient precatalyst [RuCl-
(p-cymene)(bpz*mArOH)][BPh4] (2) to be studied by NMR
Wills, M. Org. Lett. 2005, 7, 5489. (c) Morris, D. J.; Hayes, A. M.; Wills,
M. J. Org. Chem. 2006, 71, 7035. (d) Laxmi, Y. R. S.; Backvall, J.-E.
::
Chem. Commun. 2000, 611. (e) Petra, D. G. I.; Reek, J. N. H.;
Handgraaf, J.-W.; Meijer, E. J.; Dierkes, P.; Kamer, P. C. J.; Brussee,
J.; Schoemaker, H. E.; Van Leeuwen, P. W. N. M. Chem.Eur. J. 2000, 6,
::
ꢁ
2818. (f ) Pamies, O.; Backvall, J.-E. Chem.Eur. J. 2001, 7, 5052.
(40) Bennett, M. A.; Ennett, J. Organometallics 1984, 3, 1365.

ꢀ
Carrion et al.
3832 Organometallics, Vol. 28, No. 13, 2009
Table 6. Rate Constants Calculated for the Transfer Hydroge-
nation of Acetophenone Using [RuCl(p-cymene)(bpz*mArOH)]-
[BPh4] As the Precatalysta
ratiob
temperature (K)
K
1000:1c
1000:1
353
353
353
353
353
348
343
338
7.01 ꢀ 10-6
2.24 ꢀ 10-6
8.40 ꢀ 10-6
1.35 ꢀ 10-5
1.92 ꢀ 10-5
1.04 ꢀ 10-5
5.37 ꢀ 10-6
3.16 ꢀ 10-6
1000:4.4
1000:9.6
1000:18
1000:18
1000:18
1000:18
a Reaction conditions: 0.1 mmol of acetophenone in 0.5 mL of
2-propanol-d8, no base. b Ketone/precatalyst. c Reaction conditions:
2 mmol of acetophenone in 10 mL of 2-propanol, no base.
spectroscopy. The system was studied for reaction times of 2,
4, and 16 h, and in all cases the same hydride species were
observed as in the absence of substrate. At 2 and 4 h, there
were also other hydride species that were not observed
without substrate, but these were present at very low levels.
A similar experiment was carried out with the precatalyst
[RuCl(p-cymene)(bpz*mArNO2)][BPh4] (14) at 6 and 16 h,
with the expected signals obtained in each case.
Figure 5. Evolution with time of ln[acetophenone] for the
transfer hydrogenation using [RuCl(p-cymene)(bpz*mArOH)]-
[BPh4] (2) as the precatalyst.
Finally, kinetic studies for the catalytic hydrogenation of
acetophenone under base-free conditions with the precata-
lyst 2 were carried out. The goals were to obtain the reaction
order for the catalyst and to determine the possible existence
of an isotopic effect. The hydrogenation of acetophenone
was followed by 1H NMR spectroscopy in 0.5 mL of
2-propanol-d8 in an NMR tube. First, catalytic tests were
run at 80 °C at different precatalyst concentrations (sub-
strate/precatalyst ratio from 1000:1 to 1000:18, with 1 equiv
being 0.0001 mmol), and these yielded conversions between
3 and 27% at 4 h of reaction. The rate constant was
calculated from the slope for each concentration of catalyst
(see Table 6) from a plot of ln[acetophenone] versus reac-
tion time (Figure 5). Taking these values into account, and
representing ln(K) versus ln([Ru]) (Figure S14), a reaction
order of 0.8 was calculated. Considering that the catalyst
deactivation processes observed previously in the pretreat-
ment experiments may partially reduce the measured reac-
tion order, we consider that the reaction order for the
catalyst should be 1.14c
Figure 6. Eyring plot [ln(K/T) versus 1/T] for the transfer
hydrogenation of acetophenone when [RuCl(p-cymene)-
(bpz*mArOH)][BPh4] (2) is used as the precatalyst.
or some derivative of it, in the rate-determining step of the
catalytic mechanism.14c
The influence of the temperature was also assessed (65, 70,
75, and 80 °C) with a substrate/precatalyst ratio of 1000:18
(using 0.0018 mmol of 2). Once the rate constants had been
obtained for each temperature (Table 6), the activation
parameters of the process could be calculated from the
Eyring plot [ln(K/T) versus 1/T] (Figure 6). The plot is
reasonably linear (R2 = 0.98). The data obtained are ΔHq =
Mechanistic Proposal. While a detailed reaction mechan-
ism still remains to be established, a proposal will be put
forward that will take into account the information gathered
from the different experiments performed. First, the forma-
tion of hydrides from our p-cymene precursors in refluxing
2-propanol has been established, and a proposal to explain
this finding is reflected in Scheme 3. We propose an initial
partial decoordination of the NN ligand. It is necessary to
consider that a Ru-N bond rupture process has previously
been deduced from the isomer interconversion found in these
derivatives. In addition, the formation, albeit to a small
extent, of I40 in the treatment with 2-propanol of some
derivatives implies that even the total decoordination of
the NN ligand is possible. The induction period observed
for the derivative with the bpz*m ligand (without an aryl
susbstituent on the methynic carbon) may be due to a lower
level of steric hindrance, which does not facilitate this
process. The decoordination process would allow the inter-
action of a 2-propanol molecule with the Ru center (inter-
mediate A in Scheme 3), and an equilibrium, shifted to the
left, involving proton transfer from the alcohol to the free
19 ( 2 kcal mol-1, ΔSq = -26 ( 1 eu, and ΔGq298 = 27 (
3
2 kcal mol-1. The calculated ΔG2q98 value is consistent with
3
a hydridic transfer hydrogenation mechanism rather than a
direct hydrogen-transfer mechanism, which would present
lower values.41 The entropy data are in accordance with a
transition state where the variation of entropy is almost
nonexistent, suggesting that the rate-determining step is
intramolecular.14c
The rate constant in 2-propanol was also measured, and
this allowed a KH/KD ratio of 3.1 to be calculated. This
isotopic effect suggests the participation of 2-propanol,
(41) Cohen, R.; Graves, C. R.; Nguyen, S. T.; Martin, J. M. L.;
Ratner, M. A. J. Am. Chem. Soc. 2004, 126, 14796.

Article
Organometallics, Vol. 28, No. 13, 2009 3833
Scheme 3. Proposed Mechanism for the Hydride Formation in
the Catalytic Transfer Hydrogenation of Ketones under Base-
Free Conditions
the rate-determining step is the β-elimination from the iso-
propoxide group. The negative effect of an excess of p-cymene
may be due to competition with the substrate for the vacant
position of the ruthenium center.42 It is also possible that this
molecule gives rise to weak interactions with intermediates in
the catalytic cycle or with the substrate, and these could
compete with other interactions exhibited by the functional
groups of the aryl rings. In this sense, Baratta recently
proposed that weak interactions such as hydrogen bonds have
a clear effect on the catalytic process.14a,14c The positive effect
of the functional groups on the aryl rings could also be related
with the existence of weak interactions with the substrate.
Conclusions
In conclusion, we have developed a series of air-stable and
easily handled chloride derivatives that are active in the base-
free transfer hydrogenation process of ketones and alde-
hydes. The use of unstable hydrides as precursors is pre-
vented, and it is not necessary to introduce any additive to
promote the reaction. The absence of base allowed the
hydrogenation of base-sensitive substrates, and good results
were obtained with benzaldehyde. The derivatives are of the
type [RuCl(arene)(NN)][BPh4] (arene = benzene, p-cym-
ene), where NN are bis(pyrazol-1-yl)methane derivatives
with different substituents at the methynic carbon. The
derivatives with p-cymene give rise to better yields. The
activity is particularly good when a functional group is
present in the aromatic substituent in the benzyl unit of the
bis(pyrazol-1-yl)methane ligand. It is noteworthy that for
the p-cymene derivatives higher activity is obtained for the
reactions performed in the absence of base compared with
the similar processes in the presence of KOH. When the
p-cymene precatalysts are heated under reflux in 2-propanol,
the formation of hydride species was observed. A NH+
group was also detected. Kinetic studies indicate that the
reaction order is 1, and the activation parameters are con-
sistent with a hydridic transfer hydrogenation mechanism
for the catalytic reaction under base-free conditions. The
kinetic isotopic effect observed indicates that the 2-propanol,
or some derivative of it, participates in the rate-determining
step, which should be the β-elimination from the isoprop-
oxide group. A mechanism has been proposed that includes
the possible role as internal base played by the nitrogenated
ligands. This type of precatalyst could open new routes for
asymmetric hydrogenations as the presence of the base often
affects the enantioselectivity of the reduction.
pyrazolyl group can take place (formation of B). In fact, we
previously found a similar case of proton transfer from
methanol to an imidazolyl group partially decoordinated
from a Ru center.23 The observation of the NH+ group in the
NMR studies supports this proposal. In order to confirm
that the pyrazole group was a more basic center than the
ruthenium atom, a stoichiometric amount of HBF4 was
added to an NMR tube (acetone-d6) containing complex
[RuCl(p-cymene)(bpz*mPh)][BPh4]. Protonation of the pyr-
azole group was observed and not the formation of ruthe-
nium hydrides. A β-elimination from the isopropoxide group
and the elimination of acetone from B would generate the
ruthenium hydride C, from which D is obtained by the
elimination of HCl. In accordance with this proposal, HCl
is lost spontaneously in the formation of 4C that takes place
in an alcoholic medium (MeOH). The better results obtained
with the p-cymene derivatives in comparison with those for
the benzene compounds can be explained by considering that
the higher donating ability of the p-cymene group could
favor the C-H activation, the chloride elimination, and/or
the partial decoordination of the NN ligand.
The formation of the dimer I can be explained in terms of the
reaction between C and HCl according to Scheme 3. The
protonation of C with HCl can generate a dihydrogen complex
that will liberate H2 to yield E. The reaction between C and E;
with elimination of the [NNH]BPh4 species;can give rise to
the formation of I. As demonstrated previously, I can also be
obtained by treatment of [RuCl2(p-cymene)]2 under the cata-
lytic reaction conditions. However, the catalytic activity of this
dimer, and thus of I, is much lower than the ruthenium
complexes with the pyrazolyl ligands, a situation that is due
to the role of the internal base that the uncoordinated pyrazole
plays in the formation of the more active hydride species.
Concerning the catalytic cycle and considering the reaction
order of 1 for the catalysts that has been calculated, we
propose that the true catalyst is the species D and that the
catalytic cycle is that usually proposed for monohydride
species,4,15,39 with partial decoordination of the NN ligand
to accommodate the ketone molecule at the metallic center.
The different catalytic behavior observed for complexes with
distinct NN ligands points to the fact that the ligand is not
completely decoordinated from the Ru center. The kinetic
isotope effect found suggests that, as is usually proposed,14c
Acknowledgment. Financial support from the Junta de
Comunidades de Castilla-La Mancha-FEDER Funds
(PCI-08-0054) and the Ministerio de Ciencia e In-
ꢀ
novacion, MICINN, for a project (CTQ2008-0378) and
a Juan de la Cierva contract is acknowledged.
Supporting Information Available: CIF files and crystallo-
1
graphic data for 2 and 4C. H and 13C{1H} NMR tables for
complexes 1-6. Complete catalytic data including the evolution
with time for the different catalytic experiments. Eyring plot
(ln(K/T) versus (1/T)) for the reaction of transfer hydrogenation
of acetophenone with 2. This material is available free of charge
via the Internet at http://pubs.acs.org.
ꢀ
(42) Madrigal, C. A.; Garcꢀıa-Fernandez, A.; Gimeno, J.; Lastra, E. J.
Organomet. Chem. 2008, 693, 2535.