Transfer hydrogenation of ketones catalyzed by iridium-bulky phosphine complexes

Article abstract of DOI:10.1016/j.ica.2017.08.014

The complexes [Ir(COD)(PR₃)₂]PF₆ (R = PPh₃ (1); R = PBn₃ = tribenzylphosphine (2)), [Ir(COD)(PBn₃)(PAn₃)]PF₆ (3) (PAn₃ = Tri-orthoanisyl-phosphine) and cis-(P,P)-[IrH(COD)(PBn₃){η²-P,C-(C₆H₄CH₂)PBn₂}]PF₆ (4) are active in the transfer hydrogenation of ketones. However, complex (3) gives the best results in conversion toward the alcohol. Interestingly, commercial isopropanol was used as hydrogen source, without any drying treatment. In situ generated isopropoxide was used as base. An efficient conversion of a variety of ketones, aromatic or aliphatic, cyclic or linear, including molecules with conjugated or isolated C[dbnd]C moieties was achieved, thus reporting 12 examples of hydrogenated substrates. Ketones of higher steric hindrance could not be converted under the studied conditions. The experimental evidence indicates that the steric and electronic properties of the substrates are determinant in the observed conversions and performance of the system. For α,β-unsaturated ketones, preference toward the reduction of the C[dbnd]C bond was observed. However, the system shows chemoselectivity toward the carbonyl group in molecules which also bear an isolated C[dbnd]C moiety. With the results obtained, a pseudo first-order dependence of the reaction rate on the concentration of ketone was determined. Also, stoichiometric as well as in situ tests were performed to shed light into the reaction pathways possibly involved in the catalytic TH of ketones described herein (precursor 3, base and isopropyl alcohol as hydrogen source).

Full text of DOI:10.1016/j.ica.2017.08.014

Accepted Manuscript
Research paper
Transfer Hydrogenation of Ketones Catalyzed by Iridium-Bulky Phosphine
Complexes
Vanessa R. Landaeta, Abel D. Salazar-La Rosa, Rafael E. Rodríguez-Lugo
PII:
S0020-1693(17)30545-5
http://dx.doi.org/10.1016/j.ica.2017.08.014
ICA 17808
DOI:
Reference:
Inorganica Chimica Acta
To appear in:
Received Date:
Revised Date:
Accepted Date:
7 April 2017
3 August 2017
4 August 2017
Please cite this article as: V.R. Landaeta, A.D. Salazar-La Rosa, R.E. Rodríguez-Lugo, Transfer Hydrogenation of
Ketones Catalyzed by Iridium-Bulky Phosphine Complexes, Inorganica Chimica Acta (2017), doi: http://dx.doi.org/
10.1016/j.ica.2017.08.014
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.

Transfer Hydrogenation of Ketones Catalyzed by Iridium-Bulky Phosphine
Complexes
Vanessa R. Landaeta,a,* Abel D. Salazar-La Rosaa and Rafael E. Rodríguez-Lugo.a,b,*
a Departamento de Química, Universidad Simón Bolívar, Valle de Sartenejas, Baruta. Apartado 89000, Caracas
1080A, Venezuela.
b
Laboratorio de Química Bioinorgánica, Centro de Química. Instituto Venezolano de Investigaciones
Científicas (IVIC). Caracas 1020-A, Venezuela.
Corresponding authors: V. R. Landaeta and R. E. Rodríguez-Lugo. e-mail: vlandaeta@usb.ve (V. R. Landaeta);
rrodriguez@ivic.gob.ve, rerodriguez@usb.ve, (R. E. Rodríguez-Lugo).
This contribution is dedicated to Dr. Carlo Mealli in occasion of his 70th birthday.
Abstract
The complexes [Ir(COD)(PR3)2]PF6 (R = PPh3 (1); R = PBn3 = tribenzylphosphine (2)),
[Ir(COD)(PBn3)(PAn3)]PF6
(3)
(PAn3
=
Tri-orthoanisyl-phosphine)
and
cis-(P,P)-
[IrH(COD)(PBn3){η2-P,C-(C6H4CH2)PBn2}]PF6 (4) are active in the transfer hydrogenation of
ketones. However, complex (3) gives the best results in conversion toward the alcohol.
Interestingly, commercial isopropanol was used as hydrogen source, without any drying treatment.
In situ generated isopropoxide was used as base. An efficient conversion of a variety of ketones,
aromatic or aliphatic, cyclic or linear, including molecules with conjugated or isolated C=C moieties
was achieved, thus reporting 12 examples of hydrogenated substrates. Ketones of higher steric
hindrance could not be converted under the studied conditions. The experimental evidence
indicates that the steric and electronic properties of the substrates are determinant in the observed
conversions and performance of the system. For α,β-unsaturated ketones, preference toward the
reduction of the C=C bond was observed. However, the system shows chemoselectivity toward the
carbonyl group in molecules which also bear an isolated C=C moiety. With the results obtained, a
pseudo first-order dependence of the reaction rate on the concentration of ketone was determined.
Also, stoichiometric as well as in situ tests were performed to shed light into the reaction pathways
possibly involved in the catalytic TH of ketones described herein (precursor 3, base and isopropyl
alcohol as hydrogen source).
Keywords: Transfer hydrogenation, hydrogen source, Ir(I) complexes, bulky phosphines, ketones.

1. Introduction
Catalytic Transfer Hydrogenation (TH) has emerged as a simple, cheap and, mainly, safe way to
hydrogenate substrates because it avoids the use of high pressure vessels and compressed H2 or
other hazardous reagents (NaBH4, LiAlH4, etc). The hydrogen atoms transferred in the reaction
come from another molecule that usually acts as solvent (e.g. an alcohol), generally originating only
benign by-products, thus making TH an alternative to classic hydrogenation. For this reason, TH is
now widely employed for the reduction of unsaturated organic molecules [1]. In particular, ketone
hydrogenation is a reaction of interest because it has been used in the production of intermediary
alcohols, which are important building blocks in fine chemistry, agrochemistry, food and fragances,
materials, etc [2].
Even when examples of catalytic systems based on other metals have been reported [3],
ruthenium, rhodium and iridium precursors are still typically chosen for the transfer
hydrogenation of ketones. Many examples of either TH catalysis or tandem procedures including
TH have been reported [4, 5]. In particular, iridium-based systems have shown outstanding
activities in TH of ketones [6]. The most widely used precursors [7] bear phosphines or similar
ligands [8], other phosphorus-containing chelating ligands (P,N; P,P; P,N,P; P,P,P; etc.) [9], and N-
heterocyclic carbenes [10] which might also contain a second donor group such as a phosphine. In
most of these, the metal center is Ir(III). Examples of Ir(I)-based systems in transfer hydrogenation
reactions are less common [7].
In previous works, our group reported the use of iridium(I) complexes bearing bulky
phosphines as ligands (PBn3 =tribenzylphosphine, PAn3 = tris(ortho-anisyl)phosphine) in the
catalytic hydrogenation of imines and trans-cinnamaldehyde using molecular hydrogen [11, 12].
Herein we report the TH of ketones from isopropyl alcohol under basic conditions, catalyzed by
cationic Ir(I) complexes bearing the bulky phosphines mentioned above (see figure 1). Commercial
isopropanol was used, without drying treatment, which is interesting since for higher scale
applications the use of dry solvents represents an increase in operational costs. A complete catalytic
evaluation (reaction conditions, type and quantity of base, hydrogen source, temperature, scope of
substrates, etc) is presented. Also the kinetic dependence on the concentration of substrate was
determined, and stoichiometric as well as in situ tests were performed to gain some insight into the
reaction pathways involved in this transfer hydrogenation.

Figure 1. Iridium-bulky phosphine complexes used as catalyst precursors in this work.
2. Experimental
2.1.
Materials and Methods:
All reactions and manipulations were routinely performed under dry nitrogen or argon
atmosphere using standard Schlenk techniques. Complexes [Ir(COD)(PPh3)2]PF6 (1) [13],
[Ir(COD)(PBn3)2]PF6 (2) [14], [Ir(COD)(PBn3)(PAn3)]PF6 (3) [12] and cis-(P,P)-[IrH(COD)(PBn3){η2-
P,C-(C6H4CH2)PBn2}]PF6 (4) [14] were synthesized according to literature procedures. Isopropyl
alcohol (Sigma-Aldrich, 99.5%) was used as received without further purification. Other solvents
and deuterated solvents were used as received from the commercial source without further
purification. NMR samples were prepared taking 0.4 mL of the reaction mixture and using a DMSO-
d6 inner tube. Solution 1H, 13C{1H} and 31P{1H} NMR spectra were recorded either on Bruker Avance
300, 500 or 600 MHz spectrometers at Laboratorio Nacional de Resonancia Magnética Nuclear,
Instituto Venezolano de Investigaciones Científicas (IVIC). Peak positions are relative to
1
tetramethylsilane, for H and 13C{1H}, and phosphoric acid, for 31P{1H}. The chemical shifts (δ) are
measured according to IUPAC [15], expressed in parts per million (ppm) and were calibrated
against the residual solvent resonance (1H) or the deuterated solvent multiplet (13C). All NMR
spectra were recorded at room temperature (25 °C) unless otherwise stated. The multiplicity of the
signals is indicated as s, d, t, q, or m for singlets, doublets, triplets, quartets, or multiplets,
respectively. The abbreviation br. is given for broadened signals. Quaternary carbon atoms are
indicated as Cquat and aromatic units as CHar.
2.2.
Catalytic tests:
2.2.1. Catalyst screening:
The appropriate catalyst (1-4, 9 μmol, 0.5 mol%) was added to a Schlenk flask. Air was purged
by three vacuum/gas (N2) cycles. iPrOH (3.3 mL) was added with a syringe and H2 was bubbled into
the solution at atmospheric pressure for 1 minute. Then, acetophenone (1.85 mmol) and base from
a freshly prepared stock solution of KOH in iPrOH (90 µmol, 200 µL from stock solution 0.45 mol/L)
were added with a syringe. A reflux condenser was connected to the Schlenk and the system was

heated to 80 °C in an oil bath. After the specified reaction time (2 or 20 hours) the heating was
stopped and the conversion was determined by 1H NMR.
2.2.2. Optimization of reaction conditions:
The catalyst (3, 9 µmol, 0.5 mol%) was added to a Schlenk flask. Air was purged by three
vacuum/gas (N2) cycles. Then, acetophenone (1.85 mmol), the hydrogen source (3.3 mL), and the
base (90 µmol, 200 µL from a stock solution in the corresponding hydrogen source at concentration
of 0.45 mol/L) were added with a syringe. When needed, a reflux condenser was connected to the
Schlenk and the system was heated to 80 °C in an oil bath. After the specified reaction time (2 or 20
hours) the heating was stopped and the yield was determined by 1H NMR.
2.2.3. Screening of substrates:
A general procedure was applied depending on the physical state of the substrate, as follows:
a. For solid ketones:
The catalyst (3, 9 µmol) and the ketone (1.85 mmol) were added to a Schlenk flask. Air was
i
purged by three vacuum/gas (N2) cycles. Degassed PrOH (3.3 mL) was added, followed by base
i
from a freshly prepared stock solution of KOH in PrOH (90 µmol, 200 µL from stock solution 0.45
mol/L). A reflux condenser was connected to the Schlenk and the system was heated to 80 °C in an
oil bath. After the reaction time (24 hours) the heating was stopped and the yield was determined
by 1H NMR.
b. For liquid ketones:
The catalyst (3, 9 µmol) was added to a Schlenk flask. Air was purged by three vacuum/gas (N2)
i
cycles. Then, the ketone (1.85 mmol), PrOH (3.3 mL), and KOH (90 µmol, 200 µL from stock iPrOH
solution 0.45 mol/L) were added with a syringe. A reflux condenser was connected to the Schlenk
and the system was heated to 80 °C in an oil bath. After the reaction time (24 hours) the heating
was stopped and the yield was determined by 1H NMR.
2.3.
Assessment of catalyst’s stability:
The catalyst (3, 9 µmol) was added to a Schlenk flask. Air was purged by three vacuum/gas (N2)
i
cycles. Then, acetophenone (1.85 mmol), PrOH (3.3 mL), and KOH (90 µmol, 200 µL from stock
iPrOH solution 0.45 mol/L) were added with a syringe. A reflux condenser was connected to the
Schlenk and the system was heated to 80 °C in an oil bath. Every 24 hours, a sample from the

reaction media was withdrawn, placed on an NMR tube with a sealed inner tube filled with DMSO-
d6, and a 1H NMR control was taken. After the spectroscopic measurement, the sample was returned
to the Schlenk flask and a further load of acetophenone (1.85x(1-c*) mmol, where c* is the
i
normalized residual amount of unconverted ketone from the previous run) / PrOH (1.85 x(1-c*)
mmol) was added. In total, 3 additional loads of acetophenone were used.
2.4.
In situ tests, Stoichiometric Experiments and Kinetic Insights:
2.4.1. In situ monitoring of the reaction progress (acetophenone)
Under inert atmosphere, catalyst 3 (1.9 mg, 1.7 μmol), acetophenone (41.7 mg, 0.34 mmol),
i
degassed PrOH (610 μL) and a sealed inner tube filled with DMSO-d6 (ca. 0.1 mL) were placed in a
i
Young NMR tube. Then, KOH from a freshly prepared stock PrOH solution (17 µmol, 17 µL from
1
stock solution 1.0 mol/L) was added. The sample was heated to 80 °C and analyzed by H NMR
every hour during the first 8 hours, then at 22, 24 and 25 hours.
2.4.2. In situ study of the interaction between the components of the reaction in organic solvent
(acetophenone). Stoichiometric test.
Under inert atmosphere, catalyst 3 (1.9 mg, 1.7 μmol) was dissolved in CH2Cl2 in a Young NMR
tube containing a sealed inner tube filled with DMSO-d6 (ca. 0.1 mL). A control 31P{1H} NMR spectra
i
was recorded. Then, degassed PrOH (10 μL) was added, and the 31P{1H} NMR was taken. After 2
hours, another 31P{1H} NMR was recorded. Then, acetophenone (41.7 mg, 0.34 mmol) was added
and the sample was analyzed by 31P{1H} NMR. The mixture was heated at 60 °C for 2 hours, and a
31P{1H} NMR spectra was recorded. The mixture was heated overnight (16 hours) at 60 °C, and
analyzed by 31P{1H} NMR.
i
2.4.3. In situ study of the interaction between the components of the reaction in neat PrOH
(acetophenone)
i
Under inert atmosphere, catalyst 3 (1.9 mg, 1.7 μmol), degassed PrOH (610 μL) and a sealed
inner tube filled with DMSO-d6 (ca. 0.1 mL) were placed in a Young NMR tube. The 31P{1H} NMR
i
spectra of the suspension was recorded. Then, KOH from a freshly prepared stock PrOH solution
1
(17 µmol, 17 µL from stock solution 1.0 mol/L) was added. 31P{1H} and H (10 to -30 ppm) NMR
spectra of the reaction mixture were recorded immediately and after 1 hour of reaction. Then,
acetophenone (0.34 mmol) was added. 31P{1H} and 1H (10 to -30 ppm) NMR spectra of the reaction

1
mixture were recorded. The sample was heated to 60 °C and analyzed by 31P{1H} and H NMR after
1 hour. Upon heating at 60 °C, other 3 sets of spectra were taken, one every hour.
i
2.4.4 In situ study of the interaction between the components of the reaction in neat PrOH
(acetophenone and excess of 1,5-COD)
i
Under inert atmosphere, catalyst 3 (9 µmol), degassed PrOH (410 μL) and a sealed inner tube
filled with DMSO-d6 (ca. 0.1 mL) were placed in a Young NMR tube. The 31P{1H} NMR spectra of the
suspension was recorded. Then, KOH from a freshly prepared stock iPrOH solution (KOH (90 µmol,
i
1
200 µL from stock PrOH solution 0.45 mol/L) was added. 31P{1H} and H (10 to -30 ppm) NMR
spectra of the reaction mixture were recorded immediately and after 1 hour of reaction. Then,
acetophenone (1.85 mmol) was added. 31P{1H} and 1H (10 to -30 ppm) NMR spectra of the reaction
1
mixture were recorded. Then, 1,5-COD (10-fold excess, 90µmol) was added. 31P{1H} and H (10 to -
30 ppm) NMR spectra of the reaction mixture were recorded. The sample was heated to 60 °C and
analyzed by 31P{1H} and 1H NMR after 0.5 hour.
i
2.4.5. In situ study of the interaction between the components of the reaction in neat PrOH
(acetophenone and 2)
i
Under inert atmosphere, catalyst 2 (1.7 μmol), degassed PrOH (610 μL) and a sealed inner tube
filled with DMSO-d6 (ca. 0.1 mL) were placed in a Young NMR tube. The 31P{1H} NMR spectra of the
suspension was recorded. Then, KOH from a freshly prepared stock iPrOH solution (17 µmol, 17 µL
1
from stock solution 1.0 mol/L) was added. 31P{1H} and H (10 to -30 ppm) NMR spectra of the
reaction mixture were recorded immediately and after 1 hour of reaction. Then, acetophenone
1
(0.34 mmol) was added. 31P{1H} and H (10 to -30 ppm) NMR spectra of the reaction mixture were
1
recorded. The sample was heated to 60 °C and analyzed by 31P{1H} and H NMR after 1 hour. Upon
heating at 60 °C, other 3 sets of spectra were taken, one every hour.
2.4.6. Evaluation of the effect of an excess of PAn3 in the catalysis.
The catalyst (3, 9 µmol) and PAn3 (10-fold excess, 90µmol) were added to a Schlenk flask. Air
was purged by three vacuum/gas (N2) cycles. Then, acetophenone (1.85 mmol), iPrOH (3.3 mL), and
i
KOH (90 µmol, 200 µL from stock PrOH solution 0.45 mol/L) were added with a syringe. A reflux
condenser was connected to the Schlenk and the system was heated to 80 °C in an oil bath. After the
reaction time (2 or 20 hours) the heating was stopped and the yield was determined by 1H NMR.

2.4.7. Evaluation of the effect of an excess of 1,5-COD in the catalysis.
The catalyst (3, 9 µmol) and 1,5-COD (10-fold excess, 90µmol) were added to a Schlenk flask. Air
was purged by three vacuum/gas (N2) cycles. Then, acetophenone (1.85 mmol), iPrOH (3.3 mL), and
i
KOH (90 µmol, 200 µL from stock PrOH solution 0.45 mol/L) were added with a syringe. A reflux
condenser was connected to the Schlenk and the system was heated to 80 °C in an oil bath. After the
reaction time (20 hours) the heating was stopped and the yield was determined by 1H NMR.
3. Results and Discussion
3.1.
Catalytic tests:
The catalytic transfer hydrogenation of ketones was initially studied using acetophenone as
model substrate. The evaluation of complexes 1-4 in the hydrogenation of this substrate led to the
results summarized in table 1. Precursors 1 and 3 yielded the best results, with conversions toward
the corresponding alcohol (at 20hours) of 95 and 97% and TONs of 195 and 199 respectively
(corresponding to TOF = 47 h-1 and 53 h-1). Complexes 2 and 4 were less efficient than 1 and 3,
affording yields of alcohol of 88 and 81% in a 20 hours reaction. The latter are structurally related,
given the fact that 4 is the result of a C-H activation reaction on one of the benzyl groups of one of
the PBn3 ligands in 2, as previously described [14]. In this sense, the slightly lower yield of product
obtained when 2 was used as catalyst precursor, might be due to the parallel ortho-metallation
reaction, which lowers the efficiency of the transfer hydrogenation. The lower performance of
precursor 4, given the fact that the complex is coordinatively saturated, might be attributed to the
lack of a vacant site to coordinate the substrate. Given this fact, complex 4 would require previous
de-coordination of a phosphine ligand or displacement of the metallation reaction toward the Ir(I)
species [14] to perform as efficiently as 2. In light of the results obtained and considering that one
of the main purposes of this investigation was to study the effect of bulky phosphine ligands PBn3
and PAn3, complex 3 was selected as catalyst of choice for the rest of the catalytic evaluation, even
when its efficiency is only slightly better than that of the system based in 1.
Transfer hydrogenation reactions with iridium catalysts have been proposed to proceed via
three main mechanistic pathways [16], two of them involving an Ir-hydride intermediate and the
last one a direct transfer of hydrogen (Meerwein-Ponndorf-Verley type). Considering this fact, the
screening of catalysts (table 1) was carried out by performing an initial pretreatment with
molecular H2 (bubbling for 1min at atmospheric pressure) to promote the generation of hydridic
species product of oxidative addition. In this sense, once compound 3 was selected as catalyst

precursor, the relevance of this pretreatment with H2(g), as well as the rest of the conditions that
might influence the system, were studied (see table 2).
Table 1. Catalyst Screening
2 hour
20 hours
TOFd (h-1)
Complexa
TONc
conversionb conversionb
(2h)
47
1
2
46%
33%
52%
46%
95%
88%
97%
81%
195
181
199
167
34
3
53
4
47
a
Conditions: acetophenone (1.85 mmol), KOH (92 μmol), [Ir] (9
μmol, 0.5 mol%), iPrOH (3.5 mL), 80 °C, H2(g) bubbled at
atmospheric pressure for 1 minute. b Conversion determined by
1H NMR. c TON: Turn Over Number (moles of substrate / moles of
d
precursor). TOF: Turn Over Frequency (moles of substrate /
moles of precursor / time unit).
It was determined that the pretreatment with H2(g) was not necessary for this system, since the
results obtained after this pretreatment and without it are essentially identical (table 2, entries 1
and 2). This evidence suggests that the generation of a dihydride Ir(III) intermediate might not be
involved in a possible reaction pathway for the hydrogen transfer from isopropyl alcohol to ketones
using complex 3 as catalyst precursor. However, this observation would not exclude the possible
generation of hydridic (monohydridic) species resulting from β-elimination from a coordinated
ligand, e.g. an isopropoxide. In any case, considering that the previous H2 bubbling seems
unnecessary, the rest of the tests were carried out without H2(g) pretreatment. The influence of the
amount and nature of the base was also studied. If no base is added, no conversion is observed
(table 2, entry 3), thus confirming the need to use a base and, probably, shedding light into the
possible reaction pathway. When a strong inorganic base (KOH, 5 mol%) was used to generate in
situ the corresponding isopropoxide, the reaction proceeds with excellent conversion (96%) and
good efficiency (TON = 197, TOF = 47 h-1). However, if this amount of base is lowered to 1 mol%
(Cat:Base = 1:2, table 2, entry 4) the efficiency of the system diminishes considerably (47%
conversion, TON = 97). The use of a weaker inorganic base such as sodium bicarbonate (table 2,
entry 5) yielded only a 47% conversion. When an organic base (NEt3, table 2, entry 6) was used,
only traces of the product were identified after a 20 hours period. These tests confirm the need of a
strong inorganic base in sufficient amount (5 mol%) for the reaction to proceed efficiently. Several

examples of Ir-based TH systems also report the use of strong inorganic bases to ensure good
results [7, 10f].
Table 2. Screening of reaction conditions for the transfer hydrogenation of acetophenone using 3 as
catalyst precursor.
TOF (h-
2 hour
20 hours
1
Conditionsa
TON
)
conversionb conversionb
(2h)
53
47
-
1
2
H2 (Patm, 1min)
No H2 bubbling
No Base
52%
46%
-
97%
96%
199
197
-
3
No reaction
47%
4c
5c
6c
7c
8c
9c
Cat:KOH 1:2
NaHCO3
34%
-
97
97
-
35
5d
47%
NEt3
-
Traces
17%
-
Room Temp.
0.05 mol% 3
dry iPrOH
5%
15%
33%
35
596
187
5
29%
154
34
91%
iPrOH + 5%
water (v/v)
Blank run
10c
-
-
19%
9%
39
-
2d
11c
-
a
General conditions (unless otherwise stated): acetophenone (1.85
mmol), base (92 μmol), complex [Ir(COD)(PBn3)(PAn3)]PF6 (3) (9 μmol,
b
0.5 mol%), commercial iPrOH (3.5 mL), 80 °C. Conversion determined
by 1H NMR. c Without H2(g) pretreatment. d Calculated at 20 hours.
Attempts to lower the reaction temperature from 80 °C to room temperature (table 2, entry 7)
or to use a lower catalyst load (0.05 mol%, table 2, entry 8) led to poorer conversions to the alcohol.
The use of a lower catalyst load led, however, to higher values of TON and TOF (596 and 154h-1,
respectively). To assess for the possibility of using commercial isopropanol in the system, a
comparison between the model reaction (table 2, entry 2) and a catalytic run using dry solvent
(table 2, entry 9) was performed. The use of “wet” solvent from the bottle yields slightly better
results than when dry isopropanol was used. However, when a significant amount of water is
intentionally added to the reaction media (5% water, v/v, see table 2, entry 10) the efficiency of the
reaction decreases considerably, converting only a 19% of substrate. This indicates that the system
is tolerant to traces of water present in the commercial solvent, but an excess of water hampers the

catalytic reaction. Finally, the optimization of the reaction conditions also included a blank run
(table 2, entry 11), in which the catalytic nature of the system was confirmed.
Additionally, the reaction was monitored as a function of time to determine the optimum
reaction time. The screening was carried out on a Young NMR tube, in a closed system. As it can be
seen in figure 2, in the first 5 hours the reaction is faster. The reaction rate appears to decay
progressively after 6 hours. After 20 hours no significant changes in conversion were observed. In
this sense, it was decided to set 20 hours as the total reaction time for the rest of the screenings
performed. In contrast to the catalytic tests, in the study of the progress of the reaction it was
observed that the total conversion was slightly lower (88% vs. 96% observed in previous tests).
This slight decrease was attributed to the fact that the test was carried out in a closed system and
thus it became impossible to extract the product of dehydrogenation of the solvent, acetone. It is
known that in the transfer hydrogenation reaction using isopropanol as hydrogen source the
formed acetone can reversibly form back the alcohol [3]. In this sense, it has been reported [17]
that extracting the acetone from the reaction media is a key factor in terms of efficiency of the
reaction since it allows for the donor-acceptor equilibria in the transfer hydrogenation to shift
toward the products.
100
80
60
40
20
0
0
5
10 15
Reaction Time (h)
20
25
% Ketone
% Alcohol
Figure 2. Progress of the reaction (alcohol formation and ketone consumption) for the transfer
hydrogenation of acetophenone. Conditions: acetophenone (0.34 mmol), base (17 μmol), complex
[Ir(COD)(PBn3)(PAn3)]PF6 (3) (1.7 μmol, 0.5 mol%), iPrOH (0.63 mL), 80 °C.
Based on this results, the optimum conditions determined are: catalyst = complex 3; catalyst
load = 0.5 mol%; catalyst:base ratio = 1:10; base = KOH; temperature = 80 °C; reaction time = 20
hours. Under the optimized conditions, different hydrogen sources were also studied to determine

the most efficient donor molecule for the transfer hydrogenation of acetophenone with this system.
The results are shown in Table 3. Isopropyl alcohol (iPrOH), which was the solvent of choice for the
optimization of the reaction conditions, resulted in the best donor molecule using this system,
reflected in the higher conversion to alcohol and TON obtained (table 3, entry 1). When the reaction
was performed in ethyl alcohol, a lower conversion of acetophenone to the corresponding 1-
phenylethyl alcohol was obtained (52 %, TON = 107, table 3, entry 2). However, the analysis of the
mixture obtained revealed the formation of ethyl acetate, the product of dehydrogenative homo-
coupling (DHC) of ethyl alcohol, which has been previously described by several authors [18].
Clearly, a competitive reaction between the transfer hydrogenation of acetophenone and the
dehydrogenative coupling of ethanol is responsible for the lower conversions observed. This
indicates that the system using this donor molecule is not selective in the type of catalytic reaction
that occurs. Of particular interest was the choice of formic acid or formic acid/formate. It has been
proposed [3, 6] that these hydrogen sources are a suitable alternative in TH reactions to avoid the
reversible formation of the donor molecule after its dehydrogenation. For HCOOH, the
dehydrogenated product is CO2 which can be purged from the catalytic system. Unfortunately,
under acidic conditions complex 3 does not catalyze the TH reaction of acetophenone (table 3,
entry 3). When sodium formate was added as base in formic acid as media, only 10% of conversion
to the alcohol was achieved after a 20h reaction. Interestingly, all of the solvents were used directly
from the bottle without drying treatment, indicating that the system is tolerant to traces of water in
the reaction media, as stated before for iPrOH (table 2, entry 2 vs. entry 9).
Table 3. Screening of different hydrogen sources
Hydrogen
Sourcea
2 hours
20 hours
TONc
conversionb conversionb
1
2
3
iPrOH
46%
16%
-
96%
52%
197
107
-
EtOHc
HCOOHd
No Reaction
10%
4
HCOOH/HCOONae
9%
21
a
Conditions: acetophenone (1.85 mmol), KOH (92 μmol),
complex [Ir(COD)(PBn3)(PAn3)]PF6 (3) (9 μmol, 0.5 mol%),
b
Hydrogen Source (D = Donor molecule, 3.5mL), 80 °C.
c
Conversion determined by 1H NMR.
The product of
d
dehydrogenative coupling (DHC) is observed (ethyl acetate).
No base added. e Sodium Formate as base.

3.2.
Assessment of Catalyst’s stability:
As it was previously discussed, in an attempt to test the ability of the catalytic system to convert
higher quantities of ketone, the catalyst load was lowered to 0.05% mol (table 2, entry 8), obtaining
only 29% of alcohol. Given the fact that this approach was unsuccessful, it was decided to test the
tolerance and robustness of 3 when consecutive loads of substrate were added to the same amount
of catalyst. In a typical catalytic experiment, a catalyst load of 0.5 mol% was used. To assess for the
catalyst’s stability, the catalyst load was kept in 0.5 mol%, and 4 consecutive catalytic runs were
performed using the same catalyst used in the first run. Fresh loads of substrate (acetophenone)
were added after each period of 24 hours of reaction, taking into account the amount of non-
converted ketone for overall calculations. The total amount of solvent (iPrOH) was also kept fix (see
supporting information, figure S1).
Only a slight decrease in conversion was observed between the first and second runs (99% vs.
93% of alcohol). With the third amount of substrate a conversion of 85% was achieved, while with
the fourth substrate load a total of 75% of acetophenone was converted to its corresponding
alcohol. The total turnover number for this experiment is TON = 724, which is even higher than
when a catalyst load of 0.05 mol% was used. A natural consequence of the method used to perform
this experiment is the accumulation of products (benzyl alcohol and any remaining acetone)
through the different runs. In this sense, the concentration of catalyst, base and substrate changes
from one run to another. From the kinetic point of view this would explain the differences in
catalyst activity through the successive runs. Regardless of this fact, this experiment demonstrates
the robustness of the catalyst and its high activity as well, since it was possible to hydrogenate
three consecutive loads of acetophenone without a significant decay in conversion.
3.3.
Extension of the Scope of the Catalytic System:
To evaluate the scope of the catalytic system, a series of methyl ketones and one example of an
ethyl ketone were studied. The optimum conditions determined for the hydrogenation of
acetophenone, using isopropyl alcohol as hydrogen source, were used. The results are presented in
Table 4. To establish proper comparison conditions and to ensure a maximum conversion of the
substrates, these reactions were performed in a total reaction time of 24h.
The para-substitution of acetophenone has a great influence in the efficiency of the catalytic
reaction. Substituted acetophenones bearing either electron donating (table 4, entry 2) or electron
withdrawing (table 4, entry 3) groups were evaluated and, in both cases, the conversions were
lower than the one obtained for the unsubstituted substrate (table 4, entry 1). In the case of 4-

methoxyacetophenone, the substrate converted to the alcohol in moderate yield (69%). A
disfavoring effect of the methoxy group in para- position on the iridium-catalyzed TH of 4-
methoxyacetophenone has been previously reported for other systems [5]. Very likely, the electron-
donating –OMe reduces the electrophilicity of the carbon atom in the carbonyl group. On the other
hand, the electron-withdrawing NO2- group completely deactivated the substrate toward the
hydrogen transfer reaction (table 4, entry 3). It would be expected that electron-withdrawing
groups favored reduction of the ketone, as described by Crabtree for 4-CF3-PhCOMe [5]. However,
in the case of the NO2-substituted acetophenone, decomposition of the catalyst was observed upon
mixing of the reactants (the mixture turned immediately black). In other words, catalyst
decomposition took place faster than the TH reaction for this particular substrate.
Table 4. Screening of the substrate scope: Hydrogenation of methyl and ethyl ketones.
Conversion
Ketonea
Alcohol
(%)b
1
2
98
69
-
3
-
4
5
90
36
6
22
a
Conditions: ketone (1.85 mmol), KOH (92 μmol),
complex [Ir(COD)(PBn3)(PAn3)]PF6 (3) (9 μmol, 0.5
b
mol%), iPrOH (3.5 mL), 80 °C, 24 h.
determined by 1H NMR.
Conversion
On the other hand, 2-pentanone (table 4, entry 4) was converted to its corresponding alcohol in
good yield (90%), thus indicating that the system is efficient for aryl ketones as well as for alkyl

ketones. A dramatic drop in efficiency of the catalytic system was observed when 1-adamantyl-
methyl ketone was studied. Only 36% of alcohol was obtained for this substrate (table 4, entry 5),
probably due to a combination of steric hindrance and rigidity in the adamantyl substituent. The
rigidity of the cyclic adamantyl group might compromise the approach of the substrate to the active
site, which is presumably per se sterically hindered. Interestingly, changing the methyl substituent
by an ethyl (table 4 entries 1 vs. 6) markedly affects the conversion toward the alcohol, obtaining
only 22% of 1-phenyl-1-propanol. These facts indicate that the system might be sensitive to an
increase in steric hindrance in the substrate, due to the inherent bulkiness in catalyst precursor 3.
Additionally, the replacement of one of the alpha-protons in acetophenone, PhCOCH2(H), by a
methyl group in propiophenone, PhCOCH2(CH3), would make the carbonyl group less electrophilic
and consequently less prone to reduction.
In this sense, to determine the effects of the substrate steric hindrance on the performance
of the catalytic system, a series of cyclic or bulky ketones were evaluated, including substrates with
olefins as additional functional groups. The results are summarized in Table 5. Interestingly, the
transfer hydrogenation of cyclohexanone using complex 3 as catalyst precursor proceeds very
efficiently, resulting in quantitative conversion of the substrate to cyclohexyl alcohol (table 5, entry
1). Given this outcome, it resulted interesting to study adamantanone as well, which has a
structural analogy (cyclic 6-membered ketone) to cyclohexanone. However, the rigidity of the fused
cycles in adamantanone significantly affects the conversion, which drops to 35% (table 5, entry 2).
This is probably due to the difficult hybridization change at the carbon atom resulting from the
hydrogenation of the ketone, since the constrained ring might be hampering the Csp2 – Csp3
conversion. In the case of benzophenone, it was observed that substitution of a methyl group (table
4, entry 1) by a phenyl group (table 5, entry 3) does not significantly affect the catalytic system in
terms of conversion. This is in contrast with the behavior observed for the ethyl derivative, in
which a lower yield of the alcohol was obtained (table 4, entry 6). Such results underline the fact
that a carbonyl function polarized by an electron-withdrawing substituent (phenyl in this case) is
easier to reduce than a carbonyl moiety influenced by an electron-donating group (methyl in this
case). Another substrate explored in this study was fluorenone. Interestingly, regardless of the
rigidity of fluorenone, a good conversion of 9-fluorenyl alcohol was obtained (76%, table 5, entry
4). Probably, the planarity of fluorenone allows for a more suitable steric arrangement at the metal
than, for example, 1-adamantyl-methyl ketone or adamantanone, thus maintaining the efficiency of
the system. From the electronic point of view, as stated before, the carbonyl moiety in fluorenone is
activated toward the reduction reaction with respect to propiophenone.

Table 5. Screening of the substrate scope: Hydrogenation of cyclic and sterically hindered ketones
and diketones.
Conversion
Ketone
Alcohol
(%)
1
2
>99
35
95
3
4
76
94
5
6
>99% conversion
63% cyclohexanone
6% 2-cyclohexen-1-ol
31% cyclohexanol
>99% conversion
42% dihydrocarvone
58% dihydrocarveol
7
8
9
-
-
-
-
10
11
-
-
-
-
12
-
-
a
Conditions: ketone (1.85 mmol), KOH (92 μmol), complex
i
[Ir(COD)(PBn3)(PAn3)]PF6 (3) (9 μmol, 0.5 mol%), PrOH (3.5 mL), 80
°C, 24 h. b Conversion determined by 1H NMR.

Finally, a series of olefin-functionalized cyclohexanones were also studied (table 5, entries 5-7)
to evaluate the selectivity of the transfer hydrogenation toward the carbonyl moiety when another
functionality is present in the substrate. In general, the efficiency of the system is the same as in
cyclohexanone, since high conversions are obtained in every case (94% or higher conversions). The
slight decrease in conversion observed for dihydrocarvone with respect to cyclohexanone might be
attributed to the methyl substituent in ortho position to the carbonyl group, which might cause
steric hindrance. The isolated olefin moiety was not hydrogenated under these conditions, thus
obtaining the unsaturated alcohol dihydrocarveol selectively. However, the presence of conjugated
olefin moieties markedly affects the selectivity of the reaction, which proceeds mainly toward the
hydrogenation of the C=C bond. In this sense, in the hydrogenation of 2-cyclohexenone the main
product was the saturated ketone cyclohexanone, while the unsaturated alcohol was only obtained
in 6% conversion. Similarly, in the evaluation of carvone the products of hydrogenation of the
conjugated olefin as well as of both functional groups were observed, while the conjugated
unsaturated alcohol was not detected. Preference toward the olefin moiety in α,β-unsaturated
ketones has been also observed in other iridium-based TH systems [10b]. Analogously to the
hydrogenation of dihydrocarvone, the isolated C=C moiety remained intact after the reaction.
Attempts to further extend the scope of the system to other methyl ketones of natural
occurrence (such as pinacolone and 4-hydroxy-4-methyl-2-pentanone; table 5 entries 8 and 9,
respectively), highly sterically hindered aryl ketones (such as 2,2,2-triphenylacetophenone; table 5,
entry 10) or diketones (such as 2,4-pentanodione or dimedone; table 5 entries 11 and 12,
respectively) were unsuccessful. These substrates were selected due to their structural analogy to
other substrates studied in this work. For example, it resulted interesting to study the
hydrogenation of pinacolone (tert-butyl-methyl ketone) since its steric environment resembles that
of 1-adamantyl-methyl ketone (table 4, entry 5) or 4-hydroxy-4-methyl-2-pentanone to determine
the electronic influence of a methyl and hydroxyl group in comparison to 2-pentanone (table 4,
entry 4). However, in these cases, the results obtained were different than with their analogues. In
particular, in the transfer hydrogenation of pinacolone, previous reports indicate that there has
been the need to diminish steric hindrance in the catalyst, for example by using smaller ligands, for
an efficient catalysis to occur [19]. Given the fact that catalyst 3 bears two bulky phosphine ligands,
it is possible that their steric hindrance hampers the catalytic conversion of pinacolone and
analogous substrates.

3.4.
In situ tests, Stoichiometric Experiments and Kinetic Insights:
A desirable key aspect in the studied TH reaction would be to determine the nature of the active
species in the reaction. Ir-hydride complexes have been indicated as active species in two of the
mechanistic proposals discussed in the literature [7, 10f, 16, 20]. It has been already discussed that,
essentially, there is no difference in performing the catalytic reaction with a H2 pretreatment
(generation of metallic dihydrides [21], see table 2, entries 1 and 2) or without it. This could
indicate that Ir(III)-dihydride species might not be participating in the reaction. However, Ir(I)
monohydridic complexes could also result from a β-hydrogen elimination from, e.g., a coordinated
isopropoxide ligand, and their possible role in this TH cannot be excluded. This would require the
presence of a base to generate the isopropoxide ligand in situ.
In this sense, in situ tests were performed at the NMR scale to investigate the role of the base and
to assess on the possible formation of hydridic species that could be proposed as active species. In
an initial experiment, complex 3 was treated with iPrOH in an organic solvent (CH2Cl2) to determine
possible interactions between the catalyst and the solvent. No reaction was observed after 2 hours
on the 31P{1H} NMR spectra as the signals of the coordinated phosphine ligands remained
unchanged (δ = 13.6 and -10.8). After this period, acetophenone was added to the reaction mixture
and, again, no changes were observed. The reaction mixture was left overnight at room
i
temperature, and it was noticed that the complex did not react either with PrOH or the ketone.
Upon additional heating at 60 °C for two hours, the resonances of 3 persisted. No base was added in
this experiment, which suggests then that 3 does not react with the alcohol or ketone in absence of
base, thus confirming the result obtained on the catalytic run in absence of base (table 2, entry 3).
i
In another experiment, complex 3 was suspended in neat PrOH. The 31P{1H} NMR spectrum
reveals, consistently, no reaction with the alcohol. Upon addition of KOH, the color of the complex
immediately changes (from deep red to yellow) and the suspended red solid dissolves. The 31P{1H}
NMR spectrum shows complete conversion of 3 into two new species (δ = 5.4 and 12.7) and free
-
PAn3 (δ = -38.4). The PF6 resonance disappears, which could indicate that the observed species,
-
unlike 3, are not cationic and that PF6 anion has been removed as KPF6 (s). No hydrides were
1
detected by H NMR (10 to -30 ppm). After this, acetophenone was added. The resonances at 5.4
and 12.7 ppm in the 31P{1H} NMR spectrum progressively disappear leading to a new signal at 10.2
ppm, which accumulates in time. Again, no hydrides were detected by 1H NMR (10 to -30 ppm). The
fact that hydridic species were not observed does not exclude its (their) possible involvement in
this reaction, since it (they) could not be accumulating in sufficient amount to be detectable. Also,

such species could be participating in reaction (s) fast enough so it (they) cannot be seen in the
NMR timescale.
Another important feature would be to determine whether the activation of the precursor 3 for
this transfer hydrogenation occurs upon dissociation of one of the coordinated ligands. In the
classic hydrogenation of C=X (X = C, N, O) bonds catalyzed by cationic (COD)Ir-bis(phosphine)
complexes, it has been proposed that the 1,5-COD ligand is hydrogenated and, thus, it dissociates
from the metal coordination sphere producing cyclooctane [11, 12, 14, 22]. However, in those cases,
an Ir-dihydride active species has been proposed, and the evidence obtained in this work seems to
indicate that these types of species do not participate in this hydrogen transfer system.
Additionally, the spectroscopic evidence (1H NMR) obtained does not suggest decoordination of the
1,5-COD ligand.
According to the in situ test described above, upon addition of the base to an iPrOH suspension of
3, not only the complex dissolves completely, but free PAn3 is observed in the 31P{1H} NMR
spectrum. This would indicate that, very likely, one phosphine (particularly PAn3) dissociates to
generate a vacancy for the coordination of the alkoxide ligand or the substrate. However, this
observation does not, a priori, exclude the possible dissociation of another type of ligand, such as
1,5-COD, or the formation of another type of intermediate. In this sense, to investigate on these
possibilities, individual experiments in presence of either an excess of phosphine ligand or an
excess of 1,5-COD were performed.
A catalytic run in presence of an excess of phosphine (PAn3, catalyst 3:phosphine ratio = 1:10)
was carried out. Under these conditions, the conversion of acetophenone to the alcohol was
significantly affected, obtaining only 32% of alcohol. The inhibition of the reaction might indicate
that the dissociation of a phosphine ligand could be involved in the reaction pathway for this
transfer hydrogenation. Nevertheless, the possibility that the excess of phosphine could hamper
substrate/iPrO- coordination should not be neglected. In any case, considering the results of the
previously described in situ reaction along with the catalysis in presence of an excess of PAn3, it
seems more likely that the decreased conversion in this catalytic test is the result of a displacement
of the dissociation equilibrium for the phosphine ligand toward the original species.
Additionally, a catalytic test was performed in presence of an excess of 1,5-COD. In this
experiment, upon addition of an excess of 1,5-COD (catalyst 3:COD ratio = 1:10) under the
optimized conditions of the model reaction (transfer hydrogenation of acetophenone), the
conversion of the substrate was 97% (20 hours reaction). This result is essentially the same as the
one obtained in the model reaction (97% in presence of an excess of 1,5-COD vs. 96% in the model

reaction, difference within the experimental error). This could indicate that an excess of 1,5-COD
has no effect on the catalysis. However, with the intention of further excluding a possible role of an
excess of 1,5-COD in the TH of acetophenone, an in situ test on the NMR scale (1H, 31P{1H}) was
performed. In the in situ test, first, complex 3 was suspended in iPrOH. The signals of complex 3 are
1
barely detectable in the 31P{1H} spectra since it is in suspension. In the H NMR spectrum in broad
spectral window (10 to -30ppm) no hydrides were detected. Base (KOH/iPrOH) was added, upon
which the solution changed color from red to yellow, and the red solid dissolved. The 31P{1H}
1
spectrum showed two species (δ = 5.4 and 12.7). No hydrides were detected in the H NMR
spectrum. First acetophenone and then 1,5-COD (catalyst 3:COD ratio = 1:10)were added to the
reaction mixture. The sample was kept at room temperature for two hours and, again, the
resonances at δ = 5.4 and 12.7 persisted, and free PAn3 was observed (δ = -38.4). Then, the mixture
was heated at 60 °C for 0.5 h, yielding a new signal in δ = 10.2 along with the signal attributed to
OPBn3 and free PAn3. No hydrides were detected. The signal in δ = 10.2 has also been identified in
the model in situ test, and accumulates in time. The behavior observed in this in situ test is the same
as the in situ test and monitoring of the model transfer hydrogenation, which suggests that 1,5-COD
is not influencing the reaction, and confirming the results obtained in the catalytic run. This
suggests that it is not likely that a dissociation of 1,5-COD could take place in the transfer
hydrogenation of ketones using catalyst 3, since an excess of 1,5-COD has no influence on the
reaction.
In order to elaborate on the possible identity of the species at 5.4 and 12.7 ppm observed in the
i
31P{1H} NMR spectrum, an additional in situ NMR test in neat PrOH was performed. In such
experiment catalyst precursor 2, which contains exclusively PBn3, was used under identical
conditions of the corresponding test with 3. The common resonances in these two in situ reactions
are the one attributed to OPBn3 and the one at 5.4 ppm. Moreover, in the experiment with 3, when
the relative amount of free PAn3 decreases, the relative amount of the resonance at 12.7 increases.
These two facts might suggest that the species at 5.4 ppm contains solely PBn3, while the compound
at 12.7 ppm bears only PAn3 as P-based ligand. Since 1,5-COD is not fully dissociated (hemilability is
not ruled out) and given that the complexes must be neutral, it could be proposed that species
“[Ir(COD)(iOPr)(L)] (L = PBn3 or PAn3)” could be forming . With the available data, such proposition
is merely speculative and no additional discussion regarding the identity of the resonance at 10.2
ppm can be performed. The elucidation of these compounds (δ = 5.4, 10.2 and 12.7) will be matter
of further studies.

In order to investigate on the role of the P-ligands, two additional experiments were performed
using the phosphine-free complex [Ir(COD)(μ-Cl)]2 (5) as catalyst precursor. The optimum
conditions of the model reaction were reproduced for this precursor (1.85 mmol acetophenone, 92
i
μmol KOH, 4.5x2 μmol Ir, 3.5 mL PrOH, 80 °C, 20 h). In the first test, [Ir(COD)(μ-Cl)]2 (5) was used
as catalyst precursor while in the second experiment 5 in presence of a ten-fold excess of 1,5-COD
was chosen. The dimer reduced 12% and 56% acetophenone after 2 and 20 h of reaction
respectively. By adding a ten-fold excess of 1,5-COD, the reaction yielded 12% and 79%
acetophenone after 2 and 20 h of reaction, respectively. Undoubtedly, in this particular system an
excess of 1,5-COD has an significant effect on the catalytic activity upon 20 h of reaction. Regardless
of that, the catalytic performance of 3 is significantly higher than that of [Ir(COD)(μ-Cl)]2 (5) or
[Ir(COD)(μ-Cl)]2/1,5-COD: 96% vs. 56% or 79% (20 h) respectively. The other Ir(I) precursors used
for this study, 1 and 2, are more efficient than the 5/1,5-COD system as well. In this sense, the
experiments using phosphine-free precursors underline the importance of the presence of the
phosphorus-based ligands PBn3 and PAn3 in the Ir(I) catalysts for the transfer hydrogenation
reaction described herein.
It was also possible to obtain kinetic information for the transfer hydrogenation of
acetophenone catalyzed by 3. From the monitoring of the progress of the reaction (described
before, see figure 2) it was possible to obtain data for the consumption of ketone as a function of
time for the first 8 hours of catalytic reaction. A plot of the Ln([ketone]/[ketone]0) vs. time (see
supporting information, figure S3) returns a straight line, which reveals a pseudo first-order
dependence of the rate of reaction on the concentration of ketone. An apparent constant kapp
=
0.162 h-1 was obtained from the slope of the straight line, and a rate law r = kapp[ketone] can be
proposed at a constant catalyst concentration, since kapp = kcat[catalyst]a. Given the fact that the
hydrogen donor is the solvent of the reaction, its concentration is higher than needed at any time
and, thus, it can be considered constant as well.
4. Conclusions
The transfer hydrogenation of ketones was accomplished, using [Ir(COD)(PBn3)(PAn3)]PF6 (3)
as catalyst precursor and isopropyl alcohol as hydrogen source. The system converts a variety of
methyl ketones and cyclic ketones. 12 examples of ketones were hydrogenated to their
corresponding alcohols, in general with good to excellent yields (69 to >99% in 9 cases). However,
substrates with higher steric hindrance could not be converted to the corresponding alcohols. In

the case of α,β-unsaturated ketones, selectivity toward the reduction of the C=C bond was observed.
Nevertheless, the system shows chemoselectivity toward the reduction of the carbonyl group in
molecules which also bear an isolated C=C moiety. The use of a strong base is necessary for the
activation of the catalyst. The loss of a phosphine ligand during the catalysis was demonstrated. It
was also proven that an excess of 1,5-COD ligand has no effect on the reaction. A pseudo first-order
dependence of the rate of reaction on ketone concentration was determined. Further studies are
being carried out to extend the scope of the system to the hydrogenation of aldehydes and gain
more insights on the possible reaction pathway for this system when [Ir(COD)(PBn3)(PAn3)]PF6 (3)
is used as catalyst precursor.
5. Supplementary Material
Experimental details, NMR spectroscopic data of the products of the catalytic reactions, the plots
of in situ NMR experiments, and NMR spectra of the crudes of reaction are available as
supplementary information.
Acknowledgements
The authors are grateful to Decanato de Investigación y Desarrollo at Universidad Simón Bolívar
(project S1-IC DID-CB-010-15) and Instituto Venezolano de Investigaciones Científicas (project
1265) for promoting this scientific activity. Special thanks are given to Lic. Daniela Briceño for her
assistance to perform some in situ experiments at NMR scale.
References
[1] a) S. Lee, G. Nikonov, ChemCatChem 7 (2015) 107-113; b) S. H. Kim, S. H. Hong, ACS Catal. 4 (2014) 3630-
3636; c) R. V. Jagadeesh, K. Natte, H. Junge, M. Beller, ACS Catal. 5 (2015) 1526-1529; d) T. Connolly, Z.
Wang, M. A. Walker, I. M. McDonald, K. M. Peese, Org. Lett. 16 (2014) 4444-4447; e) S. Mazza, R.
Scopolliti, X. Hu, Organometallics 34 (2015) 1538-1545; f) M. Wu, T. Cheng, M. Ji, G. Liu, J. Org. Chem. 80
(2015) 3708-3713.
[2] N. Andrushko, V. Andrushko, in: N. Andrushko, V. Andrushko (Eds.), Stereoselective Synthesis of Drugs
and Natural Products, John Wiley & Sons, Inc. New Jersey, 2013, pp. 909-960.
[3] K. Z. Demmans, O. W. K. Ko, R. H. Morris, RSC Adv. 6 (2016) 88580–88587.
[4] E. M. Titova, S. M. W. Rahaman, E. S. Shubina, R. Poli, N. V. Belkova, E. Manoury, J. Mol. Catal. A: Chem. 426
(2017) 376–380
[5] J. Campos, L. S. Sharninghausen, M. G. Manas, R. H. Crabtree, Inorg. Chem. 54 (2015) 5079−5084.
[6] For examples of TH reactions using iridium catalysts bearing ligands different than phosphines or NHCs,
see: a) C. Tian, L. Gong, E. Meggers, Chem. Commun. 52 (2016) 4207- 4210; b) D. Talwar, X. Wu, O. Saidi,
N. Poyatos Salguero, J. Xiao, Chem. Eur. J. 20 (2014) 12835 – 12842; c) O. Prakash, K. N. Sharma, H. Joshi,
P. L. Gupta, A. K. Singh, Organometallics 33 (2014) 2535−2543; d) A. Robertson, T. Matsumoto, S. Ogo,
Dalton Trans. 40 (2011) 10304–10310.

[7] R. Malacea, R. Poli, E. Manoury, Coord. Chem. Rev. 254 (2010) 729–752.
[8] For some examples of transfer hydrogenation reactions with iridium-phosphine complexes, see: a) N.
Meriç, M. Aydemir, J. Organomet. Chem. 819 (2016) 120-128; b) S. A. Popoola, E. A. Jaseer, A. A. Al-Saadi,
V. Polo, M. A. Casado, L. A. Oro, Inorg. Chim. Acta 436 (2015) 146–151; c) A. Guerriero, M. Erlandsson, A.
Ienco, D. A. Krogstad, M. Peruzzini, G. Reginato, L. Gonsalvi, Organometallics 30 (2011) 1874–1884.
[9] a) S. Werkmeister, J. Neumann, K. Junge, M. Beller, Chem. Eur. J. 21 (2015) 12226 – 12250; b) X. Chen, W.
Jia, R. Guo, T. W. Graham, M. A. Gullons, K. Abdur-Rashid Dalton Trans. (2009) 1407–1410.
[10] For some recent examples of Iridium-based TH systems bearing NHCs as ligands, see: a) A. Volpe, S.
Baldino, C. Tubaro, W. Baratta, M. Basato, C. Graiff, Eur. J. Inorg. Chem. (2016) 247–251; b) J. L. Gomez-
Lopez, D. Chávez, M. Parra-Hake, A. T. Royappa, A. L. Rheingold, D. B. Grotjahn, V. Miranda-Soto,
Organometallics 35 (2016) 3148−3153. c) D. Wang, D. Astruc, Chem Rev. 115 (2015) 6621-6686; d) K.
Yoshida, T. Kamimura, H. Kuwabara, A. Yanagisawa, Chem. Commun. 51 (2015) 15442- 15445; e) N.
García, E. A. Jaseer, J. Munarriz, P. J. Sanz Miguel, V. Polo, M. Iglesias, L. A. Oro, Eur. J. Inorg. Chem. (2015)
4388–4395; f) J. Ito, H. Nishiyama, Tetrahedron Lett. 55 (2014) 3133-3146; g) A. Azua, J. A. Mata, E.
Peris, F. Lamaty, J. Martinez, E. Colacino, Organometallics 31 (2012) 3911−3919.
[11] V. R. Landaeta, B. K. Muñoz, M. Peruzzini, V. Herrera, C. Bianchini, R. A. Sánchez-Delgado, Organometallics
25 (2006) 403-409.
[12] V. R. Landaeta, F. López-Linares, R. A. Sánchez-Delgado, C. Bianchini, F. Zanobini, M. Peruzzini, J. Mol.
Catal. A: Chem. 301 (2009) 1-10.
[13] L. Haines, E. Singleton, Dalton Trans. (1972) 1891-1896
[14] V. R. Landaeta, M. Peruzzini, V. Herrera, C. Bianchini, R. A. Sánchez-Delgado, A. Goeta, F. Zanobini, J.
Organomet. Chem. 691 (2006) 1039-1050.
[15] R. K. Harris, E. D. Becker, S. M. C. de Menezes, R. Goodfellow, P. Granger, Pure and Applied Chemistry 73
(2001) 1795–1818; Solid State Nucl. Magn. Reson. 22 (2002) 458–483.
[16] a) O. Saidi, J. Williams, Top. Organomet. Chem. 34 (2011) 77-106; b) O. Pàmies, J.-E. Bäckvall, Chem. Eur.
J. 7 (2001) 5052-5058.
[17] Z. Clarke, P. Maragh, T. Dasgupta, D. Gusev, A. Lough, K. Abdur-Rashid, Organometallics 25 (2006) 4113-
4117.
[18] For some examples of homogeneous catalytic dehydrogenative coupling of ethanol to yield ethyl acetate,
see: a) J. Cheng, M. Zhu, C. Wang, J. Li, X. Jiang, Y. Wei, W. Tang, D. Xue, J. Xiao, Chem. Sci. 7 (2016) 4428-
4434; b) D. Spasyuk, D. G. Gusev, Organometallics 31 (2012) 5239–5242; c) M. Nielsen, H. Junge, A.
Kammer, M. Beller, Angew. Chem. Int. Ed. 51 (2012) 5711–5713; d) T. Zweifel, J.-V. Naubron, T. Büttner,
T. Ott, H. Grützmacher, Angew. Chem. Int. Ed. 47 (2008) 3245 –3249.
[19] T. Ohkuma, C. Sandoval, R. Srinivasan, Q. Lin, Y. Wei, K. Muñiz, R. Noyori, J. Am. Chem. Soc. 127 (2005)
8288-8289.
[20] For some studies regarding mechanistic proposals for transfer hydrogenation reactions catalyzed by
iridium complexes, see: a) J-W. Handgraaf, J. N. H. Reek, E. Jan Meijer, Organometallics 22 (2003) 3150-
3157, b) S. Bi, Q. Xie, X. Zhao, Y. Zhao, X. Kong, J. Organomet. Chem. 693 (2008) 633-638.
[21] R. H. Crabtree, P. C. Demou, D. Eden, J. M. Mihelcic, C. A. Parnell, J. M. Quirk, G. E. Morris, J. Am. Chem. Soc.
104 (1982) 6994–7001.
[22] V. Herrera, B. Muñoz, V. R. Landaeta, N. Canudas, J. Mol. Catal. A: Chem. 174 (2001) 141-149.