A Water/Toluene Biphasic Medium Improves Yields and Deuterium Incorporation into Alcohols in the Transfer Hydrogenation of Aldehydes

Article abstract of DOI:10.1002/ejic.202100022

Deuterium labeling is an interesting process that leads to compounds of use in different fields. We describe the transfer hydrogenation of aldehydes and the selective C¹ deuteration of the obtained alcohols in D₂O, as the only deuterium source. Different aromatic, alkylic and α,β-unsaturated aldehydes were reduced in the presence of [RuCl(p-cymene)(dmbpy)]BF₄, (dmbpy=4,4′-dimethyl-2,2′-bipyridine) as the pre-catalyst and HCO₂Na/HCO₂H as the hydrogen source. Moreover, furfural and glucose, were selectively reduced to the valuable alcohols, furfuryl alcohol and sorbitol. The processes were carried out in neat water or in a biphasic water/toluene system. The biphasic system allowed easy recycling, higher yields, and higher selective D incorporation (using D₂O/toluene). The deuteration took place due to an efficient effective M–H/D⁺ exchange from D₂O that allows the inversion of polarity of D⁺ (umpolung). DFT calculations that explain the catalytic behavior in water are also included.

Full text of DOI:10.1002/ejic.202100022

Full Papers
doi.org/10.1002/ejic.202100022
A Water/Toluene Biphasic Medium Improves Yields and
Deuterium Incorporation into Alcohols in the Transfer
Hydrogenation of Aldehydes
Margarita Ruiz-Castañeda,[a] Lucía Santos,[a] Blanca R. Manzano,[a] Gustavo Espino,*[b] and
Félix A. Jalón*[a]
Deuterium labeling is an interesting process that leads to
compounds of use in different fields. We describe the transfer
hydrogenation of aldehydes and the selective C1 deuteration of
the obtained alcohols in D2O, as the only deuterium source.
Different aromatic, alkylic and α,β-unsaturated aldehydes were
reduced in the presence of [RuCl(p-cymene)(dmbpy)]BF4,
(dmbpy=4,4’-dimethyl-2,2’-bipyridine) as the pre-catalyst and
HCO2Na/HCO2H as the hydrogen source. Moreover, furfural and
glucose, were selectively reduced to the valuable alcohols,
furfuryl alcohol and sorbitol. The processes were carried out in
neat water or in a biphasic water/toluene system. The biphasic
system allowed easy recycling, higher yields, and higher
selective D incorporation (using D2O/toluene). The deuteration
took place due to an efficient effective M–H/D+ exchange from
D2O that allows the inversion of polarity of D+ (umpolung). DFT
calculations that explain the catalytic behavior in water are also
included.
Introduction
deuterated drugs[12] and, in fact, in 2017 the U.S. Food and Drug
Administration (FDA) approved the first deuterated drug.[12]
Alcohols are considered to be important species since they
can be converted into valuable chemicals through a multitude
of chemical or biochemical reactions. More specifically, deute-
rium-labeled alcohols are helpful tools for the study of some
reaction mechanisms or biological transformations.[2] Some such
alcohols play pivotal roles in revealing the mechanism of cell
recognition and receptor functions.[13] Thus, the design of a
regioselective and environmentally friendly method for alcohol
deuteration is of great interest.[14–17]
Deuterium labeling permits the addition of an extra mass unit
into organic molecules without changing its chemical structure,
or physical properties, and this constitutes a unique tool to
identify and understand biological and chemical processes.[1,2]
Labeling can also be extended to tritium and this allows the
incorporation of a radioactive tag that allows multiple applica-
tions, especially in molecules and processes of biological
interest.[1,3] Deuterated compounds have commonly been used
in mechanistic,[4,5] spectroscopic, and tracer studies but the last
decade has witnessed growing interest in stable-isotope-
labeled compounds. One reason for this is the large develop-
ment of high-performance mass spectrometry[6] and the increas-
ing usage of chemical analysis based on liquid chromatogra-
phy-mass spectrometry (LCÀ MS).[7,8] In medicinal chemistry it
has been proposed that specific deuterium labeling of drugs
may result in potential beneficial properties[9,10] and it has been
found that the pharmacokinetic properties of drugs may be
altered by kinetic deuterium isotope effects.[11] Significant
clinical progress has been made recently regarding the use of
Alcohols labeled with deuterium at different positions can
be obtained by multi-step organic synthesis from deuterium-
labeled synthons.[18–21] A more time-efficient approach involves
metal-catalyzed direct H/D exchange between the target
alcohol itself as a substrate and an appropriate deuterium
source. Deuterium oxide (D2O) is the ideal D source for the
preparation of this type of compounds owing to both its low
cost and toxicity. Nevertheless, these processes suffer from
several drawbacks. For example, heterogeneous systems such
as Raney Ni,[22–24] Pd/C[25] and Pt/C,[26,27] involve harsh conditions,
moderate deuterium efficiency and low selectivity. Although
higher selectivity[16] and milder conditions[17] have been
achieved with the use of Ru/C, all of these M/C systems require
the presence of an H2 atmosphere. Different homogeneous
[a] Dr. M. Ruiz-Castañeda, Dr. L. Santos, Dr. B. R. Manzano, Dr. F. A. Jalón
Facultad de Ciencias y Tecnologías Químicas-IRICA
University of Castilla-La Mancha
Avda. C. J. Cela, 10, 13071, Ciudad Real, Spain
E-mail: Felix.Jalon@uclm.es
ruthenium catalysts[28,29] – including pincer derivatives[14,30,31]
–
have been reported but in these cases the presence of a strong
base is needed. This approach implies low selectivity as the
basic medium favors deuteration at C2 (in addition to deutera-
tion at C1), although in some cases this regioselectivity also
depends on the auxiliary ligands around the metal center.[32]
High catalyst loadings in molybdocene derivatives,[29,33,34] iridium
complexes[35] or derivatives with very sophisticated ligands have
also been reported.[36]
[b] Dr. G. Espino
Departamento de Química
Facultad de Ciencias
University of Burgos
Plaza Misael Bañuelos s/n, 09001, Burgos, Spain
E-mail: gespino@ubu.es
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/ejic.202100022
Eur. J. Inorg. Chem. 2021, 1358–1372
1358
© 2021 Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejic.202100022
An interesting alternative is the direct formation of
secondary alcohols, selectively deuterated in the C1 position,
from aliphatic and aromatic ketones. Excellent D-incorporations
were achieved using [Cp*Ir(H2O)(4,4’-OH,bpy)]SO4 (4,4’-OH,b-
py=4,4’-dihydroxy-2,2’-bipyridine) as pre-catalyst. Importantly,
this Ir-compound does not incorporate deuterium in the TH of
the more reactive benzaldehyde. The authors concluded that
substrates with high reactivity led to products with low
deuterium content or non-deuterated compounds. This high
reactivity makes also more challenging the selective TH of
aldehydes due to possible presence of secondary reactions.
We have previously reported[53,55] that the pre-catalyst [(p-
cym)RuCl(4,4’-Me2bpy)]BF4 (1) (Figure 1) with HCO2H/HCO2Na in
D2O generates a hydride that is rapidly exchanged with the D+
deuterated alcohols by the reduction of the corresponding
aldehydes or ketones. One way is by conventional methods of
organic synthesis, but this requires the use of expensive and
non-environmentally friendly deuterated reducing reagents
[39,40]
such as NaBD4,[37] LiAlD4,[38] SiDR3
or active metals.[41] The
catalytic hydrogenation of aldehydes or ketones in a homoge-
neous phase using transition metal complexes and D2 has been
reported.[42] The system H2/D2O has been explored for the
deuteration of numerous saturated and unsaturated substrates
using Pd/C.[43] The deuteration was also possible on using
alcohols as substrates.[44] The main drawback with this process
is the lack of selectivity, as different positions in the products
were deuterated. Previously, Lau and Cheng described the
deuteration of acetophenone with H2 in a homogeneous D2O/
organic medium using a Ru complex.[45] In the field of photo-
catalysis, Lloret-Fillol[46] described a dual cobalt-copper light-
driven system that is active in the hydrogenation and
deuterogenation of such substrates to give good yields
irradiating with visible light. The main drawbacks are the
complexity of the systems and the lack of selectivity in alkyl-
ketones and -aldehydes.
Transfer hydrogenation (TH) is a process that has the
advantages of operational simplicity[47] and excellent precedents
in the use of water-soluble catalysts for the TH of aldehydes
and ketones,[48–52] but this method has scarcely been explored
for deuteration processes. In fact, the deuteration of aldehydes
by a coupled methodology in which a TH and an H/D exchange
in the catalyst are employed has yet to be reported and a
similar process with ketones is rare in the literature.[53,54] Himeda
et al.[54] reported a TH process with HCO2H/D2O to obtain
$
from D2O: MÀ H+D+ MÀ D+H+ (inversion of polarity of the
water deuteron, umpolung). The displacement of this equili-
brium to the right is favored by the high number of deuterons
in the acidic D2O used as solvent although sodium formate and
formic acid are not enriched in deuterium. In this work, we have
chosen 1 as precatalyst since very few transition metal hydrides
allow an efficient process of umpolung[56,57] since the protona-
tion of a hydride can lead to unstable dihydride complexes due
to the loss of H2 or to complexes that are too stable without the
possibility of H/D exchange. We tested other halide compounds
that could behave like 1, but our efforts so far have been
unsuccessful. For example, neutral[58] or cationic[59] [Cp*IrCl
(NN’)]n+ (n=0, 1) species are able of generating hydride species
in the aqueous reaction medium, but these species do not
exchange deuterium for D2O or do so at an insufficient rate to
behave as 1. In principle, according to the proposal of Himeda
and our own experience,[53–55] the generation of the metal-D
group must be fast enough to compete kinetically with the TH
Figure 1. Substrate scope and structure of complex 1.
Eur. J. Inorg. Chem. 2021, 1358–1372
www.eurjic.org
1359
© 2021 Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejic.202100022
process to allow a high level of deuterium incorporation. Thus,
media. The substrates used and the molecular formulae of 1 are
a lowering of the hydride transfer rate to the substrate would
increase the deuteration level. Complex 1 was able to produce
deuterated alcohols from ketones efficiently.[53] However, in the
case of the TH of imines, the process in D2O took place too
rapidly to allow a good level of deuterium incorporation in the
corresponding amines. However, when the process was carried
out in the biphasic system D2O/toluene (dw/t), the hydro-
genation was slowed down, and the level of D incorporation
increased markedly.[55]
Thus, and considering the lack of examples of deuterium-
labeling of alcohols obtained by TH of aldehydes, we decided
to address this area with the aim of obtaining selective
deuteration in the C1 position of the alcohols using complex 1.
Besides, it was an aim of the work to compare the hydro-
genation and deuterogenation of aromatic, aliphatic and α-β
unsaturated aldehydes, both in water and in a biphasic water/
toluene mixture. Another important objective of the use of the
biphasic system was to facilitate the recovery and recycling of
the catalyst[60] as this has the advantages of homogeneous and
heterogeneous catalysis. Effective recycling of catalysts is of
great relevance in the applicability of these methodologies
since it affords purer products and it also allows to reduce the
generation of waste and the cost of catalysts. This biphasic
system works due to the water solubility of 1 and the toluene
solubility of the substrates and the reaction products, which
remain in the organic phase. Consequently, the separation of
the organic product from the metal catalyst is easily accom-
plished through liquid-liquid extraction. Toluene and water are
barely soluble within each other even at high temperature
indicated in Figure 1.
TH of benzaldehyde and substituted benzaldehydes in water
or a water/toluene biphasic medium
The resulting yields for the TH of benzaldehyde and its
functionalized derivatives (with para- and ortho-substituents) in
water and a biphasic medium are compiled in Table S1 and
Table S2. Catalyst 1 led to nearly full conversion for most of the
explored ortho- or para-substituted benzaldehydes after 7–
24 hours, both in aqueous and biphasic (w/t) media. Control
studies in the absence of catalyst in water/toluene for substrate
I-Br under the standard catalytic conditions reflected the
catalytic character of these transformations (entry 32 in Table S1
of the SI).
The yields for the alcohols obtained from the para-
substituted substrates after 7 hours in water (Table 1) can be
ordered according to the following sequence of R: H (93)�CF3
(92)>Me (85)>NO2 (71)>Br (61)>CN (57)>OMe (54), while in
w/t the sequence is: H (98)>OMe (91)�CF3 (90)�CN (89)>Me
(86)�NO2 (84)>Br (71). Hence, both electron-acceptor and
electron-donor substituents in para are detrimental for the TH
of the respective substrates, although this effect is considerably
dependent on the reaction medium. Indeed, the two media
produce contrasting effects over some specific substrates (e.g.,
OMe and CN). Similarly, the ortho-substituted aldehydes are
reduced more slowly as well (after 7 hours of reaction), albeit,
neither electronic or steric effects can be invoked to explain the
yields sequence: H (93)>Me (70)�NO2 (65)�OH (65)>OMe
(61) in water, and H (98)>NO2 (94)>OH (86)�OMe (86)>Me
(66) in w/t. Furthermore, we observed different degrees of
hydrogenation for substrates with the same substituent located
in different positions (para vs ortho). It is worth mentioning that
the nitro group was not reduced under the above-mentioned
conditions.
Intriguingly, in the bibliography there is no consensus on
the influence of either electronic or steric effects over the TH of
aldehydes and ketones. Fish et al. observed subtle electronic
and steric effects in the initial rates of the TH of aldehydes and
ketones employing a Rh catalyst, [Cp*Rh(bpy)(H2O)](OTf)2, in
water at pH=7 at room temperature and using HCO2Na as the
hydride source. In particular, they found that the TH of I-OMe is
slightly faster than that for I in a 20/13 ratio (favorable effect of
the OMe group, in our case the yield ratio was 54/93 in water
and 91/98 in w/t).[62] On the contrary, O’Connor et al. reported
that aryl aldehydes with electron-poor substituents in para
position are reduced faster than substrates with electron-rich
groups in the same position, using the pre-catalyst [Cp*Ir-
(pyridinesulfonamide)Cl] in isopropanol at room temperature in
the absence of base.[63] These authors attribute the observation
to the weakness of the aldehyde bond with electron-poor
groups in para position. In the same work, it was observed that
the o-MeO benzaldehyde (II-OMe) was reduced faster than its
para-substituted counterpart. Similar effects of electron-with-
drawing groups on nucleophilic reactions involving benzalde-
°
(solubility of toluene in water: 538 μg/mL at 21 C, and 1590 μg/
mL at 100 C)[61] and therefore they form a good solvent mixture
°
for a two-phase system.
The results presented here show that the incorporation of
deuterium in the final alcohol is greatly improved on using the
biphasic system in the case of nearly all substrates and good
recyclability was achieved. Interestingly, and contrary to our
expectations, it was also found that the hydrogenation process
takes place more easily in the biphasic medium when
compared to neat water.
Results and Discussion
Firstly, we decided to study the TH of several benzaldehydes
under standard conditions, to verify the activity of 1 and to
adjust the reaction times in the deuteration processes.
In particular, we studied the TH of several aldehydes using
HCO2H/HCO2Na as the hydrogen donor in the presence of 1 as
precatalyst and using water (w) or water/toluene in a 2:1 ratio
(w/t) as the solvent system. Moreover, the initial pH was
°
adjusted to 4.4, the temperature was set at 80 C and the used
catalyst/substrate/HCO2Na molar ratio was 1/200/6000
(0.5 mol% of catalyst). These experimental conditions were
optimized by us in previous reports on the hydrogenation of
unsaturated compounds in the presence of 1[53,55] in aqueous
Eur. J. Inorg. Chem. 2021, 1358–1372
www.eurjic.org
1360
© 2021 Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejic.202100022
Table 1. Yields at 7 h (if not specified) of the catalytic reactions for the substrates indicated in Figure 1.[a]
Product(s) [%] in w/t
Substrate
Alcohol [%] in w
Alcohol [%] in w/t
Substrate
I-OMe
I-Me
I
I-Br
I-CF3
I-CN
54
85
93
61
92
57
91
86
98
71
90
89
IIIa
IIIb
IVa
Va
VIa
VIIa
13
68
5
45
64
34
77
>99
1
I-NO2
71
84
II-OH
II-OMe
II-Me
62
61
70
67
86
86
66
94
II-NO2
VIIIa
95 (48 h)
°
[a] T=80 C, cat=0.626 μmols, cat/S/HCO2Na=1/200/6000, cat=0.5 mol%, biphasic conditions: 2 mL H2O:1 mL toluene, initial pH=4.4 adjusted with
HCO2H, 500 rpm. Experiments were replicated at least twice for the sake of reproducibility.
hydes have been previously described.[64] By contrast, Ogo
et al.[65,66] concluded that steric and electronic effects exert a
negligible influence in TH of ketones catalyzed by [(η6-C6Me6)
Ru(bpy)(H)]+ and [Cp*Ir(bpy)(H)]+ at pH 4.0 and 2.0, respec-
tively.
In this study, we have established that the use of a w/t
biphasic medium results in better yields for most of the
substrates, at least at short reaction times (Table S1 and
Table S2). This effect is reflected in Figure 2 for the TH of the
para-methoxybenzaldehyde (I-OMe). In the case of substrates, I-
Me, I, I-CF3 and II-Me, similar outcomes were obtained in both
media.
TH of alkylic and α,β-unsaturated aldehydes
The w/t biphasic medium was employed to carry out additional
TH experiments, due to its superior activity and the recycling
advantages. The results for some alkylic and α,β-unsaturated
aldehydes are provided in Table 2. A control experiment in the
absence of catalyst 1 for substrate Va turned out to give a null
conversion (entry 11).
The TH of cyclohexylcarbaldehyde, IIIb, (entries 12–13) did
occur, although the yields were generally lower than those
obtained under the same conditions for benzaldehydes. The
unsaturated substrate VIa (entries 14–15), with a non-conju-
Figure 2. TH of 4-methoxybenzaldehyde (I-OMe) in the presence of 1 as pre-catalyst in water (green squares) or the biphasic water/toluene (red squares)
medium.
Eur. J. Inorg. Chem. 2021, 1358–1372
www.eurjic.org
1361
© 2021 Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejic.202100022
Table 2. TH of alkylic and α,β-unsaturated aldehydes using 1 as catalyst.[a]
Entry (subs.)
Time [h]
IIIa [%]
IIIb [%]
IIIc [%]
IIId [%]
IIId/IIIb+IIIc
1 (IIIa)
2 (IIIa)
3 (IIIa)
7
24
48
36
12
0
6
6
0
13
26
31
45
56
69
2.4
1.8
2.2
R=H IVa
R=Me Va
IVb
Vb
IVc
Vc
IVd
Vd
Entry (subs.)
Time [h]
IVa [%]
IVb [%]
IVc [%]
IVd [%]
IVd/IVb+IVc
4 (IVa)
5 (IVa)
6 (IVa)
7 (IVa)
7
27
0
2
4
0
0
24
5
64
82
29
2
7.1
4.6
0.4
–
24
18
69
3
24[b]
24[c]
71
Entry (subs.)
Time [h]
Va [%]
Vb [%]
Vc [%]
Vd [%]
Vd/Vb+Vc
8 (Va)
9 (Va)
10 (Va)
11 (Va)
7
65
18
0
0
0
0
0
1
9
16
0
34
73
84
0
34
8.1
5.3
–
24
48
24[d]
100
Entry (subs.)
Time [h]
IIIc [%]
Entry (subs.)
Time [h]
VId [%]
12 (IIIb)
13 (IIIb)
7
24
68
>99
14 (VIa)
15 (VIa)
7
24
77
>99
°
[a] T=80 C, cat=0.626 μmols, cat/S/HCO2Na=1/200/6000, cat=0.5 mol%, biphasic conditions: 2 mL H2O:1 mL toluene, initial pH=4.4 adjusted with
HCO2H, 500 rpm. Experiments were replicated at least twice for the sake of reproducibility. [b] pH=5.1. [c] pH=8.8. [d] In the absence of catalyst.
gated double bond, produced only the unsaturated alcohol VId
after 24 h. This result demonstrates the chemo-selectivity of this
reaction conditions towards the hydrogenation of the aldehyde
group relative to the isolated C=C double bond, which
remained unchanged.
It is well-known that α,β-unsaturated aldehydes, such as IIIa,
IVa (cinnamaldehyde) and Va (2-methylcinnamaldehyde), can
be reduced through two different mechanisms, i.e.: 1,2- or 1,4-
addition (see Scheme 1). The direct hydrogenation of the C=C
double bond can be dismissed taking into account the result
Scheme 1. 1,2-additon or 1,4-addition as alternatives in hydrogenation processes of α,β-unsaturated aldehydes.
Eur. J. Inorg. Chem. 2021, 1358–1372
www.eurjic.org
1362
© 2021 Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejic.202100022
obtained for VIa. Moreover, we have previously proved that
Selective TH of furfural to furfuryl alcohol and glucose to
sorbitol
complex 1 cannot promote the TH of 2-cyclohexenol.[67] On the
other hand, the 1,4-addition yields an enol-intermediate which
is transformed into the saturated aldehyde (b) by tautomerism,
which in turn, is hydrogenated to the respective alcohol (c) by
mean of a 1,2-addition. The direct 1,2-addition would lead to
the allylic alcohol (d). The selective hydrogenation of α,β-
unsaturated aldehydes, such as cinnamaldehyde, is very
advantageous since it allows to obtain valuable allyl-alcohols
(cinnamyl alcohol). Unsaturated alcohols are regarded as more
valuable due to their role in synthesis, especially in the
fragrance industry. In order to obtain cinnamyl alcohol, in most
cases a heterogeneous catalyst is used[68,69] although nano-
catalysts,[70] plasmonic nanoparticles[71] or homogeneous
catalysts[72] have also been reported.
The production of renewable fuels and chemicals from biomass
is an active field of research. In line with this, the hydrogenation
of two additional aldehydes, furfural and glucose (produced in
excess from biomass) was examined under the above exper-
imental conditions.
Furfural (FF) is one of the top value-added chemicals
derived from biomass[75] and it is considered a renewable source
of C5 molecules, some of which have important added value.[76]
More specifically, furfuryl alcohol (FA) is a key precursor for the
manufacture of foundry resins, furan fiber-reinforced plastics,
lubricants, drugs, fragrances and pesticides.[77] Furfural is
generally produced by the acid-catalyzed dehydration of
pentose, which can be obtained from biomass, in aqueous
solution.[78–80] However, the extraction of furfural from the
aqueous solution is arduous and costly. Hence, the hydro-
genation of furfural in aqueous medium to generate valuable
chemicals is an attractive chemical transformation. Although
there are precedents for FF hydrogenation in the field of
heterogeneous catalysis,[81] processes in water are scarce and
they require the use of H2 under pressure.[82,83] We therefore
decided to test the behavior of 1 because it was envisaged that
the biphasic medium would allow easy separation and the use
of H2 would not be necessary. To our delight, furfural was fully
reduced to furfuryl alcohol after 7 hours (entry 1, Table 3). In
this case, evidence for a 1,4-addition was not observed and this
reflects the stability of the furan heterocycle present in this
aldehyde.
According to the US Department of Energy, sorbitol is
considered one of the most promising biomass-based platform
products.[84] Several products can be obtained from sorbitol
after aqueous-phase reforming as methanol, alkanes, and
biofuels.[85,86] Sorbitol is synthesized from either glucose (one of
the main products of cellulose hydrolysis) or sucrose through a
catalytic hydrogenation process using H2 and a nickel catalyst
at high temperatures[87–89] or other metals.[90] Sorbitol can be
alternatively obtained by electrochemical reduction of dextrose
or glucose.[87–89,91] A few microorganisms, (i.e.: three yeast strains
and bacteria such as Zymomonas mobilis and Candida boidini)
have been employed as sorbitol producers as well.[87,92,93]
The results obtained in this study for the reduction of
glucose are gathered in Table 3 (entries 2–4). In this case, the
reduction required harsher conditions than for FF, and only
55% of the aldehyde was converted into sorbitol (entry 2) upon
48 hours. Taking into account that the reduction of glucose
requires a preceding hydrolysis process, a supplement of
HCO2H was added and this improved the reaction to give
sorbitol in 95% yield (entry 3). An increase of the reaction
As shown in Table 2, the substrates IIIa, IVa and Va gave
quantitative conversions after 24 or 48 hours (entries 3, 5 and
10). Conversions at 7 hours also indicated that substrate IVa is
the most reactive (compare entries 1, 4 and 8). With regard to
selectivity at the end of the reaction, the α,β-unsaturated
alcohol (d) was the major product, in agreement with 1,2-
addition as the dominant mechanism. Nevertheless, the 1,4-
addition seems to operate as well, although in a lower extent,
because products b and c were also present. Interestingly, in
another study on the TH of 2-cyclohexenone with 1,[53] we
found that the only pathway operating was a 1,4-addition. The
difference with the result obtained in this work could be due to
the lower reactivity of a ketone when compared to an aldehyde
group. Our data are also in contrast to previous results of Joó
et al.,[73] who used a rhodium(I)-sulfonated triphenylphosphine
catalyst and found that the C=C double bond was hydro-
genated to yield 3-phenylpropanal. However, the Ru complex
[{RuCl2(mtppms)2}2] (mtppms=sodium 3-diphenylphosphino-
benzenesulfonate) exhibited selectivity towards the hydrogena-
tion of the carbaldehyde group.[74] In contrast, Himeda et al.
observed that the conversion and selectivity of this reaction
depends on temperature and pH when employing a Cp*Ir
complex with 4,4’-dihydroxy-2,2’-bipyridine as the catalyst.[49]
The ratios of the product of a 1,2-addition (d) with respect
to those that arise from a 1,4-addition (b+c) are provided in
the last column of Table 2 as an index of chemoselectivity. For
substrate IIIa this ratio is in the range 1.8–2.4 (entries 1–3).
Higher values were obtained for substrates IVa and Va (from 4.6
to 34, entries 4, 5 and 8–10) although in certain cases some of
the substrate remained. When the conversion was total, the
values of this ratio were 4.6 (substrate IVa) and 5.3 (substrate
Va). The effect of a change in the pH on this ratio was analyzed.
At pH=5.1 (entry 6), although the activity was not reduced, the
selectivity was reversed, and the most abundant product was
the fully hydrogenated alcohol IVc. At pH 8.8 the reaction
slowed down, thus reflecting in both experiments that the pH
has a marked effect on the outcome of the reaction.
°
temperature up to 100 C, without additional formic acid, was
also beneficial alternative for the yield of sorbitol (entry 4).
Eur. J. Inorg. Chem. 2021, 1358–1372
www.eurjic.org
1363
© 2021 Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejic.202100022
Table 3. Yields obtained for the respective alcohols from the TH of aldehydes VIIa and VIIIa using 1 as pre-catalyst.[a]
°
°
Entry
1
Time [h]
7
Temp. [ C]
Yield
VIId [%]
Entry
Time [h]
Temp. [ C]
Yield
VIIIb [%]
80
>99
2
48
48
48
80
80
100
55
95
95
3[b]
4
°
[a] T=80 C, cat=0.626 μmols, cat/S/HCO2Na=1/200/6000, cat=0.5 mol%, biphasic conditions: 2 mL H2O:1 mL toluene, initial pH=4.4 adjusted with
HCO2H (26 μL), 500 rpm. [b] HCO2H=130 μL. Experiments were replicated at least twice to confirm reproducibility.
Deuterium labeling in the TH of benzaldehyde derivatives
Deuterium labeling in the TH of alkylic and α,β-unsaturated
aldehydes
Deuterium labeling was evaluated in the TH of the benzalde-
hyde derivatives on using D2O (dw) or the biphasic medium
D2O/toluene (dw/t). It is important to note that in these
experiments the only deuterium source was D2O. Neither the
precatalyst, substrates nor the hydrogen source (HCOOH/
HCOONa) were enriched in deuterium, demonstrating the
sustainability and simplicity of the processes. If the proportion
of deuterium in this medium is calculated (see SI pS2), it is
estimated that the maximum deuteration level to be reached in
the products will be 99.16%. To our delight, selective incorpo-
ration of deuterium in the C1 position was observed for all of
the benzyl alcohols obtained and the results are summarized in
Table 4. In the absence of 1, hydrogenation reaction or
deuterium incorporation were not observed (tested with
substrate I-OMe, see Figure S1c). When benzyl alcohol was
Next, we decided to assess the deuterium incorporation in the
TH of alkylic and α,β-unsaturated aldehydes in the dw/t
biphasic system. Reaction times in the range 7–48 hours were
required to achieve full conversions.
The results for the alkylic aldehydes IIIb, IVb and VIa are
provided in Scheme 2. Values for D incorporation in the C1
position in the range 63–75% were found.
The expected D incorporation for substrate IIIa is depicted
in Scheme 3, for both the 1,2- and the 1,4-addition pathways.
Table 4. Selective deuteration of benzyl alcohols in C1 using
1 as
precatalyst in D2O (dw) or in the biphasic medium D2O/toluene (dw/t)
through TH of benzaldehydes.[a]
°
heated for 24 h at 80 C in dw/t in the presence of 1,
incorporation of deuterium was not observed (Figure S2). Thus,
the deuterium labeling takes place in the hydrogen transfer
process.
While the deuterium incorporation was moderate or low in
D2O (dw) (6–45%), when the process was carried out in the dw/
t two-phase medium (t=toluene) there was a notable increase
in this incorporation for all the substrates, except for I-CF3.
Values of D incorporation up to 92% were found. A similar
behavior was noticed previously by us in the selective C1
deuterogenation of amines obtained through TH of the
respective imines under similar catalytic conditions to those
used in this work.[55] In the deuterogenation of the imines, the
hydride transfer rate was slowed down in dw/t and as a result
Entry
Substrate
[%]D (dw)
[%]D (dw/t)
[%]D (dw/t)/
[%]D (dw)
1
2
3
4
5
6
7
8
I-OMe
I-Me
I
I-Br
I-CF3
I-CN
I-NO2
II-OH
II-OMe
II-Me
II-NO2
24
43
25
42(48 h)
44
2
6
16
3
91
92
83
64
18
36
52
76
63
85
45
3.8
2.1
3.3
1.5
0.4
18
8.7
4.8
21
$
the Ru-H Ru-D exchange took place in a higher degree. Hence,
9
10
11
this effect was invoked as the main cause to explain the rise in
the deuterium incorporation. Nonetheless, in the TH of
aldehydes the biphasic medium lead to an increase of the
reaction rate as above-mentioned. Thus, we have found a
different behavior and the reason for the higher deuterium
incorporation in the biphasic medium should be different.
29
<1
2.9
>45
°
[a] T=80 C, cat=0,626 μmol, cat/S/HCO2Na=1/200/6000, cat=0.5 mol%.,
with 2 mL D2O or biphasic medium (2 mL D2O:1 mL toluene), initial pD=
4.8 adjusted with HCO2H, 500 rpm; % D is the deuterium incorporation in
the C1 position of the alcohol. Unless stated, time=24 hours. Yields were
determined by 1H NMR spectroscopy. Experiments were replicated at least
twice to demonstrate reproducibility.
Eur. J. Inorg. Chem. 2021, 1358–1372
www.eurjic.org
1364
© 2021 Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejic.202100022
Scheme 2. Deuteration of the alcohols obtained in the TH of alkylic aldehydes.
Scheme 3. Deuterium incorporation in the TH of IIIa by 1,2- or 1,4-addition pathways. H/D represents a deuterium incorporation that depends on the degree
$
of the Ru-H Ru-D exchange prior to the hydride/deuteride transfer from the catalyst.
The 1,2-addition pathway gives rise to a partial deuteration
over C1. The first step in the 1,4-addition pathway implies the
partial deuteration in the C3 position. The keto-enol tautomer-
ism gives rise to an elevated D incorporation in C2 and the
successive 1,2-addition brings about the partial incorporation of
D in C1 atom of the hydrogenated alcohol.
Catalyst recycling
Catalyst recyclability reduces waste generation and improves
the efficiency of the synthetic processes. Indeed, recyclability is
one of the pillars of green catalysis.[94]
A priori, the use of a two-phase medium could facilitate the
separation of products and catalyst, since it can be accom-
plished by simple decantation, as long as the solubility of the
catalyst and the substrates and products in the solvent system
are appropriate. We first assured the complete solubility of the
aldehydes and the resulting alcohols in toluene under the
aforementioned experimental conditions, as well as the solu-
bility of the catalyst in the aqueous phase (See pS1). Hence,
catalyst and organic products were selectively solved in the
aqueous and toluene media, respectively.
The catalyst recycling experiments were performed employ-
ing p-anisaldehyde (I-OMe) as the model substrate in the w/t
biphasic medium under the standard catalytic conditions. Thus,
ten cycles of 24 hours each were completed, and the organic
phase retired. Only the corresponding amount of the retired
anisaldehyde in toluene was added at the end of cycles 1 to 9.
The pH of the water solution was experimentally determined at
the end of every cycle and changed from 4.5 to 10 (See
Figure 3). We determined experimental conversions higher than
99% for the p-OMe-benzyl alcohol at the end of the first six
The experimental deuteration and yields determined for the
α,β-unsaturated aldehydes IIIa, IVa and Va are depicted in
Scheme 4. It should be noted that the α,β-unsaturated alcohol
is the major species in all cases, which indicates that 1,2
addition is faster than 1,4, according to the mechanism
indicated in Scheme 3. Species IIIb, IVb and Vb, expected in the
1,4 addition, do not appear among the reaction products, which
indicates their fast transformation into the respective species
IIIc, IVc and Vc in the experimental reaction times. These will be
formed by the successive 1,2 addition from the first ones, which
again indicates that the 1,2 addition is favorable over the 1,4.
For alcohols IIIc, IVc and Vc, which are presumably formed by a
1,4-addition pathway involving a keto-enol tautomerism step,
we observed the expected partial deuteration in C1 and C3
along with a high degree of D labeling in C2 (lower for Vc).
Moreover, the incorporation of D in both hydrogen positions of
C2 on IVc is explained due to the reversibility of the keto-enol
tautomerism.
Eur. J. Inorg. Chem. 2021, 1358–1372
www.eurjic.org
1365
© 2021 Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejic.202100022
Scheme 4. Deuterium incorporation through the TH of α,β-unsaturated aldehydes catalyzed by 1.
Figure 3. Graphic of the obtained conversion and final pH of 10 consecutive cycles (24 h each) of the catalytic hydrogenation of I-OMe with catalyst 1 in
standard conditions. The alcohol is the sole product obtained.
Eur. J. Inorg. Chem. 2021, 1358–1372
www.eurjic.org
1366
© 2021 Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejic.202100022
cycles after simple decantation and evaporation of toluene.
added and therefore the aquo-derivative reacts to give a
Interestingly, addition of HCO2H was not needed despite this
acid is consumed in the reaction. The conversions gradually
decreased from the cycle number 7 on, up to a value of 25% in
the tenth cycle. It should be concluded that complex 1 shows
great robustness compared to other catalysts of similar
formulation.[95] Another characteristic to be highlighted is its
high activity up to a pH value of 9.5 at which other similar
compounds lose their activity to a greater extent.[57,65]
formato-complex 1(iii) that evolves to the corresponding
hydrido derivative 1(iv) (See Figure S3a for H-NMR study and
1
Figure S3b the Gibbs energy profile, according to Ref. [53]).
Chlorido,[96] aquo,[65] formato[65] and hydrido[97] complexes with
similar structures have been isolated and characterized by X-ray
diffraction. We also demonstrated that the hydrido complex
$
[RuH(p-cymene)(4,4’-Me2-bpy)]BF4 undergoes RuH RuD ex-
change in D2O.[53] We have now detected the Ru-D resonance in
2
the H NMR spectrum of complex 1(iv) in D2O solution in the
presence of HCO2Na/HCO2H (10 and 1.78 eq. respectively) after
°
Mechanistic Proposal and Computational Studies
heating at 80 C for 10 minutes (see Figure S3c). DFT studies (in
aqueous solution) support the hypothesis that this exchange is
a consequence of a reversible reaction between this hydride
derivative and the deuteron of D3O+, which evolves to a
dihydrogen intermediate 1(v) that has not been observed in
solution. Given the information outlined above, the mechanism
proposed in Figure 4 seems reasonable. We propose that once
Firstly, it is appropriate to say that, according to our previously
reported experimental studies and DFT calculations,[53] pre-
catalyst 1(i) equilibrates in D2O solution with an aqua-complex
as a result of a chloride shift by a molecule of the solvent (1(ii)
in Figure 4 and Figure S3a). An excess of formate (10 eq.) is
$
Figure 4. Proposed mechanism for the aldehyde hydrogenation involving D-labeling and for the Ru-H Ru-D exchange with 1 in D2O.
Eur. J. Inorg. Chem. 2021, 1358–1372
www.eurjic.org
1367
© 2021 Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejic.202100022
the hydrido-metal complex 1(iv) has formed (Ru-H or Ru-D), the
concluded that the increase of the number of water molecules
(4 or 5) in the cluster nearly did not affect the energetic of the
processes. Full optimization of complex+substrate+water
cluster species in the solvent allowed to locate all the
intermediates and transition states.
hydride group is transferred to the electrophilic carbaldehyde
group 1(vi), which is probably polarized in the acidic medium.
Then, the formed alcohol 1(vii) is released and the aqua-
complex 1(ii) regenerates.
Therefore, we postulate that the incorporation degree of
protium or deuterium in the carbon atom C1 of the respective
alcohol depends on the relative rates of the hydride transfer to
The hydride transfer to the C1 position of the obtained
alcohols results from the nucleophilic attack of the Ru-H species
on the C=O functional group of the aldehyde upon its proton
activation.[98,99] This means that the transfer of a proton and a
hydride takes place in a concerted manner and thus, in the
absence of a proton donor function in the catalyst, the acidic
medium is necessary in the TH process. Thus, we believe that as
1 is a coordinatively saturated complex, the hydride transfer to
the carbonyl group should take place by an outer sphere
mechanism.[98,99] An initial outer sphere intermediate for the
hydrogenation of benzaldehyde and anisaldehyde was opti-
mized, with one of these molecules located in the vicinity of
complex 1 and in the presence of a proton solvated by three
water molecules (routes in blue and red colors, respectively, in
Figure 5). In the resulting optimized intermediate RuH-ald
(3wH+), the C=O bond exhibits the appropriate orientation to
accept the two hydrogen atoms, but the proton-O and hydride-
$
the substrate – 1(vi) to give 1(vii) and that of the Ru-H Ru-D
exchange (small cycle involving iv and v). In order to evaluate
the energy differences between these processes, theoretical
calculations were carried out at the DFT level for the process in
$
water. The Ru-H Ru-D exchange process has been calculated
previously[53] (route of green color in Figure 5) and the
calculations for the hydride transfer on benzaldehyde and
anisaldehyde are included by the first time in this work. These
calculations will also help in proposing a possible explanation
for the observation of other experimental findings, such as the
general low incorporation of deuterium in the benzyl alcohols
in water and the notable difference in the hydrogenation of
benzaldehyde (I) and anisaldehyde (I-OMe) in this medium after
7 hours (93% and 54%, respectively, see Table 1).
In order to adopt the appropriate solvent model for this
study, a cluster-continuum model was used that combines the
explicit inclusion of a cluster of three water molecules in the
quantum mechanical description of the system along with the
polarizable continuum model. This methodology and the
adoption of three water molecules in the cluster as a valuable
model has been discussed elsewhere.[53] For instance, it was
C distances are rather long (1.62–1.67 and 2.46–2.73 Å,
respectively). At this point, the aldehyde is still not protonated
(Figure 5 and Table 5).
This intermediate gives rise to the transition state for the
hydrogen transfer (TS-HT-ald(3wH+), Figure 5) and this directly
evolves to the reaction products (Ru-alc(3w)). Accordingly, the
process is concerted but markedly asynchronous, since though
Figure 5. Gibbs energy profile in water (Kcal·molÀ 1) for the steps of hydride transfer steps to anisaldehyde and benzaldehyde and for the Ru-H/D+ exchange
of [(p-cym)RuH(4,4’-Me2bpy)]+. Three explicit water molecules are present in all the calculated structures. ΔG for the deprotonation of HCOOH (5.1 Kcal·molÀ 1
has been added in the hydrogenation step (see text).
)
Eur. J. Inorg. Chem. 2021, 1358–1372
www.eurjic.org
1368
© 2021 Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejic.202100022
Table 5. Distances [Å] and energies [Kcal·molÀ 1] calculated for the indicated species.
Benzaldehyde
Hydride-C
Anisaldehyde
Hydride-C
Structure
Proton-O
Energy
Proton-O
Energy
RuH-ald(3wH+)
TS-HT-ald(3wH+)
Ru-alc(3w)
2.46
1.68
1.12
1.67
1.20
0.98
5.6
8.6
À 6.2
2.73
1.67
1.14
1.62
1.17
0.99
6.4
10.2
À 5.8
the proton transfer is already completed in TS-HT-ald(3wH+)
(O··H distance=1.17–1.20 Å), the hydride transfer is still starting
(C···H distance=1.67–1.68 Å). Both the CÀ H and OÀ H bonds are
fully formed (distances, CÀ H=1.12–1.14 Å, OÀ H=0.98–0.99 Å)
in the reaction product.
methoxybenzyl alcohol by 1.1 Kcal·molÀ 1 (Figure 5 and
Scheme 5).
Concerning the mechanism of the Ru-H/D+ exchange, the
results of our previous studies indicate that the process involves
deuteronation of the Ru-hydride species by D3O+ (in fact, by
the cluster +H(OH2)3 in our calculations), which results in a
dihydrogen-bridged and a η2-HD intermediate (Scheme 6).
These compounds are 5.1 and 9.7 Kcal·molÀ 1, respectively,
above the energy of the monohydride 1(iv) (left part in green
color of Figure 5). However, these processes require an
activation energy of 12.7 Kcal·molÀ 1 over the initial hydride (TS-
H-D (2w)) due to the transformation of the dihydrogen-bridged
to the dihydrogen-coordinated species.
In order to integrate the ΔG� values obtained from our
calculations, the energy required to generate the proton that is
transferred to the substrate in this step must be considered.
Taking into account that this proton comes from formic acid, it
is necessary to add the ΔG of deprotonation of the formic acid
in water, which can be easily deduced from its pKa value (3.75).
The ΔGdeprot(HCO2H) value is 5.1 Kcal·molÀ 1.[53] In this way,
addition of this value to the calculated ΔG� of the different
structures place intermediates RuH-ald(3wH+) at 5.6 and
6.4 Kcal·molÀ 1 in the reaction coordinate for benzaldehyde and
anisaldehyde, respectively. This difference between substrates is
increased in the transition states, which are 8.6 and
10.2 Kcal·molÀ 1 above 1(iv), respectively. This difference of
1.6 Kcal·molÀ 1 for the hydride transfer can be considered as
providing support for the experimentally observed difference in
the hydrogenation of the two substrates in water (see Table 1).
In addition, according to our calculations, the thermodynamics
of the hydrogenation of these two benzaldehyde molecules
favor the formation of the benzyl alcohol over the para-
Thus, on comparing the calculations for the two processes
(Figure 5) it can be concluded that the Ru-H/Ru-D exchange is
more difficult than the hydride transfer to benzaldehyde or
anisaldehyde by 2.5 or 4.1 Kcal·molÀ 1, respectively. This differ-
ence is large enough to account for the low labeling level in
these substrates, as observed experimentally (25 and 24%
respectively, Table 4).
The corresponding DFT calculations for the process in the
biphasic system cannot be addressed. As a consequence, the
energy barriers for the different steps in this medium are not
known and it is not possible to draw any conclusions about the
Scheme 5. Free energy for the hydrogenation process of benzaldehyde and anisaldehyde.
Scheme 6. Proposed mechanism for the Ru-H/D+ exchange.
Eur. J. Inorg. Chem. 2021, 1358–1372
www.eurjic.org
1369
© 2021 Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejic.202100022
unexpected finding that the process in dw/t leads not only to a
deuterium labeling (using D2O/toluene) but also to increased
yields of the alcohols. This finding could open new ways to
increase the deuteration levels and yields of other aldehydes (or
even ketones) obtained by TH processes.
higher deuterium incorporation but also to an increased rate of
hydrogenation. We tentatively propose that the relative rates of
$
hydride transfer and Ru-H Ru-D exchange are suitable for
deuteration and other factors such as, for example, differences
in relative solubilities in the two media and thus different
concentrations of the different species may be important in Experimental Section
determining the level of D incorporation.
Experimental Details
All reactions were performed under an atmosphere of dry nitrogen
Conclusions
using Schlenk-vacuum line techniques. Pre-catalyst 1 was synthe-
sized as previously reported by us.[53] Organic solvents were distilled
in the presence of the appropriate drying agents and degassed
before use. The HCO2Na/HCO2H buffer solution was prepared by
adding HCO2Na (6.530 g, 96.0 mmol) and HCO2H (650 μL,
17.2 mmol) to water (HPLC grade). The volume was adjusted to
25 mL in a volumetric flask. The pre-catalyst solution was prepared
by dissolving complex 1 (8.5 mg, 15.65 μmol) in water (HPLC grade)
and fixing the volume to 25 mL using a volumetric flask. The fresh
solutions were bubbled with nitrogen for several minutes and then
were stored under nitrogen. The aldehydes chosen as substrates for
the catalytic tests, HCOONa and HCOOH were obtained from
Aldrich and used as received. Toluene, Et2O and H2O were supplied
by Scharlab and D2O, CDCl3 by Eurisotop.
Procedures for the unprecedented transfer hydrogenation and
deuterium labeling of different aromatic, alkylic and α,β-
unsaturated aldehydes have been developed in the presence of
[RuCl(p-cymene)(dmbpy)]BF4, 1, (dmbpy=4,4’-dimethyl-2,2’-bi-
pyridine) as the pre-catalyst and HCO2Na/HCO2H as the hydro-
gen source. In addition, two aldehydes, namely furfural and
glucose, involved in waste management were also selectively
reduced to furfuryl alcohol and sorbitol, respectively. The
processes were carried out in neat water or in a biphasic water/
toluene system where the use of a phase transfer cocatalyst is
not necessary. The aromatic aldehydes were hydrogenated
more easily and in the case of the α,β-unsaturated aldehydes
the allylic alcohols were the major products as a result of a 1,2-
addition. The biphasic water/toluene system allows very
efficient catalyst recycling without the addition of HCO2H. We
have demonstrated that the reaction is operative for I-OMe
during 6 cycles with a 100% of conversion even at pH 9.5.
For the deuterium labeling, the only D source is D2O. The
deuteration is selective in the C1 carbon – except in the case of
the saturated alcohols arising from α,β-unsaturated aldehydes,
in which case deuterium incorporation was also found in the C2
and C3 positions. The biphasic water/toluene system leads to
higher yields of the corresponding alcohols and higher
deuterium incorporation (D2O/toluene) than when neat D2O is
used.
General procedure for the catalytic hydrogenation of aldehydes
in biphasic media. A Teflon®-capped 10 mL ampoule sealed with a
Young valve was charged with the buffer solution, HCO2Na/ HCO2H
(1 mL), the catalyst solution (1 mL), and a toluene solution (1 mL)
containing 0.128 mmol of substrates: I-OMe (15.5 μL), I-Me
(15.2 μL), I (13 μL), I-Br (23,7 μL), I-CF3 (17.55 μL), I-CN (16.8 mg), I-
NO2 (19.3 mg), II-OMe (17.4 mg), II-Me (14.9 μL), II-NO2 (19.3 mg), II-
OH (14.0 μL), IIIa (14.6 μL), IIIb (15.5 μL), IVa (16.1 μL), IVb (16.9 μL),
Va (17.9 μL), VIa (15 μL), VIIa (10.6 μL), VIIIa (23.0 mg). Then, the
mixture was deaerated by bubbling nitrogen for several minutes.
The cat/substrate/HCO2Na ratio was fitted to 0.64 μmol/
0.128 mmol/3.84 mmol=1/200/6000, giving
a final pre-catalyst
concentration of 0.32 mM. The experiments were replicated by
setting two ampoules with identical concentrations, which were
simultaneously introduced into a homemade multi-hole reactor
equipped with a mineral oil bath and a temperature sensor control.
°
The reaction mixtures were preheated at 80 C and stirred magneti-
A comparative DFT analysis in aqueous solution of the
hydride transfer process to anisaldehyde and to benzaldehyde
reflects that the energy barrier is 1.6 Kcal·molÀ 1 higher in the
case of anisaldehyde, a fact in accordance with the yield
differences found for the two hydrogenation processes in water
(52 and 93% in 7 h respectively). The DFT calculations also
reflect that the deuteration takes place due to an efficient
process in which the hydride is rapidly exchanged with the D+
cally for the specified reaction times (500 r.p.m.). Next, the reaction
was stopped by cooling the ampoules in an ice/water bath and the
solutions were transferred to a 5 mL vial and kept into the fridge
until analysis. The toluene phase was separated from the aqueous
phase and the latter was extracted with diethyl ether (2×5 mL). The
combined organic phases were concentrated in air. Solid samples
were solved in CDCl3 (0.5 mL) in an NMR tube and the signals were
referenced to TMS. Conversion and yields were determined by
integration of the respective signals in the 1H NMR spectra.
$
of D2O: MÀ H+D+ MÀ D+H+ (inversion of polarity, umpo-
Control catalytic experiments. Control experiments of 24 hours
were carried out in the absence of catalyst 1 following the same
procedure as described above for substrates I-OMe, I-Br and Va
(See Figure S1 in the SI).
lung). Besides, this RuH/D exchange process is more energeti-
cally demanding than the hydride transfer of both substrates
(2.5 Kcal·molÀ 1 higher than the barrier for anisaldehyde), which
is in accordance with the low deuteration level obtained in the
monophasic medium (D2O) in both substrates (ca. 25%).
General procedure for the catalytic hydrogenation of aldehydes
in monophasic media. For these experiments, the aforementioned
protocol was followed in the absence of toluene. Substrates were
charged directly into the ampoule with a 50 μL micro-syringe for
liquid aldehydes. The total volume was 2 mL.
It was expected that the biphasic medium would lead to a
lower rate of hydrogenation and thus it would allow a higher
deuterium incorporation in the resulting alcohols because the
Ru-H/Ru-D exchange would take place before the hydride
transfer to the substrate. However, it is interesting to note that,
unexpectedly, the biphasic system led not only to a higher
Selective deuteration experiments. The solutions of buffer and
pre-catalyst were prepared in D2O (instead of H2O) as explained
previously. It is worth noting that we used non-deuterated HCO2Na,
Eur. J. Inorg. Chem. 2021, 1358–1372
www.eurjic.org
1370
© 2021 Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejic.202100022
HCO2H and toluene. In order to assure full conversions, the reaction
BU087G19). MRC thanks FEDER funds and Plan Propio de I+D+i-
times were optimized for these catalytic experiments and the
above-mentioned procedure was applied. Analyses of substrates
and products were performed by integration of the H and H NMR
signals if necessary. See examples and the analysis procedure for
these spectra in the Supporting Information (Figures S4–S20).
UCLM (2014/10340) for a predoctoral contract.
1
2
Conflict of Interest
Catalyst recycling. These experiments were performed employing
para-methoxybenzaldehyde (I-OMe) as substrate. All of the proce-
dures were done under an inert atmosphere. The experiments were
The authors declare no conflict of interest.
°
carried out following the aforementioned procedure above at 80 C
Keywords: Alcohols
Deuteration · Homogeneous catalysis · Ruthenium · Transfer
·
Density functional calculations
·
for 24 h, and after the first catalytic cycle the reaction was stopped
in an ice/water bath. Once the solution reached room temperature,
the organic phase was separated. Pasteur pipettes were employed
to separate the organic phases. The yield was experimentally
hydrogenation
1
determined through integration of specific signals in the H NMR
spectra recorded in CDCl3. Fresh para-methoxybenzaldehyde and
toluene were subsequently added to the aqueous phase and ten
additional catalytic cycles were run following the same protocol.
[1] J. Atzrodt, V. Derdau, W. J. Kerr, M. Reid, Angew. Chem. Int. Ed. 2018,
57, 1758–1784; Angew. Chem. 2018, 130, 1774–1802.
[2] A. Michelotti, M. Roche, Synth. 2019, 51, 1319–1328.
[3] J. R. Shu, A. Y. L. Saunders, D. Levinson, S. H. Landvatter, S. W.
Mahoney, A. Senderoff, S. G. Mack, J. F. Heys, J. Labelled Compd.
Radiopharm. 1999, 42, 797–807.
[4] J. E. Baldwin, J. Labelled Compd. Radiopharm. 2007, 50, 947–960.
[5] G. Parkin, J. Labelled Compd. Radiopharm. 2007, 50, 1088–1114.
[6] T.-L. Liu, R. H. Lin, D.-L. Chang, W.-T. Liu, Ch. Tsay, W.-I. Li, J.-H. Kuo,
Anal. Chem. 2002, 619 A–626 A.
[7] E. Stokvis, H. Rosing, J. H. Beijnen, Rapid Commun. Mass Spectrom.
2005, 401–407.
[8] A. K. Hewavitharana, J. Chromatogr. A 2011, 1218, 359–361.
[9] A. Katsnelson, Nat. Med. 2013, 19, 2013.
[10] T. G. Gant, J. Med. Chem. 2014, 57, 3595–3611.
[11] A. D. N. Vaz, S. L. Ripp, R. Sharma, R. S. Obach, T. J. Strelevitz, D. K.
Spracklin, A. J. Clark, K. Schildknegt, L. M. Tremaine, H. Gao, Drug
Metab. Dispos. 2011, 40, 625–634.
[12] S. H. DeWitt, B. E. Maryanoff, Biochemistry 2018, 57, 472–473.
[13] A. P. Tulloch, Prog. Lipid Res. 1983, 22, 235–256.
[14] L. Zhang, D. H. Nguyen, G. Raffa, S. Desset, S. Paul, F. Dumeignil, R. M.
Gauvin, Catal. Commun. 2016, 84, 67–70.
[15] C. M. Yung, M. B. Skaddan, R. G. Bergman, V. Di, L. Berkeley, N.
Laboratories, J. Am. Chem. Soc. 2004, 126, 13033–13043.
[16] T. Maegawa, Y. Fujiwara, Y. Inagaki, Y. Monguchi, Adv. Synth. Catal.
2008, 350, 2215–2218.
[17] Y. Fujiwara, H. Iwata, Y. Sawama, Y. Monguchi, H. Sajiki, Chem.
Commun. 2010, 46, 4977–4979.
Computational details. The software Gaussian09 (see SI) was used
for all the calculations accomplished in the present study. The
geometrical optimization and energy calculations of the complexes
were performed at the DFT level using the M06 functional including
an ultrafine integration grid. The 6-31G(d,p) basis set was used for
the H,[100] C, N, O and Cl atoms.[101] The Ru atom was described using
the scalar-relativistic Stuttgard-Dresden SDD pseudopotential and
its associated double-ζ basis set,[102] complemented with a set of f-
polarization functions.[103] Optimizations and frequency calculations
to obtain the thermal and entropic corrections to calculate ΔG
values were done using this basis set.
More accurate energies of all the species were obtained by single
point calculations on optimized structures using the extended basis
set 6-311+ +G(d,p) on all the atoms C, N, O and Cl but the
ruthenium and are described as SDD+f. To take into account both
nonspecific and specific interactions with the solvent, a mixed
continuum/discrete solvent model was used.[104] In this model, in
addition to the continuum description of the solvent (SMD
continuum model)[105] three explicit water molecules, able to
establish hydrogen-bonding interactions with the catalyst and the
substrate, have been included.
[18] I. N. Lykakis, C. Tanielian, R. Seghrouchni, M. Orfanopoulos, J. Mol.
Catal. A 2007, 262, 176–184.
[19] J. L. Abad, G. Villorbina, G. Fabriàs, F. Camps, J. Org. Chem. 2004, 69,
The structures of the reactants, intermediates, transition states and
products were optimized in water solvent (ɛ=78.35) using the
SMD continuum model. Geometries of optimized stationary points,
located on potential energy surfaces (PES) of all the complexes, are
characterized as minima (all positive frequency) or transition states
(one imaginary frequency). The corresponding structure of every
transition state is confirmed by the intrinsic reaction coordinate
(IRC) method,[106,107] to confirm they connect with the corresponding
intermediates. All the energies collected in the text are Gibbs
energies in water at 298 K. Cartesian coordinates, optimized
structures, absolute energies and Gibbs energies (Hartrees) in water
of all the calculated species can be found in the SI.
7108–7113.
[20] M. Tashiro, H. Nakamura, K. Nakayama, J. Chem. Soc.-Perkin Trans.
1988, 179–181.
[21] M. Orfanopoulos, I. Smonou, C. S. Foote, J. Am. Chem. Soc. 1990, 112,
3607–3614.
[22] E. A. Cioffi, W. S. Willis, S. L. Suib, Langmuir 1990, 6, 404–409.
[23] E. A. Cioffi, R. H. Bell, B. Le, Tetrahedron: Asymmetry 2005, 16, 471–475.
[24] E. A. Cioffi, J. H. Prestegard, Tetrahedron Lett. 1986, 27, 415–418.
[25] H. Esaki, F. Aoki, M. Umemura, M. Kato, T. Maegawa, Y. Monguchi, H.
Sajiki, Chem. A Eur. J. 2007, 13, 4052–4063.
[26] N. Ito, T. Watahiki, T. Maesawa, T. Maegawa, H. Sajiki, Adv. Synth. Catal.
2006, 348, 1025–1028.
[27] H. Sajiki, N. Ito, H. Esaki, T. Maesawa, Tetrahedron Lett. 2005, 46, 6995–
6998.
[28] S. K. S. Tse, P. Xue, C. W. S. Lau, H. H. Y. Sung, I. D. Williams, G. Jia,
Chem. A Eur. J. 2011, 17, 13918–13925.
Supporting Information (see footnote on the first page of this
article): The data that supports the findings of this study are
available in the supporting information of this article.
[29] M. Takahashi, K. Oshima, S. Matsubara, Chem. Lett. 2005, 34, 192–193.
[30] E. Khaskin, D. Milstein, ACS Catal. 2013, 3, 448–452.
[31] B. Chatterjee, C. Gunanathan, Org. Lett. 2015, 17, 4794–4797.
[32] W. Bai, K. H. Lee, S. K. S. Tse, K. W. Chan, Z. Lin, G. Jia, Organometallics
2015, 34, 3686–3698.
[33] C. Balzarek, T. J. R. Weakley, D. R. Tyler, J. Am. Chem. Soc. 2000, 122,
9427–9434.
[34] K. L. Breno, D. R. Tyler, Organometallics 2001, 20, 3864–3868.
[35] S. R. Klei, J. T. Golden, T. D. Tilley, R. G. Bergman, J. Am. Chem. Soc.
2002, 124, 2092–2093.
Acknowledgements
This work was supported by the MICIN of Spain (Grant project
RTI2018-100709-B-C21, FEDER funds), by Junta de Comunidades
de Castilla-La Mancha-FEDER (JCCM) (grant SBPLY/19/180501/
000260 and also by the Junta de Castilla y León (Project
Eur. J. Inorg. Chem. 2021, 1358–1372
www.eurjic.org
1371
© 2021 Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejic.202100022
[36] T. Nishioka, T. Shibata, I. Kinoshita, Organometallics 2007, 26, 1126–
[76] J. P. Lange, E. Van Der Heide, J. Van Buijtenen, R. Price, ChemSusChem
2012, 5, 150–166.
1128.
[37] P. L. Polavarapu, L. P. Fontana, H. E. Smith, J. Am. Chem. Soc. 1986, 108,
[77] Y. He, Y. Ding, C. Ma, J. Di, C. Jiang, A. Li, Green Chem. 2017, 19, 3844–
94–99.
3850.
[38] D. Klomp, T. Maschmeyer, U. Hanefeld, J. A. Peters, Chem. A Eur. J.
2004, 10, 2088–2093.
[39] M. Fujita, T. Hiyama, J. Org. Chem. 1988, 8, 5405–5415.
[40] M. Fujita, T. Hiyama, Tetrahedron Lett. 1987, 28, 2263–2264.
[41] N. Zhu, M. Su, W. M. Wan, Y. Li, H. Bao, Org. Lett. 2020, 22, 991–996.
[42] M. L. Clarke, G. J. Roff, Handbook of Homogeneous Hydrogenation.
Chapter 15. Wiley-VCH, Wiley-VCH, 2007.
[43] T. Kurita, F. Aoki, T. Mizumoto, T. Maejima, H. Esaki, T. Maegawa, Y.
Monguchi, H. Sajiki, Chem. A Eur. J. 2008, 14, 3371–3379.
[44] H. Esaki, R. Ohtaki, T. Maegawa, Y. Monguchi, H. Sajiki, J. Org. Chem.
2007, 72, 2143–2150.
[78] R. Karinen, K. Vilonen, M. Niemelä, ChemSusChem 2011, 4, 1002–1016.
[79] J. B. Binder, J. J. Blank, A. V. Cefali, R. T. Raines, ChemSusChem 2010, 3,
1268–1272.
[80] K. Yan, A. Chen, Energy 2013, 58, 357–363.
[81] H. Du, X. Ma, P. Yan, M. Jiang, Z. Zhao, Z. C. Zhang, Fuel Process.
Technol. 2019, 193, 221–231.
[82] R. M. Mironenko, V. P. Talsi, T. I. Gulyaeva, M. V. Trenikhin, O. B.
Belskaya, React. Kinet. Mech. Catal. 2019, 126, 811–827.
[83] D. E. Zhao, Z. Bababrik, R. Xue, W. Li, Y. Briggs, N. M. Nguyen, D.-T.
Nguyen, U. Crossley, S. P. Wang, S. Wang, B. Resasco, Nat. Catal. 2019,
2, 431–436.
[45] C. P. Lau, L. Cheng, J. Mol. Catal. 1993, 84, 39–50.
[46] A. Call, C. Casadevall, F. Acuña-Parés, A. Casitas, J. Lloret-Fillol, Chem.
Sci. 2017, 8, 4739–4749.
[47] D. Wang, D. Astruc, Chem. Rev. 2015, 115, 6621–6686.
[48] X. Wu, J. Liu, X. Li, A. Zanotti-Gerosa, F. Hancock, D. Vinci, J. Ruan, J.
Xiao, Angew. Chem. Int. Ed. 2006, 45, 6718–6722; Angew. Chem. 2006,
118, 6870–6874.
[49] Y. Himeda, N. Onozawa-Komatsuzaki, S. Miyazawa, H. Sugihara, T.
Hirose, K. Kasuga, Chem. A Eur. J. 2008, 14, 11076–11081.
[50] A. Bényei, F. Joó, J. Mol. Cat. 1990, 58, 161–163.
[51] A. N. Ajjou, J. L. Pinet, J. Mol. Catal. A 2004, 214, 203–206.
[52] F. Joó, A. Bényei, J. Organomet. Chem. 1989, 363, 19–21.
[53] M. C. Carrión, M. Ruiz-Castañeda, G. Espino, C. Aliende, L. Santos, A. M.
Rodríguez, B. R. Manzano, F. A. Jalón, A. Lledós, ACS Catal. 2014, 4,
1040–1053.
[84] T. Werpy, A. Petersen, J. Bozell, J. Holladay, A. White, D. Manheim, L.
Eliot, L. Lasure, S. Jones, U. S. Dep. Energy. 2004, 1–67.
[85] Y. Liu, L. Chen, T. Wang, Q. Zhang, C. Wang, J. Yan, L. Ma, ACS
Sustainable Chem. Eng. 2015, 3, 1745–1755.
[86] L. Vilcocq, A. Cabiac, C. Especel, S. Lacombe, D. Duprez, Catal. Today
2015, 242, 91–100.
[87] B. Kusserow, S. Schimpf, P. Claus, Adv. Synth. Catal. 2003, 345, 289–299.
[88] M. E. Ortiz, J. Bleckwedel, R. R. Raya, F. Mozzi, Appl. Microbiol.
Biotechnol. 2013, 97, 4713–4726.
[89] Barbieri G.; Barone C.; Bhagat A.; Caruso G.; Conley Z. R.; Parisi S, The
Influence of Chemistry on New Foods and Traditional Products.
Chapter 4. Sweet Compounds in Foods: Sugar Alcohols., Springer
International Publishing, Berlin, 2014.
[90] A. M. Ruppert, K. Weinberg, R. Palkovits, Angew. Chem. Int. Ed. 2012,
51, 2564–2601; Angew. Chem. 2012, 124, 2614–2654.
[54] W. H. Wang, J. F. Hull, J. T. Muckerman, E. Fujita, T. Hirose, Y. Himeda,
[91] Y. Kwon, M. T. M. Koper, ChemSusChem 2013, 6, 455–462.
[92] M. Silveira, R. Jonas, Appl. Microbiol. Biotechnol. 2002, 59, 400–408.
[93] M. M. Silveira, E. Wisbeck, C. Lemmel, G. Erzinger, J. P. Lopes Da Costa,
M. Bertasso, R. Jonas, J. Biotechnol. 1999, 75, 99–103.
Chem. A Eur. J. 2012, 18, 9397–9404.
[55] M. Ruiz-Castañeda, M. C. Carrión, L. Santos, B. R. Manzano, G. Espino,
F. A. Jalón, ChemCatChem 2018, 10, 5541–5550.
[56] B. J. Frost, C. A. Mebi, Organometallics 2004, 23, 5317–5323.
[57] C. A. Mebi, B. J. Frost, Organometallics 2005, 24, 2339–2346.
[58] M. Ruiz-Castañeda, A. M. Rodríguez, A. H. Aboo, B. R. Manzano, G.
Espino, J. Xiao, F. A. Jalón, Appl. Organomet. Chem. 2020, 34, e6003.
[59] J. Fidalgo, M. Ruiz-Castañeda, G. García-Herbosa, A. Carbayo, F. A.
Jalón, A. M. Rodríguez, B. R. Manzano, G. Espino, Inorg. Chem. 2018, 57,
14186–14198.
[60] Á. Molnár, A. Papp, Coord. Chem. Rev. 2017, 349, 1–65.
[61] Y. Yang, D. J. Miller, S. B. Hawthorne, J. Chem. Eng. Data 1997, 42, 908–
913.
[94] Benaglia M., Ed. Recoverable and Recyclable Catalysts, Wiley And Sons,
2009.
[95] C. Romain, S. Gaillard, M. K. Elmkaddem, L. Toupet, C. Fischmeister,
C. M. Thomas, J. L. Renaud, Organometallics 2010, 29, 1992–1995.
[96] L. Colina-Vegas, W. Villarreal, M. Navarro, C. R. De Oliveira, A. E.
Graminha, P. I. D. S. Maia, V. M. Deflon, A. G. Ferreira, M. R. Cominetti,
A. A. Batista, J. Inorg. Biochem. 2015, 153, 150–161.
[97] H. Hayashi, S. Ogo, T. Abura, S. Fukuzumi, J. Am. Chem. Soc. 2003, 125,
14266–14267.
[98] A. Comas-Vives, G. Ujaque, A. Lledós, Inner- and Outer-Sphere Hydro-
genation Mechanisms: A Computational Perspective, Elsevier Inc., 2010.
[99] O. Eisenstein, R. H. Crabtree, New J. Chem. 2013, 37, 21–27.
[100] W. J. Hehre, K. Ditchfield, J. A. Pople, J. Chem. Phys. 1972, 56, 2257–
2261.
[62] C. Leiva, H. Christine Lo, R. H. Fish, J. Organomet. Chem. 2010, 695,
145–150.
[63] T. M. Townsend, C. Kirby, A. Ruff, A. R. O’Connor, J. Organomet. Chem.
2017, 843, 7–13.
[64] B. Mostafa, S. M. Habibi-Khorassani, M. Shahraki, J. Phys. Org. Chem.
2017, 30, 1–11.
[65] S. Ogo, T. Abura, Y. Watanabe, Organometallics 2002, 21, 2964–2969.
[66] T. Abura, S. Ogo, Y. Watanabe, S. Fukuzumi, J. Am. Chem. Soc. 2003,
125, 4149–4154.
[67] C. Aliende, M. Pérez-Manrique, F. A. Jalón, B. R. Manzano, A. M.
Rodríguez, G. Espino, Organometallics 2012, 31, 6106–6123.
[68] X. Xu, L. Li, W. Han, J. Luo, D. Zhang, Y. Wang, G. Li, Catal. Commun.
2018, 109, 50–54.
[101] M. M. Francl, W. J. Pietro, W. J. Hehre, J. S. Binkley, M. S. Gordon, D. J.
DeFrees, J. A. Pople, J. Chem. Phys. 1982, 77, 3654–3665.
[102] D. Andrae, U. Häußermann, M. Dolg, H. Stoll, H. Preuß, Theor. Chim.
Acta 1990, 77, 123–141.
[103] A. Ehlers, A. W. Böhme, M. Dapprich, S. Gobbi, A. Höllwarth, V. Jonas,
K. F. Köhler, R. Stegmann, A. Veldkamp, G. Frenking, Chem. Phys. Lett.
1993, 208, 237–240.
[104] R. B. Sunoj, M. Anand, Phys. Chem. Chem. Phys. 2012, 14, 12715–12736.
[105] A. V. Marenich, C. J. Cramer, D. G. Truhlar, J. Phys. Chem. B 2009, 113,
6378–6396.
[69] Z. K. Gao, Y. C. Hong, Z. Hu, B. Q. Xu, Catal. Sci. Technol. 2017, 7, 4511–
4519.
[106] H. P. Hratchian, H. B. Schlegel, J. Chem. Phys. 2004, 120, 9918–9924.
[107] H. P. Hratchian, H. B. Schlegel, J. Chem. Theory Comput. 2005, 1, 61–69.
[70] N. Siddqui, B. Sarkar, C. Pendem, R. Khatun, L. N. S. Konthala, T. Sasaki,
A. Bordoloi, R. Bal, Catal. Sci. Technol. 2017, 7, 2828–2837.
[71] Y. Ma, Z. Li, Appl. Surf. Sci. 2018, 452, 279–285.
[72] S. Jiménez, J. A. López, M. A. Ciriano, C. Tejel, A. Martínez, R. A.
Sánchez-Delgado, Organometallics 2009, 28, 3193–3202.
[73] Á. Kathó, I. Szatmári, G. Papp, F. Joó, Chimia 2015, 69, 339–344.
[74] I. Szatmári, G. Papp, F. Joó, Á. Kathó, Catal. Today 2015, 247, 14–19.
[75] S. Xu, D. Pan, Y. Wu, X. Song, L. Gao, W. Li, L. Das, G. Xiao, Fuel Process.
Technol. 2018, 175, 90–96.
Manuscript received: January 12, 2021
Revised manuscript received: February 23, 2021
Accepted manuscript online: March 1, 2021
Eur. J. Inorg. Chem. 2021, 1358–1372
www.eurjic.org
1372
© 2021 Wiley-VCH GmbH