Triphosphines-containing ruthenium-acetato complexes: Synthesis, characterization, DFT, mer/fac isomerization and formic acid dehydrogenation

Article abstract of DOI:10.1016/j.molstruc.2019.127129

The triphosphines-containing ruthenium-acetato complexes [RuCl(κ²-OAc)(triphos)] (2) and the new analogue synthesized in our research group [RuCl(κ²-OAc)(etp)] (1) (triphos = 1,1,1-tris(diphenylphosphinomethyl)ethane or etp = bis[2-(

Full text of DOI:10.1016/j.molstruc.2019.127129

Journal Pre-proof
Triphosphines-containing ruthenium-acetato complexes: Synthesis, characterization,
DFT, mer/fac isomerization and formic acid dehydrogenation
Otávio Fuganti, Juliana P. da Silva, David F. Back, Márcio P. de Araujo
PII:
S0022-2860(19)31238-4
DOI:
https://doi.org/10.1016/j.molstruc.2019.127129
MOLSTR 127129
Reference:
To appear in: Journal of Molecular Structure
Received Date: 21 June 2019
Revised Date: 17 September 2019
Accepted Date: 25 September 2019
Please cite this article as: Otá. Fuganti, J.P. da Silva, D.F. Back, Má.P. de Araujo, Triphosphines-
containing ruthenium-acetato complexes: Synthesis, characterization, DFT, mer/fac isomerization
and formic acid dehydrogenation, Journal of Molecular Structure (2019), doi: https://doi.org/10.1016/
j.molstruc.2019.127129.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. 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.
©
2019 Published by Elsevier B.V.

th
September., 17 2019.
COVER LETTER
Dear Professor,
Please, following this cover letter, we are submitting the REVISED version of
the manuscript entitled “Triphosphines-containing ruthenium-acetato complexes:
Synthesis, characterization, DFT, mer/fac isomerization and formic acid
dehydrogenation” to be considered for publication in Journal of Molecular Structure.
The manuscript was revised according to the suggestions of the reviewers,
except for two points. The point-by-point list of modifications is presented below.
Reviewer #1:
Abstract, line 55: accomplished should be better replaced with checked.
Answer: corrected
Introduction, line 22: are should be changed to is.
Answer: corrected
Introduction, line 32: has high density.
Answer: corrected
Introduction, line 51: there is only one example of using...
Answer: corrected
Isolation of mer-[RuCl...], line 32: the filtered solid?
Answer: corrected
Results and Discussion, line 29: signals compatible with... were observed.
Answer: corrected
Page 13, line 34: the higher HOMO-LUMO gap is the lower the reactivity is.
Answer: corrected
Page 15, line 1: In addition, this also supports...
Answer: corrected
Scheme 2: may be, would be better to show energy differences in kcal/mol
Answer: corrected. Additionally the values were also updated in the text.

Figure 4: better to try to plot MOs with the same orientation of the molecular
frameworks.
Answer: It was tried, but the actual representation of the MO’s allows the best view of
the surfaces.
Conclusion, line 38: mainly.
Answer: corrected
Also, ref. 31 should be corrected.
Answer: corrected
Reviewer #2:
In the EXPERIMENTAL SECTION I miss a short description of the method of
identification of obtained Rh complexes.
Answer: In the experimental section (Materials and Methods) of the equipments and
conditions used for the characterization, as observed in other papers (published in J.
Mol. Struct.)
There is also a lack of information why MeOH, THF and PrOH were chosen for the
tests.
Answer: The updated sentence is as follow.
The reactions were carried out in a glass autoclave with a total volume of 194 mL. The
i
desired degassed solvent, MeOH, THF or PrOH, which are solvents commonly used in
catalysis, …
In my opinion you should also attempts to explain why the best catalytic activity was
observed for the reaction in THF.
Answer: The sentence below was included;
i
…
which can tentatively attributed to the coordination ability of the MeOH and PrOH
competing with formate to the coordination site around the ruthenium center more
strongly than the THF. Additionally, the acetato (or formate) coordination mode can
2
1
change from k to k [refs] providing a vacant site and stabilization of the alcohol
coordination by hydrogen bond with the acetato or formate. For the later, the hydrogen
bond could inhibit the β-hydride migration
I noticed two editorial errors: Page 6, Line 14 - superscript 1H NMR.
Answer: corrected
Page 18, Line 3 - PrOH instead prOH. Best regards
Answer: corrected

Reviewer #3:
Tables of the observed wavenumbers or figures of the vibrational spectra of 1 and 1"
should be included in the supplementary data.
Answer: vibrational spectra were included in the SI.
Atomic coordinates of the optimized structures of the complexes should be included in
the supplementary data.
Answer: Table containing atomic coordinates was included in the SI.
p.9 L51 6-31G++(3df,3pd) -> 6-31++G(3df,3pd)
Answer: corrected.
DFT Comparison between the experimental molecular structure and the theoretical ones
should be added in this journal's context.
Answer: The following content was included;
The geometry optimizations were performed according to details described in
the experimental section, and the computed metric parameters are summarized in Table
S1 (Supporting Information).
The computed bond lengths show good agreement with the experimental ones,
for example, the experimental Ru-Cl, Ru-O1 and Ru-O2 bond lengths are, respectively,
2
.4553, 2.209, and 2.186 Å, while the computed ones are 2.4596, 2.2300 and 2.2196 Å.
The same agreement is observed for Ru-P bond lengths. In the same way, the calculated
bond angles also have good agreement with the experimental ones.
Figure 3 [RuCl(<kappa>2-OAc)etp] (2) -> [RuCl(<kappa>2-OAc)etp] (1)
Answer: corrected.
Figure 4 Figures of 1 and 1' isomers are completely the same for both LUMO and
HOMO. Are they correct?
Answer: corrected.
References [31] What are the #xe9?
Answer: corrected.
Márcio Peres de Araujo

TITLE
Triphosphines-containing ruthenium-acetato complexes: Synthesis,
characterization, DFT, mer/fac isomerization and formic acid
dehydrogenation

“
REVISED”
Triphosphines-containing ruthenium-acetato complexes: Synthesis,
characterization, DFT, mer/fac isomerization and formic acid
dehydrogenation
a
b
c
a
Otávio Fuganti , Juliana P. da Silva , David F. Back and Márcio P. de Araujo *
a
Departamento de Química, Universidade Federal do Paraná, Centro Politécnico, CP
1
9081, CEP 81531-980 Curitiba (PR), Brazil
b
Departamento de Química, Universidade Federal de Santa Catarina, Florianópolis, SC
8040-900, Brazil
8
c
Departamento de Química, Universidade Federal de Santa Maria, CEP 97105-900
Santa Maria (RS), Brazil
Keywords: ruthenium; triphosphines; acetate; DFT; cyclic voltammetry; formic acid
decomposition
ABSTRACT
2
The triphosphines-containing ruthenium-acetato complexes [RuCl(κ -OAc)(triphos)] (2)
2
and the new analogue synthesized in our research group [RuCl(κ -OAc)(etp)] (1)
(triphos
=
1,1,1-tris(diphenylphosphinomethyl)ethane or etp
=
bis[2-
(diphenylphosphino)ethyl]phenylphosphine) were evaluated as pre-catalyst for the
2
dehydrogenation of formic acid (FA). Complex [RuCl(κ -OAc)(etp)] (1) was fully
1
13
31
1
characterized by H, C and P{ H} NMR, vibrational spectroscopy, elemental
analysis and X-ray diffraction. The electrochemical behavior of 1 and 2 was checked by
cyclic voltammetry. DFT calculations for 1, 2 and for the two isomers of 1 (1’ and 1”)
were performed and discussed. Due to the linear nature of etp ligand, three isomeric

2
forms for [RuCl(κ -OAc)(etp)] (1’ and 1”) are possible and were observed. Both
complexes were active for the decomposition of FA, in three solvents (MeOH, THF and
i
PrOH), at 80 °C, leading solely to formation of CO and H . In THF both complexes
2
2
-
achieved quantitative conversions (in times up to 125 min) and final TOF’s up to 833 h
1
,
for complex 1.
INTRODUCTION
The search for renewable sources that can meet the growing worldwide demand
for low cost and clean energy is essential. The molecular hydrogen is regarded as an
alternative fuel, because its combustion produces solely water. The development of new
ways for both, generation and storage of molecular hydrogen is one of the main goals of
the chemists nowadays. The use of organic compounds as "chemical reservoirs" has
received great attention, due to the low risks in their storage, when compared to other
methods [1]. In this way, formic acid (FA) has received considerable attention as
storage/source of molecular hydrogen. FA is liquid at room temperature, has high
-1
density (1.22 g mL ) and can be obtained catalytically, using transition metal
compounds, from CO and H [2], and the right choice of the catalyst for the
2
2
decomposition of FA can yield selectively H with CO as the only by-product [3].
2
2
Historically, triphosphine ligands are less explored in coordination chemistry
than mono and diphosphines, but in recent years the use of ruthenium triphosphine
complexes gained attention due to the works reporting, especially, the CO activation
2
and hydrogen-borrowing reactions catalyzed by such complexes [4-11]. On the other
hand, there is only one example of using such complexes for the dehydrogenation of
formic acid, with good performance and selectivity for H production [12-14]. Beller
2
and co-workers have reported the use of both, an in situ formed from the precursor

[
Ru(acac) ] and triphos ligand (triphos = 1,1,1-tris(diphenylethylphosphino) ethane) or
3
-1
the preformed [Ru(triphos)(CH CN) ](OTf) , achieving TOF values up to 1000 h [12,
3
3
2
1
4]. Recently, Gabor and co-workers carried out studies using an in situ formed Ru-
triphos with the Lewis acid (Al(OTf) ) gaining robustness, allowing up to 14 recycling
3
-1
without significant loss of activity and TOF values up to 1920 h [13].
The two ligands used in the present work 1,1,1-tris(diphenylethylphosphino)
ethane (triphos) and bis[2-(diphenylphosphino)ethyl]phenylphosphine (etp) probably
are the two simplest triphosphines commonly encountered in literature. Due to the
terminal position of the three P-atoms, the former binds to the metal center in a facial
fashion, while the latter is linear with two terminal P-atoms and one central, which
allows the coordination in both facial and meridional. Additionally, the central P-atom
differs from the terminal in terms of electronic and steric nature.
2
In this work we report the synthesis and full characterization of [RuCl(κ -
OAc)(etp)] (1), including the molecular structure determined by X-ray diffraction. The
2
complexes
[RuCl(κ -OAc)(P-P-P)]
(P-P-P
etp
=
bis[2-
1,1,1-
diphenylphosphinoethyl]phenylphosphine)
–
–
1
and
tris[diphenylphosphinomethyl]ethane – triphos – 2) were tested in catalytic
dehydrogenation of formic acid (FA).
EXPERIMENTAL SECTION
Materials and Methods
All syntheses were performed under argon atmosphere using standard Schlenk
techniques. All solvents were dried and distilled according to described methods and
saturated with argon before use. RuCl ·xH O was obtained on loan from Johnson
3
2
2
Matthey and used as received. [RuCl (PPh ) ] [15], [RuCl(κ -OAc)(PPh ) ] [16] and
2
3 3
3 3

2
[
RuCl(κ -OAc)(triphos)] (2) [17] were prepared as described in literature. All other
chemicals were purchased from Strem Chemicals (triphos, etp) or Sigma-Aldrich
sodium formate). Analytical samples and single crystals suitable for X-ray diffraction
(
were obtained from dichloromethane solutions layered with hexane at r.t.. Elemental
analyses were performed with a Thermo Scientific CHNS-O FLASH 2000 micro
analyzer coupled with an ultra-microbalance Mettler Toledo Model XP6. Infrared
-1
spectra (4000 to 400 cm ) were recorded from KBr pellets on BIO-RAD 3500 GX
–
1
equipment with spectral resolution of 1 cm . NMR spectra were recorded at room
1
temperature using a Bruker 14 Tesla spectrometer (frequencies: 600.00 MHz for H,
3
1
13
1
13
2
41.00 MHz for P and 150.00 MHz for C). H and C NMR spectra were referenced
3
1
to internal tetramethylsilane and P chemical shifts were measured relative to 85%
phosphoric acid as external reference. Cyclic voltammetry (CV) experiments were
performed on an Ivium Technologies–CompactStat potentiostat/galvanostat at room
–
1
+
–
temperature, in CH Cl solutions containing 0.100 mol L Bu N ClO (TBAP).
2
2
4
4
Working and auxiliary electrodes consisted of stationary Pt foils, while the reference
electrode was Ag/AgCl, in a Luggin capillary probe.
DFT calculations
All the calculations were performed at the DFT level in the gas phase as
implemented in the Gaussian 03 package [18]. The Perdew and Wang functional
(PW91PW91) was used due to the good results obtained by Raju et al [19] for
organometallic compounds. The DGauss DZVP [20, 21] basis set were employed for
the ruthenium atom with two additional polarization f function with the exponents
1
.9472 and 0.7489, these basis set are at the form (18s12p9d2f/[6s5p3d2f) [22]. For the
other atoms the 6-31++G(3df,3pd) was employed. The Hessian was computed to

confirm the stationary points of the potential energy surfaces (PES) as true minima. The
molecular structures were depicted by the program MOLDEN [23, 24]. Molecular
orbitals were drawn with the use of program Jmol [25]. GaussSum 3.0 [26] program
was used for population analysis.
X-Ray diffraction analysis
Data were collected on a Bruker APEX II CCD area-detector diffractometer,
employing graphite monochromatized Mo-K radiation. The structure was solved by
α
direct methods using SHELXS-97 [27]. Consecutive Fourier-difference map analyses
yielded the positions of the non-hydrogen atoms. Refinements were carried out with the
2
SHELXL-97 package [28] by full-matrix least-squares on F , with anisotropic
displacement parameters for all non-hydrogen atoms. Hydrogen atoms were included in
calculated positions. Crystal data and additional information on data collection and
refinement are contained in Table 1.
Table 1. Crystallographic Data for complex 1
empirical formula
Fw
C H ClO P Ru
36 36 2 3
730.08
cryst system
space group
a (Å)
monoclinic
P2 /n
1
8.7222(3)
17.4350(7)
21.8259(8)
90
b (Å)
c (Å)
α = γ (deg)
β (deg)
99.937(2)

3
V (Å )
3269.3(2)
4
formula units/ cell
temp, K
290
λ(Mo-K ), Å
0.71073
1.483
0.0315
0.0815
α
-3
ρ , g cm
calc
R1
wR2
PREPARATION OF THE COMPLEXES
2
The fac and mer designation for [RuCl(κ -OAc)(etp)] refers to the relative
position of the three P-atoms in the vertices of the octahedron, and the -Cl or -O in the
designation fac-Cl or fac-O refers to the atom (Cl or O) trans positioned to the P-central
of the etp ligand.
2
fac-Cl-[RuCl(κ -OAc)(etp)] (1).
2
A suspension of [RuCl(κ -OAc)(PPh ) ] (1.463 g, 1.489 mmol) and etp (816 mg, 1.527
3
3
mmol) in toluene (30 mL) was heated under reflux for 2.5 h. After cooling to room
temperature, the resulting yellow suspension was filtered, washed with hexane (3 x 10
mL) and diethyl ether (3 x 10 mL) and dried under reduced pressure. Yield: 915 mg
–
1
(
84%). Anal. Calcd for C H ClO P Ru (730.11 g mol ): C, 59.22; H, 4.97. Found: C,
36 36 2 3
1
5
9.58; H, 5.15. H NMR (CD Cl , 600 MHz): δ/ppm = 7.88-6.45 (m, 25H, -Ph), 2.91
2 2
(
m, 2H, -CH ), 2.51 (m, 2H, -CH ), 2.06 (m, 2H, -CH ), 1.75 (m, 2H, -CH ), 1.37 (m,
2 2 2 2
13
3
H, -CH ). C NMR (CD Cl , 150 MHz): δ/ppm = 186.4 (-OOCCH ), 135.4 – 128.2 (-
3 2 2 3
3
1
1
Ph), 27.6 (-CH ), 27.4 (-CH ), 25.0 (-CH ). P{ H} NMR (CD Cl , 241 MHz): δ/ppm
2
2
3
2
2

2
2
=
100.0 (t, J = 20.9 Hz, 1P, -PPh), 74.2 (d, J = 20.9 Hz, 2P, -PPh ). FTIR (KBr)
PP PP 2
-1
ν/cm : νa(COO) = 1531, νs(COO) = 1460, ∆(νa
- νs) = 71, ν Ru-P = 540 and 521. Note: A
small amount of other possible fac-isomer (named as fac-O – 1’) (~6%) was observed in
3
1
1
solution by P{ H} NMR.
Isolation of mer-[RuCl(OAc)etp] (1”)
The filtered solution of the reaction for preparation of 1 was left in freezer for
2
4h. The solid obtained was filtered off and washed with hexane (2 x 3 mL). Then the
solid was dissolved in diethyl ether and passed through a silica columm and the solution
was left again in freezer for 24h. After that, the orange crystals formed was isolated by
filtration, washed with a small amount of cold diethyl ether and dried under reduced
1
pressure for 24h. Yield: 17.4 mg (1.6%). H NMR (CDCl , 600 MHz): δ/ppm = 8.17-
3
3
1
1
6
.90 (m, 25H, -Ph), 3.27-2.32 (m, 8H, - CH ), 1.31 (m, 3H, -CH ). P{ H} NMR
2 3
2
(
CDCl , 150 MHz): δ/ppm = 126.9 ppm (t, J = 13.05 Hz, 1P, -PPh), 53,6 (d, J =
3 PP PP
-1
1
3.05 Hz, 2P, -PPh). FTIR (KBr) ν/cm : ν
= 1451, νs(COO) = 1527, ∆(ν - ν ) = 76.
s(COO) a s
Dehydrogenation of formic acid (FA)
The reactions were carried out in a glass autoclave with a total volume of 194
i
mL. The desired degassed solvent, MeOH, THF or PrOH, which are solvents
commonly used in catalysis, (20 mL) was added in the glass vessel followed by addition
of formic acid 80 % (1.62 mL, 34.8 mmol), sodium formate (237 mg, 3.48 mmol) and
the catalyst (34.8 mmol, 25.4 mg for 1 and 28.5 for 2). The system was purged with
argon (~15 min) and closed under ambient pressure and then, the glass-vessel was
placed in an oil bath at 80°C. After cooling and venting the system for releasing of the
1
dissolved gases the solution was analyzed by H NMR. For the recycling experiments,

formic acid was added (1.62 mL, 34.8 mmol), the system was purged with argon, closed
under ambient pressure and placed back in the oil bath at 80°C.
RESULTS AND DISCUSSION
Synthesis and Basic characterization
2
The
complex
fac-Cl-[RuCl(κ -OAc)(etp)]
(1)
(etp
=
bis[2-
diphenylphosphinoethyl]phenylphosphine) was synthesized according to the procedure
2
reported by Dyson et al. for the synthesis of the analogue [RuCl(κ -OAc)(triphos)] (2)
[
17]. The new complex was fully characterized, including the molecular structure
solved by X-ray crystallography.
2
The reaction of [RuCl(κ -OAc)(PPh ) ] with a slightly excess of etp ligand in
3
3
3
1
1
toluene under reflux yielded the complex 1 (isomer fac-Cl). In the P{ H} NMR
spectrum, in addition to the signals of fac-Cl, signals compatible with the fac-O-
2
[
RuCl(κ -OAc)(etp)] (1’) were observed, with fac-Cl:fac-O molar ratio of 15:1.
2
Furthermore, the third isomer mer-[RuCl(κ -OAc)(etp)] (1”) was isolated from the
filtrate (yield of 1.6%) obtained after separation of 1.
2
The proposed formulation ([RuCl(κ -OAc)(etp)]) has three possible isomeric
forms, two of them with the arrangement of the three phosphorus atoms in a facial
fashion (fac-Cl and fac-O) and one meridional. The fac-Cl has two terminal phosphorus
atoms trans positioned to the oxygen atoms of the acetato, while for the fac-O the
terminal phosphorus atoms are trans to both oxygen of the acetato and chlorido (See
Scheme 1, for the structures of the three possible isomers).

2
Scheme 1. The three possible isomeric forms for [RuCl(κ -OAc)(etp)].
1
In the H NMR spectrum of 1 the signals of the aromatic hydrogen nuclei of etp
are observed as multiplets in the region 7.9 – 6.4 ppm. The aliphatic hydrogen nuclei of
etp are observed as four multiplets at 2.9, 2.5, 2.1 and 1.8 ppm (−CH −). The –CH of
2
3
the acetato ligand is observed at 1.4 ppm. The integration of the signals is compatible
with the ratio acetato/etp = 1.
1
3
The C NMR spectrum of 1 revealed the signals of carboxylate of acetate at
86.4 ppm, the aromatic carbon nuclei of phenyl rings of the etp in the region 135.4 –
28.2 ppm, the methylenic carbon nuclei of etp at 27.6 ppm and 27.4 ppm, and the
1
1
methyl carbon atom of acetato ligand at 25.0 ppm.
3
1
1
The P{ H} spectrum of 1 (Figure 1) presents one triplet centered at 100.0.
ppm attributed to the central phosphorus nucleus and one doublet at 74.2 ppm attributed
to the two chemically equivalent terminal phosphorus nuclei, and the coupling constant
2
(
JP-P) of 20.9 Hz compatible with cis-configuration of the three phosphorus atoms.
Additionally, three unresolved double of doublets were observed at 118.75, 84.98 and
8.28 ppm attributed to the isomer fac-O (1’), which are compatible with the proposed
structure, due to the non-equivalence of the three P-nuclei (Scheme 1).
7

3
1
1
Figure 1. P{ H} NMR spectrum of 1 and 1’.
The vibrational spectrum of 1 revealed the antisymmetric and symmetric ν(COC)
-1
-1
at 1531 and 1460 cm , respectively, with ∆(ν – ν ) = 71 cm suggesting the chelated
a
s
fashion of the acetate coordination [29].
The isomer mer- (1’’), which was isolated in low yield from the filtrate of the
1
31
1
synthesis of 1, was also characterized by H, and P{ H} NMR, and FTIR.
1
In the H NMR spectrum of 1” the aromatic hydrogen nuclei of etp are observed
as multiplets in the region 8.17 – 6.90 ppm, while and the aliphatic ones in 3.27 – 2.32
ppm (−CH −). The hydrogen nuclei of methyl group of the acetate ligand are observed
2
at 1.31 ppm.
3
1
1
The P{ H} spectrum of 1” revealed, as expected the presence one triplet
centred at 126.9 ppm attribute to the central phosphorus and one doublet at 53.6 ppm
2
doublet attributed to the terminal phosphorus, and coupling constant ( J ) of 13.0 Hz.
P-P

The vibrational spectrum of 1” revealed the antisymmetric and symmetric ν(COC)
-1
-1
at 1527 and 1451 cm , with ∆(ν – ν ) = 76 cm , which is also in agreement with the
a
s
chelated fashion of the acetate coordination [29].
3
1
1
During the P{ H} NMR experiments for 1” at r.t. was observed the appearance
3
1
1
of other signals within 1 day in solution (see Figure 2, for the P{ H} NMR spectra for
2
the isomerization). The new signals, one triplet at 100.0 ppm and one doublet at 74.2 ( JP-P
=
2
0.9 Hz), are attributed to the formation of complex 1 (isomer fac-Cl) through
isomerization process mer (1”) ꢀ fac-Cl (1). After two days in solution the third isomer
1’, fac-O) begins to be observed. After six days a very small amount of 1” is still
(
observed. On the other hand, the spectrum is dominated by the presence o 1 and a small
amount of 1’ with fac-Cl:fac-O molar ratio of ~15:1. This ratio is very close to the
3
1
1
observed in the P{ H} NMR spectrum of 1 (and is constant in all independent
syntheses of 1) and may be indicative that fac-Cl is in equilibrium with the fac-O.
The HOMO-LUMO gap (see DFT section below) can be related to the reactivity
of the complexes, the higher HOMO-LUMO gap is the lower the reactivity is,
accordingly to works published [30]. In this way, the higher conversion of 1” to 1 may
be related to the lower gap in 1”.

3
1
1
Figure 2. Isomerization of 1” to 1 (and 1’) followed by P{ H} NMR.
Molecular structure of 1
The molecular structure of 1 (see Figure 3, for the structure, and caption for
selected bond length and angles) revealed the connectivity of the donor atoms around
3
1
1
the ruthenium center to be equivalent to the observed in solution, supported by P{ H}
NMR. The triphosphine etp ligand is coordinated in the facial fashion, with the two
terminal P-atoms trans positioned to the acetato, and the central one trans to the
chlorido ligand.
The Ru-P1 and Ru-P3 bond lengths (trans to acetato) are 2.2463(7) Å and
2
.2553(6) Å, respectively, are longer than Ru-P2 (2.2348(6) Å – trans to chlorido). All
Ru-P bond lengths for 1 are shorter than for 2 (2.277(3) Å and 2.282(2) Å trans to
acetato, and 2.275(2) Å, trans to chlorido) [31] and may be a result of the more flexible
nature of etp compared to triphos.
The more flexible nature of etp is reflected by the P1-Ru-P3 bond angle (P-
terminals), 95.06(2)°, compared to the observed for triphos in 2 (86.28(8)°, 89.11(9)°
and 88.37(8)°) [31].

2
Figure 3. ORTEP [32] representation of [RuCl(κ -OAc)etp] (1), thermal ellipsoids
drawn at 30% probability. The hydrogen atoms were omitted for clarity. Bond lenghts
(Å): Ru–P1 2.2463(7), Ru–P2 2.2348(6), Ru–P3 2.2553(6), Ru–O1 2.2095(16), Ru–O2
2
.1863(16), Ru–Cl 2.4553(7), C35–O1 1.258(3), C35–O2 1.259(3). Bond angles (°):
O1–Ru–O2 59.43(6), P1–Ru–P2 84.50(2), P2–Ru–P3 83.80(2), P1–Ru–P3 95.06(2),
P2–Ru–Cl 175.76(3), P2–Ru–O1 91.24(5), P2–Ru–O2 95.67(5).

DFT
The geometry optimizations were performed according to details described in
the experimental section, and the computed metric parameters are summarized in Table
S1 (Supporting Information).
The computed bond lengths show good agreement with the experimental ones,
for example, the experimental Ru-Cl, Ru-O1 and Ru-O2 bond lengths are, respectively,
2
.4553, 2.209, and 2.186 Å, while the computed ones are 2.4596, 2.2300 and 2.2196 Å.
The same agreement is observed for Ru-P bond lengths. In the same way, the calculated
bond angles also have good agreement with the experimental ones.
The relative free Gibbs energy values for the isomers (fac-Cl , fac-O and mer)
are presented in Scheme 2. As can be seen, the isomer fac-Cl (1) is more stable (0 kcal
-1
-1
mol ), while the other fac isomer, named fac-O (1’) is 2.15 kcal mol higher. In these
two fac isomers the differing point is the ligand trans-positioned (Cl or O) to the P-
central atom of the etp. The mer isomer (1”) is the lesser stable compared to the other
-1
two isomers, with free Gibbs energy 11.30 kcal mol higher compared to the isomer 1.
These results corroborate the isolation of 1 in high yield, while 1” was isolated in less
than 2% of yield. In addition, this also supports the isomerization of 1” ꢀ 1 at r.t., and
the low energy difference between 1 and 1’ may account for the possible equilibrium in
solution.

2
Scheme 2. Relative Gibbs free energy for the three possible isomers of [RuCl(κ -
OAc)(etp)] (1, 1’ and 1”).
The molecular orbitals composition (Table 2) was calculated by DFT and
revealed the predominance of the ruthenium atomic orbital with expressive contribution
of the chlorido ligand in the HOMO (complex 1: Ru – 51% / Cl – 29%; complex 1’: Ru
–
3
52% / Cl – 30%; complex 1” Ru – 52% / Cl – 27% and complex 2: Ru – 47% / Cl –
2%) (See Figure 4 for the isosurfaces of HOMO and LUMO).
The complex 2 presents the highest HOMO-LUMO gap (2.61 eV) close to the
calculated for 1 (2.56 eV). The isomer 1’ has the HOMO-LUMO gap slightly lower
than 1 (2.52 eV) while the mer-isomer (1”) has the lowest (2.34 eV).

Table 2. Frontier molecular orbitals data for complexes 1 and 2
E/eV %Ru %Cl %OAc %P-P-P
∆
HOMO/LUMO
/
eV
E /V
1/2
LUMO
HOMO
LUMO
HOMO
LUMO
HOMO
LUMO
HOMO
-1.79
-4.35
-1.79
-4.31
-1.67
-4.01
-1.79
-4.40
20
51
24
52
24
52
20
47
2
29
2
30
5
27
1
32
6
4
6
3
5
5
5
3
73
16
68
14
66
15
75
18
1
2.56
1.0
-
1
’
2.52
2.34
2.61
1
”
-
2
1.2

1
1’
1”
2
LUMO
HOMO
Figure 4. Isosurfaces of the frontier orbitals (HOMO and LUMO) of complexes 1, 1’, 1” and 2.

Cyclic Voltammetry
The cyclic voltammetry experiments of 1 and 2 were carried out in CH Cl
2
2
-
1
-1
-1
[
complexes] = 1 mmol L and TBAP (10 mol L ) at 25 °C, conditions in which
ferrocene has E1/2 = 0.58 V.
Both complexes presented one reversible and mono-electronic oxidation process
2
+
3+
attributed to the process Ru ꢀRu . The lower E value observed for 1 (1.0 V) than
1
/2
for 2 (1.2 V) is compatible with the structures of the two triphosphine ligands. The
presence of the central P-atom linked to two alkyl groups in the etp provides a more
basic ligand than triphos.
As can be seen from DFT results, the oxidation process is basically centered at
2
+
3+
the unit “Ru –Cl”, but it is expected the formation of “Ru –Cl” unit after the oxidation
process. Additionally, from the information presented in Table 2 (DFT section), the
nature of triphosphine ligands (basicity) has affected the HOMO, the lower oxidation
potential for complex 1 compared to complex 2 is supported by DFT, once the energy
of the HOMO for complex 2 is 0.05 eV lower than for 1.
Decomposition of HCOOH in CO and H
2
2
The complexes 1 and 2 were tested as pre-catalysts for the decomposition of
i
formic acid (FA) 80% in MeOH, THF and PrOH, at 80°C using the
catalyst:formate:FA molar ratio 1:100:1000. The reaction was accompanied by the
variation of the pressure as a function of time. Analysis of the gas-phase composition by
GC-TCD revealed that carbon monoxide is not formed by the concurrent dehydration
reaction of FA during the process.

The catalytic performance of the complexes was evaluated in three solvents
i
(
MeOH, THF and PrOH) (Figures 5A and 5B). As can be seen both complexes are
active for decomposition of formic acid (FA) in H and CO , but for both the best
2
2
activity is achieved when the reaction was carried out in THF. The decomposition of FA
by complexes 1 and 2 are quantitative in THF, in times up to 200 min, as evidenced by
1
H NMR analysis of the solution after pressure stabilization. It is observed equilibration
of the reactions involving the gas-phase and the gases dissolved in the solution. After
cooling the system and releasing the pressure of the vessel, upon stirring, effervescence
of the solution was observed due to the releasing of the gases dissolved.
i
The pressure increasing rate in THF, MeOH and PrOH are, respectively, 0.0152,
-
1
-1
0
.003 and 0.0015 MPa min , for 1, while for 2, 0.0078, 0.0038 and 0.0032 MPa min
Figures 5A and 5B). The pressure increasing rates are ~5 times faster in THF than in
MeOH for 1 and ~2 times for 2.
(
0.8
1.0
0.8
0.6
0.4
0.2
0.0
MeOH
THF
0.6
i
PrOH
0
.4
.2
0
MeOH
THF
iPrOH
0.0
0
25
50
75
100
125
25
50
75
100
125
Time/min
Time/min
(A)
(B)
2
Figure 5. Solvent dependence of the activity for the complexes 1 and 2: (A) [RuCl(κ -
2
OAc)(etp)] (1) and (B) [RuCl(κ -OAc)(triphos)] (2).
The best activity for both complexes was observed when the reaction was
carried out in THF (Figure 6A), which can tentatively attributed to the coordination

i
ability of the MeOH and PrOH competing with formate to the coordination site around
the ruthenium center more strongly than the THF. Additionally, the acetato (or formate)
2
1
coordination mode can change from k to k [33, 34] providing a vacant site and
stabilization of the alcohol coordination by hydrogen bond with the acetato or formate.
For the later, the hydrogen bond could inhibit the β-hydride migration. The activity of
complex 1 is higher than 2 in THF with pressure increasing rate ~2 times higher for 1
(Figure 6B).
-1
The complex 1 presents final TOF of 833 h (calculated for 100% conversion)
-1
for the first run, while complex 2, 308 h . For the subsequent two runs (only for
-1
complex 1), the TOF values decreased to 370 and 250 h , but still achieving 100% of
conversion, respectively (Table 3).
The higher activity of complex 1 may be speculatively attributed to the higher
electron density provided by the etp ligand to the metal center, as expected by its
structure and confirmed by the cyclic voltammetry experiments. It has been previously
discussed the effect of the electron density affect positively the activity of catalyst,
corroborating our observation [35-37]. In addition, the lower HOMO/LUMO gap for 1
also may account for its higher activity for the decomposition of FA.
Table 3. Dehydrogenation of FA data for 100 % of conversion in THF.
-1
Time/h
TOF/h
run
Complex
run
TON
st nd rd
st nd rd
1
/2 /3
1.2 / 2.7 / 4.0
3.2 / - / -
1 /2 /3
1
2
1000
1000
833 / 370 / 250
308/ - / -

Finally, the complex 1 were evaluated for recyclability and two recycles were
carried out (Figure 6B) and in the three runs performed, quantitative conversions were
observed in all cycles, but the activity decreases in each run.
1
.0
1.0
0.8
0.6
0.4
0.2
0.0
0.8
0.6
0.4
0.2
etp
tdpme
first run
second run
third run
0.0
0
20
40
60
80
100
120
140
160
0
50
100
150
200
250
300
Time/min
Time/min
(A)
(B)
Figure 6. Comparison of activity of complexes 1 (black square) and 2 (red circle) in
st
nd
THF (A), and recycling evaluation for 1 in THF (1 run – black squares; 2 run – red
rd
circles; 3 run - blue triangles) (B).
Conclusions
2
The complex fac-[RuCl(κ -OAc)(etp)] was prepared by established route and
fully characterized, including the molecular structure determined by X-ray diffraction.
The molecular structure of 1 was discussed and compared to the published structure of
2
. The nature of triphosphine ligands influenced mainly the energy of HOMO orbital
and the ∆HOMO/LUMO. Three isomeric forms were observed (two were isolated) for the
complex with etp (1, 1’ and 1”), and the isomer 1 was isolated in high yield. The
stability order 1 > 1’ > 1”, obtained by DFT level theory, can be used to explain the
predominance of isomer 1 and the suggested equilibrium between 1 and 1’. Cyclic

2
+
3+
voltammetry studies revealed the presence of one reversible Ru ꢀRu oxidation
process with lower E1/2 value for complex 1, which may confirm the more basic nature
of the etp ligand compared to the triphos, and it is compatible with the higher energy of
HOMO for complex 1. Both complexes are active for the decomposition of formic acid,
selectively producing H and CO . Although, both complexes are active for the
2
2
decomposition of FA, the nature of the triphosphine ligand seems to play an important
role (basicity and/or steric hindrance) in the activity of the complexes. The effect of the
triphosphine was observed in the frontier orbitals and the lower HOMO/LUMO gap for
1
may account for its higher reactivity/activity for the decomposition of FA.
Acknowledgments
The authors are grateful to CNPq for financial support, and to Johnson Matthey plc for
the loan of RuCl ·xH O (to M.P.A.). This study was also financed in part by the
3
2
Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) -
Finance Code 001.
Supplementary material
Supplementary material contains vibrational spectra of 1 and 1’’, calculated and
experimental bond lengths (Å) and angles (°) for complexes 1 and 2 and atomic
coordinates of the optimized structures of the complexes 1, 1’, 1’’ and 2.
Crystallographic data for 1 has been deposited with the Cambridge Crystallographic
Data Centre with CCDC number 1852739. The data can be obtained free of charge, via
www.ccdc.cam.ac.uk/conts/retrieving.html (or from the Cambridge Crystallographic
Data Centre, CCDC, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033;
or e-mail: deposit@ccdc.cam.ac.uk).

Competing interests
No competing interests to declare
References
[
1] C. Fellay, P.J. Dyson, G. Laurenczy, A viable hydrogen-storage system based on
selective formic acid decomposition with a ruthenium catalyst, Angewandte Chemie
International Edition, 47 (2008) 3966 –3968.
[
2] S. Fukuzumi, T. Suenobu, Hydrogen storage and evolution catalysed by metal
hydride complexes, Dalton Transactions, 42 (2013) 18-28.
3] Y. Himeda, Highly efficient hydrogen evolution by decomposition of formic acid
using an iridium catalyst with 4,4'-dihydroxy-2,2-bipyridine, Green Chemistry, 11
[
(2009) 2018–2022.
[
4] K. Thenert, K. Beydoun, J. Wiesenthal, W. Leitner, J. Klankermayer, Ruthenium-
Catalyzed Synthesis of Dialkoxymethane Ethers Utilizing Carbon Dioxide and
Molecular Hydrogen, Angewandte Chemie-International Edition, 55 (2016) 12266-
1
2269.
[
5] K. Beydoun, K. Thenert, E.S. Streng, S. Brosinski, W. Leitner, J. Klankermayer,
Selective Synthesis of Trimethylamine by Catalytic N-Methylation of Ammonia and
Ammonium Chloride by utilizing Carbon Dioxide and Molecular Hydrogen,
Chemcatchem, 8 (2016) 135-138.
[
6] S. Wesselbaum, V. Moha, M. Meuresch, S. Brosinski, K.M. Thenert, J. Kothe, T.v.
Stein, U. Englert, M. Hoelscher, J. Klankermayer, W. Leitner, Hydrogenation of carbon
dioxide to methanol using a homogeneous ruthenium-Triphos catalyst: from
mechanistic investigations to multiphase catalysis, Chemical Science, 6 (2015) 693-704.

[
7] K. Beydoun, G. Ghattas, K. Thenert, J. Klankermayer, W. Leitner, Ruthenium-
Catalyzed Reductive Methylation of Imines Using Carbon Dioxide and Molecular
Hydrogen, Angewandte Chemie-International Edition, 53 (2014) 11010-11014.
[
8] L. Zhang, Z. Han, L. Zhang, M. Li, K. Ding, Advances in Hydrogenation of
Carboxylic Acid Derivatives and CO2 Using Triphos as the Coordination Ligand,
Chinese Journal of Organic Chemistry, 36 (2016) 1824-1838.
[
9] N. Nakagawa, E.J. Derrah, M. Schelwies, F. Rominger, O. Trapp, T. Schaub,
Triphos derivatives and diphosphines as ligands in the ruthenium-catalysed alcohol
amination with NH3, Dalton Transactions, 45 (2016) 6856-6865.
[
10] E.J. Derrah, M. Hanauer, P.N. Plessow, M. Schelwies, M.K. da Silva, T. Schaub,
Ru(II)-Triphos Catalyzed Amination of Alcohols with Ammonia via Ionic Species,
Organometallics, 34 (2015) 1872-1881.
[
11] I. Mellone, F. Bertini, L. Gonsalvi, A. Guerriero, M. Peruzzini, New Wind in Old
Sails: Novel Applications of Triphos-based Transition Metal Complexes as
Homogeneous Catalysts for Small Molecules and Renewables Activation, Chimia, 69
(2015) 331-338.
[
12] I. Mellone, M. Peruzzini, L. Rosi, D. Mellmann, H. Junge, M. Beller, L. Gonsalvi,
Formic acid dehydrogenation catalysed by ruthenium complexes bearing the tripodal
ligands triphos and NP3, Dalton Transactions, 42 (2013) 2495-2501.
[
13] Z. Xin, J. Zhang, K. Sordakis, M. Beller, C.-X. Du, G. Laurenczy, Y. Li, Towards
Hydrogen Storage through an Efficient Ruthenium-Catalyzed Dehydrogenation of
Formic Acid, Chemsuschem, 11 (2018) 2077-2082.
[
14] G. Manca, I. Mellone, F. Bertini, M. Peruzzini, L. Rosi, D. Mellmann, H. Junge,
M. Beller, A. Ienco, L. Gonsalvi, Inner- versus Outer-Sphere Ru-Catalyzed Formic

Acid Dehydrogenation: A Computational Study, Organometallics, 32 (2013) 7053-
064.
15] Wilkinson,
7
[
P.S.
Hallman,
T.A.
Stephenson,
and
G.
tris(triphenylphosphine)-
Tetrakis(triphenylphosphine)dichloro-ruthenium(II)
dichlororuthenium(II), Inorganic Syntheses, 12 (1970).
[
16] V.V. Mainz, R.A. Andersen, Preparation of RuCH PMe (PMe ) Cl,
2 2 3 3
Ru(CH PMe ) (PMe ) , and Rh (CH PMe ) (PMe ) and their reactions with hydrogen,
2
2 2
3 2
2
2
2 2
3 4
Organometallics, 3 (1984) 675-678.
[
(
[
[
17] A.B. Chaplin, P.J. Dyson, Reactivity and catalytic activity of a robust ruthenium
II)−triphos complex, Inorganic Chemistry, 47 (2008) 381-390.
18] G. 03, Gaussian 03, in: M.J.e.a. Frisch (Ed.), Gaussian, Inc., Wallingford CT, 2004.
19] R.K. Raju, A.A. Bengali, E.N. Brothers, A unified set of experimental
organometallic data used to evaluate modern theoretical methods, Dalton Transactions,
5 (2016) 13766-13778.
4
[
[
20] DZVP (DFT Orbital), EMSL Basis Set Exchange Library.
21] N. Godbout, D.R. Salahub, J. Andzelm, E. Wimmer, Optimization of Gaussian-
type basis sets for local spin density functional calculations. Part I. Boron through neon,
optimization technique and validation, Canadian Journal of Chemistry, 70 (1992) 560-
5
71.
22] J.P. da Silva, F.R. Caetano, D.A. Cavarzan, F.D. Fagundes, L.L. Romualdo, J.
Ellena, M. Jaworska, P. Lodowski, A. Barison, M.P. de Araujo, fac-/mer-
RuCl (NO)(P–N)] (P–N [o-(N,N-dimethylamino)phenyl]diphenylphosphine):
Synthesis, characterization and DFT calculations, Inorganica Chimica Acta, 373 (2011)
-18.
[
[
=
3
8