Interconversion between formic acid and H2/CO2 using rhodium and ruthenium catalysts for CO2 fixation and H2 storage

Article abstract of DOI:10.1002/cssc.201000327

The interconversion between formic acid and H₂/CO₂ using half-sandwich rhodium and ruthenium catalysts with 4,4'-dihydroxy-2,2'- bipyridine (DHBP) was investigated. The influence of substituents of the bipyridine ligand was studied. Chemical shifts of protons in bipyridine linearly correlated with Hammett substituent constants. In the hydrogenation of CO ₂/bicarbonate to formate under basic conditions, significant activations of the catalysts were caused by the electronic effect of oxyanions generated by deprotonation of the hydroxyl group. Initial turnover frequencies of the ruthenium- and rhodium-DHBP complexes increased 65- and 8-fold, respectively, compared to the corresponding unsubstituted bipyridine complexes. In the decomposition of formic acid under acidic conditions, activity enhancement by the electronic effect of the hydroxyl group was observed for the ruthenium catalyst. The rhodium-DHBP catalyst showed high activity without CO contamination in a relatively wide pH range. Pressurized H₂ can be obtained using an autoclave reactor. The highest turnover frequency and number were obtained at 80°C. The catalytic system provides valuable insight into the use of CO₂ as a H₂ storage material by combining CO₂ hydrogenation with formic acid decomposition.

Full text of DOI:10.1002/cssc.201000327

DOI: 10.1002/cssc.201000327
Interconversion between Formic Acid and H2/CO2 using
Rhodium and Ruthenium Catalysts for CO2 Fixation and H2
Storage
Yuichiro Himeda,*[a] Satoru Miyazawa,[b] and Takuji Hirose[b]
The interconversion between formic acid and H2/CO2 using
half-sandwich rhodium and ruthenium catalysts with 4,4’-dihy-
droxy-2,2’-bipyridine (DHBP) was investigated. The influence of
substituents of the bipyridine ligand was studied. Chemical
shifts of protons in bipyridine linearly correlated with Hammett
substituent constants. In the hydrogenation of CO2/bicarbon-
ate to formate under basic conditions, significant activations of
the catalysts were caused by the electronic effect of oxyanions
generated by deprotonation of the hydroxyl group. Initial turn-
over frequencies of the ruthenium- and rhodium-DHBP com-
plexes increased 65- and 8-fold, respectively, compared to the
corresponding unsubstituted bipyridine complexes. In the de-
composition of formic acid under acidic conditions, activity en-
hancement by the electronic effect of the hydroxyl group was
observed for the ruthenium catalyst. The rhodium-DHBP cata-
lyst showed high activity without CO contamination in a rela-
tively wide pH range. Pressurized H2 can be obtained using an
autoclave reactor. The highest turnover frequency and number
were obtained at 808C. The catalytic system provides valuable
insight into the use of CO2 as a H2 storage material by combin-
ing CO2 hydrogenation with formic acid decomposition.
Introduction
Among the main challenges that humans will face in the
future are finding an adequate and sustainable energy supply
and protecting the global environment.[1] Promising solutions
to these challenges lie in the development of a hydrogen
economy, because molecular hydrogen offers advantages over
fossil fuels from both environmental and economical view-
points. However, the actual use of hydrogen is limited, mainly
because of problems with its storage and delivery.[2] Thus, the
development of technologies for hydrogen storage/evolution
in a reversible manner is essential.
than iridium catalysts developed in our group,[9] highly active
catalysts have been achieved by using rhodium and ruthenium
catalysts with phosphine ligands. Recently, Nozaki et al. report-
ed excellent productivity when using an iridium-pincer trihy-
dride catalyst in KOH/H2O/THF solution at 5 MPa and 2008C.[10]
The decomposition of formic acid/formate for H2 production
has been studied by using a number of heterogeneous and
homogeneous catalyst systems.[11,12] In general, formic acid and
formates can be decomposed via dehydrogenation/decarboxy-
lation and dehydration/decarbonylation pathways, which are
thermodynamically downhill under standard conditions. How-
ever, highly efficient and selective H2 production under mild
conditions is still required for the decomposition of formic
acid. Further, when H2 has to be used in fuel cells, CO-free (<
10 ppm) H2 production is essential. Recently, research reported
separately in two excellent papers by the groups of Beller[13]
and Laurenczy[14] has directed renewed attention to H2 produc-
tion.[15–19] Since then, many studies on the decomposition of
formic acid/formate have been conducted, focusing mainly on
ruthenium and rhodium catalysts.[20–25] Although the concept
of H2 storage through the use of CO2 and formic acid is well-
Formic acid (which contains 4.3 wt% of H2) and formate salt
are potential H2 storage materials because they can be han-
dled, stored, and transported easily. The concept of using CO2
as a H2 storage material by the reversible reaction between
formic acid and H2/CO2 has received renewed attention.[3] Cur-
rent research efforts are mainly devoted to the use of homoge-
nous catalysts for (1) the hydrogenation of CO2/bicarbonate
and (2) the decomposition of formic acid/formate (Scheme 1).
[a] Dr. Y. Himeda
Energy Technology Research Institute
National Institute of Advanced Industrial Science and Technology
Tsukuba Central 5-2, 1-1-1 Higashi, Tsukuba, Ibaraki 305-8565 (Japan)
Fax: (+81)29-861-6771
Scheme 1. Interconversion between formic acid and H2/CO2
E-mail: himeda.y@aist.go.jp
Catalytic activities towards the hydrogenation of CO2/bicar-
bonate for the production of formic acid/formate have been
improving steadily since the early 1990s.[4] A number of excel-
lent Reviews on this subject have been published.[5–9] Other
[b] S. Miyazawa, Prof. T. Hirose
Graduate School of Science & Engineering
Saitama University, 255 Shimo-ohkubo, Sakura-ku
Saitama, Saitama 338–8570 (Japan)
ChemSusChem 2011, 4, 487 – 493
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
487

Y. Himeda et al.
known, the hydrogenation of CO2 and the decomposition of
formic acid are, except for a few studies,[26–29] usually reported
independently.
Recently, we reported that half-sandwich iridium complexes
with 4,4’-dihydroxy-2,2’-bipyridine (DHBP) and 4,7-dihydroxy-
1,10-phenanthroline (DHPT) ligands can act as highly efficient
catalysts for the hydrogenation of CO2/bicarbonate[30,31] and
the decomposition of formic acid.[32] The high catalytic activity
of these complexes is attributed to an electron-donating effect
of the hydroxyl groups in the bipyridine ligand. In particular,
the strong electron-donating effect of oxyanions generated by
the deprotonation of the hydroxyl group in the DHBP ligand
leads to activation that is more than 1000 times stronger than
with bipyridine catalyst under basic conditions (Scheme 2).
Table 1. Chemical shifts (ppm) for DHBP complexes (1–3) in D2O.
Signal
Complex
1
2
3
D2O
KOD
6.63
7.13
8.25
D2O
KOD
6.69
7.14
8.28
D2O
KOD
6.64
7.06
8.19
3,3’-H
5,5’-H
6,6’-H
4,4’-C
7.24
7.77
8.77
7.24
7.69
8.76
7.18
7.59
8.74
170.37
178.61
170.34
178.83
170.18
178.68
changes (3,3’-H: 0.38, 5,5’-H: 0.34, and 6,6’-H: 0.33 ppm) were
observed in complex 4b (R=CO2H). On the other hand, slight
upfield shifts of the aromatic protons in 4a (R=H; up to
0.13 ppm), 4c (R=Me; up to 0.11 ppm), and 4d (R=OMe; up to
0.12 ppm) were observed because of the acid–base equilibri-
um of the aqua ligand.[36]
Scheme 2. Acid–base equilibrium of DHBP complexes.
On the other hand, we have reported that during transfer
hydrogenation in water the rhodium bipyridine complex de-
composed formic acid to evolve H2 gas via an undesirable side
reaction.[33,34] We then reported a turnover frequency (TOF) of
238 hÀ1 for the decomposition of formic acid using a rhodium
complex at 408C and pH 3.5.[31] However, we have reported
only the preliminary results obtained using rhodium and ruthe-
nium catalysts with the DHBP ligand,[30] although a number of
rhodium and ruthenium catalysts are known to be effective in
the hydrogenation of CO2/bicarbonate and decomposition of
formic acid.
The electronic substituent effects on the chemical shifts of
the complexes with the disubstituted bipyridine were investi-
gated under acidic and basic conditions. The Hammett sub-
stituent constant (sp+) values for the hydroxyl and carboxylic
acid groups under acidic conditions are À0.92 and 0.42, re-
spectively. On the other hand, under basic conditions the
values for these complexes are À2.30 and À0.02, correspond-
ing to oxyanion (OÀ) and carboxylate (CO2À), respectively. Ac-
cording to this interpretation, the Hammett plots for the iridi-
um, rhodium, and ruthenium series showed a good correlation
(Figure 1).
In this study, we elucidate the electronic substituent effects
of rhodium and ruthenium catalysts, and make a comparison
with those of the iridium catalyst, in the hydrogenation of
CO2/bicarbonate under basic conditions and decomposition of
formic acid under acidic conditions. We also investigate the in-
fluence of the central metal on the activation energies of the
catalysts.
Hydrogenation of CO2/bicarbonate under basic conditions
In a previous paper, we reported that iridium DHBP complex 1
is a highly efficient catalyst in the hydrogenation of CO2/bicar-
bonate in water.[30] In addition, we observed good correlation
between the sp+ values and the initial TOFs of the 4,4’-disubsti-
tuted-2,2’-bipyridine complexes. However, we reported only
the preliminary results for the hydrogenation of CO2/bicarbon-
ate using rhodium and ruthenium catalysts.
Results and Discussion
Syntheses and properties
DHBP aqua complexes 1–3 were prepared according to litera-
ture procedures.[35] The chemical shifts of these complexes
were examined under acidic and basic conditions in D2O
(Table 1). The addition of KOD caused upfield shifts in the
1HNMR signals of the 3,3’ (0.54–0.61 ppm), 5,5’ (0.53–
0.64 ppm), and 6,6’ protons (0.48–0.55 ppm), and downfield
shifts in the 13CNMR signals of the 4,4’-carbons (8.24–
8.50 ppm). These shifts can be attributed to an increase in the
electronic density by the deprotonation of the hydroxyl
groups in the bipyridine ligand. In a similar manner, significant
Electronic substituent effects on the hydrogenation of CO2/
bicarbonate using the iridium (5), rhodium (6), and ruthenium
series (7) were investigated in a 1m KOH solution at 808C and
1 MPa H2/CO2 (1:1). The Hammett plot for the iridium series in-
+
dicates good correlation between the initial TOFs and the sp
values and a high p-value of À1.3; these results are consistent
with previous results obtained at 4 MPa (Figure 2).[30] The initial
TOF of DHBP complex 5e was 1100 times higher than that of
bipyridine complex 5a (Table 2, entries 1 and 4). The final con-
488
www.chemsuschem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2011, 4, 487 – 493

Rhodium and Ruthenium Catalysts for CO2 Fixation and H2 Storage
Figure 1. Correlation between the Hammett substituent constant (sp+) and the chemical shifts of 3,3’-H (open squares: D2O, closed squares: KOD/D2O), 5,5’-H
(open triangles: D2O, closed triangles: KOD/D2O), and 6,6’-H (open circles: D2O, closed circles: KOD/D2O) for (a) iridium complexes (1 and 4a–d), (b) rhodium
complexes 6a–e, and (c) ruthenium complexes 7a–e.
of bipyridine complex 6a (entries 2 and 5). The degree of cata-
lytic activation caused by the electronic substituent effect was
not strong in the case of the rhodium series. The rate of for-
mate generation decreased considerably after 8 h and the final
concentration of formate was only 0.12m, probably because
the catalyst degraded. In the case of the ruthenium series too,
the Hammett plot showed a good correlation between the ini-
+
tial TOFs and the sp values, although the catalytic perfor-
mance was moderate. The initial TOF of DHBP complex 7e was
65 times higher than that of bipyridine complex 7a (entries 3
and 6). The rate of formate generation was almost constant
until 50 h and the final formate concentration was 0.54m after
100 h. This suggests that the ruthenium catalyst is relatively
stable. At 1008C, formate was generated at a higher reaction
+
Figure 2. Correlation between initial TOFs and sp values of substituent (R)
rate without significant degradation of the catalyst (entry 7).
in hydrogenation of CO2 at 1 MPa (H2/CO2 =1:1) in 1m aqueous KOH solu-
tion at 808C using (a) iridium catalysts 5a–e (0.1 mm) (open circles), (b) rho-
dium catalysts 6a--e (0.1 mm) (closed circles), and (c) ruthenium catalysts
All of the reactions involving rhodium- and ruthenium-DHBP
catalysts remained soluble during the course of the reaction.[37]
6a–e (0.2 mm) (closed squares).
From the Arrhenius plot of the temperature dependence of
the initial TOF, the apparent acti-
vation energies of the iridium-
(5e), rhodium- (6e), and rutheni-
Table 2. Hydrogenation of CO2/bicarbonate.[a]
um-DHBP catalysts (7e) were
found to be 58, 72, and
Entry
Catalyst
Amount
[mM]
Time
[h]
Temp.
[8C]
Initial
TOF [hÀ1
TON
Final conc.
of formate [M]
78 kJmolÀ1
,
respectively
]
(Figure 3). The activation ener-
gies required for the hydrogena-
tion of bicarbonate in water
using [RhCl(TPPMS)3], [RuCl2-
(PTA)4], and [(C6H6)RuCl2]2+4 PTA
(1,3,5-triaza-7-phosphaadaman-
tane) are 36,[38] 86,[39] and
1
2
3
4
5
6
7
5e
6e
7e
5a
6a
7a
7e
0.02
0.1
0.1
0.2
0.2
0.2
0.2
30
30
100
30
30
30
80
80
80
80
80
5100
160
92
4.5
20
11000
1200
5400
120
190
45
0.22
0.12
0.54
0.023
0.037
0.009
0.58
80
100
1.4
300
32
2900
[a] The reaction was carried out in a degassed 1m KOH solution (10 mL) at 1 MPa of H2/CO2 (1:1). TON: turn-
over number.
126 kJmolÀ1 [40]
,
respectively. It
seems likely that the difference
in rate-determining steps in the
hydrogenation reaction reflected
centration of formate generated at 1 MPa (0.22m) was lower
than that at 4 MPa (0.80m).[30] The rhodium-bipyridine catalyst
6a exhibited the highest catalytic activity among the iridium-
and ruthenium-bipyridine catalysts (entries 4–6). However, the
initial TOF of DHBP complex 6e was 8 times higher than that
the difference in the activation energies of the catalysts.[36]
The activity enhancement of the rhodium and ruthenium
catalysts by the electronic substituent effect of the oxyanions
was confirmed on the basis of the correlation between the ini-
+
tial TOF and the sp values of the substituents. Under aqueous
ChemSusChem 2011, 4, 487 – 493
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
489

Y. Himeda et al.
lysts are strongly dependent on pH, the pH values of the reac-
tion solution were optimized for 2 and 3. The TOFs and con-
centration of residual formate at various pH values of 1m
HCO2H/HCO2Na solution are shown in Figure 4. The initial TOFs
of 2 exceeded 1200 hÀ1 in the pH range 2.5–4.0 (Table 3,
Figure 3. Arrhenius plots of initial TOF for hydrogenation of CO2 using 5e
(open circles), 6e (closed circles), and 7e (closed squares) (0.1–0.05 mm) at
1 MPa (H2/CO2 =1:1).
Figure 4. pH dependence of gas evolution at 608C in 1m HCO2H/HCO2Na
(10 mL) at various pH values. (a) Initial TOF (closed circles) and (b) concentra-
tion of residual formate (open circles) using 2 (0.2–0.4 mm), (c) initial TOF
(closed squares) and (d) concentration of residual formate (open squares)
using 3 (0.5–1.0 mm).
conditions, the ruthenium-DHBP catalyst 3 showed higher effi-
ciency than the other ruthenium catalysts used in earlier stud-
ies.
Decomposition of formic acid/formate under acidic
conditions
entry 4), while the maximum TOF of catalyst 1 was attained in
the case of an aqueous formic acid solution (pH 1.8). The TOFs
of 3 were moderate (maximum TOF 140 hÀ1 at pH 4.0, entry 5).
The TOFs of 2 and 3 dropped significantly in pure formic acid
solution and at pH values above 5. The trends of TOFs of 2
and 3 were different from that of 1.
On the other hand, the trends of the conversions of formic
acid of 2 and 3 were similar to that of 1. At the end of the re-
action in pure formic acid solution, no formic acid was detect-
ed (Table 3, entries 2 and 3). An increase of the pH caused an
increase of the concentration of the residual formate in the re-
action solution. Moreover, negligible gas evolution was detect-
ed in the sodium formate solu-
In our previous study,[32] iridium-DHBP complex 1 acted as a
highly efficient and selective catalyst for the decomposition of
formic acid with complete conversion of formic acid in water
(Table 3, entry 1). On the other hand, the rhodium and rutheni-
um catalysts showed moderate to poor performances under
the same conditions (entries 2 and 3).
When using the series of the rhodium and ruthenium cata-
lysts gas evolution was observed. The amount of an evolved
gas increased linearly at time following a brief induction
period at the beginning of the reaction. Because DHBP cata-
tion. It was clear that formic acid
had decomposed whereas for-
Table 3. Decomposition of formic acid/formate.[a]
mate, necessary for achieving
high reaction efficiency, had not
decomposed. In addition, similar
to the reactions using 1,[32] CO
was not detected in the evolved
gas for all the reactions (gas
chromatography using a flame
ionization detector equipped
with a methanizer).
Entry
Catalyst
Amount
[mM]
Solution
Time
[h]
Temp.
[8C]
Initial
TON
Conversion[b]
[%]
TOF [hÀ1
]
1[c]
2
3
4
5
6
7
8
1
2
3
2
0.2
0.4
1.0
0.2
0.5
0.2
0.2
0.2
0.2
0.5
0.1
0.5
HCO2H
HCO2H
HCO2H
pH 2.5
pH 4.0
pH 3.0
pH 3.0
pH 3.0
pH 3.0
pH 3.0
pH 2.5
pH 3.0
5
24
48
7
9
6
5
6
5
60
60
60
60
60
60
60
60
60
60
80
80
80
2400
440
36
1340
140
1230
1350
1270
1340
94
7900
720
7700
5000
100
100
>99
92
25
73
74
75
75
74
5000
5000
4600
500
3700
3700
3800
3800
3700
9300
1600
83000
3
6a
6c
6d
6e
7e
2
Electronic substituent effects
were investigated by using the
series of rhodium and ruthenium
catalysts 6a–e and 7a–e, respec-
tively, at pH 3.0. All the catalysts
of the rhodium series exhibited
high initial TOFs (Table 3, en-
tries 6–9). However, the catalytic
9
10
11
12
13[d]
24
2.5
5
93
80
75
3
2
50
—
–
[a] The reaction was carried out in a degassed 1m formic acid/formate solution (10 mL) until gas evolution
ceased. [b] Conversion was calculated by the residual formate. [c] See Ref. [32]. [d] The reaction was carried out
using 2 (2 mmol) in a degassed 4m formic acid solution (50 mL) and 4m sodium formate solution (5 mL).
490
www.chemsuschem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2011, 4, 487 – 493

Rhodium and Ruthenium Catalysts for CO2 Fixation and H2 Storage
activity of 6b significantly decreased after 1 h, probably be-
cause of catalyst degradation, although an initial TOF of ca.
900 was observed. These results show that the series of rhodi-
um catalysts were not influenced by electronic substituent ef-
fects. In the case of the rhodium series, the rate-determining
step may be different from that of the iridium series. In the
case of the ruthenium catalyst series, the initial TOF of 7e
(entry 10) was approximately 2.9 times that of unsubstituted
analogue 7a. On the other hand, the activity for 7b was less,
because of the electron-withdrawing effect of carboxyl group
(sp+ =0.42). The TOF values for the series of ruthenium cata-
lysts show a good correlation with the Hammett substituent
constants (Figure 5).
Figure 6. Arrhenius plots of initial TOFs for decomposition of formic acid
using (a) 1 (0.2 mm) in 2m aqueous HCO2H solution (open circles), (b) 2 (0.1–
0.2 mm) in 1m aqueous HCO2H/HCO2Na solution at pH 2.5 (closed circles),
and (c) 3 (0.5–1.0 mm) in 1m aqueous HCO2H/HCO2Na solution at pH 3.0
(closed squares).
Figure 5. Hammett plot for initial TOFs vs. sp+ values of substituent (R) in
ruthenium catalysts 7a–e (0.5–1.0 mm) at 608C in 1m aqueous HCO2H/
HCO2Na solution (10 mL) at pH 3.0.
The temperature dependence of the TOFs was investigated
also. In the case of 2, the highest TOF of 7900 hÀ1 was ob-
tained at 808C and pH 2.5, with a conversion rate of 93%
(entry 11). These values are comparable to those of iridium an-
alogue 1. For 3, a satisfactory TOF of 720 hÀ1 was obtained at
808C (Table 3, entry 12). Further, in the case of 2 (2 mmol), a
highest turnover number (TON) of 83000 was obtained in a
4m HCO2H/HCO2Na solution at 808C after 50 h (entry 13).
Figure 6 shows the Arrhenius plots for 1,[32] 2, and 3. The ap-
parent activation energies of 2 and 3 are 93 kJmolÀ1 and
96 kJmolÀ1, respectively, which are higher than that of 1
(76 kJmolÀ1).[32]
Figure 7. Time course of pressure using 2 (0.2 mm) at 808C in 2m HCO2H/
HCO2Na (95:5) solution (10 mL) in glass autoclave.
in the case of 1, although a small amount of formate salt was
required for ensuring high reaction efficiency in the former
case. CO-free H2 was produced with a high TOF (up to
7900 hÀ1), high TON (up to 83000), and conversion of formic
acid (up to 100%). In addition, pressurized H2 can be obtained
in a closed reactor. The electronic substituent effects and acti-
vation energies for 2 and 3 provide valuable mechanistic in-
sight.
Pressurized gas supply will be essential in practical applica-
tions (e.g., for separation and filling of gas). A spontaneous in-
crease of the gas pressure has been observed in a closed
system, similar to the iridium catalyst.[32] When the reaction cat-
alyzed by 2 in a 2m HCO2H/HCO2Na (95:5) solution was carried
out in an autoclave, the pressure of the evolved gas exceeded
5 MPa after 4 h, and 92 mm of formate (4.6%) remained at the
end of the reaction (Figure 7). Thus, the gas pressure in the
system did not inhibit the decomposition of formic acid.
Highly efficient H2 evolution was achieved in water using 2
by the adjusting the reaction conditions. The excellent results
obtained in the case of 2 were comparable to those obtained
Conclusions
The interconversion between formic acid and H2/CO2 using
rhodium and ruthenium catalysts with DHBP in water is dem-
onstrated. Catalytic activations caused by an electronic sub-
stituent effect of the oxyanions in the rhodium and ruthenium
catalysts with DHBP ligand in the hydrogenation of CO2/bicar-
bonate are observed under basic conditions. Relatively high
TONs and TOFs are obtained using the ruthenium-DHBP cata-
lyst. In addition, an electronic substituent effect in the decom-
ChemSusChem 2011, 4, 487 – 493
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
491

Y. Himeda et al.
1.64 ppm (s, 15H); 13C NMR ([D6]DMSO): d=168.57, 155.96, 153.50,
position of formic acid/formate under acidic conditions is ob-
served in the case of the ruthenium catalyst. The rhodium-
DHBP catalyst shows efficient catalytic performance without
CO contamination in the pH range 2.5–4.0, although a small
amount of formate is required. In addition, pressurized H2 is
obtained by using an autoclave reactor. We believe that the
DHBP ligand plays an important role in both the reactions: hy-
drogenation of CO2 and decomposition of formic acid.
114.82, 110.89, 96.78 (d, JRhÀC =4.0 Hz), 57.59, 8.97 ppm; IR (KBr),
+
~
v=1615, 1560, 1493, 1338, 1252, 1233; ESIMS: m/z 489 [MÀCl] ;
Anal. calcd for C22H27Cl2N2O2Rh: C 50.30, H 5.18, N 5.33, found: C
50.05, H 5.18, N 5.22.
Preparation of ruthenium complex with 4,4’-dicarboxy-2,2’-bipyri-
dine (7b): A DMF solution (30 mL) of [(C6Me6)RuCl2]2 (334 mg,
0.50 mmol) and 4,4’- dicarboxy À2,2’-bipyridine (244 mg,
1.00 mmol) was stirred at 408C for 12 h. The reaction solution was
filtered off. The volume of the filtrate was reduced to ~5 mL in
vacuo and ethyl acetate was added to precipitate 7b as a pale
yellow solid (503 mg, 87%). An analytical sample was obtained by
chromatography on a Sephadex LH-20 (Pharmacia Fine Chemicals)
column: 1H NMR ([D6]DMSO): d=8.99 (d, J=5.7 Hz, 2H), 8.64 (d,
The DHBP catalyst system offers several excellent properties
for practical use in a H2 storage system. The aqueous reaction
without use of organic additives will be essential from a eco-
nomical and ecological point of view. The selection of the di-
rection of reactions (i.e., formation/decomposition of formic
acid) by adjusting the pH using the same catalyst offers the
possibility of H2 storage. Further investigations on the reversi-
bility of storage and evolution of H2 in the same vessel are in
progress. Furthermore, it should be noted that H2 generated
from renewable sources is required for large-scale applications,
although CO2 fixation using rhodium and ruthenium catalysts
may be employed for effective H2 storage.
1
J=1.6 Hz, 2H), 7.68 (dd, J=1.6, 5.7 Hz, 2H), 2.05 (s, 18H); H NMR
(KOD/D2O): d=9.00 (d, J=6.0 Hz, 2H), 8.65 (d, J=1.6 Hz, 2H), 8.04
(dd, J=1.6, 5.8 Hz, 2H), 2.06 ppm (s, 18H); 13C NMR (KOD/D2O): d=
173.3, 158.1, 156.1, 150.1, 129.0, 124.6, 57.0, 6.17 ppm; ESIMS; m/z
541 [MÀCl] +; Anal. calcd for C24H26Cl2N2O4Ru·CH3OH: C 49.19, H
4.95, N 4.59, found: C 49.23, H 4.76, N 4.32.
Preparation of ruthenium complex with 4,4’-dimethyl-2,2’-bipyri-
dine (7c): In the same manner as described for the preparation of
6d, [(C6Me6)RuCl2]2 (500 mg, 0.75 mmol) was treated with 4,4’-di-
methyl-2,2’-bipyridine (276 mg, 1.50 mmol) to give 7c as an
orange precipitate (662 mg, 80%): H NMR ([D6]DMSO): d=8.75 (d,
J=5.9 Hz, 2H), 8.49 (bs, 2H), 7.68 (dd, J=1.0, 5.9 Hz, 2H), 2.57 (s,
6H), 2.02 ppm (s, 18H); 13C NMR ([D6]DMSO): d=154.15, 153.11,
Experimental Section
1
All manipulations were carried out under an argon atmosphere
using standard Schlenk techniques or in a glovebox. All aqueous
solutions were degassed prior to use. HNMR and 13CNMR spectra
1
~
151.75, 128.60, 124.24, 95.32, 20.88, 15.33 ppm; IR (KBr) v=1614,
were recorded on a Varian INOVA 400 spectrometer using sodium
3-(trimethylsilyl)-1-propanesulfonate (DSS) and tetramethylsilane
(TMS) as an internal standard. ESI-MS was measured with a Micro-
mass QUATTRO II mass spectrometer. Elemental analyses were car-
ried out on an Eager 200 instrument. FTIR spectra were recorded
on a Perkin–Elmer Spectrum One spectrometer. pH values were
measured on an Orion Model 290 A pH meter with a glass elec-
trode after calibration to standard buffer solutions. The gas sam-
ples, which were obtained at various intervals with a gastight sy-
ringe through a septum, were analyzed for H2 with a TCD (thermal
conductivity detector) using an activated carbon 60/80. In case of
CO2, the samples were analyzed with an FID equipped with a
methaniser using a Porapak Q 80/100 at 508C, on a GL Science
GC390 gas chromatograph. The formate concentrations were
monitored by an HPLC on an anion-exclusion column (Tosoh
TSKgel SCX(H+)) with an aqueous phosphate solution (20 mm) as
an eluent and a UV detector (l=210 nm).[41] Research grade CO2
(>99.999%) and H2 (>99.9999%) were used. [Cp*RhCl2]2,
[Cp*IrCl2]2, [(C6Me6)RuCl2]2, 4,4’-dicarboxy-2,2’-bipyridine, 4,4’-di-
methyl-2,2’-bipyridine, and 4,4’-dimethoxy-2,2’-bipyridine were
commercially available from either Aldrich, Tokyo Kasei, or Strem.
The 4,4’-dihydroxy-2,2’-bipyridine,[42] 1--3, 4a–d,[35] 5a–e, 6a, 6e,
7a, 7e,[30] and 6b,c[43] were prepared according to literature proce-
dures.
1482, 1444, 1389, 1010, 837; ESIMS m/z 483 [MÀCl] +; Anal. calcd
for C24H30Cl2N2Ru·3/2H2O: C 52.84, H 6.10, N 5.14, found: C 52.55, H
6.14, N 5.02.
Preparation of ruthenium complex with 4,4’-dimethoxy-2,2’-bipyri-
dine (7d): In the same manner as described for the preparation of
6d, [(C6Me6)RuCl2]2 (335 mg, 0.50 mmol) was treated with 4,4’-di-
methoxy-2,2’-bipyridine (219 mg, 1.01 mmol) to give 7d as an
orange precipitate (510 mg, 93%): H NMR ([D6]DMSO): d=8.66 (d,
J=6.6 Hz, 2H), 8.24 (d, J=2.7 Hz, 2H), 7.37 (dd, J=6.6, 2.7 Hz, 2H),
4.05 (s, 6H), 2.02 ppm (s, 18H); 13C NMR ([D6]DMSO): d=167.88,
1
~
155.84, 154.49, 114.18, 111.30, 94.78, 57.26, 15.34 ppm; IR (KBr) v=
+
1610, 1492, 1416, 1340, 1280, 1230, 1044; ESIMS m/z 515 [MÀCl]
;
Anal. calcd for C24H32Cl2N2O2Ru·3/2H2O: C 49.91, H 5.76, N 4.85,
found: C 50.29, H 5.53, N 4.87.
Procedure for catalytic hydrogenation of CO2/bicarbonate: A de-
gassed aqueous 1m KOH solution (50 mL) of the complex was sa-
turated with CO2 in a 100 mL stainless steel reactor equipped with
a sampling device. The reactor was heated and then repressurized
with 1MPa (CO2:H2 =1:1). At appropriate intervals, samples were
removed and analyzed by HPLC. The initial TOF was calculated
using nonlinear least-squares fitting of the experimental data ob-
tained from the initial part of the reaction.[44]
Preparation of rhodium complex with 4,4’-dimethoxy-2,2’-bipyri-
dine (6d): A methanol solution (20 mL) of [Cp*RhCl2]2 (309 mg,
0.50 mmol) and 4,4’-dimethoxy-2,2’-bipyridine (220 mg, 0.51 mmol)
was stirred at RT for 12 h. The solvent was removed in vacuo and
the residue was dissolved in a minimum amount of acetonitrile.
Addition of ethyl acetate gave a pale yellow solid 6d (510 mg,
97%), which was collected, washed with ether, and dried in vacuo.
An analytical sample was obtained by recrystallization from
Procedure for catalytic decomposition of formic acid/formate: Typi-
cally, a 20 mm solution of catalyst (100 mL, 2 mmol) was added to a
deaerated aqueous HCO2H/HCO2Na solution, and the mixture was
stirred at the desired temperature. The volume of gas evolution
was determined by a gas meter (Shinagawa Corp., W-NK-05). The
initial TOF was calculated using linear least-squares fitting of the
experimental data obtained from the initial part of the reaction
with the exception of the brief induction period.
1
CH3CN/AcOEt: H NMR ([D6]DMSO): d=8.73 (d, J=6.6 Hz, 2H), 8.29
(d, J=2.7 Hz, 2H), 7.42 (dd, J=6.6, 2.7 Hz, 2H), 4.07 (s, 6H),
492
www.chemsuschem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2011, 4, 487 – 493

Rhodium and Ruthenium Catalysts for CO2 Fixation and H2 Storage
[23] A. Boddien, B. Loges, F. Gartner, C. Torborg, K. Fumino, H. Junge, R.
Ludwig, M. Beller, J. Am. Chem. Soc. 2010, 132, 8924–8934.
[24] S. Fukuzumi, T. Kobayashi, T. Suenobu, J. Am. Chem. Soc. 2010, 132,
1496–1497.
[25] S. Fukuzumi, T. Kobayashi, T. Suenobu, ChemSusChem 2008, 1, 827–
834.
Acknowledgements
We are grateful to the Ministry of Economy, Trade, and Industry
for providing financial support.
[26] M. L. Man, Z. Y. Zhou, S. M. Ng, C. P. Lau, Dalton Trans. 2003, 3727–
3735.
[27] Y. Gao, J. K. Kuncheria, H. A. Jenkins, R. J. Puddephatt, G. P. A. Yap, J.
Chem. Soc. Dalton Trans. 2000, 3212–3217.
Keywords: carbon dioxide fixation · formic acid · hydrogen
storage · rhodium · ruthenium
[28] W. Leitner, E. Dinjus, F. Gassner, J. Organomet. Chem. 1994, 475, 257–
266.
[29] D. Preti, S. Squarcialupi, G. Fachinetti, Angew. Chem. 2010, 122, 2635–
2638; Angew. Chem. Int. Ed. 2010, 49, 2581–2584.
[30] Y. Himeda, N. Onozawa-Komatsuzaki, H. Sugihara, K. Kasuga, Organome-
tallics 2007, 26, 702–712.
[31] Y. Himeda, N. Onozawa-Komatsuzaki, H. Sugihara, H. Arakawa, K.
Kasuga, Organometallics 2004, 23, 1480–1483.
[32] Y. Himeda, Green Chem. 2009, 11, 2018–2022.
[33] Y. Himeda, N. Onozawa-Komatsuzaki, H. Sugihara, H. Arakawa, K.
Kasuga, J. Mol. Catal. A: Chem. 2003, 195, 95–100.
[34] Y. Himeda, N. Onozawa-Komatsuzaki, H. Sugihara, H. Arakawa, K.
Kasuga, The 50th Symposium on Coordination Chemistry of Japan, Japan
Society of Coordination Chemistry, Kusatsu, 2000, 1P3K01.
[35] a) Y. Himeda, N. Onozawa-Komatsuzaki, S. Miyazawa, H. Sugihara, T.
Hirose, K. Kasuga, Chem. Eur. J. 2008, 14, 11076–11081; b) Y. Himeda, S.
Miyazawa, N. Onozawa-Komatsuzaki, T. Hirose, K. Kasuga, Dalton Trans.
2009, 6286–6288.
[36] S. Ogo, R. Kabe, H. Hayashi, R. Harada, S. Fukuzumi, Dalton Trans. 2006,
4657–4663.
[37] Y. Himeda, N. Onozawa-Komatsuzaki, H. Sugihara, K. Kasuga, J. Am.
Chem. Soc. 2005, 127, 13118–13119.
[38] A. Katho, Z. Opre, G. Laurenczy, F. Joꢁ, J. Mol. Catal. A: Chem. 2003, 204,
143–148.
[39] G. Laurenczy, F. Joꢁ, L. Nadasdi, Inorg. Chem. 2000, 39, 5083–5088.
[40] H. Horvꢂth, G. Laurenczy, A. Katho, J. Organomet. Chem. 2004, 689,
1036–1045.
[41] N. Komatsuzaki, Y. Himeda, T. Hirose, H. Sugihara, K. Kasuga, Bull. Chem.
Soc. Jpn. 1999, 72, 725–731.
[42] Y. R. Hong, C. B. Gorman, J. Org. Chem. 2003, 68, 9019–9025.
[43] U. Kçlle, B. S. Kang, P. Infelta, P. Comte, M. Gratzel, Chem. Ber. 1989, 122,
1869–1880.
[1] N. Armaroli, V. Balzani, Angew. Chem. 2007, 119, 52–67; Angew. Chem.
Int. Ed. 2007, 46, 52–66.
[2] U. Eberle, M. Felderhoff, F. Schuth, Angew. Chem. 2009, 121, 6732–6757;
Angew. Chem. Int. Ed. 2009, 48, 6608–6630.
[3] R. Williams, R. S. Crandall, A. Bloom, Appl. Phys. Lett. 1978, 33, 381–383.
[4] C. Federsel, R. Jackstell, M. Beller, Angew. Chem. 2010, 122, 6392–6395;
Angew. Chem. Int. Ed. 2010, 49, 6254–6257.
[5] W. Leitner, Angew. Chem. 1995, 107, 2391–2405; Angew. Chem. Int. Ed.
Engl. 1995, 34, 2207–2221.
[6] P. G. Jessop, T. Ikariya, R. Noyori, Chem. Rev. 1995, 95, 259–272.
[7] P. G. Jessop, F. Joꢁ, C. C. Tai, Coord. Chem. Rev. 2004, 248, 2425–2442.
[8] P. G. Jessop, Handbook of Homogeneous Hydrogenation, Vol. 1 (Eds.: J. G.
de Vries, C. J. Elsevier), Wiley-VCH, Weinheim, 2007, pp. 489–511.
[9] Y. Himeda, Eur. J. Inorg. Chem. 2007, 3927–3941.
[10] R. Tanaka, M. Yamashita, K. Nozaki, J. Am. Chem. Soc. 2009, 131, 14168–
14169.
[11] B. Loges, A. Boddien, F. Gartner, H. Junge, M. Beller, Top. Catal. 2010, 53,
902–914.
[12] T. C. Johnson, D. J. Morris, M. Wills, Chem. Soc. Rev. 2010, 39, 81–88.
[13] B. Loges, A. Boddien, H. Junge, M. Beller, Angew. Chem. 2008, 120,
4026–4029; Angew. Chem. Int. Ed. 2008, 47, 3962–3965.
[14] C. Fellay, P. J. Dyson, G. Laurenczy, Angew. Chem. 2008, 120, 4030–4032;
Angew. Chem. Int. Ed. 2008, 47, 3966–3968.
[15] H. Junge, A. Boddien, F. Capitta, B. Loges, J. R. Noyes, S. Gladiali, M.
Beller, Tetrahedron Lett. 2009, 50, 1603–1606.
[16] A. Boddien, B. Loges, H. Junge, F. Gartner, J. R. Noyes, M. Beller, Adv.
Synth. Catal. 2009, 351, 2517–2520.
[17] A. Boddien, B. Loges, H. Junge, M. Beller, ChemSusChem 2008, 1, 751–
758.
[18] W. J. Gan, P. J. Dyson, G. Laurenczy, React. Kinet. Catal. Lett. 2009, 98,
205–213.
[19] C. Fellay, N. Yan, P. J. Dyson, G. Laurenczy, Chem. Eur. J. 2009, 15, 3752–
3760.
[44] K. Kudo, N. Sugita, Y. Takezaki, Nippon Kagaku Kaishi 1977, 302–309.
[20] D. J. Morris, G. J. Clarkson, M. Wills, Organometallics 2009, 28, 4133–
4140.
Received: October 1, 2010
[21] X. L. Li, X. Y. Ma, F. Shi, Y. Q. Deng, ChemSusChem 2010, 3, 71–74.
[22] A. Majewski, D. J. Morris, K. Kendall, M. Wills, ChemSusChem 2010, 3,
431–434.
Revised: December 6, 2010
Published online on January 26, 2011
ChemSusChem 2011, 4, 487 – 493
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
493