Recyclable catalyst for conversion of carbon dioxide into formate attributable to an oxyanion on the catalyst ligand

Article abstract of DOI:10.1021/ja054236k

The catalyst recycling in the conversion of CO₂ into formate using the iridium complex with 4,7-dihydroxy-1,10-phenanthroline as a catalyst precursor is described. The catalyst precursor was dissolved in an aqueous KOH solution under CO₂ pressure prior to the reaction, but was precipitated spontaneously at the end of the reaction. The acidification by the generation of formate caused the transformation from the water-soluble deprotonated form into the water-insoluble protonated form. When the reaction was carried out at 60 °C for 20 h using 0.1 M KOH solution under 6 MPa of H₂:CO₂ (1:1), the catalyst precursor was precipitated spontaneously and the added KOH was consumed completely. The catalyst was recovered by filtration, and the product was obtained by the evaporation of the filtrate. Iridium leaching into the filtrate was found to be 0.11 ppm (< 2% of the loaded Ir). The recovered catalyst retained high catalytic activity for four cycles. Consequently, the CO₂ conversion using the complex is an environmentally benign process, whose significant features are as follows: (i) catalyst recycling by self-precipitation/filtration, (ii) waste-free process, (iii) the easy isolation of the product, (iv) high efficiency under relatively mild conditions, and (v) aqueous catalysis without the use of organic materials. Furthermore, we have demonstrated the significant roles of the oxyanion generated from the acidic phenolic hydroxyl on the catalyst ligand, which are the catalyst recovery by acid-base equilibrium, as well as the water-solubility by its polarity and the catalyst activation by its electron-donating ability. Copyright

Full text of DOI:10.1021/ja054236k

Published on Web 09/02/2005
Recyclable Catalyst for Conversion of Carbon Dioxide into Formate
Attributable to an Oxyanion on the Catalyst Ligand
Yuichiro Himeda,* Nobuko Onozawa-Komatsuzaki, Hideki Sugihara, and Kazuyuki Kasuga
National Institute of AdVanced Industrial Science and Technology, Tsukuba Central 5-2, 1-1-1 Higashi,
Tsukuba, Ibaraki 305-8565, Japan
Received June 27, 2005; E-mail: himeda.y@aist.go.jp
Scheme 1. Proposed Mechanism for the Hydrogenation of
Bicarbonate Catalyzed by the Iridium Complex with
4,7-Dihydroxy-1,10-phenanthroline
The development of efficient strategies for the recovery and
recycling of homogeneous catalysts is now recognized as an
important subject in catalysis research from an industrial point of
view.1-3 The conventional approaches are water/organic biphasic
catalysis, based on the difference in polarity between water and
organic solvents. However, to avoid the use of volatile organic
chemicals (VOCs), the new separation strategies rely on the use of
recently developed reaction media (e.g., supercritical fluids and ionic
liquids) or filtration technology for polymer- and dendrimer-
enlarged catalysts.4 The re-immobilization of a catalyst by treatment
with an acid or a base after reaction facilitated catalyst separation,5
but the formation of salts as a waste is environmentally disadvanta-
geous. More recently, the interesting finding reported by Bullock
et al. is the fact that the catalyst for the solvent-free hydrosilylation
of ketones precipitates spontaneously at the end of the reaction,6
which is based on the difference in polarity between the substrate
and product. Such a self-precipitation system helps to avoid the
use of solvent and generation of a waste in the subsequent separation
step.
of the reaction solution is affected by the pressure of CO2 and the
concentration of formate. First, the solubility of the catalyst during
the reaction was visually observed by using a glass reactor
containing 0.1 M KOH aqueous solution (50 mL) of [Cp*Ir(H2L)-
Cl]Cl (1‚Cl) (1.5 mg). The complex was completely soluble in CO2-
saturated KOH solution under 3 MPa of CO2, in which CO2 exists
as bicarbonate. When the reactor was recharged with 3 MPa of
H2:CO2 (1:1) at 60 °C, the clear pale yellow solution turned orange,
which may indicate the formation of hydride complex 3,11 and then
a yellow precipitate formed after 30 min. After 3 h of stirring, the
pressure was slowly released. The resulting suspension was
immediately filtered through a PTFE filter to give the clear pale
yellow filtrate and the yellow precipitate. The filtrate (pH ) 7.9)
was found to contain 0.05 M formate (50% conversion) and 1.0
ppm of Ir (ca. 10% of the loaded Ir), as determined by HPLC and
ICP-MS analysis, respectively. The precipitate was readily con-
verted to a deprotonated form 2 by the addition of an aqueous base,
as indicated by 1H NMR.9a It was clarified that the catalyst precursor
was precipitated with the generation of formate. The precipitation
of the catalyst even under slightly basic conditions can be
rationalized by the fact that the pH of an aqueous solution decreases
under CO2 pressure.12
Then, we investigated the influence of the KOH concentration
on the pH-dependent catalyst. Figure 1 shows the time course of
the formate concentration within an initial period of 60 min at 6
MPa of H2:CO2 (1:1) at 60 °C in various KOH concentrations. The
amount of formate generated during the initial 5 min was ap-
proximately the same for all the reactions. In other words, the
catalyst exhibited the maximum ability (TOF ) 5320 h-1) at the
beginning of the reaction under basic conditions. Subsequently,
while the reaction rate in 1.0 M KOH solution was maintained high,
those of the others decreased drastically after 5 min in 0.1 M
solution and after 20 min in 0.5 M solution. After 24 h, the reaction
rates in all cases significantly decreased with an increase in formate
concentration. It can be seen that the pH decrease caused the catalyst
precipitation by the protonation of water-soluble deprotonated form,
which was activated by oxyanion, so that the reaction rate was
decreased.
The efficient utilization of CO2 also remains the important issue
and the long-term goal.7 The formate salts generated in the
homogeneous catalytic hydrogenation of CO2 or bicarbonate are a
valuable product used in leather tanning, etc. However, one of the
problems, which must be solved, is catalyst recycling because highly
active catalysts have presently been restricted to the complexes of
precious metals, such as Pd, Rh, Ir, and Ru.7,8
In our previous report on the hydrogenation of bicarbonate,9 we
have revealed for the first time the significant roles of the oxyanion
generated from the acidic phenolic hydroxyl on the catalyst ligand,
in which its electron-donating ability and polarity lead to remark-
able catalyst activation and water-solubility, respectively. As a
result, the highest initial turnover frequency (TOF) and turnover
number (TON) in the hydrogenation of bicarbonate in water was
obtained by the use of 4,7-dihydroxy-1,10-phenanthroline (H2L)
complexes as catalyst precursors. At the same time, it was observed
that some portion of the catalyst precursor is precipitated spontane-
ously at the end of the reaction. As shown in Scheme 1, it is
suggested that the acidification by formate generation causes the
transformation from the water-soluble deprotonated form into the
water-insoluble monoprotonated [Cp*Ir(HL)Cl] and fully protonated
form.9a This finding prompted us to examine the additional role
played by an oxyanion, that is, the polarity change of the catalyst
precursor by the acid-base equilibrium. Herein, we report the
catalyst recycling by self-precipitation/filtration without waste
generation in the CO2 conversion.
The complex 1 shows the pH-dependent absorption spectral
change, due to the acid-base equilibrium of two acidic protons9,10
(see Supporting Information). However, besides the pH-dependent
water-solubility of 1, it should be noted that 1 is almost completely
insoluble in acidic aqueous solutions. On the other hand, the pH
9
13118
J. AM. CHEM. SOC. 2005, 127, 13118-13119
10.1021/ja054236k CCC: $30.25 © 2005 American Chemical Society

C O M M U N I C A T I O N S
Table 1. Batchwise Catalyst Recycling in the Conversion of CO2
into Formate Using 1‚Cl
loaded/recovered
[Ir] (ppm)a
recovery
leaching
[Ir] (ppm)a
final concn
of formate (M)b
cycle
efficiency (%)
1
2
3
4
9.0
8.4
7.7
7.0
0.11
0.22
0.42
0.61
0.105
0.104
0.103
0.103
93
92
91
a The Ir concentration was measured by ICP-MS analysis. b For all four
cycles, no bicarbonate was detected.
Figure 1. Time course of formate concentration for the hydrogenation of
bicarbonate catalyzed by 1‚Cl (0.05 mM) at 6 MPa and 60 °C in (a) 0.1 M,
(b) 0.5 M, and (c) 1.0 M aqueous KOH solutions.
to 75%. It appears that the thermal stress and exposure to air lead
to an increase in catalyst leaching.
Scheme 2. Recycling System of the Conversion of CO2/H2 and
KOH into HCO2K Using 1‚Cl in Water (g: Gas Phase, l: Liquid
Phase, s: Solid Phase, Cat.(d): Deprotonated Form, Cat.(p):
Protonated Form)
In conclusion, we have found that the CO2 conversion using the
complex 1 is an environmentally benign process whose significant
features are as follows: (i) catalyst recycling by self-precipitation/
filtration, (ii) waste-free process, (iii) the easy isolation of the
product, (iv) high efficiency under relatively mild conditions, and
(v) aqueous catalysis without the use of organic materials.
Furthermore, we have demonstrated the third additional role of an
oxyanion on the catalyst ligand, that is, the catalyst recovery by
acid-base equilibrium, in addition to the water-solubility by its
polarity and the catalyst activation by its electron-donating ability.
The attractive features of an oxyanion on the catalyst ligand may
hold significantly broader implications for the design of new
homogeneous catalysts.
Supporting Information Available: Typical procedures for catalyst
recycling and absorbance change as a function of pH for iridium
complex 1 (PDF). This material is available free of charge via the
Internet at http://pubs.acs.org.
To promote the precipitation of the catalyst precursor by the pH
decrease in the solution, the reaction in 0.1 M KOH solution was
stirred for a prolonged period of time (20 h). By filtration of the
obtained suspension, the catalyst precursor was recovered (vide
infra). As expected, iridium leaching into the filtrate decreased
drastically (0.11 ppm). The final concentration of formic acid (0.105
M) slightly exceeded the initial concentration of the added KOH,
and the pH of the filtrate was found to be 5.5.13,14 The evaporation
of the filtrate and drying under vacuum at 100 °C gave pure
potassium formate (g98% pure).15,16 It is interesting to note that
the three components (i.e., catalyst, product (formate), and solvent
(H2O)) can be easily separated by conventional filtration and
evaporation (Scheme 2). Sodium formate was also obtained by the
same procedure from NaOH with efficient recovery and conversion.
These results are indicative of the possibility of the catalyst
recycling in the conversion of CO2 into formate without waste
generation.
Finally, the reusability of the catalyst was examined in the
batchwise cycle. The above-mentioned recovered catalyst precursor
was dissolved in a 0.1 M degassed aqueous KOH, in which ICP-
MS analysis of the solution indicated that iridium could be
efficiently recovered (93% recovery; the loss of iridium was due
to sampling for assay (2%) and handling losses) (Table 1). While
the recovered catalyst retained high catalytic activity for all four
cycles, leaching increased and recovery decreased with an increased
recycling of the catalyst. The batch experiments require the careful
handling of reaction solutions containing air-sensitive active species,
but some exposure to air cannot be excluded. However, this problem
can be solved if a continuous closed system or an integrated mem-
brane reactor would be used. The thermal degradation was also
observed. When the reaction was carried out at 80 °C for 10 h,
iridium leaching increased to 0.9 ppm and recovery decreased
References
(1) Thematic Issue: Recoverable Catalysts and Reagents; Gladysz, J. A., Ed.;
Chem. ReV. 2002, 102, No. 10.
(2) Tzschucke, C. C.; Markert, C.; Bannwarth, W.; Roller, S.; Hebel, A.; Haag,
R. Angew. Chem., Int. Ed. 2002, 41, 3964-4000.
(3) Cole-Hamilton, D. J. Science 2003, 299, 1702-1706.
(4) Dijkstra, H. P.; Van Klink, G. P. M.; Van Koten, G. Acc. Chem. Res.
2002, 35, 798-810.
(5) Bayon, J. C.; Real, J.; Claver, C.; Polo, A.; Ruiz, A. J. Chem. Soc., Chem.
Commun. 1989, 1056-1057.
(6) Dioumaev, V. K.; Bullock, R. M. Nature 2003, 424, 530-532.
(7) Jessop, P. G.; Joo, F.; Tai, C.-C. Coord. Chem. ReV. 2004, 248, 2425-
2442 and references therein.
(8) Tai. C.-C.; Chang, T.; Roller, B.; Jessop, P. G. Inorg. Chem. 2003, 42,
7340-7341.
(9) (a) Himeda, Y.; Onozawa-Komatsuzaki, N.; Sugihara, H.; Arakawa, H.;
Kasuga, K. Organometallics 2004, 23, 1480-1483; (b) Stud. Surf. Sci.
Catal. 2004, 153, 263-266.
(10) Giordano, P. J.; Bock, C. R.; Wrighton, M. S. J. Am. Chem. Soc. 1978,
100, 6960-6965.
(11) Sandrini, D.; Maestri, M.; Ziessel, R. Inorg. Chim. Acta 1989, 163, 177-
180.
(12) Bonilla, R. J.; James, B. R.; Jessop, P. G. Chem. Commun. 2000, 941-
942.
(13) It is known that the hydrogenation of CO2 proceeds under slightly acidic
conditions: Laurenczy, G.; Joo, F.; Nadasdi, L. Inorg. Chem. 2000, 39,
5083-5088.
(14) We have already found that the rhodium analogue [Cp*Rh(bpy)Cl]Cl (bpy
) 2,2′-bipyridine) can hydrogenate CO2 catalytically with low TON (up
to 124) under acidic conditions: Himeda, Y.; Onozawa, N.; Sugihara,
H.; Arakawa, H.; Kasuga, K. Jpn. Kokai Tokkyo Koho JP 2004224715,
2004. Independently, Ogo et al. reported similar results using the ruthenium
analogues: Hayashi, H.; Ogo, S.; Fukuzumi, S. Chem. Commun. 2004,
2714-2715.
(15) Bicarbonate was not detected by 13C NMR analysis, in which the spectral
signal-to-noise (S/N) ratio was over 50.
(16) Formic acid may decompose during drying: mp 166-169 °C (Aldrich
catalog gives 165-168 °C). Anal. Calcd for CHO2K: C, 14.28; H, 1.20.
Found: C, 14.25; H, 1.16.
JA054236K
9
J. AM. CHEM. SOC. VOL. 127, NO. 38, 2005 13119