Kinetics and Mechanisms of Reduction of Protons and Carbon Dioxide Catalyzed by Metal Complexes and Nanoparticles

Abstract

Kinetics and mechanisms of reduction of protons and CO₂ catalyzed by metal complexes and nanoparticles have been discussed in this chapter. Kinetic studies including deuterium kinetic isotope effects on heterogeneous catalysts for hydrogen evolution by proton reduction have been demonstrated to provide essential mechanistic information on bond cleavage and formation associated with electron transfer. The rate-determining steps in the catalytic cycles are clarified by kinetic studies, providing valuable information on observable intermediates. The most important intermediates in the catalytic reduction of protons and CO₂ are metal-hydride complexes, which can reduce protons and CO₂ to produce hydrogen and formic acid, respectively. The catalytic interconversion between hydrogen and a hydrogen storage compound has been made possible by changing pH, providing a convenient hydrogen-on-demand system in which hydrogen gas can be stored as a liquid (e.g., formic acid) or solid form (NADH) and hydrogen can be produced by the catalytic decomposition of the hydrogen storage compound.

11.1 Introduction

The global annual energy consumption is increasing rapidly, whereas fossil fuels, which are currently the primary source of the energy, will be depleted eventually in the future. In addition, the burning of fossil fuels releases large amounts of carbon dioxide (CO₂) to the atmosphere, leading to global warming. Before the depletion of fossil fuels, which are the products of photosynthesis, artificial photosynthesis should be realized to produce solar fuels and to fix CO₂ using solar energy, which is the most abundant on the earth [1–4]. In an ideal artificial photosynthesis, water is split using solar energy into hydrogen (H₂) and dioxygen (O₂), which in turn can be converted into water releasing its energy as electricity in H₂ fuel cells [5–10]. Once H₂ is formed from water, CO₂ can be reduced by H₂ to produce carbon monoxide or formate, which can be further reduced to methanol and methane [11–16]. Such reduction of CO₂ by H₂ provides the possibility of storing solar energy as these C1 compounds, contributing to reduced emission of CO₂. There have so far been many reviews on each step of artificial photosynthesis, i.e., light harvesting and charge separation [17–23], proton reduction [24–28], CO₂ reduction [29–33], and water oxidation [34–42]. However, the detailed catalytic mechanisms of reduction of protons for hydrogen evolution and reduction of CO₂ have yet to be fully clarified. Kinetic studies certainly help in understanding catalytic mechanisms of proton reduction and CO₂ reduction, which are catalyzed by metal complexes and nanoparticles. Thus, in this review, we have chosen to focus on kinetics and mechanisms of reduction of protons and CO₂ catalyzed by metal complexes and nanoparticles.

11.2 Mechanisms of Catalytic Hydrogen Evolution

11.2.1 Cobalt Hydride Complexes

Platinum is currently used as the most efficient catalyst for the reduction of protons to H₂ [43]. Because of the scarcity and high cost of platinum, replacement of platinum by more earth-abundant metals such as cobalt as proton reduction catalysts has attracted much attention [44–48]. In order to develop efficient proton reduction catalysts, it is quite important to elucidate mechanisms of the catalytic proton reduction. The mechanism of the proton reduction by Co complexes has been clarified using a dinuclear Co complex with bis(pyridyl)pyrazolato (bpp⁻) and terpyridine (trpy) ligands, [Co^III ₂(trpy)₂(μ-bpp)(OH)(OH₂)](PF₆)₄(1(PF₆)₄, Fig. 11.1) [49].

Fig. 11.1

Structure of [Co^III ₂(trpy)₂(μ-bpp)(OH)(OH₂)]⁴⁺ (1 ⁴⁺)

The dinuclear cobalt(III) complex (1 ⁴⁺) undergoes step-by-step reduction by decamethylcobaltocene (Co(Cp^*)₂, Cp^* = η⁵-pentamethylcyclopentadienyl) to 1, because the one-electron oxidation potential of Co(Cp^*)₂ (E _ox = −1.53 V vs. SCE) is lower than the one-electron reduction potential of 1 ⁺ (E _red = −1.09 V vs. SCE) [49]. Addition of 4 equiv. of Co(Cp^*)₂ to a deaerated MeCN solution containing 1 ⁴⁺ resulted in rapid formation of 1, which exhibits visible and NIR absorption bands at 560 and 1,050 nm [49]. Addition of 10 equiv. of CF₃COOH to the MeCN solution of 1 resulted in the two-step decay of absorbance at 560 nm due to 1 (Fig. 11.2a) [49]. The two-step reaction of 1 with CF₃COOH suggests that the protonation of 1 affords a hydride complex, [(Co^III–H)(Co^III–H)]²⁺, which is in equilibrium with 1, followed by the reaction of the hydride complex with protons to produce H₂ [49]. The first-order dependence of k _obs with respect to the concentration of CF₃COOH in the second step indicates that the protonation of the Co(III)–H moiety is the rate-determining step to produce H₂ [49].

Fig. 11.2

Time profiles of absorbance at 560 nm due to (a) the Co^ICo^I complex (1, 0.087 mM) and (b) the Co^IICo^I complex (1 ⁺, 0.087 mM) with CF₃COOH (0.87 mM) in deaerated MeCN at 298 K. Insets: Plots of k _obs vs. concentration of CF₃COOH for the second step reaction of (a) 1 and (b) 1 ⁺ with CF₃COOH (Reprinted with the permission from Ref. [49]. Copyright 2011 American Chemical Society)

When cobaltocene (Co(Cp)₂, Cp = η⁵-cyclopentadienyl) was used as a reductant, 1 ⁺ was obtained by the three-electron reduction of 1 ⁴⁺ with Co(Cp)₂, because the one-electron oxidation potential of Co(Cp)₂ (E _ox = −0.9 V vs. SCE) is more negative than the E _red value of 1 ²⁺ (−0.78 V vs. SCE) but less negative than the E _red value of 1 ⁺ (−1.09 V vs. SCE) [49]. The reaction of 1 ⁺ with CF₃COOH also exhibited a two-step decay (Fig. 11.2b) [49].

Based on the two-step kinetics, the reaction mechanism of H₂ production from 1 ⁺ is shown in Scheme 11.1 [49]. Three-electron reduction of 1 ⁴⁺ by 3 equiv. of Co(Cp)₂ occurs to produce 1 ⁺ (Eq. 11.1). 1 ⁺ is protonated by CF₃COOH to produce the hydride complex ([Co^IICo^III–H]²⁺), which is in equilibrium with 1 ⁺ (Eq. 11.2, the first step in Fig. 11.2b). The formation of hydride complex was confirmed by the ¹H NMR spectra which exhibit a typical Co(III)–H peak at δ = −8.64 ppm [49]. The hydride complex reacts with protons to produce H₂ and 1 ³⁺ (Eq. 11.3). This is the rate-determining step for the H₂ production, because the k _obs values for the second step are proportional to the proton concentration (inset of Fig. 11.2b) [49]. 1 ³⁺ is reduced by 1 ⁺ to produce two equivalent 1 ²⁺ (Eq. 11.4) [49]:

Scheme 11.1

Reaction mechanism of H₂ production with [Co^IICo^I]⁺

$$ {\left[{\mathrm{Co}}^{\mathrm{I}\mathrm{I}\mathrm{I}}{\mathrm{Co}}^{\mathrm{I}\mathrm{I}\mathrm{I}}\right]}^{4+}+3\mathrm{C}\mathrm{o}{\left(\mathrm{C}\mathrm{p}\right)}_2\to {\left[{\mathrm{Co}}^{\mathrm{I}\mathrm{I}}{\mathrm{Co}}^{\mathrm{I}}\right]}^{+} + 3\mathrm{C}\mathrm{o}{\left(\mathrm{C}\mathrm{p}\right)}_2^{+} $$

(11.1)

$$ {\left[{\mathrm{Co}}^{\mathrm{I}\mathrm{I}}{\mathrm{Co}}^{\mathrm{I}}\right]}^{+}+{\mathrm{H}}^{+}\rightleftharpoons {\left[{\mathrm{Co}}^{\mathrm{I}\mathrm{I}}{\mathrm{Co}}^{\mathrm{I}\mathrm{I}\mathrm{I}}\hbox{--} \mathrm{H}\right]}^{2+} $$

(11.2)

$$ {\left[{\mathrm{Co}}^{\mathrm{II}}{\mathrm{Co}}^{\mathrm{II}\mathrm{I}}\hbox{--} \mathrm{H}\right]}^{2+}+{\mathrm{H}}^{+}\to {\left[{\mathrm{Co}}^{\mathrm{II}}{\mathrm{Co}}^{\mathrm{II}\mathrm{I}}\right]}^{3+}+{\mathrm{H}}_2 $$

(11.3)

$$ {\left[{\mathrm{Co}}^{\mathrm{I}\mathrm{I}}{\mathrm{Co}}^{\mathrm{I}}\right]}^{+}+{\left[{\mathrm{Co}}^{\mathrm{I}\mathrm{I}}{\mathrm{Co}}^{\mathrm{I}\mathrm{I}\mathrm{I}}\right]}^{3+}\to 2{\left[{\mathrm{Co}}^{\mathrm{I}\mathrm{I}}{\mathrm{Co}}^{\mathrm{I}\mathrm{I}}\right]}^{2+} $$

(11.4)

$$ \hbox{--} \mathrm{d}\left[{\left[{\mathrm{Co}}^{\mathrm{I}\mathrm{I}}{\mathrm{Co}}^{\mathrm{I}}\right]}^{+}\right]/\mathrm{d}t=k\left[{\mathrm{H}}^{+}\right]\left[{\left[{\mathrm{Co}}^{\mathrm{I}\mathrm{I}}{\mathrm{Co}}^{\mathrm{I}}\right]}^{+}\right]/\left(1 + K\left[{\mathrm{H}}^{+}\right]\right) $$

(11.5)

According to Scheme 11.1, the rate of decay of 1 ⁺ is given by Eq. 11.5, where k is the rate constant of protonation of the hydride complex to produce H₂ (Eq. 11.3) and K is the protonation equilibrium constant of 1 ⁺ to produce the hydride complex, [Co^IICo^III–H]²⁺ (Eq. 11.2). Based on the kinetic results in Fig. 11.2, the k and K values of 1 ⁺ at 298 K were determined to be 2.9 M⁻¹ s⁻¹ and 5.3 × 10² M⁻¹, respectively [49]. Similarly the k and K values of 1 at 298 K were also determined to be 3.3 M⁻¹ s⁻¹ and 1.1 × 10³ M⁻¹, respectively [49]. The K value of 1 is twice larger than that of 1 ⁺, because 1 has two Co^I sites as compared with 1 ⁺ which has one Co^I site. The similar k values between 1 ⁺ and 1 suggest that the two Co^I sites in 1 act rather independently in the reaction with proton.

A cobalt tetraaza-macrocyclic complex [Co^III(CR)Cl₂]⁺ (CR = 2,12-dimethyl-3,7,11,17-tetraazabicyclo(11.3.1)-heptadeca-1(17),2,11,13,15-pentaene) has been reported to act as an efficient proton reduction catalyst in photocatalytic hydrogen evolution with ascorbate (HA⁻) and ascorbic acid (H₂A) as an electron donor and a proton donor, respectively, and [Ru(bpy)₃]²⁺ as a photocatalyst in water [50]. The catalytic activity and stability of [Co^III(CR)Cl₂]⁺ were higher than that of other cobalt complexes such as cobaloxime derivatives [51–53] to afford a high turnover number (TON = 1,000) [50]. The Co(III) complex was reduced with HA⁻ to produce the Co(II) complex [50]. The Co(II) complex was further reduced by electron transfer from the excited state of [Ru(bpy)₃]²⁺ ([Ru(bpy)₃]^2+* where ^* denotes the excited state) to produce the Co(I) complex, which reacts with protons to yield H₂ and the Co(II) complex similar to 1 ⁺ in Scheme 11.1 [50]. In this case, however, formation of the Co(III)-hydride complex has not been detected [50]. The detailed photocatalytic mechanism of hydrogen evolution is discussed in the next section.

11.2.2 Rhodium Hydride Complexes

A water-soluble rhodium-aqua complex, [Rh^III(Cp^*)(bpy)(H₂O)](SO₄) (2(SO₄), bpy = 2,2′-bipyridine), acts as an efficient catalyst for H₂ evolution from HCOOH in an aqueous solution at 298 K [54]. The kinetic study revealed the catalytic mechanism of the catalytic decomposition of HCOOH to H₂ and CO₂ with 2(SO₄) as shown in Scheme 11.2 [54]. The rate of H₂ evolution increased linearly with increasing concentrations of 2(SO₄) as shown in Fig. 11.3a. On the other hand, the TOF value increased with increasing [HCOOH] to reach a limiting value as shown in Fig. 11.3b. Such a saturation behavior indicates that the formation of the formate complex is in equilibrium with HCOO⁻, followed by β-hydrogen elimination from the formate complex to produce the Rh^III-hydride complex, which becomes the rate-determining step at large concentrations of HCOO⁻. The formate complex ([Rh^III(Cp^*)(OC(O)H)(bpy)]⁺) was detected by the electrospray ionization (ESI)-mass spectrometry at m/z = 439.2 [54]. When pH was changed, the maximum TOF value was obtained at pH 3.9, which corresponds to pK _a of HCOOH. No catalytic reactivity was observed at pH higher than pK _a of 2(SO₄), indicating that the hydroxo complex [Rh^III(Cp^*)(OH)(bpy)]⁺ has no catalytic reactivity [54].

Scheme 11.2

Catalytic mechanism of decomposition of HCOOH by 2 in D₂O

Fig. 11.3

(a) Plot of rate of H₂ evolution vs. the concentration of 2(SO₄) in the decomposition of HCOOH/HCOONa (1.7 M) catalyzed by 2(SO₄) in deaerated H₂O at pH 4.1 at 293 K. (b) Plot of TOFs vs. the concentration of HCOOH/HCOONa in the decomposition of HCOOH/HCOONa catalyzed by 2(SO₄) (7.0 mM) in deaerated H₂O at pH 3.8 at 298 K (Reproduced from Ref. [54] by permission of John Wiley & Sons Ltd)

When HCOOH was replaced by DCOOH, the catalytic decomposition of DCOOH in H₂O with 2(SO₄) afforded not only HD but also H₂ [54]. A significant deuterium kinetic isotope effect was observed in the catalytic decomposition of DCOOH because the rate-determining step is the β-deuterium elimination from the formate complex to produce the Rh^III-D complex (vide supra) [54]. The formation of H₂ suggests that the deuteride species ([Rh^III(Cp^*)(D)(bpy)]⁺), formed by deuteride transfer from DCOO⁻ to [Rh^III(Cp^*)(bpy)(H₂O)]²⁺, undergoes rapid H/D exchange with H₂O to afford [Rh^III(Cp^*)(H)(bpy)]⁺ that reacts with H⁺ to produce H₂. When the decomposition of HCOOH was performed with 2(SO₄) in D₂O, D₂ was formed as a major product (73 %) together with HD (24 %) and H₂ (3 %) [54]. In this case, hydride transfer from HCOO⁻ to [Rh^III(Cp^*)(bpy)(H₂O)]²⁺ occurs to afford [Rh^III(Cp^*)(H)(bpy)]⁺ that undergoes H/D exchange with D⁺ to produce [Rh^III(Cp^*)(D)(bpy)]⁺ [54]. Then, the reaction of [Rh^III(Cp^*)(D)(bpy)]⁺ with D⁺ yields D₂ as the main product [54]. The unexchanged hydride complex [Rh^III(Cp^*)(H)(bpy)]⁺ reacts with D⁺ and a small amount of H⁺ derived from HCOOH to yield HD and a small amount of H₂, respectively [54]. Thus, rapid H/D exchange between the hydride (or deuteride) species and proton (or deuteron) occurs as shown in Scheme 11.2, suggesting that the formal hydride species has a protic character.

The protic character of the hydride species was confirmed by formation of [Rh^I(Cp^*)(bpy)] by deprotonation from [Rh^III(Cp^*)(H)(bpy)]⁺ with a base (NaOH) [54]. Such a protic character of metal-hydride species was reported for the corresponding Ir complex with the same ligand as the Rh complex, i.e., [Ir^III(Cp^*)(H)(bpy)]⁺, which undergoes efficient H/D exchange with deuteron [55, 56]. DFT calculations showed that the positive charge of metal-hydride (M-H) (+0.571) was larger for [Ir^III(Cp^*)(H)(bpy)]⁺ as compared to the value for [Rh^III(Cp^*)(H)(bpy)]⁺ (+0.481) [54].

The Rh(III) complex (2(SO₄)) can also be used as a proton reduction catalyst in photocatalytic hydrogen evolution with ascorbate (HA⁻) as an electron donor and [Ru(bpy)₃]²⁺ as a photocatalyst [57]. The photocatalytic mechanism is shown in Scheme 11.3, where photoinduced electron transfer from HA⁻ to [Ru(bpy)₃]^2+* (* denotes an excited state) occurs to produce [Ru(bpy)₃]⁺, which reduces [Rh^III(Cp^*)(bpy)]²⁺ to [Rh^II(Cp^*)(bpy)]⁺, which was detected as a transient absorption band at 750 nm in Fig. 11.4a [57]. Disproportionation of [Rh^II(Cp^*)(bpy)]⁺ occurs to produce Rh^I(Cp^*)(bpy) and [Rh^II(Cp^*)(bpy)]⁺ as indicated by the second-order decay of absorbance at 750 nm due to [Rh^II(Cp^*)(bpy)]⁺ (see the second-order plot in inset of Fig. 11.4b) [57]. Rh^I(Cp^*)(bpy) is protonated to produce the hydride complex ([Rh^III(Cp^*)(H)(bpy)]⁺), which reacts with proton to produce H₂, accompanied by regeneration of [Rh^III(Cp^*)(bpy)]²⁺ [57]. In the same manner, when a heterodinuclear iridium–ruthenium complex [Ir^III(Cp^*)(H₂O)(bpm)Ru^II(bpy)₂](SO₄)₂ (3(SO₄)₂, bpm = 2,2-bipyrimidine) was used in place of 2(SO₄), photocatalytic H₂ evolution was confirmed to proceed via disproportionation of [Ir^II(Cp^*)(H₂O)(bpm)Ru^II(bpy)₂]³⁺. Thus, disproportionation of [Rh^II(Cp^*)(bpy)]⁺ or [Ir^II(Cp^*)(H₂O)(bpm)Ru^II(bpy)₂]³⁺ is the key step to convert one-electron process induced by one photon to the two-electron process for H₂ evolution (Scheme 11.3). This shows sharp contrast to the case of [Co^III(CR)Cl₂]⁺, which is reduced by HA⁻ to produce the Co(II) complex, which is further reduced to the Co(I) complex via photoinduced electron transfer from [Ru(bpy)₃]^2+* to the Co(II) complex (vide supra) [50].

Scheme 11.3

Catalytic mechanism of hydrogen evolution with 2 or 3 as a proton reduction catalyst, ascorbate as an electron donor, and [Ru(bpy)₃]²⁺ as a photocatalyst

Fig. 11.4

(a) Transient absorption spectra of 2(SO₄) (1.6 × 10⁻⁴ M) after laser excitation of [Ru(bpy)₃]²⁺ (8.0 × 10⁻⁵ M) at λ = 455 nm in the presence of H₂A (0.8 M) and NaHA (0.3 M) in deaerated H₂O at pH 3.6 at 298 K. (b) Decay time profile of absorbance at λ = 750 nm due to [Rh^II(Cp^*)(bpy)]⁺. Inset: Second-order plot (Reproduced from Ref. [57] by permission of John Wiley & Sons Ltd)

The same type of photocatalytic H₂ evolution with ascorbate and [Ru(bpy)₃]²⁺ occurs using [Rh^III(dmbpy)₂Cl₂]Cl (dmbpy = 4,4′-dimethyl-2,2′-bipyridine) as a proton reduction catalyst [58]. The catalytic reactivity of [Rh^III(dmbpy)₂Cl₂]Cl was higher than that of [Rh^III(Cp^*)(bpy)]²⁺ to afford high TON and TOF values (1,010 and 857 h⁻¹) at pH 4.0 [58]. The high catalytic activity may result from formation of colloidal rhodium nanoparticles during the photocatalytic reaction, which are known to promote the reduction of protons into H₂ [59]. This possibility was ruled out, because the addition of a large excess of mercury had no significant effect on the catalytic activity of [Rh^III(dmbpy)₂Cl₂]Cl. Mercury is known to form amalgam with colloidal metal or to adsorb to nanoparticular metal catalysts, and mercury poisoning has been reported for rhodium colloids [60].

11.2.2.1 Iridium Hydride Complexes

A heterodinuclear iridium–ruthenium complex [Ir^III(Cp^*)(H₂O)(bpm)Ru^II(bpy)₂](SO₄)₂ (3(SO₄)₂, bpm = 2,2-bipyrimidine) also acts as an efficient catalyst for H₂ evolution from HCOOH in an aqueous solution at 298 K [61]. The maximum TOF value (426 h⁻¹) was obtained at pH 3.8 which agrees with the pK _a value of HCOOH [61]. The TOF value is much higher than that of [Rh^III(Cp^*)(bpy)(H₂O)](SO₄) under the same experimental conditions (TOF = 27 h⁻¹) [61]. The catalytic mechanism is shown in Scheme 11.4, which is similar to the case of [Rh^III(Cp^*)(bpy)(H₂O)](SO₄) in Scheme 11.2 [61]. The reaction of 3 ⁴⁺ with HCOO⁻ affords the formate complex ([Ir^III(Cp^*)(O(CO)H)(bpm)Ru^II(bpy)₂]³⁺), followed by β-hydrogen elimination to give the Ir–hydride complex ([Ir^III(Cp^*)(H)(bpm)Ru^II(bpy)₂]³⁺) which reacts with H⁺ to produce H₂, accompanied by regeneration of 3 ⁴⁺ [61]. Rapid H/D exchange between the hydride (or deuteride) species and proton (or deuteron) also occurs as the case of 2 ²⁺ in Scheme 11.2 [61]. In Scheme 11.4, however, the rate-determining step in the overall hydrogen evolution reaction is not the β-hydrogen elimination step but the reaction of the hydride complex with H⁺ to evolve H₂ at pH 3.8 [61]. The Arrhenius plots for TOF in D₂O (red circles) vs. H₂O (black circles) in Fig. 11.5 afforded A _H/A _D = 3.1 × 10⁻⁵, E _a(D)−E _a(H) = 8.2 kcal mol⁻¹, and an unusually large KIE value at 298 K (KIE = 40) [61]. Such values for the Arrhenius parameters A _H/A _D << 1 and E _a(D)−E _a(H) > 1.2 kcal mol⁻¹ together with a large KIE value at 298 K (KIE > 9) are generally taken to unambiguously demonstrate the involvement of tunneling [62–65]. Because a protic character of metal-hydride species is more enhanced for the Ir–hydride complex as compared with the Rh-hydride complex, the reaction of the Ir–hydride complex with proton becomes the rate-determining step.

Scheme 11.4

Catalytic mechanism of decomposition of HCOOH by 3 in D₂O

Fig. 11.5

Arrhenius plots of TOF for the decomposition of HCOOH/HCOONa (3.1 M) catalyzed by 3(SO₄)₂ (0.3 mM) in deaerated H₂O (closed circles) or D₂O (open circles) at pH 3.8 or pD 3.8, respectively (Reprinted with the permission from Ref. [61]. Copyright 2010 American Chemical Society)

The catalytic activity for hydrogen evolution from formic acid was further enhanced by using a C^N cyclometalated organoiridium complex, [Ir^III(Cp^*){4-(1H-pyrazol-1-yl-κN ²)benzoic acid–κC³}(H₂O)]₂SO₄ ([4]₂•SO₄, Fig. 11.6), as a catalyst with TOF value over 2,000 h⁻¹ at 298 K (Fig. 11.7) [66]. The catalytic mechanism is shown in Scheme 11.5 [66].

Fig. 11.6

ORTEP drawing of 4. Hydrogen atoms are omitted for clarity (Reprinted with the permission from Ref. [68]. Copyright 2012 American Chemical Society)

Fig. 11.7

Plot of TOF vs. concentration of HCOOH and HCOOK mixture (HCOOH/HCOOK), i.e., [HCOOH] + [HCOOK] in the decomposition of HCOOH/HCOOK catalyzed by [Ir ^A–H₂O]⁺ (0.20 mM) in deaerated H₂O at pH 2.8 at 298 K (Reproduced from Ref. [66] by permission of The Royal Society of Chemistry)

Scheme 11.5

Catalytic mechanism of H₂ evolution and decomposition of HCOO⁻ with 4 in H₂O

As pH was increased, the Ir(III) complex [Ir ^A–H₂O]⁺ released protons from the carboxy group and the aqua ligand to form the corresponding benzoate complex [Ir ^B–H₂O]⁰ and the hydroxo complex [Ir ^B–OH]⁻, respectively (Scheme 11.6). The pK _a values of [Ir ^A–H₂O]⁺ and [Ir ^B–H₂O]⁰ were determined from the spectral titration to be pK _a1 = 4.0 and pK _a2 = 9.5, respectively [66]. The saturation behavior of TOF of hydrogen evolution with increasing concentration of [HCOOH] + [HCOOK] at pH 2.8 (Fig. 11.7) indicates that hydrogen is produced via the formate complex of [Ir ^A–H₂O]⁺, followed by β-elimination to produce the hydride complex, which reacts with proton to produce H₂ (Scheme 11.6) [66]. The activation energy was determined to be 18.9 kcal mol⁻¹, which is much smaller than the activation energy of the decomposition of formic acid without catalysts (78 kcal mol⁻¹) [67].

Scheme 11.6

Acid–base equilibria of iridium aqua complexes

The C^N cyclometalated organoiridium complex [Ir ^A–H₂O]⁺ can also act as an efficient catalyst for hydrogen evolution from NADH (dihydronicotinamide adenine dinucleotide), which is a natural electron and proton source in respiration and CO₂ fixation [68], in water at pH 4.1 [69]. NADH has been frequently used as an electron and proton source in photocatalytic hydrogen evolution with a photocatalyst and a proton reduction catalyst [70–73]. Under acidic conditions, NADH can reduce thermally proton to produce H₂ and NAD⁺ by the catalysis of [Ir ^A–H₂O]⁺ (Eq.  11.6 ). The catalytic cycle is shown in Scheme 11.7 [69]. Under acidic

Scheme 11.7

Catalytic mechanism of interconversion between H₂ and NADH with 4 in H₂O

$$ \begin{array}{l}\mathrm{NADH}+{\mathrm{H}}^{+}\to {\mathrm{NAD}}^{+}+{\mathrm{H}}_2\\ {}\kern3.1em {\left[{\mathbf{Ir}}^{\mathbf{A}}-{\mathrm{H}}_2\mathrm{O}\right]}^{+}\end{array} $$

(11.6)

conditions, hydride transfer from NADH to [Ir ^A–H₂O]⁺ occurs to produce NAD⁺ and the Ir(III)-hydride complex [Ir ^A–H]⁰, which reacts with H₃O⁺ to produce H₂, accompanied by regeneration of [Ir ^A–H₂O]⁺ [69]. Under basic conditions, however, the catalytic cycle was reversed, when H₂ can reduce the deprotonated carboxylate form [Ir ^B–H₂O]⁰ to produce the Ir(III)-hydride complex, which reduces NAD⁺ to NADH, accompanied by regeneration of the deprotonated form of [Ir ^B–H₂O]⁰ [69]. Thus, interconversion between NADH and H₂ at ambient pressure and temperature can be efficiently catalyzed by [Ir ^A–H₂O]⁺ and [Ir ^B–H₂O]⁰ depending on pH.

According to Scheme 11.6, the Ir(III) complex [Ir ^A–H₂O]⁺ is converted to the hydroxo complex [Ir ^B–OH]⁻ at pH 13.6. When ethanol was added to an aqueous solution of [Ir ^B–OH]⁻ at pH 13.6, hydride transfer from ethanol to [Ir ^B–OH]⁻ occurred to produce acetaldehyde and the hydride complex [Ir ^B–H]⁻ [74]. When CD₃CD₂OH in place of CH₃CH₂OH was added to an aqueous solution of [Ir ^B–OH]⁻, a kinetic deuterium isotope effect (KIE) for the formation of [Ir ^B–D]⁻ was observed to be k _H/k _D = 2.1 [74]. The observation of KIE indicates that the reaction of ethanol with [Ir ^B–OH]⁻ involves the C–H bond cleavage. Thus, the β-hydrogen elimination of the ethoxy complex which is produced by the replacement of a hydroxy (OH) ligand of [Ir ^B–OH]⁻ by a ethoxy (CH₃CH₂O) ligand, may be the rate-determining step for formation of the hydride complex [Ir ^B–H]⁻ (Scheme 11.8). Other alcohols can also reduce [Ir ^B–OH]⁻ to produce the hydride complex [Ir ^B–H]⁻ [74].

Scheme 11.8

Dehydrogenation reaction of alcohols by [Ir^B–OH]⁻ in H₂O

The hydride complex [Ir ^B–H]⁻ is stable at pH 14. When pH was decreased to 0.8 by adding H₂SO₄, however, the hydride complex [Ir ^B–H]⁻ was converted to an aqua complex [Ir ^A–H₂O]⁺ as shown by Fig. 11.8a, accompanied by evolution of hydrogen (H₂) [74]. The conversion between the hydride complex [Ir ^B–H]⁻ and the aqua complex [Ir ^A–H₂O]⁺ accompanied by H₂ evolution was repeated by alternate change in pH between 12 and 2 in the presence of excess amount of ethanol as shown in Fig. 11.8b (Scheme 11.9) [74]. Without changing pH, however, no catalytic H₂ evolution from ethanol occurred with [Ir ^A–OH]⁰ [74].

Fig. 11.8

(a) UV-vis absorption spectral change of an aqueous solution of [Ir ^B–OH]⁻ (0.12 mM) and ethanol (82 mM) by alternate change in pH. (b) Changes of absorbance at λ = 350 nm due to the formation of a hydride complex [Ir ^B–H]⁻ in the reaction of [Ir ^B–OH]⁻ (0.12 mM) with ethanol (82 mM) in water (pH 11.8–12.2) and due to the hydrogen evolution in the reaction of the hydride complex [Ir ^B–H]⁻ with proton in water at 298 K (pH 2.0–3.3) by adding an aqueous solution of H₂SO₄ (5.0 M) or NaOH (5.0 M) (Reprinted with the permission from Ref. [74]. Copyright 2012 American Chemical Society)

Scheme 11.9

Catalytic mechanism of H₂ evolution from ethanol with [Ir^A–H₂O]⁺ in H₂O

Photoirradiation of the hydride complex [Ir ^B–H]⁻ resulted in the conversion to the [C,C] cyclometalated complex [Ir ^C–H]⁻ (Scheme 11.10) [74]. In contrast to the [C,N] cyclometalated Ir–hydride complex [Ir ^B–H]⁻, the [C,C] cyclometalated Ir–hydride complex [Ir ^C–H]⁻ can react with water to produce H₂ under basic conditions as shown in Fig. 11.9. The turnover number (TON) of H₂ evolution from isopropanol with [Ir ^C–H]⁻ (Eq. 11.7) increases linearly with time to reach 3.3 (2.5 h), whereas [Ir ^B–H]⁻ has no catalytic reactivity even at elevated temperature at 323 K (Fig. 11.9). TON for H₂ evolution from isopropanol with [Ir ^C–H]⁻ increases with increasing temperature to be 26 (1.0 h) at 353 K (Fig. 11.9) [74]. The enhanced catalytic activity of [Ir ^C–H]⁻ results from the electronic donating effect of phenylpyrazole ligand on the metal center with a [C,C] cyclometalated iridium as indicated by the upfield shift of a hydride signal bonded to Ir^III center (δ = −17.48) as compared with that of [Ir ^B–H]⁻ (δ = −14.34) [74]:

Scheme 11.10

Conversion from a [C,N] to [C,C] cyclometalated Ir complex under photoirradiation

Fig. 11.9

Time course of H₂ evolution from 2-propanol (4.3 M) catalyzed by 5 (55 μM) in water (pH 11.9) at 353 K i) and 323 K (ii) and that from ethanol (5.7 M) catalyzed by [Ir ^C–H]⁻ (55 μM) in water (pH 11.9) at 323 K (iii). Time course of H₂ evolution from 2-propanol (4.3 M) catalyzed by [Ir ^B–H]⁻ (55 μM) in water (pH 11.9) at 323 K (iv) (Reprinted with the permission from Ref. [74]. Copyright 2012 American Chemical Society)

$$ \begin{array}{l}{\left({\mathrm{CH}}_3\right)}_2\mathrm{C}\mathrm{HOH}\to {\left({\mathrm{CH}}_3\right)}_2\mathrm{C}\mathrm{O}+{\mathrm{H}}_2\\ {}\kern4em {\left[\mathbf{I}{\mathbf{r}}^{\mathbf{C}}-\mathrm{H}\right]}^{-}\end{array} $$

(11.7)

11.2.2.2 Ruthenium Hydride Complexes

The catalytic activity of hydrogen evolution from alcohols has been reported to be remarkably enhanced by using ruthenium complexes containing pincer-type ligands [75]. Catalytic hydrogen evolution occurred in methanol containing KOH (8.0 M) with [RuHCl(CO)(HN(C₂H₄PⁱPr₂)₂)] (5), which exhibited high activities up to TOF = 4,700 h⁻¹ and TON = 350,000 at 368 K [75]. Under catalytic conditions, both formate and carbonate ions were observed as traces of the reaction mixtures, indicating that the formate is an intermediate in this dehydrogenation sequence and that CO₂ is trapped as carbonate [75]. The overall stoichiometry of the hydrogen evolution from methanol with NaOH is given by Eq. 11.8. The Ru-hydride species were observed in solution under catalytic conditions [75]:

$$ \begin{array}{l}{\mathrm{CH}}_3\mathrm{O}\mathrm{H}+2\mathrm{NaOH}\to 3{\mathrm{H}}_2+{\mathrm{Na}}_2{\mathrm{CO}}_3\\ {}\kern7.2em \mathbf{5}\end{array} $$

(11.8)

Base-free hydrogen evolution from methanol without formation of CO has recently been achieved by using the ruthenium-based PNP pincer complex (6: Ru-MACHO-BH) in Eq. 11.9 [76]. The combination of Ru-MACHO-BH (6) with Ru(H)₂(dppm)₂ (7) further enhanced the catalytic activity for hydrogen evolution from neutral methanol [76]. A long-term experiment gave a 26 % yield of H₂ (relative to H₂O) and a TON > 4,200 [76]. In this case full conversion of all “available” hydrogen atoms in methanol to H₂ has been achieved by synergetic homogeneous catalysis of 6 and 7 [76]:

$$ \begin{array}{l}{\mathrm{CH}}_3\mathrm{O}\mathrm{H}+{\mathrm{H}}_2\mathrm{O}\to 3{\mathrm{H}}_2+{\mathrm{CO}}_2\\ {}\kern5.5em \mathbf{6},\kern0.28em \mathbf{7}\end{array} $$

(11.9)

Hydrogen is also produced by the electrocatalytic reduction of protons with a Ru(II) complex [Ru^II(tpy)(bpy)(S)]²⁺ (tpy = 2,2′:6′,2″-terpyridine, bpy = 2,2′-bipyridine, S = solvent) in acetonitrile (MeCN) [77]. The Ru(II)-hydride complex [Ru^II(tpy)(bpy)(H)]⁺ is produced by the reaction of the ligand-based two-electron-reduced species [Ru^II(tpy^•–)(bpy^•–)(MeCN)]⁰ with water (Eq. 11.10) [77]:

$$ {\left[{\mathrm{Ru}}^{\mathrm{II}}\left({\mathrm{tpy}}^{\bullet \hbox{-}}\right)\left({\mathrm{bpy}}^{\bullet \hbox{-}}\right)\left(\mathrm{MeCN}\right)\right]}^0+{\mathrm{H}}_2\mathrm{O}\to {\left[{\mathrm{Ru}}^{\mathrm{II}}\left(\mathrm{t}\mathrm{p}\mathrm{y}\right)\left(\mathrm{b}\mathrm{p}\mathrm{y}\right)\left(\mathrm{H}\right)\right]}^{+}+{\mathrm{OH}}^{\hbox{-} } $$

(11.10)

Further reduction of the hydride to [Ru^II(tpy^•–)(bpy)(H)]⁰ at −1.41 V (vs. NHE) is proposed to trigger the catalytic water reduction via formation of the dihydrogen–dihydride complex [Ru^II(tpy)(bpy)(H₂)]⁺ (Eqs. 11.11, 11.12, and 11.13) [77]. However, this intermediate has yet to be detected. In the presence of an acid,

$$ {\left[{\mathrm{Ru}}^{\mathrm{II}}\left({\mathrm{tpy}}^{\bullet \hbox{-}}\right)\left(\mathrm{b}\mathrm{p}\mathrm{y}\right)\left(\mathrm{H}\right)\right]}^0+{\mathrm{H}}_2\mathrm{O}\to {\left[{\mathrm{Ru}}^{\mathrm{II}}\left({\mathrm{tpy}}^{\bullet \hbox{-}}\right)\left(\mathrm{b}\mathrm{p}\mathrm{y}\right)\left({\mathrm{H}}_2\right)\right]}^{+}+{\mathrm{OH}}^{\hbox{-} } $$

(11.11)

$$ {\left[{\mathrm{Ru}}^{\mathrm{II}}\left({\mathrm{tpy}}^{\bullet \hbox{-}}\right)\left(\mathrm{b}\mathrm{p}\mathrm{y}\right)\left({\mathrm{H}}_2\right)\right]}^{+}+{\mathrm{e}}^{\hbox{-}}\to {\left[{\mathrm{Ru}}^{\mathrm{II}}\left({\mathrm{tpy}}^{\bullet \hbox{-}}\right)\left({\mathrm{bpy}}^{\bullet \hbox{-}}\right)\left({\mathrm{H}}_2\right)\right]}^0 $$

(11.12)

$$ {\left[{\mathrm{Ru}}^{\mathrm{II}}\left({\mathrm{tpy}}^{\bullet \hbox{-}}\right)\left({\mathrm{bpy}}^{\bullet \hbox{-}}\right)\left({\mathrm{H}}_2\right)\right]}^0+\mathrm{MeCN}\to {\mathrm{H}}_2+{\left[{\mathrm{Ru}}^{\mathrm{II}}\left({\mathrm{tpy}}^{\bullet \hbox{-}}\right)\left({\mathrm{bpy}}^{\bullet \hbox{-}}\right)\left(\mathrm{MeCN}\right)\right]}^0 $$

(11.13)

[Ru^II(tpy)(bpy)(H)]⁺ can react with H⁺ to produce H₂ [77]. Many other metal hydrides are known to catalyze electrochemical reduction of protons to H₂ [78–83].

11.2.2.3 Proton-Coupled Electron Transfer to Metal Nanoparticles

Pt nanoparticles (PtNPs) act as the best catalyst for catalytic reduction of protons to H₂ [43]. Photocatalytic H₂ evolution occurred efficiently using NADH as a sacrificial electron donor, 9-mesityl-10-methylacridinium ion (Acr⁺–Mes) [84] as an organic photocatalyst, and PtNPS as a proton reduction catalyst (Scheme 11.11) [85]. Photoexcitation of Acr⁺–Mes resulted in intramolecular electron transfer from the Mes moiety to the singlet excited sate of the Acr⁺ moiety to produce the electron-transfer state (Acr^•–Mes^•+) [84, 86–88]. NADH is oxidized by the Mes^•+ moiety of Acr^•–Mes^•+ to produce two equivalents of Acr^•–Mes. Electron transfer from Acr^•–Mes to PtNPs with protons resulted in H₂ evolution [85].

Scheme 11.11

Chemical structure of Acr⁺–Mes and overall photocatalytic cycle for H₂ evolution

The kinetics and mechanism of the PtNP-catalyzed hydrogen evolution by an Acr^•–Mes were studied by simultaneous determination of the rate of hydrogen evolution and the rate of electron transfer from Acr^•–Mes to PtNPs [85]. The rate of H₂ evolution in a (pH 5.0, 50 mM) CH₃COOH/CH₃COONa buffer and MeCN [1:1 (v/v)] mixed solution is virtually the same as the rate of electron transfer from Acr^•–Mes to PtNPs, which was monitored by decrease in absorbance at 520 nm due to Acr^•–Mes as shown in Fig. 11.9 [85]. This indicates that electron transfer from Acr^•–Mes to PtNPs is the rate-determining step for the catalytic H₂ evolution. The rate constant of electron transfer from Acr^•–Mes to PtNPs (k _et) is proportional to proton concentration (Fig. 11.10) [85]. When CH₃COOH/CH₃COONa buffer (pH 4.5, 50 mM) in H₂O was replaced by CH₃COOD/CH₃COONa in D₂O, an inverse kinetic isotope effect (KIE = 0.68) was observed in electron transfer from Acr^•–Mes to PtNPs [85]. Such an inverse kinetic isotope effect results from the difference in the zero-point energy for Pt–H (Pt–D) bond at the transition state as compared with that before electron transfer when the interaction between Pt and H⁺ (or D⁺) is much smaller as shown in Fig. 11.11 [85]. This indicates that proton-coupled electron transfer (PCET) from Acr^•–Mes to PtNPs producing a Pt–H bond is the rate-determining step (r.d.s.) in the catalytic hydrogen evolution. The inverse KIE (0.68) in Fig. 11.12 shows sharp contrast to the large KIE (40) observed for the hydrogen evolution from formic acid, catalyzed by an Ir–hydride complex ([Ir^III(Cp^*)(H)(bpm)Ru^II(bpy)₂]³⁺) when the heterolytic Ir–H bond cleavage by proton is the rate-determining step in Scheme 11.4 (vide supra) [61]. Based on the results in Figs. 11.10 and 11.12, the PtNP-catalyzed H₂ evolution mechanism was proposed as shown in Scheme 11.12 [85]. PCET from Acr^•–Mes to PtNPs produces the Pt–H bond, followed by rapid elimination of H₂ from two Pt–H bonds.

Fig. 11.10

(a) Time profile of electron transfer from Acr^•–Mes to spherical PtNPs with the diameter of 4.5 nm (0.1 μg), monitored by decrease in absorbance at 520 nm due to Acr^•–Mes in a (pH 5.0, 50 mM) CH₃COOH/CH₃COONa buffer and MeCN [1:1 (v/v)] mixed solution. (b) Time profile of H₂ evolution (Reproduced from Ref. [85] by permission of John Wiley & Sons Ltd)

Fig. 11.11

Illustration of the PCET pathway to produce the Pt–H or Pt–D bond

Fig. 11.12

Dependence of k _et on [H⁺] or [D⁺] for electron transfer from Acr^•–Mes to spherical PtNPs with the diameter of 4.5 nm in H₂O/MeCN [1:1 (v/v)] containing CH₃COOH/CH₃COONa buffer (50 mM) or in D₂O/MeCN [1:1 (v/v)] containing CH₃COOD/CH₃COONa buffer (50 mM) at 298 K (Reproduced from Ref. [85] by permission of John Wiley & Sons Ltd)

Scheme 11.12

Mechanism of H₂ evolution on Pt surfaces

When Acr⁺–Mes was replaced by 2-phenyl-4-(1-naphthyl)quinolinium ion (QuPh⁺–NA) [89], photocatalytic H₂ evolution also occurred efficiently with NADH and PtNPs (Scheme 11.12) [90]. However, the rate constant of electron transfer from QuPh^•–NA to PtNPs was invariant with pH [90], in contrast to the case of PCET from Acr^•–Mes to PtNPs in which the rate constant was proportional to proton concentration (Fig. 11.10) [85]. Thus, electron transfer from QuPh^•–NA to MNPs occurs without assistance of proton because of the much stronger reducing ability of QuPh^•–NA as compared with Acr^•–Mes judging from the significantly more negative oxidation potential of QuPh^•–NA (E _ox = −0.90 V vs. SCE) [89] than that of Acr^•–Mes (E _ox = −0.57 V vs. SCE) [84]. Because the rate of hydrogen evolution was much slower than the rate of electron transfer from QuPh^•–NA to MNPs and the hydrogen evolution was also pH independent at pH < 10, the rate-determining step of the catalytic H₂ evolution may be elimination of hydrogen from two Pt–H bonds [90]. Thus, the rate-determining step for the catalytic H₂ evolution is changed depending on the reducing ability of one-electron reductants (Scheme 11.13).

Scheme 11.13

Chemical structure of QuPh⁺–NA and overall photocatalytic cycle for H₂ evolution

11.2.2.4 Kinetics and Mechanisms of Catalytic CO₂ Reduction

The catalytic reduction of CO₂ by H₂ has attracted significant interest because catalytic transformation of CO₂ would be promising for the production of fuels as liquid hydrogen sources and value-added chemicals [91–96]. However, the reactions involving CO₂ are commonly carried out at high pressure [97–106], which may not be economically suitable and also poses safety concerns. In order to improve the catalytic activity for the CO₂ reduction, it is of primary importance to elucidate the catalytic mechanism.

Kinetics and mechanism of the catalytic reduction of CO₂ by H₂ to produce formic acid (HCOOH) were reported by using [Ir^III(Cp^*)(L)(H₂O)](SO₄) and [Ru^II(η⁶-C₆Me₆)(L)(H₂O)](SO₄) (L = bpy or 4,4′-OMe-bpy) as catalysts in water [107]. The rates of the catalytic reduction of CO₂ by H₂ with [Ir^III(Cp^*)(L)(H₂O)](SO₄) under acidic conditions in H₂O are affected by the pressure of H₂ and CO₂. Turnover number (TON) of the catalytic reduction of CO₂ (2.5 MPa) by H₂ with [Ir^III(Cp^*)(L)](SO₄) increased with increasing H₂ pressure at pH 3.0 at 40 °C to reach a constant value (Fig. 11.13a), whereas TON was proportional to CO₂ pressure at 5.5 MPa of H₂ (Fig. 11.13b). The reactions of [Ir^III(Cp^*)(L)(H₂O)](SO₄) with H₂ at pH 3.0 in a citrate buffer solution provide the Ir(III)–hydride complexes [Ir^III(Cp^*)(L)(H)]₂(SO₄), which were detected by the ESI-mass spectra and ¹H NMR spectra. Because TON of the catalytic reduction of CO₂ by H₂ was proportional to CO₂ pressure, the rate-determining step may be the reaction of the Ir(III)–hydride complex with CO₂ to produce the formate complex as shown in Scheme 11.14. In such a case, the rate of formation of HCOOH in the catalytic reduction of CO₂ by H₂ with [Ir^III(Cp^*)(L)(H₂O)](SO₄) is given by Eq. 11.14,

Fig. 11.13

(a) Dependence of TONs at 0.5 h on H₂ pressure for the reduction of CO₂ (2.5 MPa) by H₂ catalyzed by 6(SO₄) (20 μmol) at pH 3.0 in a citrate buffer solution (20 cm³) at 40 °C. (b) Dependence TONs at 0.5 h on CO₂ pressure for the reduction of CO₂ by H₂ (5.5 MPa) catalyzed by the 6(SO₄) (20 μmol) at pH 3.0 in a citrate buffer solution (20 cm³) at 40 °C (Reproduced from Ref. [107] by permission of The Royal Society of Chemistry)

Scheme 11.14

Catalytic mechanism of CO₂ reduction by H₂ with [Ir^III(Cp^*)(L)(H₂O)](SO₄) to form HCOOH in H₂O

$$ \mathrm{d}\left[\mathrm{HCOOH}\right]/\mathrm{d}t={k}_1{k}_2\left[{\left[\mathrm{I}\mathrm{r}\hbox{-} {\mathrm{OH}}_2\right]}^{2+}\right]{P}_{\mathrm{H}2}{P}_{\mathrm{CO}2}/\left({k}_{-1}+{k}_1{P}_{\mathrm{H}2}\right) $$

(11.14)

where k ₁ is the rate constant of the reaction of the aqua complexes [Ir^III(Cp^*)(L)(H₂O)]²⁺ with H₂, k ₋₁ is the rate constant of the back reaction, k ₂ is the rate constant of the reaction of the hydride complex [Ir^III(Cp^*(L)(H))]⁺ with CO₂, and [[Ir–OH₂]²⁺]₀ is the initial concentration of [Ir^III(Cp^*)(L)(H₂O)]²⁺ [107]. Under the conditions such that k ₁ P _H2 >> k ₋₁, the rate of formation of HCOOH becomes constant at large H₂ pressure as observed in Fig. 11.13a [107].

When [Ir^III(Cp^*)(L)(H₂O)](SO₄) was replaced by [Ru^II(η⁶-C₆Me₆)(L)(H₂O)](SO₄), TON was proportional to H₂ pressure at 40 °C (Fig. 11.14a), whereas TON exhibited a saturation behavior with increasing CO₂ pressure (Fig. 11.14b) [107]. In such a case, the rate-determining step was changed from the reaction of the Ir(III)–hydride complex with CO₂ to produce the formate complex to the reaction of the Ru(III)–aqua complex with H₂ to produce the Ru(III)–hydride complex (Scheme 11.15) [107]. The rate of formation of HCOOH is given by Eq. 11.15, which indicates that the rate becomes constant at large CO₂ pressure as observed in Fig. 11.14b [107]. The Ru(III)–hydride complex, which was prepared independently by the reaction of the Ru(III)–aqua complex with NaBH₄, reacted with CO₂ to produce the formate complex, which was detected by ESI-mass and ¹H NMR spectra [108, 109]:

Fig. 11.14

(a) Dependence of TONs at 3 h on H₂ pressure for the reduction of CO₂ (2.5 MPa) by H₂ catalyzed by 2(SO₄) (20 μmol) at pH 3.0 in a citrate buffer solution (20 cm³) at 40 °C. (b) Dependence of TONs at 3 h on CO₂ pressure for the reduction of CO₂ by H₂ (5.5 MPa) catalyzed by 2(SO₄) (20 μmol) at pH 3.0 in a citrate buffer solution (20 cm³) at 40 °C (Reproduced from Ref. [107] by permission of The Royal Society of Chemistry)

Scheme 11.15

Catalytic mechanism of CO₂ reduction by H₂ with [Ru^II(η⁶-C₆Me₆)(L)(H₂O)](SO₄) to form HCOOH in H₂O

$$ \mathrm{d}\left[\mathrm{HCOOH}\right]/\mathrm{d}t={k}_1{k}_2\left[{\left[\mathrm{R}\mathrm{u}\hbox{-} {\mathrm{OH}}_2\right]}^{2+}\right]{P}_{\mathrm{H}2}{P}_{\mathrm{CO}2}/\left({k}_{-1}+{k}_2{P}_{\mathrm{CO}2}\right) $$

(11.15)

The change in the rate-determining step in the catalytic reduction of CO₂ by H₂ between the Ir and Ru complexes results from the stronger Ir–H bond as compared with the Ru–H bond as indicated by the higher Ir–H stretching frequency (2,056 cm⁻¹) than the Ru–H stretching frequency (1,899 cm⁻¹) [107]. The stronger Ir–H bond facilitates the formation of the Ir–H bond, but decelerates the Ir–H bond cleavage by CO₂, which becomes the rate-determining step in the Ir complex-catalyzed CO₂ reduction by H₂. Conversely the weaker Ru–H bond facilitates the Ru–H bond cleavage by CO₂ but decelerates the formation of the Ru–H bond, which becomes the rate-determining step. The initial TOF for the catalytic reduction of CO₂ by H₂ with [Ir^III(Cp^*)(L)(H₂O)](SO₄) was improved from 1 h⁻¹ (L = bpy) to 27 h⁻¹ (L = 4,4′-OMe-bpy) [107]. Thus, the more electron-rich Ir–H complex exhibits the higher catalytic reactivity. The X-ray crystal structures of the Ir–H and Ru–H complexes are shown in Fig. 11.15 [55, 109]. In both cases, the hydride complexes adopt a distorted octahedral coordination which has a terminal hydride ligand.

Fig. 11.15

ORTEP drawings of (a) [Ir^III(Cp^*)(bpy)(H)](PF₆) [109] and (b) [Ru^II(η⁶-C₆Me₆)(H)](CF₃SO₃) [55]. The counter anions are omitted for clarity (Reprinted with the permission from Ref. [55, 109]. Copyright 2003 American Chemical Society)

The catalytic activity for the reduction of CO₂ to HCOOH was enhanced by using a C^N cyclometalated organoiridium complex ([Ir^III(Cp^*){4-(1H-pyrazol-1- yl-κN²)benzoic acid–κC³}(H₂O)]₂SO₄ ([Ir ^A–H₂O]⁺)), which was employed for the catalytic decomposition of HCOOH to H₂ under acidic conditions in Fig. 11.7 (vide supra) [66]. At pH 7.5, the carboxylic acid is deprotonated to produce the more electron-rich Ir complex ([Ir ^B–H₂O]⁰) when the direction of the reaction was reversed and the catalytic reduction of CO₂ by H₂ with [Ir ^B–H₂O]⁰ occurred to produce formate at ambient pressure and temperature as shown in Fig. 11.16, where TON increased linearly with time to exceed over 100 [66]. Turnover frequency (TOF) increased with decrease in pH to afford the highest value at pH 8.8 and decreased with further increase in pH to reach zero at pH 10.4 [66]. The pH dependence of TOF is similar to pH dependence of the amount ratios of [Ir ^B–H₂O]⁰ over [Ir ^B–OH]⁻ and HCO₃ ⁻ over CO₃ ²⁻ (red line and red dashed line in Fig. 11.17, respectively). Thus, the reduction of HCO₃ ⁻ by H₂ is catalyzed by [Ir ^B–H₂O]⁰ at pH 8.8. The TOF value at pH 8.8 increased linearly with increasing concentration of CO₂, which is converted to mixture of HCO₃ ⁻ and CO₃ ²⁻ (Fig. 11.18) [66]. Thus, the rate-determining step in the catalytic reduction of CO₂ to formate by H₂ is the insertion of CO₂ to the Ir–H complex [Ir ^B–H]⁰ to produce the formate complex in Scheme 11.16 [66].

Fig. 11.16

Time course of the concentration of formate and TON for the formate formation in the reduction of CO₂ by H₂ catalyzed by [Ir ^B–H₂O]⁰ (0.26 mM) under atmospheric pressure of H₂ (50 mL/min) and CO₂ (50 mL/min) in deaerated H₂O at 303 K at pH 7.5 (Reproduced from Ref. [66] by permission of The Royal Society of Chemistry)

Fig. 11.17

pH dependence of the formation rate (TOF) of formate in the catalytic generation of formate from H₂, HCO₃ ⁻, and CO₃ ²⁻ ([HCO₃ ⁻] + [CO₃ ²⁻] = 2.0 M) catalyzed by [Ir ^B–H₂O]⁰ (0.18 mM) in deaerated H₂O at 333 K (solid line). Alternate long and short dashed lines, (i) and (ii) show the amount ratios of complex [Ir ^B–H₂O]⁰ and [Ir ^B–OH]⁻, respectively, to the total amount of these complexes. Dashed lines, (i′) and (ii′) show the ratios of HCO₃ ⁻ and CO₃ ²⁻, respectively (Reproduced from Ref. [66] by permission of The Royal Society of Chemistry)

Fig. 11.18

Plot of TOF vs. the concentration of KHCO₃ and K₂CO₃ mixture (KHCO₃/K₂CO₃), i.e., [KHCO₃] + [K₂CO₃] in the hydrogenation reaction of KHCO₃/K₂CO₃ with H₂ catalyzed by [Ir ^B–H₂O]⁰ (0.18 mM) under atmospheric pressure of H₂ in deaerated H₂O at pH 8.8 at 333 K (Reproduced from Ref. [66] by permission of The Royal Society of Chemistry)

Scheme 11.16

Catalytic mechanism of CO₂ reduction by H₂ with [Ir ^A–H₂O]⁺ to form HCOO⁻ in H₂O

A dinuclear Cp^*Ir catalyst with 4,4,6,6-tetrahydroxy-2,2-bipyrimidine as a bridging ligand (see the crystal structure in Fig. 11.19) can also catalyze the reduction of CO₂ by H₂ at ambient pressure at pH 8.4 with TOF = 70 h⁻¹ at 298 K [110]. Mononuclear Cp^*Ir complexes with biazole ligands also act as efficient catalysts for reduction of CO₂ by H₂ to formate at ambient pressure and temperature [111].

Fig. 11.19

(a) X-ray crystal structure and (b) the structural formula of a dinuclear Cp^*Ir catalyst with 4,4,6,6-tetrahydroxy-2,2-bipyrimidine as a bridging ligand employed for efficient CO₂ reduction to formate by H₂ (Reprinted by permission from Macmillan Publishers Ltd: Ref. [110], copyright 2012)

Iridium complexes mentioned above act as efficient catalysts for the selective decomposition of formic acid to H₂ and CO₂ without formation of CO under acidic conditions at ambient temperature [55, 66, 110, 111]. Thus, the catalytic interconversion between hydrogen and formic acid has been made possible by changing pH with the same catalyst, providing a convenient hydrogen-on-demand system in which hydrogen (gas) can be stored as formic acid (liquid) and whenever needed hydrogen is produced by the catalytic decomposition of formic acid [11, 66, 112]. Formic acid can also be directly used as a fuel in direct formic acid fuel cells, which have recently attracted much attention due to high electromotive force, limited fuel crossover, and high practical power densities at low temperatures as compared with direct methanol fuel cells [113–116].

11.3 Conclusions

We have overviewed kinetic studies on catalytic reduction of protons and CO₂ in mainly homogeneous phase, providing valuable mechanistic insights. Kinetic studies including deuterium kinetic isotope effects on heterogeneous catalysts for hydrogen evolution have also been demonstrated to provide essential mechanistic information on bond cleavage and formation associated with electron transfer. Kinetic studies have also enabled us to determine the rate-determining steps in the catalytic cycles, providing valuable information on observable intermediates, which can be detected by various methods. The most important intermediates in the catalytic reduction of protons and CO₂ are metal-hydride complexes, which can reduce protons and CO₂ to produce hydrogen and formic acid, respectively. Metal η ¹-CO₂ complexes that are formed by a nucleophilic attack to low-valent metal complexes with the central carbon are responsible for the two-electron reduction of CO₂ to CO [33, 99, 117–119]. The key remaining challenge is not just two-electron reduction of CO₂ with two protons to formic acid or carbon monoxide (CO) but multiple proton-coupled electron transfers to produce further reduced products such as methanol and methane. Such multi-electron reduction beyond two-electron reduction of CO₂ has so far been achieved in the heterogeneous systems by photocatalysis and electrocatalysis [13, 16, 120–132]. A series of different homogeneous catalysts have also been employed to achieve the catalytic reduction of CO₂ by H₂ to methanol in a single vessel to promote the various steps of the CO₂ reduction sequence [133]. Recently the homogenously catalyzed reduction of CO₂ by H₂ to methanol has been achieved by a single ruthenium phosphine complex [134, 135]. Methanol can also be obtained by the disproportion of formic acid catalyzed by an Ir complex ([Ir^III(Cp^*)(bpy)(H₂O)](OTf)₂) [136]. Further kinetic studies on such homogeneously catalyzed multi-electron reduction of CO₂ by H₂ may elucidate the catalytic mechanisms, which will certainly help develop efficient catalysts for production of carbon-neutral alternatives to fossil fuels.