Photocatalytic CO₂ reduction with high turnover frequency and selectivity of formic acid formation using Ru(II) multinuclear complexes

Abstract

Previously undescribed supramolecules constructed with various ratios of two kinds of Ru(II) complexes—a photosensitizer and a catalyst—were synthesized. These complexes can photocatalyze the reduction of CO₂ to formic acid with high selectivity and durability using a wide range of wavelengths of visible light and NADH model compounds as electron donors in a mixed solution of dimethylformamide–triethanolamine. Using a higher ratio of the photosensitizer unit to the catalyst unit led to a higher yield of formic acid. In particular, of the reported photocatalysts, a trinuclear complex with two photosensitizer units and one catalyst unit photocatalyzed CO₂ reduction (Φ_HCOOH = 0.061, TON_HCOOH = 671) with the fastest reaction rate (TOF_HCOOH = 11.6 min^-1). On the other hand, photocatalyses of a mixed system containing two kinds of model mononuclear Ru(II) complexes, and supramolecules with a higher ratio of the catalyst unit were much less efficient, and black oligomers and polymers were produced from the Ru complexes during photocatalytic reactions, which reduced the yield of formic acid. The photocatalytic formation of formic acid using the supramolecules described herein proceeds via two sequential processes: the photochemical reduction of the photosensitizer unit by NADH model compounds and intramolecular electron transfer to the catalyst unit.

Recently, global warming and shortages of fossil fuels and carbon resources have become serious issues. The development of technologies to convert CO₂ into useful organic compounds using sunlight as an energy source would serve as an ideal solution to these problems.

Formic acid, which is the two-electron reduction product of CO₂, has recently attracted attention as a storage source of H₂ (1, 2). Formic acid itself is an important chemical. It has been employed as a preservative and an insecticide and is also a useful acid, reducing agent, and source of carbon in synthetic chemical industries.

Only a few photocatalysts for the selective formation of formic acid from CO₂ have been reported (3–8). Although oligo(p-phenylenes) (3) or a mixed system of phenazine and Co cyclam (4) successfully photocatalyzed the reduction of CO₂ to formic acid, these systems cannot work with visible light. It has been reported that [Ru(bpy)₂(CO)₂]²⁺ (bpy = 2,2′-bipyridine) acted as a catalyst for reducing CO₂ (5–8). Under basic conditions, a mixed system of this complex with [Ru(bpy)₃]²⁺ as a redox photosensitizer photocatalyzed the reduction of CO₂ to formic acid with high selectivity (6, 8). However, this photocatalytic system is limited by instability as evidenced by the fact that the catalyst decomposed following prolonged irradiation and generated black precipitates.

We have recently developed a unique architecture for constructing visible-light-driven supramolecular photocatalysts, consisting of a [Ru(N^∧N)₃]²⁺ (N^∧N = a diimine ligand)–type complex as a photosensitizer and a Re(I) diimine complex as a catalyst (9–12). These supramolecules can selectively photocatalyze the reduction of CO₂ to CO with high efficiency. The requirements for constructing efficient supramolecular photocatalysts with Ru(II) and Re(I) complexes involves two key principles: First, in the triplet metal-to-ligand-charge-transfer (³MLCT) excited state of the Ru(II) unit, the excited electron should be located at the bridging ligand. Second, a nonconjugated bridging ligand should be used because a conjugated system in the bridge ligand lowers the reducing power of the catalyst unit (9). The efficiencies and the stability of these supramolecular photocatalysts (Φ_CO = 0.12–0.15, TON_CO ∼ 200) are much greater than those of the other reported supramolecular-type photocatalysts for both CO₂ reduction (13–16) and H₂ evolution (16–19).

We sought to apply this architecture to construct unique supramolecular photocatalysts for the reduction of CO₂ to selectively form formic acid (which has not been reported to date). We successfully developed such photocatalysts having high efficiency, high selectivity for formic acid formation, high stability, and a fast reaction rate.

The structures and abbreviations of the synthesized supramolecular complexes are shown in Chart 1 and include [Ru(dmb)_n(BL)_3-n]²⁺ [dmb = 4,4′-dimethyl-2,2′-bipyridine; BL = 1,2-bis(4'-methyl-[2,2'-bipyridin]-4-yl)ethane] as the photosensitizer unit; and [Ru(dmb)_m(BL)_2-m(CO)₂]²⁺ as the catalyst unit. The abbreviations x and y in (x,y) indicate the numbers of each unit in one molecule (x: photosensitizer, y: catalyst). Their model mononuclear complexes are also abbreviated; (1,0) and (0,1) represent the photosensitizer and catalyst, respectively.

Chart 1.

Structures and abbreviations of Ru(II) complexes

Results and Discussion

For evaluation of the photocatalysis, we determined both the formation quantum yield (Φ) and the turnover number (TON) of the product along with the turnover frequency (TOF) in part. Since suitable irradiation conditions for obtaining these values are different, we chose the following reaction conditions (SI Text):

Light Irradiation Condition (LIC) 1 for determining Φ: For evaluating the number of photons absorbed by the photocatalyst, monochromic light is required, and the absorbance of the reaction solution at the wavelength of irradiation should be sufficiently high, at least 2 where 99% of the irradiated photons is absorbed by the photocatalyst. We used 480-nm monochromic light obtained using a 500-W Xe lamp with a band-pass filter (FWHM = 10 nm); the light intensity was 4.9 × 10^-8 or 4.9 × 10^-9 einstein s^-1, which was controlled using neutral density (ND) filters. The concentration of the photosensitizer unit (not the photocatalyst) was 0.3 mM, and the solution was vigorously mixed during the irradiation. The incident light flux was completely absorbed by the photosensitizer unit because the absorbance of the reaction solutions was larger than 3 at 480 nm and the light-pass length was 1 cm.

In this reaction condition, it is difficult to determine reasonable TON values because too-large amounts of the products—i.e., formic acid and BNA₂s—should be accumulated in the reaction solution after long irradiation and the products obstruct the photocatalytic reaction (see below). The photon flux determined that the rates of the photocatalytic reactions—i.e., the ordinary light source—even using a 500-W Xe lamp cannot supply enough photon flux for determining maximum TOF values. For obtaining more exact values of TON and TOF, we reduced the concentrations of the photocatalysts in the reaction solutions and used different light sources as follows.

LIC2 for determining TON: > 500-nm light was obtained using a 500-W high-pressure mercury lamp equipped with a uranyl glass and a K₂CrO₄ (30% w/w, d = 1 cm) aqueous solution filter, which supplied stronger light flux than that used in LIC1. The concentration of the photosensitizer unit was 0.05 mM. The TONs were determined after 20-h irradiation in all cases, and calculated as the produced amount of formic acid divided by the amount of supramolecule added. Using a merry-go-round irradiation apparatus, we could irradiate up to eight samples simultaneously with the same light-intensity.

LIC3 for determining TOF and measuring NMR: > 420-nm light was obtained using a 500-W high-pressure mercury lamp or a 500-W Xenon lamp with a cutoff filter. This visible-light source could irradiate the reaction solution with the highest intensity in our laboratory.

The photosensitizer unit can be selectively excited using LIC1 and LIC2, while, in the case of LIC3, the irradiated light is absorbed mainly by the photosensitizer unit and partially by the catalyst unit where slow photolysis of the photocatalysts was observed and TON obtained in LIC3 was slightly lower than that in LIC2 (Table 1). The reductants (BNAH, MeO-BNAH) and the BNA₂ scarcely absorbed > 420-nm light (Fig. S1A).

Table 1.

Photocatalytic properties of the supramolecules and the model system^*

entry	photocatalyst	reductant	products/μmol †			ΦHCOOH‡	TONHCOOH§	Φem¶	τem∥ ns	kq** 107 M-1 s-1	ηq †† %
			HCOOH	CO	H2
1	(2,1)	BNAH	30.4	1.8	1.8	0.041	562	0.085	726	1.97	59
2		MeO–BNAH ‡‡	36.8	2.4	1.9	0.061	671 (396) §§			4.72	77
3	(1,1)	BNAH	26.9	2.8	0.9	0.038	315	0.083	745	1.64	54
4	(1,2)	BNAH	8.4	7.5	1.0	0.030	353	0.089	755	2.33	64
5	(1,3)	BNAH	5.3	8.1	0.7	0.017	234	0.082	733	2.82	67
6	(1,0)+(0,1)	BNAH	5.3	1.9	0.5	−	316	0.087¶¶	766¶¶	1.02¶¶	44¶¶

^*A 4-mL CO₂-saturated dimethylformamide–triethanolamine (DMF–TEOA) (4∶1 v/v) solution containing reductant (0.1 M) and complexes was irradiated. The concentrations of all the photocatalysts were 0.3 mM for “Φ_HCOOH” and “Products” and 0.05 mM for “TON_HCOOH,” except for (2,1), of which concentrations were adjusted to half of the other complexes because (2,1) has two photosensitizer units. Therefore, the concentrations of the photosensitizer units were same in all the photocatalytic systems in the series of the experiments.

^†Irradiated in the Light Irradiation Condition (LIC) 1 for 5 h (light intensity: 4.9 × 10^-8 einstein s^-1).

^‡Irradiated in the LIC1 (light intensity: 4.9 × 10^-8 einstein s^-1).

^§Irradiated in the LIC2 for 20 h. TON_HCOOHs were calculated as [produced HCOOH]/[added supramolecule].

^¶Emission quantum yield of the complexes in DMF–TEOA (4∶1 v/v) (Excitation wavelength: 480 nm).

^∥Emission lifetime of the complexes in DMF–TEOA (4∶1 v/v) (Excitation wavelength: 456 nm).

^**Quenching rate constant of emission from the complexes by the reducing agent.

^††Quenching fractions of emission from the Ru(II) complexes by 0.1 M of the reducing agent, calculated as 0.1k_qτ_em/(1 + 0.1k_qτ_em).

^‡‡1-(4-methoxybenzyl)-1,4-dihydronicotinamide (25).

^§§Irradiated in the LIC3 for 20 h.

^¶¶Determined by emission from (1,0).

In a typical run for determining TON, a solution of dimethylformamide (DMF) and triethanolamine (TEOA) (4∶1 v/v) containing (1,1) (0.05 mM) and 1-benzyl-1,4-dihydronicotinamide (BNAH, 0.1 M) as a sacrificial electron donor was bubbled with CO₂ for 20 min and then irradiated (LIC2) to give formic acid with high selectivity. The TON of formic acid formation exceeded 300 after 20-h irradiation with small amounts of CO and H₂ (Eq. 1, Fig. 1). On the other hand, formic acid was not produced at all in the absence of BNAH or without irradiation. 1

Fig. 1.

Photocatalytic formation of formic acid (red filled circle), CO (green filled square), and H₂ (orange filled triangle) as a function of irradiation time. The (red dotted line) shows the amount of formic acid generated by photocatalysis, under the assumption that the side reaction effect by BNA₂s and the effect of the decrease of BNAH can be neglected: A CO₂ saturated DMF-TEOA (4∶1 v/v, 4 mL) solution containing BNAH (0.1 M) and (1,1) (0.05 mM) was irradiated by > 500-nm light.

In a labeling experiment using , a strong signal attributed to was observed at 168.2 ppm in the ¹³C NMR spectrum of a DMF-d₇-TEOA (4∶1 v/v) solution containing (1,1) (0.5 mM), BNAH (0.1 M), and (532 Torr) after irradiation (LIC3) for 14.5 h (TON_HCOOH = 59) (Fig. S2, Upper)*. Also, in the ¹H NMR of the same solution, a doublet (J_CH = 188 Hz) attributable to the proton coupled to the ¹³C of was observed at 8.52 ppm, but no singlet due to the proton of was detected (Fig S2, Lower)^†. These results clearly show that the carbon source of formic acid is CO₂.

Fig. 2 shows the IR spectra of (0,1) in DMF (a) and in DMF-TEOA (b). Two peaks observed in DMF are attributed to the symmetric and asymmetric stretching bands of the two CO ligands (2094 and 2040 cm^-1, respectively). On the other hand, only one peak appeared in the ν_CO region in the DMF-TEOA (4∶1 v/v) solution (1958 cm^-1). This clearly indicates that (0,1) was converted to a complex with only one carbonyl ligand in the presence of TEOA. A candidate is [Ru(dmb)₂(CO)(COOH)]⁺, which could be produced by addition of OH^- from contaminated water. However, this should not be the main complex in the reaction solution because the ν_CO of [Ru(dmb)₂(CO)(COOH)]⁺ in DMF-TEOA (1,955 cm^-1) was similar to but clearly different from that of (0,1) in the same solvent system. The structural transition of (0,1) in DMF-TEOA is attributable to the addition of deprotonated TEOA to the CO ligand giving [Ru(dmb)₂(CO){COOC₂H₄N(C₂H₄OH)₂}]⁺, (0,1)′, as shown in Eq. 2. The formation of (0,1)′ was also confirmed by an ESI-MS spectrum of the reaction solution in which the parent peak of (0,1)′ (m/z = 674) was observed as one of the main signals (Fig. S3). A similar conversion of one of the CO ligands to should occur in all multinuclear complexes because similar IR spectral changes were observed in DMF-TEOA in all cases. 2

Fig. 2.

IR spectra of (0,1) in DMF (blue line) and in DMF-TEOA (4∶1 v/v) (red line).

The UV–visible light (UV-vis) absorption spectra of (1,0), (0,1), and (0,1)′ are shown in Fig. 3. The metal-to-ligand-charge-transfer (MLCT) band of (0,1)′ was red-shifted compared with that of (0,1) because of the weak ligand field of the ligand compared with that of the CO ligand. However, because the MLCT absorption band of (1,0) was observed at a much longer wavelength than that of (0,1)′, it can be concluded that the irradiated lights absorbed only by the photosensitizer unit(s) in LIC1 (480 nm) and LIC2 (> 500 nm).

Fig. 3.

UV-vis absorption spectra of (1,0) (black line) and (0,1) (red line), which was converted to (0,1)′ (see the main text), in DMF-TEOA (4∶1 v/v) and (0,1) in DMF (blue line).

The emission quantum yields of (1,0) and the multinuclear complexes observed in DMF-TEOA are summarized in Table 1. They were very similar to the emission quantum yield of (1,0) observed in a DMF solution (Φ_em = 0.088). Therefore, in the multinuclear complexes, the intramolecular quenching of the excited state of the photosensitizer unit by the catalyst unit with the ligand(s) does not occur in DMF-TEOA (Scheme 1)^‡.

Scheme 1.

Initial process of the photocatalytic reaction using (1,1).

Under the photocatalytic reaction conditions, BNAH should function as a reductant, but TEOA cannot, as suggested by the observation that the emission from the ³MLCT excited state of the photosensitizer unit was mostly quenched by BNAH (Table 1), while quenching by TEOA was negligible (Scheme 1). However, TEOA should have other important roles in the photocatalytic reaction. It has been reported that TEOA can act as a base that captures a proton from the one-electron oxidation product of BNAH (i.e., BNAH·⁺) and suppresses the back-electron transfer from the reduced photosensitizer unit to BNAH·⁺ (process 1 in Scheme 2) (9). Another potential role for TEOA as a base involves the control of product distribution in CO₂ reduction. Tanaka et al. have reported that in the case of the electrochemical reduction of CO₂ using [Ru(bpy)₂(CO)₂]²⁺ as a catalyst, the product distribution changed depending on the pH of the reaction solution: Formic acid was the main product under basic conditions, but CO and H₂ were produced with higher yields under acidic conditions (5–7). Actually, in the case of the photocatalytic reaction using (1,1), the formation rate of formic acid was 4.7 times lower in the absence of TEOA compared with that in the presence of TEOA, and the formation selectivity of formic acid was also lower in the absence of TEOA (Fig. S5A).

Scheme 2.

Oxidation processes of BNAH.

The reducing agent BNAH can potentially act either as a two-electron or one-electron donor (20–22). The two-electron process gives BNA⁺ via an electrochemical, chemical, and electrochemical (ECE) sequence (process 2 in Scheme 2), while the one-electron process gives 4,4′- and 4,6′-BNA₂ via the coupling reaction of BNA· as oxidation products (process 3 in Scheme 2). To clarify the role of BNAH, we used high performance liquid chromatography to quantify both the decrease of BNAH and the formation of its oxidation product(s) in the photocatalytic reaction solutions (LIC2: Fig. 4). For example, 20-h irradiation induced the consumption of 141 μmol of BNAH, and the production of 63 μmol of formic acid, 4 μmol of H₂, and 3 μmol of CO. This reaction solution also contained 29 μmol of 4,4′-BNA₂^§ (23), and 25 μmol and 14 μmol of the diastereomers of 4,6′-BNA₂. On the other hand, BNA⁺ was not detected in the reaction solution. The formation of all reduction products (HCOOH, H₂, and CO) requires two electrons per molecule, and the formation of one BNA₂ molecule from two BNAH molecules donates two electrons. Thus, we conclude that BNAH acted only as a one-electron donor. The electron balance in the photocatalytic reduction of CO₂ to formic acid using (1,1) is shown in Eq. 3, below. 3

Fig. 4.

Photocatalytic production of the reduction and oxidation products—i.e., HCOOH + CO + H₂ (red filled circle) and BNA₂s (blue filled diamond)—and consumption of BNAH (black filled circle): a CO₂ saturated DMF-TEOA (4∶1 v/v, 4 mL) solutions containing BNAH (0.1 M) and (1,1) (0.05 mM) was irradiated at > 500-nm light.

Although BNA₂s are recognized as good electron donors (24), they did not act as reducing agents in this photocatalytic reaction and were the final oxidation products of BNAH. Back electron transfer is expected to occur too fast for the one-electron oxidized products of BNA₂s (BNA₂·⁺) to be irreversibly decomposed [Eq. 4]. Since the oxidation potentials of BNA₂s [for example, vs. SCE] (24) are more negative than that of BNAH [ vs. SCE] (25), it is possible that the accumulation of BNA₂s obstructs the photochemical electron transfer from BNAH to the excited state of (1,1) [k_q(BNAH) = 1.64 × 10⁷ M^-1 s^-1] in view of the much faster quenching rate constant of the emission from the excited (1,1) by BNA₂s: k_q(4,4^′-BNA₂) = 3.09 × 10⁸ M^-1 s^-1. For example, after 8-h irradiation, the reaction solution contained 291 μmol of BNAH and 51 μmol of BNA₂s where only 18% of the excited state of (1,1) was quenched by BNAH, whereas 54% could be quenched in the first stage of the photocatalytic reaction. We believe that this is the main reason why the photocatalytic efficiency decreased after the 8-h irradiation. If the effects of the side reaction and the decrease of BNAH were prevented, the formation of HCOOH would increase by 1.85-fold after 8-h irradiation (red dotted line in Fig. 1). This increased rate should be approximately the same as that in the first stage of the photocatalytic reaction. 4

The photocatalytic ability of (1,1) was much greater than that of a mixed system of the corresponding mononuclear model complexes ((1,0) + (0,1),1∶1) (Fig. 5). The irradiation of the reaction solution containing 0.3 mM of (1,1) by 480-nm monochromatic light (LIC1: light intensity: 4.9 × 10^-8 einstein s^-1) gave formic acid with Φ_HCOOH = 0.038. On the other hand, the irradiation of a solution containing (1,0) and (0,1) (0.3 mM each) instead of (1,1) gave a much smaller amount of formic acid with an induction period in the initial stage, but formation of formic acid ceased after irradiation for only 1 h. Furthermore, the irradiation of the mixed system with lower light intensity (LIC1: 4.9 × 10^-9 einstein s^-1) produced no formic acid at all, whereas (1,1) still functioned as a photocatalyst with Φ_HCOOH = 0.045 (Fig. S6). It should be emphasized that Φ_HCOOH was similar even when using a one-order difference in light intensity. During the photocatalytic reaction of the mixed system of (1,0) and (0,1), the solution changed color from red to black: This color change was more significant at lower light intensity after the same quantity of photons was introduced into the reaction solution (Fig. 6A). However, no such color change was observed when (1,1) was used as a photocatalyst (Fig. 6B).

Fig. 5.

Photocatalytic formation of formic acid as a function of irradiation time: CO₂ saturated DMF-TEOA (4∶1 v/v, 4 mL) solutions containing BNAH (0.1 M) and (2,1) (0.15 mM: purple filled circle), (1,1) (0.3 mM: red filled circle), (1,2) (0.3 mM: blue filled circle), (1,3) (0.3 mM: orange filled circle), or a mixture of (1,0) and (0,1) (0.3 mM each: black filled circle), were irradiated by 480-nm light of intensity 4.9 × 10^-8 einstein s^-1.

Fig. 6.

UV-vis absorption spectral changes of solutions containing the mixture of (1,0) and (0,1) (0.3 mM each: A) or (1,1) (0.3 mM: B) during irradiation (0–10 h at 1-h intervals): CO₂ saturated DMF-TEOA (4∶1 v/v, 4 mL) solutions containing BNAH (0.1 M) and the complex(es) were irradiated by 480-nm light of intensity 4.9 × 10^-9 einstein s^-1.

The ratio between the photosensitizer unit(s) and the catalyst unit(s) in the supramolecular system strongly influenced the outcome of photocatalysis. A higher ratio of the photosensitizer unit to the catalyst unit generated a higher yield of formic acid (LIC1: Fig. 5). In the cases in which the reaction solutions had the same absorbance at 480 nm [this light can be absorbed only by the photosensitizer unit(s)], the results of the photocatalytic reactions are summarized in Table 1. As an example, (2,1) has two photosensitizer units and one catalyst unit and generated formic acid with the highest selectivity and yield. In contrast, using (1,3) with one photosensitizer unit and three catalyst units, CO was the major product, and the photocatalytic activity was much lower with this photocatalyst than with (2,1), even though the reaction solution of (2,1) contained only 1/6 the number of catalyst units present in the (1,3) solution. All complexes produced only a small amount of H₂. The photocatalytic activities of (1,2) and (1,3) decreased rapidly (Fig. 5), and the color of the solution changed from red to black. For CO formation with (1,2) and (1,3), there was an induction period, and the production of CO accelerated after the color change (Fig. S5B). However, no such phenomenon was observed with (1,1) and (2,1). These results suggest that the actual photocatalyst for CO formation with (1,2) and (1,3) is the black product(s) produced during the photocatalytic reaction. Table 1 summarizes the emission quantum yields of (1,0) and the supramolecules using 480-nm light for excitation. Although the emission yields from the supramolecules were slightly smaller than that from (1,0), there was no correlation between the emission yield and the photocatalytic ability. Table 1 also shows the fractions of the emission quenched by 0.1 M BNAH (η_q). Clearly (1,1) and (2,1) are better photocatalysts, although emission from (1,2) and (1,3) was more efficiently quenched by BNAH than that from (1,1) and (2,1).

The photocatalytic activities of (1,2), (1,3), and the mixed system of (1,0) and (0,1) deteriorated rapidly and the reaction solution turned black even in the early stages of irradiation, as described above. As a typical example, Fig. 6A shows the changes in the UV-vis absorption spectrum of the mixed system during irradiation (LIC1). Tanaka et al. observed the precipitation of black solids during the photocatalytic reaction of the mixed system of [Ru(bpy)₃]²⁺ and [Ru(bpy)₂(CO)₂]²⁺; they suggested that the precipitation causes the deactivation of the photocatalyst (6). Deronzier and Ziessel et al. reported that the electrochemical reduction of [Ru(bpy)₂(CO)₂]²⁺ caused the dissociation of the bpy ligand to give a black polymer with ruthenium-ruthenium bonds (i.e., [Ru(bpy)(CO)₂]_n, Eq. 5) (26). As the absorption spectrum of this polymer was similar to that in Fig. 6A, the deactivation of the photocatalysts (1,0) + (0,1), (1,2), and (1,3) is presumably due to the formation of similar oligomers and/or polymers from the reduced catalyst or catalyst unit produced during irradiation^¶. Lower photocatalytic abilities of (1,0) + (0,1), (1,2), and (1,3) in the LIC1 were observed compared with those in the LIC2 (“products” vs. “TON_HCOOH” in Table 1). As described above, the formation of the oligomers and/or polymers was enhanced with the lower light intensity. Therefore, this deactivation process should be accelerated with either higher concentration of the complex(es) in solution or lower irradiated light intensity. It should be emphasized that (2,1) showed the highest photocatalytic ability in any photocatalytic reaction condition. 5

In the cases of (1,2) and (1,3), an additional absorption band was observed with a maximum at 600 nm during the irradiation (LIC1: Fig. S8B). This can be attributed to the reduced catalyst unit, because this absorption alone disappeared rapidly on introducing air into the solution. The remaining spectra after the introduction of air were very similar to that of the black polymer (Fig. 6A). The formation of formic acid involves a two-electron reduction, which could occur by the twofold reductive quenching of the excited photosensitizer unit by BNAH. Probably, the larger number of catalyst units in the supramolecule gives rise to electron donation from the one-electron reduced species of the photosensitizer unit to the different catalyst units. This should increase the lifetime of the reduced catalyst unit and trigger the oligomerization of the catalyst units. This is a major reason why (1,2) and (1,3) have low photocatalytic ability. Solutions of (1,1) and (2,1) did not undergo any similar color change during irradiation (Fig. 6B and Fig. S8A). An increase in the number of photosensitizer units in the supramolecule should increase the likelihood of injection of a second electron into the catalyst unit, thereby suppressing the process of deactivation of the catalyst unit. We checked the dependence of the photocatalytic efficiency on the concentration of (2,1) (0.02–0.15 mM) in the LIC1 (light intensity: 4.9 × 10^-8 einstein s^-1) as shown in Fig. S9A. This clearly shows that the efficiency is not affected by the concentration of (2,1) at all.

Only 59% of the emission from the photosensitizer unit of (2,1) was quenched by BNAH even in the first stage of the photocatalytic reaction (Table 1, entry 1). If this process of reductive quenching of the excited state of (2,1) can be improved, the efficiency and stability of the photocatalyst should also be improved. For this purpose, the stronger reducing agent 1-(4-methoxybenzyl)-1,4-dihydronicotinamide (MeO-BNAH)^∥, having the redox potential 0.50 V vs. SCE (25), was used as a sacrificial electron donor instead of BNAH [redox potential = 0.57 V (25) and η_q = 59%]. As a result, η_q increased to 77% (Table 1, entry 2: Fig. S1B), the photocatalytic ability of (2,1) was substantially improved (i.e., Φ_HCOOH = 0.061 in the LIC1 and TON_HCOOH = 671 in the LIC2) compared with that using BNAH (i.e., Φ_HCOOH = 0.041 and TON_HCOOH = 562). The TOF of formic acid formation (TOF_HCOOH) reached 11.6 min^-1 in the LIC3 (7.8 min^-1 in the case using BNAH)**. To the best of our knowledge, this system affords the fastest photocatalytic reduction of CO₂ reported to date.

It is noteworthy that the formation rate of formic acid using (2,1) as a photocatalyst was strongly dependent on the irradiated light flux as shown in Fig. S9B. Even in the reaction condition with the highest light flux (LIC3), TOF was still increasing, and, therefore, the rate-limiting process of the photocatalytic CO₂ reduction using (2,1) was the excitation of the photosensitizer unit; i.e., higher TOF_HCOOH should be achievable by using a stronger light source.

Finally, the reaction mechanism of formation of formic acid using the good photocatalysts; i.e., (1,1) and (2,1), is discussed. Formation of formic acid requires two-electron reduction of CO₂. It is clear that the initial one-electron transfer to the catalyst unit proceeds from the one-electron-reduced species (OER) of the photosensitizer unit, which is produced via the reductive quenching process of the excited photosensitizer unit by BNAH. The second-electron reduction of the catalyst unit, which is probably converted to the CO₂-adduct form via chemical processes after the first reduction, should also proceed through the same sequence—i.e., photochemical formation of the OER of the photosensitizer unit and then its donation of another electron to the catalyst unit—because the reductant BNAH works only as a one-electron donor. Therefore, the maximum quantum yield of formic acid formation (Φ_HCOOH) should be 0.5 using the photocatalytic systems (27). Since the quantum yield of formation of formic acid was not dependent on both the light intensity and the concentration of the photocatalyst, the sequential two-step photochemical electron transfer processes from the OER of the photosensitizer unit to the catalyst unit should proceed intramolecularly; namely, the photocatalytic formation of formic acid proceeds with participation of only one supramolecule and does not require participation of two molecules of the photocatalyst. The disproportionation of one-electron reduced supramolecules is not the main process for the formation of formic acid.

Materials and Methods

Details of synthesis of ruthenium(II) complexes, general experimental conditions, and photocatalytic reaction conditions, and analysis of formic acid are provided in SI Text.

Conclusion

We successfully developed unique supramolecular photocatalysts constructed with various ratios of two kinds of ruthenium(II) complexes for reduction of CO₂ to formic acid with high selectivity and durability. The ratio between photosensitizer units and catalyst units strongly affected the photocatalytic activities—i.e., the higher ratio of the photosensitizer unit caused a higher yield of formic acid—and (2,1) exhibited the highest photocatalytic ability (Φ_HCOOH = 0.061, TON_HCOOH = 671, TOF_HCOOH = 11.6 min^-1). The following sequence takes place twice for the formation of one formic acid molecule in the photocatalytic reaction: First, the excitation of the photosensitizer unit; second, the reductive quenching of the excited state of the photosensitizer unit by BNAH; and third, the intramolecular electron transfer from the reduced photosensitizer unit to the catalyst unit. The addition of CO₂ into the catalytic site probably occurs between the two sequences.