Cyclodextrin‐Enhanced Photocatalytic Hydrogen Evolution Based on Formate Decomposition With Bio/Metal Catalysts

ABSTRACT

Formic acid is an attractive hydrogen carrier in terms of hydrogen capacity and safety. Focusing on hydrogen release from formic acid, the developments of catalysts with easy controllability in mild conditions have been desirable. Previous study reported a hybrid catalytic system of enzymes, photosensitizers, and metal nanoparticles for controllable hydrogen production with visible light. However, methyl viologen, which is a toxic compound, was utilized as an electron mediator from photosensitizers to metal nanoparticles. 9‐Mesityl‐10‐methyl acridinium (Mes‐Acr⁺) is one of the promising candidates which hold a long‐lived charge‐separated state for the removal of methyl viologen from the hybrid system to enhance the safety and the catalytic activity. In this work, cyclodextrins (CDs) were employed to form a complex with Mes‐Acr⁺ using supramolecular interactions such as ionic and hydrophobic interactions, which improved photocatalytic activity in aqueous solution. This supramolecular photocatalyst was finally applied to photocatalytic hydrogen production with a system of formate dehydrogenase, NAD⁺, Mes‐Acr⁺, CD, and colloidal platinum nanoparticles dispersed by polyvinylpyrrolidone. An ultimate yield from formate to hydrogen reached 24% at pH 6.0, indicating a rate‐determining step would be electron transfer from Mes‐Acr radical to Pt‐PVP to reduce protons.

Hydrogen evolution from formate with a hybrid system of enzyme, donor‐acceptor linked photocatalyst‐cyclodextrin complex (Mes‐Acr⁺‐CD) and platinum nanoparticle is reported. The use of CD as a dispersant for Mes‐Acr⁺ is a perfect strategy to enhance photocatalytic activity in aqueous solution without deactivation of the enzyme. This work also reveals that ring size and substituents of CD could influence hydrogen production in the system significantly.

1. Introduction

Growing energy demand around the world poses several environmental problems, such as global warming and energy depletion, while causing mass consumption of fossil fuels [1]. Hydrogen has been drawing much attention as clean energy since water is the only by‐product after its combustion, and the gas could be produced through various processes from a wide variety of raw materials, including water, natural gas, and biomass, which would realize a stable supply of energy [2, 3, 4]. General method for hydrogen production is natural gas reforming, where methane reacts with water at around the high temperature of 800°C [5, 6]. Another promising hydrogen‐producing method is water electrolysis or photolysis using renewable energy such as sunlight [2, 7, 8, 9, 10]. However, molecular hydrogen is not suitable for large‐scale storage at ambient conditions because of its low volumetric energy density of 0.0108 MJL⁻¹ [11].

Formic acid is one of the potential candidates as a hydrogen carrier because it has a hydrogen capacity of 4.4 wt% (52 gL⁻¹) and low toxicity [12, 13, 14]. In addition, formic acid could be a renewable energy carrier based on the formic acid‐CO₂ cycle. Therefore, both homogeneous and heterogeneous metal catalysts, such as Ir, Au, Ru, and Rh for CO₂ hydrogenation or formic acid dehydrogenation have been developed [15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26]. For instance, colloidal platinum nanoparticles dispersed by polyvinylpyrrolidone (Pt‐PVP) can catalyze selective formic acid decomposition into hydrogen and carbon dioxide. There are two downsides in metal‐catalyzed hydrogen production from formic acid: a difficulty in reaction control after the addition of catalysts and a strong acidic condition of roughly pH 3.8 [27, 28, 29]. To solve these two issues, a hybrid catalytic system comprised of formate dehydrogenase (FDH), water‐soluble zinc porphyrins, Pt‐PVP, and two sorts of electron mediators was developed [30]. By using this system, hydrogen production from formate was controlled by visible‐light irradiation in a mild pH condition. Yet, methyl viologen (MV²⁺), which has the environmental toxicity derived from the cycle of MV⁺/MV²⁺ and the production of reactive oxygen species (ROS), was used as an electron carrier from porphyrins to Pt‐PVP for charge separation.

Donor‐acceptor linked photocatalysts could be useful for the removal of methyl viologen because they have a long‐lived charge‐separated state itself [31]. 9‐Mesityl‐10‐methyl acridinium (Mes‐Acr⁺) has a donor‐acceptor linked structure (Figure 1) and is one of promising photocatalysts for hydrogen production from formate in a hybrid catalytic system without methyl viologen since Mes‐Acr⁺ could become an excited state, which has the strong oxidizing ability only under light irradiation, leading to the enough reducing ability to reduce proton: ³(Acr^•–Mes^•+) (E _ox  =  −0.57 V vs. SCE) [32, 33, 34]. Therefore, the generation of this potentially toxic and highly reactive excited species could be also controlled with light irradiation even if considered for industrial scale. Furthermore, it was reported that Mes‐Acr⁺ showed the superior photocatalytic activity without MV²⁺ compared with that in the presence of MV²⁺ [33].

FIGURE 1.

Chemical structures of Mes‐Acr⁺ and CDs.

Acridinium‐based photocatalysts are often used in organic solvents such as acetonitrile because of its poor solubility in water [35]. For the utilization of Mes‐Acr⁺ in aqueous solution, where enzymes, including FDH can exhibit high catalytic activity, organic/water cosolvents or dispersants are helpful. For example, acetonitrile/water co‐solvent was used in hydrogen production with a hybrid catalytic system, and surfactants such as sodium dodecyl sulfate (SDS) could be effective dispersants of Mes‐Acr⁺ [33, 36]. Although FDH was deactivated in these solutions, cyclodextrins (CDs), which are cyclic oligosaccharides (Figure 1), do not inhibit FDH‐catalyzed formate oxidation as seen in Figure S1. Hence, CDs are expected to form an inclusion complex with Mes‐Acr⁺ by using supramolecular interaction and enhance the photocatalytic activity of Mes‐Acr⁺, generally denoted as turnover frequency (TOF) and turnover number (TON), for hydrogen production from formate in aqueous solution, to use Mes‐Acr⁺ as a catalyst and reduce the amount of used Mes‐Acr⁺ considering the cost of Mes‐Acr⁺.

In this work, fluorescence analysis on Mes‐Acr⁺ in the presence of CDs, which were α‐CD, β‐CD, M‐β‐CD, SBE‐β‐CD, and γ‐CD (Figure 1), was performed to investigate the stoichiometry and stability of an inclusion complex: Mes‐Acr⁺‐CD. Subsequently, Mes‐Acr⁺‐CD was applied to a photo‐redox system of NADH, Mes‐Acr⁺, CD, and Pt‐PVP for hydrogen production, and the effects of CDs in view of ring size and substituents were also elucidated. Finally, FDH‐catalyzed formate oxidation was combined with the photo‐redox system of NAD⁺, Mes‐Acr⁺, CD, and Pt‐PVP without methyl viologen for hydrogen production from formate shown in Figure 2.

FIGURE 2.

Hydrogen production based on formate decomposition with the hybrid system of FDH, NAD⁺, Mes‐Acr⁺‐CD, and Pt‐PVP.

2. Results and Discussion

2.1. Stoichiometry and Stability of Mes‐Acr⁺‐CD Complexes

To evaluate stoichiometry and stability of Mes‐Acr⁺‐CD inclusion complexes in aqueous solution, fluorescence response of Mes‐Acr⁺ at various CDs concentrations was investigated. α‐CD, β‐CD, M‐β‐CD, SBE‐β‐CD, and γ‐CD were employed at 0–15 mM with Mes‐Acr⁺ (0.50 µM) to form Mes‐Acr⁺‐CD complexes in 200 mM phosphate buffer (pH 7.0) at 25°C. Fluorescence intensity at 494 nm was gradually enhanced as the concentration of CDs increased in Figure S2. In the presence of 15 mM of each CD, the intensity with SBE‐β‐CD showed the strongest, whereas γ‐CD showed the weakest, ranging from 69.6 to 309 at 494 nm. Mes‐Acr⁺ has two fluorescent bands around 500 and 570 nm based on locally excited singlet state (LE^s) and charge transfer from mesityl substituent to acridinium to form singlet state (CT^s), respectively. In addition, Mes‐Acr⁺ shows the strongest fluorescent band around 570 nm in polar organic solvents such as acetonitrile at 25°C, while the maximum fluorescence intensity is observed around 500 nm in nonpolar organic solvents such as 1,2‐dichloroethane [37]. Therefore, these results reveal that Mes‐Acr⁺, especially mesityl substituent, was captured in a nonpolar environment, the cavity of CD, and LE^s of Mes‐Acr⁺ became more stable than CT^s through the formation of Mes‐Acr⁺‐CD complexes.

Based on fluorescence analysis of Mes‐Acr⁺ in the presence of CDs, the stoichiometry and stability constant of Mes‐Acr⁺ were calculated with the Benesi‐Hildebrand equation [Equation (1)].

$$\begin{array}{c}\frac{1}{I_{\mathrm{F}}-I_0}=\frac{1}{[\mathrm{Mes}-\mathrm{Ac}{\mathrm{r}}^{+}]\left(I^{\prime }-I_0\right)}+\frac{1}{\left[\mathrm{Mes}-\mathrm{Ac}{\mathrm{r}}^{+}\right]\left(I^{\prime }-I_0\right)K\left[\mathrm{CD}\right]}\end{array}$$ (1)

I _F is the fluorescence intensity of Mes‐Acr⁺ with CD, I ₀ is that without CD, I’ is that with excess of CD, and K is the stability constant [38, 39]. Figure 3 shows Benesi‐Hildebrand plots of Mes‐Acr⁺ with each CD at the concentrations ranging from 0–15 mM. 1:1 complexes with Mes‐Acr⁺ were formed in the presence of all sorts of CDs since a linear relationship between the inverse of CD concentration and fluorescence intensity was observed. Furthermore, stability of the complexes was estimated from the stability constant, which was calculated with the slope and intercept of the plots. Stability constants are summarized in Table 1. To investigate the effects of ring size of CD on the stability of the complexes, α‐CD, β‐CD, and γ‐CD should be compared since they are ring‐shaped compounds composed of 6, 7, and 8 glucose by α‐1,4‐glycosidic bond, respectively. γ‐CD formed the most stable complex with Mes‐Acr⁺ at the stability constant of 421 M⁻¹ among the three sorts of CDs, followed by α‐CD and β‐CD with 83.5 M⁻¹ and 54.0 M⁻¹, which indicates the CD with a larger ring size is favorable to include Mes‐Acr⁺ in its cavity (α‐CD: 0.57 nm; β‐CD: 0.78 nm; γ‐CD: 0.95 nm) [40]. Moreover, in terms of substituent effects of CD, SBE‐β‐CD showed the highest stability constant with 1390 M⁻¹, which is 26‐fold higher than that of β‐CD, indicating that not only hydrophobic interaction between the cavity of SBE‐β‐CD and mesityl substituent, but also electrostatic interaction between anionic sulfobutylether and cationic acridinium assumes a crucial role in the formation of Mes‐Acr⁺‐CD inclusion complex [41]. In addition, the stability constant of M‐β‐CD with 158 M⁻¹ was roughly three times as large as that of β‐CD because of the higher solubility of M‐β‐CD in aqueous solution (β‐CD:18.5 mg/mL, M‐β‐CD: ≥500 mg/mL) [42].

FIGURE 3.

Benesi‐Hildebrand plots of Mes‐Acr⁺ with CDs (0‐15 mM).

TABLE 1.

Stability constant (K) obtained from Benesi‐Hildebrand plots of CDs.

	Stability constant (K / M−1)
α‐CD	83.5
β‐CD	54.0
M‐β‐CD	158
SBE‐β‐CD	1390
γ‐CD	421

2.2. Effects of CDs on Hydrogen Production from NADH in Hybrid Catalytic System

The ring size and substituent of CDs could have an effect on the stability of the CD‐based complexes with Mes‐Acr⁺ in aqueous solution. Hence, the photocatalytic activity of Mes‐Acr⁺ for hydrogen production is likely to be subject to the types of CDs. To investigate CD‐dependence of photocatalytic activity of Mes‐Acr⁺ for hydrogen production, NADH (50 µmol) was utilized as a reducing agent in the hybrid catalytic system of Mes‐Acr⁺ (2.5 µmol), CDs (25 µmol), and Pt‐PVP (0.25 µmol) in 200 mM phosphate buffer (pH 7.0, 5.0 mL). α‐CD, β‐CD, M‐β‐CD, SBE‐β‐CD, and γ‐CD were also employed to form the complexes with Mes‐Acr⁺ in each sample. Details of the materials used and all experimental procedures are described in the Experimental Section/Methods. As seen in Figure 4, the sample solution with SBE‐β‐CD performed the highest catalytic activity for hydrogen production with the amount of 0.26 µmol h⁻¹, followed by 98, 22, 12, and 11 nmol h⁻¹ for γ‐CD, M‐β‐CD, β‐CD, and α‐CD, respectively.

FIGURE 4.

Hydrogen production velocity and the initial concentration of the complex with the hybrid catalytic system of NADH (50 µmol), Mes‐Acr⁺ (2.5 µmol), CD (25 µmol), and Pt‐PVP (0.25 µmol) in 200 mM phosphate buffer (pH 7.0). Errors were calculated as the average of multiple trials.

It is difficult to simply associate hydrogen production velocity in the hybrid system with the stability of Mes‐Acr⁺‐CD complexes to discuss the result kinetically, but the initial concentration of Mes‐Acr⁺‐CD ([Mes‐Acr⁺‐CD]₀) was calculated with the following equation [Equation (2)] since the complex of CD with Mes‐Acr⁺ forms 1:1 complex.

$$\begin{array}{c}K=\frac{{\left[\mathrm{Mes}-\mathrm{Ac}{\mathrm{r}}^{+}-\mathrm{CD}\right]}_0}{{\left[\mathrm{Mes}-\mathrm{Ac}{\mathrm{r}}^{+}\right]}_0{\left[\mathrm{CD}\right]}_0}\end{array}$$ (2)

K is the stability constant of Mes‐Acr⁺‐CD given through Benesi‐Hildebrand plots by the fluoresence analysis, [Mes‐Acr⁺]₀ and [CD]₀ were the initial concentrations of Mes‐Acr⁺ and CD. 0.14, 0.10, 0.22, 0.43, and 0.33 mM were given for α‐CD, β‐CD, M‐β‐CD, SBE‐β‐CD, and γ‐CD, respectively, as shown in Figure 4. This calculation result means 86% and 66% of Mes‐Acr⁺ was included in the cavity of SBE‐β‐CD and γ‐CD, whereas β‐CD captured 20% of Mes‐Acr⁺ in its hole, because the more negatively charged substituents or larger ring size CDs have, the more stable the complexes become. Thus, it is obvious that Mes‐Acr⁺ dispersed by CDs can work as photocatalysts in aqueous solution effectively, and electrostatic interaction and hydrophobic interaction based on the substituent and ring size of CDs could enhance the photocatalytic activity of Mes‐Acr⁺ for efficient hydrogen production in the hybrid system, which could be achieved with negatively‐charged SBE‐β‐CD.

2.3. Hydrogen Production From Formate in Hybrid Catalytic System

Formic acid is a desirable hydrogen carrier in terms of safety and efficiency for hydrogen storage and transportation, while MV²⁺ is likely to inhibit hydrogen production in the hybrid system [33]. Therefore, a sample solution containing HCOONa (50 µmol), FDH (0.95 U), NAD⁺ (25 µmol), Mes‐Acr⁺ (2.5 µmol), SBE‐β‐CD (25 µmol), Pt‐PVP (0.25 µmol), and MV²⁺ (25 µmol) in 200 mM phosphate buffer (pH 6.0) was irradiated with 250 W halogen lamp and incubated in a thermostatic chamber at 30.5°C under magnetic stirring. Figure 5 shows the time course of the amount of hydrogen production in the presence or absence of MV²⁺, and Figure S6 illustrates the chart of gas chromatography with MV²⁺. The initial hydrogen production rate with MV²⁺ largely decreased to nearly half that in the absence of MV²⁺. After 3 h irradiation, the amount of hydrogen production was 2.4 µmol and 3.3 µmol for the presence and absence of MV²⁺, respectively. This result indicates that MV²⁺ could act as an inhibitor of hydrogen production in the hybrid system with Mes‐Acr⁺.

FIGURE 5.

The time course of the amount of hydrogen production from formate with the hybrid system of HCOONa (50 µmol), FDH (0.95 U), NAD⁺ (25 µmol), Mes‐Acr⁺ (2.5 µmol), SBE‐β‐CD (25 µmol), Pt‐PVP (0.25 µmol) in the presence or absence of MV²⁺ (25 µmol) in 200 mM phosphate buffer (pH 6.0).

Furthermore, hydrogen production from HCOONa (50 µmol) in the system of FDH (0.95U), NAD⁺ (25 µmol), Mes‐Acr⁺ (2.5 µmol), SBE‐β‐CD (25 µmol), and Pt‐PVP (0.25 µmol) in 200 mM phosphate buffer (pH 7.0, 5.0 mL) was performed, monitoring the amount of formate consumed by FDH‐catalyzed formate oxidation. Figure 6 illustrates the time course of the amount of hydrogen and carbon dioxide production and formate consumption in the system. Charts of gas and ion chromatographs are also provided in Figures S7, S8 and S9, respectively. After 5 h irradiation, hydrogen was produced in the amount of 1.2 µmol, while 5.1 µmol of carbon dioxide was produced, consuming 15 µmol of formate. One point to note is that hydrogen and carbon dioxide should be produced at the same molar ratio theoretically when the same amount of formate is also consumed in the system. Focusing first on hydrogen and carbon dioxide production, the amount of hydrogen produced was around one‐fourth of that of carbon dioxide, which indicates formate oxidation catalyzed by FDH proceeds much faster than the subsequent reaction of hydrogen production from NADH; formate oxidation catalyzed by FDH proceeds at the velocity of 2.3 µM s⁻¹ [30]. Moreover, the amount of formate consumption was roughly 13 times as large as that of hydrogen production for the same reason. Regarding carbon dioxide production and formate consumption, the amount of formate consumption was triple that of carbon dioxide production. This result could be explained based on the solubility of carbon dioxide in aqueous solution. Thus, 66% of produced carbon dioxide is estimated to dissolve in phosphate buffer solution as bicarbonate or carbonate at pH 7.0.

FIGURE 6.

The time course of the amount of gas production and formate consumption in the hybrid system of HCOONa (50 µmol), FDH (0.95U), NAD⁺ (25 µmol), Mes‐Acr⁺ (2.5 µmol), SBE‐β‐CD (25 µmol), and Pt‐PVP (0.25 µmol) in 200 mM phosphate buffer (pH 7.0). Errors were calculated as the average of multiple trials.

Additionally, hydrogen production under light on/off every one hour and the lacking conditions of each component of the hybrid system was investigated to clarify if the system is constructed successfully. As seen in Figure 7, hydrogen was produced only under an irradiation condition, while carbon dioxide was produced, consuming formate in both irradiation and dark conditions. Focusing on the hydrogen production rate (v _Hydrogen) in irradiation condition, v _Hydrogen between 0–1 h was 0.21 µmol h⁻¹, whereas those between 2–3 h and 4–5 h were 0.35 µmol h⁻¹ and 0.31 µmol h⁻¹, respectively. This trend appeared because NADH produced by FDH‐catalyzed formate oxidation was concentrated under dark condition, and then it was consumed in the light‐driven hydrogen production system. Thus, the result supports FDH‐catalyzed formate oxidation and light‐driven hydrogen production system were working separately, and hydrogen production in this system was promoted by light irradiation. Table S1 also indicates that hydrogen was produced efficiently only in the presence of all components, although no or less hydrogen production was observed under the lacking condition. It is also noteworthy that TOF in the condition with all components significantly increased to 16.4‐fold that in the condition lacking SBE‐β‐CD, which could contribute to the decrease in the amount and cost of used Mes‐Acr⁺. This result confirms that all compounds are essential to produce hydrogen efficiently in the system.

FIGURE 7.

Light‐responsiveness of gas production and formate consumption in the hybrid catalytic system of HCOONa (50 µmol), FDH (0.95U), NAD⁺ (25 µmol), Mes‐Acr⁺ (2.5 µmol), SBE‐β‐CD (25 µmol), and Pt‐PVP (0.25 µmol) in 200 mM phosphate buffer (pH 7.0). Errors were calculated as the average of multiple trials.

Next, pH dependence of hydrogen production in the system was examined. A sample solution containing HCOONa (50 µmol), FDH (0.95 U), NAD⁺ (25 µmol), Mes‐Acr⁺ (2.5 µmol), SBE‐β‐CD (25 µmol), and Pt‐PVP (0.25 µmol) in 200 mM phosphate buffer (pH 6.0, 7.0, and 8.0) resulted in the production of hydrogen with pH dependence as shown in Figure 8. Charts of the gas chromatograph are also provided in Figures S7, S10 and S11. Hydrogen production velocity at pH 6.0 was 0.99 µmol h⁻¹, while those at pH 7.0 and 8.0 were 0.18 µmol h⁻¹ and 0.017 µmol h⁻¹. After 25 h irradiation at pH 6.0, a yield from formate to hydrogen reached 24% (= 12 µmol), with TONs of each catalyst being 500, 5.0, and 50 for FDH, Mes‐Acr⁺, and Pt‐PVP, respectively.

FIGURE 8.

The time course of the amount of produced hydrogen in the hybrid catalytic system of HCOONa (50 µmol), FDH (0.95 U), NAD⁺ (25 µmol), Mes‐Acr⁺ (2.5 µmol), SBE‐β‐CD (25 µmol), and Pt‐PVP (0.25 µmol) in 200 mM phosphate buffer (pH 6.0, 7.0, and 8.0). Errors were calculated as the average of multiple trials.

Here, a rate‐determining step in the hybrid system is likely to be electron transfer from Mes‐Acr radical to Pt‐PVP to reduce protons since pH dependence of hydrogen production was observed. Mes‐Acr radical is a noncharged compound, which could cancel electrostatic interaction with SBE‐β‐CD to form the stable inclusion complex. Therefore, the production of Mes‐Acr radical was investigated with UV‐vis absorption analysis monitoring around 520 nm to discuss the stability of the hybrid catalytic system [33]. NADH (50 µmol) was employed as a reducing agent in the system of Mes‐Acr⁺ (2.5 µmol) and SBE‐β‐CD (25 µmol) in 200 mM phosphate buffer (pH 7.0). Figure S12 illustrates the time course of the UV‐vis spectrum in the system. The absorbance at 520 nm gradually increased until 10 min, witnessing a modest red shift to around 550 nm. Subsequently, the absorbance at 550 nm was consistently decreased. The production of Mes‐Acr radical through the reduction of Mes‐Acr⁺ by NADH was confirmed based on the decrease in the absorption of Mes‐Acr⁺ at around 420 nm and the rise in the absorption of Mes‐Acr radical at around 520 nm. It is noteworthy that a red shift and the decrease in the absorption at 520–550 nm would be derived from a multimerization and aggregation of Mes‐Acr⁺ or Mes‐Acr radical, respectively, leading to a deactivation of the photocatalytic activity of Mes‐Acr⁺.

3. Conclusion

CDs were employed as dispersants of Mes‐Acr⁺, which has a long‐lived charge‐separated state, to enhance photocatalytic activity of Mes‐Acr⁺ in aqueous solution. The stoichiometry and stability of the Mes‐Acr⁺‐CD complexes were elucidated through fluorescence analysis. Specifically, Mes‐Acr⁺ formed 1:1 complex with SBE‐β‐CD, which showed the highest stability and photocatalytic activity for hydrogen production in the hybrid system of NADH, Mes‐Acr⁺, SBE‐β‐CD, and Pt‐PVP, as there was an ionic interaction between Mes‐Acr⁺ and SBE‐β‐CD. Mes‐Acr⁺‐CD complex was finally applied to hydrogen production from formate in the hybrid system of FDH, NAD⁺, Mes‐Acr⁺, SBE‐β‐CD, and Pt‐PVP without methyl viologen, which inhibits hydrogen production in the system. The system was successfully developed; consuming formate through FDH‐catalyzed formate oxidation reaction and subsequently producing hydrogen in the photo‐redox system from regenerated NADH with 66% of the total produced carbon dioxide captured in buffer solution; improving TOF of Mes‐Acr⁺ 16‐fold; controlling hydrogen production with light irradiation. pH dependence in hydrogen production was also observed, which indicates a rate‐determining step was electron transfer from Mes‐Acr radical to Pt‐PVP to reduce protons. The limitation will be overcome through improvements in the catalytic activity of organic photocatalysts and metal nanoparticles to promote the slowest process and suppress the deactivation of catalysts.

4. Experimental Section/Methods

4.1. Materials

NAD⁺ and NADH were purchased from Oriental Yeast Co., Ltd. Mes‐AcrClO₄ was purchased from Tokyo Chemical Industry Co., Ltd. Sodium formate, disodium hydrogenphosphate dodecahydrate, sodium dihydrogenphosphate dihydrate, α‐CD, β‐CD, M‐β‐CD, and γ‐CD were purchased from Wako Pure Chemical Co. Ltd. SBE‐β‐CD was purchased from Combi‐Blocks Inc. Pt‐PVP was purchased from Tanaka Holdings Co. Ltd. Pt concentration in Pt‐PVP solution was estimated to be 4.0 wt%. The particle size of Pt‐PVP was estimated to be around 2.0 nm using a transmission electron microscope (TEM) image shown in the analysis table published by Tanaka Holdings Co., Ltd. FDH was supplied by Merck Ltd. All chemicals had analytical grade or the highest grade available.

4.2. Fluorescence Analysis on Mes‐Acr⁺ in the Presence of CDs

A sample solution consisting of Mes‐Acr⁺ (0.50 µM) and CD (0‐15 mM) in 200 mM phosphate buffer (pH 7.0, 5.0 mL) was incubated at 25°C for 30 min. All samples were measured under an Ar atmosphere by fluorescence spectroscopy (RF‐5300 PC, SHIMADZU Corporation). The excitation wavelength was 425 nm, and data intervals were 1.0 nm. Fluorescence intensity at 494 nm after 30 min was monitored to be applied to the Benesi‐Hildebrand plot.

4.3. General Methods for Hydrogen Production with the Hybrid Catalytic System

To determine the amount of hydrogen and carbon dioxide, gas chromatography (GC‐2014, SHIMADZU Corporation) incorporated with a thermal conductivity detector was utilized. Activation charcoal column (column length was 3 mm I.D. × 2m) detected sample gas in Ar carrier gas with the flow rate adjusted to 30.0 mL min⁻¹. Sample gas was injected, flowed to column and detected at the temperature of 100.0, 70.0, and 100.0°C. Hydrogen and carbon dioxide were detected at around 0.8‐1.4 min and 13.5‐17.0 min, respectively. Chromatogram and calibration curve with good linear relationship between the amount of produced gas and peak area was prepared as shown in Figures S3 and S4. Additionally, the concentration of formate in a sample solution was determined by ion chromatography (Metrohm, EcoIC) equipped with an electrical conductivity detector and an ion exclusion column: Metrosep Organic Acids 250/7.9 Metrohm; column size: 7.8 × 250 nm; composed of 9 µm polystyrene divinylbenzene copolymer with sulfonic acid groups. 1.0 mM perchloric acid and 50 mM lithium chloride in aqueous solution were used as an eluent and a regenerant, respectively. Flow rate of eluent solution was adjusted to be 0.5 mL min⁻¹. Formate was detected around at 13.5–15.0 min. Chromatogram and calibration curve with a good linear relationship between the concentration of formate and peak area was prepared as shown in Figure S5.

4.4. Hydrogen Production From NADH With the Hybrid Catalytic System

A sample solution of NADH (50 µmol), Mes‐Acr⁺ (2.5 µmol), and Pt‐PVP (0.25 µmol) with α‐CD, β‐CD, M‐β‐CD, SBE‐β‐CD, or γ‐CD in 200 mM phosphate buffer (pH 7.0, 5.0 mL) was added to an eggplant‐type flask of 17.2‐19.3 mL with a side tube and high vacuum stopcock. Freeze‐pump‐thaw cycles (six times) were performed, and then the gas phase was flushed with Ar gas for 10 min. Finally, the sample solution was irradiated by a 250 W halogen lamp in a thermostatic chamber at 30.5°C with a magnetic stirrer.

4.5. Hydrogen Production From Formate With the Hybrid Catalytic System

5.0 mL of a sample solution including HCOONa (50 µmol), NAD⁺ (25 µmol), Mes‐Acr⁺ (2.5 µmol), SBE‐β‐CD (25 µmol), and Pt‐PVP (0.25 µmol) in 200 mM phosphate buffer (pH 7.0) was prepared in an eggplant‐type flask of 17.2‐19.3 mL with a side tube and high vacuum stopcock. Freeze‐pump‐thaw cycles (six times) were performed, and then the gas phase was flushed with Ar gas for 10 min. FDH (0.95 U) was subsequently added to the deaerated solution. Finally, the sample solution was irradiated by 250 W halogen lamp in a thermostatic chamber at 30.5°C with a magnetic stirrer.

Conflicts of Interest

The authors declare no conflict of interest.

Supporting information

Charts of UV‐vis absorption/fluorescence spectrum and gas/ion chromatograph are provided in the Supporting Information. Supporting Information is available from the Wiley Online Library or from the author.

Supporting File: chem.202502990SI‐R1.docx.

Yoshikawa S. and Amao Y., “Cyclodextrin‐Enhanced Photocatalytic Hydrogen Evolution Based on Formate Decomposition With Bio/Metal Catalysts.” Chemistry – A European Journal 32, no. 6 (2026): e02990. 10.1002/chem.202502990