Iron(II) bis(pyrazolyl)phenanthroline complexes as robust and efficient homogeneous catalysts for CO2-to-CO conversion under visible light - 2026.08.27

Imported from: /opt/downloadPDF/chemwiki_pubstore/d716ca1fc4a4befe4e1b0f04b5e6ea30/Ferreira Jr. et al. - 2026 - Iron(II) bis(pyrazolyl)phenanthroline complexes as robust and efficient homogeneous catalysts for CO_6a900f1557b78.pdf, /opt/downloadPDF/chemwiki_pubstore/d716ca1fc4a4befe4e1b0f04b5e6ea30/SI_6a900f1559124.pdf

Abstract Summary

This study describes a homogeneous molecular photocatalytic system for the reduction of CO₂ to CO under visible light using iron(II) bis(pyrazolyl)phenanthroline complexes as catalysts. The catalytic mixtures used a ruthenium polypyridyl photosensitizer, BIH as sacrificial electron donor, and mixed MeCN/H₂O solvent under CO₂ saturation.

All four iron complexes were catalytically active for CO formation. The best overall catalyst under the standard comparison conditions was Fe2, which reached a reported TON_CO of 1318 with 84% CO selectivity. Under lower catalyst loading, Fe2 reached much higher turnover numbers, up to 23,138 for CO, while maintaining high CO selectivity. The article also reports that water is essential for efficient catalysis, that the catalysis remains homogeneous according to mercury poisoning experiments, and that deactivation is mainly associated with deterioration of the photosensitizer rather than the iron catalyst.

Advances and Special Progress

A central advance is the introduction of a new family of iron(II) catalysts based on bis(pyrazolyl)phenanthroline ligands for visible-light-driven CO₂-to-CO conversion. The article presents this ligand framework as an underexplored alternative to more established iron systems based on quaterpyridine, quinquepyridine, salophen, and related ligand sets.

The work also shows that systematic ligand modification changes catalytic behavior. Neutral, electron-donating, sterically bulky, and electron-withdrawing pyrazolyl substituents were compared within a common scaffold, allowing direct structure-activity analysis. Among the tested complexes, Fe2 gave the highest activity, whereas Fe4 gave the highest CO selectivity in the series.

Another important advance is the combination of high activity with mixed-solvent water compatibility. The article explicitly shows that 7.5-10% water is beneficial and in practice essential for efficient CO₂ reduction, because anhydrous conditions give almost no CO. This is chemically significant because water both supports proton-coupled electron transfer and stabilizes intermediates according to the mechanistic interpretation presented.

The article also emphasizes durability-related insight. Control experiments, repeated irradiation studies, UV-Vis monitoring, and replenishment of the photosensitizer support the conclusion that catalyst deactivation is not mainly due to iron catalyst collapse but instead is largely linked to photosensitizer degradation. Mercury poisoning experiments further support a homogeneous catalytic regime rather than nanoparticle-mediated catalysis.

Finally, the study provides mechanistic progress by combining cyclic voltammetry, DFT, TD-DFT, and Stern-Volmer quenching analysis. The data support ligand-centered reductions on the iron complexes, identify the excited-state quenching behavior of the photosensitizer, and support a catalytic pathway in which one-electron-reduced iron species are competent for CO₂ activation.

Additional Remarks

The chemistry is relevant because CO is a useful reduced carbon product and an industrial feedstock. Photochemical CO₂-to-CO conversion is also a useful model reaction for studying multi-electron and proton-coupled small-molecule activation under mild conditions.

At the same time, the system remains a sacrificial photochemical platform rather than a fully sustainable closed cycle. The reaction requires BIH as sacrificial electron donor and [Ru(bpy)₃]²⁺ as photosensitizer. The article therefore demonstrates efficient catalytic chemistry, but it does not remove reliance on a noble-metal photosensitizer or sacrificial reagent.

The solvent system is also important to interpret correctly. The highest activity in the standard experiments occurs in mixed MeCN/H₂O rather than in dry acetonitrile. However, too much water decreases activity. The article attributes this decrease partly to the poor solubility of BIH at high water content, which limits effective quenching of the excited photosensitizer.

Competition between CO formation and H₂ evolution is present throughout the study. The catalysts are selective for CO under the reported conditions, but H₂ remains a measurable side product. No significant CH₄ or formate production was detected in the experiments described.

A further practical limitation is long-term photostability. The catalytic activity rises quickly at early irradiation times and then approaches a plateau. The article links this behavior mainly to photosensitizer photodegradation, supported by hypochromism in UV-Vis spectra and by recovery of activity after photosensitizer replenishment.

Content of the Published Article in Detail

The molecular photocatalytic system contains four closely related iron(II) complexes, designated Fe1-Fe4, each bearing a tetradentate bis(pyrazolyl)phenanthroline ligand and two coordinated water molecules. The catalysts are molecular, homogeneous iron complexes in distorted octahedral coordination environments. The standard photocatalytic mixture also contains [Ru(bpy)₃]²⁺ as photosensitizer and BIH as sacrificial electron donor in CO₂-saturated MeCN/H₂O.

The photocatalytic experiments were carried out in a borosilicate photoreactor containing 4.0 mL of solution, irradiated with blue light centered at 462 nm. Under the standard catalyst comparison conditions, the mixture contained 50 μM iron catalyst, 0.3 mM [Ru(bpy)₃]²⁺, and 0.11 M BIH in CO₂-saturated MeCN/H₂O. Product analysis of the gas phase was performed by gas chromatography. CO and H₂ were observed. The article states that no significant formate or CH₄ was detected.

The photophysical analysis shows that the strong UV absorption bands of the iron complexes are mainly ligand-centered π-π* transitions. Weak lower-energy absorptions near 450-480 nm were assigned to forbidden transitions. The article explicitly states that these data do not indicate intrinsic photosensitizing ability of the iron complexes under the catalytic conditions. Instead, light absorption is carried by the ruthenium photosensitizer.

Mechanistically, the article combines DFT and Stern-Volmer analysis to discuss excited-state electron transfer. The HOMO of BIH is calculated above that of [Ru(bpy)₃]²⁺, which supports reductive quenching of the excited ruthenium photosensitizer by BIH. Stern-Volmer measurements also show that BIH quenches the emission of [Ru(bpy)₃]²⁺ more efficiently than the iron complexes do. The article therefore identifies BIH as the dominant quencher of the excited photosensitizer.

The iron complexes can also quench the excited state of [Ru(bpy)₃]²⁺, but less efficiently. The energy-level discussion in the article indicates that oxidative quenching of the excited photosensitizer by the iron complexes is energetically feasible, whereas reductive quenching by the iron complexes is not favored. Even so, the article cautions that quenching efficiency alone should not be equated directly with catalytic efficiency.

Electrochemical studies are central to the mechanistic interpretation. All four iron complexes show two reduction waves in acetonitrile. The article considers whether these are metal-centered or ligand-centered reductions and concludes, based on DFT orbital analysis and electrochemical trends, that the reductions are predominantly ligand-centered. The lowest semi-unoccupied orbitals have limited iron character and are mainly distributed over the ligand framework.

Additional cyclic voltammetry under a restricted potential window showed that the first one-electron reduction is accessible within the reducing power available from the photosensitizer system. Under CO₂, the voltammetric responses differ from those under Ar, and the article interprets this as evidence that the one-electron-reduced iron species react with CO₂ to form intermediates that are not simply re-oxidized on the return scan. Thus, the data support the proposal that singly reduced iron complexes are catalytically relevant states for CO₂ activation.

The article does not directly observe all catalytic intermediates, but it proposes a chemically consistent pathway. After photoexcitation of [Ru(bpy)₃]²⁺, BIH reductively quenches the excited state. The reduced photosensitizer can then transfer an electron to the iron complex, generating reduced catalyst states. The electrochemical and computational data support these as ligand-centered reduced species rather than formal low-valent iron states.

CO₂ activation is proposed to occur at the reduced iron complex. The article specifically discusses the importance of proton-coupled electron transfer and the stabilizing role of water. It states that water can stabilize metal-CO₂ intermediates through hydrogen bonding and also act as a proton donor. In the mechanistic discussion, Fe-CO₂ adducts and Fe-COOH-type intermediates are invoked conceptually, but these are proposed intermediates rather than directly isolated species.

The role of water is strongly supported experimentally. In anhydrous MeCN, the system produced very little CO. Addition of 7.5% water caused a large increase in both CO production and selectivity, and 10% water gave similarly strong performance. Higher water fractions decreased activity. The article interprets this as a balance between beneficial proton delivery and intermediate stabilization on one hand, and unfavorable BIH solubility and quenching behavior on the other.

Several control experiments support the overall catalytic assignment. In the absence of light, catalyst, photosensitizer, sacrificial donor, or CO₂, no significant catalytic CO production occurred. Use of Fe(ClO₄)₂ instead of the molecular iron complex gave only minor CO and H₂ formation, showing that free Fe²⁺ is not responsible for the reported activity. Under Ar instead of CO₂, CO production was not significant.

The homogeneous nature of the catalysis was examined by mercury poisoning. The addition of a large excess of mercury did not suppress the observed CO production under the tested conditions, which the article takes as evidence against catalysis by iron nanoparticles or colloidal metallic species. The authors therefore assign the active system as homogeneous.

Catalyst robustness was examined by time-course experiments and photosensitizer replenishment. Catalytic activity increased strongly at early times and then plateaued. UV-Vis monitoring of the reaction mixture showed hypochromism attributed to photosensitizer deterioration. When additional [Ru(bpy)₃]²⁺ was added after 24 h, CO production resumed, supporting the conclusion that photosensitizer degradation is the main source of deactivation, while the iron catalyst remains comparatively robust under the reported conditions.

Catalyst

The catalysts are a series of molecular iron(II) bis(pyrazolyl)phenanthroline complexes, labeled Fe1, Fe2, Fe3, and Fe4. They are homogeneous mononuclear iron complexes containing tetradentate bis(pyrazolyl)phenanthroline ligands and two coordinated water ligands, formulated as [Fe(bpzRphen)(H₂O)₂]X₂ with different pyrazolyl substituents.

All four complexes were reported to have high-spin quintet ground states and distorted octahedral geometries. Magnetic susceptibility measurements supported high-spin Fe(II), and DFT calculations agreed with a quintet ground state for the series. The coordinated ligand framework controls the redox behavior, and the article concludes that the accessible reductions are predominantly ligand-centered.

The four ligand variants were designed to compare different substituent effects: unsubstituted pyrazolyl, dimethyl-substituted pyrazolyl, diphenyl-substituted pyrazolyl, and trifluoromethyl-substituted pyrazolyl. These modifications changed both activity and selectivity. Fe2, the dimethyl-substituted complex, gave the highest overall activity under the standard comparison conditions and also the highest reported TON_CO at low catalyst loading. Fe4, the trifluoromethyl-substituted complex, gave the highest CO selectivity in the standard series comparison.

The article further notes a structural feature for Fe4: weak intramolecular hydrogen-bonding interactions between coordinated water ligands and fluorine atoms of the CF₃ substituents. The authors suggest that such contacts may help proton management around the metal center and may contribute to the enhanced CO selectivity of Fe4. This is presented as an interpretation based on calculated structure rather than direct proof of the catalytic origin of selectivity.

The catalysts are described as robust under the reported photocatalytic conditions, but not perfectly immutable. The study indicates that deactivation mainly arises from photosensitizer deterioration rather than dominant catalyst decomposition. Mercury poisoning tests support a homogeneous molecular catalytic regime rather than conversion into heterogeneous iron particles.

Photosensitizer

The photosensitizer is [Ru(bpy)₃]²⁺. It serves as the visible-light absorber and initiates the photoredox sequence required for electron delivery to the iron catalyst. The experiments used this ruthenium complex at 0.3 mM in the photocatalytic mixtures.

The article discusses the excited state as the metal-to-ligand charge-transfer triplet state of the ruthenium complex. Stern-Volmer emission quenching experiments show that BIH efficiently quenches this excited state, with a quenching rate constant on the order of 10¹⁰ M⁻¹ s⁻¹. This supports a dominant reductive quenching pathway by BIH.

The iron catalysts also quench the excited state of [Ru(bpy)₃]²⁺, but much less efficiently than BIH. The article therefore treats BIH as the main quencher in the working catalytic system. DFT-based energy-level alignment is used to rationalize the relative feasibility of reductive and oxidative quenching pathways.

A key limitation of the photosensitizer in this system is durability. UV-Vis monitoring during irradiation showed loss of the characteristic ruthenium absorption, interpreted as photosensitizer degradation. Additional support comes from the restoration of CO production after replenishment of [Ru(bpy)₃]²⁺. Thus, the photosensitizer is chemically suitable for visible-light activation but is also identified as the main weak point in long-duration catalysis.

Investigation

	cat	cat conc [µM]	PS	PS conc [mM]	e-D	e-D conc [M]	.	.	solvent A	.	.	.	.	.	.	λexc [nm]	.	TON CO	.	.	TON CH4	TON H2	.	TON HCOOH	.	.	.	.
1.	Fe1	50	［Ru(bpy)3］2+	0.3	BIH	0.11			MeCN							462		987			not reported	187		not reported					"not reported" is not a number.
2.	Fe2	50	［Ru(bpy)3］2+	0.3	BIH	0.11			MeCN							462		1318			not reported	243		not reported					"not reported" is not a number.
3.	Fe3	50	［Ru(bpy)3］2+	0.3	BIH	0.11			MeCN							462		847			not reported	205		not reported					"not reported" is not a number.
4.	Fe4	50	［Ru(bpy)3］2+	0.3	BIH	0.11			MeCN							462		1265			not reported	133		not reported					"not reported" is not a number.
5.	Fe2	50	［Ru(bpy)3］2+	0.3	BIH	0.11			MeCN							462		311			not reported	51		not reported					"not reported" is not a number.
6.	Fe2	50	［Ru(bpy)3］2+	0.3	BIH	0.11			MeCN							462		1578			not reported	296		not reported					"not reported" is not a number.
7.	Fe2	50	［Ru(bpy)3］2+	0.3	BIH	0.11			MeCN							462		1593			not reported	300		not reported					"not reported" is not a number.
8.	Fe2	50	［Ru(bpy)3］2+	0.3	BIH	0.11			MeCN							462		2			not reported	22		not reported					"not reported" is not a number.
9.	Fe2	50	［Ru(bpy)3］2+	0.3	BIH	0.11			MeCN							462		1352			not reported	285		not reported					"not reported" is not a number.
10.	Fe2	50	［Ru(bpy)3］2+	0.3	BIH	0.11			MeCN							462		661			not reported	191		not reported					"not reported" is not a number.
11.	Fe2	50	［Ru(bpy)3］2+	0.3	BIH	0.11			MeCN							462		621			not reported	77		not reported					"not reported" is not a number.
12.	Fe2	25	［Ru(bpy)3］2+	0.3	BIH	0.11			MeCN							462		2086			not reported	123		not reported					"not reported" is not a number.
13.	Fe2	12.5	［Ru(bpy)3］2+	0.3	BIH	0.11			MeCN							462		4259			not reported	276		not reported					"not reported" is not a number.
14.	Fe2	6.25	［Ru(bpy)3］2+	0.3	BIH	0.11			MeCN							462		10168			not reported	862		not reported					"not reported" is not a number.
15.	Fe2	3.12	［Ru(bpy)3］2+	0.3	BIH	0.11			MeCN							462		23138			not reported	2177		not reported					"not reported" is not a number.
16.	Fe1	3.12	［Ru(bpy)3］2+	0.3	BIH	0.11			MeCN							462		not reported			not reported	not reported		not reported					"not reported" is not a number.
17.	Fe2	3.12	［Ru(bpy)3］2+	0.3	BIH	0.11			MeCN							462		not reported			not reported	not reported		not reported					"not reported" is not a number.
18.	Fe3	3.12	［Ru(bpy)3］2+	0.3	BIH	0.11			MeCN							462		not reported			not reported	not reported		not reported					"not reported" is not a number.
19.	Fe4	3.12	［Ru(bpy)3］2+	0.3	BIH	0.11			MeCN							462		not reported			not reported	not reported		not reported					"not reported" is not a number.

Investigation-Name: inv0ExportRefresh