Layered Double Hydroxide-Stabilized AgI Photocatalyst for Efficient Arene Arylation and Alkylation

Abstract

Developing clean and efficient methods for carbon–carbon bond formation remains a central challenge in sustainable chemistry. Here, we report a unified photocatalytic strategy for direct arene arylation and alkylation under visible light. By integrating silver iodide (AgI) nanoparticles with magnesium–aluminum-layered double hydroxide (Mg₂Al₁-LDH), we achieve efficient cross-coupling of arenes with both aryl and unactivated alkyl iodides under mild conditions and natural sunlight. This system operates without external bases or photoredox cocatalysts and tolerates base-sensitive functional groups, producing only water and iodine as byproducts. Mechanistic studies reveal a synergistic process in which photogenerated electrons reduce halides via an inner-sphere transfer mechanism, while photogenerated holes oxidize surface hydroxyl groups to generate ·OH radicals. These radicals enable hydrogen abstraction and C–H bond activation, completing the coupling process. The availability of surface hydroxyls and oxygen vacancies, along with favorable band alignment between AgI and Mg₂Al₁-LDH, is critical to catalyst performance. This work demonstrates the first example of a single catalyst system enabling both arylation and alkylation through a unified radical mechanism, offering a sustainable platform for light-driven C–H functionalization and expanding the frontiers of green synthetic chemistry.

1. Introduction

Arene arylation and alkylation are cornerstone reactions in synthetic chemistry, converting ubiquitous C–H bonds into high-value biaryls and alkylbenzenes without prefunctionalization. , These motifs are prevalent in pharmaceuticals, agrochemicals, and advanced polymers. Yet, current methods remain bifurcated: arylation typically relies on transition-metal-catalyzed cross-coupling (e.g., Suzuki, Negishi), while alkylation proceeds via Friedel–Crafts-type pathways with Brønsted- or Lewis-acid catalysts. ,

Radical-based cross-coupling offers a compelling alternative, operating under milder conditions and displaying greater functional group tolerance. However, most radical protocols require separate catalyst systems, a photoredox cocatalyst such as Ir­(ppy)₃ or Ru­(bpy)₃, , electrochemical setups, preactivated precursors, − or strong reductants. , These systems often require strong bases or acids, specialized ligands, and toxic or hard-to-obtain additives. Catalyst recovery is cumbersome, and waste streams are significant. Few platforms can unify arylation and alkylation under a single mechanism due to intrinsic differences in halide reactivity and radical generation pathways.

The major bottleneck lies in initiating the radical step. − Most reported protocols use outer-sphere electron transfer (OSET), which demands a close redox match between cocatalyst and substrate and therefore leans on strong electron donors such as potassium tert-butoxide (KOt-Bu). Weaker bases like KOH and K₃PO₄ result in minimal or no formation of the desired products, , thereby limiting the substrate scope and posing a risk to sensitive functional groups. In contrast, inner-sphere electron transfer (ISET) provides a bond-mediated pathway that is less sensitive to redox mismatch , and more amenable to weakly activated halides. Photocatalytic semiconductors that bind the halide substrate strongly therefore favor ISET and offer a unified mechanistic platform for both arylation and alkylation.

We hypothesized that a photocatalyst capable of ISET and ·OH radical generation could offer a unified mechanism for both arylation and alkylation. Building on this idea, we developed a hybrid composite of AgI nanoparticles (NPs) dispersed on MgAl-LDH (AgI/MgAl-LDH; Scheme a), which also exists as the natural clay hydrotalcite. The inherent basicity of carbonate anions on LDH surfaces is conducive to the coupling reactions. Upon visible-light activation, AgI generates electron–hole pairs. − Electrons cleave C–I bonds in aryl or alkyl iodides through ISET, yielding carbon-centered radicals that couple directly with arenes. Concurrently, holes oxidize LDH surface hydroxyls to ·OH radicals, which facilitate C–H activation and hydrogen abstraction from σ-adduct intermediates. The system maintains catalytic activity even at a low light intensity of 0.01 W cm^–2, achieving an apparent quantum yield (AQY) of 3.3%the highest reported value for AgI-based photocatalysts in organic reactions (Scheme b). This design offers an efficient method for cross-coupling under solar irradiation, overcoming long-standing mechanistic limitations.

1. (a) Direct Arylation/Alkylation of Inactive Arenes with Halides Promoted by AgI/Mg₂Al₁LDH Illuminated by Visible Light. (b) Comparison of Apparent Quantum Yields (AQYs) and Operating Light Intensities for AgI-Based Photocatalysts in Organic Reactions .

^a The structures were visualized using CIF data retrieved from the Crystallography Open Database (COD entries: 1011025 and 9012627).

^b References: a, b, c, d, e, f, g, h, i, j, k, and l.

2. Results and Discussion

2.1. Photocatalyst Structure and Characterization

The AgI/Mg₂Al₁LDH composite, containing 9.4 wt % silver (Supporting Information Table S1, entry 1), was employed as the photocatalyst to drive the arylation of benzene. After the reaction, the solution turned a deep red color (Figure a). Mass spectrometry and UV–vis spectroscopy confirmed the formation of biphenyl and I₂, respectively (Supporting Information Figures S1 and S2).

1.

Photograph of the photoreaction system and characterization of the AgI/Mg₂Al₁LDH composite. (a) Photographs of the composite, the reactant, and the product. (b) XRD patterns of the AgI/Mg₂Al₁LDH composite. (c, d) HAADF-STEM image and corresponding elemental mapping, where white islands correspond to AgI particles. (e, f) HRTEM images showing the Mg₂Al₁LDH and AgI lattice features, respectively. (g) TEM image with particle size distribution of AgI. (h) HRTEM image of the composite interface and corresponding EDS elemental mapping.

X-ray diffraction (XRD) confirmed the presence of crystalline AgI and the characteristic LDH phase (Figure b). High-angle annular dark-field scanning transmission electron microscope (HAADF-STEM) and energy-dispersive X-ray spectroscopy (EDX) elemental mapping revealed AgI NPs dispersed on the Mg₂Al₁LDH support (Figures c, d). High-resolution transmission electron microscopy (HR-TEM) resolved lattice fringes of both AgI and LDH (Figures e, f), consistent with the XRD results. The X-ray photoelectron spectroscopy (XPS) signal corresponding to surface OH groups of Mg₂Al₁LDH decreases significantly after AgI loading, suggesting that the deposited AgI covers the surface hydroxyl groups (Supporting Information Figure S3). Moreover, the (003) reflection of LDH remains unchanged upon AgI loading (Supporting Information Figure S4). Given that intercalation of species into LDH materials typically induces a shift in the (003) diffraction peak, this observation confirms that AgI nanoparticles are predominantly anchored on the external surface of the LDH support.

Further insight into the dispersion of AgI on the LDH surface was obtained. As shown in Figure g, AgI aggregates are present on the support surface, with a mean particle size of 35 nm. HRTEM imaging further revealed a well-defined interface between AgI and the Mg₂Al₁LDH support (Figure h). The lattice spacings of AgI(110) and MgAl-LDH(015) were measured to be 0.230 and 0.225 nm, respectively. The close lattice matching at this interface likely facilitates the efficient migration of photogenerated electrons and holes across the junction.

2.2. Benchmark Arylation of Benzene

Direct arylation of benzene with iodobenzene was selected to optimize the AgI/MgAl-LDH composites and reaction conditions. The yield of biphenyl varied moderately with the AgI loading and the Mg/Al ratio of the composites (Supporting Information Tables S1 and S2). Among the tested materials, 10AgI/Mg₂Al₁LDH (hereafter simplified as AgI/Mg₂Al₁LDH delivered the highest yield of biphenyl without the need for any external base or additive. Under 400 nm LED irradiation (0.1 W cm^–2), the composite afforded a 93% yield of biphenyl (Table , entry 1).

1. Screening of Reaction Conditions for Direct Arylation of Benzene .

Entry	Variation from the standard condition	Yield (%)
1	None	93
2	Without irradiation	0.1
3	Without AgI/Mg2Al1LDH	1.2
4	Without iodobenzene	0
5	Hexane instead of benzene	1.5
6	Mg2Al1LDH	2.3
7	AgI/ZrO2	10.5
8	AgBr/Mg2Al1LDH	14.6
9	AgI/ZrO2, with Na2CO3 added	11.1
10	Under an air atmosphere	56.4
11	AgI/Mg2Al1LDH-PO4 3–	60
12	AgI/Mg2Al1LDH-Cl–	90
13	Anatase TiO2 (100 nm)	0
14	Rutile TiO2 (100 nm)	0
15	CdS	0

^aStandard reaction conditions: iodobenzene (0.1 mmol), benzene (1 mL), AgI/Mg₂Al₁LDH (50 mg), N₂ atmosphere, irradiated under a 400 nm wavelength LED light with an intensity of 0.1 W cm^–2, and the reaction was conducted at 60 °C for 20 h. Mg₂Al₁LDH-PO₄ ^3– and Mg₂Al₁LDH-Cl^– were prepared using Na₃PO₄ and NaCl, respectively, following a previously reported method. Yields were determined by GC analysis.

A comparison of catalyst performance and reaction conditions for benzene arylation with iodobenzene between this work and previously reported systems is provided in Supporting Information Table S3. Notably, most reported systems require the addition of strong bases or rely on harsh reaction conditionssuch as high temperature or prolonged reaction timesto achieve comparable performance, underscoring the advantages of the present approach.

Control experiments emphasized the critical role of each reaction component. The absence of light, the composite, iodobenzene, or benzene suppressed conversion to negligible levels (Table , entries 2–5), confirming a photodriven cross-coupling pathway. Neither AgI/ZrO₂ nor Mg₂Al₁LDH alone exhibited significant activity (Table , entries 6 and 7), highlighting the essential synergy between AgI and the LDH support. Replacement of AgI with AgBr or addition of Na₂CO₃ to the AgI/ZrO₂ yielded inferior results (Table , entries 8 and 9), suggesting that interlayer CO₃ ^2– ions play a minor role in the cross-coupling process. To further elucidate the role of CO₃ ^2– species in the LDH support, we prepared two additional LDHs with different intercalated anions, namely PO₄ ^3– and Cl^–. As summarized in Table (Entries 11 and 12), when LDH-PO₄ ^3– was used as the support to load the same amount of AgI, the biphenyl yield decreased significantly. Given that PO₄ ^3– is a stronger base than CO₃ ^2–, this decrease in yield suggests that basicity alone does not account for the high activity. This inverse correlation between basicity and catalytic efficiency rules out a dominant basic-site-assisted mechanism. In contrast, replacement of CO₃ ^2– with Cl^– did not lead to a significant decrease in yield, further supporting the conclusion that the basicity of CO₃ ^2– plays a minor role in the catalytic performance of the composite.

We also evaluated the performance of TiO₂ and CdS under identical reaction conditions. As shown in Table (Entries 13–15), no biphenyl product was detected with these catalysts under the standard conditions. Notably, the AgI/Mg₂Al₁-LDH composite retained partial activity (56%) in air, demonstrating tolerance to moisture and oxygen (Table , entry 10).

2.3. Solar-Driven Photocatalysis and Optimization

The catalyst maintained high performance under natural sunlight, achieving a 60% biphenyl yield without additional thermal input (Figure a), underscoring its practical applicability in solar-driven synthesis. The high solar energy utilization efficiency represents an important advantage for producing fine chemicals on a large scale.

2.

(a) Experiment conducted under sunlight. (b) Time-course study of biphenyl production. (c) Effect of AgI/Mg₂Al₁LDH composite mass on biphenyl yield.

Time-course analysis revealed that 80% yield was reached within 10 h, progressing to 90% after 20 h (Figure b). The biphenyl yield increased slowly with catalyst loading, likely due to light screening limiting further enhancement (Figure c). This is consistent with the light-screening effect that limits charge generation at high loadings.

2.4. Substrate Scope: Arylation and Alkylation

The system accommodated a broad array of aryl iodides, including electron-rich, electron-deficient and heteroaryl substrates (Scheme ), producing biaryls in 87–99% yield. Notably, base-sensitive groups and reactive functionalities were well tolerated (e.g., compounds 17–20, Scheme ), illustrating the benefits of the mild reaction conditions. Compound 17, in particular, represents a promising intermediate for optoelectronic applications, including organic light-emitting diodes (OLEDs).

2. Substrate Scope .

^e Reaction conditions: aryl iodide or alkyl iodide (0.1 mmol), arene (1 mL), AgI/Mg₂Al₁LDH (50 mg), N₂ atmosphere, irradiated with a 400 nm wavelength LED light (0.1 W cm^–2). Reaction temperature was 60 °C for 20 h. Values in parentheses are the yields under identical conditions in the dark. Yields were determined by GC analysis. ^a80 °C, ^b48 h, ^c85 °C, 48 h, 0.2 W cm^–2, ^d0.2 W cm^–2.

The system also demonstrated reactivity trends consistent with a single-electron transfer (SET) pathway: aryl iodides bearing electron-withdrawing groups (e.g., 4-iodobenzonitrile) gave higher yields (99%) than electron-donating analogues (e.g., 4-iodoanisole, 91%). Control reactions performed in the dark yielded negligible products, confirming the photocatalytic nature of the transformation.

Heteroaryl iodides such as 3-iodothiophene and 3-iodopyridine afforded 94% and 91% yield, respectively (compounds 6 and 7, Scheme ). However, relatively harsh conditions (85 °C, 48 h) were required for these substrates. To address this limitation, we investigated the effect of LDH composition by evaluating 10AgI/Ni₂Al₁LDH for the conversion of 3-iodopyridine. Under milder conditions (60 °C, 20 h), 10AgI/Ni₂Al₁LDH afforded only 1% yield, significantly lower than that obtained with 10AgI/Mg₂Al₁LDH (21%). We attribute this decrease to catalyst poisoning, as heteroaryl iodides containing nitrogen or sulfur atoms may strongly coordinate with transition metals within the LDH support, thereby deactivating the catalytic system.

The method was extended to arenes beyond benzene. Electron-rich and electron-deficient arenes reacted smoothly, with alkyl-substituted arenes yielding 54–83%. Toluene produced a mixture of o-, m-, and p-methylbiphenyl in a 62:24:14 ratio (compound 8, Scheme ), a distribution typical for radical aromatic substitution. , No side products involving benzylic activation were observed.

Strikingly, the same photocatalytic system enabled the coupling of benzene with a variety of alkyl iodides, an achievement rarely reported in unified systems. Despite the stronger C–I bonds and weaker adsorption on solids, heterocyclic and even sterically hindered alkyl iodides yielded the desired products in moderate to excellent yields (compounds 21–25, Scheme ). The C­(sp ²)–C­(sp ³) bonds were formed under mild conditions with high selectivity, without requiring bases or cocatalysts. The particularly challenging tert-butyl 4-iodopiperidine-1-carboxylate substrate (E_red < −2.0 V vs SCE) afforded the product in 64% yield (compound 24, Scheme ), underscoring the system’s activation efficiency. No reaction occurred in the dark, confirming the necessity of light. The successful synthesis of compounds 21–25 in good yields highlights the promise of a more environmentally friendly approach to producing therapeutic agents, such as kinase inhibitors and anticancer drugs. The ability to forge C­(sp ²)–C­(sp ³) bonds under mild, base-free conditions with excellent selectivity underlines the method’s potential for sustainable synthesis.

2.5. Impact of Irradiation Properties and Reaction Kinetics

The yield increased linearly with light intensity. Even at a low intensity of 0.01 W cm^–2, the reaction reached a 40% yield (Figure a), corresponding to an apparent quantum yield (AQY) of 3.3%, which is competitive with leading photocatalytic systems in organic reactions. The action spectrum (Figure b) revealed that catalytic activity correlated with AgI’s absorption profile, with peak quantum yields observed, while minimal activity was obtained at wavelengths ≥ 550 nm. The action spectrum confirmed that the nanoparticles, rather than the LDH, harvest the incident light.

3.

Light intensity/wavelength-dependent performance of cross-coupling with AgI/Mg₂Al₁LDH composite and the kinetic study. (a) The dependence of the performance on the light intensity. The colored diamond-shaped symbols associated with biphenyl yield were obtained from three independent tests; the solid lines represent the averaged data, and the red dashed line is a guide for the eye. (b) UV–vis spectra of AgI/Mg₂Al₁LDH and Mg₂Al₁LDH, and dependence of the AQY for direct arylation of benzene with iodobenzene on the light wavelength. The light intensity was 0.1 W cm^–2 for all wavelengths. Error bars associated with AQY are the standard error of three sets of unique measurements. Ultraviolet (UV) and light-emitting diode (LED) lamps with five peak wavelengths (350 ± 5 nm, 400 ± 5 nm, 450 ± 5 nm, 550 ± 5 nm, 660 ± 5 nm) were used to drive the reaction. (c) The comparison of the calculated activation energy with that in a previous reference.

Kinetic analysis demonstrated first-order dependence on substrate concentration for the model reaction between iodobenzene and benzene over the AgI/Mg₂Al₁LDH composite under light irradiation (Supporting Information Figure S5). The Arrhenius plot (Figure c) yielded an apparent activation energy of 40.6 kJ mol^–1, markedly lower than that of the thermal analogue (61.7 kJ mol^–1; raw data presented in Supporting Information Figure S6). These findings suggest that photogenerated carriers effectively lower energy barriers and open new mechanistic channels inaccessible under thermal conditions.

2.6. Role of ·OH Radicals from Surface Hydroxyls

X-ray photoelectron spectroscopy (XPS) analysis of the AgI/Mg₂Al₁LDH composite before and after reaction revealed a substantial decrease in the surface metal–hydroxyl (M–OH) signal at 532.7 eV, concomitant with the emergence of oxygen vacancies (O_v), whereas metal oxide (M–O, 530.2 eV) and carbonate (CO₃ ^2–, 531.4 eV) signals remained largely unaffected (Figures a,b). − This transformation was accompanied by the appearance of an electron paramagnetic resonance (EPR) signal (g = 2.004) characteristic of O_v (Figure c), implying that surface hydroxyl groups are photo-oxidized to generate ·OH radicals with concomitant vacancy formation.

4.

Identification of photogenerated hydroxyl radicals in AgI/Mg _x Al _y LDH solid for direct arylation. (a) and (b) O 1s spectra of AgI/Mg₂Al₁LDH before and after the arylation reaction. (c) EPR spectra of AgI/Mg₂Al₁LDH before and after the arylation reaction. (d) Fluorescence spectra of different reaction systems: AgI/Mg₂Al₁LDH+terephthalic acid (H₂BDC)+irradiation, AgI/ZrO₂+H₂BDC+irradiation, Mg₂Al₁LDH+H₂BDC+irradiation, H₂BDC+irradiation, AgI/Mg₂Al₁LDH+H₂BDC+heating, AgI/Mg₂Al₁LDH+irradiation. (e) The recycling performance of AgI/Mg₂Al₁LDH solid. The regeneration method can be found in the Experimental section. (f) The O 1s XPS spectrum of the AgI/Mg₂Al₁LDH after regeneration. (g) Zeta potential values of the as-prepared, used, and regenerated AgI/Mg₂Al₁LDH.

To establish the correlation between oxygen vacancies and catalytic activity, we examined three low-activity compositesAgI/Al­(OH)₃, AgI/α-Al₂O₃, and AgI/In₂O₃alongside high-activity AgI/Mg₂Al₁LDH (yields are provided in Figure b). The low-activity composites exhibited no detectable O_v signals, whereas AgI/Mg₂Al₁LDH displayed a pronounced O_v signal postreaction (Supporting Information Figure S7). Quantitative analysis revealed a close correlation between catalytic performance and O_v concentration: as the biphenyl yield increased from 51% to 84% (1.64-fold), the spin concentration rose from 3.64 × 10¹² to 5.71 × 10¹² spins/mm³ (1.57-fold). This close correlation between catalytic activity and oxygen vacancy concentration provides strong evidence for the critical role of surface oxygen vacancies in the photocatalytic process.

5.

(a) Additive effect and (b) influence of composite variation on biphenyl yield in the direct arylation of benzene, with representative photographs of the composites provided. (c) Photoluminescence (PL) decays of AgI/Mg₂Al₁LDH composite with and without iodobenzene. (excitation wavelength: 398 nm). The inset table shows exponential fitting parameters for the 468 nm emission peak. High-resolution XPS spectra of Ag 3d (d) and I 3d (e) for the AgI/Mg₂Al₁LDH composite mixed with iodobenzene, recorded prior to and under irradiation. The schematic of the in situ XPS experimental setup is presented in Supporting Information Figure S15. (f) Energy diagram depicting photoinduced charge transfer. Data supporting these energy levels can be found within the main text or were derived from reported values in references. − The structures were visualized using CIF data retrieved from the Crystallography Open Database (COD entries: 1011025 and 9012627).

Surface-generated ·OH radicals are known to be effective oxidizing agents, capable of abstracting hydrogen atoms, a step essential for C–H arylation. Importantly, Mg₂Al₁LDH serves as an internal reservoir of hydroxyl groups. Among all tested LDH-based composites, Mg₂Al₁LDH exhibited the largest crystallite size (Supporting Information Figure S8), which correlates with a higher density of surface hydroxyls. , Although factors such as AgI content (Supporting Information Table S1) and specific surface area (Supporting Information Figure S9) also affect catalytic performance, hydroxyl group availability emerges as the predominant factor influencing biphenyl yields.

The production of ·OH radicals was further investigated by fluorescence spectroscopy using terephthalic acid (H₂BDC) as a fluorescent probe, which exhibits emission at 425 nm upon reaction with ·OH (Figure d). , No detectable fluorescence signal was observed in the absence of AgI NPs, Mg₂Al₁LDH, H₂BDC, or light irradiation, indicating that all these elements are essential for ·OH generation. EPR spectroscopy further confirmed the generation of the DMPO–OH· spin-adduct under irradiation (Supporting Information Figure S10). Based on these experiments, we performed quantitative EPR analysis of ·OH radicals across three different photocatalytic systems. Characteristic DMPO–OH signals were observed in all cases (Supporting Information Figure S11a), and integration of these signals (Supporting Information Figure S11b) revealed a trend that closely mirrored the corresponding biphenyl yields. This strong correlation provides compelling evidence for the involvement of the ·OH pathway in the catalytic cycle. Consistent with this conclusion, the introduction of DMPO as a radical scavenger sharply reduced the yield to 2.2% (Figure a). Furthermore, replacing Mg₂Al₁LDH with its calcined form, i.e., AgI/Mg₂Al₁LDO (metal oxide), led to a decrease in yield to 22%, highlighting the essential role of ·OH radicals in the reaction.

Repeated use of the catalyst resulted in declining yields, which was attributed to the depletion of surface hydroxyl groups. Notably, catalytic activity was fully restored upon simple regeneration using aqueous Na₂CO₃ (pH 11.2) at 85 °C, with XPS and zeta potential analyses confirming the recovery of surface hydroxyl functionality (Figures f and g). To further evaluate the structural robustness and practical sustainability of the catalyst, XRD, TEM, and ICP analyses were performed after five catalytic cycles. As shown in Supporting Information Figure S12, the characteristic diffraction peaks of Mg₂Al₁LDH exhibited decreased intensity after five cycles, whereas those of AgI remained essentially unchanged. Importantly, the peak intensity of the regenerated Mg₂Al₁LDH was largely restored to that of the fresh catalyst. Supporting Information Figure S13 presents the TEM images of AgI/Mg₂Al₁LDH for the fresh sample, the sample after five cycles, and the regenerated sample. Particle size distribution analysis revealed only a marginal increase in mean particle size from 35 to 42 nm after five catalytic cycles and subsequent regeneration. Additionally, ICP analysis (Supporting Information Table S1) showed that the Ag content in the composite decreased from 9.4% to 8.9% after five cycles and subsequent regeneration. Collectively, these results corroborate the exceptional stability of the composite under both reaction and post-treatment conditions.

Previous studies demonstrated UV-driven ·OH generation by LDH-based catalysts for pollutant degradation. − In contrast, this work provides the first experimental evidence for visible light-induced ·OH radical formation by AgI/MgAl-LDH to enable direct C–H arylation. This represents a significant environmental and mechanistic advance, leveraging benign visible light instead of energy-intensive UV sources.

2.7. Charge Transfer and Mechanistic Insights into ISET

Radical scavenger experiments clarified the contributions of photogenerated charge carriers (Figure a). Potassium iodide (KI), a hole scavenger, reduced the biphenyl yield significantly (from 80% to 40%), while manganese­(III) acetate, an electron scavenger, caused a more moderate decrease. This indicates that photogenerated holes play a dominant role in the reaction mechanism, prompting further investigation into the structural origins of this carrier-dominated reactivity. As shown in Figure b, replacing AgI with metallic Ag, or employing non-LDH supports such as α-Al₂O₃, In₂O₃, Mg­(OH)₂, and Al­(OH)₃, led to a dramatic reduction in biphenyl yield. In contrast, AgI supported on Zn₂Al₁LDH or Ni₂Al₁LDH maintained high catalytic activity. These results highlight the importance of synergistic interactions between AgI and the LDH support.

Photoluminescence (PL) lifetime measurements provided additional mechanistic insight. The introduction of iodobenzene prolonged the excited-state lifetime of AgI/Mg₂Al₁LDH (Figure c), suggesting suppressed electron–hole recombination via efficient electron transfer from photoexcited AgI to the aryl iodide substrate. Extending this analysis, we observed that mixing AgI/Mg₂Al₁LDH with iodobenzene or representative alkyl iodides (N-Boc-4-iodopiperidine and iodocyclopentane) induced a slight shift in the visible absorption spectrum relative to the pristine composite (Supporting Information Figure S14). This spectral change signifies the formation of a complex between the composite and the iodide substrates, analogous to previously reported observations supporting an ISET mechanism.

XPS analysis provided evidence for the aforementioned electron transfer between the AgI/Mg₂Al₁LDH composite and the substrate. Upon mixing with iodobenzene, the Ag 3d binding energy of the catalyst shifted to lower values (Supporting Information Figure S16), indicating partial oxidation of the Ag species. Concurrently, the I 3d binding energy of the substrate shifted toward lower binding energy (Supporting Information Figure S17), suggesting electron acceptance. These shifts reflect an intrinsic propensity for electron transfer from AgI to the aryl iodide. To further probe the electron transfer process under irradiation directly, we performed in situ XPS measurements. As shown in Supporting Information Figure S15, no binding energy shifts were observed for the Ag/Mg₂Al₁LDH composite upon irradiation (390–780 nm), confirming its photostability. In contrast, upon mixing the catalyst with iodobenzene under irradiation (Figure d), the Ag 3d peaks shifted to lower binding energy, accompanied by a corresponding shift of the I 3d peaks of iodobenzene. These observations indicate photoinduced electron transfer from AgI to iodobenzene, thereby providing direct evidence for the proposed ISET mechanism.

The feasibility of the ISET pathway was further corroborated by examining the energetic alignment between the composite and reactants. The LUMO of iodobenzene lies significantly above the conduction band of AgI (Supporting Information Figure S18), making conventional OSET energetically unfavorable. In contrast, the strong interaction between AgI and the substrate in ISET enables direct electron transfer despite this redox mismatch. Energy band alignment between the LDH support and AgI also proved critical for the ISET process. In Ag/Mg₂Al₁LDH, localized surface plasmon resonance (LSPR) excitation generates energetic electrons, but poor alignment between Ag and the LDH valence band hinders charge transfer. In contrast, AgI/Mg₂Al₁LDH features favorable valence band matching that enhances both charge separation and carrier migration (Figure f). This observation explains the dramatic drop in yield upon replacing AgI with AgBr (from 93% to 14.6%, Figure b), as AgBr possesses less favorable band positions and lower photostability, often degrading into metallic silver (see discussions in Supporting Information Figures S19–S22).

The importance of optimal band alignment was further validated through screening of alternative supports. Using Zn₂Al₁LDH and Ni₂Al₁LDH as supports retained relatively high photocatalytic activity, whereas non-LDH supports resulted in sharply diminished yields (Figure b). This trend correlates with valence band mismatches that inhibit hole transfer and consequently suppress hydroxyl radical (·OH) generation (see discussions in Section ). Collectively, these results demonstrate that achieving high photocatalytic performance relies on two synergistic factors: the ISET pathway for substrate activation and efficient hole transfer enabled by favorable valence band alignment between AgI and the Mg₂Al₁LDH support.

2.8. Reaction Mechanism

Time-dependent attenuated total reflectance Fourier-transform infrared (ATR-FTIR) spectroscopy under visible light irradiation was employed to monitor the reaction progress (Figure a). To avoid signal overlap between the LDH framework and the substrate, 4-chloroiodobenzene was selected as a model compound in place of iodobenzene, enabling clearer differentiation of infrared bands. During the reaction, a gradual decrease in the characteristic C–I stretching signal at 473 cm^–1 was observed, indicating progressive bond cleavage. In contrast, the M–O signal at 551 cm^–1 remained largely unchanged, while a slight red shift appeared near 651 cm^–1, consistent with lattice distortion induced by oxygen vacancy formation, corroborating the structural characterization results presented in Section . Control experiments confirmed that the observed transitions arose from the integrated photocatalytic system (Supporting Information Figure S23). The modest kinetic isotope effect (KIE = 1.36) observed in the reaction with benzene-d6 indicates that C–H bond activation is not the rate-determining step (Supporting Information Figure S24), supporting the hypothesis that radical generation precedes this stage. These findings, together with the roles of ·OH radicals and the ISET pathway discussed in Sections and , respectively, converge on a unified photocatalytic mechanism.

6.

Reaction mechanism study. (a) Time-dependent FTIR spectral changes of the mixture of aryl iodide and AgI/Mg₂Al₁LDH composite (the experimental setup is shown in Supporting Information Figure S23c. (b) Proposed reaction mechanism. The structures were visualized using CIF data retrieved from the Crystallography Open Database (COD entries: 1011025 and 9012627).

This mechanism integrates synergistic charge transfer and radical-mediated processes on AgI/Mg₂Al₁LDH (Figure b). Under visible light irradiation, AgI nanoparticles serve as photoactive centers, generating electron–hole pairs. Photogenerated electrons reduce aryl or alkyl iodides via an inner-sphere electron transfer (ISET) mechanism, forming reactive radical intermediates that couple with arenes (Steps I–II). Concomitantly, holes migrate to the Mg₂Al₁LDH support, where they oxidize surface hydroxyl groups to generate hydroxyl radicals (·OH) (Steps II–III), thereby suppressing electron–hole recombination. The resulting C–C bonded intermediates then undergo hydrogen abstraction by surface-generated ·OH radicals, yielding biphenyl derivatives and water. Water formation was confirmed by quantitative analysis and time-of-flight mass spectrometry (Supporting Information Table S5 and Figure S25). The photocatalytic cycle is further sustained by oxidation of iodide ions to I₂ within the oxygen vacancy-rich LDH environment. The LDH surface can be regenerated through simple alkaline treatment with aqueous Na₂CO₃, restoring the consumed hydroxyl groups (Steps III–I). Thermodynamic assessment (Supporting Information Table S6) indicates that the overall Gibbs free energy change (ΔG _r ⁰) for this photoexcited radical pathway is significantly more favorable than that for the corresponding thermal reaction, further highlighting the efficiency of this light-driven process (Scheme ).

The high selectivity for cross-coupling originates from the substantial arene-to-aryl iodide molar ratio (∼100) under standard conditions (0.1 M aryl iodide), which ensures that aryl radicals generated via C–I bond cleavage are rapidly intercepted by excess arene before undergoing self-reaction. This kinetic preference effectively suppresses homocoupling, as confirmed by the negligible byproduct formation under typical conditions. Elevating the aryl iodide concentration resulted in measurable homocoupling (Supporting Information Figure S26), corroborating the transient radical mechanism.

The preference for hydrogen abstraction from C–C bonded intermediates over direct ·OH addition to other chemicals in the reaction system was rationalized by density functional theory (DFT) calculations. The proposed pathwayC–C bond formation followed by hydrogen abstractionexhibits lower activation barriers (5.0 and 6.4 kcal·mol^–1) than competing ·OH reactions with benzene (7.9 kcal·mol^–1) or biphenyl (6.7 kcal·mol^–1) (Supporting Information Figure S27). Furthermore, the cross-coupling products exhibit significantly lower free energies than those generated via alternative pathways, indicating both kinetic and thermodynamic favorability. These results are consistent with the high selectivity observed experimentally.

3. Conclusions

This work introduces a unified, visible-light-driven strategy for direct C–H arylation and alkylation of arenes under mild conditions. By integrating AgI NPs with Mg₂Al₁LDH, we have developed a highly effective photocatalyst that operates efficiently under low-intensity visible light or natural sunlight, without requiring external bases, acids, transition-metal cocatalysts, or photoredox mediators. The system demonstrates broad functional group tolerance, including compatibility with base-sensitive moieties and inactive alkyl halides. The only byproducts, molecular iodine and water, are environmentally benign and easily separable, making the process inherently green and sustainable.

Mechanistic investigations reveal that the coupling reaction proceeds via a synergistic pathway involving photogenerated charge carriers and surface ·OH radicals, with electron transfer occurring through an inner-sphere mechanism. The formation of oxygen vacancies and the regeneration of hydroxyl groups via mild alkaline treatment further enhance catalyst recyclability and long-term performance.

This study represents the first demonstration of a single photocatalytic system enabling both arylation and alkylation through a common radical mechanism driven solely by visible light. It not only expands the scope of radical cross-coupling but also sets a precedent for clean, safe, and scalable C–H functionalization strategies powered by solar energy.

4. Experimental Section

4.1. Materials

All chemicals were purchased from commercial suppliers and used without further purification. The supplier and purity of the chemicals are indicated in parentheses as follows: iodobenzene (Adamas, 99%), 4-iodobenzene (Adamas, > 98%), 4-iodoanisole (Adamas, 99%), 1-chloro-4-iodobenzene (Adamas, > 98%), 4-iodobenzotrifluoride (Adamas, > 98%), 4-iodobenzonitrile (Adamas, > 98%), 3-iodothiophene (Adamas, > 98%), 3-iodopyridine (Adamas, > 98%), tert-butyl 4-phenylpiperidine-1-carboxylate (Adamas, > 97%), 4-iodotetrahydro-2H-pyran (Adamas, > 98%), 3-iodooxetane (Adamas, > 98%), iodocyclopentane (Adamas, 98%), iodocyclohexane (Adamas, 97%), 4-iodobenzoic acid methyl ester (Adamas, > 98%), 4-iodoaniline (Adamas, > 98%), 4-iodophenol (Adamas, > 98%), 4-iodobenzaldehyde (Adamas, > 98%), benzene (Greagent, ≥ 99.5%), benzene-d₆ (Adamas, 99.5%), biphenyl (Adamas, ≥ 99%), toluene (SCR, ≥ 99.5%), chlorobenzene (Greagent, ≥ 99.5%), diphenyl ether (Adamas, > 99%), acetophenone (Adamas, > 99%), 1,2-dichlorobenzene (Macklin, 99%), 1,3,5-trimethylbenzene (Macklin, 97%), m-xylene (Greagent, ≥ 99%), p-xylene (Greagent, ≥ 99%), ethylbenzene (Adamas, 99%), sodium carbonate (Greagent, ≥ 99.8%), methanol (Greagent, ≥ 99.5%), sodium hydroxide (Greagent, ≥ 98%), sodium iodide (Adamas, 99.99%), potassium iodide (Adamas, 99%), terephthalic acid (Adamas, 99%), silver­(I) nitrate (SCR, ≥ 99.8%), silver­(I) bromide (Adamas, 98%), aluminum­(III) nitrate nonahydrate (Greagent, ≥ 99.0%), manganese­(III) acetate dihydrate (Adamas, 98%), zinc­(II) nitrate hexahydrate (Keshi, ≥ 99.0%), nickel­(II) nitrate hexahydrate (Adamas, 99%), magnesium hydroxide (Adamas, 98.5–99%, D50:10 μm), α-aluminum­(III) oxide (Adamas, > 97.5%, D50:15 ± 2 μm), aluminum hydroxide (Aladdin, 99.9%, 2–10 μm), isopropanol (TCI, 99.5%), and N₂ (>99.999%).

4.2. Composite Preparation

Layered double hydroxides including Mg _x Al _y LDH with various x/y ratios, Ni₂Al₁LDH, and Zn₂Al₁LDH were synthesized using a coprecipitation method. A typical synthesis of Mg₂Al₁LDH involved dissolving 0.0133 mol of Mg­(NO₃)₂ 6H₂O and 0.0067 mol of Al­(NO₃)₃ 9H₂O in 20 mL of deionized water to form solution A. Separately, 0.02 mol of Na₂CO₃ was dissolved in 30 mL of deionized water to form solution B. These two solutions were simultaneously added dropwise into 15 mL of deionized water in a flask maintained at 40 °C using an injection pump at flow rates of 2 mL min^–1 for A and 3 mL min^–1 for B. After stirring for 1 h, 2 M NaOH aqueous solution was added dropwise until the pH reached 10. The resulting slurry was sealed and heated to 85 °C with stirring for 3 h, followed by aging in an oven at 80 °C for 16 h. The gel was then centrifuged at 4000 rpm and washed with deionized water until the filtrate reached neutral pH. The resulting solid was dried at 80 °C for 48 h and ground through a 40-mesh sieve, yielding approximately 1.4 g of white powder. To prepare LDHs with different metal ratios, the amounts of Mg­(NO₃)₂ 6H₂O and Al­(NO₃)₃ 9H₂O were adjusted while maintaining a total molarity at 0.02 mol.

The AgI/LDH composite was synthesized by dispersing 200 mg of LDH powder in 18.5 mL of 0.01 M AgNO₃ aqueous solution in a 100 mL beaker, followed by sonication for 5 min. Then, 1 mL of 0.222 M NaI aqueous solution was added dropwise under vigorous stirring. The mixture was stirred at room temperature for 24 h. The resulting solid was collected by centrifugation at 5000 rpm, washed three times with deionized water, and dried at 60 °C under vacuum for 20 h. The light absorption characteristics of the prepared samples are presented in Supporting Information Figure S28.The AgI/Mg₂Al₁LDO composite was prepared by grinding 100 mg of calcined Mg₂Al₁LDH (calcined at 500 °C with a heating rate of 10 °C·min^-1 for 2 h) together with 22 mg of AgI under ambient laboratory conditions for 5 minutes.

4.3. Composite Regeneration

To regenerate the composite, 100 mg of used AgI/Mg₂Al₁LDH was mixed with 30 mL of deionized water and sonicated for 5 min. Then, 60 mg of Na₂CO₃ was added, and the mixture was sealed and stirred at 85 °C for 1 h. After cooling to room temperature, the solid was washed three times with deionized water by centrifugation at 5000 rpm and dried at 60 °C under vacuum for 20 h.

4.4. C–C Cross-Coupling Reactions and Product Analysis

Photoreactions were conducted in a 10 mL quartz glass tube that does not absorb light beyond 350 nm wavelength. The composite, reactants and a magnetic stir bar were added to the tube, which was purged with nitrogen gas at 100 mL min^–1 for 1 min before sealing. The tube was stirred at 600 rpm and irradiated with an LED light of a specific wavelength (e.g., 400 nm) at a controlled temperature.

After the reaction, the mixture was filtered through a 0.22 μm Millipore filter to remove the composite. The filtrate was analyzed by gas chromatography using a Panna GC-1949 instrument equipped with an AB-5 column. Product identification was carried out with an Agilent 7890A/5975C gas chromatography–mass spectrometer. Product concentrations were quantified using an external standard method.

The yield was defined as the ratio of the final product (biaryls) to the initial amount of reactant (aryl iodide) used. The AQY value was calculated using the following equation:

$$\mathrm{AQY}=\frac{N_{light}-N_{dark}}{n}\times 100\%$$

where N _light is the number of product molecules formed under irradiation, (N _dark) is the number formed in the dark, and n is the number of incident photons.

4.5. Composite Characterization

XRD patterns were recorded using a MiniFlex600-C diffractometer (Rigaku) operated at 40 kV and 15 mA using CuKα radiation (λ = 1.541862 Å), with a step size of 0.01° in 2θ. Morphological and elemental mapping was performed using a JEOL JEM-F200 transmission electron microscope operating at 200 kV. Diffuse reflectance UV–visible spectra were recorded using a Shimadzu UV-3600 plus spectrophotometer. X-ray photoelectron spectroscopy data were collected using a Shimadzu/Krayos AXIS Ultra DLD instrument. In-situ irradiation XPS measurements were performed using a ThermoFisher ESCALAB 250Xi X-ray photoelectron spectrometer. The system utilized a monochromatic Al–Kα X-ray source (1486.6 eV) for excitation, operating under ambient conditions with illumination provided by a 300 W xenon lamp (PLS-SXE300E, Beijing Perfectlight, China) emitting light in the wavelength range of 390–780 nm. Prior to conducting the irradiation experiments, XPS spectra of all elemental components in the samples were collected in the dark as reference data. The charging effects were corrected by setting the C 1s binding energy of the adventitious carbon to 284.8 eV. Fluorescence spectra were measured on a Hitachi F7000 FL spectrophotometer with an excitation wavelength of 315 nm and emission range from 350 to 600 nm. Electron paramagnetic resonance measurements were performed on a Bruker EMXnano spectrometer operating in X-band mode with a field modulation frequency of 100 kHz. Nitrogen adsorption data were collected at –196 °C on a JW-BK2000 analyzer, and specific surface areas were calculated using the Brauner-Emmet-Teller (BET) method. Metal content was measured by inductively coupled plasma optical emission spectrometry on an Agilent 720ES. Absorption spectra of the liquid samples were measured using a LabTech BlueStarA UV–vis spectrophotometer. NMR spectra were obtained on a Bruker 400 MHz spectrometer. Attenuated total reflection Fourier transform infrared (ATR-FTIR) spectra were recorded using a PerkinElmer Spectrum Two spectrometer. Water content was measured by Karl Fischer titration on an AKF-2010 V at an oven temperature of 100 °C. Zeta potential was measured using a Microtrac nanotrac wave II instrument with the specimens dispersed in deionized water at a concentration of 1 mg mL^–1. Light intensity was measured using a probe coupled to a digital optical power meter (model S401C, PM100D, Thorlabs). Electrochemical impedance spectroscopy analysis was carried out over a frequency range from 10 kHz to 0.1 Hz at open-circuit potential with a 10 mV AC perturbation.

4.6. DFT Calculation

All calculations were performed using Gaussian 16, Revision A.03 package. All of the reactants, intermediates, transition states, products were optimized by the DFT with the M06–2X functional. For geometry optimizations and frequency calculations, BS-I basis set system was employed. In BS-I, In BS-I, we employed the Lanl2dz basis set for I with effective core potentials, 6–31G­(d) basis sets for other atoms. All the stationary structures were characterized with no imaginary frequency, and the transition state structures (TSs) were characterized with a single imaginary frequency. Intrinsic reaction coordinate (IRC) calculations were performed on the TSs. The solvent effect of toluene was evaluated through the SMD method, in which a better basis system BS-II was used. In BS-II, we employed the SDD basis set for I with effective core potentials, 6–311++G­(2d,2p) basis sets for other atoms. All reported energies are free energies at a concentration of 1 M and a temperature of 298.15 K.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/jacsau.6c00458.

• Figures S1–S32, and Tables S1–S6 (PDF)

P.H. found the activity of AgI/LDH and performed all the experiments, and drafted the manuscript. H.Y.Z., E.W., S.B., and Q.X. contributed to discussions on the mechanism and the methodology and revised the manuscript. X.Y., J.W., and C.T. contributed to NMR and other spectra analysis, as well as valuable discussions.

The authors declare no competing financial interest.