Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2026 May 20;65(29):e8133162. doi: 10.1002/anie.8133162

Frustrated Lewis Pair‐Induced Pre‐Adsorbed Oxygen Coupled With Reactant Availability as a Design Strategy for High‐Performance Single‐Atom Nanozymes

Yanjun Ji 1,2, Wenhao Miao 3, Lin Cheng 3, Huan Wang 1,✉, Zitong Zhu 1,2, Anjun Song 1,2, Junlin Ya 1,2, Ying Wang 1,2,✉, Jinsong Ren 1,2,✉, Xiaogang Qu 1,2,✉
PMCID: PMC13360751  PMID: 42160128

ABSTRACT

Single‐atom nanozymes (SANs) have recently been recognized as a promising artificial counterpart to natural enzymes. Nevertheless, their catalytic performance remains hindered by insufficient intrinsic activity, which stems from the substantial energy barriers associated with substrate activation and poor mass transport, limiting reactant availability. To address this, we designed a frustrated Lewis pair (FLP)‐based Fe‐N‐C single‐atom nanozyme (FLP‐Fe‐N‐C SAzyme) with enriched pyridinic N vacancy defects to enhance oxygen adsorption for the construction of high‐performance oxidase mimics. Density functional theory calculations demonstrate that the FLP structure, with Fe as Lewis acid sites and adjacent F regions as Lewis base sites, synergistically promotes oxygen dissociation and induces a pre‐adsorbed oxygen atom beside the catalytic center. Such a newly formed structure induced by the FLP sites showed an optimized electronic and geometric structure of the catalytic center, lowering the energy barrier. In addition, the abundant pyridinic‐N vacancies induced an enhanced oxygen adsorption, increasing the local oxygen concentration and thereby accelerating reactant availability. Hence, the FLP‐Fe‐N‐C SAzyme remarkably enhances the intrinsic activity of the oxidase‐mimicking SANs, achieving a significantly lower Km value than conventional Fe‐N‐C SANs.

Keywords: frustrated Lewis pairs, pre‐adsorbed oxygen, single‐atom nanozymes

Graphical Abstract. We propose a strategy for designing a frustrated Lewis pair‐based Fe‐N‐C single‐atom nanozyme via a one‐step chemical vapor deposition treatment of pristine Fe‐N‐C using ammonium fluoride. By integrating frustrated Lewis pairs‐induced oxygen pre‐adsorption with a reactant availability strategy, the resulting single‐atom nanozyme exhibits remarkably enhanced oxidase‐like activity compared to conventional single‐atom catalysts.

graphic file with name ANIE-65-e8133162-g001.jpg

1. Introduction

The oxidase‐mediated reduction of molecular oxygen (O2) is a crucial step in the biological energy supply and plays a vital role in sustaining the physiological functions of living organisms [1, 2]. Such a process relies on O2 activation to form highly reactive metal‐oxo species, which serve as key intermediates for a variety of reactions [3, 4]. While natural oxidases (e.g., cytochrome P450) exhibit remarkable efficacy in catalyzing the selective hydroxylation and epoxidation of organic compounds [5, 6, 7, 8], their practical application is often limited by intrinsic variability, instability, and high production costs. In this context, single‐atom nanozymes (SANs) have recently been recognized as a promising alternative to natural oxidase owing to their low cost, high stability, tunable surface properties, straightforward synthesis and storage, and relatively high biocatalytic activities [9, 10, 11]. Despite these merits, the biocatalytic performance of current oxidase‐mimicking SANs is still impeded by the following challenges: 1) the high energy barrier during the oxygen activation process [12]; 2) the limited mass transport that restricts reactant availability [13, 14]. As a result, most reported oxidase‐like SANs suffer from intrinsically slow catalytic kinetics. These challenges, specific to oxidase‐like systems, have motivated intensive efforts to develop innovative strategies for enhancing the oxidase‐like biocatalytic performance of SANs.

At present, research efforts have primarily addressed the first challenge, demonstrating that the accelerated desorption of intermediates, particularly *OH, plays a critical role in improving both peroxidase‐ and oxidase‐like catalytic processes [10, 11, 15, 16, 17, 18, 19, 20]. Consequently, design strategies for SANs have predominantly centered on enhancing *OH desorption while simultaneously maintaining favorable adsorption energies for other key intermediates [10, 15, 17, 21]. Toward this goal, various approaches such as coordination number engineering [17], edge‐site creation [20], and heteroatom doping have been employed to tailor the electronic structure of the catalytic center [12, 19, 22, 23, 24, 25, 26, 27, 28], thereby facilitating *OH desorption and enhancing overall catalytic activity. For instance, Han et al. demonstrated that Mn‐based SANs with a coordination number of three exhibited superior oxidase‐like activity compared to those with a coordination number of four [17]. Despite these advances, the oxidase‐like activity of SANs remains limited. Enhancing *OH desorption alone is insufficient to overcome the high energy barrier associated with oxygen activation. Equally critical is the parallel challenge of limited oxygen mass transport, which needs to be tackled in concert. Therefore, addressing these limitations requires a catalyst design that simultaneously overcomes the high energy barrier for oxygen activation and promotes reactant availability to construct high‐performance oxidase‐mimicking SANs.

Interestingly, the concept of frustrated Lewis pairs (FLPs), pioneered by Douglas W. Stephan and co‐workers in homogeneous catalysis, offers a talented avenue in this regard [29, 30]. FLPs comprise sterically hindered Lewis acids (LAs) and Lewis bases (LBs), where the LA acts as an electron acceptor, and the LB serves as an electron donor [29, 31, 32]. This unique combination enables the activation of a wide range of small molecules, including H2, CO2, SO2, and N2O [33, 34, 35, 36, 37]. However, conventional FLP catalysts are typically homogeneous and often rely on expensive reagents (e.g., borane‐ammonia adducts) as sacrificial agents, while also facing challenges in catalyst recovery and product purification. To overcome these limitations, researchers have shifted their focus toward the development of heterogeneous FLP catalysts, in which sterically hindered LA and LB sites are integrated within a single material [34, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47]. For instance, Qu and co‐workers successfully constructed FLP‐rich sites on CeO2 surfaces, demonstrating their efficacy in hydrogenation reactions of unsaturated substrates such as olefins, alkynes, CO2, and esters [48, 49, 50]. Nevertheless, the rational integration of FLP motifs into SANs to construct efficient enzyme‐like catalysts remains an unexplored frontier. We thus propose that constructing FLP‐based SANs constitutes a distinctive and compelling strategy to address the current limitations in oxygen activation.

In addition, previous studies have shown that oxidase‐mimicking SANs display markedly enhanced catalytic performance in O2‐saturated buffers relative to air‐saturated conditions, highlighting the pivotal influence of reactant concentration on catalytic activity [11, 17]. A comparable trend has been reported in oxygen reduction reaction (ORR) kinetics, where the pre‐adsorption of O2 molecules on Fe single‐atom catalysts significantly improves ORR efficacy [51]. Building upon these insights, we herein report the rational design of an FLP‐based Fe‐N‐C single‐atom nanozyme (denoted as FLP‐Fe‐N‐C SAzyme) synthesized via a one‐step chemical vapor deposition treatment of pristine Fe‐N‐C with ammonium fluoride (NH4F). During this in situ modification, the NH3 released from NH4F induces a controlled etching process, thereby generating abundant pyridinic‐N vacancies within the microporous framework. These vacancy defects substantially enhance O2 adsorption, thereby improving reactant availability. As a result, the FLP‐Fe‐N‐C SAzyme exhibits exceptional oxidase‐like activity, achieving a Michaelis constant (Km) significantly lower than that of pristine Fe‐N‐C and outperforming most reported state‐of‐the‐art oxidase‐mimicking nanozymes. Density functional theory (DFT) calculations elucidate that Fe and adjacent F regions function as the Lewis acid and base sites, respectively, forming an FLP motif that cooperatively promotes O2 dissociation and then induces a pre‐adsorbed oxygen atom beside the catalytic center. Such a newly formed structure optimizes both the electronic and geometric structure of the active sites, thus facilitating the oxygen reduction rate‐determining step. This study not only presents an effective route to enhance the catalytic kinetics of oxidase‐mimicking nanozymes but also opens a novel avenue for constructing heterogeneous FLP‐based single‐atom catalysts with high efficacy.

2. Results and Discussion

2.1. FLP‐Fe‐N‐C SAzyme Synthesis and Characterizations

The FLP‐Fe‐N‐C SAzyme (F‐FeSA) was synthesized via a one‐step chemical vapor deposition (CVD) treatment of pristine Fe‐N‐C using ammonium fluoride (NH4F) as the fluorine source (Figure 1a). Briefly, Fe(acac)3/ZIF‐8 was first prepared following previously reported methods [11]. Scanning electron microscopy (SEM) revealed the well‐defined polyhedral particles with uniform sizes, consistent with our earlier work, confirming the successful synthesis of Fe‐doped ZIF‐8 (Figure S1) [11]. Furthermore, the X‐ray diffraction (XRD) pattern of Fe(acac)3/ZIF‐8 closely resembled that of pristine ZIF‐8 (Figure S3), indicating that the incorporation of Fe(acac)3 did not disrupt the crystalline structure of ZIF‐8. Subsequently, the obtained Fe(acac)3/ZIF‐8 nanocrystals were pyrolyzed at 950°C for 2 h under an Ar atmosphere to yield the standard Fe‐N‐C sample (FeSA). For the CVD treatment, FeSA and NH4F powders were separately placed in two porcelain boats, with NH4F positioned upstream in the quartz tube furnace. A secondary pyrolysis was then conducted at 900°C for 1 h under a H2/Ar atmosphere to produce the final F‐FeSA catalyst. During this process, gaseous fluorine‐containing species released from NH4F decomposition were captured by the Fe‐N‐C framework, enabling the successful incorporation of F into the F‐FeSA structure. Meanwhile, NH3 produced from NH4F decomposition induced a controlled etching process that selectively removed pyridinic N, as previously reported in the literature, thereby generating abundant pyridinic‐N vacancies within the microporous network [51, 52, 53].

FIGURE 1.

FIGURE 1

Open in a new tab

The synthesis and characterization of single‐atom catalysts. (a) Schematic illustration of the synthetic procedures for the F‐FeSA and FeSA catalysts. (b, c) AC‐HAADF‐STEM image and the corresponding EDX elemental mapping of the F‐FeSA catalyst. (d) AC HAADF‐STEM image of the F‐FeSA catalyst. (e, f) AC HAADF‐STEM image and the corresponding EDX elemental mapping of the FeSA catalyst. (g) AC HAADF‐STEM image of the FeSA catalyst.

The obtained F‐FeSA and FeSA samples were subsequently characterized using a range of analytical techniques. SEM (Figure S2) and transmission electron microscopy (TEM) images (Figure 2a,b) showed that both F‐FeSA and FeSA maintained uniform polyhedral morphologies similar to the Fe(acac)3/ZIF‐8 precursor, indicating that the incorporation of F atoms did not significantly affect the overall structure of the Fe‐N‐C framework. Corresponding energy‐dispersive X‐ray (EDX) elemental mapping (Figure 1b–f) confirmed the homogeneous distribution of Fe, N, and C in both catalysts. Notably, a clear F signal was detected in F‐FeSA (Figure 1c), verifying the successful incorporation of fluorine. XRD patterns of F‐FeSA and FeSA (Figure 2d) displayed two characteristic diffraction peaks associated with graphitic carbon, with no detectable metal‐related reflections, suggesting the absence of metal crystallites and supporting the atomic dispersion of Fe. This is further supported by aberration‐corrected HAADF‐STEM (AC HAADF‐STEM) images (Figure 1d,g), in which numerous isolated bright dots corresponding to single Fe atoms were observed in both samples. As reported in previous studies, NH3 released during the CVD process can induce an etching effect, generating abundant pyridinic‐N vacancies within the carbon framework [52, 53]. To validate this, the Brunauer–Emmett–Teller (BET) surface areas (Figure S4) of the two catalysts were measured. As expected, F‐FeSA exhibited a significantly larger surface area compared to FeSA, indicating the formation of a more porous structure that facilitates oxygen availability. Consistently, Raman spectroscopy (Figure 2c) revealed a higher D/G intensity ratio for F‐FeSA, suggesting the introduction of more structural defects during the etching process. These defects may serve as additional catalytic active sites, while the enhanced porosity also contributes to improved oxygen adsorption capacity [51]. Finally, dynamic light scattering (DLS) analysis coupled with zeta potential measurements demonstrated that F‐FeSA possessed a comparable hydrodynamic size and similarly negative surface potential to FeSA, confirming that the fluorination process did not markedly alter the particle dispersion behavior (Figure S5, S6).

FIGURE 2.

FIGURE 2

Open in a new tab

(a, b) TEM images of F‐FeSA and FeSA. (c) Raman spectra of F‐FeSA and FeSA. (d) Powder XRD patterns of F‐FeSA and FeSA. (e, f) C 1s high‐resolution XPS spectra of F‐FeSA and FeSA. (g) EPR spectra of F‐FeSA and FeSA. (h, i) N 1s high‐resolution XPS spectra of F‐FeSA and FeSA.

X‐ray photoelectron spectroscopy (XPS) was employed to analyze the elemental states and compositions of the catalysts. In the C 1s spectra (Figure 2e,f), the peak at 288.94 eV showed a pronounced increase in intensity for F‐FeSA compared with FeSA, which can be attributed to the formation of C–F bonds during the pyrolysis process [54]. The high‐resolution N 1s XPS spectra (Figure 2 h,i) displayed three distinct signals attributable to pyridinic, pyrrolic, and graphitic nitrogen. As expected, the N content decreased markedly in F‐FeSA (2.6%) relative to FeSA (5.92%) (Table S4). This reduction is likely due to the NH3 released in situ during CVD, which induced a selective etching effect that removed pyridinic N sites from the carbon matrix under high‐temperature conditions. Such selective removal facilitates the formation of pyridinic‐N vacancies. Electron paramagnetic resonance (EPR) spectroscopy (Figure 2g) further confirmed the generation of N vacancies. Additionally, near‐edge X‐ray absorption fine structure (NEXAFS) (Figure S20) measurements were performed to verify the formation of pyridinic‐N vacancies. As anticipated, the C–N–C* absorption peak at 288.6 eV decreased substantially in F‐FeSA compared to FeSA, indicating successful removal of pyridinic N. These findings are consistent with the N 1s XPS results and the EPR data. We further analyzed the variation in oxygen content of F‐FeSA and FeSA. F‐FeSA exhibited a higher oxygen content than FeSA (Table S4), indicating F‐FeSA possesses a stronger oxygen adsorption capability.

We further carried out X‐ray absorption near‐edge structure (XANES) to reveal the chemical state and electronic structures of the catalyst (The detailed fitting parameters are provided in Table S1). According to near‐edge absorption in the normalized XANES curves (Figure 3a), both F‐FeSA and FeSA are located between Fe foil and Fe2O3, indicating the oxidation state of Fe in F‐FeSA and FeSA is between 0 and +3. Fourier transform extended X‐ray absorption fine structure (FT‐EXAFS) curves of the samples in R space were provided in Figure 3b–e. A dominant peak at 2.25 Å in the Fe foil (Figure 3d, Figure S7) can be assigned to the Fe‐Fe bond. Meanwhile, another two dominant peaks at 1.44 Å and 3.19 Å in Fe2O3 (Figure 3h) can be assigned to the Fe‐O pathway and the Fe‐O‐Fe pathway. Then, F‐FeSA and FeSA only possess one dominant peak near 1.5 Å with comparable location and intensity (Figure 3b–e), which were attributed to the Fe‐N bond, suggesting the absence of the Fe‐F bond in F‐FeSA. In addition, Fe‐Fe interactions were not observed in both F‐FeSA and FeSA (Figure 3d), indicating the successful construction of a single‐atom catalyst. Furthermore, EXAFS curve fitting (Figure 3b–i) reveals that the first‐shell coordination number of Fe in both F‐FeSA and FeSA is approximately 4, suggesting that the active sites adopt an Fe–N4/CF configuration. The WT‐EXAFS contour plots of F‐FeSA and FeSA (Figure 3j) display comparable R and k values, further confirming that the Fe species in these SACs predominantly exist in an Fe–N4 environment. This finding aligns with observations obtained from spherical aberration–corrected electron microscopy. Therefore, combined XPS and EXAFS analyses verify that the planar fluorination engineering of the Fe–N4 site was successfully achieved in the Fe‐N4/CF catalyst.

FIGURE 3.

FIGURE 3

Open in a new tab

Characterization of Fe‐based catalysts by XANES and EXAFS measurements. (a, d) XANES spectra and corresponding FT‐EXAFS results in R‐space for F‐FeSA, FeSA, Fe foil, and Fe2O3 as reference samples. (b, c, e, f, h, i) Fitting results of F‐FeSA, FeSA, and Fe2O3 in R‐space and k‐space. (g) k‐space EXAFS spectra at the Fe K‐edge for F‐FeSA and reference samples. (j) Wavelet transform analysis of Fe foil, F‐FeSA, and FeSA.

2.2. Catalytic Activity of F‐FeSA and FeSA

The enzyme‐like catalytic activity of F‐FeSA and FeSA was characterized by typical chromogenic reactions, with 3,3,5,5‐tetramethylbenzidine (TMB) chosen as an electron donor to catalyze the O2 reduction process. As shown in Figure 4a and Figure S16, F‐FeSA shows markedly higher absorbance at 652 nm, accompanied by a much more intense blue coloration than that of FeSA, indicating that the incorporation of F to construct a frustrated Lewis pair on a single‐atom catalyst was an ideal strategy for optimizing catalyst performance. The absorbance (Figure 4b) increases with reaction time and exhibits a linear relation in the first 100 s without an obvious decline, which could provide us a chance to test specific activity accurately. Notably, we performed a control experiment using a catalyst prepared under identical conditions but without Fe(acac)3 encapsulation (denoted as NC). As shown in Figure 4b and Figure S28, this Fe‐free catalyst exhibited negligible oxidase‐like activity, confirming that the Fe sites are indeed the essential catalytic centers. Then, we notice that nanozyme exhibited a higher performance in O2‐saturated HAc‐NaAc buffer (PH = 3.6) in previous literature [11, 17]. Thus, different gases, including N2, air, and O2‐saturated HAc‐NaAc buffer, were prepared to demonstrate the crucial role of O2 during the oxidase‐like performance. As expected, the O2‐saturated HAc‐NaAc buffer possesses the highest activity (Figure 4f). The nonlinear response to O2 concentration is expected for a bi‐substrate enzyme‐like reaction. Hence, replacing air with oxygen will naturally increase the dissolved oxygen concentration several times, but this increase does not proportionally enhance the activity. Moreover, regarding the residual catalytic activity under N2, we note that complete removal of dissolved oxygen is challenging under typical experimental conditions. We aimed to demonstrate the trend of O2 dependence rather than achieve absolute anaerobic conditions. The consistent decrease in activity from O2‐saturated to air to N2 conditions supports the O2‐driven mechanism. O2‐saturated conditions were used to ensure that oxygen is not a limiting factor, allowing us to measure the maximum catalytic activity and clearly demonstrate the critical role of oxygen in the oxidase‐like mechanism. This approach enables a more accurate comparison of the intrinsic activity differences between F‐FeSA and FeSA. Therefore, the following tests were performed in the O2‐saturated HAc‐NaAc buffer. Firstly, the specific activities of F‐FeSA and FeSA were evaluated according to standardized assay protocols to compare their catalytic performance. F‐FeSA achieved 19.67 U mg− 1, which is higher than that of FeSA (9.33 U mg− 1) and previously reported nanozymes (Figure 4c and Table S5). Additionally, increasing the incubation temperature slightly accelerated the reaction kinetics (Figure S38). Subsequently, typical Michaelis–Menten curves were obtained for both F‐FeSA and FeSA over a suitable range of TMB concentrations (Figure 4d,e), and the kinetic parameters are summarized in Table S3. The kinetic analysis reveals a substantially decreased Km value (0.1976 mM for F‐FeSA vs. 0.7156 mM for FeSA), indicating markedly enhanced substrate affinity towards TMB. This suggests that the FLP strategy primarily improves catalytic kinetics and substrate accessibility rather than merely increasing the apparent reaction rate. Meanwhile, the Kcat value for F‐FeSA is 2.33, and FeSA toward TMB is 1.20, indicating our modulation strategy successfully increases the turnover of each active site. As expected, the F‐FeSA exhibited a higher Vmax, Kcat, specific activity, and smaller Km than the FeSA. In short, these results suggest that F‐FeSA facilitates oxygen activation and possesses superior oxidase‐like activity. To further verify the substrate versatility of F‐FeSA and FeSA, two additional typical oxidase substrates, diaminobenzidine (DAB) and o‐phenylenediamine (OPD), were tested. As expected, both substrates could be efficiently oxidized by the catalysts. Notably, F‐FeSA exhibited markedly higher activity than FeSA toward the oxidation of these representative substrates, confirming its enhanced oxidase‐like catalytic performance and broad substrate applicability (Figure S33). Fluoride ions possess a strong coordination affinity toward iron centers and can preferentially bind to the exposed Fe atoms in Fe–N4 moieties. Hence, KF was used as a poisoning regent (Figure S9), the oxidase‐like activity of F‐FeSA was significantly inhibited, indicating Fe‐N4 sites are the real active sites. An electron paramagnetic resonance spectroscopy (EPR) experiment was tested to explore the active radicals. As shown in Figure 4i, characteristic peaks of DMPO‐•O2 − were detected in EPR spectra, indicating the generation of superoxide radicals (•O2 −). Meanwhile, the peak of F‐FeSA exhibited a higher intensity than FeSA, indicating F‐FeSA possesses a stronger activity to transform oxygen into reactive species. In addition, only slightly characteristic peaks of DMPO‐·OH can be detected by the EPR experiment after reaction for half an hour, indicating hydroxyl radicals were not the main reactive species (Figure 4l). Finally, the stability of the catalyst was characterized by cyclic voltammetry (CV) measurements. As shown in Figure S39, after 40 cycles, the CV curves exhibited negligible change, indicating good stability of the catalysts. This suggests that the active sites remain stable under operating conditions.

FIGURE 4.

FIGURE 4

Open in a new tab

Catalytic performance of single‐atom nanozymes. (a) Comparison of UV–vis absorption curves for different catalysts. (b) Comparison of UV–vis absorption–time course spectra for different catalysts. (c) OXD‐like specific activity of the catalysts. (d, e) Steady‐state kinetics of OXD‐like activity for F‐FeSA and FeSA using TMB as the substrate. (f) Comparison of OXD‐like activity under saturated N2, air, and O2 conditions. (g, h) Time‐dependent GSH scavenging ability by F‐FeSA (g) and FeSA (h), monitored via UV–vis spectra. (i) EPR spectra of F‐FeSA and FeSA in the presence of DMPO. (j, k) Time‐dependent NADPH scavenging ability by F‐FeSA(j) and FeSA(k), detected by UV–vis spectra. (l) EPR spectra of F‐FeSA in the presence of DMPO and HAc‐NaAc buffer (pH = 3.6). Error bars represent the standard deviation from the mean (n = 3).

Similar to oxidase‐like activity, GSHOx‐like and NOX‐like activity also require oxygen as the electron acceptor during the enzyme‐like reaction, indicating the frustrated Lewis pair strategy for boosting oxidase‐like reaction might also be useful for GSHOx‐like and NOX‐like reaction. Accordingly, DTNB was employed as a probe to assess the GSHOx‐mimicking activity of F‐FeSA and FeSA. As shown in Figure 4 g,h, the absorbance values at 412 nm decrease with reaction time for both F‐FeSA and FeSA, indicating the prepared catalyst could also act as GSHOx enzyme mimics. Of note, F‐FeSA exhibited a higher scavenging ability for glutathione (GSH) within the same reaction time (2 min), indicating the frustrated Lewis pair strategy is also adaptable for GSHOx‐like mimics (Figure S14). Analogous to its oxidase‐mimicking behavior, F‐FeSA exhibits enhanced catalytic activity in O2‐saturated buffer and achieves more efficient GSH depletion than FeSA (Figure S12–S14). Similarly, for NOX‐like reaction, thanks to the intrinsic absorbance values of NADPH at 340 nm, NOX‐like activity can be directly detected. As shown in Figure 4j,k, and Figure S15, the absorbance values at 340 nm decrease with reaction time for both F‐FeSA and FeSA, and F‐FeSA exhibited a higher scavenging ability for NADPH within the same reaction time (56 min), indicating the frustrated Lewis pair strategy is also adaptable for NOX‐like mimics.

We performed a series of control experiments to clarify the crucial role of the frustrated Lewis pair. First, the decisive role of the appropriate distance between LA and LB sites for the construction of FLP was verified by evaluating the performance of physically mixing NH4F and Fe‐N‐C (denoted as Mixing). Interestingly, physically mixing NH4F with Fe‐N‐C resulted in a slightly decreased activity compared to pristine Fe‐N‐C (Figure S31), likely due to the coordination of F‐ to Fe centers, which blocks access to the catalytic sites. This inhibitory behavior is consistent with the site‐poisoning results (Figure S9). We additionally performed control experiments with NaF added to the reaction system. As shown in Figure S32, NaF addition did not enhance the activity of the pristine FeSA. Together, these results indicate that free fluoride ions cannot reproduce the FLP effect. These results indicate that the simple mixture of LA and LB sites was not feasible to construct FLP sites. Thus, the unique spatial configuration of LA and LB in F‐FeSA caused by the in situ chemical vapor deposition strategy is crucial for the construction of FLP, which had also been confirmed by the theoretical calculations in the following section. In addition, an LB‐only control (denoted as F‐NC) was synthesized by first pyrolyzing Fe‐free ZIF‐8 to obtain NC (Figure S28), followed by treating NC with NH4F under the same chemical vapor deposition conditions. This material exhibited negligible oxidase‐like activity (with the 652 nm absorption barely detectable), confirming that isolated F sites alone cannot drive the reaction (Figure S29). In contrast, the LA‐only system (FeSA) achieved only ∼50% of the conversion of F‐FeSA. These results demonstrate that neither isolated LB nor LA sites is sufficient; the cooperative interactions between the Fe and F sites are essential for the enhanced catalytic performance. Finally, to exclude the influence of surface wettability, a water contact angle test was performed. As shown in Figure S30, water contact angle exhibits no significant difference, suggesting that surface wettability is not a determining factor. Thus, the enhanced catalytic performance can arise from the frustrated Lewis pair‐induced promoting effect. To demonstrate the generality of this FLP strategy, Co‐N‐C and F‐Co‐N‐C (F‐CoSA) were synthesized similarly to Fe‐N‐C and F‐Fe‐N‐C (Figure S34–S36), except that Fe(acac)3 was replaced with Co(acac)3. As expected, F‐CoSA exhibited a much higher activity compared to CoSA, indicating that such an FLP strategy is not limited to Fe‐N‐C but may be general for M‐N‐C catalysts (Figure S37). Both FeSA and F‐FeSA have the Fe‐N4 coordination environment, demonstrating that fluorine introduction does not alter the primary coordination structure of Fe active sites. Collectively, these results confirm that the enhanced catalytic performance could be attributed to the introduction of the frustrated Lewis pair.

2.3. Theoretical Analysis for the O2 Activation Mechanism

DFT calculations were performed to elucidate the reaction mechanism and the influence of the frustrated Lewis pair (FLP) strategy on the OXD‐like activity. Figure 5b compares the computed Gibbs free energy profiles for the oxygen reduction pathway on FeSA and F‐FeSA. Consequently, the maximum thermodynamic energy barrier for F‐FeSA is calculated to be 0.48 eV, significantly lower than that of FeSA (0.72 eV). For FeSA, the rate‐determining step (RDS) is identified as the *OH desorption step (*OH + H+ + e− → * + H2O). In contrast, the RDS for F‐FeSA shifts to the first PCET step (* + O2 + H+ + e− → *OOH). This result indicates that the F‐FeSA structure significantly lowers the energy barrier for *OH removal from the Fe center.

FIGURE 5.

FIGURE 5

Open in a new tab

Catalytic mechanism revealed by DFT calculation. (a) Calculated relative energy profiles for oxygen dissociation (Dis) and oxygen hydration (Hyd) pathways on FeSA and F‐FeSA models. (b) Gibbs free energy diagrams for the OXD‐like reaction on FeSA, F‐FeSA, and the pre‐adsorbed oxygen configuration (F‐O‐FeSA). (c) COHP analysis of the Fe‐O and C‐O interactions in the O2 dissociated configuration on F‐FeSA. PDOS analysis of the (d) Fe‐O and (e) C‐O interactions. (f) Linear correlation between the magnitude of Bader charge transfer (∆q, from Fe to *OH) and ∆G*OH. The insets display the corresponding differential charge density distributions (yellow: electron accumulation; cyan: electron depletion).

In addition to the oxygen‐hydration pathway discussed above, the oxygen‐dissociation pathway, another widely accepted ORR mechanism, was also examined. Transition state (TS) calculations (Figure 5a) indicate that, for the F‐FeSA system, the activation barrier for O2 dissociation (0.35 eV) is significantly lower than that for the oxygen hydration step (0.46 eV). This lower barrier, combined with the greater thermodynamic stability of the final state (−1.63 eV), suggests that the FLP strategy effectively promotes O2 dissociation. However, further analysis of the Gibbs free energy profile reveals a distinct feature: while O2 dissociation on F‐FeSA (resulting in one O atom adsorbed on Fe and the other on a neighboring C atom) is kinetically favored, the subsequent protonation of the C‐bound oxygen atom faces an extremely high energy barrier of 1.18 eV. The PDOS (Figure 5e) results reveal significant overlap between the C and O orbitals, indicating strong hybridization. Furthermore, COHP analysis shows that the antibonding states of the C‐O bond are located far above the Fermi level and remain largely unoccupied (Figure 5c). Consequently, bonding states dominate the C‐O interaction, resulting in a robust bond, as confirmed by the highly negative integrated COHP (ICOHP) value of −8.13 eV. This strong C‐O interaction suggests that the C‐bound oxygen acts as a stable, pre‐adsorbed species rather than a transient intermediate.

Consequently, the structure featuring the pre‐adsorbed oxygen species was identified as the catalytically active configuration (denoted as F‐O‐FeSA) for subsequent evaluation of the OXD‐like reaction. As shown in Figure 5b, the energy barrier for *OH desorption from the Fe center on F‐O‐FeSA is markedly lower than that on FeSA. While its RDS remains the first PCET step, consistent with F‐FeSA, the calculated maximum thermodynamic energy barrier is further reduced to 0.31 eV, indicating exceptional intrinsic activity. According to the Sabatier principle, an ideal catalyst requires a moderate adsorption strength for reaction intermediates to facilitate efficient product desorption. In this context, the superior performance of F‐O‐FeSA is attributed to its optimized *OH binding energy (−0.33 eV). To elucidate the electronic origin of this enhancement, differential charge density analysis and Bader charge calculations were performed to correlate the electron transfer (from Fe to *OH) with the ∆G *OH across the catalytic systems. As presented in Figure 5f, a clear linear correlation is observed between the magnitude of charge transfer and ∆G *OH. Notably, the F‐O‐FeSA structure achieves this optimal OH adsorption strength by modulating the electron transfer capacity of the Fe center, thereby explaining its enhanced OXD‐like catalytic performance. The charge transfer analysis during the O–O bond dissociation process on the catalyst model is shown in Figure S27.

2.4. Catalytic Application

We further designed a series of nanozyme‐related applications to demonstrate the superior catalytic performance of our FLP‐based Fe‐N‐C single‐atom nanozyme. Single‐atom nanozymes have recently attracted considerable attention as promising substitutes for natural oxidases due to their low cost, excellent stability, tunable surface chemistry, ease of preparation and storage, and notable catalytic performance. The strong oxidase‐mimicking activity enables rapid conversion of colorless TMB to blue oxTMB. When reductive species are introduced, oxTMB can be immediately converted back to TMB. This reversible chromogenic transition provides a versatile strategy for establishing multifunctional colorimetric sensing systems for small‐molecule detection. Ascorbic acid (AA, vitamin C) is an essential nutrient involved in tissue repair and the biosynthesis of various neurotransmitters. It is abundant in fresh fruits and leafy vegetables and serves as a crucial antioxidant capable of scavenging free radicals and mitigating oxidative damage. In many analytical scenarios, AA is also employed as an indicator for assessing the total antioxidant capacity (TAC) of real samples. Hence, we intended to use AA detection as a model demonstration of the enhanced performance and versatility of F‐FeSA. We constructed a simple colorimetric platform for quantitative AA analysis. As illustrated in Figure 6a,d, the assay relies on the oxidase‐like catalytic behavior of F‐FeSA. With increasing AA concentration, the UV‐vis absorption at 652 nm gradually decreased due to the reduction of oxTMB. A strong linear relationship (R = 0–6 µg /mL) was observed between AA concentration and absorbance within a defined range, and the limit of detection (LOD) was determined to be 0.021 µg /mL. To examine the applicability of this strategy in real matrices, AA contents in several fruit samples were measured. Among them, grapes exhibited the highest antioxidant capacity Figure 6 g,h. Considering that conventional dopamine (DA) detection methods—including HPLC, fluorescence spectroscopy, and chemiluminescence—often involve high costs and laborious sample pretreatment, we further exploited the oxidase‐like properties of F‐FeSA to construct a DA‐responsive colorimetric sensor. Experimental results (Figure 6b) revealed that DA markedly inhibits the catalytic oxidation of TMB by F‐FeSA, leading to a decrease in absorbance at 652 nm. The developed sensing system exhibited a linear response (Figure 6e) to DA concentrations ranging from 0 µg /mL to 50 µg /mL, with an LOD of 5.7 µg /mL. To further demonstrate the versatility of F‐FeSA, an additional colorimetric platform was also designed for tannic acid (TA) detection (Figure 6c,f). Notably, the performance of our F‐FeSA surpasses that of FeSA obviously (Figure S18 and S19). Finally, the selectivity and anti‐interference capability of the sensing platform were evaluated by adding potential interferences at concentrations exceeding AA. As shown in Figure 6i, except for Cysteine (Cys) with strong reducing capability, all other tested substances caused negligible signal variation, confirming the high selectivity of the proposed colorimetric system.

FIGURE 6.

FIGURE 6

Open in a new tab

Application of F‐FeSA and FeSA. (a–c) UV–vis absorption spectra of TMB + F‐FeSA mixtures in the presence of varying concentrations of AA, DA, and TA. (d–f) Linear calibration plots for the detection of AA, DA, and TA based on the F‐FeSA platform. (g) Newly prepared fruit samples. (h) Relative antioxidant capacity of the fresh fruit samples. (i) Selectivity and interference tests for AA, DA, and TA in the presence of various metal ions and amino acids. (j, k) Time‐dependent cellular uptake of F‐FeSA as detected by CLSM at the time points of 3 h and 6 h. The scale bars are 50 µm. (l) Fluorescence microscopy images of DCFH‐DA‐stained cells. Error bars represent the standard deviation from the mean (n = 3).

To demonstrate the ROS‐generation capacity of F‐FeSA and FeSA, we further conducted a series of cell‐based experiments. First, the cellular uptake of F‐FeSA was evaluated by incubating mouse breast cancer (4T1) cells with Rhodamine B (RhB)‐labeled F‐FeSA for different time periods. RhB is a commonly used fluorescent dye for labeling carbon‐based nanomaterials and has been widely employed for cellular uptake studies of single‐atom nanozymes. As shown in Figure 6j,k, the gradual increase in red fluorescence intensity over time indicates the efficient internalization of F‐FeSA by 4T1 cells (lysosomes were stained with Lyso‐Tracker Green). Herein, both F‐FeSA and FeSA are expected to exhibit enhanced catalytic activity in the acidic lysosomal environment. As anticipated, compared with the FeSA and control groups, cells treated with F‐FeSA displayed notably stronger green fluorescence (Figure 6l) by using DCFH‐DA as the ROS sensor, indicating a significantly higher ROS‐generation capability at the cellular level. These results indicate that the prepared catalyst exhibits an enhanced ROS‐generation capability, thereby offering substantial potential for improving therapeutic efficacy against deep‐seated solid tumors.

3. Conclusion

In summary, frustrated Lewis pair (FLP)‐based Fe‐N‐C single‐atom nanozymes (FLP‐Fe‐N‐C SAzymes) were successfully constructed via chemical vapor deposition. Density functional theory (DFT) calculations revealed that the Fe atoms act as Lewis acid sites, while the F regions function as Lewis base sites, synergistically promoting O2 dissociation and generating a pre‐adsorbed oxygen species adjacent to the catalytic center. This newly formed configuration exhibits a reduced energy barrier for oxidase‐like catalysis. In addition, owing to the etching effect during CVD, the FLP‐based single‐atom catalyst possesses an enhanced oxygen adsorption capacity, which further facilitates reactant availability. Overall, this work not only provides an effective strategy for optimizing oxidase‐like catalytic kinetics but also offers a novel approach for constructing heterogeneous FLP‐based single‐atom catalysts.

Author Contributions

Yanjun Ji: methodology, data curation, investigation. Wenhao Miao: methodology. Lin Cheng: methodology. Huan Wang: investigation. Zitong Zhu: methodology. Anjun Song: methodology. Junlin Ya: investigation. Ying Wang: methodology. Jinsong Ren: supervision, funding acquisition. Xiaogang Qu: supervision.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: anie72832‐sup‐0001‐SuppMat.pdf.

ANIE-65-e8133162-s001.pdf (3.1MB, pdf)

Acknowledgments

This study was supported by the National Natural Science Foundation of China (grant nos. T249526, 22437006, 22237006, 22475210, 22373097), the Jilin Province Youth Scientific and Technological Talent Support Project (QT202403), and the Science and Technology Development Project of Jilin Province (grant nos. 20230101040JC, 20260601015RC). Part of the computational time is supported by the High Performance and Computing Center of Jilin Province, the Network and Computing Center in Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, as well as the Computing Center in Jilin Normal University.

Contributor Information

Huan Wang, Email: huanwang@ciac.ac.cn.

Ying Wang, Email: ywang_2012@ciac.ac.cn.

Jinsong Ren, Email: jren@ciac.ac.cn.

Xiaogang Qu, Email: xqu@ciac.ac.cn.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

  • 1. Wikström M., Krab K., and Sharma V., “Oxygen Activation and Energy Conservation by Cytochrome c Oxidase,” Chemical Reviews 118, no. 5 (2018): 2469–2490, 10.1021/acs.chemrev.7b00664. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Wilson D. F., Harrison D. K., and Vinogradov A., “Mitochondrial Cytochrome c Oxidase and Control of Energy Metabolism: Measurements in Suspensions of Isolated Mitochondria,” Journal of Applied Physiology 117, no. 12 (2014): 1424–1430, 10.1152/japplphysiol.00736.2014. [DOI] [PubMed] [Google Scholar]
  • 3. Huang X. and Groves J. T., “Oxygen Activation and Radical Transformations in Heme Proteins and Metalloporphyrins,” Chemical Reviews 118, no. 5 (2017): 2491–2553, 10.1021/acs.chemrev.7b00373. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4. Pinakoulaki E., Daskalakis V., Ohta T., et al., “The Protein Effect in the Structure of Two Ferryl‐oxo Intermediates at the Same Oxidation Level in the heme Copper Binuclear Center of Cytochrome c Oxidase,” Journal of Biological Chemistry 288, no. 28 (2013): 20261–20266, 10.1074/jbc.M113.468488. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Ortiz de Montellano P. R., “Hydrocarbon Hydroxylation by Cytochrome P450 Enzymes,” Chemical Reviews 110, no. 2 (2010): 932–948, 10.1021/cr9002193. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Kille S., Zilly F. E., Acevedo J. P., and Reetz M. T., “Regio‐and Stereoselectivity of P450‐Catalysed Hydroxylation of Steroids Controlled by Laboratory Evolution,” Nature Chemistry 3, no. 9 (2011): 738–743, 10.1038/nchem.1113. [DOI] [PubMed] [Google Scholar]
  • 7. Espinoza R. V., Maskeri M. A., Turlik A., et al., “Epoxidation and Late‐Stage C–H Functionalization by P450 TamI are Mediated by Variant Heme‐Iron Oxidizing Species,” ACS Catalysis 12, no. 6 (2022): 3731–3742, 10.1021/acscatal.2c00364. [DOI] [Google Scholar]
  • 8. Hirao H., Kumar D., Thiel W., and Shaik S., “Two States and Two More in the Mechanisms of Hydroxylation and Epoxidation by Cytochrome P450,” Journal of the American Chemical Society 127, no. 37 (2005): 13007–13018, 10.1021/ja053847+. [DOI] [PubMed] [Google Scholar]
  • 9. Chen D., Xia Z., Guo Z., et al., “Bioinspired Porous Three‐Coordinated Single‐Atom Fe Nanozyme With Oxidase‐Like Activity for Tumor Visual Identification via Glutathione,” Nature Communications 14, no. 1 (2023): 7127, 10.1038/s41467-023-42889-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Li G., Liu H., Hu T., Pu F., Ren J., and Qu X., “Dimensionality Engineering of Single‐Atom Nanozyme for Efficient Peroxidase‐Mimicking,” Journal of the American Chemical Society 145, no. 30 (2023): 16835–16842, 10.1021/jacs.3c05162. [DOI] [PubMed] [Google Scholar]
  • 11. Ji Y., Wang Y., Wang H., Bai J., Ren J., and Qu X., “Maximizing Accessible Fe–N–C Sites on Highly Curved Surfaces via Chemical Vapor Deposition for Boosting Multienzyme‐Like Activities of Single‐Atom Nanozymes,” Angewandte Chemie International Edition 64, no. 32 (2025): e202505742, 10.1002/anie.202505742. [DOI] [PubMed] [Google Scholar]
  • 12. Xu Y., Ma Y., Chen X., et al., “Regulating Reactive Oxygen Intermediates of Fe−N−C SAzyme via Second‐Shell Coordination for Selective Aerobic Oxidation Reactions,” Angewandte Chemie International Edition 63, no. 36 (2024): e202408935, 10.1002/anie.202408935. [DOI] [PubMed] [Google Scholar]
  • 13. Jiao L., Li J., Richard L. L., et al., “Chemical Vapour Deposition of Fe–N–C Oxygen Reduction Catalysts With Full Utilization of Dense Fe–N4 Sites,” Nature Materials 20, no. 10 (2021): 1385–1391, 10.1038/s41563-021-01030-2. [DOI] [PubMed] [Google Scholar]
  • 14. Li J., Sougrati M. T., Zitolo A., et al., “Identification of Durable and Non‐Durable FeNx Sites in Fe–N–C Materials for Proton Exchange Membrane Fuel Cells,” Nature Catalysis 4, no. 1 (2020): 10–19, 10.1038/s41929-020-00545-2. [DOI] [Google Scholar]
  • 15. Li Z., Wu D., Wang Q., et al., “Bioinspired Homonuclear Diatomic Iron Active Site Regulation for Efficient Antifouling Osmotic Energy Conversion,” Advanced Materials 36, no. 46 (2024): e2408364, 10.1002/adma.202408364. [DOI] [PubMed] [Google Scholar]
  • 16. Ma C. B., Xu Y., Wu L., et al., “Guided Synthesis of a Mo/Zn Dual Single‐Atom Nanozyme With Synergistic Effect and Peroxidase‐Like Activity,” Angewandte Chemie International Edition 61, no. 25 (2022): e202116170, 10.1002/anie.202116170. [DOI] [PubMed] [Google Scholar]
  • 17. Wang Y., Cho A., Jia G., et al., “Tuning Local Coordination Environments of Manganese Single‐Atom Nanozymes With Multi‐Enzyme Properties for Selective Colorimetric Biosensing,” Angewandte Chemie International Edition 62, no. 15 (2023): e202300119, 10.1002/anie.202300119. [DOI] [PubMed] [Google Scholar]
  • 18. Choi D., Jung H., Im J., et al., “Bridging the Catalytic Turnover Gap Between Single‐Atom Iron Nanozymes and Natural Enzymes by Engineering the First and Second Shell Coordination,” Advanced Materials 36, no. 13 (2024): e2306602, 10.1002/adma.202306602. [DOI] [PubMed] [Google Scholar]
  • 19. Ji S., Jiang B., Hao H., et al., “Matching the Kinetics of Natural Enzymes With a Single‐Atom Iron Nanozyme,” Nature Catalysis 4, no. 5 (2021): 407–417, 10.1038/s41929-021-00609-x. [DOI] [Google Scholar]
  • 20. Kim K., Lee J., Park O. K., et al., “Geometric Tuning of Single‐Atom FeN 4 Sites via Edge‐Generation Enhances Multi‐Enzymatic Properties,” Advanced Materials 35, no. 19 (2023): e2207666, 10.1002/adma.202207666. [DOI] [PubMed] [Google Scholar]
  • 21. Ma L., Wang Y., Chen Y., et al., “Construction of Frustrated Lewis Pairs for Efficient Artificial Enzyme,” Advanced Functional Materials 35, no. 35 (2025): 2503358, 10.1002/adfm.202503358. [DOI] [Google Scholar]
  • 22. Ding S., Barr J. A., Lyu Z., et al., “Effect of Phosphorus Modulation in Iron Single‐Atom Catalysts for Peroxidase Mimicking,” Advanced Materials 36, no. 10 (2024): e2209633, 10.1002/adma.202209633. [DOI] [PubMed] [Google Scholar]
  • 23. Park S., Shim K. I., Nguyen P. T., et al., “Breaking the Selectivity Barrier of Single‐Atom Nanozymes Through Out‐of‐Plane Ligand Coordination,” Advanced Materials 37, no. 38 (2025): e2506480, 10.1002/adma.202506480. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Hao L., Liang X. J., Zhang Y., et al., “Fine‐Tuning the D ‐Band Center Position of Zinc to Increase the Anti‐Tumor Activity of Single‐Atom Nanozymes,” Advanced Materials 36, no. 48 (2024): e2412368, 10.1002/adma.202412368. [DOI] [PubMed] [Google Scholar]
  • 25. Wei S., Ma W., Sun M., et al., “Atom‐Pair Engineering of Single‐Atom Nanozyme for Boosting Peroxidase‐Like Activity,” Nature Communications 15, no. 1 (2024): 6888, 10.1038/s41467-024-51022-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Wei S., Sun M., Huang J., et al., “Axial Chlorination Engineering of Single‐Atom Nanozyme: Fe‐N 4 Cl Catalytic Sites for Efficient Peroxidase‐Mimicking,” Journal of the American Chemical Society 146, no. 48 (2024): 33239–33248, 10.1021/jacs.4c13335. [DOI] [PubMed] [Google Scholar]
  • 27. Zhang Z., Li F., Xi S., et al., “Atomic Metal‒Nonmetal Catalytic Pair Cooperatively Drives Efficient Enzyme‐Mimetic Catalysis,” Angewandte Chemie International Edition 64, no. 37 (2025): e202508651, 10.1002/anie.202508651. [DOI] [PubMed] [Google Scholar]
  • 28. Chen Y., Jiang B., Hao H., et al., “Atomic‐Level Regulation of Cobalt Single‐Atom Nanozymes: Engineering High‐Efficiency Catalase Mimics,” Angewandte Chemie International Edition 62, no. 19 (2023): e202301879, 10.1002/anie.202301879. [DOI] [PubMed] [Google Scholar]
  • 29. Welch G. C., Juan R. R. S., Masuda J. D., and Stephan D. W., “Reversible, Metal‐Free Hydrogen Activation,” Science 314, no. 5802 (2006): 1124–1126, 10.1126/science.1134230. [DOI] [PubMed] [Google Scholar]
  • 30. McCahill J. S., Welch G. C., and Stephan D. W., “Reactivity of “Frustrated Lewis Pairs”: Three‐Component Reactions of Phosphines, a Borane, and Olefins,” Angewandte Chemie International Edition 46, no. 26 (2007): 4968–4971, 10.1002/anie.200701215. [DOI] [PubMed] [Google Scholar]
  • 31. Stephan D. W., “Frustrated Lewis Pairs,” Journal of the American Chemical Society 137, no. 32 (2015): 10018–10032, 10.1021/jacs.5b06794. [DOI] [PubMed] [Google Scholar]
  • 32. Stephan D. W. and Erker G., “Frustrated Lewis Pair Chemistry: Development and Perspectives,” Angewandte Chemie International Edition 54, no. 22 (2015): 6400–6441, 10.1002/anie.201409800. [DOI] [PubMed] [Google Scholar]
  • 33. Das S., Laplaza R., Blaskovits J. T., and Corminboeuf C., “Engineering Frustrated Lewis Pair Active Sites in Porous Organic Scaffolds for Catalytic CO2 Hydrogenation,” Journal of the American Chemical Society 146, no. 23 (2024): 15806–15814, 10.1021/jacs.4c01890. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Li S., Wu Q., You X., et al., “Anchoring Frustrated Lewis Pair Active Sites on Copper Nanoclusters for Regioselective Hydrogenation,” Journal of the American Chemical Society 146, no. 40 (2024): 27852–27860, 10.1021/jacs.4c10251. [DOI] [PubMed] [Google Scholar]
  • 35. Mo Z., Kolychev E. L., Rit A., Campos J., Niu H., and Aldridge S., “Facile Reversibility by Design: Tuning Small Molecule Capture and Activation by Single Component Frustrated Lewis Pairs,” Journal of the American Chemical Society 137, no. 38 (2015): 12227–12230, 10.1021/jacs.5b08614. [DOI] [PubMed] [Google Scholar]
  • 36. Stephan D. W. and Erker G., “Frustrated Lewis Pairs: Metal‐Free Hydrogen Activation and More,” Angewandte Chemie International Edition 49, no. 1 (2010): 46–76, 10.1002/anie.200903708. [DOI] [PubMed] [Google Scholar]
  • 37. Wang L., Yan T., Song R., et al., “Room‐Temperature Activation of H 2 by a Surface Frustrated Lewis Pair,” Angewandte Chemie International Edition 58, no. 28 (2019): 9501–9505, 10.1002/anie.201904568. [DOI] [PubMed] [Google Scholar]
  • 38. Zhu J., Cui T., Chi J., et al., “Frustrated Lewis Pair Mediated f‐p‐D Orbital Coupling: Achieving Selective Seawater Oxidation and Breaking *OH and *OOH Scaling Relationship,” Angewandte Chemie International Edition 64, no. 2 (2025): e202414721, 10.1002/anie.202414721. [DOI] [PubMed] [Google Scholar]
  • 39. Zhong L., Liao X., Huang H., et al., “B, N Codoped Defective Reduced Graphene Oxide as a Highly Efficient Frustrated Lewis Pairs Catalyst for the Selective Hydrogenation of α,β‐Unsaturated Aldehydes to Unsaturated Alcohols,” Journal of the American Chemical Society 147, no. 4 (2025): 3840–3854, 10.1021/jacs.4c17103. [DOI] [PubMed] [Google Scholar]
  • 40. Xu B., Luo S., Hua W., et al., “Constructing Atomic Tungsten‐Based Solid Frustrated‐Lewis‐Pair Sites With D‐p Interactions for Selective CO 2 Photoreduction,” Journal of the American Chemical Society 147, no. 1 (2025): 200–210, 10.1021/jacs.4c08953. [DOI] [PubMed] [Google Scholar]
  • 41. Zou Y., Mao C., Xu M., et al., “Surface Frustrated Lewis Pairs in Titanium Nitride Enable Gas Phase Heterogeneous CO2 Photocatalysis,” Nature Communications 15, no. 1 (2024): 10604, 10.1038/s41467-024-54951-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Zhou M., Wang H., Liu R., et al., “Construction of Frustrated Lewis Pairs in Poly(heptazine Imide) Nanosheets via Hydrogen Bonds for Boosting CO 2 Photoreduction,” Angewandte Chemie International Edition 63, no. 36 (2024): e202407468, 10.1002/anie.202407468. [DOI] [PubMed] [Google Scholar]
  • 43. Yang Z., Huang T., Li M., et al., “Unveiling the Synergistic Role of Frustrated Lewis Pairs in Carbon‐Encapsulated Ni/NiO X Photothermal Cocatalyst for Enhanced Photocatalytic Hydrogen Production,” Advanced Materials 36, no. 24 (2024): e2313513, 10.1002/adma.202313513. [DOI] [PubMed] [Google Scholar]
  • 44. Li D. C., Pan Z., Tian Z., et al., “Frustrated Lewis Pair Catalyst Realizes Efficient Green Diesel Production,” Nature Communications 15, no. 1 (2024): 3172, 10.1038/s41467-024-47591-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45. Hu Z., Fan Z. W., Xu Z. M., et al., “Metal–Organic Frameworks‐Based Frustrated Lewis Pairs for Selective Reduction of Nitroolefins to Nitroalkanes,” Journal of the American Chemical Society 146, no. 26 (2024): 17924–17930, 10.1021/jacs.4c03756. [DOI] [PubMed] [Google Scholar]
  • 46. Chen W., Li S., Yi L., et al., “Precise Distance Control and Functionality Adjustment of Frustrated Lewis Pairs in Metal–Organic Frameworks,” Journal of the American Chemical Society 146, no. 17 (2024): 12215–12224, 10.1021/jacs.4c03133. [DOI] [PubMed] [Google Scholar]
  • 47. Liang Y., Zhang Z., Su X., et al., “Coordination Defect‐Induced Frustrated Lewis Pairs in Polyoxo‐Metalate‐Based Metal–Organic Frameworks for Efficient Catalytic Hydrogenation,” Angewandte Chemie International Edition 62, no. 37 (2023): e202309030, 10.1002/anie.202309030. [DOI] [PubMed] [Google Scholar]
  • 48. Zhang S., Huang Z.‐Q., Ma Y., et al., “Solid Frustrated‐Lewis‐Pair Catalysts Constructed by Regulations on Surface Defects of Porous Nanorods of CeO2,” Nature Communications 8, no. 1 (2017): 15266, 10.1038/ncomms15266. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49. Ma Y., Zhang S., Chang C.‐R., Huang Z.‐Q., Ho J. C., and Qu Y., “Semi‐Solid and Solid Frustrated Lewis Pair Catalysts,” Chemical Society Reviews 47, no. 15 (2018): 5541–5553, 10.1039/C7CS00691H. [DOI] [PubMed] [Google Scholar]
  • 50. Zhang S., Xia Z., Zou Y., et al., “Interfacial Frustrated Lewis Pairs of CeO2 Activate CO2 for Selective Tandem Transformation of Olefins and CO2 Into Cyclic Carbonates,” Journal of the American Chemical Society 141, no. 29 (2019): 11353–11357, 10.1021/jacs.9b03217. [DOI] [PubMed] [Google Scholar]
  • 51. Jia B., Xie X., Lin J., et al., “Harnessing Pyridinic N Vacancy Defect in Microporous Structures to Induce the Pre‐Adsorption of Oxygen and Boost Oxygen Reduction Reaction Kinetics,” Angewandte Chemie International Edition 64, no. 37 (2025): e202508674, 10.1002/anie.202508674. [DOI] [PubMed] [Google Scholar]
  • 52. Liu J., Liu Y., Nan B., et al., “A Two‐in‐One Strategy to Simultaneously Boost the Site Density and Turnover Frequency of Fe−N−C Oxygen Reduction Catalysts,” Angewandte Chemie International Edition 64, no. 14 (2025): e202425196, 10.1002/anie.202425196. [DOI] [PubMed] [Google Scholar]
  • 53. Zhong H., Wang J., Zhang B., et al., “Selective Introduction of Pentagon Defects Into Co‐N 4 Sites for Boosting Fenton‐Like Activity,” Advanced Functional Materials 35, no. 33 (2025): 2501208, 10.1002/adfm.202501208. [DOI] [Google Scholar]
  • 54. Bai J., Guan X., Qu H., et al., “Multiscale Structure Regulation Induced by Fluorine Coordination Enables High‐Performance and Durable PEMFC,” ACS Energy Letters 10, no. 6 (2025): 2743–2751, 10.1021/acsenergylett.5c00510. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting File: anie72832‐sup‐0001‐SuppMat.pdf.

ANIE-65-e8133162-s001.pdf (3.1MB, pdf)

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Articles from Angewandte Chemie (International Ed. in English) are provided here courtesy of Wiley

RESOURCES