X-ray diffraction is a structural characterization technique based on the diffraction phenomenon of X-rays by crystalline materials， which can be used to analyze key structural information such as unit cell parameters， crystallite size， and crystal structure types^[40]. XRD exhibits strong adaptability to sample morphology， making it applicable to powders， thin films， and even certain solution systems^[41]. Conventional ex situ XRD is typically employed to study structural changes in materials before and after reactions， but sample handling during the process may lead to the loss of structural evolution information. In contrast， in situ XRD enables real-time monitoring of dynamic changes in crystal structure during material reactions， offering superior timeliness and authenticity. By tracking variations in diffraction peaks under different reaction conditions （e.g.， temperature， electric field， atmosphere）， information such as phase transition behavior， crystal structure evolution， and microstrain changes can be obtained， providing critical insights for understanding the structure-property relationships of materials^[42,43]. In situ XRD finds broad applications across various fields， including the investigation of structural changes in minerals under high temperature and pressure to analyze ore composition， as well as the study of structural variations in protein crystals under specific conditions. It is particularly valuable for elucidating the structural evolution of electrode materials during charge/discharge processes and the phase transition behavior of electrocatalysts during reactions， thereby offering theoretical support for mechanistic studies and rational material design^[44]. For instance， in electrocatalytic UOR research， in situ XRD can be utilized to identify the true structure and active phase of catalysts under working conditions， providing evidence to clarify the dominant structural units in the reaction.

However， this technique still has certain limitations. The experimental setup demands high requirements， such as specialized in situ reaction cells or synchrotron X-ray sources. Additionally， its detection capability for amorphous or low-crystallinity components is limited， and the deconvolution of overlapping diffraction peaks remains challenging. Therefore， in situ XRD should be combined with other in situ or surface/interface characterization techniques to achieve a comprehensive understanding of complex systems.

In situ X-ray absorption spectroscopy （XAS） is a powerful technique based on synchrotron radiation that probes the local atomic and electronic structure of materials. Samples for XAS measurements include powders， thin films， and even liquids， each requiring specific preparation methods. For instance， powder samples are typically prepared by pelletizing or uniformly coating onto adhesive tape； thin films are often directly deposited on conductive substrates such as carbon paper or metal foils； liquid samples must be encapsulated in liquid cells equipped with X-ray-transparent windows for testing^[41]. The principle of XAS relies on the absorption characteristics of X-rays by atoms when interacting with matter^[45,46]. Near the absorption edge， core-level electrons are excited by X-ray energy into unoccupied orbitals. At energies approximately 50 eV above the edge， electrons are excited into the continuum， generating photoelectron waves that interfere with the scattered waves from neighboring atoms， resulting in oscillations in the absorption or fluorescence signal. Consequently， the XAS spectrum can be divided into X-ray absorption near-edge structure （XANES） and extended X-ray absorption fine structure （EXAFS）， providing insights into the electronic states， chemical bonding， coordination number， and local atomic environment of the material. XANES primarily reveals oxidation states and electronic symmetry， whereas EXAFS focuses on local structural parameters such as interatomic distances， coordination numbers， and the identity of neighboring atoms. Unlike in situ XRD， which relies on long-range ordered structures， XAS is well-suited for studying highly disordered materials due to its lower dependence on crystallinity. Moreover， XAS enables in situ measurements under real electrochemical reaction conditions， offering a more accurate representation of dynamic structural and valence state changes in catalysts^[47].

The technique primarily analyzes the local environment of the absorbing atom and faces challenges in distinguishing light elements such as C， N， and O as coordinating atoms. Therefore， complementary characterization methods are often necessary to obtain a more comprehensive understanding of the UOR.

In situ Raman spectroscopy is a technique based on the inelastic scattering phenomenon between incident light and molecules. When a sample is irradiated with laser light of a specific wavelength， electrons are initially excited from the ground state to a short-lived virtual energy state before relaxing to a lower vibrational energy level， emitting scattered light in the process^[48,49]. Depending on whether the wavelength of the scattered light differs from that of the incident light， scattering can be categorized as elastic （no wavelength change） or inelastic （wavelength change）， with Raman spectroscopy focusing on the latter. Inelastic scattering primarily includes Stokes scattering and anti-Stokes scattering： the former occurs when electrons relax to a vibrational energy level higher than the initial ground state， resulting in scattered light with a longer wavelength than the incident light， whereas the latter arises from transitions to a lower vibrational level， producing scattered light with a shorter wavelength^[50]. Since Stokes scattering is more probable under typical experimental conditions， it is usually the primary focus of Raman analysis. The characteristic vibrational modes induced by photon absorption generate distinct peaks in the Raman spectrum， reflecting molecular structural changes and granting Raman spectroscopy its “molecular fingerprinting” capability. This feature makes it an ideal tool for studying molecular structural evolution， surface intermediate formation， and reaction mechanisms in electrocatalytic processes^[51]. To enable in situ monitoring of electrocatalytic reactions， Raman measurements are often integrated with electrochemical systems， forming a confocal micro-Raman-electrochemical three-electrode platform. Typically， the working electrode is positioned at the center of the electrochemical cell， while the counter electrode and reference electrode are placed outside the laser path but close to the working electrode， with the choice of reference electrode depending on experimental requirements^[52,53]. During testing， Raman spectroscopy is frequently coupled with various electrochemical techniques， such as linear sweep voltammetry （LSV）， cyclic voltammetry （CV） and electrochemical impedance spectroscopy （EIS）， enabling synchronized multi-channel data acquisition and analysis via a potentiostat. Raman spectroscopy offers several advantages： （i） Minimal interference from water-due to its inherently weak Raman scattering signal， making it suitable for in situ detection in aqueous environments. （ii） No need for extensive sample pretreatment in most cases， allowing direct in situ characterization within the reaction system. （iii） High sensitivity toward symmetric or weakly polar functional groups （e.g.， C—C， C $\stackrel{\mathrm{ }\mathrm{ }\mathrm{ }\mathrm{ }}{=}$C， S—S）. （iv） Strong detection capability in the low-wavenumber region， facilitating the observation of low-energy processes such as lattice vibrations^[54].

However， challenges remain in Raman applications， including fluorescence interference and inherently weak signal intensity. During laser excitation， Stokes Raman scattering is often accompanied by strong fluorescence emission， which generates a broad background signal in the spectrum and significantly obscures Raman features^[55]. To mitigate fluorescence， strategies such as switching excitation wavelengths （e.g.， using near-infrared or ultraviolet lasers） are employed. Additionally， to overcome the intrinsically low Raman signal， several enhancement techniques have been developed， including surface-enhanced Raman spectroscopy （SERS）， tip-enhanced Raman spectroscopy （TERS）， and shell-isolated nanoparticle-enhanced Raman spectroscopy （SHINERS）. Among these， SERS stands out for its ultrahigh sensitivity， enabling detection at trace or even single-molecule levels， which is crucial for identifying low-concentration reaction intermediates and trace products.

Infrared absorption spectroscopy， as a vital tool for molecular structure analysis， finds extensive applications in materials science， chemistry， and catalysis. Its fundamental principle relies on transitions between vibrational or rotational energy levels within molecules. Under ambient conditions， atoms connected by chemical bonds in molecules continuously vibrate at specific frequencies， which typically fall within the range of infrared radiation. When the frequency of incident infrared light matches that of a particular vibrational mode， resonant energy transfer occurs， resulting in characteristic absorption peaks. Variations in the position， intensity， and shape of these absorption peaks collectively form molecule-specific infrared spectra， often referred to as molecular “fingerprints”^[56]. Unlike conventional dispersive infrared spectrometers that acquire spectra through wavelength-by-wavelength scanning， FTIR spectrometers obtain interferograms （interference signals） via an interferometer and subsequently convert them into frequency （or wavenumber） domain absorption spectra through Fourier transformation. Essentially， the Fourier transformation is a mathematical processing method that extracts frequency components from time-domain interferograms， thereby reconstructing complete infrared absorption spectra. This approach offers advantages such as high sensitivity， superior resolution， and rapid scanning capabilities， making it particularly suitable for studying dynamic processes in real time. In-situ FTIR spectroscopy integrates electrochemical measurement techniques with infrared spectral analysis， enabling real-time monitoring of electrocatalytic reactions at gas-liquid-solid interfaces. In UOR， this technique can track urea adsorption behavior， the formation and transformation of key intermediates （e.g.， CO， NH₂， CNO^-）， and the release of final products， thereby providing insights into reaction pathways and active sites. By capturing molecular signatures of reactants， intermediates， and products， in-situ FTIR plays a crucial role in elucidating complex reaction mechanisms at electrode surfaces， contributing significantly to the understanding of interfacial phenomena， adsorption behavior， and electrocatalytic processes.

However， FTIR spectroscopy exhibits several limitations： （i） the coexistence of multiple species during reactions leads to overlapping infrared absorption peaks， complicating spectral interpretation. （ii） Strong infrared absorption by water and organic solvents may obscure signals from important functional groups. （iii） The typically low concentrations of intermediates or products on material surfaces result in weak absorption signals， necessitating enhanced instrument sensitivity. （iv） Slight shifts in absorption peak positions may occur due to variations in the electrochemical environment of functional groups， primarily caused by electrode potential or solution pH affecting molecular electronic structures and vibrational modes， further increasing the difficulty of spectral analysis^[57-58].

In addition to commonly used characterization techniques such as in situ XRD， XAS， Raman spectroscopy， and FTIR， several novel in situ methods with unique advantages have emerged in recent years， significantly advancing our understanding of material evolution and reaction pathways during electrochemical processes^[59-63]. Among these， in situ differential electrochemical mass spectrometry （DEMS） has attracted considerable attention due to its exceptional sensitivity for detecting gaseous products and interfacial species^[64]. DEMS represents one of the few in situ techniques capable of simultaneously achieving mass， potential， spatial， and temporal resolution. This method directly couples an electrochemical cell with a mass spectrometer through a specialized semipermeable membrane， allowing gaseous or volatile species generated during reactions to enter the mass spectrometer’s vacuum system. By monitoring ion signals corresponding to different mass-to-charge ratios （m/z）， DEMS enables both qualitative and quantitative analysis of reaction products^[65,66]. With its millisecond-level temporal resolution， DEMS can track dynamic changes in real time， providing crucial information about reaction kinetics. In recent studies of the UOR， DEMS has been extensively employed to detect key gaseous products such as N₂ and CO₂， thereby elucidating reaction pathways and side processes. For instance， researchers can determine whether CO₂ is generated and subsequently deduce the carbon-nitrogen bond cleavage mechanism by monitoring gas signals at specific m/z values. Furthermore， DEMS could facilitate the calculation of Faradaic efficiency when combined with current signals， enabling highly accurate quantitative characterization of gaseous products. Despite its demonstrated advantages for in situ detection， the DEMS application faces several challenges. Primarily， the technique is mainly suitable for detecting gaseous or volatile products， with limited capability for identifying liquid-phase or solid-phase intermediates， making comprehensive analysis of all species in complex reaction systems difficult. Additionally， the construction of DEMS systems and membrane component design are relatively complex， imposing stringent requirements on experimental conditions （such as gas permeability and stability of membranes）， which consequently restricts their widespread adoption and standardized implementation. In summary， DEMS has become an increasingly important in situ characterization tool in electrochemical research for monitoring reaction products， identifying key intermediates， and elucidating reaction mechanisms.

In addition to DEMS technology for real-time monitoring of gaseous products， X-ray photoelectron spectroscopy （XPS） has emerged as a crucial surface analysis technique for investigating electrochemical systems， particularly for tracking the evolution of chemical states at electrode surfaces. Conventional XPS requires ultra-high vacuum conditions， making it incompatible with electrochemical reaction systems containing liquid phases or higher-pressure gas environments， thereby limiting its in situ detection capabilities. To overcome this limitation， researchers have developed near-ambient pressure XPS （NAP-XPS）， which enables surface analysis of catalysts under pressures ranging from millibars to several tens of millibars， better approximating realistic reaction conditions. By utilizing the tunable properties of synchrotron radiation light sources， NAP-XPS not only circumvents the constraints of vacuum environments but also facilitates depth-resolved elemental distribution analysis^[67]. This technique has been widely applied in electrochemical reactions to monitor real-time changes in element valence states and adsorbed intermediates on catalyst surfaces， providing critical insights into reaction mechanisms and the evolution of active sites. Nevertheless， NAP-XPS still faces challenges such as complex instrumentation， limited spatial resolution， and insufficient adaptability to electrolyte environments， restricting its broader application in more complex systems. Consequently， NAP-XPS is often employed in conjunction with theoretical calculations and other in situ characterization techniques to achieve multidimensional analysis of intricate catalytic systems.

Monitoring phase transitions of electrocatalysts under working conditions has become a research hotspot， as these transitions are intimately linked to catalytic activity and stability. During electrocatalytic reactions， catalysts often undergo phase transformations due to surface reconstruction processes. These structural transitions directly affect their electronic structure， surface configuration， and catalytic performance， making real-time monitoring of phase evolution crucial for understanding structure-property relationships and developing efficient， stable catalysts. To address this， researchers have recently employed various in situ/quasi-in situ characterization techniques to reveal phase transformation behaviors of materials under working conditions， spanning multiple dimensions from crystal structure changes and local coordination environment adjustments to lattice distortions. Wang et al.^[68] fabricated Ni（OH）₂/carbon paper electrodes and systematically investigated their structural evolution in alkaline media using in situ XRD coupled with electrochemical analysis. Through potential step experiments from 1.2 to 1.6 V， they simultaneously collected XRD patterns in 5 M KOH solutions with and without 1 M urea. The results demonstrated that in the absence of urea， the diffraction peak intensities of Ni（OH）₂ weakened with increasing potential while a new NiOOH （003） peak emerged at ~12.7°， confirming the electrochemical oxidation of Ni（OH）₂ to NiOOH （Fig.3a）. In contrast， with urea present， the Ni（OH）₂ peak intensities remained nearly unchanged， and no characteristic NiOOH peaks were observed， indicating that the generated NiOOH was rapidly reduced back to Ni（OH）₂ by urea with a lifetime shorter than the XRD detection timescale （Fig. 3b）. These in situ results directly verified the cyclic： Ni（OH）₂ → NiOOH → Ni（OH）₂ mechanism of the catalyst during urea electrooxidation.

While in situ XRD directly reveals the evolution of crystalline structures， in situ Raman spectroscopy provides more detailed atomic-level information about local structural rearrangements and chemical environment changes during phase transitions by monitoring molecular vibrations and chemical bond variations. Yang et al.^[69] synthesized a series of NiMn layered double hydroxides （Ni_1-xMn_x LDHs） on carbon paper via electrodeposition. Using in situ Raman spectroscopy， they clearly elucidated the phase transition pathways of Ni（OH）₂ and Ni_0.2Mn_0.8 LDH in 1 M KOH electrolyte with and without 0.33 M urea. For undoped Ni（OH）₂， characteristic NiOOH peaks （~479 and 561 cm^-1） appeared at approximately 0.38 V （vs. Hg/HgO） （Fig. 3c， d）， while Mn doping shifted this onset potential negatively to 0.30 V （Fig. 3e， f）. The Mn-O peak （603 cm^-1） weakened with increasing potential， suggesting Mn might be oxidized prior to Ni， thereby facilitating the Ni²⁺ to Ni³⁺ transition. Regardless of urea presence， NiOOH was identified as the primary catalytically active center， with its formation accompanied by intensity reduction of the urea C-N stretching peak （1004 cm^-1）， indicating urea oxidation. Further analysis revealed that Mn doping significantly lowered the energy barrier for the Ni²⁺→Ni³⁺ transition by elongating Ni-O bonds and increasing NiOOH lattice disorder， thereby activating urea electrooxidation at lower potentials. Unlike conventional XRD and Raman techniques that only provide limited static information before and after reactions， in situ XRD and Raman enable dynamic monitoring of structural evolution under actual working conditions. In situ XRD allows real-time tracking of crystal structure and phase composition changes， making it suitable for identifying new phase formation and transformation processes. In contrast， in situ Raman is more sensitive to local chemical bonds and molecular vibrations， capable of capturing short-lived intermediates and bond transformation processes. Together， these in situ techniques have become indispensable tools for investigating phase transition mechanisms in electrocatalytic reactions， providing fundamental insights for the precise characterization and rational design of complex reaction systems.

The dynamic evolution of active element valence states under working conditions holds significant importance for gaining fundamental insights into catalytic reaction mechanisms. In situ XAS has emerged as a powerful tool for investigating valence state variations during electrocatalytic processes， owing to its element-specific nature and high sensitivity to oxidation state changes. The valence state evolution of metal centers in catalysts typically exhibits strong correlations with their catalytic activity， selectivity， and stability. Duan et al.^[70] in situ constructed Mn-doped NiS₂ precursors on nickel foam substrates and systematically investigated the valence state evolution and coordination structure reconstruction of metal centers during electrochemical activation using multiple in situ spectroscopic techniques. In situ XAS results demonstrated that during potential application （OCP → 1.45 V vs. RHE）， the Ni K-edge XANES absorption edge progressively shifted to higher energies， indicating the oxidation of Ni from +2 to >+3. Concurrent EXAFS analysis revealed continuous shortening of Ni-S bond lengths （~2 Å） and their transformation into Ni-O coordination， suggesting stepwise surface oxidation of the sulfide to Ni（OH）₂ and NiOOH （Fig. 4a）. Simultaneously， Mn K-edge data showed that Mn consistently maintained high valence states between +3 and +4， with its coordination environment gradually evolving from Mn-S to Mn-O. This process was accompanied by significant Ni→Mn electron transfer， effectively stabilizing the high-valence Ni³⁺ active sites （Fig. 4b）. Furthermore， in situ ATR-SEIRAS （attenuated total reflection surface-enhanced infrared absorption spectroscopy） detected marked differences in the characteristic CO₃^2- peak at 1361 cm^-1： while Mn-NiS₂ surfaces showed nearly no signal （Fig.4d）， pure NiS₂ materials exhibited strong absorption （Fig.4c）， confirming that Mn doping significantly suppresses strong CO₂ intermediate adsorption and accelerates its desorption process. Collectively， the dynamic tracking of metal valence states and coordination environments through in situ techniques not only revealed the synergistic regulatory role of Mn during electrochemical activation but also provided direct electronic structure evidence for the material's exceptional CO₂ management capability and 200-hour stability without performance decay in the UOR. The comprehensive characterization approach combining multiple in situ spectroscopies offers profound insights into the structure-activity relationship of transition metal-based catalysts.

Building upon the strategy of modulating Ni’s electronic structure through heteroatom incorporation， researchers have also explored reconstructing highly active Ni species from waste materials， developing an approach that combines sustainability with high performance^[71]. Beyond elemental doping， manipulating oxidative environments has emerged as an effective method for electronic structure regulation， capable of inducing the dynamic transformation of Ni²⁺ into Ni（OH）₂/NiOOH and thereby stabilizing high-valence active sites to enhance catalytic performance. Recently， Alex et al.^[72] utilized spent Ni@CNTs as precursors and constructed the nickel oxyhydroxide layer through NaOCl oxidation treatment， synthesizing Ni（OH）₂-Ni@CNT composites. Using in situ Ni K-edge XAS to monitor valence state evolution during urea oxidation， they observed that at the working potential of 0.5 V vs. Hg/HgO for UOR， the Ni absorption edge energy blue-shifted from 8338.2 to 8340.2 eV， with the average oxidation state increasing from ~0.67 to ~1.41， indicating progressive oxidation of Ni²⁺ to form Ni（OH）₂/NiOOH （Fig.4e-g）. Complementary EXAFS analysis revealed significant weakening of Ni-Ni coordination alongside enhanced Ni-O bonding， quantitatively demonstrating an increase in Ni（OH）₂ content from 26% to 30% （Fig.4h）. This dynamic evolution of valence states and coordination environments confirms that the in situ cyclic generation mechanism of Ni（OH）₂/NiOOH serves as the key factor enabling the prolonged stability （60 h） of catalyst in highly alkaline media （Fig.4i）.

The dynamic evolution of catalyst electronic structure during the UOR is crucial for revealing the nature of active sites and elucidating catalytic mechanisms. Therefore， real-time monitoring of valence state changes， local coordination environments， and electron density evolution at metal centers holds significant importance for understanding reaction pathways and optimizing catalytic activity^[73]. Ji et al.^[74] constructed a locally hypermetastable CoOOH（Mo）/NiOOH structure through hydrothermal-electrodeposition and electrochemical reconstruction， where MoO₄^2- anchored via Co-O-Mo bonds facilitates electronic structure modulation. In situ Ni and Co K-edge XANES studies in 1 M KOH and 0.33 M urea demonstrated that as the potential increased from 1.2 V to 1.8 V （Fig. 5a， b）， the Ni absorption edge first exhibited a positive shift （corresponding to Ni²⁺→Ni³⁺ oxidation） followed by rapid regression， indicating only the transient existence of high-valent Ni species. In contrast， the Co absorption edge showed continuous positive shifts， reflecting its stable oxidation from Co²⁺ to Co³⁺. Complementary EXAFS analysis further revealed that the Ni-O bond length initially contracted from 2.02 Å to 1.88 Å before expanding again （Fig. 5e）， suggesting reversible evolution of its local coordination environment during the reaction. Meanwhile， the Co-O bond length shortened from 1.96 Å to 1.87 Å and remained stable， confirming Co’s persistent high-valence state. These results demonstrate that Ni’s 3d electrons undergo charge transfer to urea and its intermediates during the reaction， leading to temporary valence elevation followed by reduction， whereas Co’s 3d orbitals exhibit stronger high-valence stability through participation in intermediate oxidation. Moreover， Mo provides electron density via Co-O-Mo covalent bridges， cooperatively regulating the occupation states of Ni/Co 3d orbitals and driving their dynamic electronic evolution. In situ Raman spectroscopy provided critical corroborating evidence （Fig. 5c）： At 1.2 V， the characteristic Ni³⁺ peak （~475 cm^-1） showed a negative shift to 458 cm^-1， corresponding to NiOOH → β-Ni（OH）₂ transformation and revealing Ni 3d electron delocalization. Simultaneously， Co-related vibration peaks shifted from 505 to 575 cm^-1， indicating e_g orbital electron redistribution and sustained high oxidation states. This combination of transient electronic state evolution and stable maintenance constitutes dynamic charge transfer during the reaction， with trends fully consistent with XAS results， collectively demonstrating the pivotal role of Co-O-Mo bridges in electronic structure regulation. Further DFT calculations revealed that Mo incorporation through Co-O-Mo bridges elevates the d-band center of Co sites by approximately 0.12 eV toward the Fermi level， enhancing CO intermediate adsorption （Fig. 5d）. This reduces the free energy of the rate-determining step （second N-H cleavage） from 0.93 to 0.47 eV and lowers the subsequent CO → CO₂ conversion barrier from 1.20 to 1.08 eV. The valence state elevation and coordination bond length changes revealed by in situ characterization show excellent agreement with theoretical predictions， validating that precise electronic structure control through locally hypermetastable structures constitutes the key mechanism for achieving superior UOR activity.

Kim et al.^[75] employed Ni₃Fe alloy foam as the starting material and developed an in-situ synthesis of NiFe-oxalate metal-organic framework （O-NFF） through a one-step oxalic acid thermochemical treatment. The authors systematically monitored the electronic structure evolution of Ni during UOR by conducting in situ Ni K-edge XANES measurements with and without urea （Fig. 5f， g）. The results demonstrated that under urea-free conditions， as the potential increased from 1.45 to 1.60 V， the Ni absorption edge continuously shifted to higher energies， reflecting gradual delocalization of Ni 3d electrons accompanied by progressive accumulation of high-valence states （Ni²⁺ → Ni³⁺ → Ni^（3+δ）+）. In contrast， with urea present， the absorption edge remained nearly stationary （ΔE ≈ 0） within the 1.45-1.60 V potential range， indicating rapid consumption of Ni³⁺ by urea and maintenance of a dynamic Ni²⁺/Ni³⁺ equilibrium， thereby effectively suppressing the competing OER and achieving nearly 100% UOR selectivity^[76]. DFT calculations further elucidated the microscopic mechanism of electronic structure modulation： On the O-NFF surface， the carbonyl group of urea forms hydrogen bonds with oxalate-O atoms， exhibiting an adsorption energy of -1.14 eV. Fe doping significantly increases the Bader charge of adjacent oxalate-O， thereby strengthening hydrogen bonding and stabilizing intermediates （Fig. 5h）. The free energy diagram reveals that Fe-induced electronic rearrangement dramatically reduces the C-N bond cleavage barrier from 1.20 eV in O-NF to 0.30 eV in O-NFF （Fig. 5i）. Density of states （DOS） analysis further indicates that Fe doping shifts Ni 3d states closer to the Fermi level， enhancing electronic coupling between Ni and oxalate-O. This charge redistribution not only weakens the C-N bond strength but also accelerates reaction kinetics， thereby significantly promoting UOR activity.

The formation， transformation， and consumption of reaction intermediates constitute critical aspects in elucidating the mechanism of UOR. Dynamic monitoring of key intermediate species during the reaction process enables a profound understanding of reaction pathways， rate-determining steps， and the specific functions of active sites^[77]. With recent advances in various in situ characterization techniques， researchers can now capture the evolution of intermediates in real time under electrochemical reaction conditions， thereby providing robust support for establishing mechanistic models.

Yu et al.^[78] synthesized self-supported Fe-Co_0.85Se/FeCo-LDH heterojunctions and conducted a comprehensive analysis of UOR intermediates using both potentiostatic in situ FTIR and time-resolved in situ FTIR. The potentiostatic in situ FTIR measurements revealed the emergence of a *CNO^- absorption peak at 2169 cm^-1 when the potential exceeded 1.40 V vs. RHE， with its intensity increasing at higher potentials， indicating progressive accumulation of this intermediate under elevated potentials （Fig. 6a）. Concurrently， negative absorption signals appeared at 1680 cm^-1 （C $\stackrel{\mathrm{ }\mathrm{ }\mathrm{ }\mathrm{ }}{=}$O） and 1610 cm^-1 （N—H）， demonstrating continuous urea consumption， while the relatively weak *CO₃^2- signal at 1385 cm^-1 suggested incomplete oxidation to CO₂， with *CNO^- being the dominant product， which is indicative of a partial oxidation pathway. Time-resolved in situ FTIR further demonstrated that the *CNO^- peak intensity progressively increased from 5 to 50 s of electrolysis under constant potential at 1.45 V， accompanied by gradual attenuation of urea characteristic peaks， directly confirming the correlation between dynamic *CNO^- accumulation and urea consumption （Fig. 6b）. Based on the characteristic peak positions and their temporal evolution， the authors proposed a partial oxidation pathway on this heterostructure （Fig. 6c）： urea → CO（NH₂）₂ → NCO + NH → *CNO^- + NH₃， bypassing complete oxidation to CO₂ and thereby reducing the reaction potential. Furthermore， charge redistribution induced by Co-Se and O-Fe bonds at the heterointerface helped stabilize key intermediates like NCO， lowering the energy barrier for bond cleavage and promoting efficient UOR progression at lower overpotentials.

Beyond the remarkable capability of in situ FTIR in capturing key intermediates during the UOR， other in situ characterization techniques have equally contributed significant insights into elucidating the reaction mechanisms. Zheng et al.^[79] conducted an in-depth investigation of NiCo-LDH in alkaline UOR. Through in situ Raman spectroscopy， they monitored the structural evolution of these materials in both urea-containing and urea-free alkaline electrolytes， revealing dynamic changes in their active phases. The results demonstrated distinct behavioral patterns： In urea-free systems， the characteristic NiOOH peaks （470/553 cm^-1） intensified progressively with increasing potential， while Ni（OH）₂ peaks （502/526 cm^-1） simultaneously weakened and eventually disappeared completely （Fig.6d）. The spectral features showed reversible recovery during backward scans， indicating nearly complete and reversible conversion between Ni（OH）₂ and NiOOH in pure alkaline solutions. Conversely， NiOOH peaks exhibited only marginal enhancement at high potentials without becoming dominant in urea-containing electrolytes， with Ni（OH）₂ peaks persisting throughout the entire potential window （Fig.6e）. These observations suggest strong competition between UOR and the Ni（OH）₂ → NiOOH oxidation process， with UOR being preferentially favored， thereby limiting the generation of active NiOOH species. Complementary DFT calculations revealed that NiOOH， CoOOH， and NiCo-OOH surfaces all follow a four-step deprotonation pathway （Fig. 6f）： *CO（NH₂）₂ → *CO（N）（NH） → *CO（N）₂ → N₂ + CO₂， where removal of the fourth hydrogen atom serves as the RDS. The UOR-RDS energy barrier on NiCo-OOH （1.60 eV） was lower than the oxidation barrier for Ni（OH）₂ → NiOOH conversion （1.66 eV）， indicating that the oxidation process， rather than UOR itself， constitutes the limiting factor. Furthermore， Co doping reduced the oxidation barrier from 1.71 eV to 1.66 eV， facilitating NiCo-OOH formation and consequently enhancing overall reaction activity. Combining in situ Raman observations with DFT computations， they proposed a comprehensive reaction mechanism for Ni-based hydroxides in alkaline UOR （Fig. 6g， h）： At low potentials， the hydroxide structure remains stable. As the potential approaches the redox threshold， partial hydroxide conversion to active oxyhydroxides occurs， which catalyzes urea deprotonation while rapidly consuming interfacial OH^- species. The limited OH^- migration in electrolytes creates localized concentration depletion when UOR proceeds rapidly， thereby suppressing further hydroxide-to-oxyhydroxide conversion that equally depends on OH^- availability. This competitive relationship enables sustained high UOR current densities with minimal oxyhydroxide presence. In contrast， hydroxide-to-oxyhydroxide conversion occurs extensively only at higher potential in urea-free systems due to the higher OER energy barrier. This mechanistic framework elucidates the dynamic transformations of Ni-based materials during UOR and their regulatory effects on catalytic activity.

While the previous section discussed the dynamic evolution of intermediates during the UOR， a comprehensive understanding of the reaction mechanism requires in-depth analysis of final product formation and distribution. The types and quantities of products not only determine reaction selectivity and pathway branching but also reveal the existence of side reactions or incomplete oxidation^[80]. Conventional detection methods often fail to reflect product states under actual reaction conditions， whereas in situ characterization techniques enable real-time monitoring during electrocatalysis， directly capturing the evolution of gaseous and liquid products to provide critical evidence for mechanistic elucidation.

Han et al.^[81] prepared well-crystallized LiNiO₂ through the sol-gel synthesis followed by high-temperature annealing， subsequently obtaining catalysts LNO-x with varying degrees of delithiation via chemical delithiation. In situ ATR-FTIR played a pivotal role in clarifying UOR product formation and reaction mechanisms. In 1 M KOH and 0.5 M urea， full-spectrum measurements at 1.45 V vs. RHE captured the real-time emergence of characteristic peaks at 1430 cm^-1 （CO₃^2-） and 2390 cm^-1 （CO₂）， with intensities progressively increasing over time， directly demonstrating urea oxidation to soluble carbonates and gaseous CO₂ （Fig. 7a）. Magnified spectral regions further revealed that LNO-2 exhibited gradually intensifying signals at 2925 cm^-1 （N-H） and 1203 cm^-1 （*OCONNH） between 1.35~1.45 V （Fig.7b）， uncovering dynamic accumulation of key intermediates along the lattice oxygen-mediated （LOM） pathway. Theoretical calculations and in situ characterization complement each other in UOR mechanistic studies， particularly for verifying product formation pathways. While in situ FTIR dynamically tracks CO₂ and N₂ evolution signals， DFT analysis reveals the fundamental reasons behind their formation. Comparative analysis of the adsorbate evolution mechanism and LOM pathways shows that direct lattice oxygen participation in C-N bond cleavage within LOM enables early CO₂ desorption， avoiding accumulation of high-energy intermediates， a finding corroborated by experimental FTIR observations showing the absence of the characteristic C $\stackrel{\mathrm{ }\mathrm{ }\mathrm{ }\mathrm{ }}{=}$O band at 1720 cm^-1 （Fig. 7c）. Free energy profiles further demonstrate that LNO-2 exhibits remarkably low energy barriers for N₂ desorption （0.06 eV） and lattice oxygen regeneration （0.34 eV） along the LOM pathway， substantially lower than those of LNO-0 （Fig.7d）. These findings not only validate the kinetic advantages of LOM but also rationally explain the potential-dependent intensification of *OCONNH and N-H signals observed in FTIR spectra.

Building upon the insights into intermediates and solution-phase products provided by ATR-FTIR， in situ DEMS serves as a complementary technique that enables time-resolved detection and quantitative analysis of gaseous products， thereby offering critical evidence for elucidating UOR reaction mechanisms. Song et al.^[82] employed a two-step MOF-sulfidation in situ interfacial redox strategy to construct NiS₂@MnO_x heterojunction catalysts. In situ DEMS experiments revealed the reaction pathway and exceptional selectivity of the NiS₂@MnO_x in UOR （Fig. 7e， f）， providing experimental support for the proposed mechanism. Isotope labeling experiments using electrolytes containing ¹⁴N-urea and ¹⁵N-urea （4∶1 molar ratio） detected only ¹⁴N₂ （m/z = 28） and ¹⁵N₂ （m/z = 30） signals， with no observable ¹⁴N¹⁵N （m/z = 29）. This unambiguously demonstrated that N₂ generation occurs exclusively through intramolecular N-N coupling within individual urea molecules rather than intermolecular nitrogen recombination. Furthermore， DEMS highlighted catalytic differences： at 2.0 V vs. RHE， NiS₂@MnO_x produced negligible O₂ （m/z = 32）， indicating near 100% UOR selectivity， whereas pure NiS₂ under identical conditions exhibited significant O₂ signals， revealing competition with the OER. Based on DEMS results， they proposed that urea molecules initially coordinate strongly via N atoms to Mn（Ⅳ） sites， with nitrogen lone-pair electrons filling Mn's low-lying e_g orbitals， reducing Mn（Ⅳ） to Mn（Ⅱ） and triggering dehydrogenation. Subsequent intramolecular N-N coupling generates N₂ while C-N bond cleavage produces CO species， which OH^- oxidizes to CO₂/CO₃^2- （Fig. 5g）. Notably， Ni primarily functions as an “electron pump” in this process： Mn（Ⅱ） transfers electrons to adjacent Ni（Ⅲ）， which relays them externally before being re-oxidized to Ni（Ⅲ） under applied potential， enabling Mn/Ni cycling. This mechanism not only explains the high selectivity and N₂ generation observed by DEMS but also reveals the synergistic roles of Mn and Ni sites.

In the UOR process， the interfacial chemisorption of electrolyte anions significantly influences the electronic structure of active sites and the reaction pathway， even determining catalytic selectivity. Since this adsorption behavior exhibits strong potential dependence， conventional methods struggle to characterize it effectively. Therefore， in situ techniques are required for real-time monitoring to elucidate the role of anions in interfacial reaction mechanisms. Liu et al.^[83] employed a methanol-water mixed system and introduced 2-methylimidazole to achieve trace Co doping and synergistic modulation of defect structures on the surface of α-Ni（OH）₂ through its selective coordination-etching effect on Co²⁺， constructing WM-Ni_0.99Co_0.01（OH）₂ nanosheets （Fig.8a）. To clarify the performance enhancement mechanism， the authors further conducted a systematic analysis of interfacial adsorption behavior using in situ EIS. The Nyquist plot exhibited purely capacitive characteristics in the low-potential region （<1.30 V）， with no charge transfer kinetics observed， indicating that this stage primarily involves reversible adsorption/desorption of OH^- at surface defect sites （Fig. 8b， c）. Near the critical potential （~1.35 V）， the Nyquist plot of the WM-Ni_0.99Co_0.01（OH）₂ displayed a distinct semicircle for the first time， with a charge transfer resistance （R_ct） of 124 Ω. In contrast， the unmodified α-Ni（OH）₂ required a higher potential （≥1.40 V） to exhibit a semicircle， with a larger R_ct of 145 Ω， reflecting that the defect structure induced by Co doping significantly reduced the energy barrier for OH^- adsorption-desorption. The Bode phase angle results further corroborated this inference （Fig. 8d， e）： in the 1.30~1.35 V range， the phase angle of the modified sample rapidly decreased by approximately 7° and stabilized at 47°， indicating that OH^- could rapidly complete adsorption and deprotonation conversion at Ni³⁺/oxygen vacancy sites， promoting the formation of NiOOH. In comparison， the unmodified sample exhibited a continuous high-frequency drift in the phase angle， revealing significantly sluggish reaction kinetics. Based on these findings， the researchers proposed a UOR reaction mechanism： at the high-density Ni³⁺ defect sites induced by Co doping， OH^- first undergoes chemisorption accompanied by electron transfer from Ni²⁺ to Ni³⁺， completing deprotonation to form active NiOOH at a lower potential. Subsequently， NiOOH reacts with urea to generate N₂ and CO₂， while surface defect sites continuously participate in the catalytic cycle by capturing OH^-， undergoing deprotonation， and regenerating active sites. This recurring sequence effectively lowers the reaction overpotential and enhances catalytic stability.

While in situ characterization techniques can provide valuable information regarding structural evolution， valence state changes， and intermediate species during the UOR， they still exhibit certain limitations in spatial resolution and mechanistic interpretation. To address these shortcomings， recent advances combining experimental approaches with theoretical calculations have offered significant breakthroughs in elucidating UOR mechanisms. By employing in situ spectroscopic techniques to capture reaction intermediates in real time and complementing them with DFT calculations for energetic analysis of reaction pathways and active sites， researchers have achieved atomic-level understanding of the catalytic nature of UOR.

Liu et al.^[84] developed an amorphous Mo-Ni trithiocyanurate coordination polymer through the solvothermal method using the ligand of trithiocyanuric acid， Mo precursors， and NiCl₂ under elevated temperatures， formed a stable self-supporting architecture （Fig.9a）. To further investigate the reaction mechanism， they conducted potential-dependent monitoring via in situ Raman spectroscopy （Fig.9b）. The results revealed that starting from 1.32 V， only the vibrational peak associated with S-Ni³⁺-OH （475 cm^-1） was observed， while the characteristic NiOOH peaks （475/558 cm^-1） were absent， indicating no structural reconstruction from Ni²⁺ to Ni³⁺OOH occurred in the system. Concurrently， the characteristic urea adsorption peak （1002 cm^-1） gradually diminished with increasing potential， suggesting that Ni³⁺-OH could directly function as an electron acceptor for urea oxidation without requiring NiOOH as an intermediate state， thereby implying a “reconstruction-free” catalytic pathway. DFT calculations further corroborated these findings， demonstrating that urea preferentially adsorbs on Ni³⁺-OH sites with an adsorption energy of -1.21 eV and a Ni-O bond length of approximately 1.95 Å （Fig. 9c， d）. The potential-determining step （CON₂H₄ → CON₂H） exhibited a relatively low energy barrier （0.52 eV）， and the overall reaction pathway was strongly exothermic （releasing approximately 3.90 eV）， both thermodynamically and kinetically confirming that NiOOH is not an essential active intermediate. These results validate the possibility of Ni³⁺-OH serving as a structurally stable active center. Through synergistic analysis of in situ Raman spectroscopy and DFT calculations， the researchers established a comprehensive UOR reaction pathway model： In alkaline media， the carbonyl oxygen of urea preferentially chemisorbs on Ni³⁺-OH sites， with N/S ligands facilitating proton transfer. Ni³⁺ continuously extracts electrons from urea， guiding the molecule through a four-step dehydrogenation process to form the *CON₂ intermediate， followed by intramolecular N-N coupling， ultimately releasing N₂ and CO₂. Throughout this process， no NiOOH phase transformation occurs， and the catalyst maintains structural integrity， demonstrating a highly active and stable “reconstruction-free” mechanism.

It is important to emphasize that the proposed mechanism presents certain discrepancies with the currently widely accepted UOR paradigm. In most Ni-based systems， Ni typically undergoes oxidation to form NiOOH under alkaline conditions， and NiOOH is considered the primary contributor to catalytic activity， a process that has been repeatedly verified by various in situ characterizations. Although the stable Ni³⁺-OH configuration may be valid under specific ligand-modulated conditions， its general applicability still requires systematic validation across different electrolyte compositions， potential windows， and catalyst loading configurations. This is particularly important considering that characteristic NiOOH signals in in situ Raman spectroscopy might be obscured by interference from ligands or electrolytes.

The UOR， involving multiple electron transfer steps and dynamic evolution of intermediates， urgently requires in situ characterization techniques for real-time monitoring. In recent years， the combined application of multiple in situ characterization methods has become crucial for elucidating the UOR reaction mechanism. Using β-Ni（OH）₂ as a model electrode， Chen et al.^[85] systematically integrated in situ Raman spectroscopy， quasi-in situ XPS， DEMS， synchrotron radiation FTIR， and EIS to propose a “low-oxidation-state cycle” based UOR pathway. In situ Raman spectroscopy revealed that in alkaline conditions， β-Ni（OH）₂ sequentially transforms into β-Ni（OH）O and NiOOH with increasing potential （Fig.10a）. However， in the presence of urea， the characteristic Ni³⁺-O peaks （476 and 558 cm^-1） were significantly suppressed， indicating the reaction primarily cycles between β-Ni（OH）₂ and β-Ni（OH）O phases without forming NiOOH. Quasi-in situ XPS analysis further confirmed that after urea addition （Fig. 10b）， the Ni 2p and O 1s energy levels remained essentially unchanged， with no signals of high-valent Ni^3+δO_xH_y observed， demonstrating the electrode maintained a lower oxidation state throughout. In situ DEMS combined with ¹⁵N isotope experiments detected m/z = 30 （¹⁵N₂） signals in the evolved gases， confirming both nitrogen atoms originated from the same urea molecule and excluding intermolecular coupling pathways， thereby validating an “intramolecular N-N coupling” mechanism （Fig. 10c）. SR-FTIR showed the reaction initiates with N-H dehydrogenation of the amino group， followed by C-N bond cleavage， while the C $\stackrel{\mathrm{ }\mathrm{ }\mathrm{ }\mathrm{ }}{=}$O group remained stable throughout the process， with N₂ and CO₂ as final products （Fig. 10d）. EIS measurements revealed β-Ni（OH）O as the primary active interface for UOR， exhibiting faster electron transfer rates than the Ni³⁺^δO_xH_y pathway in OER， along with reduced interface passivation， resulting in lower overpotential and higher selectivity （Fig. 10e）. Based on these findings， the authors proposed a novel reaction pathway （Fig. 10f）： β-Ni（OH）₂ is electrochemically activated to form β-Ni（OH）O， which then reacts with urea through alternating PCET processes， undergoing dehydrogenation and rearrangement to produce N₂ and CO_₂ while regenerating β-Ni（OH）₂， establishing a stable low-oxidation-state catalytic cycle. This mechanism achieves nearly 100% N₂ selectivity while avoiding accumulation of high-valent Ni species， demonstrating exceptional catalytic performance. Through multi-technique collaborative validation， this study systematically presents the “low-oxidation-state cycle” model for the first time， providing not only new perspectives for understanding Ni-based electrocatalysts but also important theoretical guidance for designing efficient UOR catalysts.

Notably， this “low-oxidation-state cycle” pathway challenges the conventional UOR mechanism centered on high-valent Ni species， emphasizing the reversible regulation capability of Ni²⁺ and its local coordination environment. This work establishes a new interpretative framework for re-examining the true active centers in Ni-based catalysts.