Enhanced radical formation and water purification via lanthanide-nitrogen co-doped strontium titanate

Research article Enhanced radical formation and water purification via lanthanide-nitrogen co-doped strontium titanate L. Sarasino a,b , V.M. Calavita a , V. Lagostina a , P. Calza a , M.C. Paganini a,* a Department of Chemistry, University of Torino, Via Giuria 7, Turin 10125, Italy b Department of Chemistry, Biology and Biotechnology, University of Perugia, Via Elce di Sotto 8, Perugia 06123, Italy ARTICLE INFO Keywords: Photocatalysis Strontium titanate Lanthanide-nitrogen co-doping Pollutants degradation EPR ABSTRACT Traditional water treatment methods including physical, chemical, and biological approaches often fall short of achieving the required purification standards. Photocatalysis has emerged as a promising alternative due to its efficiency, sustainability, and cost-effectiveness. Among the various photocatalysts investigated, strontium titanate (SrTiO 3 ) has attracted considerable attention for its high stability and photocatalytic potential. However, its wide bandgap (3.2 eV) limits its activity to the ultraviolet region, thereby restricting its ability to utilize the broader solar spectrum. To address this limitation, doping strategies have been employed to extend light absorption into the visible range. In particular, co-doping with lanthanide and nitrogen ions has demonstrated the potential to enhance photocatalytic performance by altering the material’s electronic structure and suppressing charge carrier recombination. This study evaluates the photocatalytic activity of SrTiO 3 co-doped with various lanthanide and nitrogen ions, offering a comparative analysis across the lanthanide series. Among the tested materials, the La/N co-doped SrTiO 3 exhibited the highest efficiency in the degradation of phenol, which was used as a model pollutant. The degradation mechanism and the reactive species involved were investigated using electron paramagnetic resonance (EPR) spectroscopy combined with the spin-trapping technique. The analysis of photogenerated species in solution revealed that photogenerated holes, in synergy with superoxide radicals, play the most significant role in phenol degradation. 1. Introduction Water is indispensable for life and since few years photocatalysis has emerged as a promising solution for water purification.[1] This technique is lauded for its effectiveness, simplicity, eco-friendliness, and cost-effectiveness. Among the organic pollutants, pesticides, dyes, pharmaceutical compounds, phenolic compounds, aldehydes, and ketones are of particular concern. Notably, phenol and its derivatives are prevalent water pollutants recognized for their carcinogenic properties, persistence, and toxicity, even at minimal concentrations. These compounds typically originate from agricultural practices and industrial discharges associated with the production of resins, oils, plastics, textiles, and pharmaceuticals.[2–5] Recent advancements in photocatalysis have demonstrated significant success in the removal of contaminants from water, particularly through advanced (photo)-oxidation processes (AOPs) that generate essential oxidizing species capable of degrading these pollutants. Pollutants adsorbed onto the surface of the photocatalyst undergo degradation and oxidation, ultimately transforming into carbon dioxide CO 2 and H 2 O.[1] Strontium titanate (SrTiO 3 , STO), characterized by its perovskite structure, has garnered considerable attention for its efficacy in degrading water pollutants and its diverse photocatalytic applications. Its advantages include high efficiency, photo-stability, and easily adjustable properties.[6,7] However, a significant limitation of STO is its wide bandgap of 3.2 eV, which confines its practical applications to the ultraviolet (UV) spectrum, accounting for only 1–3 % of solar radiation.[8–10] To enhance the utilization of solar energy, extensive research efforts have been directed towards extending the absorbance edge of photocatalysts into the visible light spectrum.[11] Among the various strategies employed to achieve this goal, doping has emerged as one of the most effective methods. By introducing foreign elements into active photocatalysts with wide bandgap, it is possible to create donor or acceptor levels within the forbidden band, * Corresponding author. E-mail address: [email protected] (M.C. Paganini). Contents lists available at ScienceDirect Next Materials journal homepage: www.sciencedirect.com/journal/next-materials https://doi.org/10.1016/j.nxmate.2025.101041 Received 30 May 2025; Received in revised form 1 August 2025; Accepted 4 August 2025 Next Materials 9 (2025) 101041 Available online 14 August 2025 2949-8228/© 2025 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license ( http://creativecommons.org/licenses/by/4.0/ ). thereby broadening the light absorption capabilities of these semiconductors.[12,13] Selective substitutional doping at specific sites within the ABO 3 perovskite structure allows for the tuning of the doping effect by selecting ions with radii similar to those being replaced. However, it is important to note that the substituting ions typically possess different stable oxidation states than the ions they replace, leading to charge imbalances and the formation of bulk defects. These defects can serve as centers for electron-hole recombination, ultimately limiting the photocatalytic activity of the material.[12,14,15] To address the challenges associated with charge imbalance and defect formation, co-doping has emerged as a promising strategy.[16] This approach involves the introduction of two dopant species with oxidation states that compensate for each other’s charges, thereby mitigating the formation of defects. In particular, strontium titanate has shown enhanced photocatalytic activity when co-doped with lanthanide ions, which substitute for the Asite strontium ions (Sr 2 + ). This enhancement is attributed to the unique 4f-4f transitions associated with lanthanide ions. Additionally, nitrogen ion doping, which can substitute for oxygen ions (O 2- ) or be added interstitially, is known to introduce intra-bandgap states, similar to the effects observed in nitrogen-doped titania.[17–33] This study aims to investigate the photocatalytic activity of strontium titanate co-doped with lanthanide and nitrogen ions, comparing it to its mono-doped counterparts across the lanthanide series.[34–43] The co-doping of strontium titanate with lanthanide and nitrogen ions presents a promising avenue for enhancing its photocatalytic activity, this research seeks to fill the existing knowledge gap by providing a comparative analysis of the effects of various lanthanide ions in co-doping strategies. Among the various samples prepared and analyzed, the co-doped system with lanthanum and nitrogen exhibited the best performance in phenol degradation. The reaction mechanism was investigated using the spin-trapping technique in combination with the electron paramagnetic resonance (EPR) spectroscopy. This analytical approach enabled the identification of the reactive species generated during light irradiation of the sample, distinguishing those primarily responsible for phenol degradation from those that limit its effectiveness. Furthermore, solid-state EPR spectroscopy was employed to study the role of nitrogen, whether interstitial or substitutional, in facilitating electron transfer and suppressing electron–hole recombination. 2. Materials and methods Titanium (IV) isopropoxide (TTIP, Ti(OiPr) 4 , 97 % in iPrOH, SigmaAldrich), Strontium nitrate (Sr(NO 3 ) 2 , 99 %, Sigma-Aldrich), Ethylene Glycol (EG, C 2 H 6 O 2 , 99.8 %, Sigma-Aldrich), Hydrochloric acid (HCl, 37 %, VWR Chemicals (BDH)), Sodium Hydroxide (NaOH, 98 %, SigmaAldrich) Hexamethylenetetramine (HMT, C 6 H 12 N 4 , 99 %, SigmaAldrich), Lanthanum (III) nitrate hexahydrate (La(NO 3 ) 3 •6 H 2 O, 99.9 %, Sigma-Aldrich), Cerium (III) nitrate hexahydrate (Ce (NO 3 ) 3 •6 H 2 O, 99.9 %, Sigma-Aldrich), Praseodymium (III) nitrate hexahydrate (Pr(NO 3 ) 3 •6 H 2 O, 99.9 %, Sigma-Aldrich), Erbium (III) nitrate pentahydrate (Er(NO 3 ) 3 •5 H 2 O, 99.9 %, Sigma-Aldrich), Ytterbium (III) nitrate hexahydrate (Yb(NO 3 ) 3 •6 H 2 O, 99.9 %, SigmaAldrich), Lanthanum oxide (La 2 O 3 , 99.9 %, Sigma-Aldrich), Phenol (C 6 H 5 OH, ≥99.9 %, VWR Chemicals (BDH)), 5,5-Dimethyl-1-pyrroline N-oxide (DMPO, Sigma Aldrich), 2,2,6,6-Tetramethyl-4-piperidone (TEMP-H, Sigma-Aldrich), Sodium formate (Sigma-Aldrich) were purchased and used directly without any further purification 2.1. Synthesis of samples STO lanthanide/nitrogen co-doped (STO_Ln/N) samples were synthesized slightly modifying a previously reported procedure.[42] In 30 mL of EG at room temperature were consequently added 0.74 mL of TTIP (2.5 mmol), 5 mL of Sr(NO 3 ) 2 0.5 M, 0.025 mmol of the desired Lanthanide (1 %) precursor and 3.5 g of HMT (25 mmol). Finally, after complete dissolution of the previously added reagents, 2.5 mL of NaOH 5 M are added dropwise. The solution is stirred for 1 h at room temperature and then transferred and sealed in an autoclave, the reaction is carried out at 180◦C for 24 h. The sample is collected by centrifuge and washed with HCl 0.1 M, water and ethanol, then dried overnight at 70◦C and finally calcined at 400◦C for 2 h. For comparison samples mono-doped with lanthanides (STO_Ln), mono-doped with nitrogen (STO_N) and pure (STO) samples were prepared, with the same procedure without adding HMT, without adding any lanthanide precursor or without either, respectively. 2.2. Materials characterization X-Ray Diffraction (XRD) patterns were recorded with a PANalytical PW3040/60 X ′ Pert PRO MPD using a copper Ka radiation source (0.15418 nm) and a Bragg Brentano geometry. X ′ Pert High-Score software was used for data handling. Rietveld refinement was performed on the diffraction patterns to determine the crystallite size. Diffuse Reflectance Spectroscopy (DRS) spectra were recorded using a Varian Cary 5000 spectrometer, coupled with an integration sphere for diffuse reflectance studies, using a Carywin-UV/scan software. A sample of PTFE with 100 % reflectance was used as reference. The morphology of the particles was investigated with the high-resolution (field emission source) scanning electron microscope (FESEM, Zeiss Merlin), coupled with secondary electron and energy dispersive x-ray (EDX) detectors. Electron Paramagnetic Resonance (EPR) spectra were recorded with Xband CW-EPR Bruker EMX spectrometer equipped with a cylindrical cavity operating at 100 kHz field modulation. The measurements were performed at room temperature (RT) in liquid nitrogen (77 K) with the following experimental parameters: modulation amplitude 1 mT and attenuation 33 dB (0.1 mW) or 43 dB (0.01 mW). The effect of light on EPR spectra was investigated using a 500 W mercury/xenon lamp (Oriel Instruments) equipped with an infrared (IR) water filter. Spin-trapping experiments were carried out using a X-band benchtop EPR spectrometer (ADANI’s SPINSCAN X). Two different spin trapping molecules were used: DMPO was employed for hydroxyl radical, superoxide and, together with formate, for holes detection while TEMP-H for singlet oxygen detection. Ex situ irradiation experiments to monitor the formation of the radicals were performed using a LED with λ =365 nm. The samples were prepared suspending 10 mg of photocatalyst in i) 0.5 mL of 4-oxo-TEMP-H aqueous solution (0.044 M) for singlet oxygen detection; ii) 0.5 mL of DMPO aqueous solution 0.044 M for the hydroxyl radical detection; iii) 0.5 mL of DMPO solution (0.044 M) in acetonitrile instead of water to avoid competition with hydroxyl radicals for the superoxide species; iv) 0.5 mL of aqueous solution buffered at basic pH containing DMPO 0.044 M and sodium formate (0.5 M) for holes detection.[44,45] To test species reactivity toward degradation, in separated experiments, phenol (0.5 M) was added to i and iii. For scavenger tests isopropanol 10 % was added to iii as hole scavenger. 2.3. Photocatalytic activity The photocatalytic activity of the samples was determined by evaluating the photodegradation of phenol (5 mg/L), using 500 mg/L of the photocatalyst. The irradiation experiments were run on 5 mL of sample and catalyst suspension put in glass cells maintained under magnetic stirring and placed in a Solarbox CO.FO:MEGRA equipped with a cut-off (≥340 nm) filter. The phenol disappearance was determined with a high-pressure liquid chromatograph (HPLC) equipped with two Merck Hitachi L6200 pumps and a Merck-Hitachi L-4200 UV–vis detector. The detection wavelength was set up at 220 nm and a 70:30 aqueous phosphate buffer L. Sarasino et al. Next Materials 9 (2025) 101041 2 (pH 2.8)/acetonitrile mixture was used as eluent; flow rate was kept at 1 mL/min. 3. Results and discussion 3.1. Crystalline structure In Fig. 1 the XRD patterns of all the prepared samples are shown. The results obtained confirm the successful synthesis of crystalline samples that can be assigned to the perovskite-type structure of STO with cubic symmetry (JCPDS no. 79–0176). The crystal structure of STO exhibits cubic symmetry with Pm-3m space group. Ti 4+ ions and Sr 2+ ions are coordinated by 6 (octahedral symmetry) and 12 (cubooctahedral symmetry) oxygen ions respectively. The structure is composed of cornersharing TiO 6 octahedra, with Sr cations occupying the central positions in a cubic array of TiO 6 octahedra.[43] Substituting a Sr 2+ cation with a Ln 3+ one having slightly different ion radius and, in principle, different charge (usually +3 for Lanthanide cations) is known to cause a local imbalance of electrostatic and steric repulsion accompanied by a distortion of the structure, namely a TiO 6 octahedra tilt.[12,43,46] A similar effect is observable with nitrogen doping, interstitial or substitutional (taking oxygen place).[12,43] To verify the successful incorporation of the lanthanide ions into the structure thus excluding the formation of a Ln 2 O 3 -STO mixture during synthesis, a mechanical mixture of pristine STO and La 2 O 3 powders (corresponding to 1 % by weight of La) was prepared and its diffraction pattern compared with that of the STO_La sample doped via chemical synthesis. The comparison, shown in Fig. S1, reveals the distinct presence of three additional reflection peaks (15.6◦, 27.3◦and 27.9◦) attributable to La 2 O 3, overlapped to the main SrTiO 3 pattern and observed in the mechanical mixture only. This demonstrates that lanthanum doping of the crystalline STO structure is effectively achieved without the appreciable formation of a heterogeneous mixture. The same approach can be extended to all the other samples. As shown by Fig. 1B, the insertion of nitrogen in the structure is not straightforward as for four out of six co-doped materials the formation of a TiO 2 phase is observed, probably for greater TiO 6 octahedra tilt. Curiously in Lanthanum-Nitrogen and Praseodymium-Nitrogen codoped samples (STO_La/N and STO_Pr/N) no TiO 2 phase is detectable. On the contrary, the Ytterbium-Nitrogen co-doped sample (STO_Yb/ N) seems to be composed only of TiO 2 phase with only traces of STO typical peaks. Moreover, lanthanide and nitrogen doping/co-doping induce a shrink of the crystallite dimensions, as calculated by Scherrer equation and in reported in Table S1. The shape-controlled hydrothermal synthesis method utilized in this study, originally developed by Zhongyu Li et al.[42], enables the production of particles with a distinctive flower-like morphology. These particles are agglomerates of nano-sheets arranged in a 3D hierarchical structure, as depicted in FESEM images (Fig. 2). Only a selection of images has been reported for sake of brevity, in particular a comparative analysis between the pure sample (STO, Fig. 2A-C) and the lanthanum/nitrogen co-doped sample (STO_La/N, Fig. 2D-F) demonstrates that the doping process does not alter the morphology of the sample. This also indicates that the additional reagents do not interfere with the shape-controlled synthesis process. These images are representative for all the samples, the shape and the morphology shown are similar in all the synthesized materials, bare, doped and co-doped. The Energy Dispersive X-ray (EDX) mapping of STO_La/N (Fig. 2G) confirms the presence of lanthanum in the sample. However, nitrogen atoms are not detected by this technique likely because of their low concentration and low density. Despite this result, the presence of nitrogen in the material has been confirmed by the (more sensitive) EPR spectroscopy (Fig. 4). 3.2. Morphology None 3.3. Optical behavior: UV-Vis absorption None The influence of Ln and N doping on the optical absorption has been investigated by UV-Vis Diffuse Reflectance Spectroscopy (DRS). In Fig. 3 the percentage of reflectance and the calculated absorption spectra are shown for all the samples. The influence of lanthanide doping on the optical absorption of the materials is noticeable, though generally modest, except in the case of the cerium-doped sample (STO_Ce), which shows a markedly enhanced ability to absorb visible light beyond 380 nm. The valence band (VB) to conduction band (CB) transition, which defines the band gap energy, is only slightly affected by doping in most cases, with the notable exception of the Ce-doped material. Theoretical studies suggest that the empty or partially filled 4 f orbitals of lanthanide ions can contribute to the conduction band minimum, together with Ti 3d orbitals, leading to a narrowing of the band gap. However, experimental results indicate that the improvement in light-harvesting capability and the associated band gap reduction are generally limited, again with the exception of cerium-doped samples, [17,19,20] which is consistent with our findings (Fig. S2 A, B; Table 1 and Table S2). In fact, with respect to other Ln doping, the 4 f 1 states of Ce 3+ are Fig. 1. XRD pattern of the samples: A) pure and Lanthanide doped, B) Nitrogen doped and Lanthanide-Nitrogen co-doped. L. Sarasino et al. Next Materials 9 (2025) 101041 3 reported to mix extremely favorably with the Ti 3d states of the conduction band, lowering the energy necessary for the electronic transition upon light absorption, producing an observed enhancement in visiblelight absorption. This light absorption enhancement can also be improved by the wellknown 4 f 1 5d 0 → 4 f 0 5d 1 internal electronic transition observed for Ce 3+ ions.[44] A similar transition is expected for ytterbium-doped samples too; however, this effect is not evident in the absorption spectra of the monodoped samples. For the samples doped with praseodymium and erbium the additional electronic transitions, particularly evident in the reflectance spectra (Fig. 3B, D), are associated with internal f-f transitions of the individual lanthanide ions.[19,25] Typically, f-f transitions are parity forbidden, according to the Laporte rule.[47] However, these transitions can gain intensity and become observable through the mixing of higher energy f orbitals with 5d/6 s orbitals or via local symmetry perturbations, known as "vibronic coupling". These transitions are not present for other Ln-doped samples, Fig. 2. FESEM pictures at different magnification degree of STO (A,B,C) and STO_La/N (D,E,F) samples. An EDX map of the STO_La/N sample is also presented (G). L. Sarasino et al. Next Materials 9 (2025) 101041 4 specifically, the La 3+ ion, with the electronic configuration of [Xe], lacks electrons in the f orbitals, precluding any f-f transitions. Similarly, Ce 3+ ([Xe]4 f 1 ) and Yb 3+ ([Xe]4f 13 ) ions do not exhibit f-f transitions due to having only a single L-value, which prevents the presence of an upper 4 f state.[17,47] On the other hand, it is possible to observe a modestly increased absorption ability by the nitrogen doped samples in the range 400–600 nm (Fig. 3C, Fig. 4). As observed in the past for nitrogen doped titania the insertion of nitrogen, as doping element, creates intra-bandgap states over the valence band maximum, thereby allowing for visible light harvesting. [29,48] Also, in the case of STO_Yb/N sample it is possible to observe a deep absorption band in the region of visible light with a quite different behavior respect to the other samples, this can be explained with the fact that in this case the principal crystallographic phase is titania and not strontium titanate. Fig. 4 compares the optical absorption of the undoped, lanthanum mono-doped, nitrogen mono-doped and lanthanumnitrogen co-doped samples, highlighting the impact of doping on the material’s light-harvesting capability. The results show that nitrogen doping is more effective than lanthanum doping alone. Moreover, the enhancement observed in the co-doped sample exceeds the combined effects of the individual dopants, indicating a synergistic interaction between lanthanum and nitrogen. 3.4. EPR investigation Electron Paramagnetic Resonance (EPR) is a valuable technique for investigating electron transfer in materials upon light irradiation and therefore highlights the behavior of the photocatalyst during the reaction photo-induced. The absorption of a photon, with an appropriate energy, by the photocatalyst results in a promotion of an electron from the valence band to the conduction band of the material. This process leads to the formation of a photoexcited electron and a photogenerated hole, the socalled electron-hole (e - -h + ) couple.[49] Usually in a metal oxide photocatalyst the photoexcited electron is associated with the reduction of the metal ion (M n +e - → M n−1 ) and the hole with the oxidation of the lattice oxygen (O 2- +h + → O - ).[50] This behavior is well highlighted in Fig. 5A, that shows the EPR spectra of the pristine sample before (dark, blue curve) and after (irradiated, red curve) irradiation at 77 K with a Xenon lamp.(Fig. 6) In dark conditions no signal was observed in the pure STO according to the absence of paramagnetic species, meanwhile the light irradiation induces the arise of two well-known signals with g value 2.014 and 1.978, assignable to the well-known species O - (photogenerated hole) and Ti 3+ (photoexcited electron, Ti 4+ +e - → Ti 3+ ) respectively.[48] In order to explore the reactivity of the surface, the sample was irradiated also in oxygen atmosphere (10 mbar) and the EPR spectrum (Fig. 5B) was recorded in dark conditions at 77 K after the reaction with oxygen and after outgassing to eliminate the excess of oxygen that could Fig. 3. DRS analysis of the samples with % of Reflectance and calculated Absorption spectra of pure and lanthanide mono-doped samples (A,B) and nitrogen doped and lanthanide-nitrogen co-doped samples (C,D). In the absorption spectra (A, C) the insertions show an enlargement of the spectra in the 350–600 nm range. Table 1 Bandgap values for all the samples calculated with Tauc-plot like method. Sample Bandgap /eV Sample Bandgap /eV STO 3.25 STO-N 3.27 STO-La 3.19 STO-La/N 3.22 STO-Ce 3.12 STO-Ce/N 3.07 STO-Pr 3.20 STO-Pr/N 3.27 STO-Er 3.25 STO-Er/N 3.29 STO-Yb 3.21 STO-Yb/N 3.14 L. Sarasino et al. Next Materials 9 (2025) 101041 5 Fig. 4. DRS analysis of the samples STO, STO_N, STO_La, STO_La/N: a comparison of absorbance behavior. Fig. 5. EPR spectra of pure STO irradiated with UV lamp in A) vacuum and B) Oxygen atmosphere and EPR spectra experimental and simulated of the sample STO_La/N. L. Sarasino et al. Next Materials 9 (2025) 101041 6 affect the results being oxygen a paramagnetic molecule. The obtained EPR signal can be assigned to the superoxide radical species in orthorhombic symmetry arising from the reduction of absorbed O 2 on the surface by photoexcited electrons; these electrons, generated by the light and promoted to the surface of the material, were transferred onto the oxygen molecule in contact with the surface, being oxygen a powerful electron-scavenger.[32] The classic CW-EPR technique has been crucial in defining the nature and peculiarities of the nitrogen-based photoactive center. It was demonstrated that, in materials synthesized through wet chemistry methods, nitrogen paramagnetic centers form within the bulk of the solid (titania or zirconium titanate). In particular, in the case of TiO 2 , one of these centers plays a crucial role in the system’s photochemistry under visible light. This previously unidentified species exhibits a simple EPR signal, indicative of a single nitrogen atom. The signal appears as a hyperfine triplet, reflecting the interaction of the unpaired electron with a single ¹⁴⁴N nucleus (nuclear spin I ¼1, resulting in 2I þ1¼3 lines). [48] Later on this species has been assigned to a NO 2a 13-electrons radical species located in the bulk and generated by an interstitial nitrogen atom characterized by a rhombic g tensor whose components span over the range 2.008–2.003. The hyperfine coupling constant have been obtained by spectra simulation.[32,50–53] The EPR spectra of the STO nitrogen doped samples (Fig. 5C and Fig. S3) appear different from spectra obtained in the case of N doped Titania or N doped ZrTiO 4 previously described. In Fig. 5C the EPR spectrum of STO_La/N sample is reported within its computer simulation. It is immediately clear, observing the shape of the signal, that in this case the spectrum shows an axial symmetry instead of a rhombic one. This fact can be ascribed to the higher symmetry of SrTiO 3 with respect to TiO 2 or ZrTiO 4 . SrTiO 3 crystallizes with a centered-cubic symmetry with a Pm-3m space group while TiO 2 shows a tetragonal symmetry with a space group I4 1 /amd. ZrTiO 4 elsewhere presents an orthorhombic pseudobrookite symmetry with a PbCn space group and a lower symmetry respect to STO. For this reason, the EPR spectrum was simulated based on axial g and A hyperfine tensors having very similar g values in the range of 2.0055–2.0003. The values of g and A tensor elements obtained by the simulation are reported in Table 2 together with the data achieved from the literature. The spin density on the nitrogen atom can be derived from the hyperfine matrix A according to its experimental values reported in the Table 1. The a iso (the Fermi contact term) has been calculated in 4.4 mT and the dipolar values T respectively 2 T=2.2 and T=-1.1 mT. The spin density in the p orbital, ρ 2p , calculated by comparison of the experimental dipolar value with the corresponding atomic one ( ρ 2p =T/ T◦) is ρ 2p =0.599 (T◦=1.816 mT).[54] The isotropic Fermi contact term is expected to be positive in N-centered radical species and indicates a further amount of electron spin density (0.079) in the 2 s orbital of the nitrogen atom. The total spin density on the N atom of the observed species therefore amounts to 0.68, with the larger contribution being due to a single 2p orbital. The obtained values are in agreement with those found for N doped TiO 2 [55] and N doped ZrTiO 4 ,[51] so on the basis of both the spin density and the hyperfine constant value, it is possible to state that a fraction of the spin density is likely localized on other nuclei having zero nuclear spin. This evidence suggests that the nitrogen containing species strongly interacts with the oxide matrix as expected. Surprisingly and unlike the N_TiO 2 and N_ZrTiO 4 systems, UV irradiation did not produce any variation in the EPR spectrum of the nitrogen signal, this result can suggest that the nitrogen species does not participate in the excitation process or more probably that it plays a role in the charge transfer without the formation of any paramagnetic species.[32,44,50–52] Fig. 6. Phenol degradation curves under irradiation with UV light (≥340 nm) of STO, STO_La, STO_N and STO_La/N samples. Table 2 Spin-Hamiltonian parameters of N center in different Ti based systems. System g 1 g 2 g 3 A 1 /mT A 2 /mT A 3 /mT ρ (N.2 s) ρ (N2p) Ref. N-TiO 2 2.007 2.005 2.004 0.23 0.44 3.23 0.02 0.54 [54] N-ZrTiO 4 2.0081 2.0046 2.0038 0.22 3.40 0.16 0.02 0.67 [50] N-SrTiO 3 2.0054 2.0003 3.27 6.55 0.08 0.60 This work L. Sarasino et al. Next Materials 9 (2025) 101041 7 3.5. Photocatalytic tests The photocatalytic activity of the developed materials was assessed by using phenol as a model compound. Preliminary tests show that both direct photolysis and adsorption in the dark are negligible. Later, the disappearance of the organic compounds under simulated solar light for all the samples synthesized was monitored over time and the results are collected in Fig S4. For sake of brevity, only a short selection of results is reported. A comparison among STO, STO_La, STO_N and STO_La/N clearly shows that co-doping significantly enhances the photocatalytic efficiency, particularly for the La/N co-doped sample, which nearly completely degrades phenol in 60 min. Nitrogen doping alone significantly enhances photocatalytic activity, likely due to improved light harvesting as previously discussed. Specifically, the N mono-doped sample (STO_N) can completely degrade phenol in 2 h, outperforming the Ln mono-doped samples, which abated 50 % of phenol in 3–6 h. 3.6. Spin-trapping tests Spin-trapping tests are valuable methods to evaluate the reactive species involved in the phenol-degradation process. In particular four species formation under irradiation are investigated: superoxide radical (O 2 -• ), hydroxyl radical (OH • ), “hole” (h + ) and singlet oxygen ( 1 O 2 ), indicating the direct oxidizing effect of the photoexcited particle toward the molecule. The tested samples are the co-doped ones, being the most reactive toward degradation (according to the photocatalytic tests), additionally, for comparison, the pristine (STO), the nitrogen doped (STO_N) and the lanthanum doped (STO_La) samples are included. These reactive species production start with the absorption of a photon by the photocatalyst, promoting an electron-hole (e - -h + ) couple generation. Typically, these reactive species lifetimes are too short to allow their detection, so a spin-trap molecule is necessary, in order to form a metastable paramagnetic adduct, detectable throughout EPR. The superoxide radical (O 2 -• ) species, that arises from the transfer of the photoexcited electron to an oxygen molecule, can be detected adding to the solution DMPO, that forms the paramagnetic DMPO-O 2 - adduct (Fig. S7A).[44] The hydroxyl radical (OH • ) species derives from the electron transfer between an OH - anion and a photogenerated hole of the photocatalyst and can be detected using again DMPO, here forming the paramagnetic DMPO-OH • adduct (Fig. S7B).[44,56] To detect the hole (h + ) species a reaction with formate is needed, in order to generate the COO -• radical, that forms the paramagnetic DMPOCOO - adduct after adding DMPO to the solution (Fig. S7C).[44] The formation of the singlet oxygen species can follow these pathways: a) the e - -h + couple recombines, releasing energy, which can be absorbed by an oxygen molecule (in the triplet state) and, converts it to a singlet state ( 1 O 2 ); b) the photoexcited electron reduces an oxygen molecule to superoxide radical (O 2 -• ), then this specie reacts with the photogenerated hole, releasing the electron and converting to 1 O 2 ; c) if an intersystem crossing can occur in the photocatalyst, and a triplet state is formed, it can release energy by fluorescence and by energy transfer (as in the a pathway) and a 3 O 2 is converted to 1 O 2 . The singlet oxygen can be detected by trapping it with the specific trap 4-oxo-TEMP-H and forming the paramagnetic 4-oxo-TEMPO adduct (Fig. S7D).[45] Each signal exhibits a distinct shape determined by the couplings of the paramagnetic species within the formed adduct. In Fig. S7 the shape of the EPR signal is shown as a reference for the lanthanum/nitrogen co-doped sample. However, these signals can be quantitatively compared by evaluating the double integral of the EPR signal, as the EPR spectrum represents the first derivative of the absorbance. To assess the production of different reactive species by the samples, the double integral of the signals obtained was calculated after various irradiation periods using a 365 nm LED, subtracting the value obtained from the “dark” signal (no irradiation). Fig. S5 presents the comparison of the reactive species production across different samples over time, while Fig. S6 highlights the comparison of different reactive species production over time within each sample. The curves obtained in Fig. S5A reveals that the superoxide species O 2 -• production under irradiation decreases or remains stable in the doped materials compared to the pristine material. Fig. S5B shows that almost no OH • radical is produced under irradiation and consequently it cannot be included in the phenol degradation species. Interesting results were observed when the production of h + (Fig. S5C) was evaluated. Notably, an enhancement in h + production, with respect to the pristine material, was observed exclusively for the Lanthanum-nitrogen co-doped sample. This finding suggests that the increased h + generation could be a key factor underlying the exceptional reactivity of this sample. To verify this hypothesis the h + detection test was repeated for the STO_La/N co-doped sample adding phenol (Fig. 7A). The production of h + in this condition is totally suppressed, confirming our hypothesis: holes are the responsible for the degradation of phenol. Similarly, the production of superoxide radical is suppressed in presence of phenol (Fig. 7B), suggesting that also this reactive species is involved in the phenol degradation. On the contrary, the singlet oxygen production for the STO_La/N codoped sample remains similar to the pristine material (Fig. S5D). Repeating the test with the phenol addition, for the STO_La/N codoped sample (Fig. 7C, red curve), it showed an initial suppression of the singlet oxygen production, which is however compensated over the time. A similar test (Fig. 7C, green curve), conducted in presence of isopropanol (a hole scavenger), instead of phenol, showed a halved production of 1 O 2 , indicating that the generation of this specie involves at least two mechanisms, one of them implies the interaction with a h + (pathway a or b before detailed, or both), the other presumably is the fluorescence pathway, active even if the hole is not present. According to this evidence, summed in Fig. 8, singlet oxygen shows a detrimental effect over the degradation process, as its production implies the decreasing of the h + concentration, the effective active species. Relate to this, also the superoxide radical production seems to be detrimental toward the photodegradation ability, even being found as an active species, probably because it is involved in the pathway c of singlet oxygen production that implies the annihilation of both the reactive species (h + and O 2 -• ). 4. Conclusions Several different materials based on perovskite structure of Strontium Titanate doped with lanthanide and co-doped with nitrogen and lanthanide have been prepared and fully characterized. Some of these materials show a very interesting activity towards the abatement of a model molecule like phenol. Among the most active materials, we described which are the active species involved and why they are so active with respect to the others. The role of the dopant is crucial as well as the presence of nitrogen. The most active samples seem to be the codoped ones suggesting a crucial role of the nitrogen. We highlighted that the reactivity in solution medium follows different ways as demonstrated by spin trap experiments. We were able to individuate which are the active radical species in the abatement of phenol and which species seem to be detrimental respect to the whole process. Holes and singlet oxygens species show a crucial role in these photocatalytic reactions. L. Sarasino et al. Next Materials 9 (2025) 101041 8 Author contributions L.S.: investigation, data curation, writing—original draft, editing; V. M.C.: investigation; V.L.: data curation, editing; P.C.: funding acquisition, conceptualization, supervision, M.C.P.: conceptualization, writing—review and editing; All authors have read and agreed to the published version of the manuscripts. Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgments Authors acknowledge support from the European Union’s Horizon2020 research and innovation program under the Marie SkłodowskaCurie grant agreement No 101007578, SusWater project; the European Union’s HORIZON-MSCA-2022-DNDN-JD-IN2AQUAS grant agreement N◦101119555; European UnionNext Generation EU, Mission 4 Component 1 CUP D53D23010160006 with MUR (Italy) through PRIN Project MAPEC (N.2022599NR3) “Magnetic field assisted photo(electro) CO 2 conversion - MAPEC; the Project CH4.0 under the MUR program “Dipartimenti di Eccellenza 2023–2027“. Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at doi:10.1016/j.nxmate.2025.101041. Data Availability Data, associated metadata, and calculation tools are available from the corresponding author Fig. 7. Tests in spin-trapping measure of STO_La/N co-doped sample, with phenol for hole test (A), superoxide radical test (B) and with phenol or isopropanol for singlet oxygen tests (C). Fig. 8. A scheme of the reactions that occur at the surface of the prepared samples in solution. L. Sarasino et al. Next Materials 9 (2025) 101041 9