Journal of The Electrochemical Society       OPEN ACCESS Tailored Fabrication of Defect-Rich Ion Implanted CeO2-x Nanoflakes for Electrochemical Sensing of H2O2 To cite this article: Yueyue Luo et al 2023 J. Electrochem. Soc. 170 057519   View the article online for updates and enhancements. You may also like Identification and thermal healing of focused ion beam-induced defects in GaN using off-axis electron holography K. Ji, M. Schnedler, Q. Lan et al. - Cobalt sulfide/N,S-codoped defect-rich carbon nanotubes hybrid as an excellent bi-functional oxygen electrocatalyst Lulu Chen, Wenxiu Yang, Xiangjian Liu et al. - Constructing defect-rich Ni9S8/Fe5Ni4S8 heterostructure nanoparticles for efficient oxygen evolution reaction and overall water splitting Jinxiao Xu, Yingjun Ma, Jie Wang et al. - This content was downloaded from IP address 137.157.8.253 on 27/03/2025 at 05:52 https://doi.org/10.1149/1945-7111/acd41f /article/10.35848/1882-0786/ad163d /article/10.35848/1882-0786/ad163d /article/10.35848/1882-0786/ad163d /article/10.1088/1361-6528/aaf457 /article/10.1088/1361-6528/aaf457 /article/10.1088/1361-6528/aaf457 /article/10.1088/2515-7639/abf3ae /article/10.1088/2515-7639/abf3ae /article/10.1088/2515-7639/abf3ae /article/10.1088/2515-7639/abf3ae /article/10.1088/2515-7639/abf3ae /article/10.1088/2515-7639/abf3ae /article/10.1088/2515-7639/abf3ae /article/10.1088/2515-7639/abf3ae https://pagead2.googlesyndication.com/pcs/click?xai=AKAOjsvQ52TPalizzw6cz5B4UxF5a4mRWXCnzC1eweeQe3Gg1VVt9KdNftPUqhjErBHEpj7lccQc2Qmnev0ZON3XsetYK-WMZt2Xh-XkWCAdLsyEkZ_u8xI6Mx2CvCFsAD1LSZOVrG5YKWwSH83EOuSWeEoyWE-JjVQYtDayRiQ_F9DUs4H1M24I-tIj8JzCZFIvqhw_Xlfyx_nZKP-N8JCZvkhlPdgnsQdnPlx_A_WJR7SRWymLGeVQ9Zwn-4zkLoGBd1idgpTWaU6spM-bUpFssHXjp3A1sY5QiJFsf_5KvoruqERlZ4ETkp7PVG7UzjAl1VFCM7gzGRQeeBS68SIbej2J8CThG07NWZ__5DGLn5VjBsYw&sig=Cg0ArKJSzOJqSJqOieGT&fbs_aeid=%5Bgw_fbsaeid%5D&adurl=https://www.el-cell.com/products/pat-battery-tester/pat-tester-x/pat-tester-x-8/%3Fmtm_campaign%3Diop-pdf-advert%26mtm_kwd%3DPAT-Tester-x-8%26mtm_source%3Dpdf%26mtm_placement%3Dqr-code Tailored Fabrication of Defect-Rich Ion Implanted CeO2-x Nanoflakes for Electrochemical Sensing of H2O2 Yueyue Luo,1 Xiaoran Zheng,1 Corey Venkata Vutukuri,1 Naomi Ho,1 Armand J. Atanacio,2 Madhura Manohar,2 Hamidreza Arandiyan,3 Yuan Wang,4 C.C. Sorrell,1 Sajjad S. Mofarah,1,z and Pramod Koshy1,z 1School of Materials Science and Engineering, UNSW Sydney, Sydney, New South Wales 2052, Australia 2Centre for Accelerator Science (CAS), ANSTO, Lucas Heights, New South Wales 2234, Australia 3Centre for Applied Materials and Industrial Chemistry (CAMIC), School of Science, RMIT University, Melbourne, Victoria, Australia 4Institute for Frontier Materials, Deakin University, Melbourne, Victoria 3125, Australia As an alternative to H2O2 enzymatic biosensing devices, non-enzymatic CeO2-based biosensors have shown improved sensibility, robustness, and shelf lives. The redox capability in CeO2 and rapid switching between its oxidation states facilitate the formation of structural vacancy defects that serve as active sites. This work reports a novel approach for synthesis of defect-rich CeO2-x-based nanoflakes using a controllable electrochemical-based deposition at low temperatures (45°−65 °C) followed by low-energy ion implantation. Among the nanoflakes, Mo-implanted CeO2-x exhibited outstanding sensitivity of 4.96 × 10−5 A·mM−1 cm−2 within the linear range of 0.05–10 mM. Moreover, the ion-implanted samples yielded high sensing stability and electronic conductivity. The former was achieved through the multi-valence charge transfer between Ce and the implanted ions that caused the reduction of Gibbs free energies required for the formation/retention of the defects. The latter was due to the narrowing of the electronic bandgap of CeO2-x by creation of defect-induced midgap states. © 2023 The Author(s). Published on behalf of The Electrochemical Society by IOP Publishing Limited. This is an open access article distributed under the terms of the Creative Commons Attribution 4.0 License (CC BY, http://creativecommons.org/licenses/ by/4.0/), which permits unrestricted reuse of the work in any medium, provided the original work is properly cited. [DOI: 10.1149/ 1945-7111/acd41f] Manuscript submitted January 26, 2023; revised manuscript received April 26, 2023. Published May 24, 2023. Supplementary material for this article is available online Reactive oxygen species (ROS) are free radicals that include both oxygen-containing radicals, such as superoxide (•O2), hydroxyl (•OH), peroxyl (•RO2), and hydroperoxyl (•HO2) radicals, and certain non-radical, oxidizing agents such as hydrogen peroxide (H2O2), which have strong oxidizing ability and chemical reactivity.1–3 In general, normal cells have biological controls to prevent excessive ROS accumulation, since the ROS can oxidise the cell membrane proteins and DNA leading to oxidative stress and potentially cell mutation and/or cell death.4–6 Further, some cancer cells are known to produce hydrogen peroxide (H2O2). 4 These high concentrations of H2O2 in the cell environment are toxic to organisms and its dissociation results in the formation of highly- reactive hydroxyl free radicals (•HO‒) which are extremely destruc- tive towards macromolecules such as carbohydrates, nucleic acids, lipids, and amino acids.4,7 Although the absolute range of H2O2 in plasma remains unsettled due to the variability of the results in published literature, the likely normal range is 1–5 μM, and the concentration can reach 50 μM for some inflammatory diseases.8 For pregnant women, the plasma H2O2 level is approximately 50 ± 5.6 μM and for those with preeclampsia, the value can be go up to 77 μM.9 With regard to values in an environmental contenxt, the legal airborne permissible exposure limit is 1 ppm (29 μM) according to the Occupational Safety and Health Administration (OSHA), and an exposure amount of 75 ppm (2205 μM) is immediately dangerous to life and health.10 Therefore, there has been a keen interest in the development of an effective H2O2 detection method for use in clinical and environmental applications. Biosensors are devices that can convert a biological response to an electrical signal and have been used in food processing, fermentation processes, food safety, medical imaging, biodefence, and in the medical industry to diagnose infectious diseases.11,12 They are used to signal the occurrence of biochemical reactions between a specific substance that is to be identified and/or measured (analyte) and a biocatalyst immobilized onto the biosensor surface.12 Biosensors have extremely high sensitivity and have achieved excellent results, especially for detecting H2O2. 13,14 Enzyme-based electrochemical biosensors are commonly used due to their specific binding capabilities.15–17 However, these are expensive, time-con- suming to purify after use, must be stored in refrigerated environ- ments, lack long-term stability, and have low sensitivity due to indirect electron transfer.13–15 A promising novel alternative is non-enzymatic biosensors (metals, metal oxides, alloys, complexes, and carbon allotropes) due to their simpler design, cost-effective fabrication, longer shelf life, and higher robustness.18,19 Additional advantages include enhanced surface renewability and direct electron transfer from the analyte to the electrode. Recently, a variety of non-enzymatic sensors have been developed for sensing biomolecules, such as M- BDC metal–organic frameworks20 and NiO-MoO3 nanocomposite for sensing glucose,21 as well as CuO/black phosphorous nanohybrids22 and PBNCs/AuNPs/RGO nanocomposite for sensing H2O2. 23 A variety of nanostructured oxides, such as ZnO, Fe2O3, TiO2 and CeO2 have demonstrated desirable functional, biocompa- tible, and non-toxic properties, enabling their use in non-enzymatic electrochemical biosensors.24 Among oxides, CeO2 is a promising candidate for non-enzymatic biosensors owing to its excellent electronic conductivity and oxygen diffusivity which arises from its ability to create oxygen vacancies upon switching from Ce4+ to Ce3+.25–28 The Ce3+ ions can readily oxidise to Ce4+ through a reaction with (•HO‒) in H2O2 solution, thus providing an electro- chemical signal that can detect the presence of H2O2 at very low concentrations via a transducer.29 Through surface chemical reac- tions, Ce4+ ions can be converted back into Ce3+. The high isoelectric point (IEP) of 9.2 at pH 7.0 allows nanostructured CeO2 to immobilise low IEP biomolecules by electrostatic interactions.30 The selectivity of CeO2 is, however, limited, as is the case with other benchmark non-enzymic biosensors. Since H2O2 detection involves surface-based chemical reactions, the extent of the surface to volume ratio, the type and number of accessible active sites, as well as the surface chemical properties of biosensor materials at their interfacial region are crucial.31 Therefore, it is imperative to develop nanostructures with a maximum number of active sites and surface areas through a controlled synthesis processzE-mail: s.seifimofarah@unsw.edu.au; koshy@unsw.edu.au Journal of The Electrochemical Society, 2023 170 057519 https://orcid.org/0000-0002-7016-5260 https://orcid.org/0000-0002-4835-767X http://creativecommons.org/licenses/by/4.0/ http://creativecommons.org/licenses/by/4.0/ https://doi.org/10.1149/1945-7111/acd41f https://doi.org/10.1149/1945-7111/acd41f https://doi.org/10.1149/1945-7111/acd41f mailto:s.seifimofarah@unsw.edu.au mailto:koshy@unsw.edu.au https://crossmark.crossref.org/dialog/?doi=10.1149/1945-7111/acd41f&domain=pdf&date_stamp=2023-05-24 for use in biosensing applications.12,32 The fabrication of CeO2 and other oxide-based biosensors has been extensively studied. The most commonly utilized fabrication methods are hydrothermal, sol- vothermal, and solid-state reactions; however, these processes are complex, non-reproducible, and require binding and conductive agents, which limits their widespread use.33–36 This work reports a simplified, high-yield, controllable, low- temperature process for fabricating two-dimensional (2D) transition- metal doped (implanted) CeO2-x nanoflakes. The fabrication method involved cathodic chronopotentiometry electrodeposition (CCE) of pristine CeO2-x nanoflakes followed by introduction of Mo and V dopants using an ion-implantation method. The resultant Mo- and V- doped CeO2-x showed outstanding H2O2 detection performance. The strategy has shown promise to be applied to other materials that have multi-oxidation states, including other rare earth elements and transition metals. Results and Discussion Tailoring CeO2-x-based nanostructures.—The defect-rich CeO2-x nanoflakes were fabricated on fluorine-doped tin oxide (FTO) substrate in a three-electrode-configuration system using cathodic chronopotentiometry electrodeposition (CCE) technique. The morphological and structural optimisation of the nanoflakes was achieved by varying two key experimental factors including pH ranges of 4.0–4.5 (acute acidic) and 5.5–6.0 (mild acidic) and electrolyte temperatures (45 °C, 55 °C, and 65 °C). The selected CeO2-x nanoflakes were further optimised by implantation of single- atom transition metals (V and Mo) using low-energy ion implanta- tion. Fig. 1 shows a schematic of the two-step fabrication process that includes the electrodeposition and the metal-ion implantation. Mild acidic pH range electrodeposition.—X-ray diffraction (XRD) patterns of the nanoflakes synthesised in the mild acidic electrolyte are shown in Fig. 2a. According to the reference data (JCPDS 81–0792), the peaks were assigned to CeO2 (space group Fm3m). The data show that increase in the deposition temperature resulted in higher peak intensity and narrowing of the peak FWHM, indicating higher crystallinity of the nanoflakes. Therefore, the CeO2-x nanoflakes synthesised at 65°C exhibit the highest crystal- linity. Additionally, Raman microspectroscopy (Fig. 2b) confirmed that the intensity of the main peak at 450 cm−1, which is attributed to the F2g vibration mode of Ce-O bonds in cubic fluorite CeO2, increases with increasing temperature.37,38 In addition, an increase in deposition temperature from 45° to 55°C and then to 65°C resulted in the peak shifting from 456.44 cm−1 to 459.85 cm−1 and then to 460.59 cm−1, respectively. The peak representing the symmetrical Ce-O vibration mode in a pristine (non-defective) CeO2 is observed at the wavenumber of 465 cm−1.39 Furthermore, the non-symme- trical broadening of the F2g band in the range of 456–461 cm−1 indicates the presence of oxygen vacancies.37,38,40 Therefore, a Raman shift to lower energies (red shift) can be attributed to the combined effects of strain and phonon confinement, as well as the expansion of the lattice caused by oxygen vacancy formation.41 With increase in the electrolyte temperature, the band shifts to higher energies as a result of a decrease in oxygen vacancy concentration ([VO ••]). Additionally, the 480 cm−1 band shows higher intensity for 45°C samples, indicating higher [VO ••] at the lower deposition temperatures.42 The microstructures of the nanoflakes were analysed using scanning electron microscopy (SEM). Fig. 2c shows SEM images of the 45°C sample, where smooth nanoflakes are formed with an average thickness of ∼50 nm, while little to no cracking. Fig. 2d shows that increasing the deposition temperature to 55°C increased the nanoflake thickness to ~90 nm, resulting in a more densely packed structure with small micropores and minimal cracking. By increasing the temperature to 65°C, the nanoflakes were observed to become thicker (~230 nm) due to coalescence with the adjacent nanoflakes resulting in a sandwich-like structure (Fig. 2e). The packing density of this sample was the highest and cracks were barely visible. X-ray photoelectron spectroscopy (XPS) was used to determine the defect characteristics. Figure 2f shows the XPS spectra related to Ce 3d orbital of the CeO2-x nanoflakes and these were deconvoluted into ten individual doublets corresponding to two Ce4+ and Ce3+ oxidation states43 ; the details related to the peak allocation are as follows: (1) Ce4+ peaks at binding energies of 898.3 and 916.7 eV are attributed to the Ce(3d9 4f0) O(2p6) final states. (2) Ce4+ peaks at binding energies of 882.5, 888.7, 900.7, and 907.5 eV are attributed to the mixing of Ce(3d94f2)O(2p4) and Ce (3d9 4f1)O(2p5) final states. (3) Ce3+ peaks at 880.6, 884.9, 889.3, and 903.2 eV, are attributed to the mixing of Ce(3d9 4f1) O(2p6) and Ce(3d9 4f2) O(2p5) final states. The concentration of Ce3+ ([Ce3+]) can be determined from the areas of the Ce3+ doublet peaks centered at the binding energies of 880.6 and 884.9 eV.44–46 Considering defect equilibria and charge compensation, the formation of each requires the reduction of two Ce4+ ions to Ce3+. The quantitative analysis (atomic concentration) of Ce3+ is therefore proportional to the number of vacancies present in the system. Furthermore, these calculations can be correlated with the XPS data for the 1s orbital of the O peaks (Fig. 2g). These curves were fitted using Gaussian functions to show peaks at ∼529.5 eV, ∼531.5 eV, and ∼533 eV.43,47 Peaks 1 and 2 represent oxygens bound to Ce4+ ( OCe4[ ]+ ), and bound to Ce3+ ( OCe3[ ]+ ), respectively, and peak 3 represents adsorbed water molecules.37,42,43 The normal- ised Ce3+ concentration (i.e., Ce3+/(Ce3+ + Ce4+)) was determined Figure 1. Schematic showing fabrication of pristine CeO2-x and metal-ion implanted CeO2 nanoflakes. Journal of The Electrochemical Society, 2023 170 057519 to be 6.28, 4.79, and 5.78 at%, for the samples deposited at 45°, 55°, and 65 °C, respectively. These results were confirmed by measuring the peak area related to oxygen bound to Ce3+ which showed values of 5.62, 4.12, and 6.23 at% respectively for the same samples deposited at the three different temperatures. The small difference in the [VO ••] calculated through data for [Ce3+] and OCe3[ ]+ can be attributed to the sensitivity of the XPS beam. For the Ce, penetration depth extends to ∼1.0 nm, while for the O, it extends to ∼1.5 nm.43,48 The results showed that the [VO ••] decreased at 55°C and then showed an increase at 65°C. Such a volcano plot of VO •• variations with the temperatures can be justified from the Pourbaix diagram of Ce-H2O system proposed earlier by Mofarah et al.,49 where pH and applied potential during the electrodeposition can determine the type and extent of structural defects of CeO2-x. There are two competing factors affecting the [VO ••] including the (1) Gibbs free energy of vacancy formation ( Gf VO ••Δ − ) and (2) the concentration of dissolved oxygen determining the oxidation rate.50–54 At the lowest temperature (45°C), the Gf VO ••Δ − value is sufficient but the oxidation rate is too low to fill the vacancies. However, at 55°C, although the Gf VO ••Δ − value is adequate, someVO •• are refilled due to the partial oxidation of CeO2-x. The applied potential at 65°C led to a higher HER rate in the system and thus increased the local pH, resulting in the increased V ;O ••[ ] nevertheless, the oxidation rate was also enhanced resulting in the partial occupation of V .O •• Acute acidic pH range electrodeposition.—The XRD patterns of nanoflakes formed at different temperatures under acute acidic pH (Fig. 3a) showed that these were only composed of CeO2 and the peak intensity increased with increasing deposition temperature, similar to those obtained in mild acidic pH. The Raman spectra shown in Fig. 3b revealed that the band at 458.1 cm−1 for the CeO2-x synthesised at 45°C shifted to higher wavenumbers with increase in temperature owing to the lowering of the vacancy concentrations and increased CeO2 crystallinity. Furthermore, the SEM images of the CeO2-x electrodeposited in the acute acidic pH (shown in Figs. 3c –3e) demonstrate an extremely packed structure, in contrast to those obtained under mild acidic pH conditions. Fig. 3c illustrates the presence of coral-like morphologies without cracking at the lowest temperature (45°C); however, there appear to be gaps between the coral structure clusters. With increase in the temperature to 55°C, densely packed layer of thick nanoflakes with minimal gaps and reduced pore sizes was formed as shown in Figs. 3d, 3a . In addition, Figure 2. Structural characterisation of pure CeO2-x nanoflakes deposited at 45°C, 55°C, and 65°C under mild acidic pH (5.5–6.0) range: (a) XRD pattern, (b) Raman spectra, (c) SEM image of pure CeO2-x nanoflakes deposited at 45°C, (d) SEM image of pure CeO2-x nanoflakes deposited at 55°C, (e) SEM image of pure CeO2-x nanoflakes deposited at 65°C, (f) XPS spectra showing the Ce 3d regions and (g) XPS spectra of O 1s regions for the three samples. Journal of The Electrochemical Society, 2023 170 057519 short, thick nanorods with a thickness of ∼340 nm were spotted. The SEM images of the CeO2-x films synthesised at 65°C (Fig. 3e) exhibit a finely packed structure with small pores and uniform nanoflakes of thickness∼ 400 nm partially covered by nanoparticles. Accordingly, the VO ••[ ] of nanoflakes synthesised under acute acidic pH was studied using XPS analysis, the results of which are shown in Fig. 3f. The VO ••[ ] based on the [Ce3+] were measured to be 5.74, 5.56, and 5.55 at% for the samples deposited at 45°, 55°, and 65 °C, respectively. These values were compared with those obtained indirectly from measuring the areas associated with the peak of [OCe3+]. A general trend shows that the vacancy concentra- tion is reduced significantly when the temperature is increased to 55° C. However, when the temperature is increased to 65°C, the concentration of vacancies increased. Similar to the data for the films fabricated under mild pH ranges, the increase in vacancy concentration at the acute pH ranges can also be attributed to variations in the local pH during deposition. The morphology of the nanoflakes thickened when deposited at pH 4.0 − 4.5 (Figs. 3c–3e) compared to the case of using pH 5.5 − 6.0 (Figs 2c–2e). The SEM results showed that increasing the deposition temperature under both acidic pH conditions enhanced the thickness of the nanoflakes and this is expected due to the diffusion-controlled nature of the CCE process (Fig. S1). The mechanism behind the thinning of the deposited CeO2-x nanoflakes at higher pH levels is associated with non-faradaic formation of Ce3+ following by the deposition of CeO2-x. This has been investigated using thermodynamic analysis and the corresponding Pourbaix diagram, as shown in Fig. S2. By tailoring the CCE process, optimal structures, morphologies, and porosities of pristine CeO2-x were achieved at 55°C and pH 5.5–6.0. The suitability of this sample for established owing to it possessing combination of several desirable characteristics such as high accessible surface, thin nanoflakes (∼90 nm thick), minimal microcracks, and optimum crystallinity. Accordingly, these samples were selected for further investigation. Following the electrodeposition of the defect-rich CeO2-x samples, these samples were implanted with Mo and V using ion-implantation and these were then calcined to obtain crystalline Mo- and V-doped CeO2-x samples. The surface area of the pristine CeO2-x was measured using electrochemical surface area (ECSA). Further, the pore size distribution of the samples was investigated using image analysis of the SEM images using the software Image J. The corresponding results are shown in Fig. S3. Fabication of N2-annealed Mo- and V-doped CeO2-x nano- flakes.—In order to investigate the role of annealing atmosphere on structural, chemical, and morphological properties of the nanoflakes, Figure 3. Structural characterisation of pure CeO2-x nanoflakes deposited at 45°C, 55°C, and 65°C under acute acidic pH (4.0–4.5) range: (a) XRD patterns, (b) Raman spectra, (c) SEM image of pure CeO2-x nanoflakes deposited at 45°C, (d) SEM image of pure CeO2-x nanoflakes deposited at 55 °C, (e) SEM image of pure CeO2-x nanoflakes deposited at 65°C, (f) XPS spectra of Ce 3d region, (g) XPS spectra of O 1s region for all samples. Journal of The Electrochemical Society, 2023 170 057519 the pristine CeO2-x nanoflakes were annealed under N2 and air atmospheres and comprehensively analysed. The XRD, Raman, and XPS results of these samples are provided in Fig. S4. Then, based on these results the Mo- and V-doped CeO2-x were synthesised by ion implantation followed by annealing at 400 °C for 4 h in either air or N2. Figure 4a shows the XRD results of Mo- and V-doped CeO2-x nanoflakes. For the Mo-doped CeO2-x sample, there are new peaks at 26.6°, 39.0°, 62.6°, and 66.0° 2θ that are indexed to SnO2. This peaks arise owing to the damage caused to the samples during ion implantation, which resulted in exposure of the underlying FTO substrate. The V-doped CeO2-x did not show any peaks of FTO, and this corresponded with the absence of any visible surface damage after implantation. This difference could be due to the smaller ionic size of V (0.54 nm) compared to that of Mo (0.59 nm). Further, the Raman results (Fig. 4b) showed no significant peak shifts for the Raman bands, indicating that implantation followed by annealing did not significantly alter the concentration of defects. Figures 4c, 4d, and 4e illustrate SEM images of as-deposited, Mo-doped, and V- doped CeO2-x after annealing, respectively. Microcracks in both Mo- and V-doped CeO2-x samples increased in width after annealing and this is attributed to (1) the mismatch between the thermal expansion coefficients of CeO2 (12 × 10−6 K−1)55 and FTO (35 × 10−7 K−1)56 that results in thermal stress and widens cracks during cooling and (2) the high kinetic energy transferred from the implanted ion to the sample surface during implantation. These microcracks resulted in increased exposed surface area and could have an impact on the biosensing sensitivity of the CeO2-x nanoflakes (discussed subse- quently). The surface chemical analyses of the ion-implanted nanoflakes determined though XPS are shown in Fig. 4f and g. The results of the quantitative XPS analysis showing the Ce and O for all the as-deposited, annealed, and doped samples are tabulated are provided in Table I. According to the XPS results, the Mo- and V-doped samples showed [VO ••] of 8.13 and 8.0 at%, respectively. These results were also confirmed by XPS obtained from [OCe3+], which showed values of 13.44 and 11.65 at%, respectively for these implanted samples. The increase in [VO ••], in comparison to the pristine and annealed samples, is attributed to interstitial doping of V with lower oxidation states (V3+) in comparison to Ce4+. However, Mo dopants with the oxidation states of 5+ and 6+ are expected to yield Ce vacancies Figure 4. Structural characterisation of the non-doped, Mo- and V-doped thin films after annealing in N2.: (a) XRD patterns (b) Raman spectra (c-e) SEM images of the (c) non-doped (d) V- and (e) Mo-doped CeO2-x nanoflakes after N2 annealing; (f) XPS spectra of Ce 3d region, (g) XPS spectra of O 1s region for V- and Mo-doped CeO2-x nanoflakes. Journal of The Electrochemical Society, 2023 170 057519 (VCe″″) in the structure.43,48 The other possibility is the formation of Ce3+ through redox mechanisms to counterbalance the increased positive ion presence from the introduction of higher valence V ions (V5+). Thus, the formation of VO •• can occur through ionic, electronic, and redox compensation mechanisms.43 The XRD results of the implanted samples did not reveal any secondary phases; therefore, Mo and V ions likely occupied interstitial or substitutional positions in the CeO2 structure or the concentrations of related species were below the detection limit. The possibility of the formation of substitutional solid solutions between Ce and Mo and V was studied using Hume-Rothery rules, the results of which are shown in Table II. The differences in ionic radius between Ce (Ce3+ = 1.21 nm, Ce4+ = 1.11 nm), with respective coordination numbers (N) of VII and VIII, and Mo and V ions (with all possible N) were calculated. It was found that the difference in ionic radius was greater than 25% in a majority of the cases. However, the difference in radii between Ce4+ and Mo6+ (VII) as well as between Ce4+ and V4+ (VIII) were 21% and 22% respectively, suggesting the possibility of occurrence of substitu- tional solid solubility of Mo6+ and V4+ in the CeO2 structure. Despite the limited solid solubility of Mo and V, partial substitutional solid solution can occur at such low dopant concentrations.43,48 The solid solution formed could also undergo intervalence charge transfer (IVCT) through the reduction of Ce4+ to Ce3+ and the oxidation of implanted Mo or V substitutional ions as observed in Mo and V-doped CeO2. Equations 1 to 4 illustrate examples of possible reactions: Ce Mo Ce Mo 14 4 . 3 5+ ↔ + [ ]+ + + + Ce Mo Ce Mo 24 5 . 3 6+ ↔ + [ ]+ + + + Ce V Ce V 34 3 . 3 4+ ↔ + [ ]+ + + + Ce V Ce V 44 4 . 3 5+ ↔ + [ ]+ + + + The relative ease of occurrence of these reactions would be related to the changes in crystal radii arising from the change in valence state, resultant defect formation, and its overall impact on the stability of the crystal lattice. H2O2 sensing performance.—The detection of H2O2 molecules occurs through their adsorption and subsequent chemical reactions given in Eqs. 5–7. 2Ce H O Ce Ce O H O 2e 5 3 2 2 4 4 ads 2 + → + ( − ) + + [ ] + + + − Ce Ce O H O 2Ce H O O 2e 6 4 4 ads 2 2 3 2 2 + ( − ) + → + + + [ ] + +⋅ + − Net reaction: 2H O O 2H O 72 2 2 2→ + [ ] Where: Ce4+-Oads is the adsorbed oxygen species in the vacancies that allows for Ce3+ ions to be transformed to Ce4+ ions. The electrons that are produced as a by-product of this reaction are detected as a current increase in the CV results and therefore higher Table I. XPS data of as-deposited, N2-annealed, air-annealed, Mo-doped, and V-doped CeO2-x nanoflakes. Row Parameter (at%) pH = 5.5–6.0 Annealing Implantation 45 °C 55 °C 65 °C Air N2 Mo V 1 Ce Concentration (at%) 31.04 30.94 29.51 30.32 24.91 24.68 25.01 2 O Concentration (at%) 68.96 69.06 70.27 70.45 75.08 75.40 74.99 3 Ce/O Ratio 0.45 0.45 0.42 0.43 0.33 0.33 0.33 4 O/Ce Ratio 2.22 2.23 2.38 2.32 3.01 3.06 3.00 5 Ce3+/(Ce3+ + Ce4+) Concentration 25.11 23.81 24.81 20.55 29.43 32.50 32.08 6 O-Ce3+/O-Ce4+ Concentration 19.09 18.50 33.47 24.43 66.92 67.59 53.73 7 O-Ce3+/(O-Ce3+ + O-Ce4+) Concentration 16.03 15.60 25.01 19.63 40.09 40.33 34.95 8 Hypothetical Surface VO •• Concentration (at%) 12.56 11.91 12.41 10.28 14.72 16.25 16.04 9 Hypothetical Bulk VO •• Concentration (at%) 8.02 7.80 12.50 9.82 20.05 20.17 17.48 Table II. Ionic radii of elements involved in the fabrication of Mo- and V-doped CeO2-x. 57 Ion Coordination Number (N) Ionic Radius Ionic Differences with Ce3+ (N = 7)—(%) Ionic Differences with Ce4+ (N = 8)—(%) Mo4+ VI 0.79 0.42—(34%) 0.32—(28%) Mo5+ IV 0.60 0.61—(50%) 0.51—(46%) VI 0.75 0.46—(38%) 0.36—(32%) Mo6+ IV 0.55 0.66—(54%) 0.56—(50%) V 0.64 0.57—(47%) 0.47—(42%) VI 0.73 0.48—(39%) 0.38—(34%) VII 0.87 0.34—(28%) 0.24—(21%) V3+ VI 0.78 0.43—(0.35) 0.33—(0.29) V4+ V 0.67 0.54—(45%) 0.44—(40%) VI 0.72 0.49—(40%) 0.39—(35%) VIII 0.86 0.35—(29%) 0.25—(22%) V5+ IV 0.49 0.72—(59%) 0.62—(56%) V 0.60 0.61—(50%) 0.51—(46%) VI 0.68 0.53—(44%) 0.63—(57%) Journal of The Electrochemical Society, 2023 170 057519 [VO ••] (and consequently more Ce3+ ions) lead to higher peak redox current.58–60 Furthermore, the possible decrease in the electrode sensitivity at high H2O2 concentrations can occur and this is attributed to the interactions between O2 molecules, which occurs through H2O2 and CeO2-x reactions (Eq. 7), and VO •• formation at the surface of the nanoflakes. The H2O2 sensing performance of the defect-rich CeO2-x nano- flakes were thoroughly investigated. Initially, the sensing behaviour of the as-deposited, the N2 annealed, and the air-annealed CeO2-x nanoflakes were studied using cyclic voltammetry (CV) in PBS (as the control electrolyte) and PBS with different concentrations (0.05, 0.1, 0.5, 1, 2, 5, and 10 mM) of H2O2 at scan rates of10, 50, and 200 mV.s−1. The results are provided in Fig. S5. V-doped CeO2-x.—The H2O2 biosensing performance of the V- implanted sample nanoflakes is shown in Figs. 5a–5c, where the voltammograms at constant H2O2 concentrations of 0.05 mM, 1 mM, and 10 mM at different scan rates ranging from 10 mV.s−1 to 200 mV.s−1 are illustrated. In addition, CV tests were performed at constant scan rates of 10 mV.s−1, 50 mV.s−1, and 200 mV.s−1 while H2O2 concentration increased gradually from 0.05 mM to 10 mM (Figs. 5d–5f). Similar to the other samples, the increase in scan rates led to an increase in peak currents for redox reactions, enhancing the CV curve areas. The results of the linear trend line analysis performed on the V- doped CeO2-x nanoflakes are shown in Figs. 5g–5i. At a scan rate of 10 mV.s−1 the fitting value of the linear regression model (R2) is 0.75177. This value increased to 0.81828 at 50 mV.s−1 and remained constant at 200 mV.s−1. This indicates that the nanoflake response became more linear with increasing scan rate. At low concentrations (0.05–0.5 mM), the sensitivity and the limit of detection (LOD) of the nanoflakes were measured to be 4.3176 × 10−4 A.mM−1 and 0.015 mM, respectively. However, at higher concentrations (1 mM − 10 mM), the sensitivity and LOD values were 3.0299 × 10−5 A.mM−1 and 0.220 mM, respectively. This clarifies that the sensitivity decreases after 1 mM and therefore the LOD increases possibly owing to a reduction of [VO ••] owing to oxidation in the presence of H2O2. Mo-doped CeO2-x.—Figure 6 shows the voltammograms for the Mo-doped CeO2-x nanoflakes scanned in H2O2 concentrated PBS solution. Figures 6a–6c show the voltammograms obtained at 0.05 mM, 1 mM, and 10 mM H2O2 concentrations with the scan rates of 10, 20, 50, 100, and 200 mV.s−1. Further, the H2O2 sensing performance of the Mo-doped CeO2-x were studied at the constant scan rates of 10, 50, and 200 mV.s−1 while H2O2 concentration was varied from 0.05 to 10 mM (Figs. 6d–6f). Similar to all the samples, the area of the curve increased with increasing scan rate and concentration of H2O2. As shown in Figs. 6d–6f, peak currents for Figure 5. CV curves for V-dopedCeO2-x thin film in PBS solutions with varying concentrations (a) 0.05 mM H2O2 (b) 1 mM H2O2 (c) 10 mM H2O2 and scan rates of (d) 10 mV.s−1, (e) 50 mV.s−1, and (f) 200 mV.s−1, in 0 mM (PBS) to 10 mM H2O2, (g-i) linear detection analysis for V-doped CeO2-x nanoflakes and non-implanted thin films at H2O2 concentrations between 0.05–10 mM with scan rates of (g) 10, (h) 50, (i) 200 mV.s−1. Journal of The Electrochemical Society, 2023 170 057519 redox reactions continued to increase at high H2O2 concentrations (>5 mM), unlike those observed for pristine samples. Clearly, this behaviour indicates that Mo-doped samples demonstrate high reten- tion levels of VO •• and V ,Ce′′′′ even thoughrelatively high oxidation rates are expected to occur in the presence of high [H2O2]. Further, linear trend line analysis was performed on the Mo-doped CeO2-x nanoflakes and this is shown in Figs. 6g–6i. For the V-doped samples at 10, 50, and 200 mV.s−1, R2 values were calculated to be 0.99085, 0.98996, and 0.99208, respectively. It is evident from the graphs that there is a strong linear correlation between the peak currents of the redox reactions and the concentrations of H2O2 at all scan rates (R2 = 0.99). At the scan rate of 50 mV.s−1, a sensitivity of 7.43849 × 10−5 Am.M−1 was observed with an approximate LOD value of 0.088 mM. A linear regression plot of peak current vs H2O2 concentrations was used to measure the sensitivity of Mo-doped CeO2-x nanoflakes. With annealed samples, a weak correlation was observed across all scan rates (R2 = 0.06321, 0.07587, 0.01781), which is attributed to the loss of sensitivity to H2O2 at higher concentrations (>1 mM). A linear range is observed in the as -deposited CeO2 sample between 0.05 mM and 1 mM H2O2. In this linear range, at 50 mV.s−1, a sensitivity of 8.62996 × 10−5 A.mM−1 and an approximate LOD of 0.149 mM were observed. In addition, higher detection sensitivity was observed at lower scan rates. This was due to the diffusion-controlled nature of H2O2 detection. With a lower scan rate, where time is not a limitation, H2O2 molecules can diffuse more to and away from the electrode surface, resulting in higher sensitivity. The details of the line range, sensitivity, and LOD results for all the samples are given in Table III. Figure 7 compares the voltammograms of undoped, Mo-doped, V- doped CeO2-x after annealing in N2 for 4 h at 400 °C. It is clear that the V- and Mo-doped samples exhibit increased stability at higher H2O2 concentrations. This can be attributed to multi-valence charge transfer (MVCT) mechanisms (Eqs. 1–4) that would facilitate the formation of VO •• and V ,Ce″″ through the reduction of the Gf VO ••Δ − and G ,f VCe Δ ′′′′− and thus increase their retention during the H2O2 detection process. This results in higher sensing stability and could be related to drastically higher electronic conductivity in the implanted samples. It has previously been demonstrated that structural defects in CeO2-x narrow the bandgap by forming midgap states, thereby improving electronic conductivity by facilitating electron transitions toward the conduction band.36 In addition, the higher stability of the Mo-implanted film is likely to be attributed to the higher valence states compared to the V-doped sample. According to the Kröger-Vink defect equilibrium, these different valence charges can create defects such as oxygen vacancies, metal substitutional defects, and charge carriers such as electrons or holes. The corresponding equations for V (V3+/V5+) and for Mo (Mo5+/Mo6+) are given in the following equations. V (V3+/V5+) Figure 6. CV curves for Mo-dopedCeO2-x thin films on testing in PBS solutions with (a) 0.05 mM H2O2 (b) 1 mM H2O2 (c) 10 mM H2O2 and scan rates of (d) 10 mV.s−1, (e) 50 mV.s−1, and (f) 200 mV.s−1, in PBS solution with H2O2 concentration from 0.05 to 10 mM, (g-i) Linear detection analysis for Mo-doped CeO2-x nanoflakes and non-implanted thin films in PBS solutions with H2O2 concentrations between 0.05–10 mM with scan rates of (g) 10, (h) 50, (i) 200 mV.s−1. Journal of The Electrochemical Society, 2023 170 057519 Electronic compensation: V O 1 2 O 2V 2h 4O 8Ce2 3 2 • o′+ ↔ + + [ ] 2V O 4V 4e 8O O 9Ce o2 5 • 2↔ + ′ + + [ ] Ionic compensation: V O 2V V 3O 10Ce O o2 3 ••′↔ + + [ ] 2V O 4V V 10O 11Ce Ce2 5 • o″″↔ + + [ ] Redox compensation: V O 1 2 O g 2V 6Ce 4O Interstitial Solid Solubility 12 2 3 2 CeO i ••• Ce O X2 ′+ ( ) → + + [ ] V O Ce O 2V 2V 6O Substitutional Solid Solubility 13 2 3 2 3 CeO Ce Ce O X2 ′′ ‴+ → + + [ ] V O 1 2 O g 2V 10Ce 6O Interstitial Solid Solubility 14 2 5 2 CeO i ••••• Ce O X2 ′+ ( ) → + + [ ] 3V O Ce O 6V 10V 18O Substitutional Solid Solubility 15 2 5 2 3 CeO Ce Ce O X2 ″ ‴+ → + + [ ] Mo (Mo5+/Mo6+) Electronic compensation: Mo O 2Mo 2e 4O 1 2 O 162 5 Ce • o 2↔ + ′ + + [ ] MoO Mo 2e 2O 1 2 O 173 Ce •• o 2↔ + ′ + + [ ] Ionic compensation: 2Mo O 4Mo V 10O 182 5 Ce • Ce o″″↔ + + [ ] 2MoO Mo V 6O 193 Ce •• Ce o″″↔ + + [ ] Redox compensation: Mo O 1 2 O g 2Mo 10Ce 6O Interstitial Solid Solubility 20 2 5 2 CeO i ••••• Ce O X 2 ′+ ( ) → + + [ ] 3Mo O Ce O 6Mo 10V 18O Substitutional Solid Solubility 21 2 5 2 3 CeO Ce Ce O X 2 ″ ‴+ → + + [ ] MoO 1 2 O g Mo 6Ce 4O Interstitial Solid Solubility 22 3 2 CeO i •••••• Ce O X 2 ′+ ( ) → + + [ ] VMoO Ce O Mo 6O Substitutional Solid Solubility 23 Ce3 2 3 CeO Ce O X2 ″ ‴+ → + + [ ] Where: VCe″ = V3+ substituted on the Ce4+ site (−1 charge) VCe • = V5+ substituted on the Ce4+ site (+1 charge) MoCe • = Mo5+ substituted on the Ce4+ site (+1 charge) MoCe •• = Mo6+ substituted on the Ce4+ site (+2 charge) VCe″″ = Ce vacancy (−4 charge) VCe‴ = Ce vacancy (−3 charge) CeCe′ = Ce4+ → Ce3+ Oo = Oxygen on O site e′ = Electron (−1 charge) h• = Hole (+1 charge) The equations suggest that Mo-doping by implantation induces higher levels ofVCe″″ in the structure, whereas V-doping induces higher levels of V .O •• Based on the experimental data, however, the Mo-doped Table III. Linear range, sensitivity and LOD (S/N = 3) for each tested electrode. H2O2 Biosensor Linear Range (mM) Sensitivity (A.mM−1) Sensitivity (A.mM−1cm−2) LOD (mM) CeO2-x (low Concentration) 0.05–2 8.62996 × 10−5 5.75331 × 10−5 0.149 Mo-doped CeO2-x (max concentration) 0.05–10 7.43849 × 10–5 4.95899 × 10−5 0.088 V-doped CeO2-x (low Concentration) 0.05–0.5 43.1761 × 10–5 28.7841 × 10−5 0.015 Mo-doped CeO2-x (high concentration) 1–10 3.02987 × 10–5 2.01991 × 10−5 0.220 Figure 7. CV curves for undoped, Mo- and V-doped CeO2-x thin films annealed in N2 and tested in PBS solutions with H2O2 concentrations of (a) 0.05 mM, (b) 1 mM, and (c) 10 mM of H2O2 at a scan rate of 50 mV.s−1. Journal of The Electrochemical Society, 2023 170 057519 CeO2-x exhibited a greater number of V ,O •• suggesting that other mechanisms may be involved. The presence of Ce defects in the CeO2-x nanoflakes may contribute to the increased sensing activity of such materials by enhancing the ionic conductivity of such materials. To test the long-term stability, the pristine CeO2-x was subjected to 24 tests, and for each test, 10 cycles were measured. When not in use, the electrode was stored in dry condition. The scan rate was set to 50 mVs−1 and the voltage window used rangedfrom −0.75 to 1.0 V. The performance retention rate (i.e., stability) was measured and compared to other types of biosensors for detecting H2O2 and this is shown in Fig. 8a. The CeO2-x sensor exhibited outstanding stability of 90.4% after long- term testing (24 cycles), which is higher than most of the biosensors Figure 8. (a) Retention of stability vs number of tests for defect-rich CeO2-x in comparison with literature.61–66 (b) Selectivity of defect-rich CeO2-x for H2O2, ethanol, urea, acetone and the mixture of ethanol and acetone analytes. Journal of The Electrochemical Society, 2023 170 057519 reported to date. Since the pristine CeO2-x sample showed outstanding stability, similar stability is expected for the implanted samples owing to the dominant effect of the CeO2-x nanostructure, even though they could not be tested. Details of the sensing stability of this work and those recently published are given in the inset of Fig. 8a. Further, the selectivity performance of the defect-rich CeO2-x nanoflakes against H2O2 in the presence of including ethanol, urea acetone, and a mixture of ethanol and acetone was studied. The scan rate was set to 50 mVs−1 while the voltage window was from −0.75 to 1.5 V. The test was also conducted in PBS solution inorder to substract the PBS contribution to selectivity. The results are indicated in Fig. 8b and from these, it is clear that the defect-rich CeO2-x exhibits a significantly high sensitivity to H2O2 compared to other analytes, which makes these electrodes desirable for medical and environmental applications. Conclusions This work reports fabrication of CeO2-x nanoflakes using a simplified, controllable, cost-effective approach, i.e., cathodic chron- opotentiometry electrodeposition (CCE). The structural, morpholo- gical, chemical, and surface chemical characteristics of the nano- flakes were optimised by varying the experimental conditions including pH and temperature of the electrolyte, annealing under N2 and air atmosphere, and Mo and V doping using state-of-the-art ion implantation. Further, the sensing behavior of such nanoflakes for non-enzymatic detection of H2O2 was investigated. The results showed that the increase in surface area increased the number of available VO •• and thus increased the sensitivity and peak current densities. Further, microcracking was observed in Mo- and V-doped samples after annealing in N2 that would facilitate increased exposed surface and maximisation of the redox peak currents. Accordingly, XPS analysis showed that the [VO ••] was the highest in the Mo- and V-doped samples at 8.00 at% and 8.13 at% respectably and also these samples exhibited the highest peak currents for redox reduc- tion. The lowest [VO ••] was achieved for the air annealed sample, which also had the lowest peak current densities for the H2O2 redox reaction suggesting the critical role of VO •• on the biosensing performance. The sensing outcomes showed that the V-doped CeO2-x at low H2O2 concentrations exhibited lowest LOD and highest sensitivity among all the electrodes at 0.015 mM and 4.31761 × 10−4 A.mM−1. However, it only had a linear range of 0.05–0.5 mM limiting its applications. The Mo-doped CeO2-x performed well with a linear range across the entire tested concentrations and demonstrated a sensitivity of 7.43849 × 10−5 A.mM−1 and LOD of 0.088 mM which were a significant improvement in capabilities compared to the undoped samples. The H2O2 concentration range in the current work (0.05–10 mM) covered both of the physiological range and the environmental range.8–10 Moreover, the present biosensor is able to detect 50 nA in the presence of 1 μM of H2O2 per cm2. Thus, it can be concluded that this biosensor is suitable for the application. This shows that ion implantation can be used as an advanced technique to achieve low- levels of controlled surface doping to enhance the performance and stability of metal-oxide based nanostructured biosensors. Experimental Electrolyte fabrication.—1 M Ce(NO3)3 solution was prepared by adding 200 ml of deionised water to 86.8 g of Ce(NO3)3.6H2O salt (Sigma Aldrich, 99% purity). 1 M NH4NO3 solution was prepared by adding 200 ml of deionised water to 16.01 g of ammonium nitrate (Sigma Aldrich, 99% purity). 1 M NaCl solution was prepared by adding 200 ml of deionised water to 11.7 g of sodium chloride (Sigma-Aldrich). 1 M NaOH solution was prepared by adding 200 ml of deionised water to 36.0 g of sodium hydroxide pellets (Chem Supply). The latter was used to control the pH of the electrolyte immediately prior to electrodeposition. These concentrated 1 M solutions were diluted using deionised water (3 ml of 1 M Ce(NO3)3 solution, 3 ml of the 1 M NH4NO3 solution, and 1.2 ml of 1 M NaCl solution) to yield a 60 ml aqueous-based electrolyte. Electrodeposition.—The cathodic chronopotentiometry electro- deposition (CCE) was carried out at different temperatures of 45°, 55°, and 65 °C in two acidic pH ranges of 4.5–5.0 (acute acidic) and 5.5–6.0 (mild acidic). The CCE was done using EZStat Pro potentiostat/galvanostat (Nuvant USA), with the current resolution of −0.003 A. The CCE involved a three-electrode configuration system including FTO glass, platinum wire coil (BASi), and Ag/ AgCl electrode (BASi) which were used as working, counter, and reference electrodes, respectively. Prior to the CCE, FTO glasses were thoroughly washed with deionised water and ethanol to ensure that no residual contaminants remained after the ultrasonic cleaning. The platinum coil and reference electrode were similarly thoroughly washed with deionised water. The 60 ml electrolyte was placed on a magnetic stirrer and heated to the desired temperature. The pH and temperature of the electrolyte were constantly controlled and monitored using a Horiba LAQUA pH/ION/COND meter. To control the pH, 1 M sodium hydroxide solution was added dropwise. All the electrodes were immersed in the electrolyte to a depth of ∼20 mm. After electrodeposition, the sample was gently cleaned with deionised water, labelled with a permanent marker on the non- deposited region of the glass, and placed in a vacuum desiccator for 48 h to dry. Ion implantation.—Ion implantation was done using the low energy ion implantation (LEII) beamline attached to the SIRIUS 6 MV accelerator at the Australian Nuclear Science and Technology Organisation (ANSTO). Prior to the experiment, Dynamic-TRIM software code was used to determine the required implantation energy for a peak implantation depth of 20–40 nm. An accelerating energy of 10 keV and dosage of 1 × 1015 atoms.cm−2 was chosen for implantation. A Penning sputter source was used to generate and extract the required ions from solid cathode targets (Mo and V). The resulting ions were mass separated through a 90° electromagnet and subsequently focused by electrostatic Einzel and quadrupole lenses before the beam was electrostatic raster-scanned over the sample to perform implantation. This was performed under vacuum (1 × 10−6 Pa). Annealing.—The samples were placed on zircon crucibles and annealed in air (using Kanthal-wound muffle furnace) and in N2 (using Labec tube furnace). For annealing, the heating and cooling rates were 2 °C.min−1 and the dwell time was 4 h at 400 °C. Materials characterisation.— Scanning electron microscopy (SEM).—The morphology of the deposited samples was studied using SEM imaging by Nova NanoSEM 450 FE-SEM (FEI, USA). The samples were adhered to 20 mm sample-holder stubs using circular carbon tape. The top of the deposited CeO2-x was then attached by copper tape to the bottom of the sample holder to ensure a good electrical connection. The samples were then coated with a 15 nm thick layer of platinum using a Quorum Q300TD sputter coater with an argon gas pressure reading of 80 KPa. The SEM imaging was done under 60 Pa chamber pressure, 5 kV voltage, and working distance of 5 mm. The thickness of the average nanoflake was calculated by using the relative length compared to the scale bar. Thin film X-ray diffraction (XRD).—Mineralogical data for the nanostructures were obtained using a Malvern PANalytical, UK with CuKα radiation, 45 kV, 40 A, scan range 10°− 80° 2θ, and scan step size of 0.039°. The peaks were analysed using X’Pert High Score Plus software (Malvern, UK). A Cu manual beam attenuator was used to align each sample. Journal of The Electrochemical Society, 2023 170 057519 Raman spectroscopy.—Raman spectroscopy was also used to analyse the crystallinity of the deposited samples using Renishaw inVia confocal Raman microscope (Gloucestershire, UK) equipped with a helium-neon green laser (514 nm) and diffraction grating of 1800 grooves/mm. All Raman data were recorded over the range 100–3200 cm−1. X-Ray photoelectron spectroscopy (XPS).—XPS was used for quantitative measurement of the defects and elemental analysis of the samples. Surface analysis of the samples was conducted using a Thermo Fisher Scientific ESCALAB 250Xi spectrometer, UK, equipped with a mono-chromated Al K alpha (E = 1486.68 eV) hemispherical analyser. The chamber pressure during the analysis was kept constant at <2 × 10−9 mbar. The measured binding energies were referenced to the C1s doublet corrected to 284.80 eV and the spectra were deconvoluted using Lorentzian and Gaussian profiles. Application testing.—The H2O2 sensing tests were done using cyclic voltammetry (CV) in phosphate buffered saline (PBS) electrolyte in a similar set up as the CCE. 30 wt% hydrogen peroxide (UNIVAR) was used to make up a solution of 0.5 M H2O2. This solution was further diluted for making PBS with 0, 0.05, 0.1, 0.5, 1, 2, 5, 10 mM H2O2. These solutions were tested at scan rates of 10, 20, 50, 100, 200 mV.s−1 using the instrument Autolab PGSTAT128N potentiostat/galvanostat (Metrohm, Switzerland). The testing was done in order of ascending H2O2 concentrations to minimise contamination due to trapped H2O2 molecules on the CeO2-x/FTO surface. Additionally, the deposited samples were thoroughly washed with deionised water when changing solutions. Acknowledgments This work was supported by the Australian Research Council (DP170104130). The authors acknowledge support from the Australian Government’s National Collaborative Research Infrastructure Strategy, NCRIS, for access to the low energy ion implantation facility in the Centre for Accelerator Science at ANSTO. Further, the authors acknowledge the subsidised use of faciltiies provided by the Mark Wainwright Analytical Centre, UNSW Sydney. Author Contributions Y.L. prepared the initial draft, analysed the data, and worked on all subsequent drafts of the manuscript. X.Z. analysed the data, worked on the structure of the manuscript, and revised all the subsequent drafts of the manuscript. C.V.V. assisted with the design of the project, undertook the majority of syntheses, characterisation and data analysis. N.H. assisted with experimental work and analysis of application data. A.J.A. and M.M. designed and conducted the ion-implantation and provided the corresponding data. H.A. and Y. W. assisted with the draft preparation, data analysis, and worked on all subsequent drafts of the manuscript. C.C.S and S.S.M, and P.K. designed the project, supervised the overall project, contributed to the data analyses, and revised all drafts of the manuscript. Data Availability Statement The data supporting the findings of this study are available from the corresponding authors upon request. ORCID Xiaoran Zheng https://orcid.org/0000-0002-7016-5260 Sajjad S. Mofarah https://orcid.org/0000-0002-4835-767X References 1. A. Tauffenberger and P. J. Magistretti, Neurochem. Res., 46, 77 (2021). 2. S. H. Lee, M. K. Gupta, J. B. Bang, H. Bae, and H. Sung, Adv. Healthcare Mater., 2, 908 (2013). 3. P. P. Fu, Q. Xia, H.-M. Hwang, P. C. Ray, and H. Yu, Journal of food and drug analysis, 22, 64 (2014). 4. G.-Y. Liou and P. Storz, Free Radical Res., 44, 479 (2010). 5. K. Hensley, K. A. Robinson, S. P. Gabbita, S. Salsman, and R. A. Floyd, Free Radical Biol. Med., 28, 1456 (2000). 6. M. L. Circu and T. Y. Aw, Free Radical Biol. Med., 48, 749 (2010). 7. S. Saikolappan, B. Kumar, G. Shishodia, S. Koul, and H. K. Koul, Cancer letters, 452, 132 (2019). 8. H. J. Forman, A. Bernardo, and K. J. A. Davies, Arch. Biochem. Biophys., 603, 48 (2016). 9. K. Tsukimori, T. Yoshitomi, S. Morokuma, K. Fukushima, and N. Wake, Am J Hypertens, 21, 1343 (2008). 10. R. C. Spear and S. Selvin, Risk Analysis, 9, 579 (1989). 11. J. A. Goode, J. V. H. Rushworth, and P. A. Millner, Langmuir, 31, 6267 (2015). 12. S.-H. Yeom, B.-H. Kang, K.-J. Kim, and S.-W. Kang, Frontiers in Bioscience- Landmark, 16, 997 (2011). 13. A. K. Yagati and J. Choi, Electroanalysis, 26, 1259 (2014). 14. Y. Zhang, X. Bai, X. Wang, K. K. Shiu, Y. Zhu, and H. Jiang, Anal. Chem., 86, 9459 (2014). 15. U. Nestor, H. Frodouard, and M. Theoneste, Advances in Nanoparticles, 10, 1 (2020). 16. I. S. Kucherenko, O. O. Soldatkin, D. Y. Kucherenko, O. V. Soldatkina, and S. V. Dzyadevych, Nanoscale Advances, 1, 4560 (2019). 17. G. Luka, S. Ahmad, N. Falcone, and H.-B. Kraatz, Bioelectronics and Medical Devices, ed. K. Pal et al. (Sawston, Woodhead Publishing) p. 555 (2019). 18. S. D. Bukkitgar, S. Kumar, S. Singh, V. Singh, K. R. Reddy, V. Sadhu, G. B. Bagihalli, N. P. Shetti, C. V. Reddy, and K. Ravindranadh, Microchem. J., 159, 105522 (2020). 19. Y.-H. Wang, K.-J. Huang, and X. Wu, Biosens. Bioelectron., 97, 305 (2017). 20. G. Gumilar, Y. V. Kaneti, J. Henzie, S. Chatterjee, J. Na, B. Yuliarto, N. Nugraha, A. Patah, A. Bhaumik, and Y. Yamauchi, Chem. Sci., 11, 3644 (2020). 21. N. Salarizadeh, M. Habibi-Rezaei, and S. J. Zargar, Mater. Chem. Phys., 281, 125870 (2022). 22. K. Wang et al., Sens Actuators B Chem, 355, 131298 (2022). 23. S. Chauhan and C. Sharma, ChemistrySelect, 7, e202203223 (2022). 24. P. R. Solanki, A. Kaushik, V. V. Agrawal, and B. D. Malhotra, NPG Asia Mater., 3, 17 (2011). 25. F. Charbgoo, M. Ramezani, and M. Darroudi, Biosens. Bioelectron., 96, 33 (2017). 26. C. Fu, Y. Sun, C. Huang, F. Wang, N. Li, L. Zhang, S. Ge, and J. Yu, Talanta, 223, 121719 (2021). 27. Z.-H. Yang, S. Ren, Y. Zhuo, R. Yuan, and Y.-Q. Chai, Anal. Chem., 89, 13349 (2017). 28. Q. Zhang, G. Fan, W. Chen, Q. Liu, X. Zhang, X. Zhang, and Q. Liu, Biosens. Bioelectron., 150, 111846 (2020). 29. X. Tian, Z. Wang, M. Ding, S. Zhou, R. Ouyang, and Y. Miao, Int. J. Electrochem. Sci., 15, 310 (2020). 30. K. R. B. Singh, V. Nayak, T. Sarkar, and R. P. Singh, RSC Adv., 10, 27194 (2020). 31. W. Chen, S. Cai, Q.-Q. Ren, W. Wen, and Y.-D. Zhao, Analyst, 137, 49 (2012). 32. L. Wang, Y. Li, L. Zhao, Z. Qi, J. Gou, S. Zhang, and J. Z. Zhang, Nanoscale, 12, 19516 (2020). 33. A. Singh, A. Sharma, A. Ahmed, A. K. Sundramoorthy, H. Furukawa, S. Arya, and A. Khosla, Biosensors, 11, 336 (2021). 34. A. K. Yagati, T. Lee, J. Min, and J.-W. Choi, Biosens. Bioelectron., 47, 385 (2013). 35. Y. Zhou, S. D. Uzun, N. J. Watkins, S. Li, W. Li, A. L. Briseno, K. R. Carter, and J. J. Watkins, ACS Appl. Mater. Interfaces, 11, 1821 (2018). 36. P. Guan, Y. Li, J. Zhang, and W. Li, Nanomaterials, 6, 159 (2016). 37. S. S. Mofarah et al., Nanoscale, 13, 6764 (2021). 38. I. Kosacki, T. Suzuki, H. U. Anderson, and P. Colomban, Solid State Ionics, 149, 99 (2002). 39. C. Schilling, A. Hofmann, C. Hess, and M. V. Ganduglia-Pirovano, J. Phys. Chem. C, 121, 20834 (2017). 40. Z. Wu, M. Li, J. Howe, H. M. Meyer III, and S. H. Overbury, Langmuir, 26, 16595 (2010). 41. M. Gupta, S. C. Shirbhate, O. V. Rambadey, S. A. Acharya, and P. R. Sagdeo, ACS Appl. Energy Mater., 5, 9759 (2022). 42. S. S. Mofarah et al., Nat. Commun., 10, 1 (2019). 43. X. Zheng et al., Adv. Funct. Mater., 31, 2103171 (2021). 44. J. Feng, S. Ye, H. Xu, Y. Tong, and G. Li, Adv. Mater., 28, 4698 (2016). 45. T. Dai et al., Adv. Mater., 33, 2102593 (2021). 46. S. I. Han, S. Lee, M. G. Cho, J. M. Yoo, M. H. Oh, B. Jeong, D. Kim, O. K. Park, J. Kim, and E. Namkoong, Adv. Mater., 32, 2001566 (2020). 47. Y. Liu, C. Ma, Q. Zhang, W. Wang, P. Pan, L. Gu, D. Xu, J. Bao, and Z. Dai, Adv. Mater., 31, 1900062 (2019). 48. X. Zheng et al., ACS Appl. Mater. Interfaces, 13, 59820 (2021). 49. S. S. Mofarah et al., Adv. Mater., 31, 1905288 (2019). 50. S. Maroufi, R. Khayyam Nekouei, S. S. Mofarah, A. P. O’Mullane, Y. Yao, S. Lim, C. Cazorla, and V. Sahajwalla, Adv. Funct. Mater., 31, 2100880 (2021). 51. S. S. Mofarah et al., J. Mater. Chem. A, 8, 4753 (2020). 52. H. Arandiyan, S. S. Mofarah, C. C. Sorrell, E. Doustkhah, B. Sajjadi, D. Hao, Y. Wang, H. Sun, B. J. Ni, and M. Rezaei, Chem. Soc. Rev., 50, 10116 (2021). 53. S. S. Mofarah, R. K. Nekouei, S. Maroufi, S. Biswal, S. Lim, Y. Yao, and V. Sahajwalla, Nanoscale, 13, 3662 (2021). 54. J. S. Milićević, M. S. Ranđelović, M. Z. Momčilović, A. R. Zarubica, S. S. Mofarah, B. Matović, and C. C. Sorrel, Microchim. Acta, 187, 1 (2020). Journal of The Electrochemical Society, 2023 170 057519 https://orcid.org/0000-0002-7016-5260 https://orcid.org/0000-0002-4835-767X https://doi.org/10.1007/s11064-020-03208-7 https://doi.org/10.1002/adhm.201200423 https://doi.org/10.1016/j.jfda.2014.01.005 https://doi.org/10.1016/j.jfda.2014.01.005 https://doi.org/10.3109/10715761003667554 https://doi.org/10.1016/S0891-5849(00)00252-5 https://doi.org/10.1016/S0891-5849(00)00252-5 https://doi.org/10.1016/j.freeradbiomed.2009.12.022 https://doi.org/10.1016/j.canlet.2019.03.020 https://doi.org/10.1016/j.abb.2016.05.005 https://doi.org/10.1038/ajh.2008.289 https://doi.org/10.1038/ajh.2008.289 https://doi.org/10.1111/j.1539-6924.1989.tb01268.x https://doi.org/10.1021/la503533g https://doi.org/10.2741/3731 https://doi.org/10.2741/3731 https://doi.org/10.1002/elan.201400037 https://doi.org/10.1021/ac5009699 https://doi.org/10.4236/anp.2021.101001 https://doi.org/10.1039/C9NA00491B https://doi.org/10.1016/B978-0-08-102420-1.00029-7 https://doi.org/10.1016/B978-0-08-102420-1.00029-7 https://doi.org/10.1016/j.microc.2020.105522 https://doi.org/10.1016/j.bios.2017.06.011 https://doi.org/10.1039/C9SC05636J https://doi.org/10.1016/j.matchemphys.2022.125870 https://doi.org/10.1016/j.snb.2021.131298 https://doi.org/10.1002/slct.202203223 https://doi.org/10.1038/asiamat.2010.137 https://doi.org/10.1016/j.bios.2017.04.037 https://doi.org/10.1016/j.talanta.2020.121719 https://doi.org/10.1021/acs.analchem.7b03502 https://doi.org/10.1016/j.bios.2019.111846 https://doi.org/10.1016/j.bios.2019.111846 https://doi.org/10.20964/2020.10.12 https://doi.org/10.20964/2020.10.12 https://doi.org/10.1039/D0RA04736H https://doi.org/10.1039/C1AN15738H https://doi.org/10.1039/D0NR05746K https://doi.org/10.3390/bios11090336 https://doi.org/10.1016/j.bios.2013.03.035 https://doi.org/10.1021/acsami.8b16985 https://doi.org/10.3390/nano6090159 https://doi.org/10.1039/D0NR08097G https://doi.org/10.1016/S0167-2738(02)00104-2 https://doi.org/10.1021/acs.jpcc.7b06643 https://doi.org/10.1021/acs.jpcc.7b06643 https://doi.org/10.1021/la101723w https://doi.org/10.1021/acsaem.2c01442 https://doi.org/10.1021/acsaem.2c01442 https://doi.org/10.1038/s41467-019-10621-2 https://doi.org/10.1002/adfm.202103171 https://doi.org/10.1002/adma.201600054 https://doi.org/10.1002/adma.202102593 https://doi.org/10.1002/adma.202001566 https://doi.org/10.1002/adma.201900062 https://doi.org/10.1002/adma.201900062 https://doi.org/10.1021/acsami.1c14484 https://doi.org/10.1002/adma.201905288 https://doi.org/10.1002/adfm.202100880 https://doi.org/10.1039/C9TA11961B https://doi.org/10.1039/D0CS00639D https://doi.org/10.1039/D0NR06880B https://doi.org/10.1007/s00604-020-04406-4 55. S. SAMESHIMA, M. KAWAMINAMI, and Y. HIRATA, J. Ceram. Soc. Jpn., 110, 597 (2002). 56. M. Fukawa, K. Sato, T. Tsukamoto, K. Adachi, and H. Nishimura, Sol. Energy Mater. Sol. Cells, 49, 107 (1997). 57. R. D. t Shannon and C. T. Prewitt, Acta Crystallographica section B: Structural Crystallography and Crystal Chemistry, 26, 1046 (1970). 58. A. Uzunoglu and L. A. Stanciu, Anal. Chim. Acta, 909, 121 (2016). 59. J. Hao, S. Peng, H. Li, S. Dang, T. Qin, Y. Wen, J. Huang, F. Ma, D. Gao, and F. Li, J. Mater. Chem. A, 6, 16094 (2018). 60. H. Yu, J. Guo, C. Wang, J. Zhang, J. Liu, G. Dong, X. Zhong, and X. Diao, Electrochim. Acta, 332, 135504 (2020). 61. G. Wang, J.-J. Xu, H.-Y. Chen, and Z.-H. Lu, Biosens. Bioelectron., 18, 335 (2003). 62. V. Narwal, N. Yadav, M. Thakur, and C. S. Pundir, Biosci. Rep., 37, BSR20170194 (2017). 63. Y. Ge, Y. Wang, M. Wang, H. Wang, and Y. Huang, Int. J. Electrochem. Sci., 12, 6588 (2017). 64. Y. Zhang, C. Zhang, D. Zhang, M. Ma, W. Wang, and Q. Chen,Mater. Sci. Eng. C, 58, 1246 (2016). 65. H. Ullah, F. Usman, M. A. Rasheed, A. A. Khan, R. Ahmad, S. Y. Lee, J. Heo, G. Ali, M. Khan, and S. O. Cho, ACS Appl. Nano Mater., 5, 9096 (2022). 66. A. T. E. Vilian, S.-M. Chen, and B.-S. Lou, Biosens. Bioelectron., 61, 639 (2014). Journal of The Electrochemical Society, 2023 170 057519 https://doi.org/10.2109/jcersj.110.597 https://doi.org/10.1016/S0927-0248(97)00183-9 https://doi.org/10.1016/S0927-0248(97)00183-9 https://doi.org/10.1107/S0567740870003576 https://doi.org/10.1107/S0567740870003576 https://doi.org/10.1016/j.aca.2015.12.044 https://doi.org/10.1039/C8TA06349D https://doi.org/10.1016/j.electacta.2019.135504 https://doi.org/10.1016/S0956-5663(02)00152-5 https://doi.org/10.1042/BSR20170194 https://doi.org/10.20964/2017.07.73 https://doi.org/10.1016/j.msec.2015.09.015 https://doi.org/10.1021/acsanm.2c01397 https://doi.org/10.1016/j.bios.2014.05.023