Modulating Ce3+ Sites in Ce‐Zr Oxide Nanocatalysts through Protamine Biomineralization for Organophosphate Dephosphorylation

Catalytic is a crucial reaction for environmental detoxication of pesticides and neutralization of various molecules classified as chemical warfare agents. Herein, we report on a series of tunable Ce−Zr‐based metal oxides, (ZraCe1–aOx) prepared using a facile biomineralization technique, as catalysts for organophosphates dephosphorylation. Synchrotron scattering and spectroscopy methods showcase that ZraCe1–aOx catalysts are highly defective and exhibit an abundance of Ce3+ sites that promote oxygen vacancies needed for enhanced dephosphorylation reactions. The catalytic performance was assessed using a model para‐nitrophenyl phosphate reaction and showcases a strong dependence on Zr dopant concentration and subsequent tuning of the Ce3+/Ce4+ ratio. Analysis of synchrotron datasets allowed structure‐performance correlations between the Ce3+ concentration and associated oxygen vacancies, the dephosphorylation rate constant, and Zr concentration to be established, confirming that Ce3+ as active sites is positively correlated with the rate constant. We envision that similar biomineralization approaches can be used to fabricate Ce3+‐rich Ce−Zr oxide for environmental application in dephosphorylation and other hydrolysis reactions.


Introduction
Organophosphates (OP) are chemicals that can have exceptionally high toxicity and inertness that are widely used as agricultural insecticides and chemical warfare agents.[3] Additionally, the weaponization of highly lethal OP as chemical warfare agents necessitates strategies for capture and eventual neutralization of these compounds. [1,4]The degradation rate of OP in natural environments is low, largely relying on a limited number of microorganisms that can gradually biodegrade them to inert substances through enzymatic processes. [5][8] Among these oxides, cerium oxide demonstrates a capacity for increasing the rate of organophosphate dephosphorylation (OPD). [6,9,10]Ceria can exist in dual oxidation states as Ce 3 + and Ce 4 + , enabling the ability to accommodate a high density of oxygen vacancies (Vo), which enhances the catalytic activity for OPD due to the cooperative redox property of Ce 3 + and Ce 4 + . [1,6,9,11][22][23] Biomineralization refers to the process in natural microorganisms which utilize protein or peptide templates to synthesize metal oxides in an ambient aqueous condition and neutral pH. [24]][26] To select an inexpensive and abundant biomimetic analogue, protamine is an alluring protein scaffold, as it can be sourced from upcycled fishing industry waste.Protamine contains positively charged arginine groups with the high isoelectric point needed for condensation of metal oxides via biomineralization-based methods and has been previously used to template a range of metal oxides. [20,21,26,27]Though biomineralization pathways have been used to synthesize metal oxides, there are fewer studies on mixed metal systems especially involving ceria-based composites.Given ceria's propensity to form oxygen vacancies readily by increasing the lattice constant relative to its Ce 3 + concentration, [28,29] introducing a dopant-species like Zr can potentially stabilize these vacancy sites.Zr-dopant atoms can introduce a strain the crystal lattice due the size mismatch, making it difficult for oxygen atoms to diffuse into the vacancy sites. [30,31]Consequently, increasing the available surface-active sites for catalytic activity, making this ceria-zirconia pairing a promising candidate for the biomineralization approach.
In this study, protamine is used to template biomineralized CeÀ Zr oxide (Zr a Ce 1-a O x ) nanoparticles with varying Ce : Zr compositions.It was found that material properties, such as surface area, electronic structure, and overall structural coherency can be manipulated by Ce : Zr composition through biomineralization pathways.Extensive characterization was used to identify the presence of Ce 3 + defect centers and Vo in Zr a Ce 1-a O x which were found to directly influence OPD in aqueous conditions.para-Nitrophenyl phosphate (p-NPP) was used as the model substrate.The study also established synthesis-structure-property relationships that can be leveraged for future optimized catalyst design.We envision that similar biomineralization approaches can be used to fabricate Ce 3 + -rich oxide catalyst for a range of hydrolysis reactions and other catalytic processes as well.

Results and Discussion
Main Text Paragraph.Biomineralization is a straightforward and benign approach to synthesize oxide nanoparticles under ambient conditions.The positively charged amino groups in protamine enable controllable synthesis for oxide nanoparticles in aqueous solution via a mechanism involving amine-catalyzed polycondensation. [26,27]The resulting nanomaterials often comprise a high percentage of protamine, most notably quantified with biomineralized SiO 2 and TiO 2 . [26,27]For catalytic applications, protamine provides no added functionality and further restricts access to potential active cites for the OPD reactions.Previous work has showcased simple acid treatment protocols to remove protamine-metal oxide interactions without overly perturbing the abiotic material. [32]Here, protamine was used to biomineralize CeÀ Zr oxides (Zr a Ce 1-a O x ) as detailed in the Experimental Section.For nomenclature purposes, the original "neat" biomineralization materials include an "_N" in the title (e. g., Ce(III)O x _N), while acid treated materials include an "_A" (e. g., 1ZrCe(III,IV)O x _A).
For CeÀ Zr oxides, the nominal Zr concentration in mole percentage implemented in the synthesis is provided in front of the abbreviation (e. g., 1ZrCe(III,IV)O x _A for 1 mol % Zr).The nominal ratio of the Ce precursor is labeled as (III), (IV), and (III,IV) -for mixing 20 % Ce(III) and 80 % Ce(IV), as this approximate ratio has shown as suitable for OPD reactivity. [1]igure 1a illustrates the biomineralization synthesis protocol for the Zr a Ce 1-a O x materials, including acid treatment conditions for protamine removal, with transmission electron microscopy (TEM) images in Figure 1b and Zr a Ce 1-a O x composites display aggregate sizes between 30-40 nm, with the irregular particle shape typical of biomineralization.[20,21] These aggregates commonly consist of smaller clusters of inorganic material (~5 nm) that get constrained by the templating protamine during mineralization reactions.[22,26] Lattice fringes can be observed for biomineralized CeO x (Figure 2c, focused area), indicative of nanoscale ordering.For comparison, 50ZrCe(III,IV)O x _A appears largely amorphous, which is consistent with previous reports on biomineralized ZrO x .[21,33] The fast Fourier transform (FFT) images (Figure 1f-1i) further corroborate the visual observations of relative atomic ordering, with materials comprising a Ce majority exhibiting bright spots in the FFT and 50ZrCe(III,IV)O x _A exhibiting a diffuse ring.The TEM and FFT results imply that Zr inclusion has a substantial influence on CeO x crystallization, likely due to the formation of disordered CeÀ Zr O x ide during the biomineralization process.STEM-EDS analysis of 1ZrCe(III,IV)O x _A (Figure S1) indicates a homogeneous distribution of Ce and O species within the material, with only a small amount of Zr observed as the nominal Zr percentage is 1 mol %.In addition, nitrogen (N) and carbon(C) species are observed that likely arise from residual protamine.
The specific surface area (SSA-Brunauer-Emmett-Teller (BET)), pore volume and average pore diameter of each acidtreated material, measured using N 2 physisorption, are shown in Table 1.The SSA of Ce(III,IV)O x _A is twice that of Ce(III)O x _A and Ce(IV)O x _A (115.1 m 2 /g vs 50.8 m 2 /g and 50.6 m 2 /g, respectively).The higher SSA is likely influenced by the mineralization  of Ce(III) and Ce(IV) precursors and subsequent changes to the oxidation state caused by biomineralization or acid treatment due to the presence of different local Ce 3 + and Ce 4 + structural motifs. [34,35]For Zr a Ce 1-a O x materials, the surface areas decrease with increasing Zr concentration, with 1ZrCe(III,IV)O x _A, 10ZrCe-(III,IV)O x _A, and 50ZrCe(III,IV)O x _A exhibiting surface areas of 107.5 m 2 /g, 98 m 2 /g, and 94.6 m 2 /g, respectively, while the pure phase Zr oxide has a surface area of 111.8 m 2 /g.Pore size analysis for all materials shows in a pore size of ~6 nm, which is consistent with metal oxides made via biomineralization. [36,37]hermogravimetric analysis (TGA) was used to quantify the protamine content after acid treatment.As shown in Table 1, the Zr rich materials (i.e,.ZrO x _A, 50ZrCe(III,IV)O x _A) experience a greater than 25 % mass loss, which is higher than the other materials (~15 %).The finding suggests the ZrO x has a higher binding affinity to protamine compared to CeO x , potentially explaining the lack of order in ZrO x compared to CeO x .This likely impacts the structural order of Zr a Ce 1-a O x relative to the Zr concentration.
Synchrotron characterization methods were used to better understand the local atomic and electronic structure of the biomineralized catalysts.Given the lack of order in Zr-based biomineralized materials, high-energy X-ray diffraction (HE-XRD) coupled to atomic pair distribution function analysis was performed, as both Bragg and diffuse features are converted into a series of atomic pair distances irrespective of long range order. [38]HE-XRD patterns (Figure S2) exhibit strong diffraction peaks for Zr a Ce 1-a O x materials with the exception of 50 ZrCe-(III,IV)O x , which appears less ordered, consistent with TEM observations.HE-XRD patterns of the biomineralized materials are compared against the known XRD pattern of CeO 2 , with the dominant crystallographic plane found in Zr a Ce 1-a O x being the (111) plane.In comparison, ZrO x and 50 ZrCe(III,IV)O x show reduced diffraction peaks which suggests that the crystallinity decreased with higher Zr : Ce ratio.Compared to other Zr a Ce 1-a O x materials, the crystallinity decreases significantly for 50 ZrCe(III,IV)O x .This observation suggests the introduction of the ZrO x component increases the disorder of the CeO x component in 50 ZrCe(III,IV)O x , which is likely due to the homogeneous distribution of Ce and Zr which forms a solid solution structure in 50 ZrCe(III,IV)O x . [39]tomic PDFs account for both Bragg and diffuse features and therefore provide a complete atomic-scale depiction of the materials. [38,40]Atomic PDFs are obtained by Fourier transform reduced structure functions (F(Q), converted from HE-XRD shown in (Figure S2). Figure 2a-2b shows PDFs for all Zr a Ce 1-a O x materials, which exhibit multiple atomic pairs up to 20 Å. ZrO x _ N and ZrO x _A, in contrast, display a highly disordered structure with atomic pair peaks rapidly decaying to zero after 8 Å. Atomic pair distances are assigned against known crystal structures for CeO 2 and Ce 2 O 3 (from 1.7 to 4.5 Å) to help establish predominate local phases, as shown in Figure 2c-2d. [19]Compared to CeO 2 , the atomic pairs show two main peaks attributed to CeÀ O at 2.5 Å and Ce(IV)À Ce(IV) at 3.8 Å (red line on Figure 2), [19] where atomic pairs attributed to Ce(III)À Ce(III) are assigned to 4.2 Å. [41] For pristine materials, the ratio of Ce(III)À Ce(III) vs. Ce(IV) À Ce(IV) atomic pair peak intensity is consistent with the nominal ratio of Ce 3 + vs. Ce 4 + in the materials, which is further confirmed by XANES analysis (see below), which illustrates the ability of the biomineralization approach to produce Vo-rich CeO x . [21]or ZrO x , atomic pairs of ZrÀ O and ZrÀ Zr are assigned to 2.1 Å and 3.5 Å (black line on Figure 2) and closely resemble tetragonal ZrO 2 . [42,43]The atomic pair distances of ZrÀ O and ZrÀ Zr peaks for 50 ZrCe(III,IV)O x _N shift relative to pure ZrO x _N, hinting that 50 ZrCe(III,IV)O x _N materials exists as a ceriazirconium solid solution, consistent with the PDF analysis. [19]For the acid treated materials, different trends are shown for Zr a Ce 1-a O x materials and Ce(III,IV)O x mixed materials compared to the control materials Ce(III)O x _A and Ce(IV)O x _A.For the mixed materials, the peak intensities of Ce(III)À Ce(III) experience a significant increase compared to the pristine material, with a simultaneous decrease in Ce(IV)À Ce(IV) intensities, reflecting the reduction from Ce 4 + to Ce 3 + .
Based on a prior study, free electrons from oxygen vacancies in cerium oxides can easily transfer between the M 3 + and M 4 + species, [44] in our case the transfer is enabled by the Ce 3 + À OÀ Ce 4 + /Zr 4 + bond reducing the Ce 4 + to Ce 3 + .Following the acid treatment, there is increased surface concentration of oxygen vacancies due to the partial dissolution of the M 4 + cation.Each vacancy containing 2 unpaired electrons that can migrate to electron deficient Ce 4 + and Zr 4 + sites causing the increase in Ce(III)À Ce(III) and decrease in Ce(IV)À Ce(IV). [34,35,45]he higher pore volume and SSA could be related to the dissolution of M 4 + and the formation of Ce 3 + ions during acid treatment that cause expansion of the lattice structure. [34,35,46]e L3 X-ray absorption near edge structure spectroscopy (XANES) was employed to obtain electronic structure information of the biomineralized materials, as shown in Figure 3a-b.The white line maximum for Ce 3 + is positioned at 5725 eV, while XANES for Ce 4 + exhibit two features at 5736 eV and 5728 eV (shoulder peaks at 5728 eV), in agreement with previous reports. [47,48]As shown in Figure 3a-3b, the biomineralized materials exhibit features indicative of both Ce 3 + and Ce 4 + oxidation states.Linear combination fitting (LCF) was applied to better understand the influence of precursor composition on the final oxidation state of Ce (Figure S3). [49]As shown in Figure 4 and Table S2, the fraction of Ce 3 + in the order Ce(III)O x _N > Ce(III,IV)O x _N > Ce(IV)O x _N for the as-mineralized materials, which is consistent with observations from PDF analysis.With the increasing percentage of Zr, the percentage of Ce 3 + displays a small increase, as shown in all Zr a Ce 1-a O x materials compared to Ce(III,IV)O x _N.The enhanced reduction of Zr a Ce 1-a O x over Ce(III,IV)O x can be accredited to the Ce 3 + defect sites that are accommodated by Zr 4 + ions, which have a lower coordination number compared to Ce. [50] Theorical studies have confirmed that the oxygen vacancy formation energy decreases near the Zr cation in CeÀ Zr oxide with low Zr content, while the high Zr doping may result in a less/non-oxygen-defective CeÀ Zr oxide solid solution, as seen in our PDF data. [51,52]As such, the 1 ZrCe(III,IV)O x exhibits highest overall concentration of the Ce 3 + and Vo, as the low content Zr 4 + stabilises such defect sites in the CeÀ Zr oxide system. [21,50].After acid treatment, Zr a Ce 1-a O x mixed materials and Ce(III,IV)O x mixed materials exhibit noticeable increases in Ce 3 + concentration compared to single Ce(III)O x or Ce(IV)O x .This may lead to a greater lattice expansion as the ionic radius increases from Ce 4 + to Ce 3 + .
The trend corresponds to the BET and PDF findings further confirms the necessity of having the mixed M 3 + and M 4 + valence states and low Zr content in formation of Ce 3 + and oxygen vacancies.The increase in Ce 3 + and Vo defect sites can offer higher reducibility to enhance the catalytic performance, to accommodate a higher concentration of Vo and induce Ce 3 + -bound oxygen nucleophiles (CeÀ OÀ ). [1,53].The partial reduction of Zr(4-n) + in CeÀ Zr O x ide has been reported in the literature, [1,47] suggesting that the Zr oxidation state is influenced by the formation of the nearby oxygen vacancy sites as discussed previously.The XANES findings for 50 ZrCe(III,IV)-O x _N and 50 ZrCe(III,IV)O x _A also strongly suggest that Zr atoms are more strongly localized at positions neighboring the Vo. [1,54]ear edge X-ray absorption fine structure (NEXAFS) measurements at the O K-edge were conducted to better understand the oxygen configuration surrounding the Ce and Zr central ions for the acid treated materials, as shown in Figure 5.For the O K-edge NEXAFS, 1 s-π* electron transitions occurring between the oxygen 1 s orbital to unoccupied orbitals, where multiple features in NEXAFS are due to the ligand field splitting of the Ce or Zr d-orbital.This results in characteristic peaks related to the CeÀ O local symmetry at ~534 eV for eg orbital and ~538 eV for t 2g orbital. [55,56]It is clear the peak for t 2g orbital is only observed for Ce(III)O x _A and Ce(IV)O x _A materials, and not for Ce(III,IV)O x and Zr a Ce 1-a O x materials.
This forbidden electron transition shows that extra electrons could occupy the t 2g molecular orbital for Ce(III,IV)O x and Zr a Ce 1-a O x , which suggests the existence of a Ce 3 + -bound oxygen nucleophile (CeÀ OÀ ) in these materials.For ZrO x _A and   The catalytic performance of acid treated Zr a Ce 1-a O x and monometallic materials was evaluated by the dephosphorylation of para-nitrophenyl phosphate (p-NPP) in aqueous solution.p-NPP is a comparatively non-toxic chromogenic substrate which undergoes dephosphorylation to produce para-nitrophenol (p-NP).Both p-NPP and p-NP (converted to para-nitrophenolate in alkaline solution after adjustment to pH = 10) can display distinct UV absorbance fingerprints to evaluate the reaction rate (see Experimental Section).Figure 6a provides the UV-vis spectra corresponding to the time dependent dephosphorylation of p-NPP at 50 °C using 1 ZrCe(III,IV)O x _A as a representative case.
As the conversion of p-NPP to p-NP product proceeds, the peak intensity of p-NPP decreases at 311 nm whilst the peak intensity of p-NP (pH adjusted) gradually increases at 400 nm. [6]o side reaction was observed as the stoichiometric changes of p-NPP and p-NP were near equivalent, as shown in Figure S5, while diminished reactivity was observed for the neat materials (Figure S7), and thus will not be included in further discussion.
The concentration of p-NP and p-NPP was determined by comparing the UV absorbances to standard solutions (see Experimental Section), with the rate constant then determined (using Equations ( 1) and ( 2) in Experimental Section).substituted by Zr cations, consequently lowering the conversion towards p-NP. [1]Monometallic ZrO x _A exhibits a low rate constant as both Ce 3 + and Ce 4 + are crucial elements of the OPD reaction mechanism.
The Ce 3 + -bound oxygen nucleophile can attack the central phosphorus atom and initiate dissociation of the PÀ O bond on the p-NPP (organophosphate). [1]The Ce 4 + cation, in contrast, acts as a Lewis acid that can coordinate to the phosphoryl group and cleave the PÀ O bond, [6] although limited reactivity is observed here.Compared to the control materials, the rate constant at 25 °C, 30 °C and 40 °C follows the order 1 ZrCe(III,IV)- Figure 6c displays the p-NP yields from p-NPP dephosphorylation, which follow the order 1 ZrCe(III,IV)O x _A > Ce(III,IV)O x _ A > Ce(III)O x _A > Ce(IV)O x _A at all temperatures and is consistent with the kinetic analysis.The diminished OPD performance for Ce(III)O x _A and Ce(IV)O x _A may be due to the lower amount of Ce 3 + -bound oxygen nucleophiles on the surface (shown by NEXAFS in Figure 5).To evaluate the influence on surface area for Ce(III)O x _A, OPD experiments were performed with increased amounts of material (Figure S9).The conversion trend indicates the total surface area is not the driving force for reactivity, but increase the amount of Ce 3 + and associated Vo. Figure 6d shows the calculated activation energies for all the materials.The activation energies for 1 ZrCe(III,IV)O x _A and Ce(III,IV)O x _A were 63.12 kJ/mol and 68.02 kJ/mol, respectively, which is lower than Ce(III)O x _A at 78.39 kJ/mol.The result supports the NEXAFS finding, that 1 ZrCe(III,IV)O x _A and Ce-(III,IV)O x _A have a higher amount of metastable Ce 3 + -bound oxygen nucleophiles as active sites at 25 °C to 40 °C, and thus exhibit higher performance.The neat "N" materials' activation energies are not reported due to the protamine molecules covering the material surface and active sites which inhibit the OPD performance.A summary of the rate constants can be found in Table S3.
Table S3 shows the data and Pearson correlation (Equation (4) in Experimental Section) corresponding to the rate constant for p-NPP dephosphorylation vs. the Ce 3 + concentration and the Zr concentration. [57]Consistent with the characterization and performance results, Table S4 shows the rate constant is positively correlated to the Ce 3 + concentration with a correlation coefficient of 0.33 to 0.42 at all temperatures.The data trend demonstrates that the increase in Ce 3 + concentration invokes a higher dephosphorylation rate constant, consistent with trends reported in other studies. [1,6]The rate constant is also negatively correlated to the Zr concentration with a correlation coefficient of À 0.56 to À 0.65 at temperatures from 25 °C to 50 °C, respectively.However, Table S3 shows the Ce 3 + concentration is closely correlated to the Zr concentration with a highly positive correlation coefficient of 0.95.The findings suggest a mixed effect of the Zr doping concentration on the rate constant: (i) the direct increase in Zr dopant concentration imposes a negative impact on the rate constant, (ii) the increase in Zr dopant concentration increases the Ce 3 + concentration which indirectly benefits the rate constant.This provides an explanation for the observation that 1ZrCe(III,IV)O x _A exhibits better dephosphorylation per-formance than the other Zr a Ce 1-a O x with higher Zr concentrations.

Conclusions
Nanoscale Zr a Ce 1-a O x was synthesized by a biomineralization strategy for use as effective catalysts for organophosphate dephosphorylation.Protamine-templated Zr a Ce 1-a O x exhibits a higher surface area and pore volume compared to the ZrO x and CeO y counterparts.PDF analysis and XANES LCF results indicate that modulation of the Ce 3 + /Ce 4 + ratio could be achieved by manipulating the synthetic parameters.NEXAFS analysis confirms the existence of Ce 3 + oxygen nucleophiles on the surface of the Zr a Ce 1-a O x catalysts.Coupled with the OPD catalytic performance tests, the data shows that the Vo and Ce 3 + oxygen nucleophile is advantageous for improving catalyst activity.From the results, a correlation between the Ce 3 + concentration, the OPD rate constant, and Zr concentration was proposed.Fundamentally, a small Zr concentration could stabilize the Ce 3 + sites while excess Zr can impede the OPD rate constant and p-NP yield.As the importance of defective metal oxides as catalysts is being increasingly recognized for modulating reaction performance and selectivity, this study is significant for demonstrating a biomineralization route to catalysts for generating Ce 3 + -rich CeÀ Zr oxide for future application in dephosphorylation and other similar reactions.

Experimental Materials preparation via biomineralization
Biomineralization-driven synthesis was used to produce varying molar ratios of Ce : Zr metal oxides for optimisation.A protamine solution was prepared at 20 mg/mL adding protamine sulfate salt (protamine sulfate from herring (Grade III), Sigma-Aldrich) to deionised water and heating at 35 °C until dissolved.A phosphatecitrate buffer solution (pH 7.0) was prepared by mixing 0.1 M sodium phosphate dibasic (99.0 %, Sigma-Aldrich) and 0.05 M citric acid monohydrate (99.0 %, Sigma-Aldrich) in deionized water.3 mL of the protamine solution was added into the 27 mL of the phosphate-citrate buffer solution followed by the addition of the required molar proportions of precursors, as shown in Table S1.The precursors used for cerium were Ce 2 (SO 4 ) 3 .H 2 O (99.0 %, Sigma-Aldrich) and Ce(SO 4 ) 2 .4H 2 O (98.0 %, Sigma-Aldrich) while the precursor for zirconium was Zr(SO 4 ) 2 .H 2 O (99.00 %, Sigma-Aldrich).The mixture was then placed in the rotary mixer for 30 minutes.The subsequently formed precipitates in solution were centrifuged at 6000 rpm for 10 min.The recovered particles were acid-treated with 0.12 M H 2 SO 4 (98 %, Sigma-Aldrich, pH = 3) at 75 °C for 72 hours to remove protamine from the surface.The acid treated particles were then centrifugated and washed with deionized water five times after which they were dried in a freezer dryer.

Material Characterization
The specific surface areas of the materials were determined using nitrogen physisorption on a Tristar II.Approximately 50 mg of sample was loaded into a glass sample tube and pretreated at 150 °C for 3 hr under vacuum prior to analysis.Thermogravimetric analysis (TGA Q50, TA instruments) was undertaken to measure the protamine content.Approximately 5 mg of catalyst was loaded into the instrument and exposed to a flow of 100 % N 2 at 100 °C for 5 min to remove pre-absorbed H 2 O.The temperature was then ramped to 600 °C at 50 °C/min in a flow of 100 % dry air where it was held for 15 minutes.TEM images and STEM-EDS measurements were performed on a JEOL JEM-F200 microscope operating at 200 kV.The materials were drop-cast onto a copper supported lacey carbon grid for the analysis.
High energy X-ray diffraction (HE-XRD) experiments were performed at 11-IDÀ B of the Advanced Photon Source, Argonne National Laboratory.X-ray diffraction patterns were obtained using 86.7 keV photon, providing a Qmax of ~35 Å-1.Atomic PDFs were processed using PDFGetX3. [58]NEXAFS measurements were performed at the O K-edge using the SXR beamline of the Australian Synchrotron.NEXAFS data processing was performed with the Qant program. [59]XAS measurements at the Ce and Zr K-edges were performed at the 10-IDÀ B beamline of the Advanced Photon Source, Argonne National Laboratory.Materials were loaded into 0.0395" inner diameter thin-walled Kapton X-ray capillaries (Cole-Parmer Polyimide tubing, Johns Morries Group) and examined in a transmission geometry.Data processing and subsequent LCF fitting were performed with the Athena (Demeter XAS software package). [49]LCF analysis of Ce XANES was performed in the 5700-5750 eV energy interval using Ce(III)(NO 3 ) 3 .6H 2 O and Ce(IV)O 2 as references.The proportion of Ce 3 + and Ce 4 + identified in the linear combination analysis should correspond to the average of the Ce species in the bulk of the nanoparticles due to the nature of the characterization technique, being the distribution of Ce species different in the surface compared to the core of the nanoparticle.

Catalytic organophosphate dephosphorylation
Catalyst performance was assessed using an Agilent Cary 8454 UVvis spectroscope system to monitor the kinetics of organophosphate dephosphorylation.The model organophosphate used was para-nitrophenyl phosphate (p-NPP) The wavelength of the reactant and product used to determine the change in concentration (with time) were 311 nm for p-NP (product) and 400 nm for p-NPP (reactant).A solution comprising 6.4 mg/mL of p-NPP (99 %, Sigma Aldrich) and 1 M NaOH (99 %, Sigma Aldrich) was initially prepared.The reference cuvette (10 mm quartz, Agilent) comprising 2480 μL of deionized water and ~20 μL of NaOH was prepared to as a blank measurement.5.0 mg of the CeZrO x catalyst, 2460 μL of deionized water and a magnetic stirrer were added into the 'active' cuvette.To adjust the pH to 10, ~20 μL of NaOH was added and then the 'active' cuvette was covered with a cuvette lid to prevent evaporation.An inbuilt water circulation module (PCB1500) was used to preheat the cuvettes to the desired temperature (25 °C, 30 °C, 40 °C or 50 °C).20 μL of p-NPP was pipetted into the 'active' cuvette before UV-spectroscopy analysis commenced.Each experiment had a run time of 7200 seconds with a cycle time of 120 seconds to record the UV-vis data.
The absorbance by different 4-NPP concentrations was measured at the test temperatures (25 °C, 30 °C, 40 °C and 50 °C) to establish a standard calibration curve (Figure S6).The reactant concentration the during the degradation reaction was calculated using the slope of the calibration curve at the relevant test temperature.The product concentration was obtained by subtracting the changing reactant concentration (with reaction time) from the initial reactant concentration (p-NPP added at the start of the experiment).The rate constant, k (s-1), was calculated by monitoring the conversion of the reactant p-NPP to product p-NP with time.The rate was then determined from the slope of the linear plot (only data at steady state selected) from the below first order reaction equation: Where, n o pÀ NPP is the initial amount of p-NPP reactant (in mmol), n t;pÀ NPP is the amount of p-NPP reactant (in mmol) at reaction time, t (in seconds).
On determining the rate constant at each reaction temperatures, the activation energy, Ea (in KJ/mol), of the catalysts was evaluated using the Arrhenius equation: Where T is the reaction temperature in kelvin (K), R is the universal gas constant (8.314J/mol*K) and A is the pre-exponential factor.The activation energy Ea was than calculated by plotting In k vs. 1/ T to obtain the slope (À Ea/R) using a line of the best fit.
The Pearson correlation was calculated for the correlation coefficient based on the equation: Correl x; y ð Þ ¼ P ðx À � xÞðy À � yÞ ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi P ðx À � xÞ 2 P ðy À � yÞ 2 p (4) The correlation coefficient indicates the positive (towards + 1) or negative (towards À 1) correlation between arrays.

Figure 1 .
Figure 1.a) Illustration of the biomineralization synthesis protocol (Ce and Zr precursors were mixed with protamine in solution to form nanoparticle aggregates at pH 7) and acid treatment condition (using 0.12 M H 2 SO 4 at 75 °C for 72 hr) for protamine removal; TEM images of Zr a Ce 1-a O x materials, b) Ce(III)O x _A, c) Ce(III,IV)O x _A, d) 1 ZrCe(III,IV)O x _A, and e) 50 ZrCe(III,IV)O x _A.Fast Fourier transform image of (selected area in the orange box of Figures 1b-e), f) Ce(III)O x _A, g) Ce(III,IV)O x _A, h) 1 ZrCe(III,IV)O x _A, and i) 50 ZrCe(III,IV)O x _A

Figure 2 .
Figure 2. Atomic PDFs of a) pristine Zr a Ce 1-a O x materials, b) acid treated Zr a Ce 1-a O x materials; Atomic PDFs shown from 1.7 Å to 4.5 Å with assigned Ce and Zr paths (Ce path in red lines and Zr path in black lines) for c) pristine Zr a Ce 1-a O x materials, d) acid treated Zr a Ce 1-a O x materials.Symbols: Neat (_N); acid treated (_ A).

Figure
Figure 3c and 3d provide the Zr K-edge XANES for all biomineralized materials.They exhibit maxima at 18017 eV and 18028 eV, respectively for the Zr a Ce 1-a O x materials.50 ZrCe-(III,IV)O x _N and 50 ZrCe(III,IV)O x _A exhibit a negative shift in the E0, and a decrease in white line intensities.The observations imply a slightly lower bulk Zr oxidation state than Zr4 + compared to ZrO x , 1 ZrCe(III,IV)O x and 10 ZrCe(III,IV)O x .[54] .The partial reduction of Zr(4-n) + in CeÀ Zr O x ide has been reported in the literature,[1,47] suggesting that the Zr oxidation state is influenced by the formation of the nearby oxygen vacancy sites as discussed previously.The XANES findings for 50 ZrCe(III,IV)-O x _N and 50 ZrCe(III,IV)O x _A also strongly suggest that Zr atoms are more strongly localized at positions neighboring the Vo.[1,54] Near edge X-ray absorption fine structure (NEXAFS) measurements at the O K-edge were conducted to better understand the oxygen configuration surrounding the Ce and Zr central ions for the acid treated materials, as shown in Figure5.For the O K-edge NEXAFS, 1 s-π* electron transitions occurring between the oxygen 1 s orbital to unoccupied orbitals, where multiple features in NEXAFS are due to the ligand field splitting of the Ce or Zr d-orbital.This results in characteristic peaks related to the CeÀ O local symmetry at ~534 eV for eg orbital and ~538 eV for t 2g orbital.[55,56]It is clear the peak for t 2g orbital is only observed for Ce(III)O x _A and Ce(IV)O x _A materials, and not for Ce(III,IV)O x and Zr a Ce 1-a O x materials.This forbidden electron transition shows that extra electrons could occupy the t 2g molecular orbital for Ce(III,IV)O x and Zr a Ce 1-a O x , which suggests the existence of a Ce 3 + -bound oxygen nucleophile (CeÀ OÀ ) in these materials.For ZrO x _A and

Figure 3 .
Figure 3. XANES spectra of: a) Ce L3-edges of the pristine Zr a Ce 1-a O x materials, b) Ce L3-edges of the acid treated Zr a Ce 1-a O x materials, c) Zr K-edge of the pristine Zr a Ce 1-a O x materials, d) Zr K-edge of the acid treated Zr a Ce 1-a O x materials.Symbols; Neat (_N), acid treated (_A).

Figure 4 .
Figure 4. Linear combination fitting of Ce XANES for modelling the pristine and acid-treated materials showing a) the change in Ce 3 + fraction and b) the change in Ce 4 + fraction.Symbols; Neat (_N), acid treated (_A).

50
ZrCe(III,IV)O x _A, the peaks show ZrÀ O local symmetry at ~533 eV for e g orbital and ~538 eV for t 2g orbital, which confirms the formation of a solid solution in 50 ZrCe(III,IV)O x _A, agreeing with the PDF analysis.
Figure 6b shows the kinetics of p-NPP dephosphorylation by the Zr a Ce 1-a O x and monometallic materials vs temperature, with an example of rate constant determination for 1 ZrCe(III,IV)O x _A at 25 °C shown in Figure S8.It is evident the Zr concentration has a significant effect on the kinetics, whereby the rate constant trend order is 1 ZrCe(III,IV)O x _A > 10 ZrCe(III,IV)O x _A > 50 ZrCe-(III,IV)O x _A > ZrO x _A for all temperatures.The trend agrees with the literature where increasing the Zr concentration decreases the rate constant as surface Ce 3 + /Ce 4 + active sites are

Figure 5 .
Figure 5. Oxygen K-edge NEXAFS of the acid treated Zr a Ce 1-a O x materials; the t 2g , e g energy levels are indicated.

Figure 6 .
Figure 6.Para-Nitrophenyl phosphate (p-NPP) dephosphorylation using acid treated Zr a Ce 1-a O x catalyst a) UV-vis spectra profile for 1 ZrCe(III,IV)O x _A at 50 °C, b) rate constant at temperatures of 25 °C, 30 °C, 40 °C and 50 °C, c) percentage yield of para-nitrophenol (p-NP) at temperatures of 25 °C, 30 °C, 40 °C and 50 °C, after 2 hours of reaction d) apparent activation energy for acid treated Zr a Ce 1-a O x materials.

Table 1 .
Specific surface area (SSA), pore volume, average pore diameter and residual organic content for Zr a Ce 1-a O x materials of varying compositions.