Three-Dimensional Pinecone-like Binder-Free Pt–TiO2 Nanorods on Ti Mesh Structures: Synthesis, Characterization and Electroactivity towards Ethanol Oxidation

We report here the synthesis of binderless and template-less three-dimensional (3D) pinecone-shaped Pt/TiO2/Ti mesh structure. The TiO2 hydrothermally synthesized onto Ti mesh is composed of a mixture of flower-like nanorods and vertically aligned bar-shaped structures, whereas Pt film grown by pulsed laser deposition displays a smooth surface. XRD analyses reveal an average crystallite size of 41.4 nm and 68.5 nm of the TiO2 nanorods and Pt, respectively. In H2SO4 solution, the platinum oxide formation at the Pt/TiO2/Ti mesh electrode is 180 mV more negative than that at the Pt/Ti mesh electrode, indicating that TiO2 provides oxygeneous species at lower potentials, which will facilitate the removal of CO-like intermediates and accelerate an ethanol oxidation reaction (EOR). Indeed, the Pt/TiO2/Ti mesh catalyst exhibits current activity of 1.19 mA towards an EOR at a remarkably superior rate of 4.4 times that of the Pt/Ti mesh electrode (0.27 mA). Moreover, the presence of TiO2 as a support to Pt delivers a steady-state current of 2.1 mA, with an increment in durability of 6.6 times compared to Pt/Ti mesh (0.32 mA). Pt is chosen here as a benchmark catalyst and we believe that with catalysts that perform better than Pt, such 3D pinecone structures can be useful for a variety of catalytic or photoelectrochemical reactions.


Introduction
There is a growing need for lightweight, higher efficiency, cheap, durable and safe energy storage and conversion devices to meet special needs for next-generation highperformance portable electronic devices. Current research is focusing on flexible and wearable electronic devices such as roll-up display, light-emitting diodes, smart watches, fitness-tracking and implantable sensors that can monitor blood pressure and other health metrics [1]. Such devices are being developed rapidly and essentially require low power source systems with flexible electrodes, separators and substrates to accomplish all-inone flexible systems. Several types of lightweight power sources are being developed worldwide, including Li-ion batteries, supercapacitors, solar cells, fuel cells and biofuel cells [2][3][4][5][6].
Generally, traditional electrodes used in electrochemical power sources (EPSs) are composed of mixtures of active materials, polymeric binders and conductivity enhancers, like carbon black, graphite and carbon nanotubes, which are ink-coated over metal current collectors such as copper, aluminum or carbon paper. The additives, the metal collector elements are not only unsuitable for flexibility but add extra-weight to the EPS device. Hence, the ability to eliminate such additives and fabricate a binder and additive-free, flexible electrode would signify notable advancement for high-performance flexible EPSs.

Materials Characterization
The photograph and SEM images of the pristine Ti mesh are shown in Figure S1 (Supplementary Materials) for comparison. Figure 1 reports SEM micrographs at increasing magnifications of the TiO 2 grown by the hydrothermal technique on Ti mesh. As can be seen from SEM the images, the Ti mesh was evenly coated with TiO 2 . Higher magnifications show that the as-prepared TiO 2 is composed of a mixture of flower-like nanorods and vertically aligned bar-shaped structures.
The XRD crystalline structure corresponding to the TiO 2 /Ti mesh is reported in Figure 2. The peaks relate to rutile TiO 2 (JCPDS 21-1276). The highest peak TiO 2 (110) was employed to calculate the average crystallite size (L). This was done using the Debye-Scherrer equation: L = 0.89 λ/β cos θ, with λ corresponding to the wavelength of 1.5406 Å, β to the full-width at half-maximum (FWHM) and θ is the Bragg angle. Accordingly, the average crystallite size of the TiO 2 nanorods was estimated to be 414 Å.   Carbon contamination is practically impossible to avoid. Fortunately, it is usually restricted to the film-air interface. Figure 3b,c reveal, respectively, the high-resolution XPS Ti 2p and O 1s peaks at the surface of TiO2/Ti mesh. The positions of the Ti 2p1/2 and Ti 2p3/2 separated by 5.60 to 5.64 eV demonstrating by that Ti was in the Ti 4+ state at the surface [22][23][24]. The O 1s peak is asymmetrical, which signifies that oxygen appears in two    Carbon contamination is practically impossible to avoid. Fortunately, it is usually restricted to the film-air interface. Figure 3b,c reveal, respectively, the high-resolution XPS Ti 2p and O 1s peaks at the surface of TiO2/Ti mesh. The positions of the Ti 2p1/2 and Ti 2p3/2 separated by 5.60 to 5.64 eV demonstrating by that Ti was in the Ti 4+ state at the surface [22][23][24]. The O 1s peak is asymmetrical, which signifies that oxygen appears in two  Carbon contamination is practically impossible to avoid. Fortunately, it is usually restricted to the film-air interface. Figure 3b,c reveal, respectively, the high-resolution XPS Ti 2p and O 1s peaks at the surface of TiO 2 /Ti mesh. The positions of the Ti 2p 1/2 and Ti 2p 3/2 separated by 5.60 to 5.64 eV demonstrating by that Ti was in the Ti 4+ state at the surface [22][23][24]. The O 1s peak is asymmetrical, which signifies that oxygen appears in two different chemical states at least. The O 1s spectrum (Figure 3c) could be deconvoluted into two contributions that are OI and OII. The first contribution (OI) occurring within binding energies of 529.9~530 eV is ascribed the normal lattice sites occupied by oxygen in the TiO 2 structure. The second contribution (OII) appearing at 531~532 eV is ascribed to non-lattice oxygen [25]. This high binding energy component has been assigned to oxygen bonded to Ti +3 (O-Ti 3+ ) or to hydroxyl species that are simply created at the surface of oxide films [26].  (Figure 3c) could be deconvoluted into two contributions that are OI and OII. The first contribution (OI) occurring within binding energies of 529.9~530 eV is ascribed the normal lattice sites occupied by oxygen in the TiO2 structure. The second contribution (OII) appearing at 531~532 eV is ascribed to non-lattice oxygen [25]. This high binding energy component has been assigned to oxygen bonded to Ti +3 (O-Ti 3+ ) or to hydroxyl species that are simply created at the surface o oxide films [26].

Characterization of Pt/TiO 2 /Ti Mesh
SEM images of the Pt/Ti mesh structure revealed that the surface morphology of the Pt film is very smooth (Figure 4). The XRD indexation ( Figure 4, right side) of Pt peaks is in agreement with (111), (200), (220), (311) and (222) planes, respectively, of a face centered cubic (fcc) structure (JCPDS PDF No. 04-0802). The most intensive diffraction peak Pt (111) was selected to calculate the lattice constant (a) and the average crystallite size of Pt by Bragg's law and Debye-Scherrer equation, respectively. The average crystallite size of Pt deposited Ti mesh was found to be 932 Å. On the other hand, the lattice constant was calculated to be close to 3.924 Å, which is close to the theoretical value of 3.923 Å.  The morphology of Pt films deposited on the TiO2 arrays is unlike Pt synthesized on Ti mesh. Figure 5, at higher magnification, shows that the Pt/TiO2 resembles to 3D pinecone-shaped structure. The lattice parameter and the average crystallite size calculated from XRD of Figure 6 were found to be 3.885 Å and 685 Å, respectively. Table 1, resumes the various XRD characteristics of the Ti mesh, TiO2/Ti mesh, Pt/Ti mesh and Pt/TiO2/Ti mesh. From the table, it can be seen that the average crystallite size of Pt is smaller than that of Pt grown on Ti mesh, suggesting that TiO2 enhanced the distribution quality of the Pt nanoparticles. It can also be observed that the lattice parameter of Pt synthesized on TiO2 is lesser than that of Pt produced on Ti mesh. This might exhibit a size-induced lattice contraction in the as-prepared state with respect to bulk Pt. The interaction of oxides with metallic nanostructures has been extensively investigated with particular emphasis on the effect of surface oxygen vacancies [28]. Oxygen vacancies at the metal support interface are recognized to induce charge transfer from the oxide to the metal easing the binding of the metal to the oxide. It has been reported that Pt is strongly adsorbed at oxygen vacancy sites in TiO2 [29]. It has previously been discussed that lattice strain can dramatically affect electrochemical activity [30][31][32]. Hence it is anticipated that the support-induced lattice strain observed in Pt/TiO2 could potentially enhance electrocatalytic activity, compared with Pt/Ti mesh. The morphology of Pt films deposited on the TiO 2 arrays is unlike Pt synthesized on Ti mesh. Figure 5, at higher magnification, shows that the Pt/TiO 2 resembles to 3D pinecone-shaped structure. The lattice parameter and the average crystallite size calculated from XRD of Figure 6 were found to be 3.885 Å and 685 Å, respectively. Table 1, resumes the various XRD characteristics of the Ti mesh, TiO 2 /Ti mesh, Pt/Ti mesh and Pt/TiO 2 /Ti mesh. From the table, it can be seen that the average crystallite size of Pt is smaller than that of Pt grown on Ti mesh, suggesting that TiO 2 enhanced the distribution quality of the Pt nanoparticles. It can also be observed that the lattice parameter of Pt synthesized on TiO 2 is lesser than that of Pt produced on Ti mesh. This might exhibit a size-induced lattice contraction in the as-prepared state with respect to bulk Pt. The interaction of oxides with metallic nanostructures has been extensively investigated with particular emphasis on the effect of surface oxygen vacancies [28]. Oxygen vacancies at the metal support interface are recognized to induce charge transfer from the oxide to the metal easing the binding of the metal to the oxide. It has been reported that Pt is strongly adsorbed at oxygen vacancy sites in TiO 2 [29]. It has previously been discussed that lattice strain can dramatically affect electrochemical activity [30][31][32]. Hence it is anticipated that the support-induced lattice strain observed in Pt/TiO 2 could potentially enhance electrocatalytic activity, compared with Pt/Ti mesh.       The XPS survey spectrum of the Pt/TiO 2 material shown in Figure S2 (Supplementary Materials) reveals that the surface exclusively consists of Pt, O and C elements. No Ti metal is detected indicating that the Pt film homogenously coats the underneath TiO 2 nanorods. Figure 7a,b report the high-resolution Pt 4f core-level spectrum at the Pt/Ti mesh and Pt/TiO 2 /Ti mesh, respectively. The spectra show that Pt exists in the form of at least three oxidation states. The spectra were indeed adequately deconvoluted into three overlapping curves assigned to Pt 0 , Pt 2+ and Pt 4+ species. Table 2 shows the binding energy (BE) and relative amount of these three species assessed from the relative area of the integrated peak intensities. The peak positions of Pt 0 , Pt 2+ and Pt 4+ are in agreement with the values found in the literature [33,34]. The shift toward higher BE values compared to literature values (71.0 eV) is ascribed to the metal−support interaction and to small Pt nanoparticle sizes [35,36]. This positive shift may also imply metal−support interactions between TiO 2 and Pt as observed with XRD analyses. This interaction can change the electronic properties of Pt by increasing the Pt d-vacancy via electronic donation to Lewis acid centers such as Ti x+ at the Pt/TiO 2 interface [37][38][39].  The XPS survey spectrum of the Pt/TiO2 material shown in Figure S2 (Supplementary Materials) reveals that the surface exclusively consists of Pt, O and C elements. No Ti metal is detected indicating that the Pt film homogenously coats the underneath TiO2 nanorods. Figure 7a,b report the high-resolution Pt 4f core-level spectrum at the Pt/Ti mesh and Pt/TiO2/Ti mesh, respectively. The spectra show that Pt exists in the form of at least three oxidation states. The spectra were indeed adequately deconvoluted into three overlapping curves assigned to Pt 0 , Pt 2+ and Pt 4+ species. Table 2 shows the binding energy (BE) and relative amount of these three species assessed from the relative area of the integrated peak intensities. The peak positions of Pt 0 , Pt 2+ and Pt 4+ are in agreement with the values found in the literature [33,34]. The shift toward higher BE values compared to literature values (71.0 eV) is ascribed to the metal−support interaction and to small Pt nanoparticle sizes [35,36]. This positive shift may also imply metal−support interactions between TiO2 and Pt as observed with XRD analyses. This interaction can change the electronic properties of Pt by increasing the Pt d-vacancy via electronic donation to Lewis acid centers such as Ti x+ at the Pt/TiO2 interface [37][38][39].   From Table 2, it can be noted that in the Pt/Ti mesh, Pt 0 is widely distributed on the surface with 87.7 at% followed by slight relative concentrations (<7 at%) of Pt 2+ and Pt 4+ . It should be noted that for Pt deposition on the Ti mesh, the former has not been etched on purpose, in order to better assess the effect of TiO 2 morphology (layer vs. nanorods) on the electrochemical performance. Hence, the oxygen is due to the native TiO 2 layer present on the surface of the mesh. On the other hand, the concentration of Pt 0 decreased while those of Pt 2+ and Pt 4+ increased at the Pt/TiO 2 sample. From Table 2, compared to the Pt/Ti mesh, it was noticed that in the Pt/TiO 2 /Ti mesh the position of Pt 0 shifted by 0.45 eV toward higher BEs while the position of Pt 2+ moved by 1.26 eV toward lower BEs. These shifts can be explained by different sizes of Pt particles and different degrees of interaction with the TiO 2 support, signifying a strong interaction between the TiO 2 nanorods and the Pt film above. Other researchers have observed Pt/TiO 2 composites exhibiting ionized platinum, which was also ascribed to the strong interaction between Pt and the TiO 2 support [37,38,40]. This behavior can be assumed to the presence of oxygen vacancies at the TiO 2 support interface. The Pt 4+ (PtO 2 ) is the result of Pt cations replacing those of Ti in the TiO 2 lattice, and the Pt atoms at the surface creating Pt 2+ species. The O 1s core level peaks for the Pt/Ti mesh and Pt/TiO 2 /Ti mesh materials are shown in Figure 7c,d, respectively. A simple visual inspection of the O1s peak showed that it was wide and asymmetrical (Figure 7c,d). Therefore, the peak could be deconvoluted in two peaks. The resulting parameters are reported in Table 2. In this case, the OI element is ascribed to the bulk lattice oxygen, whereas the OII component is attributed to the surface lattice oxygen [41]. It has to be reiterated that both electrodes contained a similar amount of Pt (0.120 mg cm −2 ). At the Pt/TiO 2 /Ti mesh electrode, the CV contained the classical features of hydrogen atom adsorption (H ads ) and hydrogen atom desorption (H des ) peaks from −0.2 to 0.1 V [42][43][44][45]. On the other hand, the H ads /H des features were ill-defined at the Pt/Ti mesh electrode. In addition, the current of the H ads /H des peaks at the Pt/TiO 2 /Ti mesh were distinctively greater than those delivered by the Pt/Ti mesh, which clearly indicates the greater surface area at the former electrode [45]. This also means that Pt/TiO 2 has a larger surface area than the Pt/Ti mesh. The larger surface area signifies smaller particle size, confirming that TiO 2 improved the dispersion and utilization of the Pt nanoparticles [46], which is in line with the XRD analyses ( Table 1). The Pt oxide formation (PtOx f ) at the Pt/Ti mesh starts at 0.74 V whereas its equivalent reduction (PtOx r ) peak potential takes place at 0.56 V vs. Ag/AgCl. On the other hand, PtOx f and PtOx r were located at 0.44 V and 0.48 V, respectively at the Pt/TiO 2 /Ti mesh. This implies that TiO 2 can provide oxygeneous species at lower potentials, which will facilitate the removal of CO-like intermediates and accelerate EOR.

Electrochemical Characterization
havior. In Figure 8c, one can observe that the current decay for the EOR on the Pt/TiO2/Ti mesh catalyst is significantly slower than that on the bare Pt/Ti catalyst. The first current increase is due to the double-layer charging effect, whereas the initial decay was caused by the rapid increase of the surface coverage by intermediate species, such as adsorbed CO during EOR [48]. After 3600 s, the quasi steady-state current density (Iss) at the Pt/TiO2/Ti mesh was 6.6 times greater than that of Pt/Ti mesh (Table 3).  Subsequently, the effect of the TiO 2 arrays on the electrocatalytic behaviour of Pt was investigated towards electrooxidation of ethanol. Figure 8b shows comparative LSVs obtained in 0.5 M H 2 SO 4 + 1 M C 2 H 5 OH solution at 5 mV s −1 (quasi-steady state) at Pt/Ti mesh and Pt/TiO 2 /Ti mesh electrodes. Forward and backward CVs in the 0.5 M H 2 SO 4 + 1 M C 2 H 5 OH solution are shown in Figure S3. The LSVs at both electrodes showed characteristic EOR waves in accordance with the literature [47][48][49]. The onset potential, E onset , peak potential (E p ), I@0.50 V, and peak current I p values extracted from LSVs of Figure 8b are reported in Table 3. E onset is a criterion that provides knowledge about the kinetics of an electrochemical reaction and is identified here as the value at which a current begins for the electrooxidation of ethanol. I@0.50 V is an experimental condition near to the projected functioning potential of DEFCs devices and allows comparison of the progress of the EOR catalytic activity using different electrocatalysts. Hence from Table 3, it can be observed that E onset and E p are not significantly different at both electrodes. However, Pt/TiO 2 /Ti mesh demonstrated the best catalytic activity toward EOR in terms I p of 1.85 mA which is remarkably 3.5 times greater than the I p delivered by Pt/Ti mesh. A further outstanding performance of the Pt/TiO 2 /Ti mesh is obviously its value of I@0.5 V, which is 4.4 times greater than the Pt/Ti mesh. Chronoamperometric (CA) experiments were performed to examine the electrodes durability. Figure 8c shows the current-time (I-t) curves of the Pt/Ti mesh and Pt/TiO 2 /Ti mesh for EOR at 0.6 V upon 3600 s testing. It can be seen that in both CA profiles, the current increased abruptly, then decreased and ultimately attained quasi-steady-state behavior. In Figure 8c, one can observe that the current decay for the EOR on the Pt/TiO 2 /Ti mesh catalyst is significantly slower than that on the bare Pt/Ti catalyst. The first current increase is due to the double-layer charging effect, whereas the initial decay was caused by the rapid increase of the surface coverage by intermediate species, such as adsorbed CO during EOR [48]. After 3600 s, the quasi steady-state current density (I ss ) at the Pt/TiO 2 /Ti mesh was 6.6 times greater than that of Pt/Ti mesh (Table 3).
In summary, it is clear that incorporating TiO 2 flower-like nanorods and vertically aligned bar-shaped structures as a supporting material considerably increased the electrochemical activity of the Pt catalyst. Therefore, it can be suggested that the Pt/TiO 2 /CP catalyst offers a higher Pt utilization than the unsupported Pt/Ti mesh catalyst.

Growth of TiO 2 Arrays onto Ti Mesh
Titanium meshes of 1.5 cm × 2 cm with 0.5 mm thickness were placed in an ultrasonic bath containing acetone and washed for 20 min. After rinsing with water and drying in air, the titanium meshes were subjected to chemical etching in a solution containing 15 mL of HCl (18 wt%) solution. The chemical etching was conducted at a temperature of 80 • C and a duration of 15 min. This operation was necessary to remove the native TiO 2 layer. Afterward, the Ti meshes were placed inside a Teflon stainless steel autoclave (23 mL, Parr Instrument, Moline, IL, USA) containing 10 mL of 0.6 M of HCl aqueous solution. The hydrothermal synthesis was conducted at 180 • C and lasted 10 h. The synthesis condition effects on the microstructure of the obtained materials can be found in our previous publication [50].

Growth of Pt onto Ti Mesh and TiO 2 Arrays
Platinum films were deposited onto Ti mesh and TiO 2 /Ti mesh substrates at room temperature by PLD method employing pure Pt target (99.99%, Kurt J. Lesker Co, Jefferson Hills, PA, USA). Details on the operating principle of the PLD are reported elsewhere [51,52]. The deposition conditions were: 50 kp laser pulses, 2 Torr of helium, KrF excimer laser (λ = 248 nm), pulse width of 17 ns, repetition rate of 50 Hz, 4 Joules per cm 2 as the laser fluence and 5 cm as the distance between the substrate (Ti or TiO 2 ) and the Pt target. Prior to every deposition of Pt, evacuation of the PLD chamber was done at 4 × 10 −5 Torr by a turbo pump. The amount of the deposited Pt assessed with neutron activation analysis was 120 µg cm −2 . The PLD deposition parameters were optimal and reported in our previous publications [53,54].

Materials Characterization
An SEM (TESCAN, LYRA3) operated at 20 kV was used to analyze the surface morphology of the synthesized materials. The crystalline structure of all samples was determined by XRD using a Bruker D8 Advance diffractometer equipped with a Cu Kα source (λ = 1.5406 Å). The tube current was 40 mA with a tube voltage of 40 kV. Diffractograms were acquired with an acquisition time of 5 s per step in the Grazing Incidence Diffraction (GID) scan mode using an incident angle of 3 • and a 2θ angular of 0.04 • step size. XPS analysis was conducted to examine the concentration of the elements and their valence states at the surface of the samples with a VG Escalab 220i-XL outfitted with an Al Kα source (1486.6 eV). 10 kV, 20 mA and pass energy of the analyzer 20 eV were the conditions that operated the anode. Survey spectra were first recorded from 0 to1350 eV. Afterwards, higher resolution multiplex scans (Ti 2p, Pt 4f, C1 s and O 1s core levels) were acquired. CasaXPS software (Casa Software Ltd, Teignmouth, UK.) was employed to analyze and quantify the elements by fitting the core level spectra to a Shirley background exclusion. The metallic components of the Pt 4f and Ti 2p regions were fitted using a Gaussian/Lorentzian asymmetrically modified line shape, and a Gaussian/Lorentzian line shape was used to fit the other components. The C 1s core level located at 284.6 eV, stemming from hydrocarbon impurities present at the surface of the samples, was employed as an internal reference. All XPS spectra were readjusted with regard to the C 1s core level of accidental carbon impurity.

Electrochemical Experiments
Ethanol (100% purity) and sulfuric acid (H 2 SO 4 , 96%) were acquired from Commercial Alcohols Inc. (Toronto, ON, Canada) and Agros Organics (Fisher Scientific, Mississauga, ON, Canada), respectively. The reactants were used as received. The electrochemical properties were studied by voltammetry or LSV. The electrolytic solution was either 0.5 M H 2 SO 4 or 1 M C 2 H 5 OH + 0.5 M H 2 SO 4 . The 3-electrode cell contained Ag/AgCl (4 M NaCl) that acted as a reference electrode, a platinum coil as an auxiliary electrode and Pt/Ti mesh or Pt/TiO 2 arrays/Ti mesh as working electrodes. In this paper, the potentials are reported against Ag/AgCl. Before commencing each electrochemical experiment, argon was bubbled through the electrolytic solution for 30 min to remove dissolved oxygen. Then the surface of the Pt/Ti mesh and Pt/TiO 2 /Ti mesh were subjected to electrochemical cleaning and activation in 0.5 M H 2 SO 4 by multicycling voltammetry from −0.2 V to 1.3 V at 50 mV s −1 potential scan rate until a steady state voltammogram was reached. EOR experiments were performed with LSV using a mixture of 1 M C 2 H 5 OH and 0.5 M H 2 SO 4 within 0 V to 1.0 V at 5 mV s −1 . The Ti meshes in all cases had the same geometric size. Electrochemical measurements were conducted at ambient temperature using an Autolab, PGSTAT 20.

Conclusions
TiO 2 arrays were successfully synthesized directly on the Ti mesh by a hydrothermal method in acidic medium. The hydrothermal method does not necessitate the utilization of templates, it is easily scalable, cheap and environmentally benign. SEM observations revealed that as-prepared TiO 2 is constituted of a mixture of flower-like nanorods and vertically aligned bar-shaped structures corresponding to rutile phase, as identified by XRD. By means of XPS, the [O]/[Ti] atomic ratio was found to 3.8. This deviation from stoichiometry is ascribed to porosity that permits moisture to accrue between the voids of the perpendicularly arranged TiO 2 bars or within the arranged flower-like TiO 2 nanorods.
Afterwards, Pt catalyst as a benchmark catalyst was deposited by PLD onto synthesized TiO 2 structures in order to assess their catalytic supporting properties. SEM imaging revealed an interesting 3D pinecone-shaped Pt/TiO 2 structure. XRD analysis showed that the crystallite size of Pt in Pt/TiO 2 was smaller than that in Pt/Ti mesh, which demonstrates that the TiO 2 support enhances the dispersion quality of Pt nanoparticles. Furthermore, XPS analysis confirmed the strong interaction between Pt and the TiO 2 support, which induces ionized platinum (Pt 2+ and Pt 4+ ).
Notwithstanding having a similar amount of Pt, the three-dimensional pineconeshaped Pt/TiO 2 structure exhibited current catalytic activity towards EOR at a remarkably greater rate of 4.4 times more than unsupported Pt. Moreover, the presence TiO 2 as support enables 6.6 times increased current durability relative to the Pt/Ti mesh. As mentioned in this work, Pt was chosen here as a model catalyst and we believe that with catalysts that perform better than platinum such 3D mesh architectured electrodes are promising not only for fuel cells in general but can be useful for a variety of catalytic or photoelectrochemical reactions for other catalysts.
Supplementary Materials: The following supporting information can be downloaded online https: //www.mdpi.com/article/10.3390/molecules27061921/s1, Figure S1: SEM micrographs at increasing magnifications of pristine Ti mesh. Figure S2: XPS survey scan of Pt film grown onto TiO 2 nanorods; Figure