Hierarchically-Structured Ti/TiO₂ Electrode for Hydrogen Evolution Synthesized via 3D Printing and Anodization

Xiang Li ¹, Yuan Xue ², Ryan Dehoff ³, Costas Tsouris ³, Patricia Taboada-Serrano ^{1, 4, *}

1. Microsystems Engineering PhD Program, Rochester Institute of Technology , 77 Lomb Memorial Drive, Rochester, NY, USA
2. Materials Science Program, University of Rochester , 4011 Wegmans Hall, PO Box 270166, Rochester, NY, USA
3. Oak Ridge National Laboratory, 1 Bethel Valley Rd, Oak Ridge, TN, USA
4. Chemical Engineering, Rochester Institute of Technology, 160 Lomb Memorial Drive, Rochester, NY, USA

* Correspondence: Patricia Taboada-Serrano

Academic Editor: Alfonso Chinnici

Special Issue: Hydrogen Energy: Sustainable Production, Storage and Utilisation

Received: April 09, 2020 | Accepted: May 08, 2020 | Published: May 09, 2020

Journal of Energy and Power Technology 2020, Volume 2, Issue 2, doi:10.21926/jept.2002007

Recommended citation: Li X, Xue Y, Dehoff R, Tsouris C, Taboada-Serrano P. Hierarchically-Structured Ti/TiO2 Electrode for Hydrogen Evolution Synthesized via 3D Printing and Anodization. Journal of Energy and Power Technology 2020; 2(2): 007; doi:10.21926/jept.2002007.

© 2020 by the authors. This is an open access article distributed under the conditions of the Creative Commons by Attribution License, which permits unrestricted use, distribution, and reproduction in any medium or format, provided the original work is correctly cited.

1. Introduction

One of the most attractive renewable-energy fuel alternatives is hydrogen. Hydrogen is an ideal chemical energy carrier with energy density of 140 MJ kg^-1 and no secondary contaminants generated during its oxidation [1]. Currently, hydrogen is mainly obtained from natural gas and coal, involving consumption of fossil fuels along with greenhouse gas release [2]. Securing a clean, safe, and sustainable approach for renewable hydrogen production has become one of the biggest barriers that hinder the leap from fossil-fuel-based energy society to hydrogen energy economy.

Water is naturally the most promising source for hydrogen production due to its abundance, low cost, and easy access on earth. Consequently, a great deal of effort has been devoted to generating hydrogen from water through sustainable, energy-efficient, and cost-effective processes that have the potential to work at industrial scales, such as photocatalytic water splitting and electrolysis [3,4,5,6]. The hydrogen evolution reaction (HER), 2H⁺ + 2ē → H₂, is the basis for water splitting-based hydrogen production. In order to reach high energy-efficiency, catalysts are widely used to minimize the potential required to initiate the HER [7]. The best-known catalysts for the HER to date are noble metals, such as platinum. However, the scarcity and high price of noble metals substantially limit their large-scale deployment [8].

Developing precious-metal-free catalysts for HER is a crucial step in moving towards a hydrogen-based energy economy. Molybdenum sulfide materials are one of the typical non-noble metal catalysts explored as alternative to platinum-group catalysts due to their high electro-catalytic activity and stability for HER [9,10,11]. The synthesis of molybdenum-based catalysts, however, usually involves energy-intensive, time-consuming, and complex processes that might hinder their application at industrial scales. For example, ultra-vacuum processing [10], high-temperature treatment [12], and sulfidization under H₂S gas are some of the steps involved in the molybdenum-based-catalysts synthesis [10].

In 1972, Honda and Fujishima reported the occurrence of photocatalytic water-splitting by using crystalline titanium dioxide (TiO₂) electrode for hydrogen production. This opened the door for producing hydrogen from solar energy through a clean, low-cost, and environmentally friendly approach [13]. The finding reported by Honda and Fujishima triggered thousands of publications dealing with exploring TiO₂ photocatalytic activity for hydrogen production [14,15,16,17,18,19]. It is not surprising that TiO₂ has become one of the most common catalysts for photocatalytic hydrogen production owing to its abundance, low cost, and semiconductor properties. However, the wide band gap (3.2 ēV) limits the photocatalytic activity of TiO₂ to UV light only [20]. Therefore, the past decade has witnessed the proliferation of a wide range of studies geared towards improving the visible light absorption of TiO₂ via doping of TiO₂ with metal ions [21], coupling TiO₂ with low band gap semiconductor materials [22,23], dye sensitization of TiO₂ [19], surface modification of TiO₂ [24], and also the use of TiO₂ as support for noble metals [25,26]. Enormous efforts have been devoted to the enhancement of TiO₂ photocatalytic activity via increasingly complex chemical functionalization. For example, Ag₂S-coupled TiO₂ nanowire fabrication requires hydrothermal treatment over 72 hours [22], and cobalt-oxide loaded TiO₂/CdS composites synthesis consists of 12-hours reaction under vigorous stirring and drying under vacuum at 333 K for another 12 hours. While photocatalytic activity has significantly improved through these methods, the potential for application of these catalysts at industrial scales is lowered by their increased complexity [18].

Another methodology explored for improving the performance of TiO₂-based catalysts for photocatalytic hydrogen generation is to control catalyst architecture as a means to improve catalyst activity. Employing a core-shell structure is an efficient strategy in order to control architecture. Yang et al. [27,28,29] reported a good photocatalytic activity with a hydrogen production rate of 0.258 mmol h^-1 g^-1 for a core-shell nanostructure of rutile TiO₂ with sulfide surface (TiO₂-x:S). Kim and co-workers [28] developed a core/shell structure of magnetic nanoparticles NiFe₂O₄/TiO₂ and successfully applied it to photo-driven HER. Although core-shell structures show good catalytic performance for HER, a concern with core/shell designs is the poor contact between the core and shell materials [30]. It usually results in a dramatic decrease of catalytic activity with continuous use.

Embracing three-dimensional (3D) geometry for electrodes has become evident as another attractive way of obtaining high performing HER photo-catalysts. For instance, Xu et al. utilized a multiscale 3D TiO₂ platform decorated with graphene quantum dots for photo-electro-chemical hydrogen production. Li et al. [31] developed a 3D Pt/TiO₂ architecture via a solvo-thermal method. The 3D Pt/TiO₂ electrode exhibited high photocatalytic activity during hydrogen production tests [31]. However, a noble metal is still required to boost the photocatalytic activity of TiO₂ [32].

All the efforts detailed above were focused on improving photo-catalytic activity of TiO₂ for hydrogen production. To our knowledge, only the inclusion of TiO₂ nanoparticles on other metal catalysts has been explored for electro-catalytic HER in the absence of light; and the presence of TiO₂ has been identified as contributing to improved electrocatalytic activity for HER [33,34].

Recently, 3D printing has been introduced as the means to manufacture micro-reactors and reactor components [35]. This new approach to the manufacture of reactor components is driven by the goal of process intensification, and can certainly be incorporated into the manufacture of catalysts as well. In this work, we report on the performance of a hierarchically structured Ti-core/TiO₂-coating electro-catalyst for HER in the absence of light (no photo-catalysis involved). The electro-catalysts were produced via 3D printing and in-situ anodization, in order to superimpose highly-ordered TiO₂ nanotubes (NTs) onto hierarchically 3D Ti-metal templates used as current collectors. The reported hierarchically structured Ti/TiO₂ materials were tested in this study for HER electro-catalytic activity, and were found to be good candidates to overcome the hurdles towards renewable, industrial-scale hydrogen production. To our knowledge, this work presents the first utilization of titanium and titania for electro-catalytic HER in the absence of light. A similar synthesis approach to the one reported in this work is being pursued and can be used for other metal/metal oxide combinations, not reported in the present work.

2. Materials and Methods

2.1 Synthesis of the Hierarchical Flow-Through 3D Ti/TiO₂ NT Electrode

Macro-porous, flow-through 3D Ti templates were printed with an additive manufacturing method using ARCAM Electron Beam Melting equipment [36,37]. The templates had dimensions of 2.5 × 2.5 × 0.22 cm. Titanium dioxide nanotube (TiO₂ NT) arrays were grown on the surface of the 3D Ti templates via anodization following the methodologies described in previous work [38,39,40,41]. Two formulation of anodization baths were used to grow the TiO₂ NT layer on different 3D Ti templates: (a) an organic-solvent-based anodization solution (EG-anodization solution), and (b) an aqueous-solvent-based anodization solution (W-anodization solution). The preparation of samples and anodization solutions are described below, along with the anodization conditions in each case.

In all cases, the 3D Ti templates were thoroughly rinsed with ethanol and air-dried prior to the anodization procedures. The EG-anodization solution was prepared by mixing 147 mL ethylene glycol, 3 mL of distilled water, and 280 mg of ammonium fluoride. The EG-anodization solution was introduced into a temperature-controlled anodization cell under dry air. The anodization system consists of (a) a Princeton Applied Research K0235 flat cell with a platinum mesh counter electrode, (b) a TDK-Lambda ZUP 1203.6/U power supply, and (c) an Agilent Technologies Digital Multimeter 5 1/2-Digit Display. These three instruments are connected to a computer and controlled by a custom-developed LabVIEW code, as depicted in Figure 1. The sample was anodized at 15 °C for 1 hour in this solution with a platinum counter-electrode at a fixed voltage of 60V. After anodization, the electrode was removed from solution, rinsed with methanol, and allowed to air-dry. Electrodes synthesized in the EG-anodization solutions will be referred to as Ti/TiO₂-NT-EG in this work.

The W-anodization solution was prepared by mixing 300 mL DI water and 576.2 mg ammonium fluoride. The pH value of the W-anodization solution was increased to 3.0 by adding drops of sulfuric acid. The anodization process with W-anodization solution was carried out at 15 °C for 1 hour with a platinum counter-electrode at fixed voltages of 20V in the temperature-controlled anodization cell under dry air depicted in Figure 1. After anodization, the electrode was removed from solution, rinsed with methanol, and allowed to air-dry. Electrodes synthesized in the W-anodization solutions will be referred to as Ti/TiO₂-NT-W in this work.

Figure 1 Schematic of the custom-built anodization system used for anodization of the 3D Ti templates.

The morphology of the hierarchical flow-through Ti/TiO₂ electrode was examined by scanning electron microscopy equipped with electron dispersive spectroscopy (SEM-EDS). SEM images and EDS spectra were obtained using the MIRA3 field emission SEM (TESCAN, PA) equipped with Bruker x-flash energy-dispersion microanalysis system (Bruker, MA). Imaging was used to examine the geometry of the TiO₂ NTs for all samples and for measuring diameters, lengths, and density of packing, following the procedure to calculate specific surface area described in previous work [38,39]. Measurements of BET specific surface area were also performed, but were used only as reference since the relatively large weight of the samples added uncertainty to the results.

Finally, X-ray diffraction measurements (XRD) on the samples were also attempted, but the intricate geometry hindered the resolution of the results.

2.2 Hydrogen Evolution Reaction Testing

The electrochemical measurements were performed at room temperature using a BASi Cell stand controlled by a BASi C3 potentiostat, in which the working electrodes were the different 3D Ti/TiO₂ electrodes prepared as detailed above, the counter electrode was a BASi MW-1032 platinum electrode, and the reference electrode was BASi Ag/AgCl (BASi MW-2052). All potentials reported in this work were normalized to that of the reference hydrogen electrode (RHE). Polarization curves were obtained by linear sweep voltammetry (LSV) test, which is carried out by sweeping the potential from -0.35 V to 0.05 V with a sweep rate of 5 mV s^-1 at room temperature in 0.5 M H₂SO₄. Preliminary tests on the stability of the electrodes were performed by monitoring current during HER in a 0.5 M H₂SO₄ solution during 10 hours of operation at a constant applied potential of -150 mV and at a temperature of 25 °C. Electrochemical impedance spectroscopy (EIS) tests were performed in the same configuration at an over-potential of 500 mV with frequency ranging from 1 MHz to 1 kHz by Gamry Reference 6000 (Gamry, PA).

No human, animal, plant, or subjects were involved in the study.

3. Results

3.1 Characterization of the 3D Ti/TiO₂ Electrodes

Figure 2 presents images of the morphology of one of the 3D Ti/TiO₂-NT-EG electrodes at increasing degrees of magnification: a photographic image (Figure 2a) and several images obtained by scanning SEM (Figure 2b-f). Figure 1a is the image of 3D Ti/TiO₂ NT electrode after anodization. The 3D Ti/TiO₂ NT electrode has visible hierarchical structure, with macro-porosity and micro-porosity (nanopores). The macro-porosity provides unique flow-through features, which maximize the contact area between the electrolyte and active catalyst surface (TiO₂ NT). From Figure 2b, one can observe that the grid of 3D templates is about 1-mm and the grid-layer distance is about 0.5 mm. These features can be adjusted by modifying the 3D printing parameters. Assisted by the 3D printing technique, the 3D electrode framework can be easily produced for large-scale operations via a reliably reproducible approach. The TiO₂ NT layer grafted onto the surface of 3D Ti templates via in situ anodization can be clearly visualized in Figures 2c and 2d. The 3D Ti templates are completely wrapped by the TiO₂ NT layers. Minor charging (bright areas) is also observed due to the conductivity difference between the Ti template and the TiO₂ NT layer. Figures 2e and 2f clearly depict the nanotube structure of the TiO₂ NT layer.

Figure 2 Image and SEM micrographs of 3D Ti/TiO₂-NT-EG under various degrees of magnification.

Figure 3 depicts SEM images of identical 3D Ti templates anodized with the two solutions described in this work, the W-anodization and the EG-anodization solutions. The TiO₂ NT synthesized with the two different formulations of anodization solutions on identical 3D Ti templates showed significant variation in morphology at the nano-scale. By comparing Figures 2a and 2b, it is apparent that the morphological structures of ethylene-glycol-based TiO₂ NT (TiO₂-NT-EG) and water-based TiO₂ NT (TiO₂-NT-W) present differences in uniformity of tubular structure, cylindrical characteristics, and thickness of the nanotube walls. The TiO₂-NT-EG electrodes are uniform and have very distinct cylindrical shapes when compared to the TiO₂-NT-W electrodes. Furthermore, examination of Figures 2f, 3a, and 3b clearly shows that TiO₂-NT-EG electrodes have thicker walls.

Figure 3 Morphology of (a) 3D Ti/TiO₂-NT-EG, and (b) 3D Ti/TiO₂-NT-W.

Energy dispersive X-ray spectroscopy (EDS) results confirmed the successful anodization of TiO₂ NT on the surface of 3D Ti template (Figure 4) with the two electrolyte-formulations used in this work (i.e., ethylene glycol-based and water-based anodization solutions). Figure 4a depicts the EDS pattern of the bare 3D Ti templates, in which an oxygen peak cannot be found. Electrodes fabricated via anodization with the two anodization solutions clearly present oxygen peaks around 0.5 KēV in Figures 4b and 4c, corresponding to the TiO₂-NT-EG and TiO₂-NT-W, respectively. This confirms that the nanotube structure observed via SEM on the surface of 3D Ti templates is made of TiO₂. It is worth pointing out that the intensity of oxygen peak for TiO₂-NT-EG is slightly larger than that of TiO₂-NT-W. This fact substantiates the observations via SEM that water-based anodization formulations result in thin-walled, densely packed TiO₂ nanotubes with a smaller amount of oxide material present. Furthermore, the specific surface area for the 3D TiO₂-NT-EG samples was equal to 0.88 m²/g, while the specific surface area of the 3D TiO₂-NT-W samples was equal to 1.48 m²/g. These areas, determined via the statistical procedure described in previous work and corroborated via BET measurements, seem very low from the usual values expected for catalysts. These numbers reflect the fact that 3D Ti templates are relatively heavy. Finally, the aluminum and carbon elements found in the 3D Ti/TiO₂-NT electrodes are necessary ingredients for the 3D-printing Ti ink.

Figure 4 EDS spectra of (a) bare 3D Ti templates, (b) 3D Ti/TiO₂-NT-EG, and (c) 3D Ti/TiO₂-NT-W. The y-axis has units of cos/eV.

3.2 Hydrogen Evolution Reaction (HER) Activity

Upon successful synthesis of the 3D Ti/TiO₂ NT electrodes, their catalytic activity towards HER was carefully examined in a three-electrode test system with acid media. Figure 5 depicts the electro-catalytic performance of bare 3D Ti templates, 3D Ti/TiO₂-NT-EG, and 3D Ti/TiO₂-NT-W for HER in a 0.5 M H₂SO₄ electrolyte. Figure 5a compares the polarization curves obtained via repetitive measurements for various electrode samples. The featureless polarization curve of bare 3D Ti templates demonstrates that the 3D Ti templates do not depict catalytic activity for HER, as expected due to the position of Ti on the reactive volcano plot for HER [42]. In contrast, both 3D Ti/TiO₂-NT-EG and 3D Ti/TiO₂-NT-W show high electrocatalytic activity during the tests, evidencing a sharp increase in catalytic reduction current density (Figure 4a). The onset potential of HER was measured as -45 mV vs RHE for 3D Ti/TiO₂-NT-W and -87 mV vs RHE for 3D Ti/TiO₂-NT-EG, suggesting both types of hierarchical catalysts have good catalytic activity for HER. The dependence of over-potential (η) on the logarithmic current density (j) is depicted in Figure 5b. The linear relationships of η and log (j) observed in Figure 5b, namely the Tafel slope, can be used to hypothesize about the HER mechanism taking place in the hierarchical catalysts. Tafel slopes provide an indication of the possible mechanisms that may be taking place during an electrochemical reaction. Tafel slope values of 40 and 56 mV dec^-1 were obtained for 3D Ti/TiO₂-NT-W and 3D Ti/TiO₂-NT-EG respectively, via linear regression of the above mentioned curves. When these values are compared with the Tafel slope of the state-of-the-art Pt/C catalyst (~ 30 mV dec^-1), one can suggest that the hierarchical electrodes studied in this work exhibit slightly lower HER catalytic activities than the commercial electrode, but not significantly lower.

Figure 5 (a) Polarization curves for HER on bare 3D Ti templates, 3D Ti/TiO₂-NT-EG, and 3D Ti/TiO₂-NT-W in 0.5 M H₂SO₄. (b) Tafel plots for the various samples derived from (a) obtained via linear regression shown in dotted lines.

Figure 6 compares the Tafel slopes obtained for the electro-catalysts reported in this work. The average Tafel slope obtained for the 3D Ti/TiO₂-NT-W electrodes is smaller than that of other HER catalysts reported in recent literature (Figure 6), indicating that the 3D Ti/TiO₂-NT-W catalyst is a promising candidate for industrial application.

Figure 6 Comparison of Tafel slopes of various HER catalysts in recently published articles.

The onset potential for HER and the over-potential at a current density of 10 mA cm^-2 measured for our electro-catalysts are compared with other promising catalysts reported in the literature for HER in Table 1. The average onset potential of the 3D TiO₂-NT-W catalysts is -45 mV, which confirms that this catalyst is capable of triggering HER with energy requirements close to the commercial Pt/C catalyst. In order to reach 10 mA cm^-2 current density, the average operation over-potential for the 3D TiO₂-NT-W and the 3D TiO₂-NT-EG electrodes are 92 and 215 mV, respectively. The exchange current density (j₀) for the 3D TiO₂-NT-EG was determined to be equal to 7.68 × 10^-3 mA cm^-2; and the value corresponding to 3D TiO₂-NT-W was 2.13 × 10^-4 mA cm^-2, from extrapolation of the Tafel plots. The electrochemical tests for HER performed in this work show that the 3D Ti/TiO₂ NT-W electro-catalysts are competitive with recently documented HER catalysts, and that the catalytic activity of the 3D Ti/TiO2-NT-EG electro-catalysts is not as high as that of the 3D Ti/TiO₂-NT-W catalysts. In terms of the stability of both electrocatalysts, preliminary tests show a slight decrease of less than 5% of current density during the first hour of operation, and stabilization of the current at this level during the rest of the period of continuous operation per the test described in the methods section.

Table 1 Onset potential for HER and over-potential at 10 mA cm^-2 for various catalysts.

Electrochemical impedance spectroscopy (EIS) was employed to provide further insight into the electrode/electrolyte interface and the electrochemistry at the electrode surface. Figure 6 presents the Nyquist plots of representative EIS spectra obtained for both, 3D TiO₂-NT-EG and 3D TiO₂-NT-W electrodes, at 500 mV. The inset in Figure 7 depicts the electrical equivalent circuit (EEC) used to analyse the EIS spectra. The EEC consists of Ohmic resistance (R_s) in series with a module, in which the charge transfer resistance (R_ct) is in parallel with a constant phase element (CPE). The 3D TiO₂-NT-W electrode exhibits R_ct with the value of 9.8 Ω, which is smaller than that corresponding to 3D TiO₂-NT-EG with the value of 31.4 Ω. The high electron transfer rate on the 3D Ti/TiO₂-NT-W electrodes is particularly noteworthy.

Figure 7 EIS Nyquist plots of 3D Ti/TiO₂-NT-EG and 3D Ti/TiO₂-NT-W. Inset shows the electrical equivalent circuit used to explain the EIS spectra.

4. Discussion

The morphological differences between 3D TiO₂-NT-EG and 3D TiO₂-NT-W electrodes can be explained by the competitive effects taking place during TiO₂ NT growth. Two competitive reactions, oxidation (oxide-layer formation) and dissolution (etching) of the oxide layer, take place during the anodization process used to grow the TiO₂ NT layer on the 3D templates. Oxidation rates, and therefore mass of TiO₂ grown on 3D Ti templates, are highly dependent on the conductivity of the anodization bath and the Ohmic resistance at the surface of the electrode. Limited charge transfer during the anodization process is observed due to low electrical conductivity of the ethylene-glycol-based anodization bath, i.e., a larger resistive current drop. As a result, the mass of TiO₂ on the 3D TiO₂ NT-W electrodes obtained during the same anodization time on the same 3D Ti templates was 38% of that obtained for the 3D TiO₂-NT-EG electrodes. The differences in diameter, length, and wall thickness of the TiO₂ NTs among the electrodes stem from the fact that the water-based anodization solution promotes a higher chemical etching rate for the oxides (namely, TiO₂ NT), than the organic-based solution (ethylene glycol in the present study). The balance between the rates of oxide-layer formation and re-dissolution is offset towards re-dissolution of the oxide layer, resulting in thin-walled, shorter nanotubes in the aqueous case. At the same time, TiO₂ NTs are longer, of smaller diameters and thicker walls in the case of 3D TiO₂-NT-W electrodes. The morphological differences impact the electro-catalytic activity towards HER of both types of electrodes mainly due to the following factors: (a) the specific surface area is larger for the 3D TiO₂-NT-W electrodes, i.e., larger catalytic surface; and (b) the semi-conductor layer (TiO₂ layer) is thinner for these electrodes as well, which translates into lower resistance.

The classical HER reaction mechanisms proposed for acid media (i.e., Volmer reaction, Heyrovsky reaction, and Tafel reaction) can be used as a framework to interpret the different values of Tafel slope for the 3D Ti/TiO₂-NT electrodes with HER kinetics [42,46,47,48,49]. Specifically, in acidic and aqueous electrolyte, HER is usually comprised of three steps: (i) hydrogen adsorption (Volmer reaction); and (ii) reaction between adsorbed H and H⁺ in solution to form free H₂ (Heyrovsky reaction); or (iii) reaction between two adsorbed H to form free H₂ (Tafel reaction). These steps are presented in equations (1) – (3), where M represents an active site.

Volmer step: H₃O⁺ + ē + M ↔ M - H + H₂O                       (1)

Heyrovsky step: M - H + H₃O⁺ + ē ↔ H₂ + H₂O + M         (2)

Tafel step: 2 M - H ↔ H₂ + 2 M                                           (3)

Extensive work by Conway and others has identified the adsorption of H⁺ (Volmer reaction) as the critical mechanistic step leading to hydrogen formation [42]. Titanium (Ti), nickel (Ni), iron (Fe), and other metals are believed to exhibit poor electro-catalytic activity for HER due to their inability to readily adsorb H at low potentials [42,50,51]. Studies on single-crystal TiO₂, however, have demonstrated that TiO₂ can readily adsorb inorganic ions and molecules [50]. Tafel slopes obtained for single-crystal Pt vary between 40 mV dec^-1 and 120 mV dec^-1, depending on the crystalline structure [47,48]. Theoretically, values of Tafel slopes for HER are as follows: (a) 120 mV dec^-1 for Volmer-reaction as rate-controlling step (H-adsorption is rate-limiting); (b) 40 to 120 mV dec^-1 for Heyrovsky-reaction as rate-controlling step (reaction between adsorbed H and free H⁺ is rate limiting); and (c) 30 mV dec^-1 for Tafel-reaction as rate-controlling step (reaction between two adsorbed-H is rate limiting). HER on Pt is usually considered to occur via a Volmer-Heyrovsky mechanism, with the Heyrovsky step limiting the overall rate of reaction. The values of Tafel slopes measured for the 3D Ti/TiO₂-NT-W electrodes (40 mV dec^-1) and the 3D Ti/TiO₂-NT-EG electrodes (56 mV dec^-1) electrodes suggest that these electrodes may promote a Volmer-Heyrovsky mechanism.

Since it has been well documented that Ti displays no significant catalytic activity for HER [42,52], the competitive catalytic activities towards HER presented by the electrodes in this work can be mainly attributed to the TiO₂ NT. The 3D Ti templates contribute as effective support and current collector for the active sites due to their high conductivity and very good electrical contact with the active surface itself. In previous studies, we used highly-ordered, cylindrical TiO₂ NTs on planar Ti templates to study the formation of electrical double layer (EDL) [38,39]. Furthermore, those studies showed that electro-sorption of ions, and ion-size-exclusion effects, can be enhanced via tuning NT morphology, i.e., NT diameter, density, and length [38]. The crystalline structure of TiO₂ NTs (amorphous or anatase) can be easily tuned without changing morphology via annealing after synthesis, further enhancing electrical conductivity or even photo-catalytic-activity [38,39]. Given the fact that the TiO₂ NT properties are associated with improved HER reaction rates in this work, the combination of highly tailorable TiO₂ NT active surfaces with easily constructed 3D templates opens a door for a straight-forward method to engineer novel HER catalysts.

5. Conclusions

Summarizing, a novel strategy to fabricate hierarchically, flow-through 3D Ti/TiO₂-NT electrodes for HER was presented in this work. The 3D Ti/TiO₂-NT electrodes reported in the present work take advantage of 3D printing and in-situ anodization to achieve efficient HER electro-catalysis. Competitive catalytic activity for HER of our 3D Ti/TiO₂-NT electrodes can be ascribed to the following reasons: 1) the hierarchically flow-through 3D structure enables efficient utilization of the active catalytic surface (TiO₂ NT); 2) the TiO₂ NT structure on the 3D Ti/TiO₂-NT electrodes may trigger a Volmer-Heyrovski mechanism for HER at comparable onset potentials when compared to other emerging catalysts; and 3) the ohmic-free contact between catalyst (TiO₂ NT) and the current collector (3D Ti template) provides a continuous and efficient conducting network of charge and electron transfer. Further improvements in catalytic HER activity and performance of the present electrodes can easily be pursued via simple engineering of the highly ordered and tailorable structure and morphology of TiO₂ NTs. Finally, a clear advantage of the electrodes introduced in this work is that they are inexpensive and readily scalable for industrial applications, due to their straight-forward fabrication method. In addition, the 3D catalyst fabrication approach can be employed for a variety of metals including nickel, zirconium, and aluminum among others, as well as mixtures of metals.

Acknowledgments

The authors would like to acknowledge the following individuals and institutions:

• Richard Mayes of the Oak Ridge National Laboratory
• Richard Hailstone from the Chester F. Carlson Center for Imaging Science at Rochester Institute of Technology
• Brian McIntyre from the Institute of Optics at the University of Rochester
• Kate Gleason College of Engineering at the Rochester Institute of Technology

Author Contributions

Xiang Li performed the experimental work, analysis of data and contributed significantly to the preparation of the manuscript. Yuan Xue assisted with the EIS experiments and analysis of that set of data. Ryan Dehoff printed the 3D templates used in this work. Costas Tsouris assisted with the analysis of data, preparation, and revision of the manuscript. Patricia Taboada-Serrano supervised the work, analysed the data, and prepared and revised the manuscript.

Funding

Funding for this work was provided by the Laboratory Directed Research and Development Fund of the Oak Ridge National Laboratory and the Kate Gleason Fund.

This manuscript has been co-authored by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the US Department of Energy. The United States Government retains and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce the published form of this manuscript, or allow others to do so, for United States Government purposes. The Department of Energy will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan (http://energy.gov/downloads/doe-public-access-plan).

Competing Interests

The authors have declared that no competing interests exist.